Comparative Efficacy and Safety of Three Current Clinical Treatments for Girls with Central Precocious Puberty: A Network Meta-Analysis

Objective: The optimal treatment for girls with central precocious puberty (CPP) is unknown. We conducted a network meta-analysis to evaluate the efficacy and safety of existing treatments to provide credible clinical guidelines. Methods: We compared gonadotropin-releasing hormone analogue (GnRHa) therapy, GnRHa plus growth hormone (GH) combination therapy, and no-treatment therapy for girls with CPP by performing an electronic search for studies in PubMed, Embase, Chinese National Knowledge Infrastructure databases, and Wanfang Data from their inception until September 30, 2018. Six outcomes, including bone maturation ratio, final height, final height compared with target height, growth velocity, height gain, and gain in predicted adult height (ΔPAH), were expressed as the mean difference with 95% confidence interval. The surface under the cumulative ranking curve (SUCRA) value illustrated the rank probability of each treatment under different outcomes. Results: Twenty-two studies with 1,268 patients were included. GnRHa plus GH had the best performance on final height, final height compared with target height, growth velocity, height gain, and ΔPAH, with the highest SUCRA values of 0.919, 0.975, 0.909, 0.999, and 0.957, respectively. For bone maturation ratio, GnRHa ranked the highest, with a SUCRA value of 0.663. No severe adverse effects were reported. Conclusion: For girls with CPP, GnRHa plus GH had the highest probability of being the optimal therapy for improving final height, and no severe adverse effects were reported. Abbreviations: BMI = body mass index; CI = confidence interval; CPP = central precocious puberty; GH = growth hormone; GnRHa = gonadotropin-releasing hormone analogue; HPG = hypothalamic-pituitary-gonadal; LH = luteinizing hormone; NMA = network meta-analysis; PAH = predicted adult height; PCOS = polycystic ovary syndrome; RCT = randomized controlled trial; SUCRA = surface under the cumulative ranking curve.