Effect of letrozole cotreatment in progestin-primed ovarian stimulation on IVF/ICSI outcomes in POSEIDON group 3 and 4 poor responders: a retrospective cohort study.

As the flowchart presented in Fig.  1 , a total of 557 patients were enrolled in this research with 189 ovarian stimulation cycles for POSEIDON group 3 patients and 368 ovarian stimulation cycles for POSEIDON group 4 patients. The baseline characteristics of the different groups are shown in Table  1 and there were no significant differences in these indicators. Fig. 1 Flowchart of this study. IVF  in vitro fertilization, ICSI intracytoplasmic sperm injection, PPOS progestin-primed ovarian stimulation, FET frozen embryo transfer, LE Letrozole Table 1 Baseline information of included patients in different groups Variables POSEIDON Group 3 P value POSEIDON Group 4 P value PPOS group n = 78 PPOS + LE group n = 111 PPOS group n = 154 PPOS + LE group n = 214 Age of women (year) 31.0 (32.0, 33.0) 31.0 (29.0, 32.5) 0.811 40.0 (38.0, 43.0) 40.0 (38.0, 43.0) 0.547 BMI (kg/m 2 ) 22.03 (20.09, 24.59) 21.83 (19.89, 23.91) 0.339 22.03 (20.45, 24.03) 22.62 (20.59, 24.84) 0.111 Infertility duration (year) 3.0 (2.0, 5.0) 2.0 (1.0, 5.0) 0.164 3.0 (1.0, 5.0) 3.0 (1.0, 5.0) 0.967 Infertility type, n (%) 0.269 0.244 Primary 63/78 (80.77) 82/111 (73.87) 72/154 (46.75) 87/214 (40.65) Secondary 15/78 (19.23) 29/111 (26.13) 82/154 (53.25) 127/214 (59.35) AMH (ng/ml) 0.82 (0.37, 0.99) 0.86 (0.51, 1.31) 0.080 0.67 (0.39, 0.98) 0.65 (0.39, 1.01) 0.885 AFC 3.0 (2.0, 5.0) 3.0 (2.0, 4.0) 0.549 3.0 (2.0, 4.0) 3.0 (2.0, 4.0) 0.464 Basal FSH (mIU/mL) 9.05 (7.15, 12.74) 10.00 (7.58, 14.59) 0.143 10.71 (8.24, 14.00) 10.20 (7.97, 14.40) 0.586 Cause of infertility, n (%) 0.274 0.617 Only POR 21/78 (26.92) 26/111 (23.42) 34/154 (22.08) 41/214 (19.16) POR with female factors 42/78 (53.85) 52/111 (46.85) 80/154 (51.95) 126/214 (58.88) POR with male factors 10/78 (12.82) 16/111 (14.41) 17/154 (11.04) 19/214 (8.88) POR with mixed factors 5/78 (6.41) 17/111 (15.32) 23/154 (14.94) 28/214 (13.08) Fertilization methods # , n (%) 0.768 0.296 IVF 58/75 (77.33) 83/110 (75.45) 114/148 (77.03) 156/214 (72.90) ICSI 17/75 (22.67) 27/110 (24.55) 341/148 (22.98) 58/214 (27.10) Data were presented as mean ± SD, median (25th percent, 75th percent) and n (%) # Only include patients with oocyte retrieval BMI, body mass index; AFC, antral follicle count; AMH, anti-Müllerian hormone; FSH, follicle stimulating hormone; IVF, in vitro  fertilization; ICSI, intracytoplasmic sperm injection; POR: poor ovarian response Flowchart of this study. IVF  in vitro fertilization, ICSI intracytoplasmic sperm injection, PPOS progestin-primed ovarian stimulation, FET frozen embryo transfer, LE Letrozole Baseline information of included patients in different groups 31.0 (32.0, 33.0) 31.0 (29.0, 32.5) 40.0 (38.0, 43.0) 40.0 (38.0, 43.0) 22.03 (20.09, 24.59) 21.83 (19.89, 23.91) 22.03 (20.45, 24.03) 22.62 (20.59, 24.84) 3.0 (2.0, 5.0) 2.0 (1.0, 5.0) 3.0 (1.0, 5.0) 3.0 (1.0, 5.0) 63/78 (80.77) 82/111 (73.87) 72/154 (46.75) 87/214 (40.65) 15/78 (19.23) 29/111 (26.13) 82/154 (53.25) 127/214 (59.35) 0.82 (0.37, 0.99) 0.86 (0.51, 1.31) 0.67 (0.39, 0.98) 0.65 (0.39, 1.01) 3.0 (2.0, 5.0) 3.0 (2.0, 4.0) 3.0 (2.0, 4.0) 3.0 (2.0, 4.0) 9.05 (7.15, 12.74) 10.00 (7.58, 14.59) 10.71 (8.24, 14.00) 10.20 (7.97, 14.40) 21/78 (26.92) 26/111 (23.42) 34/154 (22.08) 41/214 (19.16) 42/78 (53.85) 52/111 (46.85) 80/154 (51.95) 126/214 (58.88) 10/78 (12.82) 16/111 (14.41) 17/154 (11.04) 19/214 (8.88) 5/78 (6.41) 17/111 (15.32) 23/154 (14.94) 28/214 (13.08) 58/75 (77.33) 83/110 (75.45) 114/148 (77.03) 156/214 (72.90) 17/75 (22.67) 27/110 (24.55) 341/148 (22.98) 58/214 (27.10) Data were presented as mean ± SD, median (25th percent, 75th percent) and n (%) # Only include patients with oocyte retrieval BMI, body mass index; AFC, antral follicle count; AMH, anti-Müllerian hormone; FSH, follicle stimulating hormone; IVF, in vitro  fertilization; ICSI, intracytoplasmic sperm injection; POR: poor ovarian response Table 2 presents the detailed cycle characteristics of POR women treated with the PPOS and PPOS + LE protocols, respectively. The total dose of Gn and duration of ovarian stimulation were comparable. On the trigger day, the serum E2 level in the PPOS + LE group was significantly lower than PPOS group [960.50 (620.75, 1244.50) vs. 773.00 (509.00, 1088.00), P  = 0.042; 873.00 (560.50, 1213.75) vs. 747.00 (380.00, 1078.75), P  = 0.011] and the LH level was remarkably higher in the PPOS + LE group [3.32 (1.67, 4.35) vs. 3.71(2.79, 4.79), P  = 0.012; 3.54 (2.30, 5.94) vs. 4.19 (2.96, 6.05), P  = 0.036] both in POSEIDON group 3 and 4. In POSEIDON group 4, PPOS + LE protocol was associated with a significantly higher number of oocytes for retrieval and maturation [3.0 (2.0, 5.0) vs. 2.0 (1.3, 4.0), P  < 0.001; 3.0 (2.0,5.0) vs. 2.0 (1.0, 3.8), P  < 0.001] as well as 2PN embryos [2.0 (1.0, 3.0) vs. 2.0 (1.0, 3.0), P  = 0.026]. The cancellation rate also was lower in PPOS + LE group (18.22% vs. 27.27%, P  = 0.039). In the POSEIDON 3 group patients, the LE cotreatment did not show an advantage in these indicators. In addition, there was no significant difference between the two groups when the oocyte retrieval rate, 2PN embryos rate, and available embryo rate were analyzed both in POSEIDON group 3 and 4 ( P  > 0.05). Table 2 Characteristics of laboratory outcomes in different groups Variables POSEIDON Group 3 P value POSEIDON Group 4 P value PPOS group n = 78 PPOS + LE group n = 111 PPOS group n = 154 PPOS + LE group n = 214 Ovarian stimulation duration (days) 9.23 ± 1.73 9.14 ± 1.41 0.705 9.16 ± 1.86 9.24 ± 1.89 0.678 Total gonadotropin dose (IU) 2718.27 ± 732.28 2906.08 ± 649.56 0.065 2899.24 ± 785.59 2968.98 ± 661.10 0.371 Premature LH surge rate, n (%) 3/78 (3.85) 1/111 (0.90) 0.383 7/154 (4.55) 9/214 (4.21) 0.875 E2 level on the trigger day (ng/ml) 960.50 (620.75, 1244.50) 773.00 (509.00, 1088.00) 0.042 * 873.00 (560.50, 1213.75) 747.00 (380.00, 1078.75) 0.011 * LH level on the trigger day (mIU/mL) 3.32 (1.67, 4.35) 3.71 (2.79, 4.79) 0.012* 3.54 (2.30, 5.94) 4.19 (2.96, 6.05) 0.036 * Endometrial thickness (mm) 7.9 (6.9, 9.6) 7.6 (6.3, 9.4) 0.228 7.2 (6.4, 8.2) 7.7 (6.1,9.0) 0.250 Dominant follicle ≥ 14 mm 3.0 (2.0,5.0) 3.0 (2.0,5.0) 0.217 3.0 (2.0, 4.0) 3.0 (2.0,4.0) 0.089 No. of oocytes retrieved 3.5 (2.0,5.0) 4.0 (3.0, 6.0) 0.127 2.0 (1.3, 4.0) 3.0 (2.0, 5.0)  < 0.001 * No. of MII oocytes 3.0 (2.0, 5.0) 3.0 (2.0,5.0) 0.176 2.0 (1.0, 3.8) 3.0 (2.0, 5.0)  < 0.001 * No. of 2PN embryos 2.0 (1.0, 4.0) 2.0 (1.0, 4.0) 0.236 2.0 (1.0, 3.0) 2.0 (1.0, 3.0) 0.026 * No. of available embryos 1.0 (1.0, 2.0) 2.0 (1.0, 3.0) 0.356 1.0 (0.0, 2.0) 1.0 (1.0, 2.0) 0.211 Mature oocyte rate, n (%) 263/308 (85.39) 411/482 (85.27) 0.963 415/479 (86.63) 760/853 (89.10) 0.182 2PN embryo rate (%), n (%) 187/263 (71.10) 292/411 (71.05) 0.987 297/415 (71.57) 504/760 (66.32) 0.065 2PN cleavage rate (%), n (%) 183/187 (97.86) 287/292 (98.29) 1.000 288/297 (96.97) 480/504 (95.24) 0.467 Available embryo rate (%), n (%) 132/183 (72.13) 203/287 (70.73) 0.744 210/288 (72.92) 318/480 (66.25) 0.054 Cycle cancellation rate (%), n (%) 17/78 (21.80) 14/111 (12.61) 0.093 42/154 (27.27) 39/214 (18.22) 0.039 * Data were presented as mean ± SD, median (25th percent, 75th percent) and n (%) * P  < 0.05 E2, Estradiol; LH, Luteinizing hormone; 2PN, two pronuclei Characteristics of laboratory outcomes in different groups 3/78 (3.85) 1/111 (0.90) 7/154 (4.55) 9/214 (4.21) 960.50 (620.75, 1244.50) 773.00 (509.00, 1088.00) 873.00 (560.50, 1213.75) 747.00 (380.00, 1078.75) 3.32 (1.67, 4.35) 3.71 (2.79, 4.79) 3.54 (2.30, 5.94) 4.19 (2.96, 6.05) 7.9 (6.9, 9.6) 7.6 (6.3, 9.4) 7.2 (6.4, 8.2) 7.7 (6.1,9.0) 3.0 (2.0,5.0) 3.0 (2.0,5.0) 3.0 (2.0, 4.0) 3.0 (2.0,4.0) 3.5 (2.0,5.0) 4.0 (3.0, 6.0) 2.0 (1.3, 4.0) 3.0 (2.0, 5.0) 3.0 (2.0, 5.0) 3.0 (2.0,5.0) 2.0 (1.0, 3.8) 3.0 (2.0, 5.0) 2.0 (1.0, 4.0) 2.0 (1.0, 4.0) 2.0 (1.0, 3.0) 2.0 (1.0, 3.0) 1.0 (1.0, 2.0) 2.0 (1.0, 3.0) 1.0 (0.0, 2.0) 1.0 (1.0, 2.0) 263/308 (85.39) 411/482 (85.27) 415/479 (86.63) 760/853 (89.10) 187/263 (71.10) 292/411 (71.05) 297/415 (71.57) 504/760 (66.32) 183/187 (97.86) 287/292 (98.29) 288/297 (96.97) 480/504 (95.24) 132/183 (72.13) 203/287 (70.73) 210/288 (72.92) 318/480 (66.25) 17/78 (21.80) 14/111 (12.61) 42/154 (27.27) 39/214 (18.22) Data were presented as mean ± SD, median (25th percent, 75th percent) and n (%) * P  < 0.05 E2, Estradiol; LH, Luteinizing hormone; 2PN, two pronuclei Detailed characteristics of FET and subsequent pregnancy outcomes in the different groups are shown in Table  3 . A total of 445 patients included in this study were followed up with