Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

Learning Objectives PN Parenteral nutrition

RCT Randomized controlled trial
1. Appropriately assess hyponatremia and SIADH Syndrome of inappropriate antidiuretic hormone
hypernatremia in a critically ill patient, and develop TBI Traumatic brain injury
an appropriate treatment plan. UUN Urine urea nitrogen
2. Discuss the causes and treatment of common
intracellular electrolyte disorders.
3. Differentiate among the causative factors for Self-Assessment Questions
metabolic acidosis and alkalosis, and construct a Answers and explanations to these questions may be
therapeutic treatment algorithm. found at the end of this chapter.
4. Specify the appropriate route (parenteral or enteral)
of nutrition administration, amount of nutrients, and 1. Which is the most appropriate indication for
particular component formulation to be provided to parenteral nutrition (PN)?
a given critically ill patient. A. Severe anorexia.
5. Identify appropriate markers for assessing the B. Lack of bowel sounds.
tolerance, safety, and efficacy of enteral or parenteral C. Ileus.
nutrition therapy. D. High gastric residual volume (GRV).
6. Describe methods for ensuring appropriate glycemic
control in critically ill patients. 2. Which is the most appropriate replacement fluid
7. Identify pertinent drug-nutrient interactions, and for a patient with significant nasogastric (NG) fluid
provide recommendations for the safe and effective drainage?
delivery of medications to patients receiving enteral
A. 0.9% sodium chloride and potassium chloride
or parenteral nutrition therapy.
20 mEq/L.
8. Discuss current controversies in the initiation,
B. 0.45% sodium chloride and potassium chloride
management, and monitoring of nutrition therapy
20 mEq/L.
for the critically ill patient.
C. 5% dextrose in 0.225% sodium chloride and
potassium chloride 20 mEq/L.
D. Lactated Ringer solution.
Abbreviations in This Chapter
3. Which trace mineral would best be increased for a
ABG Arterial blood gas PN-dependent patient with intractable diarrhea?
AG Anion gap
A. Zinc.
AKI Acute kidney injury
B. Copper.
ALI Acute lung injury
C. Iodine.
ARDS Acute respiratory distress syndrome
D. Manganese.
BEE Basal energy expenditure
BG Blood glucose
4. A 70-year-old man admitted to the intensive care
CKD Chronic kidney disease
unit (ICU) for sepsis was recently given a diagnosis
ECF Extracellular fluid
of syndrome of inappropriate antidiuresis. His
EN Enteral nutrition
serum sodium acutely fell from 130 mEq to 115
GRV Gastric residual volume
mEq/L during the past 3 days, and he recently seized
IBW Ideal body weight
secondarily to this problem. Which would be the
ICF Intracellular fluid
most appropriate treatment option?
ICU Intensive care unit
LOS Length of stay A. Intravenous 0.9% sodium chloride.
NB Nitrogen balance B. Intravenous DDAVP (desmopressin acetate).
NG Nasogastric C. Intravenous 3% sodium chloride.
D. Intravenous conivaptan.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-52
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

5. Other than the absorption/infusion rate, which best 8. A 45-year-old man (weight 90 kg) admitted to the
explains why enteral potassium administration is ICU after operative management of necrotizing
safer than parenteral potassium administration? pancreatitis is given PN consisting of 350 g of
A. Bioavailability of potassium is significantly dextrose, 130 g of amino acids, and 90 g of lipid
lower with enteral versus parenteral emulsion (20%) daily. A 24-hour urine collection
administration. for determining the nitrogen balance (NB) shows a
B. Feed-forward sensing of changes in mesenteric urine urea concentration of 900 mg/dL for a urine
potassium concentration increases urinary output of 2700 mL. He received 100% of his PN
potassium excretion. solution, and there was no significant change in his
C. Potassium chloride elixir is likely to cause blood urea nitrogen (BUN) during the NB study.
diarrhea and reduce potassium absorption. Which most accurately depicts his NB?
D. Wax matrix tablets sequester potassium release A. -15 g/day.
throughout the gastrointestinal (GI) tract. B. -7.5 g/day.
C. -2.5 g/day.
6. A 45-year-old woman with a history of celiac D. +2.5 g/day.
disease and alcoholism is admitted to the ICU. There
is no evidence of significant acute or chronic
blood loss. Her hematocrit is 30%, hemoglobin is
9 g/L, and mean corpuscular volume is 105 fL. Her
serum methylmalonic acid concentration is within
normal limits, and her serum homocysteine
concentration is elevated. Serum ferritin is within
normal limits. Which does this patient most likely
has a deficiency of?
A. Iron.
B. Thiamine.
C. Folic acid.
D. Cyanocobalamin.

7. A 40-year-old man (weight 60 kg) is admitted to

the trauma ICU after a motor vehicle accident. He
is noted to have a serum magnesium concentration
of 1.2 mg/dL, and his family states he has a history
of alcohol abuse. He is given magnesium sulfate 6 g
intravenously over 4 hours by the primary service.
His repeat serum magnesium concentration on the
following day is 1.8 mg/dL. Which would be the
most appropriate treatment for this patient?
A. No treatment is necessary because his serum
magnesium concentration is normal.
B. If a repeat serum magnesium concentration is
2 mg/dL or greater, no additional magnesium
therapy is indicated.
C. Supplemental calcium therapy should be given
concurrently with the magnesium therapy.
D. Additional magnesium therapy should be
given daily over the next 4–5 days.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-53
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

I. FLUIDS AND ELECTROLYTES

A. General Overview
1. Body water compartments
a. Total body water (TBW): About 60% of body weight for males; about 55% of body weight for
females; lower percentage for those who are obese and elderly (0.5 L/kg for males; 0.45 L/kg for
females)
b. About 60% of TBW is intracellular.
c. About 40% of TBW is extracellular water (about 80% is interstitial fluid; about 20% plasma
volume)
2. Estimating daily fluid requirements
a. 30–35 mL/kg (overestimates large person, underestimates small person)
b. 100 mL/kg for the first 10 kg, 50 mL/kg for the next 10 kg, and 20 mL/kg thereafter
c. Increased insensible losses with fever (around 10%–15% for every degree Celsius greater
than 37°C)

Table 1. Effect of Body Temperature on Insensible Fluid Losses (Surgery 1968;64:154-64)

Rectal Temperature (°C) No. of Patients Mean Fluid Loss (mL/m2/day)
36.7–37.7 205 552
37.8–38.2 160 600
38.3–38.8 48 768
38.9–40 14 840

3. Estimating electrolyte requirements

a. Approximate electrolyte concentrations in the extracellular and intracellular fluids (ECF and ICF)
(Fluid, Electrolyte, and Acid-Base Disorders, Vol 1. New York: Churchill Livingstone, 1985:1-38.)

Table 2. Electrolyte Concentrations in the ECF and the ICF

Extracellular Fluid Intracellular Fluid
Electrolyte (mEq/L) (mEq/L)
(plasma) (interstitial) (muscle)
Sodium 140 145 12
Potassium 4.5 4.8 160
Chloride 104 117 2
Bicarbonate 24 27 10
Calcium 5.0 2.8 —
Magnesium 1.5 1.0 34
Phosphate 2 2.3 54

b. “Normal” daily requirements

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-54
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

Table 3. “Normal” Daily Requirements

Sodium 50–100 mEq/day
Potassium 0.5–1.5 mEq/kg
Phosphorus 30–24 mmol
Magnesium 24–32 mEq
Calcium 10–20 mEq
Chloride a
80–120 mEq
Acetatea 80–120 mEq
Depending on the acid-base status of the patient.
a

c. Approximate electrolyte content of GI fluids (in milliequivalents per liter) (About Surgery: A
Clinical Approach. Philadelphia: Elsevier Health Sciences, 1996:5-17; Acta Chir Scand Suppl
1963;(suppl 306):301-65.

Table 4. Electrolyte Content of GI Fluids

Average Daily Sodium Potassium Chloride Bicarbonate Magnesium
Fluid Volume (mL) (mEq/L) (mEq/L) (mEq/L) (mEq/L) (mEq/L)
Stomach 1000–2000 60–90 10–15 100–130 — 0.9
Duodenum 400–600 140 5–10 90–120 80 —
Small intestine 2000–2500 140 5–10 90–120 30–40 6–12
Colon < 300 60 20–30 50 — 6–12
Pancreas 600–800 140 5–10 75 115 0.4
Bile 300–600 140 5–10 100 30 1.1

d. Electrolyte composition of common intravenous solutions (in milliequivalents per liter)

Table 5. Electrolyte Composition of Common Intravenous Solutions

Sodium Potassium Chloride Bicarbonate Calcium Magnesium
Solutions (mEq/L) (mEq/L) (mEq/L) (mEq/L) (mEq/L) (mEq/L)
5% dextrose in water — — — — — —
0.9% sodium chloride
154 — 154 — — —
(normal saline)
0.45% sodium chloride
77 — 77 — — —
(one-half normal saline)
5% dextrose in 0.225%
sodium chloride (5%
34 — 34 — — —
dextrose in one-fourth
normal saline)
3% sodium chloride
513 — 513 — — —
(hypertonic saline)
Lactated Ringer solution 130 4 109 28 2.7 —

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-55
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

4. Regulation of effective circulating volume

a. Kidney – Renin-angiotensin-aldosterone system
b. Extra-renal (carotid sinus, atrium) – Sympathetic nervous system (epinephrine and
norepinephrine) and atrial natriuretic peptide

Table 6. Hemodynamic Assessment ([Link]

Ejection Adjusted End-Diastolic Volume Index Adjusted End-Diastolic Volume Index

Fraction for Normal Subjects for Critically Ill Patients
20 200 240
30 150 180
35 125 150
40 100 120
50 50 60

5. Regulation of plasma osmolality

a. Vasopressin release
b. Thirst
c. Osmoreceptor sensitivity

B. Water and Sodium Disorders

1. Dehydration: As evidenced by decreased urine output (unless patient has glycosuria or diuretic
therapy), increased serum urea nitrogen/serum creatinine ratio (SUN/SCr greater than 20), insufficient
net fluid balance “from nursing records), poor skin turgor, dry mucous membranes, orthostatic
hypotension, “contraction alkalosis.” Increased losses
a. Fever
b. GI fluids
2. Volume excess: As evidenced by the presence of peripheral/sacral/pulmonary edema, anasarca,
congestive heart failure, acute kidney injury (AKI)
a. Excessive fluid intake
b. Impaired ability to excrete excess water and sodium (e.g., heart failure, cirrhosis with ascites,
renal failure)
3. Hyponatremia
a. Classic evaluation
i. Exclude hyperglycemia, mannitol, and glycine for unmeasured osmoles (hypertonic
hyponatremia).
ii. Exclude factitious/pseudo-hypoglycemia (isotonic hyponatremia).
iii. Evaluate ECF volume (increased, normal, decreased).
iv. Consider use of urine sodium and osmolality, if necessary.
v. Consider patient conditions/diagnoses.
b. 2014 European Society of Endocrinology guidelines (Intensive Care Med 2014;40:320-31).
i. Exclude hyperglycemia and other causes of non-hypotonic hyponatremia.
ii. Evaluate urine sodium and osmolality.
iii. Assess ECF and arterial blood volume, diuretics, presence of kidney disease
c. Treatment of hyponatremia
i. Acute or severe symptoms? – Immediate treatment with hypertonic saline
ii. ECF expanded – Fluid and sodium restriction

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-56
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

iii. ECF reduced and low urine sodium – Give sodium and fluids (treat etiologies if possible);
reduce diuretic therapy.
iv. ECF normal – Consider syndrome of inappropriate antidiuresis or secondary adrenal
insufficiency – Fluid restriction first; consider conivaptan or tolvaptan; fluid restriction with
use of 0.9% sodium chloride solution with or without diuretic therapy
4. Hypernatremia
a. Excessive sodium intake (hypertonic saline, 0.9% sodium chloride solution, lactated Ringer solution
b. Dehydration

Patient Case

Questions 1–3 pertain to the following case.

A 55-year-old woman (70 kg) admitted to the ICU for pneumonia and respiratory failure develops a serum sodium
of 125 mEq/L on her fifth day of hospital admission. Her other laboratory values include a serum potassium of 4.6
mEq/L, chloride 100 mEq/L, total carbon dioxide (CO2) content 24 mEq/L, BUN 20 mg/dL, SCr 1.1 mg/dL, and
glucose 167 mg/dL. She is currently receiving a 1-kcal/mL, 62-g/L enteral feeding formula at 60 mL/hour and a
5% dextrose in 0.45% sodium chloride infusion at 25 mL/hour. Her fluid balance has ranged from +300 to +600
mL/day during the past 3 days. She has no evidence of any significant amount of edema. Her measured serum
osmolality is 265 mOsm/kg, urine osmolality is 490 mOsm/kg, and urine sodium is 67 mEq/L.

1. Which is the most likely etiology for the patient’s hyponatremia?

A. Factitious hyponatremia.
B. Adrenal insufficiency.
C. Cerebral salt wasting.
D. Syndrome of inappropriate antidiuretic hormone (SIADH).

2. Which would be the most appropriate treatment for this woman?

A. Give sodium chloride tablets 1 g three times daily.
B. Limit fluids.
C. Change the intravenous fluid to 0.9% sodium chloride.
D. Provide a short-term intravenous infusion of 3% sodium chloride.

3. Which change in the enteral feeding formula would be best for this patient?
A. Add sodium chloride 100 mEq/L to the current formula.
B. Change the formula to a fish oil–enriched product.
C. Change the formula to a low-carbohydrate, high-fat product.
D. Change the formula to a 2-kcal/mL formula and decrease the rate.

C. Disorders of Potassium Homeostasis

1. Potassium homeostasis overview
a. 98% intracellular
b. Total body stores: 35–50 mEq/kg in normal healthy adults; 25–30 mEq/kg if significantly
undernourished
c. Normal serum concentration: 3.5–5.2 mEq/L
d. Serum concentration can be influenced by changes in pH (for every 0.1 increase in arterial pH,
serum potassium will decrease by around 0.6 mEq/L [range 0.4–1.3 mEq/L]) (J Clin Invest
1956;35:935-9), and vice versa.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-57
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

e. Average daily requirement: About 0.5–1.2 mEq/kg

f. Kidney is primary route of elimination.
g. Losses can be extensive with severe diarrhea or body fluid drainages (see Table 4 ).
h. Magnesium status can influence potassium homeostasis (Crit Care Med 1996;24:38-45; J Am Soc
Nephrol. 2007;18:2649-52; Arch Intern Med 1992;152:40-5).
i. Magnesium serves as a cofactor for the Na-K-ATPase pump.
ii. Magnesium closes potassium channels in distal nephron.
2. Hypokalemia
a. Definition: Serum potassium less than 3.5 mEq/L, though most ICUs prefer to keep patients at 4.0
mEq/L or greater, if possible
b. Signs and symptoms: Weakness, cramps, cardiac arrhythmias (ST depression, QT prolongation,
flat T wave, U wave). If severe hypokalemia (e.g., serum potassium less than 2 mEq/L), flaccid
paralysis, ileus
c. Etiologies:
i. Inadequate intake (rare; kidneys can usually adapt)
ii. Increased losses
(a) GI fluid losses (e.g., diarrhea, fistula, drainages)
(b) Hypomagnesemia
(c) Medications (diuretics, amphotericin B, mineralocorticoid excess, penicillins)
(d) Polyuria (diabetes insipidus)
(e) Renal potassium excretion (type I/distal and type II/proximal renal tubular acidosis)
(f) Diabetic ketoacidosis
iii. Increased requirements (building of new muscle/tissue – refeeding syndrome)
iv. Extracellular to intracellular shift
(a) Medications (β-adrenergic agonists, including albuterol, sodium bicarbonate or other
alkalinizing agents; insulin)
(b) Acute alkalemia
(c) Hypothermia
(d) Pentobarbital/thiopental?
d. Estimating total body potassium deficit
i. Transtubular potassium gradient to assess the contribution of the kidney to the hypokalemia
is no longer recommended because of the variability in urea reabsorption in the cortical
collecting duct, which alters solute removal. Use spot urine potassium if kidney-based
etiology of hypokalemia is unclear.
ii. Sterns equation (Medicine (Baltimore) 1981;60:339-54) (most commonly used method):
Potassium deficit (mEq) = 100 x (4.4 − serum potassium)
iii. Segal equation (Core Concepts in the Disorders of Fluid, Electrolytes, and Acid-Base
Balance. New York: Springer, 2013:49-102.)
(makes nonlinear assumption of deficit estimate): Potassium deficit (mEq per 70 kg):
3300 x exp (-serum potassium/1.5) − 200
iv. Sterns equation (Medicine (Baltimore) 1981;60:339-54) adjusted to the extent of
malnourishment:
Estimate total body potassium (mEq/kg) by:
Well nourished: 35–50 mEq/kg
Undernourished: 20–35 mEq/kg
Serum potassium of 3.0 mEq/L approximates a 10% total body deficit.
Serum potassium of 2.5 mEq/L approximates a 20% total body deficit.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-58
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

e. Treatment:
i. Treat, alleviate, or reduce the potential etiologies for hypokalemia, if possible.
ii. Ensure hypokalemia is not at least partly attributable to hypomagnesemia.
iii. The estimated deficit should be replaced over 1–3 days (depending on the extent of deficit;
the larger a deficit, the longer the repletion period) by giving boluses and increasing the
potassium content in intravenous fluids and PN/enteral nutrition (EN) solutions.
iv. Enteral or oral potassium replacement is the preferred and safer route of delivery because of
the time of absorption and feed-forward regulation of potassium homeostasis . (Ann Intern
Med 2009;150:619-25), administration of potassium chloride liquid directly into the small
bowel (by a jejunal or duodenal feeding tube) should be avoided because of its osmolality,
which can lead to abdominal cramping, distension, and diarrhea.
v. Short-term infusions of potassium chloride or potassium phosphate should be given only by
central vein. Potassium chloride can be given at 20 mEq/hour if the patient has continuous
electrocardiography (ECG) monitoring in the ICU. 10 mEq/hour is safest if the patient is
asymptomatic or not in the ICU. Potassium chloride comes prepackaged as 20- and 40-
mEq units in sterile water for injection. Peripheral intravenous solutions should not contain
potassium chloride at more than 40–60 mEq/L in an effort to reduce the pain associated with
the infusion of a concentrated potassium chloride solution and to prevent inappropriate rapid
and excessive potassium chloride dosing.
vi. Empiric intravenous potassium dosing. This algorithm (used by the Regional One Health
Nutrition Support Service) is only a guide for critically ill patients and may need to be
adjusted according to patient body size, renal function, and ongoing severity of losses.

