| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
Neuroendocrine Unit and Program in Nutritional Metabolism, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts 02114
Address all correspondence and requests for reprints to: Colleen Hadigan, M.D., M.P.H., Neuroendocrine Unit, Bulfinch 457-B, Massachusetts General Hospital, 55 Fruit Street, Boston, Massachusetts 02114. E-mail: chadigan{at}partners.org.
Abstract
The purpose of this study was to evaluate the metabolic and cardiovascular benefits of continued metformin therapy for HIV-infected patients with lipodystrophy. Eligible subjects who participated in the 3-month randomized study received an additional 6-month open label metformin treatment extension. Nineteen of the 25 potential subjects were eligible to receive open label metformin based on preestablished safety and efficacy criteria. Insulin and glucose response to oral glucose challenge, cardiovascular disease risk markers (e.g. tissue plasminogen activator), weight, and anthropometric measurements were the primary outcome measures. Continued treatment with metformin resulted in further significant reductions in tissue plasminogen activator antigen levels (P = 0.02) and body mass index (P = 0.03). Reductions in insulin levels were sustained during the 6-month treatment extension. In addition, waist circumference decreased significantly in subjects continuing metformin treatment (P = 0.01). Metformin was well tolerated and no one discontinued treatment due to side effects. These data demonstrate a sustained benefit of metformin treatment to reduce hyperinsulinemia and certain markers of cardiovascular disease risk in patients with HIV infection and lipodystrophy.
INSULIN RESISTANCE, dyslipidemia, and fat redistribution are increasingly recognized among HIV-infected patients (1, 2, 3, 4, 5). Increased truncal fat and reduced peripheral fat are seen in association with hyperlipidemia and increased risk factors for cardiovascular disease (CVD) (5, 6). Short-term administration of metformin, an insulin-sensitizing agent, has been shown to improve insulin levels and to reduce central adiposity in patients with HIV infection and fat redistribution (7, 8). In a 3-month randomized placebo-controlled trial, we demonstrated that metformin (500 mg twice daily) reduced insulin levels, weight, and diastolic blood pressure among HIV-infected subjects with lipodystrophy and evidence of insulin resistance (7). Metformin treatment was also associated with reduction in the CVD risk markers, tissue plasminogen activator (tPA), and plasminogen activator inhibitor-1 (PAI-1) antigen concentrations (6). However, the benefits of prolonged metformin therapy in this population are unknown. To assess the efficacy of continued metformin therapy in HIV lipodystrophy, we investigated the effects of metformin during a 6-month open label extension among subjects who initially participated in the 3-month randomized metformin trial.
Subjects and Methods
Subjects
Detailed descriptions of the recruitment and eligibility criteria for the initial randomized trial have been reported previously (6, 7). Briefly, subjects were required to have fat redistribution documented on physical examination and have either fasting hyperinsulinemia (insulin, >15 µU/ml; 140 pmol/liter) or impaired glucose tolerance [oral glucose tolerance test (OGTT); 2 h glucose between 140–200 mg/dl (7.8–11.1 mmol/liter)]. Written informed consent was obtained from each subject before testing in accordance with the committee on human experimentation with subjects of the Massachusetts Institute of Technology and the subcommittee on human studies at the Massachusetts General Hospital.
Protocol
After completion of the 3-month randomized trial, subjects were eligible to receive open label metformin treatment for 6 months if they met the following inclusion criteria. Subjects initially randomized to placebo began open label metformin if they continued to meet the original entry criteria outlined above. Subjects initially randomized to metformin (500 mg twice daily) were continued on metformin if glucose tolerance or fasting hyperinsulinemia improved 15% or greater after the first 3 months. Subject entry and flow-through the protocol is summarized in Fig. 1
.
|
Body fat distribution was determined by dual energy x-ray absorptiometry (DEXA) using a QDR-4500A scanner (Hologic, Inc., Bedford, MA). Regions of interest, namely arms, legs, and trunk, were standardized (1995 Users Guide, Hologic, Inc.). The DEXA scan was used to determine total body fat, and regional percent body fat, such as truncal percent fat (trunk fat grams per total trunk mass) and extremity percent fat (sum of four extremities fat grams per total sum of four extremities mass). Laboratory methods were outlined previously (6, 7).
Statistical analysis
The primary outcome variables of this study were tPA, PAI-1, insulin, and glucose AUC following standard OGTT. Secondary end points included lipid concentrations, body mass index (BMI), waist to hip ratio, waist circumference, and fat distribution as determined by DEXA. Paired t tests within each group were used to compare outcome variables before and after the 6-month open label treatment. The sustained effects of metformin on tPA were confirmed using a longitudinal random effects model (data not shown) (9). Statistical analyses were performed using SAS JMP (SAS Institute, Inc., Cary, NC), and statistical significance was defined as a two-tailed 
level of P < 0.05. All data are presented as the mean ± SEM.