frozen embryo transfer (FET). In POSEIDON group 3 and 4, the average of embryos transferred, the stage of embryo transferred, endometrial preparation protocol and endometrial thickness per cycle were similar in both groups ( P  > 0.05). As for pregnancy outcomes, the biochemical pregnancy rate (51.16% vs. 51.22%; 22.44% vs. 23.35%), clinical pregnancy rate (36.05% vs. 43.90%; 11.33% vs. 16.30%), ongoing pregnancy rate (27.91% vs. 35.77%; 10.90% vs. 11.01%), miscarriage rate (22.58% vs. 22.22%; 44.83% vs. 40.54%) and live birth rate (27.91% vs. 34.15%; 10.26% vs. 9.69%) were comparable in PPOS and PPOS + LE group both in POSEIDON group 3 and 4 patients ( P  > 0.05). Considering multiple embryo transfer in POR patients, CCPR and CLBR were also evaluated as main indicators and there was no significant difference in CCPR (39.74% vs. 48.65%, P  = 0.226; 18.83% vs. 17.29%, P  = 0.704) and CLBR (30.77% vs. 37.84%, P  = 0.316; 10.39% vs. 10.28%, P  = 0.973) between the PPOS and PPOS + LE group in POSEIDON group 3 and 4 patients. Table 3 Characteristics of embryo transfer and pregnancy outcomes in different groups Variables POSEIDON Group 3 P value POSEIDON Group 4 P value PPOS group n = 78 PPOS + LE group n = 111 PPOS group n = 154 PPOS + LE group n = 214 Transfer outcomes No. of patients 61 97 112 175 No. of total FET cycles 86 123 156 227 No. of total embryos transferred 103 138 192 270 Average of embryos transferred per cycle 1.2 ± 0.4 1.1 ± 0.3 0.136 1.2 ± 0.4 1.2 ± 0.4 0.327 No. of embryos transferred per cycle 0.135 0.326 1 69/86 (80.23) 108/123 (87.80) 120/156 (76.92) 184/227 (81.06) 2 17/86 (19.77) 15/123 (12.20) 36/156 (23.08) 43/227 (18.94) Stage of embryo transferred, n (%) 0.649 0.281 Cleavage embryo 65/103 (63.11) 91/138 (65.94) 167/192 (86.98) 225/270 (83.33) Blastocyst 38/103 (36.89) 47/138 (34.06) 25/192 (13.02) 45/270 (16.67) Endometrial thickness (mm) 9.0 (8.3, 9.7) 9.1 (8.5, 10.1) 0.207 9.0 (8.2, 9.8) 9.2 (8.4, 10.0) 0.128 Endometrial preparation, n (%) 0.526 0.193 HRT cycles 67/86 (77.91) 104/123 (84.55) 111/156 (71.15) 176/227 (77.53) Nature cycles 6/86 (6.98) 4/123 (3.25) 11/156 (7.05) 20/227 (8.81) Ovulation induction cycles 1/86 (1.16) 1/123 (0.81) 3/156 (1.92) 2/227 (0.88) Down-regulation + HRT cycles 12/86 (13.95) 14/123 (11.38) 31/156 (19.87) 29/227 (12.78) Pregnancy outcomes, n (%) Biochemical pregnancy rate 44/86 (51.16) 63/123 (51.22) 0.994 35/156 (22.44) 53/227 (23.35) 0.835 Clinical pregnancy rate 31/86 (36.05) 54/123 (43.90) 0.255 29/156 (11.33) 37/227 (16.30) 0.560 Ongoing pregnancy rate 24/86 (27.91) 44/123 (35.77) 0.232 17/156 (10.90) 25/227 (11.01) 0.972 Miscarriage rate 7/31 (22.58) 12/54 (22.22) 0.970 13/29 (44.83) 15/37 (40.54) 0.727 Live birth rate 24/86 (27.91) 42/123 (34.15) 0.340 16/156 (10.26) 22/227 (9.69) 0.856 Cumulative clinical pregnancy rate 31/78 (39.74) 54/111 (48.65) 0.226 29/154 (18.83) 37/214 (17.29) 0.704 Cumulative live birth rate 24/78 (30.77) 42/111 (37.84) 0.316 16/154 (10.39) 22/214 (10.28) 0.973 Data were presented as mean ± SD, median (25th percent, 75th percent) and n (%) FET: Frozen embryo transfer; HRT: Hormone replacement therapy * P  < 0.05 Characteristics of embryo transfer and pregnancy outcomes in different groups 69/86 (80.23) 108/123 (87.80) 120/156 (76.92) 184/227 (81.06) 17/86 (19.77) 15/123 (12.20) 36/156 (23.08) 43/227 (18.94) 65/103 (63.11) 91/138 (65.94) 167/192 (86.98) 225/270 (83.33) 38/103 (36.89) 47/138 (34.06) 25/192 (13.02) 45/270 (16.67) 9.0 (8.3, 9.7) 9.1 (8.5, 10.1) 9.0 (8.2, 9.8) 9.2 (8.4, 10.0) 67/86 (77.91) 104/123 (84.55) 111/156 (71.15) 176/227 (77.53) 6/86 (6.98) 4/123 (3.25) 11/156 (7.05) 20/227 (8.81) 1/86 (1.16) 1/123 (0.81) 3/156 (1.92) 2/227 (0.88) 12/86 (13.95) 14/123 (11.38) 31/156 (19.87) 29/227 (12.78) 44/86 (51.16) 63/123 (51.22) 35/156 (22.44) 53/227 (23.35) 31/86 (36.05) 54/123 (43.90) 29/156 (11.33) 37/227 (16.30) 24/86 (27.91) 44/123 (35.77) 17/156 (10.90) 25/227 (11.01) 7/31 (22.58) 12/54 (22.22) 13/29 (44.83) 15/37 (40.54) 24/86 (27.91) 42/123 (34.15) 16/156 (10.26) 22/227 (9.69) 31/78 (39.74) 54/111 (48.65) 29/154 (18.83) 37/214 (17.29) 24/78 (30.77) 42/111 (37.84) 16/154 (10.39) 22/214 (10.28) Data were presented as mean ± SD, median (25th percent, 75th percent) and n (%) FET: Frozen embryo transfer; HRT: Hormone replacement therapy * P  < 0.05 A total of 104 patients