0% found this document useful (0 votes)

140 views110 pages

MEHU130 - U2 - T43 - Nefropatia Diabetica

fff

Uploaded by

Stefani Atlle

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

140 views110 pages

MEHU130 - U2 - T43 - Nefropatia Diabetica

fff

Uploaded by

Stefani Atlle

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 110

37

Complications of Diabetes Mellitus

MICHAEL BROWNLEE, LLOYD P. AIELLO, JENNIFER K. SUN, MARK E. COOPER,
EVA L. FELDMAN, JORGE PLUTZKY, AND ANDREW J.M. BOULTON

CHAPTER OUTLINE
Biochemistry and Molecular Cell Biology, 1438 Diabetic Nephropathy, 1478
Insulin Resistance Increases Fatty Acid Oxidation, Causing Diabetic Neuropathies, 1488
Mitochondrial Overproduction of ROS, 1451 Diabetic Heart Disease, 1503
Diabetic Retinopathy and Other Ocular Complications of The Diabetic Foot, 1519
Diabetes, 1463

KEY POINTS
• C hronic hyperglycemia causes retinopathy and nephropathy. The Loss of heart rate variability due to autonomic neuropathy
effects of former high HbA1c levels can persist for years of subse- increases the risk of cardiac events more than fourfold. The cor-
quent lower HbA1c values. Both chronic hyperglycemia and the nerstone of treatment is blood glucose control. Specific antide-
consequences of insulin resistance cause neuropathy, atheroscle- pressants, anticonvulsants, and the γ-aminobutyric acid (GABA)
rosis, and cardiomyopathy. In some patients, chronic unresolved analogue pregabalin are used to treat painful neuropathy.
inflammation is as great a cardiovascular disease risk factor as • Prediabetes and diabetes both substantially increase risk of
high blood pressure and cholesterol. Mitochondrial dysfunction is coronary disease, early and late myocardial infarction fatality rates,
a component of many mechanisms underlying complications. and risk of congestive heart failure, even after adjustment for other
• Proliferative diabetic retinopathy and diabetic macular edema can cardiovascular risk factors. Hyperglycemia and insulin resistance in-
both cause severe visual loss. Treatment of diabetic macular edema teract synergistically with hypertension and dyslipidemia. Optimal
with intraocular injection of vascular endothelial growth factor treatment includes intensive lowering of low-density lipoprotein
(VEGF) inhibitors—medications that inhibit vascular endothelial (LDL) cholesterol with early use of statins or other agents, blood
growth factor—reduces diabetes-induced abnormal retinal thick- pressure control, and use of glucose-lowering agents that have
ening and improves long-term visual acuity outcomes. Treatment proven efficacy to reduce cardiovascular events, including met-
with panretinal photocoagulation reduces blindness from prolifera- formin, GLP1 receptor agonists, and SGLT2 inhibitors. In general,
tive diabetic retinopathy by over 90%. Intraocular anti-VEGF therapy coronary artery bypass grafting may provide better outcomes than
can now also effectively treat proliferative diabetic retinopathy. coronary stenting in patients with diabetes. Angiotensin-convert-
• Chronic diabetic kidney disease increases the risk of cardiovas- ing enzyme (ACE) inhibitors and angiotensin II receptor blockers
cular disease and death. Optimal management includes early (ARBs) reduce post–myocardial infarction mortality risks.
control of blood pressure using renin-angiotensin-aldosterone • Diabetic foot ulcers are the major cause of nontraumatic leg am-
system blockade and other agents, control of hyperglycemia putation. Risk factors are sensory neuropathic loss of propriocep-
(glucagon-like peptide-1 [GLP1]) analogs and sodium glucose tion, motor neuropathic foot deformity, and peripheral vascular
cotransporter-2 (SGLT2) inhibitors that may be more renopro- disease. Simple clinical interventions reduce amputation by up
tective, and control of dyslipidemia. to 80%. Management involves pressure off-loading, parenteral
• Diabetic neuropathies include distal symmetric polyneuropathy, antibiotics for infection, and ensuring adequate arterial inflow.
mononeuropathies, and a variety of autonomic neuropathies.

associated with accelerated atherosclerosis affecting arteries of the

Biochemistry and Molecular Cell Biology heart, brain, and lower extremities. In coronary arteries, diabetic
Clinical Overview atherosclerosis is more diffuse, with greater inflammatory infiltrate
and larger necrotic core size.1 Diabetes increases heart failure risk
All forms of diabetes mellitus are characterized by insufficient or fourfold, and doubles both early and late post–myocardial infarc-
absolute lack of insulin secretion, insulin resistance, altered fuel tion fatality rates.2 As a consequence of these complications, dia-
metabolism, and development of diabetes-specific complica- betes is the leading cause of new blindness in people 20 to 74 years
tions in the eye, kidney, and peripheral nerve. Diabetes is also old,3 the leading cause of end-stage renal disease (ESRD), and the

1438
Descargado para Paola Gavidia Davila ([email protected]) en Antenor Orrego Private University de ClinicalKey.es por Elsevier en julio 19, 2020.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2020. Elsevier Inc. Todos los derechos reservados.
Chapter 37 Complications of Diabetes Mellitus 1439

leading cause of both neuropathy and nontraumatic lower extrem- individuals with normal glucose tolerance and higher insulin resis-
ity amputations. Life expectancy for patients with diabetic ESRD tance had twice the 8-year risk of adverse cardiovascular outcomes
is under 4 years. Neuropathy affects more than 60% of patients than those with lower insulin resistance, even after adjustment for
with diabetes. Diabetic neuropathy includes distal symmetric 11 risk factors.16,17
polyneuropathy, mononeuropathies, and a variety of autonomic Chronic, unresolved systemic inflammation is a cardiovascular
neuropathies causing cardiac arrhythmias, hypotension, erectile risk factor independent of traditional lipid and nonlipid risk fac-
dysfunction, urinary incontinence, gastroparesis, and nocturnal tors. In healthy men followed prospectively as part of the Physician’s
diarrhea. Health Study, those with the highest quartile concentrations of
Overall life expectancy is about 11 to 13 years shorter for high sensitivity C-reactive protein (hsCRP) had a 2.9-fold greater
people with type 1 diabetes mellitus (T1DM), and 7 to 10 years risk of myocardial infarction than those in the lowest quartile, a
shorter for people with type 2 diabetes mellitus (T2DM) com- magnitude of independent risk as great as hypertension and cho-
pared to people without diabetes.4–6 Accelerated cardiovascular lesterol.18 Directly reducing inﬂammation with canakinumab, an
disease is the leading cause of death in people with both T1DM interleukin 1β (IL1β) neutralizing monoclonal antibody, reduced
and T2DM diabetes. T1DM patients diagnosed before the age of cardiovascular event rates in a population receiving lipid-lowering
10 years have a 30-fold increased risk of coronary heart disease and therapy at baseline. The greatest reduction occurred among those
acute myocardial infarction occurring in their early adult years.7 achieving hsCRP concentrations below 2 mg/L19 (Fig. 37.2).
While the incidence of diabetes-related ESRD, acute myocardial Individuals with and without diabetes had similar magnitude
infarction, and amputation has declined over the past 25 years, the of benefit from canakinumab suggesting the IL1β pathway is
rates are still 10-fold higher than for the overall adult population.8 not unique to diabetes. Chronic inflammation also contributes
A strong relationship between hyperglycemia and diabetic to chronic kidney disease, mediated in part by activation of the
retinopathy and nephropathy has been demonstrated in both NLRP3 (nucleotide-binding oligomerization domain-like recep-
T1DM and T2DM large prospective clinical studies—the Dia- tors [NLR] family pyrin domain containing three) inflammasome
betes Control and Complications Trial (DCCT) and the United in renal cells. Inhibition of IL1β with canakinumab also reduced
Kingdom Prospective Diabetes Study (UKPDS), respectively.9,10 major cardiovascular event rates among high-risk patients with
There is a continuous (although not linear) relationship between chronic kidney disease.20 Recent studies suggest defective resolu-
the level of glycemia and the risk of development and progres- tion of chronic inflammation is also a general feature of most
sion of these microvascular complications (Fig. 37.1). In T1DM, diabetic complications.
hyperglycemia is similarly associated with neuropathy, while in For each of the complications of diabetes, the effects of for-
T2DM, it is more closely associated with the consequences of mer high HbA1c levels can persist for years after HbA1c values
insulin resistance/hyperinsulinemia.11,12 In contrast to the benefi-
cial effects of glycemic reduction on microvascular complications,
a meta-analysis of randomized controlled trials shows reduction in CANTOS - Cardiovascular Mortality
myocardial infarction rates but limited benefits of intensive glu- 0.20
cose-lowering treatment on all-cause mortality and deaths from Placebo
HR
1.0
(95%CI)
(referent)
P
(referent)
cardiovascular causes.13 While consequences of insulin resistance/ Canakinumab, hsCRP ≥2mg/L 0.99 (0.82-1.21) 0.95
hyperinsulinemia and chronic hyperglycemia both play a role in Canakinumab, hsCRP <2mg/L 0.69 (0.56-0.85) 0.0004
the pathobiology of diabetic cardiovascular complications, the
consequences of insulin resistance/hyperinsulinemia may play 0.15
a larger role.14,15 In the San Antonio Heart Study, for example,
Cumulative Incidence

Sustained progression 0.10

20 of retinopathy
Progression to albuminuria
>300 mg/dL
15 Progression to clinical
neuropathy 0.05
Relative risk

Progession to microalbuminuria
10 >40 mg/dL

5 0.00
0 1 2 3 4 5
Years
0 • Fig. 37.2 Cumulative incidence and hazard ratios of cardiovascular
6 7 8 9 10 11 12 mortality among CANTOS participants allocated to either placebo or
canakinumab according to whether postrandomization on-treatment
HbA1c(%)
hsCRP levels were above or below 2 mg/L. Hazard ratios are adjusted
• Fig. 37.1 Relative risks for the development of diabetic complications at for age, sex, smoking status, hypertension, diabetes, body mass index,
different mean levels of glycosylated hemoglobin (HbA1c) in patients with baseline concentration of hsCRP, and baseline concentration of LDL-C.
type 1 diabetes obtained from the Diabetes Control and Complications CI, Conﬁdence interval; HR, hazard ratio. (From Aday AW, Ridker PM. Anti-
Trial. (Adapted from Skyler J. Diabetic complications: the importance of inflammatory therapy in clinical care: the CANTOS trial and beyond. Front
glucose control. Endocrinol Metab Clin North Am. 1996;25:243–254.) Cardiovasc Med. 2018;5:62.)

of retinopathy progression (%) 24 atherosclerotic plaques and plaque rupture, nor do they develop
the degree of fibrosis seen in human hearts with late-stage diabe-
20 tes–associated heart failure. Also, since much of the mechanistic
Cumulative incidence

Conventional therapy
16
data currently available from studies of cell and mouse models
reflect the earliest stages of each complication, mechanisms domi-
12 nant in the pathogenesis of later stages of each complication are
likely different from those dominant in the early stages. Recent
8 Intensive therapy proteomic, genomic, and molecular studies of tissue from patients
4 with longer duration of diabetes have contributed to enormous
recent progress in understanding the biochemical, molecular, and
0 cellular mechanisms involved in the pathogenesis of diabetic com-
plications. Many of the multiple mechanisms discussed in this
0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
section share a common element: prolonged increases in reactive
EDIC year oxygen species (ROS) production in cells critical for the develop-
• Fig. 37.3 Cumulative incidence of further progression of retinopathy 4 ment of diabetic complications in the retina, kidney, peripheral
years after the end of the Diabetes Control and Complications Trial. The nerve, arteries, and heart. In diabetic mouse models, transgenic
median glycosylated hemoglobin level was 8.2% for the conventional ther- overexpression of mitochondrial superoxide dismutases or catalase
apy group and 7.9% for the intensive therapy group. EDIC, Epidemiology prevents retinopathy,25 nephropathy,26,27 neuropathy,28 athero-
of Diabetes, Interventions, and Complications Research Group. (Modified sclerosis,29 and cardiomyopathy.30,31
from Diabetes Control and Complications Trial/Epidemiology of Diabetes
Interventions and Complications Research Group, Lachin JM, Genuth S,
et al. Retinopathy and nephropathy in patients with type 1 diabetes four Physiologic Reactive Oxygen Species
years after a trial of intensive therapy. N Engl J Med. 2000;342:381–389.) Production Is Essential for Normal Intracellular
Signaling and Cellular Homeostasis
have been lowered (called “metabolic memory” by the DCCT In normal physiology, ROS production is coupled to metabolic
and long-term observational follow-up Epidemiology of Diabe- networks and circadian clocks, and ROS species (hydrogen
tes Interventions and Complications [EDIC] study investigators peroxide, H2O2) function as signaling molecules essential for
and “a legacy effect” by the UKPDS investigators). Data from normal cellular homeostasis.32–34 Physiologic ROS (H2O2) sig-
DCCT/EDIC studies proved effects of former intensive and con- naling is essential for normal intracellular communication, cell
ventional therapy persist for at least 18 years.21,22 In the conven- differentiation, autophagy, response to insulin and growth fac-
tional therapy group, effects of previous high HbA1c on poststudy tor stimulation, and the generation of physiologic inflammatory
retinopathy, nephropathy, and major adverse cardiovascular events responses. ROS regulate protein stability, augment protein func-
(MACE; comprised of nonfatal myocardial infarction or stroke, tion, inactivate function, alter subcellular localization, and alter
or cardiovascular death) persisted as if there had been no improve- protein-protein interactions (Fig. 37.4). In normal cardiovascu-
ment in HbA1c at all (Fig. 37.3). Atherosclerotic changes not even lar function, for example, enhanced production of H2O2 from a
present at the end of the DCCT appeared subsequently in the mitochondrial source of superoxide leads to increased flow rates
previously higher HbA1c group, followed by a twofold increase in in human coronary resistance vessels. H2O2 hyperpolarizes and
heart attacks, strokes, and cardiovascular death, even though the dilates human coronary arterioles through opening of Ca2+-acti-
HbA1c level in these patients since the end of the DCCT was iden- vated K+ channels. Catalase, a scavenger of H2O2, greatly inhibits
tical to that of the formerly intensive-control group during the this flow-induced dilation.35 Similarly, H2O2 exerts a beneficial
entire time that these arterial changes developed.23 On the other effect on vasodilator function and reduces blood pressure in
hand, the beneficial effects of previous lower HbA1c persisted in transgenic mice with endothelium-targeted nicotinamide ade-
the intensive treatment group after their HbA1c increased after the nine dinucleotide phosphate (NADPH) oxidase 4 (Nox4) over-
DCCT intervention ended, as if there had been no deterioration expression.36 In the heart, low levels of hydrogen peroxide induce
in their HbA1c. proliferation of mouse embryonic stem cells as well as neonatal
Thus T1DM patients in the DCCT with long-term exposure cardiomyocytes, and ROS induce expression of cardiac-specific
to a higher level of hyperglycemia during the study became more genes, transcription factors, and growth factors in embryonic
susceptible to damage despite subsequent lower levels of hyper- stem cells. These effects are dampened by free radical scavengers.
glycemia than when they first started the trial. In contrast, lower ROS also act as transducers of mechanical strain–induced cardio-
levels of hyperglycemia during the trial made patients more resis- vascular differentiation of embryonic stem cells.37 ROS-depen-
tant to damage from subsequent higher levels. Data from long- dent activation of integrins and subsequent induction of PI3-K/
term follow-up studies from the UKPDS confirm a similar effect Akt (phosphatidylinositol-4,5-bisphosphate 3-kinase/RACα/β
for T2DM patients associated with intensive glucose control, long serine/threonine-protein kinase) signaling are also involved in
after the cessation of randomized intervention.24 cyclic strain–mediated cardiomyogenesis.38 Localized ROS pro-
Much current knowledge about diabetic complication mecha- duction contributes to stretch-induced augmentation of cardiac
nisms comes from studies in murine models and of cultured cells. contractile activity as well.39 Physiologic excitation–contrac-
Murine models are valuable tools for defining the pathogenesis tion coupling in heart muscle may also involve ROS signaling
of diabetic complications. However, it is important to recognize as physiologic stretch rapidly activates NOX2 located on sarco-
the limits of murine models. No diabetic animal model, regard- lemma and T tubule membranes in cardiomyocytes.39 The local
less of genetic background, develops the advanced stages of ROS produced sensitizes ryanodine receptors in the sarcoplasmic
human diabetic complications, such as proliferative retinopathy, reticulum. This triggers a burst of Ca2+ sparks, thereby increasing
end-stage renal failure, significant nerve fiber loss, or complex Ca2+ signaling sensitivity in healthy cardiomyocytes. Thus ROS

ROS
SH
Cys SOH
Cys
Redox-sensitive Oxidation
target protein

Regulates Augments Inactivates Alters Alters

protein stability function function subcellular protein-protein
For example, For example, For example, localization interaction
oxidation of the redox-dependent oxidation For example, For example,
NEDD8-conjugating modification of EGFR of PTPs translocation oxidation
enzyme UBC12 after EGFR-activated of NRF2 to of NRX to
ROS production the nucleus release DVL

• Fig. 37.4 The multiple ways in which oxidant signaling can affect cellular function. Oxidation of a reactive
cysteine residue SH (thiol) to SOH (sulphenic acid) leads to an alteration in the reduction-oxidation (redox)–
sensitive target that can then have myriad effects on protein stability, activity, localization, or protein-protein
interaction. Cys, Cysteine; DVL, dishevelled; EGFR, epidermal growth factor receptor; NEDD8, neural
precursor cell expressed developmentally down regulated-8; NRF2, nuclear related factor-2; NRX, nucleo-
redoxin; PTP, protein tyrosine phosphatase; ROS, reactive oxygen species. (From Holmström KM, Finkel
T. Cellular mechanisms and physiological consequences of redox-dependent signaling; UBC12, ubiquitin
carrier protein 12. Nat Rev Mol Cell Biol. 2014;15:411–421.)

production at the right time, place, level, and duration plays a corresponding alcohols. This enzyme is cytosolic and requires
crucial role in physiologic homeostasis.40 However, ROS produc- NADPH as a cofactor. In some cell types, where glucose is con-
tion at too high a level, for too long, or at an inappropriate sub- verted to the sugar alcohol sorbitol, sorbitol is then converted to
cellular location, leads to impaired cellular function and diabetic fructose by another enzyme, sorbitol dehydrogenase, using NAD+
tissue pathology. (nicotinamide adenine dinucleotide) as a cofactor. This series
Recent data suggest the site at which ROS are generated is as of reactions, termed the polyol pathway, has been implicated in
important as the amount of ROS produced.41 ROS generated the pathogenesis of several diabetic complications. However, the
by mitochondrial complex III are released into the mitochon- amount of substrate converted to product per second (Kcat) of
drial matrix, intermembrane space, outer membrane, and cyto- human aldose reductase for glucose is 0.15 [s−1]. Kcat values for
sol. In contrast, ROS generated by mitochondrial complex I are most enzymes are between 1 and 104. Moreover, because the intra-
released into the matrix. ROS from these two sites reacted with cellular glucose concentration in capillary retinal endothelial cells
distinct subsets of proteins involved in different physiologic path- incubated in 25 mM glucose is approximately 0.15 mM,46 while
ways.42 Importantly, ROS from mitochondrial complex I caused the Km of aldose reductase for glucose reported by Bohren and
a high percentage of irreversible overoxidations of cysteine thi- coworkers is 100 mM and the Kcat/Km is 1.3 [s−1 M−1], the pre-
ols, consistent with persistent oxidative damage. They also cause dicted rate of aldose reductase reduction of glucose to sorbitol in
nuclear deoxyribonucleic acid (DNA) double-strand breaks and microvascular endothelial cells would be expected to be rather low.
oxidize mitochondrial DNA, both of which have adverse con- However, aldose reductase has high affinity and enzyme activity
sequences.43–45 In contrast, ROS from mitochondrial complex for a variety of other substrates, including several glycolytic inter-
III caused reversible oxidation of cysteine thiols, consistent with mediates such as glyceraldehyde 3-phosphate, and its degradation
their function as redox switches in critical signaling pathways. Use product, methylglyoxal.47,48 AKR1B1 also efficiently reduces oxi-
of indiscriminate antioxidants, which block both the damaging dative stress-generated lipid aldehydes such as 4-hydroxynonenal
and the physiologic effects of mitochondrial ROS, likely explain with a Km in the micromolar range (10–30 μM) as compared to a
the negative findings of clinical trials with selected antioxidant Km of 50–100 mM for glucose49 (Fig. 37.5).
compounds. In vivo studies of aldose reductase inhibition in a 5-year study
in dogs showed that diabetic neuropathy was prevented, but aldose
Mechanisms of Hyperglycemia-Induced Damage reductase inhibition failed to prevent retinopathy or capillary base-
ment membrane thickening in the retina, kidney, or muscle.50 In
Increased Aldose Reductase Substrate Conversion response to this preclinical result with neuropathy, 32 clinical tri-
Aldose reductase (AKR1B1), a member of the large aldo-keto als were completed in the 1990s and early 2000s.51 None of these
reductase superfamily, catalyzes reduction of a wide variety of trials showed clinical efficacy. In diabetic apolipoprotein E (ApoE)
hydrophobic and hydrophilic carbonyl-containing compounds— knockout mice, overexpression of human aldose reductase acceler-
including glucose and several glycolytic intermediates—to their ated atherosclerosis, and pharmacologic inhibition of the enzyme

ROS

Toxic Aldose The Glyoxylase

Inactive alcohols Glucose
aldehydes reductase System
SDH
Increased glucose Sorbitol Fructose
Glc-6-P
NAD+ NADH

Fru-6-P GA3P

NADPH NADP+
Fru-1,6-bP Methylglyoxal

Glutathione reductase DHAP

GSSG GSH
D-Lactate
• Fig. 37.5 Aldose reductase (AKR1B1) catalyzes reduction of the ROS-
generated and the polyol pathway. Aldose reductase reduces reactive GSH
oxygen species (ROS)–generated toxic aldehydes such as lipid-derived
4-hydroxynonenal to inactive alcohols, and in some cell types, glucose to
sorbitol, using triphosphopyridine nucleotide (NADPH), the reduced form Glyoxalase II Glyoxalase I
of nicotinamide adenine dinucleotide phosphate (NADP+), as a cofactor.
It also reduces glyceraldehyde 3-phosphate and its reactive degradation
product methylglyoxal. GSH, Reduced glutathione; GSSG, oxidized gluta- S-D-Lactoyl-
thione; SDH, sorbitol dehydrogenase. glutathione

prevented this.52 In contrast, in mice expressing physiologic levels

of aldose reductase, knockout or pharmacologic inhibition of the • Fig. 37.6 Intracellular methylglyoxal, the major advanced glycation end
enzyme unexpectedly caused increased early atherosclerotic lesion product (AGE) precursor, is detoxified by the glyoxalase system. The
enzyme glyoxalase I, together with glyoxalase II and a catalytic amount
size in both diabetic and nondiabetic mice.53 Differences in total
of glutathione, reduces this highly reactive α-oxoaldehyde to D-lactate.
enzyme activity, cofactor levels, and levels of alternative intracellu- DHAP, Dihydroxyacetone phosphate; GA3P, glyceraldehyde 3-phosphate;
lar substrates, coupled with known differences in enzyme kinetics GSH, glutathione; Glc-6-P, glucose-6-phosphate; Fru-6-P, fructose-
for different substrates, likely explain these seemingly paradoxic 6-phosphate; Fru-1,6-bP, fructose-1,6-bisphosphate. (From Schmoch T,
observations. Uhle F, Siegler BH, et al. The glyoxalase system and methylglyoxal-derived
carbonyl stress in sepsis: glycotoxic aspects of sepsis pathophysiology.
Increased Intracellular Formation of the Major Advanced Int J Mol Sci. 2017;18.)
Glycation End Products—Precursor Methylglyoxal
The post-translational modifications of proteins called advanced high glucose concentrations in cell culture and in diabetic ani-
glycation end products (AGEs) are formed by glucose-derived mals. This hyperglycemia-induced overexpression is mediated by
dicarbonyls reacting with amino groups of unprotonated lysine ROS-induced increases in methylglyoxal, which increase binding
and arginine residues of proteins. Methylglyoxal, formed by the of the transcription factors nuclear factor-κB (NFκB) and activa-
nonenzymatic fragmentation of the glycolytic intermediate triose tor protein 1 (AP1) to the promoters of RAGE and of RAGE
phosphate, accounts for the majority of hyperglycemia-induced ligands, respectively.56
increase in AGE adducts in diabetic tissues.54 Intracellular meth- Diabetes increases levels of the major methylglyoxal-derived
ylglyoxal is detoxified by the glyoxalase system.55 The enzyme gly- adduct in retina, renal glomerulus, and sciatic nerve of rats.58,59 In
oxalase I, together with glyoxalase II and a catalytic amount of the retina, diabetic mice with knockouts of four transient receptor
glutathione, reduces this highly reactive α-oxoaldehyde to D-lac- potential cation channels (Trpc1/4/5/6 mice) have increased lev-
tate (Fig. 37.6). In cells, methylglyoxal reacts with unprotonated els of glyoxalase 1 protein and activity, which protects them from
arginine residues to form the major methylglyoxal-derived epitope the initial phenotype of diabetic retinopathy, pericyte dropout,
MG-H1 (methylglyoxal hydroimidazolone 1). Intracellular pro- and acellular capillary formation.60 In the kidneys of nondiabetic
duction of AGE precursors damages target cells by three general mice, knockdown of glyoxalase I (GLO1) increases to diabetic lev-
mechanisms (Fig. 37.7). First, AGE modification of intracellu- els both methylglyoxal modiﬁcation of glomerular proteins and
lar proteins changes their function. Second, AGE modification oxidative stress, causing alterations in kidney morphology indis-
of extracellular matrix components alters their interaction with tinguishable from those caused by diabetes, while in kidneys of
other matrix components and with integrin matrix receptors. diabetic mice, GLO1 overexpression completely prevents diabe-
Third, intracellular methylglyoxal increases expression of both tes-induced increases in methylgloxal modiﬁcation of glomeru-
the pattern recognition receptor for AGEs (RAGEs) and its major lar proteins, increased oxidative stress, and the development of
endogenous ligands, the proinflammatory S100 calgranulins.56 diabetic kidney pathology, despite unchanged levels of diabetic
Ligation of these ligands with RAGE causes cooperative interac- hyperglycemia.61
tion with the innate immune system signaling molecule toll-like Increased levels of methylglyoxal may contribute to pain-
receptor 4 (TLR4).57 Expression of RAGE, S100A8, S100A12, ful diabetic neuropathy by altering the function of the voltage-
and high mobility group box 1 (HMGB1) are all increased by gated sodium channel Nav1.8. Nav1.8 is expressed exclusively

Intracellular Protein Glycation

Glucose AGE Precursors Glucose

RAGE

Intracellular Proteins ROS

Hormones NFκB
Cytokines

Matrix

Transcription Factors

DNA

Transcription Macrophage
Integrins RNA

RAGE
Ligands

Endothelial Cell
• Fig. 37.7 Potential mechanisms by which intracellular production of advanced glycation end product
(AGE) precursors damage vascular cells. First, intracellular protein modification alters protein function. Sec-
ond, extracellular matrix modified by AGE precursors has abnormal functional properties. Third, intracel-
lular methylglyoxal increases expression of both the receptor for AGEs (RAGE) and its major endogenous
ligands. Ligand binding to RAGE induces receptor-mediated production of deleterious gene products such
as cytokines. NFκB, Nuclear factor-κB; ROS, reactive oxygen species. (Adapted from Brownlee M. Lilly
lecture 1993: glycation and diabetic complications. Diabetes. 1994;43:836–841.)

in unmyelinated, small-diameter sensory neurons called C This coordinated process of calcium cycling is critical for effi-
fibers. Post-translational modification of Nav1.8 by methylgly- cient cardiac contractions, and diabetes-induced defects caused
oxal depolarizes sensory neurons, facilitating firing of these pain by increased methylglyoxal adduct formation likely contribute to
pathway neurons. In streptozotocin-induced and genetic mouse impaired systolic function.
models of diabetes but not in Nav1.8 knockout (Scn10−/−) dia- Increased methylglyoxal production also seems to be respon-
betic mice,62 post-translation modification of Nav1.8 by meth- sible for poor cardiac stem cell–mediated repair and angiogenic
ylglyoxal increased electrical excitability and facilitated firing of capacity.66 Cardiac stem cells from biopsies of hearts from
pain pathway neurons, which in turn facilitated neurosecretion human diabetics were less able to repair postinfarction damage in
of calcitonin gene-related peptide, increasing cyclooxygenase-2 immunodeficient mice than cardiac stem cells from nondiabetic
(COX2) expression. Methylglyoxal treatment also evoked thermal patients, and conditioned medium from these cells had less angio-
and mechanical hyperalgesia, reflected by increased blood flow in genic capacity. Culture of nondiabetic murine cardiac stem cells
brain regions that are involved in pain processing. Plasma meth- in high glucose induced the same cardiac repair and angiogenesis
ylglyoxal concentrations above 600 nM discriminated between defects seen in human diabetic cells. In both human and mouse
diabetes-affected individuals with painful neuropathy and those cells, overexpression of GLO1 restored the angiogenic defects.66
without pain. In diabetic mice with defective postischemia hindlimb revascular-
Increased methylglyoxal is also an important element in the ization, overexpression of the methylglyoxal-metabolizing enzyme
pathogenesis of both diabetic atherosclerosis and diabetic car- GLO1 exclusively in bone marrow cells was sufficient to restore
diomyopathy. In nondiabetic ApoE null mice, increasing plasma bone marrow cell function and neovascularization of ischemic tis-
methylglyoxal levels to diabetic levels using a glyoxalase-1 inhibi- sue in diabetes mellitus.67
tor caused endothelial inflammation and atherogenesis similar Increased methylglyoxal also activates the unfolded protein
to that induced by diabetes mellitus.63 In human atherosclerotic response in cardiomyocytes.68 Although transient activation of the
plaques, MG-H1 (methylglyoxal hydroimidazolone 1) levels were unfolded protein response relieves endoplasmic reticulum (ER)
associated with rupture-prone plaques having increased levels of stress, prolonged activation of the unfolded protein response in
the inflammatory mediators IL8 and monocyte chemotactic pro- cardiovascular disease triggers apoptosis, mediated by the down-
tein-1 (MCP1), and higher matrix metalloproteinase-9 (MMP9) stream effector C/EBP-homologous protein (CHOP). CHOP
activity. MG-H1 was primarily found in lesion macrophages sur- plays a critical role in macrophage apoptosis, a process involved in
rounding the necrotic core, and co-localized with cleaved cas- plaque necrosis in advanced atheromata. In Chop−/−Apoe−/− mice,
pase-3.64 In the diabetic heart, methylglyoxal preferentially reacts lesion area plaque necrosis was reduced by 50%. In high fat–fed
with both ryanodine receptor 2, the major myocardial intracellu- ApoE−/− and low-density lipoprotein (LDL) receptor−/− mice,
lar mediator of calcium-induced calcium release, and with sarco- CHOP promoted plaque growth, apoptosis, and plaque necro-
plasmic reticulum Ca2+ ATPase (SERCA2a), which is responsible sis.69 In cardiomyocytes, methylglyoxal also induces apoptosis via
for the synchronized reuptake of released intracellular calcium.65 CHOP.68 Infusion of methylglyoxal in nondiabetic mice induced

Intracellular glucose

DAG

PKC
(α,β,δ, and θ isoforms)

eNOS ET1 VEGF TGFβ PAI1 NFκB NAD(P)H NFAT

oxidases
Collagen
Fibronectin
Fibrinolysis ROS

Blood-flow Vascular Capillary Vascular Proinflammatory Multiple T-cell

abnormalities permeability occlusion occlusion gene expression effects activation
Angiogenesis

• Fig. 37.8 Potential consequences of hyperglycemia-induced protein kinase C (PKC) activation. Hyper-
glycemia increases diacylglycerol (DAG) content, which activates several PKC isoforms. Activated PKC
has a number of pathogenic consequences. eNOS, Endothelial nitric oxide synthase; ET1, endothelin 1;
NAD(P)H, nicotinamide adenine dinucleotide phosphate; NFAT, nuclear factor of activated T cells; NFκB,
nuclear factor-κB; PAI, plasminogen activator inhibitor; ROS, reactive oxygen species; TGF, transforming
growth factor; VEGF, vascular endothelial growth factor. (Adapted from Koya D, Jirousek MR, Lin YW,
et al. Characterization of protein kinase C beta isoform activation on the gene expression of transforming
growth factor-beta, extracellular matrix components, and prostanoids in the glomeruli of diabetic rats. J
Clin Invest. 1997;100:115–126.)

cardiomyocyte apoptosis, inflammation, and a significant reduc- In early experimental diabetes, activation of PKCβ isoforms
tion in left ventricular fractional shortening and left ventricular mediates retinal and renal blood flow abnormalities,78 perhaps by
ejection fraction. Each of these adverse effects was prevented in depressing nitric oxide (NO) production and increasing endo-
CHOP−/− mice. In the hearts of diabetic mice, overexpression of thelin-1 activity. In the diabetic retina, hyperglycemia persis-
GLO1 in the vasculature70 prevented diabetes mellitus–induced tently activates PKC and p38a mitogen-activated protein kinase
reduction of myocardial capillary density, increased apopto- (MAPK) to increase expression of a previously unknown target
sis, and loss of cardiac function. Neuregulin production, which of PKC signaling, Src homology-2 domain-containing phospha-
transduces signals between the heart’s microvasculature and car- tase-1 (SHP1), a protein tyrosine phosphatase. This signaling cas-
diomyocytes,71 endothelial nitric oxide synthase (eNOS) dimer- cade leads to platelet-derived growth factor (PDGF) receptor-β
ization, and expression of the antiapoptotic protein BCL-2 were dephosphorylation and a reduction in downstream signaling from
also maintained in the diabetic GLO1 transgenic hearts. this receptor, resulting in pericyte apoptosis.79 Abnormal activa-
tion of PKC also has been implicated in the decreased glomerular
Activation of Protein Kinase C β, δ, and θ production of NO induced by experimental diabetes80 and in the
Protein kinase C (PKC) is a family of protein kinase enzymes decreased smooth muscle cell NO production induced by hyper-
with 15 isoforms that are involved in the regulation of protein glycemia.81 PKC activation also mediates glucose-enhanced extra-
function. Nine of these 15 PKC isoforms are activated by a lipid cellular matrix accumulation in rat glomerular mesangial cells.82
second messenger, diacylglycerol. Elevated intracellular glucose Hyperglycemia increases endothelin 1–stimulated MAPK activ-
increases diacylglycerol levels in a variety of diabetic target tissues, ity in glomerular mesangial cells by activating PKC isoforms.83
including arterial smooth muscle cells and cardiomyocytes, by de The increased endothelial cell permeability induced by high glu-
novo synthesis.72–74 Hyperglycemia primarily activates the β and δ cose concentrations in cultured cells is mediated by activation of
isoforms of PKC, but increases in activity of several other isoforms PKCα.84 Activation of PKC by elevated glucose levels also induces
have also been found (Fig. 37.8). These PKC isoforms can also expression of the permeability-enhancing vascular endothelial
be activated by intracellular ROS in the absence of diacylglycerol growth factor (VEGF) in vascular smooth muscle cells.85 In addi-
or Ca2+. The regulatory domain of these PKC isoforms contains tion to affecting hyperglycemia-induced abnormalities of blood
two pairs of zinc fingers with six cysteine residues and two zinc flow and permeability, activation of PKC contributes to increased
atoms, which can be oxidized by intracellular ROS. Oxidation matrix protein accumulation, inducing expression of transforming
alters zinc-finger conformation and activates PKC.75 Intracellular growth factor beta-1 (TGFβ1), fibronectin, and α1 type IV col-
hyperglycemia activates PKC in retina and glomeruli of diabetic lagen in glomeruli of diabetic rats.80 Many cellular abnormalities
animals.72,76 In contrast to other complication-prone tissues, total involved in diabetic cardiovascular disease have also been linked to
PKC activity is reduced in the peripheral nerve of diabetic ani- PKC activation. These include endothelial dysfunction, increased
mals. Schwann cell PKCα activity is reduced, while vascular wall vascular permeability, impaired angiogenesis, and increased apop-
PKCβII activity is increased.77 tosis. Molecular mechanisms affected by diabetes-induced PKC

Glycolytic Pathway

Glucose Glucose Glc-6-P Fruc-6-P

Gln
GFAT

Glu
Glucosamine-6-P

UDP-GlcNAc

OH O – GlcNAc
OGT
Protein Protein
OGA

• Fig. 37.9 Schematic representation of the hexosamine biosynthetic pathway (HBP). The glycolytic inter-
mediate fructose 6-phosphate (Fruc-6-P) is converted to glucosamine 6-phosphate (GlcN-6-P) by the
enzyme glutamine:fructose-6-phosphate amidotransferase (GFAT). Increased donation of N-acetylglucos-
amine moieties from UDP-GlcNAc to serine and threonine residues of proteins by O-GlcNAc transferase
(OGT) and their removal by O-GlcNAcase (OGA) affect proteins that regulate gene expression, translation,
protein degradation, signal transduction, protein localization, epigenetics, and mitochondrial bioenerget-
ics. Glc, Glucose, GlcNAc, N-acetylglucosamine; OGT, O-linked N-acetylglucosamine (GlcNAc) transfer-
ase; UDP, uridine diphosphate. (Modified from Slawson C, Copeland RJ, Hart GW. O-GlcNAc signaling: a
metabolic link between diabetes and cancer? Trends Biochem Sci. 2010;35:547–555.)

activation include alterations in functionally significant enzymatic (Rbfox2).91 Chronic activation of PKC isozymes α, β, and δ pro-
activities, such as mitogen-activated protein kinase, cytosolic motes diastolic and systolic dysfunction, fibrosis, cardiomyocyte
phospholipase A2, and Na+-K+-ATPase, and alterations in several hypertrophy, and apoptosis92 Another PKC isoform, PKCθ, plays
transcription factors.86 crucial roles in the proliferation, differentiation, and activation
Hyperglycemia-induced activation of PKCβ promotes vascular of mature T cells via activation of several transcription factors
inflammation and acceleration of atherosclerosis in diabetic ApoE in the nuclei of T cells, including nuclear factor of activated T
null mice by augmenting expression of inflammatory mediators. cells (NFAT), c-Jun, c-Fos, and AP1. Diabetes-induced cardiac
In addition, it increases macrophage expression of cluster of dif- interstitial fibrosis, reduced contractility, reduced expression of the
ferentiation 11c (CD11c), chemokines (C-C motif ligand 2), and tight junction maintaining protein ZO1, and T-cell infiltration
interleukin-1β, via increased extracellular signal-regulated kinase were all improved by treatment with an isoform-specific PKCθ
1 and 2 (ERK1 and ERK2), and c-Jun-N terminus kinase (JNK)– inhibitor.93
mitogen-activated kinase.87 In this same diabetic model, PKCβ
activation increased transcription of the proinflammatory cyto- Increased Protein Modification by O-GlcNAc
kine IL18 and inhibited transcription of IL18-binding protein The hexosamine pathway causes reversible post-translational
in the aorta. Diabetic mice showed increased plaque formation, modification of intracellular protein serine and threonine resi-
cholesteryl ester content, and macrophage infiltration. Treatment dues by N-acetylglucosamine. O-GlcNAc modifies proteins that
with a PKCβ inhibitor prevented these.88 PKCβ2 in endothe- regulate gene expression, translation, protein degradation, signal
lial cells from transgenic ApoE null mice overexpressing PKCβ transduction, protein localization, epigenetics, and mitochon-
decreased insulin- stimulated Akt/eNOS (endothelial nitric oxide drial bioenergetics.94 In cells damaged by hyperglycemia, excess
synthase) activation and increased basal and angiotensin-induced intracellular glucose provides increased fructose-6-phosphate,
expression of the vasoconstrictor endothelin-1. These dual effects which is converted to glucosamine-6-phosphate by the rate-lim-
increased endothelial dysfunction and accelerated atherosclerosis iting enzyme glutamine:fructose-6-phosphate amidotransferase.
in this model compared with ApoE−/− mice.89 Glucosamine-6-phosphate is further converted to N-acetyl-
PKC activity has been linked to myocardial dysfunction, caus- glucosamine-6-phosphate (GlcNAc-6-P) and finally to UDP-
ing cardiomyopathy and cardiac failure. Selective overexpression N-acetylglucosamine (UDP-GlcNAc). The enzyme O-GlcNAc
of PKCβ2 in the myocardium of diabetic mice increased expres- transferase (OGT) uses UDP-GlcNAc to transfer N-acetylglu-
sion of connective tissue growth factor and transforming growth cosamine to a variety of proteins, resulting in increased pro-
factor-β1, cardiomyopathy, and cardiac fibrosis.90 More recently, tein modification by N-acetylglucosamine. Another enzyme,
activation of PKCα/β in diabetic hearts has been shown to medi- N-acetylglucosaminidase (O-GlcNAcase [OGA]), removes this
ate reactivation of fetal splicing programs by phosphorylation protein modification (Fig. 37.9). Alternative splicing of the
and upregulation of the RNA-binding proteins CUGBP elav-like genes encoding the O-linked GlcNAc cycling enzymes OGT and
family member 1 (CELF1) and RNA binding fox-1 homolog 2 OGA yields isoforms targeted to discrete sites in the nucleus,

cytoplasm, and mitochondria. O-GlcNAc serves as a nutrient/ both cytoplasm and nuclear compartments of endothelial cells
stress sensor regulating cellular homeostasis by altering signal- compared with nondiabetic subjects.102 Increased O-GlcNAcyl-
ing, transcription, metabolism, organelle biogenesis, cytoskeletal ation may also contribute to diabetes-accelerated atherosclerosis
dynamics, and apoptosis.95,96 Aberrant O-GlcNAc processing by increasing ubiquitination and proteasomal degradation of the
reduced mitochondrial protein expression and respiration, mito- anti-inflammatory NFκB inhibitory protein A20 in coronary
chondrial proteins involved in the respiratory chain and the tri- endothelial and smooth muscle cells.103
carboxylic acid (TCA) cycle were decreased, and mitochondrial Chronically elevated O-GlcNAc levels also adversely affect
morphology was altered.97 myocardial function. Cardiac effects of increased O-GlcNAcyla-
In mouse coronary endothelial cells isolated from diabetic tion include decreased mitochondrial function, decreased autoph-
mice, O-GlcNAcase protein expression was significantly decreased agic signaling, and decreased contractile function.104 Ventricular
compared with coronary endothelial cells from control mice. contraction and relaxation are controlled mainly by release and
In contrast, OGT protein expression was markedly increased.98 uptake of Ca2+ by the SERCA2 pump. In hypertrophied and fail-
The resultant increased protein modification by O-GlcNAc was ing myocardium, SERCA2 protein level and its Ca2+ uptake func-
responsible for decreased endothelium-dependent relaxation of tion are depressed. Overexpression of OGT significantly reduced
the coronary arteries and reduced capillary density in the left transcription of SERCA2, causing decreased calcium reuptake
ventricle. Both of these defects were restored by overexpression and impaired diastolic relaxation.105 High glucose also increased
of O-GlcNAcase. Decreased endothelium-dependent relaxation O-GlcNAc modification of the calcium2+/calmodulin-dependent
of the coronary arteries and reduced capillary density both reflect protein kinase IIδ (CaMKIIδ), an enzyme critical for Ca2+ homeo-
inhibition of eNOS, which is required for endothelium-depen- stasis and reuptake in cardiomyocytes. O-GlcNAc-modified CaM-
dent arterial relaxation and for mobilization of stem and progeni- KII at serine 279 is increased in the heart of diabetic humans and
tor cells from the bone marrow compartment.99 In human arterial rats,106 causing autonomous activation of CaMKII. Thus CaM-
endothelial cells, activation of eNOS by phosphorylation at serine KII remains activated even after intracellular Ca2+ declines. This
1177 is inhibited directly by hyperglycemia-induced O-GlcNAc- contributes to decreased cardiac contractility and potentially fatal
ylation at this site.100 eNOS activity is also affected by several arrhythmias, such as premature ventricular complexes and delayed
other post-translational modifications, but the effect of diabetes afterdepolarizations (Fig. 37.10). Delayed afterdepolarizations
on these has not yet been determined.101 Carotid plaques from are associated with the initiation of long QT-interval arrhyth-
diabetic patients have a marked increase of O-GlcNAcylation in mias such as torsade de pointes. Overexpression of GlcNAcase or

Glucose Glc-6-P Frc-6-P

GlcN-6-P
SERCA2A transcription

UDP-GlcNAc
CaMKII
CaMKII
O-GlcNAc Transcription
O-GlcNAc
factors
RyR
Ca2+ Ca2+ OGT O-GlcNAc

Ca2+
Ca2+ Ca2+
Ca2+ Ca2+ Ca2+ PLB

Ca2+ SERCA2A

Ca2+ Delayed afterdepolarizations

• Fig. 37.10 Hyperglycemia-induced myocardial protein modification by β-linked N-acetylglucosamine

(O-GlcNAc) causes increased intracellular Ca2+ and delayed afterpolarizations. Increased O-GlcNAc modifi-
cation of calcium/calmodulin-dependent protein kinase IIδ (CaMKII) causes autonomous CaMKII activation.
CaMKII increases intracellular Ca2+ by phosphorylating ryanodine receptor 2 (RyR). O-linked N-acetylglu-
cosamine (GlcNAc) transferase (OGT) also modifies transcription complex factors regulating expression
of sarcoplasmic reticulum Ca2+-ATPase (SERCA2a), reducing SERCA2A expression and contributing to
increased intracellular Ca2+. Increased O-GlcNAc modification of these proteins causes delayed afterde-
polarizations in cardiomyocytes. PLB, Phospholamban. (Redrawn from Shah M, Brownlee M. Molecular
and cellular mechanisms of cardiovascular disorders in diabetes. Circ Res. 2016;118:1808–1829.)

inhibition of GlcNAc modification increased expression of SER- Expression of sEH was increased in diabetic mouse retinas and
CA2a, ablated sarcoplasmic reticulum Ca2+ leak, improved car- in retinas and vitreous humor of patients with diabetes.111 As a
diac contractility, and reduced arrhythmic events. Increased levels consequence, levels of the sEH diol product 19,20-dihydroxydo-
of ROS also cause autonomous activation of CaMKII by oxida- cosapentaenoic acid (19,20-DHDP) were increased. Nonprolif-
tion of adjacent methionine residues in its regulatory domain.107 erative diabetic retinopathy is characterized by pericyte loss and
Activation of this mitochondrial ROS–oxidized CaMKII pathway endothelial cell dysfunction. 19,20-DHDP prevented the associa-
increased mortality after myocardial infarction in diabetic mouse tion of presenilin 1 with N-cadherin and VE-cadherin in retinal
models.108 Mitochondrial OGT is increased in diabetic cardiac vessels, compromising pericyte-endothelial cell interactions and
mitochondria, whereas O-GlcNAcase (OGA) is reduced, causing interendothelial cell junctions.111 Inhibition of sEH prevented
increased O-GlcNAcylation of cardiac mitochondrial proteins. the early nonproliferative stage of diabetic retinopathy in animal
Inhibition of OGA and the resulting increased mitochondrial models. Similarly, sEH in glomerular podocytes is a significant
protein modification by O-GlcNAc increases oxygen consump- contributor to hyperglycemia-induced renal injury. Podocyte
tion and reduces reserve capacity.109 Reduced bioenergetic reserve sEH deficiency was associated with attenuated hyperglycemia-
capacity makes cells more sensitive to stress and cell death. induced renal endoplasmic reticulum stress, inflammation and
fibrosis, and enhanced autophagy. These effects were recapitulated
Increased Soluble Epoxide Hydrolase in immortalized murine podocytes treated with a selective sEH
Soluble epoxide hydrolase (sEH) is the dominant member of pharmacologic inhibitor.112 EETs and other epoxides also reduce
the epoxide hydrolase family in humans. It binds to specific pain caused by diabetic peripheral neuropathy.113
epoxides such as epoxyeicosatrienoic acids (EETs) and rap- In the arterial wall, anti-inﬂammatory actions of EETs include
idly converts them to less active or inactive dihydrodiols, the attenuation of cytokine-induced endothelial activation and leu-
dihydroxy-eicosatrienoic acids (diHETrEs). EETs are signaling kocyte adhesion, prevention of endothelium-dependent vascu-
molecules formed from arachidonic acid by cytochrome P450 lar remodeling, and improvement of vascular inﬂammation and
enzymes such as CYP2J2. In cell and animal models EETs have endothelial dysfunction. EETs also inhibit platelet aggregation
major anti-inflammatory activity.110 They reduce activation of and promote fibrinolysis, and may inhibit vascular smooth muscle
NFκB, causing transcriptional downregulation of the proinflam- cell proliferation. In human studies, two single nucleotide poly-
matory enzymes inducible nitric oxide synthase (iNOS), lipoxy- morphisms in the sEH coding region of the EPHX2 gene are asso-
genase 5 (LOX5), cyclooxygenase 2 (COX2), and of several ciated with increased risk of developing coronary artery disease in
pro-inflammatory cytokines (Fig. 37.11). EETs also prevent the different populations. Caucasians homozygous for the nonsynony-
signaling cascade activated by the three major unfolded protein mous K55R allele, with higher apparent soluble epoxide hydrolase
response/ER stress pathway sensors: inositol requiring protein activity in vivo, have a 3.5-fold increased risk of developing coro-
1α (IRE1α), protein kinase RNA-like endoplasmic reticulum nary heart disease.114 Low levels of EET and high levels of soluble
kinase (PERK), and activating transcription factor 6 (ATF6). epoxide hydrolase are also associated with cardiac hypertrophy.

Inflammatory Free fatty acid pool

PLA2s
Prostaglandins COX O O
Prostacyclins OH ATP CoA
Thromboxanes LA, ARA, EPA, DHA…
LOX
Leukotrienes s
Lipoxins 450
HETEs
C YP

Anti-inflammatory
20-HETE
Epoxy-fatty acids
O O PLA2s
Dihydroxy-fatty acids sEH
OH
HO OH O EpOMEs, EETs, ATP CoA
OH EpETEs, EpDPEs…
DiHOMEs, DHETs,
DiHETEs, DiHDPEs
NFκB action UPR/ER stress

• Fig. 37.11 Overview of the three proinflammatory branches of the arachidonic acid cascade generated
by cyclooxygenase (COX), lipoxygenase (LOX), and cytochrome P450 (CYP450s). Opposing these are the
anti-inflammatory epoxy fatty acids generated by P450 enzymes such as CYP2J2. These anti-inﬂammatory
epoxides are reduced to their corresponding 1,2-diols by the enzyme soluble epoxide hydrolase. ARA,
Arachidonic acid; DHA, docosahexaenoic acid; EETs, epoxyeicosatrienoic acids; EPA, eicosapentaenoic
acid; EpDPEs, DHA epoxygenase metabolites; EpETEs, epoxyeicosatetraenoic acids; EpOMEs, cis epoxy
fatty acids; ER, endoplasmic reticulum; HETE, hydroxyeicosatetraenoic acid; LA, linoleic acid; NFκB, nuclear
factor kappa-light-chain-enhancer of activated B cells; PLA2, phospholipase A2; sEH, soluble epoxide
hydrolase, UPR, unfolded protein response. (Modified from Morisseau C, Hammock BD. Impact of soluble
epoxide hydrolase and epoxyeicosanoids on human health. Annu Rev Pharmacol Toxicol. 2013;53:37–58.)

Increasing EETs by inhibition of soluble epoxide hydrolase lowers overproduction of superoxide.43,131 This can, in turn, activate other
fetal gene markers of cardiac hypertrophy, reduces heart size, and superoxide production pathways that may amplify the original dam-
decreases cardiac fibrosis in animal models.115–118 EETs appear to aging effect of hyperglycemia. Examples of amplification mecha-
act in part by reducing inflammation in cardiomyocytes. nisms include increased mitochondrial fission mediated by the
Upregulation of sEH decreases levels of the arachidonic rho-associated protein kinase 1 (ROCK1),132,133 and ROS-medi-
acid-derived inflammation stop signal lipoxin A4, while inhibi- ated uncoupling of eNOS dimers to superoxide-producing eNOS
tion of sEH increased lipoxin A4 by increasing its EET precur- monomers in endothelial cells.134 Although NADPH oxidase 4
sor.119 Lipoxins are one of four classes of proresolving mediators (Nox4) has been hypothesized to be a direct source of increased
of inflammation.120,121 Lipoxin A4 is a polymorphonuclear cell ROS in the diabetic kidney,135,136 it appears to act indirectly by
(PMN) stop signal that limits further recruitment. It also stim- increasing mitochondrial ROS production. Nox4 is constitutively
ulates macrophage efferocytosis, the phagocytosis of apoptotic active and associates with mitochondria, inhibiting mitochondrial
PMNs and debris. A failure to resolve arterial inflammation, which biogenesis and function.137 The resultant decrease in maximal res-
normally involves the suppression of inflammatory cell influx, piration suggests oxidative damage to enzymes of the tricarboxylic
effective clearance of apoptotic cells, and promotion of inflamma- acid (TCA) cycle, the electron transport chain (ETC), or inner
tory cell egress, promotes the progression of atherosclerotic lesions mitochondrial membrane lipids has occurred.138,139 Knockout of
into unstable plaques, which can trigger atherothrombotic vascu- Nox4 increases mitochondrial biogenesis and maximal respira-
lar events.122,123 tory capacity dramatically, which would prevent substrate-driven
increased mitochondrial ROS production.137 Mitochondrial ROS
Reduced Pyruvate Kinase M2 Activity cause activation of other NOX isoforms.140
The less active isoform of the glycolytic enzyme, pyruvate kinase The initiating role of mitochondrial ROS is suggested by the
M2 (PKM2), is only active as a homotetramer. Its activity can be observation that cells lacking mitochondrial electron transport
further inhibited by tyrosine kinases, acetylation, O-GlycNAcyla- chain function (ρ0 cells)43 fail to increase ROS production in
tion,124 and oxidation of cysteine residue 358.125 Although PKM2 response to high glucose.
is expressed in normal differentiated, nonproliferating cells, the Dugan and associates141 have suggested diabetic nephropa-
increased expression of this less active isoform in most tumors has thy is caused by reduced, rather than increased, mitochondrial
focused attention on the potential of PKM2 to divert glycolytic ROS production, based on the observation that AMPK activity
intermediates toward biosynthetic pathways in tumor cells. Inde- is decreased in the diabetic kidney. Protein levels of the master
pendent of its enzymatic activity, PKM2 can also regulate aerobic regulator of mitochondrial biogenesis and function, PGC1α, and
glycolysis by acting as a dual-specificity protein kinase, increasing mitochondrial density were also decreased. These authors pro-
transcription of itself, the glucose transporter GLUT1, and lactate posed a model in which these observations reflect a feed-forward
dehydrogenase A (LDLA).126–128 In postmortem kidneys from cycle initiated and maintained by decreased mitochondrial ROS.
people with T1DM for more than 50 years, an unbiased proteomic However, in human endothelial cells, silencing of AMPK causes
analysis identified expression of glomerular PKM2 as a critical increased, not decreased, ROS.142 An alternate model of a feed-
determinant of susceptibility or resistance to diabetic nephropathy forward cycle involving decreased AMPK activity and decreased
(Fig. 37.12).129 Individuals with no or mild nephropathy showed mitochondrial biogenesis caused by increased mitochondrial
significant upregulation of mitochondrial encoded cytochrome C ROS is more consistent with these and other observations. In this
oxidase II, a subunit of mitochondrial electron transport chain model, increased mitochondrial superoxide causes the release of
complex IV, the two methylglyoxal detoxifying enzymes glyoxa- Fe++ from ferritin and iron sulfur cluster–containing proteins.
lase 1 and AKRB1 (aldo/keto reductase family 1 member 1B), Interaction of this released free iron in the nucleus with mito-
and the antioxidant proteins superoxide dismutase 1 (SOD1) and chondrial superoxide–derived hydrogen peroxide forms hydroxyl
thioredoxin. Interestingly, PKM2 protein levels in the nephropa- radicals, the only ROS species capable of cleaving bonds in mac-
thy susceptible group were identical to nondiabetics, while protein romolecules (Fig. 37.13).143 This results in ROS-mediated DNA
levels in the protected group were 40-fold higher than in nondia- double-strand breaks in the nucleus. DNA double-strand breaks
betics. PK enzymatic activity in the protected group were not dif- activate DNA repair mechanisms, including the enzyme poly-
ferent from PK activity in nondiabetics, while PK activity in the ADP-ribose polymerase 1 (PARP1). Activation of PARP1 inhibits
nonprotected group was significantly lower. The median level of the key glycolytic enzyme glyceraldehyde 3-phosphate dehydroge-
sulfenylated and oxidized PKM2 was fourfold higher in the non- nase (GAPDH) by poly-ADP-ribosylation, and depletes intracel-
protected group, consistent with the observed relative differences lular NAD+ by degrading it to synthesize ADP-ribose. Inhibition
in PKM2 protein level versus activity. Mitochondrial cytochrome of GAPDH activity causes upstream accumulation of early gly-
C oxidase II protein levels in glomeruli from protected diabetic colytic intermediates, which are diverted into pathogenic signal-
individuals were also nearly 30-fold higher than in both nonpro- ing pathways.43 Diversion of glucose increases polyol pathway
tected individuals and nondiabetic controls, and protein levels of flux, whereas diversion of fructose-6-phosphate increases hexos-
the methylglyoxal degrading enzymes glyoxalase 1 and AKR1B1 amine pathway activity. Diversion of glyceraldehyde-3-phosphate
were 10-fold higher. to α-glycerol phosphate activates PKC, and reduced activity of
GAPDH diverts accumulated triose phosphates toward methylg-
Different Hyperglycemia-Induced Pathogenic Mecha- lyoxal formation. Methylglyoxal increases expression of the recep-
nisms Reflect a Single Upstream Process: Mitochondrial tor for advance glycation end products (RAGE) and its activating
Overproduction of ROS ligand S100A8/9. Hyperglycemia-induced ROS reduce intra-
A single upstream hyperglycemia–induced process—mitochon- cellular NAD+ content by 50%.144 Reduced content of NAD+
drial overproduction of superoxide—activates all the pathogenic inhibits the activity of the NAD+-dependent protein deacetylase
mechanisms described in the previous sections.130,131 Enhanced sirtuin 1 (SIRT1), which normally deacetylates and activates
intracellular glucose transport and oxidation leads to mitochondrial both the master regulator of mitochondrial biogenesis, PGC1α,

AR ↑ 1.8
Sorbitol Glucose

HK G6PC

SORD ↑ 3.2 Glucose-6-P

HK
Fructose Fructose-6-P

PFK FBP

Fructose-1, 6-BisP

ALDO A ↑ 2.5
DHAP ALDO B ↑ 3.4 Glyceraldehyde-3-P

TPI1 ↑ 2.3 GAPDH ↑ 2.5

1-3-bisphosphoglycerate

PGK

Methylglyoxal 3-phosphoglycerate

GLO1 ↑ 2.0 PGM1 ↑ 2.2

5-lactoylglutathione 2-phosphoglycerate

ENO1 ↑ 2.8
Nucleus
HAGH ↑ 2.2 Phosphoenolpyruvate
Superoxide dismutase (SOD)
PKM2 ↑ 2.7 Thioredoxin (TXN)
Glyoxalase 1 (GLO1)
Lactate Pyruvate Aldose reductase (AR)
LDHB ↑ 2.5
PDH
Acetyl-CoA

TCA cycle

NDUFB9 (NADH: ubiquinone oxidoreductase: complex I) ↑ 1.6

MT-CO2 (Cytochrome c oxidase: complex IV) ↑ 2.1

• Fig. 37.12 Schema illustrating the significant alterations of glucose metabolism and glycolysis path-
way proteins from individuals with T1DM 50 years or more protected versus nonprotected from diabetic
nephropathy. aldo A/B, aldolase A and B; AR, aldose reductase; DAG, diglyceride; DHAP, dihydroxyac-
etone phosphate; ENO1, enolase 1; FBP, fructose-1,6-diphosphatase; G6PC, glucose-6-phosphatase,
catalytic subunit; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; GLO1, glyoxalase 1; GPI, glu-
cose phosphate isomerase; HAGH, hydroxyacyl glutathione hydrolase; HK, hexokinase; LDHB, lactate
dehydrogenase; PDH, pyruvate dehydrogenase; PFK, phosphofructokinase; PGK, phosphoglycerate
kinase; PGM1, phosphoglucomutase-1; PHGDH, 3-phosphoglycerate dehydrogenase; PKM, pyruvate
kinase isoenzyme type M2; SORD, sorbitol dehydrogenase; TPI1, triosephosphate isomerase 1. (Modified
from Qi W, Keenan HA, Li Q, et al. Pyruvate kinase M2 activation may protect against the progression of
diabetic glomerular pathology and mitochondrial dysfunction. Nat Med. 2017;23:753–765.)

and of liver kinase B1 (LKB1), the kinase that activates AMPK. SIRT1 also deacetylates and thereby inactivates the p65 sub-
Thus decreased SIRT1 activity would decrease activity of LKB1, unit of NFκB, and skews macrophages to their reparative polar-
PGC1α, and AMPK, as observed by Nishikawa and cowork- ity.146 Consistent with this model, diabetic db/db mice with a
ers, but causing increased, not decreased, mitochondrial ROS conditional deletion of SIRT1 in podocytes developed more pro-
production.145 teinuria and kidney injury compared with db/db control mice,147

Diabetes

NADPH
oxidases ROS
DNA damage
ROS
eNOS PARP NAD+
eNOS ROS
eNOS
NAD+ SIRT1
Glucose
NADPH NADP+ NAD+ NADH
Increased AKR1B1 LKB1
Sorbitol Fructose substrate conversion

Glucose-6-P AMPK
Gln Glu
Fructose-6-P Glucosamine-6-P UDP-GlcNAc PGC1α
GFAT Increased
O-GlcNAcylation
NADH NAD+ Mitochondrial
biogenesis
DHAP α-Glycerol-P DAG PKC
Activation of
Glyceraldehyde-3-P protein kinase C
NFκB
GAPDH Methylglyoxal AGEs Increased AGE formation

RAGE and
1,3-Diphosphoglycerate RAGE ligands

• Fig. 37.13 Four hyperglycemia-induced pathogenic mechanisms are activated by overproduction of reac-
tive oxygen species (ROS). Increased intracellular glucose flux causes mitochondrial overproduction of ROS,
which can be further amplified by activating NADPH oxidases and uncoupling endothelial nitric oxide syn-
thase (eNOS). Stable ROS species (hydrogen peroxide, H2O2) diffuse into the nucleus, where they cause
DNA damage, which activates poly(ADPribose) polymerase (PARP). Poly-ADPribosylation of glyceraldehyde-
3-dehydrogenase (GAPDH) by PARP reduces GAPDH activity, causing upstream accumulation of early gly-
colytic intermediates, which are diverted into four pathogenic signaling pathways. PARP activity depletes
intracellular NAD+ by degrading it to synthesize ADPribose. This reduces activity of the NAD+-dependent
deacetylase sirtuin 1 (SIRT1), which reduces activity of liver kinase B1 (LKB1), the enzyme that activates
AMP-activated protein kinase (AMPK), which activates peroxisome proliferator-activated receptor gamma
coactivator 1α (PGC1α) by phosphorylation, and of PGC1α itself (which requires deacetylation for activation).
This results in decreased mitochondrial biogenesis. AGEs, Advanced glycation end products; AKR1B1,
aldose reductase; DAG, diacylglycerol; DHAP, dihydroxyacetone phosphate; eNOS, endothelial nitric oxide
synthase; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GFAT, glutamine fructose-6-phosphate
amidotransferase; Gln, glucosamine; Glu, glucose; NAD, nicotinamide adenine dinucleotide; NADPH, nico-
tinamide adenine dinucleotide phosphate; NFκB, nuclear factor-κB; PKC, protein kinase C; RAGE, receptor
for AGEs; UDP-GlcNAc, uridine diphosphate N-acetylglucosamine. (Modified from Shah M, Brownlee M.
Molecular and cellular mechanisms of cardiovascular disorders in diabetes. Circ Res. 2016;118:1808–1829.)

and PGC1α knockout diabetic mice developed a more severe neu- nephropathy–susceptible DBA/2J mice had increased tissue levels
ropathy, with mitochondrial degeneration and an increase in oxi- and urinary excretion of 8-oxo-deoxyguanosine, a major product
dative modification of intracellular proteins.148 Recently, real-time of DNA oxidation. In humans with diabetic kidney disease, biop-
in vivo studies using a mitochondrial matrix localized GFP-based sies showed increased mitochondrial DNA damage, and increased
redox biosensor in db/db mice confirmed increased mitochondrial urinary 8-oxo-deoxyguanosine.151 Glomerular podocyte-specific
ROS production in diabetic nephropathy.149 knockout of PKM2 in diabetic nephropathy–susceptible mice
Together, these diversions and pathway activations lead to cel- had reduced expression of PGC1α, reduced mitochondrial mass,
lular dysfunction, inflammation, apoptosis, and fibrosis in cells increased mitochondrial fission, and increased ROS. Increased ROS
exposed to high intracellular glucose. The central importance of caused DNA strand breaks, increased activated PKCδ, and increased
ROS in initiating each of these processes is illustrated by the fact methylglyoxal levels. Activities of PARP and GAPDH were not
that each can be prevented when hyperglycemia-mediated ROS reported. These changes were associated with accelerated podocyte
generation is curtailed by transgenic expression of the enzymes apoptosis and glomerular pathology. Analysis of cellular respiration
superoxide dismutase or catalase. using Seahorse Extracellular Flux (XF) Analyzers showed that high
Intracellular formation of ROS also upregulates soluble epoxide glucose decreased mitochondrial maximal respiration by about 36%,
hydrolase mRNA and protein in mouse endothelial cells.150 Essen- suggesting that oxidative damage to enzymes of the tricarboxylic
tial mitochondrial genes are downregulated in glomeruli of the acid cycle, the electron transport chain, or the inner mitochondrial
diabetic nephropathy–susceptible mouse strain DBA/2J, but not membrane lipids had occurred.138,139 In cultured podocytes and in
in the diabetic nephropathy–resistant strain C57BL/6J. Diabetic diabetic mice, activation of PKM2 with a small molecular weight

ALCAT1

C:18 CL DHA CL
Mitochondria-associated membrane of endoplasmic reticulum

Outer membrane

Complex I Complex II Complex III 2 Cyt C

Q0 site
2QH2 Fe-S Cyt C1
QH2
Inner QH• bL
ETF-QOR
membrane
Q Q FADH2 bH
2Q
FAD• Q1 site
FADH2 FAD
Fe-S O2
O•–2 H2O2
FADH2
H+ + NADH NAD+ + 2H+ O2 MnSOD
ETF H2O2
Succinate Fumarate FAD• O•–2
FADH2 FAD
β-keto Acetyl CoA FFA
acyl-CoA ACAD

• Fig. 37.14 Increased fatty acid oxidation, reactive oxygen species (ROS) formation, and cardiolipin (CL)
remodeling. Insulin resistance–induced increased β oxidation of free fatty acids (FFAs) causes greater
H2O2 production than does increased glucose oxidation because of increased electron leakage from the
electron transfer flavoprotein (ETF) complex. These ROS activate acyl-CoA:lysocardiolipin acyltransferase
1 (ALCAT1) transcription. ALCAT1, located in the mitochondrial-associated membrane of the endoplasmic
reticulum, causes pathologic remodeling of cardiolipin from tetra 18:2 cardiolipin to cardiolipin with highly
unsaturated fatty acid side chains and cardiolipin deficiency because of oxidative damage. This reduces
ETC (electron transport chain) electron flux, ATP synthesis, and further increases ROS. ACAD, acyl-CoA
dehydrogenase. (Redrawn from Shah M, Brownlee M. Molecular and cellular mechanisms of cardiovascu-
lar disorders in diabetes. Circ Res. 2016;118:1808–1829.)

activator increased mitochondrial biogenesis, normalized mitochon- lipoproteins by cell surface lipoprotein lipases floods the coronary
drial maximal respiration, prevented high glucose–induced elevated arteries, heart, and liver with triglyceride-derived fatty acids. In the
ROS, and restored levels of PKCδ and methylglyoxal to low glucose liver, this drives hepatic lipoprotein synthesis and secretion, and sup-
levels. These effects are consistent with increased shunting of glucose presses hepatic glucose production.152–154 Patients with T1DM are
to lactate, thereby reducing nonproductive increased flux through also insulin resistant, with insulin sensitivity reduced by about 50%.
the mitochondrial electron transport chain. T1DM patients have significant insulin resistance in adipose tissue,
as well as in liver and skeletal muscle.155,156 In T1DM patients, insu-
lin resistance predicts the extent of coronary artery calcification and
Insulin Resistance Increases Fatty Acid likely contributes to their increased risk of cardiovascular disease.157
The role of insulin itself remains controversial. In mice fed a high-fat
Oxidation, Causing Mitochondrial diet, genetically induced reduction of hyperinsulinemia decreased
Overproduction of ROS adipose tissue inflammation and increased insulin responsiveness.153
Insulin resistance/hyperinsulinemia increases mitochondrial
Insulin resistance occurs in most patients with T2DM. To isolate fatty acid oxidation in arterial endothelial cells, macrophages,
the effects of insulin resistance from those of hyperglycemia and and cardiomyocytes, causing excessive ROS production. The
diabetes, insulin resistance was evaluated in persons without diabe- major site of electron leakage from increased fatty acid oxida-
tes or impaired glucose tolerance. Those in the highest quintile of tion is electron transfer flavoprotein, which receives electrons
insulin resistance had a 2.5-fold increase in cardiovascular disease from the FADH2 formed during the first oxidation step of
risk compared with those in the lowest quintile, after correction for β-oxidation158 (Fig. 37.14). These fatty acid oxidation-derived
11 known cardiovascular risk factors, including LDL, triglycerides, ROS induce expression of the noncoding RNA gadd7, which
systolic blood pressure, and smoking.17 This observation suggests amplifies oxidative stress and its induction of the endoplasmic
that insulin resistance itself is a major cause of cardiovascular dis- reticulum stress response in a feed-forward loop.159 In two insu-
ease in T2DM. Adipocyte insulin resistance increases lipolysis and lin-resistant nondiabetic animal models, inhibition of either FFA
circulating free fatty acids (FFAs). Hydrolysis of triglyceride-rich release from adipocytes or FFA oxidation in arterial endothelium

prevented the increased production of ROS and its damaging inhibits proliferation of vascular smooth muscle cells (VSMCs).162
effects.160 In human arterial endothelial cells, this FFA-induced However, in insulin resistance, overproduction of ROS leads to oxida-
increase in ROS activates the same damaging pathways seen with tion of tetrahydrobiopterin (BH4), the essential cofactor of endothe-
high glucose. FFA-induced overproduction of superoxide also lial nitric oxide synthase. Decreased BH4 uncouples oxygen reduction
inactivates two important antiatherogenic enzymes: prostacyclin from NO synthesis, thereby converting eNOS to a superoxide-pro-
synthase and eNOS.160 ducing enzyme.163 Although this important antiatherogenic effect of
In vitro studies suggest that at the level of the vessel wall, insulin insulin is blocked by insulin resistance–induced ROS, two major pro-
has both antiatherogenic and proatherogenic effects (Fig. 37.15).161 atherogenic effects of insulin are not. Insulin both potentiates plate-
One major antiatherogenic effect is the stimulation of endothe- let-derived growth factor (PDGF)–induced VSMC proliferation and
lial NO production. NO released from endothelial cells is a potent stimulates endothelial and VSMC production of the thrombolysis-
inhibitor of platelet aggregation and adhesion to the vascular wall. inhibitor plasminogen activator inhibitor 1 (PAI1).164–166
Endothelial NO also controls the expression of genes involved in Macrophages are a central participant in atherogenesis. While
atherogenesis. It decreases expression of monocyte chemoattractant most atherosclerotic lesions are stable, insulin resistance increases
protein 1 (MCP1) and of surface adhesion molecules such as CD11/ macrophage ROS, which drive chronic inflammation and accel-
CD18, P-selectin, vascular cell adhesion molecule 1 (VCAM1), and erate their progression to unstable rupture-prone plaques. These
intercellular adhesion molecule 1 (ICAM1). Endothelial cell NO also plaques have a greater inflammatory infiltrate, a larger pro-
reduces vascular permeability and decreases the rate of oxidation of thrombotic necrotic core, and increased production of matrix
LDL to its more proatherogenic form. Finally, endothelial cell NO metaloproteinases that weaken the fibrous cap. Insulin resis-
tance in macrophages induces mitochondrial ROS production
Insulin by activating the CHOP branch of the stress unfolded protein
response (UPR).167,168 Plaque necrosis is caused by a combina-
tion of increased macrophage apoptosis and decreased phagocytic
IR
clearance of the apoptotic macrophages by a process called effe-
rocytosis, which results in post–apoptotic necrosis of apoptotic
cells and inflammation. Both of these processes are promoted by
IRS1, 2 macrophage insulin resistance.169 Apoptotic cells can inappropri-
MAPKK ately express the ROS-TNFα-NFκB-inducible “don’t eat me” cell
PI 3-Kinase surface signal CD47,170 and efferocyte receptors on phagocytes
MAP-K
such as MER tyrosine kinase (MerTK) can be cleaved and inac-
AKT
tivated by the mitochondrial ROS-activated protease ADAM17
1176
(a disintegrin and metalloproteinase).171 Blocking antibodies to
S
CD47 prevent progression of established lesions, protect against
PAI1 ET1 VCAM-1 VEGF HO1 eNOS plaque rupture, and induce regression of the necrotic core in sev-
NO eral mouse models.123,170,172,173
Fatty acid oxidation appears to be the source of mitochondrial
Vasoconstriction Vasodilation ROS production in macrophages, directly or indirectly via altered
Proliferation Monocyte Adhesion mitochondrial cardiolipin fatty acid composition.174,175 When
Migration Inflammation
Oxidative Stress fatty acids are oxidized instead of glucose, the ratio of NADH to
FADH2 changes from 5:1 to 2:1. This causes an overreduction of
the mitochondrial electron transport chain coenzyme Q (CoQ)
Proatherogenic action Antiatherogenic action pool, and an increase in reverse electron transport (RET) and
increased decreased ROS production by complex I. ROS from mitochondrial electron
transport chain complex I cause a high percentage of irreversible
In insulin resistance
overoxidation of cysteine thiols.41,42 Reverse electron transport to
• Fig. 37.15 Selective insulin resistance in vascular cells. Selective insulin complex I, driven by increased complex II oxidation of the TCA
resistance in vascular cells occurs when angiotensin II, elevated free fatty cycle intermediate succinate, drives the polarization of macro-
acids (FFAs) and glucose levels, and proinﬂammatory cytokines induced
phages to a proinflammatory phenotype.176
by diabetes and insulin resistance inhibit only the IRS/PI3K/Akt pathway.
In contrast, insulin’s stimulation of the SOS/Grb2/MAPK pathway is unaf-
In diabetic heart, increased myocardial ROS production occurs
fected or even enhanced. The selective loss of insulin’s actions via the early, before accumulation of triglycerides and subsequent syn-
IRS/PI3K/Akt pathway causes the reduction of insulin’s antiatherosclerotic thesis of C16:0 ceramides are evident.177,178 This likely reflects
action and contributes to the acceleration of atherosclerosis and other increased fatty acid oxidation and resultant oxidative damage to
cardiovascular pathologies in diabetes. AKT, protein kinase B; eNOS, mitochondrial cardiolipin,179 the specific phospholipid of mito-
endothelial nitric oxide synthase; ET1, endothelin-1; HO1, heme oxygen- chondrial membranes. Cardiolipin is important for efficient
ase-1; ICAM, intracellular adhesion molecule; IR, insulin receptor; IRS1,2, electron flux, ATP synthesis, and reduced ROS formation. In
insulin receptor substrate 1,2; MAP-K, mitogen-activated protein kinase; addition, cardiolipin is involved in mitochondrial-mediated apop-
MAPKK, MAPK kinase; NO, nitric oxide; PAI, plasminogen activator inhibi- tosis, and it plays a critical role in regulating mitochondrial fission
tor; PI, phosphatidylinositol; TNF, tumor necrosis factor; VCAM1, vascular
and fusion.180–182 In nondiabetic heart, the major species of car-
cell adhesion molecule-1; VEGF, vascular endothelial growth factor; VSMC,
vascular smooth muscle cell. (Data from King G, Brownlee M. The cellular
diolipin contains four linoleic acids (tetra 18:2 cardiolipin). This
and molecular mechanisms of diabetic complications. Endocrinol Metab unique acyl composition is not derived from de novo synthesis of
Clin North Am. 1996;25:255–270; King GL, Park K, Li Q. Selective insulin cardiolipin, but rather from a remodeling process that involves
resistance and the development of cardiovascular diseases in diabetes: phospholipases and acyltransferase-transacylases. In diabetic myo-
the 2015 Edwin Bierman Award Lecture. Diabetes. 2016;65:1462–1471.) cardium from murine models with insulin resistance (ob/ob, db/

db, and high-fat diet) and in models of severe insulin deficient db/db mice, treatment with the tetracycline antibiotic minocycline
type 1 diabetes, the fatty acyl content of the more saturated 18:2 increased Nrf2 protein levels, reduced glomerular oxidative stress
cardiolipin is dramatically reduced, whereas the content of longer markers, and ameliorated diabetic nephropathy.194
chain, more unsaturated fatty acyl cardiolipin such as 20:4 is sub- Nrf2 is expressed constitutively, and its intranuclear levels are
stantially increased.183 Because of this increase in highly unsatu- controlled post-translationally. In the absence of inducers, Nrf2
rated side chains, diabetic heart cardiolipin is more vulnerable to associates with the redox-sensitive protein Kelch-like erythroid
oxidative damage. Cardiac overexpression of cardiolipin synthase cell–derived protein with cap ‘n’ collar homology–associated pro-
increases tetra 18:2 cardiolipin in diabetic mice and prevents dia- tein 1 (Keap1), where it is rapidly polyubiquinated by Keap1-asso-
betes-induced changes in cardiolipin lipid remodeling. The cardio- ciated cullin-3–RING E2 ubiquitin ligase proteins and degraded
lipin deficiency and profound remodeling caused by diabetes and by proteasomes. Nrf2 bound to Keap1 is released by ROS oxida-
insulin resistance is caused by ROS-induced transcription of acyl- tion of critical cysteine thiols of Keap1, or by the reaction of these
CoA:lysocardiolipin acyltransferase 1 (ALCAT1) (see Fig. 37.14). thiols with ROS-generated electrophiles such as glycolysis-derived
ALCAT1 catalyzes the transfer of linoleoyl-CoA onto monoly- methylgloxal and unsaturated fatty acid peroxidation-derived
socardiolipin or dilysocardiolipin. Overexpression of ALCAT1 4-hydroxynonenal. Phosphorylation of Nrf2 by protein kinases
caused cardiolipin deficiency and fatty acid compositional changes such as casein kinase 2 (CK2) may help target Nrf2 to the nucleus.
similar to diabetes and obesity, with increased production of ROS, After forming heterodimers with small Maf proteins, Nrf2 binds
whereas ALCAT1 deficiency increased levels of tetra 18:2 cardio- to the antioxidant response element (ARE) to induce transcription
lipin in mouse heart and reduced ROS production.184 of its target genes. Export of Nrf2 from the nucleus is controlled
High levels of ROS are also a major proximal activator of the by phosphorylation. Src family members such as Fyn phosphory-
FOXO family of transcription factors. Increased ROS stimulate late Nrf2 at Tyr568, causing export from the nucleus and degrada-
FOXO translocation from the cytosol into the nucleus by increas- tion.195 Reduction of Nrf2 protein in the cytosolic compartment
ing FOXO GlcNAcylation, Jun-N-terminus kinase signaling, is mediated by β-transducin repeat–containing protein, a substrate
and CaMKII activation.185,186 Cysteine oxidation also increases adaptor for the S-phase kinase-associated protein 1−Cul1−F-box
FOXO transcriptional output. In the hearts of diabetic mice and protein E3 ubiquitin ligase, which targets GSK3β-phosphorylated
of mice with high-fat diet–induced insulin resistance, FOXO Nrf2 to the proteosome196 (Fig. 37.16).
proteins were persistently activated.187 This persistent activation Modification of critical Keap1 cysteine thiols by sulfora-
was associated with downregulation of insulin receptor substrate phane, a dietary isothiocyanate found in cruciferous vegetables,
1 (IRS1), reduced activity of IRS1 and its downstream target Akt, also releases Nrf2.199 In endothelial cells, sulforaphane prevented
and the development of cardiomyopathy. In cardiomyocyte-spe- hyperglycemia-induced activation of the hexosamine and PKC
cific FOXO1 knockout mice fed a high-fat diet, neither insulin pathways and prevented increased cellular accumulation and
resistance nor cardiomyopathy occurred. excretion of the major AGE precursor methylglyoxal.197 In the
aortae of diabetic mice, sulforaphane treatment restored aortic
Diabetes Reduces Activity of Nuclear Erythroid- levels of Nrf2 and Nrf2-dependent antioxidant gene expression,
Related Factor 2, the Master Regulator of preventing diabetes-induced increases in wall thickness, fibrosis,
inflammation (tumor necrosis factor-α [TNFα] and vascular cell
Antioxidant Gene Expression adhesion molecule-1 expression), apoptosis, and increased cell
The number of recognized ROS-regulating enzymes has increased proliferation.198 Diabetic cardiomyopathy was also prevented in
substantially over the past 15 years.40 Examples include superoxide mouse models by sulforaphane treatment.199 Normalized Nrf2
dismutases, catalases, glutathione peroxidases, glutathione reduc- activity also prevented diabetes-associated cardiac inflammation,
tase, thioredoxins, thioredoxin reductases, methionine sulfoxide fibrosis, lipid accumulation, and impaired autophagy.
reductases, and peroxiredoxins. The activity of these enzymes is In hearts from diabetic patients, Nrf2 protein is significantly
largely determined by ROS-induced changes in their transcrip- reduced. In mice, cardiac Nrf2 protein was similarly reduced after 5
tion. Increased transcription of many of these antioxidant enzymes months of diabetes mellitus.200 Hyperglycemia-induced mitochon-
is mediated by the transcription factor nuclear erythroid–related drial overproduction of ROS has been shown to increase glycogen syn-
factor 2 (Nrf2), a member of the cap ‘n’ collar subfamily of basic thase kinase 3 beta (GSK3β) activity by inhibiting Akt1-dependent
region leucine zipper transcription factors.188 By regulating oxi- phosphorylation of GSK3β at serine 9.143 Activated GSK3β targets
dant levels and oxidant signaling, Nrf2 participates in the control cytoplasmic Nrf2 for increased proteasomal degradation mediated by
of the unfolded protein response, apoptosis, mitochondrial bio- the β transducin repeat–containing protein/S-phase kinase-associated
genesis, and stem cell regulation. Nrf2 also increases transcription protein 1−Cul1−F-box protein E3 ubiquitin ligase.196
of GLO1, the rate-limiting enzyme of the glyoxalase system, which
prevents post-translational modification of proteins and histones Diabetes Activates the NLR Family Pyrin Domain
by methylglyoxal, the major AGE precursor.189,190 It also increases
transcription of the rate-controlling enzyme in the nonoxidative Containing 3 Inflammasome
branch of the pentose phosphate pathway, transketolase. Activa- Dysregulated NLRP3 inflammasome activity underlies many
tion of transketolase inhibits three of the major hyperglycemia- chronic inflammatory states. The NLRP3 inflammasome is
driven pathways implicated in the pathogenesis of diabetic vascular formed by oligomerization of inactive NLRP3, associated with
damage (the diacylglycerol-PKC pathway, the methylglyoxal-AGE apoptosis-associated speck–like protein (ASC), and procaspase-1.
formation pathway, and the hexosamine pathway) and inhibits This complex, in turn, catalyzes the conversion of procaspase-1 to
hyperglycemia-induced NFκB activation.58 Preclinical studies caspase-1, which contributes to the production and secretion of
using Nrf2 activators or Nrf2-deficient diabetic mice established mature proinflammatory IL1β and IL18.201 Activation requires
that Nrf2 is a crucial endogenous modulator of ROS and protects two steps. The first, called priming, activates the transcription fac-
from experimental diabetic nephropathy.191–193 In the kidneys of tor NFκB, which promotes transcription of NLRP3, pro-IL1β,