Results
Twenty-five subjects completed the 3-month randomized trial of metformin therapy; 11 of 14 subjects originally assigned to metformin, and 8 of 11 subjects initially assigned to placebo completed 6 months of open metformin therapy (Fig. 1
and Table 1). Three subjects increased to 850 mg, orally, twice daily; all other participants continued on 500 mg twice daily for the remainder of the study. No subject was discontinued or dropped out of the open treatment portion of the study.
|
Initiation of metformin treatment was associated with significant reductions in tPA levels (P = 0.002) and continued therapy led to further significant decreases in tPA concentrations (P = 0.02; Fig. 2). The initial reduction in PAI-1 among those initially randomized to metformin was sustained, and concentrations decreased slightly in both groups during open treatment, but this reduction was not significant. However, BMI decreased significantly for both subject groups over the 6-month open label treatment (P < 0.05 for initiators and P = 0.03 for continued treatment). Improvement in insulin AUC among subjects continuing metformin therapy for 6 months was sustained (i.e. there was no significant difference between the beginning and end of open treatment; P = 0.8; Fig. 2 and Table 2).
|
|
Two subjects changed antiretroviral regimens; one discontinued indinavir and started efavirenz 7 wk before completion, and one discontinued efavirenz and started abacavir and nevirapine 2 wk after starting metformin therapy. In addition, one subject initiated testosterone replacement therapy (5 mg transdermal patch/d) 1 month before completion. Four subjects were taking a lipid-lowering agent throughout the study. One subject started using atorvastatin after completing 6 months of metformin therapy.
Side effects, safety, and adherence
Metformin was well tolerated. Fifty percent of the subjects initiating metformin therapy reported mild diarrhea that resolved by the 6-month visit in all but two subjects. Thirty-six percent of the subjects continuing metformin therapy reported mild diarrhea, which resolved in all subjects. There were no serious adverse events, and no one discontinued medication due to side effects. No subject developed lactic acidosis. Medication compliance was 93% based on returned unused study drug.
Discussion
We demonstrate a sustained benefit of metformin therapy to reduce indexes of insulin resistance and markers of cardiovascular disease risk in patients with HIV infection and fat redistribution. After a 6-month open label treatment extension, subjects demonstrated significant continued reductions in tPA antigen levels and BMI as well as sustained benefits in insulin area under the curve. Among subjects who were initial responders to metformin, the improvement in insulin AUC seen in the 3-month randomized trial was sustained after a total of 9 months. Furthermore, our data suggest that prolonged administration of metformin is well tolerated and associated with improvement in indexes of CVD risk (i.e. insulin, BMI, waist circumference, and tPA).
Due to concerns for patient safety, the total sample size and the number of subjects who received dose increases were limited by adherence to previously established eligibility criteria. Despite the small sample size and the conservatively low dose used in most subjects, markers of CVD risk, insulin resistance, and central adiposity were improved after the 6-month open label treatment extension.
Increased levels of tPA and PAI-1 have been associated with increased risk of CVD (10, 11); therefore, it is possible that reduction in these markers with the use of metformin may improve the CVD risk profile in HIV-infected patients with fat redistribution. Longitudinal CVD outcome studies are needed to determine whether there is an increased incidence of CVD in this population and whether risk modification will be able to alter cardiovascular health. The beneficial effect of metformin on BMI and waist circumference not only persisted, but continued to improve during the 6-month treatment extension. There was an overall reduction in BMI, and total fat mass was reduced in patients continuing metformin therapy. In this group there was a similar reduction in extremity fat mass. These preliminary data suggest that fat loss associated with metformin treatment may be uniformly distributed. Metformin may therefore be most appropriate for patients with insulin resistance and truncal adiposity, as opposed to patients who have predominantly peripheral lipoatrophy in whom further overall weight loss and decreased extremity fat are not desirable.
There was no significant effect of metformin on lipid levels in either the 3-month controlled trial or the 6-month extension. Metformin has been associated with modest reductions in cholesterol and triglyceride levels in patients with type II diabetes mellitus (12, 13). The inability to show a reduction in lipid levels in the present study may be attributable to the relatively low dose of metformin used. We did not exclude patients already receiving lipid-lowering medication, and this may have limited our ability to demonstrate an additional benefit from metformin. Furthermore, the hyperlipidemia associated with HIV and fat redistribution may be partially due to direct effects of antiretroviral therapy that are not modifiable by metformin. Nonetheless, persistent hyperlipidemia may contribute to increased CVD risk in this population.