achieved live birth. In POSEIDON group 3 patients, the gestational week, average neonatal birthweight, delivery mode and gender of newborn in the PPOS group were comparable to PPOS + LE group ( P  > 0.05), while in the POSEIDON group 4 patients, newborns in PPOS group behaved lower gestational week and birthweight (37.00 ± 2.20 vs. 38.68 ± 1.09, P  = 0.003; 2921.58 ± 608.18 vs. 3332.27 ± 292.96, P  = 0.013). For maternal and neonatal complications, the occurrence of PTB, GDM, HDP, LBW, macrosomia, pneumonia, jaundice and birth defects of newborns did not differ significantly among different groups ( P  > 0.05). Detailed neonatal outcomes are shown in Table  4 . Table 4 Neonatal outcomes in different groups Variables POSEIDON Group 3 P value POSEIDON Group 4 P value PPOS group n = 78 PPOS + LE group n = 111 PPOS group n = 154 PPOS + LE group n = 214 No. of live birth cycle 24 42 16 22 No. of neonatal 26 42 19 22 Gestational age (weeks) 38.40 ± 1.53 38.01 ± 2.35 0.461 37.00 ± 2.20 38.68 ± 1.09 0.003* Mean birth weight (g) 3189.23 ± 535.54 3182.86 ± 703.01 0.969 2921.58 ± 608.18 3332.27 ± 292.96 0.013* Delivery mode, n (%) 1.000 1.000 Vaginal delivery 4/24 (16.67) 7/42 (16.67) 0 1/22 (4.54) Cesarean section 20/24 (83.33) 35/42 (83.33) 16/16 (100.00) 21/22 (95.45) Gender, n (%) 0.583 0.829 Male 16/26 (61.54) 23/42 (54.76) 11/19 (57.89) 12/22 (54.54) Female 10/26 (38.46) 19/42 (45.24) 8/19 (42.11) 10/22 (45.45) Maternal and neonatal complications, n (%) Preterm birth 1/24 (4.17) 3/42 (7.14) 1.000 4/16 (25.00) 2/22 (9.09) 0.217 Gestational Diabetes Mellitus 0 3/42 (7.14) 0.295 1/16 (6.25) 1/22 (4.54) 1.000 Hypertensive disorders of pregnancy 2/24 (8.33) 3/42 (7.14) 1.000 2/16 (12.50) 1/22 (4.54) 0.562 Placenta previa 1/24 (4.17) 3/42 (7.14) 1.000 0 0 N/A Premature rupture of membranes 0 1/42 (2.34) 1.000 0 0 N/A Macrosomia 1/26 (3.85) 2/42 (4.76) 1.000 0 1/22 (4.54) 1.000 Low birth weight 0 0 N/A 0 0 N/A Pneumonia of newborn 0 2/42 (4.76) 0.521 2/19 (10.53) 1/22 (4.54) 0.588 Jaundice of newborn 1/26 (3.85) 3/42 (7.14) 1.000 3/19 (15.79) 2/22 (9.09) 0.649 Birth defect in total 0 0 N/A 0 0 N/A Data were presented as mean ± SD, median (25th percent, 75th percent) and n (%) N/A, not applicable Neonatal outcomes in different groups Data were presented as mean ± SD, median (25th percent, 75th percent) and n (%) N/A, not applicable A univariate logistic regression model was used to study the factors affecting the clinical outcome of CLBR per oocyte retrieval cycle (Supplementary Table 1). Table 5 shows the results of multivariate logistic regression model. CLBR per oocyte was comparable with PPOS vs. PPOS + LE protocol after adjustment for age, BMI, basal AFC, AMH, duration of infertility, insemination method, the number of oocytes retrieved, number of mature oocytes, number of 2PN, and number of available embryos (aOR = 1.34, 95% CI 0.64–2.83, P  = 0.441; aOR = 0.95, 95%CI 0.43–2.09 P  = 0.892) in POSEIDON group 3 and 4 patients, respectively. The results showed that significant favorable effects of the number of available embryos on CLBR in all included patients (aOR = 2.28, 95%CI 1.37–3.78, P  = 0.001; aOR = 2.51, 95%CI 1.48–4.23, P  < 0.001). In POSEIDON group 3, the infertility duration seems to be an adverse effect on the CLBR (aOR = 0.83, 95%CI 0.71–0.98, P  = 0.026) and significant negative effects of age were also identified on the CLBR (aOR = 0.79, 95%CI 0.70–0.91, P  < 0.001) in POSEIDON group 4. Type of stimulation protocol, BMI, AFC, and number of oocytes retrieved, mature oocyte and 2PN embryos were not significant factors associated with CLBR both in POSEIDON group 3 and 4. (all P  > 0.05). Table 5 Multivariate logistic regression analysis results of the CLBR per oocyte retrieval cycle in different groups Parameters POSEIDON Group 3 POSEIDON Group 4 aOR (95% CI) P value aOR (95% CI) P value PPOS 1.00 (Reference) 1.00 (Reference) PPOS + LE 1.34 (0.64–2.83) 0.441 0.95 (0.43–2.09) 0.892 Age (year) 1.04 (0.89–1.23) 0.607 0.79 (0.70–0.91) < 0.001* BMI (kg/m 2 ) 1.000 (0.89–1.12) 0.960 0.92 (0.81–1.04) 0.188 AMH (ng/ml) 0.97 (0.66–1.44) 0.893 0.87 (0.40–1.88) 0.722 AFC 0.89 (0.73–1.09) 0.261 0.91 (0.73–1.14) 0.406 Infertility duration (year) 0.83 (0.71–0.98) 0.026 * 0.92 (0.81–1.05) 0.241 Fertilization methods IVF 1.00 (Reference) 1.00 (Reference) ICSI 0.79 (0.31–2.06) 0.633 