Metformin-sustained Weight Loss and Reduced Android Fat Tissue at 12 Months in EMPOWIR (Enhance the Metabolic Profile of Women With Insulin Resistance)

A Double-Blind, Placebo-Controlled, Randomized Trial of Normoglycemic Women With Midlife Weight Gain

Harriette Mogul, MD, MPH; Ruth Freeman, MD; Khoa Nguyen, MPH


Endocr Pract. 2016;22(5):575-586. 

In This Article

Abstract and Introduction


Objective: To assess 12-month body weight (BW) and body composition changes in normoglycemic women with midlife weight gain, after dietary and pharmacologic interventions targeting hyperinsulinemia.

Methods: EMPOWIR (Enhance the Metabolic Profile of Women With Insulin Resistance; NCT00618072) was a double-blind, placebo-controlled, 12-month trial of women with >20-pound weight gain, normal glucose tolerance test, and increased area-under-the-curve insulin. Subjects (mean ± SD, 46.7 ± 6.5 years of age; body mass index, 30.8 ± 2.8 kg/m2; 50% white) attended 4 nutrition workshops to introduce a novel carbohydrate-modified diet (CMD) and were then randomized to one of three arms for 6 months (phase 1): CMD alone (D), or in combination with metformin (M), or metformin + rosiglitazone (MR), with rerandomization of the D group to one of the active treatment arms (phase 2, months 7 through 12). Repeated measure analysis of variance was used to assess BW at baseline, 6 months, and 12 months in 32 subjects with 12-month data; paired t tests compared baseline and 12-month dual-energy X-ray absorptiometry–derived body composition.

Results: Mean (±SD) BW decreased significantly at 12 months in the M arm: 85.1 ± 8.5 kg to 79.8 ± 9.0 kg (P = .0003), with 54% of variance in weight over time explained by M treatment. Mean (±SD) percent android fat decreased significantly in the M and D arms: 53.5 ± 4.8% to 49.3 ± 7.6% (P = .010) and 52.9 ± 6.2% to 48.1 ± 8.7% (P = .021).

Conclusion: In combination with a novel carbohydrate modified diet, metformin enhanced 12-month weight loss and improved body composition in ethnically diverse normoglycemic, hyperinsulinemic women with midlife weight gain. These findings suggest that EMPOWIR's easily implemented dietary interventions, alone and in combination with pharmacotherapies that target hyperinsulinemia, merit additional investigation in larger, long-term studies.


Progressive midlife weight gain in the fourth and fifth decades, a common phenomenon among women of all ethnic groups, has profound health implications. As first demonstrated by the Nurses' Health Study, weight gain of ≥10 kg (22 pounds) after age 18 is associated with development of type 2 diabetes[1,2] with a linear effect on age-adjusted cardiovascular[3,4] and all-cause mortality.[5] Preventive interventions targeting this high-risk population are critically needed. We previously identified a distinct subgroup of healthy appearing women with midlife weight gain, increased glucose-mediated insulin response curves, and normal glucose tolerance tests[6,7] and reported that metformin, in combination with a novel hypocaloric carbohydrate-modified diet (CMD), produced significant and sustainable weight loss and normalized fasting insulin levels in this population.[8,9] The EMPOWIR (Enhance the Metabolic Profile of Women with Insulin Resistance; NCT00618072) study was conducted to evaluate these interventions in additional, more diverse populations.

The trial was designed to test the hypotheses that progressive midlife weight gain represents an early manifestation of insulin resistance and hyperinsulinemia in distinct female subsets and that strategies which decrease insulin elevations will reduce body weight (BW), improve body composition (BC), and attenuate risks of diabetes and related comorbidities.[10] The study rationale reflects the perspective that hyperinsulinemia can precede and promote obesity[11–13] and the premise that "lowering insulin secretion in hyperinsulinemic individuals may be beneficial",[14] as suggested by other investigators.[15,16]

Our initial report demonstrated significant and comparable 6-month weight loss in women randomized to CMD + placebo (D), metformin (M), and metformin + rosiglitazone (MR) groups: 4.7, 5.4, and 5.5% (P = .049, .002, and .032, respectively), with significant decline in 6-month fasting insulin, the primary study outcome variable, limited to the M group: 12.5 to 8.0 μU/mL, (P = .026). The findings were based on an intention-to-treat analysis of 44 randomized study subjects, 39 of whom completed 6 months of the study.[10]

The current report assesses BW and BC, prespecified secondary outcome variables, at 12 months in 32 subjects who completed the final 12-month visit and in 27 participants with available BC data. Dual-energy X-ray absorptiometry (DXA), a widely used surrogate marker of BC in clinical trials,[17–19] was utilized to evaluate the composite effects of the study interventions on relevant fat compartments.