Neurologic Outcomes of Surgery and Anesthesia, 1st Edition

Visit the link below to download the full version of this book:

https://medipdf.com/product/neurologic-outcomes-of-surgery-and-anesthesia-1st-ed
ition/

Click Download Now

 Dedicated to our fathers, Alexander Mashour (1929–2010) and
Jos Avidan (1929–2012)—two free spirits who played by their own rules
This page intentionally left blank
 CON T EN TS

Preface ix 8. Spinal Cord Injury and Protection 67

Contributors xi Rehan Siddiqui and Rae M. Allain
9. Neuropathic Pain 77
Robert S. Griffin and Gary J. Brenner
PA RT O N E
BRAIN
PA RT T HR E E
1. Anesthetic Neurotoxicity: Effects of Anesthetic N E RV E
Drugs on the Developing Brain 3
Ansgar M. Brambrink 10. Peripheral Nerve Injury 87
2. Delirium 13 Marnie B. Welch and Chad M. Brummett
Elizabeth L. Whitlock, Michael S. Avidan, 11. Postoperative Visual Loss and Ischemic
and Sharon K. Inouye Optic Neuropathy 97
3. Postoperative Cognitive Trajectory 21 Vijay Kumar Rimaiah and Lorri A. Lee
Robert D. Sanders, Mervyn Maze, Alex S. Evers,
and Michael S. Avidan
4. Perioperative Stroke 31 PA RT F O UR
Laurel E. Moore, Lewis B. Morgenstern, FUTURE DIRECTIONS
and George A. Mashour
5. Seizures 41
12. Standards for Intraoperative Neurophysiologic
Adam D. Niesen, Adam K. Jacob, and Monitoring 113
Sandra L. Kopp
Antoun Koht, J. Richard Toleikis, and Tod B. Sloan
6. Posttraumatic Stress Disorder Following Surgery,
13. Neurologic Biomarkers 129
Intensive Care, and Intraoperative Awareness 47
Juan P. Cata
Elliott Karren, Elizabeth L. Whitlock,
Thomas L. Rodebaugh, and Michael S. Avidan
Index 137

