Effects of Rosiglitazone on Lipids, Adipokines, and Inflammatory Markers in Nondiabetic Patients With Low High-Density Lipoprotein Cholesterol and Metabolic S

【摘要】  Background- PPAR- agonists improve insulin sensitivity and glycemic control in type 2 diabetes and may reduce atherosclerosis progression. Thus, PPAR- agonists may be an effective therapy for metabolic syndrome. However, the full spectrum of potentially antiatherogenic mechanisms of PPAR- agonists have not been fully tested in nondiabetic patients with metabolic syndrome.

Methods and Results- We performed a prospective, double-blinded, placebo-controlled study of 60 nondiabetic subjects with low high-density lipoprotein cholesterol (HDL-C) level and metabolic syndrome to rosiglitazone 8 mg daily or placebo for 12 weeks. We found no significant effect of rosiglitazone on HDL-C (+5.5% versus +5.8%, P =0.89), and an increase in total cholesterol (+8% versus -1%; P =0.03). Nevertheless, rosiglitazone significantly increased adiponectin (+168% versus +25%; P <0.001), and lowered resistin (-6% versus +4%; P =0.009), C-reactive protein (-32% versus +36%, P =0.002), interleukin (IL)-6 (-22% versus +4%, P <0.001), and soluble tumor-necrosis factor- receptor-2 (-5% versus +7%, P <0.001).

Conclusions- These findings suggest that rosiglitazone, presumably through its PPAR- agonist properties, has direct effects on inflammatory markers and adipokines in the absence of favorable lipid effects. These findings may help explain the mechanism underlying the possible antiatherosclerotic effects of rosiglitazone.

We performed a 12-week, prospective, double-blinded study of 60 nondiabetic subjects with metabolic syndrome randomized to rosiglitazone or placebo. Rosiglitazone had direct effects on inflammatory markers and adipokines in the absence of favorable lipid effects. These findings may underly the possible antiatherosclerotic effects of rosiglitazone.

【关键词】  adipocytokines lipids inflammation lipoprotein metabolism arteriosclerosis

Introduction

The metabolic syndrome is characterized by depressed high-density lipoprotein-cholesterol (HDL-C), elevated triglycerides, central obesity, impaired glucose tolerance, and elevated blood pressure. 1 These clinical factors underlie an overall insulin resistant and pro-inflammatory state. Approximately 44% of the US population older than age 50 years has the metabolic syndrome, 2 which is of concern, because it is associated with a 60% higher prevalence of coronary heart disease. 2

Synthetic peroxisome proliferator-activated receptor (PPAR)- agonists, or thiazolidinediones (TZDs), including rosiglitazone and pioglitazone, improve insulin sensitivity, at least partly through PPAR- activation in adipose tissue, 3 and improve glycemic control in type 2 diabetes. Rosiglitazone raises HDL-C levels from 3% to 13.6%, 4-9 and lowers plasma CRP levels 10 in patients with type 2 diabetes. Thus, in patients with diabetes, rosiglitazone has a variety of effects that might reduce cardiovascular risk. Rosiglitazone has been shown to reduce the progression of carotid intimal-medial thickness, 11 and pioglitazone was recently shown to lower the incidence of the combined end point of death, myocardial infarction, and stroke in patients with diabetes. 12

A major question is whether these antiatherosclerotic effects of TZDs are secondary to favorable lipid effects or to other primary effects of PPAR- activation. For example, macrophages express PPAR-, and TZD treatment of macrophages upregulates ABCA-1, enhances cholesterol efflux, and reduces the macrophage inflammatory response. 13 Studies with PPAR- agonists in mice suggest that reduction in atherosclerosis may be mediated through direct effects on macrophages. 14

Although there are substantial data on the effects of TZDs in persons with diabetes, the body of data regarding TZDs in persons without diabetes is much less complete. To investigate the potential for direct effects of rosiglitazone on lipids and inflammatory markers, we performed a prospective, double-blinded study in which we randomized nondiabetic subjects with low HDL-C and metabolic syndrome to receive either rosiglitazone 8 mg daily or a matching placebo for 12 weeks. Our goal was to test the effect of rosiglitazone on lipids, adipokines, and inflammatory proteins in this subgroup of patients.

