Administration of growth hormone improves endometrial function in women undergoing in vitro fertilization: a systematic review and meta-analysis.

BACKGROUND The positive effects of growth hormone (GH) on IVF are often attributed to improvements in oocyte and embryo quality. While emerging evidence emphasizes GH-induced improvements in the endometrium, these results are controversial. OBJECTIVE AND RATIONALE This meta-analysis aimed to evaluate whether GH administration improved endometrial function and reproductive outcomes during IVF cycles and to thus guide clinical practice. SEARCH METHODS A literature search in the Cochrane Central Register of Controlled Trials, PubMed and Embase was performed through to 30 November 2021, without language restrictions. Randomized controlled trials (RCTs) evaluating the effects of GH on IVF outcomes were included. Risk of bias and quality of evidence (QoE) were assessed according to the Cochrane Collaboration's tool and the Grading of Recommendations Assessment, Development and Evaluation system. Odds ratios (ORs) and mean differences (MDs) with 95% confidence intervals (CIs) were assessed by random-effects models. OUTCOMES A total of 25 trials with 2424 women were included. Seventeen RCTs with poor responders (n = 1723) showed that GH administration significantly increased endometrial thickness (EMT) (MD = 0.38, 95% CI: 0.18-0.59; moderate QoE), which contributed to an improved live birth rate (OR = 1.67, 95% CI: 1.13-2.49; very low QoE) and clinical pregnancy rate (CPR) (OR = 1.97, 95% CI: 1.43-2.72; low QoE). Subgroup analyses showed a dose- and time-dependent relationship between GH cotreatment and IVF outcomes; the optimal recommendation for improving CPR was consistent with that for EMT, rather than for oocytes and embryos. Hence, GH might improve fertility via effects on the endometrium. Administration of GH daily from the follicular phase of previous cycle until the hCG trigger with < 5 IU/day led to a thicker endometrium and a greater chance of becoming pregnant, while 5-10 IU/day or administration from the luteal phase of the previous cycle until the hCG trigger resulted in higher oocyte and embryo quality. Poor responders might benefit from cotreatment with the GnRH agonist long protocol more than other stimulation protocols. Pooled data from four trials (n = 354) on women with a thin endometrium indicated that improved endometrial function might be critical for improving reproductive outcomes during GH treatment, as no improvements in embryo quality were found. GH administration not only increased EMT (MD = 1.48, 95% CI: 1.21-1.75; moderate QoE) but also promoted endometrial morphology (OR = 2.67, 95% CI: 1.36-5.23; low QoE) and perfusion (OR = 5.84, 95% CI: 1.30-26.17; low QoE), thereby improving the CPR (OR = 2.71, 95% CI: 1.69-4.34; P < 0.0001; low QoE). There was insufficient evidence to reach a conclusion regarding the effects of GH in normal responders (n = 80). Due to obvious improvements in the CPR, women with a thin endometrium might be the most appropriate population to benefit from GH administration. WIDER IMPLICATIONS Improving endometrial function might be another vital mechanism by which GH improves IVF outcomes. Optimal treatment should be offered to the target population according to their personal conditions and needs. The QoE was moderate to very low, due to limited sample sizes and methodological problems; thus, the results should be interpreted with caution. More rigorous RCTs with large sample sizes are needed to confirm the effects and determine optimal GH protocols.