Long-Term Lipid Effects of Pioglitazone by Baseline Anti-Hyperglycemia Medication Therapy and Statin Use from the PROactive Experience (PROactive 14)

Article Outline

• Abstract
• Methods
• Results
• Discussion
• Acknowledgment
• References
• Copyright

Studies have shown that pioglitazone treatment in patients with type 2 diabetes mellitus can improve parameters of diabetic dyslipidemia. The aim of this study was to examine the effect of pioglitazone on triglycerides, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol levels in patients from the Prospective Pioglitazone Clinical Trial in Macrovascular Events (PROactive) to determine whether pioglitazone-induced lipid effects were altered by different baseline antihyperglycemia medication or statin use. PROactive was a long-term, randomized, double-blind, cardiovascular outcomes study in patients with type 2 diabetes at high cardiovascular risk who had pioglitazone or placebo added to existing treatment. The present post hoc study analyzed lipid results from patients who received different baseline antihyperglycemia regimens and the presence or absence of baseline statin use. Independent of antihyperglycemia medication and statin use, triglyceride levels decreased in all subgroups treated with pioglitazone (−9.9% to −12.3%), whereas little change was observed in placebo groups. High-density lipoprotein cholesterol increased nearly twice as much with pioglitazone (18.1% to 20.3%) as with placebo (8.1% to 11.8%) across all subgroups. Low-density lipoprotein cholesterol increased moderately with pioglitazone (5.2% to 9.6%) compared with placebo (3.3% to 7.6%) (placebo-adjusted range 1.11% to 4.37%). In conclusion, long-term pioglitazone therapy led to durable improvements in triglyceride and high-density lipoprotein cholesterol levels, irrespective of baseline antihyperglycemia therapy or statin use.

Diabetic dyslipidemia is characterized by hypertriglyceridemia, decreased high-density lipoprotein (HDL) cholesterol levels, and increased small low-density lipoprotein (LDL) cholesterol particles that may contribute to the high level of cardiovascular risk in patients with type 2 diabetes mellitus. Analysis of the lipid data collected during the long-term, randomized, double-blind, placebo-controlled, cardiovascular outcomes study Prospective Pioglitazone Clinical Trial in Macrovascular Events (PROactive)¹ provides a valuable opportunity to explore the durability of the lipid effects of pioglitazone in patients with different baseline antihyperglycemia medication and statin use. In PROactive, pioglitazone or placebo was added to multiple existing antihyperglycemia and cardiovascular treatments in patients with type 2 diabetes mellitus and established macrovascular disease. The average time of observation was 34.5 months. Here, we tested the hypothesis that pioglitazone affects the dyslipidemia of patients with type 2 diabetes in the same manner irrespective of their baseline antihyperglycemia medications and their use of statins. We report results from post hoc subgroup analyses of the lipid changes observed in several baseline medication cohorts of the PROactive study population. Changes in lipid values were determined for patients receiving the following antihyperglycemia treatment regimens at baseline: metformin or sulfonylurea monotherapy, metformin and sulfonylurea combination therapy, and insulin. Lipid levels were also analyzed by baseline statin use, irrespective of background antihyperglycemia medications.

Back to Article Outline

Methods 

Details of the PROactive study design and methods are provided elsewhere.¹, ² Patients were adults with type 2 diabetes (glycosylated hemoglobin [HbA_1c] >6.5%), with all patients having evidence of macrovascular events 6 months before entry. All patients provided written informed consent, and the study was conducted in accordance with the Declaration of Helsinki and the Good Clinical Practice Guidelines. Patients were randomly assigned to treatment with pioglitazone or placebo, which was added to existing therapy (including antihyperglycemia, lipid-altering, antiplatelet, and antihypertensive agents); treatment randomization was not stratified by baseline medication use. After randomization, study medication was titrated up to 45 mg in a double-blind fashion. All patients, including those who ceased study medication, were followed until death or the final visit. Throughout the study, investigators were required to adjust background therapy to meet 1999 International Diabetes Federation guidelines,³ including HbA_1c <6.5%, and to optimize lipid-altering, antiplatelet, and antihypertensive therapy.

