Comparison between the Framingham and prospective cardiovascular of Münster scores for risk assessment of coronary heart disease in human immunodeficiency virus-positive patients in Pernambuco, Brazil.

Background: The Framingham score is used in most studies on human immunodeficiency virus (HIV)-positive patients to estimate the risk for coronary heart disease; however, it may have some limitations for detecting risk among these individuals.

Objective: The aim of this study was to evaluate the agreement between the Framingham and Prospective Cardiovascular of Münster (PROCAM) scores among HIV-positive individuals and to investigate the factors associated with disagreement between the two scores.

Method: A cross-sectional study was conducted in a population of HIV/acquired immunodeficiency syndrome (AIDS) patients attending the outpatient's clinics of two reference centers for HIV/AIDS in Pernambuco, Brazil. Agreement between the Framingham and PROCAM scores was evaluated using the kappa index. From this analysis, a variable called "disagreement between scores" was created, and univariate and multivariate analysis were performed to investigate the factors associated with this variable.

Results: The prevalence of low, moderate, and high risk were, respectively, 78.7%, 13.5%, and 7.8% by Framingham score and 88.5%, 4.3%, and 7.2% by PROCAM (kappa = 0.64, P ≤ 0.0001). Agreement in the subgroup with metabolic syndrome by the International Diabetes Federation (IDF) (kappa = 0.51, P ≤ 0.0001) and the National Cholesterol Education Program (NCEP) (kappa = 0.59, P ≤ 0.0001) criteria was moderate. The Framingham score identified greater proportion of women with moderate risk. Factors independently associated with disagreement were: smoking, sex, age, low-density lipoprotein cholesterol, diastolic blood pressure, and metabolic syndrome.

Conclusion: There was a good agreement between the Framingham and PROCAM scores in HIV-positive patients, but a higher proportion of moderate-high risk was identified by the Framingham score. This disagreement should be evaluated in cohort studies to observe clinical outcomes over the course of time.