Anesthesia and neurotoxicity: The problems with animal models and why you should be skeptical
Michael E. Nemergut, MD PhD
As promised a couple of weeks ago, the PAAD returns once again to the issue of general anesthesia and neurotoxicity in the newborn. Today, I’ve asked Mike Nemergut to join the PAAD team for this review. As way of background, Dr. Nemergut is on the faculty at the Mayo Clinic and is quadruple boarded in Pediatrics, Anesthesiology, Pediatric Anesthesiology, and Pediatric Critical Care Medicine. He spends most of his time at Mayo in the PICU and OR. He was one of my students and mentees at Johns Hopkins and I continue to watch his career blossom at Mayo with great pride. While at Hopkins, Mike and I discussed/debated the neuroapoptosis research and clinical articles that have flooded our literature at length, and I asked him to review today’s article using not only his lens as a clinician but as a basic scientist as well. Myron Yaster MD
Original review article
Feng Gao, Joseph A Wahl, Thomas F Floyd. Anesthesia and neurotoxicity study design, execution, and reporting in the nonhuman primate: A systematic review Paediatr Anaesth. 2022 Apr;32(4):509-521. PMID: 35066973
Few issues have garnered more attention in anesthesiology than the potential for anesthetics to induce neurologic damage in developing children. Starting in traditional rodent animal models and culminating with studies in non-human primates (NHPs), these models have been the subject of literally hundreds of studies reporting neuroapoptosis in response to multiple anesthetic drug classes as well as neurocognitive deficits of long-standing. In response to the weight of preclinical studies as well as suggestive retrospective data from humans, the US Food and Drug Administration issued a “Drug Safety Communication” in 2016 warning that anesthetic exposures greater than 3 hours may impact neuronal development in children less than 3 years of age as well as fetuses in their 3rd trimester [1]. Several of my colleagues at the time were taken aback by this warning, concerned that the statement was premature and the data, concerning in animals but far from definitive in humans, did not merit a public warning.
The limitations of the human data remain numerous and have been subjected to much skepticism [2]. To name only a few, these limitations include the retrospective nature of much of the evidence, comparatively weak hazard ratios, and, perhaps most importantly, the inability to establish a definitive effect in prospective clinical studies. Not surprisingly, preclinical animal data have been subject to comparatively less skepticism by the medical community. In part, this is due to that fact that anesthesiologists, the physician group most-directly impacted by the warning, are largely clinicians. To use a colloquialism, while comfortable “getting into the weeds” to scrutinize the details of clinical studies that impact the verity and applicability of the data to their patients, most anesthesiologists are far removed from animal laboratories and often take the conclusions of such studies at face value. Moreover, limitations inherent to certain animal models are infrequently discussed even by those who work in such a laboratory.
To use the day 7 mouse as an example, one of the most common animal models used to establish anesthetic neurotoxicity, it is worth pondering how pup mortality might impact data interpretation. Consider the reproduction dynamics of the mouse. A typical mouse dam can give birth to ~7 pups at a time, can become pregnant within 24 hours following parturition, and can produce as many as 15 litters a year [3]; in such an animal, there is little evolutionary value of “the one.” Thus, the highly altricial day 7 mouse pup, dependent upon its mother for survival, often does not. As many as 50% of mouse pups have been reported not to survive to maternal weaning (~3-4 weeks of life) [4]. Many are thought to be actively killed, cannibalized, or die by neglect, which is a concern in any experiment that requires the dam to assist in recovery of the pup—unhealthy pups, those that have been handled, or those that have been experimented upon may not be accepted by their dam thereafter [5]. As such, a full accounting of such mortality is important for the reader. For example, if one were to conduct an experiment where a high percentage of experimental mice die after conclusion of the experiment, what might that say about the pups that survived? Are data from survivors a direct reflection of the experimental question, or are they secondary to the quality of their recovery? Did the experimental protocol treat control animals in a manner such that maternal acceptance could be evaluated? What should one conclude if such data are not reported? Of note, the above is but one variable in a single, relatively simple, small animal model.
