ORLANDO – The 2013 ACC/AHA atherosclerotic cardiovascular disease risk calculator isn’t reliably applicable to HIV-positive adults in its present form because it consistently underpredicts their MI risk, Dr. Michael J. Feinstein reported at the American Heart Association scientific sessions.
That’s the bad news. The good news is that “a simple, data-derived refit of the pooled cohort equations may improve the model’s performance in HIV-positive individuals,” said Dr. Feinstein of Northwestern University, Chicago.
Bruce Jancin/Frontline Medical News
Dr. Michael J. Feinstein
Tweaking the risk calculator to enhance its accuracy in the HIV-positive population is particularly important because this population is growing in size and aging. And as Dr. Feinstein and coinvestigators showed in another study presented at the AHA meeting, the proportion of deaths due to cardiovascular disease in HIV-positive adults is shooting upward as they live longer because of treatment advances.
The investigators’ analysis of data from the Centers for Disease Control and Prevention national Wonder database showed that the proportion of deaths due to cardiovascular disease more than doubled between 1999 and 2013. Meanwhile, proportionate cardiovascular disease mortality declined by 22% in the general population and by 28% among individuals with inflammatory polyarthropathies.
That the ACC/AHA risk calculator in its present form doesn’t perform adequately in HIV-positive individuals hadn’t previously been shown, but it doesn’t really come as a surprise, according to Dr. Feinstein. After all, it’s known that this population is at 1.5- to 2-fold increased risk for MI and roughly 5-fold increased risk for sudden cardiac death, compared to the general population, where the risk calculator works best.
“Most data suggest that even in the setting of optimally treated HIV and undetectable viral load there’s still an underlying viral reservoir that appears to be driving inflammation and atherothrombotic and even nonatherothrombotic events in this population,” he said.
Dr. Feinstein and coworkers evaluated the 2013 ACC/AHA risk calculator in 11,901 HIV-positive black or white adults enrolled in the Centers for AIDS Research Network of Integrated Clinical Systems (CNICS) database for whom 5-year follow-up was available; 52% of the subjects were aged 40 or older at baseline.
Running each of these nearly 12,000 subjects through the risk calculator, the predicted result was that 103 of them would have an acute MI during the 5-year follow-up period. In reality, 132 MIs were observed. The discrepancy between the risk calculator predictions and observed MI rates was greatest in the 63% of HIV-positive subjects deemed at low risk, with an estimated 10-year risk of atherosclerotic cardiovascular disease of less than 5%.
Among white men, the risk calculator was remarkably consistent in underpredicting MIs. Regardless of whether the risk calculator put their estimated 10-year risk at less than 5%, 5% to less than 7.5%, 7.5% to less than 10%, or at least 10%, the actual observed MI rates were 67%-68% greater across the board than predicted.
Dr. Feinstein and coworkers refit the ACC/AHA risk calculator on a trial basis by incorporating variables related to HIV-positivity into the risk equations. Then they reanalyzed the tool’s performance in the same nearly 12,000 HIV-infected subjects. They found the discrimination and calibration of the revised risk equations improved substantially and met the standard of “acceptable” by statisticians’ standards.
The next step in this ongoing CNICS project will be to validate the provisionally refit risk calculator’s performance in a separate database of HIV-infected adults with adjudicated MIs. If the results are again positive, it will be a relatively straightforward matter to introduce the revisions into the ACC/AHA risk calculator, particularly since the senior coinvestigator in this project is Dr. Donald M. Lloyd-Jones, also of Northwestern University, who played a central role in developing the 2013 risk calculator.
Dr. Feinstein reported having no financial conflicts regarding this study.