Anesth Analg 2006;102:72-74
© 2006 International Anesthesia Research Society
doi: 10.1213/01.ANE.0000181101.71587.8A
PEDIATRIC ANESTHESIA
Physiological Assessment and Control in Studies Evaluating Central Nervous System Injury: Should Size Matter?
Patrick M. Kochanek, MD,
Robert S. B. Clark, MD,
Kimberly D. Statler, MD, and
Larry W. Jenkins, PhD
Department of Critical Care Medicine, Department of Neurological Surgery, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania; Department of Pediatrics, Primary Children’s Hospital, University of Utah
Address correspondence and reprint requests to Patrick Kochanek, Safar Center for Resuscitation Research, Critical Care Med, Pittsburgh, PA 15260. Address e-mail to kochanekpm{at}ccm.upmc.edu.
In this issue of Anesthesia & Analgesia, Loepke et al. (1) examined the effect of a 1-h isoflurane exposure on extracerebral physiological variables in postnatal day (PND) 10 mice—an age suggested to mimic the brain maturation of human term infants. Studies were performed in anesthetized mice with and without controlled mechanical ventilation and in parallel groups exposed to a hypoxic-ischemic insult. Mice that were removed from the dam for an identical period of time, but not exposed to prolonged anesthesia, served as controls. Remarkably, mice anesthetized with isoflurane became hypoglycemic, as detected at the end of the 1-h anesthetic exposure. At that time, the mean blood glucose concentrations were 43 mg/dL and 38 mg/dL in spontaneously ventilating and mechanically ventilated mice, respectively. In some mice, blood glucose concentration was less than the 20 mg/dL limit of the assay used. The mean blood glucose concentration was even lower (26 mg/dL) in anesthetized, mechanically ventilated mice exposed to hypoxic-ischemic injury, and the mortality rate was 100%. The normal value for blood glucose in the control group was 108 mg/dL. Unanesthetized controls for either the anesthetic exposure alone or the hypoxic-ischemic injury did not exhibit hypoglycemia. Although the mechanism underlying the isoflurane-mediated hypoglycemia in PND 10 mice remains unclear, the findings may be important.
In 2003, Jevtovic-Todovoric et al. (2) reported in the Journal of Neuroscience, that a 6-h exposure of PND 7 rats to isoflurane, either alone or combined with midazolam and nitrous oxide, resulted in neurological deterioration in the developing brain. Needless to say, the report garnered considerable notice. The possibility that exposure to general anesthetics at a critical juncture in brain development could lead to neuronal death paralleled the concept of central nervous system injury in conditions such as the fetal alcohol syndrome, in which exposure to the N-methyl-d-aspartate receptor antagonist ethanol leads to neurodegeneration and a described phenotype (3) and raised important concerns in obstetrics, perinatal medicine, neonatology, and pediatric critical care medicine. Clearly, the findings could have far-reaching implications for clinical care, including anesthesia during pregnancy or delivery, surgery in premature infants, and possibly even sedation or brain-targeted therapies, such as "barbiturate coma" in infants with severe traumatic brain injury, in pediatric intensive care units world-wide. The findings of Jevtovic-Todovoric et al. (2) merit additional investigation in light of these potential ramifications to routine clinical care, as well as possible implications for the interpretation of experimental neonatal or pediatric brain injury studies.
The report by Loepke et al. in this issue of the journal (1) describing isoflurane’s effects in developing mice expands on the work of Jevtovic-Todovoric et al. (2), using the perspective of a clinical anesthesiologist who is probing the potential clinical ramifications of the findings. Monitoring and controlling physiological variables in the newborn or developing rodent is challenging. Endotracheal intubation, controlled mechanical ventilation, and the use of invasive continuous arterial blood pressure and oxygen saturation monitoring are technically demanding. Similarly, serial arterial blood gas sampling and assessment of blood chemistries are even more problematic because assay sample volumes are often a physiologically important fraction of the rodent’s total blood volume. Consequently, many experimental studies involving developing rodents may not effectively mirror the current clinical practice of invasive monitoring and stringent control of physiological variables in critically ill infants and young children. Loepke et al. (1) have tried to address this discrepancy in their study by monitoring and/or controlling several clinically relevant variables, including oxygenation, ventilation, and blood glucose concentrations.
