Intravenous Iron or Red Blood Cell Supplementation and Risk of Serious Infection in Left Ventricular Assist Device Patients

Introduction Patients with durable left ventricular assist devices (LVADs) are at high risk for serious infection and associated mortality. Furthermore, these patients also have a high prevalence of iron deficiency due to ongoing hemolysis, increased propensity for gastrointestinal bleeding, and dysregulated iron metabolism that accompanies mechanical circulatory support and heart failure. Intravenous (IV) iron-related supplementation [IVIRS, including IV iron infusion or red blood cell transfusion (RBC)] has been hypothesized to increase risk of infection. Hypothesis IVIRS (IV iron infusion or RBC transfusion) is associated with an increased risk of serious infection in LVAD patients. Methods This is a single center, retrospective, observational cohort study of 211 consecutive adult patients [median age 57.2 years (25th-75th percentile: 48.7-65.7 years), 19% female] who were discharged and followed after de-novo LVAD implantation between May 2008 and December 2019. Patients were followed until either end of the study period, or occurrence of death, heart transplantation, or LVAD removal/replacement, whichever came first. All administration of IVIRS and diagnosis of serious infection after discharge from index hospitalization was abstracted from the electronic medical record. Serious infection was defined as the first incident infection requiring hospitalization, divided into LVAD-associated infections (pump/cannula, pocket or driveline infection) and non-LVAD infections. Multivariate logistic regression was used to examine the association between IVIRS and serious infections. Results Over a median follow-up of 15.2 months (25th-75th percentile: 6.9-29.4 months), 69 (33%) LVAD patients received IVIRS, of whom 47 (68%) received IV iron and 44 (64%) received RBC transfusion. During follow-up, 87 (41%) patients suffered a serious infection, of which 48 (55%) were LVAD-associated infections and 39 (45%) were non-LVAD infections. After adjustment for age, sex, diabetes, chronic kidney disease, prior cardiac surgery, destination therapy and pre-IVIRS serious infection, LVAD patients who received IVIRS had 2.85 times higher odds of any first serious infection compared to those who did not receive IVIRS during the study period (95% CI 1.49-5.46, p=0.002). Patients who received IVIRS had a significant 3.03 times greater odds of non-LVAD infection (95% CI 1.37-6.70, p=0.006) and a non-significant 1.54 times greater odds of LVAD infection (95% CI 0.75-3.18, p=0.24), compared to LVAD patients who did not receive IVIRS. Conclusion IVIRS use is associated with a higher risk of serious infection in LVAD patients, primarily due to serious non-LVAD infections. These findings merit further study in larger cohorts and randomized trials to determine best practices surrounding IVIRS in LVAD patients.