- Email: [email protected]
Anesthetic Neurotoxicity: New Findings and Future Directions
THE JOURNAL OF PEDIATRICS • www.jpeds.com
COMMENTARY
Anesthetic Neurotoxicity: New Findings and Future Directions Michael C. Montana, MD, PhD1,2, and Alex S. Evers, MD1
T
he development and refinement of practices for the safe administration of anesthesia to children is a major success story in modern medicine. During the past several decades, there have been significant improvements in safety standards, cardiopulmonary monitoring, delivery systems, and airway management specific to the pediatric patient undergoing anesthesia. Millions of children receive anesthesia each year for surgical, procedural, or diagnostic purposes, and the majority of these patients receive a general anesthetic.1 Parents and care providers can be confident that the vast preponderance of these children will have a safe outcome with a low likelihood of major morbidity or mortality.2 The last several decades also have seen the discovery, and subsequent verification, that agents commonly used to induce and maintain general anesthesia in humans exhibit evidence of neurotoxicity in animal models.3 This realization dates to the early 1980s, when exposure of pregnant rat dams to chronic, low-level halothane was found to result in abnormal synaptogenesis and behavior in their offspring.4 Further concerns arose from the discovery in 1998 that the N-methyl-Daspartate (NMDA) receptor antagonist nitrous oxide (also known as laughing gas) can be neurotoxic in rodents.5 It was shown subsequently that when administered to neonatal rats during a period of critical synaptogenesis, a commonly used cocktail of anesthetics (including nitrous oxide, midazolam, and isoflurane) induces immediate widespread neuronal apoptosis and impairments in learning and memory that persist into adulthood.6 Preclinical models from Caenorhabditis elegans to nonhuman primates now suggest that multiple anesthetic agents may be neurotoxic. These include positive allosteric modulators of the GABAA receptor (benzodiazepines, propofol, and the anesthetics isoflurane, sevoflurane, and desflurane), and the NMDA receptor antagonists ketamine and nitrous oxide.3 The discovery of anesthetic neurotoxicity in animal models raises the disconcerting possibility that administration of what appears to be a safe general anesthetic may have long-lasting deleterious neurocognitive effects. These discoveries and concerns, however, represent an inversion of the traditional use of preclinical models to study human diseases. In most circumstances, a human malady is recognized clinically and is sufficiently prevalent or severe that researchers develop animal models to study disease process and to refine diagnostic and therapeutic approaches. Anesthetic neurotoxicity was first discovered in animal models, with the possibility of detriment to human patients arising from that discovery. This atypical knowledge acquisition makes it unclear what neurocognitive or behavioral components comprise the clinical syndrome of anesthetic-induced developmental neuNMDA
N-methyl-D-aspartate
rotoxicity. This uncertainty presents parents, clinicians, and researchers with a conundrum: given the millions of children that undergo general anesthesia for surgical, procedural, and diagnostic purposes each year, anesthetic neurotoxicity, although unproven in human patients, may represent a significant public health problem. Two recently published human studies that suggest a lack of harm in otherwise-healthy children following a short duration anesthetic (approximately 1 hour) deserve early mention. The first of these trials is the General Anaesthesia compared to Spinal anaesthesia (GAS) Trial, which randomized infants undergoing inguinal hernia repair to either an awake-regional technique or a general anesthetic.7 Secondary outcomes assessed at 2 years of age showed no increased risk of adverse neurodevelopment in children exposed to a general anesthetic. The Pediatric Anesthesia & Neurodevelopment Assessment (PANDA) study compared children who had undergone inguinal hernia repair with general anesthesia before 3 years of age with an unexposed sibling.8 No difference in IQ was found between exposed and unexposed siblings. Further details regarding these studies are discussed herein. The results from these trials are encouraging and suggest that a short-duration anesthetic in otherwise-healthy children may have limited effects. Nevertheless, the concerns regarding anesthetic neurotoxicity are myriad and nuanced. This commentary is intended as a review for pediatricians, anesthesiologists, and surgeons of the animal studies that first raised these concerns, the historical context of these studies, and the human studies that are either completed or ongoing.
Neuronal Damage and Behavioral Changes in Animal Models Exposure of rodents to anesthetic agents between postnatal day 7 and 14 is associated with decreased neuronal density, decreased neuronal numbers, and increased neuronal apoptosis.3,6,9 Multiple anesthetics have been implicated, including ketamine, propofol, and halogenated anesthetic gases. A cocktail of nitrous oxide, midazolam, and sevoflurane sufficient to maintain a surgical plane of general anesthesia for 6 hours results in
From the 1School of Medicine, Department of Anesthesiology, Washington University in St. Louis, St. Louis, MO; and 2Saint Louis Children’s Hospital, St. Louis, MO M.M. is supported by the National Institutes of Health/National Institute of General Medical Sciences (NIGMS) Training Program in Anesthesiology (T-32 5T32GM108539-02 [PI: A.E.]). A.E. is supported by NIGMS (RO1GM108799) and by the Taylor Family Institute for Innovative Psychiatric Research. A.E. serves on the Scientific Advisory Board of Neuroprotexeon, a pharmaceutical start-up company. M.M. declares no conflicts of interest. 0022-3476/$ - see front matter. © 2016 Elsevier Inc. All rights reserved. http://dx.doi.org10.1016/j.jpeds.2016.10.049
1 DIS 5.4.0 DTD ■ YMPD8759_proof ■ November 8, 2016
THE JOURNAL OF PEDIATRICS • www.jpeds.com apoptosis in multiple brain regions,6 and even subanesthetic doses of propofol can be apoptogenic.10 Nonhuman primates also are vulnerable to neuronal damage. When administered to postnatal day 5-6 macaques, a combination of inhaled nitrous oxide and isoflurane sufficient to maintain surgical anesthesia for 8 hours produces apoptotic and possibly necrotic neuronal damage in multiple brain regions.11 Evidence for anesthetic neurotoxicity has been found throughout the nervous system, including the hippocampus, striatum, thalamus, amygdala, cerebellum, cerebral cortex, and spinal cord.9 The damage is not limited to neurons; apoptosis also has been observed in oligodendrocytes and other glial cells. In addition to neuronal damage, early-life anesthetic exposure in rodents has been associated with long-term behavioral deficits in spatial memory3,6; however, not all studies that have assessed behavior have found a difference between exposed and control animals.9 Most studies that include behavioral outcomes have been performed in rodents; however, macaques exposed to a single 24-hour ketamine anesthetic also have longlasting impairments in learning and motivation as compared with unexposed controls.12
Volume ■■ (albeit in different brain regions) up to postnatal day 21.3 Synapse formation and elimination is thought to occur approximately between postconceptional days 10 and 60 in rats and days 25 and 1000 in humans.14 There are myriad potential confounds in correlating interspecies brain maturation to exact postconceptual dates, however extending this argument to its logical conclusion suggests that the period of maximal vulnerability in humans actually may be in the second trimester and then last until nearly 3 years of age. Importantly, our goal here is not to use this interspecies comparison to define vulnerable dates in humans but rather to illustrate the complexity in applying animal data to human patients.
The aforementioned studies represent a mere fraction of the nearly 1000 articles related to anesthetic neurotoxicity that have been published (for an excellent review, see Disma et al9). The general consensus from these studies is that anesthetic agents may be neurotoxic; however, there is significant heterogeneity in the anesthetic(s) used, the duration of exposure, the age of the animals studied, the histologic methods by which neuronal damage was assessed, and the behavioral tasks performed.
Duration of Anesthetic Exposure and Use of Multiple Agents The likelihood of apoptosis is influenced by the duration of exposure and the use of multiple vs single anesthetic agents. For example, studies have found no evidence of apoptosis when either nitrous oxide or isoflurane was administered as the sole anesthetic, only when administered in combination.6,11 Additional studies that use isoflurane as the sole anesthetic agent, however, have shown evidence of apoptosis in neurons and oligodendrocytes.6,15-17 The conflict regarding the apoptogenic nature of single-agent anesthetics is not unique to isoflurane. Single doses of ketamine also have been suggested either to cause or not to cause apoptosis.18,19 Studies that show no evidence of neuronal injury are not necessarily reassuring, because the majority of studies performed thus far indicate that the likelihood of neuronal damage increases when repeated doses of a single agent are given, when multiple agents are administered together, and especially when anesthesia is maintained for durations longer than 2-3 hours.
