Background
People with HIV (PWH) have elevated risk for heart failure (HF). However, limited multi-center data exist on presentations and etiologies of HIV-associated HF.
Methods
We adjudicated incident HF events occurring between January 1, 2010 and December 31, 2023 at University of Washington (UW) and University of Alabama-Birmingham (UAB), which are two centers within the CFAR Network of Integrated Clinical Systems (CNICS). PWH in CNICS with possible HF were first identified by a screening protocol incorporating administrative codes and biomarkers of cardiac congestion. Individuals with HF prior to baseline were excluded. Two independent physician adjudicators reviewed clinical records to adjudicate events, with confirmed HF diagnosis requiring a combination of symptoms, physician diagnosis, and HF medication use. Adjudicators also determined: (1) HF subtypes (by left ventricular ejection fraction, LVEF) based on echocardiography closest in time to HF onset, and (2) presumed etiologies (e.g., ischemic and/or non-ischemic) based on review of physician notes, procedure notes, and comorbidities present at HF onset. Using Cox proportional hazard regression, hazard ratios and 95% CIs were used to examine associations of risk factors with incident HF.
Results
After excluding 50 PWH with confirmed HF at baseline, there were 212 PWH with incident adjuciated HF, of whom 106 (50%) had HF with reduced LVEF (HFrEF, LVEF<40%), 75 (35%) had HF with preserved LVEF (HFpEF, LVEF≥50%), 28 (13%) had HF with midrange LVEF (HFmrEF, LVEF ≥40% and <50%), and 3 (1%) had unknown LVEF classification. Regarding physician-determined HF etiology, 115 (54%) had nonischemic etiology, 36 (17%) had mixed ischemic and nonischemic etiology, 24 (11%) had ischemic etiology only, and 37 (17%) had unknown etiology. The most common nonischemic etiologies of HF were hypertensive and substance use, with hypertensive etiologies more common at UAB and substance use related etiologies more common at UW. Higher HIV viral load and lower CD4 count were associated with significantly higher incidence of adjudicated HF (Table) as were older age, smoking, hypertension, diabetes mellitus, history of myocardial infarction, and renal insufficiency.
Conclusions
Nonischemic etiologies of HF are common among PWH. HIV viremia, low CD4 T cell count, traditional CVD risk factors, and renal insufficiency were associated with higher risk of incident HF. Efforts to define HIV-specific presentations and etiologies of HF are needed.