GAS, PANDA, and MASK: Reply

We would like to thank both Ing et al.¹  and Drs. Williams and Sartorelli²  for their thoughtful comments on our recent Editorial, “GAS, PANDA, and MASK: No Evidence of Clinical Anesthetic Neurotoxicity!”³  We address their concerns in turn below.

Developmental anesthesia neurotoxicity is probably one of the most debated and controversial issues of the past 2 decades in pediatric perioperative care. The comments of Ing et al.¹  are a perfect demonstration of how interpretation of any data on this topic can vary even among those individuals who have been working on this field for a long time. The most plausible explanation for this strong divergence in opinions regarding the clinical implication of the already existing data stems from the important and probably unsolvable limitation of both experimental and clinical studies in this domain. In the following lines, we address the three lines of thoughts raised by Ing et al.

Our first comments go to the interpretation of secondary outcomes. We agree that a few among the multitude of secondary outcomes may suggest a statistical association between anesthesia exposure and neurodevelopmental outcome but associations do not imply a causal relationship. One must recognize that occasional results in subgroup analysis and secondary outcomes are difficult to interpret since these studies were neither specifically designed nor sufficiently powered to obtain reliable conclusions to these questions. While they may serve as food for thought, they definitely do not provide evidence of clinical anesthetic neurotoxicity.

Our second comment goes to the dose-response relationship of anesthesia exposure. We agree that there is a biologic rationale backed by animal experimental evidence in this regard. Human data are, however, controversial. While the Mayo Anesthesia Safety in Kids (MASK) study⁴  raises the possibility that multiple anesthetic exposures and surgical episodes may lead to worse neurodevelopmental outcomes, other studies do not fully support this notion.^5,6  The large number of illnesses and confounders associated with pediatric populations necessitating repeated anesthesia and surgery at early life may explain these differences in the so-called biologic gradient.

Last but not least, Ing et al. argue to separate the clinical and the scientific questions. We respectfully disagree with this proposition since we believe that (substantial) changes in clinical practice should be driven by the results of appropriately planned clinical studies. At present, no such study convincingly supports the possibility of clinically relevant developmental anesthesia neurotoxicity. In fact, most studies, with the possible exception of the General Anesthesia Spinal (GAS) trial,⁷  are not even designed to directly answer this question. Rather, and this is a fundamental difference from experimental studies, they tackle the impact of the perioperative period as a whole, on neurobehavioral outcome. Based on current clinical data, how can we truly expect to draw specific conclusion on drug-related effects in a highly complex environment when as yet unexplained interactions among perioperative stress and numerous surgical and anesthetic factors can result in a plethora of unanticipated effects?

We fully agree with Ing et al., and we also pointed this out in our Editorial, that many unanswered questions remain. We also agree that many of these questions are intellectually stimulating. However, from a public health perspective, the elephant in the room remains the questionable feasibility of a study that would provide us with clinically relevant information on anesthetics neurotoxicity. Unfortunately, no clinical trial can ever be conducted that will conclusively prove that anesthetic neurotoxicity does not exist, because one cannot prove a negative. As said before by Dr. Ted Eger, one cannot disprove the existence of dragons.⁸  The hypothetical relevance of such investigations should also be considered in the light of epidemiologic data showing that the hypothetical impact of exposure to “perioperative period(s)” on subsequent neurodevelopment appears to be much less important than parental socioeconomic status, sex, or even the period of the year when a child is born.

Therefore, the real question we should ask at the current state of our knowledge is whether we can still convincingly justify the need and research cost for expensive clinical studies aimed to find anesthetics neurotoxicity. Or are there more important and clinically relevant questions of pediatric perioperative care aimed to promote brain health? We are convinced about the latter.

We agree with Drs. Williams and Sartorelli that awake spinal anesthesia can be considered an attractive option for appropriate surgeries in small infants. In fact, the point we wished to make in our Editorial is that the hypothetical risks of anesthetic neurotoxicity should not dictate our choice of regional versus general anesthesia. There is no evidence of the superiority of one approach over the other in terms of clinically relevant outcome. Therefore, the skills and expertise of the anesthesiologists and surgeons should be the main factors behind this strategic decision. In academic centers where teaching is a priority, the duration of even straightforward surgical procedures may often exceed the duration of a single spinal block. Given the importance of adequate analgesia during the entire procedure, general anesthesia, often in combination with a regional blockade, may have obvious advantages in these situations. As Drs. Williams and Sartorelli also point out, up to 20% of children with spinal anesthesia may need additional sedation even in experienced hands. While this situation can be easily handled by experienced pediatric anesthesiologists, failure of spinal anesthesia and the subsequent change in management plan may be more dangerous in less experienced hands. Again, it is the anesthesiologist and not the anesthetic that makes the difference.

Dr. Vutskits is an Editor of Anesthesiology. He served as consultant for Primex (Zug, Switzerland) and Regeneron (Tarrytown, New York). Dr. Culley is an Executive Editor of Anesthesiology. She serves as a Director and Secretary for the American Board of Anesthesiology, Chair of the Academic Anesthesiology Committee for the American Society of Anesthesiology, ex officio Member of the Anesthesiology Review Committee for the Accreditation Council for Graduate Medical Education and as a member of the 3C Committee for American Board of Medical Specialities.

Dr. Culley is supported by the National Institutes of Health (Bethesda, Maryland) and the Department of Anesthesiology, Perioperative, and Pain Medicine, Brigham and Women’s Hospital (Boston, Massachusetts). Dr. Vutskits is supported by the Swiss National Science Foundation (Berne, Switzerland) grant No. 31003A-130625.