Background and Aims:
The aim of the study was to compare the predictive ability of the American College of Cardiology/American Heart Association Pooled Cohort equations, the European Society of Cardiology Systematic Coronary Risk Evaluation (SCORE) equation and the Data Collection on Adverse Effects of Anti-HIV Drugs (D:A:D) study equation in persons living with human immunodeficiency virus (HIV).
Methods:
We retrospectively studied all patients with HIV infection who had at least one clinical encounter in 2008 and at that time were 40-65 years old and did not have established cardiovascular disease or diabetes. The cardiovascular risk was estimated with the Pooled Cohort equations, the SCORE equation, and the D:A:D study equation. Patients were followed-up for ten years.
Results
In 2008, 232 patients aged 40-65 years old, without diabetes or prior cardiovascular disease were enrolled in our study (76.3% males, mean age 47.1±6.6 years, 80.2% on antiretroviral treatment, mean CD4 count 607±389, mean time since diagnosis of HIV infection 7.2±5.6 years). The predicted 10-year risk for total cardiovascular events with the Pooled Cohort equation was 6.5±4.6%, whereas the observed risk was 7.8%. The predicted 10-year risk for fatal cardiovascular events with the SCORE equation was 2.5±2.5% whereas the observed risk was 0.9%. The predicted 5-year risk for total cardiovascular events was 3.5±3.0% with the D:A:D study equation, whereas the observed 5-year risk was 6.5%.
Conclusions:
The predicted risk of cardiovascular events according to the SCORE, the Pooled Cohort Equations and the D:A:D study equation, as well as the observed rates of cardiovascular events, are shown in Table 3. The SCORE showed excellent discrimination, whereas the Pooled Cohort Equations and the D:A:D study equation showed acceptable discrimination (C-statistic 0.85, 0.75 and 0.76, respectively; Table 3). The Pooled Cohort Equations and the D:A:D study equation also showed good calibration (Hosmer-Lemeshow χ2 value 3.26 and 10.52, respectively (on 8 degrees of freedom); p = 0.92 and p = 0.23, respectively). In contrast, the SCORE was not well calibrated (Hosmer-Lemeshow χ2 value 13.60 (on 4 degrees of freedom), p = 0.009). Overall, 3.0, 33.2, and 3.9% of the study population was classified as very high cardiovascular risk according to the SCORE, the Pooled Cohort Equations and the D:A:D study equation. The Pooled Cohort Equations showed poor agreement with both the SCORE and the D:A:D study equation (Cohen’s kappa coefficient 0.118 and 0.125, respectively). In contrast, the SCORE showed better agreement with the D:A:D study equation (Cohen’s kappa coefficient 0.353).

01 - A COMPARISON OF THE PREDICTIVE ABILITY OF POOLED COHORT EQUATION, SCORE EQUATION AND D:A:D STUDY EQUATION IN PEOPLE LIVING WITH HIV