Efficacy of autologous stem cell at different doses combined with core decompression in the treatment of osteonecrosis of the femoral head: a systematic review and network meta-analysis

Objective: To compare autologous stem cell (SC) at different doses combined with core decompression (CD) versus CD alone for osteonecrosis of the femoral head (ONFH), using conversion to total hip arthroplasty (THA) as the primary outcome. Methods: PubMed, Embase, Cochrane Library, Web of Science, and CNKI were systematically searched (from inception to July 2025) to identify studies comparing different SC dosages (low dose: <1×10^7; medium dose: 1×10^7-10×10^8; high dose: >1×10^8). Conventional meta-analyses were performed using Review Manager 5.3, while network meta-analyses (NMA) were conducted with Stata 16.0. Treatment efficacy was ranked using SUCRA curves. Sensitivity analyses and funnel plots were applied to assess the robustness of the findings and potential publication bias. Results: Eighteen studies involving 1,192 patients were included. Compared with CD alone, evidence from conventional meta-analysis and network meta-analysis suggested that high-dose autologous stem cell therapy (>1×10^8 cells) combined with CD was associated with a lower risk of hip failure (conversion to THA) (OR = 0.24, 95% CI: 0.12 to 0.44). The high-dose group was also associated with a lower rate of femoral head collapse (OR = 0.24, 95% CI: 0.08 to 0.74) and lower VAS score (SMD = -1.93, 95% CI: -3.64 to -0.23). However, no statistically significant advantage of the high-dose group over the low- or medium-dose groups was observed, and no clear differences in incidence of adverse events (AEs) were detected across dose categories. Conclusions: Preliminary evidence suggests that, compared with CD alone, high-dose autologous stem cell therapy (>1×10^8 cells) combined with CD is associated with a lower risk of hip failure and a lower femoral head collapse rate, with additional improvements in pain in some comparisons. However, the certainty of evidence is limited by heterogeneity in study design, follow-up, and cell dose reporting. Future studies should emphasize standardized cell processing and intervention dosing to validate the dose-response relationship and establish the optimal clinical dosage. Systematic Review Registration: https://www.crd.york.ac.uk/prospero/display_record, identfier CRD420251154025.