Low–Glycemic Index vs High–Cereal Fiber Diet in Type 2 Diabetes

To the Editor: In their randomized controlled trial, Dr Jenkins and colleagues¹ reported moderately reduced levels of hemoglobin A_1c (HbA_1c) in patients with type 2 diabetes treated with a low–glycemic index diet compared with a high–cereal fiber diet over 6 months. They concluded that low–glycemic index diets may be useful as part of the strategy to improve glycemic control in patients with type 2 diabetes.

Low–glycemic index diets are typically high-fiber diets. As the authors acknowledged, the low–glycemic index diet contained even more fiber than the high–cereal fiber diet (18.7 vs 15.7 g of fiber per 1000 kcal; P < .001). Several of the recommended foods in the high–cereal fiber group contained considerable amounts of soluble fibers (such as pectin in guava and carrots) that, unlike cereal fibers, do not appear to influence diabetes risk.² In addition, starchy high–glycemic index food such as baked potatoes, emphasized in the high–cereal fiber and restricted in the low–glycemic index group, might have further influenced the observed findings.³ However, we recognize the challenges in designing an intervention comparing effects of low–glycemic index vs high–cereal fiber diets.

We disagree with the authors' statement that “with the exception of oat and barley fiber” (which are rich in soluble β-glucan fibers), “cereal fibers are largely without metabolic effect.” The authors provided a reference supporting that additional consumption of 15 g of wheat bran per day for 3 months does not change fasting glucose and HbA_1c levels in diabetic patients.⁴ However, the dose might have been too small, and even short-term ingestion of various other insoluble fibers including purified cereal fiber in higher doses (approximately 30 g per day) shows significant meal-related metabolic effects.⁵ Improved whole-body insulin sensitivity after consumption of insoluble, nonviscous fibers has been reported, using euglycemic-hyperinsulinemic clamps.⁵

Glycemic control is incompletely reflected by HbA_1c and fasting glucose levels. The metabolic benefit of insoluble cereal fibers may exceed that of low–glycemic index or soluble fibers and may best be reflected by measures of insulin sensitivity.^{2 ,5} Although estimates of insulin sensitivity are likely to more accurately estimate hepatic rather than whole-body insulin sensitivity, to provide a complete picture the authors should report such indices including fasting insulin levels.