Effects of carnitine on glucose and lipid metabolic profiles and fertility outcomes in women with polycystic ovary syndrome: A systematic review and meta-analysis
Yi Gong,Tong Jiang,Hui He,Yu Wang,Guo-Lin Wu,Ying Shi,Qinjun Cai,Can-Li Xiong,Rong Shen,Jian Li
DOI: https://doi.org/10.1111/cen.14885
2023-01-01
Clinical Endocrinology
Abstract:ObjectivesTo quantify the effect of carnitine on glucose and lipid metabolic profiles and fertility outcomes in women with Polycystic ovary syndrome (PCOS). DesignA systematic review and meta-analysis were conducted. PatientsWomen with PCOS diagnosed by Rotterdam or Androgen Excess Society (AES) criteria and taking carnitine supplement were assessment. MeasurementsFertility outcomes (ovulation, clinical pregnancy, live birth, and miscarriage), lipid parameters (BMI, triglyceride, total cholesterol, high-density lipoprotein, low-density lipoprotein), fasting glucose and insulin, and Homeostatic Model Assessment for Insulin Resistance (HOMA-IR). ResultsIn total, 839 participants were included in this analysis. The dosage of carnitine and treatment duration reported by studies varied from 250 mg to 3000 mg daily and 84 to 90 days, respectively. The publication bias was absent. Compared with placebo, carnitine significantly improved ovulation rates (RR 3.42, 95% CI 2.39 to 4.89, I2 = 0%) and pregnancy rates (RR 11.05, 95% CI 1.21 to 100.58, I2 = 79%). None of included studies reported live birth. After treatment, carnitine resulted in significant reductions relative to baseline in body mass index (BMI, MD -0.93 kg/m2, 95% CI -1.15 to -0.70, I2 = 55.0%), insulin levels (MD -2.47 mIU/L, 95% CI -4.49 to -0.45, I2 = 0%) and the Homeostasis Model Assessment index (MD -0.67, 95% CI -1.20 to -0.14, I2 = 0%) than placebo, but not for lipid profiles including triglyceride, total cholesterol, and low-density lipoprotein. ConclusionWith the available literature, carnitine seems to improve ovulation and clinical pregnancy and insulin resistance, BMI in women with PCOS. These effects are warranted to be further validated, due to insufficient statistical power.