 |
 |
 |
 |
 |
 |
 |
|
|
|
|||||||||
|
|||||||||||
Assistant Professor of Pharmacy Practice, Massachusetts College of Pharmacy and Health Sciences–Worcester/Manchester, Manchester, NH
PharmD Student, Massachusetts College of Pharmacy and Health Sciences–Worcester/Manchester
Assistant Professor of Pharmacy Practice, Massachusetts College of Pharmacy and Health Sciences–Worcester/Manchester
Reprints: Dr. Desilets, Massachusetts College of Pharmacy and Health Sciences, 1260 Elm St., Manchester, NH 03101, fax 603/314-0209, Alicia.Desilets{at}mcphs.edu
 |
Abstract |
|---|
 Top Abstract Data Sources and SelectionTrials in AdultsTrials in AdolescentsDiscussionSummaryReferences |
|---|
DATA SOURCES: Literature was obtained through MEDLINE Ovid (1950–February week 3, 2008), EMBASE (all years), and a bibliographic review of relevant articles. Key words included metformin, obesity, overweight, and weight loss.
STUDY SELECTION/DATA EXTRACTION: All studies published in the English language that evaluated the effects of metformin on weight in obese or overweight individuals were critically analyzed. Relevant articles were selected for inclusion in this review.
DATA SYNTHESIS: Metformin is first-line pharmacotherapy in the treatment of overweight or obese patients with type 2 diabetes, with beneficial effects on weight in this population. Multiple trials have evaluated the effect of metformin on weight and other metabolic parameters in adults and adolescents without diabetes. Five of 12 trials in adults evaluated weight loss as a primary endpoint. Significant weight reduction was found in 4 of these studies; however, the trials were small and of weak design. Weight reduction was significant in 5 of the 6 adolescent trials; similarly, these studies were limited by weak study design and small patient population. Metabolic parameters (blood pressure, waist circumference, cholesterol parameters, insulin/glucose levels) often showed varying results. Metformin was well tolerated; gastrointestinal effects were the most frequently reported adverse effects.
CONCLUSIONS: The weight loss effects of metformin in overweight or obese adults and adolescents without diabetes appear promising; however, trials have been limited by small patient populations and weak design. Metformin may also have a positive effect on metabolic parameters such as waist circumference, fasting insulin and glucose levels, and triglycerides. Further research involving large-scale trials that evaluate weight loss as a primary outcome is necessary to firmly establish the role of metformin in this population.
Key Words: metabolic syndrome, metformin, obesity
Published Online, May 13, 2008. www.theannals.com, DOI 10.1345/aph.1K656
Many health consequences are associated with overweight and obesity including heart disease, diabetes, asthma, sleep apnea, arthritis, reproductive complications, gallbladder disease, incontinence, increased risk of surgical complications, depression, and certain types of cancer. Moreover, when compared with patients at a healthy weight, individuals who are obese are at a 50–100% increased risk of premature death.6 Also of concern in this population is the high risk of the metabolic syndrome, characterized by an increase in waist circumference, blood pressure, blood glucose, and triglycerides, as well as reduced high-density lipoprotein cholesterol (HDL-C) levels.7,8 Patients with the metabolic syndrome are at increased risk of cardiovascular disease and type 2 diabetes mellitus.9 Even moderate weight loss (5–15% reduction in total body weight) in patients who are overweight or obese can reduce many of these health sequelae.6
Lifestyle modifications, including caloric restriction and increased physical activity, are first-line therapy for weight loss; however, patients are often unable to achieve clinically meaningful weight loss with lifestyle modifications alone.10,11 Weight regain is another area for concern; patients who lose weight while implementing lifestyle modifications commonly regain the lost weight within one year.12 Pharmacotherapy may be a necessary addition in some obese patients to achieve long-term weight loss.4 Orlistat and sibutramine are the only agents currently indicated by the Food and Drug Administration (FDA) for the long-term treatment of overweight and obesity in adults.4 Orlistat is the only pharmacotherapeutic agent indicated by the FDA for the treatment of overweight adolescents13; however, gastrointestinal effects often limit its use in both adults and adolescents.12 Sibutramine may increase blood pressure and heart rate, limiting its use in patients with cardiovascular risk factors.12 Due to these limitations, alternative pharmacotherapeutic options for overweight or obese adults and adolescents require further investigation.
