Diminished ovarian reserve (DOR) is one of the leading problems affecting women’s reproductive health. It is a clinical syndrome characterized by reduced anti-Müllerian hormone (AMH) levels, elevated follicle-stimulating hormone (FSH) levels, and decreased antral follicle count (AFC). DOR is also highly associated with low fertility and even infertility due to the concomitant decrease in oocyte quality with quantity (1, 2). Therefore, assisted reproductive technology (ART) is often considered for infertile women with DOR. Statistics derived from 181,536 ART cycles showed that the incidence of DOR was approximately 19% to 26% in the US ART population (3). Women with DOR are closely correlated with a reduced probability of pregnancy in ART as well (4). Meanwhile, the negative consequences of DOR include a higher incidence of preeclampsia (5), elevated odds of recurrent pregnancy loss (6), and a higher risk of miscarriage (7). An increase in the prevalence of infertility caused by DOR has tremendously influenced quality of life in women. Although various adjuvant reagents and stimulation protocols have been applied to promote outcomes in infertile women with DOR undergoing IVF, effective treatment remains a clinical challenge.

Growth hormone (GH), secreted by the pituitary gland, is involved in metabolism, cell growth, and development (8). Growing evidence in the literature demonstrated the expression of GH in female oocytes, ovarian granulosa, placenta, and uterus, which reveals that GH may play a distinct role in women’s reproduction (9, 10). Clinically, a decrease in GH levels tends to correlate with poor ovarian response, low oocyte quality and cleavage rate in IVF (11). The beneficial role of GH supplementation during ovarian stimulation may facilitate the implantation process by improving endometrial receptivity and promoting the maturation process of luteinization through regulating the number of receptors in granulosa cells of patients with DOR, such as bone morphogenetic hormone receptor, FSH receptor, and LH receptor (11, 12). Recently, the number of randomized controlled trials (RCTs) of GH for infertile patients with DOR receiving IVF has been increasing. However, several clinical studies have yielded controversial results. For example, Shi et al. (13) reported that the GH co-treatment with IVF was related to promoting the clinical pregnancy rate but not to reducing the dose and the duration of gonadotropin application, which was contrary to Zhao’s study (14). Furthermore, Kang et al. (15) confirmed that a greater number of oocytes were retrieved in the GH group compared to the control group. Still, the optimal embryos rate between the two groups illustrated no difference. While Zhang et al. (16) demonstrated that though the number of optimal embryos was improved by GH adjuvant therapy, the number of oocytes retrieved was not enhanced. Therefore, this meta-analysis of RCTs was conducted to inform clinical practice. The study’s concerns were as follows: (1) Does GH improve the outcomes in infertile women with DOR undergoing IVF?; (2) Is the GH co-treatment associated with a reduction in gonadotropin required during ovarian stimulation?

We conducted this study according to the preferred reporting items for systematic review and meta-analysis (PRISMA) statement (17) and registered on PROSPERO (registration number: CRD42023421739)

Infertility is a widespread reproductive health problem globally. Approximately 12.7% of reproductive-age females suffered from infertility in the US, and 25% of cases of infertility were diagnosed as ovulatory disorders (2). One of the major causes of infertility is DOR (2). Nevertheless, the etiology of DOR is uncertain due to the various complicated factors involved. Abundant literature has elucidated the association of DOR with some pathogenic factors, including environmental influences (triclosan and arsenic exposure, per- and polyfluoroalkyl substances intake) (26–28), unhealthy habits (cigarette, alcohol, and addictive drug consumption) (29), medical complications (gynecological surgery, radiotherapy, and chemotherapy) (30, 31), maternal exposures (intrauterine nutrition restriction, chronic gestational hypoxia) (32, 33), autoimmune diseases (HIV infection, thyroid hormone imbalance) (34, 35), and so forth. Notably, the deterioration of gamete quality and the progressive loss of oocytes occur naturally with older age. Therefore, a burgeoning number of women with DOR are seeking ART treatment for infertility yearly (36). Despite recent advances in ART, the outcomes of natural cycle IVF for DOR are unsatisfying due to the high cycle cancellation rate. For this reason, diverse stimulation interventions and protocols have been recommended to ameliorate IVF outcomes in infertile women with DOR. These schemes used for ovulation induction include the addition of adjuvant therapy, increased gonadotropin dosages, and different procedures for pituitary suppression during ovarian stimulation (37–39). However, high doses of gonadotrophins are often associated with a range of unexpected side effects (40). Hence, exploring an optimal adjuvant treatment remains an ongoing clinical challenge.

GH treatment in conjunction with ovarian stimulation has been considered an efficient strategy for enhancing the outcomes of IVF in infertile women (41). The molecular mechanisms of GH in the reproductive field are rather complex. Substantial evidence has demonstrated that GH could improve oocyte developmental competence by alleviating apoptosis and elevating mitochondrial membrane potential through activating the PI3K-AKT and the Sirt3-Sod2 signaling pathway in granulosa cells; meanwhile, GH may also promote cell proliferation and participate in the steroidogenic process via PI3K-AKT and PLC-PKC signaling to interact with LHR and FSHR and then regulate the functions of granulosa cells and the follicle development (42). Additionally, Liu et al., using in vitro and in vivo experiments, revealed that GH could significantly ameliorate the decline of oocyte quality and depletion of ovarian reserve associated with advanced age via decreasing the expression of γH2AX and inhibiting Fos and Jun signaling pathways in oocytes (43). In terms of uterine receptivity, GH may mediate endometrial thickness and increase endometrial blood perfusion by upregulating IGF-1 and VEGF together with ITGB3 in the uterus (44, 45).