Methods

Study Participants

The study was approved by the Institutional Review Committees at the University of Pennsylvania and the Philadelphia Veterans Affairs Medical Center, and all participants provided written informed consent. Subjects were enrolled from outpatient practices between April 2003 and October 2004. The enrolled subjects included men and women between the ages of 18 and 75 years, with low HDL-C (<40 mg/dL for men and <50 mg/dL for women), and at least 2 of the following criteria: (1) abdominal obesity (waist circumference 40 inches in men and 35 inches in women); (2) blood pressure 130/85 mm Hg or the ongoing use of an antihypertensive agent; (3) fasting serum glucose level 110 mg/dL; and (4) fasting serum triglyceride level 150 mg/dL. Subjects were excluded if they had an LDL-C level 190 mg/dL, diabetes (defined by the use of antihyperglycemic medication or a fasting glucose 125 180/100 mm Hg), a serum 800 mg/dL, hepatic disease, vascular disease, a chronic inflammatory disorder, surgery within 30 days, active substance abuse, or any lipid-lowering therapy within the preceding 6 weeks.

Study Protocol

Subjects were randomized in a double-blind fashion to either rosiglitazone 8 mg or an identical matching placebo once daily for 12 weeks (GlaxoSmithKline, Inc). Randomization was performed by an unblinded investigational pharmacist, using a random number generator (Rando; Hawkeye Softworks). Participants were instructed to maintain their usual dietary and exercise habits. Study assessments were performed at baseline, 6 weeks, and 12 weeks for the following: (1) weight (Scale-Tronix digital scale; Scale-Tronix) and waist circumference; (2) blood pressure; (3) diet stability (through dietary records, obtained for 3 days before each visit, and analyzed with The Minnesota Nutrition Data System version 4.06 (University of Minnesota, Minneapolis, Minn); (4) exercise frequency (categorized as daily, 2 to 3 times per week, once per week, once per month, or never); (5) alcohol use; (6) bioelectrical impedance assessment of body composition (Quantum II Analyzer, RJL Systems); and (6) fasting blood samples for laboratory analyses.

Laboratory Analyses

Lipoprotein analyses were performed at baseline and 12 weeks on samples obtained after a 12-hour fast in a Centers for Disease Control-standardized lipid laboratory. Levels of total cholesterol, HDL-C, and triglycerides were measured with a Cobas Fara II autoanalyzer (Roche Diagnostic Systems, Inc) using Sigma reagents (Sigma Chemical Co). Very low-density lipoprotein cholesterol (VLDL-C) levels were determined after ultracentrifugation at a density of 1.006 g/mL. Levels of lipoprotein(a) [Lp(a)] were measured using DiaSorin reagents (DiaSorin Inc). Levels of free fatty acids, phospholipids, free cholesterol, apolipoprotein (apo)A-I, apoA-II, apoB, and apoC-III were measured on a Hitachi 912 autoanalyzer using Wako reagents (Wako Pure Chemical Industries). Insulin, leptin, and adiponectin levels were measured by radioimmunoassay (Linco Research, Inc). Lipoprotein subfractions were measured by proton nuclear magnetic resonance spectroscopy (LipoScience, Inc). Insulin resistance was estimated with the homeostasis model assessment or HOMA (plasma insulin [µIU/mL] x plasma glucose ÷22.5). Samples were assayed for CRP with an ultra-high sensitivity latex turbidometric immunoassay (Wako Pure Chemical Industries Ltd) on a Hitachi 912 autoanalyzer. Plasma resistin levels were measured by enzyme immunoassay (Linco Research, Inc). 15 Plasma levels of IL-6 and sTNF- R 2 (the soluble cleavage product of the activated tumor necrosis factor - receptor) were measured with a commercially available enzyme immunoassays (R&D Systems).

Statistical Analysis

We estimated a 10%±14% greater increase in HDL-C with rosiglitazone versus placebo. 4,5,7 To have 80% power to detect this difference, with a 2-tailed =0.05, we estimated the need for a sample size of 52 subjects. This sample size also provided 90% power, with a 2-tailed of 0.05, to detect a 25% greater decrease in CRP with rosiglitazone, compared with placebo. 10 Based on an estimated 15% dropout rate, we chose to enroll 60 total subjects. Only subjects who completed the study were included in the primary analyses.