Table 7. Empiric Intravenous Potassium Dosing

Serum Potassium Potassium Chloride Dosage
(mEq/L) (mEq)a Laboratory Tests
3.5–3.9 40 mEq x 1; increase in IV/PN/EN Obtain BMP, magnesium next AM
Obtain BMP, magnesium next AM; may wish to
3–3.4 40 mEq x 2; increase in IV/PN/EN get stat K 1–2 hours after second 40-mEq bolus,
especially if losses are thought to be high. Reassess
Obtain repeat serum K 1–2 hours after second
40-mEq bolus and reassess; may need one or two
2–2.9 40 mEq x 3+; increase in IV/PN/EN
additional boluses; repeat. Check serum magnesium
next AM. Reassess
a
Potassium phosphate may be considered in lieu of potassium chloride if concurrent hypokalemia and hypophosphatemia (very common in critically
ill patients receiving EN/PN). Thirty millimoles of potassium phosphate is equivalent to 44 mEq of K.
AM = morning; BMP = basic metabolic panel; IV = intravenous; K = potassium.

vii. The accuracy of the historical assumption of “a 0.5 to 0.6 mEq/L increase in serum potassium
will occur for every 40 mEq of intravenous potassium administered” (Arch Intern Med
1990;150:613-7; J Clin Pharmacol 1994;34:1077-82)
viii. Serum potassium concentrations are equilibrated within 1–2 hours after completion of the
intravenous potassium chloride infusion (Crit Care Med 1991;19:694-9; J Clin Pharmacol
1994;34:1077-82) and are recommended for patients with severe and/or complicated cases of
hypokalemia.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-59
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

3. Hyperkalemia
a. Definition: Serum potassium greater than 5.2 mEq/L, although usually not a significant problem
until serum potassium approaches 6 mEq/L
i. Rule out factitious hyperkalemia (hemolysis of blood sample, white blood cell count greater
than 10 x 103 cells/mm3, platelet count greater than 400,000/mm3).
ii. Assess arterial blood gas (ABG) (severe acidosis).
iii. Immediately after a large blood transfusion?
b. Signs and symptoms: Peaked and tented T waves on ECG, symptoms similar to those of
hypokalemia (weakness, paralysis)
c. Etiologies:
i. Drugs – Potassium-sparing diuretics (spironolactone, amiloride, triamterene), angiotensin-
converting enzyme inhibitors, angiotensin receptor blockers, nonsteroidal anti-inflammatory
drugs, heparin, trimethoprim, octreotide
ii. Excessive intake (usually in combination with compromised renal function) – Be sure to
examine all intravenous fluids, enteral and PN regimens, penicillin G (1.7 mEq of potassium
per million units), packed red blood cells (potassium 7.5–13 mEq/L)
iii. Renal dysfunction (chronic kidney disease [CKD], AKI)
iv. Hyporeninemic hypoaldosteronism
v. Tissue catabolism (chemotherapy, rhabdomyolysis)
vi. Severe acidemia
vii. Elderly patients
d. Treatment
i. Calcium gluconate (10%) 1- to 2-g intravenous slow push (especially if ECG changes) to
stabilize the myocardium
ii. Regular human insulin 10 units intravenously plus or minus 50 g of dextrose intravenously
(results in only temporary redistribution)
iii. Sodium bicarbonate 50–100 mEq intravenously (especially if acidemic – results in only
temporary redistribution)
iv. Sodium polystyrene sulfonate 25–50 g (intragastric administration preferred) – Increases
potassium elimination
v. Albuterol (results in only temporary redistribution)
vi. Loop diuretics – Increase potassium elimination
vii. Hemodialysis – Increases potassium elimination
viii. Make sure no exogenous sources of potassium (e.g., intravenous fluids, EN, PN); to reduce
intake with EN or PN; use a “renal” (no or low-electrolyte formulation, if necessary)

D. Disorders of Magnesium Homeostasis

1. Magnesium homeostasis overview
a. 99% intracellular (17% of total body content is in the muscle) or in the skeleton
b. Total body stores: Around 2000 mEq
c. Normal serum concentration: 1.8–2.4 mg/dL (about 30% bound to protein)
d. Average daily requirement: Around 24–40 mEq/day
e. Kidney is primary route of elimination (around 70% reabsorbed in ascending loop of Henle) and is
without any hormonal regulation of renal magnesium reabsorption.
f. Losses can be extensive with severe diarrhea or body fluid drainages (see Table 4).
g. Magnesium depletion can influence potassium and calcium homeostasis.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-60
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

2. Hypomagnesemia
a. Definition: Although the lower limit of normal for serum magnesium concentrations is 1.8 mg/
dL (1.5 mEq/L), most clinicians define significant hypomagnesemia as 1.5 mg/dL (1.3 mEq/L) or
less. Many ICUs have a target serum magnesium concentration greater than 2 mg/dL (1.8 mEq/L).
Serum concentrations of magnesium may be slightly falsely lowered in the presence of significant
hypoalbuminemia.
b. Signs and symptoms: Muscle weakness, cramping, paresthesias, Chvostek and Trousseau signs,
tetany, QT prolongation, hypokalemia, hypocalcemia
c. Etiologies:
i. GI losses (especially diarrhea) – Average stool loss of about 6 mEq/L; up to 10–12 mEq/L or
greater for secretory diarrheal losses
ii. Alcohol (increased renal excretion; impaired absorption; poor nutritional status of patients
who abuse alcohol)
iii. Sepsis/critical illness (increased urinary excretion – several factors)
iv. Pancreatitis (partly attributable to calcium-magnesium soap formation in peritoneum)
v. Thermal injury/traumatic brain injury (TBI) (increased urinary excretion – several factors)
vi. Drugs – Diuretics, amphotericin B, cyclosporin/tacrolimus, foscarnet, pentamidine, cisplatin/
carboplatin/ifosfamide/cetuximab, lactulose/orlistat, aminoglycosides (?), digoxin (?), proton
pump inhibitors (?)
vii. Polyuria (osmotic agents, hypercalcemia, ureagenesis)
d. Estimating magnesium deficit: For a serum magnesium concentration of less than 1.5 mg/dL (1.3
mEq/L), a 1- to 2-mEq/kg deficit can be expected.
e. Treatment:
i. Treat the etiology (if possible).
ii. Successful treatment of hypomagnesemia usually takes 3–5 days of intravenous therapy.
Intramuscular magnesium therapy for replacement therapy is inadvisable because of the limit
on volume per injection site with respect to dosage requirements and tissue irritation.
iii. Intravenous magnesium sulfate 32–48 mEq/day (4–6 g/day) – Suggested to be sufficient to
maintain serum magnesium within 2–2.5 mg/dL for most magnesium-deficient patients (Crit
Care Med 1996;24:38-45; Annu Rev Med 1981;32:245-259).
iv. Empiric intravenous magnesium sulfate dosing. This algorithm (used by Regional One
Health’s Nutrition Support Service) (Nutrition 1997;13:303-8) was designed primarily for
trauma and thermally injured patients but can likely be universally applied to other critically
ill patient populations. The therapy may need to be adjusted according to renal function
and ongoing severity of losses. Our empiric approach to dosing intracellular electrolytes
for patients with significant renal impairment is to give one-half the recommended dose
(see Table 7). However, electrolyte therapy for patients with renal impairment must be
individualized according to individual response.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-61
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

Table 8. Empiric Intravenous Magnesium Sulfate Dosing

Serum Magnesium (mg/dL) Dose (g/kg)
1.6–1.8 0.05
1–1.5 0.1
<1 0.15
For ease of use and preparation, intravenous magnesium sulfate should be ordered in 2 g units (e.g., 2 g, 4 g, 6 g). The drug should be mixed in
100–250 mL of normal saline or 5% dextrose and given at a rate no faster than 1 g (8 mEq) per hour (Nutrition 1997;13:303-8). Maximum dose
should be held at a ceiling of 8–10 g per administration. Magnesium concentrations are often elevated for several hours or longer after an infusion
because it takes about 48 hours for the magnesium to fully redistribute to the body tissues (Nutrition 1997;13:303-8)

v. Oral magnesium: Difficult to successfully accomplish in critically ill patients because of the
adverse GI effects of oral magnesium (e.g., diarrhea) and the high elemental magnesium
doses required to achieve repletion. Although it has been inferred that certain oral magnesium
products are better tolerated than others (e.g., gluconate vs. oxide), this tolerability likely
pertains to the elemental magnesium content of the products. The lower the elemental
magnesium content, the more tolerable the oral product. However, the lower the magnesium
content, the more difficult it is to achieve magnesium repletion for a patient with significant
magnesium depletion.

Table 9. Common Oral Magnesium Products

Strength Elemental Mg Content

Salt Form (mg) (mEq) Usual Dosing
Oxide 400 19.8 1–2 tablets twice or three times daily
140 6.9

Gluconate 500 2.2 1–2 tablets twice or three times daily

Chloride 100 2.6 1–2 tablets twice or three times daily
Mg = magnesium.

Table 10. Oral Magnesium Content

Salt Form % Elemental Magnesium Content
Oxide 60
Carbonate 45
Hydroxide 42
Citrate 16
Lactate 12
Chloride 12
Sulfate 10
Gluconate 5

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-62
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

3. Hypermagnesemia
a. Definition: Serum magnesium concentration greater than 2.4 mg/dL; patients usually do not
experience symptoms until the serum magnesium concentration exceeds about 4 mg/dL
b. b. Signs and symptoms: Hypotension, decreased deep tendon reflexes, cardiovascular
manifestations (e.g., bradycardia, somnolence, muscle paralysis, arrhythmias) generally do not
occur until serum concentrations are greater than 4 mg/dL
c. Etiologies: Renal failure or impairment early post-infusion elevation of serum magnesium
concentration, excessive dosing of magnesium/antacids, post-cathartic use (e.g., magnesium citrate)
– To develop hypermagnesemia, these events usually occur together with renal impairment.
d. Treatment:
i. Remove source of magnesium intake.
ii. Intermittent slow bolus doses of calcium gluconate (2 g) over 5–10 minutes until severe
symptoms abate (the effect of calcium is transient, and repeat therapy may be needed as
frequently as every hour)
iii. Ventilate the patient, if necessary.
iv. Hemodialysis

Patient Case

Questions 4 and 5 pertain to the following case.

A 55-year-old man (weight 70 kg) is admitted to the ICU after a total colectomy and hepatic resection for stage IV
colon cancer. He has a history of an unintentional 20-lb weight loss during the past several months before hospital
admission. He is a 2 pack/day tobacco smoker and has a history of frequent alcohol consumption. He is currently
weaning from the ventilator; however, his NG volume output is greater than 2 L/day, and PN is initiated. The
intent is to achieve goal caloric and protein intake within 3–4 days. His serum laboratory values are as follows:
sodium 140 mEq/L, potassium 3.2 mEq/L, chloride 102 mEq/L, total CO2 content 25 mEq/L, BUN 14 mg/dL,
SCr 0.9 mg/dL, calcium 8.1 mg/dL, phosphorus 2 mg/dL, magnesium 1.4 mg/dL, and albumin 2.5 g/dL.

4. Which potassium-phosphorus dosing regimen would be most appropriate for this patient?
A. Potassium chloride liquid 40 mEq per NG tube for two doses, 2 Neutra-Phos capsules in water
per NG tube.
B. Potassium phosphate 30 mmol intravenously x 1 dose, followed by two 40-mEq potassium chloride
doses per NG tube.
C. Potassium chloride 40 mEq intravenously x 1 dose and potassium phosphate 45 mmol intravenously x
1 dose.
D. Potassium chloride 40 mEq intravenously x 1 dose and potassium phosphate 30 mmol intravenously x
1 dose.

5. In addition to potassium and phosphorus supplementation, the patient is given magnesium sulfate 6 g
intravenously over 6 hours. His repeat serum magnesium the next day is 1.9 mg/dL. Which therapeutic option
would be best for this patient?
A. Give magnesium oxide 500 mg twice daily for 4–5 days.
B. Give magnesium sulfate 2–4 g intravenously daily for 4–5 days.
C. Give a second dose of 6 g of magnesium sulfate intravenously.
D. No additional treatment is necessary.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-63
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

E. Disorders of Calcium Homeostasis

1. Calcium homeostasis overview
a. Most prevalent intracellular cation in body; 99% of body’s calcium in bone; highly protein bound
in plasma
b. Total body stores: About 1–1.2 kg of calcium
c. Normal serum concentration: 8.5–10.5 mg/dL; normal serum ionized concentration 1.12–1.32
mmol/L
d. Serum concentration can be influenced by:
i. Changes in plasma albumin concentration – For every 1 g/dL in serum albumin below 4 g/
dL, serum calcium will decrease by around 0.8 mg/dL (Clin Chim Acta 1971;35:483-9); do
not use in critically ill patients (inaccurate) (JPEN J Parenter Enteral Nutr 2004;28:133-41).
Use ionized calcium concentration for critically ill patients. However, most (85%) critically ill
patients with a total serum calcium concentration less than 7 mg/dL are hypocalcemic (ionized
serum calcium of 1.12 mmol/L or less) (Nutr Clin Pract 2007;22:323-8).
ii. Changes in pH (for every 0.1-unit increase in arterial pH, serum ionized calcium will decrease
by about 0.05 mmol/L (Arch Pathol Lab Med 2002;126:947-50) because of increased protein
binding
e. Average daily requirement: 15 mEq/day intravenously with PN (Ann Surg 1983;197:1-6); 1–3 g/
day orally
f. Kidney is primary route of elimination.
g. Magnesium status can influence calcium homeostasis.
i. Hypomagnesemia results in end-organ resistance to parathyroid hormone.
ii. Hypomagnesemia may impair parathyroid hormone secretion.
iii. Hypocalcemia will correct within 2 days after hypomagnesemia is corrected.
2. Hypocalcemia
a. Definition: Corrected serum total calcium less than 8.5 mg/dL (non-ICU patients); ionized serum
calcium concentration less than 1.12 mmol/L
b. Signs and symptoms: Tingling, paresthesias, hyperactive deep tendon reflexes, Chvostek and
Trousseau signs, prolonged QT interval
c. Etiologies
i. Critical illness
ii. Continuous renal replacement therapy (CRRT) (citrate anticoagulation)
iii. Massive blood transfusion
iv. Hypomagnesemia
v. Hyperphosphatemia
vi. Pancreatitis
vii. Drugs (amphotericin B, cisplatin, cyclosporine, foscarnet, bisphosphonates, loop diuretics)
viii. Malabsorption
ix. Hypoparathyroidism
x. Chronic kidney injury/AKI
xi. Vitamin D deficiency
xii. Severe alkalemia
d. Treatment: Regional One Health Nutrition Support Service intravenous calcium gluconate
dosing guidelines (JPEN J Parenter Enteral Nutr 2005;29:436-4; JPEN J Parenter Enteral Nutr
2007;31:228-33; Nutrition 2007;23:9-15)

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-64
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

Table 11. Regional One Health Nutrition Support Service Intravenous Calcium Gluconate Dosing Guidelines
Ionized Calcium
(mmol/L) Intravenous Calcium Gluconate Dosage (mEq)
1–1.12 2 g (9.3 mEq) calcium gluconate in 100 mL of 0.9% NaCl or D5W over 2 hours
≤ 0.99 4 g (18.6 mEq) calcium gluconate in 100 or 250 mL of 0.9% NaCl or D5W over 4 hours
For ease of use and preparation, calcium gluconate should be ordered in gram increments. Calcium chloride should be used (preferably) only in code
situations, not for routine replacement therapy, because the chloride salt contains about 2.5 times the amount of elemental calcium and can cause
tissue necrosis when given peripherally in contrast to calcium gluconate. However, during extreme circumstances, such as a national drug shortage
of intravenous calcium gluconate, calcium chloride can be given in 0.67- and 1.3-g doses in lieu of 2- and 4-g calcium gluconate doses. In addition,
calcium chloride should never be added to the PN solution unless there is no phosphate in the PN solution and the commercial amino acids used in
the PN solution do not contain phosphorus (some amino acid products do contain phosphate). A serum ionized calcium concentration determination
should be repeated several hours after completing the calcium gluconate infusion to allow equilibration (Nutrition 2007;23:9-15). Therapy that is
more aggressive may need to be considered for patients with tetany or life-threatening cardiac arrhythmias caused by hypocalcemia.
Intravenous calcium administration should be used with extreme caution in patients with severe hypokalemia or in those receiving digoxin or other
digitalis alkaloids.
Always check the serum phosphorus concentration because hyperphosphatemia can induce hypocalcemia, given the metastatic precipitation of
calcium phosphate in the soft tissues and lungs (usually associated with renal disease).
D5W = 5% dextrose in water; NaCl = sodium chloride.

