Methods: We identified all adult HIV+ (n=226) and HIV- (n=86,321) Kaiser Permanente Northern California members with a first known admission for ACS, defined as ST elevation myocardial infarction (STEMI), non-ST elevation MI (NSTEMI), or unstable angina, between 1996-2010. Deaths were ascertained from hospital records, California death certificates, and Social Security Administration vital status files. 1- and 3-year all-cause mortality by HIV status was assessed by Kaplan-Meier. Adjusted hazard ratios (HR) for death by HIV status were obtained from Cox models adjusting for baseline age, sex, race, diagnosis year, admitting diagnosis, ever smoking status, and LDL, HDL, and TG levels. Finally, adjusted HRs were obtained for HIV+ patients stratified by CD4 and antiretroviral therapy (ART) class compared with HIV- patients.
Results: HIV+ patients were younger (54.6 vs. 67.4 years), more likely male (94.4% vs. 62.5%), and smoked more (71.5% vs. 55.3%) than HIV- patients. The admitting diagnosis for HIV+ vs. HIV- patients was more often STEMI (37.0% vs. 26.8%), and less often unstable angina (28.1% vs. 39.1%). Unadjusted survival after the ACS hospitalization was similar for HIV+ compared with HIV- patients at 1 year (92.4% vs. 90.2%; P=0.26) and 3 years (83.4% vs. 81.4%; P=0.43). However, being HIV+ was associated with higher adjusted HRs for death at 1 year (HR=2.2; 95% CI=1.3-3.5) and 3 years (HR=2.5; 95% CI=1.8-3.5). At 3 years, HIV+ patients with CD4≥500 cells/μL had similar mortality compared with HIV- patients with an HR of 0.7 (95% CI=0.2-2.1), while mortality was elevated for HIV+ patients with CD4 201-499 and CD4<200, with HRs of 2.5 (95% CI=1.6-4.0) and 5.6 (95% CI=3.2-9.7), respectively. HRs were similar for HIV+ patients compared with HIV- patients regardless of ART class.
Conclusion: Within a large, community-based sample of adults hospitalized for ACS, HIV patients were at greater risk of all-cause mortality, although no difference in mortality was observed at 3 years for HIV+ patients with high CD4+ cell counts. Future studies should investigate other outcomes, including ACS recurrence and CVD-specific mortality.
A. Prasad, None
J. Zaroff, None
D. Klein, None
M. Horberg, None
A. Go, None
J. Lo, None