Real world cardiovascular risk assessment using reduced DAD, SCORE and Framingham equations in a German HIV cohort

Abstract

Abstract Introduction People living with HIV are at increased risk for cardiovascular disease (CVD). Therefore, prevention of CVD should be integrated in HIV care. Although several risk scores exist to predict the risk of developing CVD, none of this specifically addresses the fact of HIV. Previously, the reduced Data Collection on Adverse Events of Anti-HIV Drug score (D:A:D-R) model has been proposed to assess the CVD risk for HIV patients. However, this score has not been validated in an independent cohort, or compared to traditional CVD prediction models established in the non-HIV population. Purpose We aim to compare the predictive value of the CVD risk scores D:A:D-R, SCORE and Framingham CVD in a German HIV population and evaluate their CVD risk in a real world setting. Methods This is a prospective, cross-sectional, single center study. Patients with HIV and older than 30 years were enrolled between January 2018 and December 2019. Exclusion criteria were preexisting CVD and statin therapy. CVD risk estimates were calculated using the D:A:D-R, Framingham CVD (FRS), and SCORE models. Agreement between estimates was assessed using Cohen's kappa coefficient. The indication for statin therapy was calculated using the AHA and EACS guidelines. Results 488 patients with HIV and a median age of 47.8 years were included. Median time since HIV diagnosis was 9.9 years. All, but one Patient were on ART for a median time of 5.9 years. 61% were female. Using the D:A:D-R score, the highest proportion of patients with high or very high risk for CVD were identified (17.8%). In comparison, using SCORE and FRS only 4.7% and 13.7% of patients were identified as high or very high risk, respectively. There was a poor agreement between D:A:D-R and SCORE (k=0.11), and D:A:D-R and FRS (k=0.33). Based on the calculated CVD risk using the three models, indication for primary statin medication varied between 34.8% to 92%. Conclusion Our data demonstrate that standard CVD risk scores underestimate the CVD risk in patients with HIV. Using SCORE and FRS, a high proportion of patients with HIV and at high risk for CVD would be underestimated and miss the indication for statin medication. Therefore, using scores including HIV specific parameters should be considered in this patient population. Funding Acknowledgement Type of funding source: None