Patient Case

Questions 6 and 7 pertain to the following case.

A 24-year-old man (weight 90 kg) is admitted to the trauma ICU postoperatively from repair of his duodenal,
jejunal, ileal, and colon injuries, hepatorrhaphy, and splenectomy after several gunshot wounds to the abdomen.
He also received 10 units of packed red blood cells. He is noted to have a serum ionized calcium concentration of
0.86 mmol/L, potassium of 4.6 mEq/L, and magnesium of 1.8 mg/dL. His SCr concentration is 0.8 mg/dL, and
his urine output is greater than 0.5 mL/kg/hour.

6. Which is the most likely etiology for his hypocalcemia?

A. Hypomagnesemia.
B. Excessive urinary diuresis.
C. Blood transfusion.
D. Critical illness.

7. Which therapeutic regimen would be best for this patient?

A. Calcium gluconate 2 g intravenously over 2 hours.
B. Calcium gluconate 4 g intravenously over 4 hours.
C. Calcium chloride 1 g intravenous push over 5–10 minutes.
D. No calcium therapy necessary.

3. Hypercalcemia
a. Definition: Corrected serum calcium greater than 10.5 mg/dL or ionized calcium greater than 1.32
mmol/L; signs and symptoms are more evident when total serum calcium of 12 mg/dL or greater
or ionized calcium of 1.5 mmol/L or greater
b. Signs and symptoms: Mental status changes, polyuria, shortened QT interval, bradycardia,
atrioventricular block
c. Etiologies:
i. Immobilization
ii. Chronic critical illness–associated metabolic bone disease

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-65
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

iii. Excessive calcium intake

iv. Hyperparathyroidism
v. Granulomatous diseases (tuberculosis, sarcoidosis)
vi. Malignancy
vii. Drugs (thiazide diuretics, vitamin D)
viii. Dehydration
d. Treatment:
i. Mobilize the patient (if possible); discontinue calcium from the PN solution.
ii. Intravenous fluids with 0.9% sodium chloride (if dehydrated) at 200 to 300 mL/hour x 48
hours or until rehydrated with or without furosemide 40–80 mg intravenously every 12 hours
iii. Calcitonin 4 units/kg intramuscularly every 12 hours; can be increased to 8 units/kg every
12 hours as needed
iv. Pamidronate 90 mg intravenously once for acute hypercalcemia not related to etiologies i or ii
v. Pamidronate 30 mg intravenously daily for 3 days if hypercalcemia caused by etiologies i or ii
vi. This author’s practice: Salmon calcitonin 200 units intramuscularly every 12 hours x 48
hours (rapid tachyphylaxis often limits therapy duration). Two hundred units is selected as the
dosage because it comes from the manufacturer in a 200-IU/mL vial and is usually near the
initial appropriate dosage range for many patients. If the patient is thought to have chronic
critical illness–associated metabolic bone disease or hypercalcemia from immobilization, it
is suggested to simultaneously add pamidronate 30 mg intravenously daily for 3 consecutive
days because of the delay in bisphosphonate’s onset of action (Chest.2000;118:761-6).
Monitor for hypocalcemia. Do not use bisphosphonates in patients with renal impairment.
vii. Parathyroidectomy for patients with primary hyperparathyroidism
viii. Prednisone 40 mg/day and greater for 10 days for patients with granulomatous diseases
(e.g., sarcoidosis, tuberculosis)

F. Disorders of Phosphorus Homeostasis

1. Phosphorus homeostasis overview
a. 99% intracellular, of which 85% is bound to bone
b. Extracellular pool of phosphorus: Around 600 mg (about 20 mmol), 10% protein bound
c. Normal serum concentration: 2.5–4.5 mg/dL
d. Serum concentration can be influenced by parathyroid hormone (increased parathyroid hormone
leads to increased urinary excretion of phosphorus), and alkalemia can decrease serum phosphorus
concentration.
e. Average daily requirement: Around 20 mg/kg/day
f. Kidney is primary route of elimination.
2. Hypophosphatemia
a. Definition: Serum phosphorus less than 2.5–3 mg/dL; severe hypophosphatemia less than 1–1.5
mg/dL
b. Signs and symptoms: Weakness, paresthesias; severe depletion can lead to congestive
cardiomyopathy, cardiac arrest, seizures, coma, respiratory arrest, rhabdomyolysis
c. Etiologies:
i. Alcoholism
ii. Malnutrition/refeeding syndrome
iii. Critical illness (especially trauma, TBI, thermal injury)
iv. Diabetic ketoacidosis
v. Hepatic resection
vi. Drugs – Insulin, catecholamines, antacids, sucralfate, calcium

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-66
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

vii. Alkalemia
viii. Malabsorption – Chronic diarrhea
ix. Hyperparathyroidism
x. Cancer (phosphatonins [e.g., fibroblast growth factor-23])
d. Treatment
i. Target serum phosphorus concentration when patient is in the ICU and ventilator-dependent:
4 mg/dL (N Engl J Med 1985;313:420-4; Inten Care Med 1995;21:826-31)
ii. Intravenous dosing guidelines

Table 12. Intravenous Phosphorus Dosing Guidelines

General Medical-Surgical Population
Serum Phosphorus Dosage (mmol/kg) (Crit Care Med High Requirementsa Population (JPEN J
(mg/dL) 1995;23:1504-11) Parenter Enteral Nutr 2006;30:209-14)
2.3–3 0.16 0.32
1.6–2.2 0.32 0.64
< 1.6 0.64 1
a
Patients with thermal injury (JPEN J Parenter Enteral Nutr 2001;25:152-9), those with trauma (especially those with a TBI) (JPEN J Parenter
Enteral Nutr 2006;30:209-14; Nutrition 2010;26:784-90), those malnourished with evidence of significant complications from refeeding syndrome,
or those with hepatic resections.
The drug should be mixed in 100–250 mL of normal saline or 5% dextrose in water and given at a rate no faster than 7.5 mmol/hour. Phosphorus
should always be ordered in millimoles for ease of use and preparation. For ease of use and preparation in the pharmacy, phosphorus should be
ordered in units divisible by 3 mmol (e.g., 15, 30, 45, 60) whenever possible.
Potassium phosphate salt can be used for patients with a serum potassium less than 4 (3 mmol P = 4.4 mEq potassium). Sodium phosphate salt
should be used for patients with a serum potassium of 4 or greater (3 mmol P = 4 mEq Na).

iii. Oral or enteral phosphorus dosing:

(a) Difficult to accomplish in patients not receiving EN because of single-entity products and
doses needed for repletion and adverse GI effects (e.g., diarrhea)
(b) Neutra-Phos and Neutra-Phos K (only 8 mmol of phosphorus per tablet/packet)
(c) Can add potassium phosphate/sodium phosphate injection to enteral feeding solution
(e.g., 30 mmol/L) or oral sodium phosphate solution 5–10 mL per liter of EN
(5 mL = 20 mmol of phosphorus)
3. Hyperphosphatemia
a. Definition: Serum phosphorus concentration greater than 5 mg/dL, usually not clinically relevant
until serum phosphorus is greater than 6 mg/dL
b. Signs and symptoms: Hypocalcemia and metastatic calcification (e.g., neuromuscular irritability,
prolonged QT interval, tetany) – Usually does not occur until the serum calcium-phosphorus
product approaches 55 or greater (Adv Exp Med Biol 1978;103:195-201).
c. Etiologies:
i. Renal failure
ii. Immobility
iii. Chronic critical illness–associated metabolic bone disease
iv. Excessive phosphorus intake
v. Vitamin D toxicity
vi. Tumor lysis syndrome
vii. Hypoparathyroidism

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-67
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

d. Treatment:
i. Reduce phosphorus intake (omit from PN solution, plus or minus reduce lipid content of PN
solution if a high-fat formulation [controversial as phosphorus in organic form: phospholipids
and not inorganic such as sodium phosphate]; change to “low- or no-electrolyte” renal enteral
formula).
ii. Phosphate binders (Am J Health Syst Pharm 2005;62:2355-61)

Table 13. Phosphate Binders

Phosphorus-Binding Recommended Empiric
Drug Available Strength Capacity Initial Dose
Calcium carbonatea 500, 750, 1000 mg per tablet Calcium 43 mg/g 1 g QID
Calcium acetatea 667 mg per tablet Calcium 106 mg/g 1334–2001 mg TID
800 mg per capsule or
Sevelamerb Sevelamer 80 mg/g 800–1600 mg TID
packet
250, 500, 750, 1000 mg per
Lanthanum Data not available 500 mg TID
tablet
Do not use if the patient is hypercalcemic.
a

Carbonate comes in a powder; easier for administering to tube-fed patients; less likely to worsen metabolic acidosis than gel capsule in patients
b

with renal failure because gel capsule is in hydrochloric acid salt form.
QID = four times daily; TID = three times daily.

II. ACID-BASE DISORDERS

A. Normal Homeostasis
1. Normal values

Table 14. Normal Blood Gas Values

Variables Arterial Blood Mixed Venous Blood
pH 7.35–7.45 7.31–7.41
Pco2 35–45 41–51
Po2 80–100 35–40
HCO3 22–26 22–26
Base excess -2 to +2 -2 to +2
O2 saturation > 95% 70%–75%
HCO3 = bicarbonate.

2. Interpreting ABGs
a. Acidemia (pH less than 7.35) versus alkalemia (pH greater than 7.45)
b. Acidemia and alkalemia refer to an abnormal pH being either low or high, respectively. Acidosis
and alkalosis refer to the metabolic or respiratory processes that led to the abnormal pH. Although
the terms emia and osis are similar, they are uniquely different.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-68
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

Table 15. Adverse Effects of Severe Acidemia (pH 7.25 or less)

Impaired cardiac output Increased metabolic demands
Peripheral ischemia (centralization of blood) Insulin resistance
Increased pulmonary vascular resistance Decreased ATP synthesis
Hypotension Increased protein breakdown
Increased risk of arrhythmias Hyperkalemia
Decreased catecholamine responsiveness Diaphragmatic fatigue/dyspnea
Obtundation/coma Hyperventilation

Table 16. Adverse Effects of Severe Alkalemia (pH of 7.55 or greater)

Arteriolar constriction Hypokalemia
Hypotension Hypophosphatemia
Increased risk of arrhythmias Hypocalcemia/tetany
Decreased coronary blood flow/decreased angina Organic acid production
threshold
Seizures Hypoventilation/hypercapnia/hypoxemia
Lethargy, delirium, stupor
For simple acid-base disorders, identify pH, Pco2, and HCO3 in that order. Whichever side of 7.40 the pH is on, the respiratory or metabolic
processes that coincide with that pH abnormality are the primary etiology. If the pH is less than 7.40, an elevated Pco2 (respiratory acidosis) or a
decreased HCO3 (metabolic acidosis) is the primary etiology. If the pH is greater than 7.40, a decreased Pco2 (respiratory alkalosis) or an increased
HCO3 (metabolic alkalosis) is the primary etiology. An easy introductory overview to acid-base disorders by Haber is provided in the references
(West J Med 1991;155:146-51).
However, sometimes more than one primary abnormality is present, or the anticipated compensatory process (metabolic or respiratory) is inadequate
and may be contributing to the acid-base disorder. As a result, various formulas have been developed to predict what may be considered adequate
compensation. However, many of these mathematical equations have limitations in their clinical utility and accuracy (Crit Care Med 2007;35:1264-
70; Am J Resp Crit Care Med. 2000;162:2246-51; J Trauma Acute Care Surg 2012;73:27-32; Clin J Am Soc Nephrol 2007;2:162-74; Arch Intern
Med 1992;152:1625-9) and can be difficult to memorize (West J Med 1991;155:146-51).
In addition, the issue of mixed acid-base disorders is confounded by several factors that can lead to errors in the interpretation of acid-base disorders.
This would include non–steady-state conditions as well as the inability for the patient to adequately compensate through the respiratory pathway
because of mechanical ventilator restrictions. Some of the more common equations for assessing acid-base disorders are discussed later in this
chapter.

3. Use of base excess

a. Reflects the amount of base needed in vitro to return the plasma pH to 7.40 at standard conditions
(Pco2 40 mm Hg, 37°C)
b. Base excess reflects the metabolic component to interpreting the ABG. Some have described
its use as a means for almost freeing the clinician from memorizing the “acid-base” correction
formulas (described as follows). Despite its simplicity, it has limitations (J Trauma Acute Care
Surg 2012;73:27-32). Crystalloid resuscitation (leading to hyperchloremic acidosis), exogenous
bicarbonate administration, ethanol ingestion, and acetate/bicarbonate buffer in hemodialysis or
CRRT solutions can lead to erroneous base excess calculations and errors in interpretation
(J Trauma Acute Care Surg 2012;73:27-32).
4. Compensatory response to acid-base disorders

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-69
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

Table 17. Anticipated compensation to acid-base disorders (Crit Care 2000;4:6-14)

Serum HCO3
Primary Disorder (mEq/L) Anticipated Pco2 (mm Hg)
Metabolic acidosis ≤ 22 (1.5 x HCO3) + 8 (± 2)
Metabolic alkalosis ≥ 28 (0.7 x HCO3) + 21
Primary disorder Pco2 (mm Hg) Anticipated serum HCO3 (mEq/L)
Respiratory acidosis (acute) a
> 45 [(Pco2 − 40)/10] + 24
[(Pco2 − 40)/3] + 24
Respiratory acidosis (chronic)a > 45
Or
HCO3 should increase by ~4 mEq/L per 10-mm Hg
increase in Pco2 > 40
Respiratory alkalosis (acute)a < 35 24 − [(40 − Pco2)/5]
24 − [(40 − Pco2)/2]
Respiratory alkalosis (chronic)a < 35
Or
For each 10-mm Hg decrease in Pco2, HCO3 should
decrease by ~5 mEq/L
a
Compensation is different for acute versus chronic respiratory disorders because it takes about 2 days for the kidneys to adapt to a persistent change
in respiratory status.

B. Respiratory Acidosis
1. Make sure it is not caused by excessive sedation/analgesia or overfeeding with EN/PN.
2. Metabolic compensation – See Table 17 .

C. Respiratory Alkalosis
1. Make sure the patient is getting adequate sedation/analgesia, fever/pneumonia is being treated; nicotine
and drug withdrawal regimen is/are appropriate
2. Metabolic compensation – See Table 17

D. Metabolic Acidosis
1. Use of the serum anion gap (AG)
a. Used to determine the etiology for the metabolic acidosis. AG is the difference between major
cations and anions in blood (trying to detect whether there is an abundance of unmeasured anions).
AG = (Na + K) – (Cl + HCO3)
b. Normal range is around 3–12 mEq/L. Because serum potassium is small, it is not often used when
calculating cations (e.g., Na+)
c. Some clinicians adjust AG for serum albumin (Crit Care Med 1998;26:1807-10) and phosphorus
(Crit Care Med 2007;35:2630-6) (serum albumin concentration should be multiplied by 2–2.3, and
serum phosphorus [milligrams per deciliter] should be multiplied by 0.5, and both should be added
to the anions [e.g., chloride plus bicarbonate]). Using this method, the adjusted AG (or sometimes
called the strong ion gap when referring to the physicochemical methodology for interpreting acid-
base disorders) should be close to 0 (± 2) if the patient does not have an AG acidosis.
d. Causes of an AG acidosis: One easy pneumonic to remember (there are others) is A MUD PIE:
A = aspirin (or other salicylates)
M = methanol
U = uremia (including rhabdomyolysis)
D = diabetes (diabetic ketoacidosis)

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-70
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

P = paraldehyde
I = infection or ischemia (lactic acidosis)
E = ethylene glycol or ethanol toxicity
e. Types of lactic acidosis (lactate greater than 18/dL and pH less than 7.35)
i. Type A: Hypoperfusion (cardiogenic or septic shock, regional ischemia, severe anemia)
ii. Type B: Metabolic – No tissue hypoxia
(a) B1 = sepsis without shock, liver disease, leukemia, lymphoma, AIDS (acquired
immunodeficiency syndrome)
(b) B2 = drugs/toxins (metformin, didanosine/stavudine/zidovudine, ethanol, linezolid,
propofol, propylene glycol toxicity caused by intravenous lorazepam or pentobarbital),
nitroprusside (cyanide) toxicity
(c) B3 = inborn errors of metabolism (pyruvate dehydrogenase deficiency)
f. Causes of a normal AG acidosis
2. Another easy pneumonic to remember (there are others) is ACCRUED.
A = Ammonium chloride/acetazolamide (urine bicarbonate loss)
C = Chloride intake (PN, intravenous solutions)
C = Cholestyramine (GI bicarbonate loss)
R = Renal tubular acidosis: Types I, II, and IV
U = Urine diverted into the intestine (e.g., ileal conduit, vesicoenteric fistula)
E = Endocrine disorders (e.g., aldosterone deficiency)
D = Diarrhea or small/large bowel fluid losses (e.g., enterocutaneous fistulas)
3. Use of the delta ratio for determining mixed acid-base disorders
Delta ratio =
ΔAG/ΔHCO3 = (measured AG−normal AG)/(normal HCO3−measured HCO3) =
(AG−14)/(24−measured HCO3)

Table 18. Interpreting delta ratio

Delta Ratio Assessment
< 0.4 Hyperchloremic normal AG acidosis
<1 High AG acidosis and normal AG acidosis
1–2 AG acidosis
High AG acidosis and concurrent metabolic alkalosis OR a preexisting compensated
>2
respiratory alkalosis

4. An alternative method (and perhaps a simpler approach) to the delta ratio is to calculate the “excess
gap” compared with the AG (West J Med 1991;155:146-51).
Excess gap = AG − 12 (12 being the upper limit of normal for AG).
5. The excess gap is then added to the measured serum bicarbonate concentration. If the sum is less than
a normal serum bicarbonate concentration (e.g., 28–30 mEq/L), a mixed AG and non-AG acidosis is
present. If the sum is greater than a normal bicarbonate concentration, the patient likely has an AG
acidosis and concurrent metabolic alkalosis.
6. Evaluation of respiratory compensation: See Table 17
7. Treatment
a. Aggressive interventional therapy unnecessary until pH less than 7.20–7.25
b. Treat primary etiology! This should be the focus of treating the acid-base disorder.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-71
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

c. Intravenous sources of alkali – Done conservatively in conjunction with treating primary disorder
whenever possible. The intent is not to normalize the pH but to improve the pH (definitely avoid
overcorrection).
i. Sodium bicarbonate – Most commonly used
ii. Sodium acetate – Available in PN solutions
iii. Sodium citrate – Used orally for patients with chronic kidney injury
iv. THAM (0.3 N tromethamine) – Possibly indicated for patients with chronic and severe
hypercapnia or extreme volume overload (very rarely used in our clinical practice; however,
some institutions have had to use THAM more often in recent years because of intravenous
crystalloid/bicarbonate shortages)
d. Total bicarbonate dose (mEq) = 0.5 x Wt (kg) x (24 − HCO3)
i. Give one-third to one-half of the calculated total dose (or 1–2 mEq/kg) over several hours to
achieve a pH of around 7.25 (avoid boluses if possible).
ii. Once the pH is around 7.25 or greater, slower correction without increasing bicarbonate more
than 4–6 mEq/L to avoid exceeding the target pH
iii. Serial ABGs (e.g., every 6 hours), watch rate of decrease in serum potassium
iv. Use of sodium bicarbonate injection is controversial in patients with lactic acidosis (Curr
Opin Crit Care 2008;14:379-83).
e. Adverse effects of sodium bicarbonate excess:
i. Hypernatremia, hyperosmolality, volume overload
ii. Hypokalemia, hypocalcemia, hypophosphatemia
iii. Paradoxical worsening of the acidosis (if the fractional increase in Pco2 production exceeds
the fractional bicarbonate change)
iv. Over-alkalinization
f. THAM – Shorter duration of action than sodium bicarbonate and appears not to decrease serum
potassium. (J Nephrol 2005;18:303-7)

Patient Case

Questions 8 and 9 pertain to the following case.

