Triglycerides and cardiovascular disease: meta-analysis of individual data on 600 000 participants in 60 studies

Triglycerides are specific fats that are increased in the blood of people who smoke, eat fatty foods and have high blood pressure. Widely available tests that can measure blood triglycerides concentrations are frequently used to help predict heart attack risk. Although dozens of long-term studies of triglycerides and heart attack risk have been conducted over the past 50 years, it is still uncertain whether measurement of these blood fats is actually worthwhile or whether they are likely to be causative of heart attacks. Individual studies have tended to be too small, and previous reviews of such studies have been insufficiently detailed to give clear results. To help provide much more reliable evidence about the relevance of triglycerides to heart disease, we will conduct detailed combined analyses of the available studies of triglycerides and future heart attack risk in a total of about 600 000 individuals from about 60 long-term studies. Previously, this approach helped clarify the evidence about other predictors of heart attack (such as blood pressure) and it should also do so for triglycerides.

Despite almost fifty years of epidemiological research, there is uncertainty about the relevance of triglycerides to coronary heart disease (CHD) in general populations. Individual studies have been too limited and previous reviews of such studies have been insufficiently detailed. We will conduct a meta-analysis of individual data on 600 000 participants from 60 studies, which have recorded a total of 25 000 incident cases of non-fatal myocardial infarction or coronary death. Such an analysis will reliably characterize several features of the association between triglycerides and CHD (and other major vascular and non-vascular outcomes) in appropriate detail and with precision, such as the magnitude of associations in different circumstances (e.g., in women and men, at different levels of triglycerides, at different ages, at different levels of known risk factors, at different levels of markers of insulin sensitivity [notably HDL]) after making appropriate allowances for within-person variability in triglycerides values and for possible confounding factors. These findings will contribute compelling evidence that will help to determine the potential usefulness of: (i) measurement of triglycerides in the prediction of CHD and (ii) strategies that lower triglycerides in cardiovascular disease prevention.