Loepke et al. (1) studied the effects of a 1-h exposure to isoflurane, with or without mechanical ventilation, alone or in concert with hypoxic-ischemic injury. They observed respiratory acidosis and marked hypoglycemia after a 1-h exposure to an isoflurane concentration of 1.8% in oxygen. In contrast, Jevtovic-Todovoric et al. (2) studied the effects of a 6-h exposure to anesthetics in PND 7 rats. Although body temperature was controlled, the rats were not mechanically ventilated, were maintained on a Fio2 of either  0.25 (N2O/O2 3:1) with isoflurane or room air plus isoflurane, were not administered routine IV fluids with dextrose, and were not subjected to continuous arterial blood pressure, oxygen saturation, or blood glucose monitoring. The isoflurane concentration ranged among 0.75%, 1.0%, and 1.5%. In 8 separate rats treated with the 3-drug anesthetic combination (although the specific isoflurane concentration in these 8 rats was not provided), a single blood gas was measured at the end of the 6-h period. The results of the blood gas determinations were surprisingly good, with no metabolic acidosis or hypoxemia, but a mild respiratory alkalosis. Blood glucose levels were not measured, but the findings of Loepke et al. (1) suggest that hypoglycemia would likely occur during a 6-h isoflurane exposure in a developing rodent. However, there are some important differences to consider when comparing these two reports. The PND 10 mice exposed to 1-h of isoflurane (1) do not mirror the PND 7 rats exposed to 6 h of isoflurane (2). Moreover, Loepke et al. (1) used an isoflurane concentration of 1.8%, contrasting the somewhat smaller concentrations (i.e., a maximum concentration of 1.5%) used by Jevtovic-Todovoric et al. (2). The possible role of hypoglycemia, if it occurred, in the widespread neurodegeneration observed by Jevtovic-Todovoric et al. (2) is not clear, but at least deserves careful consideration.
Unfortunately, we have more questions than answers as to what level of blood glucose would be deleterious in a PND 7 rat or PND 10 mouse subjected to prolonged isoflurane anesthesia. Loepke et al. (1) discussed blood glucose concentrations in PND 10 mice from a purely clinical perspective. The critical value for blood glucose concentration in contributing to brain injury in developing rodents, with or without anesthesia, remains unclear. Classic studies of insulin-induced hypoglycemia have been reported in newborn rats as early as 1953 (5) and sustained glucose levels of 20 mg/dL for at least 3 h produced substantial neuronal injury in unanesthetized rats between the ages of PND 3 and 8 (6). Yager et al. (7) more recently reported that normal blood glucose for PND 7 rats was 97 mg/dL and that insulin induced hypoglycemia to a value of 77 mg/dL increased mortality during a 2-h hypoxia-ischemia exposure from 4% to 30%. In contrast, PND 7 rats fasted for 12 h (without anesthesia) before hypoxia-ischemia appeared behaviorally normal, had blood glucose concentrations of 60 mg/dL, and did not have enhanced neuronal damage versus controls. In this study, high levels of ketones were posed to be either neuroprotective or at least to provide an alternative fuel in the fasted rats. The comparison of these studies with the current work of Loepke et al. (1) is complicated, however, because fasting preceded anesthesia. Also, classic studies in developing primates suggest that the duration of hypoglycemia needed to produce brain injury may be rather long and surprisingly variable (8). Thus, despite the findings of Loepke et al. (1), the critical level of blood glucose that would be deleterious in the PND 7 rat or PND 10 mouse exposed to prolonged isoflurane anesthesia and the potential contribution of hypoglycemia to the findings of Jevtovic-Todovoric et al. (2) remain unclear.