PA RT T WO
S P I NA L C O R D

7. Cervical Spine Injury 59

Mazen A. Maktabi
This page intentionally left blank
 PR EFACE

The latter half of the twentieth century saw a dramatic a single organ for the purposes of outcomes studies. Finally,
improvement in perioperative outcomes, in large part there are still a number of fundamental mysteries regarding
because of the development of safer drugs and the stand- neural function. Indeed, one reason we do not have a stan-
ardization of intraoperative monitoring. It is striking, how- dard monitor for the brain is that there is an incomplete
ever, that the standard monitors in current use are limited understanding of what is best to measure and where.
to the cardiovascular and respiratory systems, when it is the Despite the challenges—or, rather, because of them—
nervous system that is the primary target of the therapeu- neurologic and psychiatric outcomes of surgery, as well as
tic effects of general anesthesia. Indeed, during the twenti- the potentially neurotoxic effects of general anesthetics, are
eth century attention was focused on the cardiopulmonary a subject of intense and often controversial inquiry. Where
outcomes of surgery, with significantly less attention paid can the interested clinician or investigator turn for a concise
to the brain and other neural structures. The consequent introduction to the issues at hand? Neurologic Outcomes of
improvement in cardiopulmonary outcomes led to a new Surgery and Anesthesia provides, for the first time, a con-
standard of safety and a new horizon of outcomes research. cise and clear resource for information on virtually all major
The perioperative physicians and investigators of the neurologic and psychiatric outcomes, including delirium,
twenty-first century are now recognizing the need for more stroke, posttraumatic stress disorder, spinal cord injury,
extensive study of neurologic outcomes. However, this postoperative visual loss, and more. Furthermore, general
endeavor is proving to be a significant challenge, for several topics of interest such as neurotoxicity, neuromonitoring,
reasons. First, there are no standard intraoperative monitors and neurologic biomarkers are addressed. Internationally
for the brain or nervous system. Thus, unlike the electrocar- recognized experts who have made important contribu-
diogram, which can detect intraoperative myocardial isch- tions to the primary literature of the field have been selected
emia, we do not routinely use an electroencephalogram or to write high-yield chapters in a standardized format that
other brain monitor that might detect intraoperative cere- enhances clarity. We thank these scholars and clinicians for
bral ischemia. Second, there is a notable lack of biomarkers their outstanding work, which we believe has resulted in a
for the nervous system, which limits the large-scale study cutting-edge and readable textbook that will bring together
of neurologic outcomes. There are a number of commonly in a single volume the many neurologic complications of
assessed quantitative markers for heart or kidney damage, surgery and anesthesia that previously were treated as indi-
but there is nothing comparable to troponin or creatinine vidual topics.
for the brain or spinal cord. Furthermore, some outcomes The brain in particular—and the nervous system in
relate to subtle functional abilities that may not even be general—is often regarded as the final frontier of the bio-
reflected in abnormal neural biomarker levels. More subtle logical sciences because of its complexity, functional het-
deficits in cognition, for example, require careful, costly, erogeneity, and sensitivity to injury. Similarly, we feel that
and labor-intensive examinations by trained professionals, the nervous system is the final frontier of perioperative
creating obstacles for large-scale studies. The third challenge medicine and hope that Neurologic Outcomes of Surgery and
is that the nervous system is not functionally homogenous. Anesthesia creates new awareness of this exciting field as well
Unlike the heart that serves primarily as a pump, the nervous as a renewed sense of importance in protecting the nervous
system (even the brain itself ) has tremendous functional system from injury during surgery and beyond.
specialization, rendering it almost impossible to treat it as — George A. Mashour and Michael S. Avidan

ix
This page intentionally left blank
 CON T R IBU TOR S

Rae M. Allain, MD Juan P. Cata, MD

Associate Director, Surgical Critical Care Assistant Professor, Department of Anesthesiology and
Department of Anesthesiology and Perioperative Medicine
Pain Medicine Section of Neuroanesthesia
St. Elizabeth’s Medical Center University of Texas
Boston, Massachusetts MD Anderson Cancer Center
Houston, Texas
Michael S. Avidan, MB BCh, FCASA
Professor of Anesthesiology and Surgery Alex S. Evers, MD
Director, Institute of Quality Improvement, Research & Henry E. Mallinckrodt Professor, Anesthesiology
Informatics Head, Department of Anesthesiology
Division Chief, Cardiothoracic Anesthesiology Washington University School of Medicine
Washington University School of Medicine St. Louis, Missouri
St. Louis, Missouri
Robert S. Griffin, MD, PhD
Ansgar M. Brambrink MD, PhD Department of Anesthesiology
Professor, Department of Anesthesiology & Perioperative Division of Musculoskeletal and Interventional Pain
Medicine Management
Oregon Health & Science University Hospital For Special Surgery
Portland, Oregon New York, New York

Gary J. Brenner, MD, PhD Sharon K. Inouye, MD, MPH

Director of the MGH Pain Medicine Professor, Department of Medicine
Fellowship Harvard Medical School; and
Massachusetts General Hospital Director, Aging Brain Center
Department of Anesthesia, Critical Care and Pain Institute for Aging Research
Medicine; and Hebrew SeniorLife
Assistant Professor in Anesthesia Boston, Massachusetts
Harvard Medical School
Boston, Massachusetts Adam K. Jacob, MD
Assistant Professor, Department of Anesthesiology
Chad M. Brummett, MD Mayo Clinic
Assistant Professor Rochester, Minnesota
Director, Pain Research
Department of Anesthesiology Elliott Karren, MD
Division of Pain Medicine Department of Anesthesiology
University of Michigan Washington University School of Medicine
Ann Arbor, Michigan St. Louis, Missouri

xi
Antoun Koht, MD Adam D. Niesen, MD
Professor of Anesthesiology, Neurological Surgery and Assistant Professor, Department of Anesthesiology
Neurology Mayo Clinic
Northwestern University Medical School Rochester, Minnesota
Chicago, Illinois
Vijay Kumar Rimaiah, MBBS, MD
Sandra L. Kopp, MD Acting Instructor and Senior Fellow, Department of
Associate Professor, Department of Anesthesiology Anesthesiology and Pain Medicine
Mayo Clinic University of Washington
Rochester, Minnesota Seattle, Washington

