LAUSR

Stroke can be a devastating complication of surgery, with few effective strategies for detection or prevention. Overt stroke (ie, stroke with clinically apparent manifestations) occurs in approximately 0·1–1·0% of patients undergoing non-cardiac surgery and is associated with a higher mortality than that in patients who did not have a stroke. In the non-surgical setting, covert stroke is more common than overt stroke and is associated with cognitive decline. However, the incidence and sequelae of covert stroke in the perioperative setting remain unknown. Furthermore, current neuromonitoring strategies and serum biomarkers are not useful for identifying cerebrovascular vulnerability or injury. Thus, covert perioperative stroke could represent an insidious threat to brain health and there is an esssential need to advance our understanding of incidence and pathogenesis to ultimately inform preventive strategies. To address these issues, the NeuroVISION investigators did a prospective, observational, multicentre, international study of 1114 patients (mean age 73 years, 43% women, 57% men) undergoing non-cardiac surgery to characterise covert stroke and related outcomes, with a particular focus on cognitive decline, reported in The Lancet. Covert stroke was detected by MRI in the postoperative period and cognitive decline (the primary outcome) was assessed at 1 year, as defined by a decrease of 2 points or more on the Montreal Cognitive Assessment (MoCA) with reference to the preoperative baseline. NeuroVISION built on the findings of a 100-patient proof-of-concept study that found a surprising 10·0% (95% CI 5·5–17·4) incidence of perioperative covert stroke. In reproducing this finding in a large, multicentre, international study, NeuroVISION’s foundational contribution is to confirm, with precision by neuroimaging criteria, that there is indeed a very high incidence (7% [95% CI 6–8]) of perioperative covert stroke in patients aged 65 years and older presenting for non-cardiac surgery. In fact, this incidence is an order of magnitude higher than the incidence of overt stroke in patients at moderate risk undergoing non-cardiac surgery. Furthermore, nearly a quarter of enrolled patients had chronic brain infarction based on MRI findings. In terms of the primary outcome of the study, cognitive decline (as determined by MoCA) 1 year after surgery was identified in 29 (42%) patients undergoing surgery with covert stroke and 274 (29%) patients undergoing surgery without (adjusted odds ratio 1·98, 95% CI 1·22–3·20, absolute risk increase 13%). Methodological strengths of the study encourage confidence in the results related to stroke. This was an international, multicentre study that used a proportional sampling strategy for balancing patient representation across sites. This study design confers high external validity. Stroke was also diagnosed objectively, with MRI and independent evaluation by neuroradiologists. Patients, clinicians, research teams, and outcome adjudicators were masked to neuroimaging results. Some limitations also warrant discussion, particularly regarding the cognitive function assessment. Although the MoCA is a widely used screening tool for detecting cognitive impairment, it was not designed for diagnostic use after stroke, and concerns have been raised regarding reliability and validity in this setting. On the basis of the original strategy described by Tan and colleagues, a MoCA score decrease of 2 or more is associated with a positive predictive value of 20% with respect to the clinically relevant decline based on neuropsychological testing. Thus, false positives were likely to be present, which weakens the conclusion that perioperative covert stroke carries a cognitive cost. Additionally, as the authors mention, the absence of a non-surgical control group also precludes more definitive conclusions regarding cognitive trajectory. This aspect of the study design is particularly relevant to the primary outcome because the low preoperative MoCA scores in many patients suggest that they were probably already declining cognitively before their surgery. Apart from the possible effects on cognition, it would be highly informative to patients, family members, and clinicians to know whether perioperative covert stroke might impact quality of life, symptoms of depression, and patient functionality. These outcomes were ascertained, but they were not reported in the current manuscript. Overall, the NeuroVISION study suggests that there could be neuropathological contributors to postoperative cognitive decline, in contrast to past hypotheses focusing on anaesthetic drugs or surgical insult. In the NeuroVISION study, the type of anaesthetic or surgical risk was not associated with cognitive decline. Given the ageing surgical population and the paucity of validated Published Online August 15, 2019 http://dx.doi.org/10.1016/ S0140-6736(19)31770-2