877    

Background: In the pre-HAART era, infective endocarditis has been shown to occur more frequently in HIV⁺ injection drug users (IDU) than in HIV^- IDU.  Few data exist on the incidence of and risk factors for IE in HIV patients in the post-HAART era

Methods: We assessed the incidence of  first episodes of infective endocarditis in a large Maryland clinical cohort between 1990-1995 and 1996-2002 in which comprehensive demographic, clinical and therapeutic data have been longitudinally collected. A nested case control analysis was used to assess risk factors associated with infective endocarditis between 1996-2002.  4 non- infective endocarditis controls were randomly selected from the overall cohort for each case  Controls were matched on cohort enrollment date and duration of follow-up. Multivariate conditional logistic regression analyses were used to assess risk factors for infective endocarditis.

Results: Of 3995 patients followed after January 1, 1990, there were 57 infective endocarditis episodes  between January 1, 1990 and December 31, 1995, for an incidence rate of 20.9 infective endocarditis /1000 person-years and 92 episodes between January 1, 1996 and December 31, 2002 for an incidence of 7.5 infective endocarditis events per 1000 person-years (p <0.001). For first cases of infective endocarditis between 1996 and 2002, factors associated (p <0.05) with having a infective endocarditis event included history of IDU (adjusted odds ratio [AOR] 10.3, 95% CI 4.1, 26.0), female gender (AOR 2.81, 95% CI 1.37, 5.76), CD4 count <50 cells/mm³ at infective endocarditis episode (AOR 2.29, 95% CI 1.01, 5.22), and viral load >100,000 copies/mL at infective endocarditis episode (OR 3.96, 95% CI 1.48, 10.6).  There was a non-significant association of age>40 years (AOR 1.89, 0.95, 3.75) and African American race (AOR 2.64, 95% CI 0.88, 7.98) with infective endocarditis episode. Hepatitis C⁺ (OR 4.43, 95% CI 2.41, 8.11), income <10K (OR 3.38, 95% CI 1.13, 10.1) were associated with infective endocarditis in bivariate analysis, and but not in multivariate analysis. Treatment with HAART, insurance, and homelessness were not associated with risk of infective endocarditis.

Conclusions:  In our urban cohort, the incidence rate of infective endocarditis decreased significantly in the post-HAART era. The discrepancy in the risk of infective endocarditis between HIV⁺ men and women is striking and should be a focus of future investigation.

Keywords: Endocarditis; Gender; Injection Drug Use