Lorri A. Lee, MD Thomas L. Rodebaugh, PhD

Professor, Departments of Anesthesiology and Pain Assistant Professor, Department of Psychology
Medicine, and Neurologic Surgery (adjunct) Washington University School of Medicine
University of Washington St. Louis, Missouri
Seattle, Washington
Robert Sanders, MD
Mazen A. Maktabi, MD Senior Clinical Research Associate, Imaging
Chief, Division of General Surgery Anesthesia Neuroscience
Department of Anesthesia, Critical Care, and Pain Institute of Neurology
Medicine Faculty of Brain Sciences
Massachusetts General Hospital University College London
Harvard Medical School London, England
Boston, Massachusetts
Rehan Siddiqui, MD
George A. Mashour, MD, PhD Department of Anesthesia, Critical Care, and Pain
Associate Chair for Faculty Affairs Medicine
Associate Professor of Anesthesiology and Neurosurgery Massachusetts General Hospital
Faculty, Neuroscience Graduate Program Harvard Medical School
University of Michigan Medical School Boston, Massachusetts
Ann Arbor, Michigan
Tod B. Sloan, MD, MBA, PhD
Mervyn Maze, MD Professor of Anesthesiology
Professor and Chair, Department of Anesthesia and University of Colorado School of Medicine
Perioperative Care Aurora, Colorado
University of California, San Francisco
San Francisco, California J. Richard Toleikis, PhD
Associate Professor of Anesthesiology
Laurel E. Moore, MD Rush University Medical School
Clinical Assistant Professor, Department of Chicago, Illinois
Anesthesiology
Assistant Clinical Director, University Hospital Operating Marnie B.Welch, MD
Rooms Department of Anesthesiology
Director, Neuroanesthesiology Dartmouth Hitchcock Medical Center
University of Michigan Medical School Lebanon, New Hampshire
Ann Arbor, Michigan
Elizabeth L. Whitlock, MD, MSc
Lewis B. Morgenstern, MD Department of Anesthesia and Perioperiative Care
Professor, Neurology University of California
University of Michigan Medical School San Francisco Medical Center
Ann Arbor, Michigan San Francisco, California

xii • C O N T R I B U TO R S
PA RT I

BRAIN
This page intentionally left blank
 1.
ANESTHETIC NEUROTOXICIT Y
E F F E C T S O F A N E S T H ET I C D RU G S O N T H E D E VE L O P I N G B R A I N

Ansgar M. Brambrink

I N T R O D U C T I O N TO hyperalgesia,5 developmental brain alterations,6 and inap-

