ABSTRACT
BACKGROUND AND PURPOSE: The coronavirus disease 2019 (COVID-19) outbreak has added challenges to providing quality acute stroke care due to the reallocation of stroke resources to COVID-19. Case series suggest that patients with COVID-19 have more severe strokes; however, no large series have compared stroke outcomes with contemporary non-COVID-19 patients. Purpose was to analyze the impact of COVID-19 pandemic in stroke care and to evaluate stroke outcomes according to the diagnosis of COVID-19. METHODS: Retrospective multicenter cohort study including consecutive acute stroke patients admitted to 7 stroke centers from February 25 to April 25, 2020 (first 2 months of the COVID-19 outbreak in Madrid). The quality of stroke care was measured by the number of admissions, recanalization treatments, and time metrics. The primary outcome was death or dependence at discharge. RESULTS: A total of 550 acute stroke patients were admitted. A significant reduction in the number of admissions and secondary interhospital transfers was found. COVID-19 was confirmed in 105 (19.1%) patients, and a further 19 patients were managed as suspected COVID-19 (3.5%). No differences were found in the rates of reperfusion therapies in ischemic strokes (45.5% non-COVID-19, 35.7% confirmed COVID-19, and 40% suspected COVID-19; P=0.265). However, the COVID-19 group had longer median door-to-puncture time (110 versus 80 minutes), which was associated with the performance of chest computed tomography. Multivariate analysis confirmed poorer outcomes for confirmed or suspected COVID-19 (adjusted odds ratios, 2.05 [95% CI, 1.12-3.76] and 3.56 [95% CI, 1.15-11.05], respectively). CONCLUSIONS: This study confirms that patients with COVID-19 have more severe strokes and poorer outcomes despite similar acute management. A well-established stroke care network helps to diminish the impact of such an outbreak in stroke care, reducing secondary transfers and allowing maintenance of reperfusion therapies, with a minor impact on door-to-puncture times, which were longer in patients who underwent chest computed tomography.