SH SH ROS

Keap1 βTrCP
Cul3 Cul1
P
Nrf2 Nrf2

Ub E2 Akt Ub E3

GSK3β Proteosomal
Nrf2 Nrf2 P degradation

Cellular protection and cell survival

Antioxidant proteins

Other
proteins CBP/
Nrf2 p300
sMaf Antioxidant gene
ARE expression

• Fig. 37.16 Diabetes mellitus reduces nuclear erythroid–related factor 2 (Nrf2) protein in diabetic kidney
and heart. Nrf2, the master regulator of antioxidant gene expression, associates with the redox-sensitive
protein Kelch-like erythroid cell–derived protein with cap ‘n’ collar homology–associated protein 1 (Keap1),
where it is rapidly polyubiquitinated by Keap1-associated cullin-3 (Cul3)–RING E2 ubiquitin ligase proteins
and degraded by proteasomes. Reactive oxygen species (ROS) oxidation of critical cysteine thiols of Keap1
causes release of bound Nrf2 protein. Phosphorylation of Nrf2 by protein kinases such as CK2 help target
Nrf2 to the nucleus. Nrf2 forms heterodimers with small Maf proteins, which bind to the antioxidant response
element (ARE) in its target gene promoters. After export of Nrf2 from the nucleus, cytosolic Nrf2 is phos-
phorylated by GSK3β. This phosphorylated Nrf2 is recognized by β-transducin repeat–containing protein
(βTrCP), a substrate adaptor for the S-phase kinase–associated protein-1−Cul1−F-box protein E3 ubiquitin
ligase, which targets Nrf2 phosphorylated by GSK3β to the proteasome. Akt, protein kinase B; CBP/p300,
CREB-binding protein and its homologue p300; CK2, casein kinase II; E2, ubiquitin-conjugating enzymes;
GSK3β, glycogen synthase kinase-3β; NRF2, nuclear factor (erythroid-derived 2)–like 2; sMaf, small mus-
culoaponeurotic fibrosarcoma proteins; Ub, ubiquitin. (Redrawn from Shah M, Brownlee M. Molecular and
cellular mechanisms of cardiovascular disorders in diabetes. Circ Res. 2016;118:1808–1829.)

and pro-IL18.202 In mononuclear cells from diabetic patients and mitochondrial DNA. Oxidized DNA fragments then exit the
in vascular endothelial cells from diabetic rats, NFκB is chronically mitochondria, bind to the NLRP3 inflammasome, and activate
active.203 This likely results from TLR4 or TLR4/RAGE heterodi- it (Fig. 37.17).
mer signaling stimulated by binding of ROS-induced S100A8/9 Increased expression of the NLRP3 inflammasome compo-
calgranulins. A second signal activates the NLRP3 inflammasome nents Nlrp3 and apoptosis-associated speck–like protein was
by facilitating the oligomerization of inactive NLRP3, apoptosis- found in monocytes from new, untreated patients with T2DM.
associated speck–like protein, and procaspase-1. Although diverse Along with increased expression, there was increased inflamma-
molecular cues can trigger this step, several studies had suggested some activation. Consistent with this, drug-naive T2DM patients
that these might all act through a mitochondrial pathway asso- had significantly higher serum levels of the proinflammatory cyto-
ciated with high levels of mitochondrial ROS. Blocking ROS kines IL1β and IL18 than did healthy individuals.207 In kidneys
generation by mitochondria has been shown to abolish NLRP3 from diabetic humans and in murine models of both T2DM and
inflammasome activation, whereas artificial induction of mito- T1DM, inflammasome activation was detected in glomerular
chondrial ROS can spontaneously induce NLRP3-mediated IL1β endothelial cells and podocytes.208 Nlrp3 deficiency protected
secretion.204,205 Mitochondrial ROS stimulate calcium influx via these mice from experimental diabetic nephropathy. Analysis of
the TRPM2 channel (transient receptor potential cation channel gene expression data in the Nephromine database revealed persis-
subfamily M member 2), and macrophages deficient in TRPM2 tently elevated glomerular expression of inflammasome markers
have drastically impaired NLRP3 inflammasome activation and in patients with diabetic nephropathy and in murine models.209
IL1β secretion.206 Synthesis and oxidation of mitochondrial DNA NLRP3 inflammasome activation also occurs in diabetic renal
also drive this activation step. TLR4 signaling increases levels of tubulointerstitial cells. Tubulointerstitial fibrosis is the final com-
the human mitochondrial CMP kinase enzyme CMPK2, which mon pathway for loss of renal function in diabetic nephropathy,
increases nucleotide cytidine triphosphate, causing synthesis of and tubuloepithelial to mesenchymal transdifferentiation is one

Ca2+
S100A8/12 Glucose/fatty acids

RAGE TLR4
Newly made oxidized SIGNAL 2
ROS mitochondrial DNA
Ca2+ influx
Priming
Procaspase-1

ASC
NLRP3
Inactive NLRP3 ASC Inflammasome
SIGNAL 1 Procaspase-1

Active caspase-1

NFκB
Pro-IL1β IL1β

Pro-IL18 IL18

• Fig. 37.17 NOD-like receptor family, pyrin domain-containing 3 (NRLP3) inflammasome activation in dia-
betic monocytes, kidney, coronary arteries, and heart. Hyperglycemia and increased fatty acids induce
reactive oxygen species (ROS). These increase receptor for advanced glycation end products (RAGE)
expression, which heterodimerizes with toll-like receptor 4 (TLR4). Signaling from this complex causes
nuclear factor-κB (NFκB)–mediated transcription of inactive NRLP3, prointerleukin 1β (Pro-IL1β), and Pro-
IL18. Increased intracellular Ca2+ and/or newly made oxidized mitochondrial DNA triggers oligomerization
of inactive NLRP3, associated with apoptosis-associated speck–like protein (ASC), and procaspase-1.
This activated inflammasome complex catalyzes the conversion of procaspase-1 to caspase-1, and of
pro-IL1β and proIL-18 to mature IL1β and IL18. S100A8/12, Calgranulin A/B heterodimer ligand for RAGE.
(Data from Shah M, Brownlee M. Molecular and cellular mechanisms of cardiovascular disorders in dia-
betes. Circ Res. 2016;118:1808–1829; Murphy MP. Newly made mitochondrial DNA drives inflammation.
Nature. 2018;560:176–177.)

source of renal interstitial fibrogenic myofibroblasts. In diabetic atherosclerosis, increased NLRP3 was found both in macrophages
kidney disease, the degree of tubulointerstitial fibrosis is probably of advanced lesions and in endothelial cells and smooth muscle
a stronger predictor of GFR decline than glomerular changes.210 A cells. The changes found in porcine diabetic atherosclerosis were
critical mediator of renal fibrosis is the ROS-activated redox-sen- also present in coronary atherosclerosis samples from diabetic
sitive kinase apoptosis signal-regulating kinase 1 (ASK1), which humans.
activates p38 and JNK downstream signaling pathways. In renal
biopsies from patients with diabetic kidney disease, ASK1 acti- Diabetes Activates the Transcription Factor NFAT
vation was associated with myofibroblasts, and in db/db eNOS-
/- diabetic mice, inhibition of ASK1 arrested progressive GFR The transcription factor NFAT plays a role in the development of
decline, glomerulosclerosis, and proteinuria.211 diabetic retinopathy, nephropathy, atherosclerosis, and cardiomy-
In a T2DM rat model, excessive activation of NLRP3 was opathy. In resting cells, NFAT proteins are phosphorylated and
associated with cardiac inflammation, cell death, disorganized are located in the cytoplasm. In diabetes, intracellular calcium is
ultrastructure, and fibrosis. NLRP3 gene silencing ameliorated increased by increased ROS. Increased intracellular calcium then
cardiac inflammation, apoptosis, fibrosis, and left ventricular activates NFATc1-c4 by increasing NFAT dephosphorylation by
cardiac dysfunction.212 In a well-established porcine model of the Ca2+/calmodulin-dependent serine/threonine phosphatase cal-
diabetic atherosclerosis, which develops complex atherosclerotic cineurin, which facilitates NFAT translocation into the nucleus.
plaques resembling those of humans, the cleavage processing of Once in the nucleus, NFAT interacts with coregulators to achieve
the sterol regulatory element-binding protein transcription fac- optimal NFAT activation.215 In the nucleus, ADP-ribosylation
tors (SREBP1 and SREBP2) and expression of their target genes, mediated by PARP1 acts as a molecular switch to positively regulate
which increase fatty acid synthesis and cholesterol biosynthesis, NFAT-dependent cytokine gene transcription216 (Fig. 37.18). In
were increased in endothelial cells and infiltrating macrophages diabetes, nuclear PARP1 is activated by intracellular ROS-induced
of both fatty streaks and advanced lesions with fibrous caps, DNA strand breaks (see Fig. 37.13).
necrotic cores, and cholesterol cores. SIRT1 and AMPK activity In diabetic retinopathy, upregulation of proinflammatory
were also reduced.213 Increased SREBP1a in macrophages directly cytokines such as TNFα and upregulation of retinal microvascu-
upregulates transcription of NLRP3,214 and in porcine diabetic lar endothelial cell adhesion proteins are important components

Glucose/free fatty acids L-type calcium channel

RyR

ROS Calcium Ca2+

Calpain Ca2+/CaM
Increases osteopontin,
Calcineurin MCP1, ICAM1, IL6,
and atherosclerosis
NFAT P

RP Stimulates
PA R
NFAT hypertrophy,
fibrosis, and
heart failure
Nuclear transcription

• Fig. 37.18 Increased mitochondrial oxidation of glucose or fatty acids activates nuclear factor of acti-
vated T cells (NFAT)–mediated transcription of genes promoting diabetic retinopathy, nephropathy, ath-
erosclerosis, and cardiomyopathy. Mitochondrial overproduction of reactive oxygen species (ROS) causes
increased intracellular Ca2+, which activates the calcium-activated neutral cysteine protease calpain. Cal-
pain then activates the Ca2+/calmodulin-dependent (CaM; Ca2+) serine/threonine phosphatase calcineu-
rin. Dephosphorylation facilitates nuclear translocation of the transcription factor NFAT. In the nucleus,
NFAT interacts with poly-ADPribose polymerase (PARP), which increases NFAT transcriptional activity via
NFAT poly-ADPribosylation. ICAM1, Intercellular adhesion molecule-1; IL6, interleukin-6; MCP1, monocyte
chemoattractant protein-1; RyR, ryanodine receptor. (Modified from Shah M, Brownlee M. Molecular and
cellular mechanisms of cardiovascular disorders in diabetes. Circ Res. 2016;118:1808–1829.)

of chronic inflammation.217 In human retinal microvascular a role in cardiac hypertrophy, fibrosis, and cardiomyocyte apop-
endothelial cells, TNFα activates NFAT signaling. Activated tosis.223 In the diabetic heart, NFAT is activated by the calcium-
NFAT specifically upregulated expression of the adhesion protein activated neutral cysteine protease calpain, which in turn activates
VCAM1, which increases adhesion of lymphocytes and macro- calcineurin. In cardiomyocytes, increased calpain activity activates
phages to endothelium, the inflammatory cytokine CX3CL1 NFAT-dependent cardiac hypertrophy and heart failure.224 In
(C-X3-C motif chemokine ligand 1), which promotes leukocyte two mouse models of diabetes, cardiac-specific deletion of calpain
adhesion, the neutrophil chemotactic cytokine CXCL6, and the reduced myocardial hypertrophy and fibrosis, leading to improve-
activated T-cell chemotactic cytokine CXC11.218 In human renal ment of myocardial function. Calpain activation correlated with
glomerular podocytes exposed to sera from diabetic kidney disease increased activity of NFAT and NFκB, consistent with calpain’s
patients, increased TNFα production caused NFATc1-mediated role in activation of calcineurin and degradation of the cytosolic
repression of the cholesterol efflux regulator ATP-binding cas- NFκB inhibitor, NFκB inhibitor α.225,226
sette transporter A1 (ABCA1), and reduced activity of sterol-O-
acyltransferase 1 (SOAT1), resulting in free cholesterol-mediated Diabetes Increases Neutrophil Extracellular
podocyte injury.219
In arteries of diabetic mice, activated NFATc3 induces arte- Traps
rial smooth muscle cell expression of the proinflammatory matrix In atherosclerotic plaques, neutrophils prime macrophages for
protein osteopontin, a cytokine that promotes atherosclerosis and proinflammatory responses.227 This priming is mediated by extra-
diabetic vascular disease. NFAT inhibition effectively reduced cellular webs of DNA bound to cytotoxic histones, which are
arterial wall osteopontin, IL6, MCP1, intercellular adhesion mol- released by activated neutrophils, called neutrophil extracellular
ecule-1, CD68, and tissue factor expression, and lowered plasma traps (NETs).228 This process follows a coordinated multistep pro-
IL6 in diabetic mice.220 In diabetic ApoE−/−mice, inhibition of cess: histone citrullination, chromatin decondensation, migration
NFAT-signaling completely suppressed a 2.2-fold increase in ath- of elastase and other granule enzymes into the nucleus, disinte-
erosclerotic plaque area. Inhibition of NFAT also reduced lipid gration of the nuclear membrane, and release of DNA, histones,
content in the plaque of diabetic mice independent of plasma and granule proteins into the extracellular space.229 The release of
glucose and lipid levels.221 NFATc3 activated by increased mito- NETs primes macrophages to produce pro-IL1β, which is cleaved
chondrial ROS production also increases arterial vasoconstrictor to mature proinflammatory IL1β by caspase-1. Caspase-1, in
reactivity to endothelin-1.222 NFAT activation also appears to play turn, is secreted by macrophages in response to activation of the

1 Activation of neutrophil PAD4, histone

citrullination and chromatin decondensation

Glucose/free fatty acids

2 Release of nuclear
ROS
material and
enzymes (NETs)

PAD4
Cit

3 Priming of macrophage
H4R3 H4cit3

4 Secretion of
Pro-IL1β and
IL1β
cleavage by
5 Propagation of inflammation inflammasome
caspase-1

• Fig. 37.19 Diabetes increases neutrophil extracellular traps (NETs), priming macrophages for inflammation.
Increased reactive oxygen species (ROS) increase transcription and activation of peptidylarginine deimi-
nase 4 (PAD4), the enzyme that initiates formation and release of NETs by citrullination (Cit) of histones.
Released NETs prime macrophages to produce pro-interleukin 1β (Pro-IL1β), which is cleaved to mature
proinflammatory IL1β by caspase-1 secreted by macrophages in response to NOD-like receptor family,
pyrin domain-containing 3 inflammasome activation. H4cit3, Histone 4 with arginine residue 3 converted to
citrulline; H4R3, histone 4 arginine 3; ROS, reactive oxygen species. (Modified from Shah M, Brownlee M.
Molecular and cellular mechanisms of cardiovascular disorders in diabetes. Circ Res. 2016;118:1808–1829.)

NLRP3 inflammasome.227 Both ROS and increased intracellular by monocytes, which differentiate into proinflammatory mac-
Ca2+ activate NET formation (Fig. 37.19).230 Atherosclerosis- rophages. These cells engulf the inflammatory stimulus and
prone Apoe−/− mice with deletions of two serine proteases that remove damaged tissue debris. Resolution of inflammation is an
localize to NETs developed dramatically smaller atherosclerotic active process involving a switch from synthesis of arachidonic
lesions compared with Apoe−/− control animals, despite similar acid–derived proinflammatory prostaglandins and leukotri-
lipid concentrations and leukocyte counts in blood. These triple enes to synthesis of at least four families of specialized prore-
mutant mice had no NETs, lower systemic IL1β concentration, solving mediators (SPMs). These SPM families are arachidonic
and fewer lesional IL17-producing T cells.231 Quenching mito- acid–derived lipoxins, and omega-3 polyunsaturated fatty acid–
chondrial ROS in a mouse model of atherosclerosis (aged myeloid derived resolvins, protectins, and maresins.121 SPMs decrease
cell mCAT→LDL receptor null [Ldlr−/−] chimeric mice) decreased further recruitment of PMNs, stimulate macrophage uptake
lesional NETs compared with age-matched Ldlr−/− mice, and also and clearance of apoptotic cells (efferocytosis), shift macrophage
decreased total macrophage content in atherosclerotic lesions.232 phenotype from proinflammatory to proresolution, and initiate
Oxidized LDL (oxLDL) and 7-keto-cholesterol, the most abun- tissue repair processes. Obesity and diabetes delay PMN apop-
dant oxysterol in human oxLDL, induced mitochondrial ROS tosis and impair macrophage efferocytosis. In human arteries,
production and highly proinflammatory oxidized mitochondrial high levels of resolvin D1 and low levels of the proinflammatory
DNA-bound NETs.233 Neutrophils from both type 1 and type leukotriene B4 were associated with plaques having less necrosis
2 diabetic humans have elevated expression of peptidyl arginine and thicker fibrous caps. In contrast, high levels of leukotriene
deiminase 4, the enzyme critical for histone citrullination-medi- B4 and low levels of resolvin D1 were associated with plaques
ated chromatin decondensation and NET formation.234 having large necrotic cores and thin fibrous caps.241
Current evidence suggests diabetes causes defective SPM bio-
Nonresolving Inflammation in Metabolic synthesis downstream of their fatty acid precursors.238,239 The
resultant nonresolving inflammation causes continuous leuko-
Syndrome, Diabetes, and Atherosclerosis cyte influx into atherosclerotic lesions, inflammatory macrophage
Chronic low-grade inflammation is associated with meta- polarization, and defective efferocytosis. These drive the progres-
bolic syndrome, diabetes, and atherosclerosis.235–240 In normal sion of atherosclerosis to the development of clinically dangerous
physiology, inflammation has two temporal phases: an initia- lesions with large necrotic, procoagulant cores and thin fibrous
tion phase of acute inflammation and a resolution phase. The caps. SPMs limit plaque progression by suppressing inflammation,
initiation phase involves recruitment of neutrophils followed enhancing efferocytosis, and promoting an increase in thickness

Advanced SPMs Stable LTs

plaque plaque
LTs SPMs

Continuous Platelet Decreased No platelet

monocyte aggregation monocyte aggregation
influx influx

Thin Endothelium Thick Smaller

fibrous cap
fibrous cap necrotic
Necrotic
core core

Inflammatory Impaired Resolving Increased

creased
macrophage efferocytosis macrophage efferocytosis
rocytosis
polarization polarization
Smooth muscle cells

Nonresolving inflammation Resolving inflammation

A B

• Fig. 37.20 Nonresolving inflammation in atherosclerosis. Reduced specialized proresolving mediators

(SPMs) and increased leukotrienes (LTs) promotes instability of atherosclerotic plaques. When the SPM:LT
ratio is low (A), resolution of inflammation is impaired, leading to sustained inflammatory monocyte influx,
platelet aggregation, proinflammatory macrophage polarization, impaired efferocytosis, large necrotic
cores, and thin fibrous caps. Conversely, when the SPM:LT ratio is high (B), resolution of inflammation is
characterized by decreased monocyte influx, proresolving macrophage polarization, enhanced efferocy-
tosis, decreased necrotic core formation, and thicker fibrous caps. (Redrawn from Kasikara C, Doran A,
Cai B, et al. The role of non-resolving inflammation in atherosclerosis. J Clin Invest. 2018;128:2713–2723.)

of the fibrous cap (Fig. 37.20). Enhanced efferocytosis promotes phagosome formation around secondarily encountered apoptotic
macrophage production of several SPMs, including lipoxin A4, cells is impaired.244 The primary regulators of outer mitochondrial
resolvin E1, and protectin D1. Lipoxins also recruit monocytes as fusion are dynamin-related guanosine triphosphate hydrolases
macrophages further enhance efferocytosis in the resolving lesions. (GTPases), termed mitofusin 1 (Mfn1) and mitofusin 2 (Mfn2),
Efferocytosis requires specific phagocytotic receptors, which bind while inner mitochondrial fusion and cristae stabilization involve
to apoptotic cells. Deletion of one of these receptors, Mer tyrosine optic atrophy protein 1 (Opa1). The primary regulator of mamma-
kinase (MerTK), caused impaired efferocytosis, accumulation of lian mitochondrial fission is the GTPase dynamin-related protein
apoptotic cells, and large necrotic cores in murine atherosclerotic 1 (Drp1), which is recruited from the cytosol to the mitochondrial
lesions. It also caused impaired SPM production by macrophages, outer membrane where it binds to four Drp1 receptors: mitochon-
because MerTK signaling decreases proinflammatory leukotriene drial fission factor (MFF), mitochondrial dynamics protein of 49
synthesis and increases synthesis of proresolving lipoxin A4.239 kDa (miD49) and 51 kDa (miD51), and mitochondrial fission 1
MerTK can be inactivated by the metalloproteinase ADAM17, protein (FIS1).245 Oligomerization of Drp1 is thought to provide
which, when activated by macrophage overproduction of mito- the mechanical force to constrict mitochondrial membranes and to
chondrial ROS, cleaves the receptor.171 High levels of soluble fragment the organelle. Mitochondrial fission occurs at mitochon-
MerTK are found in advanced human atherosclerotic lesions, dria-endoplasmic reticulum contact sites, with involvement of the
suggesting that this is one mechanism impairing efferocytosis. mitochondrial phospholipid cardiolipin and calcium transfer.246
Increased levels of ROS cause autonomous activation of CaM- Increased mitochondrial fission caused by increased mitochon-
KII by oxidation of adjacent methionine residues in its regulatory drial recruitment of Drp1 is associated with key features of diabetic
domain.107 In arteries, activated CaMKII suppresses efferocytosis nephropathy in diabetic mouse models133 (Fig. 37.21). In diabetic
in macrophages and promotes atherosclerotic plaque necrosis.242 mice lacking Drp1, decreased fission protected against progres-
sion of diabetic nephropathy.247 Increased mitochondrial fission
also occurs in mitochondria from mouse coronary endothelial
Diabetes Alters Mitochondrial Dynamics cells, where the level of the fusion protein Opa1 is decreased and
Mitochondria dynamics are the continuous processes of mitochon- the level of Drp1 is increased.248 This response may represent an
drial fusion, fission, biogenesis, and mitophagy, which together important proresolving response in atherosclerotic lesions, since
maintain optimal cellular bioenergetics and ROS homeostasis.243 mice lacking myeloid Drp1 have defective efferocytosis.235,244
In normal physiology, fission helps facilitate mitophagy, which Persistent increased fission is also associated with myocardial dys-
degrades and recycles damaged mitochondria and mitochondrial function in humans with type 2 diabetes, due in part to decreased
fragments, although fission also increases ROS production due expression of the fusion protein mitofusin 1.249 Drp1 recruitment
to incomplete transfer of electrons through the electron transport to mitochondria is regulated by its phosphorylation at serine 616
chain. Mitochondrial fission is a critical process that enables mac- by either CaMKII or ROCK1. In the diabetic heart and kidney,
rophages to clear multiple apoptotic cells. When fission is disabled, ROS can activate both of these kinases.108,133

Glucose/free fatty acids

RyR

ROS Calcium Ca2+

Calpain Ca2+/CaM

Calcineurin

P
DRP1 DRP1
ROS

DRP1 DRP1
DRP1 DRP1
MFF MID49
FIS1 MID51

• Fig. 37.21 Diabetes induces increased mitochondrial fission in kidney, coronary artery, and myocardium.
Increased reactive oxygen species (ROS) from either excess glucose or fatty acids increases intracel-
lular Ca2+. Increased Ca2+ activates the calcium-activated neutral cysteine protease calpain, which acti-
vates the Ca2+/calmodulin-dependent (CaM; Ca2+) serine/threonine phosphatase calcineurin. Calcineurin
dephosphorylates the GTPase dynamin–related protein 1 (DRP1), which is then recruited from the cytosol
to the mitochondrial outer membrane where it binds to four DRP1 receptors—mitochondrial fission fac-
tor (MFF), mitochondrial dynamics protein of 49 kDa (MID49), and 51 kDa (MID51), and mitochondrial
fission 1 protein (FIS1). Drp1 oligomerization provides the mechanical force to constrict mitochondrial
membranes and fragment the organelle (mitochondrial fission). Increased fission causes further increases
in ROS production and mitochondrial dysfunction. (Modified from Shah M, Brownlee M. Molecular and
cellular mechanisms of cardiovascular disorders in diabetes. Circ Res. 2016;118:1808–1829.)

Genetic Determinants of Susceptibility to 80

Study 1
Microvascular Complications Study 2
Prevalence of diabetic

Clinicians have long observed that different patients with simi- 60

nephropathy (%)

lar duration and degree of hyperglycemia differ markedly in their

susceptibility to microvascular complications. Such observations 40
suggested that genetic differences exist that affect the pathways
by which hyperglycemia damages microvascular cells. The leveling
of risk of overt proteinuria after 30 years’ duration of T1DM at 20
27% is evidence that only a subset of patients are susceptible to
development of diabetic nephropathy.250 A role for genetic deter- 0
minants of susceptibility to diabetic nephropathy is more directly Siblings of Siblings of
supported by familial clustering, with an estimated heritability of probands with probands without
at least 40%. In two studies of families in which two or more sib- diabetic nephropathy diabetic nephropathy
lings had T1DM, the risk of nephropathy in a diabetic sibling was • Fig. 37.22 Familial clustering of diabetic nephropathy. Prevalence of dia-
83% or 72% if the proband diabetic sibling had advanced diabetic betic nephropathy in two studies of diabetic siblings of probands with or
nephropathy, but only 17% or 22% if the index patient did not without diabetic nephropathy. (Adapted from Seaquist ER, Goetz FC, Rich
have diabetic nephropathy (Fig. 37.22).251,252 Familial clustering S, et al. Familial clustering of diabetic kidney disease: evidence for genetic
for the risk of severe diabetic retinopathy was also reported by the susceptibility to diabetic nephropathy. N Engl J Med. 1989;320:1161–
1165; Quinn M, Angelico MC, Warram JH, et al. Familial factors determine
DCCT. Likewise, familial aggregation of coronary artery calcifi-
the development of diabetic nephropathy in patients with IDDM. Diabeto-
cation occurs in families with T2DM.253 Numerous associations logia. 1996;39:940–945.)
have been made between various candidate gene polymorphisms
and the risk of diabetic complications. However, in a meta-analy-
sis of three large cohorts from the Genetics of Nephropathy—an While genome-wide association studies (GWAS) performed in
International Effort (GENIE), no associations remained signifi- large, well-characterized study populations have identified a few
cant after correction for multiple testing or application of strin- susceptibility loci for diabetic retinopathy, nephropathy, and car-
gent significance thresholds.254 diovascular disease,255 the cumulative effects of disease-associated

single nucleotide polymorphisms (SNPs) have failed to account with complications.44 Induced pluripotent stem cells (iPSCs) from
for the majority of complex-trait heritability for a large number fibroblasts of patients with no complications had low miR200 lev-
of diseases, including diabetes and its complications (the “missing els, while iPSCs from fibroblasts of patients with complications
heritability” problem). Ultimately, the central unanswered ques- (retinopathy, nephropathy, neuropathy, and cardiovascular disease)
tion is: How is coordinated gene expression regulatory informa- had high miR200 levels. miR200 was found to inhibit DNA dam-
tion conveyed—the information that determines when, where, at age checkpoint protein expression, causing accumulation of DNA
what level, in what combinations, and for how long, genes are double-strand breaks, inflammation, and apoptosis (Fig. 37.23).
expressed? With protein coding sequences representing approxi- Knockdown of miR200 in fibroblasts from patients with compli-
mately 2% of the genome, research in the post-GWAS era is cations rescued checkpoint protein expression and reduced DNA
currently focused on noncoding RNAs, genetic-epigenetic inter- damage. In differentiated neurons derived from all clinical back-
actions, and long-range chromatin 3D looping architecture.256,257 grounds, exogenous overexpression of miR200 produced the DNA
damage mark γH2AX phosphorylation (pH2AX). These data are
consistent with the model of ROS-mediated DNA damage activat-
Noncoding RNAs and Diabetic Complications ing multiple hyperglycemia/IR-induced pathogenic mechanisms
Although piwi-interacting RNAs (piRNAs), endogenous small discussed in previous sections.
interfering RNAs (siRNAs), intron-derived microRNAs (miR-
NAs), and a host of long noncoding RNAs all have regulatory Molecular Basis for Metabolic Memory
roles, the best understood noncoding RNAs with respect to
diabetic complications are the miRNAs, which regulate several Stable cellular phenotypes are maintained by a multitude of inter-
key biologic pathways and cellular functions involved in dia- connected multilevel positive and negative feedback loops,270,271
betic complications. Most individual miRNAs target hundreds interacting with combinatorial extracellular signals from adja-
of specific mRNAs,258 thereby coordinately regulating complex cent cells, from receptor ligands secreted by cells in other tissues,
gene networks. In the retina, diabetes increased levels of several and from changes in metabolite flux and concentration. Famil-
microRNA mediators of inflammation (miR146, miR155, and iar examples of stable alterations in cellular phenotype are the
miR132),259 while decreasing levels of the anti-inflammation induction of pluripotency in terminally differentiated cells, and
mediator miR146a. Levels of miR146a are also reduced in sci- the differentiation of these iPS cells into a variety of cell types.272
atic nerve of db/db mice. Treatment with a miR146a mimetic In preclinical models of T1DM and T2DM, diabetes irrevers-
reduced macrophage activation and improved nerve conduction ibly depletes two bone marrow–derived mesenchymal progenitor
velocity, intraepidermal nerve fiber density, and axonal degen- subpopulations having provasculogenic expression profiles, and
eration. miR146a was also decreased in kidneys and hearts of this deficit in vascular progenitor cells is not correctable by restor-
type 1 and type 2 diabetic animals.260 In diabetic mouse kid- ing glucose homeostasis in vitro or in vivo.273 These data suggest
ney cortex, levels of the profibrotic miR192 are also increased, that diabetes causes permanent modifications in subpopulations
and miR192 knockdown prevented diabetes-induced protein- of mesenchymal progenitor cells that are passed to daughter cells
uria and renal fibrosis.261 Diabetes also increased miR29c in the with each division.
kidney, which induces cell apoptosis and increases extracellular Similarly, when fibroblasts from patients with T1DM for more
matrix protein accumulation. Knockdown of miR29c prevented than 50 years with complications were used to make iPS cells,
high glucose–induced cell apoptosis and significantly reduced the miR200 family was upregulated in fibroblasts from patients
albuminuria and kidney mesangial matrix accumulation in db/ with complications, in the reprogrammed iPS cells from these
db mice.262 fibroblasts, and in neurons differentiated from iPS cells from the
miR33 regulates cellular functions associated with cardiovas- cohort with complications. These neurons had increased suscep-
cular disease. It increases macrophage activation, inhibits mito- tibility to genotoxic stressors, including exposure to high glucose
chondrial biogenesis, and represses autophagy.263 In mice placed and various sources of increased ROS. The miR200 family target
on an atherogenic diet for 16 weeks and then made diabetic by transcripts encoding DNA damage checkpoint proteins, and there
streptozotocin injection, anti-miR33 treatment decreased ath- was a concomitant increase in DNA damage and a loss of DNA
erosclerotic plaque macrophage content and inflammatory gene damage checkpoint proteins.44 These findings were confirmed in
expression. The decreased macrophage content in anti-miR33– aortic tissue from patients with complications and were reversed
treated diabetic mice was associated with a blunting of hypergly- in fibroblasts and iPS cells by knockdown of miR200. In fibro-
cemia-induced monocytosis and reduced monocyte recruitment blasts from patients with complications, there was also a striking
to the plaque.264 Diabetes increases circulating numbers of neu- impairment in reprogramming efficiency, and the induced iPCs
trophils and monocytes through an ROS-mediated mechanism had impaired cellular growth and differentiation. In tissue from
and impairs regression of early atherosclerosis by increasing mac- patients with complications, there was increased inflammation due
rophage entry into the lesion in mouse models.265,266 to macrophage infiltration. There was also increased apoptosis due
MicroRNAs may also play a role in the pathogenesis of diabetic to elevated levels of DNA damage, which was not compensated
cardiomyopathy. miR499, miR133a, and miR373 are downregu- by differentiation of progenitor cells. Taken together, the obser-
lated in diabetic cardiomyocytes.267 Downregulation of miR133a vations from both these studies could reflect underlying genetic
in a normal adult genetic background was sufficient to induce car- or epigenetic causes. However, since the DCCT study patients
diac hypertrophy, and its downregulation is a prerequisite for the were randomly assigned to one of two different treatment groups,
development of apoptosis, fibrosis, and prolongation of the QT treatment-related epigenetic causes are a more likely explanation
interval in animal models.268 The protective capacity of cardiac for the long-term metabolic memory observed subsequently in the
progenitor cells after myocardial infarction is also enhanced by EDIC study.
miR133a.269 Epigenetic causes are those that change DNA or RNA struc-
In patients with over 50 years of T1DM, expression levels of ture and function, but do not change DNA-RNA sequence. Along
miR200 differentiated those with no complications from those with the activities of microRNAs and other noncoding RNAs, a

Genotoxic stress

DNA double strand break

Controls or Medalists Medalists

Without Complications With Complications
High DNA repair Low DNA repair
machinery machinery

Low miR200 High miR200

ATM DNA-Pk ATM DNA-Pk

Ub Ub
Ub ATR Ub ATR
Ub PARP1 MRE11 Ub PARP1 MRE11
Me Co-regulators POL II RAD50 MDC1 Ac Me Co-regulators POL II RAD50 MDC1 Ac
Me Ac Me Ac
Chromatin NBS1 pHA2X Transcription Chromatin NBS1 pHA2X Transcription
modifiers MED15/17 modifiers MED15/17

EZH2, SUZ12, Chromatin EZH2, SUZ12, Chromatin

DNMTs, HDACs, etc DNMTs, HDACs, etc

Fixing of DNA errors leading to Accumulation of DNA errors,

normal cell cycle and growth inflammation leading to apoptosis
• Fig. 37.23 Expression levels of microRNA (miR) miR200 from patients with over 50 years of T1DM differ-
entiate those with no complications from those with complications. Induced pluripotent stem cells (iPSCs)
from fibroblasts of patients with no complications had low miR200 levels, while iPSCs from fibroblasts
of patients with complications (retinopathy, nephropathy, neuropathy, and cardiovascular disease) had
high miR200 levels. miR200 was found to inhibit DNA damage checkpoint protein expression, caus-
ing accumulation of DNA double-strand breaks, inflammation, and apoptosis. ATM, Ataxia telangiectasia
mutated; ATR, ataxia telangiectasia and Rad3‐related protein; DNA-PK, DNA-dependent protein kinase;
DNMTs, DNA methyltransferases; EZH2, enhancer of zeste homolog 2; HDACs, histone deacetylases;
MDC1, mediator of DNA damage checkpoint 1; Me, methyl; MED 15/17, mediator of RNA polymerase II
transcription subunits 15 and 17; MRE11, MRE11 homolog, double-strand break repair nuclease; NBS1,
Nijmegen breakage syndrome protein 1; PARP1, poly-ADPribose polymerase 1; pH2AX, phosphorylated
H2A histone family member X; POL II, RNA polymerase II; RAD 50, DNA repair protein RAD50; SUZ12,
suppressor of zeste 12 protein homolog; Ub, ubiquitin. (Modified from Bhatt S, Gupta MK, Khamaisi M,
et al. Preserved DNA damage checkpoint pathway protects against complications in long-standing type 1
diabetes. Cell Metab. 2015;22:239–252.)

variety of enzymes modify chromatin by adding and removing though the post-DCCT HbA1c values for both intensive and
chemical groups to DNA (primarily cytosine methylation) and to standard treatment groups became identical at the end of the
the protruding N-terminal tails of nucleosome histone proteins DCCT (∼8%), and remained identical for the duration of the
(monomethylation, dimethylation, and trimethylation of lysine EDIC study. The most significant persistent DNA methylation
and arginine residues, acetylation of lysine residues, and modifica- change was hypomethylation of thioredoxin-interacting protein
tion of arginines by methylglyoxal and of serines and threonines (TXNIP).278 TXNIP binds to and inhibits thioredoxin, increas-
by O-GlcNAc).190,274 In conjunction with remodeling complexes, ing ROS production and inhibiting renal tubular autophagy.279
these modifications mediate the availability or nonavailability of The importance of this DNA methylation profiling study is that
DNA sequences for transcription (Fig. 37.24). Increased ROS it shows, for the first time, that differential methylation at several
modulate several epigenetic mechanisms in different cell types and loci persists for more than 16 to 17 years in circulating leukocytes
contexts.275–277 from the same diabetic patient cohort. Every diabetic complica-
DNA methylation can repress or activate transcription, tion is a complex heterocellular process, however, and each dif-
depending on its location near promoters or near enhancers in ferent cell type involved has a distinct epigenome regulating gene
intergenic regions.275 Similarly, methylation of histone tails can expression and cell type–specific pathologies despite sharing a
repress or activate transcription, depending on the position of common genetic sequence.275 A similar but technically more lim-
the modified residue in a specific histone tail, and on the degree ited study of selected histone modifications in monocytes showed
of methylation. Acetylation of histone tails generally facilitates enrichment of activating histone 3 lysine 9 acetylation (H3K9Ac)
transcription. in promoters of more than 15 genes related to the NFκB inflam-
In whole blood and isolated monocyte samples from DCCT/ matory pathway.280
EDIC type 1 diabetic patients at the EDIC study baseline and Transient hyperglycemia, at a level sufficient to increase mito-
in samples from the same patients obtained during EDIC years chondrial ROS production, induces long-lasting activating epi-
16 to 17, DNA methylation profiling revealed strong connec- genetic changes (increased monomethylation of histone 3 lysine
tions with networks associated with diabetic complications, even 4) in the proximal promoter of the NFκB subunit p65 in vitro in

Glucose/fatty acids

Reactive oxygen species

Epigenetic landscape

DNA Histone noncoding RNAs ATP-dependent

modification modification (miRNA, IncRNA) chromatin remodelling

Ac
5hmC Met

ADP
8OG 5mC O-GlcNAc Ribo
MGO

Gene expression (nuclear/mitochondrial)

Diabetic complications

• Fig. 37.24 Increased metabolite-generated reactive oxygen species (ROS) modulate the epigenetic landscape,
altering histone modifications, DNA modifications, expression of noncoding RNAs, and ATP-dependent chro-
matin remodeling. These changes subsequently affect gene expression patterns implicated in the pathogenesis
of diabetic complications. DNA modifications include cytosine methylation (5mC), hydroxymethylation (5hmC),
or 8-oxo-2’-deoxyguanosine (8OG) formation. Histone modifications include methylation (Met), acetylation
(Ac), ADP-ribosylation (ADP-Ribo), phosphorylation (P), glycation by methylglyoxal (MGO), and glycosylation by
O-linked N-acetylglucosamine (O-GlcNAc). Noncoding RNAs include microRNAs (miRNA) and long noncoding
RNAs (lncRNA). ATP-dependent chromatin remodeling includes moving and adding/removing nucleosomes by
ATPase-containing complexes. (Modified from Kietzmann T, Petry A, Shvetsova A, et al. The epigenetic landscape
related to reactive oxygen species formation in the cardiovascular system. Br J Pharmacol. 2017;174:1533–1554.)

human aortic endothelial cells (16-hour exposure), and in vivo in and facilitate the formation of transcriptional complexes. In endo-
aortic cells in nondiabetic mice (6-hour exposure). These epigen- thelial cells, proinflammatory activation of NFκB led to rapid
etic changes caused sustained increases in p65 gene expression and recruitment of NFκB to newly formed superenhancers, and rapid
in the expression of p65-dependent proinflammatory genes. Both large-scale redistribution of the epigenetic reader bromodomain
the epigenetic changes and the gene expression changes persist for and extraterminal domain 4 (abbreviated BET4 or BRD4) from
at least 6 days of subsequent normal glycemia in cultured cells and decommissioned preexisting basal enhancers to the newly formed
for months in previously diabetic mice whose beta-cell function superenhancers.286 This recruitment caused increased levels of
had recovered.281,282 Hyperglycemia-induced epigenetic changes H3K27ac and immediate transcription of inflammatory genes.
and increased p65 expression are prevented by normalizing mito- Blocking the association of BRD4 with acetylated lysines in LDL
chondrial superoxide production or superoxide-induced methylg- receptor–deficient hypercholesterolemic mice attenuated both
lyoxal. Demethylation of another histone lysine residue, histone early and late atherosclerosis development.
3 lysine 9, is also induced by hyperglycemia-induced overpro- BRD4 and other members of the BET family are also critical
duction of ROS. This reduces inhibition of p65 gene expression, effectors of pathologic cardiac remodeling and heart failure via
and therefore acts synergistically with the activating methylation their ability to coactivate multiple master transcription factors
of histone 3 lysine 4. The specific H3K4m1 enrichment is medi- known to initiate and promote heart failure, including NFκB
ated by the lysine methyltransferase Set7, which is recruited to the and NFAT.287 BET inhibition suppresses pathologic cardiac
nucleus by high glucose–induced mechanisms.283 Genome-wide gene expression programs and arrests pathologic hypertrophy
analysis of Set7 function in human vascular endothelial cells con- and heart failure in vivo. BRD4 (most potently its splice isoform
firmed its role in regulating NFκB-dependent pathways, affecting BRD4B) also functions as an endogenous inhibitor of DNA dam-
a large number of genes associated with vascular function by both age response signaling, which enhances radiation ROS-induced
histone lysine methylation and lysine methylation on nonhistone cell death. In contrast, knockdown of this isoform caused rapid
substrates such as transcription factors.284,285 The demethylation cell-cycle checkpoint recovery and enhanced cell survival.288
of histone 3 lysine 9 is mediated by the histone demethylase LSD1 Much has been learned since the original clinical description
(lysine-specific histone demethylase 1) (Fig. 37.25). of metabolic memory, but much more research remains to be
Another component of epigenetic regulation is the action of done before the molecular basis for metabolic memory is clearly
epigenetic reader proteins that bind to modified histone residues understood.

K
of care for most eyes with visual impairment from center-involved
T Q T DME.290,291 Currently, more than 90% of severe vision loss result-
R K
A ing from PDR can be prevented, and approximately 50% of eyes
m m with visual loss from DME can resolve their retinal thickening
and/or recover vision of 20/20 or better with appropriate medi-
Set 7 LSD1
cal and ophthalmologic care.292,292a Thus the primary clinical care
Histone methyl Histone emphasis for the prevention of vision loss in the diabetic patient
transferase demethylase is on early identification, accurate classification, and timely treat-
ment of retinopathy through lifelong routine ophthalmologic fol-
low-up appointments and optimized treatment of hyperglycemia,
hypertension, and hyperlipidemia.
K
R
T Q T
K
Epidemiology and Impact of Proliferative
A
m Diabetic Retinopathy and Diabetic Macular
• Fig. 37.25 Transient hyperglycemia induces long-lasting activating epi-
Edema
genetic changes in the proximal promoter of the NFκB subunit p65 in There is a higher risk of more frequent and severe ocular com-
endothelial cells. Hyperglycemia-induced reactive oxygen species (ROS) plications in T1DM than T2DM.293 Approximately 25% of
and methylglyoxal cause activating modifications of histone 3 lysine 4 patients with T1DM have retinopathy after 5 years, and this
(monomethylation) and derepressing modifications of histone 3 lysine 9 figure increases to 60% and 80% after 10 and 15 years, respec-
(removal of two methyl groups) at the NFκB p65 proximal promoter. K is
tively. The most visually threatening form of retinopathy (PDR)
the symbol for the amino acid lysine. The chains of circled letters are the
N-terminal tails of histone H3. LSD1, lysine-specific histone demethylase
is present in about 67% of T1DM patients who have had dia-
1A; Set 7, SET domain-containing protein 7. (From Brasacchio D, Okabe J, betes for 35 years.294 However, because T2DM accounts for
Tikellis C, et al. Hyperglycemia induces a dynamic cooperativity of histone 90% to 95% of the diabetic population in the United States,
methylase and demethylase enzymes associated with gene-activating type 2 disease accounts for a higher fraction of patients with
epigenetic marks that co-exist on the lysine tail. Diabetes. 2009;58:1229– vision loss.
1236; El-Osta A, Brasacchio D, Yao D, et al. Transient high glucose causes In the United States, an estimated 700,000 people have PDR,
persistent epigenetic changes and altered gene expression during subse- 130,000 of whom have high-risk PDR. Among people with
quent normoglycemia. J Exp Med. 2008;205:2409–2417.) diabetes, 500,000 have macular edema, of whom 325,000 have
clinically significant macular edema (CSME).295–299 An estimated
63,000 cases of PDR, 29,000 cases of high-risk PDR, 80,000 cases
Diabetic Retinopathy and Other Ocular of macular edema, 56,000 cases of CSME, and 12,000 to 24,000
Complications of Diabetesa new cases of legal blindness occur each year as a result of diabetic
retinopathy.295,296,300 Blindness is estimated to be 25 times more
Diabetic retinopathy is a well-characterized sight-threatening common in persons with diabetes than in those without the dis-
chronic microvascular complication that eventually afflicts vir- ease.301,302 Rates of progression to PDR or severe visual loss have
tually all patients with diabetes mellitus.289 Diabetic retinopa- declined over the last four decades in developed countries due to
thy is characterized by gradually progressive alterations in the improvements in medical management, earlier identification, and
retinal microvasculature, leading to areas of retinal nonperfusion, treatment of diabetic retinopathy with laser photocoagulation.303
increased vascular permeability, and pathologic intraocular pro- However, given the dramatically increasing prevalence of diabetes
liferation of retinal vessels. The complications associated with worldwide, with 629 million individuals expected to be affected
retinal neovascularization, termed proliferative diabetic retinopathy by diabetes by the year 2045,304 a proportionally larger number
(PDR), and increased vascular permeability, termed diabetic mac- of individuals will be at risk for vision loss from diabetic eye com-
ular edema (DME), can result in severe and permanent vision loss. plications over the next few decades. Threefold increases in vision-
Early-stage diabetic retinopathy eventually occurs in nearly threatening diabetic retinopathy will affect 3.4 million Americans
all patients, although early intensive treatment of hyperglycemia aged 40 and older by 2050, as estimated by the Centers for Dis-
delays its onset and progression in T1DM. In selected patients ease Control and Prevention (CDC).306
with T2DM, systemic treatment with fenofibrate, ACE inhibi- Many individuals with diabetes and vision-threatening dia-
tors, and angiotensin II receptor blockers (ARBs) may affect the betic retinopathy are not aware of the presence or the severity of
development and progression of diabetic retinopathy. Multiple their disease. The 2010 National Health and Nutrition Exami-
phase 3 clinical trials over the past 5 years have now established nation Survey revealed that only 45% of Americans with DME
intravitreously delivered anti-VEGF therapy as the new standard were aware that diabetes had affected their eyes and nearly 60% of
these individuals had not had a dilated eye examination within the
a Portions of this section draw on, among others, Aiello LM, Cavallerano JD, past year. These findings reflect both a lack of awareness among
Aiello LP. Diagnosis, management, and treatment of nonproliferative diabetic patients at risk for vision loss from diabetic eye complications and
retinopathy and diabetic macular edema. In: Albert DM, Jakobiec FA, eds. insufficient evaluation for many patients with vision-threatening
Principles and Practice of Ophthalmology. 2nd ed. Philadelphia: WB Saunders; retinopathy.306
2000:1900–1914; Aiello LP, Cavallerano J, Klein R. Diabetic eye disease. In
DeGroot LJ, James JL, eds. Endocrinology, 5th ed. Philadelphia: WB Saun-
ders; 2005:1305–1317; Aiello LP, Gardner TW, King GL, et al. Diabetic Pathophysiology of Diabetic Retinopathy
retinopathy: technical review. American Diabetes Association. Diabetes Care.
1998;21:143–156; Aiello LP, Cavallerano J. Diabetic retinopathy. In John- A detailed discussion of the pathophysiologic mechanisms under-
stone MT, Veves A, eds. Contemporary Cardiology: Diabetes and Cardiovascular lying diabetic retinopathy and other diabetes-related compli-
Disease. Totowa, NJ: Humana Press; 2001:385–398. cations has been presented earlier in this chapter. The earliest

Diabetes

No
High glucose diabetic
retinopathy

Pericyte loss
Basement membrane
thickening
Retinal blood flow change

Microaneurysm formation Retinal vascular

leakage

Macular edema Mild

(may occur at any stage) nonproliferative
diabetic
retinopathy
Moderate visual loss (NPDR)
Retinal hemorrhages

Capillary dropout Moderate

NPDR

Retinal ischemia

Increased growth factors (e.g., VEGF) Severe

NPDR

Retinal neovascularization Iris neovascularization

Vascular fibrosis Vitreous hemorrhage Neovascular glaucoma

Retinal traction
Proliferative
Mild to severe Severe
diabetic
visual loss visual loss
Retinal detachment retinopathy

Severe
visual loss

• Fig. 37.26 Pathogenesis of diabetic retinopathy. This schematic flow chart represents the major pre-
clinical and clinical findings associated with the full spectrum of diabetic retinopathy and macular edema.
VEGF, Vascular endothelial growth factor.

histologic effects of diabetes mellitus in the eye include loss of blood and serum from the retinal vessels results in retinal hemor-
retinal vascular pericytes (supporting cells for retinal endothelial rhages, retinal edema, and hard exudates (see Fig. 37.27A and C).
cells), thickening of vascular endothelium basement membrane, Vision loss can follow if the fovea is affected by the leakage.316
and alterations in retinal blood flow (Fig. 37.26).307–310 With With time, increasing sclerosis and endothelial cell loss lead to
increasing loss of retinal pericytes, the retinal vessel wall develops narrowing of the retinal vessels, which decreases vascular perfusion
outpouchings (microaneurysms) and becomes fragile. and can ultimately lead to obliteration of the capillaries and small
Clinically, microaneurysms and small retinal hemorrhages vessels (see Fig. 37.27B). The resulting retinal ischemia is a potent
might not always be readily distinguishable from each other inducer of angiogenic growth factors. Several angiogenic growth
and are usually evaluated together as “hemorrhages and micro- factors have been isolated from eyes with diabetic retinopathy,
aneurysms” (Fig. 37.27A). Rheologic changes occur in diabetic including insulin-like growth factors, basic fibroblast growth fac-
retinopathy and result from increased platelet aggregation, integ- tor (bFGF), and VEGF.317,318 These factors promote the develop-
rin-mediated leukocyte adhesion, and endothelial damage.311–313 ment of new vessel growth and retinal vascular permeability.319,320
Disruption of the blood-retina barrier can ensue, characterized Indeed, inhibition of angiogenic molecules such as VEGF and
by increased vascular permeability.314,315 Subsequent leakage of their signaling pathways can suppress the development of retinal

Non-perfusion

^ ^
Ma Non-perfusion
IRMA Hard exudate>
^ <Fovea Fovea>
IRMA ^ <Optic disc
^
VB
VB
Non-perfusion Macular
edema
<Hard exudate
^
Ma Non-perfusion
A B C

NVD>
Extensive Vitreous
NVE hemorrhage
NVD

VH VH
D E F

<Fibrous tissue
Fibrous> <Fovea Optic disc>
tissue ^Optic disc Fovea>

Fibrous
tissue

G H I
• Fig. 37.27 Clinical features of diabetic retinopathy: Some typical findings in human diabetic retinopathy.
(A) Findings in severe nonproliferative diabetic retinopathy, including microaneurysms (Ma), venous bead-
ing (VB), and intraretinal microvascular abnormalities (IRMA). (B) Fluorescein angiogram showing marked
capillary nonperfusion. (C) Clinically significant macular edema with retinal thickening and hard exudates
involving the fovea. (D) Extensive neovascularization of the optic disc (NVD), illustrating high-risk prolifera-
tive diabetic retinopathy. (E) Neovascularization elsewhere (NVE) and two small vitreous hemorrhages (VH),
also illustrating high-risk proliferative diabetic retinopathy. (F) Extensive vitreous hemorrhage arising from
severe neovascularization of the disc (NVD). (G) Severe fibrovascular proliferation surrounding the fovea.
(H) Traction retinal detachment from extensive fibrovascular proliferation. (I) Scars from scatter (panreti-
nal) laser photocoagulation. The macula, fovea, and optic disc are not treated to preserve central vision.
Laser burns are evident as white retinal lesions. (Adapted from Aiello LP. Eye complications of diabetes. In
Korenman SG, Kahn CR, eds. Atlas of Clinical Endocrinology. Vol 2: Diabetes. Philadelphia, PA: Blackwell
Scientific; 1999.)

neovascularization and retinal vascular permeability.321,322 VEGF- concentration in the anterior chamber to cause abnormal new ves-
independent endogenous inhibitors of angiogenesis and vascular sel proliferation on the iris and the anterior chamber angle.318,325
permeability such as pigment epithelial-derived factor (PEDF), Uncontrolled anterior segment neovascularization can result in
plasma kallikrein, and erythropoietin have also been found in the neovascular glaucoma because the fibrovascular proliferation in
eye, and these have therapeutic potential.323,324 the angle of the eye causes blockage of aqueous outflow through
New vessels tend to grow in regions of strong vitreous adhe- the trabecular meshwork.305
sion to the retina, such as at the optic disc and major vascular Proliferating new vessels in diabetic retinopathy are fragile and
arcades (see Fig. 37.27D and E). The posterior vitreous face also have a tendency to bleed, which results in preretinal and vitreous
serves as a scaffold for pathologic neovascularization, and the hemorrhages (see Fig. 37.27E and F). Although the presence of a
new vessels commonly arise at the junctions between perfused large amount of blood in the preretinal space or vitreous cavity is
and nonperfused retina. When the retina is severely ischemic, the not damaging to the retina, these intraocular hemorrhages often
concentration of angiogenic growth factors can reach sufficient cause vision loss by blocking the visual axis. Vitreous hemorrhage

can clear spontaneously without intervention, but eyes in which to narrow, and became statistically nonsignificant from 5 years
hemorrhage is nonclearing may need vitrectomy surgery to restore onward (8.1% vs. 8.2%, p = 0.09). The rate of further progression
vision. Vitreous hemorrhage can also decrease the ability to visual- of complications from their levels at the end of the DCCT remains
ize the retina and thereby limit the ability to adequately diagnose less in the former intensive treatment group. Thus the benefits of
and treat other retinal disease. Membranes on the retinal surface 6.5 years of intensive treatment continue to accrue well beyond the
can be induced by blood and result in wrinkling and traction on period of initial HbA1c difference.9,22,330,333–335 Beyond retinopa-
the retina. Although all retinal neovascularization, given sufficient thy progression itself, the risk for diabetes-related ocular surgery
time, eventually becomes quiescent, as with most scarring pro- was reduced 48% in the intensive therapy group336 over a median
cesses there is progressive fibrosis of the new vessel complexes that follow-up of 23 years. Applying DCCT intensive insulin therapy
is associated with contraction. In the eye, such forces can exert to all persons in the United States with T1DM would result in
traction on the retina, leading to tractional retinal detachment and a gain of 920,000 person-years of sight,331 although the costs of
retinal tears that can cause severe and permanent vision loss if left intensive therapy are three times that of conventional therapy.337
untreated (see Fig. 37.27G and H). Presence of renal disease, as manifested by microalbuminuria
In short, causes of vision loss from complications of diabetes and proteinuria, is yet another significant risk factor for onset and
mellitus include retinal ischemia involving the fovea, macular progression of diabetic retinopathy.338,339 Hypertension is associ-
edema at or near the fovea, preretinal or vitreous hemorrhage, ated with PDR and is an established risk factor for the develop-
retinal detachment, and neovascular glaucoma. Vision loss can ment of macular edema.340 Independent of renal disease, elevated
also result from more indirect effects of vasculopathy in diabetic serum lipid levels are associated with extravasated lipid in the
patients, such as retinal vessel occlusion, accelerated atheroscle- retina (hard exudates) and vision loss.341
rotic disease, and embolic phenomena.
Clinical Findings
Clinical Features of Diabetic Retinopathy Clinical findings associated with early and progressing diabetic reti-
nopathy include microaneurysms, hemorrhages, cotton-wool spots,
Risk Factors hard exudates, intraretinal microvascular abnormalities, and venous
Duration of diabetes is closely associated with the onset and sever- abnormalities such as venous loops, venous tortuosity, and venous
ity of diabetic retinopathy. Nearly all patients with T1DM develop beading (see Fig. 37.27A and C). Microaneurysms are outpouch-
some degree of retinopathy after 20 years.293,296,326 In the United ings of the capillary walls that can leak fluid and result in intrareti-
States approximately 20% of patients with newly diagnosed nal edema and hemorrhages. The intraretinal hemorrhages can be
T2DM have retinopathy at the time of diagnosis,326 and more flame shaped or dot-blot–like in appearance, reflecting the archi-
than 60% develop some degree of retinopathy over subsequent tecture of the retinal layer in which they occur. Flame-shaped hem-
decades. In the UKPDS study of T2DM, 35% of female subjects orrhages occur in inner retina closer to the vitreous, and dot-blot
and 39% of male subjects had some level of diabetic retinopathy hemorrhages occur deeper in the retina. Intraretinal microvascular
at the time of diabetes diagnosis.327 abnormalities are either new vessel growth within the retinal tissue
Age of onset of diabetes is another risk factor. Diabetic reti- itself or shunt vessels through areas of poor vascular perfusion. It is
nopathy is rare in prepubescent patients with T1DM.87 In the common for intraretinal microvascular abnormalities to be located
Wisconsin Epidemiologic Study of Diabetic Retinopathy, approx- adjacent to cotton-wool spots. Cotton-wool spots are caused by
imately 4% of patients younger than 30 years of age at diagnosis microinfarcts in the nerve fiber layer of the retina. Venous caliber
and nearly 2% of patients older than 30 years of age at diagnosis abnormalities, also known as venous beading, are generally a sign of
were legally blind. In the younger-onset group, 86% of blindness severe retinal hypoxia. In some cases of extensive vascular loss, how-
was attributable to diabetic retinopathy. In the older-onset group, ever, the retina may actually appear free of nonproliferative lesions
in which other eye diseases were also common, 33% of the cases of on clinical examination. Such areas are termed featureless retina and
legal blindness were due to diabetic retinopathy.294,326,328 are a sign of severe retinal hypoxia.
Lack of appropriate glycemic control is the most significant Vision loss from diabetic retinopathy generally results from per-
known risk factor for the onset and progression of diabetic reti- sistent nonclearing vitreous hemorrhage, traction retinal detach-
nopathy. The DCCT demonstrated a clear relationship between ment, retinal nonperfusion, or DME (see Figs. 37.26 and 37.27).
hyperglycemia and higher risk of diabetic microvascular complica- Neovascularization with fibrous tissue contraction can distort the
tions, including retinopathy, in 1441 patients with T1DM.9,329–332 retina and lead to traction retinal detachment. The new vessels can
Intensive insulin therapy reduced or prevented the development bleed, causing preretinal or vitreous hemorrhage. The most com-
of retinopathy by 27% as compared with conventional therapy in mon cause of vision loss from diabetes, however, is macular disease
patients monitored for 4 to 9 years in the DCCT. Additionally, and macular edema. Macular edema is more likely to occur in
intensive insulin therapy reduced the progression of diabetic reti- patients with T2DM. In diabetic macular disease, macular edema
nopathy by 34%, to 76%, and had a substantial beneficial effect involving the fovea or nonperfusion of the capillaries in the central
over the entire range of retinopathy severity. These results under- macula is responsible for the loss of vision.
score the fact that although intensive therapy might not prevent
retinopathy completely, it reduces the risk of retinopathy onset Classification Systems
and progression. Classification of Diabetic Retinopathy
Interestingly, the effect of reducing the HbA1c in this group, Diabetic retinopathy is broadly classified into nonproliferative
from 9.1% for conventional treatment to the 7.3% for intensive diabetic retinopathy (NPDR) and proliferative diabetic retinopathy
treatment, resulted in a benefit maintained through 18 years of categories.342,343 Macular edema can be present in eyes with non-
subsequent follow-up in the EDIC study, even though the dif- proliferative or proliferative disease and is not used in the clas-
ference in mean HbA1c levels of the two former randomized sification of level of retinopathy. The historic terms background
treatment groups was only 0.4% at 1 year (p < 0.001), continued retinopathy and preproliferative diabetic retinopathy have been

replaced to reflect the specific characteristics and risk stratification such as hemorrhages or microaneurysms, cotton-wool spots, and
of the prognostically important subgroups in NPDR (Table 37.1). increased vascular permeability. Moderate and severe NPDR is
Generally, diabetic retinopathy progresses from no retinopa- characterized by increasing severity of hemorrhages or microan-
thy through mild, moderate, severe, and very severe NPDR and eurysms, venous caliber abnormalities, intraretinal microvascular
eventually to PDR. The level of NPDR is determined by the abnormalities, and vascular closure. The level of NPDR establishes
extent and location of clinical manifestations of retinopathy. Mild the risk of progression to sight-threatening retinopathy and dic-
NPDR is characterized by limited microvascular abnormalities tates appropriate clinical management and follow-up.

TABLE 37.1 G
lossary and Abbreviations Pertinent to Diabetic Eye Disease
Term Definition
Antivascular endothelial growth For the purposes of this chapter, these refer to VEGF inhibitors (including aflibercept, bevacizumab, and ranibizumab)
factor (anti-VEGF) therapies; that are given intravitreously for the treatment of diabetic macular edema and proliferative diabetic retinopathy.
VEGF inhibitors
Background diabetic retinopathy An outdated term referring to some stages of NPDR; not closely associated with disease progression; replaced by the
(BDR) various levels of NPDR.
Center-involved (or central- Abnormal thickening of the central retina (usually 1 mm in diameter central retinal subfield) due to increased vascular
involved) diabetic macular permeability. Because central involvement is more likely to cause visual impairment, it is commonly used as a
edema (ciDME) threshold for treatment.
Clinically significant macular Thickening of the retina in the macular region of sufficient extent and location to threaten central visual function.
edema (CSME)
Cotton-wool spot A gray or white area lesion in the nerve fiber layer of the retina resulting from stasis of axoplasmic flow caused by
microinfarcts of the retinal nerve fiber layer.
Diabetes Control and Complications A multicenter, randomized clinical trial designed to address whether intensive insulin therapy could prevent or slow
Trial (DCCT) the progression of systemic complications of diabetes mellitus.
Diabetic retinopathy (DR) Retinal damage related to the underlying systemic disease of diabetes mellitus.
Diabetic Retinopathy Study (DRS) The first multicenter, randomized clinical trial to demonstrate the value of scatter (panretinal) photocoagulation in
reducing the risk of vision loss among patients with all levels of diabetic retinopathy.
Diabetic Retinopathy Vitrectomy A multicenter clinical trial evaluating early vitrectomy for patients with very advanced diabetic retinopathy or nonre-
Study (DRVS) solving vitreous hemorrhage.
Early Treatment Diabetic Retinopathy A multicenter, randomized clinical trial that addressed at what stage of retinopathy scatter (panretinal) photocoagula-
Study (ETDRS) tion was indicated, whether focal photocoagulation was effective for preventing moderate vision loss due to clini-
cally significant macular edema, and whether aspirin therapy altered the progression of diabetic retinopathy.
Focal or grid laser photocoagulation A type of laser treatment whose main goal is to reduce vascular leakage, either by focal treatment of leaking retinal
microaneurysms or by application of therapy in a gridlike pattern for patients with clinically significant macular
edema.
Hard exudate Lipid accumulation within the retina as a result of increased vasopermeability.
High-risk characteristic proliferative Proliferative diabetic retinopathy of defined extent, location, or clinical findings that is particularly associated with
diabetic retinopathy (HRC-PDR) severe vision loss.
Microaneurysm An early vascular abnormality consisting of an outpouching of the retinal microvasculature.
Neovascular glaucoma (NVG) Elevation of intraocular pressure caused by the development of neovascularization in the anterior segment of the eye.
Neovascularization at the disc (NVD) Retinal neovascularization occurring at or within 1500 μm of the optic disc.
Neovascularization elsewhere (NVE) Retinal neovascularization that is located more than 1500 μm away from the optic disc.
Neovascularization of the iris (NVI) Neovascularization occurring on the iris (rubeosis iridis), usually as a result of extensive retinal ischemia.
No light perception (NLP) The inability to perceive light.
Nonproliferative diabetic retinopathy Severities (mild, moderate, severe) of clinically evident diabetic retinopathy that precede the development of PDR.
(NPDR)
Preproliferative diabetic retinopathy An outdated term referring to more advanced levels of NPDR; not closely associated with disease progression;
(PPDR) replaced by the various levels of NPDR.
Proliferative diabetic retinopathy An advanced level of diabetic retinopathy in which proliferation of new vessels or fibrous tissue occurs on or within
(PDR) the retina.
Rubeosis iridis Neovascularization of the iris.

PDR is characterized by vasoproliferation of the retina and its vitreous hemorrhage, or preretinal hemorrhage). Table 37.2 com-
complications, including new vessels on the optic disc (NVD), pares levels of retinopathy in the international classification to
new vessels elsewhere on the retina (NVE), preretinal hemor- those defined by the landmark ETDRS.
rhage (PRH), vitreous hemorrhage, and fibrous tissue prolifera- In regard to DME, the international classification identifies two
tion (FP). On the basis of the extent and location of these lesions, broad categories: macular edema apparently absent (no apparent
PDR is classified as early PDR or high-risk PDR. Larger areas of retinal thickening or hard exudates in the posterior pole) and macu-
these complications as well as new vessels that are near the optic lar edema apparently present (some apparent retinal thickening or
disc are associated with greater risks of vision loss. hard exudates in the posterior pole). Macular edema is subclassified
as mild (some retinal thickening or hard exudates in the posterior
Classification of Diabetic Macular Edema pole but distant from the center of the macula), moderate (retinal
DME can be present at any severity of diabetic retinopathy. His- thickening or hard exudates approaching the center of the macula
torically, eyes with DME were categorized as having nonclini- but not involving the center), or severe (retinal thickening or hard
cally significant or clinically significant macular edema. The term exudates involving the center of the macula). Table 37.3 compares
CSME was first introduced in the Early Treatment Diabetic Reti- levels of DME in the international classification to ETDRS levels
nopathy Study (ETDRS) to indicate an increased risk for moder- of DME.
ate visual loss and was used as a threshold to determine need for As compared with ETDRS retinopathy grading, the Interna-
laser treatment. CSME exists if there is retinal thickening at or tional Classification of Diabetic Retinopathy and Diabetic Macu-
within 500 μm of the fovea, hard exudates at or within 500 μm of lar Edema reduces the number of levels of diabetic retinopathy,
the fovea with adjacent retinal thickening, or an area or areas of simplifies descriptions of the categories, and describes the levels
retinal thickening one disc area or more in size, any part of which
is within 1500 μm of the fovea.342,344,345 CSME is a clinical diag-
nosis that is not dependent on visual acuity or results of ancillary TABLE 37.2 L
evels of Diabetic Retinopathy
testing such as fluorescein angiography and can be present even
when vision is 20/20 or better. International
In the recent era, with the advent of ocular coherence tomogra- Classification Level ETDRS Level
phy (OCT) and its objective quantitative measurement of retinal
No apparent retinopathy Level 10: DR absent
thickness, clinical care and clinical trial endpoints have shifted to
evaluating whether the center of the macula is involved or not. Mild NPDR Level 20: very mild NPDR
Data from the ETDRS evaluating eyes with macular edema have
Moderate NPDR Levels 35, 43, and 47: moderate NPDR
shown the presence or absence of thickening involving the center
of the macula, now termed center-involved DME, which is highly Severe NPDR Levels 53A–E: severe to very severe NPDR
associated with short-term and long-term visual acuity outcomes. PDR Levels 61, 65, 71, 75, 81, 85: PDR, high-risk
In the ETDRS, eyes with center-involved DME had nearly a PDR, very severe or advanced PDR
10-fold greater risk for developing moderate visual loss compared
to eyes without center involvement. The identification of center- DR, Diabetic retinopathy; ETDRS, Early Treatment Diabetic Retinopathy Study; NPDR, nonpro-
involved DME is critical because it will generally indicate how liferative diabetic retinopathy; PDR, proliferative diabetic retinopathy.
likely the DME is to cause visual impairment and is a key crite- From Grading diabetic retinopathy from stereoscopic color fundus photographs—an exten-
sion of the modified Airlie House classification. ETDRS report number 10. Early Treatment
ria for determining the need for treatment. Noncenter-involved
Diabetic Retinopathy Study Research Group. Ophthalmology. 1991;98:S786–S806.
DME does not typically cause visual loss or symptoms, and fre-
quently does not progress to center-involved DME. Thus if the
center of the macula is not involved there is often not a compel-
ling reason to treat. As a result, recent clinical studies and guide-
TABLE 37.3 International Classification of Diabetic
lines for clinical care have generally used the presence or absence
of thickening of the central 1-mm diameter of retina, along with Macular Edema (DME)
visual status, as the defining criteria to consider initiation of treat- Ophthalmic Findings (Retinal
ment for eyes with DME. Disease Thickening or Hard Exudates ETDRS Scale
Severity Level in Posterior Pole) Equivalent
International Classification of Diabetic Retinopathy and
Diabetic Macular Edema DME apparently None apparent
The American Academy of Ophthalmology initiated a project to absent
establish a consensus International Classification of Diabetic Reti- DME apparently Some apparent
nopathy and Diabetic Macular Edema in an effort to simplify clas- present
sification and standardize communication among diabetes health Mild DME Some findings present but distant DME but not
care providers.346,347 This international classification describes five from center of the macula CSME
clinical levels of diabetic retinopathy: no apparent retinopathy
(no abnormalities), mild NPDR (microaneurysms only), moder- Moderate DME Findings approaching the center CSME
ate NPDR (more than microaneurysms only but less than severe but not involving the center
NPDR), severe NPDR (any of the following: >20 intraretinal Severe DME Findings involving the center of the CSME
hemorrhages in each of four retinal quadrants, definite venous macula
beading in two or more retinal quadrants, prominent intraretinal
microvascular abnormalities in one or more retinal quadrants, and CSME, Clinically significant macular edema; ETDRS, Early Treatment Diabetic Retinopathy Study.
no PDR), and PDR (one or more of retinal neovascularization,

without relying on reference to the standard photographs of the primary therapy for neovascular glaucoma and provides durable
Airlie House Classification of diabetic retinopathy. This approach regression of neovascularization of the iris and angle for most
makes clinical use easier and more uniform among practitioners eyes. Other approaches such as goniophotocoagulation, topical or
not versed in the complexities of the ETDRS grading system. systemic antiglaucoma medications, and antiglaucomatous filtra-
However, because of this simplification, the International Clas- tion surgery are available when needed.356–358 Intravitreal admin-
sification of Diabetic Retinopathy and Diabetic Macular Edema istration of VEGF inhibitors is now frequently utilized in eyes
is not a replacement for ETDRS levels of diabetic retinopathy in with acute neovascular glaucoma and results in remarkably rapid
large-scale clinical trials or studies for which precise retinopathy regression of the neovascularization with resulting normalization
classification is required. of the intraocular pressure.359 However, the effects of anti-VEGF
agents are often transient, and there is frequently recurrence of
Other Ocular Manifestations of Diabetes neovascularization unless injections are given on a monthly basis.
All structures of the eye are susceptible to complications of dia- Thus a more durable therapy such as scatter laser photocoagula-
betes. The consequence of these changes can range from being tion is often performed.
unnoticed by both patient and physician, to symptomatic but not The cornea of the diabetic person is more susceptible to
sight threatening, to requiring evaluation to rule out potentially injury and slower to heal after injury than is the nondiabetic
life-threatening underlying causes other than diabetes. cornea.359a,360 The diabetic cornea is also more prone to infec-
Mononeuropathies of the third, fourth, or sixth cranial nerves tious corneal ulcers, which can lead to rapid loss of vision,
can arise in association with diabetes; mononeuropathy of the need for corneal transplant, or loss of the eye if it is not treated
fourth cranial nerve is least likely associated with diabetes and aggressively. Consequently, diabetic patients using contact lenses
warrants workup for other causes.348–350 Nerve palsies present a should exercise caution to avoid contact lens overwear and to
significant diagnostic challenge because misdiagnosis can result in maintain careful monitoring.
a life-threatening lesion remaining untreated. In one review of cra- Open-angle glaucoma is 1.4 times more common in the dia-
nial nerve palsies treated in a diabetic patient population in 1967, betic population than in the nondiabetic population.361 The prev-
42% of mononeuropathies were not diabetic in origin.349 This alence of glaucoma increases with age and duration of diabetes,
finding underscores the danger of routinely attributing mononeu- but medical therapy for open-angle glaucoma is generally effective.
ropathies to the diabetic condition itself without carefully ruling In a study of 76,318 women enrolled in the Nurses’ Health Study,
out other potential causes. The percentage of all extraocular mus- Pasquale and coworkers found that T2DM is associated with an
cle palsies attributable to diabetes mellitus is estimated at 4.5% to increased risk of primary open-angle glaucoma in women.362
6%.350 Mononeuropathies may be the initial presenting sign of Diabetes effects on the crystalline lens can result in transient
new-onset diabetes, and diabetes should therefore be considered refractive changes, alterations in accommodative ability,363 and
in the differential diagnosis of any mononeuropathy affecting the cataracts. Refractive change can be significant and is related to
extraocular muscles, even in patients who do not claim a history of fluctuation of blood glucose levels with osmotic lens swelling.365
diabetes. Diabetes-induced third-nerve, fourth-nerve, and sixth- Cataracts can occur earlier in life and progress more rapidly in the
nerve palsies are usually self-limited and should resolve spontane- presence of diabetes.364,365 Cataracts are 1.6 times more common
ously in 2 to 6 months. Palsies can recur or subsequently develop in people with diabetes than in those without diabetes.364,365 In
in the contralateral eye. patients with earlier onset diabetes, duration of diabetes, retinopa-
The optic disc can be affected by diabetes in a variety of ways thy status, diuretic use, and HbA1c levels are risk factors.366 In
other than vasoproliferation. Diabetic papillopathy is a diagnosis patients with later onset diabetes, age of the patient, lower intraoc-
of exclusion and must be distinguished from other causes of disc ular pressure, smoking, and lower diastolic BP may be additional
swelling such as true papilledema from increased intracranial pres- risk factors.367,368 Diabetic patients undergoing simultaneous
sure, pseudopapilledema such as optic nerve head drusen, toxic kidney and pancreas transplantation are at an increased risk of
optic neuropathies, neoplasms of the optic nerve, and hyperten- developing all types of cataracts, independent of the use of corti-
sion.351 Optic disc pallor can occur following spontaneous remis- costeroids after transplantation.369 Both phacoemulsification and
sion of proliferative retinopathy or remission following scatter extracapsular cataract extraction with intraocular lens implanta-
(panretinal) laser photocoagulation (see Fig. 37.27I). Because dia- tion are appropriate surgical therapies. The principal determinant
betes poses an increased risk for developing open-angle glaucoma, of postoperative vision and progression of retinopathy is related
the disc pallor following remission of retinopathy or laser photo- to the preoperative presence of DME and the severity of diabetic
coagulation must be considered when evaluating the optic nerve retinopathy.370,371
head for glaucoma. Other findings with higher incidence among patients with
A diabetic ocular complication with potentially serious con- diabetes include xanthelasma,348 microaneurysms of the bulbar
sequences is neovascularization of the iris. Neovascularization of conjunctiva,372 posterior vitreous detachment,373 and the rare but
the iris occurs in 4% to 7% of diabetic eyes and may be present in often fatal orbital fungal infection Mucorales phycomycosis.352,353
40% to 60% of eyes with proliferative retinopathy. Usually new Prompt diagnosis and treatment of phycomycosis caused by
iris vessels are first observed at the pupillary border, followed by Mucor species are crucial, although the survival rate remains at
a fine network of vessels over the iris tissue progressing into the only 57%.353,374
filtration angle of the eye. Closure of the angle by the fibrovascular
network results in neovascular glaucoma, which can lead to irre- Monitoring and Treatment of Diabetic
versible optic nerve damage and vision loss due to the rise in intra-
ocular pressure.354 Neovascular glaucoma is difficult to manage Retinopathy
and requires aggressive treatment. Diabetes is the second leading Appropriate clinical management of diabetic retinopathy has been
cause of neovascular glaucoma, accounting for 32% of cases.355 defined by results of major randomized, multicenter clinical trials
When possible, scatter (panretinal) laser photocoagulation is the (Fig. 37.28): the Diabetic Retinopathy Clinical Research Network

No diabetic Mild Moderate Severe Very severe Early High risk Severe
retinopathy Nonproliferative diabetic retinopathy PDR PDR PDR

Sorbinil retinopathy trial

DCCT/UKPDS

ETDRS

DRS

DRVS

• Fig. 37.28 Schematic representation of the major multicenter clinical trials of diabetic retinopathy and the
severity levels of diabetic retinopathy that they primarily addressed. DCCT, Diabetes Control and Com-
plications Trial; DRS, Diabetic Retinopathy Study; DRVS, Diabetic Retinopathy Vitrectomy Study; ETDRS,
Early Treatment Diabetic Retinopathy Study; PDR, proliferative diabetic retinopathy; UKPDS, United King-
dom Prospective Diabetes Study.

