Pioglitazone has been shown to reduce fasting triglyceride levels. The mechanisms of this effect have not been fully elucidated, but decreased lipolysis may contribute to blunt the hypertriglyceridemic response to a meal. To test this hypothesis, we studied 27 type 2 diabetes mellitus (T2DM) patients and 7 sex-, age-, and body mass index-matched nondiabetic controls. Patients were randomized to pioglitazone (45 mg/day) or placebo for 16 wk. Whole body lipolysis was measured [as the [(2)H(5)]glycerol rate of appearance (R(a))] in the fasting state and for 6 h following a mixed meal. Compared with controls, T2DM had higher postprandial profiles of plasma triglycerides, free fatty acid (FFA), and beta-hydroxybutyrate, and a decreased suppression of glycerol R(a) (P < 0.04) despite higher insulin levels [268 (156) vs. 190 (123) pmol/l, median (interquartile range)]. Following pioglitazone, triglycerides and FFA were reduced (P = 0.05 and P < 0.04, respectively), and glycerol R(a) was more suppressed [-40 (137) vs. +7 (202) mumol/min of placebo, P < 0.05] despite a greater fall in insulin [-85 (176) vs. -20 (58) pmol/l, P = 0.05]. We conclude that, in well-controlled T2DM patients, whole body lipolysis is insulin resistant, and pioglitazone improves the insulin sensitivity of lipolysis.