Metformin maintains the weight loss and metabolic benefits following rimonabant treatment in obese women with polycystic ovary syndrome (PCOS)

Summary Objective Rimonabant has been shown to reduce weight, free androgen index (FAI) and insulin resistance in obese patients with polycystic ovary syndrome (PCOS) compared to metformin. Studies have shown that significant weight regain occurs following the cessation of rimonabant therapy. This study was undertaken to determine if subsequent metformin treatment after rimonabant would maintain the improvement in weight, insulin resistance and hyperandrogenaemia in PCOS. Design An extension study for 3 months with the addition of metformin to the randomised open labelled parallel study of metformin and rimonabant in 20 patients with PCOS with a body mass index ≥ 30 kg/m2. Patients who were on 3 months of rimonabant were changed over to metformin for 3 months, whereas those on 3 months of metformin were continued on metformin for another 3 months. Measurements The primary end-point was a change in weight; secondary end-points were a change in FAI and insulin resistance. Results The mean weight loss of 6·2 kg associated with 3 months of rimonabant treatment was maintained by 3 months of metformin treatment (mean change +0·2 kg, P = 0·96). Therefore, the percentage reduction in weight remained significantly higher in the rimonabant/metformin group compared to metformin only subjects at 6 months compared to baseline (–6·0 ± 0·1%vs. –2·8 ± 0·1%, P = 0·04). The percentage change in testosterone and FAI from baseline to 6 months was also greater in the rimonabant/metformin group. [Testosterone (–45·0 ± 5·0%vs. –16 ± 2·0%, P = 0·02); FAI (–53·0 ± 5·0%vs. –17·0 ± 12·2%, P = 0·02)]. HOMA-IR continued to fall significantly in the rimonabant/metformin group between 0, 3 and 6 months (4·4 ± 0·5 vs. 3·4 ± 0·4 vs. 2·7 ± 0·3, respectively, P