Objective: In this post hoc analysis, we examined insulin requirements and regimens, glycemic control, cardiovascular outcomes, and safety in the patients treated with insulin at baseline in the Prospective Pioglitazone Clinical Trial in Macrovascular Events study.
Design: The Prospective Pioglitazone Clinical Trial in Macrovascular Events study was a double-blind, placebo-controlled outcome study (mean follow-up 34.5 months) in 5238 high-risk patients with type 2 diabetes randomized to pioglitazone (force titrated to 45 mg) or placebo. One third of the total population (pioglitazone 864; placebo 896) were receiving insulin at baseline.
Results: A rapid and sustained decrease in insulin dose was observed with pioglitazone vs. a progressive increase with placebo. By study end, the mean insulin dose was lower with pioglitazone (42 vs. 55 U/d with placebo; P < 0.0001). The insulin regimen (number on insulin, need for multiple injections, and reduction in oral agents) had been simplified vs. placebo; nevertheless, a greater glycosylated hemoglobin reduction was observed with pioglitazone (-0.93%) vs. placebo (-0.45%; P < 0.0001). At the final visit, insulin had been discontinued in 9% of pioglitazone vs. 2% of placebo patients (P < 0.0001). More insulin-resistant patients (defined as poorly controlled type 2 diabetes despite high doses of insulin) in the pioglitazone plus insulin group showed the greatest glycosylated hemoglobin decline. There were nonsignificant reductions with pioglitazone relative to placebo in the cardiovascular primary (hazard ratio 0.86; 95% confidence interval 0.71, 1.04; P = 0.1198) and main secondary (hazard ratio 0.85; 95% confidence interval 0.67, 1.08; P = 0.1831) end points in insulin-treated patients. The rates of overall heart failure, edema, and hypoglycemia were higher with pioglitazone [13.5 vs 10.5% (P = 0.0489); 30.8 vs. 18.2% (P < 0.0001); and 42.1 vs 29.0% (P < 0.0001), respectively], but there were no significant differences in serious events.
Conclusions: Pioglitazone use in combination with insulin resulted in a sustained improved glycemic control and allowed the treatment regimens to be simplified and the insulin doses reduced.
Trial registration: ClinicalTrials.gov NCT00174993.