Comparison of efficacy, safety, and economy of recombinant and urinary follicle-stimulating hormone in women with a predicted normal response undergoing assisted reproductive technology.

A total of 3,966 eligible cycles were included in the analysis. Baseline characteristics of the rFSH group and the uFSH group were detailed in Table  1 . Prior to matching, significant differences were observed between the two groups in several parameters including maternal age, duration of infertility, type of infertility, tubal factor, endometriosis, uterine factor, AFC, AMH, basal FSH, and year of treatment ( P  < 0.05). After PSM, each group retained 1,133 cycles and the baseline characteristics between the groups were similar. Table 1 Baseline characteristics in rFSH group and uFSH group before and after propensity score matching rFSH ( n  = 1785) uFSH ( n  = 2181) P  value rFSH ( n  = 1133) uFSH ( n  = 1133) P  value Age (years) 31.6 ± 4.4 33.2 ± 4.3 < 0.001 32.6 ± 4.3 33.2 ± 4.3 0.886 Body mass index (kg/m2) 22.2 ± 3 22.3 ± 3 0.052 22.2 ± 3.1 22.3 ± 3 0.913 Infertility duration (years) 4.0 ± 3.1 4.6 ± 3.5 < 0.001 4.3 ± 3.4 4.6 ± 3.5 0.536 Type of infertility, n (%) < 0.001 0.746  Primary 697 (39.1) 515 (23.6) 334 (29.5) 327 (28.9)  Secondary 1088 (61.0) 1666 (76.4) 799 (70.5) 806 (71.1) Infertility diseases, n (%)  Tubal factor 1209 (67.7) 1587 (72.8) 0.001 804 (71.0) 793 (70.0) 0.612  Male factor 440 (24.7) 567 (26.0) 0.332 280 (24.7) 300 (26.5) 0.336  Ovulatory dysfunction 85 (4.8) 85 (3.9) 0.181 42 (3.7) 44 (3.9) 0.826  Endometriosis 152 (8.5) 148 (6.8) 0.040 84 (7.4) 95 (8.4) 0.392  Uterine factor 240 (13.5) 241 (11.1) 0.022 133 (11.7) 146 (12.9) 0.406 Antral follicle count 10.0 ± 2.7 9.2 ± 2.7 < 0.001 9.5 ± 2.7 9.2 ± 2.7 0.969 AMH (ng/mL) 2.4 ± 1.0 2.2 ± 1.0 < 0.001 2.2 ± 0.9 2.2 ± 1.0 0.674 Basal FSH (IU/L) 5.9 ± 1.6 6.1 ± 1.7 0.003 6.0 ± 1.6 6.1 ± 1.7 0.844 Previous IVF attempts, n (%) 0.664 0.397  0 1338 (75.0) 1661 (76.2) 855 (75.5) 853 (75.3)  1–2 361 (20.2) 423 (19.4) 231 (20.4) 220 (19.4)  ≥3 86 (4.8) 97 (4.5) 47 (4.2) 60 (5.3) Year of treatment, n (%) < 0.001 0.759  2017 196 (11.0) 448 (20.5) 166 (14.7) 178 (15.7)  2018 336 (18.8) 464 (21.3) 227 (20.0) 240 (21.2)  2019 401 (22.5) 403 (18.5) 238 (21.0) 217 (19.2)  2020 420 (23.5) 389 (17.8) 237 (20.9) 230 (20.3)  2021 432 (24.2) 477 (21.9) 265 (23.4) 268 (23.7) Notes : Numeric variables are reported as mean value ± standard deviation and categorical variables are reported as number (percentage) Abbreviations : FSH, follicle stimulating hormone; AMH, anti-Müllerian hormone; IVF, in vitro fertilization Baseline characteristics in rFSH group and uFSH group before and after propensity score matching Notes : Numeric variables are reported as mean value ± standard deviation and categorical variables are reported as number (percentage) Abbreviations : FSH, follicle stimulating hormone; AMH, anti-Müllerian hormone; IVF, in vitro fertilization Table  2 displays the characteristics of COS and embryo culture for two groups following PSM. In the matched cohort, the uFSH group exhibited a higher total Gn dose and a longer stimulation duration compared to the rFSH group (all P  < 0.05). Conversely, the rFSH group exhibited elevated levels