Age of Exposure in Animals vs Humans In mammals, NMDA- and GABA-mediated neuronal activity are important for synaptogenesis during a critical period of brain maturation.13 The sequence of events involved in brain maturation including neurogenesis, synaptogenesis, myelination, and increases in brain weight, along with the development of behavioral milestones, proceed in an orderly and similar fashion in all mammalian species studied. These species include mice, rats, guinea pigs, cats, sheep, nonhuman primates, and humans.14 The postconceptual and postnatal age at which synaptogenesis begins and its duration, however, vary significantly between species. Even within the same species, different brain regions are maximally vulnerable to anesthetic neurotoxicity at different developmental time points.3 This interspecies variation and differential vulnerability underscores an important challenge in conducting and interpreting experiments involving animal models of neurodevelopment, that of determining what postconceptual age in a human patient corresponds to a given postconceptual age in another species, and what brain region(s) may be most vulnerable at that time point. For example, peak vulnerability to anesthetic neurotoxicity in rodents occurs at postnatal day 7 and remains present
Use of Apoptosis to Define Neuronal Toxicity Another challenge in extrapolating data from animal studies to humans lies in the use of immunohistochemical markers of neuronal apoptosis to signify neurotoxicity. Apoptosis during brain maturation is a physiologic process that is necessary for the removal of excess neurons produced during normal development.20 It can be argued that anesthetic exposure results in an exaggerated, pathologic increase in neuronal apoptosis of otherwise-healthy cells,3,6,21 although the possibility exists that the increased apoptosis seen following anesthetic exposure in animals merely represents an acceleration of the death of neurons that would eventually have undergone physiologic pruning.13 Regardless of whether it occurs under physiologic or pathologic conditions, the final steps of the apoptotic cascade include the activation of proteolytic effector caspases.20 Many animal studies of anesthetic neurotoxicity use staining for activated caspase-3 to mark neurons that have passed the point of no return and are committed to cell death; however, caspase-3 staining is only able to reveal neurons in the early stages of apoptosis (approximately 6-12 hours after cascade initiation), as eventual phagocytosis of the dead neurons leaves no substrate for immunoreactivity. Furthermore, neuronal apoptosis may actually be one of the least-sensitive ways to assess altered synaptogenesis. Short durations of anesthetic
Caveats of Animal Models
2
Montana and Evers DIS 5.4.0 DTD ■ YMPD8759_proof ■ November 8, 2016
COMMENTARY
2016 exposure that do not induce neuronal death can still affect the density of dendritic spines,22 which may have a persistently measurable phenotype in the absence of neuronal cell death. In future studies, magnetic resonance imaging and positron emission tomography may provide data on changes in brain structure, gray matter and axonal density, and surrogate markers for brain inflammation in both humans and animals23,24; to date, a definitive phenotypic signature for late and persistent neurodegenerative changes has not been accepted universally. Use of Animal Behavioral Tasks as Surrogates for Human Cognition Although it is accepted generally that exposure to anesthetics during vulnerable periods results in neurodegenerative histopathologic changes, how much this affects the animal on an organismal level is less certain. As discussed, neurocognitive deficits have been found in adult animals exposed to anesthetics as neonates; however, there is some question as to how much the most commonly used rodent tasks, including locomotor activity, and assessments of spatial learning and memory, actually apply to human children.6,9,25 Studies of anesthetic neurotoxicity in nonhuman primates, which are more phylogenetically related to humans, may allow future preclinical models to be more directly applicable to our species. The National Center for Toxicological Research has used the Operant Test Battery to assess the effects of nonanesthetic drug exposure on monkeys for several decades, and the behavior of sufficiently trained rhesus monkeys can be comparable with that of children.26 The use of this battery in preclinical models of anesthetic neurotoxicity may therefore allow for a closer comparison with human subjects. Lack of Surgical Stimuli and Physiologic Monitoring in Animal Models Studies on anesthetic neurotoxicity in many animal models are performed in a manner inconsistent with the modern operating suite and in the absence of surgical stimuli. The stressors of surgery themselves may be detrimental to neurodevelopment, and there is evidence that anesthesia and analgesia blunt these responses. For example, preterm infants administered fentanyl while undergoing patent ductus ligation have demonstrably improved postoperative outcomes compared with infants to whom opioid analgesia is not administered.27 In addition, neonates who received deep intraoperative anesthesia have improved hormonal and metabolic responses to surgery and lower mortality than infants anesthetized with a lighter technique.28 It should be noted that opioids were used for analgesia in both of these human studies and that, despite popular belief, halogenated inhaled anesthetics do not provide any significant measure of analgesia. There is also concern that surgical stressors may actually worsen anesthetic neurotoxicity. Studies performed in rats in the absence of opioid analgesia suggest that the combination of surgical stressors and inhaled anesthesia may be more detrimental than either alone. Both chemically induced and surgically induced pain exacerbated the immediate apoptosis and subsequently delayed cognitive impairment of a combined
nitrous oxide/isoflurane anesthetic in postnatal day 7 rats.29 Surgery without anesthesia is clearly detrimental, and it is entirely possible that surgical stressors without adequate analgesia may exacerbate anesthetic neurotoxicity. What specific role adequate analgesia may play in blunting these effects remains to be elucidated. Physiologic monitoring of human subjects undergoing anesthesia and surgery in the modern era also includes continuous monitoring of oxygenation, ventilation, temperature, and circulation. Human patients can be assessed rigorously for hypoglycemia and electrolyte imbalances preoperatively, intraoperatively, and postoperatively. This detailed level of physiologic monitoring generally is not performed in preclinical testing of anesthetic neurotoxicity, although there are notable exceptions.9 It is possible that unrecognized deviations from physiologic norms (eg, hypoxia, hypoglycemia, hypotension) may compound or confound evidence for, or against, anesthetic neurotoxicity. Recently conducted studies have moved toward full physiologic monitoring and improved maintenance of homeostatic variables, including endotracheal intubation and mechanical ventilation, between exposed and control groups.17,30 These efforts should be applauded and continued in future studies so as to obtain data more relevant to human subjects.
Conclusions from Animal Models When administered to neonatal animals, multiple anesthetic agents commonly used in clinical practice have been shown to cause both neuronal damage and long-term behavioral changes. These effects have been repeated by multiple research groups and in multiple species, including nonhuman primates. There is good evidence that long-duration exposures, simultaneous exposures to multiple anesthetic agents, and repeated anesthetic exposures cause histologic damage. The degree of histologic damage appears to correlate with observed behavioral changes; however, evidence for neuronal damage, and especially for long-term behavioral changes, following short-duration and single-agent exposures is less robust. The question of how applicable animal studies are to human subjects also remains, and before focusing on the risks that anesthetics may pose to human patients, one must remember that many surgeries and procedures performed on children each year are demonstrably beneficial, and we do not have viable alternatives to the anesthetics currently in clinical use that are necessary to perform these procedures. These caveats aside, we feel that the data from animal studies raise and reaffirm the concern that anesthetics, although necessary for the safe conduct of a multitude of surgeries, may in fact be neurotoxic, especially when given for long durations and at high doses.
Data from Observational Human Studies Before 2016, data from human studies on the topic of anesthetic neurotoxicity had been limited to retrospective and observational studies. Several of these studies deserve mention
Anesthetic Neurotoxicity: New Findings and Future Directions DIS 5.4.0 DTD ■ YMPD8759_proof ■ November 8, 2016
3
THE JOURNAL OF PEDIATRICS • www.jpeds.com (Table). A 2011 matched cohort study of more than 8000 children born between January 1, 1976, and December 31, 1982, in Rochester, Minnesota, included 350 children exposed to anesthesia before age 2 years. Children who were exposed to multiple, but not single, anesthetics/surgical procedures had a significantly greater incidence of learning disabilities compared with unexposed controls, even when accounting for general health.31 The Western Australian Pregnancy Cohort (Raine) Study of nearly 3000 children born between 1989 and 1991 also has been used to assess for cognitive changes following anesthesia. In this cohort, 321 children were exposed to anesthesia before age 3 years. At 10 years of age, exposed children were administered individually neuropsychological tests to assess a variety of specific behavioral modalities, something the Mayo study was not able to do. Assessed retrospectively, children in the Raine cohort demonstrated deficits in language and abstract reasoning at age 10 years even following a single exposure; however, there were no significant differences observed in assessments of fine or gross motor function, or in a standardized questionnaire completed by parents for the purposes of evaluating behavioral problems.32 An additional retrospective study of a subset of the Raine cohort found an increased incidence of deficits in individual language assessments at 10 years of age in children who had anesthesia and surgery before age 3 years. In exposed children, there also was an increased incidence of International Classification of Diseases, Ninth Revision codes for language and cognitive disorders. No differences, however, were found when exposed vs unexposed children were assessed with a group test of academic achievement.33 A study of Dutch twins that used a group-administered assessment of academic achievement as its primary outcome also failed to show any differences at age 12 years between twin pairs discordant for anesthesia administration before age 3 years (ie, when one twin received an anesthetic and the other did not).34 Interestingly in this study, twin pairs in which both twins were exposed to anesthesia before age 3 years were found to have significantly lower educational achievement scores and more cognitive deficits compared with unexposed twin pairs. The lack of difference between the discordant twins, however, does not support a relationship between anesthesia and later-life learning deficits. It has been noted previously that studies of potential anesthetic neurotoxicity relying on broader measures of cognition such as academic achievement tend to be negative, whereas those that perform individualized assessments of specific modalities are more likely to be positive.35 Reflecting what is seen in animal models, the individual modality assessed is important in interpreting these studies. This is illustrated by a 2016 study from the Raine cohort that found no differences in visual acuity, refractive error, or thickness of the retinal nerve fiber between children exposed to anesthesia before age 3 years and unexposed individuals.36
Data from Prospective Trials Unfortunately, data from animal studies and retrospective observational human studies do not provide definitive evi-
Volume ■■ dence for or against anesthetic neurotoxicity. These studies have generated sufficient concern to warrant large-scale, randomized clinical trials to assess whether general anesthetics impair neurocognitive development in human subjects. Results from the first of these studies recently have been published. An observer-blinded, international, multisite, randomized, controlled, equivalence study comparing 294 infants who received general anesthesia with 238 infants who received awake regional anesthesia for inguinal hernia repair found no differences in a prespecified secondary neurodevelopmental outcomes assessment between the 2 groups at 2 years of age.7 Although primary outcome assessments will be performed when children are 5 years old, these initial results represent the first data from a randomized, controlled trial in human patients designed to assess for developmental anesthetic neurotoxicity. In this study, only an awake regional technique or a sevoflurane inhalational general anesthetic was permitted. In the awake regional group, the use of any pharmacologic supplemental sedation was considered a protocol violation. At age 2 years, children were assessed by both an as-per-protocol and an intention-to-treat analysis in 5 distinct scales of developmental function, including cognitive, communication, physical, social/emotional, and adaptive measures. The mean duration of sevoflurane administration was only 54 minutes, and patient recruitment was restricted to relatively healthy infants. No evidence for adverse neurodevelopmental outcomes was found between the groups that received an awake regional or a general anesthetic. A similar ambidirectional sibling-matched cohort study of 105 healthy sibling pairs presenting for elective inguinal hernia repair also was completed recently.8 In this study, exposed subjects who had received a single anesthetic exposure before age 36 months for inguinal hernia repair were identified retrospectively. Neurocognitive and behavioral outcomes were then assessed prospectively by the use of unexposed siblings whose age was within 36 months of the exposed sibling as controls. Like the awake regional vs general anesthesia study previously discussed, the duration of anesthesia in exposed subjects was relatively short, with a median duration of 80 minutes (range 20-240 minutes). Like the previously discussed study, the subjects were all relatively healthy. Also like the previous study, there were no significant differences found between exposed and unexposed siblings in either the primary outcome of IQ scores or secondary domain-specific neuropsychological outcome tests.