Diabetes

Has patient No Diabetes Type I Age No

had eye exam type ≥10 years
since diagnosis
of diabetes? Yes

Yes Is patient
Yes pregnant?
Type 2
No
Ophthalmic
follow-up per Is diabetes No
patient’s diabetic duration
retinopathy level ≥5 years?

Refer to eye Yes Initial

care provider for ophthalmic
comprehensive exam not yet
eye exam required

• Fig. 37.29 Schematic flow chart of major principles involved in determining the timing of initial ophthalmic
examination after a diagnosis of diabetes mellitus. These are minimal recommended times. Ocular symp-
toms, complaints, or other associated medical issues can necessitate earlier evaluation. Guidelines are
regularly reevaluated based on new study results.

Protocols I, S, and T,290,291,375 Diabetic Retinopathy Study and timely diagnosis and therapy. Unfortunately, many diabetic
(DRS),376 ETDRS,316 Diabetic Retinopathy Vitrectomy Study patients do not receive adequate eye care at an appropriate stage
(DRVS),377 DCCT,378 and UKPDS.379 These studies have elu- of their disease.380,381 In fact, 11% of T1DM and 7% of T2DM
cidated the progression rates of each level of diabetic retinopathy, patients with high-risk PDR necessitating prompt treatment
guided follow-up intervals, and demonstrated the proper delivery, had not been examined by an ophthalmologist within the past
timing, and resulting effectiveness of glycemic control, laser pho- 2 years.381 In one study, 55% of patients with high-risk PDR or
tocoagulation, and intravitreous anti-VEGF therapy (Figs. 37.29, CSME had never had laser photocoagulation.380
37.30, 37.31, and 37.32). They have also established recommen- The comprehensive eye examination is the mainstay of such
dations for vitrectomy surgery for diabetic eye complications. evaluation and is necessary on a repetitive, lifelong basis for patients
with diabetes.342,382 Such an evaluation has four major compo-
Comprehensive Eye Examination nents: history, examination, diagnosis, and treatment as needed.
An accurate ocular examination detailing the extent and loca- Annual retinal evaluation to assess the presence and level of dia-
tion of retinopathy-associated findings is critical for determining betic retinopathy and DME is essential to guide patient care. The
monitoring and treatment decisions in patients with diabetic reti- fundamentals of a comprehensive eye examination for the non-
nopathy. As detailed later, most of the blindness associated with diabetic patient have been detailed by the American Academy of
advanced stages of retinopathy can be averted with appropriate Ophthalmology382 and the American Optometric Association.383

Diabetes

Is DME Yes Is CSME Yes Is vision

present? present excellent?

No
No

Minimum follow-up Yes No

Is HRC PDR every 3 months;
present? sooner if indicated Focal/grid laser
Consider
by DR photocoagulation
focal/grid laser
probably
photocoagulation
indicated

Yes No

Panretinal Is severe NPDR or Yes Consider panretinal Follow-up in 4

photocoagulation worse present? photocoagulation months or less
probably
indicated No

Is moderate Yes Follow-up every

NPDR present? 4–8 months

Is mild Yes Follow-up every

NPDR present? 8–12 months

Follow-up
every 12 months

• Fig. 37.30 Diabetic retinopathy and macular edema examination and treatment flow chart: nonpregnant
patients. The schematic flow chart presents major principles involved in determining routine ophthalmic
follow-up and indications for treatment in nonpregnant patients with diabetes. These intervals are only
general, minimal recommended frequencies. Ocular symptoms, complaints, or other associated ophthal-
mic or medical issues can necessitate earlier evaluation or an altered approach. Guidelines are regularly
reevaluated based on new study results. CSME, Clinically significant macular edema; DME, diabetic macu-
lar edema; DR, diabetic retinopathy; HRC PDR, high-risk characteristic proliferative diabetic retinopathy;
NPDR, nonproliferative diabetic retinopathy; PDR, proliferative diabetic retinopathy.

The examination of the patient with diabetes should be similar, when interpreted by a trained eye care provider can also be appro-
with additional emphasis on portions of the examination that priate for retinal evaluation.387,388 Furthermore, ocular telehealth
relate to problems particularly relevant to diabetes. programs for diabetic retinopathy that utilize validated means of
Dilated ophthalmic examination is superior to undilated evalu- retinal imaging have the potential to expand access to highly effec-
ation because only 50% of eyes are correctly classified as to pres- tive evidence-based diabetes eye care and provide cost-effective
ence and severity of retinopathy through undilated pupils.384,385 alternative methods of care.389
Appropriate ophthalmic evaluation entails pupillary dilation, slit-
lamp biomicroscopy, examination of the retinal periphery with Initial Ophthalmic Evaluation
indirect ophthalmoscopy or mirrored contact lens, and sometimes The recommendation for initial ocular examination in persons with
gonioscopy.382,383 Because of the complexities of the diagnosis diabetes is based on the prevalence rates of retinopathy and the inci-
and treatment of PDR and CSME, ophthalmologists with spe- dence of subsequent vision threatening diabetic eye complications (see
cialized knowledge and experience in the management of diabetic Fig. 37.29). Approximately 80% of T1DM patients have retinopathy
retinopathy are required to determine and provide appropriate after 15 years of disease, but only about 25% have any retinopathy
surgical intervention.386 Thus it is recommended that all patients after 5 years of diabetes.385 The prevalence of PDR is less than 2% at
with diabetes should have dilated ocular examinations by an expe- 5 years and 25% by 15 years. For T2DM, the onset date of diabetes
rienced eye care provider (ophthalmologist or optometrist), and is usually unknown, and severe retinal disease can be observed even at
diabetic patients should be under the direct or consulting care of the time of diabetes diagnosis. Up to 3% of patients whose diabetes
an ophthalmologist experienced in the management of diabetic is first diagnosed after age 30 years (T2DM) have CSME or high-risk
retinopathy at least by the time moderate diabetic retinopathy or PDR at the time of initial diagnosis of diabetes.390 Thus in patients
DME is present. Retinal imaging that has demonstrated equiva- older than 10 years, initial ophthalmic examination is recommended
lency to dilated retinal fundus examination or the accepted stan- beginning 5 years after the diagnosis of T1DM and immediately after
dard of ETDRS seven-standard-field stereoscopic retinal imaging diagnosis of T2DM (see Fig. 37.29).342,391

Diabetes

Is patient No Is patient No Standard eye evaluation

pregnant? planning protocol for patient’s level
pregnancy? of diabetic retinopathy

Yes
Yes
Refer to eye care
provider for comprehensive
ophthalmic evaluation

Follow-up per No Yes Yes

Does patient Is DME Is CSME
eye care provider
have DR or DME? present? present?
or every 3 months
No No Yes

Panretinal Yes Consider Focal/grid laser

Is HRC PDR
photocoagulation focal/grid laser photocoagulation
present?
probably indicated photocoagulation probably indicated
No

Consider Yes Is severe

panretinal NPDR or worse
photocoagulation present?

Eye care provider

follow-up every 3
months or less

• Fig. 37.31 Diabetic retinopathy and macular edema examination and treatment flow chart: pregnant
patients. The schematic flow chart shows major principles involved in determining routine ophthalmic
follow-up and indications for treatment in pregnant patients with diabetes. These intervals are only gen-
eral, minimal recommended frequencies. Ocular symptoms, complaints, or other associated ophthalmic
or medical issues can necessitate earlier evaluation or an altered approach. Because retinopathy can
progress rapidly in pregnant patients with diabetes, careful and more frequent evaluation is often indicated.
Guidelines are regularly reevaluated based on new study results. CSME, Clinically significant macular
edema; DME, diabetic macular edema; DR, diabetic retinopathy; HRC PDR, high-risk characteristic prolif-
erative diabetic retinopathy; NPDR, nonproliferative diabetic retinopathy.

Puberty and pregnancy can accelerate retinopathy progression. Follow-Up Ophthalmic Examination
The onset of vision-threatening retinopathy is rare in children Follow-up ocular examination is determined from the risk of
prior to puberty, regardless of the duration of diabetes391,392; how- disease progression at any particular retinopathy severity (see
ever, significant retinopathy can arise within 6 years of disease if Fig. 37.30). NPDR is clinically categorized into four levels
diabetes is diagnosed between the ages of 10 and 30 years.295 Dia- of severity based on clinical findings compared to stereo fun-
betic retinopathy can become particularly aggressive during preg- dus photographic standards: mild, moderate, severe, and very
nancy in patients with diabetes.393,394 In the past, the prognosis severe.397 Progression of nonproliferative retinopathy to the visu-
for pregnancy in the diabetic patient with microvascular complica- ally threatening level of high-risk PDR is closely correlated with
tions was so poor that pregnant diabetic patients were commonly baseline NPDR severity (Table 37.4). Progression rates from
advised to avoid or terminate pregnancies.395 With recognition of each individual NPDR level to any other retinopathy level are
the importance of glycemic control, diabetic patients in the child- also known. These are used to define standard minimal follow-
bearing age now can experience safe pregnancy and childbirth up intervals as detailed in Fig. 37.30 and Table 37.5. Because
with minimal risk to both the mother and the baby.396 significant sight-threatening retinopathy can initially occur with
Ideally, patients with diabetes who are planning pregnancy no or minimal visual symptoms, patients with no clinically evi-
should have a comprehensive eye examination within 1 year prior dent diabetic retinopathy and no known ocular problems still
to conception (see Fig. 37.31). Treatment may be indicated prior require annual comprehensive ophthalmic examinations even if
to conception for patients who are at risk for vision loss if their dia- they are totally asymptomatic.
betic retinopathy were to worsen during pregnancy. Patients who
become pregnant should have a comprehensive eye examination Evaluation and Treatment of Proliferative Diabetic
in the first trimester of pregnancy. Close follow-up throughout Retinopathy
pregnancy is indicated, with subsequent examinations determined The extent and location of neovascularization determine the level
by the findings present at the first-trimester examination.342 This of PDR.398,399 PDR is best evaluated by dilated fundus exami-
recommendation does not apply to women who develop gesta- nation using slit-lamp biomicroscopy combined with indirect
tional diabetes because they are not at increased risk of developing ophthalmoscopy or stereo fundus photography. Fluorescein angi-
diabetic retinopathy. ography and optical coherence tomography angiography can also

Follow-up Is treatment for Follow-up in 3 months

dependent on diabetic retinopathy or or as indicated by
ocular status macular edema indicated? ocular status

Yes

Yes Is treatment of
Focal/grid
macular edema PRP for DR
laser for DME
indicated?
No
Is treatment for No Retinopathy
DR indicated? still active or
progressive?

Is further laser Yes

Follow-up in No
apparently
3-4 months futile? Is there No Consider
retina novel
Yes left to treat? therapies
No Is treatment
of DME still Consider Yes
indicated? novel
therapies
Re-treat
Yes
PRP

Re-treat
DME Follow-up as
indicated by
ocular status

• Fig. 37.32 Photocoagulation flow chart. This schematic flow chart details general photocoagulation treat-
ment approaches in patients with diabetic retinopathy or diabetic macular edema. These are only general
guidelines, and actual treatment choices can be affected by numerous other factors, including findings in
the same eye or in the contralateral eye and systemic issues. DME, Diabetic macular edema; DR, diabetic
retinopathy; PRP, scatter (panretinal) photocoagulation.

TABLE 37.4 Progression to Proliferative Diabetic treatment, eyes with high-risk PDR have a 28% risk of severe
Retinopathy by Level of Nonproliferative vision loss within 2 years. This risk compares with a 7% risk of
Diabetic Retinopathy severe vision loss after 2 years for eyes with PDR but without
high-risk characteristics.398
CHANCE OF HIGH-RISK PDR (%) Until this decade, scatter (panretinal) laser photocoagulation
Retinopathy Level 1 Year 5 Years was the sole therapy shown to be effective against PDR. The DRS
demonstrated that panretinal laser photocoagulation was effec-
Mild NPDR 1 16 tive in reducing the risk of severe vision loss from PDR by 50%
Moderate NPDR 3–8 27–39 or more. The ETDRS further demonstrated that panretinal laser
photocoagulation applied when an eye approaches or just reaches
Severe NPDR 15 56 high-risk PDR reduces the risk of severe vision loss to less than
Very severe NPDR 45 71 4%. Based on these results, prompt scatter photocoagulation is
considered appropriate for all patients with high-risk PDR, usu-
PDR with fewer high-risk 22–46 64–75
ally appropriate for patients with PDR less than high risk, and
characteristics
may be advisable for patients with severe or very severe NPDR,
NPDR, Nonproliferative diabetic retinopathy; PDR, proliferative diabetic retinopathy. especially in the setting of T2DM (see Fig. 37.30).316,398-401
From Aiello LP, Gardner TW, King GL, et al. Diabetic retinopathy. Diabetes Care. 1998;21:143–156. Recent progression of eye disease, status of the fellow eye, com-
pliance with follow-up, concurrent health concerns such as hyper-
tension or kidney disease, and other factors must be considered in
determining if laser surgery should be performed in these patients.
aid in identifying small, subtle patches of retinal neovasculariza- In particular, patients with T2DM should be considered for scat-
tion. However, these imaging modalities are usually not required ter photocoagulation before high-risk PDR develops because the
for the diagnosis of PDR because these findings are clinically evi- risk of severe vision loss and the need for pars plana vitrectomy
dent in most cases (Fig. 37.33). (PPV) can be reduced by 50% in these patients, especially when
The presence of PDR in an eye substantially increases the macular edema is present.401
risk for severe vision loss. Severe visual loss is defined as best In scatter photocoagulation, 1200 to 1800 laser burns are
corrected acuity of 5/200 or worse on two consecutive visits 4 applied to the peripheral retinal tissue, actually focally destroy-
months apart. This represents vision loss substantially worse than ing the outer photoreceptor and retinal pigment epithelium of the
the 20/200 or worse limit defined as legal blindness. Without retina (see Fig. 37.27I). Large vessels are avoided, as are areas of

TABLE 37.5 R
ecommended General Management of Diabetic Retinopathy
RISK OF PROGRESSION (%) EVALUATION TREATMENT
To High-Risk Color Scatter Laser Focal Intravitreal Follow-Up
Level of DR To PDR (1 yr) PDR (5 yr) Photo OCT FA (PRP) Laser VEGF Inhibitors (mo)

Mild NPDR
All 5 15
No DME No No No No No No 12
Non-ciDME Yes Yes Occ No Occ Occ 4–6
ciDME Yes Yes Yes No Occ Yes 1–4
Moderate NPDR
All 12–27 33
No DME Yes No No No No No 6–8
Non-ciDME Yes Yes Occ No Occ Occ 4–6
ciDME Yes Yes Yes No Yes Yes 1–4
Severe NPDR
All 52 60
No DME Yes No No Rarely No No 3–4
Non-ciDME Yes Yes Occ Occ after DME Occ Occ 2–3
treatment
ciDME Yes Yes Yes Occ after DME Yes Yes 1–3
treatment
Very Severe NPDR
All 75 75
No DME Yes No No Occ No No 2–3
Non-ciDME Yes Yes Occ Occ after DME Occ Occ 2–3
treatment
ciDME Yes Yes Yes Occ after DME Yes Yes 1–3
treatment
PDR < High Risk
All — 75
No DME Yes No No Occ No Occ 2–3
Non-ciDME Yes Yes Occ Occ after DME Occ Occ 2–3
treatment
ciDME Yes Yes Yes Occ after DME Yes Yes 1–3
treatment
PDR With High-Risk Characteristics
All — —
No DME Yes No No Yes No Occ 2–3
Non-ciDME Yes Yes Occ Yes Usually Occ 1–2
ciDME Yes Yes Yes Yes Yes Yes 1–2

ciDME, Center-involved diabetic macular edema; DME, diabetic macular edema; FA, fluorescein angiography; ME, macular edema; NPDR, nonproliferative diabetic retinopathy; Occ, occasionally; OCT,
optical coherence tomography; PDR, proliferative diabetic retinopathy; VEGF, vascular endothelial growth factor.

Diabetes Assessment 1 Month

After Initial
Injections(s)
No Does patient
have macular edema?

Yes
Will patient NV Stable or Yes No Injection and
treatment be altered No Is focal/grid laser Resolved? Return in 1 Month
by angiogram planned, if possible?
results (rare)?

No
No
Fluorescein
Yes angiogram usually Yes
Re-inject and Return Yes NV Worsens
not indicated
in 1 Month or Recurs
Yes

Is patient pregnant
or allergic to No
fluorescein dye?
Double Follow-Up
No Intervals Up to 4
Months
Fluorescein
angiogram
often indicated
• Fig. 37.34 Principles of the Diabetic Retinopathy Clinical Research
Network (DRCR.net) antivascular endothelial growth factor treatment
• Fig. 37.33 Fluorescein angiogram flow chart. The schematic flow chart algorithm for proliferative diabetic retinopathy. NV, Neovascularization.
details a general algorithm for appropriate use of fluorescein angiogra- (Redrawn from Sun JK, Glassman AR, Beaulieu WT, et al. Rationale and
phy in the ocular evaluation of patients with diabetes mellitus. In unusual application of the protocol S anti-vascular endothelial growth factor algo-
cases, confounding factors can alter the appropriate approach. rithm for proliferative diabetic retinopathy. Ophthalmol. 2018;125.)

preretinal hemorrhage. The treatment is thought to exert its effect is highly effective at reducing retinal neovascularization. In Pro-
by increasing oxygen delivery to the inner retina, decreasing viable tocol S, 38% of the eyes treated with ranibizumab were eligible
hypoxic growth factor–producing cells, and increasing the rela- to defer monthly injections due to sustained stability of neovas-
tive perfusion per area of viable retina. The total treatment is usu- cularization at least once over the first 2 years of follow-up, and
ally applied over one to three sessions, spaced 1 to 2 weeks apart. retinal neovascularization was completely resolved at 44% of visits
Follow-up evaluation usually occurs at 3 months. through 2 years.405 Nonetheless, eyes randomized to ranibizumab
The response to scatter photocoagulation varies. The most treatment in Protocol S continued to need a median of three injec-
desirable effect is to see a regression of the new vessels, although tions per year through 2 to 5 years. It should also be realized that
stabilization of the neovascularization with no further growth can patients with PDR often have systemic comorbidities that may
result. This latter situation requires careful clinical monitoring. In contribute to missed or rescheduled office visits. Thus patient
some cases, new vessels continue to proliferate, requiring addi- compliance with follow-up and treatment recommendations is a
tional scatter photocoagulation or adjuvant treatment with intra- critical aspect of ensuring successful outcomes with anti-VEGF
vitreous anti-VEGF agents (see Fig. 37.32). therapy for PDR. If compliance is expected to be a problem for
Recent studies have shown that anti-VEGF treatment with a specific patient, then treatment with laser panretinal photoco-
aflibercept or ranibizumab is an effective first-line alternative agulation alone, or in combination with a VEGF inhibitor, is the
to scatter photocoagulation in carefully selected patients with preferred therapeutic approach.
PDR375,402,403 Neovascular processes are exquisitely sensitive to Surgical intervention with PPV is usually reserved for eyes
anti-VEGF agents, and eyes with severe neovascularization of with nonclearing vitreous hemorrhage or traction detachment
the retina or anterior segment have demonstrated dramatic and from PDR. The DRVS, completed in 1989, demonstrated that
rapid improvement with anti-VEGF therapy.322,404 The Diabetic early PPV in persons with severe fibrovascular proliferation was
Retinopathy Clinical Research (DRCR) Network Protocol S more likely to result in better vision and less likely to result in
demonstrated that visual acuity outcomes at 2 and 5 years were poor vision, particularly in patients with T1DM.377 PPV aims
noninferior in eyes treated with intravitreal ranibizumab to those primarily to remove abnormal fibrovascular tissue, alleviate retinal
obtained with scatter photocoagulation. Anti-VEGF treatment traction, allow the retina to obtain a more anatomically normal
with ranibizumab was associated with several advantages over laser position, and remove vitreous opacities such as vitreous hemor-
therapy. There was greater average visual gain over the course of rhage. The actual outcome data from this study is not entirely
the first 2 years, reduced rates of visual impairment from DME applicable today due to dramatic advances in surgical techniques
onset, less loss of peripheral visual field, and decreased need for and the advent of laser endophotocoagulation that have occurred
vitrectomy surgery at both 2 and 5 years in the ranibizumab group. in the intervening years. Nevertheless, it is clear that PPV can save
When delivered by a standardized retreatment algorithm such and restore vision in many cases of severe retinal disease that are
as that developed for Protocol S (Fig. 37.34), anti-VEGF therapy not amenable, or not responsive, to laser photocoagulation.

Treatment of Nonproliferative Diabetic Retinopathy

Assessment 1 Month
Studies of intravitreous therapy with VEGF inhibitors and ste- After Initial
roids have demonstrated beneficial effects of these medications on Injection(s)
DR severity and rates of progression to PDR-related complica-
tions such as vitreous hemorrhage and need for vitrectomy. Nearly
40% of eyes achieve two or more step regression of diabetic reti-
nopathy severity after 3 years of monthly continuous anti-VEGF
therapy for DME.406–408 Decreased rates of PDR onset and need DME No No Injection and
for PDR treatment are also observed after ranibizumab therapy as Improving Return in 1 Month
compared to sham treatment.407 After an average of 4.4 injections
of aflibercept over 6 months, 58% of eyes experienced a two or
more step improvement in diabetic retinopathy severity as com- Yes
pared to 6% of sham-treated eyes.408 Steroid therapy, whether
delivered by the peribulbar or intravitreal route, also resulted in
Re-inject and Return Yes DME Worsens
decreased risk of diabetic retinopathy worsening. Over 2 years,
in 1 Month or Recurs
intravitreal steroid administration resulted in a 32% relative risk
reduction for retinopathy progression as compared to sham treat-
ment.407 Although these results are promising, there is no treat-
No
ment mandate at this time for eyes with nonproliferative diabetic
retinopathy since long-term visual outcomes have not been defini-
tively shown to be superior after treatment with any of these intra- Double Follow-up
Interval Up to 4
ocular therapeutic approaches versus medical care alone. Months

Treatment of Diabetic Macular Edema

• Fig. 37.35 Diagram showing the Diabetic Retinopathy Clinical Research
Untreated CSME is associated with an approximately 25% chance Network (DRCR.net) rationale for treatment and follow-up of center-
of moderate vision loss after 3 years (defined as at least doubling involved diabetic macular edema (DME) with antivascular endothelial
the visual angle; e.g., 20/40 reduced to 20/80).316 Macular edema growth factor therapy. (Modified from Aiello LP, Beck RW, Bressler NM,
is best evaluated clinically by dilated examination using slit-lamp et al. Rationale for the Diabetic Retinopathy Clinical Research Network
biomicroscopy or stereo fundus photography. The newer diagnos- intravitreal anti-VEGF treatment and follow-up protocol for center-involved
tic ophthalmic imaging technique of OCT has provided a means diabetic macular edema. Ophthalmol. 2011;118:e5–e14.)
to objectively quantify retinal thickening and is currently the
objective method of choice. When used in conjunction with visual
acuity measurement, OCT has been used to monitor response to throughout years 2 to 5.411,412 No increased systemic events or
treatment and help determine timing of intervention.409,410 As serious ocular adverse events were attributed to the treatment
discussed earlier, the current standard for macular edema evalu- other than the known small risk of endophthalmitis (1 per 1000
ation focuses on determining the presence or absence of edema injections) associated with intravitreal injections themselves.
involving the center of the macula. A head-to-head comparative effectiveness trial demonstrated
Current first-line therapy for most eyes with center-involved overall efficacy of all three currently commercially available intra-
DME and DME-related visual impairment of 20/32 or worse is ocular VEGF inhibitors (aflibercept, bevacizumab, and ranibi-
intravitreous injections of VEGF inhibitors following a defined zumab) in improving visual acuity and reducing retinal thickening
protocol such as that described by the DRCR Network291 (Fig. in eyes with center-involved DME.291 Eyes with only mild visual
37.35). Typically, injections are performed monthly with a load- impairment from DME and baseline vision of 20/32 to 20/40
ing dose of at least four to six injections. On average, eight to nine had similar visual acuity outcomes on average regardless of thera-
injections are performed in the first year of treatment. The average peutic group assignment. However, in eyes with baseline vision of
number of injections required to maintain the beneficial visual 20/50 or worse, aflibercept provided visual gains superior to those
acuity gains declines substantially to three or four in the second achieved with bevacizumab at 1 and 2 years, and vision outcomes
year, one or two in the third year, and zero to one in years 4 and 5. were superior to ranibizumab at 1 but not 2 years.
Results from an initial multicenter randomized controlled Although VEGF inhibitors are highly successful in many eyes,
clinical trial evaluating intravitreal administration of ranibizumab approximately 40% to 50% of eyes still have persistent DME
either with immediate laser or with deferred laser show a marked or visual impairment despite chronic monthly therapy. Thus
benefit of the anti-VEGF agent compared to laser alone. After 1 there is a continued unmet need for novel therapies that address
year, ranibizumab as applied in the trial resulted in a nine-letter VEGF-independent mechanisms for DME onset, persistence, and
mean gain (p < 0.001) when combined either with prompt laser worsening.
or deferral of laser for at least 24 weeks. This was more effective Focal laser photocoagulation may still be indicated for some
than prompt laser alone (three-letter gain) for the treatment of patients with DME (see Figs. 37.27C and 37.30) without center
center-involving DME in eyes with central thickening and vision involvement or in eyes with good vision or in patients who cannot
reduced to 20/32 to 20/320. The number of eyes gaining two or tolerate a regimen of intravitreal injections. The ETDRS demon-
more lines of vision almost doubled in the anti-VEGF groups as strated focal laser photocoagulation for CSME reduced the 5-year
compared with laser alone. Conversely, eyes losing two or more risk of moderate vision loss by 50%, but 15% of patients con-
lines of vision were about one-third as many in the anti-VEGF tinue to experience vision loss.345 In focal laser photocoagulation,
groups as compared to laser alone. Results were sustained over lesions from 500 to 3000 μm from the center of the macula that
5 years despite a substantially decreased number of injections are contributing to thickening of the macular area are generally

directly photocoagulated. These lesions are identified clinically or The pathogenesis of DME is highly complex, and a variable
by fluorescein angiography and consist primarily of leaking micro- response to treatment modalities has been observed in many
aneurysms. When leakage is diffuse or microaneurysms are exten- patients. Investigations of varying drug dosages to identify the
sive, photocoagulation may be applied to the macula in a grid optimal treatment concentration and development of sustained
configuration, avoiding the fovea region. drug delivery devices to limit risks, costs, and inconvenience
If macular laser is planned, fluorescein angiography can be a associated with repeated intraocular injections are underway.
useful test for guiding the focal treatment of microaneurysm that Multiple potential therapeutics are also being tested in clinical
are leaking fluid into the retina and identifying areas of macular trials for the treatment of DME, including a bispecific antibody
capillary nonperfusion that might benefit from grid laser treat- against VEGF and angiotensin II, plasma kallikrein inhibition,
ment (see Fig. 37.33). Because there are risks associated with fluo- photobiomodulation therapy, and other novel approaches.
rescein angiography, including nausea, urticaria, hives, and rarely
death (1 in 222,000 patients) or severe medical sequelae (1 in 2000 Control of Systemic Disorders and Effect of Systemic
patients),413–415 fluorescein angiography is not otherwise part of Medications
the routine examination of a diabetic patient without diabetic reti- In addition to the importance of intensive glycemic control in
nopathy, and the procedure is usually contraindicated in patients reducing the onset and progression of diabetic retinopathy as dis-
with known allergy to fluorescein dye or during pregnancy. cussed earlier, it is critical for the optimal ocular health of diabetic
Follow-up evaluation after focal laser treatment generally patients that several other systemic considerations be optimized.
occurs after 3 months (see Fig. 37.30). In cases in which mac- Elevated blood pressure exacerbates the development and
ular edema persists, further treatment may be necessary. In the progression of diabetic retinopathy. Concomitant hypertension
presence of macular edema, patients with PDR or severe NPDR is common in diabetes. Patients with T1DM have a 17% preva-
should be considered for anti-VEGF or laser treatment of macu- lence of hypertension at baseline and a 25% incidence after 10
lar edema whether or not the macular edema is center involved years.420 There is a 38% to 68% prevalence in T2DM.421–423 In
because they are likely to require scatter laser photocoagulation in most studies, hypertension correlates with other retinopathy risk
the near future and because scatter photocoagulation, while ben- factors, including duration of diabetes, higher HbA1c level, pres-
eficial for PDR, can exacerbate existing macular edema. ence of proteinuria, and male gender. The risk of PDR is associ-
The ophthalmic use of corticosteroids administered either through ated with the presence of hypertension at the baseline visit, higher
the periocular or intravitreal routes for the treatment of DME gained HbA1c levels, and presence of more severe levels of retinopathy
widespread use due to early case reports and uncontrolled clini- at the initial visit.424 Patients with hypertension are more likely
cal trials documenting its rapid and often dramatic effect on retinal to develop retinopathy, diffuse macular edema, and more severe
thickening. Two multicenter randomized prospective clinical trials levels of retinopathy425–427 and have more rapid progression of
were undertaken to address both the effectiveness and safety of both retinopathy when compared with diabetic patients who do not
routes of steroid administration. In the first, peribulbar steroid injec- have hypertension.427–429
tions were found to have no significant benefit for the treatment of The large randomized, prospective UKPDS in 1148 patients
DME.416 The 3-year results of a second multicenter randomized con- with T2DM demonstrated a 34% (p = 0.0004) and 47% (p <
trolled trial comparing intravitreal steroids to focal laser therapy have 0.004) reduction in risk of diabetic retinopathy progression and
shown that despite an initial rapid reduction in retinal thickness and moderate visual acuity loss, respectively, in patients assigned to
improvement in vision with the intravitreal steroid injection, by 1 intensive blood control.430 Effects were independent of glycemic
year the results were no better than laser photocoagulation, and at control, and risk reductions were similar regardless of whether
2 years through 3 years, steroid was inferior to the laser treatment the hypertension was controlled with an angiotensin-converting
in both visual outcome and retinal thickness improvement.417,418 enzyme (ACE) inhibitor (captopril) or a beta blocker (atenolol).
Intravitreal steroid injections were associated with an approximately Overall, hypertension appears to be a significant risk factor in the
fourfold increase in the rate of intraocular pressure complications and development and progression of diabetic retinopathy and should
fourfold increase in need for cataract surgery compared to laser treat- be rigorously controlled. Until the results of specific trials inves-
ment. The development of cataract was likely a large contributor to tigating the blood pressure levels required to minimize end-organ
declining visual acuity in the steroid-treated group after the first 6 damage in patients with diabetes are known,431 target blood pres-
months. This hypothesis is supported by the fact that patients who sure should most likely be maintained as low as safely possible.
had already undergone cataract surgery prior to initiating intravitreal Associations between renal and retinal angiopathy are numer-
steroid treatment demonstrated vision improvement comparable to ous. Both microalbuminuria and proteinuria are associated with
that seen in the anti-VEGF-treated group.375 retinopathy.432–435 The presence and severity of diabetic retinopa-
Intravitreous steroid therapy has also been investigated as a thy are indicators of the risk of gross proteinuria,436,437 and con-
potential additive therapy to intravitreous VEGF inhibitors in versely, proteinuria predicts PDR.433,438,439 Half of all T1DM
patients who have persistent retinal thickening and visual impair- patients with PDR and 10 or more years of diabetes have concom-
ment despite at least six injections of anti-VEGF. Although retinal itant proteinuria.432 In T1DM, the prevalence of PDR increases
thickening is significantly reduced in eyes after combined steroid from 7% at onset of microalbuminuria to 29% 4 years after onset
and anti-VEGF therapy compared to those receiving anti-VEGF of albuminuria, compared with 3% and 8%, respectively, in
treatment alone, visual acuity outcomes do not appear to be patients without persistent microalbuminuria.440 The Appropriate
improved.419 Thus, currently, intravitreal steroid alone is not the Blood Pressure Control in Diabetes (ABCD) trial found both the
preferred primary therapy for DME or a recommended adjuvant severity and progression of retinopathy were associated with overt
therapy in eyes that have not responded successfully to anti-VEGF albuminuria.441–443 The presence of gross proteinuria at baseline is
therapy. Intraocular steroid treatment may have a role, however, associated with 95% increased risk of developing macular edema
for treatment of patients with DME who cannot receive anti- among patients with T1DM,424 and dialysis can improve macular
VEGF agents or who are pseudophakic before treatment. edema in diabetic patients with renal failure.

Despite these associations, a causal relationship between diabetic It is estimated that 25% to 40% of patients with T1DM and
kidney disease and diabetic retinopathy has not been established. 5% to 40% of patients with T2DM ultimately develop some fea-
The frequent coexistence of retinal and renal microangiopathies tures of diabetic kidney disease.454,455 Over 20% of patients with
with factors that affect both complications, such as associated T2DM already have diabetic kidney disease when they are diag-
hypertension and disease duration, make it difficult to establish.444 nosed with diabetes,456 and a further 20% to 40% develop dia-
Overall, it is important to carefully consider the renal status of any betic nephropathy, mostly within 10 years of diagnosis. Although
patient with diabetes and to ensure that the patient is receiving opti- nephropathy appears be more common in T1DM, because of the
mal care in this regard. In addition, rapidly progressive retinopathy, large and increasing number of persons with T2DM,449,457 more
especially in a patient with a long history of diabetes where retinop- than 80% of diabetic patients in renal replacement programs have
athy previously has been stable, should suggest the need for renal T2DM.
evaluation.
Low hematocrit was an independent risk factor in the Natural History of Nephropathy in Type 1
ETDRS analysis of baseline risk factors for development of
high-risk PDR and severe vision loss.445 A cross-sectional study Diabetes
involving 1691 patients revealed a twofold increased risk of any Nephropathy and specifically proteinuria in the setting of diabetes
retinopathy in patients with a hemoglobin level less than 12 g/ have been known for more than 100 years, and the classic struc-
dL compared to those with a higher hemoglobin concentration, tural features of glomerulosclerosis were described more than 70
using multivariate analyses controlling for serum creatinine, years ago.458 However, it is only since the 1980s that the natu-
proteinuria, and other factors.446 In patients with retinopathy, ral history of this condition has been extensively delineated. This
those with low hemoglobin levels have a fivefold increased risk is partly because significantly more patients are surviving to see
of severe retinopathy compared with those with higher hemo- the full presentation of this condition. For example, in 1971, the
globin levels. There have been limited reports of resolution of median survival time of patients with T1DM and overt nephropa-
macular edema and hard exudate with improvement or stabiliza- thy was 5 years, with less than 10% surviving more than 10 years.
tion of visual acuity in erythropoietin-treated patients after an Consequently, few patients were able to survive the course of their
increase in mean hematocrit.447 In view of the potential associa- renal disease. By comparison, in 1996, the median survival in
tion of low hematocrit and diabetic retinopathy, it is important an equivalent population was more than 17 years. Not surpris-
to ensure that patients with diabetic retinopathy and anemia are ingly, nearly 10 times more patients with T1DM are now entering
receiving appropriate management. ESRD programs.
In summary, diabetes is clearly a multisystem disease requir- Diabetic nephropathy is characterized clinically as a triad of
ing a comprehensive medical team approach. Even with regard to hypertension, proteinuria, and ultimately renal impairment.459
ocular health, this necessitates the involvement of multiple health The classic five stages of nephropathy as described by Mogensen
care specialists for optimal patient care. and colleagues,460 although not totally accurate, remain the best
way of describing this condition (Fig. 37.36). This description
Diabetic Nephropathy relies on functional evaluation of the renal disease and is based
on serial measurements of glomerular filtration rate (GFR) and
Diabetic nephropathy remains a major cause of morbidity and albuminuria.
death for persons with either T1DM or T2DM. In Western
countries, diabetes is the leading single cause of end-stage renal Stage 1: Hyperfiltration
disease.210 Indeed, in many countries such as the United States, The initial phase has been termed the hyperfiltration phase. It is
more than 50% of patients in renal replacement therapy programs associated with an elevation of glomerular filtration rate460 and
have diabetes as the major cause of their renal failure. However, presumably an increase in capillary glomerular pressure. Although
the full impact of diabetic nephropathy is far greater.448 Glob- invariably present in animal models of T1DM,461 an elevation of
ally most patients with diabetes are in developing countries that GFR occurs in up to 40% of type 1 diabetic patients. Hyper-
do not have the resources or health infrastructure to provide uni- filtration is considered to occur as a result of concomitant renal
versal renal replacement therapy.449 Even in developed countries, hypertrophy462 and partly due to a range of intrarenal hemody-
for every 20 patients with diabetes and chronic kidney disease, namic abnormalities that occur in the diabetic milieu that con-
less than one will survive to end-stage renal disease, succumbing tribute to glomerular hypertension.461 The pathophysiology of
instead to atherosclerotic cardiovascular disease, heart failure, or
infection. Furthermore, the presence and severity of diabetic renal
disease significantly contributes to the risk of these conditions. Glomerular Blood
filtration rate pressure
For example, almost all of the excess in cardiovascular deaths in
persons with diabetes younger than 50 years can be attributed to Hyperfiltration =/
nephropathy.450 Indeed, in T1DM subjects without nephropathy
there may be a lower risk of premature death,451 although a recent / Silent
review of the FinnDiane cohort indicates that in nonalbuminuric
T1DM subjects there is still evidence of increased mortality.452 In / Microalbuminuria
patients with T2DM, microalbuminuria is associated with a two-
fold to fourfold increase in the risk of death. In patients with overt =/ Macroalbuminuria
proteinuria and hypertension, the risk is even higher.453 Conse-
quently, the goal to reduce ESRD in patients with diabetes is only Renal impairment
one component as part of the overall benefit in preventing diabetic
kidney disease. • Fig. 37.36 The phases (natural history) of diabetic nephropathy.

renal hypertrophy associated with diabetes remains unexplained, structural changes, including basement membrane thickening and
although specific growth factors such as the growth hormone mesangial expansion. More recently, loss of the glomerular cells
(GH)/insulin-like growth hormone (IGF1) system and transform- known as podocytes has also been described as a relatively early
ing growth factor β (TGFβ) have been implicated.463,464 Notably, structural change in the diabetic kidney. Indeed, by performing
there is not only glomerular but also tubular hypertrophy. Indeed, detailed quantitative studies of renal morphology it is often possible
the tubular hypertrophy explains the increased kidney weight in to detect those who will develop renal damage.475 This is a very
diabetes because tubules make up more than 90% of the kidney important phase clinically because it is hoped that investigators will
weight.465 In addition, increased salt reabsorption associated with be able to develop new tests, such as biomarkers in plasma or urine
proximal tubular hypertrophy can also contribute to glomerular or sophisticated assessments from renal biopsy material, to identify
hyperfiltration via tubuloglomerular feedback.463 which patients will progress to more advanced renal disease. Because
The second explanation for the increase in GFR associated overall less than 40% of T1DM subjects will progress, it is critical
with diabetes relates to hemodynamic changes within the kidney. that we detect those who are the potential progressors and could
Although not directly tested in humans, micropuncture studies be candidates for early prevention and treatment strategies to avoid
in rodents, particularly by Brenner’s group in the 1980s, revealed end-stage renal disease. As yet, no reliable clinically translatable sur-
that experimental diabetes was associated with a range of intrare- rogate markers or predictors have been identified as worth pursuing
nal hemodynamic changes.461 Alongside hyperfiltration, there is in clinical practice during this silent phase of the disease.
an increase in effective renal plasma flow, and thus some investi- Extensive studies have tested various plasma markers, such
gators call this the hyperperfusion-hyperfiltration phase of diabetic as prorenin,476 tumor necrosis factor α (TNFα) receptor,477
nephropathy. At the same time, intraglomerular capillary pressure advanced glycation end products,478 and precursors such as meth-
is increased, reflecting relative efferent versus afferent arteriolar ylglyoxal,479 but the predictive value has not been conclusive.
vasoconstriction461 with activation of the intrarenal renin-angio- Although diabetic kidney disease probably has at least in part a
tensin system (RAS) and reduced synthesis of the vasodilator genetic basis, few potentially relevant gene polymorphisms such as
nitric oxide. the angiotensin-converting enzyme gene have been identified.480
The importance of this hyperfiltration phase in predict- Recent genome-wide association studies have not generally been
ing and leading to diabetic nephropathy remains controversial. very helpful.481,482 The measurement of albumin fragments (ghost
Several groups have confirmed the initial relationship between albumin) in the urine of patients with diabetes may be another,
elevated GFR and later development of proteinuria as described albeit unproven, approach.483 Serial prospective ambulatory
by Mogensen and Christensen.466 However, this has not been a blood pressure monitoring studies have also demonstrated mod-
universal finding. Nevertheless, subsequent studies with anti- est rises in blood pressure in patients in this silent phase up to
hypertensive agents, and in particular agents that interrupt the 5 years before urinary albumin excretion begins to increase.484
renin-angiotensin system, have shown attenuation of some of However, none of these markers has been proved to be sensitive
these glomerular hemodynamic abnormalities.467 This provides or specific enough on further clinical evaluation for widespread
justification to consider that at least some of these intrarenal clinical application.
hemodynamic changes in diabetes play a role in the development
and progression of nephropathy. However, in insulin-dependent Stage 3: Microalbuminuria
diabetic patients the ACE inhibitor captopril had the smallest The third phase is known as microalbuminuria or the stage of
effect (17%) in reducing the risk of doubling serum creatinine in incipient nephropathy. At this stage, often 5 to 15 years after the
patients with serum creatinine of 1 mg/dL and the largest effect initial diagnosis of T1DM, the urinary albumin excretion rate has
(76%) in patients whose serum creatinine was 2 mg/dL. These increased into the microalbuminuric range of 20 to 200 μg per
data do not seem to support a primary role for early hyperfiltra- minute or 30 to 300 mg per 24 hours.485 In the past, microal-
tion in the development of early diabetic nephropathy. Results buminuria was considered to be a predictor rather than a mani-
suggest that in patients with type 2 diabetes with advanced renal festation of diabetic kidney disease. Increasingly it has been
disease, a link to hyperfiltration was not apparent.468,469 appreciated, particularly when based on interpretation of renal
This field has recently been reactivated with the advent of morphologic studies, that in the microalbuminuric phase there is
sodium glucose cotransporter-2 (SGLT2) inhibitors, which influ- often but not always widespread evidence of advanced glomerular
ence tubuloglomerular feedback and have been shown to reduce structural changes.486,487 Concomitant with these changes, sys-
intraglomerular pressure via effects on afferent arteriolar dila- tolic and diastolic blood pressures are increased. Furthermore, the
tion.470 Indeed, these agents have impressive effects in attenuating nocturnal dip in blood pressure seen in normal persons is often
diabetes-associated hyperfiltration, as reported initially in T1DM lost with the development of microalbuminuria.488 Renal func-
but subsequently in T2DM.471,472 Whether these hemodynamic tion during this phase may be increased, normal, or reduced.
benefits explain, at least in part, the renoprotection seen with The best approach to screen for microalbuminuria remains
these agents in large clinical studies, including Empagliflozin controversial. The original studies used 24-hour or overnight urine
Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus sampling methods. However, a spot urine albumin:creatinine ratio
Patients–Removing Excess Glucose (EMPA-REG OUTCOME) in an early-morning urine specimen has been validated and appears
and Canagliflozin Cardiovascular Assessment Study (CANVAS), to be a practical option for routine clinical practice.210,485 Because
remains to be determined.473,474 the onset of persistent microalbuminuria, if left untreated, is often
a reliable harbinger of overt nephropathy,466 it is incumbent on
Stage 2: The Silent Stage clinicians to perform serial measurements of this parameter and to
The next stage is known as the silent stage, where, from a clini- repeat the measurement if there is an isolated elevation in urinary
cal point of view, there is no overt evidence of any form of renal albumin excretion.
dysfunction. Patients usually have normal GFR with no evidence Studies suggest that in many patients with T1DM, microalbu-
of albuminuria. However, this phase is associated with significant minuria can be transient and can reverse to normoalbuminuria.489

Thus the onset of microalbuminuria does not irrevocably seal the Natural History of Nephropathy in Type 2
fate of the patient. A study of 386 patients with persistent micro-
albuminuria at study entry showed that regression of microalbu- Diabetes
minuria occurred in 58% of patients,489 although other groups The natural history of diabetic nephropathy in patients with
have reported much lower rates of this phenomenon.490 Notably, T2DM is less well understood than in patients with T1DM.
in that study, microalbuminuria of short duration, optimal levels This partly reflects the fact that T2DM is largely a disease of an
of HbA1c (<8%), low systolic blood pressure (<115 mm Hg), and older population, with associated obesity, hypertension, dyslip-
low levels of both cholesterol and triglycerides were independently idemia, and high rates of cardiovascular disease, which are all
associated with the regression of microalbuminuria. Therefore factors associated with increased risk of renal disease. In addi-
screening of diabetic patients for nephropathy is now recom- tion, approximately 7% of patients with T2DM already have
mended to include at least twice-annual measurements of urinary microalbuminuria at the time of diagnosis. This may be partly
albumin concentrations in T1DM patients. related to the fact that most of these patients have had untreated
diabetes for 5 to 10 years (on average) before diagnosis. Within
Stage 4: Macroalbuminuria 5 years of a diagnosis of T2DM, up to 18% of patients have
The next stage is the macroalbuminuria phase or overt nephropathy. microalbuminuria, especially those with poor metabolic control
This stage represents the phase that has been previously described and high blood pressure. This has led some investigators to sug-
as diabetic nephropathy and is highly predictive of subsequent gest that nephropathy in T2DM is different from that seen in
renal failure if left untreated. It is characterized by a urinary albu- patients with T1DM.
min excretion rate greater than 300 mg per 24 hours (200 μg/ However, the natural history of nephropathy in T2DM has
minute). This phase usually occurs after 10 to 15 years of diabetes, more similarities than differences from that seen in T1DM.
but the risk of overt renal disease never truly disappears and can Hyperfiltration does occur in T2DM,494 although it has been
appear after 40 or 50 years of T1DM. reported to be less commonly observed than in T1DM. This
There are at least two peaks of incidence of overt nephropathy, observation must be interpreted with caution because GFR
and this has been termed by some investigators as representing normally declines with age, and hyperfiltration can still exist,
slow and fast trackers.491 The key contributors to this marked although the GFR remains in the normal adult range. Micro-
variation in the timing of onset of proteinuria, independent of albuminuria also occurs in T2DM. However, the finding of
glycemic or blood pressure control, remain elusive, although microalbuminuria in T2DM might not be as specific for diabetic
a range of genetic, molecular, and environmental factors have renal disease as described in the seminal studies in T1DM. In
been proposed. In association with this increase in proteinuria, the context of a very high prevalence of cardiovascular disease,
more than two-thirds of patients have overt systemic hyper- microalbuminuria may be more closely associated with nonre-
tension.492 During this phase, if left untreated, blood pressure nal events such as stroke and myocardial infarction.439 Further-
continues to rise, accelerating the decline in GFR, which pro- more, incipient or overt cardiac failure, urinary tract infection,
motes a further rise in blood pressure, creating a vicious cycle of and urinary obstruction (e.g., enlarged prostate) can also lead to
progressive renal impairment that ultimately leads to end-stage microalbuminuria.
renal disease. Many patients with T2DM and microalbuminuria also
progress to overt proteinuria. However, it is increasingly
Stage 5: Uremia appreciated that the situation has become more complex, and
The final uremic phase, which historically occurred in up to 40% many groups have now described subjects with T1DM495 as
of T1DM subjects,454 requires the institution of renal replacement well as T2DM496 who develop renal impairment with progres-
therapy. As recently as the 1970s, patients with diabetes were not sive decline in GFR in the absence of significant proteinuria.
considered candidates for renal replacement therapy because of The exact explanation for this phenomenon is unknown, and
their abysmal prognosis. However, improvements in the manage- ongoing studies are exploring if these patients have different
ment of cardiovascular disease and renal replacement options have renal morphologic changes from those with the more clas-
seen the survival on dialysis approach that of patients with renal sic syndrome of diabetic nephropathy: overt proteinuria and
disease from other causes. Many patients with diabetes and end- declining GFR. Preliminary studies suggest a prominent vascu-
stage renal disease are also now considered candidates for renal lar component for this form of nonproteinuric renal dysfunc-
transplantation, which is associated with better outcomes than tion. Nonetheless, it appears that the risk of end-stage renal
remaining on dialysis. However, there is evidence that the renal disease in patients with T2DM and renal impairment is simi-
lesions of diabetes often recur in the transplanted kidney, though lar in the presence or absence of microalbuminuria, underlin-
the lead time to develop ESRD means that few kidneys are lost ing the importance of an estimated GFR in the management
through recurrent disease. of patients with T2DM. This has led to many of the national
Increasingly, single pancreas-kidney (SPK) and pancreas-after- and international guidelines now recommending the inclusion
kidney (PAK) transplantation have become therapeutic options of regular measurements of serum creatinine and determina-
for patients with T1DM and end-stage renal disease, and these tion of estimated GFR using a variety of different formulas.
newer options appear to offer advantages over kidney-alone trans- The frequency of these measurements differs among the vari-
plantation. In particular there is some evidence that maintaining ous guidelines, but at a minimum these measurements should
euglycemia following pancreas transplantation can lead to reso- be performed on a yearly basis. Importantly, in T2DM, many
lution of many diabetes-related renal lesions such as mesangial subjects will develop a progressive decline in GFR without the
expansion.493 This reversal is often not apparent until after 10 development of albuminuria. Indeed, recent renal biopsy stud-
years of euglycemia, emphasizing the slow turnover of matrix and ies although not conclusive suggest that there may be a dif-
the potential long-term effects of hyperglycemic memory on the ference in renal morphologic lesions between albuminuric and
kidney. nonalbuminuric forms of diabetic kidney disease.497

Metabolic Hemodynamic In addition to the mechanisms described here, the diabetic

kidney appears to be readily modulated by a range of vaso-
Glucose active hormones. Indeed, it is increasingly appreciated that
Advanced glycation Flow/pressure
Oxidative stress Renin angiotensin there may be important interactions between metabolic path-
ways and various hemodynamic factors, including vasoactive
Intracellular signaling molecules
hormones such as angiotensin II in mediating renal injury
in diabetes (see Fig. 37.37).504,505 Although many drugs that
modulate hormone levels or action might not be specific for
Growth factors and cytokines diabetic kidney disease, interruption of the renin-angiotensin
system appears to be an excellent approach not only for reduc-
ing blood pressure but also for correcting many of the cellular,
Diabetic complications biochemical, hemodynamic, and structural abnormalities seen
in the diabetic kidney. These agents appear to be very powerful
• Fig. 37.37 Interactions between metabolic and hemodynamic factors in antiproteinuric agents at all stages of renal disease, including
promoting diabetic complications, including nephropathy. microalbuminuria and macroalbuminuria, although the exact
mechanism of action remains to be fully defined. Furthermore,
Pathogenesis of Diabetic Nephropathy these agents influence renal structure although these findings
are primarily deduced from preclinical studies. In addition, as
It is likely that many of the mechanisms implicated in dia- seen in the Reduction in Endpoints in Noninsulin-Dependent
betic microvascular complications, in general, play a central Diabetes Mellitus with the Angiotensin II Antagonist Losartan
role in the development and progression of diabetic nephropa- (RENAAL) study,506 they also attenuate decline in renal func-
thy (Fig. 37.37).43 It is evident that hyperglycemia is neces- tion in type 2 subjects with advanced renal disease. Based on
sary for the initiation of renal injury, because patients without the discovery in the late 1990s that proteinuria in a range of
diabetes do not develop this type of nephropathy. Moreover, nephropathies could occur as a result of molecular and struc-
intensive therapy designed to achieve improved glycemic con- tural abnormalities in a highly specialized structure known as
trol is able to attenuate the development of nephropathy, as the slit diaphragm within the glomerular epithelial cell (podo-
assessed by urinary albumin excretion, although it is not fully cyte), a number of experimental studies, subsequently con-
prevented.334 However, it is now clear that other factors must firmed in humans, showed that depletion of one of these slit
also be involved because continuous florid hyperglycemia is pore proteins, nephrin, could be attenuated or prevented by
not necessarily required for diabetic hyperfiltration and kidney agents that interrupt the RAS.507
growth to occur. Indeed, glomerular hyperfiltration and tubu- In addition to promoting glomerular nephrin depletion, angio-
lar hypertrophy can persist in patients with T1DM even after tensin II also appears to have other actions that promote the devel-
euglycemia is achieved through aggressive insulin therapy.498 opment of proteinuria, including trophic effects on the kidney
Other pathways that may be involved in diabetic nephropa- and increasing glomerular membrane pore size.508 Although many
thy include generation of mitochondrial ROS, accumulation of investigators have focused on the RAS and, in particular, the vaso-
AGEs, and activation of intracellular signaling molecules such as constrictor angiotensin II, it is increasingly appreciated that other
PKC.131,499 Many of the seminal studies performed in endothe- vasoconstrictors may be important. These include endothelin and
lial cells demonstrating a central role for mitochondrial ROS in a number of vasodilators, such as nitric oxide, bradykinin, atrial
activating pathways implicated in diabetic vascular complications natriuretic peptide, and vasodilatory angiotensins such as angio-
have been reproduced in mesangial cells.500 Advanced glycation, tensin 1-7.509 This exploration of the role of vasoactive hormones
which occurs at an accelerated rate in diabetic patients, is a promi- and their respective receptors in the diabetic kidney is critical for
nent phenomenon in the kidney. Not only is the kidney the major designing new treatments for this condition, because these path-
site for excretion of AGEs, but also many of the proteins with a ways are ideal targets for drug development. This point has already
long life, such as collagen, are extensively glycated in patients with been demonstrated for agents that interrupt the renin-angiotensin
diabetes.501 Furthermore, various AGE receptors (e.g., RAGE) system, including ACE inhibitors and ARBs. However, the use
have been described in the kidney, which appear to play a role in of dual RAS inhibition, specifically the combination of an ACE
mediating some of the deleterious effects of AGEs, such as stim- inhibitor and an ARB, despite positive effects on albuminuria,510
ulation of growth factor expression and induction of important is not recommended due to increased risks of hyperkalemia and
phenotypic changes within certain renal cell populations to pro- acute kidney injury, as was also seen with RAS blockade with a
mote scarring.56 Preliminary studies have used various approaches combination of a renin inhibitor and an ARB.511
to inhibit renal AGE accumulation and action, including a form
of soluble RAGE (sRAGE). A range of pharmacologic agents has
shown promising results, but clinical translation of these findings
Pathology of Diabetic Renal Disease
remains to be fully defined.501 Selective PKC isoform inhibitors Diabetic renal disease was originally described as a glomeru-
have been evaluated in small clinical trials, but their role in renal lopathy associated with diffuse or nodular glomerulosclerosis.458
disease remains controversial.502 Some exciting pilot studies evalu- Subsequent studies using electron microscopy have revealed
ating a number of cytosolic sources of oxidative stress, such as nic- that glomerular basement membrane thickening and mesangial
otinamide adenine dinucleotide phosphate oxidase, suggest that expansion are prominent glomerular abnormalities in diabe-
certain NADPH oxidase isoforms such as Nox4 may be excellent tes475 (Fig. 37.38). Indeed, prospective studies have shown that
targets for new renoprotective therapies.503 This is strengthened these changes predict to a certain degree the development of
by the advent of orally bioavailable NADPH oxidase inhibitors,136 overt renal disease in patients with T1DM. However, fewer than
which are now under clinical investigation. one-third of diabetic patients with microalbuminuria have the

Tubulopathy

Tubular hyperplasia and hypertrophy

Glomerulopathy
Progressive and cumulative atrophy
Mesangial expansion
Thickening of the TBM
Glomerular hypertension
Epithelial mesenchymal transition
Diffuse thickening of the GBM
Accumulation of lysosomal bodies
Broadening of foot processes
Armani-Ebstein lesion
Podocyte loss
Reduced tubular brush border
Reduced slit pore proteins
Increased tubular salt reabsorption
Glomerulomegaly
Increased Na+/H+ antiporter activity
Kimmelstiel-Wilson lesion
Impaired tubular acidification
Adhesions to Bowman’s capsule
Abnormal tubuloglomerular feedback
Neovascularization
Decreased endocytosis of protein
Nodular and diffuse
glomerulosclerosis
Abnormal lysosomal processing

Impaired uptake of organic ions

• Fig. 37.38 Glomerular and tubular manifestations of diabetic nephropathy. GBM, Glomerular basement
membrane; TBM, tubular basement membrane.

typical glomerulopathy described by Kimmelsteil and Wilson Other Renal Manifestations of Diabetes
in 1936.458,512 Although initial studies emphasized the mesan-
gial cell changes in the glomerulus, glomerular epithelial cell Renal Artery Stenosis
abnormalities represent new areas of active research.507 Podo- Because diabetic patients have, in general, an increased burden of
cyte dysfunction and subsequent apoptosis, ultimately leading atherosclerosis, they appear to have a higher risk of renal artery ste-
to depletion of podocytes within the glomerulus appears to play nosis. However, although angiographic studies have demonstrated
a pivotal role in the development of proteinuria in diabetes.513 a high prevalence of renal artery stenosis in diabetic patients, these
Although most of the focus has been on glomerular changes in lesions are often of no hemodynamic significance. Nevertheless, a
the diabetic kidney, more recent studies have identified important small subgroup will have a hemodynamically significant stenosis,
changes in the other sites within the kidney, including the tubules, enhancing hypertension, increasing the risk of acute pulmonary
interstitium, medulla, and papilla.465 Diabetic tubulopathy is edema, and inducing progressive renal impairment.515 In such
characterized by a variety of structural and functional changes, subjects, specific interventions such as surgery or angioplasty need
including tubuloepithelial cell hypertrophy, tubular basement to be considered.516 Furthermore, some patients have bilateral
membrane thickening, epithelial-mesenchymal transition,514 and renal artery stenosis that, on commencement of an agent such as
the accumulation of glycogen (see Fig. 37.38). There is also an an ACE inhibitor, can lead to acute renal failure.517 Fortunately,
expansion of the interstitial space with infiltration of various cell in most patients, if the renal failure is diagnosed early, cessation
types, including myofibroblasts and macrophages. of the ACE inhibitor leads to rapid restoration of renal function
These tubular changes represent more than just the aftermath in this situation.
of diabetic nephropathy. The dysregulation of tubular functions
in diabetes can precede or at least accompany the changes in Renal Papillary Necrosis
the renal glomerulus and the onset of albuminuria.512 Indeed, Renal papillary necrosis involves a severe destructive process,
the functional and structural changes in the proximal tubule presumably as a result of ischemia to the medulla and papilla.518
may be key to contributing to the development and progression Beethoven’s final illness might have been papillary necrosis
of diabetic nephropathy.465 For example, it has been suggested in the context of diabetes.519 The papilla is very sensitive to
that tubuloglomerular feedback mechanisms can drive hyper- these ischemic changes because even in the normal setting it
filtration associated with diabetes462 and that tubular dysfunc- is exposed to a relatively hypoxic environment. Concomitant
tion can contribute to albuminuria due to defective uptake and exacerbating factors include urinary tract infection and analge-
lysosomal processing.483 Indeed, renal function and prognosis sic abuse. The importance of ischemia and possibly angiotensin
correlate better with structural lesions in the tubules and corti- II in this disorder has been suggested in experimental stud-
cal interstitium than with classic glomerular changes of diabetic ies in transgenic rats that overexpress renin and angiotensin
nephropathy. II in their kidney after induction of diabetes.520 In these rats,

Microalbuminuria: 31–299 mg/day

Immediate Treatment targets

• Treat hypertension • BP <135/75 mm Hg (ACE inhibitor)
• Strive for euglycemia • Hemoglobin A1c <7%
• Reduce hyperlipidemia • LDL cholesterol <100 mg/dL (statin)

Baseline/periodic Monitoring
• Electrocardiogram • Urinary protein
• Echocardiography • Creatinine clearance
• Dobutamine stress test • Retinopathy (cataracts)
• Urine culture • Cardiac integrity
• Fluorescein angiography • Bone density
• Doppler limb flow • Peripheral perfusion
• Neurologic stability
• Psychosocial adjustment

Assess comorbid conditions

• Persistent angina
• Congestive heart failure, cardiomyopathy
• Respiratory disease
• Autonomic neuropathy: gastroparesis, obstipation,
diarrhea, cystopathy, orthostatic hypotension
• Neurologic: cerebrovascular accident or stroke residual
• Musculoskeletal disorders, renal bone disease
• Infections: HIV, hepatitis, indolent ulcers
• Hematologic problems other than anemia
• Vision impairment (decreased acuity to blindness) loss

• Fig. 37.39 Flow chart illustrating the management of diabetic nephropathy before the onset of renal fail-
ure. ACE, Angiotensin-converting enzyme; BP, blood pressure; HIV, human immunodeficiency virus; LDL,
low-density lipoprotein.

diabetes was associated with development of papillary necrosis, Contrast-Induced Nephropathy

which was prevented by blockade of the renin-angiotensin sys- Because many diabetic patients have impaired renal function, they
tem. Clinically, papillary necrosis often manifests as flank pain, are at high risk of increased renal impairment from certain neph-
hematuria, and fever. Urinalysis reveals red and white blood rotoxic agents. One of the most important risks relates to radio-
cells, bacteria, and papillary fragments. Ureteric obstruction contrast dyes.524 Where possible, patients with diabetes and renal
can occur as a result of these fragments and must be addressed impairment should avoid imaging studies that involve contrast
as an emergency. and, in particular, multiple studies performed in rapid succession.
Where intravenous contrast forms an indispensable tool to man-
Renal Tubular Acidosis agement, low-osmolality, nonionic, or gadolinium-based contrast
A well-known functional abnormality associated with diabetic media may be less nephrotoxic in patients with renal failure.525,526
tubulopathy is renal tubular acidosis (type 4), manifesting as hyper- It has been suggested that patients who require such procedures
kalemia and hyperchloremic metabolic acidosis.521 This is thought should be well hydrated before, during, and after the procedure.
to be a manifestation of hyporeninemic hypoaldosteronism asso- However, a recent report suggests that this approach may not be as
ciated with diabetes, resulting in proximal tubule ammonia pro- effective as previously presumed.527 The role of N-acetylcysteine,
duction being reduced to levels inadequate to buffer acid in the a thiol-containing antioxidant, shows promise to protect against
distal nephron. The precise cause of this abnormality remains contrast-induced nephropathy.528 The oral hypoglycemic drug
to be established. In some patients, there appears to be a defect metformin should also be discontinued before contrast procedures
in the conversion of prorenin to active renin.522 It has also been to prevent life-threatening lactic acidosis.
suggested that damage to the tubular cells of the juxtaglomeru-
lar apparatus associated with diabetes can contribute to impaired Management of Diabetic Kidney Disease
renin release, possibly due to reduced renal prostaglandin produc-
tion and elevated vasopressin levels.523 Blood pressure and glycemic control represent the major corner-
A major risk associated with hyporeninemic hypoaldosteron- stones for preventing and treating diabetic nephropathy (Figs.
ism is the development of life-threatening hyperkalemia. This is 37.39 and 37.40). In the early 1980s, a number of Scandinavian
an increasingly important issue with the widespread use of ACE researchers found that aggressive blood pressure reduction reduces
inhibitors and ARBs, often in combination, in this population. the rate of progression of diabetic nephropathy,529,530 and in the
This is further exacerbated by the use of potassium-sparing diuret- 1990s other researchers found that intensified glycemic control has
ics (such as spironolactone) and beta blockers. a similar benefit in both T1DM (DCCT)9 and T2DM (UKPDS)

Proteinuria: >299 mg/day

Nephrosis: >3.5 g/day

Immediate Treatment targets

• Treat hypertension • Attain dry weight: diuretics
• Extract edema • BP <135/75 mm Hg: ACE inhibitors,
• Correct anemia multiple drugs
• Strive for euglycemia • Hematocrit >35%: erythropoietin
• Reduce hyperlipidemia • Hemoglobin A1c <7%: metabolic control
• LDL cholesterol <100 mg/dL: diet + control
• Educate patient re: regimen applied

Assess comorbid conditions Monitoring

• Persistent angina • Urinary protein

• Congestive heart failure, cardiomyopathy • Creatinine clearance
• Respiratory disease • Retinopathy (cataracts)
• Autonomic neuropathy: gastroparesis, obstipation, • Cardiac integrity
diarrhea, cystopathy, orthostatic hypotension • Bone density
• Neurologic: cerebrovascular accident or stroke residual • Peripheral perfusion
• Musculoskeletal disorders, renal bone disease • Neurologic stability
• Infections: HIV, hepatitis, indolent ulcers • Psychosocial adjustment
• Hematologic problems other than anemia
• Vision impairment (decreased acuity to blindness, loss)

• Fig. 37.40 Flow chart illustrating the management of diabetic nephropathy after the onset of clinical pro-
teinuria. ACE, Angiotensin-converting enzyme; BP, blood pressure; HIV, human immunodeficiency virus;
LDL, low-density lipoprotein.

diabetic subjects. These findings have led to the view that optimi- It remains to be determined how useful the intensification of
zation of blood pressure and plasma glucose levels should be the glycemic control is in the setting of overt nephropathy as a last-
mainstay of therapy for diabetic nephropathy. ditch strategy to delay the onset of end-stage renal disease. Aggres-
sive management of hypertension and lipid lowering are clearly
Glycemic Control in Diabetic Kidney Disease more important than glycemic control in reducing cardiovascular
The importance of glucose as a factor in the progression of dia- events and slowing renal disease progression at this stage of rela-
betic kidney disease, as initially suggested from epidemiologic tively advanced disease, although some studies suggest that poor
and preclinical studies, was clearly demonstrated in the DCCT glycemic control can accelerate the loss of renal function in dia-
study in patients with T1DM.9 In both the primary and sec- betic nephropathy.536 However, a number of large studies have
ondary prevention arm of the study, any decrease in HbA1c was failed to show any evidence that strict glycemic control per se
strongly associated with a reduction in the risk of development retards renal progression once overt nephropathy is present.537 In
of microalbuminuria as well as a decrease in the risk of progres- fact, in the Veterans Administration Diabetes Trial (VADT) tight
sion to overt nephropathy. The follow-up EDIC study has con- glycemic control reduced albuminuria but failed to demonstrate
firmed long-lasting benefits of this therapeutic approach.531 The any effect on GFR.538 In addition, as renal function fails, tight
UKPDS clearly demonstrated a role for intensified glycemic con- glycemic control becomes more hazardous, with an increased risk
trol in newly diagnosed T2DM subjects when treatment led to a of hypoglycemia. Nonetheless, because there is sufficient evidence
reduction in HbA1c from 7.9% to 7%. The Action in Diabetes that glycemic control can reduce both macrovascular events and
and Vascular Disease: Preterax and Diamicron MR Controlled microvascular complications of diabetes at other sites, it is reason-
Evaluation (ADVANCE) study has demonstrated that a further able to suggest that optimization of metabolic control in patients
reduction of HbA1c to an average of 6.5% was associated with a with overt nephropathy remains worthwhile.
further reduction in renal events, as assessed by the development The choice of agent, however, remains controversial. Certainly,
and progression of microalbuminuria.532 A subsequent evaluation several types of drugs are able to improve glycemic control in
of this study533 demonstrated intensified glycemic control that patients with T2DM539; however, the particular advantages of one
reduced the development of end-stage renal disease, emphasizing class over another for preventing and treating diabetic nephropa-
the fact that tighter glycemic control continues to confer renal thy remain to be established. Nevertheless, this issue appears to
benefits even in the setting of more advanced renal disease. This have become more clinically relevant with the widespread use of
renal benefit persisted, as seen in the follow-up study known as newer classes of glucose-lowering drugs, including dipeptidyl pep-
ADVANCE-ON.534 Thus despite the ongoing controversy as to tidase 4 (DDP4) inhibitors, GLP1 analogs, and SGLT2 inhibi-
the appropriate HbA1c target to reduce macrovascular disease as a tors.539 With respect to DPP4 inhibitors, modest renal benefits
result of the findings from the Action to Control Cardiovascular have been suggested, particularly with saxagliptin in the Saxa-
Risk in Diabetes (ACCORD) study,535 no such controversy as to a gliptin Assessment of Vascular Outcomes Recorded in Patients
possible deleterious effect of intensified glycemic control has been with Diabetes Mellitus (SAVOR) study540 and in pooled analy-
reported with respect to nephropathy. ses with linagliptin.541 Pending results from studies such as the

Cardiovascular and Renal Microvascular Outcome Study With patients with overt nephropathy. In these patients, it has been sug-
Linagliptin in Patients With Type 2 Diabetes Mellitus (CARME- gested that optimal blood pressure control is less than 125/75 mm
LINA)542 will further clarify the potential renoprotective role of Hg.547 Indeed, in a subanalysis from the blood pressure arm of
these agents. Recent studies have been performed with GLP1 ana- the ADVANCE study, no blood pressure threshold was detected.
logues, including liraglutide and semaglutide.543–545 Both agents, Specifically, in those subjects in whom achieved blood pressure
in addition to conferring cardiovascular protection, afforded renal was reduced to levels even lower than currently recommended in
benefits. In the Liraglutide Effect and Action in Diabetes: Evalu- national and international guidelines, there was a further decrease
ation of Cardiovascular Outcome Results (LEADER) trial, lira- in renal events.548 Thus it is possible that if individuals can toler-
glutide-treated patients had a 22% lower incidence of new onset ate lower blood pressures without major side effects such as diz-
of persistent macroalbuminuria and a slower decline in the eGFR ziness and syncope it may be worth considering treating some
over time, particularly in the subgroup of patients who had evi- T2DM subjects to blood pressure levels below those currently rec-
dence of kidney damage at baseline.545 In the Trial to Evaluate ommended. However, syncope is often a side effect limiting the
Cardiovascular and Other Long-term Outcomes with Semaglu- options to intensify blood pressure–lowering therapy, particularly
tide in Subjects with Type 2 Diabetes (SUSTAIN 6), semaglu- in those diabetic subjects with systolic hypertension as a result of
tide-treated patients had a similar reduction in renal outcomes. diabetes-associated vascular stiffness.
However, semaglutide-treated patients also had increased risk of There is good evidence that tight blood pressure control, no mat-
serious diabetic retinopathy, including 1.91-fold increased risk of ter how it was achieved, is associated with a significant reduction
retinal photocoagulation and a 2.29-fold increased risk of vitre- in the risk of microalbuminuria (primary prevention). Although
ous hemorrhage,543,544 which may be due to more severe retinal blood pressure reduction appears paramount, there is also evidence
disease at baseline and rapid glycemic improvements. Finally, that ACE inhibitors549 have renoprotective actions beyond their
SGLT2 inhibitors have been shown to have strong renoprotec- antihypertensive effects for primary prevention.550,551 However, if
tive effects. For example, the EMPA-REG OUTCOME study treatment should commence in the normoalbuminuric stage, such
demonstrated an approximately 50% decrease in composite renal a strategy would involve treating the majority of patients who are
outcomes, including benefits on reducing hard renal endpoints not at risk of nephropathy. Ideally, it would be useful to be able
such as end-stage renal failure.474 Similar benefits have also been to identify patients, still normoalbuminuric, whose likelihood of
reported in the CANVAS study using the SGLT2 inhibitor cana- progression is increased. As of yet, no such markers of predisposi-
gliflozin.473 Of major interest is the finding that renoprotection tion to renal disease are available, although serum prorenin476 and
is observed even in the absence of glucose lowering with these modest elevations in urinary albumin excretion albeit still within
agents as seen in subjects with eGFR below 45 mL per minute.546 the normal range (borderline microalbuminuria)489 might ulti-
This suggests that these agents confer renoprotection at least in mately be examples of such markers.
part in a glucose-independent manner, but the mechanism(s) are The issue of primary prevention has been readdressed in two
unknown. Changes in sodium balance, intrarenal hemodynam- studies in which subjects with normoalbuminuria were treated
ics, plasma volume, decreased insulin resistance, and triglyceride- with agents that interrupt the RAS. In the first smaller trial, which
rich lipoprotein production and modest actions in reducing blood included the performance of sequential renal biopsies, no benefit
pressure, body weight, and uric acid have all been suggested as of early institution of either the ACE enalapril or the angiotensin
possible mechanisms. II receptor antagonist losartan was observed, as defined not only
A number of differences in the side effect profiles should influ- as a lack of effect on albuminuria but also no significant retarda-
ence prescribing habits. In patients with renal impairment, particular tion in progression of renal morphologic injury despite benefits
care must be exercised in selecting and dosing oral diabetic therapy on retinopathy.552 In the second, much larger trial known as the
because an accumulation of either the drug or active metabolites Diabetic Retinopathy Candesartan Trials (DIRECT) study, the
can lead to hypoglycemia (e.g., with glyburide) and other serious angiotensin II antagonist candesartan, despite some modest reti-
adverse effects such as lactic acidosis (with metformin). Thiazolidin- noprotective effects, had no major impact on reducing the new
ediones (TZDs), such as pioglitazone and rosiglitazone, should be onset of microalbuminuria.553
used with caution in patients with advanced nephropathy who have In secondary prevention studies, the additional benefits achieved
or are at risk of heart failure, although renal impairment per se is not from blocking the renin-angiotensin system are clearer. A meta-anal-
a contraindication for this class of drugs. ysis incorporating the findings of more than 10 studies in patients
with microalbuminuria has demonstrated the ability of ACE inhibi-
Blood Pressure Control in Diabetic Kidney Disease tors not only to retard the development of overt proteinuria but
A sustained reduction in blood pressure appears to be one of the also to decrease urinary albumin excretion by more than 30%. In
most important interventions to prevent progressive nephropathy some patients with microalbuminuria, ACE inhibition can reduce
in T1DM and T2DM. For example, in the UKPDS, a reduction urinary albumin excretion into the normoalbuminuric range.550 In
in blood pressure from 154 to 144 mm Hg was associated with patients with T1DM and overt proteinuria, aggressive blood pres-
a 30% reduction in microalbuminuria. All national and interna- sure reduction reduced proteinuria by up to 50% and retarded the
tional guidelines now emphasize the importance of blood pressure rate of decline in renal function.529,530
reduction in the diabetic patient. Although many guidelines sug- Similar studies have been performed in T2DM patients. Two
gest that specific targets should be achieved, no such threshold landmark trials, RENAAL and Irbesartan in Diabetic Nephropa-
appears to exist for any renal endpoint in patients with diabetes. thy Trial (IDNT), examined the renoprotective effects of the
In particular, the risk of progressive diabetic nephropathy contin- ARBs losartan and irbesartan, respectively.506,469 In both stud-
ues to decrease, with blood pressure reductions into the normal ies, when compared to various alternative antihypertensive agents
range and below, meaning that the lowest achievable blood pres- such as calcium antagonists (but not ACE inhibitors), ARB treat-
sure is associated with the best clinical outcomes. This is particu- ment was associated with a reduction in end-stage renal failure, a
larly important in those with the greatest risk of renal damage, greater than 30% decrease in proteinuria, and a major reduction

in hospitalization for heart failure. As a result of these studies, The role of lipid-lowering agents as renoprotective drugs
ARBs are recommended as first-line treatment for blood pressure remains controversial. Although in rodents a large body of evi-
reduction in T2DM patients with overt proteinuria.554 dence suggests that lipids promote renal injury and that various
Although ACE inhibitors have not been as extensively stud- lipid-lowering drugs reduce nephropathy, even in the setting of no
ied in this population, the Diabetics Exposed to Telmisartan and or minimal effect on lipids,565 the data in humans are variable.566
Enalapril (DETAIL) trial suggested similar renoprotective actions However, in a study of fenofibrate in T2DM, known as the Feno-
for both drug classes.555 Similar findings comparing the angioten- fibrate Intervention and Event Lowering in Diabetes (FIELD)
sin II antagonist telmisartan to the ACE inhibitor, ramipril, are study, there was an impressive reduction in albuminuria.567,568
also reported in the much larger ONTARGET study, albeit that Furthermore, in the Heart Protection Study, simvastatin appeared
trial was not performed exclusively in diabetic subjects.556 Thus to retard the decline in renal function, although this analysis was
from a clinical perspective no clear difference between these two not confined to the diabetic subgroup.569 Another group has also
drug classes has been identified. The one exception is cough, which reported a potential renoprotective effect of a statin,570 although
occurs in 5% to 30% of patients taking ACE inhibitors, depend- this effect has not been observed in all studies. Indeed, several
ing on ethnicity, which is higher in Asian subjects. In microal- meta-analyses have generally reported at best modest benefits of
buminuric T2DM subjects, ARBs have also been demonstrated statins on renal disease.571–573 Nevertheless, because cardiovascu-
to have a role. For example, in the Irbesartan Microalbuminuria lar disease is so prominent in diabetic patients, particularly those
Type 2 (IRMA2) trial, irbesartan dose-dependently reduced the with incipient or overt renal disease, lipid-lowering treatment
risk of development of macroproteinuria,557 confirming the find- should be considered in most patients independent of its putative
ings seen predominantly with ACE inhibitors in microalbumin- renoprotective actions.574
uric T1DM subjects.550 Other approaches to consider include correction of anemia
Another approach to inhibit the renin-angiotensin system has with agents such as erythropoietin.575 The role of these agents as
involved the use of the recently introduced renin inhibitors such renoprotective drugs remains to be clarified,576 but the potential
as aliskiren. For example, in the Safety and Efficacy of Aliskiren benefits on general patient well-being and in reducing left ven-
When Added to Standardized Losartan and Optimal Antihyper- tricular hypertrophy577 provide a rationale for using such agents
tensive Therapy in Patients With Hypertension, Type 2 Diabe- judiciously in diabetic patients. However, the Trial to Reduce
tes and Proteinuria (AVOID) study in T2DM subjects,558 this Cardiovascular Events With Aranesp Therapy (TREAT) study,
agent appeared to have an additional effect on albuminuria when although focusing on cardiovascular events and mortality using
administered with the angiotensin II antagonist losartan. Unfortu- the erythropoietin analogue darbepoetin, did not demonstrate
nately, as reported in the larger subsequent Aliskiren Trial in Type that darbepoetin is renoprotective. Furthermore, the drug was
2 Diabetes Using Cardiorenal Endpoints (ALTITUDE) study,511 unfortunately associated with a twofold increase in cerebrovascu-
aliskiren has failed to demonstrate superior cardiovascular or renal lar events.578 With the advent of new hemoglobin-raising drugs
protection, and thus this approach of add-on renin inhibition is such as the hypoxia-inducible factor prolyl hydroxylase inhibi-
not recommended. Other approaches focusing on blood pressure tors,579 there is likely to be a new approach for correcting anemia
reduction continue to be examined in these populations. Pilot in diabetes-associated chronic kidney disease.
studies with various mineralocorticoid (MC) antagonists, albeit Over the past decade, several clinical trials targeting diabetic
not studied extensively, have revealed that spironolactone and nephropathy with novel agents have yielded disappointing results.
eplerenone reduce albuminuria.559,560 Unfortunately, these agents For example, PKCβ inhibition with ruboxistaurin, which had
are associated with hyperkalemia, particularly in subjects with renal renal benefits in experimental diabetes, failed to show any major
impairment and/or type 4 renal tubular acidosis. However, the benefits on albuminuria in T2DM subjects.580 Another promising
advent of a newer MC antagonist, finerenone, which has a lower agent, sulodexide, postulated to restore the glomerular charge by
risk of hyperkalemia, has led to renewed interest in this treatment repleting the loss of glycosaminoglycans581 and thereby act as an
approach. Indeed, in the mineralocorticoid Receptor Antagonist antiproteinuric and ultimately renoprotective drug, also failed to
Tolerability Study—Diabetic Nephropathy (ARTS-DN) trial, a demonstrate any evidence of renoprotection in several large trials.
dose-dependent effect of finerenone was associated with a progres- An endothelin antagonist, avosentan, was assessed in the Assess
sive decrease in albuminuria.561 These positive findings have led to the Effect of the Endothelin Receptor Antagonist Avosentan on
larger trials with this drug that are currently in progress. Time to Doubling of Serum Creatinine, End Stage Renal Disease
or Death in Patients With Type 2 Diabetes Mellitus and Diabetic
Evaluation of Additional Approaches to the Nephropathy (ASCEND) trial.582 Although this drug was asso-
ciated with impressive reductions in albuminuria, the associated
Management of Diabetic Kidney Disease side effect of fluid retention reduced the enthusiasm for this agent.
Low-protein diets (0.75 g/kg/day) have been shown to retard the However, another endothelin antagonist, atrasentan, with fewer
progression of renal disease, although the data are not totally con- side effects and being antiproteinuric,583 was developed. However,
vincing for diabetic nephropathy, per se. A meta-analysis of five this endothelin antagonist also unfortunately failed at a relatively
studies in T1DM subjects supported a minor renoprotective role advanced stage of clinical development. Finally, bardoxolone, an
for these diets,562 but this has not been a universal finding.563 There agonist of the transcription factor Nrf2, which appears to act as
are even fewer data in T2DM subjects with overt nephropathy.564 an antioxidant, has been investigated in T2DM subjects with
However, the expected benefits that can be achieved through pro- impaired renal function. In the initial Bardoxolone Methyl Treat-
tein restriction in patients with diabetic nephropathy are at best ment: Renal Function in CKD/Type 2 Diabetes (BEAM) trial584
modest in comparison with adequate blood pressure control and an improvement in renal function was reported. However, a subse-
blockade of the renin-angiotensin system. Moreover, the nutri- quent larger study in T2DM subjects with stage 4 chronic kidney
tional impact of such interventions must be carefully considered, disease known as the Bardoxolone Methyl Evaluation in Patients
particularly in patients with brittle glycemic control. with Chronic Kidney Disease and Type 2 Diabetes Mellitus: the

Occurrence of Renal Events (BEACON) study was prematurely of access catheters and lines. Delay in referral can result in a more
terminated because of increased cardiovascular events.585 Never- precipitous start to renal replacement and usually a bad prognostic
theless, this drug continues to be evaluated,586 including in new outcome.588
studies, albeit with a lower dose, such as in the Phase II Study Many options are now available for the diabetic patient requir-
of Bardoxolone Methyl in Patients With Chronic Kidney Disease ing renal replacement therapy.589 These include home or facility
and Type 2 Diabetes (TSUBAKI) study where positive findings hemodialysis, including overnight dialysis, peritoneal dialysis,
have been reported. renal transplantation (cadaveric or living related), or combined
pancreas-kidney transplantation. Most patients choose hemodi-
alysis rather than peritoneal dialysis, although data are conflicting
Treatment of the Diabetic Uremic Patient regarding which approach leads to better survival (Table 37.6).
Renal impairment in a patient with diabetes necessitates changes Some patients opt for withdrawal of treatment because their qual-
in therapy. Often, blood glucose control becomes more brittle ity of life, with advanced cardiovascular disease, visual impair-
because the half-life of insulin is prolonged and the renal response ment, and amputations, is poor.
to hypoglycemia is impaired. High swinging blood glucose levels
in a patient with nephropathy can often mistakenly lead to an The Burden of Nephropathy
increase in oral therapy. However, in patients with renal impair-
ment, particular care must be exercised in the selection and dosing One must never consider renal disease in a diabetic patient in
of oral hypoglycemic therapy. Fortunately, DPP4 inhibitors can isolation. Proteinuria, per se, is strongly associated with other
be used safely in such individuals, albeit often the dose needs to be complications such as macrovascular disease, heart failure, and ret-
reduced in subjects with low GFR. Importantly, one DPP4 inhibi- inopathy. Furthermore, treatments directed toward one complica-
tor, linagliptin, which is not excreted by the kidney, can be given tion may be useful for the other complications. Indeed, intensified
without dose reduction, even in subjects on dialysis.587 Nonste- glycemic control has been shown to be particularly useful for other
roidal anti-inflammatory drugs (NSAIDs) and COX2 inhibitors microvascular complications.9 In addition, the various antihyper-
should be avoided when possible because their use is associ- tensive regimens, particularly those using agents that interrupt the
ated with inadequate blood pressure control, often as a result of renin-angiotensin system as well as glucose and blood pressure
reduced efficacy of antihypertensive drug therapy. Patients at high lowering with SGLT2 inhibitors, also confer important cardio-
risk for progressive deterioration in their renal function should vascular benefits such as reducing heart failure.469,506,590 Recent
be considered for an early referral to a nephrology service for data also support use of angiotensin receptor neprilysin inhibi-
management of renal failure (Fig. 37.41). This facilitates access to tors (ARNi). A post hoc analysis of the Prospective Comparison
erythropoietin, control of calcium phosphate balance, and plan- of ARNI With ACEI to Determine Impact on Global Mortal-
ning for renal replacement therapy with the preemptive placement ity and Morbidity in Heart Failure (PARADIGM-HF) trial using

Azotemia: Serum creatinine >2.0 mg/dL

Immediate Treatment objectives

Asses comorbid conditions Monitoring

• Persistent angina • Urinary protein

•Fig. 37.41 Flow chart illustrating the management of diabetic nephropathy after onset of renal failure.
ACE, Angiotensin-converting enzyme; BP, blood pressure; HIV, human immunodeficiency virus; LDL, low-
density lipoprotein; PD, peritoneal dialysis.