Antidiabetic agents, including acarbose,14 exenatide,15-19 and metformin,20-22 have demonstrated a potential for weight reduction in patients with diabetes. Metformin, a biguanide oral hypoglycemic agent, has been shown to reduce weight, hyperinsulinemia, and hyperglycemia in adult patients with type 2 diabetes.21,23 It is first-line pharmacotherapy in overweight and obese patients with type 2 diabetes.21,22 Further, metformin has demonstrated additional health benefits independent of its effects on glucose and insulin levels; these include inhibition of platelet aggregation, antioxidant activity, weight reduction, effects on lipid parameters (total cholesterol, HDL-C, low-density lipoprotein cholesterol, triglycerides), and arterial hypertension.24,25
Several of metformin's mechanisms have been evaluated for weight loss effects. These include reduction in gastrointestinal absorption of carbohydrates and insulin resistance,24,26 induction of an anorectic and lipolytic effect,27 and decreased levels of leptin (a hormone integral to body weight regulation).24,28 Metformin can be given safely to euglycemic patients, as it does not induce hypoglycemia.29
|
Data Sources and Selection |
|---|
|
TopAbstractData Sources and SelectionTrials in AdultsTrials in AdolescentsDiscussionSummaryReferences |
|---|
|
Trials in Adults |
|---|
|
TopAbstractData Sources and SelectionTrials in AdultsTrials in AdolescentsDiscussionSummaryReferences |
|---|
|
To date, trials primarily evaluating metformin's effects on weight have been small. Metformin 850 mg 3 times a day was evaluated in a trial (N = 31) that enrolled obese (BMI >30 kg/m2) adults over the course of 28 weeks.24 Unlike most trials evaluating metformin's effects on weight, patients were instructed to maintain their pre-study lifestyle because the authors wished to evaluate the effects of metformin in the absence of dietary restriction and exercise regimens. At study end, patients had a significant decrease in weight (Table 1). Leptin levels also declined significantly.
In another small-scale trial (N = 26), Tankova et al.32 evaluated the effects of metformin (titrated to 850 mg 2–3 times daily) on weight in obese adults with normal glucose tolerance. After 6 months, a nonsignificant decrease in weight of 3.24% was found; 15.4% of participants had more than a 5% weight loss. Of note, a significant correlation was found between weight loss and insulin resistance. Metformin was well tolerated, with 4 subjects reporting gastrointestinal adverse effects during the first week of treatment only. The small sample size and lack of a control group limit application of these results; however, these small-scale trials do provide preliminary evidence of metformin's potential benefit in overweight patients.24,32
In a 6-month trial, the effects of metformin 850 mg twice daily (n = 50), sibutramine 10 mg twice daily (n = 50), or orlistat 120 mg 3 times daily (n = 50) were evaluated in female patients with a BMI greater than 30 kg/m2.33 All study groups had significant reductions in weight and BMI. Of note, patients in the sibutramine group had the greatest reduction in BMI compared with the other 2 groups. Metformin's gastrointestinal adverse effects were better tolerated than those of orlistat; 22 patients reported adverse events in the orlistat group versus 14 in the metformin group. Results from this trial are important in that metformin, when compared with 2 agents already established as effective weight loss agents, was equally as effective as orlistat in reducing weight and BMI.
In a trial of 57 obese women, the combination of metformin 850 mg twice daily with orlistat 120 mg 3 times daily (n = 27) was compared with orlistat 120 mg 3 times daily alone (n = 30).34 Patients were randomized to 1 of the 2 treatment groups for 3 months of therapy. At study end, both groups had a significant decrease in weight; however, there was no significant difference between groups. Both therapies were well tolerated, and no participants discontinued therapy due to an adverse event. Results from this study do not support the addition of metformin in patients receiving orlistat; however, the small sample size should be considered when interpreting these results.
In another trial, obese and overweight women with a BMI of greater than 30 kg/m2 or a BMI greater than 27 kg/m2 with other cardiovascular risk factors received either fluoxetine 20 mg daily plus metformin 500 mg 3 times a day (case group; n = 66) or no treatment (control group; n = 25).35 Weight and BMI changes in the case group were significantly different from baseline and when compared with the control group. Neither weight reduction nor BMI decreased significantly in the control group. Thirty-two patients in the case group were treated for more than 6 months and 12 were treated for more than 12 months; results continued to show significant weight reductions of 9.43 and 9.83 kg, respectively. Twelve percent of patients experienced transient gastrointestinal adverse effects. Clinically meaningful results from this trial are questionable due to the open-label design and the lack of a placebo group. Moreover, the decision to take drug therapy at trial initiation was left to the patient; this lack of randomization could lead to study bias favoring weight loss in patients who were willing to use drug therapy.