Conclusions and Questions for Future Study Taken at face value, the recently published results from these 2 controlled clinical studies would seem reassuring, and in fact they are, but only as far as the results apply to relatively healthy patients receiving a short-duration anesthetic. Data from preclinical models have not demonstrated conclusively that shortduration, or single-agent, anesthetic exposures consistently cause neurodegeneration, much less significant behavioral effects. Findings from animal models are therefore consistent with
4
Montana and Evers DIS 5.4.0 DTD ■ YMPD8759_proof ■ November 8, 2016
2016
DIS 5.4.0 DTD ■ YMPD8759_proof ■ November 8, 2016
Anesthetic Neurotoxicity: New Findings and Future Directions
Table. Findings from selected human studies on anesthesia-related neurocognitive assessments Year
Study design
Anesthesia Exposure
Bartels et al34
2009
Retrospective monozygotic concordant-discordant twin study
Exposure to anesthesia before age 3 y was determined based on parent surveys.
A total of 1143 monozygotic twin pairs
Population studied
Davidson et al7
2016
Assessor-masked randomized controlled equivalence trial
Randomized to receive either awake regional or general anesthesia.
Flick et al31
2011
Matched cohort study
A procedure requiring general anesthesia before 2 y of age. Median anesthetic exposure was 75 min.
Ing et al32
2012
Retrospective cohort study
Determined based on whether children had undergone a surgical procedure that requires anesthesia.
Ing et al33
2014
Retrospective cohort study
Determined based on whether children had undergone a surgical procedure that requires anesthesia.
Infants < 60 wk postmenstrual age having inguinal hernia repair. A total of 238 children in the awake regional and 294 in the general anesthesia group. A total of 350 children exposed to anesthesia and 700 matched unexposed controls (2 controls per case). All children were from Olmsted County, Minnesota. Western Australian Pregnancy Cohort (Raine) Study. 321 children who had a surgical procedure before age 3 y were compared with 2287 who did not. A total of 781 children from the Western Australian Pregnancy Cohort (Raine) Study; 112 children had anesthesia exposure before age 3 y.
Sun et al8
2016
Ambidirectional sibling matched cohort
Median duration of 80 min of inhaled anesthesia. Twentyeight children also received IV agents.
Yazar et al36
2016
Cohort study
Determined based on whether children had undergone a surgical procedure that requires anesthesia.
A total of 105 children exposed to anesthesia before age 3 y compared with unexposed siblings within 36 mo of age; 18 exposed and 23 unexposed children also received anesthesia after 36 mo. Western Australian Pregnancy Cohort (Raine) Study. A total of 127 children who had a surgical procedure before age 3 y were compared with 707 who did not.
Outcomes measured
Findings
Comments
EA assessed by Dutch-CITO elementary standardized testing. CP assessed by Connors' Teacher Rating Scale. Secondary outcomes assessment at 2 y of age using the composite cognitive score from the Bayley Scale of Infant and Toddler Development III. Need for an IEP; presence of learning disabilities in reading, writing, or math; and group administered tests of achievement and cognition.
Twins exposed to anesthesia before age 3 y had lower EA and more CPs, however discordant twins did not differ from each other. No differences in the composite cognitive scale between the 2 groups.
Specific anesthetics used and duration of anesthesia were not known to the authors.
Neuropsychological tests were performed at 10 y of age and designed to assess language, listening, cognitive function, motor skills, and behavior.
Children exposed to anesthesia before age 3 y had an increased incidence of deficits in language and abstract reasoning.
Authors did not have direct access to medical records to determine specific anesthetics used, dose, or duration.
Neuropsychological tests were performed at 10 years of age. ICD-9–coded mental, behavioral, and neurodevelopment disorders were extracted from followup visits. Academic performance on a statewide fifth grade achievement test was assessed. Global cognitive function (IQ) assessed at 10 y of age was the primary outcome. Secondary outcomes included domain-specific neurodevelopment tests.
Exposure to anesthesia was associated with more deficits in individual language assessments and ICD-9 codes for language disorders. No difference in academic achievement testing was seen.
Authors did not have direct access to medical records to determine specific anesthetics used, dose, or duration.
No differences were seen between exposed and unexposed siblings in either the primary or secondary outcomes.
More than 75% of parents of test subjects had completed at least 2 y of college and more than 90% of parents were married.
At age 20 y, a comprehensive ophthalmologic examination including visual acuity, refractive error, and optic disc assessment was performed.
No association between anesthesia exposure and myopia, reduced visual acuity, or retinal nerve thinning.
Authors did not have direct access to medical records to determine specific anesthetics used, dose, or duration.
CP, cognitive problems; EA, educational achievement; ICD-9, International Classification of Diseases, Ninth Revision; IEP, individualized education plan; IV, intravenous.
Median anesthetic duration was 54 min. Primary outcomes assessment is planned for 5 y of age.
Children who had 2 or more anesthetics, but not one, were more likely to have learning disabilities, and an IEP for speech and language.
5
COMMENTARY
Authors
THE JOURNAL OF PEDIATRICS • www.jpeds.com findings from recently published human studies, namely that short-duration, single-agent anesthetics likely do not pose a significant risk to relatively healthy children. A sobering counterpoint to this conclusion is the possibility that longer duration, or repetitive, anesthetic exposures may pose a risk for neurocognitive impairment in human subjects. Healthy children generally do not require multiple anesthetics; thus, it will be extremely challenging for future studies to tease out whether any observed neurocognitive deficits are due to the anesthetic regimen or the pathophysiologic background of patients that require the anesthetic and the associated procedures. Thankfully, there is a way forward. Limitation, mitigation, and alternative neuroprotective strategies are being tested that may reduce the risk of anesthetic neurotoxicity following long exposures. In rats the inert gas xenon, which is an NMDA antagonist and anesthetic, has been found not be apoptogenic and to limit isoflurane-induced apoptosis.37 Similarly, the finding that the alpha 2 -adrenergic receptor agonist dexmedetomidine reduces isoflurane-induced caspase-3 activation in postnatal day 7 rats suggests that it may be a neuroprotective sedating agent.38 In mice the AMPAkine CX546, which potentiates a-amino-3-hydroxy-5-methyl-4isoxazolepropionic acid activity during emergence from anesthesia, recently was shown to diminish synaptic changes and behavioral deficits following ketamine anesthesia.39 In nonhuman primates, coadministration of a lithium infusion protects against isoflurane-induced neuronal apoptosis.40 In humans, the T REX Pilot Study currently is recruiting to test the feasibility of using a combination of dexmedetomidine, opioids, and regional nerve blocks as an alternative technique to gaseous inhaled anesthetics (ClinicalTrials.gov: NCT02353182). Further research is required to determine whether these strategies are neuroprotective in long-duration anesthetics. Studies in animals designed to test these strategies must be rigorous and use appropriate physiologic monitoring. If viable alternative anesthetic strategies are found, they should be used to inform the design of randomized, controlled clinical trials. Additional relevant questions remain. Are specific patient populations more at risk than others? Are patients who undergo specific surgical procedures more likely than others to have neurocognitive deficits? Are infants born premature who require sedation or anesthesia more vulnerable than infants born at term? Can diagnostic imaging modalities or biomarkers noninvasively diagnose neurotoxicity? If cumulative or repeat dosing increases the risk for neurotoxicity in humans, can we define cut-offs below which detectable effects are unlikely? Even though gaseous inhaled anesthetics are used almost exclusively in the operating room or diagnostic/procedural suite, thousands of infants and children worldwide in intensive care unit settings are sedated with other agents, including benzodiazepines and opioids, to facilitate life-saving practices such as mechanical ventilation. Are these neonatal intensive care unit and pediatric intensive care unit patients at risk for sedative neurotoxicity following prolonged sedation regimens? Definitive determination of the risks that anesthetics and sedatives may pose to the developing human brain will require a con-