Comparisons between groups for continuous variables were performed with analysis of covariance models, each containing the baseline value of the response variable and an indicator variable for treatment group. All data were assessed for a normal distribution before analyses. Levels of triglycerides, free cholesterol, free fatty acids, apolipoprotein C III, Lp (a), insulin, certain LDL and HDL subfractions, CRP, IL-6, sTNF-, adiponectin, and insulin resistance (HOMA) were skewed, and thus log-transformed before analyses. Resistin levels were not completely normalized by log-transformation and were thus analyzed using the Mann-Whitney test (Wilcoxon rank sum test). Dichotomous variables were compared by 2 analysis or logistic regression analyses. Comparisons of dietary nutrient composition between groups were made using the unpaired t test. Pearson correlation analyses were used to test correlations between non-normally distributed variables. Normally distributed continuous variables are given as means with SD, and non-normally distributed variables are given as medians with inter-quartile ranges. All analyses were performed using SPSS 11.5 (SPSS, Inc).

Results

The final study group included 30 subjects assigned to receive rosiglitazone and 27 subjects assigned to receive placebo ( Figure ). One subject was missing data for measurements of lipoprotein particle sizes, adipokines, and inflammatory markers, because of a laboratory error.

Diagram of the flow of subjects through the study.

Fifty-three percent of the participants were black. There were no significant differences in baseline characteristics between the study groups ( Table 1 ). There was no significant difference in exercise frequency between the groups at baseline ( P =0.24) or 12 weeks ( P =0.70) (data not shown). No subjects started lipid-lowering therapy during the study.

TABLE 1. Baseline Characteristics of the Participants

Changes in Weight, Blood Pressure, Insulin Resistance, and Free Fatty Acids

There was no significant effect of rosiglitazone on body weight ( P =0.51), waist circumference ( P =0.15), dietary intake (all P 0.41), or percent fat mass ( P =0.085) (data not shown). Systolic blood pressure decreased by 9±15 mm Hg on rosiglitazone, but also decreased by 5±13 mm Hg in the placebo group ( P =0.68). Diastolic blood pressure decreased by 2±10 mm Hg on rosiglitazone group and increased by 3±11 mm Hg on placebo ( P =0.16).

The decrease in insulin resistance (as estimated by the HOMA equation) (-26% in both groups; P =0.23) and in free fatty acid levels (-16% versus -4%, respectively; P =0.16) ( Table 2 ) were not significantly different between the rosiglitazone and placebo groups.

TABLE 2. The Response of Insulin, Glucose, and HOMA-IR to Treatment

Changes in High-Density Lipoproteins

There was no significant difference in the change in HDL-C level between the rosiglitazone and placebo groups (+5.5% versus +5.8%, respectively; P =0.89) or apolipoprotein A-I levels (-1.4% versus +0.6%; P =0.10), but a significantly greater increase in apolipoprotein A-II (+16% versus +3%; P <0.001) in the participants receiving rosiglitazone, compared with placebo ( Table 3 ). The subjects receiving rosiglitazone experienced a greater decrease in large HDL ( P =0.038) particles and a greater increase in medium-sized HDL particles ( P =0.033). There was no significant change in small HDL particle concentration ( P =0.092) ( Table 4 ).

TABLE 3. The Response of Lipid and Lipoprotein Levels to Treatment

TABLE 4. Response of lipoprotein particle distribution by NMR spectroscopy

Changes in ApoB-Containing Lipoproteins

As shown in Table 3, there were greater increases in levels of total cholesterol (+9% versus -2%; P =0.030), free (unesterified) cholesterol (+14% versus -2%; P =0.015), and non-HDL-C (+9% versus -2%; P =0.028) in the group receiving rosiglitazone compared with placebo. ApoB also increased more on rosiglitazone, (+10% versus -2%; P =0.021). Neither LDL-C (+0% versus -4%; P =0.42), VLDL-C (+40% versus +2%; P =0.076) or triglyceride levels (+4% versus -3%; P =0.40) responded significantly differently in those receiving rosiglitazone compared with placebo. We did not find any significant effects of rosiglitazone treatment on LDL particle size ( Table 4 ).

