Background:Patients with implanted left ventricular assist device (LVAD) have been shown in multiple studies to have significantly increased bleeding rates, and yet independent risk factors for such complications remain poorly characterized.
Objective:This study sought to assess factors associated with increased bleeding risks after LVAD implantation.
Methods: The case-control study was represented by a retrospective electronic medical chart review for all adult patients above age 18, who had the LVAD implanted at Maimonides Medical Center from 2013-2018. The study comprised of a follow up period of 24 months after the LVAD implantation.
84 adult patients with implanted LVADs at Maimonides Medical Center were included in the study. The predictors among patients with any late postoperative bleeding (>7 days post-surgery) within the study period were compared to those without any events of interest. The outcome measure was a composite variable reflecting any bleeding event, such as upper and lower gastrointestinal (GI) bleed, intracranial hemorrhage or hemorrhage in any other organ. Patients were considered to have GI bleed if they had one or more of the following symptoms: guaiac-positive stool with hemoglobin drop >2g/dL, hematemesis, melena, active bleeding or blood within the GI tract at the time of endoscopy or colonoscopy. Intracranial or other organ bleeding was defined as appropriate clinical presentation and findings on imaging with hemoglobin levels reduced by more than or equal to 2 g/dl with no alternative explanation for anemia.
Logistic regression was used to create a multivariable model to identify predictors associated with an increased risk of all cause bleeding within 24 months after LVAD implantation.
Results: The study population consisted of 43 cases and 41 controls. Baseline characteristics were similar in both groups. A total of 43 (51%) patients had at least 1 episode of any type of bleed within study period.
Multivariable analyses showed that blood urea nitrogen (BUN) >20 before LVAD implantation and Creatinine >1.2 at the time of bleeding were significantly associated with all cause bleeding risk within 24 months after surgery with odds ratio (OR): 4.46, 95% confidence interval (CI): 1.78 to 11.15 and OR 3.55, 95% CI 1.13 to 11.15 respectively.
Conclusions: Perioperative BUN >20 (p=0.001) and postoperative Creatinine >1.2 (p=0.03) are associated with higher incidence of all cause bleeding within 24 months after surgery among patients with LVAD. Future studies should evaluate whether there are other factors accurately predicting bleeding events in order to improve outcomes.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.