Volume ■■ certed global effort and coordination between parents, scientists, and healthcare providers. The task will not be easy and will require years to complete; however, we owe it to our youngest and most vulnerable patients, as well as their parents, and their caregivers, to answer these questions. ■ Submitted for publication Jul 22, 2016; last revision received Sep 9, 2016; accepted Oct 17, 2016 Reprint requests: Alex S. Evers, MD, Washington University School of Medicine, Washington University Medical Center, Campus Box 8054; 660 South Euclid Ave, St Louis, MO 63110-1093. E-mail: [email protected]
References 1. Wanderer JP, Rathmell JP. Complex information for anesthesiologists presented quickly and clearly. Anesthesiology 2014;120:A23. 2. van der Griend BF, Lister NA, McKenzie IM, Martin N, Ragg PG, Sheppard SJ, et al. Postoperative mortality in children after 101,885 anesthetics at a tertiary pediatric hospital. Anesth Analg 2011;112:1440-7. 3. Jevtovic-Todorovic V, Absalom AR, Blomgren K, Brambrink A, Crosby G, Culley DJ, et al. Anaesthetic neurotoxicity and neuroplasticity: an expert group report and statement based on the BJA Salzburg Seminar. Br J Anaesth 2013;111:143-51. 4. Uemura E, Levin ED, Bowman RE. Effects of halothane on synaptogenesis and learning behavior in rats. Exp Neurol 1985;89:520-9. 5. Jevtovic-Todorovic V, Todorovic SM, Mennerick S, Powell S, Dikranian K, Benshoff N, et al. Nitrous oxide (laughing gas) is an NMDA antagonist, neuroprotectant and neurotoxin. Nat Med 1998;4:460-3. 6. Jevtovic-Todorovic V, Hartman RE, Izumi Y, Benshoff ND, Dikranian K, Zorumski CF, et al. Early exposure to common anesthetic agents causes widespread neurodegeneration in the developing rat brain and persistent learning deficits. J Neurosci 2003;23:876-82. 7. Davidson AJ, Disma N, de Graaff JC, Withington DE, Dorris L, Bell G, et al. Neurodevelopmental outcome at 2 years of age after general anaesthesia and awake-regional anaesthesia in infancy (GAS): an international multicentre, randomised controlled trial. Lancet 2016;387:23950. 8. Sun LS, Li G, Miller TL, Salorio C, Byrne MW, Bellinger DC, et al. Association between a single general anesthesia exposure before age 36 months and neurocognitive outcomes in later childhood. JAMA 2016;315:231220. 9. Disma N, Mondardini MC, Terrando N, Absalom AR, Bilotta F. A systematic review of methodology applied during preclinical anesthetic neurotoxicity studies: important issues and lessons relevant to the design of future clinical research. Paediatr Anaesth 2016;26:6-36. 10. Cattano D, Young C, Straiko MM, Olney JW. Subanesthetic doses of propofol induce neuroapoptosis in the infant mouse brain. Anesth Analg 2008;106:1712-4. 11. Zou X, Liu F, Zhang X, Patterson TA, Callicott R, Liu S, et al. Inhalation anesthetic- induced neuronal damage in the developing rhesus monkey. Neurotoxicol Teratol 2011;33:592-7. 12. Paule MG, Li M, Allen RR, Liu F, Zou X, Hotchkiss C, et al. Ketamine anesthesia during the first week of life can cause long-lasting cognitive deficits in rhesus monkeys. Neurotoxicol Teratol 2011;33:220-30. 13. Loepke AW, McGowan FX Jr, Soriano SG. CON: the toxic effects of anesthetics in the developing brain: the clinical perspective. Anesth Analg 2008;106:1664-9. 14. Workman AD, Charvet CJ, Clancy B, Darlington RB, Finlay BL. Modeling transformations of neurodevelopmental sequences across mammalian species. J Neurosci 2013;33:7368-83. 15. Creeley CE, Dikranian KT, Dissen GA, Back SA, Olney JW, Brambrink AM. Isoflurane- induced apoptosis of neurons and oligodendrocytes in the fetal rhesus macaque brain. Anesthesiology 2014;120:626-38. 16. Brambrink AM, Back SA, Riddle A, Gong X, Moravec MD, Dissen GA, et al. Isoflurane-induced apoptosis of oligodendrocytes in the neonatal primate brain. Ann Neurol 2012;72:525-35.
6
Montana and Evers DIS 5.4.0 DTD ■ YMPD8759_proof ■ November 8, 2016
COMMENTARY
2016 17. Brambrink AM, Evers AS, Avidan MS, Farber NB, Smith DJ, Zhang X, et al. Isoflurane-induced neuroapoptosis in the neonatal rhesus macaque brain. Anesthesiology 2010;112:834-41. 18. Rudin M, Ben-Abraham R, Gazit V, Tendler Y, Tashlykov V, Katz Y. Singledose ketamine administration induces apoptosis in neonatal mouse brain. J Basic Clin Physiol Pharmacol 2005;16:231-43. 19. Scallet AC, Schmued LC, Slikker W Jr, Grunberg N, Faustino PJ, Davis H, et al. Developmental neurotoxicity of ketamine: morphometric confirmation, exposure parameters, and multiple fluorescent labeling of apoptotic neurons. Toxicol Sci 2004;81:364-70. 20. Dekkers MP, Nikoletopoulou V, Barde YA. Cell biology in neuroscience: death of developing neurons: new insights and implications for connectivity. J Cell Biol 2013;203:385-93. 21. Jevtovic-Todorovic V, Olney JW. PRO: anesthesia-induced developmental neuroapoptosis: status of the evidence. Anesth Analg 2008;106:165963. 22. Briner A, De Roo M, Dayer A, Muller D, Habre W, Vutskits L. Volatile anesthetics rapidly increase dendritic spine density in the rat medial prefrontal cortex during synaptogenesis. Anesthesiology 2010;112:546-56. 23. Backeljauw B, Holland SK, Altaye M, Loepke AW. Cognition and brain structure following early childhood surgery with anesthesia. Pediatrics 2015;136:e1-12. 24. Zhang X, Liu S, Newport GD, Paule MG, Callicott R, Thompson J, et al. In vivo monitoring of sevoflurane-induced adverse effects in neonatal nonhuman primates using small-animal positron emission tomography. Anesthesiology 2016;125:133-46. 25. Loepke AW, Istaphanous GK, McAuliffe JJ 3rd, Miles L, Hughes EA, McCann JC, et al. The effects of neonatal isoflurane exposure in mice on brain cell viability, adult behavior, learning, and memory. Anesth Analg 2009;108:90-104. 26. Paule MG, Green L, Myerson J, Alvarado M, Bachevalier J, Schneider JS, et al. Behavioral toxicology of cognition: extrapolation from experimental animal models to humans: behavioral toxicology symposium overview. Neurotoxicol Teratol 2012;34:263-73. 27. Anand KJ, Sippell WG, Aynsley-Green A. Randomised trial of fentanyl anaesthesia in preterm babies undergoing surgery: effects on the stress response. Lancet 1987;1:62-6. 28. Anand KJ, Hickey PR. Halothane-morphine compared with high-dose sufentanil for anesthesia and postoperative analgesia in neonatal cardiac surgery. N Engl J Med 1992;326:1-9.
29. Shu Y, Zhou Z, Wan Y, Sanders RD, Li M, Pac-Soo CK, et al. Nociceptive stimuli enhance anesthetic-induced neuroapoptosis in the rat developing brain. Neurobiol Dis 2012;45:743-50. 30. Whitaker EE, Bissonnette B, Miller AD, Koppert TL, Tobias JD, Pierson CR, et al. A novel, clinically relevant use of a piglet model to study the effects of anesthetics on the developing brain. Clin Transl Med 2016;5:2. 31. Flick RP, Katusic SK, Colligan RC, Wilder RT, Voigt RG, Olson MD, et al. Cognitive and behavioral outcomes after early exposure to anesthesia and surgery. Pediatrics 2011;128:e1053-61. 32. Ing C, DiMaggio C, Whitehouse A, Hegarty MK, Brady J, von UngernSternberg BS, et al. Long-term differences in language and cognitive function after childhood exposure to anesthesia. Pediatrics 2012;130:e47685. 33. Ing CH, DiMaggio CJ, Whitehouse AJ, Hegarty MK, Sun M, von UngernSternberg BS, et al. Neurodevelopmental outcomes after initial childhood anesthetic exposure between ages 3 and 10 years. J Neurosurg Anesthesiol 2014;26:377-86. 34. Bartels M, Althoff RR, Boomsma DI. Anesthesia and cognitive performance in children: no evidence for a causal relationship. Twin Res Hum Genet 2009;12:246-53. 35. Flick RP, Nemergut ME, Christensen K, Hansen TG. Anesthetic-related neurotoxicity in the young and outcome measures: the devil is in the details. Anesthesiology 2014;120:1303-5. 36. Yazar S, Hewitt AW, Forward H, Jacques A, Ing C, von Ungern-Sternberg BS, et al. Early anesthesia exposure and the effect on visual acuity, refractive error, and retinal nerve fiber layer thickness of young adults. J Pediatr 2016;169:256-9, e1. 37. Ma D, Williamson P, Januszewski A, Nogaro MC, Hossain M, Ong LP, et al. Xenon mitigates isoflurane-induced neuronal apoptosis in the developing rodent brain. Anesthesiology 2007;106:746-53. 38. Sanders RD, Sun P, Patel S, Li M, Maze M, Ma D. Dexmedetomidine provides cortical neuroprotection: impact on anaesthetic-induced neuroapoptosis in the rat developing brain. Acta Anaesthesiol Scand 2010;54:710-6. 39. Huang L, Cichon J, Ninan I, Yang G. Post-anesthesia AMPA receptor potentiation prevents anesthesia-induced learning and synaptic deficits. Sci Transl Med 2016;8:344ra85. 40. Noguchi KK, Johnson SA, Kristich LE, Martin LD, Dissen GA, Olsen EA, et al. Lithium protects against anaesthesia neurotoxicity in the infant primate brain. Sci Rep 2016;6:22427.
