Abstract T P393: New Pooled Cohort Risk Equations: Application to a Recent Stroke Patient Population
Background: In 2013, Pooled Cohort Risk (PCR) equations, which incorporates new sex- and race-specific estimates of the 10-year risk for atherosclerotic cardiovascular disease (ASCVD) including stroke, for ASCVD-free adults was introduced. Given the similarity of primary and secondary stroke risk factors, and benefit of a potential tool to readily identify stroke patients at high intermediate-term vascular risk for appropriate treatment, we evaluated the prediction and discrimination of the PCR and Framingham Cardiovascular Risk (FCR) equations after a recent stroke.
Method: We conducted an analysis of Vitamin Intervention for Stroke Prevention dataset of 3555 recent ischemic stroke patients aged ≥35 years and followed for 2 years. The PCR and FCR scores were calculated using the coefficients for the equations for calculating an estimate of an individual’s 10-year vascular risk. Subjects were categorized as having low-PCR/-FCR (<20%), high-PCR/-FCR (≥20%), and known-ASCVD. Independent associations of high-PCR/-FCR with recurrent stroke (primary outcome) and stroke/coronary heart disease (CHD)/vascular death (secondary outcomes) were assessed. Model discrimination (c-statistic) was assessed using the areas under the receiver operating characteristic curves.
Results: Both the PCR and FCR were independently related to both outcomes: compared with low-PCR, high-PCR was associated with stroke (adjusted hazard ratio [AHR], 1.79; 95% CI, 1.25–2.57) and stroke/CHD/vascular death (2.05; 1.55–2.70). Compared with low-FCR, high-FCR was associated with stroke (2.06; 1.34–3.16) and stroke/CHD/vascular death (1.57; 1.12–2.20). The c-statistic of PCR/FCR as a continuous variable for stroke was 0.56 (95% CI, 0.54–0.58) and 0.56 (0.54–0.57), respectively and for stroke/CHD/vascular death was 0.62 (0.60–0.63) and 0.61 (0.59–0.63), respectively.
Conclusions: Both the PCR and FCR are significant predictors of recurrent vascular events, but neither one of them appear to be an optimal model to discriminate intermediate-term ASCVD prediction among recent stroke patients already receiving secondary stroke prevention.
Author Disclosures: J. Park: None. H. Kwon: None. B. Ovbiagele: None.
- © 2015 by American Heart Association, Inc.