TABLE 37.6 O
ptions in Therapy for End-Stage Renal Disease in Diabetic Patients
Variable Peritoneal Dialysis Hemodialysis Kidney Transplantation
Extensive extrarenal disease No limitation No limitation except for hypotension Excluded in cardiovascular insufficiency
Geriatric patients Limited by frailty, cognitive No limitation Arbitrary exclusion as determined by
dysfunction; patient treatment program
satisfaction greater with assisted
peritoneal dialysis
Complete rehabilitation Rare, if ever Very few patients Common as long as graft functions
Death rate Much higher than for nondiabetic Much higher than for nondiabetic About the same as for nondiabetic
patients patients patients
First-year survival rate ∼75% ∼75% >90%
Morbidity during first year ∼15 days in hospital ∼12 days in hospital Weeks to months hospitalized
Survival to second decade Almost never <5% ∼1 in 5
Progression of complications Usual and unremitting; hyperglyce- Usual and unremitting; might benefit Interdicted by functioning pancreas plus
mia and hyperlipidemia from metabolic control kidney; partially ameliorated by correc-
tion of azotemia
Special advantage Can be self-performed; avoids Can be self-performed; efficient Cures uremia; freedom to travel
swings in solute and level of extraction of solute and water
intravascular volume in hours
Disadvantages Peritonitis; hyperinsulinemia; Blood access a hazard for clotting, Cosmetic disfigurement, hypertension,
hyperglycemia, hyperlipidemia; hemorrhage, and infection; cyclic personal expense for cytotoxic drugs;
long hours of treatment; more hypotension, weakness, aluminum induced malignancy; HIV (human
days hospitalized than with either toxicity, amyloidosis immunodeficiency virus) transmission
hemodialysis or transplantation
Patient acceptance Variable, usual compliance with Variable; often noncompliant with Enthusiastic during periods of good
passive tolerance for regimen dietary, metabolic, or antihyper- renal allograft function; exalted when
tensive components of regimen pancreas proffers euglycemia
Relative cost Most expensive over long run Less expensive than kidney trans- Pancreas plus kidney engraftment most
plantation in the first year; subse- expensive uremia therapy for diabetics;
quent years more expensive after first year, kidney transplantation
alone is lowest cost option

sacubitril/valsartan, a combination of neprilysin inhibition and neuropathies, including cardiac autonomic neuropathy (CAN),
angiotensin II receptor blockade, was reported to slow the rate of and autonomic neuropathy of the gastrointenstinal, urogenital,
reduction of eGFR in patients with stage 2 to 4 heart failure with and sudomotor systems. Focal neuropathies are much less common
reduced ejection fraction, a finding which was more prominent in than diffuse neuropathies and include isolated mononeuropathies
the participants with T2DM at study entry.591 Interestingly this of one or more peripheral nerves; even rarer are focal neuropathies
agent is also associated with modest glucose-lowering effects.592 of one or more nerve roots, classified as radiculopathy or polyra-
Thus as clearly expounded in the Intensified Multifactorial diculopathy, respectively (Tables 37.7 and 37.8).
Intervention in Patients With Type 2 Diabetes and Microalbu-
minuria (Steno-2) study, the multifactorial approach in microal- Epidemiology and Impact of Diabetic
buminuric subjects will not only lead to renal benefits as confirmed
in a follow-up study,593 but will also confer other advantages to Neuropathies
the diabetic patient, including reducing all-cause mortality.594 Diabetic neuropathies are the most common complications of
Because those with renal disease have the greatest risk for nonrenal T1DM and T2DM. Distal symmetric polyneuropathy repre-
complications, it stands to reason that they also are likely to have sents the most commonly diagnosed and highly morbid of the
the greatest absolute benefit from risk-reduction strategies. diabetic neuropathies. While prevalence estimates vary depending
on how the diagnosis is made, DSPN occurs in at least half if
Diabetic Neuropathies not more of all diabetic patients over time.595 Prevalence as deter-
mined by clinical examination was approximately 45% after 25
Diabetic neuropathies are a group of clinical syndromes with dis- years of diabetes in the classic study by Pirat following a cohort
tinct presentations secondary to different underlying pathogenetic of 4400 diabetic patients between 1947 and 1973.596 However,
mechanisms. These syndromes are most commonly classified into when more contemporary quantitative sensory testing or nerve
two broad groups: diffuse versus focal neuropathies. Among the conduction studies are used as part of the diagnostic criteria,
diffuse diabetic neuropathies, distal symmetric polyneuropathy the prevalence of DSPN increases from 60%597 to greater than
(DSPN) is the most common, followed by the family of autonomic 75%.598 Multiple studies now confirm that DPSN also occurs in

TABLE 37.7 C
lassification for Diabetic Neuropathies
Diabetic Neuropathies Sudomotor dysfunction
A. Diffuse neuropathy • Distal hypohydrosis/anhidrosis
DSPN • Gustatory sweating
• Primarily small fiber neuropathy Hyoglycemia unawareness
• Primarily large fiber neuropathy
Abnormal pupillary function
• Mixed small and large fiber neuropathy (most common)
B. Mononeuropathy (mononeuritis multiplex) (atypical forms)
Autonomic
Isolated cranial or peripheral nerve (e.g., CN III, ulnar, median,
Cardiovascular
femoral, peroneal)
• Reduced HRV
Mononeuritis multiplex (if confluent may resemble
• Resting tachycardia polyneuropathy)
• Orthostatic hypotension
• Sudden death (malignant arrhythmia) C. Radiculopathy or polyradiculopathy (atypical forms)
Radiculoplexus neuropathy (a.k.a. lumbosacral polyradiculopathy,
Gastrointestinal proximal motor amyotrophy)
• Diabetic gastroparesis (gastropathy) Thoracic radiculopathy
• Diabetic enteropathy (diarrhea) Nondiabetic neuropathies common in diabetes
• Colonic hypomotility (constipation) Pressure palsies
Urogenital Chronic inflammatory demyelinating polyneuropathy
• Diabetic cystopathy (neurogenic bladder) Radiculoplexus neuropathy
• Erectile dysfunction Acute painful small-fiber neuropathies (treatment induced)
• Female sexual dysfunction

DSPN, Distal symmetric polyneuropathy.

From Pop-Busui R, Boulton AJM, Feldman EL, et al. Diabetic neuropathies: a position statement by the American Diabetes Association. Diabetes Care. 2017;40:136–154. Copyright and all rights
reserved. Material from this publication has been used with the permission of the American Diabetes Association.

TABLE 37.8 M
ononeuritis and Entrapment Syndromes patients experience some form of disrupted sleep, depression,
anxiety, and poor work productivity.603 Impairment of physical
Feature Mononeuritis Entrapment functioning in DSPN is associated with a 15-fold increase in the
Onset Sudden Gradual
likelihood of falling and fractures, particularly in older diabetics.604
Foot ulceration leading to amputation is one of the most serious
Nerves Usually single but may Single nerves exposed to and adverse effects of DSPN. Foot ulcers are a strong predictor of
be multiple trauma early mortality in diabetic patients605 and precede nearly all ampu-
Common C3, C6, C7, ulnar, Median, ulnar, peroneal, tations.606 Of patients with new foot ulcers, 5% die within 1 year
nerves median, peroneal medial and lateral plantar of their first ulcer and 42.2% of people with existing ulcers die
within 5 years, as shown in a recent study of 414,523 patients with
Progression Not progressive; resol Progressive
diabetes.607 A 2017 systematic review reported that the global foot
ves spontaneously
ulcer prevalence is 6.3%, with the highest prevalence, 13%, occur-
Treatment Symptomatic Rest, splints, diuretics, ring in North America.608 Worldwide DSPN accounts for more
steroid injections, surgery hospitalizations than all the other diabetic complications com-
for paralysis bined and is the root cause of 50% to 75% of nontraumatic ampu-
C3, C6, C7, Cervical spinal nerves 3, 6, and 7.
tations in North America.609,610 Though rates of lower extremity
Modified from Vinik AI, Mehrabyan A. Diabetic neuropathies. Med Clin North Am. amputation among Medicare-enrolled diabetics have decreased by
2004;88:947–999. 28.8% between 2000 and 2010,611 the cost of caring for DSPN
complications has not decreased because of delay in diagnosis.612
Like DSPN, autonomic neuropathies adversely affect function
in patients with T1DM and T2DM, affecting a wide variety of
patients with prediabetes and the metabolic syndrome599–601 and organ systems, including cardiovascular, gastrointestinal, urogeni-
while higher in those metabolic syndrome patients with hypergly- tal, and sudomotor/thermoregulatory systems. Of these, cardiac
cemia, ranging from prediabetic to T2DM levels, DSPN can also autonomic neuropathy is the best studied. Multiple studies con-
occur independent of glycemic status.601 With the alarming rise of firm early mortality and sudden death in diabetic CAN patients,
prediabetes, T2DM, and the metabolic syndrome, estimates range with mortality estimates up to 56% over a 10-year period.613 One-
from 15 million to 30 million individuals in the United States are third of patients with autonomic neuropathy have impaired qual-
living with DSPN.602 ity of life,614 with erectile dysfunction predicting not only poor
DSPN leads to a poor quality of life, limiting a patient’s func- quality of life in the affected patient615 but also in the patient’s
tion, especially DSPN associated with pain. Up to 70% of DSPN sexual partner.616

Dorsal root Myelin

ganglion channels

Motor Metabolites,
neurons trophic factors,
signaling molecules

Remak
C fibers bundle
(unmyelinated)
Schwann
cell-axon
transport

Axon
Schwann cells/
myelin

• Fig. 37.42 The peripheral nervous system. The peripheral nervous system (PNS) is comprised of both
neurons and Schwann cells (SCs), and the structure, location, and interaction of these components have
important implications for PNS function. Efferent axons of motor neurons, whose cell bodies are located
in the ventral horn of the spinal cord, carry signals from the central nervous system (CNS) to muscles
and glands, whereas afferent axons of sensory neurons, whose cell bodies are located in the dorsal root
ganglia, relay information from peripheral sensory receptors to the CNS. Thin and unmyelinated sensory
axons, also known as C fibers or small fibers, are associated with nonmyelinating Schwann cells and are
grouped as Remak bundles. These represent a large portion of the PNS neurons. Myelinated sensory
axons, on the other hand, are surrounded by myelin sheaths made by Schwann cells that form distinct
nodal domains important for saltatory conduction. They also form a tubular network of myelinic channels
that connect the SC cytoplasm with the periaxonal space, providing substrate for energy generation to the
axonal compartment. (Redrawn from Feldman EL, Nave KA, Jensen TS, et al. New horizons in diabetic
neuropathy: mechanisms, bioenergetics, and pain. Neuron. 2017;93:1296–1313.)

While cost estimates vary, the total annual cost of DSPN is burning pain, although patients also report tingling and prickly
estimated to be greater than $25 billion,617 with total annual med- sensations.621 With time, large fiber involvement occurs, with cor-
ical costs for a diabetic patient with painless DSPN estimated at responding numbness and poor position sense.
$12,492 compared to $6632 for a diabetic patient without DSPN. How diabetes changes the interaction between Schwann cells
Costs are even greater with painful DSPN, from $27,931 up to and axons is not yet understood, and it remains unclear if the
$30,755 per patient, depending on the severity of the pain617 and primary insult in DSPN lies in the sensory neuron, axon, or
the use of generic versus nongeneric medications.618 Importantly, Schwann cell; logically, it is likely that all three components are
little money is spent in the United States on patient education or adversely affected, in both distinct and overlapping ways. In the
preventative foot care, an investment that has been proven to be diabetic microenvironment, glycolytic intermediates are depleted
effective in decreasing the incidence of DSPN complications.619 in Schwann cells622 while fatty acid oxidation is increased,623 pro-
ducing a state of cellular oxidative stress and excess cytosolic bio-
active lipids.624 Emerging evidence reveals that axon-Schwann cell
Pathophysiology of Diabetic Neuropathies communication, via myelinic channels between the two tissues,
While the pathophysiology of diabetic neuropathy has been dis- allows for the two-way transport of these glycolytic intermedi-
cussed earlier in this chapter, an understanding of nerve structure ates and bioactive lipids.602,625 The concept of crosstalk between
(Fig. 37.42) provides additional insight into neuropathy patho- Schwann cells and axons is further supported by a recent finding
genesis and clinical presentation. In the peripheral nervous system, that delivery of exosomes from Schwann cells cultured in high
afferent axons from sensory receptors send information to the dor- glucose to sciatic nerves of mice resulted in murine DSPN.626
sal root ganglia of the central nervous system (CNS), while effer- Finally, the idea of axon-Schwann cell crosstalk and recipro-
ent axons relay CNS signals to muscles and joints. Schwann cells cal supply of critical intermediary metabolites is supported by
envelope and myelinate large and midsize sensory axons, knowns Schwann cell–specific knockout experiments of a serine-threonine
as large fibers. In contrast, the cytoplasm from one Schwann cell kinase required for Schwann cell energy regulation. This knockout
can envelope 30 or more unmyelinated small axons, known as caused degeneration of sensory greater than motor axons, espe-
small fibers. These large and small fibers carry specific informa- cially in the small unmyelinated fibers.627
tion: Large fibers relay vibratory, proprioception, and tactile sensa- Clinical trials have yielded insights into potential differences
tion; small fibers relay thermal sensation and pain, and they also underlying the pathogenesis of DSPN in T1DM and T2DM.
regulate microvascular blood flow. Small fiber damage usually pre- Glucose control has a significant effect on DSPN onset and pro-
cedes large fiber damage in diabetes620 and frequently presents as gression in T1DM but no effect or a modest 3% to 5% effect at

Hyperglycaemia Type 1 diabetes

Dyslipidaemia
Type 2 diabetes
Glucose Triglycerides Both
LDL
Protein AGE-LDL Oxidised HDL
AGEs LDL FFA

Insulin
RAGE LOX1 TLR4
Insulin + C-peptide
Hexosamine
Glucose Inflammatory signals FFAs
pathway
Gly NADPH oxidase * PI3K
pathway

col Insulin
Polyol

ysi signalling
s

Electron Akt
Osmotic transport Insulin
Cholesterol
stress * overload * resistance

ROS (*) Oxysterols

Mitochondrial Loss of
complex neurotrophic
DNA damage ER stress dysfunction Apoptosis signals

A Mechanisms of cell damage

Macrophage
activation

Neurons

Glial cells

Vascular endothelial cells

B Cell damage nerve dysfunction

• Fig. 37.43 Mechanisms of diabetic neuropathy. Factors linked to type 1 diabetes (red), type 2 diabetes
(blue), and both (green) cause DNA damage, endoplasmic reticulum stress, mitochondrial complex dys-
function, apoptosis, and loss of neurotrophic signaling (A). These mechanisms can damage neurons, glial
cells, and vascular endothelial cells and can trigger macrophage activation. Damage to each of these cell
types can lead to nerve dysfunction and neuropathy (B). The relative importance of the pathways in this
network will vary with cell type, disease profile, and time. AGE, Advanced glycation end products; ER,
endoplasmic reticulum; FFA, free fatty acids; HDL, high-density lipoprotein; LDL, low-density lipoprotein;
LOX1, oxidized LDL receptor 1; PI3K, phosphatidylinositol-3-kinase; RAGE, receptor for advanced gly-
cation end products; ROS, reactive oxygen species (red star); TLR4, toll-like receptor 4. (Redrawn from
Callaghan BC, Cheng HT, Stables CL, et al. Diabetic neuropathy: clinical manifestations and current treat-
ments. Lancet Neurol. 2012;11:521–534.)

best on DSPN in T2DM.628 The idea that DSPN is actually two nerve dysfunction in people with diabetes after the exclusion of
distinct disorders, DSPN T1DM and DSPN T2DM, is a source other causes.”631,632 Alternate causes of DSPN include vitamin
of active investigation,11 with the emerging concept that neurons B12 deficiency, multiple myeloma, and hereditary neuropathies.
can develop insulin resistance629 that contributes to DSPN in The onset of DSPN is usually insidious but can be acute, as dis-
T2DM (Fig. 37.43).630 cussed later. DSPN is primarily a sensory neuropathy and can
involve small fibers, large fibers, or both.633 Motor involvement is
Clinical Features of Diabetic Distal Symmetrical only seen late in the disease, if at all.
Polyneuropathy Symptoms of DSPN secondary to diabetes depend on the
affected fiber types. Early in the course of the disease, small fibers
Clinical Symptoms are preferentially involved, leading to symptoms of pain, hyper-
DSPN is the most common and widely recognized form of dia- algesia and allodynia in the lower limbs, followed over time by a
betic neuropathy. An international consensus meeting defined loss of thermal sensitivity and pain perception.621 These same early
DSPN as “the presence of symptoms and/or signs of peripheral symptoms of DSPN are also the main constellation of presenting

symptoms in prediabetic neuropathy.600 Symptoms often are exac- The Semmes-Weinstein 10-g filament as an assessment of touch
erbated at night and are manifested in the feet more than the hands. pressure, a large fiber function, can be used as a screening tool for
Spontaneous episodes of pain can be severely disabling. The pain DSPN.634 The filament is bent to a C-shape applying pressure to
varies in intensity and character and has been variably described by the plantar surface of the great toe. Because of the variability in
patients as lancinating, stabbing, or sharp. Paresthesias or episodes the sensitivity and specificity of the 10-g filament,635 it is recom-
of distorted sensation, such as pins and needles, tingling, coldness, mended that a three-site test be done involving the plantar aspects
numbness, or burning, often accompany the pain.633 As a whole, of the great toe, the third metatarsal, and the fifth metatarsal. The
these multiple symptoms are classified as positive symptoms, as the ability to sense the 10-gram filament on the soles of the feet is
affected diabetic patient actively experiences discomfort. also frequently used by podiatrists as a screening tool for DSPN,
In contrast, the symptoms of large fiber neuropathy are most although it is generally held that this is not sufficiently sensitive to
commonly denoted as negative symptoms, with the patient pre- diagnose DSPN.634
senting with a numb, insensate foot. Diabetic patients with large In advanced cases of DSPN, examination will reveal wasting of
fiber predominant DSPN are frequently unaware of their neurop- the small muscles of the feet, with hammertoes and weakness of
athy, and are surprised when a physical examination reveals absent toe extension; in extreme cases, there is weakness of the anterior
sensation in their feet. As large fibers mediate position sense, poor tibialis and the intrinsic hand muscles.
balance with frank falling is a common presenting symptom in
diabetic patients with a large fiber predominant DSPN. While less Differential Diagnosis of Diabetic Neuropathy
common, these patients can also experience a deep-seated, dull, The American Academy of Neurology suggests checking a serum
toothache-like sensation in their feet and, in very severe cases, will protein electrophoresis with immunofixation, a B12 level, and thy-
experience symptoms of distal weakness of the toes and ankles, roid function tests in all patients suspected of having DSPN. If
secondary to late-onset large motor fiber involvement. these are normal, and there are no atypical signs, the most likely
While the distinction is classically made between small and diagnosis is prediabetic neuropathy or DSPN.631 The importance
large fiber predominant DSPN, a common presentation of DSPN of performing these screening tests was confirmed in the Roches-
in diabetic patients is a combination of symptoms from both fiber ter Diabetic Neuropathy Study, in which up to 10% of peripheral
types. A few clinical “pearls” can assist the clinician in identify- neuropathy in diabetic patients was determined to not be second-
ing symptoms that are atypical for DSPN secondary to diabetes. ary to diabetes.636
These include motor greater than sensory symptoms and a report- Importantly, the symptoms and signs of DSPN are sensory and
able asymmetry. If the physical examination confirms these symp- symmetrical, and any convincing asymmetry evidence of motor
toms, an alternative diagnosis to DSPN must be pursued. greater than sensory involvement, or a rapid and aggressive clinical
course, should lead the clinician to suspect an alternative etiology
Clinical Signs of Diabetic Neuropathy for the patient’s clinical presentation. In these atypical cases, it is
A simple examination can distinguish between small and large essential to exclude other causes of neuropathy in patients with
fiber impairment. Inspection of the skin is an essential first part of diabetes (Table 37.9).637
the clinical examination because small fibers also subserve sudo-
motor function. The foot is inspected for dry skin, fissures, and Classification of Diabetic DSPN
callouses. Small fiber function is then assessed using a simple dis- The Toronto Consensus panel convened in 2009 and redefined the
posable safety pin, a cotton wisp, or by assessing the ability of a minimal criteria for diagnosis of typical diabetic neuropathy638:
patient to distinguish temperatures between a proximal location 1. Possible DSPN. The presence of symptoms or signs of DSPN
(the face) and a distal location (the dorsum of the great toe). A may include the following: symptoms—decreased sensation,
simple bedside approach is to determine if the patient can dif- positive neuropathic sensory symptoms (e.g., “asleep numb-
ferentiate between the sharp and dull end of the safety pin on ness,” prickling or stabbing, burning or aching pain) predomi-
the dorsum of the toe, and if he or she can perceive a cotton nantly in the toes, feet, or legs; or signs—symmetric decrease
wisp in the same anatomic location. To assess temperature in the of distal sensation or unequivocally decreased or absent ankle
most simple manner, a cool tuning fork is placed on the patient’s reflexes.
cheek, then the same instrument is placed on the great toe, and 2. Probable DSPN. The presence of a combination of symptoms
the patient is asked to report on perceiving the same cool tem- and signs of neuropathy, including any two or more of the fol-
perature in both anatomic locations or not. Large fiber function is lowing: neuropathic symptoms, decreased distal sensation, or
initially assessed by measuring the time in seconds that a patient unequivocally decreased or absent ankle reflexes.
can perceive vibration sensation on the great toe after a health 3. Confirmed DSPN: The presence of an abnormality of nerve
care provider places a 128-Hz vibrating tuning fork on the dor- conduction and a symptom(s) or sign(s) of neuropathy con-
sum of the patient’s great toe. Proprioception is assessed by minor firm DSPN. If nerve conduction is normal, a validated mea-
movements of the first metatarsophalangeal joint of the great toe. sure of small fiber neuropathy (with Class 1 evidence) may be
Ankle reflexes are an essential part of assessing large fiber function used. To assess for the severity of DSPN, several graded clinical
and are best produced with the examiner mildly dorsiflexing the scores can be used that include various continuous measures of
patient’s foot and striking the Achilles tendon with a Tromner or sum scores of neurologic signs, symptoms, or composite clini-
Queens Square reflex hammer. cal and electrophysiologic scores; scores of function of activities
Given that the glove and stocking distribution of the DSPN of daily living; or scores of predetermined tasks or of disability.
reflects nerve fiber length, when sensory loss reaches the mid- 4. Subclinical DSPN. The presence of no signs or symptoms of
calves, patients begin to perceive symmetric sensory loss of the neuropathy are confirmed with abnormal nerve conduction or
fingertips. Examination of the upper extremity is most commonly a validated measure of small fiber neuropathy (with Class 1 evi-
performed as outlined earlier for the lower extremity, except the dence). Definitions 1, 2, or 3 can be used for clinical practice,
dorsum of the index finger is used by the examiner. and definitions 3 or 4 can be used for research studies.

TABLE 37.9 C
ommon Causes of Rare Subtypes of Peripheral Neuropathy
Localization Condition
Diffuse, nonlength-dependent AIDP; CIDP; CIDP variants: POEMS syndrome,
neuropathy; demyelinating IgM anti-MAG neuropathy, Waldenstrom acroglobulinemia, and MGUS; diphtheria; and toxic exposures (hexane, arsenic, and
sensory motor amiodarone)
Demyelinating sensory Sensory CIDP or AIDP; and DADS (IgM anti-MAG neuropathy)
Demyelinating motor MMN
Axonal sensory motor Toxic exposures; ASMAN; and AIP
Axonal sensory Paraneoplastic (Hu, CRMP5, and amphyphisin); Sjögren syndrome; chemotherapy (platinum based, bortezomib); vitamin
B6 toxicity; idiopathic; HIV, HTLV; autoimmune hepatitis; celiac disease; HSAN; Friedrich ataxia; CANVAS; SANDO; and
CANOMAD
Axonal motor ALS; PMA; postpolio syndrome; HIV, HTLV, WNV, enterovirus D68; MMN without conduction block; radiation; monomelic
amyotrophy; HMN; SMA (including Kennedy syndrome); and complicated HSP
Multiple mononeuropathies Systemic vasculitic neuropathy: microscopic polyangiitis, Wegener granulomatosis, polyarteritis nodosa, Churg-Strauss syn-
drome, cryoglobulinemia, Sjögren syndrome, rheumatoid arthritis, and SLE; nonsystemic vasculitic neuropathy; neoplasm
(malignant and benign); HNPP; sarcoidosis; amyloidosis; MMN; and MADSAM
Polyradiculopathy Compressive: disc herniation/spondylosis, osteomyelitis, and neoplasm; noncompressive: infection (CMV, VZV, Lyme, and
tuberculosis), inflammatory (sarcoidosis), neoplastic (leukemia and lymphoma), and radiation
Plexopathy Compressive: neoplasm and hemorrhage;
noncompressive: infection (VZV, HSV, CMV, and Lyme), inflammatory (sarcoidosis), neoplastic (leukemia, lymphoma), and
radiation
Radiculoplexus neuropathy Diabetic lumbar (diabetic amyotrophy); diabetic cervical; postsurgical inflammatory; nondiabetic lumbar or cervical; infection
(VZV, HSV, CMV, and Lyme); inflammatory (sarcoidosis); neoplastic (leukemia and lymphoma); and radiation

AIDP, Acute inflammatory demyelinating polyneuropathy; AIP, acute intermittent porphyria; ALS, amyotrophic lateral sclerosis; ASMAN, acute sensory motor axonal neuropathy; CANOMAD, chronic ataxic
neuropathy, ophthalmoplegia, monoclonal IgM protein, cold agglutinins, disialosyl antibodies; CANVAS, cerebellar ataxia neuropathy vestibular areflexia syndrome; CIDP, chronic inflammatory demyelinat-
ing polyneuropathy; CMV, cytomegalovirus; CRMP5, collapsing response mediator protein 5; DADS, distal-acquired demyelinating syndrome; HIV, human immunodeficiency virus; HMN, hereditary motor
neuropathy; HNPP, hereditary neuropathy with liability to pressure palsies; HSAN, hereditary sensory autonomic neuropathy; HSP, hereditary spastic paraplegia; HSV, herpes simplex virus; HTLV, human
T-lymphotrophic virus; MADSAM, multifocal acquired demyelinating sensory and motor neuropathy; MGUS, monoclonal gammopathy of unclear significance; MMN, multifocal motor neuropathy; PMA,
progressive muscular atrophy; POEMS, polyneuropathy, organomegaly, endocrinopathy, monoclonal gammopathy, and skin changes; SANDO, sensory ataxia neuropathy dysarthria ophthalmoplegia; SLE,
systemic lupus erythematosus; SMA, spinal muscular atrophy; VZV, varicella zoster virus; WNV, West Nile virus.
Reproduced with permission from Callaghan BC, Price RS, Chen KS, et al. The importance of rare subtypes in diagnosis and treatment of peripheral neuropathy: A review. JAMA Neurol. 2015; 72:1510–
1518. Copyright ©2015 American Medical Association. All rights reserved.

5. Small fiber neuropathy (SFN). SFN should be graded as fol- sensitivities and specificities are listed in Table 37.10. Electrophys-
lows: (1) possible: the presence of length-dependent symptoms iologic testing or nerve conduction studies along with a referral to
and/or clinical signs of small fiber damage; (2) probable: the a neurologist is normally not needed, unless the patient has either
presence of length-dependent symptoms, clinical signs of small atypical symptoms or signs.
fiber damage, and normal sural nerve conduction; and (3) The importance of the skin biopsy as a diagnostic tool for
definite: the presence of length-dependent symptoms, clinical DSPN is increasingly recognized when no clear objective evidence
signs of small fiber damage, normal sural nerve conduction, of small or large fiber dysfunction is present on examination.621
and altered intraepidermal nerve fiber (IENF) density at the This technique quantitates small IENFs through antibody stain-
ankle and/or abnormal thermal thresholds at the foot (both of ing of the neuronal antigen protein gene product (PGP) 9.5. It is
the later provide Class 1 evidence). minimally invasive (3-mm-diameter punch biopsies) but enables
direct study of small fibers that cannot be evaluated by standard
Graded Scores for the Classification of DSPN nerve conduction studies. Prediabetic neuropathy and DSPN are
While DSPN is diagnosed at the bedside, as outlined earlier, there accompanied by loss of these small fibers in a distal-to-proximal
are several scores that provide structure to the bedside examina- gradient.
tion and range from simple assessments of large and small fibers
to more complex assessments requiring quantitative thermal test- Treatment of DSPN
ing, electrophysiology, or skin biopsies. These scores include the Treatment of DSPN is commonly divided into either direct dis-
Neuropathy Impairment Score of the lower limb (NIS-LL),639 ease-modifying therapies or treatment of pain, a common occur-
Michigan Diabetic Neuropathy Score (MDNS),640 Modified rence in patients with DSPN. Accumulating evidence supports
Toronto Clinical Neuropathy Score (mTCNS),641 Total Neuropa- the idea that disease-modifying therapies are distinct for DSPN,
thy Score–Clinical (TNS-C),642 the Utah Early Neuropathy Score depending if the patient has T1DM or T2DM.645 In contrast,
(UENS),643 and the Neuropathy Disability Score (NDS).644 The treatment of pain is not personalized to the type of diabetes,

TABLE 37.10 S
ensitivity and Specificity for Tested Neuropathy Scales
Cutoff Sensitivity Specificity PPV NPV LRP LRN
Modified Toronto Clinical Neuropathy Score (mTCNS) 3.00 98.00 97.00 0.99 0.94 31.20 0.03
Total Neuropathy Score (Clinical) (TNSc) 5.00 81.00 97.00 0.99 0.66 25.90 0.20
Utah Early Neuropathy Scale (UENS) 3.00 85.00 97.00 0.99 0.72 27.20 0.15
Early Neuropathy Score (ENS) 5.00 83.00 97.00 0.99 0.67 26.67 0.17
Michigan Diabetic Neuropathy Score (MDNS) 5.00 80.00 100.00 1.00 0.65 — 0.20
Neuropathy Impairment Score of the Lower Limb (NISLL) 3.00 83.00 97.00 0.98 0.69 26.47 0.18
Neuropathy Disability Score (NDS) 4.00 89.00 100.00 1.00 0.78 — 0.11

LRN, Likelihood ratio negative; LRP, likelihood ratio positive; NPV, negative predictive value; PPV, positive predictive value.
Modified from Zilliox LA, Ruby SK, Singh S, et al. Clinical neuropathy scales in neuropathy associated with impaired glucose tolerance. J Diabetes Complications. 2015;29:372–377.

although emerging research indicates there may be differences in TABLE 37.11 American Diabetes Association
the pain phenotypes among the different types of diabetes.602 Recommendations for Distal Symmetric
Polyneuropathy (DSPN)
Disease-Modifying Therapy in DSPN
There are no US Food and Drug Administration (FDA)–approved • T ight glucose control targeting near-normal glycemia in patients
with type 1 diabetes dramatically reduces the incidence of distal
therapies that directly address DSPN as a disease entity. In the symmetric polyneuropathy and is recommended for distal symmetric
absence of pharmaceuticals, dietary and lifestyle interventions polyneuropathy prevention in type 1 diabetes. A
are recommended to modify the course of the illness. A 2012 • In patients with type 2 diabetes with more advanced disease and
Cochrane review reported that glucose control has only modest multiple risk factors and comorbidities, intensive glucose control
effects, if any at all, on DSPN in T2DM, while it is an impor- alone is modestly effective in preventing distal symmetric polyneu-
tant factor in determining the onset and progression of DSPN ropathy, and patient-centered goals should be targeted. B
in T1DM.645 In both prediabetes and T2DM, the components • Lifestyle interventions are recommended for distal symmetric
of the metabolic syndrome are collectively drivers of neuropathy. polyneuropathy prevention in patients with prediabetes/metabolic
Multiple studies confirm that dyslipidemia, hypertension, obesity, syndrome and type 2 diabetes. B
and low hip:waist ratios are present in patients with prediabetic
ADA recommendations are assigned ratings of A, B, or C depending on the quality of e vidence.
neuropathy and T2DM patients with DSPN. Modification of
From Pop-Busui R, Boulton AJM, Feldman EL, et al. Diabetic neuropathies: a position state-
these factors, beyond glucose control alone, is required to effect
ment by the American Diabetes Association. Diabetes Care. 2017;40:136–154. Copyright
a clinically meaningful change in neuropathy in both patient and all rights reserved. Material from this publication has been used with the permission of
groups. Of interest, obese patients with the metabolic syndrome the American Diabetes Association.
but normoglycemia also have neuropathy, further confirming that
glucose control alone is insufficient to address neuropathy patho-
genesis in the prediabetic or T2DM patient.600
The 2017 American Diabetes Association (ADA) recommen-
dations are outlined in Table 37.11.646 Both the American Academy of Neurology (AAN) and the
European Federation of Neurological Societies (EFNS) have
Treatment of Painful Neuropathy reviewed the Class 1 and 2 studies for the treatment of painful
Pain is a common presentation of prediabetic neuropathy and DSPN (Table 37.12). In general, three classes of therapies appear
present in approximately 20% of patients with DSPN.647 There effective in the treatment of DSPN: the voltage-gated α2δ ligands
are no compelling studies that show glucose control is helpful in (pregabalin, gabapentin), the serotonin-norepinephrine reuptake
the treatment of pain. In contrast, aggressive glucose control over inhibitors (duloxetine, venlafaxine), and the secondary amine tri-
a short period of time can result in debilitating pain syndromes in cyclic antidepressants (amytriptyline, nortriptyline, desipramine).
both T1DM and T2DM patients.648 These pain syndromes are Fig. 37.45 provides a management algorithm for the patient with
known as treatment-induced neuropathies. It is now well estab- painful DSPN; Table 37.13 outlines the commonly used doses
lished that the degree of a patient’s impairment correlates with and range of side effects. A simple rule is to begin with one drug
the rate of change of the HbA1c (Fig. 37.44),649 cautioning the from one of the three classes, and maximize the dose of this drug
clinician to achieve glycemic control at a rate of less than 1% per to either side effects or known therapeutic dosage; if there is some
month HbA1c.649 clinical benefit, although not complete, then a second agent from
There are three FDA-approved drugs for the treatment of pain- a different drug class should be added for more symptom relief.
ful DSPN, with better evidence for duloxetine and pregabalin. This agent is again titrated slowly to either side effects or maxi-
The opioid tapentadol also is FDA approved, but with weaker mal known therapeutic dose. While figures vary, this approach
evidence. In light of the current opioid epidemic in both North generally provides relief for at least two-thirds, if not more, of
America and worldwide, opioid use is discouraged in the treat- patients618; there is also an emerging literature that coupling pain
ment of painful DSPN. treatment with an exercise regimen provides even greater symptom

10
relief.650 This approach to pain is recommended in the 2011 AAN
guidelines and the 2017 ADA position statement.

Pain scores
8
6 While not part of the ADA position statement, the AAN
4 guidelines discuss the addition of selected therapies, which are
2 briefly described later, for patients who do not achieve pain
20 0 control with metabolic control, combination therapy, and exer-
cise.651 Alpha-lipoic acid is an antioxidant that is used in Europe
15 at a dose of 600 mg per day as an adjunct to the treatment of
HbA1c

painful DSPN.652,653 Capsaicin is a natural substance that locally

10 depletes substance P, resulting in a loss of small fibers. Capsaicin
5
cream applied topically (0.075% four times/day) to the feet in
patients with painful DSPN was reported to provide a modest
–15 –10 –5 0 5 10 15 improvement in pain.654 Similarly local lidocaine applied to
A Months D painful areas has been reported to improve pain in selected small
studies,655 although a Cochrane systematic review found no role
10 for lidocaine in the treatment of painful neuropathy.656 Studies
on the use of transelectric stimulation (TENS) units are limited
Pain scores

8
6 for painful DSPN, but do support its use as an adjuvant therapy
4 in treatment-resistant patients.657–659
2
There is no role for opioids in the treatment of chronic pain-
20 0
ful DSPN, and limited use, if at all, in acute painful DSPN,
possibly for the pain experienced with treatment-induced neu-
15
HbA1c

ropathy. If pain management is not achieved without opioids,

10 then referral to a pain clinic for the patient is the next sug-
gested step.
5
–15 –10 –5 0 5 10 15 Other Diabetic Somatic Neuropathies: Clinical
B Months E Features and Treatment Paradigms
As presented in Table 37.7, there are other diabetic neuropa-
10 thies. Cranial mononeuropathies occur primarily in the older
Pain scores

8 population. Onset is usually acute and associated with pain,

6 and the course is self-limited, resolving within 6 to 8 weeks.
4
It is generally believed these neuropathies are secondary to
2
0
vascular insufficiency.660 The most common cranial mononeu-
20
ropathy secondary to diabetes is a mononeuropathy of the third
15
cranial nerve, presenting as a complete third nerve palsy with
HbA1c

pupillary sparing. Less common are mononeuropathies of cra-

10 nial nerves 4, 6, and 7. Because it is essential to distinguish
these mononeuropathies from a more serious central etiology,
5 including cranial aneurysms and stroke, neuroimaging is fre-
–15 –10 –5 0 5 10 15 quently required to confirm that the symptoms and signs are
C Months F due soley to diabetes.
Mononeuropathies must be distinguished from entrapment
• Fig. 37.44 Complications and risks associated with treatment-induced
syndromes, which start slowly, progress, and persist without
neuropathy of diabetes (TIND). (A–F) The 104 individuals with treatment-
induced neuropathy of diabetes are grouped by change in glycosylated
intervention (see Table 37.8). The most common entrapment
HbA1c scores. (A) The 27 individuals with a decrease in HbA1c of 2% to site in diabetic patients is the median nerve, although patients
3.9% are shown. (B) The 52 individuals with a decrease in HbA1c of 4% also experience ulnar, lateral cutaneous femoral, and peroneal
to 7% are shown. (C) The 25 individuals with a decrease in HbA1c of more nerve entrapment. Median nerve entrapment, commonly known
than 7% are shown. (A–C) The lower portion of the graph (left y-axis) reveals as the carpal tunnel syndrome, occurs three times as often in per-
the glycosylated haemoglobin (HbA1c) scores over time. The mean value is sons with diabetes than in a normal, healthy population,661,662
shown in blue and the standard deviation in light purple. The upper portion and its increased prevalence in diabetes may be related to dia-
of the graph (right y-axis) reveals neuropathic pain scores during the same betic cheiroarthropathy,663 repeated undetected trauma (due to
time frame. The mean value is shown in black and the standard deviation repetitious use of the hands), metabolic changes, or accumula-
in dark purple. The representative distribution of neuropathic is shown in
tion of fluid or edema within the confined space of the carpal
(D–F), with the area in dark purple representing pain common to all individu-
als, and the area in light purple common to many individuals. (D) The least
tunnel.664 Diagnosis can be confirmed by nerve conduction
widespread pain distribution in the individuals with the smallest change in studies or ultrasound.665 The mainstays of nonsurgical treat-
HbA1c (corresponding to A). (E) The pain distribution in the individuals with ment are resting of the wrist, aided by the placement of a wrist
moderate decreases in HbA1c (B). (F) The group with the largest decrease splint in a neutral position for day and night use, and the addi-
in HbA1c. (C) has widespread neuropathic pain. (Redrawn from Gibbons tion of anti-inflammatory medications. Surgical treatment con-
CH, Freeman R. Treatment-induced neuropathy of diabetes: an acute, iat- sists of sectioning the volar carpal ligament.666 The decision to
rogenic complication of diabetes (part 1). Brain. 2015;138:43–52.) proceed with surgery should be based on several considerations,

1496 SE C T I O N V I I I Disorders of Carbohydrate and Fat Metabolism

Societies Guidelines on the Treatment of Painful Diabetic Distal Symmetric Polyneuropathy
PATIENTS WITH >50% PAIN
REDUCTION
No. Receiving Mean Pain Reduction
Descargado para Paola Gavidia Davila ([email protected]) en Antenor Orrego Private University de ClinicalKey.es por Elsevier en julio 19, 2020.

Evidence Study Treatment/ Total on 0-10 Rating Scale Treatment Placebo

Sourcec Treatment per Day Classa Duration (wk) Sampleb vs. Placebo (95% Cl) Effect (%) Effect (%) Common Adverse Effects
Lesser et al., 2004 Pregabalin, 300 mg 5 81/337 −1.26 (−1.86 to −0.65) 46 18 Dizziness, somnolence,
peripheral edema, confu-
sion, blurry vision
Rosenstock et al., 2004 Pregabalin, 300 mg I 8 76/146 −1.47 (−2.19 to −0.75) 40 14.5
Lesser et al., 2004 Pregabalin, 600 mg I 5 82/337 −1.45 (−2.06 to −0.85) 48 18
Richter et al., 2005 Pregabalin, 600 mg I 6 72/223 −1.26 (−1.89 to −0.64) 39 15
Freynhagen et al., 2005 Pregabalin, II 12 82/209 Approximately −1.4 to 48-52 24
300–600 mg 1.6 (p = 0.002)
Backonja et al., 1998 Gabapentin, I 8 70/135 −1.2 (−1.9 to −0.6) Not reported; 60% treated with Dizziness, somnolence,
900–3600 mg gabapentin had at least moder- confusion
ate improvement (>30%) vs
33% treated with placebo

Gorson et al., 1999 Gabapentin, 900 mg II 6 19/30 No difference Not reported; 42.5% treated with
gabapentin reported moderate
or excellent pain relief vs 22.5%
treated with placebo

Simpson, 2001 Gabapentin, II 8 27/54 −1.9 (Not reported; Not reported; 55.5% treated with
900-3600 mg p < 0.01) gabapentin reported much to
moderate improvement vs 25.9%
treated with placebo

Vrethem et al., 1997 Amitriptyline, 75 mg I 4 33/99 −1.8 (Not reported; Not reported; 63% of patients Dry mouth, sedation, vertigo
p < 0.001) treated with amitriptyline had at
least 20% improvement vs 22%
treated with placebo

Max et al., 1987 Amitriptyline, II 6 29 (Crossover) Not reported Not reported; 65.5% treated with
25–150 mg amitriptyline reported moderate
to complete improvement vs
3.5% treated with placebo

Raskin et al., 2005 Duloxetine, 60 mg I 12 116/348 −0.9 (−1.39 to −0.42) 50 30 Nausea, somnolence, hyper-
hidrosis, anorexia
Goldstein et al., 2005 Duloxetine, 60 mg II 12 86/344 −1.17 (−1.84 to −0.5) 49 26
Wernicke et al., 2006 Duloxetine, 60 mg II 12 85/248 −1.32 (−1.95 to −0.69) 43 27
Raskin et al., 2005 Duloxetine, 120 mg I 12 116/348 −0.87 (−1.36 to −0.39) 39 30
Descargado para Paola Gavidia Davila ([email protected]) en Antenor Orrego Private University de ClinicalKey.es por Elsevier en julio 19, 2020.

Goldstein et al., 2005 Duloxetine, 120 mg II 12 80/344 −1.45 (−2.13 to −0.78) 52 26

Wernicke et al., 2006 Duloxetine, 120 mg II 12 78/248 −1.44 (−2.08 to −0.81) 53 27

Rowbotham et al., 2004 Venlafaxine, I 6 82/242 −0.7 (Not reported; 56 34 Nausea, dyspepsia, sweating,
150–225 mg p < .001) somnolence, insomnia,
blood pressure and cardia
rhythm changes
aClassI randomized controlled trials must have allocation concealment, clearly defined primary outcomes, and inclusion and exclusion criteria with greater than 80% of patients completing the study. Class II randomized controlled trials lack one or more of the require-
ments listed for class I studies.
bNumber of participants receiving the dosage in column 2 out of the total number of participants in the trial. Many trials had multiple intervention groups.
cRefer to source article for full reference listings.
Modified with permission from Callaghan BC, Price RS, Chen KS, et al. Distal symmetric polyneuropathy: a review. JAMA. 2015;14:2172–2181. Copyright ©2015 American Medical Association. All rights reserved.

Chapter 37 Complications of Diabetes Mellitus

1497
1498 SE C T I O N V I I I Disorders of Carbohydrate and Fat Metabolism

Is pain due to DSPN confirmed? No/not sure

Yes

Assess comorbidities, potential for AEs, Refer to

(Table 37.13), drug interactions, costs to select neurology/pain
initial therapy from the 3 choices below clinic

#Secondary amine tricyclic

*Voltage gated α2-δ ligand **Serotonin-norepinephrine
(pregabalin, gabapentin) reuptake inhibitor antidepressant (nortriptyline
(duloxetine, venlafaxine) desipramine)
Titrate dose as shown in Titrate dose as shown in Titrate dose as shown in
Table 37.13 Table 37.13 Table 37.13

No clinically meaningful effect

c. May add tramadol or

a. Switch to another agent b. Try combining agents
tapentadol
from above from above
if a and b fail

No clinically meaningful
Refer to pain clinic
effect/not tolerated

• Fig. 37.45 Algorithm for management of the patient with pain because of distal symmetric polyneuropa-
thy (DSPN). AE, Adverse events. *Pregabalin is FDA approved for painful DSPN, whereas gabapentin is
not. Pharmacokinetic profile, spectrum of AEs, drug interactions, comorbidities, and costs to be consid-
ered in selecting the agent of choice. **Duloxetine is FDA approved for painful DSPN, whereas venlafax-
ine is not. Pharmacokinetic profile, spectrum of AEs, drug interactions, comorbidities, and costs should
be considered in selecting the agent of choice. #None is FDA approved for painful DSPN. Spectrum of
AEs, drug interactions, and comorbidities need be considered if selecting these agents. (Modified from
Pop-Busui R, Boulton AJ, Feldman EL, et al. Diabetic neuropathy: a position statement by the American
Diabetes Association. Diabetes Care. 2017;40:136–154.)

including severity of symptoms, appearance of motor weakness, Clinical Features of Diabetic Autonomic
and failure of nonsurgical treatment.667
Another distinct class of diabetic neuropathies includes the Neuropathy
diabetic polyradiculoneuropathies. This disorder is also known by Diabetic autonomic neuropathy is secondary to dysfunction
many other names, including proximal motor neuropathy, diabetic in the sympathetic and parasympathetic nervous systems. The
amyotrophy, diabetic femoral neuropathy, and diabetic radiculo- pathogenesis underlying diabetic autonomic neuropathy is gener-
plexus neuropathy. It primarily affects older T2DM male patients. ally considered to be similar to DSPN, with a distal-to-proximal
The onset of the disease can be gradual but is more commonly dying back of axons due to multiple metabolic insults.671 Because
abrupt, presenting with excruciating pain in the thighs and hips or of the widespread innervation pattern of the autonomic nervous
buttocks, followed by significant weakness of the proximal muscles system to critical organs, the symptoms and signs of autonomic
of the lower limbs. The neuropathy begins unilaterally, but fre- dysfunction are varied, dependent on the organ system. These are
quently spreads bilaterally, and coexists with DSPN. Electrophysi- summarized in Table 37.14.646
ologic evaluation reveals lumbosacral plexopathy most commonly Cardiac autonomic neuropathy is clinically the most impor-
superimposed on DSPN.668 Patients usually clinically resolve with tant of the autonomic neuropathies because of its association with
time, and the general course of treatment is supportive. Use of intra- early cardiac arrhythmia, silent myocardial infarctions, and early
venous gammaglobulin is reported to accelerate the resolution of mortality.672 CAN may present as lightheadedness upon stand-
symptoms,669 but must be used with caution in diabetic patients ing, palpitations, frank syncope, and a generalized sense of weak-
because of the greater risk of inducing renal failure.670 ness. In very early stages, CAN may be asymptomatic and only

identified by abnormal heart rate variability, either with deep controlling all aspects of the metabolic syndrome. These rec-
breathing or upon standing. Later signs include resting tachycar- ommendations are presented in Table 37.16 and parallel those
dia with a heart rate above 100 beats per minute and orthostatic for DSPN (see Table 37.11).
hypotension defined as a reduction of more than 20 mm Hg in Diabetic autonomic neuropathy can involve the gastrointes-
systolic or 10 mm Hg in diastolic pressure when going from a tinal tract, resulting in esophageal dysmotility, delayed gastric
lying to a standing position. emptying, and both constipation and diarrhea with frank fecal
The 2017 ADA guidelines suggest screening for CAN in incontinence. Among these disorders, the best studied is diabetic-
all diabetic patients with DSPN or other microvascular com- mediated delayed gastric emptying, also known as gastroparesis.
plications of diabetes (Table 37.15).646 As with DSPN, there While estimates vary, the disorder is more prevalent in T1DM
are no disease-modifying therapies, and T1DM patients are than T2DM and has a prevalence of approximately 5% in T2DM
instructed to monitor glucose control, while T2DM patients patients.673 The 2017 ADA recommendations for the screen-
are instructed to focus more on general lifestyle parameters and ing and diagnosis of gastroparesis is presented in Table 37.17.646

TABLE 37.13 T
reatment for Pain Associated With Distal Symmetric Polyneuropathy
DOSE NNT Range 30–50% Common
Drug Class Agenta Initial Effective Improvementc Adverse Events Major Adverse Events
Anticonvulsants Pregabalinb 25–75 mg, 300–600 mg/day 3.3–8.3 Somnolence Angioedema
1–3×/day Dizziness Hepatotoxicity
Peripheral edema Rhabdomyolysis
Headache Suicidal thoughts and behavior
Ataxia Seizures after rapid
Fatigue discontinuation
Xerostomia Thrombocytopenia
Weight gain
Gabapentin 100–300 mg, 900–3600 mg/day 3.3–7.2 Somnolence Stevens-Johnson syndrome
1–3×/day Dizziness Suicidal thoughts and behavior
Ataxia Seizures after rapid
Fatigue discontinuation
Antidepressants
Serotonin- Duloxetineb 20–30 mg/day 60–120 mg/day 3.8–11 Nausea Stevens-Johnson syndrome
norepinephrine Somnolence Hepatotoxicity
reuptake Dizziness Hypertensive crisis
inhibitors Constipation Gastrointestinal hemorrhage
Dyspepsia Delirium
Diarrhea Myocardial infarction
Xerostomia Cardiac arrhythmias
Anorexia Glaucoma
Headache Suicidal thoughts and behavior
Diaphoresis Shift to mania in patients with
Insomnia bipolar disorder
Fatigue Seizures
Decreased libido Severe hyponatremia
Fragility bone fractures
Serotonin syndrome
Neuroleptic malignant syndrome
Venlafaxine 37.5 mg/day 75–225 mg/day 5.2–8.4 Nausea Same as duloxetine
Somnolence
Dizziness
Constipation
Dyspepsia
Diarrhea
Xerostomia
Anorexia
Headache
Diaphoresis
Insomnia
Fatigue
Decreased libido
Continued

TABLE 37.13 T
reatment for Pain Associated With Distal Symmetric Polyneuropathy—cont’d
DOSE NNT Range 30–50% Common
Drug Class Agenta Initial Effective Improvementc Adverse Events Major Adverse Events
Tricyclic antide- Amitriptyline 10–25 mg/day 25–100 mg/day 2.1–4.2 Xerostomia Delirium
pressants Somnolence Cardiac arrhythmias
Fatigue Conduction abnormalities
Headache Myocardial infarction
Dizziness Heart failure exacerbation
Insomnia Stroke
Orthostatic Seizures
hypotension Hepatotoxicity
Anorexia Bone marrow suppression
Nausea Suicidal thoughts and behavior
Urinary retention Shift to mania in bipolar
Constipation disorder
Blurred vision Neuroleptic malignant syndrome
Accommodation Serotonin syndrome
Disturbance Severe hyponatremia
Mydriasis Fragility bone fractures
Weight gain
Desipramine Same as above Same as above
Nortriptyline Same as above Same as above
Opioids Tramadol 50 mg 1–2×/ 210 mg/day 3.1–6.4 Somnolence Confusion
day Nausea Seizures
Vomiting Cardiac arrhythmias
Constipation Hypertension
Lightheadedness Hypersensitivity reactions
Dizziness Stevens-Johnson syndrome
Headache
Tapentad- Immediate Immediate release: N/A Somnolence Respiratory depression
olb release: Day 1: 700 mg; Nausea Serotonin syndrome
50–100 mg after day 1,
4–6×/day 60 mg/day
Extended Extended release: Vomiting Seizures
release: 50 50 mg 2×/day Constipation Hypertension
mg 2×/day Dizziness Neonatal opioid withdrawal
syndrome
aRefer to source article for specific studies referenced for each agent.
bFDA-approved.

cFDA considers 30% to 50% improvement to be significant.

NNT, Number needed to treat.
Modified from Pop-Busui R, Boulton AJM, Feldman EL, et al. Diabetic neuropathies: a position statement by the American Diabetes Association. Diabetes Care. 2017;40:136–154. Copyright and all rights
reserved. Material from this publication has been used with the permission of the American Diabetes Association.

Treatment of gastroparesis is multifaceted. Patients are instructed both with long-term diabetes. Erectile dysfunction is common in
to eat multiple small meals with lower fat and fiber content and men and may occur in up to 50% of men over the age of 40.
to discontinue any drugs that decrease gastric motility, such as Men are more likely to experience erectile dysfunction if they have
tricyclic antidepressants or opioids. If these interventions do not T1DM or T2DM, with an increase in dysfunction with a longer
provide sufficient symptomatic relief, patients can begin meto- duration of diabetes. Erectile dysfunction is further exacerbated
clopramide, the only FDA-approved drug for the treatment of by obesity, smoking, and hypertension, along with excessive alco-
gastroparesis. Patients should not take the drug for more than 5 hol use and selected medications.674 While there is likely a com-
consecutive days, secondary to the risk of developing extrapyrami- ponent of autonomic dysfunction in most affected individuals, it
dal symptoms, and even this short regimen should only be under- is essential to address the modifiable risks factors that are outlined
taken when all other interventions have failed.646 earlier. The phosphodiesterase type 5 inhibitors provide a first line
Sexual and bladder function are also under the control of the of therapy for erectile dysfunction; these drugs include sildenafil,
autonomic nervous system, and dysfunction occurs commonly in tadalafil, and vardenafil.675 Additional therapeutic approaches

TABLE 37.14 S
ymptoms and Signs Associated With Diabetic Autonomic Neuropathy
Cardiac Autonomic Neuropathy Gastrointestinal Urogenital Sudomotor
Resting tachycardia Gastroparesis (Gastropathy) Bladder dysfunction Dry skin
Abnormal blood pressure regulation • Nausea • Frequency • Anhidrosis
• Nondipping • Bloating • Urgency • Gustatory sweating
• Reverse dipping • Loss of appetite • Nocturia
• Early satiety • Hesitancy
• Postprandial vomiting • Weak stream
• Brittle diabetes • Dribbling
• Urinary incontinence
• Urinary retention
Orthostatic hypotension (all with standing) Esophageal dysfunction Male sexual dysfunction
• Lightheadedness • Heartburn • Erectile dysfunction
• Weakness • Dysphagia for solids • Decreased libido
• Faintness • Abnormal ejaculation
• Visual impairment
• Syncope
Orthostatic tachycardia or bradycardia and Diabetic diarrhea Female sexual dysfunction
chronotropic incompetence (all with standing) • Profuse and watery diarrhea • Decreased sexual desire
• Lightheadedness • Fecal incontinence • Increased pain during
• Weakness • May alternate with constipation intercourse
• Faintness • Decreased sexual arousal
• Dizziness • Inadequate lubrication
• Visual impairment
• Syncope
Exercise intolerance Constipation
• May alternate with explosive diarrhea

From Pop-Busui R, Boulton AJM, Feldman EL, et al. Diabetic neuropathies: A position statement by the American Diabetes Association. Diabetes Care. 2017;40:136–154. Copyright and all rights reserved.
Material from this publication has been used with the permission of the American Diabetes Association.

TABLE 37.15 American Diabetes Association TABLE 37.16 American Diabetes Association
Recommendations for Cardiac Autonomic Recommendations for Cardiac Autonomic
Neuropathy Screening and Diagnosis Neuropathy
• S ymptoms and signs of autonomic neuropathy should be assessed • O
ptimize glucose control as early as possible to prevent or delay the
in patients with microvascular and neuropathic complications. E development of cardiovascular autonomic neuropathy in people with
type 1 diabetes. A
• In the presence of symptoms or signs of cardiovascular autonomic
neuropathy, tests excluding other comorbidities or drug effects/ • C
onsider a multifactorial approach targeting glycemia among other
interactions that could mimic cardiovascular autonomic neuropathy risk factors to prevent cardiovascular autonomic neuropathy in
should be performed. E people with type 2 diabetes. C
• C
onsider assessing symptoms and signs of cardiovascular auto- • C
onsider lifestyle modifications to improve cardiovascular autonomic
nomic neuropathy in patients with hypoglycemia unawareness. C neuropathy in patients with prediabetes. C
ADA recommendations are assigned ratings of A, B, or C depending on the quality of evi- ADA recommendations are assigned ratings of A, B, or C depending on the quality of evi-
dence. Expert opinion E is a separate category for recommendations in which there is no dence.
evidence from clinical trials, in which clinical trials may be impractical, or in which there is
From Pop-Busui R, Boulton AJM, Feldman EL, et al. Diabetic neuropathies: a position state-
conflicting evidence.
ment by the American Diabetes Association. Diabetes Care. 2017;40:136–154. Copyright
From Pop-Busui R, Boulton AJM, Feldman EL, et al. Diabetic neuropathies: a position state- and all rights reserved. Material from this publication has been used with the permission of
ment by the American Diabetes Association. Diabetes Care. 2017;40:136–154. Copyright the American Diabetes Association.
and all rights reserved. Material from this publication has been used with the permission of
the American Diabetes Association.

hyperplasia (BPH), but are frequently exacerbated by diabetes.
include intracavernosal injections, vacuum devices, psychosexual While estimates vary, studies suggest up to a twofold increase in
counseling, and cognitive behavioral therapy.675 Generally, a com- lower urinary tract symptoms in men with diabetes.676 Urinary
bination of therapies provides the best therapeutic responses. incontinence occurs in 50% of middle-aged women with dia-
Bladder and lower urinary tract dysfunction can occur in men betes; in several studies, the presence of diabetes is reported to
and women with diabetes. Symptoms include frequent urination significantly increase the risk of female urinary incontinence.676
with urinary urgency, nocturia, and a weak urinary stream. In This can be multifactorial, due to multiple pregnancies, obesity,
older men, these symptoms can be attributed to benign prostatic and medications, in addition to diabetes. The studies that directly

TABLE 37.17 American Diabetes Association link bladder and lower urinary tract dysfunction to autonomic
Recommendations for Gastrointestinal neuropathy are listed in Table 37.18676 and suggest an association
Neuropathies with bladder and lower urinary tract dysfunction to both CAN
• E valuate for gastroparesis in people with diabetic neuropathy, and DSPN. In both men and women, bladder and lower urinary
retinopathy, and/or nephropathy by assessing for symptoms of tract dysfunction increase the risk of urinary tract infections.677
unexpected glycemic variability, early satiety, bloating, nausea, and As with all forms of diabetic autonomic neuropathy, patients are
vomiting. C counseled to achieve glucose control and address other comor-
• Exclusion of other causes documented to alter gastric emptying, bidities, including obesity, hypertension, and hyperlipidemia, that
such as use of opioids or glucagon-like peptide 1 receptor agonists may contribute to the disorder.678
and organic gastric outlet obstruction, is needed before performing Hypoglycemic unawareness is another form of diabetic auto-
specialized testing for gastroparesis. C nomic neuropathy with significant patient morbidity. Hypogly-
• To test for gastroparesis, either measure gastric emptying cemia is defined as a plasma glucose level of less than 70 mg/dL
with scintigraphy of digestible solids at 15-min intervals for (3.9 mmol/L) without autonomic warning signs.679,680 In most
4 h after food intake or use a 13C-octanoic acid breath patients, two categories of symptoms occur with hypoglycemia:
test. B neuroglycopenic and autonomic symptoms. Neuroglycopenic
ADA recommendations are assigned ratings of A, B, or C depending on the quality of e vidence.
From Pop-Busui R, Boulton AJM, Feldman EL, et al. Diabetic neuropathies: a position state-
symptoms are directly correlated with lack of available glucose for
ment by the American Diabetes Association. Diabetes Care. 2017;40:136–154. Copyright brain function, including hunger, altered mental status, confusion,
and all rights reserved. Material from this publication has been used with the permission of perioral paresthesias, difficulty speaking, diffuse weakness, and
the American Diabetes Association. syncope. In more severe cases, patients develop seizures, become
comatose, and (in extreme cases) die. Autonomic symptoms

TABLE 37.18 S
tudies of Autonomic Neuropathy and Bladder Dysfunction in Diabetes
Overall Population Definition of Bladder
Authors (Year)b (Diabetes Type) Definition of Autonomic Neuropathy Dysfunction Findings
Male + Female
Ueda et al., 1997 63 diabetesa Sympathetic skin response Volume at first desire Mean vol. at first desire to void,
to void max bladder capacity lower for
Max bladder capacity sympathetic skin response absent.
Bladder pressure Mean bladder pressure and residual
Residual urine urine greater for sympathetic skin
response absent
Low et al., 2004 231 T1DM/T2DM Autonomic Symptom Profile (ASP) ASP urinary domain: Significant correlations between ASP
Composite Autonomic Severity Score bladder dysfunction, urinary domain and overall CASS
(CASS) sexual dysfunction and domain scores
(males only)
Kebapci et al., 54 T2DM CAN: deep breathing, Valsalva, stand test LUTS: IPSS, urinary QT prolongation associated with
2007 27 males incontinence, uro- increased postvoid residual urine
27 females dynamic studies OR 2.33 (0.16–34.89)
Pavy-Le Traon 684 T1DM CAN severity Ewing Score (0–5): deep Bladder dysfunction Bladder dysfunction independently
et al., 2010 breathing, Valsalva, stand test, HRV, SBS symptoms associated with CAN
Males
Pop-Busui et al., 635 T1DM DCCT/ CAN: R-R variation <15, or R-R varia- LUTS: AUASI 8–35 LUTS prevalence: 158 (25 %)
2015 EDIC study tion 15–19.9 plus Valsalva ratio ≤1.5, Odds of ED + LUTS: 2.65 (1.47–4.79)
10 mmHg drop in DBP
Bansal et al., 52 diabetesa Sympathetic skin response: Medtronic LUTS: IPSS 8–35 uro- Diabetic cystopathy correlated with
2011 electromyographic system dynamic studies abnormal motor and sensory nerve
conduction velocity studies and
abnormal sympathetic skin responses
Females
Hotaling et al., 571 T1DM DCCT/ CAN: R-R variation <15, or R-R varia- UI: Sandvik Severity UI prevalence: 172 (30 %)
2016 EDIC study tion 15–19.9 plus Valsalva ratio ≤1.5, Index 3–12
10 mm Hg drop in DBP
aType of diabetes not indicated.
bRefer to source article for full reference listing of each study.
AUASI, American Urological Association Symptom Index; CAN, cardiovascular autonomic neuropathy; DBP, diastolic blood pressure; DCCT/EDIC, Diabetes Control and Complications Trial/Epidemiology
of Diabetes Interventions and Complications; ED, erectile dysfunction; HRV, heart rate variability; IIEF, International Index of Erectile Dysfunction; IPPS, International Prostate Symptom Score; LUTS, lower
urinary tract symptoms; SBS, spontaneous baroreflex slope; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus; UI, urinary incontinence.
Modified from Braffett BH, Wessells H, Sarma AV. Urogenital autonomic dysfunction in diabetes. Curr Diabetes Rep. 2016;16:119.

include diaphoresis, tremulousness, pallor, palpitations, and gen- self-reported assessment tool, the global Hypoglycemia Assess-
eralized anxiety. Hypoglycemic unawareness is defined as the pres- ment Tool (HAT), reported the global incidence of hypoglycemic
ence of neuroglycopenic symptoms prior to or in the absence of unawareness in T1DM and T2DM patients on insulin therapy as
autonomic warning symptoms.681 97.4% and 95.3%, respectively, with approximately 6.9 events per
Counterregulatory responses under the control of the auto- month in T1DM patients and 2.4 events in T2DM patients.685
nomic nervous system are required to restore normal glycemia in Regardless of the exact percentages, it is clear that hypoglyce-
response to hypoglycemia. These essential responses include stimu- mic unawareness is common and a source of significant patient
lation of catecholamines (both norepinephrine and epinephrine), morbidity. There is a strong association between recurrent hypo-
glucagon, cortisol, and growth hormone, while inhibiting insulin glycemia and an increase in fatal cardiac arrhythmias.686 An analy-
secretion.682 These counterregulatory responses stimulate hepatic sis of unrecognized hypoglycemia in 10,096 participants from the
glucose production while decreasing glucose utilization in the ACCORD trial revealed a relationship between the number of
periphery, and collectively restore euglycemia (Fig. 37.46). Recur- episodes of hypoglycemic unawareness and death, leading to early
rent hypoglycemic episodes decrease the plasma glucose level that discontinuation of the trial.687 While hypoglycemic unawareness
serves as the setpoint for activating the counterregulatory response does not result in significant permanent cognitive impairment
in both T1DM and T2DM patients.682 Severe hypoglycemia with in adults,688,689 it represents a major and potentially disabling
life-threatening neuroglycopenic symptoms is more common in event in children and adolescents.690 Recurrent episodes in young
T1DM patients with hypoglycemic unawareness and correlates children can result in behavioral problems and neurocognitive
with longer duration of diabetes, intensive control, and advanc- dysfunction691,692 associated with changes in brain structure.693
ing age.683,684 While a commonly cited statistic is that 40% of In adults who drive, it can result in potentially fatal automobile
T1DM patients experience one or more episodes of hypoglyce- accidents.
mic unawareness,682 a recent multinational study using a patient Management of hypoglycemic unawareness begins with pre-
vention, with careful blood glucose monitoring and reasonable
Sensor
blood glucose targets personalized to each patient,694 in parallel
Hypoglycemia with patient education.695 This approach can be augmented with
Brain glucose β2-adrenergic agents, caffeine, and serotonin reuptake inhibitors,
Brain sensing although the effects are small and use of these agents is controver-
Hypothalamus sial.682 Most therapeutic efforts focus on glycemic control. Con-
brain stem tinuous glucose monitoring (CGM) to identify hypoglycemia can
decrease the number and severity of adverse hypoglycemic events.
In a group of T1DM patients with hypoglycemic unawareness,
Autonomic nervous system CGM reduced severe hypoglycemic events from 8.1 to 0.6 events
per year.696 Pharmacologic treatment options begin with optimiz-
ing insulin regimens; for example, rapid-acting insulin analogs
Effector afford more flexibility for patients with exercise regimens and
organs
varying mealtimes,697 while long-acting insulin analogs provide
a flatter, less variable glycemic profile that decreases the rate of
Pancreas Adrenal
overall and nocturnal hypoglycemia.698 The use of continuous
gland
subcutaneous insulin infusion (insulin pump) is very effective in
Counter- decreasing the number of hypoglycemic events,699 with reported
regulatory improvements in patient-reported quality of life assessments.700
hormones Regardless of the insulin regimen chosen by the patient, person-
D-cell E-cell
alized insulin treatment with appropriate glycemic targets is the
cornerstone of treatment.682,701
n Glucagon p Insulin n Cortisol n Epinephrine

Diabetic Heart Disease

Coronary Artery Disease
Target
organs In recent decades, cardiovascular disease mortality rates have
declined substantially in the general population in the United
Muscle Liver States. However, improvement in cardiovascular mortality rates
has been significantly less in people with diabetes.8 Diabetes
remains associated with excess morbidity and mortality rates even
after adjustment for traditional risk factors. Most studies of car-
diovascular risk in diabetes have been in the T2DM population,
who represent more than 90% of all patients with diabetes.
p Glucose utilization n Glucose output
Effect of Diabetes on Risk of Coronary Heart Disease
• Fig. 37.46 Counterregulatory
response to hypoglycemia in type 2 dia-
(CHD)
betes. (From Martín-Timón I, del Cañizo-Gómez FJ. Mechanisms of The Framingham Heart Study showed a twofold to threefold eleva-
hypoglycemia unawareness and implications in diabetic patients. World J tion in risk of clinically evident atherosclerotic disease in patients
Diabetes. 2015;6:912–926.) with T2DM compared to those without diabetes.702 Diabetic

men in the Multiple Risk Factor Intervention Trial (MRFIT) had of coronary heart disease in this population. Studies suggest that
an absolute risk of cardiovascular death more than three times proteinuria is a marker of generalized vascular damage that predis-
higher than that of the nondiabetic cohort, even after adjustment poses to atherosclerosis and coronary events.
for established risk factors.703 Seminal work from Finland showed It is unclear whether there has been any recent decline in mor-
that patients with T2DM without a previous myocardial infarc- tality or morbidity rates from coronary artery disease associated
tion have a risk of myocardial infarction over 7 years as high as that with T1DM. The Pittsburgh Epidemiology of Diabetes Compli-
of nondiabetic patients with a history of a myocardial infarct.704 cations (EDC) study reported no difference in the cumulative inci-
In this study, the case fatality rate after myocardial infarction was dence of cardiovascular disease with 20, 25, or 30 years of disease
also substantially higher in patients with diabetes. In women, dia- duration according to year of diagnosis (1950 through 1980).713
betes eliminates the cardioprotective effects of the premenopausal The benefits of improved treatment of hyperglycemia and con-
period, and women with diabetes have a cardiovascular mortality comitant hypertension and other comorbid conditions with stan-
rate as high as that of diabetic men. The follow-up of the Diabe- dard care therefore do not appear to have reduced cardiovascular
tes UK cohort, a group of 23,751 subjects with insulin-treated mortality rates associated with T1DM, in contrast to dramatic
diabetes diagnosed before the age of 30 years, also showed similar reductions realized for diabetic retinopathy and nephropathy.
mortality rates for men and women, and the size of this cohort
permitted robust gender-specific estimates of standardized mortal- Aggregation of Traditional CHD Risk Factors in Diabetes
ity ratios.705 For example, among those aged 30 to 39 years, the Multiple traditional cardiovascular risk factors (e.g., hyperten-
standardized mortality ratios for ischemic heart disease mortal- sion, dyslipidemia, obesity, insulin resistance) occur together in
ity were 8 and 41.6, respectively. Other forms of cardiovascular patients with diabetes.714 Approximately 50% of patients with
diseases, such as hypertension, valvular disease, cardiomyopathy, T2DM have hypertension, and more than 30% have hypercho-
heart failure, and stroke, are also increased. lesterolemia at the time of diagnosis. As in nondiabetic patients,
The risks of cardiovascular events and death conferred by these risk factors independently predict cardiovascular mortal-
T2DM have been examined in several prospective and observa- ity.703 However, even in the presence of one or more concomitant
tional trials with various populations of patients. The adjusted risk factors, diabetes further increases the cardiovascular death rate
relative risk is generally increased by 1.3 to 1.6 for total mortality, (Fig. 37.47). It also appears that diabetes interacts synergistically
mortality attributed to cardiovascular events, nonfatal myocardial with other risk factors to more sharply increase risk as the number
infarction or stroke, and congestive heart failure.706–709 Many epi- of total risk factors increases.
demiologic studies observe the hazard ratio for cardiovascular mor- The UKPDS further confirmed the importance of risk factor
tality risk in patients with diabetes is similar to that in nondiabetic aggregation in T2DM as well as the relationship between specific
patients who had experienced a previous myocardial infarction. risk factors and future cardiovascular events. In this large popula-
Whether or not the diagnosis of diabetes itself is a risk equivalent tion of patients with newly diagnosed T2DM, higher LDL choles-
to having a history of prior myocardial infarction remains con- terol levels, lower HDL cholesterol concentrations, higher HbA1c
troversial.709,710 However, all studies demonstrate diabetes results levels and systolic blood pressure measurements, and a history of
in a significantly increased risk of cardiovascular disease. Thus smoking measured at baseline were all associated with a risk of
the strategy of considering diabetes as a CHD risk equivalent is future cardiovascular disease.10
appropriate for purposes of assessing risk and defining an aggres- Cardiovascular risk associated with T2DM may be a conse-
sive multirisk treatment regimen. With more intensive manage- quence of insulin resistance and its associated abnormalities that
ment of cholesterol, blood pressure, and decreased smoking, rates are often present in the prediabetic state.715 In patients spanning
of myocardial infarction, stroke, amputation, and mortality are
decreasing among people with diabetes, especially in high-income
Influence of multiple risk factors* on CVD
countries. Yet the global burden of cardiovascular disease attrib- death rates in diabetic and nondiabetic men
utable to diabetes is not declining due to major increases in the 140
Age-adjusted CVD death rate

number of people developing diabetes.711

per 10,000 person-years

120 No diabetes
Coronary heart disease risk has also been evaluated in smaller Diabetes
numbers of patients with T1DM. In the Framingham study, the 100
cumulative cardiovascular mortality rate in patients with T1DM 80
was approximately four times that of nondiabetic patients by age 60
55 years.710 Patients with T1DM and a HbA1c level of 6.9% or
40
lower, considered at goal by most management guidelines, had a
risk of death from any cause or from cardiovascular causes that was 20
twice as high as the risk for matched controls.712 As in patients 0
with T2DM, the early deaths related to coronary heart disease in None One only Two only All three
patients with T1DM may occur by the fourth decade of life, and
*Serum cholesterol >200 mg/dL,
the cumulative mortality rate increases at a similar rate in both smoking, SBP >120 mm Hg
T1DM and T2DM in the subsequent decades. The increased car-
diovascular mortality rate with age in patients with T1DM is sub- • Fig. 37.47 Age-adjusted cardiovascular disease (CVD) death rates by
number of risk factors for men with and without diabetes at baseline
stantially even higher in those patients who also have nephropathy,
screened for the Multiple Risk Factor Intervention Trial. In the presence
with gradations of cardiovascular risk evident across the spectrum of diabetes, the cardiovascular death rate steeply rises at any level of
from microalbuminuria to macroalbumuria to end-stage renal dis- concomitant risk factors. SBP, Systolic blood pressure. (From Stamler J,
ease, with a 2.8-fold, 9.2-fold, and 18.3-fold increase in mortality Vaccaro O, Neaton JD, et al. Diabetes, other risk factors, and 12-year
compared to the general population across these groups.451 There- cardiovascular mortality for men screened in the Multiple Risk Factor Inter-
fore persistent proteinuria is a strong predictor of the development vention Trial. Diabetes Care. 1993;16:434–444.)