A 70-year-old woman (weight 50 kg) who underwent a radical cystectomy with ileal conduit urinary diversion
for bladder cancer develops a postoperative ileus and requires PN. Her NG fluid output is about 1–1.5 L/day,
and she has been receiving esomeprazole. Her current intravenous solutions include 0.45% sodium chloride
with potassium chloride 20 mEq/L at 50 mL/hour in addition to her PN. Total daily electrolyte intake from the
PN is as follows: sodium chloride 60 mEq/day, potassium acetate 40 mEq/day, sodium phosphate 15 mmol/
day, magnesium 24 mEq/day, and calcium gluconate 10 mEq/day. After several days of PN therapy, she has the
following serum electrolyte perturbations: sodium 141 mEq/L, potassium 3.9 mEq/L, chloride 117 mEq/L, total
CO2 content 22 mEq/L, BUN 14 mg/dL, and SCr 0.8 mg/dL. Her ABG revealed the following: pH 7.29, Po2 95
mmHg, Pco2 35 mm Hg, and bicarbonate 21 mEq/L.

8. Which best describes the patient’s type of acid-base disorder?

A. Hyperchloremic, normal AG acidosis.
B. AG acidosis.
C. AG acidosis with hyperchloremia.
D. Respiratory acidosis with concurrent metabolic alkalosis.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-72
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

Patient Case (continued)

9. Which is the most appropriate treatment algorithm for this patient?
A. Substitute the sodium chloride in the PN solution for sodium acetate.
B. Sodium bicarbonate 100 mEq intravenously.
C. Change the supplemental intravenous solution to Ringer lactate solution.
D. Add 100 mEq of sodium bicarbonate to the PN solution.

E. Metabolic Alkalosis: pH greater than 7.45; symptoms are not usually severe until pH is greater than
7.55–7.60
1. Assessment (to help guide treatment) based on urinary chloride
a. Saline responsive (urinary chloride less than 10 mEq/L)
i. Excessive gastric fluid losses
ii. Diuretic therapy (especially loop diuretics)
iii. Dehydration (contraction alkalosis)
iv. Hypokalemia
v. (Over-) Correction of chronic hypercapnia
b. Saline resistant (urinary chloride greater than 20 mEq/L)
i. Excessive mineralocorticoid activity (e.g., hydrocortisone)
ii. Excessive alkali intake
iii. Profound potassium depletion (serum potassium less than 3 mEq/L)
iv. Excess licorice (mineralocorticoid) intake
v. Massive blood transfusion
c. Respiratory compensation (highly variable and may not be possible for ventilator-dependent
patients)
d. Intravascular volume status (important for saline-responsive alkalemia)
2. Treatment – Saline-responsive alkalemia
a. Treat underlying cause (if possible).
b. Decreased intracellular volume? Give intravenous 0.9% sodium chloride infusion (with potassium
chloride, if necessary).
c. Increased intracellular volume? Acetazolamide 250–500 mg orally or intravenously once to four
times daily plus potassium chloride if necessary (Intensive Care Med 2010;36:859-63; Acta
Anaesthesiol Scand 1983;27:252-4; Crit Care Med 1999;27:1257-61).
i. Hydrochloric acid therapy if alkalosis persistent or initial pH greater than 7.6
(a) N or 0.2 N of hydrochloric acid (use 0.2 N for patients requiring fluid restriction).
Hydrochloric acid should be given by central venous administration, and it requires
delivery in a glass bottle.
ii. Dosage of hydrochloric acid:
(a) Chloride deficit (Arch Surg 1975;110:819-21):
Dose (mEq) = 0.2 L/kg x Wt (kg) x (103 − serum chloride)
(b) Bicarbonate excess (J Am Soc Nephrol 2000;11:369-75):
Dose (mEq) = 0.5 L/kg x Wt (kg) x (serum HCO3- 24)
(c) Dickerson’s empiric approach: Give one-half of calculated dose over 12 hours, repeat
ABG at 6 and 12 hours after initiating hydrochloric acid infusion, and readjust infusion
rate if necessary; continue therapy and monitoring until pH less than 7.5; then stop and
reassess

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-73
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

3. Treatment – Saline-unresponsive alkalosis: Treat underlying cause (if possible).

a. Exogenous corticosteroids – Decrease dose or use drug with less mineralocorticoid effect.
b. Excessive alkali intake – Alter regimen.
c. Profound hypokalemia (serum potassium less than 3 mEq/L) – Aggressive potassium supplementation
d. Rare causes: Endogenous mineralocorticoid excess (Bartter or Gitelman syndrome) –
Spironolactone, amiloride, or triamterene; consider surgery
e. Liddle syndrome: Amiloride or triamterene

F. The Stewart or Physicochemical Approach to Acid-Base Disorders – This emerging approach to acid-base
disorders is based on charge differences between ions. Essentially, acid-base disorders can be defined
by differences between the activity of abundant cations (sodium, potassium, ionized calcium, ionized
magnesium) and the activity of all abundant anions (chloride, lactate). The influence of albumin and serum
phosphate is also considered. Although this technique has advantages over the traditional method discussed
herein, the Stewart approach has been criticized by others because it offers no diagnostic or prognostic
advantages over the traditional/bicarbonate analysis (Crit Care Med 2007;35:1264-70). The reader is
referred elsewhere for additional information on this methodology (Crit Care Med 2004;32:1120-4; J
Trauma 2009;66:1045-51; J Crit Care.1995;10:51-5).

III. NUTRITION SUPPORT

A. Nutritional Assessment
1. Classes of malnutrition
a. “New” American Society for Parenteral and Enteral Nutrition (ASPEN) international consensus
nomenclature (JPEN J Parenter Enteral Nutr 2010;34:156-9)
i. Starvation-related malnutrition (e.g., anorexia nervosa)
ii. Chronic disease–related malnutrition (e.g., Crohn disease, organ failure)
iii. Acute disease or injury-related malnutrition (e.g., major infection, burns, trauma)
b. “Classic” definition
i. Marasmus (e.g., decreased fat/muscle protein stores but normal serum proteins)
ii. Kwashiorkor (e.g., normal fat, decreased muscle protein, decreased serum proteins)
iii. Kwashiorkor-Marasmus mix (decreased fat, muscle protein, and serum proteins)
c. Based on weight loss – A 10% unintentional weight loss within a 6-month period is considered
significant.
d. Based on current weight
i. Mild malnutrition: 80%–89% ideal body weight (IBW)*
ii. Moderate malnutrition: 70%–79% IBW*
iii. Severe malnutrition: Less than 70% IBW*
iv. Obese: Greater than 130% IBW*
*
IBW: Female = 45.5 kg/5 ft plus 2.3 kg per inch above 5 ft
Male = 50 kg/5 ft plus 2.3 kg per inch above 5 ft
e. Based on body mass index (BMI) = weight (kg)/height2 (m2)
i. Less than 18.5: Underweight
ii. 18.5–24.9: Normal
iii. 25–29.9: Overweight
iv. 30–34.9: Class I obesity
v. 35–39.9: Class II obesity
vi. Greater than 40: Class III obesity
f. Empiric weight adjustment for amputations

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-74
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

Table 19. Body compartments contribution to body weight (J Am Diet Assoc 1995;95:215-8)
Body Part Amputation Approximate Contribution to Body Weight (%)
Foot 1.5
Calf, foot 5.9
Leg (from hip) 16
Hand 0.7
Hand and forearm 2.3
Arm 5

2. Serum proteins used in nutritional assessment

a. Albumin: Half-life 20 days
i. Depletion: Mild 2.8–3.5 g/dL, moderate 2.1–2.7 g/dL, severe less than 2.1 g/dL
ii. Limitations: Long half-life, large body pool; is a negative acute-phase reactant protein that
decreases in response to infection, inflammation, surgery, injury, or other acute event
b. Transferrin: Half-life 7 days
i. Depletion: Mild 150–250 mg/dL, moderate 100–150 mg/dL, severe less than 100 mg/dL
ii. Limitations: Increased with iron deficiency; is a negative acute-phase reactant protein that
decreases in response to infection inflammation, surgery, injury or other acute event
c. Prealbumin: Half-life 2 days
i. Depletion: Mild 10–15 mg/dL, moderate 7–10 mg/dL, severe less than 7 mg/dL
ii. Limitations: Increased with CKD; is a negative acute-phase protein that decreases in response
to infection, inflammation, surgery, injury, or other acute event
d. Physical examination: Loss of subcutaneous body fat, muscle atrophy (including temporal
wasting), presence of lower extremity edema and/or ascites
e. Subjective Global Assessment (JPEN J Parenter Enteral Nutr 1987;11:8-13): Incorporates overall
evaluation by incorporating five elements of the patient’s history (presence of weight loss, dietary
intake change, presence of significant adverse GI symptoms persistent for more than 2 weeks,
physical functional capacity, and metabolic demands of the patient’s disease state) and physical
examination

B. Energy Requirements
1. Assessing caloric requirements: Indirect calorimetry – Measured energy expenditure by oxygen
consumption and CO2 production – The “gold standard”
a. Respiratory quotient (Vco2/Vo2); 1 for carbohydrate oxidation; 0.7 for fat oxidation; 0.8 for
protein oxidation; greater than 1 usually implies overfeeding (net fat synthesis), less than
0.7 suggests ketosis or an error in measurement (too much fraction of inspired oxygen [Fio2]
variability at higher Fio2 concentrations)
b. Organization guideline recommendations

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-75
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

Table 20. Guideline recommendations for caloric intake

Organization Recommendation
American College of Chest
25 kcal/kg/day; increase by 10%–20% with SIRS
Physicians (Chest 1997;111:769-78)
SCCM/ASPEN (JPEN J Parenter
25–30 kcal/kg/day or use of indirect calorimetry
Enteral Nutr 2009;33:277-316)
ESPEN PN (Clin Nutr
25 kcal/kg/day (in the absence of measured REE)
2009;28:387-400)
ESPEN EN (Clin Nutr 20–25 kcal/kg/day (during acute phase)
2006;25:210-23) 25–30 kcal/kg/day (during recovery)
ASPEN Obesity (JPEN J Parenter < 14 kcal/kg actual weight/day or 50%–70% of estimated requirements
Enteral Nutr 2013;37:714-44) when given with a high protein intake
25–30 kcal/kg/day or 1.2–1.4 x BEE (Harris-Benedict equations) or
Eastern Association for Surgery of 1.2–1.4 x BEE (Harris-Benedict equations), 30 kcal/kg/day (1.4 x BEE)
Trauma (J Trauma 2004;57:660-9) for patients with TBI, 22–25 kcal/kg/day for paraplegics, 20–22 kcal/kg
for quadriplegics
BEE = basal energy expenditure; ESPEN = European Society for Clinical Nutrition and Metabolism; REE = resting energy expenditure; SCCM =
Society of Critical Care Medicine.

c. Predictive methods
i. Mifflin-St. Jeor equations (preferred for non–ventilator-dependent obese patients with AKI or
CKD or hepatic encephalopathy when a hypocaloric, high-protein regimen is not possible)
(a) Females = (10 x Wt) + (6.25 x Ht) – (5 x A) − 161*
(b) Males = (10 x Wt) + (6.25 x Ht) − (5 x A) + 5*
*
Age (years); Ht (centimeters); Wt = actual body weight (kilograms).
ii. Penn State equation (preferred for ventilator-dependent obese patients with AKI or CKD or
hepatic encephalopathy when a hypocaloric, high-protein regimen is not possible):
REE = (Mifflin x 0.96) + (Tmax x 167) + (Ve x 31) − 6212*
*
REE = resting energy expenditure; Tmax = maximum temperature in degrees Celsius;
Ve = minute ventilation, L/min.
iii. Modified Penn State equation (preferred for ventilator-dependent obese patients 60 years or
older with AKI or CKD or hepatic encephalopathy when a hypocaloric, high-protein regimen
is not possible)
REE = (Mifflin x 0.71) + (Tmax x 85) + (Ve x 64) − 3085*
*
Tmax = maximum temperature in degrees Celsius; Ve = minute ventilation, L/min.
iv. Basal energy expenditure (BEE) – Harris-Benedict equations (preferred for small adults and
elderly patients)
(a) Females = 655 + (9.6 x Wt) + (1.7 x Ht) − (4.7 x A)*
(b) Males = 66 + (13.7 x Wt) + (5 x Ht) − (6.8 x A)*
*
Wt (kilograms), Ht (centimeters), Age (years).
d. Adverse effects of overfeeding – Do not exceed 5 mg/kg/minute of glucose/carbohydrate (Ann
Surg 1979;190:274-85); limit total caloric intake (do not exceed 1.3–1.5 x measured resting
energy expenditure); do not exceed intravenous fat intake of 2.5 g/kg/day (most clinicians limit
intravenous fat to around 1.5 g/kg/day or less).

