Log in to MyACC
Performance of ACC/AHA Pooled Cohort CV Risk Equations
Quick Takes
- Performance of the ACC/AHA Pooled Cohort Equation (PCE) for atherosclerotic cardiovascular disease (ASCVD) risk in the real-world community-based settings showed higher accuracy among non-White women and White men.
- Importantly, the predictive capability of the PCE for future ASVCD events was not impacted by including individuals outside the arbitrary cut-offs for age, blood pressure, and dyslipidemia in the equation and escalation of statin medications.
- These findings might be reassuring in clinical practice on the continued use of PCE when estimating ASCVD in individuals outside the ranges and while escalating medications .
Study Questions:
What is the performance of the American College of Cardiology/American Heart Association (ACC/AHA) pooled cohort equation (PCE) for atherosclerotic cardiovascular disease (ASCVD) in real-world clinical practice?
Methods:
The investigators validated the PCE in a community-based cohort of consecutive patients that sought primary care in Olmsted County, MN; between 1997-2000, followed through 2016. Inclusion criteria were similar to those of PCE derivation. The study also included individuals with values outside the PCE range (i.e., age, blood pressure, cholesterol) cut-offs, or those on statin therapy initiation over follow-up to test if including those would significantly affect PCE predictive capabilities. As the original PCE did, they excluded those that, at baseline, had known history of clinical coronary artery disease, defined as history of stable or unstable angina, myocardial infarction (MI), coronary artery bypass surgery or percutaneous coronary revascularization, history of stroke, atrial fibrillation, or heart failure.
Patient information was ascertained using the record linkage system of the Rochester Epidemiology Project. ASCVD events (nonfatal and fatal MI and ischemic stroke) were validated in duplicate. The primary outcome, as in the PCE derivation, was ASCVD defined as the composite of fatal and nonfatal MI and fatal and nonfatal ischemic stroke. Calculated and observed ASCVD risk and c-statistics were compared across predefined groups.
Results:
The study included 30,042 adults, mean age 48.5 ± 12.2 years, 46% males; median follow-up was 16.5 years, truncated at 10 years for this analysis. Mean ASCVD risk was 5.6 ± 8.73%. There were 1,555 ASCVD events (5.2%). The PCE revealed good performance overall (c-statistic, 0.78), and in sex and race subgroups, which was highest among non-White females (c-statistic, 0.81) and lowest in White males (c-statistic, 0.77). Out-of-range values and statin medication initiation did not affect model performance. Overall, the PCE accurately estimated the risk for ASCVD in the community, with only a minimal absolute difference between the predicted risk and the observed rate of events (5.68% vs. 5.22%), with similarly small differences for sex and race subgroups. There was significant overestimation of risk in the subgroup of subjects with predicted risk >10% (4.3% absolute difference for all), and among White men and women, and to a lesser degree in other sex and race subgroups. Among those with low predicted risk (<5%), the risk was slightly underestimated.
Conclusions:
Performance of the ACC/AHA PCE for ASCVD risk was assessed in a real-world community-based setting of adults seeking primary care. The PCE demonstrated favorable overall performance, with higher accuracy observed among non-White women and White men. Notably, the predictive capability of the PCE was not significantly affected by including individuals with out-of-range values (i,e., age, blood pressure, cholesterol) or initiating statin therapy over follow-up.
Perspective:
There have been very few studies validating the PCE in the contemporary era where there are better prescriptions and adherence to statin medications and more emphasis on lifestyle modification. Perhaps the most important findings of this current study are that statin usage did not impact predicted risk and including individuals outside of the arbitrary ranges for age, blood pressure, cholesterol, etc., set by PCE did not specifically affect the performance. These findings might be reassuring in clinical practice on the continued use of PCE when estimating ASCVD in individuals outside the ranges and while escalating medications in this group. There are several individuals that are outside the ranges of age, blood pressure, etc., in PCE that might have elevated long-term risks. Demonstrating that risk associated with these values is likely linear or quasilinea,r thus less likely to impact performance of PCE, can be reassuring to clinicians.
The PCE was developed in the pre-statin era and since then there have been significant changes in the management of CVD. This might explain why in previous validation studies, there was an overprediction of events in the high risk >10% of ASCVD group. This finding was also seen in the current study. The current study also had a slight underestimation of risk in the low-risk group (<5%). Importantly, these findings suggest that there are still factors that are not accounted for in the PCE in the evaluation of low-risk women, especially non-White women, which result in pervasive underestimation of ASCVD risk in this demographic. The reasons for this disparity might be the younger age and differential impact of traditional ASCVD risk factors in women, and nontraditional sex-specific ASCVD risk factors such as pre-eclampsia, gestational diabetes, history of polycystic ovary syndrome, or pregnancies resulting in pre-term births.
Clinical Topics: Cardiovascular Care Team, Prevention, Vascular Medicine
Keywords: Atherosclerosis, Ischemic Stroke, Myocardial Infarction, Risk Assessment, Secondary Prevention
< Back to Listings