of E2 and progesterone, as well as an increased numbers of ≥ 14 mm follicles on trigger day (all P  < 0.05). Additionally, laboratory outcomes, including the number of retrieved oocytes, oocyte retrieval rate, metaphase II (MII) oocytes for intracytoplasmic sperm injection (ICSI), two-pronuclei (2PN) oocytes, cleaved embryos, good-quality embryos on day 3, and transferable embryos, were all significantly higher in the rFSH group (all P  < 0.05). Table 2 Clinical treatment characteristics in rFSH group and uFSH group after propensity score matching rFSH ( n  = 1133) uFSH ( n  = 1133) P  value Ovarian stimulation protocol, n (%) 0.515  EF-GnRH-a 981 (86.6) 969 (85.5)  ML-GnRH-a 9 (0.8) 14 (1.2)  GnRH-ant 143 (12.6) 150 (13.2) Initial dose of Gn (IU) 234.8 ± 27.8 237.2 ± 28.9 0.204 Total Gn dose (IU) 2540.3 ± 633.8 2733.9 ± 752.4 < 0.001 Stimulation duration (days) 10.2 ± 1.7 10.6 ± 2.0 < 0.001 Endometrial thickness on trigger day (mm) 10.8 ± 2.6 10.7 ± 2.6 0.652 E2 level on trigger day (pg/mL) 1775.0 ± 992.0 1691.5 ± 1065.1 0.003 LH level on trigger day (IU/L) 1.2 ± 1.0 1.2 ± 1.0 0.188 P level on trigger day (ng/mL) 0.6 ± 0.4 0.5 ± 0.3 < 0.001 No. of ≥ 14 mm follicles on trigger day 8.0 ± 3.1 7.3 ± 3.0 < 0.001 No. of oocytes retrieved 10.4 ± 4.5 9.2 ± 4.4 < 0.001 Oocyte retrieval rate (%) 134.7 ± 42.9 132.4 ± 46.6 0.036 Fertilization type, n (%) 0.204  Insemination 2 (0.2) 6 (0.5)  IVF 876 (77.3) 868 (76.6)  ICSI 197 (17.4) 215 (19.0)  IVF + ICSI 58 (5.1) 44 (3.9) No. of MII oocytes for ICSI 8.4 ± 4.1 7.4 ± 3.7 0.009 No. of 2PN oocytes 6.6 ± 3.6 5.8 ± 3.4 < 0.001 No. of cleaved embryos 6.3 ± 3.5 5.6 ± 3.4 < 0.001 No. of good-quality embryos on day 3 1.9 ± 1.9 1.7 ± 1.7 0.018 No. of transferable embryos 3.1 ± 2.0 2.9 ± 1.9 0.009 Cycles with freezing-all, n (%) 189 (16.7) 178 (15.7) 0.531 The reason for freezing-all, n (%) 0.465  GnRH-ant protocol 78 (41.3) 86 (48.3)  Prevention of OHSS 49 (25.9) 46 (25.8)  Hydrosalpinx 41 (21.7) 31 (17.4)  Uterine factors 15 (7.9) 13 (7.3)  Premature progesterone elevation 6 (3.2) 2 (1.1) Cycles without transferable embryos, n (%) 65 (5.7) 83 (7.3) 0.126 Notes : Numeric variables are reported as mean value ± standard deviation and categorical variables are reported as number (percentage) Abbreviations : EF-GnRH-a, early-follicular phase long-acting gonadotropin-releasing hormone (GnRH) agonist; ML-GnRH-a, mid-luteal phase short-acting GnRH agonist; GnRH-ant, flexible GnRH antagonist; Gn, gonadotropin; E2, estradiol; LH, luteinizing hormone; P, progesterone; IVF, in vitro fertilization; ICSI, intracytoplasmic sperm injection; MII, metaphase II; 2PN, two pronuclei; OHSS: ovarian hyperstimulation syndrome Clinical treatment characteristics in rFSH group and uFSH group after propensity score matching Notes : Numeric variables are reported as mean value ± standard deviation and categorical variables are reported as number (percentage) Abbreviations : EF-GnRH-a, early-follicular phase long-acting gonadotropin-releasing hormone (GnRH) agonist; ML-GnRH-a, mid-luteal phase short-acting GnRH agonist; GnRH-ant, flexible GnRH antagonist; Gn, gonadotropin; E2, estradiol; LH, luteinizing hormone; P, progesterone; IVF, in vitro fertilization; ICSI, intracytoplasmic