Anesthetic Neurotoxicity: New Findings and Future Directions
7
DIS 5.4.0 DTD ■ YMPD8759_proof ■ November 8, 2016
COMMENTARY
Anesthetic Neurotoxicity: New Findings and Future Directions Michael C. Montana, MD, PhD1,2, and Alex S. Evers, MD1
T
he development and refinement of practices for the safe administration of anesthesia to children is a major success story in modern medicine. During the past several decades, there have been significant improvements in safety standards, cardiopulmonary monitoring, delivery systems, and airway management specific to the pediatric patient undergoing anesthesia. Millions of children receive anesthesia each year for surgical, procedural, or diagnostic purposes, and the majority of these patients receive a general anesthetic.1 Parents and care providers can be confident that the vast preponderance of these children will have a safe outcome with a low likelihood of major morbidity or mortality.2 The last several decades also have seen the discovery, and subsequent verification, that agents commonly used to induce and maintain general anesthesia in humans exhibit evidence of neurotoxicity in animal models.3 This realization dates to the early 1980s, when exposure of pregnant rat dams to chronic, low-level halothane was found to result in abnormal synaptogenesis and behavior in their offspring.4 Further concerns arose from the discovery in 1998 that the N-methyl-Daspartate (NMDA) receptor antagonist nitrous oxide (also known as laughing gas) can be neurotoxic in rodents.5 It was shown subsequently that when administered to neonatal rats during a period of critical synaptogenesis, a commonly used cocktail of anesthetics (including nitrous oxide, midazolam, and isoflurane) induces immediate widespread neuronal apoptosis and impairments in learning and memory that persist into adulthood.6 Preclinical models from Caenorhabditis elegans to nonhuman primates now suggest that multiple anesthetic agents may be neurotoxic. These include positive allosteric modulators of the GABAA receptor (benzodiazepines, propofol, and the anesthetics isoflurane, sevoflurane, and desflurane), and the NMDA receptor antagonists ketamine and nitrous oxide.3 The discovery of anesthetic neurotoxicity in animal models raises the disconcerting possibility that administration of what appears to be a safe general anesthetic may have long-lasting deleterious neurocognitive effects. These discoveries and concerns, however, represent an inversion of the traditional use of preclinical models to study human diseases. In most circumstances, a human malady is recognized clinically and is sufficiently prevalent or severe that researchers develop animal models to study disease process and to refine diagnostic and therapeutic approaches. Anesthetic neurotoxicity was first discovered in animal models, with the possibility of detriment to human patients arising from that discovery. This atypical knowledge acquisition makes it unclear what neurocognitive or behavioral components comprise the clinical syndrome of anesthetic-induced developmental neuNMDA
N-methyl-D-aspartate
rotoxicity. This uncertainty presents parents, clinicians, and researchers with a conundrum: given the millions of children that undergo general anesthesia for surgical, procedural, and diagnostic purposes each year, anesthetic neurotoxicity, although unproven in human patients, may represent a significant public health problem. Two recently published human studies that suggest a lack of harm in otherwise-healthy children following a short duration anesthetic (approximately 1 hour) deserve early mention. The first of these trials is the General Anaesthesia compared to Spinal anaesthesia (GAS) Trial, which randomized infants undergoing inguinal hernia repair to either an awake-regional technique or a general anesthetic.7 Secondary outcomes assessed at 2 years of age showed no increased risk of adverse neurodevelopment in children exposed to a general anesthetic. The Pediatric Anesthesia & Neurodevelopment Assessment (PANDA) study compared children who had undergone inguinal hernia repair with general anesthesia before 3 years of age with an unexposed sibling.8 No difference in IQ was found between exposed and unexposed siblings. Further details regarding these studies are discussed herein. The results from these trials are encouraging and suggest that a short-duration anesthetic in otherwise-healthy children may have limited effects. Nevertheless, the concerns regarding anesthetic neurotoxicity are myriad and nuanced. This commentary is intended as a review for pediatricians, anesthesiologists, and surgeons of the animal studies that first raised these concerns, the historical context of these studies, and the human studies that are either completed or ongoing.
Neuronal Damage and Behavioral Changes in Animal Models Exposure of rodents to anesthetic agents between postnatal day 7 and 14 is associated with decreased neuronal density, decreased neuronal numbers, and increased neuronal apoptosis.3,6,9 Multiple anesthetics have been implicated, including ketamine, propofol, and halogenated anesthetic gases. A cocktail of nitrous oxide, midazolam, and sevoflurane sufficient to maintain a surgical plane of general anesthesia for 6 hours results in
From the 1School of Medicine, Department of Anesthesiology, Washington University in St. Louis, St. Louis, MO; and 2Saint Louis Children’s Hospital, St. Louis, MO M.M. is supported by the National Institutes of Health/National Institute of General Medical Sciences (NIGMS) Training Program in Anesthesiology (T-32 5T32GM108539-02 [PI: A.E.]). A.E. is supported by NIGMS (RO1GM108799) and by the Taylor Family Institute for Innovative Psychiatric Research. A.E. serves on the Scientific Advisory Board of Neuroprotexeon, a pharmaceutical start-up company. M.M. declares no conflicts of interest. 0022-3476/$ - see front matter. © 2016 Elsevier Inc. All rights reserved. http://dx.doi.org10.1016/j.jpeds.2016.10.049
1 DIS 5.4.0 DTD ■ YMPD8759_proof ■ November 8, 2016
THE JOURNAL OF PEDIATRICS • www.jpeds.com apoptosis in multiple brain regions,6 and even subanesthetic doses of propofol can be apoptogenic.10 Nonhuman primates also are vulnerable to neuronal damage. When administered to postnatal day 5-6 macaques, a combination of inhaled nitrous oxide and isoflurane sufficient to maintain surgical anesthesia for 8 hours produces apoptotic and possibly necrotic neuronal damage in multiple brain regions.11 Evidence for anesthetic neurotoxicity has been found throughout the nervous system, including the hippocampus, striatum, thalamus, amygdala, cerebellum, cerebral cortex, and spinal cord.9 The damage is not limited to neurons; apoptosis also has been observed in oligodendrocytes and other glial cells. In addition to neuronal damage, early-life anesthetic exposure in rodents has been associated with long-term behavioral deficits in spatial memory3,6; however, not all studies that have assessed behavior have found a difference between exposed and control animals.9 Most studies that include behavioral outcomes have been performed in rodents; however, macaques exposed to a single 24-hour ketamine anesthetic also have longlasting impairments in learning and motivation as compared with unexposed controls.12
Volume ■■ (albeit in different brain regions) up to postnatal day 21.3 Synapse formation and elimination is thought to occur approximately between postconceptional days 10 and 60 in rats and days 25 and 1000 in humans.14 There are myriad potential confounds in correlating interspecies brain maturation to exact postconceptual dates, however extending this argument to its logical conclusion suggests that the period of maximal vulnerability in humans actually may be in the second trimester and then last until nearly 3 years of age. Importantly, our goal here is not to use this interspecies comparison to define vulnerable dates in humans but rather to illustrate the complexity in applying animal data to human patients.
The aforementioned studies represent a mere fraction of the nearly 1000 articles related to anesthetic neurotoxicity that have been published (for an excellent review, see Disma et al9). The general consensus from these studies is that anesthetic agents may be neurotoxic; however, there is significant heterogeneity in the anesthetic(s) used, the duration of exposure, the age of the animals studied, the histologic methods by which neuronal damage was assessed, and the behavioral tasks performed.
Duration of Anesthetic Exposure and Use of Multiple Agents The likelihood of apoptosis is influenced by the duration of exposure and the use of multiple vs single anesthetic agents. For example, studies have found no evidence of apoptosis when either nitrous oxide or isoflurane was administered as the sole anesthetic, only when administered in combination.6,11 Additional studies that use isoflurane as the sole anesthetic agent, however, have shown evidence of apoptosis in neurons and oligodendrocytes.6,15-17 The conflict regarding the apoptogenic nature of single-agent anesthetics is not unique to isoflurane. Single doses of ketamine also have been suggested either to cause or not to cause apoptosis.18,19 Studies that show no evidence of neuronal injury are not necessarily reassuring, because the majority of studies performed thus far indicate that the likelihood of neuronal damage increases when repeated doses of a single agent are given, when multiple agents are administered together, and especially when anesthesia is maintained for durations longer than 2-3 hours.
