Abstract
We compared the Framingham risk score (FRS) for 10-year coronary heart disease (CHD) risk in age- and race-matched hepatitis C virus (HCV)-infected and HCV-uninfected persons: 114,073 HCV-infected (111,436 HCV-monoinfected and 2,637 HIV/HCV-coinfected) and 122,996 HCV-uninfected (121,380 HIV and HCV-uninfected and 1,616 HIV-monoinfected) males without cardiovascular disease, diabetes, or hepatitis B. In unadjusted analyses, FRS was similar between the HCV-infected and HCV-uninfected groups [median (interquartile range, IQR) risk points 13 (10-14) vs. 13 (10-14), p=0.192]. Cholesterol levels were lower and current smoking more prevalent in the HCV groups (both HCV and HIV/HCV) compared with the uninfected groups (p<0.001 for both). Prevalence of non-FRS CHD risk factors, such as substance abuse and chronic kidney disease, in the cohort was high, and differed by HCV and HIV status. Adjusting for age, race/ethnicity, body mass index, chronic kidney disease, drug and alcohol use, and HIV status, HCV infection was associated with minimally lower FRS (β=-0.095 risk points, p<0.001), suggesting a small but significant difference in 10-year CHD risk estimation in HCV-infected as compared to HCV-uninfected persons when measuring risk by FRS. Given the complex relationship between HCV, HIV, and CHD risk factors, some of which are not captured by the FRS, the FRS may underestimate CHD risk in HCV-monoinfected and HIV/HCV-coinfected persons. HCV- and HIV/HCV-specific risk scores may be needed to optimize CHD risk stratification.