Results of use of metformin and replacement of starch with saturated fat In diets of patients with type 2 diabetes

Abstract

To improve glycemic control by substituting saturated fat for starch, to identify any adverse effect on lipids masked by the extensive use of metformin and lipid-lowering drugs, and to attempt to separate dietary effects from effects of multiple drugs. We undertook a retrospective review of medical records of patients who completed 1 year of follow-up after dietary prescription. The study subjects included 151 patients in the diet group (whose dietary instructions included high saturated fat but starch avoidance) and 132 historical control subjects (who were allowed unlimited monounsaturated fat but had restriction of starch in their diets). Hemoglobin A1c (HbA1c) levels improved in both study groups (-1.4 +/- 0.2% [P<0.001]; 95% confidence interval [CI], -1.9 to -0.9). Use of metformin was associated with a decrease in HbA1c (-0.12 +/- 0.003%/mo [P<0.001]; 95% CI, -0.17 to -0.07). The diet group had an additional decrease of -0.7 +/- 0.2% (P<0.001; 95% CI, -1.1 to -0.3). Weight increase was associated with the use of insulin (+0.3 +/- 0.07 kg/mo [P<0.001]; 95% CI, 0.2 to 0.5), sulfonylurea (+0.18 +/- 0.06 kg/mo [P<0.01]; 95% CI, 0.05 to 0.30), and troglitazone (+0.7 +/- 0.2 kg/mo [P<0.005]; 95% CI, 0.3 to 1.2). Although not statistically significant, metformin therapy showed a trend for weight loss (-0.14 +/- 0.08 kg/mo; P = 0.07). An additional weight loss was noted in the diet group (-2.65 +/- 0.62 kg [P<0.001]; 95% CI, -3.87 to -1.44). Hydroxymethylglutaryl-coenzyme A reductase inhibitor use was associated with reduced total cholesterol level (-1.7 +/- 0.6 mg/dL per month [P<0.005]; 95% CI, -2.9 to -0.5). The diet group had an additional decrease of -13.0 +/- 4.5 mg/dL (P<0.001; 95% CI, -21.9 to -4.1). No significant effect of the diet on triglyceride, low-density lipoprotein, or high-density lipoprotein levels was detected. Addition of saturated fat and removal of starch from a high-monounsaturated fat and starch-restricted diet improved glycemic control and were associated with weight loss without detectable adverse effects on serum lipids.