Age of Exposure in Animals vs Humans In mammals, NMDA- and GABA-mediated neuronal activity are important for synaptogenesis during a critical period of brain maturation.13 The sequence of events involved in brain maturation including neurogenesis, synaptogenesis, myelination, and increases in brain weight, along with the development of behavioral milestones, proceed in an orderly and similar fashion in all mammalian species studied. These species include mice, rats, guinea pigs, cats, sheep, nonhuman primates, and humans.14 The postconceptual and postnatal age at which synaptogenesis begins and its duration, however, vary significantly between species. Even within the same species, different brain regions are maximally vulnerable to anesthetic neurotoxicity at different developmental time points.3 This interspecies variation and differential vulnerability underscores an important challenge in conducting and interpreting experiments involving animal models of neurodevelopment, that of determining what postconceptual age in a human patient corresponds to a given postconceptual age in another species, and what brain region(s) may be most vulnerable at that time point. For example, peak vulnerability to anesthetic neurotoxicity in rodents occurs at postnatal day 7 and remains present
Use of Apoptosis to Define Neuronal Toxicity Another challenge in extrapolating data from animal studies to humans lies in the use of immunohistochemical markers of neuronal apoptosis to signify neurotoxicity. Apoptosis during brain maturation is a physiologic process that is necessary for the removal of excess neurons produced during normal development.20 It can be argued that anesthetic exposure results in an exaggerated, pathologic increase in neuronal apoptosis of otherwise-healthy cells,3,6,21 although the possibility exists that the increased apoptosis seen following anesthetic exposure in animals merely represents an acceleration of the death of neurons that would eventually have undergone physiologic pruning.13 Regardless of whether it occurs under physiologic or pathologic conditions, the final steps of the apoptotic cascade include the activation of proteolytic effector caspases.20 Many animal studies of anesthetic neurotoxicity use staining for activated caspase-3 to mark neurons that have passed the point of no return and are committed to cell death; however, caspase-3 staining is only able to reveal neurons in the early stages of apoptosis (approximately 6-12 hours after cascade initiation), as eventual phagocytosis of the dead neurons leaves no substrate for immunoreactivity. Furthermore, neuronal apoptosis may actually be one of the least-sensitive ways to assess altered synaptogenesis. Short durations of anesthetic
Caveats of Animal Models
2
Montana and Evers DIS 5.4.0 DTD ■ YMPD8759_proof ■ November 8, 2016
COMMENTARY
2016 exposure that do not induce neuronal death can still affect the density of dendritic spines,22 which may have a persistently measurable phenotype in the absence of neuronal cell death. In future studies, magnetic resonance imaging and positron emission tomography may provide data on changes in brain structure, gray matter and axonal density, and surrogate markers for brain inflammation in both humans and animals23,24; to date, a definitive phenotypic signature for late and persistent neurodegenerative changes has not been accepted universally. Use of Animal Behavioral Tasks as Surrogates for Human Cognition Although it is accepted generally that exposure to anesthetics during vulnerable periods results in neurodegenerative histopathologic changes, how much this affects the animal on an organismal level is less certain. As discussed, neurocognitive deficits have been found in adult animals exposed to anesthetics as neonates; however, there is some question as to how much the most commonly used rodent tasks, including locomotor activity, and assessments of spatial learning and memory, actually apply to human children.6,9,25 Studies of anesthetic neurotoxicity in nonhuman primates, which are more phylogenetically related to humans, may allow future preclinical models to be more directly applicable to our species. The National Center for Toxicological Research has used the Operant Test Battery to assess the effects of nonanesthetic drug exposure on monkeys for several decades, and the behavior of sufficiently trained rhesus monkeys can be comparable with that of children.26 The use of this battery in preclinical models of anesthetic neurotoxicity may therefore allow for a closer comparison with human subjects. Lack of Surgical Stimuli and Physiologic Monitoring in Animal Models Studies on anesthetic neurotoxicity in many animal models are performed in a manner inconsistent with the modern operating suite and in the absence of surgical stimuli. The stressors of surgery themselves may be detrimental to neurodevelopment, and there is evidence that anesthesia and analgesia blunt these responses. For example, preterm infants administered fentanyl while undergoing patent ductus ligation have demonstrably improved postoperative outcomes compared with infants to whom opioid analgesia is not administered.27 In addition, neonates who received deep intraoperative anesthesia have improved hormonal and metabolic responses to surgery and lower mortality than infants anesthetized with a lighter technique.28 It should be noted that opioids were used for analgesia in both of these human studies and that, despite popular belief, halogenated inhaled anesthetics do not provide any significant measure of analgesia. There is also concern that surgical stressors may actually worsen anesthetic neurotoxicity. Studies performed in rats in the absence of opioid analgesia suggest that the combination of surgical stressors and inhaled anesthesia may be more detrimental than either alone. Both chemically induced and surgically induced pain exacerbated the immediate apoptosis and subsequently delayed cognitive impairment of a combined
nitrous oxide/isoflurane anesthetic in postnatal day 7 rats.29 Surgery without anesthesia is clearly detrimental, and it is entirely possible that surgical stressors without adequate analgesia may exacerbate anesthetic neurotoxicity. What specific role adequate analgesia may play in blunting these effects remains to be elucidated. Physiologic monitoring of human subjects undergoing anesthesia and surgery in the modern era also includes continuous monitoring of oxygenation, ventilation, temperature, and circulation. Human patients can be assessed rigorously for hypoglycemia and electrolyte imbalances preoperatively, intraoperatively, and postoperatively. This detailed level of physiologic monitoring generally is not performed in preclinical testing of anesthetic neurotoxicity, although there are notable exceptions.9 It is possible that unrecognized deviations from physiologic norms (eg, hypoxia, hypoglycemia, hypotension) may compound or confound evidence for, or against, anesthetic neurotoxicity. Recently conducted studies have moved toward full physiologic monitoring and improved maintenance of homeostatic variables, including endotracheal intubation and mechanical ventilation, between exposed and control groups.17,30 These efforts should be applauded and continued in future studies so as to obtain data more relevant to human subjects.
Conclusions from Animal Models When administered to neonatal animals, multiple anesthetic agents commonly used in clinical practice have been shown to cause both neuronal damage and long-term behavioral changes. These effects have been repeated by multiple research groups and in multiple species, including nonhuman primates. There is good evidence that long-duration exposures, simultaneous exposures to multiple anesthetic agents, and repeated anesthetic exposures cause histologic damage. The degree of histologic damage appears to correlate with observed behavioral changes; however, evidence for neuronal damage, and especially for long-term behavioral changes, following short-duration and single-agent exposures is less robust. The question of how applicable animal studies are to human subjects also remains, and before focusing on the risks that anesthetics may pose to human patients, one must remember that many surgeries and procedures performed on children each year are demonstrably beneficial, and we do not have viable alternatives to the anesthetics currently in clinical use that are necessary to perform these procedures. These caveats aside, we feel that the data from animal studies raise and reaffirm the concern that anesthetics, although necessary for the safe conduct of a multitude of surgeries, may in fact be neurotoxic, especially when given for long durations and at high doses.
Data from Observational Human Studies Before 2016, data from human studies on the topic of anesthetic neurotoxicity had been limited to retrospective and observational studies. Several of these studies deserve mention
Anesthetic Neurotoxicity: New Findings and Future Directions DIS 5.4.0 DTD ■ YMPD8759_proof ■ November 8, 2016
3
THE JOURNAL OF PEDIATRICS • www.jpeds.com (Table). A 2011 matched cohort study of more than 8000 children born between January 1, 1976, and December 31, 1982, in Rochester, Minnesota, included 350 children exposed to anesthesia before age 2 years. Children who were exposed to multiple, but not single, anesthetics/surgical procedures had a significantly greater incidence of learning disabilities compared with unexposed controls, even when accounting for general health.31 The Western Australian Pregnancy Cohort (Raine) Study of nearly 3000 children born between 1989 and 1991 also has been used to assess for cognitive changes following anesthesia. In this cohort, 321 children were exposed to anesthesia before age 3 years. At 10 years of age, exposed children were administered individually neuropsychological tests to assess a variety of specific behavioral modalities, something the Mayo study was not able to do. Assessed retrospectively, children in the Raine cohort demonstrated deficits in language and abstract reasoning at age 10 years even following a single exposure; however, there were no significant differences observed in assessments of fine or gross motor function, or in a standardized questionnaire completed by parents for the purposes of evaluating behavioral problems.32 An additional retrospective study of a subset of the Raine cohort found an increased incidence of deficits in individual language assessments at 10 years of age in children who had anesthesia and surgery before age 3 years. In exposed children, there also was an increased incidence of International Classification of Diseases, Ninth Revision codes for language and cognitive disorders. No differences, however, were found when exposed vs unexposed children were assessed with a group test of academic achievement.33 A study of Dutch twins that used a group-administered assessment of academic achievement as its primary outcome also failed to show any differences at age 12 years between twin pairs discordant for anesthesia administration before age 3 years (ie, when one twin received an anesthetic and the other did not).34 Interestingly in this study, twin pairs in which both twins were exposed to anesthesia before age 3 years were found to have significantly lower educational achievement scores and more cognitive deficits compared with unexposed twin pairs. The lack of difference between the discordant twins, however, does not support a relationship between anesthesia and later-life learning deficits. It has been noted previously that studies of potential anesthetic neurotoxicity relying on broader measures of cognition such as academic achievement tend to be negative, whereas those that perform individualized assessments of specific modalities are more likely to be positive.35 Reflecting what is seen in animal models, the individual modality assessed is important in interpreting these studies. This is illustrated by a 2016 study from the Raine cohort that found no differences in visual acuity, refractive error, or thickness of the retinal nerve fiber between children exposed to anesthesia before age 3 years and unexposed individuals.36
Data from Prospective Trials Unfortunately, data from animal studies and retrospective observational human studies do not provide definitive evi-
Volume ■■ dence for or against anesthetic neurotoxicity. These studies have generated sufficient concern to warrant large-scale, randomized clinical trials to assess whether general anesthetics impair neurocognitive development in human subjects. Results from the first of these studies recently have been published. An observer-blinded, international, multisite, randomized, controlled, equivalence study comparing 294 infants who received general anesthesia with 238 infants who received awake regional anesthesia for inguinal hernia repair found no differences in a prespecified secondary neurodevelopmental outcomes assessment between the 2 groups at 2 years of age.7 Although primary outcome assessments will be performed when children are 5 years old, these initial results represent the first data from a randomized, controlled trial in human patients designed to assess for developmental anesthetic neurotoxicity. In this study, only an awake regional technique or a sevoflurane inhalational general anesthetic was permitted. In the awake regional group, the use of any pharmacologic supplemental sedation was considered a protocol violation. At age 2 years, children were assessed by both an as-per-protocol and an intention-to-treat analysis in 5 distinct scales of developmental function, including cognitive, communication, physical, social/emotional, and adaptive measures. The mean duration of sevoflurane administration was only 54 minutes, and patient recruitment was restricted to relatively healthy infants. No evidence for adverse neurodevelopmental outcomes was found between the groups that received an awake regional or a general anesthetic. A similar ambidirectional sibling-matched cohort study of 105 healthy sibling pairs presenting for elective inguinal hernia repair also was completed recently.8 In this study, exposed subjects who had received a single anesthetic exposure before age 36 months for inguinal hernia repair were identified retrospectively. Neurocognitive and behavioral outcomes were then assessed prospectively by the use of unexposed siblings whose age was within 36 months of the exposed sibling as controls. Like the awake regional vs general anesthesia study previously discussed, the duration of anesthesia in exposed subjects was relatively short, with a median duration of 80 minutes (range 20-240 minutes). Like the previously discussed study, the subjects were all relatively healthy. Also like the previous study, there were no significant differences found between exposed and unexposed siblings in either the primary outcome of IQ scores or secondary domain-specific neuropsychological outcome tests.
