Prediction of future cardiovascular events by Framingham, SCORE and asCVD risk scores is less accurate in HIV-positive individuals from the HIV-HEART Study compared with the general population

C-A Schulz,L Mavarani,N Reinsch,S Albayrak-Rena,A Potthoff,N Brockmeyer,M Hower,R Erbel,K-H Jöckel,B Schmidt,S Esser,HIV HEART Aging Study Group and Heinz Nixdorf Recall Investigative Group

DOI: https://doi.org/10.1111/hiv.13124

Abstract:Objectives: Cardiovascular diseases (CVDs) occur more often in people living with HIV (PLWH) than in the general population. It has been reported that CVD risk scores developed for the general population underestimate the CVD risk in PLWH. Performances of the Framingham Risk Score (FRS), the Systematic Coronary Risk Evaluation (SCORE) and the atherosclerotic cardiovascular disease (asCVD) risk score in PLWH were compared with the general population to quantify score-specific differences in risk prediction. Methods: HIV-positive outpatients from the HIV-HEART (HIVH) study (n = 567) were compared with participants from the population-based Heinz Nixdorf Recall (HNR) study (n ~ 4440) both recruited from the German Ruhr area. During a follow-up time of around 5 years, the associations between the FRS and incident CVD and peripheral artery disease (CVD_pAD), SCORE and coronary heart disease (CHD), and asCVD and incident CVD were examined using logistic regression. Score performances were assessed by comparing the areas under the curve (AUCs). Results: The mean ages were 52.9 ± 6.7 and 59.1 ± 7.7 years in the HIVH and HNR studies, respectively. There were fewer incident CVD events in the HNR study than in the HIVH study (CVD_pAD: 3.9% vs. 12.1%; CHD: 2.1% vs. 7.8%; CVD: 3.5% vs. 9.9%). Age- and sex-adjusted CVD risk was greater with increasing FRS, SCORE and asCVD in both cohorts, but the scores performed more accurately in the HNR than in HIVH study (AUCs FRS: 0.71 vs. 0.65; SCORE: 0.70 vs. 0.62; asCVD: 0.74 vs. 0.62). Conclusions: Associations between risk scores and future CVD were observed in both cohorts, but the score performances were less reliable in PLWH than in the general population.