Background Stroke systems of care employ a hub-and-spoke model, with fewer centers performing mechanical thrombectomy (MT) compared with stroke-receiving centers, where a higher number offer high-level, centralized treatment to a large number of patients.

Objective To characterize rates and outcomes of readmission to index and non-index hospitals for patients with ischemic stroke who underwent MT.

Methods This study leveraged a population-based, nationally representative sample of patients with stroke undergoing MT from the Nationwide Readmissions Database between 2010 and 2014. Descriptive, logistic regression analyses, and univariate and multivariate logistic regression models were carried out to determine patient- and hospital-level factors, mortality, complications, and subsequent readmissions associated with index and non-index hospitals' 90-day readmissions.

Results In the study, 2111 patients with a stroke were treated with MT, of whom 534 were readmitted within 90 days. The most common reasons for readmission were: septicemia (5.9%), atrial fibrillation (4.8%), and cerebral artery occlusion with infarct (4.8%). Among readmitted patients, 387 (74%) were readmitted to index and 136 (26%) to non-index hospitals. On multivariable logistic regression analysis, non-index hospital readmission was not independently associated with major complications (p=0.09), mortality (p=0.34), neurological complications (p=0.47), or second readmission (p=0.92).

Conclusion One-quarter of patients with a stroke treated with MT were readmitted within 90 days, and one quarter of these patients were readmitted to non-index hospitals. Readmission to a non-index hospital was not associated with mortality or increased complication rates. In a hub-and-spoke model it is important that follow-up care for a specialized procedure can be performed effectively at a vast number of non-index hospitals covering a large geographic area.