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-76
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

i. Hypercapnia: It was traditionally thought that excessive glucose intake alone was responsible
for hypercapnia observed during overfeeding. However, studies of acutely ill patients
showed that aggressive feeding resulted in marked increases in CO2 production (Ann Surg
1980;191:40-6; JAMA 1980;243:1444-7). Substitution of glucose kilocalories with lipid
decreases CO2 production (Anesthesiology 1981;54:373-7) when overfeeding but does not
alter CO2 production if not overfeeding (e.g., 1.3 x BEE) (Chest 1992;102:551-5). Because
most institutions lack the ability to measure energy expenditure, estimates are used. If the
patient experiences hypercapnia without a known cause, the nutrition therapy should be
suspected and the caloric intake empirically decreased (especially if the patient is having
trouble weaning from the ventilator).
ii. Hyperglycemia: In a retrospective study of 102 PN-fed patients not predisposed to
hyperglycemia, dextrose intakes in excess of 5 mg/kg/minute resulted in substantial
hyperglycemia (blood glucose [BG] greater than 200 mg/dL) in 18 of 37 patients (Nutr Clin
Pract 1996;11:151-6). Patients with stress-induced hyperglycemia or diabetes are even more
susceptible to hyperglycemia with EN or PN.
iii. Fatty infiltration of the liver: May be owing to overfeeding with fat or carbohydrate.
Usually presents as a cholestatic liver disease (increased γ-glutamyltransferase, alkaline
phosphatase, and ultimately bilirubin) after at least 1 week to 10 days of overfeeding (Arch
Surg 1978;113:504-8). May be transient or reversible or can progress to end-stage liver
disease. Over a few weeks, patients can appear jaundiced. Patients with critical illness and/
or infections tend to be more susceptible to hepatic steatosis compared with non–critically ill
patients (possibly because of an exaggerated inflammatory process). Although treatment with
fish oil appears promising in infants and children (Ann Surg 2009;250:395-402; Nutr Clin
Pract 2013;28:30-9), data for adults are lacking. Usual management for adult patients with
suspected PN-associated liver disease is first to ensure the patient is not being overfed and
given a mixed-fuel PN solution, followed by cyclic PN (PN is infused over part of the day).
Reinstitution of EN as soon as possible (if possible) is of utmost importance.
e. Recommendations for caloric requirements*
i. Maintenance OR elective surgery: 25 kcal/kg/day
ii. Malnourished, nutritionally depleted: 1.4–1.5 x BEE
iii. Medical ICU patients: 25–30 kcal/kg/day
iv. Minor infection or surgery: 25–30 kcal/kg/day
v. Major surgery/trauma/sepsis: 30–32 kcal/kg/day
vi. Obese (hypocaloric) nutrition: 22–25 kcal/kg IBW/day or less
vii. Older (older than 65 years): 1.3–1.5 x BEE
viii. Smaller patients (weight 50 kg or less): 1.3–1.5 x BEE
*
These are general recommendations, based on the lack of measured energy expenditure, from
this author’s practice, which are subject to exception depending on prevailing disease states,
organ failures, extent of malnourishment, provision of paralytic drugs/pentobarbital/propofol,
and types of injuries.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-77
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

C. Protein Requirements
1. Guideline recommendations

Table 21. Guideline recommendations for protein intake

Organization Recommendation
American College of Chest 1.2–1.5 g/kg/day; 1.5–2 g/kg/day not to exceed 2 g/kg/day with SIRS;
Physicians (Chest 1997;111:769-78) routine NB determinations recommended
SCCM/ASPEN (JPEN J Parenter 1.2–2 g/kg/day; higher amounts are likely needed for multiple trauma
Enteral Nutr 2009;33:277-316) or burns
ESPEN PN (Clin Nutr
1.3–1.5 g/kg IBW/day
2009;28:387-400)
ESPEN EN (Clin Nutr
Not given
2006;25:210-23)
ASPEN Obesity (JPEN J Parenter 1.2 g/kg actual weight/day or 2–2.5 g/kg IBW/day when given with a
Enteral Nutr 2013;37:714-44) hypocaloric regimen
Eastern Association for Surgery of
1.25–2 g/kg/day; 2 g/kg/day for burns
Trauma (J Trauma 2004;57:660-9)

2. Recommendations for protein requirements*

a. Maintenance or elective surgery: 0.8–1 g/kg/day
b. Minor infection or surgery: 1.2–1.5 g/kg/day
c. Malnourished, nutritionally depleted: 1.5 g/kg/day
d. Medical ICU patients: 1.5–2 g/kg/day
e. Major surgery/trauma/sepsis: 2–2.5 g/kg/day
f. Renal failure
i. AKI: 0.6–1 g/kg/day
ii. CKD: 0.6–1 g/kg/day
iii. Hemodialysis: 1–1.5 g/kg/day
iv. CRRT: 2–2.5 g/kg/day
g. Obese, hypocaloric
i. BMI less than 40: 2 g/kg IBW/day
ii. BMI of 40 or greater: 2.5 g/kg IBW/day
*
Ideally, this intake should be adjusted according to the results of an NB determination, when
possible.

Patient Case
10. A 40-kg woman admitted to the trauma ICU receives a PN solution containing 350 g of dextrose, 160 g of amino
acids, and 80 g of lipid daily. She has normal renal and hepatic function. Her most recent ABG from the morning
shows a pH of 7.30, Pco2 of 55 mm Hg, Po2 of 96 mm Hg, and bicarbonate of 31. Her fingerstick BG values from
the past 24 hours range from 150 to 180 mg/dL. Which would be best to recommend concerning her PN?
A. Decrease dextrose to 175 g/day, and increase lipid to 120 g/day.
B. Add 20 units of regular human insulin per day to the PN solution.
C. Decrease all the macronutrients by about one-half.
D. Increase the acetate content of the PN solution.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-78
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

3. Assessing protein requirements: NB

a. NB = nitrogen in (Nin) – nitrogen out (Nout)
b. To calculate Nin:
i. Add all daily protein intake sources together, including the protein in EN, liquid protein
supplements, oral dietary supplements, and/or PN solution.
ii. Convert protein intake (grams per day) to nitrogen intake (grams per day):
Nin (g/day) = protein (g/day)/6.25 (assumes good-quality protein as 16% nitrogen content)
iii. To calculate Nout: Nout (g/day) = (UUN + 4 g)* + BUN change correction (if necessary), where
UUN is 24-hour urine urea nitrogen excretion (grams per day).
iv. *For patients with high UUN excretion (e.g., greater than 15 g/day), UUN/0.85 + 2 may be a
more reliable estimate of total urinary nitrogen excretion loss and stool/integumentary losses
(Nutrition 2005;21:332-8).

BUN Change Estimation

BUN adjustment (g/day) = [change in BUN (mg/dL) x 0.01] x [body water (L/kg) x Wt (kg)]
Use only if BUN change is 5 mg/dL or greater.
Body water = 0.6 L/kg for males; 0.55 L/kg for females
v. Calculate the NB: NB (g/day) = Nin g/day – Nout g/day.
vi. Measured versus predicted creatinine clearance (for gross assessment of adequacy in the
24-hour urine collection)
vii. Serum prealbumin changes are unreliable because of the influence of stress/inflammation;
some clinicians will obtain a serum C-reactive protein concentration together with a serum
prealbumin concentration for assessment.
4. Does more protein really make a difference?
a. Weijs (2012) (JPEN J Parenter Enteral Nutr 2012;36:60-8): 28-day mortality improved in those
who received an average of 1.3 g/kg/day versus 1.1 or 0.8 g/kg/day (886 mixed ICU patients).
b. Allingstrup (2012) (Clin Nutr 2012;31:462-8): 28-day mortality improved in those who received
an average of 1.5 g/kg/day versus 1.1 or 0.8 g/kg/day (113 mixed ICU patients).

Patient Case

Questions 11–13 pertain to the following case.

A 45-year-old man with obesity (BMI 35 kg/m2, IBW 75 kg; current body weight 114 kg), acute pancreatitis, and
sepsis was given a PN solution containing 200 g of dextrose, 150 g of amino acids, and 50 g of 20% lipids daily.

11. Which best depicts the kilocalories and protein this regimen will provide (normalized to IBW)?
A. 24 kcal/kg IBW/day and 2 g/kg IBW/day.
B. 27 kcal/kg IBW/day and 2 g/kg IBW/day.
C. 16 kcal/kg IBW/day and 1.3 g/kg IBW/day.
D. 20 kcal/kg IBW/day and 1.3 g/kg IBW/day.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-79
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

Patient Case (continued)

A 24-hour urine collection was done to determine the NB. The UUN concentration was 900 mg/dL, and the urine
volume output was 3000 mL. The patient’s BUN was essentially unchanged during the NB determination.

12. Using the “classic nitrogen balance equation” (NB = Nin – UUN – 4), which most accurately depicts the
patient’s NB?
A. +4 g/day.
B. -4 g/day.
C. -7 g/day.
D. -10 g/day.

13. Which changes would be best to make to the patient’s PN regimen?

A. Increase the protein and non-protein energy content.
B. Increase the protein content, and decrease the non-protein energy content.
C. Increase the protein content.
D. Increase the non-protein energy content.

D. Principles of EN and PN
1. Indications for EN: If the patient is unable to eat adequate amounts to achieve goal nutritional intakes.
EN is preferred to PN because EN has less infectious complications (Ann Surg 1992;215:503-13;
JPEN J Parenter Enteral Nutr 2009;33:277-316). This position is universally accepted among all
guideline sources.
a. Lack of bowel sounds, flatus, or bowel movement is not a contraindication for EN because these
are non-specific indicators of GI function (SCCM/ASPEN 2009; (JPEN J Parenter Enteral Nutr
2009;33:277-316).
b. Evidence of ileus (e.g., dilated loops of bowel on abdominal radiography) is, however, a
contraindication for EN.
c. High NG output (greater than around 800 mL NG output) in a 24-hour period might indicate
delayed gastric emptying, and the patient might not be ready for EN when fed into the stomach
and post-pyloric feeding is not possible. If the patient’s GI function appears to be improving,
clamp NG tube for 4 hours and check GRV (if GRV is less than around 250 mL and the abdomen
is not distended, patient is probably ready for gastric feeding). Note: This is our Nutrition Support
Service’s empiric practice and cannot be found in any guideline recommendations. See section
G2, “GRVs and prokinetic agents,” that follows for an expanded discussion on when/how to use
prokinetic pharmacotherapy for patients with gastric feeding intolerance.
d. Refusal to eat/anorexia is not an absolute contraindication for EN. Ensure appropriate dietary
preferences, and add high-calorie/protein liquid supplements to meals and bedtime snack first.
2. EN formulas

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-80
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

Table 22. EN Formulas

Additional
Product Category Indication Macronutrients Comments Examples
Standard tube 1–1.2 kcal/mL Polymeric, 300–500 Jevity, Nutren 1.0,
Minimal stress
feeding Protein 40–50 g/L mOsm/kg, fiber Fibersource HN
Congestive heart
2 kcal/mL Polymeric, 700–800 Nutren 2.0, Resource
Volume restricted failure, fluid-
Protein 60–80 g/L mOsm/kg, fiber 2.0, TwoCal HN
restricted patients
AKI (predialysis), High calorie, low
renal dysfunction protein, no or low
2 kcal/mL
Renal with increased electrolytes, 600 Renalcal, Suplena
Protein 30–40 g/L
serum potassium, mOsm/kg, volume
Phos, Mg restricted, fiber
High calorie, modest
Renal failure with 2 kcal/mL electrolytes, volume- Novasource Renal,
Renal
hemodialysis Protein 80–90 g/L restricted, 1000 Nepro
mOsm/kg, fiber
Critically ill patients
Increased protein 1 kcal/mL High-protein con-
(especially trauma, Replete, Promote
needs Protein 62 g/L tent, fiber, isotonic
surgical, burns)
Hyperglycemia, 1.2 kcal/mL Low-carbohydrate Diabetisource AC,
Glucose intolerance
diabetes mellitus Protein 60 g/L content, fiber Glucerna 1.2
Critically ill surgical
and trauma patients, 1.5 kcal/mL Additional arginine, Impact Peptide 1.5,
Immune enhancing
perioperative GI Protein 90–94 g/L glutamine; fish oil Pivot
cancer
Obese patients with 1 kcal/mL High protein, low
Bariatric Peptamen Bariatric
good renal function Protein 93 g/L calories
Low fat/MCT, di/tri-
Malabsorption, fat 1–1.5 kcal/mL
Elemental diet peptides/free AA, no Vivonex RTF Vital
intolerance Protein 50–68 g/L
fiber, low residue
Volume restricted,
1.2–1.5 kcal/mL
Pulmonary ARDS, ALI fish oil, low Oxepa Peptamen AF
Protein 63–76 g/L
omega-6 fats
Cirrhosis High branched
1.2–1.5 kcal/mL NutriHep Hepatic-
Hepatic with hepatic chain, low aromatic
Protein 40–44 g/L Aid II
encephalopathy AA
Protein supplement High protein needs Protein Liquid or powder
AA = amino acids; ALS = acute lung injury; ARDS = acute respiratory distress syndrome; MCT = medium-chain triglyceride.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-81
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

3. Tube feeding (TF) – Medication interactions: Hold TF 1 hour before and after drug administration
(Pharmacotherapy: A Pathophysiologic Approach, 8th ed. New York: McGraw-Hill, 2011:2493-503).
*
Increase TF rate to account for time off TF.
a. Warfarin (Pharmacotherapy 2008;28:308-13)
b. Phenytoin (Nutr Clin Pract 1996;11:28-31)
c. Levothyroxine (Nutr Clin Pract 2010;25:646-52; J Endocrinol Invest 2014;37:583-7)
d. Itraconazole (Antimicrob Agents Chemother 1997;41:2714-8)
e. Fluoroquinolones (J Antimicrob Chemother 1996;38:871-6)?
*
Some clinicians have empirically increased the dosage of these drugs while giving continuous
enteral feeding rather than holding the EN for 1 hour before and after drug administration.
This author discourages this practice, especially for warfarin and phenytoin, because the doses
necessary to overcome the effects of drug binding to the continuous EN are potentially toxic
when the EN is held or discontinued without a dose adjustment. We have had limited success
with preventing the development of subclinical or overt hypothyroidism by increasing the
levothyroxine dose by 25 mcg/day temporarily while the patient is receiving continuous EN
(Nutr Clin Pract 2010;25:646-52). Others have increased the ciprofloxacin dose to 750 mg twice
daily during continuous EN to achieve therapeutic plasma concentrations well above the MIC
(minimum inhibitory concentration) for a gram-negative urinary tract infection (J Antimicrob
Chemother 1996;38:871-6).
4. Indications for PN
a. European Society for Clinical Nutrition and Metabolism (ESPEN) PN guidelines (2009) (Clin
Nutr 2009;28:387-400): Patients who are not expected to receive EN within 3 days should receive
PN within 24–48 hours if EN is contraindicated or if they cannot tolerate EN.
b. SCCM/ASPEN (2009) (JPEN J Parenter Enteral Nutr 2009;33:277-316): PN indicated only after
the first 7 days of hospitalization when EN is not feasible or available.
c. Dickerson’s interpretation: The approach depends on several factors (e.g., if the patient is
malnourished or wellnourished before ICU admission, patient population). Early nutrition (defined
as within 24–72 hours according to published studies) appears to be beneficial for those with
prolonged ICU stays and a high level of catabolism, including trauma, TBI, thermal injury, and
some surgical subpopulations. Impact of early nutrition appears more variable with respect to
clinical outcome for medical ICU patients and is likely related to a shorter duration in ICU stay
and a lower level of catabolism for many patients.
5. PN formulations
a. Peripheral versus central venous administration
i. Osmolality of peripheral administration is limited to about 800 mOsm/kg.
ii. Because of the osmolality issue, peripheral PN solutions are “diluted,” requiring large
volumes (contraindicated for fluid-restricted patients and difficult for older patients).
iii. Estimating the osmolality of PN solutions:
Approximate osmolality = [glucose g/L x 5] + [AA% x 100] + [lipid % x 15] + 200*
*
Accounts for electrolytes/vitamins and can be variable, depending on amounts provided.
AA = amino acids.
iv. Phlebitis common with peripheral PN and difficult to use beyond 2–3 days
b. Safe practice guidelines for prescribing PN solutions – Should be prescribed in total amount per
day (e.g., glucose 200 g/day, amino acids 150 g/day, lipid 30 g/day, fluid volume 2500 mL/day,
sodium chloride 60 mEq/day, potassium acetate 80 mEq/day), NOT by concentrations (e.g., 20%
dextrose in water, 8% amino acids) or by compounding techniques (e.g., 500 mL of 50% dextrose
in water plus 500 mL of 10% amino acids).

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-82
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

c. Glucose requirements
i. Obligatory requirements for central nervous system, renal medulla, bone marrow, leukocytes,
etc.: Around 130 g/day
ii. Surgical wound about 80–150 g/day (based on atrioventricular differences and blood flow
from a burned limb)
iii. Caloric contribution of glucose: 3.4 kcal/g (as opposed to carbohydrate 4 kcal/g)
iv. Mean glucose oxidation rate in critically ill patients is around 5 mg/kg/day (or about 25 kcal/
kg/day as glucose). In general, most clinicians avoid exceeding this glucose intake.
d. Lipid requirements
i. Main source is soybean oil – May be given separately from the PN admixture or as part
of the PN solution. When given separately from the dextrose/amino acid formulation, the
maximum allowable hang time according to the U.S. Food and Drug Administration (FDA)
is 12 hours. With recent lipid shortages, olive oil–based intravenous lipid (which provides a
more favorable cardiovascular lipid profile in patients on long-term PN) is now available from
Canada. Intravenous fish oil requires permission from the FDA.
ii. Caloric contribution of intravenous fat emulsion: 10% = 1.1 kcal/mL; 20% = 2 kcal/mL;
30% = 3 kcal/mL or around 11 kcal/g for 10% emulsion, 10 kcal/g for 20% and 30% emulsion
iii. Dosage: About 100–150 g weekly (or 1–1.5 g/kg weekly) is enough to prevent essential fatty
acid deficiency (EFAD). The FDA states a maximum upper limit of 2.5 g/kg/day in adults,
though most clinicians try to keep the daily dose to 1.5 g/kg/day or less (see discussion on
lipid emulsion and immune function in Controversial Topics section).
iv. Biochemical evidence for EFAD (the “classic definition” is an increased triene/tetraene
[icosatrienoic acid/arachidonic acid] ratio greater than 0.4) occurs in 30%, 66%, 83%, and
100% of patients after 1, 2, 3, and 4 weeks of fat-free “full-calorie, continuous” PN (Surgery
1978;84:271-7). Clinical signs and symptoms of EFAD usually do not occur until about 2
weeks after biochemical evidence in adults. Since the investigators initiated intravenous lipid
emulsion soon after the biochemical appearance of EFAD, only 2 of 32 patients developed
clinical evidence suggestive of EFAD. EFAD can occur much sooner for infants and children.
Obese patients receiving hypocaloric high-protein therapy can maintain normal plasma fatty
acid profiles for up to 5 weeks (J Nutr Biochem 1994;5:243-7). Cyclic PN has been suggested
to mobilize lipid from endogenous depots, but conclusive data are lacking.
v. Clinical symptoms (dry, scaly skin; hair loss; poor wound healing) occur about 2 weeks after
biochemical evidence of deficiency in adults. Therefore, in most adults, the earliest you will
see EFAD is after about 3 weeks of fat-free full-calorie continuous PN.
vi. Serum triglyceride concentration should be monitored at least weekly and more often for
those with proven or suspected impaired triglyceride clearance.
vii. Predisposing conditions that may result in impaired clearance of triglycerides:
(a) Excessive lipid intake (often caused by propofol therapy)
(b) Acute pancreatitis
(c) Uncontrolled diabetes
(d) Liver failure
(e) Kidney failure (decreased lipoprotein lipase activity, carnitine deficiency with long-term
hemodialysis patients)
(f) End-stage sepsis (multisystem organ failure)
(g) History of hyperlipidemia
(h) Obesity
(i) HIV (human immunodeficiency virus) (occurred even before current antiretroviral
therapy) (Am J Med 1989;86:27-31)

