Incidence of Infection and Antimicrobial Consumption in Ventricular Assist Device (VAD) Recipients at the Prince Charles Hospital (TPCH): A Retrospective Analysis

Abstract

Ventricular assist devices (VADs) are frequently used as a bridge to heart transplant; however, infections are a common cause of increased morbidity and mortality. The optimal prophylactic antimicrobial regimen has not been effectively evaluated in literature. Forty-three (43) patients received a VAD over the 5-year study period (2012-2017) at The Prince Charles Hospital (TPCH), Brisbane Australia. Of these, 41 patients were followed from implantation until transplantation or death. Antimicrobial prophylactic regimens and individual episodes of infection were recorded. The infection profiles, including types and incidence were compared to published literature using definitions from the International Society for Heart and Lung Transplantation (ISHLT) guidelines for consistency. Median duration of VAD insertion was 79 days (IQR: 36-167). Patients received aztreonam, fluconazole and vancomycin (median duration 8 days). Twenty-two (22) (53.6%) patients experienced a VAD-specific and/or a VAD-related infective episode. Incidence of infection in the study cohort was 0.60 infections per 100 patient days. Thirteen (13) patients (31.7%) experienced 16 VAD-specific infections which were all driveline infections. Thirteen (13) patients (31.7%) experienced 14 VAD-related infections. The predominant VAD-related infection type was bacteraemia (36%). Predominant bacterial profiles of VAD-specific as well as VAD related infections were gram positive. Only three episodes had a gram negative as a causative pathogen which occurred much later post VAD insertion. Median time till VAD-specific or VAD-related infection was 46 and 15 days respectively. Obesity was significantly associated with increased risk of infection (HR: 3.2; 95% CI: 1.3-7.4). Infection is a common complication of VAD implantation. In our study population gram positive bacteria were the predominant causative pathogen. Based on the micro-organism profile there may be scope for a narrowing of the antibiotic regimen. A larger, multicentre study would be able to accurately guide a change. The information gathered in our study offers a strong foundation for such a multicentre study.