Fasting serum lipid and HbA_1c values were measured centrally at ICON Laboratories (Dublin, Ireland). Lipid levels were determined at baseline and at 6-month intervals thereafter. Direct quantitative enzymatic methods were used to measure HDL cholesterol and LDL cholesterol, and triglyceride (TG) levels were measured with a glycerol-banked enzymatic assay. Methods used to measure HDL cholesterol and TGs were accredited by the Centers for Disease Control and Prevention's Lipid Standardization Program. Wilcoxon's rank-sum test was used to compare percentage change from baseline in TGs, HDL cholesterol, LDL cholesterol, LDL cholesterol/HDL cholesterol ratio, and HbA_1c in each baseline medication cohort.

Back to Article Outline

Results 

The entire PROactive study population has been described previously.¹, ² A total of 5,238 patients from 19 European countries were enrolled, and all but 2 patients were followed until the final visit or death. At baseline, 10% and 20% of patients were receiving metformin and sulfonylurea monotherapy, respectively; 25% were receiving metformin and sulfonylurea combination therapy; and 34% were receiving insulin plus oral agents. Forty-three percent of the study population was receiving statin therapy at baseline. More than 80% of patients in each cohort completed the study while receiving study medication, and the average time of observation was 34.5 months. The mean age ranged from 60 to 63 years, and body mass index ranged from 30 to 32 kg/m² in all cohorts. Patients at entry had LDL cholesterol levels that were at or near the International Diabetes Federation's contemporary definition of “low risk” (115.8 mg/dl) in each antihyperglycemia medication cohort and among patients who were not receiving statins at baseline. TG and HDL cholesterol levels were less well controlled: they were within the “at-risk” and “high-risk” ranges for all cohorts. Baseline statin use ranged from 39% to 53% across the antihyperglycemia medication cohorts, but within each cohort, statin use was similar in the pioglitazone and placebo arms. Baseline fibrate use was also well balanced (Table 1).

Table 1. Baseline characteristics

Group	Men	Duration of Type 2 Diabetes (yrs)	Statin Use	Fibrate Use	HbA1c	Triglycerides, mg/dl	HDL–C, mg/dl	LDL–C, mg/dl
All patients
Pioglitazone (n=2,605)	67%	9.4±6.94	43%	10%	n=2,561; 7.8% (7.0%–8.9%)	n=2,561 160.3; 113.4–232.1	n=2,561 42.5; 36.4–50.7	n=2,562 111.8; 88.6–136.5
Placebo (n=2,633)	66%	9.6±7.09	43%	11%	n=2,597; 7.9% (7.1%–8.9%)	n=2,586 162.1; 114.3–233.0	n=2,585 42.9; 36.4–51.0	n=2,583 110.6; 88.9–136.9
Metformin
Pioglitazone (n=253)	70%	5.1±5.05	49%	12%	n=249; 7.4% (6.8%–8.2%)	n=250 165.2; 119.6–221.4	n=249 41.8; 36.4–49.9	n=249 109.8; 90.5–128.8
Placebo (n=261)	67%	5.6±5.41	53%	8%	n=257; 7.4% (6.8%–8.3%)	n=257 170.1; 125.8–238.3	n=257 42.2; 36.4–50.7	n=257 108.3; 85.5–131.9
Sulfonylurea
Pioglitazone (n=508)	68%	7.3±5.95	37%	7%	n=502; 7.6% (6.8%–8.4%)	n=501 155.0; 114.3–240.0	n=501 42.5; 36.4–50.3	n=501 118.7; 95.1–141.9
Placebo (n=493)	71%	6.9±6.07	39%	9%	n=489; 7.5% (6.7%–8.6%)	n=487 155.9; 111.6–223.2	n=487 43.3; 37.1–50.7	n=487 117.2; 94.4–141.5
Metformin plus sulfonylurea
Pioglitazone (n=654)	72%	9.5±6.30	46%	12%	n=644; 7.9% (7.2%–8.9%)	n=640 166.1; 113.4–227.6	n=639 42.5; 36.0–49.9	n=640 108.7; 86.2–131.5
Placebo (n=660)	66%	9.5±6.42	42%	12%	n=649; 7.9% (7.2%–8.9%)	n=650 165.6.1; 118.7–234.7	n=650 42.2; 36.0–51.0	n=649 109.1; 87.0–131.1
Insulin
Pioglitazone (n=864)	58%	12.8±7.05	39%	11%	n=851; 8.2% (7.4%–9.2%)	n=849 160.3; 111.6–238.3	n=852 43.3; 36.4–51.4	n=852 111.4; 86.2–138.1
Placebo (n=896)	61%	13.1±7.12	42%	13%	n=887; 8.3% (7.5%–9.3%)	n=880 161.7; 112.5–235.6	n=878 42.9; 36.0–52.2	n=878 110.2; 86.2–136.9
Statin
Pioglitazone (n=1,108)	74%	9.2±6.91	100%	6%	n=1,095; 7.8% (7.0%–8.9%)	n=1,096 160.8; 114.7–229.4	n=1,096 41.8; 35.6–49.5	n=1,095 96.7; 78.9–117.2
Placebo (n=1,137)	74%	9.4±7.20	100%	4%	n=1,124; 7.9% (7.1%–8.9%)	n=1,122 159.9; 113.4–232.1	n=1,121 42.5; 36.0–50.3	n=1,120 95.1; 78.1–114.7
No statin
Pioglitazone (n=1,497)	62%	9.7±6.97	0	14%	n=1,473; 7.9% (7.0%–8.8%)	n=1,465 160.3; 111.6–233.8	n=1,465 43.7; 37.1–51.4	n=1,467 123.7; 102.1–145.8
Placebo (n=1,496)	59%	9.8±7.01	0	16%	n=1,473; 7.9% (7.0%–9.0%)	n=1,464 164.8; 116.0–233.4	n=1,464 43.3; 36.7–51.8	n=1,463 125.7; 102.5–148.5