Effects on Adipocytokines and Inflammatory Markers

We observed a highly significant 168% increase in adiponectin in the rosiglitazone group, compared with a 25% increase in the placebo group ( P <0.001) ( Table 5 ). We also observed a 6% decrease in resistin in the rosiglitazone group, compared with a 4% increase in the placebo group ( P =0.01) ( Table 5 ). Those assigned to rosiglitazone experienced a greater decrease in CRP level (-32% versus +36%; P =0.002), IL-6 (-22% versus +4%; P =0.027), and sTNF- R 2 (-5% versus +7%; P <0.001) ( Table 5 ). In the group receiving rosiglitazone, there were significant correlations between the changes in resistin and changes in both IL-6 ( r =0.55, P =0.003) and sTNF- R 2 ( r =0.40, P =0.033), but not with the change in CRP ( r =0.22, P =0.26). There were no significant correlations between changes in HOMA-IR and any of the inflammatory markers (all r <0.09), resistin ( r =-0.26, P =0.18) or adiponectin ( r =0.05) in the rosiglitazone group

TABLE 5. The Response of Adipokines and Inflammatory Markers to Treatment (n=54)

Safety and Tolerability

The average adherence with study medication was 95%±5% for rosiglitazone and 95%±5% for placebo ( P =0.66). There were 30 mild to moderate adverse events reported by 17 individuals receiving rosiglitazone and 30 mild to moderate events by 14 individuals receiving placebo ( P =0.35). One subject assigned to rosiglitazone experienced a substantial increase in total cholesterol and triglycerides, requiring termination from the study 1 week early.

Discussion

We investigated the effects of rosiglitazone on lipoproteins, adipokines, and inflammatory markers in patients with low HDL-C and metabolic syndrome. Our subjects were obese, highly insulin-resistant, and included 53% blacks. Rosiglitazone modestly increased total cholesterol, LDL-C, and non-HDL-C, had no significant effect on HDL-C, but had favorable effects on adipokines and inflammatory markers.

Our finding that rosiglitazone did not raise HDL-C in patients with metabolic syndrome are consistent with the findings from 3 previous studies of patients without diabetes, 2 involving patients with coronary artery disease 11,16 and 1 involving nonobese patients with metabolic syndrome. 17 We found no effect of rosiglitazone on apoA-1 levels, but an increase in apoA-II levels. ApoA-II has been associated with visceral fat accumulation and impaired catabolism of large VLDL particles. 18 It is thus intriguing to speculate that upregulation of apoA-II may contribute to an increase in VLDL-C levels, the most readily available clinical measure of triglyceride-rich remnant lipoproteins. In our study, mean VLDL-C concentration increased by 40% on rosiglitazone, although this change did not differ significantly from the placebo group.

In contrast, we found overall favorable effects of rosiglitazone on adipokines, including a significant increase in adiponectin levels. This increase in adiponectin may be attributable to a direct effect of rosiglitazone on adipocytes, and possibly macrophages. 19 Adiponectin has been shown to play a role in modulating insulin sensitivity 20 and to be increased by rosiglitazone in patients with diabetes. 20 Furthermore, increasing quintiles of adiponectin levels have been associated with decreased risk of myocardial infarction. 21

Our study demonstrates that rosiglitazone lowers resistin levels in patients with metabolic syndrome, as recently demonstrated in one small study of 14 patients with type 2 diabetes. 22 Resistin was originally found to be secreted by adipocytes in mice, in which it caused impaired insulin action. 23 In humans, however, resistin is predominantly produced by macrophages in response to inflammatory stimuli and is almost undetectable in adipose tissue. 15 Rosiglitazone may thus have a direct effect on resistin expression, such as through macrophage PPAR- activation. We have recently shown that resistin levels independently correlate with degree of coronary artery calcification. 15

Rosiglitazone also significantly reduced levels of CRP, IL-6, and sTNF- R 2. Similar responses for CRP and IL-6 were previously found in patients with diabetes. 10 We found a significant correlation between the rosiglitazone-induced changes in resistin and changes in the inflammatory markers IL-6 and sTNF- R 2. These findings are consistent with our previous findings of a significant correlation between baseline levels of resistin and sTNF- R 2. 24 Rosiglitazone has been shown to lower CRP in nondiabetic patients with coronary artery disease 16 and in nonobese Taiwanese patients with metabolic syndrome. 17

