External validation of pooled cohort equations using systolic blood pressure intervention trial data

Area of Science:

• Cardiology
• Preventive Medicine
• Clinical Epidemiology

Background:

• Cardiovascular disease risk is commonly estimated using the ACC/AHA Pooled Cohort Equations (PCEs).
• The accuracy of PCEs is debated, especially for individuals recommended for statin therapy based on ACC/AHA guidelines.
• External validation is crucial to assess the real-world performance of risk prediction tools.

Purpose of the Study:

• To externally validate the ACC/AHA PCEs in a population eligible for statin therapy.
• To assess the calibration and discrimination of PCEs in predicting atherosclerotic cardiovascular disease (ASCVD) events.
• To determine if PCEs accurately estimate ASCVD risk in the SPRINT trial cohort.

Main Methods:

• Utilized data from 4057 participants in the Systolic Blood Pressure Intervention Trial (SPRINT).
• Compared predicted 10-year ASCVD risks from PCEs with observed events over a median follow-up of 3.3 years.
• Employed the Hosmer-Lemeshow test for calibration and calculated the area under the curve (AUC) for discrimination.

Main Results:

• PCEs demonstrated poor calibration, with a statistically significant deviation from observed events (p < 0.001).
• The equations consistently overestimated the predicted probability of ASCVD events.
• Discrimination was moderate, with an AUC of 0.65 (95% CI, 0.60-0.69).

Conclusions:

• The ACC/AHA PCEs may overestimate ASCVD risk in patients recommended for statin therapy.
• Findings suggest potential limitations of current risk prediction tools in this specific patient group.
• Further research is needed to refine ASCVD risk assessment for statin-eligible individuals.