Clinical Impact of Hematoma Expansion in Left Ventricular Assist Device Patients

• Published in: World neurosurgery 2020-11, Vol.143, p.e384-e390
• Main Authors: Melmed, Kara R., Mondellini, Giulio, Roh, David, Boehme, Amelia, Park, Soojin, Yuzefpolkya, Melana, Naka, Yoshifumi, Uriel, Nir, Agarwal, Sachin, Connolly, E. Sander, Claassen, Jan, Colombo, Paolo C., Willey, Joshua Z.
• Format: Article
• Language: English
• Subjects: Aged; Aged, 80 and over; Cerebral Hemorrhage - diagnostic imaging; Cerebral Hemorrhage - mortality; Cerebral Hemorrhage - therapy; Cohort Studies; Female; Heart Ventricles - diagnostic imaging; Heart-Assist Devices - adverse effects; Heart-Assist Devices - trends; Hematoma - diagnostic imaging; Hematoma - mortality; Hematoma - therapy; Hematoma expansion; Hospital Mortality - trends; Humans; Intracerebral hemorrhage; Left ventricular assist device; Male; Middle Aged; Prospective Studies; Retrospective Studies; Stroke
• ISSN: 1878-8750; 1878-8769
• DOI: 10.1016/j.wneu.2020.07.169

Hematoma expansion (HE) is associated with poor outcome in patients with intracerebral hemorrhage (ICH), but the impact on patients with an left ventricular assist device (LVAD) is unknown. We aimed to define the occurrence of HE in the LVAD population and to determine the association between HE and mortality. We performed a retrospective cohort study of LVAD patients and intentionally matched anticoagulated controls without LVAD admitted to Columbia University Irving Medical Center with ICH between 2008 and 2019. We compared HE occurrence between patients with an LVAD and those without an LVAD using regression modeling, adjusting for factors known to influence HE. We evaluated pump thrombosis following anticoagulation reversal. We examined the association between HE and hospital mortality using Poisson regression modeling adjusting for factors associated with poor outcome. Among 605 patients with an LVAD, we identified 28 patients with ICH meeting the study’s inclusion criteria. Our LVAD ICH cohort was predominantly male (71%), with a mean age of 56 ± 10 years. The median baseline hematoma size was 20.1 mL3 (interquartile range [IQR], 8.6–46.9 mL3), and the median ICH score was 1 (IQR, 1–2). There was no significant difference in occurrence of HE in LVAD patients and matched non-LVAD patients (adjusted odds ratio [OR], 1.3; 95% confidence interval [CI], 0.4–4.2). There was an association between HE and in-hospital mortality in LVAD patients (adjusted OR, 4.8; 95% CI, 1.4–6.2). HE occurrence appears to be similar in LVAD and non-LVAD patients. HE has a significant impact on LVAD ICH mortality, underscoring the importance of adequate coagulopathy reversal and blood pressure management in these patients.