1.20 (0.48–3.02) 0.694 No. of oocytes retrieved 0.71 (0.45–1.11) 0.130 0.64 (0.37–1.11) 0.114 No. of MII oocytes 1.30 (0.75–2.26) 0.346 1.38 (0.67–2.86) 0.384 No. of 2PN embryos 1.25 (0.79–1.98) 0.332 1.08 (0.66–1.78) 0.747 No. of available embryos 2.28 (1.37–3.78) 0.001* 2.51 (1.48–4.23) < 0.001* Adjustments were made for age, BMI, duration of infertility, AMH, AFC and fertilization method, the number of oocytes retrieved, number of MII oocytes, number of 2PN embryos, and number of available embryos * P  < 0.05 BMI body mass index, AFC antral follicle count, AMH anti-Müllerian hormone, IVF  in vitro fertilization, ICSI intracytoplasmic sperm injection, CLBR cumulative live birth rate, OR odds ratio, CI confidence interval, 2PN two pronuclei Multivariate logistic regression analysis results of the CLBR per oocyte retrieval cycle in different groups Adjustments were made for age, BMI, duration of infertility, AMH, AFC and fertilization method, the number of oocytes retrieved, number of MII oocytes, number of 2PN embryos, and number of available embryos * P  < 0.05 BMI body mass index, AFC antral follicle count, AMH anti-Müllerian hormone, IVF  in vitro fertilization, ICSI intracytoplasmic sperm injection, CLBR cumulative live birth rate, OR odds ratio, CI confidence interval, 2PN two pronuclei

This retrospective cohort study was performed in POR patients who underwent IVF/ICSI at the Reproductive Medicine Center of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology in China from January 1, 2018 to December 31, 2021. Written informed consent was obtained from all patients prior to participation and all associated data were extracted from the electronic medical record database. Patients who underwent the first PPOS protocol and met the POSEIDON group 3 and 4 criteria were enrolled in this study. POSEIDON group 3 consists of women aged < 35 years with poor ovarian reserve (AFC < 5, AMH < 1.2 ng/mL), and POSEIDON group 4 consists of women aged ≥ 35 years with poor ovarian reserve (AFC < 5, AMH < 1.2 ng/mL). Each patient was followed for 2 years from their ovarian stimulation cycle. Exclusion criteria include (1) chromosomal abnormalities in either spouse; (2) preimplantation genetic testing (PGT); (3) oocyte cryopreservation, oocyte and sperm donation; (4) abnormal uterine anatomical structure, intrauterine adhesion, uterine endometrial polyps and subserous myoma; (5) severe endometriosis [(Stages III–IV following the revised American Society for Reproductive Medicine (r-ASRM) Classification [ 21 ] during laparoscopic surgery)] and adenomyosis; (6) patients with history of recurrent spontaneous abortion (RSA); (7) patients with history of tuberculosis and other serious systemic diseases, such as hypertension, diabetes mellitus, severe thyroid dysfunction, and malignant tumors; (8) combined transfer of embryos from other different stimulated cycles; and (9) no live birth but with surplus embryo by the end of 2 years of follow-up. For the PPOS protocol, patients received Gn daily at a dose of 150–300 IU/day and medroxyprogesterone acetate (MPA) (Zhejiang Xianju Pharmaceutical Co., Ltd.) 10 mg/day simultaneously from day 2 or 3 of their menstrual cycle until the trigger day. In the PPOS + LE group, LE (Fu Rui, Jiangsu Hengrui Pharmaceutical Co., Ltd.) was administered 2.5–5 mg/day for 5 consecutive days, beginning with ovarian stimulation. During the stimulation process, Gn doses were adjusted depending on patients’ follicular development as identified by serial transvaginal sonography (TVS) and serum hormone levels. Once two or more leading follicles reached 18 mm in mean diameter, ovulation was triggered by the injection of 250 μg recombinant human chorionic gonadotrophin (rhCG) (Ovidrel, Merck-Serono, Switzerland) or 10,000 IU hCG (Lizhu Pharmaceutical Trading Co. Ltd., Zhuhai, China) or combined with 0.1 mg triptorelin (Decapeptyl, Ferring Pharmaceuticals, Germany). Oocyte retrieval was performed approximately 36 h later under the TVS guidance. Based on semen parameters, oocytes were fertilized by IVF or ICSI. All embryos were screened on the morning of day 3 after oocyte retrieval and with surplus embryos being cryopreserved or continuously cultured to the blastocyst stage. Due to the effect of high progesterone levels on endometrial receptivity, all embryos obtained