This study demonstrates that the benefits of metformin on hyperinsulinemia, CVD risk marker (tPA antigen), weight, and waist circumference were sustained in subjects who completed 6–9 months of therapy. Furthermore, additional reductions in tPA antigen levels, BMI, and waist circumference were noted with continued therapy. Metformin was well tolerated and was not associated with serious adverse events. Subjects in this study were carefully screened for preexisting impairment in liver function, and therefore our results may not be generalized to other HIV-infected populations. Future investigation is necessary to determine the potential benefits of higher doses of metformin and to evaluate the ability of metformin to mediate the development of CVD or type II diabetes mellitus in patients with metabolic complications of HIV infection and fat redistribution.
Acknowledgments
The investigators thank the nursing and dietary staff of the Massachusetts Institute of Technology General Clinical Research Center for their dedicated patient care, Dr. Izabella Lipinska and Gregory Neubauer for performance of assays, Dr. Hang Lee for biostatistical support, and Drs. Nesli Basgoz, Benjamin Davis, and Paul Sax for critical review of the manuscript.
Footnotes
This work was supported in part by NIH Grants M01-RR-300088, T32-DK-07703, K23-DK-02844, and R01-DK-59535.
S.G. is a recipient of an unrestricted educational grant from Bristol-Meyers Squibb.
Abbreviations: BMI, Body mass index; CVD, cardiovascular disease; DEXA, dual energy x-ray absorptiometry; OGTT, oral glucose tolerance test; PAI-1, plasminogen activator inhibitor-1; tPA, tissue plasminogen activator.
Received May 6, 2002.
Accepted July 1, 2002.
References
This article has been cited by other articles:
 |
|
 |
|
  J. E. Mallewa, E. Wilkins, J. Vilar, M. Mallewa, D. Doran, D. Back, and M. Pirmohamed HIV-associated lipodystrophy: a review of underlying mechanisms and therapeutic options J. Antimicrob. Chemother., October 1, 2008; 62(4): 648 - 660. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. H. Scarpello and H. C. Howlett Metformin therapy and clinical uses Diabetes and Vascular Disease Research, September 1, 2008; 5(3): 157 - 167. [Abstract] [PDF]
|
|
|
|
 |
|
 T. T. Brown Approach to the Human Immunodeficiency Virus-Infected Patient with Lipodystrophy J. Clin. Endocrinol. Metab., August 1, 2008; 93(8): 2937 - 2945. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. F Wiernsperger Review: 50 years later: is metformin a vascular drug with antidiabetic properties? The British Journal of Diabetes & Vascular Disease, September 1, 2007; 7(5): 204 - 210. [Abstract] [PDF]
|
|
|
|
 |
|
 J. P.H. van Wijk, E. J.P. de Koning, M. C. Cabezas, J. op't Roodt, J. Joven, T. J. Rabelink, and A. I. Hoepelman Comparison of Rosiglitazone and Metformin for Treating HIV Lipodystrophy: A Randomized Trial Ann Intern Med, September 6, 2005; 143(5): 337 - 346. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. C. Rudorf and S. A. Krikorian Adverse Effects Associated With Antiretroviral Therapy and Potential Management Strategies Journal of Pharmacy Practice, August 1, 2005; 18(4): 258 - 277. [Abstract] [PDF]
|
|
|
|
 |
|
 S. Grinspoon and A. Carr Cardiovascular Risk and Body-Fat Abnormalities in HIV-Infected Adults N. Engl. J. Med., January 6, 2005; 352(1): 48 - 62. [Full Text] [PDF]
|
|
|
|
 |
|
 P. Koutkia, B. Canavan, J. Breu, M. Torriani, J. Kissko, and S. Grinspoon Growth Hormone-Releasing Hormone in HIV-Infected Men With Lipodystrophy: A Randomized Controlled Trial JAMA, July 14, 2004; 292(2): 210 - 218. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Benavides and M. C Nahata Pharmacologic Therapy for HIV-Associated Lipodystrophy Ann. Pharmacother., March 1, 2004; 38(3): 448 - 457. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Lee Altered Glucose Metabolism in HIV-infected Patients Treated With HAART Journal of Pharmacy Practice, February 1, 2004; 17(1): 80 - 86. [Abstract] [PDF]
|
|
|
|
 |
|
 H. Yki-Jarvinen, J. Sutinen, A. Silveira, E. Korsheninnikova, R. M. Fisher, K. Kannisto, E. Ehrenborg, P. Eriksson, and A. Hamsten Regulation of Plasma PAI-1 Concentrations in HAART-Associated Lipodystrophy During Rosiglitazone Therapy Arterioscler Thromb Vasc Biol, April 1, 2003; 23(4): 688 - 694. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
, Subjects initially randomized to placebo (PBO; n = 8) who started metformin (MET) at the 3 month visit. 
, Subjects initially randomized to metformin who continued on open therapy (n = 11). - - -, The double-blind portion of the study; —, open label treatment. By paired t test comparing 3 months to 9 months in each group: **, P = 0.002; *, P = 0.02.