TRIALS THAT EVALUATED WEIGHT LOSS AS A SECONDARY OUTCOME
Numerous trials have evaluated metformin and its potential benefit in
preventing diabetes and/or its effects on the metabolic syndrome (or its
components).25,36-48
Because weight loss reduces the risk of these important health consequences,
it is often evaluated as a secondary outcome in these trials
(Table 2).
|
The DPP (Diabetes Prevention Program)36 was a large-scale (N = 3234) trial that primarily evaluated the incidence of diabetes at the end of the treatment period among patients treated with either metformin, standard lifestyle intervention (placebo group), or intensive lifestyle modification. Weight loss was reported as minimal, with the intensive lifestyle modification group, having the best results, followed by the metformin group then the placebo group.
A recent publication reported on a subset of patients (N = 758) from the DPP who did not develop diabetes after the first year.37 Weight results in this subgroup were similar to those of the original population; intensive lifestyle modification resulted in the greatest weight reduction (–8.3 kg males; –7.0 kg females), followed by metformin (–2.8 kg males; –3.0 kg), then placebo (–0.3 kg males; –0.01 kg females). This landmark trial shows the potential for metformin to cause significant weight reduction in patients at risk for diabetes. However, intensive lifestyle modification demonstrated the best results, further establishing the drug's role as first-line therapy in overweight or obese adults.
The BIGPRO (Biguanides and Prevention of the Risks in Obesity) trial evaluated the relationship between insulin resistance and cardiovascular disease by initiating metformin (an insulin sensitizer) in insulin-resistant patients.38,39 The preliminary stage was reported in the BIGPRO1 trial.38 This trial randomized patients to receive metformin 850 mg daily or placebo; metformin-treated patients had a nonsignificant decrease in weight after 1 year. All patients were contacted concerning adverse events, regardless of whether they completed the trial. Diarrhea was the most reported adverse event in the metformin-treated group, with 5.3% reporting severe diarrhea (2.2% in the placebo group) and 14.5% reporting moderate diarrhea (2.2% in the placebo group). Nausea and vomiting were the only other adverse events that were reported more frequently in the metformin group (2.2% severe, 4% moderate) than in the placebo group (0.9% severe, 1.7% moderate).
In a trial of similar design, the BIGPRO 1.2 evaluated the effects of
metformin 850 mg twice daily or placebo in male patients with elevated blood
pressure (systolic blood pressure 
140 and 
180 mm Hg or diastolic
90 and 105 mm Hg or in treatment for hypertension), waist-to-hip
circumference ratio 0.95, and triglycerides (150 and 600
mg/dL).25
The main objective of this trial was to evaluate the effects of metformin on
triglycerides and blood pressure; a nonsignificant change in weight was found
at the end of the 3-month treatment period.
Metformin was evaluated in a trial involving 60 patients without diabetes who had previously undergone coronary artery bypass surgery or angioplasty.40 Fifty-nine patients were included in the data analysis. All patients received lovastatin 40 mg and metformin (average daily dose 1759 mg) for 12 weeks. A significant decrease in weight was observed in patients with a BMI greater than 27 kg/m2 (–2.3 kg) versus baseline and patients with a BMI less than 27 kg/m2 (–0.7 kg). Similarly, fasting insulin levels decreased more in patients with a BMI greater than 27 kg/m2 (–1.41 µIU/mL) than in those with a BMI less than 27 kg/m2 (–0.11 µIU/mL); however, this was not a significant decrease from baseline.
A small trial evaluated the effects of metformin 500 mg or placebo twice daily on the components of the metabolic syndrome in patients at risk for type 2 diabetes.41 A nonsignificant decrease in weight was found in patients taking metformin versus the placebo group.
A smaller (N = 19) crossover trial evaluated the use of metformin for 6 weeks (850 mg for 1 wk, titrated to 850 mg twice daily for 5 wk) in obese (BMI >30.0 kg/m2) patients with impaired glucose tolerance.42 A small, nonsignificant decrease in weight was observed (–0.6 kg); results from this trial are limited by its short duration.
In a retrospective analysis of hyperinsulinemic, euglycemic women (N = 21) with syndrome W (an early marker for the metabolic syndrome characterized by midlife weight gain, appetite dysregulation, and abnormal glucose-mediated insulin response curves), metformin (mean dose 2000 mg daily) showed positive results.43 Mean body weight and mean waist circumference decreased significantly at all visits from baseline (–11.8 kg and –9.4 cm, respectively). Metformin was well tolerated, with no reported adverse effects. Both the population size and the retrospective nature of this study limit its applicability to other populations.