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-83
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

(j) Pregnancy
(k) Small-for-gestational-age neonates (carnitine synthesis is maturational-dependent)
viii. Propofol – A hidden source of lipids (10% soybean emulsion containing 1.1 kcal/mL)
e. Electrolyte requirements (see Fluid and Electrolytes section)
f. Vitamins (multivitamin infusion or multivitamin complex 10 mL/day; extra vitamins if patient has
any vitamin deficiencies)
g. Trace minerals (MTE-5 cocktail for normal requirements)
i. Zinc 3 mg/day normal requirements; 5 mg/day during critical illness; increased requirements
for patients with diarrhea, intestinal fistulae. An additional 10 mg/day for a total of 13 mg/day
is usually sufficient to meet increased intestinal losses (Gastroenterology 1979;76:458-67).
Deficiency is characterized by loss of hair; erythematous rash, especially in periorbital regions
of face; poor wound healing. Classic zinc deficiency is termed acrodermatitis enteropathica.
ii. Copper 0.3–0.5 mg/day is usually sufficient (Gastroenterology 1981;81:290-7). Copper
deficiency is rare but is becoming more apparent in obese patients after gastric bypass
procedures. Classic presentation of copper deficiency includes a “microcytic anemia
unresponsive to iron therapy” or pancytopenia.
iii. Chromium 10–12 mcg/day normal requirements, up to 20 mcg/day for diarrhea. Rare. Classic
presentation for deficiency is hyperglycemia.
iv. Manganese 150–300 mcg/day is probably enough. Some studies state that the amount
of manganese contamination in the compounding of PN may be adequate as opposed to
supplementation. Deficiency is very rare. Deficiency has been reported to present as a “diaper
rash.” Several case reports of manganese toxicity associated with liver disease and high
manganese intake (800 mcg – 1 mg/day) (Nutrition 2001;17:689-93). Signs and symptoms of
toxicity emulate those of Parkinson disease.
v. Selenium 60 mcg/day up to 120 mcg/day for patients with diarrhea or short bowel syndrome.
Deficiency results in extreme muscle weakness and congestive cardiomyopathy. Classic
presentation with cardiomyopathy has been termed Keshan disease (named after a province in
China where the first cases of selenium deficiency with cardiomyopathy were discovered).
vi. It is common clinical practice to withhold copper and manganese in the PN formulation
for patients with hepatobiliary/cholestatic liver disease or a direct (conjugated) bilirubin
concentration greater than 2 mg/dL.
vii. Some clinicians will withhold selenium for patients with significant renal disease who do not
receive hemodialysis or CRRT, though the data in support of this practice are lacking.
6. Should supplemental PN be given to patients intolerant of EN?
a. ESPEN PN (2009) (Clin Nutr 2009;28:387-400): All patients receiving less than their target EN
after 2 days should be considered for supplemental PN.
b. ESPEN EN (2006) (Clin Nutr 2006;25:210-23): For patients intolerant of EN, supplemental PN
should be considered. Overfeeding should be avoided.
c. SCCM/ASPEN (2009) (JPEN J Parenter Enteral Nutr 2009;33:277-316): If patient is unable to
meet energy requirements after 7–10 days of EN alone, consider initiating supplemental PN.
d. Canadian Practice Guidelines Update (2014) (Nutr Clin Pract 2014;29:29-43): It is strongly
recommended that early supplemental PN or large volumes of hypertonic dextrose solutions not
be used in unselected critically ill patients (i.e., low-risk patients with short stay in ICU). In the
patient who is not tolerating adequate EN, data are insufficient to put forward a recommendation
about when PN should be initiated.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-84
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

e. Caesar/EPaNIC study (N Engl J Med 2011;365:1-17) – 4640 mostly surgical patients, randomized
controlled trial (RCT): EN only x 7 days; then PN initiated (hypertonic dextrose solutions x 2
days; then PN) versus supplemental PN in addition to whatever EN patients can tolerate during the
first 7 days
i. Worsened survival (72% vs. 75%), greater infections (26% vs. 23%), ICU length of stay
(LOS) greater than 3 days (51% vs. 48%) with early supplemental PN
ii. The patient population was limited in that malnourished (BMI less than 17 kg/m2) patients
were excluded. Additionally, about 60% of the population were cardiac surgery patients
which the indication for PN is questionable, 50% of patients were extubated by ICU day 2,
and 70% of patients had an ICU LOS of only 3–4 days (which would imply a questionable
severity of critical illness). Finally, only 58% of patients in the early PN group were even
given PN (for 1 to 2 days) and only 25% patients in the late PN group ever received PN.
f. Heidegger (2013) (Lancet 2013;381:385-93): 307 patients, two medical centers, RCT: Patients
who received less than 60% target from EN by day 3 with an anticipated ICU stay greater than
5 days received supplemental PN or EN alone. Supplemental PN was discontinued by day 8.
i. Supplemental PN group had decreased infections (27% vs. 38%).
ii. Smaller study compared with Caesar study. Not all patients had resting energy expenditure
measured (some were predicted resting energy expenditure). Protein target was only 1.2 g/kg/
day. No difference in ICU/hospital LOS, mortality
g. Dickerson’s interpretation: Many of the patients in the Caesar study did not require PN to
start with. PN should not be given indiscriminately to patients because of increased infectious
complications in those who were not malnourished versus improved morbidity in perioperative
malnourished patients with GI cancer (VA Cooperative Trial, 1991). Heidegger’s data showed a
benefit from supplemental PN in a sicker patient population than that in the Caesar study. Short-
term supplemental PN may be indicated for patients anticipated to have prolonged duration of
inability to use GI tract, anticipated prolonged duration of ICU stay, malnourished, and high level
of catabolism, but the routine use of supplemental PN, for all patients, is discouraged. Effective
use of prokinetic pharmacotherapy may also reduce the need for supplemental PN.

E. Timing of Initiation of Nutrition Support: Early or not for ICU patients?

1. ESPEN EN (2006) (Clin Nutr 2006;25:210-23): Questionable benefit of early versus delayed EN;
however, it was recommended that critically ill patients, who are hemodynamically stable and have a
functioning GI tract, be fed early (less than 24 hours), if possible
2. ESPEN PN (2009) (Clin Nutr 2009;28:387-400): All patients not expected to be on EN within 3 days
should receive PN within 24–48 hours if EN is contraindicated or not tolerated.
3. SCCM/ASPEN (2009) (JPEN J Parenter Enteral Nutr 2009;33:277-316): Early EN should be initiated
within 24–48 hours after admission.
4. Surviving Sepsis Campaign guidelines (2013) (Crit Care Med 2013;41:580-637): Administer oral
nutrition or EN within the first 48 hours after severe sepsis is diagnosed.
5. Dickerson’s interpretation: The data are confusing because several studies and a meta-analysis in
the literature have used different times for the definition of early nutrition therapy: 24, 36, 48, and
72 hours. Most evidence-based clinicians would suggest that nutrition therapy be initiated for most
patients within 48 hours of ICU admission and definitely no later than 72 hours. Surgical ICU patients,
including those with trauma and thermal injury, have been more consistently shown to benefit from
early EN as opposed to more variable results for medical ICU patients.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-85
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

F. Glycemic Control
1. Definition of the appropriate BG target range
a. Society of Critical Care Medicine (SCCM) guidelines (2012) (Crit Care Med 2012;40:3251-76):
A BG of 150 mg/dL or greater should trigger initiation of insulin therapy to keep BG less than 150
mg/dL for most patients and maintain BG absolutely less than 180 mg/dL.
b. ASPEN guidelines (2013) (JPEN J Parenter Enteral Nutr 2013;37:23-36): A target BG range of
140–180 mg/dL is recommended.
c. Surviving Sepsis Campaign guidelines (2013) (Crit Care Med 2013;41:580-637): An insulin
dosing protocol to keep BG less than 180 mg/dL, rather than an upper target of 110 mg/dL, is
recommended when the patient has two consecutive BG measurements greater than 180 mg/dL.
d. American Diabetes Association (2014) (Diabetes Care 2014;37(suppl 1):S14-80): An insulin
infusion should be used to control hyperglycemia, starting with a threshold no higher than 180 mg/
dL. BG should be maintained between 140 and 180 mg/dL. Although strong evidence is lacking,
lower glucose targets may be appropriate in select patients. BG targets less than 110 mg/dL are not
recommended.
e. Dickerson’s interpretation: Many evidence-based clinicians use a target BG range of 140–180
mg/dL when caring for patients in a mixed medical-surgical ICU. A growing amount of evidence
from smaller studies shows that certain subpopulations such as trauma, traumatic brain injury,
cardiothoracic surgery, and thermal injury may benefit from tighter BG (e.g., less than 140–150
mg/dL) control if it can be done safely without hypoglycemia.
2. BG monitoring frequency
a. SCCM guidelines (2012) (Crit Care Med 2012;40:3251-76): BG should be monitored every
1–2 hours for most patients receiving an insulin infusion; monitoring every 4 hours is not
recommended because of the risk of unrecognized hypoglycemia.
b. Surviving Sepsis Campaign guidelines (2013) (Crit Care Med 2013;41:580-637): BG should
be monitored every 1–2 hours during the insulin infusion and then extended to every 4 hours
thereafter once stability in BG control achieved
c. American Diabetes Association (2014) (Diabetes Care 2014;37(suppl 1):S14-80): BG should be
monitored every ½–2 hours during the insulin infusion.
d. Dickerson’s interpretation: We monitor BG every hour and then extend it to every 2 hours once
there is stability in BG control and the intravenous insulin infusion rate is demonstrated. Although
some of the major trials show that BG monitoring could be extended to every 4 hours once two
consecutive BG concentrations in the target range are achieved, we believe our patients are too
labile to safely extend monitoring to every 4 hours. If they are stable enough to warrant BG
monitoring every 4 hours, we will then begin to transition the EN-fed patient to intermediate- or
long-acting subcutaneous insulin therapy (JPEN J Parenter Enteral Nutr 2013;37:506-16).
3. Point-of-care BG monitoring
a. SCCM guidelines (2012) (Crit Care Med 2012;40:3251-76): Point-of-care glucose meters are
acceptable but not optimal for routine BG testing during an insulin infusion. Arterial or venous
blood sampling, instead of fingerstick capillary BG testing, is suggested for patients in shock, on
vasopressor therapy, or with severe peripheral edema.
b. At BG concentrations greater than 80 mg/dL, there was strong agreement (73% and 85%) in
glycemic interventions when using either capillary or arterial BG measurements in critically ill
patients (Intensive Care Med 2004;30:804-10).

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-86
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

c. High doses of drugs such as acetaminophen, ascorbic acid, dopamine, or mannitol may
interfere with the accuracy of BG measurements, and false elevation of results by point-of-care
glucose meters that use the glucose-oxidase method may occur (Jacobi et al., 2012). Glucose-
dehydrogenase–based assays are sensitive to interference and false elevation of results if the
patient receives medications containing maltose (immunoglobulins) or icodextrin (used in some
dialysis solutions).
d. Dickerson’s interpretation: Point-of-care meters with capillary BG testing are probably acceptable
for most critically ill patients as long as the patient’s BG concentrations are not often less than
80 mg/dL. For low BG concentrations, point-of-care meters may erroneously report higher BG
concentrations than measured (Crit Care Med 2009;37:2691-6). Patients with significant peripheral
edema may also have erroneous BG concentrations.
4. Hypoglycemia
a. Most guidelines define hypoglycemia as a BG less than 70 mg/dL because increased glucagon,
catecholamine, and growth hormone production occurs when the BG falls below this
concentration. Mild to moderate hypoglycemia is usually defined as a BG concentration of 40–60
mg/dL (because autonomic symptoms usually appear) and severe (life threatening) hypoglycemia
as less than 40 mg/dL.
b. Most common risk factors for hypoglycemia during insulin (Crit Care Med 2007;35:2262-7; Crit
Care Med 2006;34:96-101)
i. Excessive insulin dose
ii. Abrupt discontinuation of EN or PN without an adjustment in the insulin therapy (purported
to cause 62% of severe hypoglycemic events in the 2006 van den Berghe trial of medical ICU
patients) (Crit Care Med 2012;40:3251-76)
iii. Renal failure (half-life of insulin is prolonged, ? impaired renal gluconeogenesis in response
to hypoglycemia)
iv. Advanced age
v. Inotropes, vasopressor agents, octreotide with insulin therapy
vi. Sepsis
5. Transitioning from a continuous intravenous insulin infusion
a. Lack of a transition plan has been shown to result in loss of glycemic control. Different methods
have been described in the literature, and the best approach depends on whether the patient is
transitioning to an oral diet, bolus EN, or continuous EN (Crit Care Med 2012;40:3251-76;
Diabetes Care 2014;37(suppl 1):S14-80; JPEN J Parenter Enteral Nutr 2013;37:506-16; The
ASPEN Adult Nutrition Support Core Curriculum, 2nd ed. Silver Spring, MD: American Society
for Parenteral and Enteral Nutrition, 2012:580-602):
b. Example for transitioning from a continuous intravenous regular human insulin infusion to a
subcutaneous intermediate acting insulin (neutral protamine Hagedorn [NPH]) for patients who
receive continuous enteral feeding (JPEN J Parenter Enteral Nutr 2013;37:506-16)*
i. Give 30%–50% of the daily required intravenous regular human insulin, divided into two
separate doses and given every 12 hours. Continue the graduated intravenous insulin infusion
according to the algorithm.
ii. BG measurements are continued every 1–2 hours and are mapped to the timing of the
subcutaneous NPH therapy to ascertain its pharmacokinetic peak and duration.
iii. The NPH dosage and interval are adjusted accordingly daily until the intravenous regular
human insulin infusion is “auto-weaned” to about 1–2 units/hour.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-87
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

iv. The regular human insulin infusion is then discontinued with BG determinations, with sliding-
scale regular human insulin coverage every 3–4 hours. The NPH is further titrated daily
according to the required amount of corrective regular human insulin required to maintain BG
within the target BG range.
v. Frequency of BG measurements is decreased because the patient has stability in glycemic
control.
*
This method was evaluated in 32 patients who transitioned from a continuous intravenous
regular human insulin to subcutaneous NPH and intermittent sliding-scale regular human
insulin coverage (JPEN J Parenter Enteral Nutr 2013;37:506-16). BG concentrations were
maintained in the target range for 18±3 hours/day. Eighteen patients (56%) experienced at
least a single episode of moderate hypoglycemia (40–59 mg/dL), and three patients (9%) had
an episode of severe hypoglycemia (BG less than 40 mg/dL). The overall rates of moderate
and severe hypoglycemia were 1.3% and 0.1% of all BG determinations. Patients older than
60 years were at a higher risk of hypoglycemia than were younger patients.

Patient Case
14. A 55-year-old woman (75 kg) without diabetes is given PN after a major GI resection. She has been weaned
from mechanical ventilation and is being transferred from the ICU to the floor. Her current PN formulation
is 200 g of dextrose (1.8 mg/kg/minute), 110 g of amino acids, and 80 g of lipids (1.1 g/kg/day), which meets
her goal requirements of protein at 26 kcal/kg/day and 1.5 g/kg/day. It contains regular human insulin at 20
units/day. During the past 24 hours, her fingerstick BG measurements have been between 170 and 210 mg/dL,
and her serum glucose concentration is 182 mg/dL. She has received 14 units of sliding-scale regular human
insulin coverage. Which would be best to suggest for optimal glycemic control?
A. Increase regular insulin to 30 units/day.
B. Decrease dextrose to 100 g/day.
C. Increase regular insulin to 50 units/day.
D. Do not change the current regimen.