were cryopreserved by vitrification using the Cryotop system. Embryos were graded according to the Istanbul consensus and Gardner score [ 22 ]. Frozen-thawed embryo transfers were performed in subsequent cycles with different endometrial preparation protocols. For women with regular menstrual cycles, follicular growth was monitored by TVS starting on days 10–12 of the menstrual cycle. Oral dydrogesterone (20 mg/d) is usually given on the day of ovulation, and cleavage stage embryos or blastocysts were transferred 3 or 5 days after ovulation with up to 2 embryos, respectively. For patients with irregular menstruation who underwent hormone replacement therapy (HRT), oral estradiol was administered on days 2 or 3 of the menstrual cycle until the endometrial thickness was ≥ 7 mm and then endometrial transformation was initiated. For patients using gonadotropin-releasing hormone agonist (GnRH-a) downregulation combined with HRT (GnRH-a + HRT), 3.75 mg GnRH-a was given subcutaneously on the days 1–2 of menstruation, and the HRT was started 28–30 days later to prepare the endometrium. Embryo transfer was performed as before. Luteal support was maintained until 8–10 weeks of gestation or negative β-hCG detection 2 weeks after transfer. The primary outcome of this study was the cumulative live birth rate (CLBR) and cumulative clinical pregnancy rate (CCPR) per oocyte retrieval cycle defined as live birth or clinical pregnancy that occurs during the subsequent FET cycle after the same ovarian stimulation cycle. Multiple births in a single pregnancy were considered a single live birth. The secondary outcomes were the outcomes of response to ovarian stimulation including the number of retrieved oocytes, MII oocytes, two pronuclei (2PN) embryos, available embryos which referred to the number of embryos available for cryopreservation, the oocyte maturation rate (MII oocytes/total retrieved oocytes), the normal fertilization rate (2PN embryos/MII oocytes), the available embryo rate (the available embryo/2PN cleavage embryos), the premature LH surge rate (serum LH level higher than 10 mIU/mL or rising above twice the basal level before the trigger day) and cycle cancellation rate. Cycle cancellation was defined as not obtaining oocytes or no available embryos to be frozen for later transfer. Meanwhile, the pregnancy outcomes including the biochemical pregnancy, clinical pregnancy, ongoing pregnancy, miscarriage, live birth. Maternal and neonatal complications also measured. Biochemical pregnancy was defined as serum β-hCG ≥ 10 mIU/mL. Clinical pregnancy was confirmed as the presence of one or more gestational sacs in the uterine through TVS of 4 weeks after embryo transfer. Ongoing pregnancy refers to the presence of at least 1 fetus with heartbeat in the uterine cavity confirmed by ultrasound beyond 12 weeks. Miscarriage was defined as the loss of pregnancy before 28 gestational weeks. Live birth was considered as the delivery of at least an infant after 28 weeks of gestation. Maternal and neonatal complications include preterm birth (PTB) (live birth between 28 and 37 weeks of gestation), gestational diabetes mellitus (GDM), hypertensive disorders of pregnancy (HDP), placenta previa, premature rupture of membranes (PROM), low birth weight ( 4000 g), pneumonia of newborn, jaundice of newborn and birth defect in total. Information regarding all perinatal outcomes was collected separately by special follow-up staff from telephone interviews after delivery. Data collected were analyzed by SPSS 27.0 (IBM, Chicago, IL). According to the normality of the distribution, continuous variables were presented as mean ± SD or medians (first and third quartile), and categorical variables were described as the number of cases and percentage (%). The significance of differences between means between the two groups were compared using the T test or Mann–Whitney test, as appropriate. Categorical variables were compared through the Pearson’s chi-squared test or Fisher’s exact test. Meanwhile, to explore the relationship among variables, multivariate logistic regression models were conducted. We calculated crude regression estimates as well as estimates adjusted for related baseline covariates with 95% confidence intervals (CI). Two-tailed  P  values < 0.05 were considered statistically significant.

