Ischemic Stroke Occurs Less Frequently in Patients With COVID-19

A Multicenter Cross-Sectional Study

Kimon Bekelis, MD; Symeon Missios, MD; Javaad Ahmad, MD; Nicos Labropoulos, PhD; Clemens M. Schirmer, MD, PhD; Daniel R. Calnan, MD, PhD; Jonathan Skinner, PhD; Todd A. MacKenzie, PhD


Stroke. 2020;51(12):3570-3576. 

In This Article


Using a comprehensive all-payer cross-section of patients from a large healthcare system in New York State, we did not identify an increased likelihood of stroke on presentation among patients of all ages with COVID-19. Patients with stroke with concurrent SARS-CoV-2 infection demonstrated increased case-fatality and a trend more discharge to rehabilitation. Our results were robust when considering several observational techniques to account for confounders. We also did not identify an increased frequency of MI among patients with COVID-19, a disease with similar risk factors and epidemiology to stroke. Our report is contributing to an ongoing debate about possible systemic complications of SARS-CoV-2 infection.

These findings do not support the concern for a stroke epidemic in young adults,[2,3] fueled by a prior case study.[1] The authors of the latter report described their experience with five patients with stroke infected with SARS-CoV-2, aged 33 to 49. COVID-19 is a mild disease in most, but occasionally it progresses to acute respiratory distress syndrome, multiorgan dysfunction, cytokine storm, inflammation, coagulation, and death.[16,17] Proposed mechanisms contributing to strokes in this patient population include coagulopathy and endothelial dysfunction.[1] Although the authors shed light into the clinical characteristics of young adults with these 2 pathologies, they did not investigate a possible association between stroke and COVID-19.

Experience from other viral respiratory epidemics has uncovered a possible association of severe influenza with cardiovascular disease.[10] However, prior observational studies on this issue have offered mixed results. Some have demonstrated that admissions for stroke and heart disease are increasing during influenza epidemics.[11] These were supported by multiple small and large international observational studies demonstrating a protective effect of flu vaccination against cardiovascular disease, stroke, and all-cause mortality.[18] Exacerbation of chronic disease processes, and possible endothelial dysfunction, have been postulated as the contributing factors to some of these associations.[10,18] However, other extensive reports have not identified a protective effect of influenza vaccination on cardiovascular disease.[12]

In our analysis, we did not observe a positive association of COVID-19 and ischemic stroke. On the contrary, a sharp decrease in stroke admissions was observed around the peak of the pandemic. This is consistent with anecdotal international reports describing a phenomenon of "vanishing strokes and heart attacks" during this period.[4–6] The decrease was so severe in some countries that multiple public service announcements were released,[6] urging patients to seek immediate care for signs and symptoms of stroke.[7] Most stroke experts attributed this phenomenon to the unwillingness of patients to be exposed to SARS-CoV-2 in an overwhelmed emergency room.[6]

Italian researchers[4] have additionally hypothesized a pathophysiologic mechanism behind this decreased stroke occurrence, based on the controversial role of IL (interleukin)-6 in stroke.[19] There is experimental evidence that IL-6, which is elevated in severe COVID-19, has a neuroprotective effect and enhances angiogenesis.[19] Alternate explanations proposed by this group are based on the thrombocytopenia encountered even in patients with mild COVID-19.[20] It is likely that low platelets prevent the formation of large clots in the intracranial circulation.[20] Lastly, the widespread mitigation measures, which have minimized the prevalence of influenza in the community, could have decreased the negative impact of the flu on cardiovascular disease and stroke.[18,21] Further research into the cause of the observed associations is warranted.

Regardless of the decreased occurrence of stroke among patients with COVID-19, some strokes will undoubtedly occur in this population.[13–15] Although recent investigations have demonstrated that cardiovascular disease is a negative predictive factor of outcomes for COVID-19,[22] the inverse association has not been uncovered before. We identified a higher risk of unfavorable outcomes in patients with ischemic stroke infected with SARS-CoV-2, independent of traditional stroke risk factors and surrogates of stroke severity. We observed a mortality for COVID-19 positive patients with stroke to be 31.8% as compared with 4.6% for COVID-19–negative patients with stroke. We hypothesize this to be due to the respiratory impact of COVID-19 on an already compromised patient with stroke. This observation can offer invaluable insight to epidemiologists, and clinicians treating stroke in the frontline, especially in areas of high COVID-19 prevalence. This knowledge can be integrated into the recommendations of professional societies, who already have developed algorithms[5,23,24] to guide stroke management in the setting of the pandemic.

Our study has several limitations. Residual confounding can account for some of the observed associations. We used multiple sensitivity analyses and a wide selection of covariates to minimize this bias. Coding inaccuracies will undoubtedly occur and can affect our estimates, through misclassification. However, several reports have demonstrated that coding for stroke has shown nearly a perfect association with medical record review.[25,26] Coding for COVID-19 has not been validated yet, nevertheless, the creation of a specific code for this disease and the heightened awareness during the pandemic are expected to minimize coding inaccuracies. Although our data include all hospitals from a healthcare system that spans Nassau and Suffolk counties in NY State, some of the areas impacted the most during the COVID-19 pandemic, the generalization of this analysis to the entire US population is uncertain.

Additionally, we lack posthospitalization and long-term data on our patients. Quality metrics (ie, modified Rankin Scale), and clinical information on the functional status of the patients (National Institutes of Health Stroke Scale), which reflect stroke severity, were not available. However, using IV tPA and mechanical thrombectomy as surrogates for stroke severity can partially control for this bias, when assessing the impact of COVID-19 on stroke case-fatality. The definitive impact of COVID-19 on stroke outcomes can only be assessed in prospective registries. In this direction, the NeuroPoint Alliance has created the first module for a cerebrovascular registry, with results expected in the near future.[27] We also have not subdivided the observed strokes into subtypes, meaning it is possible that one particular cause of stroke may be a more frequent finding in patients with COVID-19. Finally, causality cannot be definitively established based on observational data, despite the use of advanced techniques.