G. Specialized and/or Controversial Topics in Nutrition Support Therapy for Critically Ill Patients
1. Trophic versus full EN—OR how much is enough to achieve a therapeutic benefit?
a. SCCM/ASPEN (2009) (JPEN J Parenter Enteral Nutr 2009;33:277-316): Efforts to provide greater
than 50%–65% of goal calories should be made to achieve the clinical benefit of EN during the
first week of hospitalization.
b. Surviving Sepsis Campaign guidelines (2013) (Crit Care Med 2013;41:580-637): Avoid
mandatory full-caloric feeding in the first week; instead, suggest low-dose feeding (e.g., up to 500
kcal/day), advancing only as tolerated
c. EDEN (2012) (JAMA 2012;307:795-803): 1000 medical ICU patients with acute lung injury
(ALI) randomized to receive “trophic” feeds at 10–20 mL/hour versus “full feeds” for the
first 6 days of feeding. After 6 days, the feedings were increased to the “full-feeding” rates.
Malnourished patients were excluded from study entry. The initial 272 patients were part of the
OMEGA study (JAMA 2011;306:1574-81) investigating the influence of fish oil supplementation
compared with protein-containing control supplement. The “full-feeding” group received 1300
kcal/day, whereas the “trophic” group received 400 kcal/day. Protein intakes for the groups
were not given. There were no differences in clinical outcomes between groups; however, the
“full-feeds” group experienced significantly greater GI complications and a higher daily BG
concentration.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-88
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

d. Two smaller studies (Am J Clin Nutr 2011;93:569-77; Crit Care Med 2011;39:967-74), also
implied a benefit from “trophic feeds.” For the Rice study (Crit Care Med 2011;39:967-74),
patients received an average of 1418 kcal/day versus 300 kcal/day. No difference was noted in
clinical outcomes, but gastric feeding tolerance was improved with the lower caloric intake.
However, patients in both groups averaged only 5.5 versus 5.1 days of required EN. Conversely,
Arabi (Am J Clin Nutr 2011;93:569-77) provided an average of 1067 kcal/day to the “trophic
feeding” group and 1252 kcal/day to the “target feeding” group, a difference of 185 kcal/day (the
caloric equivalent of about 1 L of 5% dextrose in water). Their data showed a reduction in 28-day
all-cause mortality with permissive underfeeding (23% vs. 18%); however, their data were also
confounded by different insulin therapy strategies (target BG 80–110 mg/dL vs. 180–200 mg/dL).
e. Dickerson’s interpretation: It is unclear what exact minimum amount of calories and protein
are necessary to gain a therapeutic benefit. Early “full” EN and early “trophic” feedings had
similar clinical outcomes in the EDEN study, but some clinicians would suggest that this was
a comparison of hypocaloric feeding (about 15 kcal/kg/day), not “full feeding” to “trophic” or
semi-starvation feeding (about 5 kcal/kg/day). Most clinical studies involving EN in critically ill
patients generally achieve intakes of only about two-thirds of what is prescribed for the first week
of therapy. Recent efforts for improving EN intake include prokinetic pharmacotherapy, raising of
GRV thresholds, and short-term increased EN infusion rates in response to feeding interruptions.
These efforts may improve overall caloric and protein intake for “full-feeds” versus “trophic-
feeds” comparisons and ultimately provide a clearer answer to this controversial area. Observed
differences in clinical outcomes with early EN are likely related to the patient population (e.g.,
medical ICU vs. surgical/trauma/burn patients), gastric feeding tolerance, duration of EN and ICU
stay, actual caloric and protein intakes, and efficacy of glycemic control. The reader is cautioned
not to apply the findings from one select patient population (e.g., well-nourished medical ICU
patients with acute lung injury) to all critically ill patients.
2. GRVs and prokinetic agents
a. North American Summit on Aspiration in the Critically Ill Patient (2002) (JPEN J Parenter Enteral
Nutr 2002;26(6 suppl):S80-85): GRV of 500 mL or greater indicates need to withhold gastric
feeding and reassess tolerance. GRV in the range of 200–500 mL should prompt careful bedside
evaluation and initiation of an algorithmic approach to reduce risk.
b. SCCM/ASPEN (2009) (JPEN J Parenter Enteral Nutr 2009;33:277-316): To reduce the risk of
aspiration: Elevate the head of bed 30–45 degrees. GRVs in the range of 200–500 mL should raise
concern and lead to the implementation of measures to reduce the risk of aspiration, but automatic
cessation of feeding should not occur for GRVs less than 500 mL in the absence of other signs of
intolerance.
c. REGANE study (Intensive Care Med 2010;36:1386-93): About 300 patients (about 80% medical
ICU) were randomized to hold tube feeds for a GRV of 200 mL versus 500 mL. ALL patients
received “prophylactic metoclopramide therapy 10 mg intravenously q8h” for the first 3 days of
EN. Increasing the GRV limit to 500 mL (from 200 mL) was associated with an increase in EN
volume received and was not associated with adverse clinical outcomes. There was no difference
in days on the ventilator, ICU LOS, mortality, pneumonia (no microbial confirmation was required
for the diagnosis of pneumonia), or GI complications between groups.
d. Reignier (2013) (JAMA 2013;309:249-56): About 450 patients (more than 90% medical ICU)
were randomized to GRV monitoring or no GRV monitoring during EN. Tube feeds were held
for GRV greater than 250 mL (in monitoring group) or for vomiting/regurgitation (both groups).
No difference was found in ventilator days, ICU or hospital LOS, ICU-acquired infections,
or mortality. There was a higher incidence of vomiting (42% vs. 27%) in those without GRV
monitoring. Others have suggested that decreased EN delivery may have occurred for the no-GRV

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-89
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

monitoring group (either intentionally by the over-reporting of emesis/regurgitation or

unintentionally by nursing personnel) because there was only a 200-kcal difference in intake
between the groups for the week (Nutrition 2013;29:1075-9).
e. Metheny (2006) (Crit Care Med 2006;34:1007-15): Observational study of 360 patients (around
75% trauma/surgery/neurosurgery/cardiac surgical ICU patients). More patients with pneumonia
(n=173) had pepsin-positive tracheal secretions than did those without pneumonia (n=187): 42%
vs. 21% (p<0.001). Those with vomiting or diagnosed gastroesophageal reflux disease had more
pepsin-positive tracheal aspirates (p=0.01; 0.033). The frequency of aspiration increased for a
GRV greater than 250 mL or two consecutive GRVs greater than 200 mL.
f. Nguyen (2007) (Crit Care Med 2007;35:2561-7; Crit Care Med 2007;35:483-9): Dual
metoclopramide/erythromycin therapy is more effective than erythromycin therapy, which is more
effective than metoclopramide therapy.
g. Dickerson (2009) (JPEN J Parenter Enteral Nutr 2009;33:646-55): Combination prokinetic therapy
is preferred to metoclopramide therapy alone (because of marked tachyphylaxis) for trauma
patients with brain injury; single-drug therapy with metoclopramide is suitable for trauma patients
without brain injury.
h. Dickerson’s interpretation: The “key” to interpreting gastric feeding tolerance is routine and
frequent abdominal examinations and assessment of the patient for evidence of vomiting, cramping,
nausea, regurgitation, or distension. GRV alone as a GI assessment is inadequate and prone to error.
Patients with a mild or modestly elevated GRV but a normal abdominal examination or lack of GI
symptoms can continue EN. If the patient has symptoms or is distended/tympanic, feedings should
be interrupted, and either prokinetic therapy or advancement of the feeding tube into the small bowel
should be considered. Also of note, a GRV can sometimes be difficult to accurately assess with a
small-bore feeding tube because of the tube collapsing on aspiration or because of its location in
the stomach. We empirically prefer this technique to waiting for the patient to regurgitate or vomit
because the risk of aspiration increases when these events occur (Crit Care Med 2006;34:1007-15):.
3. Immune-enhancing diets – What makes them unique? Omega-3 fatty acids, glutamine, arginine, nucleotides
a. Populations in which their use appears beneficial in reducing infectious complications
i. GI cancer surgery (pre- and perioperative)
ii. Trauma
iii. General surgical?
b. SCCM/ASPEN (2009) (JPEN J Parenter Enteral Nutr 2009;33:277-316): Immune-modulating
enteral formulations should be used for the appropriate patient population (patients with major
elective surgery, trauma, burns, and head and neck cancer and those who are critically ill on
mechanical ventilation). Surgical patients – grade A; medical ICU patients – grade B. Use with
caution in patients with severe sepsis.
c. ESPEN EN (2006) (Clin Nutr 2006;25:210-23): Indicated for patients requiring elective upper GI
surgery and trauma. Improves outcome in less severe sepsis. Not recommended for patients with
severe sepsis or burns
d. Surviving Sepsis Campaign guidelines (2013) (Crit Care Med 2013;41:580-637): Suggest using
nutrition without specific immune-modulating supplementation
e. Dickerson’s interpretation: It would make sense not to use an immune enhancing diet (IED) if
the patient is already infected because that is the primary goal for using IEDs. Data from patients
undergoing GI cancer surgery and trauma patients appear promising. ESPEN guidelines are
primarily based on the Galban study (Crit Care Med 2000;28:643-8), which used APACHE II
(Acute Physiology and Chronic Health Evaluation II) scores to show the severity of illness/sepsis.
Unfortunately, the number of patients in the higher APACHE II–stratified groups was insufficient
to show a statistically significant difference (e.g., beta error) in outcomes, despite trends.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-90
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

4. EN and vasopressors/hemodynamic instability

a. The concern: Blood shunting away from the GI tract may predispose patients to reported
complications of intestinal ischemia, perforation, and necrosis.
b. SCCM/ASPEN (2009) (JPEN J Parenter Enteral Nutr 2009;33:277-316): In the setting of
hemodynamic compromise (high-dose catecholamines, fluid resuscitation), EN should be withheld
until the patient is fully resuscitated and stable.
c. Mancl (2013) (JPEN J Parenter Enteral Nutr 2013;37:641-51): Retrospectively evaluated the
tolerance and safety of concurrent EN and intravenous vasopressor therapy in 259 patients.
An inverse relationship was found between maximum norepinephrine equivalent dose and EN
tolerability: 12.5 mcg/minute for patients who tolerated EN compared with 19.4 mcg/minute for
those who were intolerant (p<0.001). Serum lactate was elevated in 31% of patients, and 0.9%
experienced bowel ischemia or perforation. Tolerability was related to the cumulative vasopressor
dose and possibly related to the specific vasopressor administered. The norepinephrine-equivalent
dose was 12.5 mcg/minute for those who tolerated EN in contrast to 19.4 mcg/minute for those
who did not tolerate EN for the Mancl study.
d. Khalid (2010) (Am J Crit Care 2010;19:261-8): A retrospective analysis to evaluate early compared
with delayed EN in 1174 critically ill patients who required EN, mechanical ventilation, and
vasopressors. Most of the patients were admitted to the ICU for respiratory failure. Unfortunately,
few details were given regarding which vasopressor agents were used or what their doses were.
e. Dickerson’s interpretation: Most cases of intestinal ischemia and bowel perforation during EN
with hemodynamic instability stem from surgical or trauma patients in whom intra-abdominal
surgery was indicated. Only 6% of the population in the Mancl study were surgical/trauma
patients. Although an improvement in cardiac index with pharmacotherapy generally leads to
improved splanchnic blood perfusion during shock, most studies suggest—albeit with conflicting
data—that epinephrine, norepinephrine, and vasopressin tend to decrease splanchnic blood flow.
Intragastric administration of EN may be possible for some patients during low, stable doses of
vasopressors. However, close monitoring for feeding intolerance is mandatory.
5. EN and pancreatitis
a. American College of Gastroenterology guidelines (2013) (Am J Gastroenterol 2013;108:1400-16):
In mild acute pancreatitis, oral feeding can be initiated immediately if the patient has no nausea
or Am J Gastroenterol 2013;108:1400-16. In severe acute pancreatitis, EN is recommended to
prevent infectious complications, whereas PN should be avoided.
b. International consensus guidelines (2012) (JPEN J Parenter Enteral Nutr 2012;36:284-91): EN is
preferred to PN. EN may be used in the presence of pancreatic pseudocysts, fistulas, and ascites.
Early nutrition is indicated for severe pancreatitis. Post-pyloric feeding is not necessarily required.
For EN, consider a small peptide-based medium-chain triglycerides (MCT) oil formula to improve
tolerance. Use PN if EN is contraindicated or not well tolerated. Intravenous fat emulsions are
generally safe and well tolerated as long as the baseline triglycerides are below 400 mg/dL and the
patient has no history of hyperlipidemia.
c. SCCM/ASPEN (2009) (JPEN J Parenter Enteral Nutr 2009;33:277-316): Patients with severe
acute pancreatitis should have an NG tube placed and be initiated on EN as soon as fluid volume
resuscitation is complete. Patients with mild to moderate acute pancreatitis do not require nutrition
support therapy (unless an unexpected complication develops or the patient does not advance to
an oral diet within 7 days). Patients with severe acute pancreatitis may be fed by the gastric or
jejunal route. For the patient with severe acute pancreatitis, PN use should be considered when
EN is not feasible.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-91
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

d. Dickerson’s interpretation: Most clinicians will try to introduce a low-fat diet or EN formula (with
fat content predominantly as MCT) for patients with mild to moderate acute pancreatitis when the
abdominal pain and nausea are substantially reduced. Many clinicians will opt for PN for patients
with severe acute pancreatitis (with a prolonged course and NPO [nothing by mouth] for a few
to several days) if jejunal feeding not possible. When giving EN, a low-fat diet (which includes a
substantial portion of the fat kilocalories as MCT) is usually preferred.
6. Fish oil and acute respiratory distress syndrome (ARDS)
a. SCCM/ASPEN (2009) (JPEN J Parenter Enteral Nutr 2009;33:277-316): Patients with ARDS
and severe ALI should be placed on an enteral formulation characterized by an anti-inflammatory
profile (i.e., omega-3 fish oils, borage oil) and antioxidants.
b. ESPEN EN (2006) (Clin Nutr 2006;25:210-23): Patients with ARDS should receive EN enriched
with omega-3 fatty acids and antioxidants.
c. Canadian Practice Guidelines update (2014) (Nutr Clin Pract 2014;29:29-43): Downgraded
their initial 2009 recommendations from “recommend” (the specialized formula) to “should be
considered”
d. Rice (2011) (JAMA 2011;306:1574-81): 272 medical ICU patients with ALI/ARDS were
randomized to receive either a twice-daily omega-3 supplementation or an isocaloric, protein-
containing supplement together with their EN. The supplements were calorically equivalent
but differed in protein content (4 g/dose vs. 20 g/dose, respectively). Patients who received
fish oil bolus supplements had fewer ventilator-free days (implying worse outcomes) but no
difference in mortality or other organ failures compared with those who received the protein
control supplement. Those who received fish oil also experienced more diarrhea (29 vs. 21%,
respectively). The investigators concluded that twice-daily supplementation of fish oil/γ-linoleic
acid/antioxidant did not improve clinical outcomes and might be harmful.
e. Dickerson’s interpretation: The first three RCTs (Crit Care Med 1999;27:1409-20; Crit Care Med
2006;34:2325-33; Crit Care Med 2006;34:1033-8) and the subsequent meta-analysis (JPEN J
Parenter Enteral Nutr 2008;32:596-605) from those original studies examined a conventional
compared with a modified diet for patients with ALI and ARDS. The meta-analysis results showed
that the specialized diet improved ICU LOS, ventilator days, organ failure, and mortality compared
with the conventional diet. The problem with those studies was that they potentially compared a
treatment formula with “the worst formula.” In essence, the studies compared the modified diet
to another diet of similar high-fat content but as an omega-6 fat source (e.g., the worst formula,
given the hypothesis that ARDS is related to the inflammatory process, which can be worsened or
improved by the fat source). Rice’s data (JAMA 2011;306:1574-81) are difficult to interpret because
patients received a bolus supplement in addition to their EN, instead of continuously substituting a
portion of the fat kilocalories as fish oil. The control group likely received more protein (40 g/day
plus EN) than did the fish oil group (8 g/day plus EN). These patients were also part of the EDEN
(JAMA 2012;307:795-803 trial regarding “trophic vs. full feeds,” which may have also confounded
their data. Because the control group had improved outcomes compared with the fish oil supplement
group, it may be asked whether the difference in protein received by the control group contributed
to the improved outcome of the control group. Until more data are available, we continue to use a
modified diet for patients with ALI and ARDS, even though its benefit was questioned by one RCT.
7. Should lipids be held during the first week of PN?
a. SCCM/ASPEN (2009) (JPEN J Parenter Enteral Nutr 2009;33:277-316): In the first week of
hospitalization in the ICU when PN is required and EN is not feasible, patients should be given a
parenteral formulation without soy-based lipids.
b. ESPEN PN (2009) (Clin Nutr 2009;28:387-400): Lipid emulsions should be an integral part of PN
to provide energy and ensure essential fatty acid provision in long-term ICU patients.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-92
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

c. Dickerson’s interpretation: The SCCM/ASPEN recommendation was based on a single RCT of 57

trauma patients who received 30 kcal/kg IBW/day in which 25% of the non-protein kilocalories
were provided as an intravenous fat emulsion or glucose for the first 10 days of PN (J Trauma
1997;43:52-60). The soybean fat emulsion was infused over a 10- to 12-hour period. Calories
lost to the omission of lipid emulsion were not replaced in the control (no-lipid) group, with
each group receiving an average of 34 kcal/kg/day compared with 27 kcal/kg/day. Protein intake
was maintained at 1.6 g/kg/day. The lipid-containing PN group had more infections, including
differences in pneumonia and line sepsis (72 infections in 30 patients), than did the non-lipid
group (29 infections in 27 patients). Hospital LOS, ICU LOS, and ventilator days were all greater
for the lipid group as well (J Trauma 1997;43:52-60. Short-term (10 hour) infusion of lipid can
decrease reticuloendothelial (RES) system clearance after 3 days of intermittent intravenous lipid
emulsion infusion at a dosage of 0.13 g/kg/hour (JPEN J Parenter Enteral Nutr 1989;13:614-9).
In contrast, continuous infusion of intravenous lipid emulsion at the same total daily dosage
was not associated with impaired RES clearance (JPEN J Parenter Enteral Nutr 1990;14:467-
71). Unfortunately, serum triglycerides were not measured in the Battistella study (J Trauma
1997;43:52-60). This is important because RES clearance is adversely affected for patients
unable to clear the lipid emulsion, as evidenced by hypertriglyceridemia. Some clinicians have
also questioned whether overfeeding could have contributed to the observed poorer outcomes.
Many clinicians are skeptical of Battistella’s findings (J Trauma 1997;43:52-60) and await results
from a larger controlled trial. Two recent small retrospective studies have shown no difference in
infectious complications after early lipid emulsions (Surg Infect (Larchmt) 2011;12:43-7; Nutr
Clin Pract 2014;29:355-9). Most clinicians provide fat emulsion as part of the PN early in the
course of the patient’s hospitalization.
8. Glutamine
a. SCCM/ASPEN (2009) (JPEN J Parenter Enteral Nutr 2009;33:277-316): Adding enteral glutamine
to an EN regimen (not already containing glutamine) should be considered in burn, trauma, and
mixed ICU patients. When PN is used in the critical care setting, supplementation with parenteral
glutamine should be considered.
b. ESPEN EN (2006) (Clin Nutr 2006;25:210-23): Glutamine should be added to a standard
enteral formula in burn and trauma patients. Data are insufficient to support enteral glutamine
supplementation in surgical or heterogeneous critically ill patients.
c. ESPEN PN (2009) (Clin Nutr 2009;28:387-400): When PN is indicated in ICU patients, the
amino acid solution should contain l-glutamine 0.2–0.4 g/kg/day (e.g., alanyl-glutamine dipeptide
0.3–0.6 g/kg/day).
d. Canadian Practice Guidelines update (2014) (Nutr Clin Pract 2014;29:29-43): When PN
is prescribed to critically ill patients, parenteral supplementation with glutamine should be
considered (downgraded from “strongly recommended”). We strongly recommend that high-dose
glutamine not be used in critically ill patients with shock and multiorgan failure.
e. REDOXS study (2013) (N Engl J Med 2013;368:1489-97): Critically ill patients (n=1223) were
randomized to receive placebo, glutamine (0.35 g/kg/day intravenously plus 30 g/day enterally),
an antioxidant cocktail (500 mcg of intravenous selenium plus enteral administration of selenium
at 300 mcg/day, 20 mg of elemental zinc, beta-carotene, vitamin E, or 1500 mg of vitamin C), or
both glutamine and the antioxidant cocktail. One of these treatment regimens was given within 24
hours after admission to the ICU for a maximum of 28 days or until discharge from the ICU. There
was a trend toward increased 28-day mortality for those who received glutamine compared with
those who did not (32% vs. 27%; adjusted odds ratio of 1.28, p=0.05). Antioxidants had no effect
on 28-day mortality.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-93
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

