To the Editor:—
“Anesthesia in Infancy Linked to Later Disabilities” is a provocative, if not sensational headline published by Time magazine Tuesday, March 24, 2009,1regarding the findings of a retrospective cohort study of anesthetic exposure and learning disabilities between 1967 and 1982 by Dr. Wilder et al. 2The articles in the April 2009 issue of Anesthesiology regarding anesthesia and the developing brain are of great interest to practitioners of pediatric anesthesia. The alarms are ever increasing regarding the risk of anesthesia for the developing human brain. But the significance of the animal studies to clinical practice is uncertain, and there is little to support a causal link between anesthesia and learning disabilities. There does seem to be an association between anesthetic exposures and learning disabilities, but a similar correlation undoubtedly exists between hospital admission, intravenous fluid administration, and repeated invasive and/or noninvasive hemodynamic monitoring and these same learning disabilities. A few comments regarding both the animal research and the retrospective studies will, I hope, provide some perspective on the issue of anesthetic neurotoxicity.
Previous animal studies do not evaluate anesthetic effect in the presence of surgical or medical stressors. The tail clamp model of Stratmann et al. 3more closely resembles the response to surgery, and they are to be applauded for detailing the effects of hypercapnia and acidosis on outcome.4However, they report a mortality of 25%, including deaths in the animals exposed to “only” 2 h of anesthesia. Although the phrase “clinically relevant doses of anesthetics” is now commonly used, I would remind readers that the life expectancy of a rat is only 9 months. One might ask what a comparable anesthetic exposure in humans is. Simple mathematics would suggest that 4 h in the life of a rat might represent as many as 16 days for humans with a life expectancy of 75 yr. Interestingly, in an early study, Jevtovic-Todorovic et al. 5demonstrated a threshold response to cerebrocortical injury and reported that inclusion of isoflurane (1%), halothane, pentobarbital, and diazepam all prevented neurotoxic reactions in adult rats during a 3-h exposure to nitrous oxide and/or ketamine. These specimens demonstrated histologically normal neurons. It is unclear why subsequent studies of anesthetic neurotoxicity in rodent pups subjected the animals to longer exposures when a threshold effect was seen with various anesthetic agents. Perhaps an animal model with mortality statistics that resemble outcomes in anesthetized neonates would be more appropriate for evaluating the long-term effects of anesthesia on the developing brain. One must also be aware that exposure of the developing brain to increased oxygen concentrations produces similar neuropathologic changes.6
As the parent of a 17 year old with moderately severe learning disabilities and a history of multiple anesthetic exposures before the age of 4 yr, I found the article by Dr. Wilder et al. linking early exposure to anesthesia and learning disabilities both intriguing and troubling. They do provide some interesting data, most of which they do not address in the discussion. To their credit, they admit that one cannot determine whether the results reflect exposure to anesthesia or the need for anesthesia. However, in the discussion, despite controlling for birth weight, sex, and gestational age, they do not address the confounders cited, including prolonged labor and hemorrhagic complications of pregnancy. They do not speak to the comorbidities of children presenting to the operating room for multiple procedures. One would expect this information to be available in their hospital database. Certainly, one should analyze the data for the effects of factors such as perioperative hemorrhage, sepsis, seizure disorders, severe lung disease and its associated episodes of hypoxemia and prolonged ventilation, neurologic malformations, and cerebral palsy. I would suggest that the complex medical history of my own son is representative of the learning disabled who have had multiple anesthetic exposures. His American Society of Anesthesiologists physical status never exceeded II during his four anesthetics as a young child, despite the fact that during that critical period of development, he experienced neonatal sepsis, disseminated osteomyelitis, hepatic insufficiency with attendant coagulopathy, postfebrile partial complex seizure disorder, perioperative hemorrhage, and postoperative anemia. I have searched for explanations for his learning disabilities, but not once in 17 yr have I thought to attribute them to his anesthetic exposures.
A rational understanding of the potential neurotoxicity of anesthetic and sedative agents in the developing brain requires an animal model that closely mimics the clinical reality of serious illness and surgical stress requiring intervention. Clinical reviews, both retrospective and prospective, that address the association of anesthetic exposure and compromised neurodevelopment in young children are critical to our understanding. A threshold toxic dose should be sought and the possibility of a biphasic response of the developing brain should be considered, much like oxygen exposure where both hypoxia and hyperoxia result in permanent deficits. Paracelsus stated that “Poison is in everything, and no thing is without poison. The dosage makes it either a poison or a remedy.” His observation is as relevant today as it was nearly 500 yr ago.
Children's Hospital of Wisconsin, Medical College of Wisconsin, Milwaukee, Wisconsin. sutaylor@mcw.edu