The HDL effects of rosiglitazone differ from those of pioglitazone. The recently published GLAI study directly comparing rosiglitazone and pioglitazone in patients with diabetes showed a 14.9% increase in HDL-C with pioglitazone versus a 7.8% increase with rosiglitazone. 25 Furthermore, we recently performed a study with pioglitazone in patients with metabolic syndrome that was similar in design to the current study, and showed a 14% increase in HDL-C with pioglitazone. 26 In patients with diabetes, pioglitazone also reduced triglycerides to a greater extent than rosiglitazone, 25 although this could have been influenced by their selection of patients with elevated baseline triglyceride levels. Similarly, in our 2 separate studies in nondiabetic subjects with metabolic syndrome, pioglitazone reduced triglycerides more than rosiglitazone. The mechanism underlying these differential effects between agents considered to be ligands for the same PPAR- receptor remains unclear. However, a previous study has shown incompletely overlapping transcriptional regulation by TZDs. 27 Pioglitazone has been suggested to have modest effects on activating PPAR-, which could contribute to its more potent effects on HDL-C and triglyceride levels. 28 Whereas neither apoA-I production rate nor fractional catabolic rate appears to be affected by pioglitazone, apoC-III was reduced by pioglitazone, consistent with a PPAR- effect. 29 This effect may account for the decrease in triglyceride concentration with pioglitazone, and could indirectly increase HDL-C via CETP-mediated exchange of VLDL triglycerides for HDL cholesterol. Interestingly, pioglitazone has been shown to increase expression of fatty acid oxidation enzymes in adipose tissue, as well as increase expression of PPAR- itself, which might also contribute to changes in triglycerides and/or HDL. 30 Both rosiglitazone and pioglitazone exerted potent antiinflammatory effects and more than doubled adiponectin levels in our studies.

Although randomized, double-blinded, and placebo-controlled, our study had important limitations. We estimated insulin resistance through the use of the HOMA index, whereas dynamic measures, such as the frequently sampled intravenous glucose tolerance test, are better validated. Our observed 26% reduction in HOMA-IR with rosiglitazone is consistent with that found in previous studies in patients without diabetes. 16,17 The fasting glucose levels in the control group were higher at baseline than in the rosiglitazone group, and the comparable decrease in HOMA-IR in the control group may be attributable to the decrease in fasting glucose level in this group between baseline and follow-up, representing regression to the mean. There was no measurable change in weight, exercise, or other lifestyle that would otherwise have accounted for this change. Third, we used a 12-week treatment period, and 1 previous study suggested that the metabolic response to rosiglitazone may evolve over a 1-year period of time. 9 Last, our study was also limited by the inclusion of a small number of women.

In this prospective, double-blinded study, in which 60 nondiabetic patients with metabolic syndrome were randomized to rosiglitazone 8 mg or placebo for 12 weeks, we found that rosiglitazone had modestly unfavorable overall effects on plasma lipids but significantly raised adiponectin levels and reduced resistin as well as other markers of inflammation. Thus, our data support the concept that the potential antiatherosclerotic effects of rosiglitazone may be partly related to direct effects on adipokines and inflammation. The ongoing Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of Glycemia in Diabetes (RECORD) Trial will test whether rosiglitazone lowers the incidence of cardiovascular events. A reduction in cardiovascular events with rosiglitazone in this trial would more likely reflect rosiglitazone?s favorable effects on adipokines and inflammation than its lipid-related effects.

Acknowledgments

This work was supported by a grant from Glaxo Smith Kline, Inc, and in part by grant MO1-RR00040 from the National Center for Research Resources (NCRR)/National Institutes of Health, supporting the University of Pennsylvania General Clinical Research Center. D.J.R. is a recipient of a Burroughs Wellcome Fund Clinical Scientist Award in Translational Research and a Doris Duke Charitable Foundation Distinguished Clinical Scientist Award. We are indebted to the nursing and dietary staff at the University of Pennsylvania GCRC for their assistance in performing this study.

