Outcomes of cardiovascular magnetic resonance imaging in people living with HIV

Abstract

Funding Acknowledgements Type of funding sources: None. Background/Introduction People living with human immunodeficiency virus (HIV, PLWH) are at increased risk of cardiovascular disease (CVD). HIV infection and accelerated traditional risk factors due to highly-active antiretroviral therapy (HAART) are proposed mechanisms for increased rate of heart failure (HF). The pathophysiological drivers of myocardial dysfunction and worse cardiovascular outcome in HIV remain poorly understood. Purpose To examine prognostic relationships of cardiac imaging measures with cardiovascular outcome in PLWH on HAART. Methods This is a prospective observational longitudinal study using cardiac magnetic resonance (CMR) imaging in consecutive PLHWH on long-term HAART who were screened for underlying CVD and followed up clinically for adjudicated adverse cardiovascular events (cardiovascular mortality, non-fatal acute coronary syndrome, an appropriate device discharge, or a documented HF hospitalization). Imaging protocol included routine assessment of cardiac volumes and function, scar by late gadolinium enhancement, myocardial perfusion and native T1 /T2 mapping. Time-to-event analysis was performed from the index CMR exam to the first single event per patient Systematic risk scores for CVD (Framingham risk score (FRS), Data Collection on Adverse effects of anti-HIV Drugs score, D:A:D and MAGGIC integer score) were calculated using original online calculators. Results 156 participants (males 62%, 50 [42-57] years of age) were included. 24 events were observed (4 HF deaths, 1 sudden cardiac death, 2 non-fatal acute myocardial infarction, 1 appropriate device discharge and 16 HF hospitalizations) during a median follow-up of 13 [9-19] months. Patients with events had higher native T1 (ms, 1149 [1115-1163] ms vs. 1110 [1075-1138] ms), native T2 (ms, 40 [38-41] vs. 37 [36-39]), LV mass index (g/m², 65 [49-77] vs. 57 [49-64]) p < 0.05 for all). In multivariable analyses, native T1 was independently predictive of adverse events (ChiSq 15.9, p < 0.001, native T1 (10 ms) hazard ratio (95% confidence interval) 1.20 (1.08-1.33), p = 0.001), followed by a model that also included LV mass (ChiSq 17.1, p < 0.001). Traditional cardiovascular risk scores were not predictive of the adverse events. Conclusions Native myocardial T1 and LV mass by CMR, as opposed to traditional cardiovascular risk scores, predict cardiovascular outcome in PLWH, together reflecting the pathological myocardial remodeling of myocardial fibrosis and inflammation that potentially explain higher rates of HF in PLWH as compared to the non-infected population. These findings may inform personalized approaches to screening and early intervention to reduce the burden of HF.