Covert Perioperative Stroke: Hiding in Plain Sight

To the Editor –

Cui and colleagues recently reported the results of a large, prospective cohort study demonstrating an alarming 11.9% incidence of perioperative covert stroke in non-cardiac surgical patients.¹ Moreover, participants experiencing a stroke were more likely to experience early postoperative delirium and lower scores on telephone-based neurocognitive assessments one year later. These results align with the international NeuroVISION study,² and the reproducibility of these findings increases confidence in the results. This astonishingly high stroke incidence should serve as a “call to arms” for anesthesiologists to identify modifiable risk factors under our purview.

In this context, anesthesiologists may be interested in intraoperative physiologic contributions to cerebral hypoperfusion and stroke. The study team did perform a post hoc analysis to test the associations between hypotension, defined as mean arterial pressure (MAP) <65 mmHg, and subsequent stroke (odds ratio 1.30 [95% CI 0.65 – 2.59], p=0.464). Additional clarity would be helpful with respect to the hypotension definition. It is not clear whether this definition meant any occurrence of MAP <65 mmHg, or if the mean MAP for the entirety of the case needed to be <65 mmHg. This is relevant because there is emerging evidence that stroke risk may increase with cumulative time and depth below a given hypotension threshold.³ If possible, within the study dataset, it would thus be informative to test associations between stroke and both time and depth (i.e., area under the curve) below a MAP of 65 mmHg. Indeed, while no significant association was found in the current analysis, it is worth noting that the upper bound of the confidence interval has a >4-fold distance above 1 (1.59) compared to the lower bound distance below 1 (0.35).⁴ Likewise, while mean arterial partial pressure carbon dioxide values were similar between patients with and without a stroke, area under the curve analyses may be more informative than descriptive, bivariable comparisons of means.

Perioperative covert stroke is a problem worthy of our attention, and the authors are to be commended for conducting and following through with this study, particularly given the coronavirus-related challenges. Hopefully studies like this will prompt clinical and scientific action to address this alarming yet insidious issue that has perhaps been hiding in plain sight.