Cardiovascular risk in HIV-infected individuals : a comparison of three risk prediction algorithms

Abstract

Introduction: Cardiovascular (CV) risk is known to be increased in HIV-infected individuals. Our aim was to assess CV risk in HIV-infected adults.

Methods: CV risk was estimated for each patient using three different risk algorithms: SCORE, the Framingham risk score (FRS), and DAD. Patients were classified as at low, moderate or high CV risk. Clinical and anthropometric data were collected.

Results: We included 571 HIV-infected individuals, mostly male (67.1%; n=383). Patients were divided into two groups according to antiretroviral therapy (ART): naïve (7.5%; n=43) or under ART (92.5%; n=528). The mean time since HIV diagnosis was 6.7±6.5 years in the naive group and 13.3±6.1 years in the ART group. Metabolic syndrome (MS) was identified in 33.9% (n=179) and 16.3% (n=7) of participants in the ART and naïve groups, respectively. MS was associated with ART (OR=2.7; p=0.018). Triglycerides ≥150 mg/dl (OR=13.643, p<0.001) was one of the major factors contributing to MS. Overall, high CV risk was found in 4.4% (n=23) of patients when the SCORE tool was used, in 20.5% (n=117) using the FRS, and in 10.3% (n=59) using the DAD score. The observed agreement between the FRS and SCORE was 55.4% (k=0.183, p<0.001), between the FRS and DAD 70.5% (k=0.465, p<0.001), and between SCORE and DAD 72.3% (k=0.347, p<0.001).

Conclusion: On the basis of the three algorithms, we detected a high rate of high CV risk, particularly in patients under ART. The FRS was the algorithm that classified most patients in the high CV risk category (20.5%). In addition, a high prevalence of MS was identified in this patient group.