We read with interest the article by Mohanlal and Holman (1) on the oral triglyceride tolerance test, showing that the test meal they propose is palatable, acceptable to subjects, and provides reproducible evaluation of postchallenge triglyceride profiles.
Although there is a tendency to consider postprandial hypertriglyceridemia as an independent risk factor for cardiovascular disease, no test meal for its evaluation has been universally accepted or normal ranges for postprandial triglycerides have been identified. Moreover, a clear distinction should be made between postprandial, which is the daily situation, and postchallenge hypertriglyceridemia.
The test meal proposed in the above study is not a regular meal and, as a test meal, has the disadvantage of a high carbohydrate content (almost to the level of an oral glucose tolerance test) eliciting insulin secretion, which modulates triglyceride removal from the circulation. However, insulin secretion varies greatly not only among diabetic subjects, but also among nondiabetic subjects, and it is very different between diabetic and nondiabetic subjects.
The total area under the curve (AUC) of triglycerides and postprandial triglyceride levels are not the best index of postprandial triglyceride responses because they are strongly influenced by fasting triglyceride concentrations. A much better index uses incremental AUC (IAUC) and incremental triglyceride levels because they reflect only the postchallenge triglyceride changes (2). Therefore, the difference in postprandial triglycerides between diabetic and nondiabetic subjects in the study of Mohanlal and Holman (1) is significant by the criterion of AUC because diabetic subjects have much higher fasting triglyceride levels (1.35 vs. 0.77 mmol/l), but when the criterion of IAUC is used, diabetic subjects do not have a significantly higher postprandial triglyceride response. Thus the greater “exposure” of diabetic subjects suggested in this study is mainly due to higher fasting triglycerides than excessive postprandial increase.
The extremely high (r = 0.909) correlation found between postchallenge and fasting triglyceride levels is misleading because fasting levels contribute >70% to the postprandial levels.
We have recently shown (3), using a test meal containing 40 g fat, 19 g protein, and only 10 g carbohydrates in nondiabetic subjects, that the triglyceride increment over the fasting value and the IAUC were not correlated with the fasting triglyceride concentration. Thus postprandial hypertriglyceridemia can occur irrespective of the fasting triglyceride concentrations and is better described by incremental values.
We think that each test meal has its advantages and disadvantages and that until a universally accepted oral triglyceride tolerance test (in analogy to the oral glucose tolerance test) is developed, test meals should be in each study and, moreover, incremental rather than absolute triglyceride values should be used for the evaluation of postchallenge hypertriglyceridemia.
