Back

Mechanical Circulatory Assist Devices Stroke Subtype Classification: A Novel Stroke Classification System in Patients with Ventricular Assist Devices

Pinto, C. B.; Owens, C. D.; Alvarado-Dyer, R.; Loggini, A.; Ortiz Garcia, J. G.; Heldner, M. R.; Prabhakaran, S.; Saleh Velez, F. G.

2025-10-27 neurology
10.1101/2025.10.22.25338591 medRxiv
Show abstract

BackgroundLeft ventricular assist devices (LVADs) have emerged as the standard of care for bridging-destination therapy in refractory heart failure, however, due to their inherent characteristics, they carry a higher risk of cerebrovascular events. Despite the increased incidence of associated cerebrovascular events in this population, the underlying pathophysiological mechanisms remain poorly understood, hindering effective management and prognosis. We aimed to (1) develop and apply a novel etiological classification system for mechanical circulatory assist device (MCAD)-related strokes, and (2) evaluate the interrater reliability of this classification when applied to cerebrovascular events in LVAD recipients. MethodsThe MCAD stroke classification divides LVAD-related strokes in four categories: 1. Management-related (M); 2. Clinically related (C); 3. Acquired (A); 4. Device-related (D). We evaluated the interrater agreement between two independent board-certified vascular neurologists who classified strokes in LVAD patients based on the proposed MCAD classification. ResultsWe retrospectively reviewed 192 LVAD patients at the University of Chicago Medical Center and identified 32 cerebrovascular events from 2010 to 2020. Among those, 59.4% were ischemic and 40.6% hemorrhagic strokes. The interrater agreement was extremely high (intraclass correlation coefficient = 0.96 {+/-} 0.03). The raters agreed in the primary classification of 32 subjects with the following distribution of mechanisms: 62.6% M, 15.6% A, 15.6% D, 6.2% C. Four subjects fit into more than one category. ConclusionWe propose an MCAD-focused stroke classification that integrates etiology with therapeutic and prognostic decision making. Although evaluated in LVAD patients, this classification can potentially be extrapolated to other mechanical assist devices.

Published in Journal of Stroke and Cerebrovascular Diseases (predicted rank #2) · training set

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.