THE CLINICAL PROBLEM propriate behaviors.7–9 Nevertheless, prolonged exposure to
some of the very agents that are available to treat pain, pro-
While over many decades the field of anesthesiology has vide sedation, and maintain anesthesia may be harmful to
developed safe strategies to provide anesthesia to infants human fetuses and infants. Such exposure could potentially
and children, thereby radically improving perioperative produce the outcomes that occur following the experience
care and surgical outcomes in the young, there currently is of pain and stress without alleviating medication. This par-
growing concern that most of today’s typically used anes- adox is stimulating discussions regarding the risk-to-benefit
thetics might be harmful to the developing brain. Evidence relationship of providing human neonates and young infants
is mounting that exposure to sedative and anesthetic agents with anesthetic and sedative medications for painful or
injures brain cells and could potentially cause persistent stressful procedures.10 The relevance of anesthetic-induced
neurocognitive deficits that present as inappropriate behav- neurotoxicity to developing humans is currently unclear;
iors, leaning difficulties, and specific neuropsychological therefore, to date no practice changes have been advocated.
syndromes.1 However, both in society and among health care practitio-
Pediatric anesthesiologists and intensive care clinicians ners, awareness is growing about the potential neurotoxic-
face a difficult dilemma. On the one hand, strong evidence ity of anesthetic agents to developing humans. It is essential
from the laboratory promotes the view that many sedatives that the clinical relevance of anesthetic neurotoxicity be
and anesthetics are harmful to the developing brain of sev- clarified and that the safety of specific sedative, analgesic,
eral species, including nonhuman primates.2,3 In contrast, and anesthetic drugs be explored. Two prospective clin-
comparable data pertaining to humans are not available, ical studies are under way, which aim to answer some of
and the clinical evidence is ambiguous and currently based the burning questions related to this issue11,12 (for review,
on retrospective epidemiologic studies.4 The conundrum see Stratmann [2011]3), such as whether regional anesthe-
for health care providers and patients is that virtually all sia is safer than general anesthesia. Consortiums between
currently available anesthetic agents appear to cause harm government and scientific organizations have been created
in animal models, and there are no proven safe alternatives to guide the discussions between professionals and soci-
on the horizon. ety, as well as to support research and development aimed
It is now accepted that infants and neonates experience at resolving this perceived public health problem as expe-
pain and stress. Providing them adequate analgesia and seda- ditiously as possible13,14 (for more detailed information see
tion during surgery reduces the risk that they will acquire http://www.smarttots.org).

3
 I N C I D E N C E , P R E VA L E N C E , of longer than 120 minutes was associated with the highest
A N D O U TC O M E S risk.42 Adjustment for comorbidity confirmed these results
and suggested a two-fold higher hazard ratio for developing
Since the first descriptions of toxic effects of neuroactive a learning disability with two or more anesthesia exposures
drugs in the developing brain,15–17 strong evidence has been (median duration 75 minutes) before the age of 2 years.43
provided by multiple studies that anesthetic drugs cur- Most recently, the same group using the same data sources
rently used in humans cause widespread neuroapoptosis in showed that repeated exposure to general anesthesia before
immature mice, rats, pigs, guinea pigs, and nonhuman pri- age 2 years more than doubles the risk for the diagnosis of
mates.15–36 This injury occurs in developing animal brains attention-deficit/hyperactivity disorder until age 19 years,
at doses and durations that are clinically relevant and is even after adjusting for comorbidities.44
associated with long-term neurocognitive deficits weeks to Thomas and co-workers from the University of Iowa