Data are expressed as mean±SD for duration of type 2 diabetes and median (interquartile range) for HbA_1c and lipid parameters. International Diabetes Federation 1999 definitions of “low risk,” “at risk,” and “high risk” were as follows: for TG, 150.5, 150.5 to 194.7, and >194.7 mg/dl; for HDL-C, 46.3, 38.6 to 46.3, and <38.6 mg/dl; and for LDL-C, 115.8, 115.8 to 154.4, and >154.4 mg/dl. Millimoles per liter converted to milligrams per deciliter: value/conversion factor, where conversion factor = 0.01129 for TG and 0.02586 for HDL-C and LDL-C.

HDL-C = HDL cholesterol; LDL-C = LDL cholesterol.

In all baseline medication cohorts, pioglitazone was associated with substantial decreases in TG levels, whereas little change occurred with placebo treatment. This divergence appeared at the first postbaseline visit (month 6) and persisted throughout the study (Figure 1, Figure 2). At the final visit, the changes in TGs for the overall population were −11.4% with pioglitazone and 1.8% with placebo (Table 2). In all cohorts, pioglitazone treatment was also associated with significant increases in HDL cholesterol seen at 6 months and persistent throughout the study (Figure 1, Figure 2). At the final visit, the change in HDL cholesterol in the overall population was nearly twice that observed with placebo (Table 2). LDL cholesterol levels were increased at the final visit in the 2 treatment arms of all cohorts, and the increases were moderately higher with pioglitazone than with placebo. A greater decrease in the LDL cholesterol/HDL cholesterol ratio was associated with pioglitazone treatment overall and within each cohort (Table 2).

• 
  • View Large Image
  • Download to PowerPoint

• Figure 1. 

  Time course of TG and HDL cholesterol (HDL-C) levels for all patients by baseline antihyperglycemia medication use. * p value for treatment group comparison (pioglitazone vs placebo) <0.01 for each cohort; ^† p <0.05 for each cohort; ^‡ p <0.0001 for each cohort. For all patients (black diamonds), p <0.0001 at all time points for TGs and HDL cholesterol. MET = metformin; SU = sulfonylurea.

• 
  • View Large Image
  • Download to PowerPoint

• Figure 2. 

  Time course of TG and HDL cholesterol (HDL-C) levels by baseline statin use. * p value for the treatment group comparison (pioglitazone vs placebo) <0.0001 for the 2 cohorts.

