Abstract
Considerable controversy exists regarding the use of chromium (Cr) supplementation to modulate carbohydrate metabolism in subjects with diabetes. Recently, we reported that Cr supplementation, provided as 1000 microg/d as Cr picolinate, enhanced insulin sensitivity in subjects with type 2 diabetes mellitus. Our data agreed with some, but not all, studies that evaluated a similar dose and formulation in type 2 diabetes mellitus and suggested that subject selection and characteristics may be important considerations when assessing the clinical response. Thus, the goal of this study was to assess which metabolic or clinical characteristics, when obtained at baseline, best determine a clinical response to Cr when assessing changes in insulin sensitivity. Seventy-three subjects with type 2 diabetes mellitus were assessed in a double-blinded, randomized, placebo-controlled study. Subjects were assessed at baseline for glycemic control with glycated hemoglobin measures, oral glucose tolerance tests, and body weight and body fat measures (dual-energy x-ray absorptiometry). After baseline, insulin sensitivity in vivo was assessed with the use of hyperinsulinemic-euglycemic clamps. After the baseline clamp, subjects were randomized to receive Cr supplementation (1000 microg Cr/d provided as Cr picolinate) or placebo daily for 6 months. All study parameters were repeated after 6 months. The relationship of the baseline characteristics of the study subjects to the change in insulin sensitivity was determined. Sixty-three percent of the subjects with type 2 diabetes mellitus responded to the Cr treatment as compared with 30% with placebo. The only subject variable significantly associated with the clinical response to Cr was the baseline insulin sensitivity, as assessed with the hyperinsulinemic-euglycemic clamp (partial R(2) = .4038) (P = .0004). Subject phenotype appears to be very important when assessing the clinical response to Cr because baseline insulin sensitivity was found to account for nearly 40% of the variance in the clinical response to Cr.