Premature ovarian failure has manifested a trend of younger, and stem cell therapy has been progressively implemented in clinical practice in recent years. Nevertheless, given the extensive range of sources and variegated existence of stem cells in diverse tissues, certain disparities prevail in their biological characteristics and functions. In this paper, the therapeutic efficacies of dissimilar sources of mesenchymal stem cells on animal models of premature ovarian failure were contrasted, with the aim of providing a basis for the clinical application of stem cells.

The animal model experiments of mesenchymal stem cell therapy for premature ovarian failure were retrieved from PubMed, The Cochrane Library, and EMbase, as well as Chinese databases such as CNKI, WanFang, VIP, and China Biomedical Literature Service. The search period extended from the inception to December 31, 2023. Two researchers independently screened the literature, extracted and analyzed the data. The quality of the included studies was evaluated by means of the SYRCLE animal experiment bias risk assessment table. Main outcome measures: Follicle stimulating hormone, estradiol, luteinizing hormone, the quantity of follicles at all levels. Secondary outcome measure: Pregnancy rate. Network meta-analysis, mapping, and tabulation were executed using Stata 17.0 software after assessing the risk of bias in the included studies.

Totally 24 animal experiment studies were incorporated, and the overall quality of the literature was mediocre, encompassing 7 distinct sources of mesenchymal stem cells. They were umbilical cord-derived mesenchymal stem cells, menstrual blood-derived mesenchymal stem cells, placenta-derived mesenchymal stem cells, human cord blood-derived mesenchymal stem cells, bone marrow-derived mesenchymal stem cells, adipose-derived mesenchymal stem cells, and amnio-derived mesenchymal stem cells. The network meta-analysis demonstrated that (1) in contrast to the blank group, mesenchymal stem cells from various sources were effective in enhancing the pregnancy rate and estradiol, reducing follicle-stimulating hormone and luteinizing hormone, augmenting the number of follicles at all levels, and diminishing the number of atretic follicles. (2) According to the area map under the cumulative sequencing curve, the three stem cells with the most prominent efficacy in improving estradiol levels were umbilical cord-derived mesenchymal stem cells (72.7%) > adipose-derived mesenchymal stem cells (72.6%) > menstrual blood-derived mesenchymal stem cells (71.7%). (3) The three kinds of stem cells with the highest efficacy in reducing follicle-stimulating hormone levels were the adipose-derived mesenchymal stem cells (96.3%) > human cord blood-derived mesenchymal stem cells (65.4%) > umbilical cord-derived mesenchymal stem cells (63.9%). (4) The three kinds of stem cells with the highest efficacy in reducing luteinizing hormone levels were adipose-derived mesenchymal stem cells (100.0%) > umbilical cord-derived mesenchymal stem cells (51.6%) > human cord blood-derived mesenchymal stem cells (46.8%). (5) The top three kinds of stem cells for increasing the number of primordial follicles were human cord blood-derived mesenchymal stem cells (76.3%) > umbilical cord-derived mesenchymal stem cells (75.5%) > menstrual blood-derived mesenchymal stem cells (57.5%). (6) The top three kinds of stem cells for increasing the number of primary follicles were umbilical cord-derived mesenchymal stem cells (75.3%) > adipose-derived mesenchymal stem cells (53.0%) > the placenta-derived mesenchymal stem cells (51.7%). (7) The top three kinds of stem cells for increasing the number of secondary follicles were adipose-derived mesenchymal stem cells (76.1%) > menstrual blood-derived mesenchymal stem cells (66.8%) > umbilical cord-derived mesenchymal stem cells (66.5%). (8) The top three kinds of stem cells in reducing the number of atretic follicles were adipose-derived mesenchymal stem cells (99.9%) > bone marrow-derived mesenchymal stem cells (68.1%) > umbilical cord-derived mesenchymal stem cells (53.4%).

(1) For animal models of premature ovarian failure, the results of the network meta-analysis disclosed that various stem cell transplantation treatments were preponderant over the blank or placebo group to varying extents, and the efficacies were comparable. (2) The results indicated that umbilical cord-derived mesenchymal stem cells were the most frequently utilized and adipose-derived mesenchymal stem cells were the most potent. More high-quality experimental study data are requisite in the future for further validation.