months after the exposure. reported data from a smaller sample that produced results
While initially challenged regarding validity and rele- along the same line. Brief anesthesia before 1 year of age
vance,37–39 the evidence from animal studies pointing toward doubled the risk of scoring below the 5th percentile in
a significant problem has gained acceptance, first because school achievement tests compared to the population
several independent groups have reproduced the findings norm. Moreover, the poorest academic performance was
of neuroapoptosis, and second as evidence is consistent observed in those children exposed to the longest duration
across mammalian species, including nonhuman primates. of anesthesia.45
Nevertheless, it remains a matter of discussion whether the Kalkman and co-workers from the University of
animal findings are relevant to the human situation, and it Utrecht in The Netherlands also reported a higher inci-
will take several years before results from these prospective dence of subsequent neurocognitive impairments when
clinical trials are available. anesthesia was provided to infants than that among child-
The best available human evidence comes from epidemi- ren older than 1 year at the time of exposure.46
ologic studies. While the data are inconclusive, several ret- Hansen and co-workers from the Odense University
rospective analyses suggest that anesthesia exposure during in Denmark reported from a large cohort of Danish ado-
infancy may increase risk for neurobehavioral disturbances. lescents that there was no evidence of impaired academic
Emerging evidence suggests that age at exposure, number of achievement in those who had undergone one brief anes-
exposures, and total duration of anesthesia may all be asso- thetic (30–60 minutes) at 1 year of age or younger.47
ciated with worse neurobehavioral outcomes. Bartels and co-workers from the Vrije University in
DiMaggio and co-workers at Columbia University Amsterdam analyzed data from The Netherlands Twin
analyzed data of children enrolled in Medicaid in order to Registry and found that monozygotic twins who were
identify those who were exposed to brief general anesthesia exposed to anesthesia before age 3, according to parental
before 3 years of age. Their results demonstrate that expo- report, had significantly lower educational achievement
sure to anesthesia more than doubles the risk of being diag- scores and significantly more cognitive problems than
nosed with behavioral disorders during later development twins not exposed to anesthesia. However, in those twin
compared to matched children who were not exposed. pairs in which one twin was exposed to anesthesia and the
Moreover, the risk was even higher in children exposed other twin was not, the academic achievements were similar
twice or more during the first 3 years of life.40 Analysis of between the two.48
a birth cohort of siblings nested in the this study suggested It is remarkable that several large and reasonably
that the risk of being diagnosed with developmental and well-controlled, yet still retrospective, studies document an
behavioral disorders was 60% greater than that of a simi- association between brief exposure to anesthesia in early life
lar group of siblings who did not undergo surgery before and neurobehavioral disturbances later in life. While these
3 years of life.41 studies certainly prove relevance for further investigation in
Wilder, Flick, and co-workers at the Mayo Clinic eval- humans, at this point it is not possible to prove a causal rela-
uated county-based data for children who had received tionship based on theses retrospective studies because of their
anesthesia prior to 4 years of age. Their analysis indicated inherent methodologic limitations and their limited focus.
no measurable risk for learning disabilities when young For instance, all of these studies are based on second-
children received a single general anesthetic. In contrast, ary analyses of observational data that had been collected
children anesthetized twice or more were at increased risk usually for other purposes, and some include information
for learning disabilities. Additionally, anesthesia duration from parental interviews. Another concern is that all data