sperm injection; MII, metaphase II; 2PN, two pronuclei; OHSS: ovarian hyperstimulation syndrome Table  3 presents a comparison of efficacy, safety, and economy between the groups after PSM. For both fresh and frozen embryo transfers, no significant differences were observed in clinical pregnancy rates, miscarriage rates, and live birth rates between the groups. The CDR (rFSH: 56.1% vs. uFSH: 55.0%; P  = 0.612), and TTLB (days, rFSH: 364.4 ± 157.0 vs. uFSH: 354.5 ± 151.4; P  = 0.939) were also similar between the groups (all P  > 0.05). Figure  2 shows the Kaplan-Meier curve for CDR, with the log-rank test indicating no significant difference ( P  = 0.454). In the Cox proportional hazard model, increased age and BMI, decreased AFC, primary infertility, non-male factor, and previous IVF attempts were associated with lower CDR, while treatment protocol and other confounders did not show significant effects on CDR in the model (Table  4 ). Table 3 The efficacy, safety, and economy indicators in rFSH group and uFSH group after propensity score matching rFSH ( n  = 1133) uFSH ( n  = 1133) P  value Efficacy indicators Fresh embryo transfer, n (%)  No. of embryos transferred 0.707   1 194/879 (22.1) 186/872 (21.3)   2 685/879 (77.9) 686/872 (78.7)  Period of embryos transferred 0.288   Cleavage embryo 785/879 (89.3) 792/872 (90.8)   Blastocyst 94/879 (10.7) 80/872 (9.2)  Transfer at least 1 good-quality embryo 673/879 (76.6) 659/872 (75.6) 0.627  Clinical pregnancy rate 531/879 (60.4) 531/872 (60.9) 0.836  Miscarriage rate 82/531 (15.4) 75/531 (14.1) 0.545  Live birth rate 442/879 (50.3) 451/872 (51.7) 0.548 Frozen embryo transfer, n (%)  No. of embryos transferred 0.894   1 161/437 (36.8) 151/405 (37.3)   2 276/437 (63.2) 254/405 (62.7)  Period of embryos transferred 0.956   Cleavage embryo 249/437 (57.0) 230/405 (56.8)   Blastocyst 188/437 (43.0) 175/405 (43.2)  Transfer at least 1 good-quality embryo 259/437 (59.3) 231/405 (57.0) 0.512  Clinical pregnancy rate 253/437 (57.9) 233/405 (57.5) 0.915  Miscarriage rate 60/253 (23.7) 61/233 (26.2) 0.530  Live birth rate 193/437 (44.2) 172/405 (42.5) 0.620 Cumulative delivery rate, n (%) 635/1133 (56.1) 623/1133 (55.0) 0.612 TTLB (days) 364.4 ± 157.0 354.5 ± 151.4 0.939 Safety indicators Incidence of OHSS, n (%)  Mild 12/1133 (1.1) 12/1133 (1.1) 1.000  Moderate 11/1133 (1.0) 6/1133 (0.5) 0.224  Severe 0/1133 (0.0) 1/1133 (0.1) 1.000 Incidence of moderate-to-severe OHSS, n (%) 11/1133 (1.0) 7/1133 (0.6) 0.344 Cumulative multiple pregnancy rate, n (%) 177/748 (23.7) 179/730 (24.5) 0.700 Economy indicators The cost of COS (RMB) GnRH agonist 797.6 ± 308.8 795.7 ± 321.6 0.961 GnRH antagonist 240.4 ± 1282.2 195.9 ± 523.1 0.776 Total Gn 6246.4 ± 1374.5 3267.9 ± 773.1 < 0.001 Transvaginal ultrasonography 597.2 ± 132.3 617.5 ± 142.4 0.001 Endocrine examination 1065.9 ± 226.5 1081.0 ± 238.3 0.038 Total cost of COS 8947.6 ± 1888.0 5958.0 ± 1057.4 < 0.001 Notes : Numeric variables are reported as mean value ± standard deviation and categorical variables are reported as number (percentage) Abbreviations : TTLB, time from controlled ovarian stimulation to live birth; OHSS: ovarian hyperstimulation syndrome; COS: controlled ovarian