Second International Perioperative Neurotoxicity Workshop Summary

Perioperative neurotoxicity continues to garner considerable interest from the medical community and our patients. Accordingly, a 1-day conference was held in Toronto, Canada on June 15, 2010 for investigators from multiple perspectives to share current knowledge on this important topic. At the conference, 20 speakers addressed the effects of perioperative events on the neonatal to the aging brain with data derived from both fundamental and clinical protocols. Evidence in support of a correlation between surgery and subsequent neurophysiological changes has accumulated,1–12 but at present, causality cannot be concluded at either extreme of age. Nor have any molecular, cellular, or pathophysiological steps linking perioperative events with cognitive outcomes been discerned in human data. This nascent status is attributable in part to study design, elements of which are now being addressed in several ongoing and planned multicenter studies in children in conjunction with the Food and Drug Administration (SmartTotsa and the GAS study b) and in adults in association with Alzheimer Disease Research Centers. Challenges include the choice and standardization of the cognitive domain to be tested, the timing of testing, and controlling for anesthetic, surgery, intercurrent medications, and comorbidities. Despite these challenges, clinical studies of the effects of neonatal and perinatal surgery on cognition are likely to yield answers sooner than those at the other extreme of age, where issues of test timing become more important and are less well understood. For example, late-onset Alzheimer disease may take decades to reach the point at which cognitive symptoms first appear, and modulation at this point may not be possible. Thus, the vulnerable period is not well defined, nor is the expected time course of symptom appearance. This may explain why clinical studies on the influence of perioperative factors on late-onset Alzheimer disease onset and progression have been either negative or inconclusive to date.13–17 However, the clinical entity called postoperative cognitive dysfunction in the elderly is fairly well recognized,6,7,9,14,18–20 but much, including causation and possible relationships to other forms of age-related dementias, remains entirely unknown. Retrospective and prospective studies are currently in progress, some using established Alzheimer Center databases. Perioperative biomarker studies, only a few of which have been reported,10,18,21,22 may help detect neurotoxicity in the elderly, but the technology is still unproven (imaging biomarkers) or invasive (cerebrospinal fluid biomarkers). Genetic testing (apolipoprotein E4 [ApoEe4], presenilin-1 [PS-1], translocase of outer mitochondrial membrane 40 homolog [TOMM40])23 may also allow focus on the most vulnerable groups of patients to reveal interactions more quickly. A host of mechanistic studies have been helpful in providing a rationale for the overall concern of perioperative neurotoxicity, teasing apart the causalities, refining hypotheses, and suggesting clinical strategies to test for the problem in patients. These have ranged from cell culture to histopathology to animal behavioral studies, including the nonhuman primate.16,24–36 Although most such studies have examined effects of the anesthetic in isolation, a few are beginning to incrementally model other aspects of the perioperative scenario, including surgery. The stimulation of inflammatory cascades by even uncomplicated orthopedic or abdominal surgery may accelerate or induce pathogenesis, especially in a vulnerable brain.22,37–39 It is fair to say at this point that considerable experimental evidence provides plausibility for anesthesia and surgery to independently cause durable cognitive decline after either neonatal or elderly exposure, and that many molecular mechanisms may be operant. Some evidence points to potency differences among general anesthetics for the different mechanisms of toxicity. Although no halogenated agent has consistently emerged as less “neurotoxic” in these mechanistic studies, the more target-specific injectable anesthetic adjuvants such as dexmedetomidine,40 or simple anesthetic gases such as xenon,41 may be lower risk. However, μ-opioid agonists seem to have potential for specific hippocampal toxicity.42 To complicate interpretation further, general anesthetics might be both protective and toxic,43,44 depending on a host of preexisting, or concurrent cellular vulnerabilities, anesthetic type, anesthetic concentration/dose and exposure duration, and factors such as opiates and other drugs, hypoxia, hypercarbia, and hypothermia. These mechanistic studies have also provided for potential cotherapies to antagonize “neurotoxicity,” such as dantrolene,45 lithium,46,47 carnitine,48,49 and anakinra.50 Most investigators are convinced that perioperative neurotoxicity occurs, especially in the setting of a brain made vulnerable by developmental or pathological processes or perhaps by genomic predisposition, but that causality is presently unclear and probably multifactorial. For now, identification and characterization of a clinical phenotype of durable cognitive sequelae, caused by surgery, at any age remains a major challenge for the field of perioperative research. Until then, clinicians should heed the Food and Drug Administration's Anesthesia and Life-Support Drugs Advisory Committee assessment of March 2007: “… there are not adequate data to extrapolate the animal findings to humans …” and “… existing and well-understood risks of anesthesia continue to be the overwhelming considerations in designing an anesthetic, and the understood risks of delaying surgery are the primary reasons to determine the timing.”c Anesthesiologists and surgeons often inform their patients of potential complications that may in fact be very rare, e.g., death. Presently available clinical data, incomplete as they may be, indicate that deleterious postoperative cognitive sequelae may not be rare. Fully informed consent requires that significant risks of particular importance to the patient be disclosed, and even transient alterations in cognition might be considered important by the patient, particularly after elective procedures for non–life-threatening conditions. Indeed, many patients and their families are already aware of the issue, and may believe, or have been told (not necessarily by anesthesiologists), that it is due to the anesthetic. Thus, a clear and consistent message needs to be understood and articulated by anesthesiologists during the consent process: altered cognitive states after surgery have been reported, and may last for weeks to months or longer. The incidence and causes of these changes are unknown, and probably vary depending on both patient-specific and procedural characteristics. We must also emphasize that further research is urgently needed to determine which patients and procedures incur greater risk of subsequent cognitive effects, whether in the young or the elderly, and what steps may be taken to ameliorate the risk. Perioperative Neurotoxicity Working Group Avidan MS, Baranov D, Bickler PE, Cibelli M, Crosby GJ, Crowder CM, Culley DJ, Daiello LA, Deiner S, Eckenhoff MF, Eckenhoff RG, Evers AS, Flick RP, Gunstad J, Hogan KJ, Jankowski CJ, Jevtovic-Todorovic V, Johns RA, Kofke WA, Maze M, Miller KW, Mintz C, Olney JW, Orser BA, Patel PM, Palotás A, Perouansky M, Planel E, Saager L, Sanders RD, Sessler DI, Silverstein JH, Slikker W, Soriano SG, Sprung J, Stapelberg F, Stratmann G, Warner DS, Wei H, Whittington RA, Williams JW, Xie Z, Zuo Z. Roderic G. Eckenhoff, MD Department of Anesthesiology and Critical Care University of Pennsylvania School of Medicine Philadelphia, Pennsylvania [email protected]

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.