Conclusions and Questions for Future Study Taken at face value, the recently published results from these 2 controlled clinical studies would seem reassuring, and in fact they are, but only as far as the results apply to relatively healthy patients receiving a short-duration anesthetic. Data from preclinical models have not demonstrated conclusively that shortduration, or single-agent, anesthetic exposures consistently cause neurodegeneration, much less significant behavioral effects. Findings from animal models are therefore consistent with
4
Montana and Evers DIS 5.4.0 DTD ■ YMPD8759_proof ■ November 8, 2016
2016
DIS 5.4.0 DTD ■ YMPD8759_proof ■ November 8, 2016
Anesthetic Neurotoxicity: New Findings and Future Directions
Table. Findings from selected human studies on anesthesia-related neurocognitive assessments Year
Study design
Anesthesia Exposure
Bartels et al34
2009
Retrospective monozygotic concordant-discordant twin study
Exposure to anesthesia before age 3 y was determined based on parent surveys.
A total of 1143 monozygotic twin pairs
Population studied
Davidson et al7
2016
Assessor-masked randomized controlled equivalence trial
Randomized to receive either awake regional or general anesthesia.
Flick et al31
2011
Matched cohort study
A procedure requiring general anesthesia before 2 y of age. Median anesthetic exposure was 75 min.
Ing et al32
2012
Retrospective cohort study
Determined based on whether children had undergone a surgical procedure that requires anesthesia.
Ing et al33
2014
Retrospective cohort study
Determined based on whether children had undergone a surgical procedure that requires anesthesia.
Infants < 60 wk postmenstrual age having inguinal hernia repair. A total of 238 children in the awake regional and 294 in the general anesthesia group. A total of 350 children exposed to anesthesia and 700 matched unexposed controls (2 controls per case). All children were from Olmsted County, Minnesota. Western Australian Pregnancy Cohort (Raine) Study. 321 children who had a surgical procedure before age 3 y were compared with 2287 who did not. A total of 781 children from the Western Australian Pregnancy Cohort (Raine) Study; 112 children had anesthesia exposure before age 3 y.
Sun et al8
2016
Ambidirectional sibling matched cohort
Median duration of 80 min of inhaled anesthesia. Twentyeight children also received IV agents.
Yazar et al36
2016
Cohort study
Determined based on whether children had undergone a surgical procedure that requires anesthesia.
A total of 105 children exposed to anesthesia before age 3 y compared with unexposed siblings within 36 mo of age; 18 exposed and 23 unexposed children also received anesthesia after 36 mo. Western Australian Pregnancy Cohort (Raine) Study. A total of 127 children who had a surgical procedure before age 3 y were compared with 707 who did not.
Outcomes measured
Findings
Comments
EA assessed by Dutch-CITO elementary standardized testing. CP assessed by Connors' Teacher Rating Scale. Secondary outcomes assessment at 2 y of age using the composite cognitive score from the Bayley Scale of Infant and Toddler Development III. Need for an IEP; presence of learning disabilities in reading, writing, or math; and group administered tests of achievement and cognition.
Twins exposed to anesthesia before age 3 y had lower EA and more CPs, however discordant twins did not differ from each other. No differences in the composite cognitive scale between the 2 groups.
Specific anesthetics used and duration of anesthesia were not known to the authors.
Neuropsychological tests were performed at 10 y of age and designed to assess language, listening, cognitive function, motor skills, and behavior.
Children exposed to anesthesia before age 3 y had an increased incidence of deficits in language and abstract reasoning.
Authors did not have direct access to medical records to determine specific anesthetics used, dose, or duration.
Neuropsychological tests were performed at 10 years of age. ICD-9–coded mental, behavioral, and neurodevelopment disorders were extracted from followup visits. Academic performance on a statewide fifth grade achievement test was assessed. Global cognitive function (IQ) assessed at 10 y of age was the primary outcome. Secondary outcomes included domain-specific neurodevelopment tests.
Exposure to anesthesia was associated with more deficits in individual language assessments and ICD-9 codes for language disorders. No difference in academic achievement testing was seen.
Authors did not have direct access to medical records to determine specific anesthetics used, dose, or duration.
No differences were seen between exposed and unexposed siblings in either the primary or secondary outcomes.
More than 75% of parents of test subjects had completed at least 2 y of college and more than 90% of parents were married.
At age 20 y, a comprehensive ophthalmologic examination including visual acuity, refractive error, and optic disc assessment was performed.
No association between anesthesia exposure and myopia, reduced visual acuity, or retinal nerve thinning.
Authors did not have direct access to medical records to determine specific anesthetics used, dose, or duration.
CP, cognitive problems; EA, educational achievement; ICD-9, International Classification of Diseases, Ninth Revision; IEP, individualized education plan; IV, intravenous.
Median anesthetic duration was 54 min. Primary outcomes assessment is planned for 5 y of age.
Children who had 2 or more anesthetics, but not one, were more likely to have learning disabilities, and an IEP for speech and language.
5
COMMENTARY
Authors
THE JOURNAL OF PEDIATRICS • www.jpeds.com findings from recently published human studies, namely that short-duration, single-agent anesthetics likely do not pose a significant risk to relatively healthy children. A sobering counterpoint to this conclusion is the possibility that longer duration, or repetitive, anesthetic exposures may pose a risk for neurocognitive impairment in human subjects. Healthy children generally do not require multiple anesthetics; thus, it will be extremely challenging for future studies to tease out whether any observed neurocognitive deficits are due to the anesthetic regimen or the pathophysiologic background of patients that require the anesthetic and the associated procedures. Thankfully, there is a way forward. Limitation, mitigation, and alternative neuroprotective strategies are being tested that may reduce the risk of anesthetic neurotoxicity following long exposures. In rats the inert gas xenon, which is an NMDA antagonist and anesthetic, has been found not be apoptogenic and to limit isoflurane-induced apoptosis.37 Similarly, the finding that the alpha 2 -adrenergic receptor agonist dexmedetomidine reduces isoflurane-induced caspase-3 activation in postnatal day 7 rats suggests that it may be a neuroprotective sedating agent.38 In mice the AMPAkine CX546, which potentiates a-amino-3-hydroxy-5-methyl-4isoxazolepropionic acid activity during emergence from anesthesia, recently was shown to diminish synaptic changes and behavioral deficits following ketamine anesthesia.39 In nonhuman primates, coadministration of a lithium infusion protects against isoflurane-induced neuronal apoptosis.40 In humans, the T REX Pilot Study currently is recruiting to test the feasibility of using a combination of dexmedetomidine, opioids, and regional nerve blocks as an alternative technique to gaseous inhaled anesthetics (ClinicalTrials.gov: NCT02353182). Further research is required to determine whether these strategies are neuroprotective in long-duration anesthetics. Studies in animals designed to test these strategies must be rigorous and use appropriate physiologic monitoring. If viable alternative anesthetic strategies are found, they should be used to inform the design of randomized, controlled clinical trials. Additional relevant questions remain. Are specific patient populations more at risk than others? Are patients who undergo specific surgical procedures more likely than others to have neurocognitive deficits? Are infants born premature who require sedation or anesthesia more vulnerable than infants born at term? Can diagnostic imaging modalities or biomarkers noninvasively diagnose neurotoxicity? If cumulative or repeat dosing increases the risk for neurotoxicity in humans, can we define cut-offs below which detectable effects are unlikely? Even though gaseous inhaled anesthetics are used almost exclusively in the operating room or diagnostic/procedural suite, thousands of infants and children worldwide in intensive care unit settings are sedated with other agents, including benzodiazepines and opioids, to facilitate life-saving practices such as mechanical ventilation. Are these neonatal intensive care unit and pediatric intensive care unit patients at risk for sedative neurotoxicity following prolonged sedation regimens? Definitive determination of the risks that anesthetics and sedatives may pose to the developing human brain will require a con-
