Division of Clinical Research, Biomedical Center, Edogawa-ku, Tokyo, Japan.
Postgrad Med. 2011 Jan;123(1):45-52. doi: 10.3810/pgm.2011.01.2244.
The aim of this study was to evaluate the effects of pioglitazone on lipid profiles in relation to glycemic control. Eighty-one treatment-naïve patients with type 2 diabetes mellitus received pioglitazone monotherapy. Subjects who had ≥ 1% reduction in hemoglobin A(1c) (HbA(1c)) levels were defined as responders (n = 47) and those with < 1% reduction as nonresponders (n = 34). At 3 months, the HbA(1c) levels and several lipid parameters were compared with baseline values. Because it is known that the response to some antihyperglycemic agents is proportional to baseline HbA(1c) levels, the changes (Δ) in these parameters were compared for 2 groups based on their ΔHbA(1c)/baseline HbA(1c) ratio. The lowest tertile was called super-responders (n = 25) and highest tertile was called extreme nonresponders (n = 24). At baseline, HbA(1c) levels and body mass index (BMI) were significantly higher in responders; no significant differences were observed in total cholesterol (TC), triglycerides (TGs), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), LDL-C/HDL-C ratio, or non-HDL-C between the 2 groups. At 3 months, significant decreases in TGs and increases in HDL-C were observed in both groups. In contrast, TC, non-HDL-C, and LDL-C/HDL-C ratio significantly decreased in responders. Low-density lipoprotein cholesterol had a tendency to decrease in responders. However, these parameters were nonsignificantly increased in nonresponders. Body mass index significantly increased in responders, while it slightly increased in nonresponders. Analysis of covariance revealed that significant intergroup differences existed with TC, LDL-C, non-HDL-C, and LDL-C/HDL-C ratio, while no such differences were observed with TGs, HDL-C, and BMI. Very similar results were obtained with super-responders and extreme nonresponders. These results suggest that approximately 40% of the treatment-naïve Japanese subjects with type 2 diabetes mellitus were nonresponders to pioglitazone, and differential regulations of lipid parameters exist between responders and nonresponders treated with pioglitazone. Bad cholesterols (eg, TC, LDL-C, and non-HDL-C) were reduced only in responders. Irrespective of its efficacy on glycemic control, pioglitazone can significantly downregulate TGs and upregulate HDL-C levels.
本研究旨在评估吡格列酮对血脂谱的影响与血糖控制的关系。81 例初治 2 型糖尿病患者接受吡格列酮单药治疗。将血红蛋白 A1c（HbA1c）水平降低≥1%的患者定义为应答者（n=47），将 HbA1c 水平降低<1%的患者定义为无应答者（n=34）。在 3 个月时，比较 HbA1c 水平和几项血脂参数与基线值的差异。由于已知一些抗高血糖药物的反应与基线 HbA1c 水平成正比，因此根据他们的ΔHbA1c/基线 HbA1c 比值，将这两组的这些参数的变化（Δ）进行比较。最低三分位数称为超级应答者（n=25），最高三分位数称为极端无应答者（n=24）。基线时，应答者的 HbA1c 水平和体重指数（BMI）显著较高；两组间总胆固醇（TC）、甘油三酯（TGs）、高密度脂蛋白胆固醇（HDL-C）、低密度脂蛋白胆固醇（LDL-C）、LDL-C/HDL-C 比值或非高密度脂蛋白胆固醇（non-HDL-C）无显著差异。3 个月时，两组 TG 均显著降低，HDL-C 均显著升高。相反，TC、非 HDL-C 和 LDL-C/HDL-C 比值在应答者中显著降低。LDL-C 有在应答者中降低的趋势，但在无应答者中无显著增加。协方差分析显示，TC、LDL-C、非 HDL-C 和 LDL-C/HDL-C 比值存在显著的组间差异，而 TGs、HDL-C 和 BMI 则无差异。超级应答者和极端无应答者也得到了非常相似的结果。这些结果表明，约 40%的初治日本 2 型糖尿病患者对吡格列酮无应答，而接受吡格列酮治疗的应答者和无应答者之间存在血脂参数的差异调节。坏胆固醇（如 TC、LDL-C 和非 HDL-C）仅在应答者中降低。无论其对血糖控制的疗效如何，吡格列酮均可显著降低 TGs 水平并升高 HDL-C 水平。