4 • N E U R O L O G I C O U TC O M E S O F S U R G E RY A N D A N E S T H E S I A
sources describe anesthesia-relevant information that dates Recent evidence suggests that some nonanesthetic/
back several decades (1970–1980s). Since then, anesthesia nonsedatives that are used clinically, such as caffeine, may
practice has changed significantly. For example, the volatile be able to augment the deleterious effects of NMDA antag-
anesthetic agent halothane is rarely used today in developed onists and GABAA-mimetic drugs. Caffeine, a nonselective
countries, and monitoring devices that were unavailable 30 adenosine receptor antagonist and an inhibitor of acetyl-
years ago are standard care today (e.g., quantitative mea- choline esterase, increased apoptotic cell death associated
surements of anesthetic gases; capnometry and capnogra- with exposure to alcohol,56 anesthetics like isoflurane,57,58
phy; peripheral oxygen saturation measurement via pulse and sedatives like benzodiazipines59 in several experimental
oximetry). The focus of several investigations was limited models. The clinical relevance of these findings is that some
to subjects exposed only briefly to anesthetic drugs. While preterm infants are treated for weeks with caffeine to stimu-
their population-based approach removes certain biases, late their respiratory drive in the intensive care environment
none of them provides information about the indication to treat apnea.60 At the same time many of these infants also
for anesthesia and the drugs used, and some provide no receive sedatives to relieve distress or anesthetics to induce
information about the number of exposures, the duration tolerance for necessary procedures. Still other preterm
of exposure, or whether investigators attempted to control infants are treated short term in the context of general anes-
for confounders such as socioeconomic status or family thesia in order to reduce the risk for postoperative apnea.
conditions. Prolonged durations of exposure, combinations A probable key risk factor for neurotoxic effects of
of different drugs, and developmental age at exposure are anesthetics is age at exposure. Evidence from the labora-
all likely important determinants of severity and pattern of tory suggests that the period of peak vulnerability coin-
neuropathologic and neurobehavioral outcome after early cides with the “brain growth spurt,” which varies between
anesthesia exposure, but these factors require more detailed species and is characterized by rapid synaptogenesis.15,16 In
evaluation. Two large, prospective human trials are under rodents this developmental period is considered to occur in
way, and their results might help clarify some of the risk the early postnatal period. In contrast, in humans synapto-
factors for bad outcomes and answer whether and to what genesis continues from about mid-gestation to several years
extent early exposure to general anesthesia might cause after birth,61,62 thus vulnerability to anesthetics in humans
long-term adverse neurobehavioral sequelae.49–52 could extend long beyond infancy. Retrospective studies in
humans suggest an increased risk for neurobehavioral defi-
cits when anesthesia is provided to patients under the age of
R I S K FAC TO R S 1–4 years. Studies in nonhuman primates support the con-
cerns about an extended period of vulnerability. Ketamine
Key risk factors are the type of drugs and the drug com- anesthesia in the pregnant rhesus monkey at the beginning
binations to which the developing brain is exposed. of the third trimester induced fetal neuroapoptosis.29,35
Anesthesia-induced developmental neuroapoptosis was Anesthesia with isoflurane,31,32,36 ketamine,30,32,35 or nitrous
first observed after exposure of infant rodents to drugs that oxide plus isoflurane33 resulted in apoptotic cell death in
block the N-methyl-D-aspartate (NMDA) subtype of glu- 5- to 6-day-old rhesus macaques.
tamate receptors.15 Subsequent research in the same model When comparing neurodevelopmental milestones, a
showed an even more pronounced apoptogenic potency third-trimester rhesus fetus corresponds approximately to a
for ethanol, which both blocks NMDA receptors and pro- human neonate. A 5- to 6-day-old rhesus infant is approx-
motes neurotransmission at γ-aminobutyric acid (GABAA) imately equivalent to a 6-month old human infant.62,63 It is
receptors.16 Research was quickly extended to include other unclear whether the window of vulnerability indeed closes
drugs that interact with GABAA and NMDA receptors at that time or is species dependent and for certain sub-
as agonist and antagonist, respectively, known targets for groups of brain cells remains open for a significantly longer
many anesthetics and sedatives. Studies showed similar del- period of brain development.
eterious effects when clinically relevant doses of ketamine,17 Another issue that has not been systematically studied
midazolam,17 propofol,23 isoflurane,22,26,27 sevoflurane,24,25 is whether the developing brain has a differential sensi-
desflurane,27,53 and chloral hydrate,54 as well as antiepilep- tivity to the apoptogenic action of anesthetic drugs dur-
tic drugs, including those that block voltage-gated sodium ing the fetal and neonatal periods. Our own preliminary
channels,55 were administered to infant rodents. Moreover, findings in nonhuman primates indicate that isoflurane
combinations of NMDA antagonists with GABAA-mimetic induces more severe damage in the neonatal brain than in
drugs potentiated the neuroapoptotic effects.17,19,20 the fetal brain. In contrast, ketamine is more toxic for the