the spectrum of glucose tolerance, from normal to impaired glu-

Ischemic heart disease mortality (%)
cose tolerance to frank diabetes, insulin resistance positively cor-
relates with atherosclerosis as assessed by carotid intimal medial 50 Cardiovascular mortality (%)
thickening.716 The San Antonio Heart Study, a population-based All-cause mortality
study of diabetes and cardiovascular disease in Mexican Americans 40
and non–Latin American whites, showed those who converted to
diabetes from a prediabetic state and who were more insulin resis-

Mortality (%)
tant had higher blood pressure, higher triglyceride levels, and lower 30
HDL cholesterol levels.17 Strikingly, individuals with normal glu-
cose tolerance and high insulin resistance had twice the 8-year 20
risk of adverse cardiovascular outcomes compared to those with
low insulin resistance, even after adjustment for 11 risk factors.
Although it is not possible to fully distinguish hyperinsulinemia 10
from hyperglycemia, hyperinsulinemia has been hypothesized to
contribute mechanistically to the relationship between hypergly-
0
cemia and coronary heart disease in patients with T2DM,717 and
a number of studies have shown hyperinsulinemia to indepen- 4.7–<6.7 6.7–<7.8 7.8–<8.8 8.8–9.9 =>10.0
dently predict cardiovascular risk. These results suggest treatment Quintiles of average FBG values
strategies that increase insulin sensitivity in these patients can • Fig. 37.48 All-cause mortality, cardiovascular mortality, and ischemic
reduce cardiovascular risk. However, no significant difference in heart disease mortality rates in patients with type 2 diabetes mellitus by
the rates of death and major cardiovascular events were observed quintiles of average fasting blood glucose (FBG) values. Cardiovascular
using insulin provisional compared to insulin sparing therapeutic mortality and all-cause mortality rates increase throughout the range of
strategies in patients with both T2DM and heart disease in the fasting plasma glucose in a graded fashion. (From Andersson DK, Svärd-
Bypass Angioplasty Revascularization Investigation 2 Diabetes sudd K. Long-term glycemic control relates to mortality in type II diabetes.
Trial (BARI 2D).718 Furthermore, when insulin therapies were Diabetes Care. 1995;18:1534–1543.)
used prospectively to lower HbA1C in either patients with T2DM
or prediabetes, they have not been shown to worsen cardiovascular captured by HbA1c may also explain excess cardiovascular risk in
outcomes. This suggests that hyperinsulinemia is a marker of insu- prediabetes as well as in diabetes. Hyperglycemia increases insulin
lin resistance but not a cause of the adverse cardiovascular effects resistance, and circadian variation in insulin resistance causes fast-
of insulin resistance. ing hyperglycemia. One study showed that mortality rates from
Given the multifactorial nature of atherogenic risk in patients all causes, cardiovascular mortality, and ischemic heart disease
with T2DM, it is reasonable to conclude that an aggressive mul- increased progressively across quintiles of fasting blood glucose
tifactorial intervention could significantly reduce cardiovascu- levels in patients with T2DM (Fig. 37.48).719 Other data sup-
lar risk. The value of such a treatment regimen was tested in the port a dose-response relationship between fasting hyperglycemia
Steno-2 study,595 in which 160 patients with T2DM and microal- and cardiovascular mortality rate in diabetes, with patients with
buminuria were randomized to receive either conventional treat- the highest levels of fasting blood glucose having a cardiovascular
ment in accordance with national guidelines or intensive therapy mortality rate almost five times higher than patients with the two
that included behavior modification and targeted pharmacologic lowest levels combined.720
therapy for hyperglycemia, hypertension, dyslipidemia, and Prediabetes refers to the presence of impaired fasting glucose
microalbuminuria along with secondary prevention of cardio- (≥100 mg/dL [≥5.5 mmol/L] and <120 mg/dL [<6.7 mmol/L]),
vascular disease with aspirin. Over a mean follow-up period of and impaired glucose tolerance, defined as a 2-hour post–75-g oral
7.8 years, patients who received intensive treatment had greater glucose load value of 140 to 199 mg/dL (7.8–11 mmol/L) with
improvements in HbA1c, blood pressure, fasting serum choles- the fasting level below 100 mg/dL (<5.5 mmol/L) and/or HbA1c
terol and triglyceride values, and urinary albumin excretion than between 5.7% and 6.5%. Impaired glucose tolerance (IGT)—a
did patients receiving conventional therapy. The greater degree prediabetic state associated with insulin resistance—is also a risk
of improvement in risk factors with intensive therapy was also factor for mortality. In a large study of the relationship between
reflected in outcomes. Patients receiving intensive therapy had 2-hour postload blood glucose and cardiovascular fatality, 17,869
an approximately 50% reduction in events comprising a multiple male civil servants enrolled in the Whitehall Study from 1967 to
cardiovascular endpoint, and in retinopathy, nephropathy, and 1969 were monitored, and outcomes were correlated with base-
autonomic neuropathy.594 line measurements of the 2-hour postload blood glucose (2hBG)
level after a 50-g oral glucose load.721 The hazard ratio for car-
Insulin Resistance, Plasma Glucose, and Excess CHD Risk diovascular mortality increased as a linear function of 2hBG for
in Diabetes all values of 2hBG greater than 83 mg/dL (>4.6 mmol/L). With
Insulin resistance and its consequences increase cardiovascular 2hBG values between 83 and 200 mg/dL (4.6 and 11.1 mmol/L),
risk in people with normal glucose tolerance. Yet diabetes confers the age-adjusted hazard ratio for cardiovascular disease was 3.62
increased risk that is not explained by the known risk factors asso- (95% CI, 2.3–5.6).
ciated with insulin resistance. Newer proatherogenic consequences Tominaga and colleagues examined survival rates in a cohort
of pathway-specific insulin resistance are discussed in the first sec- of participants in a diabetes prevalence trial in Japan722 and
tion of this chapter. Although a meta-analysis of randomized con- concluded the risk of cardiovascular fatality is associated with
trolled trials showed limited benefits of intensive glucose-lowering impaired glucose tolerance rather than impaired fasting glucose.
treatment on all-cause mortality and deaths from cardiovascular Further substantiation of the role of impaired glucose tolerance
causes, using HbA1c as an indicator,13 effects of hyperglycemia not in cardiovascular mortality risk was provided by an analysis of

data from the Diabetes Epidemiology: Collaborative Analysis of inflammatory markers (C-reactive protein, PAI1, and fibrinogen)
Diagnostic Criteria in Europe (DECODE) study.723 In this study, than converters to diabetes with predominantly low insulin secre-
more than 25,000 men and women were monitored for a mean tion or nonconverters. Thus insulin resistance is associated with a
of 7.3 years, and outcomes were correlated with measurements of proinflammatory state that can contribute to the atherogenic risk
fasting glucose and 2hBG after a 75-mg glucose load at baseline. profile in prediabetic patients. Many aspects of insulin resistance
The results indicated that the oral glucose tolerance test provides and the proatherogenic and proinflammatory state of prediabetes
the best index of risk of mortality associated with impaired glucose are associated with increased visceral fat.731
tolerance. By 2012, more than one-third of all US adults met the defi-
The Nurses’ Health Study also implicated the prediabetic state nition and criteria for metabolic syndrome agreed to jointly by
as a cardiovascular risk factor.724 In this large cohort of women, several international organizations.732 Although there continues
5894 developed diabetes over a 20-year follow-up. The age- to be debate about the validity and significance of identifying the
adjusted relative rate for myocardial infarction was 3.75 (95% CI, metabolic syndrome or prediabetes in patients, clinicians, includ-
3.10–4.53) in the period before diagnosis of diabetes and 4.57 ing primary care physicians, often find the construct of value in
(95% CI, 3.87–5.39) after diagnosis, compared with women who understanding the relevant underlying issues and explaining these
did not develop diabetes, even after adjustment for other cardio- to patients. Behind some of the reluctance to identify metabolic
vascular risk factors. The risk of stroke was also increased before syndrome/prediabetic states in patients is the clinical question of
onset of diabetes. what to do after such a diagnosis is made. Perhaps new data show-
The continuum of cardiovascular risk with rising glucose levels ing cardiovascular risk reduction with newer glucose-lowering
has also been identified in patients with T1DM725 and in people therapies that do not cause hypoglycemia (discussed later) may
without clinically overt diabetes but with varying levels of glucose shift this perspective. These new drug options, along with effective
intolerance. medications for reducing cholesterol and hypertension, support
the clinical utility of continuing to identify prediabetes and the
Cardiovascular Disease in Patients With metabolic syndrome.
Metabolic Syndrome Cardiovascular Consequences of Metabolic Syndrome
Definitions and Diagnosis Metabolic syndrome, prediabetes, and other overlapping defini-
Metabolic syndrome is a term referring to the frequent clustering tions from the NCEP, WHO, and others have been evaluated
of the cardiovascular risk factors of hypertension, obesity, insulin in multiple studies regarding the incidence and mortality risk of
resistance, dyslipidemia, and dysglycemia in a single patient. Sepa- CHD.733–735 In general, age-adjusted cardiovascular mortality
rate diagnostic criteria for different risk components have been put and all-cause mortality increase twofold to threefold in patients
forth by the National Cholesterol Education Program (NCEP) with metabolic syndrome, regardless of the diagnostic criteria used
Adult Treatment Panel III (ATP III),726 the World Health Orga- to define the syndrome. Increased risk is not seen in every study.
nization (WHO),727 and the International Diabetes Federation Differences among these studies exist and involve when and how
(IDF),728 with clinical overlap among these definitions and the analyses were done and the metabolic syndrome definitions used,
designation of prediabetes.729 including key issues such as whether frank diabetes is included or
According to prior NCEP guidelines,729a the metabolic syn- excluded from metabolic syndrome definition and the hypergly-
drome is based on the presence of three of the following five risk cemia threshold used.
factors726: In a definitive meta-analysis of 87 studies involving 951,083
• Abdominal obesity (waist circumference >40 inches (>101.6 patients, the metabolic syndrome was found to be associated with
cm) in men and >35 inches (>88.9 cm) in women) an increased risk of cardiovascular disease (relative risk [RR]: 2.35;
• Plasma triglycerides 150 mg/dL (1.7 mmol/L) or higher 95% CI: 2.02–2.73), cardiovascular mortality (RR: 2.40; 95%
• Plasma HDL cholesterol less than 40 mg/dL (<1.04 mmol/L) CI: 1.87–3.08), all-cause mortality (RR: 1.58; 95% CI: 1.39–
in men, less than 50 mg/dL (<1.30 mmol/L) in women 1.78), myocardial infarction (RR: 1.99; 95% CI: 1.61–2.46), and
• Blood pressure 130/85 mm Hg or higher stroke (RR: 2.27; 95% CI: 1.80–2.85); metabolic syndrome in
• Fasting plasma glucose 110 mg/dL (6.1 mmol/L) or higher the absence of diabetes was still associated with increased cardio-
The NCEP criteria give precedence to obesity as a contributor vascular risk,732 although it is based on fewer studies.
to the metabolic syndrome and apply cutoff points for triglycer- Another approach to assessing cardiovascular risk of metabolic
ides and HDL that are probably less stringent than would be used syndrome involves consideration of this cohort in clinical tri-
to identify a categoric risk factor, reflecting the fact that many als. For example, in the West of Scotland Coronary Prevention
marginal risk factors can result in a significant risk for CHD. The Study,735 those with the metabolic syndrome had an increased
NCEP criteria do not require explicit demonstration of insulin cardiovascular risk, along with elevated C-reactive protein levels
resistance for diagnosis of the metabolic syndrome, and patients versus the rest of the cohort, which added prognostic value for
with diabetes are not excluded from the diagnosis.727 There are both cardiovascular disease and diabetes.
modest differences between the NCEP criteria and those devel- Controversies over metabolic syndrome and other prediabetic
oped by the WHO727 and the IDF.728 One intriguing aspect of states exist.736 One issue has been whether the cardiovascular risk
the IDF criteria is the recognition that individuals of Asian origin associated with the metabolic syndrome differs from the additive
may have increased visceral adiposity with lower waist circumfer- risk conferred by the presence of each metabolic syndrome com-
ence than what is observed in Western countries and hence the ponent. Another has been what therapeutic intervention should
difference in waist circumference criteria for such individuals. ensue once a diagnosis of metabolic syndrome is made. These and
Studies support the concept that insulin resistance during other issues have limited the extent to which this issue has been
the prediabetic state contributes to atherogenic risk.730 Pre- studied in trials or further considered by guidelines.737 Evidence
diabetic subjects who were insulin resistant had higher levels of continues to emerge regarding the pathologic issues associated

with prediabetes, including intimal medial thickening and plaque limitations. Although relatively large (N = 1441), the DCCT fol-
in carotid arteries, and adverse functional cardiac parameters such lowed a relatively young (mean age 27 years) population of patients
as reduced diastolic filling (now referred to as heart failure with with T1DM for approximately 6 years. At the end of follow-up,
preserved ejection fraction).738 For now, specific relatively subtle dif- few events had occurred.9 Intensive therapy reduced the risk of
ferences among diagnostic criteria of prediabetic states may be less coronary artery and peripheral vascular disease by 41% compared
relevant for the clinician than the recognition of the increased risk with conventional therapy, but the difference was not statistically
of cardiovascular disease and diabetes associated with metabolic significant. Similarly, in the VA trial, intensive blood glucose con-
syndrome, the commonly encountered clustering of these risk fac- trol in patients with T2DM trended to but did not significantly
tors, and the impact lifestyle and modest weight reduction can reduce cardiovascular endpoints.739 Both studies lacked adequate
have of many component risk factors. New data for cardiovascular power to detect a difference in macrovascular events between
event reduction in patients with diabetes with agents that do not treatment groups because of the small number of events in each
cause hypoglycemia as well as weight loss agents may stimulate group, small patient populations, and relatively short follow-up.
reexamination of these issues. In contrast, the 17-year follow-up of the 1441 patients from the
DCCT trial (EDIC) demonstrated the benefit of early intensive
The Role of Glycemic Control in Improving glycemic control in T1DM on cardiovascular risk in the subsequent
post-DCCT, EDIC observational setting.740 This study and its
Cardiovascular Outcomes findings are discussed in the sections on biochemistry and molecu-
The UKPDS confirmed the positive association between plasma lar cell biology and retinopathy earlier in the chapter. During fol-
glucose levels and CHD risk for HbA1c levels greater than 6.2% low-up, those who had previously been randomized to intensive
in patients with diabetes.717 Each percentage point elevation in treatment had an emergent reduction in cardiovascular risk of 42%
HbA1c increased cardiovascular risk by 11%. Multiple other lines (p = 0.02) and a reduction in the composite risk of nonfatal myo-
of evidence, as noted, confirm the relationship between hypergly- cardial infarction, nonfatal stroke, or cardiovascular death of 57%
cemia, diabetes, and increased cardiovascular risk. Despite this, (p = 0.02). Of note, this separation was evident even though after
it has been difficult to establish that glucose control in diabetes 11 years of follow-up, HbA1c, as well as blood pressure and lipid risk
could decrease future macrovascular events, including trials inves- factors, were essentially identical between the two arms. Patients
tigating intensive versus standard diabetes treatment and different in the conventional treatment group had more albuminuria and
strategies of achieving glycemic control. Although meta-analyses microalbuminuria than intensively treated patients, but the differ-
of randomized controlled trials of diabetes treatments revealed ences in risk remained significant after adjustment for these factors.
improved glucose control was associated with reductions in non- These findings indicate early intensive glycemic control reduced
fatal myocardial infarction rates, there has been no evidence for the long-term risk of cardiovascular disease in patients with T1DM
decreases in either cardiovascular death or all-cause mortality.13 while also raising intriguing mechanistic questions regarding meta-
Given the clear increased cardiovascular risk with diabetes, the bolic memory (discussed in the first section of this chapter).
inability to reduce cardiovascular events in response to treating For T2DM, the UKPDS was larger and adequately powered to
diabetes represented a clinical paradox, a contrast to contempora- detect differences between groups in terms of macrovascular events
neous studies showing the increased cardiovascular risk of hyper- in regular versus intensive glycemic control.379 Those receiving
cholesterolemia could be reversed by lowering LDL with statin intensive glycemic control showed a trend toward lower myocardial
therapy, including in patients with diabetes. Various hypotheses infarction rates compared to conventional treatment, which was not
have been raised regarding this inability of diabetes treatment to statistically significant (p = 0.052).379 Despite the lack of overall
show improved cardiovascular outcomes, including trial issues efficacy of intensive treatment for management of macrovascular
such as study design, sample size, duration, and poorly selected complications of diabetes in the UKPDS, there were indications
primary endpoints, and disease issues such as disease duration or that metformin as a specific therapy may be effective in reducing
treatment-specific factors (e.g., drugs used). cardiovascular events.379,741 In a retrospective analysis of an over-
In 2008, the FDA mandated that novel glucose-lowering thera- weight subset (n = 342) of the UKPDS cohort who were random-
pies seeking approval must demonstrate cardiovascular safety. The ized to metformin, significant reductions in the occurrence of any
resulting data, involving over nine cardiovascular outcome trials diabetes-related endpoint (32%), diabetes-related death (42%), and
reported to date involving 200,000 patients, have led to a striking all-cause mortality (36%) were seen compared with conventionally
pivot for the fields of both diabetes and cardiology with data dem- treated patients.741 Although this is subset data in a limited num-
onstrating not only cardiovascular safety but also cardiovascular ber of patients, these results, along with registry data suggesting
benefit for glucagon-like peptide 1 receptor agonists (GLP1RA) decreased mortality with metformin use,742 promoted acceptance
and sodium glucose transporter 2 inhibitors (SGLT2i). Other of metformin having potential cardiovascular benefit.
studies established cardiovascular safety but also identified poten- Subsequent trials examined whether lowering the target for
tial concerns over an increased risk for inducing heart failure, as glycemic control (to HbA1c <7%) would reduce cardiovascular
seen with some dipeptidyl-peptidase 4 inhibitors (DPP4i). Given events. The ACCORD trial randomized patients with T2DM who
the landmark nature of these trials, their impact on clinical man- had established cardiovascular disease or multiple risk factors to
agement of diabetes, and the new questions posed by these find- a treatment-directed HbA1c value lower than 7% or to standard
ings, a review of the prior studies with older antidiabetic agents therapy with an HbA1c target value between 7% and 7.9%. Over
warrants review and provides context before the results with these 3.5 years of follow-up, total mortality rate increased in the inten-
newer agents are considered. sively treated group without any significant reduction in cardiovas-
Earlier studies, such as the DCCT in T1DM and the smaller cular events.535 The VADT, which randomized patients to either
Veterans Affairs Diabetes Feasibility (VA) trial in T2DM, did intensive treatment (median HbA1c 6.9%) or standard therapy
not show a reduction in cardiovascular endpoints with intensive (median HbA1c 8.4%) also failed to demonstrate cardiovascular
metabolic control. These well-conducted studies had inherent benefit through tighter HbA1c control.538 The ADVANCE trial,

which randomly assigned patients to an intensively treated group, of similar glucose control in both arms.745 After a mean of 34.5
who attained HbA1c of 6.5% versus a standard control group who months, there was no significant difference in the two treatment
had HbA1c values of 7.3%, demonstrated a relative reduction of groups in the study’s primary endpoint—a seven-component com-
10% in the combined outcome of major macrovascular and micro- posite of all-cause mortality, nonfatal myocardial infarction, stroke,
vascular events that was not statistically significant.532 Of note, acute coronary syndrome, surgical intervention in the leg or coro-
this statistically insignificant 10% relative decrease in cardiovascu- nary arteries, and amputation above the ankle. This broad primary
lar events was driven largely by a 21% reduction in nephropathy. endpoint, which extended beyond the more standard objective car-
Taken together, these three studies do not support HbA1c reduction diovascular primary endpoint used in cardiovascular trials, was used
beyond 7%, improving cardiovascular outcomes in patients with to increase the prospect of demonstrating cardiovascular benefit,
established coronary heart disease and treated with these agents. under the assumption that peripheral arterial disease behaves like
An additional perspective on the prior data showing limited if coronary disease. However, improvements in peripheral arterial dis-
any cardiovascular benefit through older glucose-lowering treat- ease endpoints have been difficult to demonstrate, even with potent
ments is that HbA1c represents a flawed surrogate marker for cardiovascular risk-reducing agents such as statins. Patients treated
identifying risk of future macrovascular outcomes responsive to with pioglitazone had significantly lower risk for the prespecified
therapy, contrasting with the ability of HbA1c to track efficacy secondary endpoint in this trial, which was more consistent with
in predicting microvascular responses to glucose-lowering treat- the standard primary cardiovascular trial endpoint of combined all-
ment. Given that HbA1c is used to define the disease of diabetes, cause mortality, nonfatal myocardial infarction, and nonfatal stroke
enrolling patients based on HbA1c may not consistently enroll (HR: 0.84; 95% CI: 0.80–1.02; p = 0.027).
those more likely to benefit in terms of macrovascular outcomes Two issues have complicated the use of TZDs. A meta-analysis
in response to the treatment under study. of rosiglitazone effects on cardiovascular outcomes examining short-
Overall, many studies provide strong epidemiologic data show- term trials of glycemic control suggested rosiglitazone was associ-
ing that diabetes is associated with a large increase in cardiovascular ated with an increased risk of myocardial infarction. Total event
risk. In contrast to studies of microvascular disease, where the asso- rates were low in the included studies and they did not examine car-
ciated risk is high per unit change in HbA1c, the associated cardio- diovascular events as primary endpoints nor were the cardiovascular
vascular risk per unit change in HbA1c is lower, once corrected for events adjudicated.746 In contrast, other meta-analyses, including
triglycerides, LDL cholesterol, hypertension, and smoking. ones using only randomized clinical trials with prespecified cardio-
While the benefit of glucose lowering with metformin to vascular endpoints and/or adjudication, found no increase in car-
reduce myocardial infarction was shown in the substudy of the diovascular events or cardiovascular death with either rosiglitazone
UKPDS,741 as previously mentioned, little data was available to or pioglitazone.747 This conclusion was also supported by other
support other therapeutic approaches or specific agents. subsequent studies and an FDA consensus review panel.748 Differ-
ences do exist between pioglitazone and rosiglitazone. A second fac-
tor limiting use of these agents has been their known side effects,
Studies Using Insulin-Sensitizer Medications including fluid retention that may increase heart failure (although
The BARI 2D trial randomized 2368 patients with T2DM and without myocardial changes or increased mortality), increased bone
coronary heart disease to either prompt revascularization or inten- fractures, significant increases in weight among a subset of patients,
sive medical therapy alone and to either insulin sensitization or and increases in bladder cancer.749 Whether these untoward effects
insulin provision diabetes therapy. Randomization was stratified are offset by clinical cardiovascular benefit may require further con-
by the proposed revascularization method. The 5-year survival sideration, especially given more recent positive clinical trial data
and major cardiovascular event rates were similar in all study discussed later in the chapter.
subgroups except for those patients undergoing coronary artery Prior inability to conclusively demonstrate cardiovascular
bypass grafting who had fewer major cardiovascular events after event reduction with specific diabetes approaches or therapies
revascularization. There was less hypoglycemia and weight gain could have been due to not having the right agent or targeting the
and greater apparent benefit from coronary artery bypass grafting most impactful mechanism rather than other explanations such as
in the insulin sensitization group. patient characteristics or study design.750,751
To evaluate insulin versus other therapies for lowering glucose As a result of the 2008 FDA-instituted guidance that new
as a strategy to decrease cardiovascular events, in the Outcome antidiabetic drugs must provide cardiovascular safety data in their
Reduction With Initial Glargine Intervention (ORIGIN) trial, approval process, a large amount of new data regarding cardiovas-
insulin or noninsulin therapies were used in patients with either cular effects of newer antidiabetic agents has emerged and con-
prediabetes or early T2DM to achieve a fasting plasma glucose of tinues to do so, including new data showing cardiovascular risk
95 mg/dL versus 123 mg/dL (5.3 mmol/L vs 6.8 mmol/L) with reduction for patients treated with GLP1RA or SGLT2 inhibi-
other therapies.743 No difference in cardiovascular outcomes was tors. These results have focused thinking on specific agents and
evident in the more aggressively treated arm. potential mechanisms that are likely more complex than simply
Thiazolidinediones are direct insulin sensitizers, improving lowering blood glucose concentration.
insulin sensitivity rather than increasing insulin levels. TZDs bind
to and activate the nuclear receptors known as peroxisomal pro- Studies Using Newer Antidiabetic Medications: A
liferator-activated receptor gamma (PPARγ) and to mitoNEET, New Era?
a mitochondrial protein-regulating energy metabolism.744 TZD
effects in vascular and inflammatory cells limit atherosclerosis and Dipeptidyl Peptidase 4 Inhibitors and Cardiovascular
inflammatory responses. In the PROactive study (i.e., Prospec- Disease
tive Pioglitazone Clinical Trial in Macrovascular Events), 5238 DPP4i lower glucose by inhibiting degradation of the incre-
patients with T2DM and evidence of macrovascular disease were tin GLP1. The SAVOR–Thrombolysis in Myocardial Infarction
randomized to either pioglitazone or usual care with the objective (TIMI) 53 (SAVOR-TIMI 53) trial established safety for this

agent with no difference in major adverse cardiovascular events further questioning regarding mechanisms explaining these results.
(MACE) (the composite of nonfatal myocardial infarction, non- Nevertheless, the findings stand. The fact that two different agents
fatal stroke, or cardiovascular death); however, a 27% increase in within each drug class have shown a positive cardiovascular pri-
hospital admissions for heart failure was seen.752 The Examination mary endpoint reinforces the general validity of the data. Fur-
of Cardiovascular Outcomes With Alogliptin Versus Standard of thermore, both GLP1RAs and SGLT2 inhibitors improved renal
Care (EXAMINE) similarly established cardiovascular safety for outcomes (see “Glycemic Control in Diabetic Kidney Disease”).
the MACE outcome with this DPP4i.753 Although EXAMINE GLP1RAs have GLP1 receptor-dependent and receptor-inde-
investigators reported that alogliptin had no impact on heart fail- pendent effects on the cardiovascular system.758,759 GLP1RAs can
ure,754 differences in the design and endpoint measurements in induce satiety, promote weight loss, lower blood pressure, and
this trial as compared to SAVOR-TIMI 53 may limit this conclu- have other effects in preclinical and smaller translational studies
sion.755 In the Trial to Evaluate Cardiovascular Outcomes After that might promote cardiovascular benefit.760 In the LEADER
Treatment With Sitagliptin (TECOS), sitagliptin showed no trial543 (Table 37.19), 9340 patients with T2DM and increased
cardiovascular benefit but also no adverse effects, including no cardiovascular risk received random assignment to either lira-
increased admissions for heart failure, all as compared to usual glutide (titrated to 1.8 mg once daily) or placebo, with median
care. Other approaches to the issue of DPP4i and heart failure, follow-up of 3.8 years. The primary time-to-event analysis, which
like observational studies and meta-analyses, have varied in con- was the combined results of first occurrence of cardiovascular
clusion about whether specific agents differ in heart failure risk. death, nonfatal myocardial infarction, or nonfatal stroke, was
The FDA has included heart failure warnings in DPP4i labeling. significantly decreased among liraglutide (overall incidence 13%)
While none of these agents increase mortality, it is prudent to versus placebo group (14.9%, p < 0.001 for noninferiority; p =
examine DPP4i-treated patients with diabetes for signs and symp- 0.01 for superiority).543
toms of heart failure. Notably, prior to these outcome studies, less Semaglutide, a GLP1RA administered once a week, was stud-
rigorous forms of data had suggested cardiovascular benefits with ied in the SUSTAIN 6 trial, which enrolled 3297 patients with
DPP4i, underscoring the primacy of randomized, prospective, T2D, of whom the majority had established cardiovascular dis-
blinded clinical trial data. ease, chronic kidney disease, or both. Participants received either
In the CARMELINA study,756,756a the effect of linagliptin (5 semaglutide (0.5 or 1 mg) or placebo once weekly for 2 years. The
mg once daily) was compared to placebo (both added to standard primary composite outcome, which was the first occurrence of the
of care) on cardiovascular outcomes in 6979 adults with T2DM same 3-point major adverse cardiovascular event, was significantly
and high cardiovascular risk, the majority of whom also had kid- reduced in those on semaglutide (p < 0.001 for noninferiority),
ney disease. Linagliptin demonstrated a similar cardiovascular with nonsignificantly decreased nonfatal myocardial infarction
safety profile compared to placebo when added to standard of care. and significantly decreased nonfatal stroke (see Table 37.19).544
There was a similar kidney safety profile and no increase in heart In further support of GLP1RAs having cardiovascular benefit, the
failure compared to placebo.756,756a The companion Cardiovascu- Harmony Outcomes trial demonstrated that 9463 patients with
lar Outcome Study of Linagliptin Versus Glimepiride (CARO- T2DM and cardiovascular disease receiving albiglutide 30 to 50
LINA)757 is comparing linagliptin to the sulfonylurea glimepiride mg based on glycemia and tolerability had a 22% reduction in
in patients with T2DM and higher cardiovascular risk (known cardiovascular events compared to those receiving placebo in con-
cardiovascular disease or T2DM-associated end-organ damage or junction with usual care761 over a median follow-up of 1.5 years.
older age or two or more cardiovascular risk factors). In contrast, other GLP1RAs have shown cardiovascular safety
but no benefit in outcome trials. In the Evaluation of Lixisenatide
GLP1RAs and Cardiovascular Disease in Acute Coronary Syndrome (ELIXA) trial, 6068 patients with
In contrast to the DPP4 inhibitor data, studies with both GLP1RAs T2DM with recent (<180 days) myocardial infarction or unsta-
and SGLT2 inhibitors provided landmark data showing decreased ble angina hospitalization received lixisenatide 10 to 20 μg once
cardiovascular events with glucose-lowering therapy. Not surpris- daily or placebo762 with usual care with median follow-up of 25
ingly, these positive findings have prompted intense scrutiny and months. The primary composite endpoint of cardiovascular death,

TABLE 37.19 K
ey Findings From GLP1 Receptor Agonist Cardiovascular Outcome Trials
Primary CV Hospitalization
Sample Follow-Up Composite Myocardial Nonfatal All-Cause for Heart
Trial GLP1RA Size Years Outcomea Infarction Stroke CV Mortality Mortality Failure
LEADER Liraglutide 9340 3.5 0.87 ↓14% p = 0.046 ↓11% ↓22% ↓15% ↓13%
(0.78-0.97); p = 0.30 p = 0.007 p = 0.02 p = 0.14
p = 0.01 NS NS
SUSTAIN-6 Semaglutide 3297 2.0 0.74 (0.58-0.95); ↓15% ↓39% → → →
p = 0.02 p = 0.38 p = 0.04 p = 0.92 p = 0.79 p = 0.57
NS NS NS NS

Results are shown as hazard ratio or odds ratio with (95% confidence intervals), or as percent change.
Full trial names and references are given in the text.
aCV death, myocardial infarction, or stroke.
CV, Cardiovascular; GLP1RAs, glucagon-like peptide-1 receptor agonists.

myocardial infarction, stroke, or unstable angina hospitalization that most participants did not have a heart failure history, this
showed no superiority, while also establishing safety given non- effect highlights the extent to which subclinical heart failure as
inferiority versus placebo. The Exenatide Study of Cardiovascular well as risk for heart failure is a major issue in T2DM while also
Event Lowering (EXSCEL) trial, with extended-release exenatide suggesting new directions for SGLT2 inhibitor use now being
also showed safety but no benefit when studied in 14,752 patients investigated. The rapid separation of curves in terms of mortality
with T2DM treated with either extended-release exenatide 2 mg and heart failure, both occurring after approximately 6 months,
weekly or placebo over 3.2 years; no difference was seen versus was also noteworthy.
placebo on the primary composite outcome of first occurrence of Canagliflozin, another SGLT2 inhibitor, was studied in the
cardiovascular death, nonfatal myocardial infarction, or nonfatal CANVAS473 program (see Table 37.20), which involved two car-
stroke (HR: 0.91; 95% CI: 0.83–1.00; p < 0.001 for noninferior- diovascular trials, including CANVAS-R that focused on patients
ity; p = 0.06 for superiority).763 with renal disease. These studies involved 10,142 subjects with
The basis for positive cardiovascular outcomes with liraglutide, T2DM and either a history of cardiovascular disease (approxi-
semaglutide, and albiglutide but not lixisenatide or extended- mately two-thirds) or just an additional cardiovascular risk factor
release exenatide remains obscure. Differences between clinical (one-third). After a mean of 3.6 years (median 2.4 years), the pri-
trials such as enrollment criteria, subtle differences in patient pop- mary major adverse cardiovascular endpoint (myocardial infarc-
ulations, study locations, trial duration, and concomitant medi- tion, stroke, cardiovascular mortality) was significantly reduced.
cation use may be underlying factors. It remains possible some Although each of these components trended toward benefit, none
members of the same drug class may have more potent effects on of them were significant on their own, including cardiovascular
discrete cardiovascular risk parameters, such as blood pressure mortality or total mortality, in contrast to the EMPA-REG OUT-
or weight loss. Given the prior literature, differences in glucose COME trial. A signal for an increased risk of amputations was
control itself seem unlikely to explain variable outcome benefits seen in CANVAS, without clear evidence for similar issues in the
within drug classes or between older and newer agents. Pharma- EMPA-REG OUTCOME trial. Other analyses have also raised
cologic differences among the agents is difficult to exclude (e.g., concerns regarding possible increased risk for amputations with
short-acting vs long-acting activity with long-acting exenatide canagliflozin. The extent to which this finding is valid involves
nearly reaching statistical significance). other SGLT2 inhibitors, and the potential underlying mechanism
Initial concerns about increased risk of pancreatitis, pancreatic for this putative issue remains unresolved and awaiting further
cancer, and thyroid cancer with GLP1RAs have not been borne data, although the significant reduction in cardiovascular events
out in clinical trials, although a prior history of such issues is is evident. In Dapagliflozin Effect on Cardiovascular Events
viewed as a potential contraindication to use. Given that GLP1 (DECLARE-TIMI 58), the effects of dapagliflozin in 17,150
provides a gut-derived satiety signal to the brain, the gastrointes- patients with inadequately controlled diabetes and either known
tinal side effects seen with these drugs, mainly nausea, are not cardiovascular disease (secondary prevention cohort) or at least two
surprising, may contribute to beneficial weight loss, and are not risk factors for cardiovascular disease (primary prevention cohort)
a barrier for most patients. GLP1RAs do not pose concerns for was studied. In announced but not yet published topline results,
patients with chronic kidney disease, within the limits of the dapagliflozin demonstrated cardiovascular safety and a reduction
populations enrolled. A decrease in progression of chronic kidney in a composite of cardiovascular death or hospitalization for heart
disease has been noted. For some patients, the injectable nature failure but without a significant difference in the major adverse
of GLP1RAs may be an issue. These agents on their own do not composite outcome of combined cardiovascular death, nonfatal
cause hypoglycemia, although caution may be needed when com- myocardial infarction, or ischemic stroke.
bined with insulin provisional diabetes treatments (e.g., sulfonyl- Other large cardiovascular outcome trials are being done with
ureas or insulin). other SGLT2 inhibitors. Ertugliflozin is also being investigated in
a large outcome cardiovascular trial. Future studies in other specific
SGLT2 Inhibitors and Cardiovascular Disease populations will provide additional insight into SGLT2 inhibi-
By blocking glucose absorption in the proximal tubule, SGLT2 tor effects. Other real-world data analyses, performed outside of
inhibitors lower glucose levels, while also wasting an energy sub- a randomized clinical trial setting, have supported cardiovascular
strate and having a diuretic effect, with resulting weight loss and benefits with SGLT2 inhibitors. For example, CVD-REAL766 (see
blood pressure lowering.764 Table 37.20), which reviewed medical claims, hospital records,
In the EMPA-REG OUTCOME trial765 (Table 37.20), and registries from 309,056 patients in six countries taking SGLT2
7020 patients with T2DM and established cardiovascular dis- inhibitors, found SGLT2 inhibitor use was associated with a 39%
ease received empagliflozin, 10 mg or 25 mg, or placebo once lower risk of heart failure hospitalization (p < 0.001) and a 51%
daily in addition to usual care. After a median duration of 3.1 reduction in total death (p < 0.001) independent of country.
years, the primary major cardiovascular event composite outcome The mechanism for cardiovascular benefits with SGLT2 inhibi-
among those receiving empagliflozin was significantly decreased. tors remains to be defined. The multitude of prior negative trials
Especially impressive was that this positive outcome was driven with glucose-lowering argues against the modest improvement in
primarily by a major decrease in death, which was seen in both HbA1c as underlying the observed benefit. Reductions in blood
cardiovascular (38%) and all-cause mortality (32%). Hospitaliza- pressure, weight and volume, with concomitant increases in hema-
tion for heart failure was significantly reduced in those on empa- tocrit as well as unique aspects to tissue-based volume shifts, have
gliflozin (−35%, p = 0.002), as evident in both those with and been hypothesized to contribute to the decrease in cardiovascular
without a history of heart failure, as possibly expected for an agent risk. However, the relative small changes in these parameters and
with diuretic effects. An antihypertensive effect was also evident. the lack of similar effects in trials with diuretics or other antihy-
However, a similar magnitude of reduction in hospitalization for pertensives suggest the involvement of other factors. Serum uric
heart failure has not been seen in clinical trials of diuretics or other acid levels, previously linked to cardiovascular risk, are decreased
antihypertensive drugs, making this explanation less likely. Given with SGLT2 inhibition, but the rapidity with which the curves

TABLE 37.20 K
ey Findings From SGLT2 Inhibitor Cardiovascular Outcome and Observational Studies
SGLT2 Inhibitor SGLT2i vs History Median Primary CV Myocardial Hospitalization
Daily Dose vs Placebo or of CVD Follow-Up Composite Infarction (Fatal Stroke (Fatal or All-Cause for Heart
Studies Comparator Comparator, N Patients, % Years Outcomea or Nonfatal) Nonfatal) CV Mortality Mortality Failure
Randomized Controlled Trials vs Placebo
EMPA-REG Empagliflozin 10 4687 vs 2333 99 3.1 0.86 (0.74-0.99; 0.87 (0.70-1.09); 1.18 (0.89-1.56); 0.62 (0.49-0.77); 0.68 (0.57-0.82); 0.65 (0.50-0.85);
OUTCOME1 or 25 mg vs p = 0.04 p = 0.23 p = 0.26 p < 0.001 p < 0.001 p = 0.002
placebo
CANVAS2 Canagliflozin 5795 vs 4347 65 2.4 0.86 (0.75-0.97); 0.85 (0.69-1.05); 0.90 (0.71-1.15); 0.87 (0.72-1.06); 0.87 (0.74-1.01); 0.67 (0.52-0.87)b
100–300 mg p = 0.02 NS NS NS NS
vs placebo
Observational Studies vs Active Comparator
CVD-REAL3 SGLT2i (dapa- 154528 vs 13 0.6–0.7 NA 0.85 (0.72-1.00); 0.83 (0.71-0.97); NA 0.49 (0.41-0.57); 0.61 (0.51-0.73);
gliflozin 42%, 154528 p = 0.05 p = 0.02 p < 0.001 p < 0.001
canagliflozin (propensity

Chapter 37 Complications of Diabetes Mellitus

53%) matching)

Results are shown as hazard ratio or odds ratio (95% confidence intervals). p values are provided as available.
aCardiovascular death, nonfatal myocardial infarction, or nonfatal strokes.
bHospitalization for heart failure was prespecified exploratory cardiovascular outcome.
NA, Not available; NS, nonsignificant; SGLT2, sodium glucose cotransporter type 2.
Data from 1Zinman B, Wanner C, Lachin JM, et al. Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes. N Engl J Med. 2015;373:2117–2128; 2Neal B, Perkovic V, Mahaffey KW, et al. Canagliflozin and cardiovascular and renal events in type
2 diabetes. N Engl J Med. 2017;377:644–657; 3Kosiborod M, Cavender MA, Fu AZ, et al. Lower risk of heart failure and death in patients initiated on sodium-glucose cotransporter-2 inhibitors versus other glucose-lowering drugs: the CVD-REAL study (comparative
effectiveness of cardiovascular outcomes in new users of sodium-glucose cotransporter-2 inhibitors). Circulation. 2017;136:249–259.

1511
1512 SE C T I O N V I I I Disorders of Carbohydrate and Fat Metabolism

separate in the SGLT2 trials make this explanation less likely. leaves unresolved the question of changing therapies for patients
Other proposed mechanisms include more efficient fuel switching who have reasonable glycemic control but use antidiabetic drugs
in patients with T2DM and ischemia, through increased plasma with no proven cardiovascular benefit. Separately, should cardiolo-
β-hydroxybutyrate as a better energy substrate, changes in dys- gists prescribe glucose-lowering agents, and how should previously
regulated neurohormonal signaling, and alterations in sodium- encountered barriers to cardiologist engagement with diabetes
hydrogen exchange, among others. treatment be overcome? Greater insight into the mechanisms
Untoward effects with SGLT2 inhibition align in part with its underlying cardiovascular benefit from these newer agents may
known mechanism of action. Induced glucosuria can promote yield further refinement in their use, while also advancing con-
myocotic genital infections, which can be offset by good hygienic tinued drug development. Finally, the positive clinical trial data
practices, with consideration of avoiding these agents in patients in T2DM patients with established cardiovascular disease or high
with a history of these issues. Increased diabetic ketoacidosis and cardiovascular risk have prompted further consideration of how
euglycemic ketoacidosis have been reported with SGLT2 inhibi- the benefits seen with these therapies might be extended to other
tors although not seen in meta-analyses or the randomized trials indications and patient groups, particularly T2DM patients with-
discussed earlier. The question of increased amputation risk with out current cardiovascular disease. Clinical trials of these agents in
canagliflozin was noted previously. Hypoglycemia is not an issue people with T2DM and hypertension or heart failure are already
with this class of drugs, unless added to a regimen that includes under study. These agents, which do not cause hypoglycemia, may
insulin provisional therapies. be especially appealing for reducing cardiovascular risk in those
The positive cardiovascular outcome trials with both GLP1RAs with prediabetes.
and SGLT2 inhibitors (Fig. 37.49) raise fundamental questions
regarding optimal T2DM management: How should these data Characteristic Features and Treatment of
be incorporated into guidelines, including the choice between
GLP1RAs and SGLT2 inhibitors? Most guidelines have adopted Dyslipidemia in Diabetic Patients
an approach of recommending these agents in ways that align with Dyslipidemia has been well characterized as a risk factor that
the trials, for use mainly in patients with inadequately controlled increases atherosclerosis in patients with T2DM. Various pat-
HbA1c and established or high risk for cardiovascular disease. This terns of dyslipidemia are often encountered in diabetes and may

Differences in CV risk reduction within

4 CVOTs demonstrating MACE superiority

GLP1RAs SGLT2i

LEADERa Favors treatment Favors placebo Favors treatment Favors placebo EMPA-REGc
MACE
CV death
Nonfatal stroke
Nonfatal MI

0.3 2.0 0.3 2.0

Hazard ratio (95% CI) Hazard ratio (95% CI)

SUSTAIN 6b
Favors treatment Favors placebo Favors treatment Favors placebo CANVASd

MACE
CV death
Nonfatal stroke
Nonfatal MI
0.3 2.0
0.3 2.0 Hazard ratio (95% CI)
Hazard ratio (95% CI)
• Fig. 37.49 Components of positive primary endpoint responses in landmark glucagon-like peptide 1
receptor agonist (GLP1RA) and sodium-glucose cotransporter-2 (SGLT2) inhibitor trials. Data shown for
GLP1RAs (LEADER/liraglutide, SUSTAIN 6/semaglutide) and SGLT2 inhibitor (EMPA-REG/empagliflozin,
CANVAS/canafliglozin) cardiovascular outcome trials (CVOT). CI, Confidence interval; MACE, major
adverse cardiovascular events; MI, myocardial infarction. Full study names are noted in text. aMarso SP,
Daniels GH, Brown-Frandsen K, et al. Liraglutide and cardiovascular outcomes in type 2 diabetes. N
Engl J Med. 2016;375:311–322; bMarso SP, Bain SC, Consoli A, et al. Semaglutide and cardiovascular
outcomes in patients with type 2 diabetes. N Engl J Med. 2016;375:1834–1844; cZinman B, Wanner C,
Lachin JM, et al. Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes. N Engl J Med.
2015;373:2117–2128; dNeal B, Perkovic V, Mahaffey KW, et al. Canagliflozin and cardiovascular and renal
events in type 2 diabetes. N Engl J Med. 2017;377:644–657.

increase atherogenic risk (also see Chapter 41). Although patients Several studies have further validated the use of statins in
with diabetes tend not to have marked elevations in plasma LDL patients with diabetes and related conditions.567,771–774 In the
cholesterol levels, their LDL cholesterol levels remain predictive Heart Protection Study, a large (N = 20,536), randomized, pla-
of cardiovascular risk. LDL particles in diabetes are smaller and cebo-controlled trial of the use of simvastatin 40 mg in high-risk
more dense than normal LDL. These small, dense triglyceride- patients, roughly 29% of the study participants had T2DM.775
rich LDL particles are especially pathologic. Small dense LDL par- Over the 5-year course of the study, treatment with simvastatin
ticles are more susceptible to oxidation, particularly in the setting resulted in a significant reduction in the occurrence of major vas-
of poor glucose control. Other evidence suggests that glycation cular events in patients with T2DM who had established coronary
of LDL may be enhanced in diabetes, impairing recognition of heart disease (33.4% vs 37.8% in simvastatin-treated and placebo-
the lipoprotein by its hepatic receptor and extending its half-life. treated patients, respectively), in T2DM with no prior coronary
Conversely, HDL cholesterol levels are decreased in patients with heart disease (13.8% vs 18.6%), and in both categories combined
T2D.767 (20.2% vs 25.1%). Overall, the study also demonstrated a highly
A key feature of diabetic dyslipidemia is an increase in hepatic significant 12% relative risk reduction in all-cause mortality and
very low-density lipoprotein (VLDL) production in response to an 18% reduction in coronary mortality among all subjects treated
elevations in FFA flux often seen in diabetes with insulin-resis- with simvastatin.
tant adipocytes. Although insulin and lipases such as lipoprotein The Collaborative Atorvastatin Diabetes Study (CARDS) had
lipase mediate the uptake of triglyceride-derived FFAs by striated a fundamental impact on the management of diabetes because of
muscle, reducing the levels presented to the liver, insulin resistance the magnitude of beneficial effects seen in the statin-treated group.
results in the opposite effect, increasing the levels of FFAs pre- CARDS was a randomized, placebo-controlled trial that tested
sented to the liver. The metabolic syndrome, with its characteristic the effect of starting low dose atorvastatin (10 mg/day) in T2DM
abdominal obesity and insulin resistance, also increases FFA deliv- patients with no prior history of heart disease, with plasma LDL
ery to the liver. In addition, reduced lipoprotein lipase activity in levels lower than 160 mg/dL (<4.1 mmol/L) (N = 2838) on pre-
T2DM leads to an accumulation of triglyceride-rich lipoproteins venting acute coronary heart events, coronary revascularization,
in the plasma of these patients and may also result in decreased or stroke.574 The trial was terminated 2 years early because the
physiologic delivery of lipolytically derived, biologically active prespecified efficacy criteria were met. After a median of 3.9 years
molecules to cells. Such models align with genetic evidence in the follow-up, patients treated with atorvastatin had a relative risk
general population. Gain-of-function variants with increased lipo- reduction for first cardiovascular event of 37% (95% CI: 52–17%
protein lipase activity, as well as those with genetic loss of function reduction; p = 0.001), compared with placebo-treated patients.
of inhibitors of the lipoprotein lipase pathway, appear to have less Assessed separately, acute coronary heart disease events, coronary
cardiovascular risk. Triglyceride-rich lipoproteins also play a role revascularizations, and stroke were significantly reduced, by 36%,
in the reduced levels of HDL cholesterol by increasing the transfer 31%, and 48%, respectively. The CARDS trial showed that in
of cholesterol from these particles. patients with T2DM, a threshold LDL cholesterol level should
A number of landmark trials have established that lowering not be the sole determinant of whether a statin is prescribed. Of
LDL cholesterol levels produces major clinical benefits in terms of note, stroke was reduced in CARDS, addressing prior questions
reducing cardiovascular events in patients with and without a his- about whether LDL lowering with statins might or might not
tory of cardiovascular disease at baseline. These findings have now reduce stroke to the same extent as they lower coronary heart dis-
been extended to the population of subjects with T2DM and dys- ease. Currently, it is generally accepted that statins are effective in
lipidemia. Although LDL levels are often within the average range reducing cerebrovascular disease.776 As a result of CARDS and
in patients with diabetes, treatment with 3-hydroxy-3-methylglu- other studies discussed, most if not all guidelines have endorsed
taryl coenzyme A (HMG-CoA) reductase inhibitors (statins) has statin use in most appropriately aged individuals with diabetes.
consistently shown improved outcomes among both those with The Treating to New Targets (TNT) study compared the
and without diabetes, including those with and without a prior effects of atorvastatin 10 mg or 80 mg daily for a median follow-
history of heart disease. Statins are now incorporated into cur- up period of 4.9 years in patients with clinically evident coro-
rent diabetes management guidelines, discussed further later in nary heart disease who also met the NCEP criteria for diagnosis
the chapter. of the metabolic syndrome.772 The study included 778 patients
In the Cholesterol and Recurrent Events (CARE) trial, T2DM with T2DM, who constituted 22% of the study population.
patients with a prior cardiovascular event receiving pravastatin had Treatment with atorvastatin 80 mg reduced major cardiovascular
a 25% reduction in the incidence of cardiovascular death, nonfatal events 29% more than atorvastatin 10 mg, presumably due to the
myocardial infarction, coronary artery bypass graft, and revascu- greater reduction in LDL cholesterol seen with higher dosing of
larization procedures compared to placebo.768 In the Long-Term atorvastatin.
Intervention With Pravastatin in Ischemic Disease (LIPID) study, Pooling of data from the multiple statin trials further sup-
patients with diabetes had a 19% reduction in major coronary ports statin benefits in those with diabetes. In the Cholesterol
heart disease events (fatal cardiovascular disease and nonfatal myo- Treatment Trialists’ (CTT) Collaboration,777 analysis of 18,646
cardial infarction).769 In a post hoc subgroup analysis of secondary patients with diabetes in 14 different studies revealed a 9% pro-
prevention in a large cohort of patients with diabetes, impaired portional reduction in all-cause mortality per mmol/L reduction
glucose tolerance, or normal glucose tolerance, simvastatin nor- in LDL cholesterol (RR: 0.91, 99% CI: 0.82–1.01; p = 0.02) as
malized associated elevations in total cholesterol and triglycerides compared to a similar 13% reduction in those without diabe-
across the range of glucose values.770 Treatment also significantly tes (RR: 0.87, 99% CI: 0.82–0.92; p < 0.0001).778 A clinically
reduced major coronary events and revascularizations in patients and statistically significant 21% proportional reduction in major
with diabetes and reduced major coronary events, revasculariza- vascular events per mmol/L reduction in LDL cholesterol was
tions, and total and coronary mortality in patients with impaired observed in people with diabetes (RR: 0.79, 99% CI: 0.72–0.86;
glucose tolerance. p < 0.0001).

In 2013, revised cholesterol treatment guidelines were pro- reduced LDL levels by 59% (92 mg/dL to 30 mg/dL [2.4 mmol/L
vided by the American College of Cardiology (ACC) and the to 0.78 mmol/L]) and was associated with a statistically significant
American Heart Association (AHA).779 These recommendations reduction in cardiovascular events of 15%.784 The ODYSSEY trial
depart from prior guidelines by focusing initiation of lipid-low- with alirocumab (Praluent) in patients with recent acute coronary
ering therapy on four clinically defined patient groups who war- syndrome also showed a statistically significant reduction in car-
ranted either high-intensity, moderate-intensity, or low-intensity diovascular events, as presented but not yet published.785 Patients
statin intervention while also abandoning use of LDL targets as with diabetes are receiving attention in terms of PCSK9 inhibitor
a means of guiding therapy. Patients with diabetes were one of use as a function of their increased risk for cardiovascular events.
these four treatment groups, as supported by the data from pri- Post hoc studies on outcomes in those with diabetes was a pre-
mary and secondary prevention trials noted previously. A con- specified analysis in FOURIER and demonstrated similar benefit
troversial aspect of these guidelines was their abandonment of and no adverse issues among these patients (40%) compared to
specific LDL targets. Both TNT and the subsequent Improved those who did not have diabetes at baseline (60%). In ODYS-
Reduction of Outcomes: Vytorin Efficacy International Trial SEY, 29% of subjects had diabetes, indicating that there should
(IMPROVE-IT), which showed that adding the nonstatin be forthcoming data about this group. Other studies indicate no
LDL-lowering agent ezetimibe to stable statin therapy further issues with alirocumab use in those with diabetes, with presenta-
decreased cardiovascular events in patients with coronary heart tions suggesting an augmented impact on event reduction in acute
disease,780,781 support the clinical utility of using specific LDL coronary syndrome patients. An ongoing scientific question is evi-
targets. The 2013 guidelines might be best viewed as a scientific dence suggesting PCSK9 loss of function may be associated with
statement given their precise focus on what had been proven in increased risk of hyperglycemia and perhaps diabetes,786 although
clinical trials, which explains the specific age cutoff points for no increased signal for new diabetes has been seen in the clinical
statin treatment of patients with diabetes. trials published to date.
Clinical trials have also revealed statins are associated with an Outcomes of treatment with other nonstatin therapies
increase in new-onset diabetes among a small subset of patients have been variable. With niacin, data have been disappoint-
at high risk for development of T2DM. In the Justification for ing, with no additional benefit seen when added to statins in
the Use of Statins in Prevention: An Intervention Trial Evaluating several trials.773,787,788 Although the lower HDL levels found in
Rosuvastatin (JUPITER) study, which investigated rosuvastatin patients with diabetes might make niacin an appealing treat-
20 mg versus placebo in 17,603 patients with no history of ath- ment option, niacin’s lack of proven efficacy combined with its
erosclerotic events, baseline mean LDL levels of 108 mg/dL, and side effects of increasing insulin resistance and relatively poor
an elevated C-reactive protein of 4.2 mg/L, physician-reported tolerability have limited its use. Recent genetic studies using
new-onset diabetes was found in 270 patients on rosuvastatin Mendelian randomization have demonstrated that variants
versus 216 on placebo.782 In the JUPITER population 41% of associated with higher HDL levels often do not protect against
participants had metabolic syndrome at study entry. New-onset CV events.789 Current attention is focused on the concept that
diabetes was found in 28% in patients with one or more risk fac- the key element in the inverse relationship between HDL and
tors for developing diabetes, who at the same time benefited by cardiovascular disease is HDL functionality, which may not be
a 39% reduction in cardiovascular events and a 17% reduction reflected in total HDL levels.790
in total mortality. Among those with no risk factors for diabe- Fibric acid derivatives, which lower high triglycerides and
tes, there was no increase in new diagnoses of diabetes.783 Thus raise HDL, have also been proposed as a treatment in people
the modest increase in diabetes in response to statin treatment is with diabetes given the usual nature of diabetic dyslipidemia. In
markedly offset by major decreases in cardiovascular events. For the VA High-Density Lipoprotein Cholesterol Intervention Trial
patients in whom statin treatment is appropriate, concerns over (VA-HIT), men given gemfibrozil had a significant decrease in
increased conversion to diabetes should not be a factor in the deci- coronary events and strokes in the absence of statin use.774 Post
sion to use statins. While increases in new diabetes has been seen hoc analyses suggesting these benefits were driven by the diabetes
with multiple statins, the mechanism underlying this statin effect subgroup focused further attention on fibrates. The FIELD study
remains obscure. assessed the effect of long-term fenofibrate therapy on cardiovas-
Additional data are now available regarding LDL cholesterol cular events in patients with T2DM.567 Patients were random-
levels and cardiovascular risk associated with using additional new ized to receive either micronized fenofibrate 200 mg per day (n =
LDL-lowering therapies. In population studies, individuals with 4895) or placebo (n = 4900). During 5-year follow-up, 5.9% of
lifelong low LDL cholesterol levels were found to have genetic loss the placebo-treated patients versus 5.2% of the fenofibrate-treated
of function in the Proprotein Convertase Subtilisin/Kexin Type patients had a major coronary event (coronary heart disease death
9 (PCSK9) gene, which was associated with decreased atheroscle- or nonfatal myocardial infarction), the primary endpoint of the
rosis and cardiovascular events but no other untoward effects. trial, a difference that was not statistically significant. Fenofibrate
Further supporting PCSK9 as a drug target, genetic variants with therapy significantly reduced some secondary endpoints, includ-
increased PCSK9 function resulted in a lipid and clinical picture ing total cardiovascular events (HR: 0.89, 95% CI: 0.75–1.05,
that replicated familial hypercholesterolemia (FH). PCSK9 action p = 0.035), coronary revascularization, progression of albumin-
targets the LDL receptor for proteosomal degradation rather than uria, and the need for laser treatment of retinopathy. Other sec-
allowing it to return to the cell surface for another cycle of LDL ondary endpoints such as cardiovascular mortality were higher,
clearance. In an impressive glimpse into future drug development although not statistically significant, with fenofibrate. Statistical
prospects, in less than 10 years, PCSK9 inhibitors in the form significance for the primary endpoint of the study might have
of monoclonal antibodies were being tested for their clinical effi- been missed because a greater percentage of patients in the placebo
cacy in cardiovascular event reduction. In FOURIER, adding the group began statin therapy during the study period compared to
PCSK9 inhibitor evolocumab (Repatha) to statin therapy versus the fenofibrate group, thus modulating their risk and potentially
statin alone in 27,546 patients with established coronary disease masking the treatment effect.

The ACCORD trial addressed the use of fibrates in combina- dL (<1.81 mg/dL) in higher risk patients even when the high-risk
tion with statins in patients with T2DM. This trial investigated patient had a baseline LDL cholesterol level lower than 100 mg/
whether combination therapy with simvastatin plus fenofibrate, dL (<2.57 mmol/L).795 The most recent ACC/AHA cholesterol
compared with simvastatin alone, would reduce cardiovascular guidelines took a completely different approach. The guidelines
events in diabetic patients at high risk for cardiovascular disease. abandon specific LDL targets, instead identifying four specific
In this randomized clinical trial comparing fenofibrate to placebo patient groups in whom clinical trial data supported statin use.
in statin-treated patients, combination therapy did not reduce the Since clinical trials had not specifically targeted more-intensive
rate of cardiovascular mortality, nonfatal myocardial infarction, or versus less-intensive LDL goals with a single statin, the recom-
nonfatal stroke, although other secondary parameters such as first mendations also focused on the appropriate statin dose intensity
myocardial infarction were improved. In a prespecified subgroup for given patient groups. The presence of diabetes was one of these
with a high triglyceride ratio (triglycerides ≥204 mg/dL [≥5.3 four patient groups, with the recommendation that designated
mmol/L] and HDL ≤34 mg/dL [≤0.88 mmol/L]), fenofibrate did patients with diabetes (age 40–75 years), as defined by subjects
reduce the primary outcome as compared to the placebo group. A enrolled in trials, receive a high-intensity statin (atorvastatin 40 or
pattern has emerged in fibrate trials of greater benefit in those with 80 mg, rosuvastatin 20 or 40 mg) if their 10-year calculated risk
significantly elevated triglycerides and lower HDL—the popula- was greater than or equal to 7.5%, and a moderate-intensity statin
tion in which fibrates would be commonly used. At this time, no (these same statins at their other lower doses and remaining other
formal recommendation exists that patients with significant dys- statins at their mid to highest doses) if calculated 10-year risk was
lipidemia receive combination statin-fibrate therapy for further less than 7.5%. This approach has been controversial, especially
risk reduction on the basis of T2DM alone.791 In those with a the abandonment of target LDL levels. However, data from mul-
significant history of CVD, elevated triglycerides, and low HDL, tiple clinical trials, including those with ezetimibe, supported
expert opinion might endorse fibrates as an adjunct to statin ther- benefits from lower LDL levels. The subsequent PCSK9 inhibitor
apy once LDL levels are appropriately controlled as a reasonable data has further strengthened evidence for additional cardiovas-
if not definitively established option. Importantly, improved gly- cular benefit from achieving lower LDL levels as mentioned in
cemic control and decreased insulin resistance, whether through further guideline updates and will likely be incorporated in an
drug therapy or improved lifestyle interventions and modest upcoming AHA/ACC guideline release.
weight loss, can improve diabetic dyslipidemia. Topline results
from the Reduction of Cardiovascular Events With EPA—Inter- Characteristic Features and Treatment of
vention Trial (REDUCE-IT),792 using a specific eicosapentaenoic
acid only from fish oil in patients with established cardiovascular Hypertension in Diabetic Patients
disease or diabetes mellitus and at least one other cardiovascu- Approximately 50% of patients with newly diagnosed diabetes
lar risk factor indicate a 25% reduction in the primary endpoint also have hypertension. As with dyslipidemia, hypertension inter-
composite of the first occurrence of major adverse cardiovascu- acts with diabetes to amplify the risk of cardiac mortality (see Fig.
lar events, including cardiovascular death, nonfatal myocardial 37.47). Although the cause of hypertension is multifactorial, the
infarction, nonfatal stroke, coronary revascularization, or unstable insulin-resistant state is one factor postulated to predispose patients
angina requiring hospitalization. to develop hypertension. In addition to its negative effects on the
Genetic variants linked to low triglyceride levels, like apoli- cardiovascular system, high blood pressure is a key contributor to
poprotein CIII (apoCIII) loss of function, have been reported to the development of microvascular disease in diabetes. Based on the
predict future cardiovascular events.793 ApoCIII loss of function guidelines of the Joint National Committee on Prevention, Detec-
reduces plasma triglyceride, and the atherogenicity of chylomi- tion, Evaluation, and Treatment of High Blood Pressure (JNC 7),
crons, VLDL, LDL, and HDL particles. These findings provide a blood pressure should be reduced to less than 130/85 mm Hg in
rationale for pursuing ApoCIII inhibition and genetic knockdown patients with diabetes.547 A longer duration of diabetes has been
as a therapeutic strategy, which continues to be pursued despite linked to increased arterial stiffness, decreased vasomotor function,
initial FDA rejection of volanesorsen, an apoCIII antisense ther- and changes in pulse pressure, which can contribute to patient
apy, out of concerns for thrombocytopenia.794 symptoms in response to blood pressure treatment. Patients with
Sources for cholesterol guidelines have transitioned from the diabetes may also lose the usual nocturnal blood pressure dip.
prior NCEP Adult Treatment Panel recommendations, the last Results of the most recent clinical trials underscore the benefits
being the NCEP ATPIII, to joint guidance from the American of aggressive treatment of hypertension in patients with diabetes,
College of Cardiology and the American Heart Association. Con- although achieving mean blood pressure reductions to currently
sidering both the ATPIII recommendations and the most recent recommended targets is challenging both in clinical trials and in
ACC/AHA approaches is valuable and provides additional per- real-world settings. This may also help explain disconcordant trial
spective. The importance of dyslipidemia as a contributor to car- results. Data have continued to support achieving a systolic blood
diovascular risk in patients with diabetes was incorporated in the pressure between 120 and 140 mm Hg in those with diabetes,
NCEP ATPIII guidelines, which identified diabetes as being equal while additional benefit is less clear with systolic blood pressure
in cardiovascular risk to having had prior coronary heart events.726 less than 120 mm Hg.796
According to the NCEP ATPIII guidelines, patients with T2DM Use of a long-acting dihydropyridine calcium channel blocker
would receive cholesterol-lowering therapy if the LDL cholesterol in the Systolic Hypertension in Europe (Syst-Eur) study resulted
level were higher than 130 mg/dL (>3.36 mmol/L), with the in substantial reductions in rates of total mortality (55%), cardio-
goal of reducing LDL cholesterol to less than 100 mg/dL (<2.57 vascular mortality (76%), and cardiovascular events (69%) in the
mmol/L).726 In practice, more aggressive approaches were often diabetic subgroup, greater benefits than were seen in the subgroup
used, instituting drug treatment if LDL cholesterol level were without diabetes. In the Heart Outcomes Prevention Evaluation
higher than 100 mg/dL (>2.57 mmol/L). Subsequent guideline (HOPE) study, in which almost 40% of patients had diabetes and
updates endorsed an LDL cholesterol goal of less than 70 mg/ one other cardiovascular risk factor, ramipril reduced the primary

outcome by 24% and total mortality risk by 25%.797 Even in nor- A major unresolved question has been whether treatment of
motensive patients with diabetes, some benefit was seen with a hypertension to lower targets than currently recommended would
drop in blood pressure (2–4 mm Hg) with ACE inhibitor therapy, further reduce cardiovascular risk in patients with type 2 diabetes.
raising questions about what mechanisms were responsible for the Most trials establishing the remarkable benefit of blood pressure
benefit seen with even modest blood pressure changes. Other rig- lowering, regardless of the medications used, have studied people
orously designed studies, such as the UKPDS798 and the Hyper- whose systolic blood pressure was higher than 140 mm Hg. In
tension Optimal Treatment (HOT) study,799 suggested that tight the ACCORD trial, the effect of lowering systolic blood pressure
blood pressure control in patients with diabetes caused cardiovas- to two different levels on cardiovascular risk was assessed.802 One
cular benefits of even greater magnitude. In the ADVANCE trial, group was randomized to a systolic blood pressure lower than
over 11,000 type 2 diabetes patients received either a fixed-dose 120 mm Hg (intensive therapy); the other group was treated to
combination of perindopril and indapamide or placebo along a systolic pressure lower than 140 mm Hg (standard therapy).
with usual care.800 Those on perindopril/indapamide had on aver- Targeting the lower blood pressure level did not reduce the rate
age 5.6–mm Hg lower systolic blood pressures and 2.2–mm Hg of fatal and nonfatal cardiovascular events. Based on this result,
diastolic blood pressures, with a 9% decrease in the relative risk there is no recommendation to treat patients with T2DM and
of a major macrovascular or microvascular event as compared to hypertension to a systolic blood pressure lower than the currently
16.8% for those on placebo, with a hazard ratio 0.91. The mac- recommended target of 130 mm Hg to decrease cardiovascular
rovascular and microvascular events responded similarly but not events. However, further blood pressure reduction may reduce the
independently, highlighting the pathogenic role of hypertension incidence of diabetic nephropathy.548,802 In contrast, in the Sys-
in both large and small vessel disease. tolic Blood Pressure Intervention Trial (SPRINT), which did not
In the Losartan Intervention For Endpoint Reduction in enroll patients with diabetes, more aggressive blood pressure low-
Hypertension (LIFE) study, patients with diabetes, hypertension, ering was beneficial,803 prompting subsequent discussions about
and signs of left ventricular hypertrophy were randomly assigned potential discrepancies between SPRINT and these other studies,
to treatment with losartan-based (n = 586) or atenolol-based including differences in hypertension management in the presence
(n = 609) treatment for hypertension.800a Despite similar blood or absence of diabetes.804
pressure reductions, losartan was more effective than atenolol for The HOT trial also studied intensive versus standard control
reducing rates of cardiovascular morbidity and mortality while in 18,790 participants, including 1501 subjects who had diabe-
also demonstrating fewer conversions to new diabetes. The ability tes, with a focus on enrollment and targeted blood pressure based
of losartan to reduce events more effectively than atenolol may be on diastolic measurements. While more intensive blood pressure
related to the ability of angiotensin receptor blockers to reverse left control did not make a difference in the primary cardiovascular
ventricular hypertrophy more effectively than beta blockers. outcome in the study as a whole, the subgroup with diabetes expe-
Although beta blockers are thought to worsen glycemic control rienced a 51% decrease in events, with optimal cardiovascular risk
in patients with diabetes, it is not clear whether this is a property reduction found in those achieving a diastolic blood pressure of
of all members of this drug class or whether this property persists 82.6 mm Hg and systolic blood pressure of 138.5 mm Hg.799
if beta blockers are given in combination with renin-angiotensin
system inhibitors that are known to increase insulin sensitivity. In Acute Coronary Syndromes in Diabetes Mellitus
the Glycemic Effects in Diabetes Mellitus: Carvedilol-Metoprolol
Comparison in Hypertensives (GENINI) trial, patients with doc- The case fatality rate from myocardial infarction is almost twice
umented T2DM and hypertension who were taking a stable dose as high in patients with diabetes as in nondiabetic patients. This
of either an ARB or an ACE inhibitor were randomized to receive excess risk is seen both during the acute phase and in the early and
either carvedilol, a nonselective beta blocker, which also blocks late postinfarction periods. A number of mechanisms have been
α1-adrenergic receptors, or metoprolol, a β1-selective adrenergic hypothesized to be responsible for worse outcomes in patients
blocker. Although the degree of blood pressure control was similar with diabetes, including the following:
with both beta blockers, HbA1c and insulin resistance increased • Increased risk of heart failure due to maladaptive remodeling of
significantly with metoprolol but not with carvedilol. Therefore the left ventricle805–807
carvedilol appears not to cause the adverse effects that metopro- • Increased risk of sudden death due to sympathovagal imbal-
lol does on glucose levels when used in combination with RAS ance as a consequence of autonomic neuropathy808–810
inhibitors, although this conclusion needs to be confirmed in a • Increased likelihood of early reinfarction due to impaired fibri-
longer term outcome trial. nolysis811–813
The investigators of the Antihypertensive and Lipid-Lowering • Extensive underlying atherosclerosis814,815
Treatment to Prevent Heart Attack Trial (ALLHAT) compared • Changes in myocardial cell metabolism, including a shift from
outcomes during first-step treatment of hypertension in 31,512 glucose oxidation to FFA oxidation, with less generation of
patients with T2DM, impaired fasting glucose (IFG) levels, or ATP at any level of oxygen consumption816,817
normoglycemia with a calcium channel blocker (amlodipine 2.5– • Associated cardiomyopathy818
10 mg/day) or ACE inhibitor (lisinopril 10–40 mg/day) compared Collective data provide strong evidence that a variety of treat-
with a thiazide-type diuretic (chlorthalidone 12.5–25 mg/day).801 ment modalities can improve outcomes from myocardial infarc-
There was no significant difference in the occurrence of the pri- tion in patients with diabetes. In terms of interventions, patients
mary outcome (fatal cardiovascular disease or nonfatal myocardial with diabetes experiencing an acute infarct respond as favorably
infarction) in patients with T2DM treated with a calcium chan- to fibrinolytic therapy as do nondiabetic patients.814,815 Glycemic
nel blocker or an ACE inhibitor compared with chlorthalidone. control is an essential component of overall management. Glucose
Patients with impaired fasting glucose treated with a calcium levels at hospital admission have been independently correlated
channel blocker had a significantly higher relative risk for the pri- with early and late fatality after myocardial infarct in patients with
mary outcome than patients receiving chlorthalidone. and without diabetes mellitus.820–823

