Context: Reduction of 50% excess body weight, using a very low-calorie diet (VLCD; 450 kcal/d) improves insulin sensitivity in obese type 2 diabetes mellitus patients.
Objective: The objective of the study was to evaluate whether adding exercise to the VLCD has additional benefits.
Design: This was a randomized intervention study.
Setting: The study was conducted at a clinical research center in an academic medical center.
Subjects: Twenty-seven obese [body mass index 37.2 ± 0.9 kg/m(2) (mean ± sem)] insulin-treated type 2 diabetes mellitus patients.
Intervention: Patients followed a 16-wk VLCD. Thirteen of them simultaneously participated in an exercise program (E) consisting of 1-h, in-hospital training and four 30-min training sessions on a cycloergometer weekly.
Outcome measures: Insulin resistance was measured by a hyperinsulinemic euglycemic clamp. Insulin signaling, mitochondrial DNA (mtDNA) content, and intramyocellular lipid content was measured in skeletal muscle biopsies.
Results: Baseline characteristics were identical in both groups. Substantial weight loss occurred (-23.7 ± 1.7 kg VLCD-only vs. -27.2 ± 1.9 kg VLCD+E, P = NS within groups). The exercise group lost more fat mass. Insulin-stimulated glucose disposal increased similarly in both study groups [15.0 ± 0.9 to 39.2 ± 4.7 μmol/min(-1) · kg lean body mass (LBM(-1)) VLCD-only vs. 17.0 ± 1.0 to 37.5 ± 3.5 μmol/min(-1) · kg LBM(-1) in VLCD+E], as did phosphorylation of the phosphatidylinositol 3-kinase-protein kinase B/AKT insulin signaling pathway. In contrast, skeletal muscle mtDNA content increased only in the VLCD+E group (1211 ± 185 to 2288 ± 358, arbitrary units, P = 0.016 vs. 1397 ± 240 to 1196 ± 179, P = NS, VLCD-only group). Maximum aerobic capacity also only increased significantly in the VLCD+E group (+6.6 ± 1.7 ml/min(-1) · kg LBM(-1) vs. +0.7 ± 1.5 ml/min(-1) · kg LBM(-1) VLCD-only, P = 0.017).
Conclusion: Addition of exercise to a 16-wk VLCD induces more fat loss. Exercise augments maximum aerobic capacity and skeletal muscle mtDNA content. These changes are, however, not reflected in a higher insulin-stimulated glucose disposal rate.