【参考文献】
  Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA. 2001; 285: 2486-2497.

Alexander CM, Landsman PB, Teutsch SM, Haffner SM. NCEP-defined netabolic syndrome, diabetes, and prevalence of coronary heart disease among NHANES III participants age 50 years and older. Diabetes. 2003; 52: 1210-1214.

Evans RM, Barish GD, Wang Y-X. PPARs and the complex journey to obesity. Nature Medicine. 2004; 10: 1-7.

Wolffenbuttel BHR, Gomis R, Squatrito S, Jones NP, Patwardhan RN. Addition of low-dose rosiglitazone to sulphonylurea therapy improves glycaemic control in Type 2 diabetic patients. Diabet Med. 2000; 17: 40-47.

Fonseca V, Rosenstock J, Patwardhan R, Salzman A. Effect of Metformin and Rosiglitazone Combination Therapy in Patients With Type 2 Diabetes Mellitus: A Randomized Controlled Trial. JAMA. 2000; 283: 1695-1702.

Raskin P, Rendell M, Riddle MC, Dole JF, Freed MI, Rosenstock J; for the Rosiglitazone Clinical Trials Study Group. A randomized trial of rosiglitazone therapy in patients with inadequately controlled insulin-treated type 2 diabetes. Diabetes Care. 2001; 24: 1226-1232.

Lebovitz HE, Dole JF, Patwardhan R, Rappaport EB, Freed MI. Rosiglitazone monotherapy is effective in patients with type 2 diabetes. J Clin Endocrinol Metab. 2001; 86: 280-288.

Phillips LS, Grunberger G, Miller E, Patwardhan R, Rappaport EB, Salzman A. Once- and twice-daily dosing with rosiglitazone improves glycemic control in patients with type 2 diabetes. Diabetes Care. 2001; 24: 308-315.

Derosa G, Cicero AFG, Gaddi A, Ragonesi PD, Fogari E, Bertone G, Ciccarelli L, and Piccinni MN. Metabolic effects of pioglitazone and rosiglitazone in patients with diabetes and metabolic syndrome treated with glimepiride: a twelve-month, multicenter, double-blind, randomized, controlled, parallel-group trial. Clinical Therapeutics. 2004; 26: 744-754.

Haffner SM, Greenberg AS, Weston WM, Chen H, Williams K, Freed MI. Effect of rosiglitazone treatment on nontraditional markers of cardiovascular disease in patients with type 2 diabetes mellitus. Circulation. 2002; 106: 679-684.

Sidhu JS, Kaposzta Z, Markus HS, Kaski JC. Effect of rosiglitazone on common carotid intima-media thickness progression in coronary artery disease patients without diabetes mellitus. Arterioscler Thromb Vasc Biol. 2004; 24: 930-934.

Dormandy JA, Charbonnel B, Eckland DJ ea. 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.

Barbier O, Pineda Torra I, Duguay Y, Blanquart C, Fruchart J-C, Glineur C, Staels B. Pleiotropic actions of peroxisome proliferator-activated receptors in lipid metabolism and atherosclerosis. Arterioscler Thromb Vasc Biol. 2002; 22: 717-726.

Chawla A, Boisvert WA, Lee CH, Laffitte BA, Barak Y, Joseph SB, Liao D, Nagy L, Edwards PA, Curtiss LK, Evans RM, Tontonoz P. A PPAR gamma-LXR-ABCA1 pathway in macrophages is involved in cholesterol efflux and atherogenesis. Mol Cell. 2001; 7: 161-171.

Reilly MP, Lehrke M, Wolfe ML, Rohatgi A, Lazar MA, Rader DJ. Resistin is an inflammatory marker of atherosclerosis in humans. Circulation. 2005; 111: 932-939.

Sidhu JS, Cowan D, Kaski J-C. The effects of rosiglitazone, a peroxisome proliferator-activated receptor-gamma agonist, on markers of endothelial cell activation, C-reactive protein, and fibrinogen levels in non-diabetic coronary artery disease patients. J Am Coll Cardiol. 2003; 42: 1757-1763.