Table 2. Lipid changes for all patients by baseline antidiabetic medication use or by statin use (percentage change from baseline at final visit)

Cohort	TG		HDL-C		LDL-C		LDL-C/HDL-C
	Pioglitazone	Placebo	Pioglitazone	Placebo	Pioglitazone	Placebo	Pioglitazone	Placebo
All patients	n=2,201	n=2,196	n=2,201	n=2,195	n=2,200	n=2,194	n=2,199	n=2,194
Baseline (mg/dl)	160.32	162.98	42.92	42.92	111.76	109.82	2.57	2.53
Change from baseline (%)	−11.37⁎	1.81	19.00⁎	10.07	7.20‡	4.90	−9.45⁎	−4.17
Metformin	n=215	n=222	n=214	n=222	n=214	n=222	n=214	n=222
Baseline (mg/dl)	165.63	172.28	41.96	42.54	109.24	108.08	2.49	2.44
Change from baseline (%)	−12.28†	4.38	18.42⁎	8.06	7.80	4.86	−7.63	−2.62
Sulfonylurea	n=434	n=409	n=435	n=410	n=435	n=410	n=435	n=410
Baseline (mg/dl)	157.22	155.89	42.92	42.92	119.49	114.66	2.76	2.66
Change from baseline (%)	−9.90⁎	−0.70	18.99⁎	11.03	5.23‡	2.61	−10.00	−7.91
Metformin + sulfonylurea	n=559	n=551	n=558	n=551	n=557	n=551	n=556	n=551
Baseline (mg/dl)	162.09	168.29	42.73	41.76	108.28	108.28	2.52	2.53
Change from baseline (%)	−10.33†	−0.70	19.92⁎	11.76	7.95‡	3.58	−9.98‡	−4.74
Insulin	n=718	n=746	n=720	n=743	n=720	n=743	n=720	n=743
Baseline (mg/dl)	159.88	160.32	43.31	43.31	110.79	109.05	2.51	2.46
Change from baseline (%)	−12.02⁎	2.83	18.12⁎	8.87	9.60	7.56	−6.78†	−1.27
Statin	n=950	n=971	n=950	n=970	n=948	n=969	n=948	n=969
Baseline (mg/dl)	159.43	159.43	41.76	42.54	96.29	94.35	2.27	2.20
Change from baseline (%)	−12.28⁎	3.08	20.25⁎	10.00	8.20	7.09	−9.13⁎	−1.44
No statin	n=1,251	n=1,225	n=1,251	n=1,225	n=1,252	n=1,225	n=1,251	n=1,225
Baseline (mg/dl)	160.32	167.40	43.70	43.31	123.74	124.52	2.83	2.85
Change from baseline (%)	−10.90⁎	0.00	18.42⁎	10.13	6.07§	3.34	−9.66§	−6.35

Data are expressed as median percentage change from baseline. Millimoles per liter converted to milligrams per deciliter: value/conversion factor, where conversion factor=0.01129 for TG and 0.02586 for HDL-C and LDL-C.

Abbreviations as in Table 1.

Statin use increased by 9% to 14% throughout the study across different antihyperglycemia medication cohorts, and the balance observed at baseline was maintained. In the baseline statin cohort, >90% of patients were receiving statins at the final visit. Statin use was reported at the final visit for approximately 1/4 of those patients who were not receiving statins at baseline. Overall, 56% of patients were receiving statins at study end, compared with 43% at baseline. Fibrate use also increased slightly in all cohorts but remained well balanced between treatment groups (Table 1).

Glycemic control improved irrespective of randomized treatment, background antihyperglycemia medication, or statin use. However, pioglitazone consistently resulted in significantly greater HbA_1c improvements than placebo at the final visit (p <0.0001). In each cohort, the median decreases in HbA_1c were 0.7% to 0.8% with pioglitazone and 0.2% to 0.4% with placebo. In the metformin, sulfonylurea, and metformin plus sulfonylurea combination therapy cohorts, higher proportions of pioglitazone-treated patients remained on their baseline regimens, with approximately half as many pioglitazone-treated patients requiring the addition of another oral antidiabetic medication or insulin. In the insulin cohort, 9.5% of pioglitazone-treated patients were able to discontinue insulin by the final visit, compared with only 2% of placebo-treated patients.

In the overall population, there was a 3.6-kg increase in mean body weight in the pioglitazone arm and a 0.4-kg decrease in the placebo arm, with edema and heart failure (5.7% pioglitazone, 4.1% placebo) more commonly reported with pioglitazone treatment.