Metformin's effects on glucose metabolism were studied in African American subjects at risk for type 2 diabetes.44 Subjects were randomized to receive either metformin 500 mg daily or placebo; metformin-treated individuals experienced a modest weight reduction, while those in the placebo group had a significant increase in weight (p < 0.02) and BMI (p < 0.01; Table 2). Although metformin was well tolerated, the low dose should be noted; the nonsignificant weight reduction in people treated with metformin could be due, in part, to this dose.
|
Trials in Adolescents |
|---|
|
TopAbstractData Sources and SelectionTrials in AdultsTrials in AdolescentsDiscussionSummaryReferences |
|---|
|
A trial of 24 nondiabetic, hyperinsulinemic, normoglycemic overweight adolescents (BMI >30 kg/m2) demonstrated that metformin, titrated to 850 mg twice daily, combined with a hypocaloric diet is associated with significant weight loss and reduction in body fat versus placebo.50 The study further demonstrated reduction in plasma leptin, fasting insulin, cholesterol, and triglyceride levels. Minor and transient adverse effects were reported in 5 of the 12 subjects; these included mild nausea (n = 5), dizziness (n = 2), and loose stools (n = 2). This preliminary study demonstrates the potential benefits of metformin in the management of overweight hyperinsulinemic adolescents.
Another trial evaluated the effects of metformin 500 mg twice daily on overweight (BMI >30 kg/m2) adolescent patients aged 12–19 years.51 Additional inclusion criteria were a fasting insulin concentration greater than 15 µU/mL (hyperinsulinemia) and at least 1 first- or second-degree relative with type 2 diabetes. Patients treated with metformin experienced a reduction in BMI, whereas BMI in the placebo group increased (p < 0.02) versus baseline. Fasting blood glucose and insulin levels also improved in the metformin group; however, these changes were not significant in the placebo group. In contrast, benefits concerning insulin sensitivity, HbA1c, and serum lipids were not significant. Reduction in leptin was significant only in female patients (5.5%; p < 0.05 vs placebo). Metformin was well tolerated by most patients; 3 experienced abdominal discomfort/diarrhea that resolved by the second week of therapy, 1 patient reported intermittent nausea that resolved after dose reduction, and 1 patient had a migraine exacerbation. Limitations of these data include small sample size, short duration, and the fact that the treatment and control groups were not matched precisely for ethnic background, sex, or initial BMI. Despite these limitations, this trial provides preliminary evidence of metformin's benefit in adolescents at increased risk of type 2 diabetes.
The role of metformin was evaluated in overweight (mean baseline BMI 35.2 kg/m2) nondiabetic adolescents aged 9–18 years.28 The participants received standardized information on lifestyle modifications along with metformin 1 g twice daily and placebo for 6 months each, with a 2-week washout period. Treatment with metformin had a significant beneficial effect over placebo on weight, BMI, and waist circumference. Metformin also showed benefits in fasting insulin and a small but significant beneficial effect on fasting glucose. Compared with placebo, there was no significant benefit associated with metformin on insulin sensitivity; this may be suggestive of inadequate visceral fat loss, which is implicated in the development of insulin resistance in adolescents.55,56 In general, metformin was well tolerated; only 2 participants were unable to tolerate metformin 1 g twice a day due to nausea.28 These participants were able to tolerate 750 mg twice daily.
Lustig et al.52 analyzed the effects of metformin in overweight (BMI >95th percentile for age) children and adolescents aged 2–18 years (mean 12.5) whose BMI continued to increase after a 3-month trial of lifestyle modifications. An oral glucose tolerance test was conducted on each patient to determine corrected insulin response at glucose peak (CIRgp) and composite insulin sensitivity index (CISI). Subgroup analysis determined that reduction in BMI was significant after 4 months only in white pubertal females with CIRgp of 1.0 and CISI of 1.0 (–2.7 kg/m2; p < 0.001); this reduction was not sustained after 16 months (–1.6 kg/m2; p = 0.32). Increasing insulin sensitivity (as determined by CISI) was associated with less weight loss. While this study was limited by its sample size and study design (open-label, possible placebo effects), it supports the thought that metformin may be more effective for weight loss in overweight adolescents with insulin resistance.
In a subgroup of Chinese adolescents, metformin 500 mg twice daily was
evaluated for its effect on metabolic
parameters.53
Overweight adolescents aged 10–16 years who were diagnosed with
metabolic syndrome (defined as 3 of the following: central obesity,
systolic or diastolic blood pressure >95th percentile for age and sex,
triglycerides >95th percentile for age and sex, HDL-C <5th percentile,
impaired glucose tolerance, type 2 diabetes) underwent treatment with
metformin and lifestyle modifications. There was a significant reduction in
BMI in the 20 patients who completed 3 months of therapy. This weight loss was
greater than that seen in other trials; this may be due, in part, to the
shorter trial duration. There were also significant reductions in all other
metabolic parameters (Table 3).
