Dual stimulation for low ovarian responders in ART: a systematic review and meta-analysis

Background: Response to ovarian stimulation varies among patients. Low ovarian responders (LORs) represent a particularly challenging subgroup, and numerous strategies have been explored to improve their outcomes. Dual stimulation has emerged as a potentially promising approach to increase the number of retrieved oocytes, mature oocytes, and consequently, the number of available embryos and the likelihood of achieving a live birth in this population. However, current evidence regarding the efficacy of dual stimulation in LORs remains limited, warranting further high-quality studies. Objective: This systematic review and meta-analysis aimed to compare live birth rates between dual ovarian stimulation and single or two conventional stimulation cycles in LORs undergoing ART. Materials and Methods: A systematic review and meta-analysis was performed comparing dual ovarian stimulation (two stimulation phases separated by <7 days between oocyte retrieval of the first ovarian stimulation and the initiation of the second ovarian stimulation) with either a single or two consecutive conventional stimulation cycles in LOR undergoing ART. A literature search was conducted until March 2025 in MEDLINE, Embase, Cochrane CENTRAL, Scopus, Web of Science, and ClinicalTrials.gov to identify eligible randomized controlled trials (RCTs). The primary outcome was the live birth rate. Secondary outcomes included clinical pregnancy, ongoing pregnancy, miscarriage, total and mature oocytes retrieved, number of available embryos, number of euploid embryos, number of good- or top-quality embryos, and euploidy rate. The risk of bias and study trustworthiness were assessed using the RoB2 Tool and TRACT checklist. Meta-analyses used fixed or random effects models based on statistical heterogeneity (I² >50%), results were combined with the intention-to-treat (ITT) principle [supplementary per protocol (PP) analyses were performed] and are reported as relative risk (RR) or weighted-mean-difference (WMD) with 95% confidence intervals (CI). Subgroup analysis was performed based on whether dual ovarian stimulation was compared with one or two consecutive conventional stimulation cycles. Main outcomes: Six RCTs (n=428) were identified. Two studies were excluded from the meta-analysis because of low trustworthiness. No significant difference was observed in live birth (RR: 0.78, 95% CI: 0.48-1.26; I2=22.7%, three studies, n=216), ongoing pregnancy (RR: 1.08, 95% CI: 0.54-2.18; I2 =0%, one study, n=120), clinical pregnancy (RR: 0.90, 95% CI: 0.43-1.88; I2=69.8%, three studies, n=256) and miscarriage (RR: 1.24, 95% CI: 0.50-3.08; I2=21.4%, four studies, n=336). The subgroup analyses suggested a significant difference between the two subgroups (p=0.04) in clinical pregnancy rates: dual ovarian stimulation versus single conventional stimulation (RR: 1.21, 95% CI: 0.68-2.17) and dual ovarian stimulation versus two consecutive conventional cycles (RR: 0.44, 95% CI: 0.20-0.96). One study (n=80) reported significantly fewer total oocytes (WMD: -4.20; 95% CI: -5.22 to -3.18), MII oocytes (WMD: -1.90; 95% CI: -2.67 to -1.13), and available embryos (WMD: -0.60; 95% CI: -0.84 to -0.36) in the dual stimulation group. Conversely, PP analysis, including three studies (n=233), did not identify any significant differences between groups in the number of oocytes, MII oocytes, and available embryos.