Effects of metformin withdrawal after long and short term treatment in PCOS: observational longitudinal study.

BACKGROUND Metformin plays a consolidated role in the management of polycystic ovary syndrome (PCOS). However, there is no clear answer on how long we should treat and on how long its beneficial impact sustain after we stop treatment. We compared the effects of metformin withdrawal after long-term (LT) and short term (ST) treatment in PCOS women that had previously well responded to metformin. METHODS We conducted observational longitudinal study including 44 PCOS women (31 (28-36) years and BMI 32.5 (27.7-34.9) kg/m2) that were followed for 6 months after metformin withdrawal. Prior inclusion, ST group had been treated with metformin on average for 1.03 ± 0.13 year, LT group for 5.07 ± 2.52 years. We followed anthropometric, metabolic, reproductive parameters and eating behavior as assessed by TFEQ-R18. RESULTS After metformin withdrawal, ST group gained significant amount of weight (from 92 (75.5-107.3) kg to 96 (76-116) kg; p = 0.019). Weight tended to increase also in LT users (from 87 (75-103) to 87 (73-105) kg; p = 0.058). More women in LT group maintained stable weight (27% in LT group vs 15% in ST group). Eating behavior deteriorated in both groups. Withdrawal of metformin resulted in a decrease of menstrual frequency (6 (6-6) to 6 (4-6) menstrual bleeds per 6 months; p = 0.027) and in borderline increase of androstenedione (6.4 (4.6-7.6) to 7.8 (4.8-9.6) nmol/L; p = 0.053) in LT group. Waist circumference, HOMA and glucose homeostasis remained stable in both groups. There were no differences between groups at 6-month follow up. CONCLUSION Collectively, present study implies some metabolic and endocrine treatment legacy in both groups as well as some group-specific deteriorations in clinical parameters 6 months after metformin withdrawal. TRIAL REGISTRATION The study is registered at Clinical Trials with reference No. NCT04566718.