A recent study by McClaine et al. (4) further explored the question of whether anesthetic exposure during brain development is deleterious. Using a clinically relevant paradigm including blood gas and arterial blood pressure monitoring, McClaine et al. (4) exposed near-term pregnant sheep to a combination of midazolam, sodium thiopental, and isoflurane for 4 h. A 4-h exposure was chosen to mimic what would be the longest potentially anticipated anesthetic need for clinical use in obstetrical surgery. Anesthetic exposure did not lead to enhanced neuronal death as assessed 6 days after treatment. However, the study included only three control ewes, and some potentially relevant differences between groups may have been missed as a result of inadequate statistical power. Unfortunately, the financial trade-off of using a large animal model with a clinically relevant paradigm often leads to potential problems with adequate sample size.
Reports on the effects of anesthetic exposure on the developing brain are conflicting and demand further investigation. From the current study by Lempke et al. (1), we have once again learned that every effort should be made to include physiological monitoring and control in developing rodent models at a level commensurate with clinical care. We believe that comprehensive physiological monitoring is needed during "exposures" and "insults" and anticipate additional research in this important area. In work from our center in the area of developmental cardiopulmonary arrest and resuscitation, we have tried to live up to this creed. Fink et al. (9) incorporated a clinically relevant paradigm with comprehensive monitoring and physiological control in PND 17 rats, and we believe that this is of paramount importance to interpreting results in testing of therapies in cerebral resuscitation.
In light of the small cost of rodent research relative to large animal work, the many molecular tools available in rodents, and the wealth of established models of brain injury in developing rodents, the value of these models is obvious. Nevertheless, we continue to lack answers to the important questions of whether anesthetic exposure is injurious to the developing brain. It is quite likely that there is truth in both camps (neuroprotection versus neurodegeneration) on the question of the effect of anesthetic exposure in the developing brain. Future studies, both in experimental models and using clinical databases tracking anesthetic exposure in obstetrical and neonatal practice or prospective comparisons in these populations, are needed to further clarify this important matter.
 |
Footnotes
|
|---|
Accepted for publication July 26, 2005.
|
References
|
|---|
- Loepke AW, McCann JC, Kurth CD, McAuliffe JJ. Physiologic effects of isoflurane anesthesia in neonatal mice. Anesth Analg 2006;102:75–80.[Abstract/Free Full Text]
- Jevtovic-Todorovic V, Hartman RE, Izumi Y, et al. Early exposure to common anesthetic agents causes widespread neurodegeneration in the developing rat brain and persistent learning deficits. J Neuroscience 2003;23:876–82.[Abstract/Free Full Text]
- Dikranian K, Qin YQ, Labruyere J, et al. Ethanol-induced neuroapoptosis in the developing rodent cerebellum and related brain stem structures. Brain Res Dev Brain Res 2005;155:1–13.[Medline]
- McClaine RJ, Uemura K, de la Fuente S, et al. General anesthesia improves fetal cerebral oxygenation without evidence of subsequent neuronal injury. J Cereb Blood Flow Metab 2005;25:1060–9.[Web of Science][Medline]
- Hicks SP. Developmental brain metabolism. Effects of cortisone, anoxia, fluoroacetate, radiation insulin and other inhibitors on the embryo, newborn and adult. AMA Arch Pathol 1953;55:302–7.[Web of Science][Medline]
- Jones EL, Smith WT. Hypoglycaemic brain damage in the neonatal rat. In: Brierley JB, Meldrum BS, eds. Brain hypoxia. Philadelphia: JB Lippincott, 1971;231–41.
- Yager JY, Heitjan DF, Towfighi J, Vannucci RC. Effect of insulin-induced and fasting hypoglycemia on perinatal hypoxic-ischemic brain damage. Pediatr Res 1992;31:138–142.[Web of Science][Medline]
- Kahn KJ, Myers RE. Insulin-induced hypoglycaemia in the non-human primate. I. Clinical consequences. In: Brierley JB, Meldrum BS, eds. Brain hypoxia. Philadelphia: JB Lippincott, 1971;185–94.
- Fink EL, Alexander H, Marco CD, et al. Experimental model of asphyxial cardiopulmonary arrest in rats. Pediatr Crit Care Med 2004;5:139–44.[Medline]
|
Top
References