Studies such as the Diabetes and Insulin-Glucose Infusion remodeling. In addition to these hemodynamic benefits, ACE
in Acute Myocardial Infarction (DIGAMI) study have assessed inhibitors can also improve outcomes in diabetes by improving
the impact of intensive glycemic control in patients with diabe- endothelial function,831 improving fibrinolysis,832 and decreasing
tes during the acute phase of myocardial infarction. Patients in insulin resistance.833
this study were randomized to either intensive insulin therapy In a retrospective analysis of the GISSI-3 (Gruppo Italiano per
(insulin-glucose infusion for 24 hours, followed by subcutaneous lo Studio della Sopravvivenza nell’Infarto Miocardico) study,834
insulin injection for 3 months) or standard glycemic control.824 lisinopril administration within 24 hours after hospital admis-
The intensive insulin regimen lowered blood glucose level dur- sion for acute myocardial infarction reduced both 6-week and
ing the first hour after admission and at discharge compared with 6-month mortality rates more in patients with diabetes than in
conventional therapy. The 1-year mortality rate was reduced with patients without diabetes.
the insulin infusion group compared with the control group, a dif- Similarly, in the Trandolapril Cardiac Evaluation Study
ference that was maintained after 3.4 years of follow-up. (TRACE), which investigated trandolapril use in patients with
Although the mechanisms responsible for the potential ben- left ventricular dysfunction after prior anterior myocardial infarct,
efit shown in the original DIGAMI study are not entirely clear, a subgroup analysis of diabetes patients revealed trandolapril treat-
experimental data suggest that strict glycemic control can improve ment greatly improved outcomes, with a 36% decrease in mortal-
myocardial cell metabolism by increasing the availability of glu- ity from any cause and a 62% reduction in progression to severe
cose as a substrate for ATP generation and reducing the formation heart failure as compared to placebo in those with diabetes, ben-
of FFAs, thereby shifting cardiac metabolism from FFA oxidation efits that were more pronounced than the effects seen in TRACE
to glycolysis and glucose oxidation. Intensive glycemic control subjects without diabetes (18% reduction in mortality, no effect
can also reverse the impaired fibrinolysis that is typically seen in on progression to severe CHF).834a
patients with diabetes. Beta blockers are now widely accepted for the treatment of
However, DIGAMI 2, a prospective, randomized, open-label acute coronary syndrome in patients with diabetes. Older, noncar-
trial that followed up the DIGAMI trial, comparing outcomes in dioselective beta blockers might have adversely affected the lipid
patients with either T1DM or T2DM, failed to corroborate the profile and inhibited the metabolic response to hypoglycemia, but
earlier reported improvement in outcomes with intensive insulin more recent data with cardioselective beta blockers suggest that
treatment.825 The lack of effect of long-term insulin treatment on these agents have less negative effect on metabolic indices, perhaps
outcomes may have been perhaps partially influenced by protocol because they increase peripheral blood flow and improve glucose
violations, including the patient group assigned to standard meta- delivery.835,836 Clinical trial data confirm that beta blockers reduce
bolic management, without insulin or glucose infusions, having the rates of mortality and reinfarction in patients with myocar-
such interventions, with 14% receiving insulin-glucose infusions dial infarction in the presence of diabetes. In fact, the magnitude
and as many as 41% receiving extra insulin injections. As a result, of their effects in patients with diabetes appear to exceed those
the blood glucose levels in all three groups were not significantly seen in nondiabetic patients. A large review of data from more
different after treatment. than 45,000 patients, 26% of whom had diabetes, showed beta
Results from the CREATE-ECLA (Clinical Trial of Revipa- blocker therapy was associated with a lower 1-year mortality rate
rin and Metabolic Modulation in Acute Myocardial Infarction in patients with diabetes than in those without diabetes, with no
Treatment Evaluation—Estudios Cardiologicos Latin America), evidence of an increase in diabetes-related complications.837
a randomized, controlled trial of 20,201 patients who presented Postulated mechanisms for the benefit of beta blockers in
with ST-segment elevation myocardial infarction within 12 hours patients with diabetes include dampening of the sympathetic ner-
after onset of symptoms, are consistent with DIGAMI 2 findings. vous system overactivity that arises as a consequence of autonomic
Patients were randomized to either high-dose glucose-insulin- neuropathy. Beta blockers can also reduce FFA levels and thereby
potassium (GIK) infusion (i.e., 25% glucose, 50 U/L regular reduce myocardial oxygen requirements. Carvedilol, although not
insulin, and 80 mEq KCl) administered over 24 hours or to usual a cardioselective beta blocker, has unique α1-adrenergic recep-
care.826 Roughly 18% of the patients in both treatment arms had tor blocking properties along with purported antioxidant effects.
T2DM. After 30 days, there were no differences in the rate of These actions may underlie the decreased insulin resistance and
occurrence of mortality, cardiac arrest, cardiogenic shock, or rein- greater blood pressure effects of this agent, which may be of par-
farction in the two treatment groups. ticular benefit in patients with type 2 diabetes.838
Sulfonylureas have been implicated as increasing cardiovascular Aspirin has been a cornerstone of therapy for the primary or
mortality rate, particularly in patients undergoing revasculariza- secondary prevention of acute coronary syndrome in patients
tion for acute myocardial infarction.827 The UKPDS did not show with T1DM and T2DM who do not have contraindications to its
a deleterious effect of these agents on the incidence of sudden use. Aspirin significantly lowers the risk of myocardial infarction
death or myocardial infarction379 or over 10 years of follow-up.24 without increasing the risk of vitreous or retinal bleeding, even
The sulfonylureas act through the sulfonylurea receptor compo- in patients with retinopathy.839 Use of aspirin therapy (75–162
nent of ATP-sensitive potassium channels in the pancreatic beta mg/day) as a secondary prevention strategy in those with dia-
cell. In the heart, ATP-sensitive potassium channels are involved betes and a history of atherosclerotic cardiovascular is currently
in ischemic preconditioning and coronary vasodilation.828–830 It recommended by the ADA,840 with presumed benefits through
is not clear whether the sulfonylureas modulate these channels in modulation of the enhanced platelet aggregation seen with both
the heart or vascular system or whether they significantly increase T1DM and T2DM.841 Despite this, controversy has arisen about
risk in patients with diabetes who are experiencing an acute myo- the ideal dose of aspirin, if enteric coating might interfere with
cardial infarct. aspirin’s effects,842 and more recently, if aspirin benefits in pri-
ACE inhibitors dramatically reduce the mortality rate after mary prevention are offset by the risk of bleeding. The Aspirin
a myocardial infarction in patients with diabetes, ostensibly in Reducing Events in the Elderly (ASPREE) trial studied aspirin
through their effects to reduce infarct size and limit ventricular 100 mg versus placebo in 19,114 subjects 70 years of age or older

with no known cardiovascular disease, including about 10% of Results of the BARI study showed that coronary artery bypass
participants who had diabetes, and found no benefit on cardio- grafting provides better outcomes than percutaneous transluminal
vascular outcomes, including the diabetes subgroup, but a signifi- coronary angioplasty in patients with diabetes, possibly because
cant increase in bleeding.843 In contrast, in ASCEND, in which it addresses the extensive coronary vascular disease in these
15,480 participants with diabetes received either aspirin 100 mg patients.850 This study did not employ stents or glycoprotein IIb/
or placebo for a mean of 7.4 years, the aspirin group had fewer IIIa inhibitors, two modalities that, when used together, appear
serious vascular events (658 participants [8.5%] vs 743 [9.6%]; to improve outcomes after percutaneous transluminal coronary
RR: 0.88, 95% CI: 0.79–0.97, p = 0.01) than placebo, but at angioplasty in patients with diabetes.
the expense of more major gastrointestinal or other extracranial In terms of revascularization strategies, the Future Revasculariza-
bleeding (4.1% on aspirin group vs 3.2% on placebo group, RR: tion Evaluation in Patients With Diabetes Mellitus: Optimal Man-
1.29, p = 0.003).844 Aspirin use in patients with diabetes must be agement of Multivessel Disease (FREEDOM) trial had a major
tailored to their particular overall cardiovascular and bleeding risk. impact on management decisions. This trial was an appropriately
Antiplatelet therapy with the antagonist of the P2Y12 subclass powered, randomized comparison of percutaneous coronary inter-
of ADP receptor clopidogrel also benefits patients with diabetes. ventions (with drug-eluting stents) and coronary artery bypass graft
The Clopidogrel Versus Aspirin in Patients at Risk of Ischaemic (using arterial grafting) in patients with diabetes and multivessel
Events (CAPRIE) trial compared outcomes in patients with non– coronary artery disease on concomitant optimal medical therapy. It
ST-segment elevation myocardial infarction treated with aspirin provided strong evidence that that coronary bypass resulted in fewer
or with clopidogrel and included 3866 patients with diabetes.845 major cardiovascular events, and reduced 5-year mortality compared
Although the event rate was higher among the diabetic patients to percutaneous coronary interventions.851 Continued progress in
than in the overall study population, the response to treatment stent technology and adjunctive therapies, as noted, continues to
was also better. The event rate for the primary endpoint (vascular make this a challenging decision. Patients with diabetes with less
death, ischemic stroke, myocardial infarction, or rehospitalization extensive coronary disease for whom invasive intervention is war-
for ischemia or bleeding) was 17.7% for diabetic patients treated ranted can certainly be managed with percutaneous interventions.
with aspirin and 15.6% for those randomized to clopidogrel, a
significant RR of 12.5%. Newer antiplatelet therapies are also Cardiomyopathy in Patients With Diabetes
now available, including P2Y12 receptor antagonists with greater
potency. In the Trial to Assess Improvement in Therapeutic Out- Mellitus
comes by Optimizing Platelet Inhibition With Prasugrel-Throm- Diabetes is associated with a fourfold increase in the risk of conges-
bolysis In Myocardial Infarction 38 (TRITON-TIMI 38) study, tive heart failure, even after adjustment for other cardiovascular risk
participants with diabetes had a 30% relative risk reduction in factors such as age, blood pressure, cholesterol level, obesity, and
the primary endpoint (cardiovascular death, nonfatal myocardial history of coronary heart disease.852 Contributing factors include
infarct, or nonfatal stroke) without evidence for increases in major diabetes-induced cardiomyocyte dysfunction, impaired microvas-
bleeding.846 In the Platelet Inhibition and Patient Outcomes cular perfusion because of defective endothelial function, increased
(PLATO) trial, ticagrelor equally reduced the rate of ischemic collagen deposition with fibrosis, and maladaptive remodeling after
events in acute coronary syndrome patients both with and with- myocardial infarction, leading to both diastolic and systolic heart
out diabetes; in the diabetes subgroup, reductions were seen in the failure (also discussed in “Biochemistry and Molecular Cell Biol-
primary composite endpoint (18%), all-cause mortality (18%), ogy”). Current nomenclature853 organizes heart failure into two
and stent thrombosis (35%) with no increase in major bleeding. broad categories: with preserved ejection fraction (HFpEF) and
Patients with diabetes are felt to have an increased risk for com- with reduction ejection fraction (HFrEF).854 Patients with diabetes,
plications with percutaneous interventions like stent thrombo- who commonly present with either form, also experience higher
sis, and may warrant more intensive oral antiplatelet therapy in rates of heart failure than do nondiabetic patients after an acute
appropriate patients. myocardial infarction, regardless of the size of the infarct zone.806,855
Newer adjunct therapies, such as the platelet glycoprotein IIb/ These findings suggest that diabetes itself causes deleterious effects
IIIa receptor antagonists that antagonize platelet action, have on the myocardium, leading to poorer outcomes.
also been assessed in diabetic patients who present with unstable A number of key structural, functional, and metabolic factors
angina or non–Q-wave infarction. Overall, these agents appear to in diabetes have been implicated in the increased risk of maladap-
work equally well, or perhaps slightly better, in patients with dia- tive remodeling that leads to heart failure. For example, evidence of
betes compared with nondiabetic patients. In the Platelet Recep- silent myocardial infarction is found in up to 40% of patients with
tor Inhibition in Ischemic Syndrome Management in Patients diabetes who present with a clinically apparent infarct and can lead
Limited by Unstable Signs and Symptoms (PRISM-PLUS) study, to unrecognized regional and global ventricular dysfunction.834,856
the addition of tirofiban to heparin therapy reduced the 7-day As many as 50% of patients with diabetes and coronary heart disease
composite endpoint, compared with heparin alone. This effect have cardiac autonomic neuropathy, which is known to contribute
was greater in patients with diabetes than in patients without to both systolic and diastolic dysfunction.857 Like hypertension,
diabetes.847 diabetes can cause fibrosis of the myocardium and increased col-
In one study of patients undergoing percutaneous transluminal lagen deposition.858,859 These effects are even more pronounced in
coronary angioplasty, glycoprotein IIb/IIIa antagonist therapy was patients with coexisting hypertension and diabetes and may con-
associated with fewer acute events but a higher rate of long-term tribute to the finding of diastolic dysfunction common in patients
target-vessel revascularization in the diabetic cohort compared with diabetes.860 Enhanced myocardial endothelial dysfunction in
with the nondiabetic cohort.848 However, in another trial, in diabetes has also been described as a pathophysiologic pathway to
which stents were used, the rate of target vessel revascularization impaired microvascular perfusion and ischemia.861,862
at 6 months was significantly decreased with the addition of a gly- On a cellular level, both hyperglycemia and insulin resistance
coprotein IIb/IIIa antagonist compared with placebo.849 have direct negative effects on myocardial metabolism. Depression

of myocardial GLUT4 levels in the setting of diabetes inhibits to 2.5% in Western countries: Higher incident rates are found
glucose entry and glycolysis in the heart. As a result, intracellu- in select diabetic populations, including US Medicare recipi-
lar metabolism shifts from glycolysis to FFA oxidation, thereby ents (6%), US veterans (5%), and the global diabetes popula-
reducing glycolytic ATP generation, a major source of energy tion (6.3%).870 Approximately 5% to 10% of diabetic patients
under anaerobic (i.e., ischemic) conditions.816 The production of have had past or present foot ulceration, and 1% have under-
oxygen free radicals can also be enhanced in this situation, fur- gone amputation.866 Although between 1990 and 2010, a sig-
ther depressing myocardial contractile function.862 Mitochondrial nificant reduction in certain diabetes-related complications was
dysfunction is a feature of ischemia-reperfusion injury, leading to reported from the United States,8 unfortunately there was no
generation of reactive oxygen species. In experimental models of reduction in rates of amputations reported in this time frame.
ischemia-reperfusion, mitochondrial reactive oxygen species gen- A large community-based study in the United Kingdom showed
eration is mediated by reversal of succinate dehydrogenase, lead- an annual incidence of ulceration of approximately 2%; this
ing to succinate accumulation. On reperfusion, restoration of rose to 7% with known diabetic neuropathy and to as high as
succinate dehydrogenase activity leads to rapid succinate oxida- 50% with a past history of ulceration.871 The lifetime risk for
tion, causing reactive oxygen species generation by reverse electron development of a foot ulcer in a diabetic patient is estimated to
transport at mitochondrial complex I.863 Increased myocardial be as high as 25%.868,872 Multiple factors likely contribute to
production of succinate is a feature of acute ST-segment–eleva- different rates of foot ulceration, including diabetes duration,
tion myocardial infarction in humans.864 smoking, and other neurologic and cardiovascular risk factors.
A major advance in the heart failure field in general was the find- Foot ulceration is most common in patients with a history of
ing in PARADIGM-HF that the combination of an angiotensin similar problems. In experienced diabetic foot clinics, more
receptor blocker (valsartan) and a neprilysin inhibitor (sacubitril) than 50% of patients with new foot ulcers give a past ulcer his-
proved superior to the ACE inhibitor enalapril in patients with tory. Patients with other complications of diabetes, including
reduced ejection fraction (≤40%), significantly reducing cardiovas- retinopathy and renal dysfunction, are at increased risk for foot
cular death or hospitalization for heart failure, and doing so against ulceration. Patients on dialysis are among those at the very high-
a medicine previously shown to have a mortality benefit.865 Pro- est risk for ulceration.872a
gression of heart failure and renal dysfunction were also decreased. More than 80% of amputations are preceded by foot ulcers.
Of note, among the nearly 45% of patients in this study who had Diabetes is the most frequent cause of nontraumatic lower limb
diabetes, there was evidence of greater decreases in HbA1c and less amputation in the United States, and rates are 15 times greater
progression to insulin use, raising intriguing mechanistic questions, than those in the nondiabetic population. Reduced amputation
while also potentially supporting the use of sacubitril/valsartan in rates occur following implementation of foot screening and edu-
patients with diabetes and reduced ejection fractions.592 cation programs.873,874 Implementation of strategies to reduce
Knowledge about the pathophysiology underlying increased ulceration and amputation are necessary and may benefit from
coronary artery disease in people with prediabetes and type 2 dia- additional education on behavioral aspects of diabetic foot care,
betes has increased enormously in recent years. That knowledge including awareness of ulceration risk, compliance with daily foot
has directed seminal clinical studies, which in turn have enlarged inspection, and the importance of nonweightbearing with early
our understanding of pathophysiology. These seminal studies foot infections.875 Treatments to reduce cardiovascular and renal
provide the rationale for an optimized multifactorial approach to risk likely further reduce amputation rates, but this has been less
the prevention and treatment of diabetic cardiovascular disease, clearly established.
including choice of drugs to manage hyperglycemia, dyslipidemia,
and hypertension, and postinfarction sequelae. Data from ongoing Pathogenesis of Foot Ulceration
clinical trials and new pharmaceutical developments will further
improve the cardiovascular prognosis for people with diabetes. Foot ulceration results from multiple interacting component
causes, none of which alone is sufficient to cause ulceration but
The Diabetic Foot when combined may lead to skin breakdown. A combination
of two or more risk factors promotes diabetic foot ulceration,
Foot complications of ulceration and amputation are associated and the triad of component causes—neuropathy, deformity,
with high mortality, yet diabetic foot ulcers are the most preventable and trauma—is present in 63% of incident ulcers. Edema and
late adverse outcome of diabetes.866–869 Elliott Joslin correctly noted ischemia are also common component causes. These risk fac-
in 1934 “diabetic gangrene is not heaven-sent, but earth-born.” The tors were demonstrated in a North American/United Kingdom
neuropathic foot does not spontaneously ulcerate; insensitivity, collaborative study of more than 150 consecutive foot ulcer
deformity, and unperceived trauma (e.g., inappropriate footwear) cases.876
promote skin breakdown. Most amputations are preceded by foot
ulcers. Increased appreciation of the complex pathogenesis of foot Diabetic Neuropathy and the Foot Ulcer
ulcers permits design of screening programs for risk identification, All three components of neuropathy—sensory, motor, and
preventive education, and design of multidisciplinary foot care pro- autonomic—can contribute to foot ulceration. The risk of foot
grams. Much progress has been made, but suboptimal adherence to ulceration in patients with neuropathy is sevenfold higher than
current professional guidelines has prevented translation of this into in those without this complication of diabetes.871 Chronic sen-
a universal decline in amputation rates. sorimotor neuropathy is common, affecting at least one-third of
older diabetic patients in Western countries. Its onset is gradual
and insidious, and symptoms may be so minimal that they go
Epidemiology of Diabetic Foot Ulceration unnoticed. Although uncomfortable, painful, and paresthetic
Foot ulceration is common and occurs in both T1DM and symptoms predominate in most patients, some never experience
T2DM. The annual incidence of foot ulceration is about 2% symptoms. Clinical examination usually reveals a sensory deficit

in a glove-and-stocking distribution, with signs of motor dys- Peripheral Vascular Disease, Diabetic Foot Ulcers,
function such as small muscle wasting in the feet, and absent and Amputation Risk
ankle reflexes. Although a history of typical symptoms strongly Peripheral vascular disease in isolation rarely causes ulceration.
suggests a diagnosis of neuropathy, absence of symptoms does However, the common combination of vascular disease with minor
not exclude the diagnosis and must never be equated with a lack trauma can lead to ulceration. Minor injury and subsequent infec-
of foot ulcer risk. Therefore assessment of foot ulcer risk must tion increase the demand for blood supply beyond the circulatory
always include a careful foot examination, including assessment capacity, and ischemic ulceration and risk of amputation develop.
of proprioception (i.e., pressure perception using a 10-g mono- Early identification of patients at risk for peripheral vascular dis-
filament) regardless of history. Loss of sensation and propriocep- ease is essential, and appropriate investigation involving noninvasive
tion reduce capillary perfusion and increase ischemia at sites of Doppler studies, together with arteriography, often leads to revascu-
increased pressure load. larization procedures to improve lower extremity blood flow. Pres-
Sympathetic autonomic neuropathy affecting the lower limbs ence or absence of a dorsalis pedis or posterior tibial pulse may be
results in reduced sweating, dry skin, and development of cracks the simplest and most reliable indicator of significant ischemia that
and fissures. In the absence of large-vessel arterial disease, there can be elicited at the bedside.878 However, Doppler-derived ankle
may be increased blood flow to the foot, with arteriovenous shunt- pressure can be misleadingly high in patients with long-standing
ing leading to the warm but at-risk foot. diabetes, so the use of the toe-brachial index together with Doppler
waveforms of distal arteries can be useful in deciding which patients
Callus, Deformity, and High Foot Pressures need further diagnostic intervention.879
Motor neuropathy, with imbalance of the flexor and extensor Distal bypass surgery or endovascular interventions have
muscles in the foot, commonly results in foot deformity, with good short-term but mixed long-term results in terms of limb
prominent metatarsal heads and clawing of the toes (Fig. 37.50). salvage.878,880 Advances in techniques and devices for both dis-
The combination of proprioceptive loss due to sensory neuropathy tal arterial bypass surgery and endovascular interventions have
and the prominence of metatarsal heads lead to increased pressures been made, and it is unclear which approach is preferred. A
and loads under the diabetic foot. Neuropathy is only one cause of recent systematic review was unable to conclude any particu-
high foot pressures. High pressures, together with dry skin, often lar method was superior to any other.881 However, a recent
result in the formation of callus under weightbearing areas of the study in all patients with diabetic foot ulcers demonstrates that
metatarsal heads. Presence of plantar callus is a highly significant prompt vascular evaluation and revascularization within 2 weeks
marker for foot ulcer risk. Conversely, removal of plantar callus is reduces the rate of amputations to the rate of nondiabetics with
associated with reduced foot pressures and thereby a reduced foot peripheral artery disease, supporting a change in clinical disease
ulcer risk.877 management.882

A B
• Fig. 37.50 The high-risk neuropathic foot. (A, B) Two lateral views of a patient with typical signs of a
high-risk neuropathic foot. Notice the small-muscle wasting, clawing of the toes, and marked prominence
of the metatarsal heads. At presentation with type 2 diabetes mellitus, this patient had severe neuropathy
with foot ulceration on both the right foot (shown here) and the left foot. (From Andersson DK, Svärdsudd
K. Long-term glycemic control relates to mortality in type II diabetes. Diabetes Care. 1995;18:1534–1543.)

Prevention of Foot Ulceration and Amputation The Diabetic Foot Care Team
Relatively simple interventions can reduce amputations by up to Patients identified as being at high risk for foot ulceration should be
80%.873,874 Therefore strategies for early identification of patients managed by a team of specialists with expertise in the diabetic foot.
at potential risk for ulceration are required, and education pro- The podiatrist usually takes responsibility for care of the skin and
grams that can be adapted for widespread application need to nails and, together with the specialist nurse or diabetes educator,
be developed. Because foot ulcers precede most amputations, provides foot care education. The orthotist, or shoe fitter, is invalu-
are among the most common causes of hospital admission for able to advise about and sometimes design footwear to protect high-
patients with diabetes, and account for much morbidity and even risk feet. Team members should work closely with the diabetologist
fatality, the widespread application of preventive foot care strate- and the vascular and orthopedic surgeons. Patients with risk factors
gies is urgently required. for ulceration require preventive foot care education, including self-
Patients with any type of diabetes require regular foot exami- foot care at home, frequent review by a foot surgeon or podiatrist,
nation for evidence of risk factors for ulceration, irrespective of good foot wear with orthotics when indicated, and avoidance of
disease duration. Neuropathy, vascular disease, and even foot potentially dangerous over-the-counter foot products.874,877
ulceration may be the presenting feature of T2DM, so there can be
no exception to the rule of screening (Fig. 37.51). At a minimum, Classification of Foot Ulcers
such screening should be carried out annually. Of all the long-term
complications of diabetes, foot problems and their risk factors are Many different classification systems for grading diabetic foot
probably the easiest to detect. No expensive equipment is required, ulcers have been reported.868,877 One developed by Wagner885
and feet can be examined for evidence of neuropathic and vascular (Table 37.22) is widely used and accepted. More recently, the
deficits in the office setting using simple equipment.871 University of Texas (UT) group developed an alternative classifica-
In 2008, a task force of the ADA published a report on the tion system that, in addition to ulcer depth (as in the Wagner sys-
examination components that should be included in the annual tem), accounts for presence or absence of infection and ischemia
Comprehensive Diabetic Foot Examination (CDFE) (Table (Table 37.23). A prospective study from 2001 compared these two
37.21).884 The most important message is for practitioners to have wound classification systems and concluded that the UT scheme
patients remove shoes and socks to examine feet for the presence is a better predictor of outcome than the older Wagner system.885
of callus, deformity, muscle wasting, and dry skin, all of which
are clearly visible on inspection. A simple neurologic examination Management of Diabetic Foot Ulcers
includes assessment of pressure perception using a 10-g mono-
filament. Sensory neuropathy with clinical evidence of peripheral Basic principles of wound healing apply equally to diabetic foot
arterial disease are the strongest predictors of future foot ulcer- ulcers as to wounds in any other site or condition. In general, a
ation. Absence of the ability to perceive pressure from a 10-g
monofilament, inability to perceive a vibrating 128-Hz tuning
fork over the hallux, and absent ankle reflexes all have been shown TABLE 37.21 Key Components of the Comprehensive
to be predictors of foot ulceration.868,869 Diabetic Foot Examination
Dermatologic
Skin status: color, thickness, dryness, cracking
Diabetic patient Sweating
Infection: check between toes for fungal infection
Ulceration
Risk factors for ulceration? Calluses/blistering: hemorrhage into callus?
• Peripheral neuropathy
• Peripheral vascular disease Musculoskeletal
• Foot deformity Deformity (e.g., claw toes, prominent metatarsal heads, Charcot joint)
• Edema Muscle wasting (guttering between metatarsals)
• Past ulcer history Assess whether shoes are appropriate for the feet (e.g., size, width)
• Other complications
Neurologic
Ability to perceive pressure from a 10-g monofilament plus one of the
following:
No risk factor Yes risk factor Vibration using 128-Hz tuning fork
Pinprick sensation
Ankle reflexes
Vibratory perception threshold
Provide general advice • Foot care education
• Nail care, hygiene • Regular podiatry Vascular
• Footwear • Possible special Foot pulses
• Podiatry footwear, hosiery, etc.
Ankle-brachial index, if indicated

Adapted from Boulton AJ, Armstrong DG, Albert SF, et al. Comprehensive foot examination and
Review risk status Much more frequent review, risk assessment: a report of the Task Force of the Foot Care Interest Group of the American
at least annually always inspecting feet Diabetes Association, with endorsement by the American Association of Clinical Endocrinolo-
gists. Diabetes Care. 2008;31:1679–1685.

• Fig. 37.51 Simple algorithm for risk screening in the diabetic foot.

diabetic foot ulcer has high likelihood of healing if the following treated with TCC before biopsy. These important observations
three conditions are satisfied: support that repetitive pressure on a neuropathic wound contrib
• Arterial inflow is adequate. utes to the chronicity of the wound, whereas pressure relief results
• Infection is treated appropriately. in a reparative phase.
• Pressure is removed from the wound and the immediate sur- The next most common error is inappropriate management
rounding area. of infection. Topical applications are usually unhelpful, and if
Although this approach appears simplistic, failure of diabetic clinical infection is present, then it must be treated urgently with
foot ulcers to heal is usually a result of failure to pay sufficient antibiotics. Most infections are polymicrobial, with gram-positive
attention to one or more contributing conditions, including cocci, gram-negative rods, and anaerobes, with or without multi-
pressure on the wound, infection, ischemia, and inadequate drug-resistant organisms (see later discussion).
debridement. Another common management error is failure to appreciate
The most common cause of nonhealing neuropathic foot ulcers ischemic symptoms that are atypical due to altered pain sensation
is failure to remove pressure from the wound and immediate sur- as a result of neuropathy. The most difficult ulcer to heal is the
rounding area. Patients who are advised not to put pressure over neuroischemic ulcer. Symptoms and even signs of ischemia may
an ulcer find it especially difficult to adhere to such advice when be altered in the diabetic state. Therefore appropriate noninvasive
peripheral sensation is lost or reduced. Pain results in protection of investigation and arteriography are indicated for patients with a
an injured area; the lack of pain permits pressure to be put directly nonhealing diabetic foot ulcer if there is any question about the
onto the ulcer and results in nonhealing. A patient with normal vascular status.
sensation and a foot wound will limp to avoid putting pressure on Inappropriate wound debridement contributes to slow healing
the wound because doing so is painful. A patient who walks on a or nonhealing of a diabetic foot ulcer. Appropriate debridement
plantar wound without limping must have neuropathy. and removal of all callus, dead, and macerated tissue is essential
The effect of pressure relief on the histopathologic features of local treatment of a diabetic foot ulcer, which results in more rapid
neuropathic ulcers was assessed in a randomized study886 with healing compared with inadequately debrided wounds.
biopsy either at presentation or after 20 days of off-loading in a The principles of management of neuropathic and neuroisch-
total-contact cast (TCC). Histologic features of chronic inflam- emic foot ulcers are considered in the following sections referenc-
mation, with mononuclear infiltration, cellular debris, and scarce ing both the UT and the Wagner grading systems.
evidence of angiogenesis or granulation, were seen in patients who
underwent biopsy at presentation; whereas granulation, neoangio- Neuropathic Foot Ulcer Without Osteomyelitis (Wagner
genesis, and a predominance of fibroblasts were seen in patients Grades 1, 2; University of Texas Grades 1a, 1b, 2a, 2b)
Provision of adequate pressure relief is the most important feature
in the management of neuropathic foot ulcers that occur under
TABLE 37.22 Wagner Diabetic Foot Ulcer Classification weightbearing areas such as the metatarsal heads and great toe.
System This is usually achieved by a TCC or a removable Scotch cast
Grade Description boot.868,877
The TCC has long been recognized as the gold standard for
0 No ulcer, but high-risk foot (e.g., deformity, callus, insensitivity) off-loading a foot wound. The TCC was associated with the short-
1 Superficial full-thickness ulcer
est healing time in a randomized, controlled trial conducted by
Anderson and colleagues comparing three off-loading techniques
2 Deeper ulcer, penetrating tendons, no bone involvement demonstrated.887 When any cast device is used, regular removal of
3 Deeper ulcer with bone involvement, osteitis the cast is essential, because regular debridement of the wound by
a specialist is essential, and any casting device could injure insensi-
4 Partial gangrene (e.g., toes, forefoot) tive skin, especially over bony prominences. As the TCC requires
5 Gangrene of whole foot a specially trained casting technician to apply it, and frequent
removal is required for wound assessment, recent research has
Modified from Oyibo S, Jude EB, Tarawneh I, et al. A comparison of two diabetic foot ulcer focused on alternative reusable devices that can only be removed
classification systems: the Wagner and the University of Texas wound classification systems.
in the specialist’s office.
Diabetes Care. 2001;24:84–88.
The removable cast walker (RCW) resulted in slower heal-
ing than the TCC, in the aforementioned trial by Armstrong

TABLE 37.23 U
niversity of Texas Wound Classification System
Stage Grade 0 Grade 1 Grade 2 Grade 3
A Preulcer or postulcer lesion; no skin break Superficial ulcer Deep ulcer to tendon or capsule Wound penetrating bone or joint
B + Infection + Infection + Infection + Infection
C + Ischemia + Ischemia + Ischemia + Ischemia
D + Infection and ischemia + Infection and ischemia + Infection and ischemia + Infection and ischemia

Modified from Armstrong DG, Lavery LA, Harkless LB. Validation of a diabetic wound classification system. The contribution of depth, infection, and ischemia to risk of amputation Diabetes Care.
1998;12:855–859.

et al,887 even though prior gait laboratory studies suggested they that are without cellulitis, spreading infection, or discharge can
are equally efficacious at off-loading. The reason for this dispar- occasionally be left to spontaneously auto-amputate. The presence
ity may be explained by the observation that although patients of more extensive gangrene requires urgent hospital admission;
were instructed to wear the RCW at all times, patients used these treatment of infection, often with multiple antibiotics; glycemic
devices for only 28% of all footsteps during a 24-hour period. To control, usually with intravenous insulin; and detailed vascular
overcome this limitation the RCW, which can be applied by any assessment. It is in this situation that the team approach is most
clinic personnel and does not require specialist training, can be important, with close collaboration among the diabetes specialist,
rendered irremovable by wrapping it in casting material. A con- the vascular surgeon, and the radiologist.
trolled trial showed the irremovable RCW was as effective at heal-
ing neuropathic foot wounds as the TCC.888 Charcot Neuroarthropathy
Theoretically, complete healing of all superficial and neuro-
pathic ulcers should be possible without the need for amputation. Charcot neuroarthropathy is a disabling condition affecting the
In the treatment of neuropathic ulcers with a good peripheral joints and bones of the foot in the general diabetic population. It
circulation, antibiotics are not indicated unless there are clear is much more frequent (10–13%) in high-risk diabetic patients.
clinical signs of infection, including prominent discharge, local Permissive features for development of this condition include the
erythema, and cellulitis. The presence of any infective features in presence of severe peripheral neuropathy and autonomic dysfunc-
Wagner grade 1 or 2 ulcers would warrant reclassification in the tion with increased blood flow to the foot. The peripheral circula-
UT system from 1a or 2a to 1b or 2b. In such cases, deep wound tion is usually intact. In the Western world, diabetes is the most
specimens should be taken and broad-spectrum oral antibiotic common cause of a Charcot foot, and increased awareness of this
treatment should be started with, for example, either an amoxi- condition can enable earlier diagnosis and treatment to prevent
cillin–clavulanic acid combination (Augmentin) or clindamy- severe deformity and disability.
cin. The antibiotic may need to be altered after sensitivity results The pathogenesis of the Charcot process remains poorly
become available.889 understood; however, the patient with peripheral insensitivity and
autonomic dysfunction, with increased blood flow reflecting the
Neuroischemic Ulcers (Wagner Grades 1, 2; University of autonomic dysfunction, to the foot is vulnerable to trauma, which
Texas Grades 1c, 1d) the patient may not recall. Repetitive trauma results in increased
The principles of management of neuroischemic Wagner grade 1 blood flow through the bone, increased osteoclastic activity, and
and 2 ulcers are similar to those for neuropathic ulcers, with the fol- bone remodeling. In some cases, patients walk on a fracture,
lowing important differences. TCCs are not usually recommended which leads to continuing bone and joint destruction in that area.
for management of neuroischemic ulcers, although removable casts Recent evidence suggests acute Charcot neuropathy may be trig-
and pneumatic cast boots (Aircast) may be used in cases without gered in the susceptible (i.e., neuropathic) individual by any event
infection. Antibiotic therapy is required for most neuroischemic that leads to localized inflammation in the affected foot, triggering
ulcers. Investigation of the circulation is indicated, including non- a vicious cycle in which there is increasing inflammation, increas-
invasive assessment and, if required, arteriography with appropriate ing expression of receptor activator of nuclear factor-κΒ ligand
subsequent surgical management or angioplasty.878 (RANKL), a member of the tumor necrosis factor superfamily,
and increasing bone breakdown.892 Targeting activation of the
Osteomyelitis (Wagner Grade 3; University of Texas RANKL/osteoprotegerin pathway might lead to novel, future
Grades 3b, 3d) treatments.
Wagner or UT grade 3 ulcers are deeper and involve underlying Charcot neuropathy can be difficult to distinguish from
bone, often with abscess formation. Osteomyelitis is a serious osteomyelitis or an inflammatory arthropathy.890 A unilateral
complication of foot ulceration and may be present in as many swollen, hot foot without an ulcer in a patient with neuropathy
as 50% of diabetic patients with moderate to severe foot infec- must be considered to be a Charcot foot until proved otherwise.
tions.868 If the physician can probe down to bone in a deep ulcer, Charcot arthropathy can be diagnosed in most patients by plain
the presence of osteomyelitis is strongly suggested. Plain radio- radiography and a high index of suspicion (Fig. 37.52). Radio-
graphs are indicated for any nonhealing foot ulcer and are use- graphs reveal bone and joint destruction, fragmentation, and
ful in the diagnosis of osteomyelitis in more than two-thirds of remodeling, although radiographic findings may be normal early
patients, although radiologic changes may be delayed. In difficult in disease. In such cases, three-phase bisphosphonate bone scans
cases, further investigation, such as magnetic resonance imaging show increased bone uptake, while 111In-labeled bone scans will
(MRI), bone scans, or an indium-111 (111In)–labeled white blood be negative in the absence of infection. In patients where osteo-
cell scan can be useful in diagnosing bone infection.890 myelitis is suspected, 18F-fluorodeoxyglucose–positron emission
Although osteomyelitis treatment is traditionally surgical and tomography–computed tomography (FDG-PET-CT) has the
involves resection of the infected bone, successful long-term treat- highest sensitivity for distinguishing Charcot arthropathy from
ment can occasionally be effective with antibiotics. Staphylococcus osteomyelitis.
aureus is the most common infective bacteria. Therefore agents After diagnosis, management of the acute phase involves
such as clindamycin (which penetrates bone well) or flucloxacillin immobilization, usually in a TCC.893 There is little evidence
are often used. Antibiotic therapy for 90 days was equally effica- to support use of any pharmacologic treatment in the manage-
cious when compared to local surgery for diabetic foot osteomy- ment of this condition. Although Charcot neuroarthropathy is
elitis in a recent randomized trial.891 rare, it should be suspected in any patient with unexplained
swelling and heat in a neuropathic foot. Early intervention with
Gangrene (Wagner Grades 4, 5) immobilization may halt progression that, in the untreated
The presence of gangrene is always a serious sign in the diabetic state, can lead to marked foot deformity and require local or
foot. However, localized areas of gangrene, especially in the toes, major amputations.

No signs
of trauma

Shortened
first ray (first
metatarsal bone)
Gapping
Pain, swelling,
redness in
the midfoot
Bony
fragmentation
Bony
consolidation,
osteosclerosis,
fusion of joints

Normal Charcot neuroarthopathy

• Fig. 37.52 Some of the key features of Charcot neuroarthropathy, an often overlooked complication of
diabetes. (Redrawn from Botek G, Anderson MA, Taylor R. Charcot neuroarthropathy: an often overlooked
complication of diabetes. Cleve Clin J Med. 2010;77:593–599.)

Achilles Tendon Lengthening Procedures

Adjunct Treatments for Foot Ulcers
Although some centers advocate and perform tendo-Achilles
Tissue-Engineered Skin and Platelet-Derived Growth lengthening as an adjunct treatment to prevent recurrent diabetic
Factors foot ulcers posthealing, there are no adequate data supporting use
Genetically derived growth factors and novel bioengineered skin of this procedure in the diabetic foot. Given the high risk of any
substitutes have been developed as adjunctive treatments for dia- foot surgery in the diabetic foot, the nonsurgical approach of pre-
betic foot ulcers, including Apligraf, a bilayered living human scribing footwear designed to reduce pressure on the metatarsal
skin equivalent; Dermagraft, a human fibroblast-derived dermal heads is preferable unless and until adequately powered prospec-
substitute; and human platelet-derived growth factors. Although tive randomized clinical trials of Achilles tendon lengthening pro-
there are some benefits for each of these agents, they are costly894 cedures demonstrate superior outcomes.
and should be reserved for ulcers that fail to respond to standard
treatments. Any newer treatments should be considered an addi- SGLT2 Inhibitors and Diabetic Foot Disease
tion to good wound care, which must always include adequate An increased incidence of distal lower extremity amputations,
off-loading and regular debridement. There is little objective evi- mainly at the toe and metatarsal levels, was observed in the CAN-
dence to date in support of routine use of these therapies, as well VAS trial473 designed to evaluate the cardiovascular safety of cana-
as others under study, including stem cell therapy and hyperbaric gliflozin. Numerous potential mechanisms have been suggested,
oxygen.895 but the cause of this finding remains enigmatic. However, a recent
pharmacovigilance analysis confirmed that use of canagliflozin,
Negative-Pressure Wound Therapy (NPWT) but not dapagliflozin or empagliflozin, might be associated with
NPWT, also known as vacuum-assisted closure, is increas- an increased risk of amputations. The use of canagliflozin should
ingly used to treat large and complex diabetic foot wounds. The probably be avoided in patients with distal neuropathy or periph-
treatment may stimulate development of granulation tissue in eral arterial disease or with a history of foot disease. Although this
previously nonhealing wounds and can also be helpful in the post- does not appear to be a class effect,897 use of other SGLT2 inhibi-
operative management of diabetic foot wounds, as supported by tors should be used with caution in the abovementioned groups.
randomized, controlled trials in complex nonhealing diabetic foot
ulcers896 and postoperative cases.895 Use of NPWT remains con- References
troversial, as discussed in a systematic review. The strongest evi-
dence for efficacy is with postoperative diabetic foot wounds.895 The complete list of references is available online at ExpertConsult.com.

Descargado para Paola Gavidia Davila ([email protected]) en Antenor Orrego Private University de ClinicalKey.es por Elsevier en julio 19, 2020.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2020. Elsevier Inc. Todos los derechos reservados.
References 20. Ridker PM, MacFadyen JG, Glynn RJ, et al. Inhibition of interleu-
kin-1beta by canakinumab and cardiovascular outcomes in patients
1. Virmani R, Burke AP, Kolodgie F. Morphological characteristics with chronic kidney disease. J Am Coll Cardiol. 2018;71(21):2405–
of coronary atherosclerosis in diabetes mellitus. Can J Cardiol. 2414.
2006;22(suppl B):81b–84b. 21. Nathan DM, Bayless M, Cleary P, et al. Diabetes control and com-
2. Donahoe SM, Stewart GC, McCabe CH, et al. Diabetes and plications trial/epidemiology of diabetes interventions and com-
mortality following acute coronary syndromes. J Am Med Assoc. plications study at 30 years: advances and contributions. Diabetes.
2007;298(7):765–775. 2013;62(12):3976–3986.
3. Kempen JH, O’Colmain BJ, Leske MC, et al. The prevalence of 22. Diabetes Control and Complications Trial (DCCT)/Epidemiology
diabetic retinopathy among adults in the United States. Arch Oph- of Diabetes Interventions and Complications (EDIC) Research
thalmol. 2004;122(4):552–563. Group, Lachin JM, White NH, et al. Effect of intensive diabetes
4. Livingstone SJ, Levin D, Looker HC, et al. Estimated life expec- therapy on the progression of diabetic retinopathy in patients with
tancy in a Scottish cohort with type 1 diabetes, 2008-2010. J Am type 1 diabetes: 18 years of follow-up in the DCCT/EDIC. Diabe-
Med Assoc. 2015;313(1):37–44. tes. 2015;64(2):631–642.
5. Secrest A, Washington RE, Orchard TJ. Mortality in type 1 diabe- 23. Lachin JM, Orchard TJ, Nathan DM, DCCT/EDIC Research
tes. In: National Institutes of Health, ed. Diabetes in America. 3rd ed. Group. Update on cardiovascular outcomes at 30 years of the
Bethesda, MD: National Institutes of Health; 2018. diabetes control and complications trial/epidemiology of dia-
6. Rosenquist K, Fox CS. Mortality in type 2 diabetes. In: National betes interventions and complications study. Diabetes Care.
Institutes of Health, ed. Diabetes in America. 3rd ed. Bethesda, MD: 2014;37(1):39–43.
National Institutes of Health; 2018. 24. Holman RR, Paul SK, Bethel MA, Matthews DR, Neil HA. 10-
7. Rawshani A, Sattar N, Franzén S, et al. Excess mortality and car- year follow-up of intensive glucose control in type 2 diabetes. N
diovascular disease in young adults with type 1 diabetes in relation Engl J Med. 2008;359(15):1577–1589.
to age at onset: a nationwide, register-based cohort study. Lancet. 25. Kowluru RA, Kowluru V, Xiong Y, Ho YS. Overexpression of
2018;392(10146):477–486. mitochondrial superoxide dismutase in mice protects the retina
8. Gregg EW, Li Y, Wang J, et al. Changes in diabetes-related from diabetes-induced oxidative stress. Free Radic Biol Med.
complications in the United States, 1990-2010. N Engl J Med. 2006;41(8):1191–1196.
2014;370(16):1514–1523. 26. Craven PA, Melhem MF, Phillips SL, DeRubertis FR. Overex-
9. Nathan DM, Genuth S, Lachin J, et al. The effect of intensive treat- pression of Cu2+/Zn2+ superoxide dismutase protects against
ment of diabetes on the development and progression of long-term early diabetic glomerular injury in transgenic mice. Diabetes.
complications in insulin-dependent diabetes mellitus. N Engl J 2001;50(9):2114–2125.
Med. 1993;329(14):977–986. 27. DeRubertis FR, Craven PA, Melhem MF, Salah EM. Attenuation
10. Turner RC. The U.K. prospective diabetes study. A review. Diabetes of renal injury in db/db mice overexpressing superoxide dismutase:
Care. 1998;21(suppl 3):C35–C38. evidence for reduced superoxide-nitric oxide interaction. Diabetes.
11. Callaghan BC, Hur J, Feldman EL. Diabetic neuropathy: one dis- 2004;53(3):762–768.
ease or two? Curr Opin Neurol. 2012;25(5):536–541. 28. Vincent AM, Russell JW, Sullivan KA, et al. SOD2 protects neu-
12. Ang L, Jaiswal M, Martin C, Pop-Busui R. Glucose control and rons from injury in cell culture and animal models of diabetic neu-
diabetic neuropathy: lessons from recent large clinical trials. Curr ropathy. Exp Neurol. 2007;208(2):216–227.
Diabetes Rep. 2014;14(9):528. 29. Wang Y, Wang GZ, Rabinovitch PS, Tabas I. Macrophage mito-
13. Boussageon R, Bejan-Angoulvant T, Saadatian-Elahi M, et al. chondrial oxidative stress promotes atherosclerosis and nuclear
Effect of intensive glucose lowering treatment on all cause mor- factor-kappaB-mediated inflammation in macrophages. Circ Res.
tality, cardiovascular death, and microvascular events in type 2 2014;114(3):421–433.
diabetes: meta-analysis of randomised controlled trials. BMJ. 30. Shen X, Zheng S, Metreveli NS, Epstein PN. Protection of cardiac
2011;343:d4169. mitochondria by overexpression of MnSOD reduces diabetic car-
14. Ginsberg HN. Insulin resistance and cardiovascular disease. J Clin diomyopathy. Diabetes. 2006;55(3):798–805.
Invest. 2000;106(4):453–458. 31. Ye G, Metreveli NS, Donthi RV, et al. Catalase protects cardio-
15. Bornfeldt KE, Tabas I. Insulin resistance, hyperglycemia, and ath- myocyte function in models of type 1 and type 2 diabetes. Diabetes.
erosclerosis. Cell Metab. 2011;14(5):575–585. 2004;53(5):1336–1343.
16. Wagenknecht LE, D’Agostino RB Jr, Haffner SM, Savage PJ, Rew- 32. Finkel T. Signal transduction by reactive oxygen species. J Cell Biol.
ers M. Impaired glucose tolerance, type 2 diabetes, and carotid wall 2011;194(1):7–15.
thickness: the Insulin Resistance Atherosclerosis Study. Diabetes 33. Burgoyne JR, Mongue-Din H, Eaton P, Shah AM. Redox signaling
Care. 1998;21(11):1812–1818. in cardiac physiology and pathology. Circ Res. 2012;111(8):1091–
17. Hanley AJ, Williams K, Stern MP, Haffner SM. Homeostasis 1106.
model assessment of insulin resistance in relation to the incidence 34. Holmstrom KM, Finkel T. Cellular mechanisms and physiological
of cardiovascular disease: the San Antonio Heart Study. Diabetes consequences of redox-dependent signalling. Nat Rev Mol Cell Biol.
Care. 2002;25(7):1177–1184. 2014;15(6):411–421.
18. Ray KK, Cannon CP, Cairns R, Morrow DA, Ridker PM, 35. Liu Y, Zhao H, Li H, Kalyanaraman B, Nicolosi AC, Gutterman
Braunwald E. Prognostic utility of apoB/AI, total cholesterol/ DD. Mitochondrial sources of H2O2 generation play a key role in
HDL, non-HDL cholesterol, or hs-CRP as predictors of clinical flow-mediated dilation in human coronary resistance arteries. Circ
risk in patients receiving statin therapy after acute coronary syn- Res. 2003;93(6):573–580.
dromes: results from PROVE IT-TIMI 22. Arterioscler Thromb Vasc 36. Ray R, Murdoch CE, Wang M, et al. Endothelial Nox4 NADPH
Biol. 2009;29(3):424–430. oxidase enhances vasodilatation and reduces blood pressure in vivo.
19. Ridker PM, MacFadyen JG, Everett BM, et al. Relationship Arterioscler Thromb Vasc Biol. 2011;31(6):1368–1376.
of C-reactive protein reduction to cardiovascular event reduc- 37. Schmelter M, Ateghang B, Helmig S, Wartenberg M, Sauer H.
tion following treatment with canakinumab: a secondary analy- Embryonic stem cells utilize reactive oxygen species as transducers
sis from the CANTOS randomised controlled trial. Lancet. of mechanical strain-induced cardiovascular differentiation. FASEB
2018;391(10118):319–328. J. 2006;20(8):1182–1184.

1524.e1
Descargado para Paola Gavidia Davila ([email protected]) en Antenor Orrego Private University de ClinicalKey.es por Elsevier en julio 19, 2020.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2020. Elsevier Inc. Todos los derechos reservados.
1524.e2 References

38. Heo JS, Lee JC. beta-Catenin mediates cyclic strain-stimulated 59. Karachalias N, Babaei-Jadidi R, Ahmed N, Thornalley PJ. Accu-
cardiomyogenesis in mouse embryonic stem cells through ROS- mulation of fructosyl-lysine and advanced glycation end products
dependent and integrin-mediated PI3K/Akt pathways. J Cell Bio- in the kidney, retina and peripheral nerve of streptozotocin-induced
chem. 2011;112(7):1880–1889. diabetic rats. Biochem Soc Trans. 2003;31(Pt 6):1423–1425.
39. Prosser BL, Ward CW, Lederer WJ. X-ROS signaling: rapid mech- 60. Sachdeva R, Schlotterer A, Schumacher D, et al. TRPC proteins
ano-chemo transduction in heart. Science. 2011;333(6048):1440– contribute to development of diabetic retinopathy and regulate
1445. glyoxalase 1 activity and methylglyoxal accumulation. Mol Metab.
40. Nathan C, Cunningham-Bussel A. Beyond oxidative stress: an 2018;9:156–167.
immunologist’s guide to reactive oxygen species. Nat Rev Immunol. 61. Giacco F, Du X, D’Agati VD, et al. Knockdown of glyoxalase
2013;13(5):349–361. 1 mimics diabetic nephropathy in nondiabetic mice. Diabetes.
41. Scialo F, Fernandez-Ayala DJ, Sanz A. Role of mitochondrial 2014;63(1):291–299.
reverse electron transport in ROS signaling: potential roles in 62. Bierhaus A, Fleming T, Stoyanov S, et al. Methylglyoxal modifica-
health and disease. Front Physiol. 2017;8:428. tion of Nav1.8 facilitates nociceptive neuron firing and causes hyper-
42. Bleier L, Wittig I, Heide H, Steger M, Brandt U, Drose S. Gener- algesia in diabetic neuropathy. Nat Med. 2012;18(6):926–933.
ator-specific targets of mitochondrial reactive oxygen species. Free 63. Tikellis C, Pickering RJ, Tsorotes D, et al. Dicarbonyl stress in
Radic Biol Med. 2015;78:1–10. the absence of hyperglycemia increases endothelial inflammation
43. Brownlee M. The pathobiology of diabetic complications: a unify- and atherogenesis similar to that observed in diabetes. Diabetes.
ing mechanism. Diabetes. 2005;54(6):1615–1625. 2014;63(11):3915–3925.
44. Bhatt S, Gupta MK, Khamaisi M, et al. Preserved DNA damage 64. Hanssen NM, Wouters K, Huijberts MS, et al. Higher levels of
checkpoint pathway protects against complications in long-stand- advanced glycation endproducts in human carotid atherosclerotic
ing type 1 diabetes. Cell Metab. 2015;22(2):239–252. plaques are associated with a rupture-prone phenotype. Eur Heart
45. Shimada K, Crother TR, Karlin J, et al. Oxidized mitochondrial J. 2014;35(17):1137–1146.
DNA activates the NLRP3 inflammasome during apoptosis. 65. Tian C, Alomar F, Moore CJ, et al. Reactive carbonyl species and
Immunity. 2012;36(3):401–414. their roles in sarcoplasmic reticulum Ca2+ cycling defect in the dia-
46. Zhang JZ, Gao L, Widness M, Xi X, Kern TS. Captopril inhibits betic heart. Heart Fail Rev. 2014;19(1):101–112.
glucose accumulation in retinal cells in diabetes. Invest Ophthalmol 66. Molgat AS, Tilokee EL, Rafatian G, et al. Hyperglycemia inhibits
Vis Sci. 2003;44(9):4001–4005. cardiac stem cell-mediated cardiac repair and angiogenic capacity.
47. Bohren KM, Grimshaw CE, Gabbay KH. Catalytic effectiveness of Circulation. 2014;130(11 suppl 1):S70–S76.
human aldose reductase. Critical role of C-terminal domain. J Biol 67. Vulesevic B, McNeill B, Geoffrion M, et al. Glyoxalase-1 overex-
Chem. 1992;267(29):20965–20970. pression in bone marrow cells reverses defective neovascularization
48. Baba SP, Barski OA, Ahmed Y, et al. Reductive metabolism of in STZ-induced diabetic mice. Cardiovasc Res. 2014;101(2):306–
AGE precursors: a metabolic route for preventing AGE accumula- 316.
tion in cardiovascular tissue. Diabetes. 2009;58(11):2486–2497. 68. Nam DH, Han JH, Lee TJ, et al. CHOP deficiency prevents meth-
49. Dixit BL, Balendiran GK, Watowich SJ, et al. Kinetic and struc- ylglyoxal-induced myocyte apoptosis and cardiac dysfunction. J
tural characterization of the glutathione-binding site of aldose Mol Cell Cardiol. 2015;85:168–177.
reductase. J Biol Chem. 2000;275(28):21587–21595. 69. Thorp E, Li G, Seimon TA, Kuriakose G, Ron D, Tabas I.
50. Engerman RL, Kern TS, Larson ME. Nerve conduction and aldose Reduced apoptosis and plaque necrosis in advanced atherosclerotic
reductase inhibition during 5 years of diabetes or galactosaemia in lesions of Apoe-/- and Ldlr-/- mice lacking CHOP. Cell Metab.
dogs. Diabetologia. 1994;37(2):141–144. 2009;9(5):474–481.
51. Chalk C, Benstead TJ, Moore F. Aldose reductase inhibitors for the 70. Vulesevic B, McNeill B, Giacco F, et al. Methylglyoxal-induced
treatment of diabetic polyneuropathy. Cochrane Database Syst Rev. endothelial cell loss and inflammation contribute to the develop-
2007;(4):CD004572. ment of diabetic cardiomyopathy. Diabetes. 2016;65(6):1699–
52. Vedantham S, Noh H, Ananthakrishnan R, et al. Human aldose 1713.
reductase expression accelerates atherosclerosis in diabetic apolipo- 71. Parodi EM, Kuhn B. Signalling between microvascular endothe-
protein E-/- mice. Arterioscler Thromb Vasc Biol. 2011;31(8):1805– lium and cardiomyocytes through neuregulin. Cardiovasc Res.
1813. 2014;102(2):194–204.
53. Srivastava S, Vladykovskaya E, Barski OA, et al. Aldose reductase 72. Craven PA, Davidson CM, DeRubertis FR. Increase in diacylglyc-
protects against early atherosclerotic lesion formation in apolipo- erol mass in isolated glomeruli by glucose from de novo synthesis
protein E-null mice. Circ Res. 2009;105(8):793–802. of glycerolipids. Diabetes. 1990;39(6):667–674.
54. Thornalley PJ, Battah S, Ahmed N, et al. Quantitative screening 73. Inoguchi T, Battan R, Handler E, Sportsman JR, Heath W, King
of advanced glycation endproducts in cellular and extracellular GL. Preferential elevation of protein kinase C isoform beta II and
proteins by tandem mass spectrometry. Biochem J. 2003;375(Pt diacylglycerol levels in the aorta and heart of diabetic rats: differ-
3):581–592. ential reversibility to glycemic control by islet cell transplantation.
55. Thornalley PJ. The glyoxalase system: new developments towards Proc Natl Acad Sci U S A. 1992;89(22):11059–11063.
functional characterization of a metabolic pathway fundamental to 74. Shiba T, Inoguchi T, Sportsman JR, Heath WF, Bursell S, King
biological life. Biochem J. 1990;269(1):1–11. GL. Correlation of diacylglycerol level and protein kinase C activ-
56. Yao D, Brownlee M. Hyperglycemia-induced reactive oxygen spe- ity in rat retina to retinal circulation. Am J Physiol. 1993;265(5 Pt
cies increase expression of the receptor for advanced glycation end 1):E783–E793.
products (RAGE) and RAGE ligands. Diabetes. 2010;59(1):249– 75. Cosentino-Gomes D, Rocco-Machado N, Meyer-Fernandes JR.
255. Cell signaling through protein kinase C oxidation and activation.
57. Rong LL, Gooch C, Szabolcs M, et al. RAGE: a journey from the Int J Mol Sci. 2012;13(9):10697–10721.
complications of diabetes to disorders of the nervous system - strik- 76. Derubertis FR, Craven PA. Activation of protein kinase C in glo-
ing a fine balance between injury and repair. Restor Neurol Neurosci. merular cells in diabetes. Mechanisms and potential links to the
2005;23(5–6):355–365. pathogenesis of diabetic glomerulopathy. Diabetes. 1994;43(1):1–8.
58. Hammes HP, Du X, Edelstein D, et al. Benfotiamine blocks three 77. Mizukami H, Ogasawara S, Yamagishi S, Takahashi K, Yagihashi S.
major pathways of hyperglycemic damage and prevents experimen- Methylcobalamin effects on diabetic neuropathy and nerve protein
tal diabetic retinopathy. Nat Med. 2003;9(3):294–299. kinase C in rats. Eur J Clin Invest. 2011;41(4):442–450.

78. Ishii H, Jirousek MR, Koya D, et al. Amelioration of vascular dys- 98. Makino A, Dai A, Han Y, et al. O-GlcNAcase overexpression
functions in diabetic rats by an oral PKC beta inhibitor. Science. reverses coronary endothelial cell dysfunction in type 1 diabetic
1996;272(5262):728–731. mice. Am J Physiol Cell Physiol. 2015;309(9):C593–C599.
79. Geraldes P, Hiraoka-Yamamoto J, Matsumoto M, et al. Activation 99. Aicher A, Heeschen C, Mildner-Rihm C, et al. Essential role of
of PKC-delta and SHP-1 by hyperglycemia causes vascular cell endothelial nitric oxide synthase for mobilization of stem and pro-
apoptosis and diabetic retinopathy. Nat Med. 2009;15(11):1298– genitor cells. Nat Med. 2003;9(11):1370–1376.
1306. 100. Du XL, Edelstein D, Dimmeler S, Ju Q, Sui C, Brownlee M.
80. Craven PA, Studer RK, DeRubertis FR. Impaired nitric Hyperglycemia inhibits endothelial nitric oxide synthase activ-
oxide-dependent cyclic guanosine monophosphate genera- ity by posttranslational modification at the Akt site. J Clin Invest.
tion in glomeruli from diabetic rats. Evidence for protein kinase 2001;108(9):1341–1348.
C-mediated suppression of the cholinergic response. J Clin Invest. 101. Qian J, Fulton D. Post-translational regulation of endothelial nitric
1994;93(1):311–320. oxide synthase in vascular endothelium. Front Physiol. 2013;4.
81. Ganz MB, Seftel A. Glucose-induced changes in protein kinase C 102. Long MT, Fox CS. The Framingham Heart Study—67 years of dis-
and nitric oxide are prevented by vitamin E. Am J Physiol Endocri- covery in metabolic disease. Nat Rev Endocrinol. 2016;12:177–183.
nol Metab. 2000;278(1):E146–E152. 103. Shrikhande GV, Scali ST, da Silva CG, et al. O-glycosylation
82. Pugliese G, Pricci F, Pugliese F, et al. Mechanisms of glucose- regulates ubiquitination and degradation of the anti-inflammatory
enhanced extracellular matrix accumulation in rat glomerular protein A20 to accelerate atherosclerosis in diabetic ApoE-null
mesangial cells. Diabetes. 1994;43(3):478–490. mice. PLoS One. 2010;5(12):e14240.
83. Glogowski EA, Tsiani E, Zhou X, Fantus IG, Whiteside C. High 104. Marsh SA, Collins HE, Chatham JC. Protein O-GlcNAc-
glucose alters the response of mesangial cell protein kinase C iso- ylation and cardiovascular (patho)physiology. J Biol Chem.
forms to endothelin-1. Kidney Int. 1999;55(2):486–499. 2014;289(50):34449–34456.
84. Hempel A, Maasch C, Heintze U, et al. High glucose concentra- 105. Clark RJ, McDonough PM, Swanson E, et al. Diabetes and the
tions increase endothelial cell permeability via activation of protein accompanying hyperglycemia impairs cardiomyocyte calcium
kinase C alpha. Circ Res. 1997;81(3):363–371. cycling through increased nuclear O-GlcNAcylation. J Biol Chem.
85. Williams B, Gallacher B, Patel H, Orme C. Glucose-induced 2003;278(45):44230–44237.
protein kinase C activation regulates vascular permeability fac- 106. Erickson JR, Pereira L, Wang L, et al. Diabetic hyperglycaemia acti-
tor mRNA expression and peptide production by human vates CaMKII and arrhythmias by O-linked glycosylation. Nature.
vascular smooth muscle cells in vitro. Diabetes. 1997;46(9): 2013;502(7471):372–376.
1497–1503. 107. Daniels L, Bell JR, Delbridge LM, McDonald FJ, Lamberts RR,
86. Geraldes P, King GL. Activation of protein kinase C isoforms and Erickson JR. The role of CaMKII in diabetic heart dysfunction.
its impact on diabetic complications. Circ Res. 2010;106(8):1319– Heart Fail Rev. 2015;20(5):589–600.
1331. 108. Luo M, Guan X, Luczak ED, et al. Diabetes increases mortality
87. Kong L, Shen X, Lin L, et al. PKCbeta promotes vascular inflam- after myocardial infarction by oxidizing CaMKII. J Clin Invest.
mation and acceleration of atherosclerosis in diabetic ApoE null 2013;123(3):1262–1274.
mice. Arterioscler Thromb Vasc Biol. 2013;33(8):1779–1787. 109. Banerjee PS, Ma J, Hart GW. Diabetes-associated dysregulation of
88. Durpes MC, Morin C, Paquin-Veillet J, et al. PKC-beta activation O-GlcNAcylation in rat cardiac mitochondria. Proc Natl Acad Sci
inhibits IL-18-binding protein causing endothelial dysfunction and U S A. 2015;112(19):6050–6055.
diabetic atherosclerosis. Cardiovasc Res. 2015;106(2):303–313. 110. Morisseau C, Hammock BD. Impact of soluble epoxide hydrolase
89. Li Q, Park K, Li C, et al. Induction of vascular insulin resistance and epoxyeicosanoids on human health. Annu Rev Pharmacol Toxi-
and endothelin-1 expression and acceleration of atherosclerosis by col. 2013;53:37–58.
the overexpression of protein kinase C-beta isoform in the endothe- 111. Hu J, Dziumbla S, Lin J, et al. Inhibition of soluble epoxide hydro-
lium. Circ Res. 2013;113(4):418–427. lase prevents diabetic retinopathy. Nature. 2017;552(7684):248–
90. Way KJ, Isshiki K, Suzuma K, et al. Expression of connective tis- 252.
sue growth factor is increased in injured myocardium associated 112. Bettaieb A, Koike S, Hsu MF, et al. Soluble epoxide hydrolase
with protein kinase C beta2 activation and diabetes. Diabetes. in podocytes is a significant contributor to renal function under
2002;51(9):2709–2718. hyperglycemia. Biochim Biophys Acta. 2017;1861(11 Pt A):2758–
91. Verma SK, Deshmukh V, Liu P, et al. Reactivation of fetal splicing 2765.
programs in diabetic hearts is mediated by protein kinase C signal- 113. Inceoglu B, Wagner KM, Yang J, et al. Acute augmentation of
ing. J Biol Chem. 2013;288(49):35372–35386. epoxygenated fatty acid levels rapidly reduces pain-related behav-
92. Rask-Madsen C, King GL. Vascular complications of diabe- ior in a rat model of type I diabetes. Proc Natl Acad Sci U S A.
tes: mechanisms of injury and protective factors. Cell Metab. 2012;109(28):11390–11395.
2013;17(1):20–33. 114. Lee CR, North KE, Bray MS, et al. Genetic variation in soluble
93. Li Z, Abdullah CS, Jin ZQ. Inhibition of PKC-theta preserves car- epoxide hydrolase (EPHX2) and risk of coronary heart disease:
diac function and reduces fibrosis in streptozotocin-induced dia- the Atherosclerosis Risk in Communities (ARIC) study. Hum Mol
betic cardiomyopathy. Br J Pharmacol. 2014;171:2913–2924. Genet. 2006;15(10):1640–1649.
94. Zachara N, Akimoto Y, Hart GW. The O-GlcNAc modification. 115. Xu D, Li N, He Y, et al. Prevention and reversal of cardiac hyper-
In: Varki A, Cummings RD, Esko JD, et al., eds. Essentials of Gly- trophy by soluble epoxide hydrolase inhibitors. Proc Natl Acad Sci
cobiology. 3rd ed. Cold Spring Harbor, NY: Cold Spring Harbor U S A. 2006;103(49):18733–18738.
Laboratory Press; 2017:239–252. 116. Ai D, Pang W, Li N, et al. Soluble epoxide hydrolase plays an essen-
95. Hardiville S, Hart GW. Nutrient regulation of signaling, tran- tial role in angiotensin II-induced cardiac hypertrophy. Proc Natl
scription, and cell physiology by O-GlcNAcylation. Cell Metab. Acad Sci U S A. 2009;106(2):564–569.
2014;20(2):208–213. 117. Zhao TT, Wasti B, Xu DY, Shen L, Du JQ, Zhao SP. Soluble
96. Bond MR, Hanover JA. A little sugar goes a long way: the cell biol- epoxide hydrolase and ischemic cardiomyopathy. Int J Cardiol.
ogy of O-GlcNAc. J Cell Biol. 2015;208(7):869–880. 2012;155(2):181–187.
97. Tan EP, Villar MT, E.L., et al. Altering O-linked beta-N-acetyl- 118. Kompa AR, Wang BH, Xu G, et al. Soluble epoxide hydrolase inhi-
glucosamine cycling disrupts mitochondrial function. J Biol Chem. bition exerts beneficial anti-remodeling actions post-myocardial
2014;289(21):14719–14730. infarction. Int J Cardiol. 2013;167(1):210–219.

119. Ono E, Dutile S, Kazani S, et al. Lipoxin generation is related to species with other sources of oxidative stress. Antioxid Redox Signal.
soluble epoxide hydrolase activity in severe asthma. Am J Respir Crit 2014;20(2):308–324.
Care Med. 2014;190(8):886–897. 141. Dugan LL, You YH, Ali SS, et al. AMPK dysregulation promotes
120. Buckley CD, Gilroy DW, Serhan CN. Proresolving lipid mediators diabetes-related reduction of superoxide and mitochondrial func-
and mechanisms in the resolution of acute inflammation. Immu- tion. J Clin Invest. 2013;123(11):4888–4899.
nity. 2014;40(3):315–327. 142. Wang S, Song P, Zou MH. AMP-activated protein kinase, stress
121. Serhan CN, Levy BD. Resolvins in inflammation: emergence responses and cardiovascular diseases. Clin Sci. 2012;122(12):555–
of the pro-resolving superfamily of mediators. J Clin Invest. 573.
2018;128(7):2657–2669. 143. Giacco F, Du X, Carratu A, et al. GLP-1 cleavage product reverses
122. Tabas I. Macrophage death and defective inflammation resolution persistent ROS generation after transient hyperglycemia by disrupt-
in atherosclerosis. Nat Rev Immunol. 2009;10:36. ing an ROS-generating feedback loop. Diabetes. 2015;64(9):3273–
123. Yurdagul A Jr, Doran AC, Cai B, Fredman G, Tabas IA. Mecha- 3284.
nisms and consequences of defective efferocytosis in atherosclerosis. 144. Garcia Soriano F, Virag L, Jagtap P, et al. Diabetic endothelial dys-
Front Cardiovasc Med. 2017;4:86. function: the role of poly(ADP-ribose) polymerase activation. Nat
124. Wang Y, Liu J, Jin X, et al. O-GlcNAcylation destabilizes the active Med. 2001;7(1):108–113.
tetrameric PKM2 to promote the Warburg effect. Proc Natl Acad 145. Nishikawa T, Brownlee M, Araki E. Mitochondrial reactive oxygen
Sci U S A. 2017;114(52):13732–13737. species in the pathogenesis of early diabetic nephropathy. J Diabetes
125. Dayton TL, Jacks T, Vander Heiden MG. PKM2, cancer metabo- Investig. 2015;6(2):137–139.
lism, and the road ahead. EMBO Rep. 2016;17(12):1721–1730. 146. Sack MN. Mitochondrial fidelity and metabolic agility control
126. Luo W, Hu H, Chang R, et al. Pyruvate kinase M2 is a PHD3- immune cell fate and function. J Clin Invest. 2018;128(9):3651–
stimulated coactivator for hypoxia-inducible factor 1. Cell. 3661.
2011;145(5):732–744. 147. Liu R, Zhong Y, Li X, et al. Role of transcription factor acetylation
127. Wang HJ, Hsieh YJ, Cheng WC, et al. JMJD5 regulates PKM2 in diabetic kidney disease. Diabetes. 2014;63(7):2440–2453.
nuclear translocation and reprograms HIF-1alpha-mediated glu- 148. Choi J, Chandrasekaran K, Inoue T, Muragundla A, Russell JW.
cose metabolism. Proc Natl Acad Sci U S A. 2014;111(1):279–284. PGC-1alpha regulation of mitochondrial degeneration in experi-
128. Lu Z, Hunter T. Metabolic kinases moonlighting as protein mental diabetic neuropathy. Neurobiol Dis. 2014;64:118–130.
kinases. Trends Biochem Sci. 2018;43(4):301–310. 149. Galvan DL, Badal SS, Long J, et al. Real-time in vivo mitochondrial
129. Qi W, Keenan HA, Li Q, et al. Pyruvate kinase M2 activation may redox assessment confirms enhanced mitochondrial reactive oxy-
protect against the progression of diabetic glomerular pathology gen species in diabetic nephropathy. Kidney Int. 2017;92(5):1282–
and mitochondrial dysfunction. Nat Med. 2017;23(6):753–762. 1287.
130. Nishikawa T, Edelstein D, Du XL, et al. Normalizing mitochon- 150. Zhang D, Xie X, Chen Y, Hammock BD, Kong W, Zhu Y. Homo-
drial superoxide production blocks three pathways of hyperglycae- cysteine upregulates soluble epoxide hydrolase in vascular endothe-
mic damage. Nature. 2000;404(6779):787–790. lium in vitro and in vivo. Circ Res. 2012;110(6):808–817.
131. Brownlee M. Biochemistry and molecular cell biology of diabetic 151. Qi H, Casalena G, Shi S, et al. Glomerular endothelial mitochon-
complications. Nature. 2001;414(6865):813–820. drial dysfunction is essential and characteristic of diabetic kidney
132. Aghajanian A, Wittchen ES, Campbell SL, Burridge K. Direct acti- disease susceptibility. Diabetes. 2017;66(3):763–778.
vation of RhoA by reactive oxygen species requires a redox-sensitive 152. Brown MS, Goldstein JL. Selective versus total insulin resistance: a
motif. PLoS One. 2009;4(11):e8045. pathogenic paradox. Cell Metab. 2008;7(2):95–96.
133. Wang W, Wang Y, Long J, et al. Mitochondrial fission triggered by 153. Czech MP. Insulin action and resistance in obesity and type 2 dia-
hyperglycemia is mediated by ROCK1 activation in podocytes and betes. Nat Med. 2017;23:804.
endothelial cells. Cell Metab. 2012;15(2):186–200. 154. Titchenell PM, Quinn WJ, Lu M, et al. Direct hepatocyte insu-
134. Crabtree MJ, Tatham AL, Al-Wakeel Y, et al. Quantitative regu- lin signaling is required for lipogenesis but is dispensable for the
lation of intracellular endothelial nitric-oxide synthase (eNOS) suppression of glucose production. Cell Metab. 2016;23(6):1154–
coupling by both tetrahydrobiopterin-eNOS stoichiometry and 1166.
biopterin redox status: insights from cells with tet-regulated 155. Bergman BC, Howard D, Schauer IE, et al. The importance of
GTP cyclohydrolase I expression. J Biol Chem. 2009;284(2): palmitoleic acid to adipocyte insulin resistance and whole-body
1136–1144. insulin sensitivity in type 1 diabetes. J Clin Endocrinol Metab.
135. Jha JC, Banal C, Chow BS, Cooper ME, Jandeleit-Dahm K. Dia- 2013;98(1):E40–E50.
betes and kidney disease: role of oxidative stress. Antioxid Redox 156. Bergman BC, Howard D, Schauer IE, et al. Features of hepatic and
Signal. 2016;25(12):657–684. skeletal muscle insulin resistance unique to type 1 diabetes. J Clin
136. Jha JC, Gray SP, Barit D, et al. Genetic targeting or pharma- Endocrinol Metab. 2012;97(5):1663–1672.
cologic inhibition of NADPH oxidase nox4 provides renopro- 157. Schauer IE, Snell-Bergeon JK, Bergman BC, et al. Insulin resis-
tection in long-term diabetic nephropathy. J Am Soc Nephrol. tance, defective insulin-mediated fatty acid suppression, and
2014;25(6):1237–1254. coronary artery calcification in subjects with and without type 1
137. Bernard K, Logsdon NJ, Miguel V, et al. NADPH oxidase 4 diabetes: the CACTI study. Diabetes. 2011;60(1):306–314.
(Nox4) suppresses mitochondrial biogenesis and bioenergetics 158. Rosca MG, Vazquez EJ, Chen Q, Kerner J, Kern TS, Hoppel CL.
in lung fibroblasts via a nuclear factor erythroid-derived 2-like Oxidation of fatty acids is the source of increased mitochondrial
2 (Nrf2)-dependent pathway. J Biol Chem. 2017;292(7):3029– reactive oxygen species production in kidney cortical tubules in
3038. early diabetes. Diabetes. 2012;61(8):2074–2083.
138. Adam-Vizi V, Chinopoulos C. Bioenergetics and the formation 159. Brookheart RT, Michel CI, Listenberger LL, Ory DS, Schaf-
of mitochondrial reactive oxygen species. Trends Pharmacol Sci. fer JE. The non-coding RNA gadd7 is a regulator of lipid-
2006;27(12):639–645. induced oxidative and endoplasmic reticulum stress. J Biol Chem.
139. Divakaruni AS, Paradyse A, Ferrick DA, et al. Analysis and inter- 2009;284(12):7446–7454.
pretation of microplate-based oxygen consumption and pH data. 160. Du X, Edelstein D, Obici S, Higham N, Zou M-H, Brownlee M.
Methods Enzymol. 2014;547:309–354. Insulin resistance reduces arterial prostacyclin synthase and eNOS
140. Schulz E, Wenzel P, Munzel T, Daiber A. Mitochondrial activities by increasing endothelial fatty acid oxidation. J Clin
redox signaling: interaction of mitochondrial reactive oxygen Invest. 2006;116(4):1071–1080.

161. Jiang ZY, Lin YW, Clemont A, et al. Characterization of selective 184. Li J, Romestaing C, Han X, et al. Cardiolipin remodeling by
resistance to insulin signaling in the vasculature of obese Zucker ALCAT1 links oxidative stress and mitochondrial dysfunction to
(fa/fa) rats. J Clin Invest. 1999;104(4):447–457. obesity. Cell Metab. 2010;12(2):154–165.
162. Li H, Forstermann U. Nitric oxide in the pathogenesis of vascular 185. Eijkelenboom A, Burgering BM. FOXOs: signalling integrators for
disease. J Pathol. 2000;190(3):244–254. homeostasis maintenance. Nat Rev Mol Cell Biol. 2013;14(2):83–
163. Li H, Forstermann U. Pharmacological prevention of eNOS 97.
uncoupling. Curr Pharm Des. 2014;20(22):3595–3606. 186. Ozcan L, Wong CC, Li G, et al. Calcium signaling through CaM-
164. Banskota NK, Taub R, Zellner K, King GL. Insulin, insulin-like KII regulates hepatic glucose production in fasting and obesity. Cell
growth factor I and platelet-derived growth factor interact addi- Metab. 2012;15(5):739–751.
tively in the induction of the protooncogene c-myc and cellular 187. Battiprolu PK, Hojayev B, Jiang N, et al. Metabolic stress-induced
proliferation in cultured bovine aortic smooth muscle cells. Mol activation of FoxO1 triggers diabetic cardiomyopathy in mice. J
Endocrinol. 1989;3(8):1183–1190. Clin Invest. 2012;122(3):1109–1118.
165. Stolar MW. Atherosclerosis in diabetes: the role of hyperinsu- 188. Ma Q. Role of nrf2 in oxidative stress and toxicity. Annu Rev Phar-
linemia. Metab Clin Exp. 1988;37(2 suppl 1):1–9. macol Toxicol. 2013;53:401–426.
166. Du XL, Edelstein D, Rossetti L, et al. Hyperglycemia-induced 189. Xue M, Rabbani N, Momiji H, et al. Transcriptional control of
mitochondrial superoxide overproduction activates the hexosamine glyoxalase 1 by Nrf2 provides a stress-responsive defence against
pathway and induces plasminogen activator inhibitor-1 expres- dicarbonyl glycation. Biochem J. 2012;443(1):213–222.
sion by increasing Sp1 glycosylation. Proc Natl Acad Sci U S A. 190. Galligan JJ, Wepy JA, Streeter MD, et al. Methylglyoxal-derived
2000;97(22):12222–12226. posttranslational arginine modifications are abundant histone
167. Tabas I, Ron D. Integrating the mechanisms of apoptosis induced marks. Proc Natl Acad Sci U S A. 2018;115(37):9228–9233.
by endoplasmic reticulum stress. Nat Cell Biol. 2011;13(3):184– 191. Jiang T, Huang Z, Lin Y, Zhang Z, Fang D, Zhang DD. The pro-
190. tective role of Nrf2 in streptozotocin-induced diabetic nephropa-
168. Han J, Back SH, Hur J, et al. ER-stress-induced transcriptional thy. Diabetes. 2010;59(4):850–860.
regulation increases protein synthesis leading to cell death. Nat Cell 192. Bai Y, Cui W, Xin Y, et al. Prevention by sulforaphane of dia-
Biol. 2013;15(5):481–490. betic cardiomyopathy is associated with up-regulation of Nrf2
169. Tabas I, Tall A, Accili D. The impact of macrophage insulin resis- expression and transcription activation. J Mol Cell Cardiol.
tance on advanced atherosclerotic plaque progression. Circ Res. 2013;57:82–95.
2010;106(1):58–67. 193. Zheng H, Whitman SA, Wu W, et al. Therapeutic potential of
170. Kojima Y, Weissman IL, Leeper NJ. The role of efferocytosis in Nrf2 activators in streptozotocin-induced diabetic nephropathy.
atherosclerosis. Circulation. 2017;135(5):476–489. Diabetes. 2011;60(11):3055–3066.
171. Myers TJ, Brennaman LH, Stevenson M, et al. Mitochondrial 194. Shahzad K, Bock F, Al-Dabet MM, et al. Stabilization of endog-
reactive oxygen species mediate GPCR-induced TACE/ADAM17- enous Nrf2 by minocycline protects against Nlrp3-inflammasome
dependent transforming growth factor-alpha shedding. Mol Biol induced diabetic nephropathy. Sci Rep. 2016;6:34228.
Cell. 2009;20(24):5236–5249. 195. Niture SK, Khatri R, Jaiswal AK. Regulation of nrf2-an update.
172. Cai B, Thorp EB, Doran AC, et al. MerTK receptor cleavage pro- Free Radic Biol Med. 2014;66:36–44.
motes plaque necrosis and defective resolution in atherosclerosis. J 196. Rada P, Rojo AI, Chowdhry S, McMahon M, Hayes JD, Cuadrado
Clin Invest. 2017;127(2):564–568. A. SCF/{beta}-TrCP promotes glycogen synthase kinase 3-depen-
173. Kojima Y, Volkmer JP, McKenna K, et al. CD47-blocking anti- dent degradation of the Nrf2 transcription factor in a Keap1-inde-
bodies restore phagocytosis and prevent atherosclerosis. Nature. pendent manner. Mol Cell Biol. 2011;31(6):1121–1133.
2016;536(7614):86–90. 197. Xue M, Qian Q, Adaikalakoteswari A, Rabbani N, Babaei-Jadidi
174. Kanter JE, Kramer F, Barnhart S, et al. Diabetes promotes R, Thornalley PJ. Activation of NF-E2-related factor-2 reverses
an inflammatory macrophage phenotype and atherosclero- biochemical dysfunction of endothelial cells induced by hyper-
sis through acyl-CoA synthetase 1. Proc Natl Acad Sci U S A. glycemia linked to vascular disease. Diabetes. 2008;57(10):2809–
2012;109(12):E715–E724. 2817.
175. Kanter JE, Bornfeldt KE. Inflammation and diabetes-accelerated 198. Wang Y, Zhang Z, Sun W, et al. Sulforaphane attenuation of type
atherosclerosis: myeloid cell mediators. Trends Endocrinol Metab. 2 diabetes-induced aortic damage was associated with the upreg-
2013;24(3):137–144. ulation of Nrf2 expression and function. Oxid Med Cell Longev.
176. Mills EL, Kelly B, Logan A, et al. Succinate dehydrogenase sup- 2014;2014:123963.
ports metabolic repurposing of mitochondria to drive inflamma- 199. Zhang Z, Wang S, Zhou S, et al. Sulforaphane prevents the devel-
tory macrophages. Cell. 2016;167(2):457–470.e13. opment of cardiomyopathy in type 2 diabetic mice probably by
177. Turpin SM, Nicholls HT, Willmes DM, et al. Obesity-induced reversing oxidative stress-induced inhibition of LKB1/AMPK path-
CerS6-dependent C16:0 ceramide production promotes weight way. J Mol Cell Cardiol. 2014;77:42–52.
gain and glucose intolerance. Cell Metab. 2014;20(4):678–686. 200. Tan Y, Ichikawa T, Li J, et al. Diabetic downregulation of Nrf2
178. Chaurasia B, Summers SA. Ceramides - lipotoxic inducers of meta- activity via ERK contributes to oxidative stress-induced insu-
bolic disorders. Trends Endocrinol Metab. 2015;26(10):538–550. lin resistance in cardiac cells in vitro and in vivo. Diabetes.
179. Han X, Yang J, Yang K, et al. Alterations in myocardial cardiolipin 2011;60(2):625–633.
content and composition occur at the very earliest stages of diabe- 201. Shao BZ, Xu ZQ, Han BZ, Su DF, Liu C. NLRP3 inflammasome
tes: a shotgun lipidomics study. Biochemistry. 2007;46(21):6417– and its inhibitors: a review. Front Pharmacol. 2015;6:262.
6428. 202. Guo H, Callaway JB, Ting JP. Inflammasomes: mechanism of
180. Ren M, Phoon CK, Schlame M. Metabolism and function of mito- action, role in disease, and therapeutics. Nat Med. 2015;21(7):677–
chondrial cardiolipin. Prog Lipid Res. 2014;55:1–16. 687.
181. Paradies G, Paradies V, De Benedictis V, Ruggiero FM, Petrosillo 203. Bierhaus A, Schiekofer S, Schwaninger M, et al. Diabetes-associ-
G. Functional role of cardiolipin in mitochondrial bioenergetics. ated sustained activation of the transcription factor nuclear factor-
Biochim Biophys Acta. 2014;1837(4):408–417. kappaB. Diabetes. 2001;50(12):2792–2808.
182. Frohman MA. Role of mitochondrial lipids in guiding fission and 204. Nakahira K, Haspel JA, Rathinam VA, et al. Autophagy proteins
fusion. J Mol Med. 2015;93(3):263–269. regulate innate immune responses by inhibiting the release of
183. He Q, Han X. Cardiolipin remodeling in diabetic heart. Chem Phys mitochondrial DNA mediated by the NALP3 inflammasome. Nat
Lipids. 2014;179:75–81. Immunol. 2011;12(3):222–230.