stimulation; GnRH, gonadotropin-releasing hormone; Gn: gonadotropin; FSH: follicle-stimulating hormone The efficacy, safety, and economy indicators in rFSH group and uFSH group after propensity score matching Notes : Numeric variables are reported as mean value ± standard deviation and categorical variables are reported as number (percentage) Abbreviations : TTLB, time from controlled ovarian stimulation to live birth; OHSS: ovarian hyperstimulation syndrome; COS: controlled ovarian stimulation; GnRH, gonadotropin-releasing hormone; Gn: gonadotropin; FSH: follicle-stimulating hormone Fig. 2 Kaplan-Meier curves of the cumulative delivery rate in women treated with recombinant follicle-stimulating hormone (rFSH) and urinary follicle-stimulating hormone (uFSH) after propensity score matching Kaplan-Meier curves of the cumulative delivery rate in women treated with recombinant follicle-stimulating hormone (rFSH) and urinary follicle-stimulating hormone (uFSH) after propensity score matching Table 4 Cox proportional hazard models for cumulative delivery rate Independent covariates Covariate strata Parameter estimation Standard error Chi-square Adjusted hazard ratio 95% Confidence intervals P  value Treatment protocol uFSH vs. rFSH -0.02 0.06 0.16 0.98 0.88–1.09 0.692 Age -0.05 0.01 53.24 0.95 0.94–0.96 < 0.001 Body mass index -0.02 0.01 5.21 0.98 0.96-1.00 0.022 Infertility duration 0.00 0.01 0.04 1.00 0.98–1.02 0.836 Antral follicle count 0.04 0.01 11.95 1.04 1.02–1.06 0.001 AMH 0.05 0.03 2.90 1.05 0.99–1.12 0.089 Basal FSH -0.03 0.02 3.17 0.97 0.94-1.00 0.075 Type of infertility Secondary vs. primary 0.16 0.07 5.93 1.18 1.03–1.34 0.015 Tubal factor Yes vs. no 0.12 0.07 3.48 1.13 0.99–1.29 0.062 Male factor Yes vs. no 0.14 0.07 4.30 1.15 1.01–1.31 0.038 Ovulatory dysfunction Yes vs. no -0.11 0.16 0.46 0.90 0.65–1.23 0.497 Endometriosis Yes vs. no -0.09 0.11 0.60 0.92 0.74–1.14 0.439 Uterine factor Yes vs. no 0.01 0.09 0.01 1.01 0.85–1.20 0.922 Previous IVF attempts 1–2 vs. 0 -0.29 0.08 13.64 0.75 0.65–0.87 0.000 ≥ 3 vs. 0 -0.47 0.16 8.72 0.63 0.46–0.85 0.003 Year of treatment 2018 vs. 2017 0.02 0.10 0.03 1.02 0.84–1.23 0.872 2019 vs. 2017 -0.09 0.10 0.77 0.92 0.76–1.11 0.381 2020 vs. 2017 -0.03 0.10 0.08 0.97 0.80–1.18 0.776 2021 vs. 2017 0.01 0.09 0.00 1.01 0.84–1.21 0.946 Abbreviations: FSH, follicle-stimulating hormone; AMH, anti-Müllerian hormone; IVF, in vitro fertilization Cox proportional hazard models for cumulative delivery rate Abbreviations: FSH, follicle-stimulating hormone; AMH, anti-Müllerian hormone; IVF, in vitro fertilization For safety, there were no significant differences in the incidence of moderate-to-severe OHSS (rFSH: 1.0% vs. uFSH: 0.6%; P  = 0.344) between the groups. In terms of the cost of COS, the total cost in the rFSH group was higher than that in the uFSH group (8947.6 ± 1888.0 vs. 5958.0 ± 1057.4; P  < 0.001). In details, the cost of Gn (6246.4 ± 1374.5 vs. 3267.9 ± 773.1; P  < 0.001) in the rFSH group was significantly higher than in the uFSH group, while the costs for transvaginal ultrasonography (597.2 ± 132.3 vs. 617.5 ± 142.4, P  = 0.001) and endocrine examination (1065.9 ± 226.5 vs. 1081.0 ± 238.3, P  = 0.038) were slightly lower than that in the uFSH group.

