Modified clomiphene citrate protocol versus gonadotropin-releasing hormone antagonist protocol in in vitro fertilization: a propensity score-matched retrospective study on cost and efficacy.

BACKGROUND: The rising prevalence of infertility has substantially increased the demand for in vitro fertilization (IVF). While various controlled ovarian stimulation (COS) protocols exist, the gonadotropin-releasing hormone antagonist (GnRH-ant) protocol is valued for its efficacy in preventing premature luteinizing hormone (LH) surges. However, it involves high medication costs and requires frequent injections, which impose a significant burden on patients. Clomiphene citrate (CC), a classic oral ovulation induction agent, offers advantages such as low cost and convenient administration. With the widespread adoption of the "freeze-all" strategy, its potential adverse impact on the endometrium can be mitigated. Recent modifications to the CC protocol, combining it with adequate-dose gonadotropins, aim to balance efficacy and cost. However, well-designed studies directly comparing this modified CC protocol with the conventional GnRH-ant protocol in a general IVF population are lacking, particularly those incorporating comprehensive cost-effectiveness analyses. METHODS: 3,157 patients undergoing IVF or intracytoplasmic sperm injection (ICSI) cycles using either the CC protocol or the GnRH-ant protocol were matched 1:1 using propensity score matching (PSM) based on female age, basal follicle-stimulating hormone (FSH), anti-Müllerian hormone (AMH), antral follicle count (AFC), body mass index (BMI), fertilization method, and sperm source. A total of 1,858 patients (929 per group) were included after matching. Embryological parameters and pregnancy outcomes were compared between the two groups, along with cost-effectiveness and sensitivity analyses. RESULTS: Baseline characteristics were well-balanced after PSM (with almost all standardized mean differences (SMD) < 0.1). The clinical pregnancy rate was comparable between the CC group and the GnRH-ant group (46.26% vs. 41.07%, P = 0.113), and the cumulative clinical pregnancy rate was also similar (51.60% vs. 51.79%, P = 1.000). The live birth rate (35.48% vs. 31.95%, P = 0.268) and cumulative live birth rate (35.48% vs. 38.29%, P = 0.398) were also comparable between the two groups. Compared to the GnRH-ant group, the CC group exhibited a higher Day 3 embryo formation rate (66.67% vs. 62.50%, P = 0.012), despite having lower numbers of follicles on the day of hCG trigger and retrieved oocytes (9.00 vs. 10.00 and 6.00 vs. 8.00, P < 0.0001). No significant differences were observed in the oocyte retrieval rate, oocyte maturation rate, fertilization rate, and blastulation rate between two groups. The cost of COS in the CC group was significantly lower than that in the GnRH-ant group (1,414.70 [741.50-2,915.50] CNY [~198 USD] vs. 7,266.24 [5,747.64-9,011.05] CNY [~1,017 USD], P < 0.0001), representing a saving of approximately 5,851.54 CNY (~819 USD; 80.53%). Cost-effectiveness analysis indicated that the CC protocol was superior across all outcome measures, including incremental cost-effectiveness ratios (ICERs) for oocyte, embryo, and clinical pregnancy. One-way sensitivity analysis (OWSA) indicated that the clinical pregnancy rate between the two groups was the most critical factor influencing the ICER for clinical pregnancy. Probabilistic sensitivity analysis (PSA) also suggested that in 98.50% of simulations, the CC protocol was more cost-saving. CONCLUSIONS: The modified CC protocol was associated with ovarian stimulation effects and clinical outcomes comparable to those of the GnRH-ant protocol. Furthermore, it represented a more cost-effective ovarian stimulation strategy, significantly reducing treatment costs while maintaining a clinical pregnancy rate comparable to that of the GnRH-ant protocol.