Wang T-D, Chen W-J, Lin J-W, Chen M-F, Lee Y-T. Effects of rosiglitazone on endothelial function, C-reactive protein, and components of the metabolic syndrome in nondiabetic patients with the metabolic syndrome. Am J Cardiol. 2004; 93: 362-365.

van ?t Hooft FM, Ruotolo G, Boquist S, de Faire U, Eggertsen G, Hamsten A. Human evidence that the apolipoprotein a-II gene is implicated in visceral fat accumulation and metabolism of triglyceride-rich lipoproteins. Circulation. 2001; 104: 1223-1228.

Chinetti G, Zawadski C, Fruchart JC, Staels B. Expression of adiponectin receptors in human macrophages and regulation by agonists of the nuclear receptors PPARalpha, PPARgamma, and LXR. Biochem Biophys Res Commun. 2004; 314: 151-158.

Tiikkainen M, Hakkinen A-M, Korsheninnikova E, Nyman T, Makimattila S, Yki-Jarvinen H. Effects of rosiglitazone and metformin on liver fat content, hepatic insulin resistance, insulin clearance, and gene expression in adipose tissue in patients with type 2 diabetes. Diabetes. 2004; 53: 2169-2176.

Pischon T, Girman CJ, Hotamisligil GS, Rifai N, Hu FB, Rimm EB. Plasma adiponectin levels and risk of myocardial infarction in men. JAMA. 2004; 291: 1730-1737.

Junga HS, Younb B-S, Choa YM, Yub K-Y, Parkb HJ, Shina CS, Kima SY, Leea HK, and Park KS. The effects of rosiglitazone and metformin on the plasma concentrations of resistin in patients with type 2 diabetes mellitus. Metabolism. 2005; 54: 314-320.

Steppan CM, Bailey ST, Bhat S, Brown EJ, Banerjee RR, Wright CM, Patel HR, Ahima RS, Lazar MA. The hormone resistin links obesity to diabetes. Nature. 2001; 409: 307-312.

Lehrke M, Reilly M, Millington S, Iqbal N, Rader D, and Lazar MA. An inflammatory cascade leading to hyperresistinemia in humans. PLoS Med. 2004; 1: 161-168.

Goldberg RB, Kendall DM, Deeg MA, Buse JB, Zagar AJ, Pinaire JA, Tan MH, Khan MA, Perez AT, Jacober SJ, for the GLAI Study Investigators. 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.

Szapary PO, Bloedon LT, Samaha FF, Duffy D, Wolfe ML, Reilly MP, Chittams J, Rader DJ. Effects of pioglitazone on lipoproteins, inflammatory markers, and adipokines in non-diabetic patients with the metabolic syndrome. Arterioscler Thromb Vasc Biol. 2005;in press.

Camp HS, Li O, Wise SC, Hong YH, Frankowski CL, Shen X, Vanbogelen R, Leff T. Differential activation of peroxisome proliferator-activated receptor-gamma by troglitazone and rosiglitazone. Diabetes. 2000; 49: 539-547.

Sakamoto J, Kimura H, Moriyama S, Odaka H, Momose Y, Sugiyama Y, Sawada H. Activation of human peroxisome proliferator-activated receptor (PPAR) subtypes by pioglitazone. Biochem Biophys Res Comm. 2000; 278.

Nagashima K, Lopez C, Donovan D, Ngai C, Fontanez N, Bensadoun A, Fruchart-Najib J, Holleran S, Cohn JS, Ramakrishnan R, Ginsberg HN. Effects of the PPAR{gamma} agonist pioglitazone on lipoprotein metabolism in patients with type 2 diabetes mellitus. J Clin Invest. 2005; 115: 1323-1332.

Bogacka I, Xie H, Bray GA, Smith SR. Pioglitazone induces mitochondrial biogenesis in human subcutaneous adipose tissue in vivo. Diabetes. 2005; 54: 1392-1399.

作者单位：Department of Medicine (F.F.S., P.O.S., M.M.W., A.K., D.J.R.), Cardiovascular Division, and the Cardiovascular Institute, University of Pennsylvania Medical Center; The Philadelphia Veterans Affairs Medical Center (F.F.S., N.I., M.M.W.); The Institute for Translational Medicine and Therapeutics (F.F