From January 2017 to December 2021, a retrospective analysis was conducted on women with a predicted normal response undergoing IVF/ICSI treatment at the Reproductive Medicine Center of Jiangxi Maternal and Child Health Hospital affiliated with Nanchang Medical College (Nanchang, China). A total of 3,966 treatment cycles were included; however, 2,266 treatment cycles were actually compared after propensity score matching (PSM). Women with an expected normal ovarian response were typically characterized by a FSH level  1.1 ng/mL. The exclusion criteria were as follows: (i) Mild stimulation cycles ( n  = 820); (ii) Over 40 years old ( n  = 910); (iii) Preimplantation genetic testing cycles ( n  = 185); (iv) Sperm or oocyte donor cycles ( n  = 79) (Fig.  1 ). The study was approved by the Reproductive Medicine Ethics Committee of Jiangxi Maternal and Child Health Hospital (Approval No. 2024-15) and conducted in accordance with the Declaration of Helsinki. Data were sourced from the hospital’s database and are available upon reasonable request. Fig. 1 Flowchart of selection for the study. FSH, follicle stimulating hormone; AFC, antral follicle count; AMH, anti-Müllerian hormone Flowchart of selection for the study. FSH, follicle stimulating hormone; AFC, antral follicle count; AMH, anti-Müllerian hormone Controlled ovarian stimulation (COS) was carried out in an early-follicular phase long-acting gonadotropin-releasing hormone (GnRH) agonist (EF-GnRH-a) protocol, a mid-luteal phase short-acting GnRH agonist (ML-GnRH-a) protocol, or a flexible GnRH antagonist (GnRH-ant) protocol. In the EF-GnRH-a protocol, a 3.75 mg long-acting GnRH agonist (Leuprorelin Acetate Microspheres for Injection, Livzon, China) is injected on day 2 or 3 of the menstrual cycle, with endometrial thickness < 5 mm, FSH < 5 IU/L, LH < 5 IU/L, and estradiol (E2) < 50 pg/mL for 28–35 days after downregulation to check the status of downregulation. In the ML-GnRH-a protocol, a daily injection of 0.1 mg short-acting GnRH agonist (Triptorelin Acetate Injection, Ferring, Switzerland) in the mid-luteal phase (Days 21–23) of the menstrual cycle was used for pituitary downregulation. In the GnRH-ant protocol, daily morning injections of 0.25 mg of GnRH antagonist (Cetrorelix, Merck-Serono, Switzerland) are administered to prevent premature ovulation under specific conditions. These conditions include the presence of at least one follicle measuring greater than 14 mm, E2 serum levels exceeding 300 pg/mL, or LH serum levels exceeding 10 IU/L. The starting dose of Gn was determined based on a comprehensive assessment of the maternal age, body mass index (BMI), ovarian reserve, and previous medical history, and subsequently adjusted according to the number and growth of follicles, serum hormone levels, and endometrial thickness. During the study period, the FSH received by patients included rFSH (Gonal-F ® , Merck Serono, Switzerland) or uFSH (Urofollitropin for Injection, Lizhu Pharmaceutical Trading Co., China). When at least 2 dominant follicles reached a diameter of ≥ 18 mm, 0.25 mg of recombinant human chorionic gonadotropin (rhCG, Azer, Merck Serono, Germany) was administered to trigger. Oocyte retrieval was performed under transvaginal ultrasound guidance 36–38 h after trigger administration. Fertilization was carried out via conventional IVF or ICSI, selected based on semen parameters, with pronuclei formation assessed 16–18 h post-insemination. The fertilized zygotes were cultured in G1-plus medium (Vitrolife, Sweden) until the cleavage stage (Day 3), where embryo quality was evaluated at 67–69 h for cell number, fragmentation, and symmetry [ 16 ]. Depending on developmental potential, embryos were either transferred at this stage or further cultured to the blastocyst stage (Day 5/6) and graded using the Gardner scoring system, with only those scoring ≥ 3BC considered suitable for vitrification [ 17 ]. Embryo transfer procedures were performed under transvaginal ultrasound guidance, primarily utilizing soft Cook Sydney embryo transfer catheters (model: K-JETS-7019, Cook Medical, USA). If insertion difficulties arose, operators switched to the Family Transfert Embryonnaire (model: 1321600, Prodimed S.A.S, France). The number of embryos transferred was determined collaboratively by the clinician and embryologist after discussion with the patient. While single embryo transfer was recommended for younger patients with favorable prognoses, transfer of two embryos was permitted if explicitly requested by the patient following comprehensive counseling. For patients undergoing fresh embryo transfer, endometrial transformation was initiated on the day of oocyte retrieval using intramuscular progesterone (60 mg/d; Xianju Pharma, China). Luteal phase support was administered concomitantly via vaginal progesterone gel (90 mg/d; Crinone, Merck Serono, Switzerland) and oral dydrogesterone (20 mg/d; Duphaston, Abbott Biologicals, USA). This combined regimen was continued until 10 weeks of gestation upon confirmed pregnancy. The primary evaluation indicators for efficacy, safety, and economy were the CDR per initiated cycle, the incidence of moderate-to-severe OHSS, and the cost of COS, respectively. The CDR per initiated cycle was defined as the number of deliveries with at least one live birth resulting from one initiated ART cycle, including all cycles in which fresh and/or frozen embryos are transferred, until one delivery with a live birth occurs or until all embryos are used, whichever occurs first. The delivery of a singleton, twin, or other multiples is registered as one delivery [ 14 ]. A cycle is considered to have no deliveries if there are remaining embryos and no deliveries occur within two years. A live birth is characterized by the delivery of a viable infant after a gestational period of at least 24 weeks. Time to live birth (TTLB) is defined as the period from the start date of