Background COVID-19 has demonstrated its potential effects spread well beyond respiratory symptoms. Large reviews of patients presenting with COVID have demonstrated increases in risk for acute ischemic events. These are not limited to the extracranial space, with potential links to ischemic stroke as well. While the cause is likely multifactorial, neuro-interventionalists must determine best practice treating acute ischemic stroke in COVID patients. Sequelae of the viral illness may challenge established guidelines in choosing medical management versus endovascular thrombectomy (EVT) for these patients. Here we perform a matched analysis of the National Inpatient Sample (NIS), comparing potential functional outcome differences between patients treated with EVT versus medical management for ischemic stroke when concomitantly presenting with COVID.

Methods Weighted discharge NIS data from March through December of 2020 was surveyed for patients admitted with primary International Classification of Disease version 10 (ICD-10) diagnosis of ischemic stroke, COVID-19, and a reported National Institute of Health Stroke Score (NIHSS). Due to the non-random nature of the NIS sampling design, we utilized nearest neighbor matching to adjust confounding between patient undergoing medical management and EVT based on age, gender, comorbidities, and NIHSS of the patients. Univariate and multivariate logistic regression was performed, accounting for the sampling design of the NIS. We estimated differences in favorable functional outcomes defined as a routine discharge or discharge to a rehabilitation hospital in patients undergoing EVT or medical management for ischemic stroke when also diagnosed with COVID.

Results 990 patients were analyzed. Two matched groups were created with 495 undergoing EVT and 495 treated by medical management. After nearest neighbor matching, there were no differences in age, comorbidities, gender or NIHSS. Patients treated with EVT had higher absolute, although nonsignificant, rates of favorable functional outcomes compared to those managed medically (19.19% vs. 10.10%, p = 0.090; OR, 2.11 [95% CI, 0.87-5.10]; p = 0.095). TPA was utilized at a higher rate in patients receiving EVT compared to medical management (35.35% vs. 17.17%, p = 0.005). There were higher rates of intracerebral hemorrhage (ICH) in the EVT group (20.20% vs. 8.08%, p = 0.025). Inpatient mortality rates were not different between the two groups (20.20% vs. 18.18%, p = 0.730).

Conclusions In this matched NIS analysis of patients treated for ischemic stroke with a concomitant COVID diagnosis, we demonstrate that EVT trends towards, although nonsignificant, higher rates of favorable functional outcomes when compared to medically managed patients.

Disclosures C. Heskett: None. L. Fry: None. A. Brake: None. F. De Stefano: None. K. Morey: None. C. Lei: None. A. Alkiswani: None. K. Le: None. J. Peterson: None. K. Ebersole: None.