A N E S T H ET I C N E U R OTOX I C I T Y • 5
fetal brain than for the neonatal brain. These observations Exposure to anesthetics apparently also affects neuronal
underscore the interdependence of both the developmen- dendrites and synapsis. Isoflurane has been shown to reduce
tal age at the time of exposure and the class of anesthetic the number of immature dendritic spines in vitro and the
drug administered. Evidence that ketamine (and perhaps number of synapses in vivo of infant mice, which is likely
other NMDA antagonists, including alcohol64) may be mediated by isoflurane-induced p75 receptor signaling.74,75
more toxic when exposure occurs prenatally suggests that Furthermore, Briner and co-workers showed that isoflu-
we need to focus future human studies on in utero fetuses rane, sevoflurane, desfurane, and propofol all reduce syn-
and premature infants. aptic spine density in newborn rodents; but when given to
A precise delineation of the peak vulnerability also older infant rodents (16–25 days old), the same drugs actu-
requires an understanding of the key mechanisms that ally increased spine density and synaptogenesis.76–78
lead to persistent neurobehavioral deficits at a later age. Proliferation and differentiation of new neurons appear
While several studies in rodents17,18,20,25,65–67 and one sensitive to anesthetics in an age-dependent fashion. Rodents
“proof-of-concept” study in nonhuman primates34 docu- exposed at a young age showed a persistent impairment of
ment that exposure to anesthesia early in life has long-term neurogenesis that was associated with neurocognitive defi-
neurocognitive consequences, it remains unknown whether cits later in life.79,80 In particular, multiple brief isoflurane
the functional deficits result from injury to neurons, to glia, exposures to rats and mice on postnatal days 14–17 caused
or to the proliferating stem cell pool, or whether an impaired a reduction in dentate hippocampal neurons and neurocog-
ability to create cell-to-cell contact is responsible. nitive impairment that became progressively more severe
Neuroapoptosis has been shown to result from expo- with advancing age.80 Other investigators have provided
sure to different anesthetics as well as to alcohol during fetal evidence that neonatal exposure of rodents to ethanol81 or
and neonatal life in rodents and in other species, including to various sedatives and anticonvulsants82 suppresses neuro-
guinea pig,28 piglet,68 and nonhuman primate.29–36,64 No genesis in the dentate hippocampal gyrus, which is associ-
brain region is completely spared, but the brain regions ated with a permanent reduction in the number of dentate
sustaining the most severe neuronal losses are regions that hippocampal neurons and with persistent neurocognitive
receive and integrate sensory information through both vis- deficits as the animals mature. Interestingly, the vulnera-
ual and auditory association pathways, which are critically bility for anesthesia-induced impairment of neurogenesis
important for normal neurocognitive function. Our own appears to extend past the second postnatal week in mice80
data showed that thalamus and basal ganglia were severely and thereby lasts longer than the window of vulnerability
damaged in fetal brain of nonhuman primates exposed to currently assumed for anesthesia-induced neuroapoptosis.
anesthetics in utero. This closely mirrors findings in human It seems likely that each of these injury mechanisms has
children69–71 who were exposed in utero to alcohol or antiep- a specific period of peak vulnerability; further research is
ileptic drugs, both of which are neurotoxic to the develop- needed to provide more evidence.
ing brain. Anesthetic exposure appears to affect particularly
glutamatergic, GABAergic, and dopaminergic neurons in
the developing brain but spares cholinergic neurons in the P R E VE N T I VE S T R AT E G I E S A N D
basal forebrain.72 T R E AT M E N T
Glia cells also have been found to be sensitive, as evi-
denced from recent observations of robust apoptosis Researchers have identified several compounds and strate-
involving young oligodendroglia following exposure to gies that limit the extent of anesthesia-induced neuroapop-
alcohol73 or anesthetic drugs32,36 in the developing brain tosis in vitro and in vivo experimental models. However, no
of rodent and nonhuman primates. Deletion of young oli- evidence is available from clinical studies.
godendrocytes that are just beginning to myelinate axons, It is accepted by many that the anesthesia-induced cell
which interconnect neurons throughout the developing death morphologically and in terms of temporal manifestation
brain, could potentially have adverse long-term neurobe- is consistent with programmed cell death (apoptosis) and
havioral consequences, which might be additive to those involves mechanisms shown as characteristic for apoptotic
of anesthesia-induced neuroapoptosis.36 It also remains cell death induced by other injury mechanisms.15–17,83–85
unclear whether the window of oligodendroglial vulner- In vitro and in vivo studies have demonstrated that
ability to anesthetic exposure parallels that of neurons or exposure to anesthetics is associated with a reduction of
is further extended, given the ongoing myelination in later elements involved in prosurvival signaling cascades such as
childhood. phosphorylated extracellular signal-regulated protein kinase

6 • N E U R O L O G I C O U TC O M E S O F S U R G E RY A N D A N E S T H E S I A