Presenting Author:

Jes Sanders, B.S.

Principal Investigator:

Matthew Feinstein , M.D.

Department:

Medicine

Keywords:

Atrial fibrillation, atrial flutter, human immunodeficiency virus, health disparities, risk factors

Location:

Ryan Family Atrium, Robert H. Lurie Medical Research Center

C37 - Clinical

Atrial Fibrillation and Flutter in Persons with Human Immunodeficiency Virus

Background Persons infected with human immunodeficiency virus (HIV) have elevated risks for various manifestations of cardiovascular disease (CVD) including myocardial infarction, arrhythmia, heart failure, and sudden cardiac death. Atrial fibrillation (AF) is the most commonly diagnosed arrhythmia, but no studies to our knowledge have compared AF and atrial flutter (AFL) prevalence and associated risk factors for HIV+ persons and uninfected controls. Methods Patients from an electronic data repository of 5,052 HIV+ patients and 10,121 uninfected controls (frequency-matched 1:2 on demographics, zip code, and clinic location) receiving care at an urban academic medical center were screened for possible AF/AFL using ICD-9, ICD-10, and CPT codes. Physician chart review subsequently adjudicated AF/AFL based on electrocardiographic evidence and/or physician notes. Prevalence of AF/AFL was compared for HIV+ versus uninfected persons and among HIV+ persons across levels of immunodeficiency. Multivariate analyses were used to determine associations of known CVD risk factors with prevalent AF/AFL for HIV+ persons. Results There were 101 confirmed AF/AFL cases among 5,052 HIV+ patients (2.00%) and 159 confirmed AF/AFL cases among 10,121 uninfected controls (1.57%) [Odds Ratio (OR) 1.28, 95% Confidence Interval (CI) 0.99-1.64; p=0.056] (Figure 1). The association between HIV status and AF/AFL was attenuated after adjustment for demographics and CVD risk factors and qualitative effect modification was observed between HIV status and hypertension; among non-hypertensive patients, HIV+ patients were 1.55 times more likely than uninfected patients to have AF/AFL (OR 1.55, 95% CI 0.97-2.51), whereas among hypertensive patients, HIV+ patients were less likely than uninfected patients to have AF/AFL (OR 0.70, 95% CI 0.52-0.95) (Figure 2). Compared with HIV+ persons, those with nadir CD4+ T cell count <200 cells/mm3 were significantly more likely than those with nadir CD4+ T cell count ≥500 to have AF/AFL after adjustment for demographics and CVD risk factors (OR 3.18, 95% CI 1.22-8.23; p=0.02) (Figure 1). Conclusion AF/AFL appear to be more common among HIV+ persons compared with uninfected matched controls. The association between HIV status and AF/AFL may be modified by hypertension. One potential explanation for this finding is HIV+ patients who are diagnosed with AF/AFL may often have acute and/or critical illnesses as AF/AFL triggers and may not necessarily have diagnoses of hypertension; meanwhile, uninfected patients with AF/AFL may be more likely to have long-standing hypertension and traditional CVD risk factors as the trigger for AF/AFL. This study represents the first to our knowledge to compare prevalence and risk factors associated with physician-adjudicated diagnoses of AF/AFL for HIV+ and uninfected persons. Accordingly, our findings require validation in subsequent analyses of large cohorts of HIV+ and matched uninfected control patients.