Back to Article Outline

Discussion 

The results of this post hoc analysis demonstrate that treatment with pioglitazone led to rapid (within 6 months) and durable improvements in TG and HDL cholesterol levels, and these changes were consistent in each baseline medication cohort analyzed. Although there was an unexpectedly large increase in HDL cholesterol in the placebo arm of each cohort, the increase in HDL cholesterol observed with pioglitazone was consistently and statistically significantly greater. Changes in LDL cholesterol and LDL cholesterol/HDL cholesterol ratio were also similar in each cohort. Although LDL cholesterol increased in pioglitazone-treated patients, previous studies have shown that this was not associated with an increase in apolipoprotein B and was accompanied by a shift toward larger LDL particles and a reduction in LDL particle concentration.⁴ Particular LDL subfractionations, however, were not measured in PROactive.⁵, ⁶

Epidemiologic studies have shown a positive relation between TG levels and coronary artery disease,⁷, ⁸ while an inverse relation between HDL cholesterol levels and cardiovascular risk has been seen.⁹, ¹⁰ PROactive was not designed to elucidate mechanisms of action, but improvements in the lipid profiles by pioglitazone may have contributed to a reduced risk for cardiovascular events in the pioglitazone arm. Pioglitazone treatment was associated with a 10% risk reduction (hazard ratio 0.90, 95% confidence interval 0.80 to 1.02, p = 0.095) in the composite primary end point of time to the first event of all-cause mortality, nonfatal myocardial infarction (including silent infarction), stroke, acute coronary syndrome, or various macrovascular interventional procedures.

This analysis extends the findings of previous pioglitazone combination therapy studies¹¹ and shows that similar lipid changes can be expected when pioglitazone is added to a variety of antihyperglycemia treatment regimens. The results clearly show the lipid-altering properties of pioglitazone independent of baseline antihyperglycemia or lipid medication. Similarly, pioglitazone treatment was associated with consistent lipid changes in the presence and absence of baseline statin use, suggesting that the lipid effects of pioglitazone may complement the inhibition of hydroxy-3-methylglutaryl coenzyme A reductase. The TG-lowering effect of pioglitazone has been attributed to increased plasma TG clearance, likely caused by increased lipoprotein lipase-mediated lipolysis, which in turn may be a response to pioglitazone-induced apolipoprotein C-III reduction.¹² The means by which pioglitazone influences other lipid levels are less completely understood.

Throughout PROactive, changes in HbA_1c were greater with pioglitazone than with placebo in all cohorts. Although lowering glucose concentrations may contribute to improvements in lipid abnormalities in patients with type 2 diabetes mellitus, the lipid-altering properties of pioglitazone appear to be independent of glycemic control. In a 24-week double-blind comparison of pioglitazone and rosiglitazone, the 2 agents improved baseline HbA_1c and insulin resistance levels by the same degree, yet only pioglitazone significantly reduced TG levels and increased HDL cholesterol.¹² Similar results were obtained in studies that compared pioglitazone (as monotherapy or in combination therapy regimens) with metformin or sulfonylurea.¹¹, ¹³ In PROactive, the lipid-altering properties of pioglitazone were not influenced by differences in baseline HbA_1c across the antihyperglycemia medication cohorts. Despite higher baseline HbA_1c levels and longer diabetes duration in the insulin and metformin and sulfonylurea combination therapy cohorts, the addition of pioglitazone had similar lipid effects in these patients.

PROactive was designed as a cardiovascular outcomes study in the context of best-of-care treatment. As such, randomization was not stratified by baseline antihyperglycemia medications, and these therapies were not controlled during the study. However, the 2 treatment arms in each cohort were well balanced in terms of sample size and baseline demographics. The shifts in antihyperglycemia medication use were expected and consistent with target-driven modifications. The greatest shift in the sulfonylurea and metformin monotherapy cohorts involved the addition of a third oral antihyperglycemia drug, while the greatest shift in the metformin and sulfonylurea combination therapy cohort involved the addition of insulin. Most patients in the insulin cohort continued to receive insulin, but more of those who were able to reduce or discontinue insulin were in the pioglitazone group.

As with background antihyperglycemia medications, lipid-altering therapy was not strictly controlled in PROactive. However, statin and fibrate use was well balanced at baseline, and as the study progressed, the use of these medications changed proportionally as investigators adjusted therapy to attain and maintain International Diabetes Federation lipid targets. Therefore, the favorable effects of pioglitazone on lipid levels were likely driven by the addition of pioglitazone, not shifts in background lipid-altering therapy.