Volume ■■ certed global effort and coordination between parents, scientists, and healthcare providers. The task will not be easy and will require years to complete; however, we owe it to our youngest and most vulnerable patients, as well as their parents, and their caregivers, to answer these questions. ■ Submitted for publication Jul 22, 2016; last revision received Sep 9, 2016; accepted Oct 17, 2016 Reprint requests: Alex S. Evers, MD, Washington University School of Medicine, Washington University Medical Center, Campus Box 8054; 660 South Euclid Ave, St Louis, MO 63110-1093. E-mail: [email protected]
References 1. Wanderer JP, Rathmell JP. Complex information for anesthesiologists presented quickly and clearly. Anesthesiology 2014;120:A23. 2. van der Griend BF, Lister NA, McKenzie IM, Martin N, Ragg PG, Sheppard SJ, et al. Postoperative mortality in children after 101,885 anesthetics at a tertiary pediatric hospital. Anesth Analg 2011;112:1440-7. 3. Jevtovic-Todorovic V, Absalom AR, Blomgren K, Brambrink A, Crosby G, Culley DJ, et al. Anaesthetic neurotoxicity and neuroplasticity: an expert group report and statement based on the BJA Salzburg Seminar. Br J Anaesth 2013;111:143-51. 4. Uemura E, Levin ED, Bowman RE. Effects of halothane on synaptogenesis and learning behavior in rats. Exp Neurol 1985;89:520-9. 5. Jevtovic-Todorovic V, Todorovic SM, Mennerick S, Powell S, Dikranian K, Benshoff N, et al. Nitrous oxide (laughing gas) is an NMDA antagonist, neuroprotectant and neurotoxin. Nat Med 1998;4:460-3. 6. Jevtovic-Todorovic V, Hartman RE, Izumi Y, Benshoff ND, Dikranian K, Zorumski CF, et al. Early exposure to common anesthetic agents causes widespread neurodegeneration in the developing rat brain and persistent learning deficits. J Neurosci 2003;23:876-82. 7. Davidson AJ, Disma N, de Graaff JC, Withington DE, Dorris L, Bell G, et al. Neurodevelopmental outcome at 2 years of age after general anaesthesia and awake-regional anaesthesia in infancy (GAS): an international multicentre, randomised controlled trial. Lancet 2016;387:23950. 8. Sun LS, Li G, Miller TL, Salorio C, Byrne MW, Bellinger DC, et al. Association between a single general anesthesia exposure before age 36 months and neurocognitive outcomes in later childhood. JAMA 2016;315:231220. 9. Disma N, Mondardini MC, Terrando N, Absalom AR, Bilotta F. A systematic review of methodology applied during preclinical anesthetic neurotoxicity studies: important issues and lessons relevant to the design of future clinical research. Paediatr Anaesth 2016;26:6-36. 10. Cattano D, Young C, Straiko MM, Olney JW. Subanesthetic doses of propofol induce neuroapoptosis in the infant mouse brain. Anesth Analg 2008;106:1712-4. 11. Zou X, Liu F, Zhang X, Patterson TA, Callicott R, Liu S, et al. Inhalation anesthetic- induced neuronal damage in the developing rhesus monkey. Neurotoxicol Teratol 2011;33:592-7. 12. Paule MG, Li M, Allen RR, Liu F, Zou X, Hotchkiss C, et al. Ketamine anesthesia during the first week of life can cause long-lasting cognitive deficits in rhesus monkeys. Neurotoxicol Teratol 2011;33:220-30. 13. Loepke AW, McGowan FX Jr, Soriano SG. CON: the toxic effects of anesthetics in the developing brain: the clinical perspective. Anesth Analg 2008;106:1664-9. 14. Workman AD, Charvet CJ, Clancy B, Darlington RB, Finlay BL. Modeling transformations of neurodevelopmental sequences across mammalian species. J Neurosci 2013;33:7368-83. 15. Creeley CE, Dikranian KT, Dissen GA, Back SA, Olney JW, Brambrink AM. Isoflurane- induced apoptosis of neurons and oligodendrocytes in the fetal rhesus macaque brain. Anesthesiology 2014;120:626-38. 16. Brambrink AM, Back SA, Riddle A, Gong X, Moravec MD, Dissen GA, et al. Isoflurane-induced apoptosis of oligodendrocytes in the neonatal primate brain. Ann Neurol 2012;72:525-35.
6
Montana and Evers DIS 5.4.0 DTD ■ YMPD8759_proof ■ November 8, 2016
COMMENTARY
2016 17. Brambrink AM, Evers AS, Avidan MS, Farber NB, Smith DJ, Zhang X, et al. Isoflurane-induced neuroapoptosis in the neonatal rhesus macaque brain. Anesthesiology 2010;112:834-41. 18. Rudin M, Ben-Abraham R, Gazit V, Tendler Y, Tashlykov V, Katz Y. Singledose ketamine administration induces apoptosis in neonatal mouse brain. J Basic Clin Physiol Pharmacol 2005;16:231-43. 19. Scallet AC, Schmued LC, Slikker W Jr, Grunberg N, Faustino PJ, Davis H, et al. Developmental neurotoxicity of ketamine: morphometric confirmation, exposure parameters, and multiple fluorescent labeling of apoptotic neurons. Toxicol Sci 2004;81:364-70. 20. Dekkers MP, Nikoletopoulou V, Barde YA. Cell biology in neuroscience: death of developing neurons: new insights and implications for connectivity. J Cell Biol 2013;203:385-93. 21. Jevtovic-Todorovic V, Olney JW. PRO: anesthesia-induced developmental neuroapoptosis: status of the evidence. Anesth Analg 2008;106:165963. 22. Briner A, De Roo M, Dayer A, Muller D, Habre W, Vutskits L. Volatile anesthetics rapidly increase dendritic spine density in the rat medial prefrontal cortex during synaptogenesis. Anesthesiology 2010;112:546-56. 23. Backeljauw B, Holland SK, Altaye M, Loepke AW. Cognition and brain structure following early childhood surgery with anesthesia. Pediatrics 2015;136:e1-12. 24. Zhang X, Liu S, Newport GD, Paule MG, Callicott R, Thompson J, et al. In vivo monitoring of sevoflurane-induced adverse effects in neonatal nonhuman primates using small-animal positron emission tomography. Anesthesiology 2016;125:133-46. 25. Loepke AW, Istaphanous GK, McAuliffe JJ 3rd, Miles L, Hughes EA, McCann JC, et al. The effects of neonatal isoflurane exposure in mice on brain cell viability, adult behavior, learning, and memory. Anesth Analg 2009;108:90-104. 26. Paule MG, Green L, Myerson J, Alvarado M, Bachevalier J, Schneider JS, et al. Behavioral toxicology of cognition: extrapolation from experimental animal models to humans: behavioral toxicology symposium overview. Neurotoxicol Teratol 2012;34:263-73. 27. Anand KJ, Sippell WG, Aynsley-Green A. Randomised trial of fentanyl anaesthesia in preterm babies undergoing surgery: effects on the stress response. Lancet 1987;1:62-6. 28. Anand KJ, Hickey PR. Halothane-morphine compared with high-dose sufentanil for anesthesia and postoperative analgesia in neonatal cardiac surgery. N Engl J Med 1992;326:1-9.
29. Shu Y, Zhou Z, Wan Y, Sanders RD, Li M, Pac-Soo CK, et al. Nociceptive stimuli enhance anesthetic-induced neuroapoptosis in the rat developing brain. Neurobiol Dis 2012;45:743-50. 30. Whitaker EE, Bissonnette B, Miller AD, Koppert TL, Tobias JD, Pierson CR, et al. A novel, clinically relevant use of a piglet model to study the effects of anesthetics on the developing brain. Clin Transl Med 2016;5:2. 31. Flick RP, Katusic SK, Colligan RC, Wilder RT, Voigt RG, Olson MD, et al. Cognitive and behavioral outcomes after early exposure to anesthesia and surgery. Pediatrics 2011;128:e1053-61. 32. Ing C, DiMaggio C, Whitehouse A, Hegarty MK, Brady J, von UngernSternberg BS, et al. Long-term differences in language and cognitive function after childhood exposure to anesthesia. Pediatrics 2012;130:e47685. 33. Ing CH, DiMaggio CJ, Whitehouse AJ, Hegarty MK, Sun M, von UngernSternberg BS, et al. Neurodevelopmental outcomes after initial childhood anesthetic exposure between ages 3 and 10 years. J Neurosurg Anesthesiol 2014;26:377-86. 34. Bartels M, Althoff RR, Boomsma DI. Anesthesia and cognitive performance in children: no evidence for a causal relationship. Twin Res Hum Genet 2009;12:246-53. 35. Flick RP, Nemergut ME, Christensen K, Hansen TG. Anesthetic-related neurotoxicity in the young and outcome measures: the devil is in the details. Anesthesiology 2014;120:1303-5. 36. Yazar S, Hewitt AW, Forward H, Jacques A, Ing C, von Ungern-Sternberg BS, et al. Early anesthesia exposure and the effect on visual acuity, refractive error, and retinal nerve fiber layer thickness of young adults. J Pediatr 2016;169:256-9, e1. 37. Ma D, Williamson P, Januszewski A, Nogaro MC, Hossain M, Ong LP, et al. Xenon mitigates isoflurane-induced neuronal apoptosis in the developing rodent brain. Anesthesiology 2007;106:746-53. 38. Sanders RD, Sun P, Patel S, Li M, Maze M, Ma D. Dexmedetomidine provides cortical neuroprotection: impact on anaesthetic-induced neuroapoptosis in the rat developing brain. Acta Anaesthesiol Scand 2010;54:710-6. 39. Huang L, Cichon J, Ninan I, Yang G. Post-anesthesia AMPA receptor potentiation prevents anesthesia-induced learning and synaptic deficits. Sci Transl Med 2016;8:344ra85. 40. Noguchi KK, Johnson SA, Kristich LE, Martin LD, Dissen GA, Olsen EA, et al. Lithium protects against anaesthesia neurotoxicity in the infant primate brain. Sci Rep 2016;6:22427.
Anesthetic Neurotoxicity: New Findings and Future Directions
7
DIS 5.4.0 DTD ■ YMPD8759_proof ■ November 8, 2016