f. Dickerson’s interpretation: The REDOXS study showed that aggressive glutamine dosing might
be harmful when used in the critically patient with multiple organ dysfunction (40% of patients
had renal dysfunction). The glutamine dose was about 0.7–0.8 g/kg/day for many patients, which
exceeded the 0.3- to 0.5-g/kg/day doses that previously were shown to be of benefit (Crit Care Med
2003;31:2444-9; Crit Care Med 2001;29:2075-80). The REDOXS data may have also been skewed
because the number of patients with more than two failing organs at baseline was higher in the groups
receiving glutamine than in those not receiving glutamine (187 vs. 148), which likely contributed
to a higher mortality. In addition, 79% of the patients were medical ICU patients. The current body
of literature suggests that glutamine supplementation appears to be beneficial for thermally injured
and trauma patients (Nutr Clin Pract 2011;26:479-94). Only 3% of the study population were trauma
patients, and those with significant thermal injury were excluded. Because of the unfavorable results
obtained in the REDOXS study, the principal investigator in a separate publication recommended that
glutamine supplementation be reserved for trauma or thermally injured patients at a dosage of 0.35–
0.5 g/kg/day ( JPEN J Parenter Enteral Nutr 2013;37:442-3). It is also recommended that glutamine
administration be avoided in patients with significant renal or hepatic dysfunction.
9. Selenium
a. SCCM/ASPEN (2009) (JPEN J Parenter Enteral Nutr 2009;33:277-316): A combination of
antioxidant vitamins and trace minerals (specifically including selenium) should be provided to all
critically ill patients receiving specialized nutrition therapy.
b. ESPEN PN (2009) (Clin Nutr 2009;28:387-400): All PN prescriptions should include a daily
dose of multivitamins and trace elements. Selenosis has been observed in the healthy population
with long-term intakes greater than 750 mcg/day; therefore, doses of 750–1000 mcg/day should
probably not be exceeded in the critically ill, and administration of supraphysiologic doses should
perhaps be limited to 2 weeks.
c. Surviving Sepsis Campaign guidelines (2013) (Crit Care Med 2013;41:580-637): Use of selenium
to treat severe sepsis is not recommended.
d. Canadian Practice Guidelines update (2014) (Nutr Clin Pract 2014;29:29-43): Given the outcome
of reduced infections, evidence is sufficient to upgrade the recommendation for the use of
intravenous/PN selenium supplementation from “insufficient data” to “should be considered.”
e. Dickerson’s interpretation: Although the emerging data look promising (excluding the REDOXS
trial), one problem is the practicality of delivering selenium therapy to non-infected critically
ill patients. Most studies that showed improved outcomes administered a bolus selenium dose,
followed by a continuous insulin infusion. Because many critically ill patients require other
continuous infusions (fentanyl/midazolam/insulin/vasopressors/propofol, PN), the practicality
of providing all of these infusions is complex given the availability of ports for infusion. In
addition, there is the problem of inadequate dosing information in AKI or renal insufficiency.
Thus, given selenium’s emerging but not universally accepted proof of benefit, it has become more
of a quagmire regarding what to do with selenium, and selenium is not usually provided beyond
conventional doses in PN solutions at many institutions.
10. Arginine: There is a risk of arginine causing hypotension because it is a precursor to the vasodilator
nitric oxide. In one small study of vasopressor-dependent critically ill patients with infections, an
intravenous bolus dose of 200 mg/kg of l-arginine resulted in transient hypotension and hemodynamic
changes that lasted for 10–15 minutes (Crit Care Med 1993;21:1287-95). However, given the amount
of arginine contained in nutritional formulations and its method of delivery (continuous infusion), it
is unlikely that EN and PN solutions could significantly contribute to hemodynamic instability (Nutr
Metab (Lond) 2012;9:54). In addition, the use of arginine-supplemented enteral diets has been shown
to improve infectious complications and decrease hospital LOS when used perioperatively for patients
requiring GI surgery (J Am Coll Surg 2011;212:385-99).

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-94
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

ANSWERS AND EXPLANATIONS TO PATIENT CASES

1. Answer: D take about 4–5 days to replete this patient’s magnesium

Hyperglycemia and other causes of non-hypotonic deficiency (presumably caused by chronic alcohol
hyponatremia have been excluded. Urine osmolality is ingestion). Supplemental magnesium would be added to
greater than 100 mOsm/kg, which rules out psychogenic the PN solution in addition to daily doses of intravenous
polydipsia, and a large amount of hypotonic fluids magnesium sulfate.
were not being given. Urine sodium was greater than
30 mEq/L, and the patient did not receive diuretic 6. Answer: C
therapy or have kidney disease. The patient appeared Although critical illness and fluid resuscitation
to be normovolemic without evidence of significant therapy may have factored into the development of
edema (expansion of the ECF compartment). Because his hypocalcemia, massive blood transfusion is the
the patient also has pneumonia (a common cause of most profound cause. Citrate, added to the blood as
SIADH), all of these factors indicate that the patient has an anticoagulant, readily binds calcium and can cause
hyponatremia caused by SIADH. hypocalcemia. Previous studies have shown that
hypocalcemia is common when patients are given more
2. Answer: B than 5 units of blood at a time.
Fluid restriction in which excess water is retained relative
to sodium is the most appropriate treatment of SIADH. 7. Answer: B
The “vaptans” may also be considered; however, this A short-term intravenous infusion of 4 g of calcium
was not a choice. gluconate (1 g = 4.6 mEq) over 4 hours has previously
been shown to be a safe and effective therapeutic
3. Answer: D regimen for moderate to severe hypocalcemia (ionized
The best way to fluid-restrict an enterally fed patient is calcium less than 1 mmol/L). A bolus dose of calcium
to use the most concentrated formulas, which are the chloride (1 g = 13.6 mEq) would be an effective means
2-kcal/mL formulations that are specifically designed for treating symptomatic hyperkalemia, but it would be
for patients with congestive heart failure. Unfortunately, unnecessarily aggressive in this patient scenario.
protein intake may be inadequate with the use of these
formulations in certain populations, and supplemental 8. Answer: A
protein may have to be provided. Calculation of the AG (141 + 4 − 117 − 21 = 7) shows
that no AG is present. The pH of 7.29 with a low-normal
4. Answer: D Pco2 suggests mild metabolic acidosis. Respiratory
These dosages should be selected as the correct answer compensation should have been better, but this may not
as they follow the dosing guidelines. However, in prac- be possible for the patient, depending on the ventilator
tice, because this patient is at risk of refeeding syndrome settings. The serum chloride of 117 mEq/L indicates
(substantial recent weight loss) and hypophosphatemia hyperchloremia. A non-AG hyperchloremic metabolic
(weight loss, glucose-based PN solution, liver resection), acidosis is common for patients with an ileal conduit
a more aggressive dosing approach would probably be where urine is diverted into a short loop of ileum (urinary
better. Because he has an NG volume output of 2 L, chloride reabsorption occurs) that acts as ostomy for
enteral administration of electrolytes would not be prac- urine output.
tically feasible. Supplemental potassium and phosphorus
would be added to the PN solution, in addition to daily 9. Answer: A
intravenous doses of potassium and phosphorus. Because the severity of the patient’s acidemia is mild
(pH 7.29), aggressive therapy with sodium bicarbonate
5. Answer: B is not indicated. In addition, sodium bicarbonate is
Because it takes about 48 hours for serum magnesium incompatible with calcium-containing PN solutions
to redistribute, the next day’s serum magnesium and cannot be added to the PN solution. Calculation of
concentration is “falsely elevated.” In general, it will the bicarbonate deficit (0.5 x 50 x (24 − 21) = 75 mEq)

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-101
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

suggests that changing the sodium chloride to sodium 6.4 g of nitrogen, which should place the patient close
acetate would correct the bicarbonate deficit within 24–48 to nitrogen equilibrium (assuming that most of the
hours. However, patients with ileal conduits tend to have additional nitrogen is retained). Increasing the protein
persistent hyperchloremic metabolic acidosis (likely intake without altering the non-protein intake will
from other chloride solutions given to the patient, which increase the total caloric intake to 26 kcal/kg/day. If the
are then excreted in the urine and reabsorbed in the ileal goal is to keep the caloric intake the same at around 24
conduit), and more prolonged therapy may be required. kcal/kg IBW/day, reducing the fat intake to 35 g/day
(from 50 g/day) would meet that goal.
10. Answer: C
The current PN regimen provides 61 kcal/kg/day total 14. Answer: A
(glucose 6.1 mg/kg/minute and lipid emulsion 1.5 Because the target BG should be 140–180 mg/dL for
g/kg/day) and protein 4 g/kg/day. The PN regimen this surgical patient being transferred to the floor, a
represents gross overfeeding of this small woman and modest improvement in glycemic control is indicated.
can explain her hyperglycemia and hypercapnia. Cutting Ideally, obligatory glucose requirements should be met
all macronutrients by about one-half would result in a (e.g., about 130 g/day plus about 80–150 g/day for
more reasonable regimen for this patient: 30 kcal/kg/ wound healing) to prevent the use of amino acids for
day (glucose 3 mg/kg/minute and lipid emulsion 0.8 gluconeogenesis. Thus, decreasing the glucose intake
g/kg/day) and protein 2 g/kg/day. Because she is so to 100 g/day is not desirable, given the mild increases
small (40 kg), it would be important to double-check in BG concentration. The easiest method to achieve
the weight-based calculation to see whether this new glycemic control and meet caloric needs is to modestly
regimen is appropriate to meet her caloric needs without increase the regular human insulin by 10 units/day. The
overfeeding by calculating the BEE using the Harris- patient is unlikely to experience hypoglycemia with the
Benedict equation for females (caloric intake should provision of insulin at 30 units/day when given 200 g of
not exceed 1.3–1.5 x BEE for a critically ill patient with intravenous dextrose concurrently. As the stress resolves
traumatic injuries). and glycemic control improves, insulin can be decreased
or eliminated from the PN solution.
11. Answer: A
Total kilocalories per day = (200 g x 3.4 kcal/g dextrose)
+ (150 g x 4 kcal/g protein) + (50 x 10 kcal/g lipid
emulsion) = 1780 total kcal/75 kg = 23.7 kcal/kg IBW/
day = 24 kcal/kg IBW/day. Protein intake is 150 g/75 kg
= 2 g/kg IBW/day. This caloric and protein intake would
be appropriate for hypocaloric, high-protein nutrition
therapy in critically ill patients with obesity without
significant renal or hepatic dysfunction.

12. Answer: C
Nin = 150 g of amino acids/6.25 = 24 g.
UUN = 900 mg/dL = 9 g/L; 9 g/L x 3 L = 27 g.
NB = 24 – 27 – 4 = -7.

13. Answer: B
In an effort to avoid overfeeding yet meet the demands
of increased catabolism, the appropriate modification
of the PN formulation is to increase the protein content
while decreasing the non-protein energy content.
Increasing the protein intake from 150 g to 190 g/day
(around 2.5 g/kg IBW/day) will provide an additional

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-102
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

ANSWERS AND EXPLANATIONS TO SELF-ASSESSMENT QUESTIONS

1. Answer: C 5. Answer: B
An ileus, usually detected on radiologic examination Studies show that increases in mesenteric potassium
of the lower abdomen, indicates lack of motility and concentrations detected by potassium sensors in the
presence of distention and air within the small bowel. This splanchnic vascular bed evoke increased renal potassium
is usually depicted as “dilated loops of bowel.” Patients excretion (feed-forward regulation of potassium
cannot be fed safely or efficaciously by the enteral route homeostasis), even before regulation by aldosterone
during an ileus. A feeding tube can be placed for enteral (classic feedback regulation). Some clinicians may
feeding of the patient with anorexia, and PN is not have erroneously selected bioavailability, but the
indicated. Presence or absence of bowel sounds is not bioavailability of enteral potassium is 95%–100% in
an accurate marker for assessing bowel function. A high the absence of aberrations in GI motility, function,
GRV during enteral feeding, combined with abdominal or anatomy. A major difference between enteral and
distension, bloating, emesis, or regurgitation, can often parenteral potassium is that the rate of absorption is
be efficaciously treated with prokinetic pharmacotherapy slower with enteral potassium. The rate of intravenous
or advancement of the feeding tube into the small bowel potassium administration can be inadvertently infused
with resumption of enteral feeding. too quickly (it is acceptable to infuse potassium at 10
mEq/hour for patients without a cardiac monitor and up
2. Answer: B to 20 mEq/hour for those with a monitor).
The correct answer is B, “0.45% sodium chloride and
potassium chloride 20 mEq/L,” based on the average 6. Answer: C
electrolyte composition of gastric fluid (see Table 5 Folic acid deficiency is correct because the patient’s
regarding the electrolyte composition of GI fluids). homocysteine concentration is elevated, whereas her
methylmalonic acid concentration is normal. If both were
3. Answer: A elevated, it would likely be a vitamin B12 deficiency,
With significant diarrhea, intravenous zinc requirements although a combined B12-folate deficiency is possible
from GI fluid losses during critical illness will increase (but less common than a B12 deficiency alone). If her
from the normal requirements of 3–5 mg/day. Data show methylmalonic acid concentration were elevated and
that most patients with increased intestinal losses can her homocysteine were normal (rare), she would likely
achieve a positive zinc balance on 13 mg of intravenous have a vitamin B12 deficiency. If both are normal, the
zinc daily Gastroenterology 1979;76:458-67). As a patient’s macrocytosis is caused by other factors such as
result, most clinicians will provide additional zinc liver disease or alcohol.
supplementation for patients with short bowel syndrome,
intestinal fistulas, or prolonged and sustained diarrhea. 7. Answer: D
The correct answer is D, “additional magnesium therapy
4. Answer: C should be given daily over the next 4–5 days,” because it
Given the severity of the patient’s condition (recent sei- take 48 hours for magnesium to equilibrate after a short-
zure from severe hyponatremia) and likely diagnosis term infusion. Treatment of significant hypomagnesemia
of syndrome of inappropriate diuresis or SIADH. The usually takes a few to several days of repletion therapy.
immediate goal should be to achieve a serum sodium Hypocalcemia should autocorrect with magnesium
concentration of greater than 120 mEq/L by short term supplementation with 48 hours of magnesium therapy, but
infusion of 3% sodium chloride. Conivaptan could then calcium therapy can be given concurrently, if necessary
be given to correct the hyponatremia limiting the increase (symptomatic or ionized calcium concentration less than
in serum sodium concentration to less than 10–12 mEq/L/ 1 mmol/L).
day. Fluid restriction is imperative and is the primary
overall management technique for this patient.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-103
 Fluids, Electrolytes, Acid-Base Disorders, and Nutrition Support

8. Answer: B
Described are the calculations for determining an NB:
NB = Nin – UUN – 4
Nin = protein in (g/day)/6.25 = 20.8 g
UUN (g/day): 900 mg/dL = 9000 mg/dL = 9 g/L; 2700
mL/day = 2.7 L/day
9 g/L x 2.7 L/day = 24.3 g/day
NB = 20.8 – 24.3 – 4 = -7.5 g/day
If asked, “what adjustments would you make to the par-
enteral nutrition regimen?”, the best choice would be to
increase the protein intake (to about 2 g/kg/day) because
the current regimen provides only around 1.4 g/kg/day.
Although the NB is usually negative for a critically ill
patient because the anabolic effect of nutrition cannot
completely overcome the catabolism of critical illness,
most patients can achieve close to nitrogen equilibrium
(an NB of around -4 to +4 g/day). Increasing the protein
intake by 50 g/day should provide about 8 g of additional
nitrogen, which should be sufficient to achieve an NB
close to nitrogen equilibrium.

ACCP Updates in Therapeutics® 2015: Critical Care Pharmacy Preparatory Review Course

1-104