Models help predict ASCVD risk in adults with HIV

In 2013, the American College of Cardiology (ACC) and the American Heart Association (AHA) developed the pooled cohort equations to predict the 10-years risk of atherosclerotic cardiovascular disease (ASCVD). The equations were incorporated into the ACC/AHA guidelines.

A new retrospective analysis shows for the first time that the equations discriminated MI risk adequately and exhibited moderate calibration. Two other data-derived MI risk estimation models that incorporated HIV-specific factors had similar discrimination as the pooled cohort equations, but they had worse calibration.

Lead researcher Matthew J. Feinstein, MD, of Northwestern University’s Feinberg School of Medicine, and colleagues published their results online in JAMA Cardiology on Dec. 21.

Feinstein said in a news release that the risk for MI for adults with HIV was approximately 50 percent higher than the pooled cohort equations predicted.

The researchers noted that more than 1.2 million adults in the U.S. and 25 million adults worldwide have HIV. They added that people with HIV have nearly twice the risk for MI and greater risks for heart failure and sudden death.

For this analysis, the researchers used the Centers for AIDS Research Network of Integrated Clinical Systems cohort, which includes adults with HIV who received care at one of eight centers in the U.S. The patients initially enrolled in 1995 or later.

Of the 11,288 patients in this study, 61.2 percent were white, 81.9 percent were men and 70 percent were taking antiretroviral therapy.

Based on the pooled cohort equations, the mean 10-year ASCVD risk estimates were highest for black men and lowest for white women. The researchers defined ASCVD as nonfatal MI, coronary heart disease death and stroke.

After a mean follow-up period of 4.1 years, the MI rates were 6.9 per 1,000 person-years in black men, 4.4 per 1,000 person-years in white men, 7.2 per 1,000 person-years in black women and 3.3 per 1,000 person-years in white women.

The MI rates were 7.5 per 1,000 person-years in adults who were at least 40 years old versus 2.2 per 1,000 person-years in adults who were younger than 40 years old. Meanwhile, the MI rates were 6.3 per 1,000 person-years in participants who were not virally suppressed and 4.7 per 1,000 person-years in those who were virally suppressed.

The Harrell C statistics, which measure discrimination for MI risk, were 0.75 for the pooled cohort equations and 0.72 and 0.73 for the two models that incorporated HIV-specific factors. The researchers mentioned that the pooled cohort equations were moderately calibrated but predicted consistently lower MI rates.

The researchers cited a few limitations of the study, including that they could not assess strokes and that the trial had a relatively short mean follow-up period. They also only analyzed five of the eight sites because the other three sites did not have adjudicated MI data. In addition, people with HIV in the U.S. are typically younger than the cohorts that were used to develop the pooled cohort equations. Further, a few of the patients were taking statins, which could have decreased the observed MI rates for higher-risk patients more than for lower-risk patients.

Heidi Crane, MD, one of the study’s authors from the University of Washington, said in a news release that adults with HIV have chronic inflammation and higher rates of traditional risk factors such as smoking, which contributes to higher MI and stroke rates.

“Despite these differences, we found that risk scores developed in the general population — while not as accurate as we would like — are still useful in assessing risk in HIV populations,” Crane said. “More research is needed to develop better ways to assess risk in HIV.”