ovulation induction to the date of delivery. OHSS was defined using the Golan criteria [ 18 ]. Mild OHSS involves symptoms of abdominal distension that resolve spontaneously. Moderate OHSS is diagnosed by the ultrasonographic evidence of ascites, accompanied by nausea, vomiting, and diarrhea. Severe OHSS is identified by clinical signs of ascites and/or hydrothorax, possibly along with breathing difficulties, hemoconcentration, coagulation abnormalities, and reduced renal function. Clinical pregnancy is confirmed by the ultrasonographic identification of one or more gestational sacs within the uterine cavity 30 days post-embryo transfer. Termination of pregnancy before the 24-week gestation period is classified as a miscarriage. For laboratory outcomes, the oocyte retrieval rate was calculated as the number of retrieved oocytes divided by the number of ≥ 14 mm follicles on trigger day. Good-quality embryos on Day 3 should consist of 7–10 blastomeres with a uniform size, and the fragment proportion should be less than 20%. Good-quality blastocysts should be graded 4BB or higher according to the Gardner grading system [ 17 ]. Transferable embryos are considered to be cleavage embryos with 6 or more blastomeres and less than 40% fragmentation, or blastocysts that are graded 3BB or higher. The PSM method was employed to control for potential disparities between the rFSH and uFSH groups. A propensity score was generated through multivariate logistic regression, incorporating variables such as maternal age, BMI, duration of infertility, AFC, anti-Müllerian hormone (AMH), basal FSH, type of infertility, infertility diseases, previous IVF attempts, and the year of treatment. Matching was executed using the nearest neighbor method without replacement at a 1:1 ratio. Additionally, the Cox proportional hazards model was employed to evaluate the relative impact of treatment protocol and associated confounding factors on the CDR, calculating the adjusted hazard ratio (AHR) and the 95% confidence interval (CI). Continuous variables were presented as mean ± standard deviation and assessed for normality using the Shapiro-Wilk and Kolmogorov-Smirnov tests. Since all continuous variables deviated from a normal distribution, comparisons were conducted using the Mann-Whitney U test or the Kruskal-Wallis test. Categorical variables were presented as counts and percentages and compared using the Pearson chi-square test; when expected frequencies were below 5, Fisher’s exact test was employed. Kaplan-Meier curves were utilized to analyze the cumulative delivery events for the rFSH and uFSH groups, with comparisons made using the log-rank test. Statistical analysis was performed in a two-sided setting, with p  < 0.05 considered statistically significant. All statistical analyses were conducted using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA).

In conclusion, for women with predicted normal response undergoing IVF/ICSI, the application of recombinant and urinary FSH leads to a similar CDR and incidence of moderate-to-severe OHSS. However, the application of rFSH leads to higher costs of COS than uFSH. Considering the comparable efficacy and safety of the two FSH formulations, it is recommended that the choice between rFSH and uFSH should be primarily influenced by non-clinical factors, such as cost considerations, previous clinical experiences, and patient preferences.

This single-center retrospective study comprehensively evaluated the efficacy, safety, and economy of rFSH and uFSH in women with a predicted normal response undergoing ART. The findings revealed that rFSH generated a higher number of retrieved oocytes and transferable embryos compared to uFSH. However, this did not result in a higher CDR. In terms of safety and economy, both rFSH and uFSH exhibited similar incidences of moderate-to-severe OHSS; however, the cost of COS was significantly higher for rFSH than for uFSH. Current research comparing rFSH and uFSH is mostly limited to over ten years ago [ 19 ], with only a few related studies emerging in the past decade [ 8 – 12 , 20 – 22 ]. Given the rapid development of ART in recent years, previous studies may not accurately reflect the current capabilities and practices in ART. Furthermore, recent studies have reported conflicting outcomes, necessitating a contemporary evaluation of the efficacy of these two types of FSH in COS. Currently, only one study employs CDR as its primary outcome [ 8 ]. This research indicates that while rFSH and uFSH generated comparable pregnancy and live birth rates per transfer cycle, rFSH led to a greater number of transferable embryos and a higher CDR, contradicting our findings. Possible explanations for this discrepancy could involve variations in study populations, sample sizes, and improvements in IVF practice over time, such as better embryo culture systems, laboratory techniques, freezing protocols, and individualized stimulation strategies. Moreover, despite significant differences observed in the number of oocytes retrieved and transferable embryos, the numerical differences were only 1.2 and 0.2, respectively, which may not be sufficient to influence the CDR. In our study, the CDR for the rFSH and uFSH groups were 56.1% and 55.0%, respectively, which were conservatively estimated [ 23 ]. The actual CDR might be slightly higher, as we classified censored data (with remaining embryos but no live birth) as unsuccessful. Nonetheless, there was no significant difference in CDR between the groups after removing the censored data (rFSH: [635/1021] 62.2% vs. uFSH: [623/1043] 59.7%; P  = 0.252). Our study revealed a higher proportion of tubal infertility (approximately 70%) compared to international cohorts [ 24 , 25 ], consistent with China’s epidemiological profile [ 26 ]. The sum of infertility diagnosis percentages exceeding 100% occurs because patients with multiple diagnoses were counted across all applicable