Serum Total and Non-High-Density Lipoprotein Cholesterol and the Risk Prediction of Cardiovascular Events - The JALS-ECC

Background: Few Japanese studies have compared serum non-high-density lipoprotein (non-HDL) cholesterol with serum total cholesterol as factors for predicting risk of cardiovascular events. Currently, few tools accurately estimate the probability of developing cardiovascular events for the Japanese general population. Methods and Results: A total of 22,430 Japanese men and women (aged 40-89 years) without a history of cardiovascular events from 10 community-based cohorts were followed. In an average 7.6-year follow up, 104 individuals experienced acute myocardial infarction (AMI) and 339 experienced stroke. Compared to serum total cholesterol, serum non-HDL cholesterol was more strongly associated with risk of AMI in a dose-response manner (multivariable adjusted incidence rate ratio per 1 SD increment [95% confidence interval]=1.49 [1.24-1.79] and 1.62 [1.35-1.95], respectively). Scoring systems were constructed based on multivariable Poisson regression models for predicting a 5-year probability of developing AMI; the non-HDL cholesterol model was found to have a better predictive ability (area under the receiver operating curve [AUC]=0.825) than the total cholesterol model (AUC=0.815). Neither total nor non-HDL serum cholesterol levels were associated with any stroke subtype. Conclusions: The risk of AMI can be more reliably predicted by serum non-HDL cholesterol than serum total cholesterol. The scoring systems are useful tools to predict risk of AMI. Neither total nor non-HDL serum cholesterol can predict stroke risk in the Japanese general population. (Circ J 2010; 74: 1346-1356)