Minor gastrointestinal adverse effects were reported early in treatment: 7 of
the metformin-treated patients experienced appetite reduction, 2 patients
reported slight nausea, and 2 patients experienced increased bowel movements.
Although this was a short-term, open-label trial with a small sample size, it
provides further evidence of metformin's positive effects on metabolic
parameters.
A recent retrospective chart review evaluated the effects of lifestyle
interventions with and without metformin on components of the metabolic
syndrome and relative BMI (actual BMI divided by BMI for age and sex
multiplied by 100) in 63 overweight (BMI 95th percentile for age and sex)
adolescents aged 11–18
years.54
Twelve of the 37 patients receiving metformin had type 2 diabetes; no patients
in the non-metformin group had diabetes. Reduction in relative BMI was
observed in all patients, although it was significant only in patients treated
with metformin. However, it is important to note that patients in the
metformin group weighed more at baseline. Metformin also had positive effects
on metabolic parameters; Table
3 indicates results in patients without diabetes. Data were
analyzed in both treatment groups and in patients with and without diabetes;
interestingly, those without diabetes were more likely to lose weight. These
results call to question the theory that insulin resistance leads to greater
weight reduction with metformin. Although the retrospective design and small
sample size limit generalization of study results, this trial corroborates
weight loss and improvement in metabolic parameters found by other trials.
|
Discussion |
|---|
|
TopAbstractData Sources and SelectionTrials in AdultsTrials in AdolescentsDiscussionSummaryReferences |
|---|
To date, 5 adult trials have evaluated metformin's effects on weight as a primary outcome; in general, these trials are small (26–66 metformin-treated patients), of short duration (3–7 mo), and of weak study design (3 are open-label, none are randomized and placebo-controlled). Weight loss with metformin in these trials ranged from 2.9 to 9.0 kg versus baseline. However, these results must be interpreted with caution, as many of these trials utilize lifestyle modification, and without a placebo group, it is unknown whether metformin is responsible for additional weight loss.
Other trials evaluating weight loss as a secondary outcome were of longer duration; placebo-subtracted weight loss in randomized, placebo-controlled trials lasting 12 months or longer was 1.2–2.5 kg. The significance of this weight loss was inconsistent. Also, because weight was evaluated as a secondary outcome, these trials may have been insufficiently powered to evaluate this finding. In trials comparing metformin with other agents for weight loss, metformin was found to be equally effective as orlistat in reducing weight and BMI, but metformin was inferior to the centrally acting anorectics sibutramine and fenfluramine (which has been withdrawn from the market). Further, the addition of metformin to orlistat does not appear to provide any additional benefit. Our findings are consistent with those of a systematic review of metformin for weight loss in adults57; insufficient evidence is available to recommend metformin for treatment of overweight or obese adults.
The 6 trials in overweight adolescents have limitations similar to those of the studies conducted in adults: small sample size (12–45 metformin-treated patients) and weak study design (2 open-label, 1 retrospective chart review). In general, the duration of trials in adolescents was short (2–12 mo), although one open-label trial lasted 16 months, and the retrospective chart review reported data up to 61 months (but does not report the number of patients treated for this length of time). Weight reduction was significant in all but one trial; however, the manner of reporting weight loss was inconsistent, as some trials reported kilograms lost, while others reported percentage lost or change in BMI, and one trial reported change in relative BMI. Due to the difference in reported results, it is impossible to draw comparisons. In the 2 randomized, placebo-controlled trials, placebo-subtracted weight loss ranged from 1.4 to 2.9 kg over 2–6 months.
|
Summary |
|---|
|
TopAbstractData Sources and SelectionTrials in AdultsTrials in AdolescentsDiscussionSummaryReferences |
|---|
|
References |
|---|
|
TopAbstractData Sources and SelectionTrials in AdultsTrials in AdolescentsDiscussionSummaryReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. Diamanti-Kandarakis, C. D Christakou, E. Kandaraki, and F. N Economou Metformin: an old medication of new fashion: evolving new molecular mechanisms and clinical implications in polycystic ovary syndrome Eur. J. Endocrinol., February 1, 2010; 162(2): 193 - 212. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Casazza and O. Thomas Do Dietary Modifications Made Prior to Pubertal Maturation Have the Potential to Decrease Obesity Later in Life? A Developmental Perspective ICAN: Infant, Child, & Adolescent Nutrition, October 1, 2009; 1(5): 271 - 281. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
 |
 |
 |
 |
