Features of infective endocarditis in a contemporary cohort of persons who inject drugs: a matched comparison analysis of long-term prognostic factors.

Abstract

The objective of the study was to assess the short- and long-term mortality of infective endocarditis (IE) among people who inject drugs (PWID). Using prospectively collected data on hospitalized patients (years 2000 through 2021) with IE, PWID were identified and included in this study. Survival analysis was performed to analyze short- and long-term mortality and study their risk factors among PWID and a matched group of non-intravenous drug users (N-IDU). In a study of 485 patients admitted for IE, 55 (11%) of them were PWID. These PWID patients were 1:1 age- and sex- matched to an N-IDU group (N = 55 per group). Both groups had similar baseline comorbid conditions, including congestive heart failure, type 2 diabetes, and neoplastic diseases. However, PWID were more likely to have HCV co-infection (62% vs 16%, respectively, p < 0.001) and advanced liver disease/cirrhosis (52% vs 7.9%, respectively, p < 0.001). IE in PWID more often affected the tricuspid valve (42% vs 22%, respectively, p = 0.024) and presented with more embolic events (66% vs 35%, respectively, p < 0.01). S. aureus was the primary cause of IE in PWID (44% vs 21%, respectively, p = 0.01). After adjusting for other variables, PWID (HR = 2.99, 95% CI [1.06, 8.43], p = 0.038) and valve bioprosthetic replacement (HR = 5.37, 95% CI [1.3, 22.1], p = 0.02) were independently associated with increased mortality risk, whereas IE caused by tricuspid valve infection was associated with reduced mortality risk (HR = 0.25, 95% CI [0.06, 0.97], p = 0.046). In this cohort, PWID had increased risk of long-term mortality after hospital discharge for IE, when compared to matched N-IDU with similar baseline characteristics. The reasons behind the significant increase in mortality warrant further investigation.