Back to Article Outline

Acknowledgment 

We thank Andrew Roberts and Amy Yuping Xia for their assistance with manuscript development. Editorial assistance was provided by Absolute Health Care Communications and funded by Takeda Pharmaceutical Company, Deerfield, Illinois.

Back to Article Outline

References 

1. Dormandy JA, Charbonnel B, Eckland DJ, Erdmann E, Massi-Benedetti M, Moules IK, et al. Secondary prevention of macrovascular events in patients with type 2 diabetes in the PROactive study (Prospective Pioglitazone Clinical Trial in Macrovascular Events): a randomised controlled trial. Lancet. 2005;366:1279–1289
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (165 KB)
   • CrossRef
2. Charbonnel B, Dormandy J, Erdmann E, Massi-Benedetti M, Skene A PROactive Study Group. The Prospective Pioglitazone Clinical Trial in Macrovascular Events (PROactive): can pioglitazone reduce cardiovascular events in diabetes? (Study design and baseline characteristics of 5238 patients). Diabetes Care. 2004;27:1647–1653
   • View In Article
   • MEDLINE
   • CrossRef
3. European Diabetes Policy Group. A desktop guide to type 2 diabetes mellitus. Diabet Med. 1999;16:716–730
   • View In Article
   • MEDLINE
   • CrossRef
4. Deeg MA, Buse JB, Goldberg RB, Kendall DM, Zagar AJ, Jacober SJ, et al. Pioglitazone and rosiglitazone have different effects on serum lipoprotein particle concentrations and sizes in patients with type 2 diabetes and dyslipidemia. Diabetes Care. 2007;30:2458–2464
   • View In Article
   • CrossRef
5. Goldberg RB, Kendall DM, Deeg MA, Buse JB, Zagar AJ, Pinaire JA, et al. A comparison of lipid and glycemic effects of pioglitazone and rosiglitazone in patients with type 2 diabetes and dyslipidemia. Diabetes Care. 2005;28:1547–1554
   • View In Article
   • MEDLINE
   • CrossRef
6. Goldberg RB. Impact of thiazolidinediones on serum lipoprotein levels. Curr Atheroscler Rep. 2006;8:397–404
   • View In Article
   • MEDLINE
   • CrossRef
7. Austin MA, Hokanson JE, Edwards KL. Hypertriglyceridemia as a cardiovascular risk factor. Am J Cardiol. 1998;81(suppl):7B–12B
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (114 KB)
   • CrossRef
8. Assmann G, Schulte H, Funke H, von Eckardstein A. The emergence of triglycerides as a significant independent risk factor in coronary artery disease. Eur Heart J. 1998;19(suppl):M8–M14
   • View In Article
9. Abbott RD, Wilson PW, Kannel WB, Castelli WP. High density lipoprotein cholesterol, total cholesterol screening, and myocardial infarction (The Framingham study). Arteriosclerosis. 1988;8:207–211
   • View In Article
   • MEDLINE
10. Birjmohun RS, Hutten BA, Kastelein JJ, Stroes ES. Efficacy and safety of high-density lipoprotein cholesterol-increasing compounds: a meta-analysis of randomized controlled trials. J Am Coll Cardiol. 2005;45:185–197
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (179 KB)
    • CrossRef
11. Khan M, Xu Y, Edwards G, Urquhart R, Mariz S. Effects of pioglitazone on the components of diabetic dyslipidaemia: results of double-blind, multicenter, randomised studies. Int J Clin Pract. 2004;58:907–912
    • View In Article
    • MEDLINE
    • CrossRef
12. Nagashima K, Lopez C, Donovan D, Ngai C, Fontanez N, Bensadoun A, et al. Effects of the PPARgamma agonist pioglitazone on lipoprotein metabolism in patients with type 2 diabetes mellitus. J Clin Invest. 2005;115:1323–1332
    • View In Article
    • MEDLINE
    • CrossRef
13. Al Majali K, Cooper MB, Staels B, Luc G, Taskinen MR, Betteridge DJ. The effect of sensitisation to insulin with pioglitazone on fasting and postprandial lipid metabolism, lipoprotein modification by lipases, and lipid transfer activities in type 2 diabetic patients. Diabetologia. 2006;49:527–537
    • View In Article
    • MEDLINE
    • CrossRef