categories. This methodology allows independent statistical comparison of each infertility factor between groups and aligns with common methodological approaches in existing literature [ 27 ]. In our study, uFSH typically required higher total doses of Gn and a longer stimulation duration. This finding is consistent with several previous studies [ 8 , 11 , 28 ]. Gn preparations vary in their biological properties, metabolic clearances, and half-lives [ 6 ]. rFSH is often used as a standard control group in efficacy studies, and it is generally reported that rFSH requires lower doses than uFSH to achieve effective ovarian stimulation, reflecting its enhanced biological activity [ 28 ]. Consistent with previous reports, the rFSH group exhibited elevated progesterone levels on trigger day, likely attributable to increased follicular development [ 9 ]. Consequently, 6 cycles in the rFSH group adopted a freeze-all policy due to progesterone elevation with a cut-off of 2.0 ng/mL, compared to only 2 cycles in the uFSH group. Our study also supported the advantage of rFSH in achieving higher numbers of retrieved oocytes, 2PN oocytes, and transferable embryos. Several studies have assessed embryo and oocyte quality with differing results [ 8 , 29 – 32 ]. For instance, Yang et al. reported that rFSH led to more transferable embryos compared to uFSH [ 8 ]. Cheon et al. observed a significantly higher count of good-quality oocytes with rFSH [ 30 ]. Conversely, Hedon et al. found no significant difference in oocyte and embryo quality between the two types of FSH [ 31 ]. Selman et al. reported that uFSH generated a significantly higher Grade 1 embryo score compared to rFSH [ 32 ]. In terms of safety, our research found comparable incidences of OHSS between the two groups. The latest meta-analysis also indicates no significant difference in the incidence of OHSS between recombinant and urinary Gn (32 trials; N  = 7740; OR = 1.18; 95% CI, 0.86–1.61) [ 33 ]. Despite ongoing concerns regarding long-term safety risks associated with urinary gonadotropins, such as potential contamination with infectious pathogens, for over fifty years, millions of women have safely used urinary gonadotropins without any reported cases of infectious diseases caused by these treatments [ 13 ]. When choosing between rFSH and uFSH, economy factors play a crucial role. Our research indicates that rFSH is more expensive, largely because uFSH costs less per unit. A multicenter retrospective cohort study in China, involving 102,061 IVF cycles, found that the average total cost of Gn was lower in the uFSH group compared to the rFSH group, with estimated costs of 6,734 RMB for rFSH and 3,880 RMB for uFSH [ 34 ]. Patki et al. emphasized that cost should be a primary factor in the selection of specific Gn, as urinary Gn provide a more economical choice compared to their recombinant preparations [ 33 ]. A randomized study confirmed that both types of FSH were equally effective in ovarian stimulation, while uFSH is more cost-effective due to lower production costs [ 35 ]. Another study produced similar results, demonstrating the cost-effectiveness of urinary preparations, given the cost per international unit [ 36 ]. This study provides critical insights into the comparative applications of rFSH and uFSH in ART. Our findings indicate that despite the advantages conferred by advanced recombinant technologies (ensuring high purity and consistent bioactivity), these benefits of rFSH do not translate into superior clinical outcomes in predicted normal response women in terms of CDR when compared with uFSH. These findings may assist clinicians and policymakers in developing evidence-based guidelines to optimize clinical outcomes and cost-efficiency in ART treatments. It supports a growing body of evidence suggesting that the choice between rFSH and uFSH may hinge more on non-clinical factors such as cost considerations, availability, and patient or provider preferences based on past experiences or institutional protocols. The primary strength of this study lies in the comprehensive assessment of the efficacy, safety, and economy of recombinant and urinary FSH in ART. We calculated both the CDR, the ultimate indicator of ART efficacy, and the costs associated with COS, which required considerable human and temporal resources to collect. Moreover, we used both the Cox proportional hazards model and PSM to ensure higher reliability of the obtained outcome. Particularly for PSM, this post hoc randomization technique is especially crucial in studies that feature multiple parallel evaluation criteria. Additionally, the large sample size enhances the statistical power of the study, allowing for more definitive conclusions about the effects observed. It is imperative to acknowledge certain limitations inherent in the current study. Firstly, as a retrospective cohort study, it may introduce potential for inherent bias and residual confounding. Secondly, the study population consisted of women with a predicted normal response, which, although enhancing the robustness of the outcomes, concurrently restricts their generalizability. Thirdly, given the difficulty in collecting data on treatment costs, our study was limited to the costs associated with COS, rather than including the costs of the entire fresh cycle and subsequent frozen cycles. Furthermore, this cohort predominantly underwent two cleavage-stage embryo transfers, which contrasts with the current mainstream practice of extended culture and single blastocyst transfer. While the number and type of embryos transferred may not significantly impact CDR [ 37 , 38 ], they critically influence OHSS and multiple pregnancy risk, limiting the generalizability of our safety comparisons. Lastly, the findings, derived from a single-center cohort, necessitate validation through further multicenter studies to ascertain their applicability to broader populations.

