This study included 325 women (325 cycles) who underwent GnRH antagonist ovarian stimulation and 374 patients (733 cycles) who underwent multiple minimal ovarian stimulation (MOS) including natural cycles. Baseline demographic and clinical characteristics were similar between GnRH-antagonist and MOS groups, though basal FSH in MOS group is higher than in GnRH-antagonist (p < 0.001), as shown in Table 1. GnRH-antagonist cycles were characterized by significant longer duration of gonadotropin(Gn) stimulation days, more total dose of Gn, higher peak E2, higher P level, lower LH level and thicker endometrium at trigger day, compared to MOS cycle (Table 2). GnRH-antagonist resulted in more oocytes retrieved, more fertilized oocytes and more viable embryos than both first MOS and the cumulative ones of multiple MOS. (p < 0.001) (Table 2).
Table 1
Baseline demographic and clinical characteristics according to different protocols
| GnRH-antagonist | Minimal ovarian stimulation | P |
Maternal age (y) | 38.46 ± 4.64 | 38.83 ± 4.75 | 0.328 |
BMI | 23.57 ± 2.9 | 23.78 ± 3.1 | 0.056 |
Primary infertility | 143 | 154 | 0.451 |
Infertility years | 5 (2,8) | | 4 (2,7) | 0.134 |
Primary cause of infertility | | | | |
Male | 117 | 129 | 0.677 |
Tubal | 209 | 231 | 0.487 |
Poor ovary response | 6 | 25 | 0.002 |
Endometriosis | 14 | 21 | 0.429 |
Anovulatory | 10 | 4 | 0.105 |
Unexplained | 12 | 4 | 0.039 |
Other causes | 8 | 15 | 0.252 |
Basal E2 level (pmol/L) | 145(95.5, 211.5) (N = 315) | 134(88.59 211)(N = 355) | 0.220 |
Basal FSH level (mIU/ml) | 9(7.2, 15.1) (N = 315) | 11.7(8.6, 17.725) (N = 354) | < 0.001 |
Table 2
Cycle characteristics according to different protocols
| First GnRH-antagonist (325 cycles) | Minimal ovarian stimulation | |
| First (374 cycles) | Multiple (733 cycles) |
Duration of Gn stimulation(days) | 9 (8, 10) | 6 (4, 8) | / | < 0.001 | / |
Total dose of Gn(IU) | 2400(1800,2925) | 900(600,1256.25) | / | < 0.001 | / |
Peak E2 level at trigger day (pmol/L) | 7585(4213.5, 11666) (N = 320) | 2707(1630, 4815) (N = 365) | / | < 0.001 | / |
P level at trigger day (nmol/L) | 2.64(1.623, 3.683) (N = 68) | 1.36(1.032, 3.105) (N = 96) | / | 0.005 | / |
LH level at trigger day(U/L) | 3.14(2.205, 5.21) (N = 67) | 7.96(5.318, 13.858) (N = 96) | / | < 0.001 | / |
Endometrial thickness at trigger day(mm) | 9(8.5, 10.4) | 6(5, 8.2) | / | < 0.001 | / |
ICSI/IVF | 100/225 | 104/270 | / | 0.390 | / |
No. of oocytes retrieved | 7(4, 10) | 2(1, 4) | 4 (2, 7) | < 0.001 | < 0.001 |
No. of fertilized oocytes | 5(3, 7) | 2(1, 3) | 3 (1, 5) | < 0.001 | < 0.001 |
No. of viable embryos | 2(1, 4) | 1(0, 2) | 2 (1,3) | < 0.001 | < 0.001 |
a : First GnRH-antagonist vs First minimal ovarian stimulation |
b : First GnRH-antagonist vs Multiple minimal ovarian stimulation |
As for clinical result (Table 3), the CLBR for both groups of patients was generally low. For the first IVF cycle, GnRH-antagonist demonstrated higher CLBR per aspiration than first MOS from both univariate analysis (12.92% versus 4.54%, Crude OR 3.117; 95%CI 1.737, 5.592, p < 0.001) and multivariate analysis after adjusting for female age, BMI, basal FSH, basal E2, infertility years and primary infertility (vs secondary infertility) (Adjusted OR 2.606; 95%CI 1.386, 4.899, p = 0.003). Female age and basal FSH and infertility years were also independent factors negatively associated with likelihood of CLBR per aspiration (Supplemental Fig. 1). In the MOS group, clustering of multiple treatment cycles per woman has to be considered. So we also measured the CLBR per person in this group of patients. GnRH-antagonist displayed higher CLBR per aspiration than the CLBR per person of multiple MOS from univariate analysis (12.92% versus 7.22%, Crude OR 1.907; 95%CI 1.147, 3.171, p < 0.001), while the type of treatment (GnRH-antagonist vs MOS) was not associated with CLBR in multivariate logistic regression after adjusting the same factors as above (Adjusted OR 1.702; 95%CI 0.971, 2.982, p = 0.063). Female age and basal FSH were only independent factors negatively associated with likelihood of CLBR per person (Supplemental Fig. 2).
Table 3
, Clinical outcomes according to different protocols
| First GnRH-antagonist (325 cycles, 325 persons) (per aspiration) | First minimal ovarian stimulation (374 cycles, 374 persons) (per aspiration) | Multiple minimal ovarian stimulation (733cycles, 374 persons) (per ITT) | Pa | Pb | Adjusted OR (95%CI)a Pa | Adjusted OR (95%CI)b Pb |
CLBR | 42 (12.92%) | 17 (4.54%) | 27 (7.22%) | < 0.001 | 0.012 | 2.606 (1.386, 4.899) 0.003 | 1.702 (0.971, 2.982) 0.063 |
Cost | 20838 (17953, 23422) | 12254 (9612.5, 14875.5) | 21261.5 (15892.5, 35140.25) | < 0.001 | 0.13 | / | / |
Time to First Live Birth | 9(8, 10.75) | / | 11 (9, 14) | / | 0.014 | / | / |
a : First GnRH-antagonist vs First minimal ovarian stimulation |
b : First GnRH-antagonist vs Multiple minimal ovarian stimulation |
Adjusted OR: Adjusting for age, BMI, basal FSH, basal E2, infertility years, primary infertility (vs secondary infertility). |
In the economy-effectiveness analysis, when considering only the first cycle of ovarian stimulation, the cost of GnRH-antagonist was higher than MOS (20838 (17953, 23422) ¥ versus 12254 (9612.5, 14875.5) ¥, p < 0.001). But the cumulative financial expenditure was statistically similar between one time GnRH-antagonist and multiple MOS (20838 (17953, 23422) ¥ versus 21261.5 (15892.5, 35140.25) ¥, p = 0.13). When it comes to the time to first live birth, GnRH-antagonist showed obviously shorter time than repeated modified natural cycles (9 (8, 10.75) months versus 11(9, 14) months, p = 0.014).