P5083Assessing cardiovascular risk in people with atherosclerotic vascular disease on intensive statin therapy
Borislava Mihaylova, the HPS3/TIMI55–REVEAL Collaborative Group
Abstract
Borislava Mihaylova, the HPS3/TIMI55–REVEAL Collaborative Group
Abstract
Background: The HPS3/TIMI55-REVEAL trial involved 30 449 adults with atherosclerotic vascular disease all of whom were receiving intensive atorvastatin therapy, a population typically considered for novel cardiovascular protective therapies. We present prediction models for risk of myocardial infarction or coronary death [MI/CHD] and myocardial infarction, ischaemic stroke or cardiovascular death [MI/IS/CVD] in this population. Methods: Among the 15 224 participants allocated to placebo in REVEAL, 15 080 with fully available baseline data contributed to the development of the risk models. During a median 4 years of follow-up, 1036 (6.9%) participants experienced MI/CHD and 1548 (10.3%) experienced MI/IS/CVD. Cox proportional hazard methods with forward-backward stepwise selection of baseline characteristics (p<0.05 for inclusion and p<0.01 for retaining) were used. The risk models were assessed among 15 086 anacetrapib-allocated participants in REVEAL and an external cohort of 12 756 placebo-allocated participants in HPS2-THRIVE, using Harrell's c-index for discrimination. The relative risk reductions with anacetrapib across categories of participant by cardiovascular risk were studied. Results: In addition to socio-demographic characteristics (region, age, sex, and smoking status), lipid levels (LDL, HDL), creatinine, albuminuria, diastolic blood pressure, previous disease history (coronary, peripheral vascular, cerebrovascular, diabetes, heart failure and atrial fibrillation) and use of anticoagulant/ antiplatelet contributed important information to coronary and/or vascular risk scoring. The REVEAL MI/CHD risk score achieved discrimination (c-index) of 0.66 (95% CI 0.64–0.68) in REVEAL and 0.67 (0.65–0.69) in HPS2-THRIVE cohorts and the REVEAL MI/IS/CVD risk score achieved discrimination 0.66 (0.65, 0.67) in REVEAL and 0.66 (0.64, 0.68) in HPS2-THRIVE cohorts. Among placebo-allocated participants in REVEAL, the 4-year MI/CHD risk ranged from 2% to 16% and the 4-year MI/IS/CVD risk from 4% to 26% across deciles of predicted risk. The use of anacetrapib resulted in similar relative risk reductions across categories of respective cardiovascular risk. Conclusions: REVEAL coronary and vascular risk models successfully stratify patients with established atherosclerotic vascular disease managed on intensive statin therapy into higher and lower risks with several-fold risk difference. In REVEAL, similar relative vascular risk reductions were observed with use of anacetrapib across risk strata. Acknowledgement/Funding: Merck, the British Heart Foundation, the Medical Research Council and the National Institute for Health Research.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Background: The HPS3/TIMI55-REVEAL trial involved 30 449 adults with atherosclerotic vascular disease all of whom were receiving intensive atorvastatin therapy, a population typically considered for novel cardiovascular protective therapies. We present prediction models for risk of myocardial infarction or coronary death [MI/CHD] and myocardial infarction, ischaemic stroke or cardiovascular death [MI/IS/CVD] in this population. Methods: Among the 15 224 participants allocated to placebo in REVEAL, 15 080 with fully available baseline data contributed to the development of the risk models. During a median 4 years of follow-up, 1036 (6.9%) participants experienced MI/CHD and 1548 (10.3%) experienced MI/IS/CVD. Cox proportional hazard methods with forward-backward stepwise selection of baseline characteristics (p<0.05 for inclusion and p<0.01 for retaining) were used. The risk models were assessed among 15 086 anacetrapib-allocated participants in REVEAL and an external cohort of 12 756 placebo-allocated participants in HPS2-THRIVE, using Harrell's c-index for discrimination. The relative risk reductions with anacetrapib across categories of participant by cardiovascular risk were studied. Results: In addition to socio-demographic characteristics (region, age, sex, and smoking status), lipid levels (LDL, HDL), creatinine, albuminuria, diastolic blood pressure, previous disease history (coronary, peripheral vascular, cerebrovascular, diabetes, heart failure and atrial fibrillation) and use of anticoagulant/ antiplatelet contributed important information to coronary and/or vascular risk scoring. The REVEAL MI/CHD risk score achieved discrimination (c-index) of 0.66 (95% CI 0.64–0.68) in REVEAL and 0.67 (0.65–0.69) in HPS2-THRIVE cohorts and the REVEAL MI/IS/CVD risk score achieved discrimination 0.66 (0.65, 0.67) in REVEAL and 0.66 (0.64, 0.68) in HPS2-THRIVE cohorts. Among placebo-allocated participants in REVEAL, the 4-year MI/CHD risk ranged from 2% to 16% and the 4-year MI/IS/CVD risk from 4% to 26% across deciles of predicted risk. The use of anacetrapib resulted in similar relative risk reductions across categories of respective cardiovascular risk. Conclusions: REVEAL coronary and vascular risk models successfully stratify patients with established atherosclerotic vascular disease managed on intensive statin therapy into higher and lower risks with several-fold risk difference. In REVEAL, similar relative vascular risk reductions were observed with use of anacetrapib across risk strata. Acknowledgement/Funding: Merck, the British Heart Foundation, the Medical Research Council and the National Institute for Health Research.
Key concepts: Medicine, Atherosclerotic cardiovascular disease, ATHEROSCLEROTIC VASCULAR DISEASE, Statin, Cardiology, Disease, Vascular disease, Internal medicine