205. Zhou R, Yazdi AS, Menu P, Tschopp J. A role for mitochondria in 228. Brinkmann V, Reichard U, Goosmann C, et al. Neutrophil
NLRP3 inflammasome activation. Nature. 2011;469(7329):221– extracellular traps kill bacteria. Science. 2004;303(5663):1532–
225. 1535.
206. Zhong Z, Zhai Y, Liang S, et al. TRPM2 links oxidative stress to 229. Fadini GP, Menegazzo L, Scattolini V, Gintoli M, Albiero M, Avo-
NLRP3 inflammasome activation. Nat Commun. 2013;4:1611. garo A. A perspective on NETosis in diabetes and cardiometabolic
207. Lee HM, Kim JJ, Kim HJ, Shong M, Ku BJ, Jo EK. Upregulated disorders. Nutr Metab Cardiovasc Dis. 2016;26(1):1–8.
NLRP3 inflammasome activation in patients with type 2 diabetes. 230. Rohrbach AS, Slade DJ, Thompson PR, Mowen KA. Activation of
Diabetes. 2013;62(1):194–204. PAD4 in NET formation. Front Immunol. 2012;3:360.
208. Shahzad K, Bock F, Dong W, et al. Nlrp3-inflammasome activa- 231. Nahrendorf M, Swirski FK. Immunology. Neutrophil-macrophage
tion in non-myeloid-derived cells aggravates diabetic nephropathy. communication in inflammation and atherosclerosis. Science.
Kidney Int. 2015;87(1):74–84. 2015;349(6245):237–238.
209. Shahzad K, Bock F, Al-Dabet MM, et al. Caspase-1, but not 232. Wang Y, Wang W, Wang N, Tall AR, Tabas I. Mitochondrial oxi-
caspase-3, promotes diabetic nephropathy. J Am Soc Nephrol. dative stress promotes atherosclerosis and neutrophil extracellular
2016;27(8):2270–2275. traps in aged mice. Arterioscler Thromb Vasc Biol. 2017;37(8):e99–
210. Thomas MC, Brownlee M, Susztak K, et al. Diabetic kidney dis- e107.
ease. Nat Rev Dis Primers. 2015;1:15018. 233. Lood C, Blanco LP, Purmalek MM, et al. Neutrophil extracellular
211. Liles JT, Corkey BK, Notte GT, et al. ASK1 contributes to fibro- traps enriched in oxidized mitochondrial DNA are interferogenic
sis and dysfunction in models of kidney disease. J Clin Invest. and contribute to lupus-like disease. Nat Med. 2016;22(2):146–
2018;128(10):4485–4500. 153.
212. Luo B, Li B, Wang W, et al. NLRP3 gene silencing ameliorates 234. Wong SL, Demers M, Martinod K, et al. Diabetes primes neutro-
diabetic cardiomyopathy in a type 2 diabetes rat model. PLoS One. phils to undergo NETosis, which impairs wound healing. Nat Med.
2014;9(8):e104771. 2015;21(7):815–819.
213. Li Y, Xu S, Jiang B, Cohen RA, Zang M. Activation of sterol 235. Saltiel AR, Olefsky JM. Inflammatory mechanisms linking obesity
regulatory element binding protein and NLRP3 inflammasome and metabolic disease. J Clin Invest. 2017;127(1):1–4.
in atherosclerotic lesion development in diabetic pigs. PLoS One. 236. Hotamisligil GS. Inflammation and metabolic disorders. Nature.
2013;8(6):e67532. 2006;444(7121):860–867.
214. Im SS, Yousef L, Blaschitz C, et al. Linking lipid metabolism to the 237. Rocha VZ, Libby P. Obesity, inflammation, and atherosclerosis.
innate immune response in macrophages through sterol regulatory Nat Rev Cardiol. 2009;6:399.
element binding protein-1a. Cell Metab. 2011;13(5):540–549. 238. Spite M, Claria J, Serhan CN. Resolvins, specialized proresolving
215. Hogan PG, Chen L, Nardone J, Rao A. Transcriptional regulation by lipid mediators, and their potential roles in metabolic diseases. Cell
calcium, calcineurin, and NFAT. Genes Dev. 2003;17(18):2205–2232. Metab. 2014;19(1):21–36.
216. Olabisi OA, Soto-Nieves N, Nieves E, et al. Regulation of tran- 239. Kasikara C, Doran AC, Cai B, Tabas I. The role of non-resolving
scription factor NFAT by ADP-ribosylation. Mol Cell Biol. inflammation in atherosclerosis. J Clin Invest. 2018;128(7):2713–
2008;28(9):2860–2871. 2723.
217. Tang J, Kern TS. Inflammation in diabetic retinopathy. Prog Retin 240. Rahman K, Fisher EA. Insights from pre-clinical and clinical stud-
Eye Res. 2011;30(5):343–358. ies on the role of innate inflammation in atherosclerosis regression.
218. Savage SR, Bretz CA, Penn JS. RNA-Seq reveals a role for NFAT- Front Cardiovasc Med. 2018;5:32.
signaling in human retinal microvascular endothelial cells treated 241. Fredman G, Hellmann J, Proto JD, et al. An imbalance between
with TNFalpha. PLoS One. 2015;10(1):e0116941. specialized pro-resolving lipid mediators and pro-inflammatory
219. Pedigo CE, Ducasa GM, Leclercq F, et al. Local TNF causes leukotrienes promotes instability of atherosclerotic plaques. Nat
NFATc1-dependent cholesterol-mediated podocyte injury. J Clin Commun. 2016;7:12859.
Invest. 2016;126(9):3336–3350. 242. Doran AC, Ozcan L, Cai B, et al. CAMKIIgamma suppresses an
220. Nilsson-Berglund LM, Zetterqvist AV, Nilsson-Ohman J, et al. efferocytosis pathway in macrophages and promotes atherosclerotic
Nuclear factor of activated T cells regulates osteopontin expression plaque necrosis. J Clin Invest. 2017;127(11):4075–4089.
in arterial smooth muscle in response to diabetes-induced hypergly- 243. Vasquez-Trincado C, Garcia-Carvajal I, Pennanen C, et al. Mito-
cemia. Arterioscler Thromb Vasc Biol. 2010;30(2):218–224. chondrial dynamics, mitophagy and cardiovascular disease. J
221. Zetterqvist AV, Berglund LM, Blanco F, et al. Inhibition of nuclear Physiol. 2016;594(3):509–525.
factor of activated T-cells (NFAT) suppresses accelerated atheroscle- 244. Wang Y, Subramanian M, Yurdagul A Jr, et al. Mitochondrial fis-
rosis in diabetic mice. PLoS One. 2014;8(6):e65020. sion promotes the continued clearance of apoptotic cells by macro-
222. Friedman JK, Nitta CH, Henderson KM, et al. Intermittent phages. Cell. 2017;171(2):331–345.e22.
hypoxia-induced increases in reactive oxygen species activate 245. Mishra P, Chan DC. Mitochondrial dynamics and inheritance
NFATc3 increasing endothelin-1 vasoconstrictor reactivity. Vascul during cell division, development and disease. Nat Rev Mol Cell
Pharmacol. 2014;60(1):17–24. Biol. 2014;15(10):634–646.
223. Letavernier E, Zafrani L, Perez J, Letavernier B, Haymann JP, Baud 246. Tilokani L, Nagashima S, Paupe V, Prudent J. Mitochondrial
L. The role of calpains in myocardial remodelling and heart failure. dynamics: overview of molecular mechanisms. Essays Biochem.
Cardiovasc Res. 2012;96(1):38–45. 2018;62(3):341–360.
224. Bourajjaj M, Armand AS, da Costa Martins PA, et al. NFATc2 is 247. Ayanga BA, Badal SS, Wang Y, et al. Dynamin-related pro-
a necessary mediator of calcineurin-dependent cardiac hypertrophy tein 1 deficiency improves mitochondrial fitness and protects
and heart failure. J Biol Chem. 2008;283(32):22295–22303. against progression of diabetic nephropathy. J Am Soc Nephrol.
225. Li Y, Ma J, Zhu H, et al. Targeted inhibition of calpain reduces 2016;27(9):2733–2747.
myocardial hypertrophy and fibrosis in mouse models of type 1 248. Makino A, Scott BT, Dillmann WH. Mitochondrial fragmentation
diabetes. Diabetes. 2011;60(11):2985–2994. and superoxide anion production in coronary endothelial cells from
226. Wang JC, Zhao Y, Li XD, Zhou NN, Sun H, Sun YY. Proteolysis a mouse model of type 1 diabetes. Diabetologia. 2010;53(8):1783–
by endogenous calpain I leads to the activation of calcineurin in 1794.
human heart. Clin Lab. 2012;58(11–12):1145–1152. 249. Montaigne D, Marechal X, Coisne A, et al. Myocardial contractile
227. Warnatsch A, Ioannou M, Wang Q, Papayannopoulos V. Inflamma- dysfunction is associated with impaired mitochondrial function
tion. Neutrophil extracellular traps license macrophages for cytokine and dynamics in type 2 diabetic but not in obese patients. Circula-
production in atherosclerosis. Science. 2015;349(6245):316–320. tion. 2014;130(7):554–564.

250. Krolewski AS, Warram JH, Freire MB. Epidemiology of late 273. Januszyk M, Sorkin M, Glotzbach JP, et al. Diabetes irreversibly
diabetic complications. A basis for the development and evalua- depletes bone marrow-derived mesenchymal progenitor cell sub-
tion of preventive programs. Endocrinol Metab Clin North Am. populations. Diabetes. 2014;63(9):3047–3056.
1996;25(2):217–242. 274. Hardiville S, Hart GW. Nutrient regulation of gene expres-
251. Seaquist ER, Goetz FC, Rich S, Barbosa J. Familial clustering of sion by O-GlcNAcylation of chromatin. Curr Opin Chem Biol.
diabetic kidney disease. Evidence for genetic susceptibility to dia- 2016;33:88–94.
betic nephropathy. N Engl J Med. 1989;320(18):1161–1165. 275. Keating ST, Plutzky J, El-Osta A. Epigenetic changes in diabetes
252. Quinn M, Angelico MC, Warram JH, Krolewski AS. Familial fac- and cardiovascular risk. Circ Res. 2016;118(11):1706–1722.
tors determine the development of diabetic nephropathy in patients 276. Kietzmann T, Petry A, Shvetsova A, Gerhold JM, Gorlach A. The
with IDDM. Diabetologia. 1996;39(8):940–945. epigenetic landscape related to reactive oxygen species formation
253. Wagenknecht LE, Bowden DW, Carr JJ, Langefeld CD, Freedman in the cardiovascular system. Br J Pharmacol. 2017;174(12):1533–
BI, Rich SS. Familial aggregation of coronary artery calcium in 1554.
families with type 2 diabetes. Diabetes. 2001;50(4):861–866. 277. Keating ST, van Diepen JA, Riksen NP, El-Osta A. Epigenetics
254. Williams WW, Salem RM, McKnight AJ, et al. Association testing in diabetic nephropathy, immunity and metabolism. Diabetologia.
of previously reported variants in a large case-control meta-analysis 2018;61(1):6–20.
of diabetic nephropathy. Diabetes. 2012;61(8):2187–2194. 278. Chen Z, Miao F, Paterson AD, et al. Epigenomic profiling reveals
255. Dahlstrom E, Sandholm N. Progress in defining the genetic basis of an association between persistence of DNA methylation and meta-
diabetic complications. Curr Diab Rep. 2017;17(9):80. bolic memory in the DCCT/EDIC type 1 diabetes cohort. Proc
256. Henikoff S, Greally JM. Epigenetics, cellular memory and gene Natl Acad Sci U S A. 2016;113(21):E3002–E3011.
regulation. Curr Biol. 2016;26(14):R644–R648. 279. Huang C, Lin MZ, Cheng D, Braet F, Pollock CA, Chen XM.
257. Do C, Shearer A, Suzuki M, et al. Genetic-epigenetic interac- Thioredoxin-interacting protein mediates dysfunction of tubular
tions in cis: a major focus in the post-GWAS era. Genome Biol. autophagy in diabetic kidneys through inhibiting autophagic flux.
2017;18(1):120. Lab Invest. 2014;94(3):309–320.
258. Friedman RC, Farh KK-H, Burge CB, Bartel DP. Most mam- 280. Miao F, Chen Z, Genuth S, et al. Evaluating the role of epigenetic
malian mRNAs are conserved targets of microRNAs. Genome Res. histone modifications in the metabolic memory of type 1 diabetes.
2009;19(1):92–105. Diabetes. 2014;63(5):1748–1762.
259. Kato M, Castro NE, Natarajan R. MicroRNAs: potential media- 281. El-Osta A, Brasacchio D, Yao D, et al. Transient high glu-
tors and biomarkers of diabetic complications. Free Radic Biol Med. cose causes persistent epigenetic changes and altered gene
2013;64:85–94. expression during subsequent normoglycemia. J Exp Med.
260. Feng B, Chen S, McArthur K, et al. miR-146a-Mediated extracel- 2008;205(10):2409–2417.
lular matrix protein production in chronic diabetes complications. 282. Brasacchio D, Okabe J, Tikellis C, et al. Hyperglycemia induces
Diabetes. 2011;60(11):2975–2984. a dynamic cooperativity of histone methylase and demethylase
261. Putta S, Lanting L, Sun G, Lawson G, Kato M, Natarajan R. enzymes associated with gene-activating epigenetic marks that
Inhibiting microRNA-192 ameliorates renal fibrosis in diabetic coexist on the lysine tail. Diabetes. 2009;58(5):1229–1236.
nephropathy. J Am Soc Nephrol. 2012;23(3):458–469. 283. Okabe J, Orlowski C, Balcerczyk A, et al. Distinguishing hyper-
262. Long J, Wang Y, Wang W, Chang BH, Danesh FR. MicroRNA-29c glycemic changes by Set7 in vascular endothelial cells. Circ Res.
is a signature microRNA under high glucose conditions that targets 2012;110(8):1067–1076.
Sprouty homolog 1, and its in vivo knockdown prevents progression 284. Keating ST, El-Osta A. Transcriptional regulation by the Set7
of diabetic nephropathy. J Biol Chem. 2011;286(13):11837–11848. lysine methyltransferase. Epigenetics. 2013;8(4):361–372.
263. Aryal B, Singh AK, Rotllan N, Price N, Fernandez-Hernando 285. Keating ST, Ziemann M, Okabe J, Khan AW, Balcerczyk A, El-
C. MicroRNAs and lipid metabolism. Curr Opin Lipidol. Osta A. Deep sequencing reveals novel Set7 networks. Cell Mol Life
2017;28(3):273–280. Sci. 2014;71(22):4471–4486.
264. Distel E, Barrett TJ, Chung K, et al. miR33 inhibition overcomes 286. Brown JD, Lin CY, Duan Q, et al. NF-kappaB directs dynamic
deleterious effects of diabetes mellitus on atherosclerosis plaque super enhancer formation in inflammation and atherogenesis. Mol
regression in mice. Circ Res. 2014;115(9):759–769. Cell. 2014;56(2):219–231.
265. Nagareddy PR, Murphy AJ, Stirzaker RA, et al. Hyperglycemia 287. Anand P, Brown JD, Lin CY, et al. BET bromodomains mediate
promotes myelopoiesis and impairs the resolution of atherosclero- transcriptional pause release in heart failure. Cell. 2013;154(3):569–
sis. Cell Metab. 2013;17(5):695–708. 582.
266. Brownlee M. Hyperglycemia-stimulated myelopoiesis causes 288. Floyd SR, Pacold ME, Huang Q, et al. The bromodomain protein
impaired regression of atherosclerosis in type 1 diabetes. Cell Brd4 insulates chromatin from DNA damage signalling. Nature.
Metab. 2013;17(5):631–633. 2013;498(7453):246–250.
267. Zhou Q, Lv D, Chen P, et al. MicroRNAs in diabetic cardiomy- 289.
National Diabetes Data Group. Diabetes in America. 2nd ed.
opathy and clinical perspectives. Front Genet. 2014;5:185. Bethesda, MD: National Institute of Diabetes and Digestive and
268. Matkovich SJ, Wang W, Tu Y, et al. MicroRNA-133a protects Kidney Diseases; 1995.
against myocardial fibrosis and modulates electrical repolarization 290.
Diabetic Retinopathy Clinical Research Network. Randomized
without affecting hypertrophy in pressure-overloaded adult hearts. trial evaluating ranibizumab plus prompt or deferred laser or triam-
Circ Res. 2010;106(1):166–175. cinolone plus prompt laser for diabetic macular edema. Ophthal-
269. Izarra A, Moscoso I, Levent E, et al. miR-133a enhances the protec- mology. 2010;117(6):1064–1077.e35.
tive capacity of cardiac progenitors cells after myocardial infarction. 291. Diabetic Retinopathy Clinical Research Network, Wells JA, Glass-
Stem Cell Rep. 2014;3(6):1029–1042. man AR, et al. Aflibercept, bevacizumab, or ranibizumab for dia-
270. Davidson EH, Levine MS. Properties of developmental gene regu- betic macular edema. N Engl J Med. 2015;372(13):1193–1203.
latory networks. Proc Natl Acad Sci U S A. 2008;105(51):20063– 292. Ferris FL 3rd, Davis MD, Aiello LM. Treatment of diabetic reti-
20066. nopathy. N Engl J Med. 1999;341:667–678.
271. Burrill DR, Silver PA. Making cellular memories. Cell. 292a Diabetic Retinopathy Clinical Research Network; Writing Com-
2010;140(1):13–18. mittee, Aiello LP, Beck RW, Bressler NM, et al. Rationale for the
272. Macarthur BD, Ma’ayan A, Lemischka IR. Systems biology of diabetic retinopathy clinical research network treatment proto-
stem cell fate and cellular reprogramming. Nat Rev Mol Cell Biol. col for center-involved diabetic macular edema. Ophthalmology.
2009;10(10):672–681. 2011;118:e5–e14.

293. Harris MI, Hadden WC, Knowler WC, Bennett PH. Prevalence of 314. Stitt AW, Gardiner TA, Archer DB. Histological and ultrastruc-
diabetes and impaired glucose tolerance and plasma glucose levels tural investigation of retinal microaneurysm development in dia-
in U.S. population aged 20-74 yr. Diabetes. 1987;36(4):523–534. betic patients. Br J Ophthalmol. 1995;79(4):362–367.
294. Klein R, Klein BE, Moss SE, Davis MD, DeMets DL. The Wis- 315. Cunha-Vaz J, Faria de Abreu JR, Campos AJ. Early break-
consin epidemiologic study of diabetic retinopathy. II. Prevalence down of the blood-retinal barrier in diabetes. Br J Ophthalmol.
and risk of diabetic retinopathy when age at diagnosis is less than 1975;59(11):649–656.
30 years. Arch Ophthalmol. 1984;102(4):520–526. 316. Early photocoagulation for diabetic retinopathy. ETDRS report
295. Klein R, Klein BE, Moss SE, Cruickshanks KJ. The Wisconsin epi- number 9. Early treatment diabetic retinopathy study research
demiologic study of diabetic retinopathy. XV. The long-term inci- group. Ophthalmology. 1991;98(suppl 5):766–785.
dence of macular edema. Ophthalmology. 1995;102(1):7–16. 317. Meyer-Schwickerath R, Pfeiffer A, Blum WF, et al. Vitreous lev-
296. Javitt JC, Canner JK, Sommer A. Cost effectiveness of current els of the insulin-like growth factors I and II, and the insulin-like
approaches to the control of retinopathy in type I diabetics. Oph- growth factor binding proteins 2 and 3, increase in neovascular eye
thalmology. 1989;96(2):255–264. disease. Studies in nondiabetic and diabetic subjects. J Clin Invest.
297. Javitt JC, Aiello LP, Bassi LJ, Chiang YP, Canner JK. Detecting 1993;92(6):2620–2625.
and treating retinopathy in patients with type I diabetes melli- 318. Aiello LP, Avery RL, Arrigg PG, et al. Vascular endothelial growth
tus. Savings associated with improved implementation of current factor in ocular fluid of patients with diabetic retinopathy and
guidelines. American Academy of Ophthalmology. Ophthalmology. other retinal disorders. N Engl J Med. 1994;331(22):1480–1487.
1991;98(10):1565–1573; discussion 1574. 319. Miller JW, Adamis AP, Aiello LP. Vascular endothelial growth fac-
298. Javitt JC, Aiello LP, Chiang Y, Ferris FL 3rd, Canner JK, Green- tor in ocular neovascularization and proliferative diabetic retinopa-
field S. Preventive eye care in people with diabetes is cost-saving to thy. Diabetes Metab Rev. 1997;13(1):37–50.
the federal government. Implications for health-care reform. Dia- 320. Ozaki H, Hayashi H, Vinores SA, Moromizato Y, Campochiaro
betes Care. 1994;17(8):909–917. PA, Oshima K. Intravitreal sustained release of VEGF causes retinal
299. Javitt JC, Aiello LP. Cost-effectiveness of detecting and treat- neovascularization in rabbits and breakdown of the blood-retinal
ing diabetic retinopathy. Ann Intern Med. 1996;124(1 Pt 2): barrier in rabbits and primates. Exp Eye Res. 1997;64(4):505–517.
164–169. 321. Aiello LP, Pierce EA, Foley ED, et al. Suppression of retinal neo-
300. Centers for Disease Control and Prevention. National diabetes sta- vascularization in vivo by inhibition of vascular endothelial growth
tistics report. National Estimate on Diabetes. 2014. factor (VEGF) using soluble VEGF-receptor chimeric proteins.
301. Kahn HA, Hiller R. Blindness caused by diabetic retinopathy. Am J Proc Natl Acad Sci U S A. 1995;92(23):10457–10461.
Ophthalmol. 1974;78(1):58–67. 322. Adamis AP, Shima DT, Tolentino MJ, et al. Inhibition of vascu-
302. Palmberg PF. Diabetic retinopathy. Diabetes. 1977;26:703–709. lar endothelial growth factor prevents retinal ischemia-associated
303. Wong TY, Mwamburi M, Klein R, et al. Rates of progression in iris neovascularization in a nonhuman primate. Arch Ophthalmol.
diabetic retinopathy during different time periods: a systematic 1996;114(1):66–71.
review and meta-analysis. Diabetes Care. 2009;32(12):2307–2313. 323. Duh EJ, Yang HS, Suzuma I, et al. Pigment epithelium-derived
304. International Diabetes Federation. Global fact sheet. In: IDF Dia- factor suppresses ischemia-induced retinal neovascularization and
betes Atlas. 8th ed. 2017. Available at: https://www.idf.org/about- VEGF-induced migration and growth. Invest Ophthalmol Vis Sci.
diabetes/what-is-diabetes/facts-figures.html. 2002;43(3):821–829.
305. Tolentino MJ, Miller JW, Gragoudas ES, Chatzistefanou K, Fer- 324. Gao BB, Clermont A, Rook S, et al. Extracellular carbonic
rara N, Adamis AP. Vascular endothelial growth factor is sufficient anhydrase mediates hemorrhagic retinal and cerebral vascu-
to produce iris neovascularization and neovascular glaucoma in a lar permeability through prekallikrein activation. Nat Med.
nonhuman primate. Arch Ophthalmol. 1996;114(8):964–970. 2007;13(2):181–188.
306. Bressler NM, Varma R, Doan QV, et al. Underuse of the health 325. Tripathi RC, Li J, Tripathi BJ, Chalam KV, Adamis AP.
care system by persons with diabetes mellitus and diabetic macular Increased level of vascular endothelial growth factor in aqueous
edema in the United States. JAMA Ophthalmol. 2014;132(2):168– humor of patients with neovascular glaucoma. Ophthalmology.
173. 1998;105(2):232–237.
307. Cogan DG, Toussaint D, Kuwabara T. Retinal vascular patterns. 326. Klein R, Klein BE, Moss SE, Davis MD, DeMets DL. The Wis-
IV. Diabetic retinopathy. Arch Ophthalmol. 1961;66:366–378. consin epidemiologic study of diabetic retinopathy. III. Prevalence
308. Bursell SE, Clermont AC, Kinsley BT, Simonson DC, Aiello LM, and risk of diabetic retinopathy when age at diagnosis is 30 or more
Wolpert HA. Retinal blood flow changes in patients with insulin- years. Arch Ophthalmol. 1984;102(4):527–532.
dependent diabetes mellitus and no diabetic retinopathy. Invest 327. Kohner EM, Aldington SJ, Stratton IM, et al. United Kingdom
Ophthalmol Vis Sci. 1996;37(5):886–897. Prospective Diabetes Study, 30: diabetic retinopathy at diagnosis
309. Sosula L, Beaumont P, Hollows FC, Jonson KM. Dilatation and of non-insulin-dependent diabetes mellitus and associated risk fac-
endothelial proliferation of retinal capillaries in streptozotocin- tors. Arch Ophthalmol. 1998;116(3):297–303.
diabetic rats: quantitative electron microscopy. Invest Ophthalmol. 328. Centers for Disease Control and Prevention. Vision Health Initia-
1972;11(11):926–935. tive (VHI) Report: Projection of diabetic retinopathy and other
310. Speiser P, Gittelsohn AM, Patz A. Studies on diabetic retinopathy. major eye diseases among people with diabetes mellitus United
3. Influence of diabetes on intramural pericytes. Arch Ophthalmol. States, 2005-2050. 2015. Available at: https://www.cdc.gov/vision-
1968;80(3):332–337. health/publications/diabetic_retinopathy.htm.
311. Barouch FC, Miyamoto K, Allport JR, et al. Integrin-mediated 329. Effect of intensive therapy on the development and progression of
neutrophil adhesion and retinal leukostasis in diabetes. Invest Oph- diabetic nephropathy in the diabetes control and complications
thalmol Vis Sci. 2000;41(5):1153–1158. trial. The diabetes control and complications (DCCT) research
312. Miyamoto K, Khosrof S, Bursell SE, et al. Vascular endothelial group. Kidney Int. 1995;47(6):1703–1720.
growth factor (VEGF)-induced retinal vascular permeability is 330. The relationship of glycemic exposure (HbA1c) to the risk of devel-
mediated by intercellular adhesion molecule-1 (ICAM-1). Am J opment and progression of retinopathy in the diabetes control and
Pathol. 2000;156(5):1733–1739. complications trial. Diabetes. 1995;44(8):968–983.
313. Miyamoto K, Khosrof S, Bursell SE, et al. Prevention of leukostasis 331. Lifetime benefits and costs of intensive therapy as practiced in
and vascular leakage in streptozotocin-induced diabetic retinopathy the diabetes control and complications trial. The Diabetes Con-
via intercellular adhesion molecule-1 inhibition. Proc Natl Acad Sci trol and Complications Trial Research Group. J Am Med Assoc.
U S A. 1999;96(19):10836–10841. 1996;276(17):1409–1415.

332. Progression of retinopathy with intensive versus conventional treat- 353. Blitzer A, Lawson W, Meyers BR, Biller HF. Patient survival
ment in the Diabetes Control and Complications Trial. Diabetes factors in paranasal sinus mucormycosis. The Laryngoscope.
control and complications trial research group. Ophthalmology. 1980;90(4):635–648.
1995;102(4):647–661. 354. Gartner S, Henkind P. Neovascularization of the iris (rubeosis iri-
333. Writing team for the diabetes control and complications trial/ dis). Surv Ophthalmol. 1978;22(5):291–312.
epidemiology of diabetes interventions and complications 355. Brown GC, Magargal LE, Schachat A, Shah H. Neovascular glau-
research group. Effect of intensive therapy on the microvascu- coma. Etiologic considerations. Ophthalmology. 1984;91(4):315–
lar complications of type 1 diabetes mellitus. J Am Med Assoc. 320.
2002;287(19):2563–2569. 356. Wand M, Dueker DK, Aiello LM, Grant WM. Effects of
334. Lachin JM, Genuth S, Cleary P, Davis MD, Nathan DM. Retinop- panretinal photocoagulation on rubeosis iridis, angle neovas-
athy and nephropathy in patients with type 1 diabetes four years cularization, and neovascular glaucoma. Am J Ophthalmol.
after a trial of intensive therapy. N Engl J Med. 2000;342(6):381– 1978;86(3):332–339.
389. 357. Aiello LM, Wand M, Liang G. Neovascular glaucoma and vitre-
335. Aiello LP. Diabetic retinopathy and other ocular findings in the ous hemorrhage following cataract surgery in patients with diabetes
diabetes control and complications trial/epidemiology of dia- mellitus. Ophthalmology. 1983;90(7):814–820.
betes interventions and complications study. Diabetes Care. 358. Simmons RJ, Dueker DK, Kimbrough RL, Aiello LM. Goniopho-
2014;37(1):17–23. tocoagulation for neovascular glaucoma. Trans Sect Ophthalmol Am
336. DCCT/EDIC Research Group, Aiello LP, Sun W, et al. Intensive Acad Ophthalmol Otolaryngol. 1977;83(1):80–89.
diabetes therapy and ocular surgery in type 1 diabetes. N Engl J 359. Mason JO 3rd, Albert MA Jr, Mays A, Vail R. Regression of neo-
Med. 2015;372(18):1722–1733. vascular iris vessels by intravitreal injection of bevacizumab. Retina.
337. Resource utilization and costs of care in the diabetes control and 2006;26(7):839–841.
complications trial. Diabetes Care. 1995;18(11):1468–1478. 359a Regillo CD, Brown GC, Savino PJ, et al. Diabetic papillopathy.
338. Kroc Collaborative Study Group. Blood glucose control and the Patient characteristics and fundus findings. Arch Ophthalmol.
evolution of diabetic retinopathy and albuminuria. A preliminary 1995;113:889–895.
multicenter trial. N Engl J Med. 1984;311(6):365–372. 360. Hyndiuk RA, Kazarian EL, Schultz RO, Seideman S. Neuro-
339. Chase HP, Jackson WE, Hoops SL, Cockerham RS, Archer PG, trophic corneal ulcers in diabetes mellitus. Arch Ophthalmol.
O’Brien D. Glucose control and the renal and retinal complica- 1977;95(12):2193–2196.
tions of insulin-dependent diabetes. J Am Med Assoc. 1989;261(8): 361. Klein BE, Klein R, Moss SE. Intraocular pressure in diabetic per-
1155–1160. sons. Ophthalmology. 1984;91(11):1356–1360.
340. Krolewski AS, Canessa M, Warram JH, et al. Predisposition to 362. Pasquale LR, Kang JH, Manson JE, Willett WC, Rosner BA,
hypertension and susceptibility to renal disease in insulin-depen- Hankinson SE. Prospective study of type 2 diabetes mellitus and
dent diabetes mellitus. N Engl J Med. 1988;318(3):140–145. risk of primary open-angle glaucoma in women. Ophthalmology.
341. Chew EY, Klein ML, Ferris FL 3rd, et al. Association of elevated 2006;113(7):1081–1086.
serum lipid levels with retinal hard exudate in diabetic retinopathy. 363. Marmor MF. Transient accommodative paralysis and hyperopia in
Early Treatment Diabetic Retinopathy Study (ETDRS) Report 22. diabetes. Arch Ophthalmol. 1973;89(5):419–421.
Arch Ophthalmol. 1996;114(9):1079–1084. 364. Klein BE, Klein R, Moss SE. Prevalence of cataracts in a popula-
342. Aiello LP, Gardner TW, King GL, et al. Diabetic retinopathy. Dia- tion-based study of persons with diabetes mellitus. Ophthalmology.
betes Care. 1998;21(1):143–156. 1985;92(9):1191–1196.
343. Four risk factors for severe visual loss in diabetic retinopathy. The 365. Ederer F, Hiller R, Taylor HR. Senile lens changes and diabetes
third report from the Diabetic Retinopathy Study. The diabetic retin two population studies. Am J Ophthalmol. 1981;91(3):381–
inopathy study research group. Arch Ophthalmol. 1979;97(4):654– 395.
655. 366. Bursell SE, Baker RS, Weiss JN, Haughton JF, Rand LI. Clini-
344. Treatment techniques and clinical guidelines for photocoagulation cal photon correlation spectroscopy evaluation of human diabetic
of diabetic macular edema. Early Treatment Diabetic Retinopathy lenses. Exp Eye Res. 1989;49(2):241–258.
Study Report Number 2. Early treatment diabetic retinopathy 367. Klein BE, Klein R, Moss SE. Incidence of cataract surgery in the
study research group. Ophthalmology. 1987;94(7):761–774. Wisconsin epidemiologic study of diabetic retinopathy. Am J Oph-
345.
Photocoagulation for diabetic macular edema: early treatment thalmol. 1995;119(3):295–300.
diabetic retinopathy study report no. 4. The early treatment dia- 368. Klein BE, Klein R, Wang Q, Moss SE. Older-onset diabetes and
betic retinopathy study research group. Int Ophthalmol Clin. lens opacities. The beaver dam eye study. Ophthalmic Epidemiol.
1987;27(4):265–272. 1995;2(1):49–55.
346. Wilkinson CP, Ferris FL 3rd, Klein RE, et al. Proposed interna- 369. Pai RP, Mitchell P, Chow VC, et al. Posttransplant cataract: les-
tional clinical diabetic retinopathy and diabetic macular edema dis- sons from kidney-pancreas transplantation. Transplantation.
ease severity scales. Ophthalmology. 2003;110(9):1677–1682. 2000;69(6):1108–1114.
347. Chew EY. A simplified diabetic retinopathy scale. Ophthalmology. 370. Dowler JG, Hykin PG, Hamilton AM. Phacoemulsification versus
2003;110(9):1675–1676. extracapsular cataract extraction in patients with diabetes. Ophthal-
348. Waite JH, Beetham WP. Visual mechanisms in diabetes mellitus: mology. 2000;107(3):457–462.
comparative study of 2002 diabetics and 437 nondiabetics for con- 371. Borrillo JL, Mittra RA, Dev S, et al. Retinopathy progression and
trol. N Engl J Med. 1935;212:367–379. visual outcomes after phacoemulsification in patients with diabetes
349. Zorrilla E, Kozak GP. Ophthalmoplegia in diabetes mellitus. Ann mellitus. Trans Am Ophthalmol Soc. 1999;97:435–445; discussion
Intern Med. 1967;67(5):968–976. 445–449.
350. Rush JA, Younge BR. Paralysis of cranial nerves III, IV, and 372. Funahashi T, Fink AI. The pathology of the bulbar conjunc-
VI. Cause and prognosis in 1,000 cases. Arch Ophthalmol. tiva in diabetes mellitus: I. Microaneurysms. Am J Ophthalmol.
1981;99(1):76–79. 1963;55:504–511.
351. Khodadoust AA, Silverstein AM, Kenyon DR, Dowling JE. Adhe- 373. Tagawa H, McMeel JW, Trempe CL. Role of the vitreous in dia-
sion of regenerating corneal epithelium. The role of basement betic retinopathy. II. Active and inactive vitreous changes. Ophthal-
membrane. Am J Ophthalmol. 1968;65(3):339–348. mology. 1986;93(9):1188–1192.
352. Schwartz JN, Donnelly EH, Klintworth GK. Ocular and orbital 374. Fleckner RA, Goldstein JH. Mucormycosis. Br J Ophthalmol.
phycomycosis. Surv Ophthalmol. 1977;22(1):3–28. 1969;53(8):542–548.

375. Writing Committee for the Diabetic Retinopathy Clinical Research 397. Grading diabetic retinopathy from stereoscopic color fundus pho-
Network, Gross JG, Glassman AR, et al. Panretinal photocoagula- tographs—an extension of the modified Airlie House classification.
tion vs intravitreous ranibizumab for proliferative diabetic retinopa- ETDRS report number 10. Early Treatment Diabetic Retinopathy
thy: a randomized trial. J Am Med Assoc. 2015;314(20):2137–2146. Study Research Group. Ophthalmology. 1991;98(suppl 5):786–
376. Photocoagulation treatment of proliferative diabetic retinopathy. 806.
Clinical application of Diabetic Retinopathy Study (DRS) find- 398. Indications for photocoagulation treatment of diabetic retinopathy:
ings, DRS Report Number 8. The diabetic retinopathy study diabetic retinopathy study report no. 14. The diabetic retinopathy
research group. Ophthalmology. 1981;88(7):583–600. study research group. Int Ophthalmol Clin. 1987;27(4):239–253.
377. Two-year course of visual acuity in severe proliferative diabetic 399. Photocoagulation treatment of proliferative diabetic retinopathy:
retinopathy with conventional management. Diabetic Reti- the second report of diabetic retinopathy study findings. Ophthal-
nopathy Vitrectomy Study (DRVS) report #1. Ophthalmology. mology. 1978;85(1):82–106.
1985;92(4):492–502. 400.
Photocoagulation for diabetic macular edema. Early treatment
378.
The effect of intensive diabetes treatment on the progression diabetic retinopathy study report number 1. Early treatment
of diabetic retinopathy in insulin-dependent diabetes mellitus. diabetic retinopathy study research group. Arch Ophthalmol.
The diabetes control and complications trial. Arch Ophthalmol. 1985;103(12):1796–1806.
1995;113(1):36–51. 401. Ferris F. Early photocoagulation in patients with either type I or
379.
Intensive blood-glucose control with sulphonylureas or insulin type II diabetes. Trans Am Ophthalmol Soc. 1996;94:505–537.
compared with conventional treatment and risk of complications 402. Gross JG, Glassman AR, Liu D, et al. Five-year outcomes of pan-
in patients with type 2 diabetes (UKPDS 33). UK Prospective Dia- retinal photocoagulation vs intravitreous ranibizumab for pro-
betes Study (UKPDS) Group. Lancet. 1998;352(9131):837–853. liferative diabetic retinopathy: a randomized clinical trial. JAMA
380. Klein R, Klein BE, Moss SE, Davis MD, DeMets DL. The Wiscon- Ophthalmol. 2018;136(10):1138–1148.
sin epidemiologic study of diabetic retinopathy. VI. Retinal photo- 403. Sivaprasad S, Prevost AT, Vasconcelos JC, et al. Clinical efficacy
coagulation. Ophthalmology. 1987;94(7):747–753. of intravitreal aflibercept versus panretinal photocoagulation for
381. Witkin SR, Klein R. Ophthalmologic care for persons with diabe- best corrected visual acuity in patients with proliferative diabetic
tes. J Am Med Assoc. 1984;251(19):2534–2537. retinopathy at 52 weeks (CLARITY): a multicentre, single-blinded,
382. American Academy of Ophthalmology. Comprehensive Adult Medi- randomised, controlled, phase 2b, non-inferiority trial. Lancet.
cal Eye Evaluation: Preferred Practice Pattern; 2015. 2017;389(10085):2193–2203.
383. American Optometric Association. Comprehensive Adult Eye and 404. Avery RL. Regression of retinal and iris neovascularization
Vision Examination: Optometric Clinical Practice Guideline. 2nd ed. after intravitreal bevacizumab (Avastin) treatment. Retina.
2015. 2006;26(3):352–354.
384. Klein R, Klein BE, Neider MW, Hubbard LD, Meuer SM, Broth- 405. Sun JK, Glassman AR, Beaulieu WT, et al. Diabetic Retinopathy
ers RJ. Diabetic retinopathy as detected using ophthalmoscopy, a Clinical Research Network. Rationale and application of the pro-
nonmydriatic camera and a standard fundus camera. Ophthalmol- tocol S anti-vascular endothelial growth factor algorithm for prolif-
ogy. 1985;92(4):485–491. erative diabetic retinopathy. Ophthalmology. 2019;126:87–95.
385. Moss SE, Klein R, Kessler SD, Richie KA. Comparison between 406. Ip MS, Domalpally A, Sun JK, Ehrlich JS. Long-term effects of
ophthalmoscopy and fundus photography in determining severity therapy with ranibizumab on diabetic retinopathy severity and
of diabetic retinopathy. Ophthalmology. 1985;92(1):62–67. baseline risk factors for worsening retinopathy. Ophthalmology.
386. Sussman EJ, Tsiaras WG, Soper KA. Diagnosis of diabetic eye dis- 2015;122(2):367–374.
ease. J Am Med Assoc. 1982;247(23):3231–3234. 407. Bressler SB, Qin H, Melia M, et al. Exploratory analysis of effect
387. Cavallerano JD, Aiello LP, Cavallerano AA, et al. Nonmydriatic of intravitreal ranibizumab or triamcinolone on worsening of dia-
digital imaging alternative for annual retinal examination in per- betic retinopathy in a randomized clinical trial. JAMA Ophthalmol.
sons with previously documented no or mild diabetic retinopathy. 2013;131(8):1033–1040.
Am J Ophthalmol. 2005;140(4):667–673. 408. Brown D. Intravitreal aflibercept injection (IAI) for moderately
388. American Diabetes Association. Standards of medical care in diabe- severe to severe nonproliferative diabetic retinopathy (NPDR):
tes—2014. Diabetes Care. 2014;37(suppl 1):S14–S80. The Phase 3 PANORAMA Study. Presented at the Association for
389. Silva PS, Cavallerano JD, Aiello LM, Aiello LP. Telemedicine and Research in Vision and Ophthalmology meeting. 2018. Poster C0289.
diabetic retinopathy: moving beyond retinal screening. Arch Oph- 409. Krzystolik MG, Strauber SF, Aiello LP, et al. Reproducibility
thalmol. 2011;129(2):236–242. of macular thickness and volume using Zeiss optical coherence
390. Klein R, Moss SE, Klein BE. New management concepts for timely tomography in patients with diabetic macular edema. Ophthalmol-
diagnosis of diabetic retinopathy treatable by photocoagulation. ogy. 2007;114(8):1520–1525.
Diabetes Care. 1987;10(5):633–638. 410. Browning DJ, Glassman AR, Aiello LP, et al. Relationship between
391. Krolewski AS, Warram JH, Rand LI, Christlieb AR, Busick EJ, optical coherence tomography-measured central retinal thick-
Kahn CR. Risk of proliferative diabetic retinopathy in juvenile- ness and visual acuity in diabetic macular edema. Ophthalmology.
onset type I diabetes: a 40-yr follow-up study. Diabetes Care. 2007;114(3):525–536.
1986;9(5):443–452. 411. Elman MJ, Bressler NM, Qin H, et al. Expanded 2-year follow-
392. Kostraba JN, Klein R, Dorman JS, et al. The epidemiology of dia- up of ranibizumab plus prompt or deferred laser or triamcinolone
betes complications study. IV. Correlates of diabetic background plus prompt laser for diabetic macular edema. Ophthalmology.
and proliferative retinopathy. Am J Epidemiol. 1991;133(4):381– 2011;118(4):609–614.
391. 412. Elman MJ, Qin H, Aiello LP, et al. Intravitreal ranibizumab
393. Sunness JS. The pregnant woman’s eye. Surv Ophthalmol. for diabetic macular edema with prompt versus deferred laser
1988;32(4):219–238. treatment: three-year randomized trial results. Ophthalmology.
394. Klein BE, Moss SE, Klein R. Effect of pregnancy on progression of 2012;119(11):2312–2318.
diabetic retinopathy. Diabetes Care. 1990;13(1):34–40. 413. LaPiana FG, Penner R. Anaphylactoid reaction to intrave-
395. White P. Diabetes mellitus in pregnancy. Clin Perinatol. nously administered fluorescein. Arch Ophthalmol. 1968;79(2):
1974;1(2):331–347. 161–162.
396. Best RM, Chakravarthy U. Diabetic retinopathy in pregnancy. Br J 414. Butner RW, McPherson AR. Adverse reactions in intravenous fluo-
Ophthalmol. 1997;81(3):249–251. rescein angiography. Ann Ophthalmol. 1983;15(11):1084–1086.

415. Wittpenn JR, Rapoza P, Sternberg P Jr, Kuwashima L, Saklad J, 433. Klein R, Moss SE, Klein BE. Is gross proteinuria a risk factor for
Patz A. Respiratory arrest following retrobulbar anesthesia. Oph- the incidence of proliferative diabetic retinopathy? Ophthalmology.
thalmology. 1986;93(7):867–870. 1993;100(8):1140–1146.
416. Chew E, Strauber S, Beck R, et al. Randomized trial of peribulbar 434. Savage S, Estacio RO, Jeffers B, Schrier RW. Urinary albumin
triamcinolone acetonide with and without focal photocoagulation excretion as a predictor of diabetic retinopathy, neuropathy, and car-
for mild diabetic macular edema: a pilot study. Ophthalmology. diovascular disease in NIDDM. Diabetes Care. 1996;19(11):1243–
2007;114(6):1190–1196. 1248.
417. Diabetic Retinopathy Clinical Research Network. A randomized 435. Cruickshanks KJ, Ritter LL, Klein R, Moss SE. The association
trial comparing intravitreal triamcinolone acetonide and focal/ of microalbuminuria with diabetic retinopathy. The Wisconsin
grid photocoagulation for diabetic macular edema. Ophthalmology. Epidemiologic Study of Diabetic Retinopathy. Ophthalmology.
2008;115(9):1447–1449, 1449.e1–e10. 1993;100(6):862–867.
418. Beck RW, Edwards AR, Aiello LP, et al. Three-year follow-up of a 436. Park JY, Kim HK, Chung YE, Kim SW, Hong SK, Lee KU. Inci-
randomized trial comparing focal/grid photocoagulation and intra- dence and determinants of microalbuminuria in Koreans with type
vitreal triamcinolone for diabetic macular edema. Arch Ophthalmol. 2 diabetes. Diabetes Care. 1998;21(4):530–534.
2009;127(3):245–251. 437. Klein R, Klein BE, Moss SE, Cruickshanks KJ. Ten-year inci-
419. Maturi RK, Glassman AR, Liu D, et al. Effect of adding dexa- dence of gross proteinuria in people with diabetes. Diabetes.
methasone to continued ranibizumab treatment in patients with 1995;44(8):916–923.
persistent diabetic macular edema: a DRCR Network Phase 2 ran- 438. Mogensen CE, Chachati A, Christensen CK, et al. Microalbu-
domized clinical trial. JAMA Ophthalmol. 2018;136(1):29–38. minuria: an early marker of renal involvement in diabetes. Uremia
420. Klein R, Klein BE, Lee KE, Cruickshanks KJ, Moss SE. The inci- Invest. 1985;9(2):85–95.
dence of hypertension in insulin-dependent diabetes. Arch Intern 439. Damsgaard EM, Froland A, Jorgensen OD, Morgensen CE.
Med. 1996;156(6):622–627. Prognostic value of urinary albumin excretion rate and other
421. Vascular complications in non-insulin dependent diabetics in Thai- risk factors in elderly diabetic patients and non-diabetic control
land. Thai multicenter research group on diabetes mellitus. Diabe- subjects surviving the first 5 years after assessment. Diabetologia.
tes Res Clin Pract. 1994;25(1):61–69. 1993;36(10):1030–1036.
422. Klein R, Klein BE, Moss SE, DeMets DL. Blood pressure and 440. Mathiesen ER, Ronn B, Storm B, Foght H, Deckert T. The natu-
hypertension in diabetes. Am J Epidemiol. 1985;122(1):75–89. ral course of microalbuminuria in insulin-dependent diabetes: a
423. Fujimoto WY, Leonetti DL, Kinyoun JL, Shuman WP, Stolov WC, 10-year prospective study. Diabet Med. 1995;12(6):482–487.
Wahl PW. Prevalence of complications among second-generation 441. Villarosa IP, Bakris GL. The appropriate blood pressure control in
Japanese-American men with diabetes, impaired glucose tolerance, diabetes (ABCD) trial. J Hum Hypertens. 1998;12(9):653–655.
or normal glucose tolerance. Diabetes. 1987;36(6):730–739. 442. Nelson RG, Knowler WC, Pettitt DJ, Hanson RL, Bennett PH.
424. Klein R, Klein BE, Moss SE, Cruickshanks KJ. The Wisconsin Epi- Incidence and determinants of elevated urinary albumin excretion
demiologic Study of Diabetic Retinopathy: XVII. The 14-year inci- in Pima Indians with NIDDM. Diabetes Care. 1995;18(2):182–
dence and progression of diabetic retinopathy and associated risk 187.
factors in type 1 diabetes. Ophthalmology. 1998;105(10):1801– 443. Gomes MB, Lucchetti MR, Gazzola H, et al. Microalbuminuria
1815. and associated clinical features among Brazilians with insulin
425. Zander E, Heinke P, Herfurth S, Reindel J, Ostermann FE, Kerner dependent diabetes mellitus. Diabetes Res Clin Pract. 1997;35(2–
W. Relations between diabetic retinopathy and cardiovascular neu- 3):143–147.
ropathy—a cross-sectional study in IDDM and NIDDM patients. 444. Predictors of the development of microalbuminuria in patients
Exp Clin Endocrinol Diabetes. 1997;105(6):319–326. with Type 1 diabetes mellitus: a seven-year prospective study.
426. Prevalence of small vessel and large vessel disease in diabetic patients The Microalbuminuria Collaborative Study Group. Diabet Med.
from 14 centres. the world health Organisation multinational study 1999;16(11):918–925.
of vascular disease in diabetics. Diabetes Drafting Group. Diabeto- 445. Davis MD, Fisher MR, Gangnon RE, et al. Risk factors for high-
logia. 1985;28(suppl):615–640. risk proliferative diabetic retinopathy and severe visual loss: early
427. Agardh CD, Agardh E, Torffvit O. The association between reti- treatment diabetic retinopathy study report #18. Invest Ophthalmol
nopathy, nephropathy, cardiovascular disease and long-term Vis Sci. 1998;39(2):233–252.
metabolic control in type 1 diabetes mellitus: a 5 year follow-up 446. Qiao Q, Keinanen-Kiukaanniemi S, Laara E. The relationship
study of 442 adult patients in routine care. Diabetes Res Clin Pract. between hemoglobin levels and diabetic retinopathy. J Clin Epide-
1997;35(2–3):113–121. miol. 1997;50(2):153–158.
428. Lopes de Faria JM, Jalkh AE, Trempe CL, McMeel JW. Diabetic 447. Friedman EA, Brown CD, Berman DH. Erythropoietin in dia-
macular edema: risk factors and concomitants. Acta Ophthalmol betic macular edema and renal insufficiency. Am J Kidney Dis.
Scand. 1999;77(2):170–175. 1995;26(1):202–208.
429. Marshall G, Garg SK, Jackson WE, Holmes DL, Chase HP. Fac- 448. Stephenson JM, Kenny S, Stevens LK, Fuller JH, Lee E. Protein-
tors influencing the onset and progression of diabetic retinopathy uria and mortality in diabetes: the WHO multinational study of
in subjects with insulin-dependent diabetes mellitus. Ophthalmol- vascular disease in diabetes. Diabet Med. 1995;12(2):149–155.
ogy. 1993;100(8):1133–1139. 449. Zimmet P, Alberti KG, Shaw J. Global and societal implications of
430. Tight blood pressure control and risk of macrovascular and micro- the diabetes epidemic. Nature. 2001;414(6865):782–787.
vascular complications in type 2 diabetes: UKPDS 38. UK Pro- 450. Ismail N, Becker B, Strzelczyk P, Ritz E. Renal disease and hyper-
spective Diabetes Study Group. BMJ. 1998;317(7160):703–713. tension in non-insulin-dependent diabetes mellitus. Kidney Int.
431. Schrier RW, Estacio RO, Jeffers B. Appropriate blood pressure con- 1999;55(1):1–28.
trol in NIDDM (ABCD) Trial. Diabetologia. 1996;39(12):1646– 451. Groop PH, Thomas MC, Moran JL, et al. The presence and sever-
1654. ity of chronic kidney disease predicts all-cause mortality in type 1
432. Klein R, Klein BE, Moss SE, et al. The Wisconsin Epidemiology diabetes. Diabetes. 2009;58(7):1651–1658.
Study of Diabetic Retinopathy: V. Proteinuria and retinopathy in a 452. Groop PH, Thomas M, Feodoroff M, Forsblom C, Harjutsalo V,
population of diabetic persons diagnosed prior to 30 years of age. FinnDiane Study Group. Excess mortality in patients with type 1
In: Friedman EA, L’Esperance FAJ, eds. Diabetic Renal-Retinal Syn- diabetes without albuminuria-separating the contribution of early
drome. 3rd ed. Orlando, FL: Grune & Stratton; 1986:245–264. and late risks. Diabetes Care. 2018;41(4):748–754.

453. Standl E, Stiegler H. Microalbuminuria in a random cohort of 474. Wanner C, Inzucchi SE, Lachin JM, et al. Empagliflozin and
recently diagnosed type 2 (non-insulin-dependent) diabetic patients progression of kidney disease in type 2 diabetes. N Engl J Med.
living in the greater Munich area. Diabetologia. 1993;36(10):1017– 2016;375(4):323–334.
1020. 475. Mauer SM, Steffes MW, Ellis EN, Sutherland DE, Brown DM,
454. Andersen AR, Christiansen JS, Andersen JK, Kreiner S, Deckert T. Goetz FC. Structural-functional relationships in diabetic nephrop-
Diabetic nephropathy in Type 1 (insulin-dependent) diabetes: an athy. J Clin Invest. 1984;74(4):1143–1155.
epidemiological study. Diabetologia. 1983;25(6):496–501. 476. Allen TJ, Cooper ME, Gilbert RE, Winikoff J, Skinni SL, Jerums
455. Parving HH, Lewis JB, Ravid M, Remuzzi G, Hunsicker LG. G. Serum total renin is increased before microalbuminuria in dia-
DEMAND investigators. Prevalence and risk factors for microal- betes. Kidney Int. 1996;50(3):902–907.
buminuria in a referred cohort of type II diabetic patients: a global 477. Niewczas MA, Gohda T, Skupien J, et al. Circulating TNF recep-
perspective. Kidney Int. 2006;69(11):2057–2063. tors 1 and 2 predict ESRD in type 2 diabetes. J Am Soc Nephrol.
456. Gatwood J, Chisholm-Burns M, Davis R, et al. Evidence of chronic 2012;23(3):507–515.
kidney disease in veterans with incident diabetes mellitus. PLoS 478. Saulnier PJ, Wheelock KM, Howell S, et al. Advanced glycation
One. 2018;13(2):e0192712. end products predict loss of renal function and correlate with
457. Thomas MC, Cooper ME, Zimmet P. Changing epidemiology of lesions of diabetic kidney disease in American Indians with type 2
type 2 diabetes mellitus and associated chronic kidney disease. Nat diabetes. Diabetes. 2016;65(12):3744–3753.
Rev Nephrol. 2016;12(2):73–81. 479. Hanssen NMJ, Scheijen J, Jorsal A, et al. Higher plasma methyl-
458. Kimmelstiel P, Wilson C. Intercapillary lesions in the glomeruli of glyoxal levels are associated with incident cardiovascular disease in
the kidney. Am J Pathol. 1936;12(1):83–98.7. individuals with type 1 diabetes: a 12-year follow-up study. Diabe-
459. Cooper ME. Pathogenesis, prevention, and treatment of diabetic tes. 2017;66(8):2278–2283.
nephropathy. Lancet. 1998;352(9123):213–219. 480. Parving HH, de Zeeuw D, Cooper ME, et al. ACE gene poly-
460. Mogensen CE, Christensen CK, Vittinghus E. The stages in dia- morphism and losartan treatment in type 2 diabetic patients with
betic renal disease. With emphasis on the stage of incipient diabetic nephropathy. J Am Soc Nephrol. 2008;19(4):771–779.
nephropathy. Diabetes. 1983;32(suppl 2):64–78. 481. Ma RC, Cooper ME. Genetics of diabetic kidney disease-from
461. Hostetter TH, Troy JL, Brenner BM. Glomerular hemodynamics the worst of nightmares to the light of dawn? J Am Soc Nephrol.
in experimental diabetes mellitus. Kidney Int. 1981;19(3):410– 2017;28(2):389–393.
415. 482. van Zuydam NR, Ahlqvist E, Sandholm N, et al. A genome-wide
462. Thomson SC, Vallon V, Blantz RC. Kidney function in early dia- association study of diabetic kidney disease in subjects with type 2
betes: the tubular hypothesis of glomerular filtration. Am J Physiol diabetes. Diabetes. 2018;67(7):1414–1427.
Renal Physiol. 2004;286(1):F8–F15. 483. Comper WD, Osicka TM, Clark M, MacIsaac RJ, Jerums G. Ear-
463. Sharma K, Jin Y, Guo J, Ziyadeh FN. Neutralization of TGF-beta lier detection of microalbuminuria in diabetic patients using a new
by anti-TGF-beta antibody attenuates kidney hypertrophy and the urinary albumin assay. Kidney Int. 2004;65(5):1850–1855.
enhanced extracellular matrix gene expression in STZ-induced dia- 484. Poulsen PL, Hansen KW, Mogensen CE. Ambulatory blood
betic mice. Diabetes. 1996;45(4):522–530. pressure in the transition from normo- to microalbumin-
464. Segev Y, Landau D, Rasch R, Flyvbjerg A, Phillip M. Growth hor- uria. A longitudinal study in IDDM patients. Diabetes.
mone receptor antagonism prevents early renal changes in non- 1994;43(10):1248–1253.
obese diabetic mice. J Am Soc Nephrol. 1999;10(11):2374–2381. 485. Mogensen CE, Keane WF, Bennett PH, et al. Prevention of dia-
465. Thomas MC, Burns WC, Cooper ME. Tubular changes in early betic renal disease with special reference to microalbuminuria. Lan-
diabetic nephropathy. Adv Chronic Kidney Dis. 2005;12(2):177– cet. 1995;346(8982):1080–1084.
186. 486. Brocco E, Fioretto P, Mauer M, et al. Renal structure and function
466. Mogensen CE, Christensen CK. Predicting diabetic nephropathy in non-insulin dependent diabetic patients with microalbuminuria.
in insulin-dependent patients. N Engl J Med. 1984;311(2):89–93. Kidney Int Suppl. 1997;63:S40–S44.
467. Anderson S, Rennke HG, Brenner BM. Therapeutic advantage 487. Caramori ML, Fioretto P, Mauer M. The need for early predictors
of converting enzyme inhibitors in arresting progressive renal dis- of diabetic nephropathy risk: is albumin excretion rate sufficient?
ease associated with systemic hypertension in the rat. J Clin Invest. Diabetes. 2000;49(9):1399–1408.
1986;77(6):1993–2000. 488. Lurbe E, Redon J, Kesani A, et al. Increase in nocturnal blood pres-
468. Lewis EJ, Hunsicker LG, Bain RP, Rohde RD. The effect of angio- sure and progression to microalbuminuria in type 1 diabetes. N
tensin-converting-enzyme inhibition on diabetic nephropathy. The Engl J Med. 2002;347(11):797–805.
collaborative study group. N Engl J Med. 1993;329(20):1456– 489. Perkins BA, Ficociello LH, Silva KH, Finkelstein DM, Warram JH,
1462. Krolewski AS. Regression of microalbuminuria in type 1 diabetes.
469. Lewis EJ, Hunsicker LG, Clarke WR, et al. Renoprotec- N Engl J Med. 2003;348(23):2285–2293.
tive effect of the angiotensin-receptor antagonist irbesartan in 490. Hovind P, Tarnow L, Rossing P, et al. Predictors for the development
patients with nephropathy due to type 2 diabetes. N Engl J Med. of microalbuminuria and macroalbuminuria in patients with type 1
2001;345(12):851–860. diabetes: inception cohort study. BMJ. 2004;328(7448):1105.
470. Heerspink HJ, Perkins BA, Fitchett DH, Husain M, Cherney 491. Bilous RW, Mauer SM, Sutherland DE, Steffes MW. Mean glo-
DZ. Sodium glucose cotransporter 2 inhibitors in the treat- merular volume and rate of development of diabetic nephropathy.
ment of diabetes mellitus: cardiovascular and kidney effects, Diabetes. 1989;38(9):1142–1147.
potential mechanisms, and clinical applications. Circulation. 492. Parving HH, Andersen AR, Smidt VM, Oxenbøll B, Edsberg B,
2016;134(10):752–772. Christiansen JS. Diabetic nephropathy and arterial hypertension.
471. Cherney DZ, Perkins BA, Soleymanlou N, et al. Renal hemody- Diabetologia. 1983;24(1):10–12.
namic effect of sodium-glucose cotransporter 2 inhibition in patients 493. Fioretto P, Steffes MW, Sutherland DE, Goetz FC, Mauer M.
with type 1 diabetes mellitus. Circulation. 2014;129(5):587–597. Reversal of lesions of diabetic nephropathy after pancreas trans-
472. Fioretto P, Zambon A, Rossato M, Busetto L, Vettor R. SGLT2 plantation. N Engl J Med. 1998;339(2):69–75.
inhibitors and the diabetic kidney. Diabetes Care. 2016;39(suppl 494. Vora JP, Leese GP, Peters JR, Owens DR. Longitudinal evaluation
2):S165–S171. of renal function in non-insulin-dependent diabetic patients with
473. Neal B, Perkovic V, Mahaffey KW, et al. Canagliflozin and car- early nephropathy: effects of angiotensin-converting enzyme inhi-
diovascular and renal events in type 2 diabetes. N Engl J Med. bition. J Diabetes Complications. 1996;10(2):88–93.
2017;377(7):644–657.

495. Lane PH, Steffes MW, Mauer SM. Glomerular structure in IDDM 517. Hricik DE, Browning PJ, Kopelman R, Goorno WE, Madias
women with low glomerular filtration rate and normal urinary NE, Dzau VJ. Captopril-induced functional renal insufficiency in
albumin excretion. Diabetes. 1992;41(5):581–586. patients with bilateral renal-artery stenoses or renal-artery stenosis
496. Tsalamandris C, Allen TJ, Gilbert RE, et al. Progressive decline in in a solitary kidney. N Engl J Med. 1983;308(7):373–376.
renal function in diabetic patients with and without albuminuria. 518. Griffin MD, Bergstralhn EJ, Larson TS. Renal papillary necrosis—a
Diabetes. 1994;43(5):649–655. sixteen-year clinical experience. J Am Soc Nephrol. 1995;6(2):248–
497. Ekinci EI, Jerums G, Skene A, et al. Renal structure in normo- 256.
albuminuric and albuminuric patients with type 2 diabetes and 519. Davies PJ. Beethoven’s nephropathy and death: discussion paper. J
impaired renal function. Diabetes Care. 2013;36(11):3620–3626. R Soc Med. 1993;86(3):159–161.
498. Wiseman MJ, Mangili R, Alberetto M, Keen H, Viberti G. Glo- 520. Kelly DJ, Wilkinson-Berka JL, Allen TJ, Cooper ME, Skinner
merular response mechanisms to glycemic changes in insulin- SL. A new model of diabetic nephropathy with progressive renal
dependent diabetics. Kidney Int. 1987;31(4):1012–1018. impairment in the transgenic (mRen-2)27 rat (TGR). Kidney Int.
499. Forbes JM, Cooper ME. Mechanisms of diabetic complications. 1998;54(2):343–352.
Physiol Rev. 2013;93(1):137–188. 521. DeFronzo RA. Hyperkalemia and hyporeninemic hypoaldosteron-
500. Kiritoshi S, Nishikawa T, Sonoda K, et al. Reactive oxygen spe- ism. Kidney Int. 1980;17(1):118–134.
cies from mitochondria induce cyclooxygenase-2 gene expression 522. Lush DJ, King JA, Fray JC. Pathophysiology of low renin syn-
in human mesangial cells: potential role in diabetic nephropathy. dromes: sites of renal renin secretory impairment and prorenin
Diabetes. 2003;52(10):2570–2577. overexpression. Kidney Int. 1993;43(5):983–999.
501. Forbes JM, Cooper ME, Oldfield MD, Thomas MC. Role of 523. Nadler JL, Lee FO, Hsueh W, Horton R. Evidence of prostacyclin
advanced glycation end products in diabetic nephropathy. J Am Soc deficiency in the syndrome of hyporeninemic hypoaldosteronism.
Nephrol. 2003;14(8 suppl 3):S254–S258. N Engl J Med. 1986;314(16):1015–1020.
502. Tuttle KR, Bakris GL, Toto RD, McGill JB, Hu K, Anderson PW. 524. Parfrey PS, Griffiths SM, Barrett BJ, et al. Contrast material-
The effect of ruboxistaurin on nephropathy in type 2 diabetes. Dia- induced renal failure in patients with diabetes mellitus, renal insuf-
betes Care. 2005;28(11):2686–2690. ficiency, or both. A prospective controlled study. N Engl J Med.
503. Gorin Y, Block K, Hernandez J, et al. Nox4 NAD(P)H oxidase 1989;320(3):143–149.
mediates hypertrophy and fibronectin expression in the diabetic 525. Weisberg LS, Kurnik PB, Kurnik BR. Risk of radiocontrast
kidney. J Biol Chem. 2005;280(47):39616–39626. nephropathy in patients with and without diabetes mellitus. Kidney
504. Cooper ME. Interaction of metabolic and haemodynamic factors Int. 1994;45(1):259–265.
in mediating experimental diabetic nephropathy. Diabetologia. 526. Schieda N, Blaichman JI, Costa AF, et al. Gadolinium-based con-
2001;44(11):1957–1972. trast agents in kidney disease: comprehensive review and clinical
505. Thomas MC, Tikellis C, Burns WM, et al. Interactions between practice guideline issued by the Canadian Association of Radiolo-
renin angiotensin system and advanced glycation in the kidney. J gists. Can Assoc Radiol J. 2018;69(2):136–150.
Am Soc Nephrol. 2005;16(10):2976–2984. 527. Nijssen EC, Rennenberg RJ, Nelemans PJ, et al. Prophylactic hydra-
506. Brenner BM, Cooper ME, de Zeeuw D, et al. Effects of losartan on tion to protect renal function from intravascular iodinated contrast
renal and cardiovascular outcomes in patients with type 2 diabetes material in patients at high risk of contrast-induced nephropathy
and nephropathy. N Engl J Med. 2001;345(12):861–869. (AMACING): a prospective, randomised, phase 3, controlled, open-
507. Cooper ME, Mundel P, Boner G. Role of nephrin in renal disease label, non-inferiority trial. Lancet. 2017;389(10076):1312–1322.
including diabetic nephropathy. Semin Nephrol. 2002;22(5):393–398. 528. Tepel M, van der Giet M, Schwarzfeld C, Laufer U, Liermann
508. Andersen S, Blouch K, Bialek J, Deckert M, Parving HH, Myers D, Zidek W. Prevention of radiographic-contrast-agent-induced
BD. Glomerular permselectivity in early stages of overt diabetic reductions in renal function by acetylcysteine. N Engl J Med.
nephropathy. Kidney Int. 2000;58(5):2129–2137. 2000;343(3):180–184.
509. Wassef L, Langham RG, Kelly DJ. Vasoactive renal factors 529. Mogensen CE. Long-term antihypertensive treatment
and the progression of diabetic nephropathy. Curr Pharm Des. inhibiting progression of diabetic nephropathy. Br Med J.
2004;10(27):3373–3384. 1982;285(6343):685–688.
510. Mogensen CE, Neldam S, Tikkanen I, et al. Randomised controlled 530. Parving HH, Andersen AR, Smidt UM, Hommel E, Mathiesen ER,
trial of dual blockade of renin-angiotensin system in patients with Svendsen PA. Effect of antihypertensive treatment on kidney func-
hypertension, microalbuminuria, and non-insulin dependent dia- tion in diabetic nephropathy. Br Med J. 1987;294(6585):1443–
betes: the candesartan and lisinopril microalbuminuria (CALM) 1447.
study. BMJ. 2000;321(7274):1440–1444. 531. DCCT/EDIC Research Group. Effect of intensive diabetes treat-
511. Parving HH, Brenner BM, McMurray JJ, et al. Cardiorenal end ment on albuminuria in type 1 diabetes: long-term follow-up of
points in a trial of aliskiren for type 2 diabetes. N Engl J Med. the Diabetes Control and Complications Trial and Epidemiology
2012;367(23):2204–2213. of Diabetes Interventions and Complications study. Lancet Diabe-
512. Gilbert RE, Cooper ME. The tubulointerstitium in progressive diates Endocrinol. 2014;2(10):793–800.
betic kidney disease: more than an aftermath of glomerular injury? 532. Patel A, MacMahon S, Chalmers J, et al. Intensive blood glucose
Kidney Int. 1999;56(5):1627–1637. control and vascular outcomes in patients with type 2 diabetes. N
513. Pagtalunan ME, Miller PL, Jumping-Eagle S, et al. Podocyte loss Engl J Med. 2008;358(24):2560–2572.
and progressive glomerular injury in type II diabetes. J Clin Invest. 533. Perkovic V, Heerspink HL, Chalmers J, et al. Intensive glucose
1997;99(2):342–348. control improves kidney outcomes in patients with type 2 diabetes.
514. Oldfield MD, Bach LA, Forbes JM, et al. Advanced glycation end Kidney Int. 2013;83(3):517–523.
products cause epithelial-myofibroblast transdifferentiation via 534. Zoungas S, Chalmers J, Neal B, et al. Follow-up of blood-pressure
the receptor for advanced glycation end products (RAGE). J Clin lowering and glucose control in type 2 diabetes. N Engl J Med.
Invest. 2001;108(12):1853–1863. 2014;371(15):1392–1406.
515. Gilbert RE, Cooper ME, Krum H. Drug administration in 535. Gerstein HC, Miller ME, Byington RP, et al. Effects of
patients with diabetes mellitus. Safety considerations. Drug Saf. intensive glucose lowering in type 2 diabetes. N Engl J Med.
1998;18(6):441–455. 2008;358(24):2545–2559.
516. Salifu MO, Haria DM, Badero O, Aytug S, McFarlane SI. Chal- 536. Björck S. Clinical trials in overt diabetic nephropathy. In: Mogensen
lenges in the diagnosis and management of renal artery stenosis. CE, ed. The Kidney and Hypertension in Diabetes Mellitus. 3rd ed.
Curr Hypertens Rep. 2005;7(3):219–227. London: Kluwer Academic; 1996:375–384.

537. Parving HH. Renoprotection in diabetes: genetic and non-genetic 557. Parving HH, Lehnert H, Brochner-Mortensen J, Gomis R, Ander-
risk factors and treatment. Diabetologia. 1998;41(7):745–759. sen S, Arner P. The effect of irbesartan on the development of dia-
538. Duckworth W, Abraira C, Moritz T, et al. Glucose control and betic nephropathy in patients with type 2 diabetes. N Engl J Med.
vascular complications in veterans with type 2 diabetes. N Engl J 2001;345(12):870–878.
Med. 2009;360(2):129–139. 558. Parving HH, Persson F, Lewis JB, Lewis EJ, Hollenberg NK.
539. Kahn SE, Cooper ME, Del Prato S. Pathophysiology and treatment Aliskiren combined with losartan in type 2 diabetes and nephropa-
of type 2 diabetes: perspectives on the past, present, and future. thy. N Engl J Med. 2008;358(23):2433–2446.
Lancet. 2014;383(9922):1068–1083. 559. Epstein M. Aldosterone as a mediator of progressive renal dis-
540. Mosenzon O, Leibowitz G, Bhatt DL, et al. Effect of saxagliptin ease: pathogenetic and clinical implications. Am J Kidney Dis.
on renal outcomes in the SAVOR-TIMI 53 trial. Diabetes Care. 2001;37(4):677–688.
2017;40(1):69–76. 560. Schjoedt KJ, Rossing K, Juhl TR, et al. Beneficial impact of spi-
541. Cooper ME, Perkovic V, McGill JB, et al. Kidney disease end ronolactone in diabetic nephropathy. Kidney Int. 2005;68(6):2829–
points in a pooled analysis of individual patient-level data from a 2836.
large clinical trials program of the dipeptidyl peptidase 4 inhibitor 561. Bakris GL, Agarwal R, Chan JC, et al. Effect of finerenone on albu-
linagliptin in type 2 diabetes. Am J Kidney Dis. 2015;66(3):441– minuria in patients with diabetic nephropathy: a randomized clini-
449. cal trial. J Am Med Assoc. 2015;314(9):884–894.
542. Rosenstock J, Perkovic V, Alexander JH, et al. Rationale, design, 562. Pedrini MT, Levey AS, Lau J, Chalmers TC, Wang PH. The
and baseline characteristics of the CArdiovascular safety and Renal effect of dietary protein restriction on the progression of diabetic
Microvascular outcomE study with LINAgliptin (CARMELINA®): and nondiabetic renal diseases: a meta-analysis. Ann Intern Med.
a randomized, double-blind, placebo-controlled clinical trial in 1996;124(7):627–632.
patients with type 2 diabetes and high cardio-renal risk. Cardiovasc 563. Hansen HP, Tauber-Lassen E, Jensen BR, Parving HH. Effect of
Diabetol. 2018;17(1):39. protein restriction on prognosis in type 1 diabetic patients with
543. Marso SP, Daniels GH, Brown-Frandsen K, et al. Liraglutide diabetic nephropathy. Kidney Int. 2002;62(1):220–228.
and cardiovascular outcomes in type 2 diabetes. N Engl J Med. 564. Waugh NR, Robertson AM. Protein restriction for diabetic renal
2016;375(4):311–322. disease. Cochrane Database Syst Rev. 2000;(2):CD002181.
544. Marso SP, Bain SC, Consoli A, et al. Semaglutide and cardiovas- 565. Jandeleit-Dahm K, Cao Z, Cox AJ, Kelly DJ, Gilbert RE, Cooper
cular outcomes in patients with type 2 diabetes. N Engl J Med. ME. Role of hyperlipidemia in progressive renal disease: focus on
2016;375(19):1834–1844. diabetic nephropathy. Kidney Int Suppl. 1999;71:S31–S36.
545. Mann JFE, Orsted DD, Brown-Frandsen K, et al. Liraglutide and 566. Fried LF, Forrest KY, Ellis D, Chang Y, Silvers N, Orchard TJ.
renal outcomes in type 2 diabetes. N Engl J Med. 2017;377(9):839– Lipid modulation in insulin-dependent diabetes mellitus: effect on
848. microvascular outcomes. J Diabetes Complications. 2001;15(3):113–
546. Wanner C, Lachin JM, Inzucchi SE, et al. Empagliflozin and clini- 119.
cal outcomes in patients with type 2 diabetes mellitus, established 567. Keech A, Simes RJ, Barter P, et al. Effects of long-term fenofibrate
cardiovascular disease, and chronic kidney disease. Circulation. therapy on cardiovascular events in 9795 people with type 2 diabe-
2018;137(2):119–129. tes mellitus (the FIELD study): randomised controlled trial. Lancet.
547. Chobanian AV, Bakris GL, Black HR, et al. Seventh report of the 2005;366(9500):1849–1861.
joint national Committee on prevention, detection, evaluation, and 568. Davis TM, Ting R, Best JD, et al. Effects of fenofibrate on renal
treatment of high blood pressure. Hypertension. 2003;42(6):1206– function in patients with type 2 diabetes mellitus: the Fenofibrate
1252. Intervention and Event Lowering in Diabetes (FIELD) Study. Dia-
548. de Galan BE, Perkovic V, Ninomiya T, et al. Lowering blood pres- betologia. 2011;54(2):280–290.
sure reduces renal events in type 2 diabetes. J Am Soc Nephrol. 569. Collins R, Armitage J, Parish S, Sleigh P, Peto R. MRC/BHF Heart
2009;20(4):883–892. Protection Study of cholesterol-lowering with simvastatin in 5963
549. Strippoli GF, Craig M, Craig JC. Antihypertensive agents for people with diabetes: a randomised placebo-controlled trial. Lancet.
preventing diabetic kidney disease. Cochrane Database Syst Rev. 2003;361(9374):2005–2016.
2005;4:CD004136. 570. Athyros VG, Papageorgiou AA, Elisaf M, Mikhailidis DP. Statins
550. ACE. Inhibitors in Diabetic Nephropathy Trialist Group. Should and renal function in patients with diabetes mellitus. Curr Med Res
all patients with type 1 diabetes mellitus and microalbuminuria Opin. 2003;19(7):615–617.
receive angiotensin-converting enzyme inhibitors? A meta-analysis 571. Qin X, Dong H, Fang K, Lu F. The effect of statins on renal out-
of individual patient data. Ann Intern Med. 2001;134(5):370–379. comes in patients with diabetic kidney disease: a systematic review
551. Kvetny J, Gregersen G, Pedersen RS. Randomized placebo-con- and meta-analysis. Diabetes Metab Res Rev. 2017;33(6).
trolled trial of perindopril in normotensive, normoalbuminuric 572. Shen X, Zhang Z, Zhang X, et al. Efficacy of statins in patients with
patients with type 1 diabetes mellitus. QJM. 2001;94(2):89–94. diabetic nephropathy: a meta-analysis of randomized controlled tri-
552. Mauer M, Zinman B, Gardiner R, et al. Renal and retinal effects als. Lipids Health Dis. 2016;15(1):179.
of enalapril and losartan in type 1 diabetes. N Engl J Med. 573. Su X, Zhang L, Lv J, et al. Effect of statins on kidney disease out-
2009;361(1):40–51. comes: a systematic review and meta-analysis. Am J Kidney Dis.
553. Bilous R, Chaturvedi N, Sjolie AK, et al. Effect of candesartan on 2016;67(6):881–892.
microalbuminuria and albumin excretion rate in diabetes: three 574. Colhoun HM, Betteridge DJ, Durrington PN, et al. Primary
randomized trials. Ann Intern Med. 2009;151(1):11–20; W3–W4. prevention of cardiovascular disease with atorvastatin in type
554. Arauz-Pacheco C, Parrott MA, Raskin P. Treatment of hyperten- 2 diabetes in the Collaborative Atorvastatin Diabetes Study
sion in adults with diabetes. Diabetes Care. 2003;26(suppl 1):S80– (CARDS): multicentre randomised placebo-controlled trial. Lan-
S82. cet. 2004;364(9435):685–696.
555. Barnett AH, Bain SC, Bouter P, et al. Angiotensin-receptor block- 575. Thomas MC, Cooper ME, Tsalamandris C, MacIsaac R, Jerums G.
ade versus converting-enzyme inhibition in type 2 diabetes and Anemia with impaired erythropoietin response in diabetic patients.
nephropathy. N Engl J Med. 2004;351(19):1952–1961. Arch Intern Med. 2005;165(4):466–469.
556. Mann JF, Schmieder RE, McQueen M, et al. Renal outcomes with 576. Jungers P, Choukroun G, Oualim Z, Robino C, Nguyen AT, Man
telmisartan, ramipril, or both, in people at high vascular risk (the NK. Beneficial influence of recombinant human erythropoietin
ONTARGET study): a multicentre, randomised, double-blind, therapy on the rate of progression of chronic renal failure in predi-
controlled trial. Lancet. 2008;372(9638):547–553. alysis patients. Nephrol Dial Transplant. 2001;16(2):307–312.