Enter the task at hand; the paper by Gao et al attempts to scrutinize the preclinical literature, focusing on data from the most sentient and controversial of all animal models – the non-human primate (NHP) [6]. They conducted a systematic review of the literature utilizing NHPs to study anesthetic neurotoxicity, focusing on the “rigor of design, execution, and reporting.” Looking at over 15 year’s worth of data from 23 studies, they assessed data quality on the basis of adherence to ASA basic monitoring requirements as well as a standardized 38-point checklist from the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines. Originally published in 2010 and updated in 2020 [7], the ARRIVE guidelines are the result of an international collaboration of animal researchers to ensure that animal studies report sufficient information to allow for interpretation. In short, the authors report critical failings regarding study design, reporting and execution. Their concerns were far-reaching, ranging from the relatively benign “documentation of anesthesia provider” to more serious concerns related to absence of blinding, protocol violations resulting in comingling of ventilated and non-ventilated animals, as well as the reporting of adverse events such as failed intubations. In sum, the authors report that, despite being present for over a decade, approximately 66% of the ARRIVE guideline criteria were adhered to in the NHP studies, similar to that described in the literature at large [8]. Despite their endorsement by many leading institutions and journals, acceptance of the ARRIVE guidelines is not universal. Indeed, while many prominent basic science journals (such as Nature and Cell) have endorsed the ARRIVE guidelines, many clinical journals (such as Anesthesiology and Anesthesia & Analgesia) [9] have not. As journals such as these frequently publish our most translational ideas, it is imperative that reporting standards for animal data are comprehensive, consistent, and clear for the benefit of the researcher, the reader, and the public at large. Until there is consistency for reporting standards between such journals, one can expect controversary surrounding data content and quality to abound.
References
1. FDA. FDA Drug Safety Communication: FDA review results in new warnings about using general anesthetics and sedation drugs in young children and pregnant women. Secondary FDA Drug Safety Communication: FDA review results in new warnings about using general anesthetics and sedation drugs in young children and pregnant women 2016. https://www.fda.gov/drugs/drug-safety-and-availability/fda-drug-safety-communication-fda-review-results-new-warnings-about-using-general-anesthetics-and.
2. Nemergut ME, Aganga D, Flick RP. Anesthetic neurotoxicity: what to tell the parents? Paediatr Anaesth 2014;24(1):120-6 doi: 10.1111/pan.12325[published Online First: Epub Date]|.
3. Katherine E. Quesenberry D, MPH, DABVP (Avian);, Thomas M. Donnelly B, DVP, DACLAM, DABVP(ECM). Breeding and Reproduction of Mice
. Secondary Breeding and Reproduction of Mice
4. Weber EM, Algers B, Würbel H, Hultgren J, Olsson IA. Influence of strain and parity on the risk of litter loss in laboratory mice. Reprod Domest Anim 2013;48(2):292-6 doi: 10.1111/j.1439-0531.2012.02147.x[published Online First: Epub Date]|.
5. Leidinger CS, Thöne-Reineke C, Baumgart N, Baumgart J. Environmental enrichment prevents pup mortality in laboratory mice. Lab Anim 2019;53(1):53-62 doi: 10.1177/0023677218777536[published Online First: Epub Date]|.
6. Gao F, Wahl JA, Floyd TF. Anesthesia and neurotoxicity study design, execution, and reporting in the nonhuman primate: A systematic review. Paediatr Anaesth 2022;32(4):509-21 doi: 10.1111/pan.14401[published Online First: Epub Date]|.
7. Kilkenny C, Browne WJ, Cuthill IC, Emerson M, Altman DG. Improving bioscience research reporting: the ARRIVE guidelines for reporting animal research. PLoS Biol 2010;8(6):e1000412 doi: 10.1371/journal.pbio.1000412[published Online First: Epub Date]|.
8. Leung V, Rousseau-Blass F, Beauchamp G, Pang DSJ. ARRIVE has not ARRIVEd: Support for the ARRIVE (Animal Research: Reporting of in vivo Experiments) guidelines does not improve the reporting quality of papers in animal welfare, analgesia or anesthesia. PLoS One 2018;13(5):e0197882 doi: 10.1371/journal.pone.0197882[published Online First: Epub Date]|.
9. NC3Rs. Journals. Secondary Journals. https://arriveguidelines.org/supporters/journals.