Since the relation between serum triglyceride level and coronary heart disease after adjustment for cholesterol and other risk factors remains controversial, the authors have tested the hypothesis that it could be different according to the level of serum cholesterol, i.e., that there may be an interaction between cholesterol and triglyceride level in prediction of coronary heart disease risk. The data of the Paris Prospective Study were used to carry out a survival analysis, by using the Cox model, with coronary heart disease death as the end point and triglyceride and different risk factors as the predictor variables. In this study, during a mean follow-up of 11.4 +/- 2.2 years, 157 coronary heart disease deaths occurred. Serum triglyceride level is not an independent predictor of coronary heart disease death after adjustment for serum cholesterol. However, when cholesterol, triglyceride, and their interaction term are introduced in the regression equation, all variables contribute significantly to the risk. This is also true when other risk factors are taken into account. As a consequence of this interaction, among the 3,585 subjects with a serum cholesterol level lower than 220 mg/100 ml, serum triglyceride level is an independent predictor of risk, even after adjustment for all other risk factors.