Metformin Effects on Metabolic, Vitamin B12, Iron, Erythropoietin, and Hematological Parameters in Women With PCOS

Polycystic ovary syndrome (PCOS) is a common and heterogeneous endocrine and metabolicdisorder characterized by reproductive, endocrine, and metabolic abnormalities. Insulinresistance, hyperinsulinemia, dyslipidemia, excess weight, and disturbances inreproductive function are important components of the PCOS phenotype. Metformin is aninsulin‐sensitizing medication widely used to address metabolic abnormalities associatedwith PCOS.This double‐blind, placebo‐controlled, parallel‐group randomized controlled trial wasdesigned to evaluate longitudinal changes in metabolic, vitamin B12, iron‐status,erythropoietic, and hematological parameters among women with PCOS during six months ofmetformin therapy. The study also examined selected associations among demographic,clinical, reproductive, familial, metabolic, vitamin B12, iron‐status, erythropoietic,and hematological characteristics.Women aged 18‐40 years with PCOS confirmed by a specialist in Obstetrics and Gynecologywere eligible for participation. Participants were excluded if they had chronic hemolyticanemia, chronic kidney disease, endocrine disorders other than PCOS, hepatic disorders,metabolic disorders, cardiovascular disease, or other concurrent medical conditions thatcould influence glucose metabolism, iron status, vitamin B12, erythropoietin, orhematological parameters. Women who had received insulin‐sensitizing agents, ironsupplementation, hormonal therapy, vitamin B12 supplementation, or other medicationsknown to substantially affect the study outcomes during the predefined period beforeenrollment were also excluded.Following baseline assessment, eligible participants were randomly allocated in a 1:1ratio using a computer‐generated random allocation sequence to either the metforminintervention group or the matching placebo group. Participants in the intervention groupreceived metformin at a total daily dose of 1500 mg, administered in three divided doseswith meals. Participants in the control group received a corresponding matching placeboaccording to the same administration schedule. Both groups continued to receive routineclinical care.The double‐blind design was maintained throughout the study. Participants were blinded totreatment allocation, and investigators involved in participant assessment and outcomeevaluation were also blinded to treatment assignment. Matching preparations were used tominimize the possibility of identifying treatment allocation based on appearance.Clinical, anthropometric, biochemical, iron‐status, vitamin B12, erythropoietic, andhematological assessments were performed at three predefined time points: baseline, 3months, and 6 months. Metabolic assessments included body mass index, fasting bloodglucose, serum insulin, HOMA‐IR, total cholesterol, triglycerides, LDL‐C, and HDL‐C.Iron‐status assessments included serum iron, total iron‐binding capacity, ferritin, andtransferrin saturation. Serum vitamin B12 and erythropoietin were also assessed.Hematological assessments included hemoglobin, hematocrit, red blood cell count, meancorpuscular volume, mean corpuscular hemoglobin, mean corpuscular hemoglobinconcentration, red cell distribution width, white blood cell count, lymphocytes,neutrophils, eosinophils, basophils, monocytes, and platelet count.The study was conducted in obstetrics and gynecology clinics affiliated with the Ministryof Health in Jordan. The study period extended from August 2025 to February 2026. A totalof 400 women were initially enrolled and randomly allocated, with 200 participantsassigned to each treatment group. During the six‐month follow‐up period, 100 participantsdid not complete the required assessments, leaving 300 participants in the finallongitudinal analysis, comprising 150 participants in the metformin group and 150participants in the matching placebo group.No additional structured educational, behavioral, lifestyle, or self‐managementintervention was introduced as part of the study protocol. Participants continued toreceive routine clinical care. Lifestyle behaviors, including physical activity, dietaryintake, and adherence to lifestyle recommendations, were not systematically quantifiedusing validated assessment instruments.Longitudinal changes across baseline, 3 months, and 6 months were evaluated usingrepeated‐measures analysis of variance. Associations between selected continuousvariables were examined using Pearson's correlation coefficient when appropriate, whileSpearman's rank‐order correlation coefficient was used for ordinal variables or variablesthat did not meet the assumptions required for Pearson correlation. Correlation analyseswere considered exploratory and hypothesis‐generating because multiple associations wereexamined.The study received ethical approval from the Ethical Board of the Jordanian Ministry ofHealth, Amman, Jordan (Approval No. 99923), and the Institutional Review Board ofAl‐Ahliyya Amman University (Approval No. 2026‐2025/30/1). Written informed consent wasobtained from all participants before enrollment. The study was conducted in accordancewith the ethical principles of the Declaration of Helsinki.