577. Mix TC, Brenner RM, Cooper ME, et al. Rationale—Trial to 597. Weisman A, Bril V, Ngo M, et al. Identification and prediction of
Reduce Cardiovascular Events with Aranesp Therapy (TREAT): diabetic sensorimotor polyneuropathy using individual and simple
evolving the management of cardiovascular risk in patients with combinations of nerve conduction study parameters. PLoS One.
chronic kidney disease. Am Heart J. 2005;149(3):408–413. 2013;8(3). e58783.
578. Pfeffer MA, Burdmann EA, Chen CY, et al. A trial of darbepoetin 598. Shabeeb D, Najafi M, Hasanzadeh G, Hadian MR, Musa AE,
alfa in type 2 diabetes and chronic kidney disease. N Engl J Med. Shirazi A. Electrophysiological measurements of diabetic periph-
2009;361(21):2019–2032. eral neuropathy: a systematic review. Diabetes Metab Syndr.
579. Gupta N, Wish JB. Hypoxia-inducible factor prolyl hydroxylase 2018;12(4):591–600.
inhibitors: a potential new treatment for anemia in patients with 599. Callaghan BC, Gao L, Li Y, et al. Diabetes and obesity are the main
CKD. Am J Kidney Dis. 2017;69(6):815–826. metabolic drivers of peripheral neuropathy. Ann Clin Transl Neurol.
580. Tuttle KR, McGill JB, Haney DJ, Lin TE, Anderson PW. Kidney 2018;5(4):397–405.
outcomes in long-term studies of ruboxistaurin for diabetic eye dis- 600. Callaghan BC, Xia R, Reynolds E, et al. Association between meta-
ease. Clin J Am Soc Nephrol. 2007;2(4):631–636. bolic syndrome components and polyneuropathy in an obese pop-
581. Achour A, Kacem M, Dibej K, Skhiri H, Bouraoui S, El May M. ulation. JAMA Neurol. 2016;73(12):1468–1476.
One year course of oral sulodexide in the management of diabetic 601. Callaghan BC, Xia R, Banerjee M, et al. Metabolic syndrome com-
nephropathy. J Nephrol. 2005;18(5):568–574. ponents are associated with symptomatic polyneuropathy indepen-
582. Wenzel RR, Littke T, Kuranoff S, et al. Avosentan reduces albumin dent of glycemic status. Diabetes Care. 2016;39(5):801–807.
excretion in diabetics with macroalbuminuria. J Am Soc Nephrol. 602. Feldman EL, Nave KA, Jensen TS, Bennett DLH. New horizons in
2009;20(3):655–664. diabetic neuropathy: mechanisms, bioenergetics, and pain. Neuron.
583. Kohan DE, Pritchett Y, Molitch M, et al. Addition of atrasentan to 2017;93(6):1296–1313.
renin-angiotensin system blockade reduces albuminuria in diabetic 603. Alleman CJ, Westerhout KY, Hensen M, et al. Humanistic
nephropathy. J Am Soc Nephrol. 2011;22(4):763–772. and economic burden of painful diabetic peripheral neuropa-
584. Pergola PE, Raskin P, Toto RD, et al. Bardoxolone methyl and thy in Europe: a review of the literature. Diabetes Res Clin Pract.
kidney function in CKD with type 2 diabetes. N Engl J Med. 2015;109(2):215–225.
2011;365(4):327–336. 604. Strotmeyer ES, Cauley JA, Schwartz AV, et al. Reduced periph-
585. de Zeeuw D, Akizawa T, Audhya P, et al. Bardoxolone methyl in eral nerve function is related to lower hip BMD and calcaneal
type 2 diabetes and stage 4 chronic kidney disease. N Engl J Med. QUS in older white and black adults: the Health, Aging, and
2013;369(26):2492–2503. Body Composition Study. J Bone Miner Res. 2006;21(11):
586. Chin MP, Bakris GL, Block GA, et al. Bardoxolone methyl 1803–1810.
improves kidney function in patients with chronic kidney disease 605. Martins-Mendes D, Monteiro-Soares M, Boyko EJ, et al. The
stage 4 and type 2 diabetes: post-hoc analyses from Bardoxolone independent contribution of diabetic foot ulcer on lower extrem-
Methyl Evaluation in Patients with Chronic Kidney Disease and ity amputation and mortality risk. J Diabetes Complications.
Type 2 Diabetes Study. Am J Nephrol. 2018;47(1):40–47. 2014;28(5):632–638.
587. Gallwitz B. Safety and efficacy of linagliptin in type 2 diabetes 606. Jeon BJ, Choi HJ, Kang JS, Tak MS, Park ES. Comparison of five
patients with common renal and cardiovascular risk factors. Ther systems of classification of diabetic foot ulcers and predictive fac-
Adv Endocrinol Metab. 2013;4(3):95–105. tors for amputation. Int Wound J. 2017;14(3):537–545.
588. Jungers P, Zingraff J, Page B, Albouze G, Hannedouche T, Man NK. 607. Walsh JW, Hoffstad OJ, Sullivan MO, Margolis DJ. Association of
Detrimental effects of late referral in patients with chronic renal fail- diabetic foot ulcer and death in a population-based cohort from the
ure: a case-control study. Kidney Int Suppl. 1993;41:S170–S173. United Kingdom. Diabet Med. 2016;33(11):1493–1498.
589. Pirson Y. The diabetic patient with ESRD: how to select the modality 608. Zhang P, Lu J, Jing Y, Tang S, Zhu D, Bi Y. Global epidemiology
of renal replacement. Nephrol Dial Transplant. 1996;11(8):1511–1513. of diabetic foot ulceration: a systematic review and meta-analysis
590. Zinman B, Lachin JM, Inzucchi SE. Empagliflozin, cardiovas- (dagger). Ann Med. 2017;49(2):106–116.
cular outcomes, and mortality in type 2 diabetes. N Engl J Med. 609. Holzer SE, Camerota A, Martens L, Cuerdon T, Crystal-Peters J,
2016;374(11):1094. Zagari M. Costs and duration of care for lower extremity ulcers in
591. Packer M, Claggett B, Lefkowitz MP, et al. Effect of neprilysin patients with diabetes. Clin Ther. 1998;20(1):169–181.
inhibition on renal function in patients with type 2 diabetes and 610. Caputo GM, Cavanagh PR, Ulbrecht JS, Gibbons GW, Karchmer
chronic heart failure who are receiving target doses of inhibitors of AW. Assessment and management of foot disease in patients with
the renin-angiotensin system: a secondary analysis of the PARA- diabetes. N Engl J Med. 1994;331(13):854–860.
DIGM-HF trial. Lancet Diabetes Endocrinol. 2018;6(7):547–554. 611. Belatti DA, Phisitkul P. Declines in lower extremity amputa-
592. Seferovic JP, Claggett B, Seidelmann SB, et al. Effect of sacubi- tion in the US Medicare population, 2000-2010. Foot Ankle Int.
tril/valsartan versus enalapril on glycaemic control in patients with 2013;34(7):923–931.
heart failure and diabetes: a post-hoc analysis from the PARA- 612. Driver VR, Fabbi M, Lavery LA, Gibbons G. The costs of diabetic
DIGM-HF trial. Lancet Diabetes Endocrinol. 2017;5(5):333–340. foot: the economic case for the limb salvage team. J Vasc Surg.
593. Oellgaard J, Gaede P, Rossing P, Persson F, Parving HH, Pedersen 2010;52(suppl 3):17s–22s.
O. Intensified multifactorial intervention in type 2 diabetics with 613. Freeman R. Diabetic autonomic neuropathy. Handb Clin Neurol.
microalbuminuria leads to long-term renal benefits. Kidney Int. 2014;126:63–79.
2017;91(4):982–988. 614. Bakkers M, Faber CG, Hoeijmakers JG, Lauria G, Merkies IS.
594. Gaede P, Vedel P, Larsen N, Jensen GV, Parving HH, Pedersen O. Small fibers, large impact: quality of life in small-fiber neuropathy.
Multifactorial intervention and cardiovascular disease in patients Muscle Nerve. 2014;49(3):329–336.
with type 2 diabetes. N Engl J Med. 2003;348(5):383–393. 615. Malavige LS, Jayaratne SD, Kathriarachchi ST, Sivayogan S,
595. International Diabetes Federation. IDF clinical practice recom- Ranasinghe P, Levy JC. Erectile dysfunction is a strong predictor
mendations for managing Type 2 diabetes in primary care. 2017. of poor quality of life in men with Type 2 diabetes mellitus. Diabet
Available at: www.idf.org/managing-type2-diabetes. Med. 2014;31(6):699–706.
596. Pirart J. [Diabetes mellitus and its degenerative complications: 616. Li HJ, Bai WJ, Dai YT, Xu WP, Wang CN, Li HZ. An analysis of
a prospective study of 4,400 patients observed between 1947 treatment preferences and sexual quality of life outcomes in female
and 1973 (3rd and last part) (author’s transl)]. Diabete Metab. partners of Chinese men with erectile dysfunction. Asian J Androl.
1977;3(4):245–256. 2016;18(5):773–779.

617. Sadosky A, Mardekian J, Parsons B, Hopps M, Bienen EJ, Mark- 637. Callaghan BC, Price RS, Chen KS, Feldman EL. The importance
man J. Healthcare utilization and costs in diabetes relative to the of rare subtypes in diagnosis and treatment of peripheral neuropa-
clinical spectrum of painful diabetic peripheral neuropathy. J Dia- thy: a review. JAMA Neurol. 2015;72(12):1510–1518.
betes Complications. 2015;29(2):212–217. 638. Tesfaye S, Boulton AJ, Dyck PJ, et al. Diabetic neuropathies:
618. Callaghan BC, Feldman EL. Painful diabetic neuropathy: many update on definitions, diagnostic criteria, estimation of severity,
similarly effective therapies with widely dissimilar costs. Ann Intern and treatments. Diabetes Care. 2010;33(10):2285–2293.
Med. 2014;161(9):674–675. 639. Bril V. NIS-LL: the primary measurement scale for clinical
619. Adiewere P, Gillis RB, Imran Jiwani S, Meal A, Shaw I, Adams trial endpoints in diabetic peripheral neuropathy. Eur Neurol.
GG. A systematic review and meta-analysis of patient education 1999;41(suppl 1):8–13.
in preventing and reducing the incidence or recurrence of adult 640. Feldman EL, Stevens MJ, Thomas PK, Brown MB, Canal N,
diabetes foot ulcers (DFU). Heliyon. 2018;4(5):e00614. Greene DA. A practical two-step quantitative clinical and electro-
620. Javed S, Petropoulos IN, Tavakoli M, Malik RA. Clinical and diag- physiological assessment for the diagnosis and staging of diabetic
nostic features of small fiber damage in diabetic polyneuropathy. neuropathy. Diabetes Care. 1994;17(11):1281–1289.
Handb Clin Neurol. 2014;126:275–290. 641. Bril V, Perkins BA. Validation of the Toronto clinical scoring system
621. Terkelsen AJ, Karlsson P, Lauria G, Freeman R, Finnerup NB, for diabetic polyneuropathy. Diabetes Care. 2002;25(11):2048–
Jensen TS. The diagnostic challenge of small fibre neuropathy: 2052.
clinical presentations, evaluations, and causes. Lancet Neurol. 642. Cornblath DR, Chaudhry V, Carter K, et al. Total neuropathy
2017;16(11):934–944. score: validation and reliability study. Neurology. 1999;53(8):1660–
622. Hinder LM, Vivekanandan-Giri A, McLean LL, Pennathur S, Feld- 1664.
man EL. Decreased glycolytic and tricarboxylic acid cycle interme- 643. Singleton JR, Bixby B, Russell JW, et al. The Utah Early Neuropa-
diates coincide with peripheral nervous system oxidative stress in a thy Scale: a sensitive clinical scale for early sensory predominant
murine model of type 2 diabetes. J Endocrinol. 2013;216(1):1–11. neuropathy. J Peripher Nerv Syst. 2008;13(3):218–227.
623. Sas KM, Kayampilly P, Byun J, et al. Tissue-specific metabolic 644. Young MJ, Boulton AJ, MacLeod AF, Williams DR, Sonksen PH.
reprogramming drives nutrient flux in diabetic complications. JCI A multicentre study of the prevalence of diabetic peripheral neu-
Insight. 2016;1(15):e86976. ropathy in the United Kingdom hospital clinic population. Diabe-
624. Sas KM, Lin J, Rajendiran TM, et al. Shared and distinct lipid- tologia. 1993;36(2):150–154.
lipid interactions in plasma and affected tissues in a diabetic mouse 645. Callaghan BC, Little AA, Feldman EL, Hughes RA. Enhanced
model. J Lipid Res. 2018;59(2):173–183. glucose control for preventing and treating diabetic neuropathy.
625. Viader A, Sasaki Y, Kim S, et al. Aberrant Schwann cell lipid Cochrane Database Syst Rev. 2012;(6):CD007543.
metabolism linked to mitochondrial deficits leads to axon degen- 646. Pop-Busui R, Boulton AJ, Feldman EL, et al. Diabetic neuropathy:
eration and neuropathy. Neuron. 2013;77(5):886–898. a position statement by the American Diabetes Association. Diabe-
626. Jia L, Chopp M, Wang L, Lu X, Szalad A, Zhang ZG. Exosomes tes Care. 2017;40(1):136–154.
derived from high-glucose-stimulated Schwann cells promote 647. Cortez M, Singleton JR, Smith AG. Glucose intolerance, metabolic
development of diabetic peripheral neuropathy. FASEB J. 2018. syndrome, and neuropathy. Handb Clin Neurol. 2014;126:109–
fj201800597R. 122.
627. Beirowski B, Babetto E, Golden JP, et al. Metabolic regulator 648. Gibbons CH. Treatment-induced neuropathy of diabetes. Curr
LKB1 is crucial for Schwann cell-mediated axon maintenance. Nat Diab Rep. 2017;17(12):127.
Neurosci. 2014;17(10):1351–1361. 649. Gibbons CH, Freeman R. Treatment-induced neuropathy of
628. Callaghan BC, Cheng HT, Stables CL, Smith AL, Feldman EL. diabetes: an acute, iatrogenic complication of diabetes. Brain.
Diabetic neuropathy: clinical manifestations and current treat- 2015;138(Pt 1):43–52.
ments. Lancet Neurol. 2012;11(6):521–534. 650. Singleton JR, Smith AG, Marcus RL. Exercise as therapy for dia-
629. Kim B, Feldman EL. Insulin resistance in the nervous system. betic and prediabetic neuropathy. Curr Diab Rep. 2015;15(12):120.
Trends Endocrinol Metab. 2012;23(3):133–141. 651. Bril V, England J, Franklin GM, et al. Evidence-based guideline:
630. Grote CW, Groover AL, Ryals JM, Geiger PC, Feldman EL, treatment of painful diabetic neuropathy: report of the Ameri-
Wright DE. Peripheral nervous system insulin resistance in ob/ob can Academy of Neurology, the American association of Neu-
mice. Acta Neuropathol Commun. 2013;1:15. romuscular and Electrodiagnostic medicine, and the American
631. England JD, Gronseth GS, Franklin G, et al. Practice Parameter: Academy of physical medicine and Rehabilitation. Neurology.
evaluation of distal symmetric polyneuropathy: role of autonomic 2011;76(20):1758–1765.
testing, nerve biopsy, and skin biopsy (an evidence-based review). 652. Ziegler D, Ametov A, Barinov A, et al. Oral treatment with alpha-
Report of the American Academy of Neurology, American Associa- lipoic acid improves symptomatic diabetic polyneuropathy: the
tion of Neuromuscular and Electrodiagnostic Medicine, and Amer- SYDNEY 2 trial. Diabetes Care. 2006;29(11):2365–2370.
ican Academy of Physical Medicine and Rehabilitation. Neurology. 653. Ziegler D, Nowak H, Kempler P, Vargha P, Low PA. Treatment of
2009;72(2):177–184. symptomatic diabetic polyneuropathy with the antioxidant alpha-
632. Boulton AJ, Gries FA, Jervell JA. Guidelines for the diagnosis and lipoic acid: a meta-analysis. Diabet Med. 2004;21(2):114–121.
outpatient management of diabetic peripheral neuropathy. Diabet 654. Treatment of painful diabetic neuropathy with topical capsaicin. A
Med. 1998;15(6):508–514. multicenter, double-blind, vehicle-controlled study. The Capsaicin
633. Bird SJ, Brown MJ. The clinical spectrum of diabetic neuropathy. Study Group. Arch Intern Med. 1991;151(11):2225–2229.
Semin Neurol. 1996;16(2):115–122. 655. Barbano RL, Herrmann DN, Hart-Gouleau S, Pennella-Vaughan
634. Feng Y, Schlösser FJ, Sumpio BE. The Semmes Weinstein mono- J, Lodewick PA, Dworkin RH. Effectiveness, tolerability, and
filament examination as a screening tool for diabetic peripheral impact on quality of life of the 5% lidocaine patch in diabetic poly-
neuropathy. J Vasc Surg. 2009;50(3):675–682, 682.e1. neuropathy. Arch Neurol. 2004;61(6):914–918.
635. Wang F, Zhang J, Yu J, et al. Diagnostic accuracy of monofilament 656. Derry S, Wiffen PJ, Moore RA, Quinlan J. Topical lidocaine
tests for detecting diabetic peripheral neuropathy: a systematic for neuropathic pain in adults. Cochrane Database Syst Rev.
review and meta-analysis. J Diabetes Res. 2017;2017:8787261. 2014;7:CD010958.
636. Dyck PJ, Kratz KM, Karnes JL, et al. The prevalence by staged 657. Forst T, Nguyen M, Forst S, Disselhoff B, Pohlmann T, Pfutzner
severity of various types of diabetic neuropathy, retinopathy, and A. Impact of low frequency transcutaneous electrical nerve stimula-
nephropathy in a population-based cohort: the rochester diabetic tion on symptomatic diabetic neuropathy using the new Salutaris
neuropathy study. Neurology. 1993;43(4):817–824. device. Diabetes Nutr Metab. 2004;17(3):163–168.

658. Kumar D, Marshall HJ. Diabetic peripheral neuropathy: ameliora- 683. Geddes J, Schopman JE, Zammitt NN, Frier BM. Prevalence of
tion of pain with transcutaneous electrostimulation. Diabetes Care. impaired awareness of hypoglycaemia in adults with Type 1 diabe-
1997;20(11):1702–1705. tes. Diabet Med. 2008;25(4):501–504.
659. Naderi Nabi B, Sedighinejad A, Haghighi M, et al. Comparison 684. Cryer PE. The barrier of hypoglycemia in diabetes. Diabetes.
of transcutaneous electrical nerve stimulation and pulsed radiofre- 2008;57(12):3169–3176.
quency sympathectomy for treating painful diabetic neuropathy. 685. Emral R, Pathan F, Cortes CAY, et al. Self-reported hypoglycemia
Anesth Pain Med. 2015;5(5):e29280. in insulin-treated patients with diabetes: results from an interna-
660. Vinik AI, Suwanwalaikorn S. Autonomic neuropathy. In: DeFronzo tional survey on 7289 patients from nine countries. Diabetes Res
RA, ed. Current Therapy of Diabetes Mellitus. St. Louis: Mosby; Clin Pract. 2017;134:17–28.
1997:165–176. 686. Zoungas S, Patel A, Chalmers J, et al. Severe hypoglycemia and risks
661. Karpitskaya Y, Novak CB, Mackinnon SE. Prevalence of smoking, of vascular events and death. N Engl J Med. 2010;363(15):1410–
obesity, diabetes mellitus, and thyroid disease in patients with car- 1418.
pal tunnel syndrome. Ann Plast Surg. 2002;48(3):269–273. 687. Seaquist ER, Miller ME, Bonds DE, et al. The impact of frequent
662. Perkins BA, Olaleye D, Bril V. Carpal tunnel syndrome in patients and unrecognized hypoglycemia on mortality in the ACCORD
with diabetic polyneuropathy. Diabetes Care. 2002;25(3):565–569. study. Diabetes Care. 2012;35(2):409–414.
663. Chaudhuri KR, Davidson AR, Morris IM. Limited joint mobil- 688.
Diabetes Control and Complications Trial/Epidemiology of
ity and carpal tunnel syndrome in insulin-dependent diabetes. Br J Diabetes Interventions and Complications Study Research
Rheumatol. 1989;28(3):191–194. Group, Jacobson AM, Musen G, et al. Long-term effect of dia-
664. Vinik AI, Holland MT, Le Beau JM, Liuzzi FJ, Stansberry KB, Colen betes and its treatment on cognitive function. N Engl J Med.
LB. Diabetic neuropathies. Diabetes Care. 1992;15(12):1926– 2007;356(18):1842–1852.
1975. 689. Reichard P, Pihl M. Mortality and treatment side-effects during
665. Calandruccio JH, Thompson NB. Carpal tunnel syndrome: mak- long-term intensified conventional insulin treatment in the Stock-
ing evidence-based treatment decisions. Orthop Clin North Am. holm Diabetes Intervention Study. Diabetes. 1994;43(2):313–317.
2018;49(2):223–229. 690. Ly TT, Gallego PH, Davis EA, Jones TW. Impaired awareness of
666. Dawson DM. Entrapment neuropathies of the upper extremities. hypoglycemia in a population-based sample of children and adoles-
N Engl J Med. 1993;329(27):2013–2018. cents with type 1 diabetes. Diabetes Care. 2009;32(10):1802–1806.
667. Vinik A, Mehrabyan A, Colen L, Boulton A. Focal entrapment 691. Perantie DC, Lim A, Wu J, et al. Effects of prior hypoglycemia and
neuropathies in diabetes. Diabetes Care. 2004;27(7):1783–1788. hyperglycemia on cognition in children with type 1 diabetes mel-
668. Sander HW, Chokroverty S. Diabetic amyotrophy: current con- litus. Pediatr Diabetes. 2008;9(2):87–95.
cepts. Semin Neurol. 1996;16(2):173–178. 692. Northam EA, Anderson PJ, Jacobs R, Hughes M, Warne GL,
669. Laughlin RS, Dyck PJ. Diabetic radiculoplexus neuropathies. Werther GA. Neuropsychological profiles of children with type 1
Handb Clin Neurol. 2014;126:45–52. diabetes 6 years after disease onset. Diabetes Care. 2001;24(9):1541–
670. Orbach H, Katz U, Sherer Y, Shoenfeld Y. Intravenous immuno- 1546.
globulin: adverse effects and safe administration. Clin Rev Allergy 693. Ho MS, Weller NJ, Ives FJ, et al. Prevalence of structural central
Immunol. 2005;29(3):173–184. nervous system abnormalities in early-onset type 1 diabetes mel-
671. Pop-Busui R. What do we know and we do not know about cardio- litus. J Pediatr. 2008;153(3):385–390.
vascular autonomic neuropathy in diabetes. J Cardiovasc Transl Res. 694. Cryer PE, Axelrod L, Grossman AB, et al. Evaluation and manage-
2012;5(4):463–478. ment of adult hypoglycemic disorders: an endocrine Society clinical
672. Pop-Busui R. Cardiac autonomic neuropathy in diabetes: a clinical practice guideline. J Clin Endocrinol Metab. 2009;94(3):709–728.
perspective. Diabetes Care. 2010;33(2):434–441. 695. de Zoysa N, Rogers H, Stadler M, et al. A psychoeducational pro-
673. Choung RS, Locke GR 3rd, Schleck CD, Zinsmeister AR, Melton gram to restore hypoglycemia awareness: the DAFNE-HART pilot
LJ 3rd, Talley NJ. Risk of gastroparesis in subjects with type 1 study. Diabetes Care. 2014;37(3):863–866.
and 2 diabetes in the general population. Am J Gastroenterol. 696. Choudhary P, Ramasamy S, Green L, et al. Real-time continuous
2012;107(1):82–88. glucose monitoring significantly reduces severe hypoglycemia in
674. Saigal CS, Wessells H, Pace J, Schonlau M, Wilt TJ. Urologic hypoglycemia-unaware patients with type 1 diabetes. Diabetes Care.
Diseases in America Project. Predictors and prevalence of erec- 2013;36(12):4160–4162.
tile dysfunction in a racially diverse population. Arch Intern Med. 697. Home PD. The pharmacokinetics and pharmacodynamics of rapid-
2006;166(2):207–212. acting insulin analogues and their clinical consequences. Diabetes
675. Khera M, Goldstein I. Erectile dysfunction. BMJ Clin Evid. Obes Metab. 2012;14(9):780–788.
2011;2011. 698. Stailey M, Conway SE. Review of the next generation of long-act-
676. Braffett BH, Wessells H, Sarma AV. Urogenital autonomic dys- ing basal insulins: insulin degludec and insulin glargine. Consult
function in diabetes. Curr Diabet Rep. 2016;16(12):119. Pharm. 2017;32(1):42–46.
677. McKibben MJ, Seed P, Ross SS, Borawski KM. Urinary tract infec- 699. Quiros C, Gimenez M, Rios P, et al. Long-term outcome of insulin
tion and neurogenic bladder. Urol Clin North Am. 2015;42(4):527– pump therapy: reduction of hypoglycaemia and impact on glycae-
536. mic control. Diabet Med. 2016;33(10):1422–1426.
678. Yuan Z, Tang Z, He C, Tang W. Diabetic cystopathy: a review. J 700. Linkeschova R, Raoul M, Bott U, Berger M, Spraul M. Less severe
Diabetes. 2015;7(4):442–447. hypoglycaemia, better metabolic control, and improved quality
679. Desouza CV, Bolli GB, Fonseca V. Hypoglycemia, diabetes, and of life in Type 1 diabetes mellitus with continuous subcutaneous
cardiovascular events. Diabetes Care. 2010;33(6):1389–1394. insulin infusion (CSII) therapy; an observational study of 100
680. Ortiz MR. Hypoglycemia in diabetes. Nurs Clin North Am. consecutive patients followed for a mean of 2 years. Diabet Med.
2017;52(4):565–574. 2002;19(9):746–751.
681. de Galan BE, Schouwenberg BJ, Tack CJ, Smits P. Pathophysiol- 701.
American Diabetes Association. 6. Glycemic targets: standards
ogy and management of recurrent hypoglycaemia and hypoglycae- of medical care in diabetes-2018. Diabetes Care. 2018;41(suppl
mia unawareness in diabetes. Neth J Med. 2006;64(8):269–279. 1):S55–S64.
682. Martin-Timon I, Del Canizo-Gomez FJ. Mechanisms of hypogly- 702. Scott J, Huskisson EC. Graphic representation of pain. Pain.
cemia unawareness and implications in diabetic patients. World J 1976;2(2):175–184.
Diabetes. 2015;6(7):912–926.

703. Stamler J, Vaccaro O, Neaton JD, Wentworth D. Diabetes, other 723. Glucose tolerance and mortality: comparison of WHO and Ameri-
risk factors, and 12-yr cardiovascular mortality for men screened can Diabetes Association diagnostic criteria. The DECODE study
in the Multiple Risk Factor Intervention Trial. Diabetes Care. group. European Diabetes Epidemiology Group. Diabetes Epide-
1993;16(2):434–444. miology: collaborative analysis of diagnostic criteria in europe. Lan-
704. Haffner SM, Lehto S, Ronnemaa T, Pyorala K, Laakso M. Mortal- cet. 1999;354(9179):617–621.
ity from coronary heart disease in subjects with type 2 diabetes and 724. Hu FB, Stampfer MJ, Haffner SM, Solomon CG, Willett WC,
in nondiabetic subjects with and without prior myocardial infarc- Manson JE. Elevated risk of cardiovascular disease prior to clini-
tion. N Engl J Med. 1998;339(4):229–234. cal diagnosis of type 2 diabetes. Diabetes Care. 2002;25(7):1129–
705. Laing SP, Swerdlow AJ, Slater SD, et al. Mortality from heart dis- 1134.
ease in a cohort of 23,000 patients with insulin-treated diabetes. 725. Shankar A, Klein R, Klein BE, Moss SE. Association between gly-
Diabetologia. 2003;46(6):760–765. cosylated hemoglobin level and cardiovascular and all-cause mor-
706. Malmberg K, Yusuf S, Gerstein HC, et al. Impact of diabetes on tality in type 1 diabetes. Am J Epidemiol. 2007;166:393–402.
long-term prognosis in patients with unstable angina and non-Q- 726. National Cholesterol Education Program (NCEP) Expert Panel
wave myocardial infarction: results of the OASIS (Organization on Detection, Evaluation, and Treatment of High Blood Choles-
to Assess Strategies for Ischemic Syndromes) Registry. Circulation. terol in Adults (Adult Treatment Panel III). Third Report of the
2000;102(9):1014–1019. National Cholesterol Education Program (NCEP) Expert Panel
707. Mukamal KJ, Nesto RW, Cohen MC, et al. Impact of diabetes on Detection, Evaluation, and Treatment of High Blood Choles-
on long-term survival after acute myocardial infarction: compa- terol in Adults (Adult Treatment Panel III) final report. Circulation.
rability of risk with prior myocardial infarction. Diabetes Care. 2002;106(25):3143–3421.
2001;24(8):1422–1427. 727. Grundy SM, Brewer HB Jr, Cleeman JI, et al. Definition of meta-
708. Evans JM, Wang J, Morris AD. Comparison of cardiovascular risk bolic syndrome. Report of the national heart, lung, and blood
between patients with type 2 diabetes and those who had had a Institute/American heart association conference on scientific issues
myocardial infarction: cross sectional and cohort studies. BMJ. related to definition. Circulation. 2004;109:433–438.
2002;324(7343):939–942. 728. Alberti KG, Zimmet P, Shaw J. Metabolic syndrome—a new
709. Lee CD, Folsom AR, Pankow JS, Brancati FL. Cardiovascular world-wide definition. A consensus statement from the Interna-
events in diabetic and nondiabetic adults with or without history tional Diabetes Federation. Diabet Med. 2006;23(5):469–480.
of myocardial infarction. Circulation. 2004;109(7):855–860. 729. American Diabetes Association. Standards of medical care in diabe-
710. Krolewski AS, Warram JH, Rand LI, Kahn CR. Epidemiologic tes—2011. Diabetes Care. 2011;34(suppl 1):S11–S61.
approach to the etiology of type I diabetes mellitus and its compli- 729a Grundy SM, Bilheimer D, Chait A, et al. Summary of the second
cations. N Engl J Med. 1987;317(22):1390–1398. report of the National Cholesterol Education Program (NCEP)
711. Harding JL, Pavkov ME, Magliano DJ, Shaw JE, Gregg EW. Expert Panel on Detection, Evaluation, and Treatment of High
Global trends in diabetes complications: a review of current evi- Blood Cholesterol in Adults (Adult Treatment Panel II). JAMA.
dence. Diabetologia. 2019;62:3–16. 1993;269:3015–3023.
712. Lind M, Svensson AM, Kosiborod M, et al. Glycemic con- 730. Festa A, Hanley AJ, Tracy RP, D’Agostino R Jr, Haffner SM.
trol and excess mortality in type 1 diabetes. N Engl J Med. Inflammation in the prediabetic state is related to increased insu-
2014;371(21):1972–1982. lin resistance rather than decreased insulin secretion. Circulation.
713. Pambianco G, Costacou T, Ellis D, Becker DJ, Klein R, Orchard 2003;108(15):1822–1830.
TJ. The 30-year natural history of type 1 diabetes complications: 731. Lim S, Meigs JB. Links between ectopic fat and vascular disease in
the Pittsburgh Epidemiology of Diabetes Complications Study humans. Arterioscler Thromb Vasc Biol. 2014;34(9):1820–1826.
experience. Diabetes. 2006;55(5):1463–1469. 732. Moore JX, Chaudhary N, Akinyemiju T. Metabolic syndrome
714. Wilson PW, Kannel WB, Silbershatz H, D’Agostino RB. Cluster- prevalence by race/ethnicity and sex in the United States, national
ing of metabolic factors and coronary heart disease. Arch Intern health and nutrition examination survey, 1988–2012. Prev Chronic
Med. 1999;159(10):1104–1109. Dis. 2017;14:E24.
715. Haffner SM, Mykkanen L, Festa A, Burke JP, Stern MP. Insulin- 733. Lakka HM, Laaksonen DE, Lakka TA, et al. The metabolic syn-
resistant prediabetic subjects have more atherogenic risk factors drome and total and cardiovascular disease mortality in middle-
than insulin-sensitive prediabetic subjects: implications for pre- aged men. J Am Med Assoc. 2002;288(21):2709–2716.
venting coronary heart disease during the prediabetic state. Circu- 734. Malik S, Wong ND, Franklin SS, et al. Impact of the metabolic
lation. 2000;101(9):975–980. syndrome on mortality from coronary heart disease, cardiovas-
716. Howard G, O’Leary DH, Zaccaro D, et al. Insulin sensitivity and cular disease, and all causes in United States adults. Circulation.
atherosclerosis. The insulin resistance atherosclerosis study (IRAS) 2004;110(10):1245–1250.
investigators. Circulation. 1996;93(10):1809–1817. 735. Sattar N, Gaw A, Scherbakova O, et al. Metabolic syndrome with
717. Deedwania PC. The deadly quartet revisited. Am J Med. and without C-reactive protein as a predictor of coronary heart
1998;105(1a):1s–3s. disease and diabetes in the West of Scotland Coronary Prevention
718. BARI 2D Study Group, Frye RL, August P, et al. A randomized Study. Circulation. 2003;108(4):414–419.
trial of therapies for type 2 diabetes and coronary artery disease. N 736. Kahn R, Buse J, Ferrannini E, Stern M. American diabetes associa-
Engl J Med. 2009;360(24):2503–2515. tion, European association for the study of diabetes. The metabolic
719. Andersson DK, Svardsudd K. Long-term glycemic control relates to syndrome: time for a critical appraisal: joint statement from the
mortality in type II diabetes. Diabetes Care. 1995;18(12):1534–1543. American diabetes association and the European association for the
720. Wei M, Gaskill SP, Haffner SM, Stern MP. Effects of diabetes and study of diabetes. Diabetes Care. 2005;28(9):2289–2304.
level of glycemia on all-cause and cardiovascular mortality. The San 737. Grundy SM, Cleeman JI, Daniels SR, et al. Diagnosis and manage-
Antonio Heart Study. Diabetes Care. 1998;21(7):1167–1172. ment of the metabolic syndrome: an American heart association/
721. Brunner EJ, Shipley MJ, Witte DR, Fuller JH, Marmot MG. Rela- national heart, lung, and blood Institute scientific statement. Cir-
tion between blood glucose and coronary mortality over 33 years in culation. 2005;112(17):2735–2752.
the Whitehall Study. Diabetes Care. 2006;29(1):26–31. 738. Bamberg F, Hetterich H, Rospleszcz S, et al. Subclinical disease
722. Tominaga M, Eguchi H, Manaka H, Igarashi K, Kato T, Sekikawa burden as assessed by whole-body MRI in subjects with prediabetes,
A. Impaired glucose tolerance is a risk factor for cardiovascular dis- subjects with diabetes, and normal control subjects from the gen-
ease, but not impaired fasting glucose. The funagata diabetes study. eral population: the KORA-MRI Study. Diabetes. 2017;66(1):158–
Diabetes Care. 1999;22(6):920–924. 169.

739. Abraira C, Colwell J, Nuttall F, et al. Cardiovascular events and 757. Rosenstock J, Kahn SE, Johansen OE, et al. Effect of linagliptin vs
correlates in the veterans affairs diabetes Feasibility trial. Vet- glimepiride on major adverse cardiovascular outcomes in patients
erans affairs cooperative study on glycemic control and com- with type 2 diabetes: The CAROLINA Randomized Clinical Trial.
plications in type II diabetes. Arch Intern Med. 1997;157(2): JAMA. 2019;322:1155–1166.
181–188. 758. Drucker DJ. The cardiovascular biology of glucagon-like peptide-1.
740. Nathan DM, Cleary PA, Backlund JY, et al. Intensive diabetes Cell Metab. 2016;24(1):15–30.
treatment and cardiovascular disease in patients with type 1 diabe- 759. Drucker DJ. The ascending GLP-1 road from clinical safety to reduc-
tes. N Engl J Med. 2005;353(25):2643–2653. tion of cardiovascular complications. Diabetes. 2018;67(9):1710–1719.
741. Effect of intensive blood-glucose control with metformin on com- 760. Smilowitz NR, Donnino R, Schwartzbard A. Glucagon-like pep-
plications in overweight patients with type 2 diabetes (UKPDS tide-1 receptor agonists for diabetes mellitus: a role in cardiovascu-
34). UK Prospective Diabetes Study (UKPDS) Group. Lancet. lar disease. Circulation. 2014;129(22):2305–2312.
1998;352(9131):854–865. 761. Hernandez AF, Green JB, Janmohamed S, et al. Albiglutide and
742. Roussel R, Travert F, Pasquet B, et al. Metformin use and mortality cardiovascular outcomes in patients with type 2 diabetes and cardio-
among patients with diabetes and atherothrombosis. Arch Intern vascular disease (Harmony Outcomes): a double-blind, randomised
Med. 2010;170(21):1892–1899. placebo-controlled trial. Lancet. 2018. ([Epub ahead of print]).
743. Bosch J, Gerstein HC, Dagenais GR, et al. Basal insulin and car- 762. Pfeffer MA, Claggett B, Diaz R, et al. Lixisenatide in patients
diovascular and other outcomes in dysglycemia. N Engl J Med. with type 2 diabetes and acute coronary syndrome. N Engl J Med.
2012;367):319–328. 2015;373(23):2247–2257.
744. Geldenhuys WJ, Leeper TC, Carroll RT. mitoNEET as a novel 763. Holman RR, Bethel MA, Mentz RJ, et al. Effects of once-weekly
drug target for mitochondrial dysfunction. Drug Discov Today. exantide on cardiovascular outcomes in type 2 diabetes. N Engl J
2014;19(10):1601–1606. Med. 2017;377(13):1228–1239.
745. Dormandy JA, Charbonnel B, Eckland DJ, et al. Secondary pre- 764. Inzucchi SE, Zinman B, Wanner C, et al. SGLT-2 inhibitors and
vention of macrovascular events in patients with type 2 diabetes cardiovascular risk: proposed pathways and review of ongoing out-
in the PROactive Study (PROspective pioglitAzone Clinical Trial come trials. Diab Vasc Dis Res. 2015;12(2):90–100.
in macroVascular Events): a randomised controlled trial. Lancet. 765. Zinman B, Wanner C, Lachin JM, et al. Empagliflozin, cardiovas-
2005;366(9493):1279–1289. cular outcomes, and mortality in type 2 diabetes. N Engl J Med.
746. Nissen SE, Nicholls SJ, Wolski K, et al. Comparison of piogli- 2015;373(22):2117–2128.
tazone vs glimepiride on progression of coronary atherosclerosis in 766. Birkeland KI, Jørgensen ME, Carstensen B, et al. Cardiovascu-
patients with type 2 diabetes: the PERISCOPE randomized con- lar mortality and morbidity in patients with type 2 diabetes fol-
trolled trial. J Am Med Assoc. 2008;299(13):1561–1573. lowing initiation of sodium-glucose co-transporter-2 inhibitors
747. Lago RM, Singh PP, Nesto RW. Congestive heart failure and car- versus other glucose-lowering drugs (CVD-REAL Nordic): a
diovascular death in patients with prediabetes and type 2 diabetes multinational observational analysis. Lancet Diabetes Endocrinol.
given thiazolidinediones: a meta-analysis of randomised clinical tri- 2017;5(9):709–717.
als. Lancet. 2007;370(9593):1129–1136. 767. Gowri MS, Van der Westhuyzen DR, Bridges SR, Anderson
748. Mitka M. Panel recommends easing restrictions on rosiglitazone JW. Decreased protection by HDL from poorly controlled type
despite concerns about cardiovascular safety. J Am Med Assoc. 2 diabetic subjects against LDL oxidation may Be due to the
2013;310(3):246–247. abnormal composition of HDL. Arterioscler Thromb Vasc Biol.
749. Kung J, Henry RR. Thiazolidinedione safety. Exp Opin Drug Saf. 1999;19(9):2226–2233.
2012;11(4):565–579. 768. Goldberg RB, Mellies MJ, Sacks FM, et al. Cardiovascular events
750. Duckworth WC, Abraira C, Moritz TE, et al. The duration and their reduction with pravastatin in diabetic and glucose-intoler-
of diabetes affects the response to intensive glucose control in ant myocardial infarction survivors with average cholesterol levels:
type 2 subjects: the VA Diabetes Trial. J Diabetes Complications. subgroup analyses in the cholesterol and recurrent events (CARE)
2011;25(6):355–361. trial. The care investigators. Circulation. 1998;98(23):2513–2519.
751. Reaven PD, Moritz TE, Schwenke DC, et al. Intensive glucose- 769.
Long-Term Intervention with Pravastatin in Ischaemic Dis-
lowering therapy reduces cardiovascular disease events in veterans ease (LIPID) Study Group. Prevention of cardiovascular events
affairs diabetes trial participants with lower calcified coronary ath- and death with pravastatin in patients with coronary heart dis-
erosclerosis. Diabetes. 2009;58(11):2642–2648. ease and a broad range of initial cholesterol levels. N Engl J Med.
752. Scirica BM, Bhatt DL, Braunwald E, et al. Saxagliptin and cardio- 1998;339(19):1349–1357.
vascular outcomes in patients with type 2 diabetes mellitus. N Engl 770. Haffner SM, Alexander CM, Cook TJ, et al. Reduced coronary
J Med. 2013;369(14):1317–1326. events in simvastatin-treated patients with coronary heart disease
753. White WB, Cannon CP, Heller SR, et al. Alogliptin after acute and diabetes or impaired fasting glucose levels: subgroup analyses
coronary syndrome in patients with type 2 diabetes. N Engl J Med. in the Scandinavian Simvastatin Survival Study. Arch Intern Med.
2013;369(14):1327–1335. 1999;159(22):2661–2667.
754. Zannad F, Cannon CP, Cushman WC, et al. Heart failure and 771. Ballantyne CM, Olsson AG, Cook TJ, Mercuri MF, Peder-
mortality outcomes in patients with type 2 diabetes taking alo- sen TR, Kjekshus J. Influence of low high-density lipoprotein
gliptin versus placebo in EXAMINE: a multicentre, randomised, cholesterol and elevated triglyceride on coronary heart disease
double-blind trial. Lancet. 2015;385(9982):2067–2076. events and response to simvastatin therapy in 4S. Circulation.
755. Scirica BM, Braunwald E, Raz I, et al. Heart failure, saxagliptin, 2001;104(25):3046–3051.
and diabetes mellitus: observations from the SAVOR-TIMI 53 ran- 772. Deedwania P, Barter P, Carmena R, et al. Reduction of low-den-
domized trial. Circulation. 2014;130(18):1579–1588. sity lipoprotein cholesterol in patients with coronary heart disease
756. Rosenstock J, Perkovic V, Johansen OE, et al. Effect of linagliptin and metabolic syndrome: analysis of the Treating to New Targets
vs placebo on major cardiovascular events in adults with type 2 study. Lancet. 2006;368(9539):919–928.
diabetes and high cardiovascular and renal risk: The CARMELINA 773. Taylor AJ, Sullenberger LE, Lee HJ, Lee JK, Grace KA. Arterial
Randomized Clinical Trial. JAMA. 2019;321:69–79. Biology for the Investigation of the Treatment Effects of Reduc-
756a McGuire DK, Alexander JH, Johansen OE, et al. Linagliptin ing Cholesterol (ARBITER) 2: a double-blind, placebo-controlled
effects on heart failure and related outcomes in individuals with study of extended-release niacin on atherosclerosis progression in
type 2 diabetes mellitus at high cardiovascular and renal risk in secondary prevention patients treated with statins. Circulation.
CARMELINA. Circulation. 2019;139:351–361. 2004;110(23):3512–3517.

774. Rubins HB, Robins SJ, Collins D, et al. Gemfibrozil for the second- 792. Bhatt DL, Steg PG, Miller M, et al. Cardiovascular risk reduc-
ary prevention of coronary heart disease in men with low levels of tion with icosapent ethyl for hypertriglyceridemia. N Engl J Med.
high-density lipoprotein cholesterol. Veterans Affairs High-Density 2019;380:11–22.
Lipoprotein Cholesterol Intervention Trial Study Group. N Engl J 793. Jorgensen AB, Frikke-Schmidt R, Nordestgaard BG, Tybjaerg-
Med. 1999;341(6):410–418. Hansen A. Loss-of-function mutations in APOC3 and risk of isch-
775. Heart Protection Study Collaborative Group. MRC/BHF Heart emic vascular disease. N Engl J Med. 2014;371(1):32–41.
Protection Study of cholesterol lowering with simvastatin in 20,536 794. U.S. Food and Drug Administration. FDA Briefing Document.
high-risk individuals: a randomised placebo-controlled trial. Lan- Endocinologic and Metabolic Drugs Advisory Committee Meeting;
cet. 2002;360(9326):7–22. 2018. Available at: https://www.fda.gov/downloads/AdvisoryCom-
776. Sarikaya H, Ferro J, Arnold M. Stroke prevention—medical and mittees/CommitteesMeetingMaterials/Drugs/Endocrinologicand-
lifestyle measures. Eur Neurol. 2015;73(3–4):150–157. MetabolicDrugsAdvisoryCommittee/UCM606857.pdf.
777.
Cholesterol Treatment Trialists’ (CTT) Collaboration, Fulcher 795. Grundy SM, Cleeman JI, Merz CN, et al. Implications of recent
J, O’Connell R, et al. Efficacy and safety of LDL-lowering ther- clinical trials for the national cholesterol education program adult
apy among men and women: meta-analysis of individual data treatment panel III guidelines. Circulation. 2004;110(2):227–239.
from 174,000 participants in 27 randomised trials. Lancet. 796. Ó Hartaigh B, Szymonifka J, Okin PM. Achieving target SBP for
2015;385(9976):1397–1405. lowering the risk of major adverse cardiovascular events in persons
778.
Cholesterol Treatment Trialists’ (CTT) Collaborators, Kearney with diabetes mellitus. J Hypertens. 2018;36(1):101–109.
PM, Blackwell L, et al. Efficacy of cholesterol-lowering therapy in 797. Yusuf S, Sleight P, Pogue J, Bosch J, Davies R, Dagenais G.
18,686 people with diabetes in 14 randomised trials of statins: a Effects of an angiotensin-converting-enzyme inhibitor, ramipril,
meta-analysis. Lancet. 2008;371(9607):117–125. on cardiovascular events in high-risk patients. N Engl J Med.
779. Goff DC Jr, Lloyd-Jones DM, Bennett G, et al. 2013 ACC/AHA 2000;342(3):145–153.
guideline on the assessment of cardiovascular risk: a report of the 798. Efficacy of atenolol and captopril in reducing risk of macrovascular
American College of cardiology/American heart association task and microvascular complications in type 2 diabetes: UKPDS 39.
force on practice guidelines. J Am Coll Cardiol. 2014;63(25 Pt UK prospective diabetes study group. BMJ. 1998;317(7160):713–
B):2935–2959. 720.
780. Masana L, Pedro-Botet J, Civeira F. IMPROVE-IT clinical impli- 799. Hansson L, Zanchetti A, Carruthers SG, et al. Effects of inten-
cations. Should the “high-intensity cholesterol-lowering therapy” sive blood pressure lowering and low-dose aspirin in patients
strategy replace the “high-intensity statin therapy”? Atherosclerosis. with hypertension: principal results of the Hypertension Optimal
2015;240(1):161–162. Treatment (HOT) randomised trial. HOT Study Group. Lancet.
781. Cannon CP, Blazing MA, Giugliano RP, et al. Ezetimibe added 1998;351:1755–1762.
to statin therapy after acute coronary syndromes. N Engl J Med. 800. Patel A, ADVANCE Collaborative Group, MacMahon S, et al.
2015;372(25):2387–2397. Effects of a fixed combination of perindopril and indapamide on
782. Ridker PM, Danielson E, Fonseca FA, et al. Rosuvastatin to pre- macrovascular and microvascular outcomes in patients with type 2
vent vascular events in men and women with elevated C-reactive diabetes mellitus (the ADVANCE trial): a randomised controlled
protein. N Engl J Med. 2008;359:2195–2207. trial. Lancet. 2007;370(9590):829–840.
783. Ridker PM, Pradhan A, MacFadyen JG, Libby P, Glynn RJ. 800a Dahlöf B, Devereux RB, Kjeldsen SE, et al. Cardiovascular mor-
Cardiovascular benefits and diabetes risks of statin therapy in bidity and mortality in the Losartan Intervention For Endpoint
primary prevention: an analysis from the JUPITER trial. Lancet. reduction in hypertension study (LIFE): a randomised trial against
2012;380(9841):565–571. atenolol. Lancet. 2002;23:995–1003.
784. Sabatine MS, Giugliano RP, Keech AC, et al. Evolocumab and 801. Whelton PK, Barzilay J, Cushman WC, et al. Clinical outcomes
clinical outcomes in patients with cardiovascular disease. N Engl J in antihypertensive treatment of type 2 diabetes, impaired fasting
Med. 2017;376(18):1713–1722. glucose concentration, and normoglycemia: antihypertensive and
785. Gonzalez PE, Joshi P. ODYSSEY Outcomes. American College of Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALL-
Cardiology (Website); 2018. Available at: https://www.acc.org/lat- HAT). Arch Intern Med. 2005;165(12):1401–1409.
est-in-cardiology/articles/2018/07/05/13/22/odyssey-outcomes. 802. ACCORD Study Group, Cushman WC, Evans GW, et al. Effects
786. Lotta LA, Sharp SJ, Burgess S, et al. Association between low-density of intensive blood-pressure control in type 2 diabetes mellitus. N
lipoprotein cholesterol-lowering genetic variants and risk of type 2 Engl J Med. 2010;362(17):1575–1585.
diabetes: a meta-analysis. J Am Med Assoc. 2016;316(13):1383– 803. SPRINT Research Group, Wright JT Jr, Williamson JD, et al. A
1391. randomized trial of intensive versus standard blood-pressure con-
787. Landray MJ, Haynes R, Hopewell JC, et al. Effects of extended- trol. N Engl J Med. 2015;373(22):2103–2116.
release niacin with laropiprant in high-risk patients. N Engl J Med. 804. Huang C, Dhruva SS, Coppi AC, et al. Systolic blood pressure response
2014;371(3):203–212. in SPRINT (systolic blood pressure intervention trial) and ACCORD
788. AIM-HIGH Investigators. The role of niacin in raising high-density (action to control cardiovascular risk in diabetes): a possible explana-
lipoprotein cholesterol to reduce cardiovascular events in patients tion for discordant trial results. J Am Heart Assoc. 2017;6(11).
with atherosclerotic cardiovascular disease and optimally treated 805. Nesto RW, Zarich S. Acute myocardial infarction in diabe-
low-density lipoprotein cholesterol Rationale and study design. tes mellitus: lessons learned from ACE inhibition. Circulation.
The Atherothrombosis Intervention in Metabolic syndrome with 1998;97(1):12–15.
low HDL/high triglycerides: impact on Global Health outcomes 806. Jacoby RM, Nesto RW. Acute myocardial infarction in the diabetic
(AIM-HIGH). Am Heart J. 2011;161(3). 471–477.e2. patient: pathophysiology, clinical course and prognosis. J Am Coll
789. Voight BF, Peloso GM, Orho-Melander M, et al. Plasma HDL Cardiol. 1992;20(3):736–744.
cholesterol and risk of myocardial infarction: a mendelian ran- 807. Iwasaka T, Takahashi N, Nakamura S, et al. Residual left ventricu-
domisation study. Lancet. 2012;380(9841):572–580. lar pump function after acute myocardial infarction in NIDDM
790. Rader DJ, Hovingh GK. HDL and cardiovascular disease. Lancet. patients. Diabetes Care. 1992;15(11):1522–1526.
2014;384(9943):618–625. 808. Bernardi L, Ricordi L, Lazzari P, et al. Impaired circadian modula-
791. Ginsberg HN, Elam MB, Lovato LC, et al. Effects of combination of sympathovagal activity in diabetes. A possible explanation
tion lipid therapy in type 2 diabetes mellitus. N Engl J Med. for altered temporal onset of cardiovascular disease. Circulation.
2010;362(17):1563–1574. 1992;86(5):1443–1452.

809. Zarich S, Waxman S, Freeman RT, Mittleman M, Hegarty P, 828. Cleveland JC Jr, Meldrum DR, Cain BS, Banerjee A, Harken AH.
Nesto RW. Effect of autonomic nervous system dysfunction on the Oral sulfonylurea hypoglycemic agents prevent ischemic precondi-
circadian pattern of myocardial ischemia in diabetes mellitus. J Am tioning in human myocardium. Two paradoxes revisited. Circula-
Coll Cardiol. 1994;24(4):956–962. tion. 1997;96(1):29–32.
810. Muller JE, Tofler GH, Stone PH. Circadian variation and triggers 829. Katsuda Y, Egashira K, Ueno H, et al. Glibenclamide, a selective
of onset of acute cardiovascular disease. Circulation. 1989;79(4): inhibitor of ATP-sensitive K+ channels, attenuates metabolic coro-
733–743. nary vasodilatation induced by pacing tachycardia in dogs. Circula-
811. Imperatore G, Riccardi G, Iovine C, Rivellese AA, Vaccaro O. tion. 1995;92(3):511–517.
Plasma fibrinogen: a new factor of the metabolic syndrome. A pop- 830. Davis CA 3rd, Sherman AJ, Yaroshenko Y, et al. Coronary vas-
ulation-based study. Diabetes Care. 1998;21(4):649–654. cular responsiveness to adenosine is impaired additively by block-
812. Sobel BE, Woodcock-Mitchell J, Schneider DJ, Holt RE, Marutsuka ade of nitric oxide synthesis and a sulfonylurea. J Am Coll Cardiol.
K, Gold H. Increased plasminogen activator inhibitor type 1 in coro- 1998;31(4):816–822.
nary artery atherectomy specimens from type 2 diabetic compared 831. O’Driscoll G, Green D, Maiorana A, Stanton K, Colreavy F, Taylor
with nondiabetic patients: a potential factor predisposing to throm- R. Improvement in endothelial function by angiotensin-converting
bosis and its persistence. Circulation. 1998;97(22):2213–2221. enzyme inhibition in non-insulin-dependent diabetes mellitus. J
813. Meigs JB, Mittleman MA, Nathan DM, et al. Hyperinsulinemia, Am Coll Cardiol. 1999;33(6):1506–1511.
hyperglycemia, and impaired hemostasis: the Framingham Off- 832. Vaughan DE, Rouleau JL, Ridker PM, Arnold JM, Menapace FJ,
spring study. J Am Med Assoc. 2000;283(2):221–228. Pfeffer MA. Effects of ramipril on plasma fibrinolytic balance in
814. Woodfield SL, Lundergan CF, Reiner JS, et al. Angiographic find- patients with acute anterior myocardial infarction. HEART Study
ings and outcome in diabetic patients treated with thrombolytic Investigators. Circulation. 1997;96(2):442–447.
therapy for acute myocardial infarction: the GUSTO-I experience. 833. Torlone E, Britta M, Rambotti AM, et al. Improved insulin action
J Am Coll Cardiol. 1996;28(7):1661–1669. and glycemic control after long-term angiotensin-converting
815. Mak KH, Moliterno DJ, Granger CB, et al. Influence of diabe- enzyme inhibition in subjects with arterial hypertension and type
tes mellitus on clinical outcome in the thrombolytic era of acute II diabetes. Diabetes Care. 1993;16(10):1347–1355.
myocardial infarction. GUSTO-I investigators. Global utilization 834. Zuanetti G, Latini R, Maggioni AP, Franzosi M, Santoro L,
of Streptokinase and tissue plasminogen activator for Occluded Tognoni G. Effect of the ACE inhibitor lisinopril on mortality in
coronary arteries. J Am Coll Cardiol. 1997;30(1):171–179. diabetic patients with acute myocardial infarction: data from the
816. Sun D, Nguyen N, DeGrado TR, Schwaiger M, Brosius FC 3rd. GISSI-3 study. Circulation. 1997;96(12):4239–4245.
Ischemia induces translocation of the insulin-responsive glucose 834a. Gustafsson I, Torp-Pedersen C, Køber L, Gustafsson F, Hildeb-
transporter GLUT4 to the plasma membrane of cardiac myocytes. randt P. Effect of the angiotensin-converting enzyme inhibitor
Circulation. 1994;89(2):793–798. trandolapril on mortality and morbidity in diabetic patients with
817. Oliver MF, Opie LH. Effects of glucose and fatty acids on myocar- left ventricular dysfunction after acute myocardial infarction. Trace
dial ischaemia and arrhythmias. Lancet. 1994;343(8890):155–158. Study Group. J Am Coll Cardiol. 1999;34(1):83–89.
818. Stone PH, Muller JE, Hartwell TY, et al. The effect of diabetes 835. Lakshman MR, Reda DJ, Materson BJ, Cushman WC, Freis ED.
mellitus on prognosis and serial left ventricular function after acute Diuretics and beta-blockers do not have adverse effects at 1 year
myocardial infarction: contribution of both coronary disease and on plasma lipid and lipoprotein profiles in men with hypertension.
diastolic left ventricular dysfunction to the adverse prognosis. The Department of Veterans Affairs Cooperative Study Group on Anti-
MILIS Study Group. J Am Coll Cardiol. 1989;14:49–57. hypertensive Agents. Arch Intern Med. 1999;159(6):551–558.
819. Deleted in review. 836. Shorr RI, Ray WA, Daugherty JR, Griffin MR. Antihypertensives
820. Bellodi G, Manicardi V, Malavasi V, et al. Hyperglycemia and and the risk of serious hypoglycemia in older persons using insulin
prognosis of acute myocardial infarction in patients without diabe- or sulfonylureas. J Am Med Assoc. 1997;278(1):40–43.
tes mellitus. Am J Cardiol. 1989;64(14):885–888. 837. Chen J, Marciniak TA, Radford MJ, Wang Y, Krumholz HM. Beta-
821. Oswald GA, Smith CC, Betteridge DJ, Yudkin JS. Determinants blocker therapy for secondary prevention of myocardial infarction
and importance of stress hyperglycaemia in non-diabetic patients in elderly diabetic patients. Results from the National Cooperative
with myocardial infarction. Br Med J. 1986;293(6552):917–922. Cardiovascular Project. J Am Coll Cardiol. 1999;34(5):1388–1394.
822. Fava S, Aquilina O, Azzopardi J, Agius Muscat H, Fenech FF. The 838. Giugliano D, Acampora R, Marfella R, et al. Metabolic and cardio-
prognostic value of blood glucose in diabetic patients with acute vascular effects of carvedilol and atenolol in non-insulin-dependent
myocardial infarction. Diabet Med. 1996;13(1):80–83. diabetes mellitus and hypertension. A randomized, controlled trial.
823. Capes SE, Hunt D, Malmberg K, Gerstein HC. Stress hypergly- Ann Intern Med. 1997;126(12):955–959.
caemia and increased risk of death after myocardial infarction in 839. Aspirin effects on mortality and morbidity in patients with diabetes
patients with and without diabetes: a systematic overview. Lancet. mellitus. Early Treatment Diabetic Retinopathy Study report 14.
2000;355(9206):773–778. ETDRS Investigators. J Am Med Assoc. 1992;268(10):1292–1300.
824. Malmberg K, Ryden L, Efendic S, et al. Randomized trial of insulin- 840. American Diabetes Association. 9. Cardiovascular disease and risk
glucose infusion followed by subcutaneous insulin treatment in dia- management: standards of medical care in diabetes-2018. Diabetes
betic patients with acute myocardial infarction (DIGAMI study): Care. 2018;41(suppl 1):S86–S104.
effects on mortality at 1 year. J Am Coll Cardiol. 1995;26(1):57–65. 841. Davi G, Catalano I, Averna M, et al. Thromboxane biosynthesis
825. Malmberg K, Ryden L, Wedel H, et al. Intense metabolic control and platelet function in type II diabetes mellitus. N Engl J Med.
by means of insulin in patients with diabetes mellitus and acute 1990;322(25):1769–1774.
myocardial infarction (DIGAMI 2): effects on mortality and mor- 842. Grosser T, Fries S, Lawson JA, Kapoor SC, Grant GR, FitzGerald
bidity. Eur Heart J. 2005;26(7):650–661. GA. Drug resistance and pseudoresistance: an unintended conse-
826. Mehta SR, Yusuf S, Diaz R, et al. Effect of glucose-insulin-potas- quence of enteric coating aspirin. Circulation. 2013;127(3):377–
sium infusion on mortality in patients with acute ST-segment 385.
elevation myocardial infarction: the CREATE-ECLA randomized 843. McNeil JJ, Wolfe R, Woods RL, et al. Effect of aspirin on cardio-
controlled trial. J Am Med Assoc. 2005;293(4):437–446. vascular events and bleeding in the healthy elderly. N Engl J Med.
827. Garratt KN, Brady PA, Hassinger NL, Grill DE, Terzic A, Holmes 2018;379(16):1509–1518.
DR Jr. Sulfonylurea drugs increase early mortality in patients with 844. ASCEND Study Collaborative Group, Bowman L, Mafham M,
diabetes mellitus after direct angioplasty for acute myocardial et al. Effects of aspirin for primary prevention in persons with dia-
infarction. J Am Coll Cardiol. 1999;33(1):119–124. betes mellitus. N Engl J Med. 2018;379(16):1529–1539.

845. Hirsh J, Bhatt DL. Comparative benefits of clopidogrel and aspi- 865. McMurray JJ, Packer M, Desai AS, et al. Angiotensin-nepri-
rin in high-risk patient populations: lessons from the CAPRIE and lysin inhibition versus enalapril in heart failure. N Engl J Med.
CURE studies. Arch Intern Med. 2004;164(19):2106–2110. 2014;371(11):993–1004.
846. Wiviott SD, Braunwald E, McCabe CH, et al. Prasugrel versus 866. Boulton AJM, Cavanagh PR, Rayman G. The Foot in Diabetes. 4th
clopidogrel in patients with acute coronary syndromes. N Engl J ed. Chichester, UK: John Wiley & Sons; 2006.
Med. 2007;357(20):2001–2015. 867. Bowker JH, Pfeifer MA. Levin & O’Neal’s the Diabetic Foot. 7th ed.
847. Platelet receptor inhibition in ischemic syndrome management in St. Louis: Mosby; 2008.
patients limited by unstable signs and symptoms (PRISM-PLUS) 868. Hinchliffe RJ, Andros G, Apelqvist J, et al. A systematic review
study investigators. Inhibition of the platelet glycoprotein IIb/IIIa of the effectiveness of revascularization of the ulcerated foot in
receptor with tirofiban in unstable angina and non-Q-wave myo- patients with diabetes and peripheral arterial disease. Diabetes
cardial infarction. N Engl J Med. 1998;338(21):1488–1497. Metab Res Rev. 2012;28(suppl 1):179–217.
848.
EPILOG Investigators. Platelet glycoprotein IIb/IIIa receptor 869. Boulton AJM. The diabetic foot. Med Clin North Am.
blockade and low-dose heparin during percutaneous coronary 2013;97(5):775–992.
revascularization. N Engl J Med. 1997;336(24):1689–1696. 870. Armstrong DG, Boulton AJM, Bus SA. Diabetic foot ulcers and
849. Lincoff AM, Califf RM, Moliterno DJ, et al. Complementary clin- their recurrence. N Engl J Med. 2017;376(24):2367–2375.
ical benefits of coronary-artery stenting and blockade of platelet 871. Abbott CA, Carrington AL, Ashe H, et al. The North-West Diabe-
glycoprotein IIb/IIIa receptors. Evaluation of Platelet IIb/IIIa Inhi- tes Foot Care Study: incidence of, and risk factors for, new diabetic
bition in Stenting Investigators. N Engl J Med. 1999;341(5):319– foot ulceration in a community-based patient cohort. Diabet Med.
327. 2002;19(5):377–384.
850. Bypass Angioplasty Revascularization Investigation (BARI) Investi- 872. Bowling FL, Rashid ST, Boulton AJ. Preventing and treating foot
gators. Comparison of coronary bypass surgery with angioplasty in complications associated with diabetes mellitus. Nat Rev Endocri-
patients with multivessel disease. N Engl J Med. 1996;335(4):217– nol. 2015;11(10):606–616.
225. 872a Ndip A, Lavery LA, Lafontaine J, et al. High levels of foot ulcer-
851. Farkouh ME, Domanski M, Sleeper LA, et al. Strategies for mulation and amputation risk in a multiracial cohort of diabetic
tivessel revascularization in patients with diabetes. N Engl J Med. patients on dialysis therapy. Diabetes Care. 2010;33(4):878–880.
2012;367(25):2375–2384. 873. van Houtum WH, Rauwerda JA, Ruwaard D, Schaper NC, Bak-
852. Kannel WB, Hjortland M, Castelli WP. Role of diabetes in ker K. Reduction in diabetes-related lower-extremity amputations
congestive heart failure: the Framingham study. Am J Cardiol. in The Netherlands: 1991-2000. Diabetes Care. 2004;27(5):1042–
1974;34(1):29–34. 1046.
853. Burchfield JS, Xie M, Hill JA. Pathological ventricular remodeling: 874. Krishnan S, Nash F, Baker N, Fowler D, Rayman G. Reduction in
mechanisms: part 1 of 2. Circulation. 2013;128(4):388–400. diabetic amputations over 11 years in a defined U.K. population:
854. Seferovic PM, Paulus WJ. Clinical diabetic cardiomyopathy: a two- benefits of multidisciplinary team work and continuous prospec-
faced disease with restrictive and dilated phenotypes. Eur Heart J. tive audit. Diabetes Care. 2008;31(1):99–101.
2015;36(27):1718–1727. 1727a–1727c. 875. Vileikyte L. Psychosocial and behavioral aspects of diabetic foot
855. Aronson D, Rayfield EJ, Chesebro JH. Mechanisms determining lesions. Curr Diab Rep. 2008;8(2):119–125.
course and outcome of diabetic patients who have had acute myo- 876. Reiber GE, Vileikyte L, Boyko EJ, et al. Causal pathways for inci-
cardial infarction. Ann Intern Med. 1997;126(4):296–306. dent lower-extremity ulcers in patients with diabetes from two set-
856. Cabin HS, Roberts WC. Quantitative comparison of extent of tings. Diabetes Care. 1999;22(1):157–162.
coronary narrowing and size of healed myocardial infarct in 33 nec- 877. Boulton AJ. The diabetic foot: from art to science. The 18th
ropsy patients with clinically recognized and in 28 with clinically Camillo Golgi lecture. Diabetologia. 2004;47(8):1343–1353.
unrecognized (“silent”) previous acute myocardial infarction. Am J 878. Jude EB, Eleftheriadou I, Tentolouris N. Peripheral arterial disease
Cardiol. 1982;50(4):677–681. in diabetes—a review. Diabet Med. 2010;27(1):4–14.
857. Vinik AI, Ziegler D. Diabetic cardiovascular autonomic neuropa- 879. Vriens B, D’Abate F, Ozdemir BA, et al. Clinical examination and
thy. Circulation. 2007;115(3):387–397. non-invasive screening tests in the diagnosis of peripheral artery
858. van Hoeven KH, Factor SM. A comparison of the pathological disease in those with diabetes related foot ulceration. Diabet Med.
spectrum of hypertensive, diabetic, and hypertensive-diabetic heart 2018;35(7):902.
disease. Circulation. 1990;82(3):848–855. 880. Albayati MA, Shearman CP. Peripheral arterial disease and
859. Kawaguchi M, Techigawara M, Ishihata T, et al. A comparison bypass surgery in the diabetic lower limb. Med Clin North Am.
of ultrastructural changes on endomyocardial biopsy specimens 2013;97(5):821–834.
obtained from patients with diabetes mellitus with and without 881. Hinchliffe RJ, Brownrigg JR, Andros G, et al. Effectiveness of
hypertension. Heart Vessel. 1997;12(6):267–274. revascularization of the ulcerated foot in patients with diabetes and
860. Paulus WJ, Tschope C. A novel paradigm for heart failure with pre- peripheral artery disease: a systematic review. Diabetes Metab Res
served ejection fraction: comorbidities drive myocardial dysfunc- Rev. 2016;32(suppl 1):136–144.
tion and remodeling through coronary microvascular endothelial 882. Noronen K, Saarinen E, Alback A, Venermo M. Analysis of the
inflammation. J Am Coll Cardiol. 2013;62(4):263–271. elective treatment process for critical limb ischaemia with tissue
861. Nahser PJ Jr, Brown RE, Oskarsson H, Winniford MD, Ros- loss: diabetic patients require rapid revascularisation. Eur J Vasc
sen JD. Maximal coronary flow reserve and metabolic coro- Endovasc Surg. 2017;53(2):206–213.
nary vasodilation in patients with diabetes mellitus. Circulation. 883. Deleted in review.
1995;91(3):635–640. 884. Boulton AJ, Armstrong DG, Albert SF, et al. Comprehensive foot
862. Depre C, Vanoverschelde JL, Taegtmeyer H. Glucose for the heart. examination and risk assessment: a report of the task force of the
Circulation. 1999;99(4):578–588. foot care interest group of the American Diabetes Association, with
863. Chouchani ET, Pell VR, Gaude E, et al. Ischaemic accumulation of endorsement by the American Association of Clinical Endocrinolo-
succinate controls reperfusion injury through mitochondrial ROS. gists. Diabetes Care. 2008;31(8):1679–1685.
Nature. 2014;515(7527):431–435. 885. Oyibo SO, Jude EB, Tarawneh I, Nguyen HC, Harkless LB, Boul-
864. Kohlhauer M, Dawkins S, Costa ASH, et al. Metabolomic profil- ton AJ. A comparison of two diabetic foot ulcer classification sys-
ing in acute ST-segment-elevation myocardial infarction identifies tems: the Wagner and the University of Texas wound classification
succinate as an early marker of human ischemia-reperfusion injury. systems. Diabetes Care. 2001;24(1):84–88.
J Am Heart Assoc. 2018;7(8).

886. Piaggesi A, Viacava P, Rizzo L, et al. Semiquantitative analysis of 892. Ndip A, Williams A, Jude EB, et al. The RANKL/RANK/OPG
the histopathological features of the neuropathic foot ulcer: effects signaling pathway mediates medial arterial calcification in diabetic
of pressure relief. Diabetes Care. 2003;26(11):3123–3128. Charcot neuroarthropathy. Diabetes. 2011;60(8):2187–2196.
887. Armstrong DG, Nguyen HC, Lavery LA, van Schie CH, Boulton 893. Petrova NL, Edmonds ME. Acute Charcot neuro-osteoarthropa-
AJ, Harkless LB. Off-loading the diabetic foot wound: a random- thy. Diabetes Metab Res Rev. 2016;32(suppl 1):281–286.
ized clinical trial. Diabetes Care. 2001;24(6):1019–1022. 894. Langer A, Rogowski W. Systematic review of economic evalu-
888. Katz IA, Harlan A, Miranda-Palma B, et al. A randomized trial of ations of human cell-derived wound care products for the treat-
two irremovable off-loading devices in the management of plantar ment of venous leg and diabetic foot ulcers. BMC Health Serv Res.
neuropathic diabetic foot ulcers. Diabetes Care. 2005;28(3):555– 2009;9:115.
559. 895. Armstrong DG, Lavery LA. Negative pressure wound therapy after
889. Lipsky BA. New developments in diagnosing and treating diabetic partial diabetic foot amputation: a multicentre, randomised con-
foot infections. Diabetes Metab Res Rev. 2008;24(Suppl 1):S66– trolled trial. Lancet. 2005;366(9498):1704–1710.
S71. 896. Blume PA, Walters J, Payne W, Ayala J, Lantis J. Comparison of
890. Teh J, Berendt T, Lipsky BA. Rational imaging: investigating sus- negative pressure wound therapy using vacuum-assisted closure
pected bone infection in the diabetic foot. BMJ. 2010;340:415– with advanced moist wound therapy in the treatment of diabetic
417. foot ulcers: a multicenter randomized controlled trial. Diabetes
891. Lazaro-Martinez JL, Aragon-Sanchez J, Garcia-Morales E. Anti- Care. 2008;31(4):631–636.
biotics versus conservative surgery for treating diabetic foot 897. Fadini GP, Avogaro A. SGLT2 inhibitors and amputations in the
osteomyelitis: a randomized comparative trial. Diabetes Care. US FDA adverse event reporting system. Lancet Diabetes Endocri-
2014;37(3):789–795. nol. 2017;5(9):680–681.

First Page PDF
No ratings yet
First Page PDF
1 page
Diabetes
No ratings yet
Diabetes
17 pages
Case Studies Diabetes and Complications 003 002
No ratings yet
Case Studies Diabetes and Complications 003 002
57 pages
Physical Evaluation I (Dent 5121) : Endocrine System: Diabetes Mellitus
No ratings yet
Physical Evaluation I (Dent 5121) : Endocrine System: Diabetes Mellitus
47 pages
Readings
No ratings yet
Readings
24 pages
5 Endocrine
No ratings yet
5 Endocrine
20 pages
Chapter 405 - Diabetes Mellitus - Complications
No ratings yet
Chapter 405 - Diabetes Mellitus - Complications
15 pages
Publication
No ratings yet
Publication
6 pages
Week 2 Complications of Diabetes Mellitus
No ratings yet
Week 2 Complications of Diabetes Mellitus
33 pages
4th Year. General Med. DM Lec 1 and 2
No ratings yet
4th Year. General Med. DM Lec 1 and 2
40 pages
Chapter 405 - Diabetes Mellitus - Complications
No ratings yet
Chapter 405 - Diabetes Mellitus - Complications
15 pages
Endocrine Disorders
No ratings yet
Endocrine Disorders
113 pages
Aquifer InternalMedicine08 - 55
No ratings yet
Aquifer InternalMedicine08 - 55
8 pages
l3 - Diabetes Sequele and Treatment Beretta
No ratings yet
l3 - Diabetes Sequele and Treatment Beretta
66 pages
Sapna Proj
No ratings yet
Sapna Proj
21 pages
Diabetes Mellitus David FINAL
No ratings yet
Diabetes Mellitus David FINAL
104 pages
23.farmaki, P., Et Al., Complications of The Type 2 Diabetes Mellitus. Curr Cardiol Rev, 2020
No ratings yet
23.farmaki, P., Et Al., Complications of The Type 2 Diabetes Mellitus. Curr Cardiol Rev, 2020
3 pages
Complications of Diabetes
No ratings yet
Complications of Diabetes
3 pages
Type 2 Diabetes Mellitus
No ratings yet
Type 2 Diabetes Mellitus
5 pages
Drug Therapy Prevention and Treatment Diabetes
No ratings yet
Drug Therapy Prevention and Treatment Diabetes
8 pages
Chronic Complications of Diabetes Mellitus
No ratings yet
Chronic Complications of Diabetes Mellitus
48 pages
Complications of Diabetes Mellitus Type 1
No ratings yet
Complications of Diabetes Mellitus Type 1
7 pages
Diabetes Mellitus
No ratings yet
Diabetes Mellitus
24 pages
مح١٨
No ratings yet
مح١٨
44 pages
Diabetes Mellitus Pathophysiology
No ratings yet
Diabetes Mellitus Pathophysiology
5 pages
Case Stud
No ratings yet
Case Stud
5 pages
11.5 Rehabilitation Medicine II
No ratings yet
11.5 Rehabilitation Medicine II
18 pages
Diabetes: Symptoms and Management
No ratings yet
Diabetes: Symptoms and Management
8 pages
5 Diabetes
No ratings yet
5 Diabetes
41 pages
ENDO 6C. Chronic Complications of Diabetes Mellitus
No ratings yet
ENDO 6C. Chronic Complications of Diabetes Mellitus
7 pages
Complications of Diabetes Mellitus: A Review
No ratings yet
Complications of Diabetes Mellitus: A Review
6 pages
Pathology of Diabetes
No ratings yet
Pathology of Diabetes
41 pages
Diabetes Mellitus: Signs and Symptoms
No ratings yet
Diabetes Mellitus: Signs and Symptoms
2 pages
DM Introduction
No ratings yet
DM Introduction
3 pages
Diabetes - in Class - Student 2
No ratings yet
Diabetes - in Class - Student 2
41 pages
Diabetes and Hypoglycemia
No ratings yet
Diabetes and Hypoglycemia
64 pages
Diabetes Mellitus2 2022 Recent
No ratings yet
Diabetes Mellitus2 2022 Recent
6 pages
Biochem PPT Case Study Group 2
No ratings yet
Biochem PPT Case Study Group 2
24 pages
FM Case 6 - Diabetes Care Visit
No ratings yet
FM Case 6 - Diabetes Care Visit
9 pages
Type 1 Type 2: Immune System Destroyed Resistance
No ratings yet
Type 1 Type 2: Immune System Destroyed Resistance
8 pages
Vascular Problems in Diabetes
No ratings yet
Vascular Problems in Diabetes
27 pages
Diabetes Work
No ratings yet
Diabetes Work
7 pages
Diabetes Mellitus
No ratings yet
Diabetes Mellitus
5 pages
The Complications of Diabetes
No ratings yet
The Complications of Diabetes
1 page
Nursing Case Study: Diabetic Amputation
100% (2)
Nursing Case Study: Diabetic Amputation
61 pages
Diabetes Mellitus
No ratings yet
Diabetes Mellitus
37 pages
Understanding Diabetes: A Medical Overview
No ratings yet
Understanding Diabetes: A Medical Overview
62 pages
10.7556 Jaoa.2000.100.10.621
No ratings yet
10.7556 Jaoa.2000.100.10.621
14 pages
Lecture 17 Diabetes
No ratings yet
Lecture 17 Diabetes
51 pages
Chronic Diabetic Complications
No ratings yet
Chronic Diabetic Complications
46 pages
Diabetes Mellitus: Type 1 Diabetes (Formerly Known As Insulin-Dependent) in Which The Pancreas Fails To Produce The
No ratings yet
Diabetes Mellitus: Type 1 Diabetes (Formerly Known As Insulin-Dependent) in Which The Pancreas Fails To Produce The
3 pages
Pharma Lit Review
No ratings yet
Pharma Lit Review
31 pages
Diabetes Mellitus - Complications
No ratings yet
Diabetes Mellitus - Complications
36 pages
Type 2 Diabetes: Pathophysiology & Management
No ratings yet
Type 2 Diabetes: Pathophysiology & Management
58 pages
Diabetes
100% (1)
Diabetes
7 pages
Diabetes and Adult Surgical Inpatients
No ratings yet
Diabetes and Adult Surgical Inpatients
5 pages
Em 3910
No ratings yet
Em 3910
4 pages
Prosthetic Implications With The Diabetic Patient: Lawrence R. Lange, B.S., C.P.O
No ratings yet
Prosthetic Implications With The Diabetic Patient: Lawrence R. Lange, B.S., C.P.O
7 pages
Passive Voice - Exercises
No ratings yet
Passive Voice - Exercises
13 pages
5698-Article Text-11456-11958-10-20241124
No ratings yet
5698-Article Text-11456-11958-10-20241124
8 pages
Future For, MS
No ratings yet
Future For, MS
6 pages
MEHU130 U2 T74 Colorectal Cancer
No ratings yet
MEHU130 U2 T74 Colorectal Cancer
14 pages
Salmonella y Salmonellosis
No ratings yet
Salmonella y Salmonellosis
33 pages
Pathophysiology of Cardiogenic Pulmonary Edema - UpToDate
No ratings yet
Pathophysiology of Cardiogenic Pulmonary Edema - UpToDate
14 pages
Fficacy and Safety of Injectable and Oral Antibiotics
No ratings yet
Fficacy and Safety of Injectable and Oral Antibiotics
17 pages
Structural Organisation in Animals
100% (1)
Structural Organisation in Animals
61 pages
Kseeb I Puc Biology - EM Model Paper 2022
No ratings yet
Kseeb I Puc Biology - EM Model Paper 2022
4 pages
Muscular Soreness (Immediate Versus Delayed Onset) : Lifting The Weight Only
No ratings yet
Muscular Soreness (Immediate Versus Delayed Onset) : Lifting The Weight Only
9 pages
Grand - Tests For Neet Ug Examination
No ratings yet
Grand - Tests For Neet Ug Examination
27 pages
Day-by-Day Blood Pressure Variability and Functional Outcome After Acute Ischemic Stroke
No ratings yet
Day-by-Day Blood Pressure Variability and Functional Outcome After Acute Ischemic Stroke
8 pages
Cell Commitment N Determination
No ratings yet
Cell Commitment N Determination
41 pages
THE CARDIOVASCULAR SYSTEM Review
No ratings yet
THE CARDIOVASCULAR SYSTEM Review
34 pages
Pediatric Cardiology Case Summaries
No ratings yet
Pediatric Cardiology Case Summaries
2 pages
This Study Resource Was Shared Via: Sinus Bradycardia
No ratings yet
This Study Resource Was Shared Via: Sinus Bradycardia
3 pages
DNA Polymerase 2 Termination
No ratings yet
DNA Polymerase 2 Termination
12 pages
November 2009 Final Edition
No ratings yet
November 2009 Final Edition
45 pages
5008 Therapy System PDF
No ratings yet
5008 Therapy System PDF
8 pages
FINAL EXAM ENGLISH 600 For Students
No ratings yet
FINAL EXAM ENGLISH 600 For Students
10 pages
Coaching & Educational Academy (Chiniot) : Objective Type
No ratings yet
Coaching & Educational Academy (Chiniot) : Objective Type
14 pages
MHC: Structure, Function, and Interaction
No ratings yet
MHC: Structure, Function, and Interaction
18 pages
Restorative Dentistry and It's Related To Pulp Health: DR - Ahmed Al-Jobory
No ratings yet
Restorative Dentistry and It's Related To Pulp Health: DR - Ahmed Al-Jobory
17 pages
Somera Case 1928
75% (4)
Somera Case 1928
68 pages
Sanity Table
No ratings yet
Sanity Table
2 pages
Plasma پلازما
No ratings yet
Plasma پلازما
7 pages
Kalish 2017
No ratings yet
Kalish 2017
8 pages
5 A Student Investigates How Light Affects Photosynthesis by Looking at Changes in
No ratings yet
5 A Student Investigates How Light Affects Photosynthesis by Looking at Changes in
2 pages
Objectives of The Operative Dentistry
No ratings yet
Objectives of The Operative Dentistry
5 pages
Fitness Plan for Muscle Building
No ratings yet
Fitness Plan for Muscle Building
2 pages
CPR Review Worksheet Ans Key
No ratings yet
CPR Review Worksheet Ans Key
9 pages
Three Dimensional Finite Element Analysis of Pelvic Bone
No ratings yet
Three Dimensional Finite Element Analysis of Pelvic Bone
6 pages
Chapter 8 & 9 Photosynthesis and Cellular Respiration Test: Multiple Choice
100% (1)
Chapter 8 & 9 Photosynthesis and Cellular Respiration Test: Multiple Choice
4 pages
Biological Sciences: General Biology GENB (0404C) : 1. Basic Program 15-16 Credits 2. Supporting Courses 32 Credits
No ratings yet
Biological Sciences: General Biology GENB (0404C) : 1. Basic Program 15-16 Credits 2. Supporting Courses 32 Credits
2 pages
Q.P. Code: 801131: AUGUST 2019
No ratings yet
Q.P. Code: 801131: AUGUST 2019
8 pages
MP Module 2 PDF
No ratings yet
MP Module 2 PDF
32 pages
Humoral Theory for Health Professionals
100% (2)
Humoral Theory for Health Professionals
12 pages