The administration of exogenous gonadotropins (Gn) to facilitate the development of multiple follicles represents a pivotal component in the assisted reproductive technology (ART) [ 1 ]. At present, the selection among Gn preparations frequently alternates between recombinant follicle-stimulating hormone (rFSH), highly purified urinary follicle-stimulating hormone (uFSH) and human menopausal gonadotropin (HMG) [ 2 ]. Due to variations in their sources, production, and purification processes, urinary and recombinant FSH products display unique characteristics concerning purity, degradation, and glycosylation profiles [ 3 ]. rFSH, synthesized via advanced recombinant DNA technology, offers benefits such as high purity, consistent potency, and the absence of urinary proteins, potentially reducing the risk of allergic reactions [ 4 , 5 ]. Conversely, uFSH, derived from the urine of postmenopausal women, might contain < 0.1 IU luteinizing hormone (LH) and < 5% other protein impurities, leading to inconsistencies between batches [ 6 ]. Despite its biological origin and the inherent batch-to-batch variability, uFSH proves effective in stimulating follicular development and continues to be a viable option for woman [ 7 ]. Consequently, the variations in clinical efficacy, safety, and economy of these Gn preparations are a central concern for both clinicians and patients. Retrospective studies comparing the clinical efficacy of recombinant and urinary FSH have produced inconsistent findings [ 8 – 12 ]. Several studies have found no significant difference in the outcomes of ART treatment between rFSH and uFSH [ 11 , 12 ], while others have identified disparities in the cumulative delivery rate (CDR) per initiated cycle [ 8 ], clinical pregnancy rage [ 9 ], and rate of no transferable embryos [ 10 ]. The most recent meta-analysis of randomized controlled trials indicates that there is little or no difference in clinical pregnancy rate (21 studies; N  = 4165; RR = 1.03; 95% CI, 0.94–1.13) or live birth rate (12 studies; N  = 2458; RR = 1.03; 95% CI, 0.90–1.18) between women treated with rFSH and uFSH [ 2 ]. The European Society of Human Reproduction and Embryology (ESHRE) guidelines equally recommend the use of rFSH and uFSH in gonadotropin-releasing hormone (GnRH) agonist protocol [ 7 ]. Additionally, a global network survey revealed that the majority of respondents noted no significant difference in safety or efficacy between the two FSH preparations [ 13 ]. The efficacy of most studies is usually measured by indicators such as the number of retrieved oocytes, fertilization rate, implantation rate, and pregnancy outcome. However, with the increasing use of embryo freezing and thawing, it is essential to consider the CDR per initiated cycles as the ultimate measure of success [ 14 , 15 ]. This is particularly critical when comparing ovulation-stimulating drugs that influence the development of multiple follicles and the number of transferable embryos. Currently, there is only one study that compares the CDR between rFSH and uFSH, indicating that rFSH achieves a higher CDR [ 8 ]. Given the existing controversies of exogenous Gn selection, this study aims to provide a comprehensive comparison of rFSH and uFSH in terms of efficacy, safety, and economy. By conducting a retrospective cohort study including 3,966 in vitro fertilization or intracytoplasmic sperm injection (IVF/ICSI) cycles over a five-year period, we seek to contribute valuable insights into the ongoing debate and aid in the formulation of evidence-based clinical guidelines.