The effect of pioglitazone on glycemic and lipid parameters and adverse events in elderly patients with type 2 diabetes mellitus: A post hoc analysis of four randomized trials

Abstract

Objective: This analysis was conducted to evaluate the impact of pioglitazone (PIO), both as monotherapy and as part of combination therapy, on glycemic and lipid parameters and adverse events in elderly patients with type 2 diabetes. Methods: This was a post hoc analysis of pooled data, truncated at 1 year, from patients aged ≥65 years with type 2 diabetes in 4 multicenter, randomized, double-blind, parallel-group trials. For inclusion in these trials, patients were required to be between the ages of 35 and 75 years and to have had poorly controlled type 2 diabetes, a glycosylated hemoglobin (HbA 1c) value between 7.5% and 11.0%, and stable or worsening glycemic control for at least 3 months. Blood samples were obtained at baseline and every 4 to 10 weeks thereafter for determination of HbA 1c, fasting plasma glucose (FPG), and lipid parameters (high-density lipoprotein cholesterol [HDL-C], low-density lipoprotein cholesterol [LDL-C], triglycerides [TG], total cholesterol [TC], TC:HDL-C ratio, and free fatty acids). Results: Data from 891 elderly patients (age range, 69.1–69.8 years) were included: 282 who received PIO, 123 metformin (MET), 142 sulfonylurea (SU), 105 SU + PIO, 107 SU + MET, 63 MET + PIO, and 69 MET + SU. With a few exceptions, all treatment groups were similar at baseline. From baseline to week 52, none of the changes in HbA 1c and FPG between each treatment group and its comparator were significant. The adjusted mean (SE) percent changes in HDL-C for the monotherapies were 17.95% (1.11) for PIO, 10.71% (1.70) for MET, and 5.17% (1.51) for SU (both comparisons, P < 0.05). For the combination therapies, the adjusted mean percent changes in HDL-C were 16.77% (1.84) for SUPIO versus 7.87% (1.75) for SUMET ( P < 0.05), and 16.34% (2.34) for MET + PIO versus 0.11% (2.19) for METSU ( P < 0.05). The adjusted mean percent changes in LDL-C for the monotherapies were 7.00% (1.28) for PIO, −0.68% (1.91) for MET, and −6.77% (1.73) for SU (both comparisons, P < 0.05). For the combination therapies, the adjusted mean percent change in LDL-C was significant for METPIO compared with METSU (13.62% [2.69] vs −4.32% [2.58], respectively; P < 0.05). The adjusted mean percent change in TG was significant for MET + PIO compared with MET + SU (−10.93% [4.44] vs 8.37% [4.15], respectively; P < 0.05). The adjusted mean percent changes in TC for the monotherapies were 6.16% (0.88) for PIO, −1.77% (1.35) for MET, and −6.90% (1.19) for SU (both comparisons, P < 0.05). For the combination therapies, the adjusted mean percent changes in TC were 2.67% (1.45) for SUPIO versus −1.40% (1.39) for SUMET ( P < 0.05) and 7.89% (1.85) for METPIO compared with −1.19% (1.73) for METSU ( P < 0.05). The differences in change in the TC:HDL-C ratio were not significant between groups. The adjusted mean changes in free fatty acids for the monotherapies were −0.14 (0.02) mmol/L for PIO, −0.001 (0.03) mmol/L for MET, and −0.07 (0.02) mmol/L for SU (both comparisons, P < 0.05). For the combination therapies, the adjusted mean change in free fatty acids was significant for SU + PIO compared with SUMET (−0.12 [0.03] vs 0.06 [0.03] mmol/L, respectively; P < 0.05). PIO monotherapy was associated with the lowest incidence of hypoglycemia (1.4%) among the 7 treatment groups. The SUPIO group had the highest incidence of weight gain (4.8%). The rate of deaths was <2% in all the treatment groups; no adverse event associated with death was considered related to study medication. Conclusions: In this post hoc analysis of data from elderly patients participating in 4 randomized clinical trials, PIO effectively controlled glycemic and lipid parameters over 52 weeks and was well tolerated. The effects seen in this analysis were comparable to those in the overall study populations.