点击显示 收起
Center for Research in Nutrition and Health Disparities
Arnold School of Public Health
University of South Carolina
2718 Middleburg Drive
Columbia, SC 29208
E-mail: EJMayer{at}gwm.sc.edu
Dear Sir:
A recent editorial by Pereira (1) in the Journal presented several important perspectives on studies of the glycemic index (GI) and glycemic load in relation to weight status. Most of these perspectives related to the interpretation of findings reported thus far from epidemiologic studies, and Pereira looks forward to studies yet to be conducted. It is important also to recognize that, in addition to various analytic issues, numerous potentially serious methodologic problems exist with respect to the application of the GI to studies of usual diet; these problems have to do with the index itself. Such concerns were noted both during the 1980s (2) and more recently (3). They include, but are not limited to, the derivation of the underlying GI values for specific foods from studies of glucose excursion in response to ingestion of the food after an overnight fast, whereas much of the food consumed as part of a usual diet is consumed during the postprandial interval. And, as Pereira appropriately noted, habitual diets that have a low GI may simply be generally prudent diets, with frequent consumption of nutrient-rich and fiber-rich foods, as recently described (4). Dietary fiber, the type of carbohydrate, and the processing of carbohydrate-containing food do seem to matter to glucose and insulin metabolism and related health outcomes. It is extremely important that our understanding of these processes is advanced. Pereira noted inconsistencies in the literature and called for longer, high-quality, randomized controlled trials. I suggest that, before such trials are conducted, new work be conducted to address methodologic problems and to advance our understanding of the underlying construct of the GI. Without further work to better characterize aspects of carbohydrate-containing foods that affect health, we risk spending limited research dollars to produce studies that will continue to be sometimes positive and sometimes negative, but we will still lack the underlying knowledge required to understand those findings.
ACKNOWLEDGMENTS
The author had no conflict of interest.
REFERENCES