A previous meta-analysis (46) focusing on GH for patients with poor ovarian responders undergoing IVF suggested that GH is beneficial for women in increasing the number of oocytes retrieved and MII oocytes, along with the clinical pregnancy rate and the number of embryos available to transfer. However, they failed to estimate the association of GH supplementation with cycle cancellation rate, cleavage rate, fertilization rate, E₂ levels on the day of hCG, endometrial thickness on the day of hCG, total days of Gn used, and total dose of Gn applied. Furthermore, the previous study didn’t distinguish the effects of GH for women with different ovarian reserves. Therefore, to provide a more specific reflection on the significance of GH in a unique population, we focused our study on women with DOR.

In this systematic review, we estimated a potential association of GH treatment in infertile women with DOR undergoing IVF to provide evidence for informing clinical practice. Our research extended and replicated former reviews on the effectiveness of GH in fertility, which, however, generally focused on poor responders. According to our meta-analysis, adjuvant GH not only increased the clinical pregnancy rate, the number of oocytes retrieved, the optimal embryos rate, and the number of optimal embryos but also decreased the cycle cancellation rate of infertile women with DOR, which was the most critical outcome for women undergoing IVF. Moreover, GH co-treatment was more beneficial in improving the fertilization rate and embryo implantation rate. In particular, the administration of GH resulted in an obvious increase in the E₂ levels and endometrial thickness on the day of hCG, which may be a novel therapeutic option for women with a thin endometrium. Similarly, a significant decrease in the total days and dose of gonadotropin applied was observed when GH was added during ovarian stimulation. Based on subgroup analyses, we also noted there was no dose-dependent connection between adjuvant GH and the number of oocytes retrieved, whereas an increased dose of GH (≥4.5 IU/d) was superior in improving the number of optimal embryos and E₂ levels on the day of hCG. Furthermore, the combination of the GnRH antagonist protocol with GH significantly promoted the number of oocytes retrieved compared to the GnRH agonist protocol.

To the best of our knowledge, this is the first systematic review and meta-analysis evaluating the impact of GH on the outcomes in infertile women with DOR undergoing IVF. Some limitations ought to be noted in the interpretation of our findings. First, although 10 RCTs were included in this analysis, two studies (20, 24) adopted the median (25th percentile, 75th percentile) for continuous variables and one study (14) just reported the percentages without detailed data on clinical pregnancy rate and cleavage rate. As a result, we were unable to integrate the corresponding results into our analysis. The second limitation is that the ovarian stimulation protocols varied in each study, including GnRH antagonist protocol, minimal stimulation protocol, long GnRH agonist protocol, and short GnRH agonist protocol. Therefore, we were forced to divide these protocols into GnRH antagonist protocol and GnRH agonist protocol for subgroup analysis. For this reason, we could not estimate the correlation of GH plus different ovarian stimulation protocols with the outcomes of IVF. The third limitation is that all studies incorporated poor descriptions of their methodologies. For example, they did not report blinding of the participants, investigators, and assessors throughout the trial, which substantially impaired the strength of the evidence. However, since study populations are diverse in the real world of clinical practice, non-blinded pragmatic studies have been proposed to obtain clinically relevant outcomes (47). The fourth limitation is that publication bias cannot be evaluated by Begg’s and Egger’s tests, as less than ten studies were included in each outcome. Nevertheless, to accommodate this limitation, we conducted a sensitivity analysis for each estimate, and it verified that our results were robust. The fifth limitation is that the age of patients is believed to lead to differences in IVF efficacy, with younger infertile women with DOR possibly achieving greater improvement. Unfortunately, none of the studies reported outcomes based on different ages. Therefore, we can’t perform subgroup analysis to compare the clinical significance of GH co-treatment among different age groups. Lastly, our meta-analysis focused on GH in infertile women with DOR undergoing IVF. Hence, our pooled results may not be appropriate for those not diagnosed with DOR.

In conclusion, GH supplementation has significant clinical efficacy in improving the outcomes in infertile women with DOR undergoing IVF, including increasing the clinical pregnancy rate, number of oocytes retrieved, optimal embryos rate and number of optimal embryos, fertilization rate, embryo implantation rate, E₂ levels, and endometrial thickness on the day of hCG, along with decreasing cycle cancellation rate, total days, and total dose of gonadotropin applied. However, our findings are insufficient to offer clinical recommendations due to the high risk of bias and heterogeneity. Therefore, well-designed studies are needed to certify our results in the future.

The original contributions presented in the study are included in the article. Further inquiries can be directed to the corresponding author.

GYL and LWX designed the study. GYL implemented the methods and conducted the experiments. GYL, XFZ, and SNL analyzed and presented the results. GYL completed an initial draft. All authors contributed to the article and approved the submitted version.