Currently, developing clinical strategies for POR patients is still challenging. POR patients are given the PPOS protocol gradually, and ongoing improvements have been noted, including the type of exogenous progesterone (dydrogesterone and medroxyprogesterone) [ 29 ], flexible and conventional protocol [ 30 ], changing MPA dose [ 31 , 32 ] to better implement this protocol. Research is still underway to discuss the role of LE addition on IVF/ICSI outcomes in POR patients. The potential benefit of LE combination therapy was initially documented by Goswami et al. [ 23 ] in POR patients undergoing IVF/ICSI and subsequent retrospective studies have also demonstrated the benefits of LE combination for POR patients, but mainly in GnRH antagonist and mild stimulation protocols, that include reducing the dose of Gn [ 17 ], decreasing the rate of cancellations and increasing pregnancy and live birth rates [ 24 – 26 ]. Given disparities in pituitary suppression, there is inadequate information to determine if the effects of LE in the PPOS protocol are equivalent to those of other modified regimens. To our knowledge, this is the first study evaluating the effect of LE cotreatment with PPOS in expected POR patients, and our findings showed that in POSEIDON group 4 patients, more oocytes were retrieved as well as mature oocytes, 2PN embryos in the PPOS + LE group. Meanwhile, the CCPR and CLBR of the PPOS group were similar to that of the PPOS + LE group both in POSEIDON group 3 and 4 patients, as were other reproductive outcomes in FET cycles. Further analysis showed LE combination did not significantly alter the majority of neonatal outcomes within the study population. Based on our research, advanced-aged POR patients may benefit more from using LE as an advantageous adjuvant agent in the PPOS protocol to enable to obtain more oocytes and embryos in clinical practice. Actually, young POR patients (POSEIDON group 3) require special consideration due to their age-specific characteristics and stringent fertility requirements. This study revealed that early addition of LE had no discernible improvement in IVF/ICSI outcomes in young POR patients but with a better tendency of reproductive outcomes, consistent with several previous studies which also found enhanced trends for pregnancy outcome in young POR administered LE in comparison with controls without reaching statistical significance [ 17 , 27 ]. Kahraman et al. [ 28 ] reported further encouraging results, that is, in GnRH antagonist protocol with LE, POSEIDON group 3 patients were capable of acquiring higher number of 2PN embryos with significantly higher clinical pregnancy and live birth. To be sure, there is still debate concerning whether or not young POR patients' pregnancy prospects can be improved by taking adjuvant LE administration in PPOS protocol. A recent review discussed the ovarian steroidogenesis in POSEIDON 3 poor responders and attempted to find the expected benefits from improving the androgenic environment in poor responders, discovering that systemic androgen supplementation is likely to be beneficial not only on stimulation parameters but, most importantly, in terms of live birth rate, and discovered that no significant improvement when hCG in combination with aromatase inhibitors are prescribed in these patients regardless of “short or long-term intra-ovarian androgen priming” [ 9 ] and the significant deficit in basal/stimulated ovarian steroidogenesis in young poor responders could definitely explain the undesirable results. Due to the criteria used to define or categorize poorly responding women and the lack of subgroup comparisons, whether the addition of LE improves ovarian steroidogenesis deficits in young POR patients still needs further discussion to find the optimal way to counteract this defect. Differently, for POSEIDON group 4 patients, the poor endometrial receptivity and increased incidence of embryonic aneuploidy may account for their unfavorable results. Research has indicated that the concurrent administration of LE may enhance endometrial receptivity [ 29 ] and effective methods for enhancing oocyte and embryo outcomes for this population are also crucial. Lin et al. [ 30 ] found that CC and LE combined with PPOS protocol and GnRH antagonist protocol can increase the high-quality oocyte ratio and yield comparable fertilization rate and pregnancy outcomes as well as achieve higher number of oocyte retrieved and 2PN embryos although without significant difference in POSEIDON group 4 patients. The higher yields of oocytes, mature oocytes and improved embryo quality were observed in a study with extended use of 5 mg/d LE from the first day of gonadotropin stimulation to the day of triggering although without the POSEIDON stratification system [ 26 ], but considering the small study population and subjective biases on embryo assessment, results should be interpreted with caution. Research is still underway to discuss the role of LE addition on IVF/ICSI outcomes in POR patients. In this study, POSEIDON group 4 patients also achieved high number of mature oocytes, 2PN embryos and achieved comparable pregnancy outcomes with the LE cotreatment in PPOS protocol. With both differences and similarities to previous studies, we postulated that the different results between these studies, most which reported higher oocytes and embryos but without improved pregnancy outcomes might be due to the crowd characteristics with high aneuploidy rate, different methodology and using different criteria and cutoff values for ovarian reserve tests to define the POR. Moreover, it may be attributed to the variations in sample sizes, different doses of LE used (2.5 vs. 5 mg) and the different maintenance days. In addition, the unknown effects of different ranges of LH windows on folliculogenesis and oocyte maturation also warrant consideration [ 31 ]. However, these promising results also emphasize the potential and clinical value of LE in combination with PPOS in advanced-aged POR patients. Notably, our findings showed that LH levels on the trigger day were higher in the LE-supplemented group, which may be induced by LE by blocking estrogen production. The application of MPA in conventional PPOS protocol may lead to greater LH suppression, and low endogenous LH negatively affects follicular development and pregnancy outcome [ 32 , 33 ], the addition of LE attenuates this suppression without affecting the MPA effect. On the other hand, despite the increase in LH levels in the group with added LE, the rate of premature LH surge was similar in both groups and the FET pregnancy outcome was similar between the two groups, suggesting that the embryos in both groups had the same developmental potential. Changes in LH trends during ovulation induction may be balanced between LH and MPA. Consisted with several studies showing that CLBR significantly increased with the number of oocytes and embryos retrieved [ 34 , 35 ], our analysis also indicated that the number of retrieved oocytes, mature oocytes, 2PN embryos and available embryos were positive with CLBR, and after adjustments for confounding factors, the available embryos was the main factor both in POSEIDON group 3 and 4 patients. Furthermore, age was inversely correlated with CLBR in POSEIDON 4 patients, which also because the quantity and quality of oocytes usually decreased with age and the probability of aneuploid embryos increased with age [ 36 ]. However, there was no difference in the CLBR between the PPOS + LE and control groups in POSEIDON group 3 and 4 patients, respectively. This was likely because only FET results of one IVF/ICSI cycle were analyzed in this study, but for most POR patients, multiple IVF/ICSI cycles are usually required to improve the chances of pregnancy and live birth and it was difficult to obtain statistical differences based on only one oocyte retrieval cycle. Correspondingly, this study suggested that LE cotreatment with PPOS protocol exhibited no significant association with CLBR. For PPOS protocol, fewer studies have been reported on the application of different combined oral medications on the basis of this regimen. The positive effects of CC application on IVF/ICSI outcomes have been focused on, especially in POR patients [ 31 , 37 , 38 ]. LE and CC act similarly and their application in other populations suggests similar positive effects, but reports on the use of CC and LE in combination with PPOS regimens in patients with POR, respectively, are very limited, which gives us new directions to ponder. Finally, the potential effects of MPA on fetal growth have been safely investigated in recent years [ 39 ] and nor is it clear from some evidence that LE can increase the risk of fetal harm [ 40 ]. It could be questioned whether the combined use of LE and MPA poses any teratogenic risk. Although our neonatal data suggest that the combination of LE and MPA did not increase the risk of adverse birth outcomes, the limited data still requires in-depth consideration of any questions of safety. To our knowledge, this is the first study to evaluate the impact of LE cotreatment in PPOS protocol among POR patients with POSEIDON criteria for detailed subgroups, which provides a solid basis for the development of clinical protocols and individualized fertility treatments. Furthermore, to eliminate the possible impact of multiple previous IVF attempts and more comprehensively evaluated the entire ovulation induction cycle, this study exclusively included women undergoing their first IVF/ICSI cycle with PPOS and PPOS + LE protocol and used CCPR and CLBR as the final observations. However, there are certain restrictions on this study. First, this study was a retrospective cohort study with small sample size. Despite participants have similar baseline characteristics and were modified with logistic regression analysis, confounding variables interfered with the results. Meanwhile, as our study population consisted of POR patients who required multiple ovulation induction to obtain a limited number of embryos and, therefore, a relatively small number of FET cycles. Finally, no detailed subgrouping of our early LE dosage was performed due to the limited sample size and further studies with larger sample sizes would be necessary to confirm our findings regarding the disparities in IVF/ICSI outcomes of the two protocols.

For POSEIDON group 3 and 4 patients, the CCPR and CLBR of the PPOS and PPOS + LE protocols were comparable. For POSEIDON group 4, the number of oocytes retrieved, mature oocytes, 2PN embryos was higher in the PPOS + LE group. Both in the POSEIDON 3 and 4 group, the number of available embryos showed a positive correlation with CLBR. More large-scale randomized controlled studies are necessary to further evaluate the effects of LE in PPOS protocol for POR patients.

With a prevalence of 9–24%, poor ovarian response (POR) is the primary factor responsible for infertility in assisted reproductive technology (ART) characterized by the recruitment of fewer follicles than the desired target, high gonadotropin (Gn) doses, increased cycle cancellation rates as well as lower pregnancy and live birth rates (LBR) [ 1 , 2 ]. Recently, a new stratification of POR patients based on combination of qualitative (age), quantitative [serum anti-Müllerian hormone (AMH), antral follicle count (AFC)], and ovarian response to prior stimulation treatments based on POSEIDON criteria was described. This allowed for the identification of four distinct groups [ 3 , 4 ], which is more meaningful since grabbing into account the number of acquired oocytes, age-related embryonic aneuploidy rate, and ovarian sensitivity to exogenous Gn compared with the Bologna criteria [ 5 ]. For POR patients, obtaining sufficient oocytes through personalized ovarian stimulation protocol is essential to achieve higher number of transferable and high-quality embryos to fulfill their reproductive goals. The alternative protocol known as progestin-primed ovarian stimulation (PPOS), proposed in 2015 by Dr. Kuang and gradually applied to POR patients, can not only more effectively suppress the premature luteinizing hormone (LH) surge and increase the number of oocytes retrieved through using the exogenous progesterone, but also be more flexible and affordable [ 6 , 7 ]. In addition, advances in frozen-thawed embryo technology and its progressively proven safety have also facilitated the progress of PPOS [ 8 ]. The pathogeny underlying POR remains mostly unclear and often recommended strategies, such as prescription of higher follicle stimulating hormone (FSH) doses and/or addition of products with LH-activity proved to be partly ineffective in improving follicular recruitment and/or the number of good quality embryos [ 9 ]. Numerous studies have demonstrated how ovarian paracrine factors affect follicular development and stressed the androgens’ role in ovarian physiology [ 10 , 11 ]. Androgen treatment can improve ovarian response by increasing granulosa cell proliferation, attenuating follicular atresia, as well as enhancing preantral follicular growth [ 12 ] and increasing the antral follicle sensitivity to Gn [ 13 ]. Research on the value of systemic or intra-ovarian androgen priming in a clinical setting is ongoing. Letrozole (LE), a third-generation aromatase inhibitor, can promote follicular development by enhancing ovarian response to Gn through temporarily increasing accumulation of androgens in the ovarian microenvironment and promoting the increase of endogenous gonadotropin secretion to facilitate follicular growth and development [ 14 ]. LE cotreatment with GnRH antagonist strategy during ovarian stimulation has been reported in several studies with positive effects on in vitro fertilization/intracytoplasmic sperm injection (IVF/ICSI) outcomes among POR patients [ 15 – 18 ]. PPOS protocol combined with early LE was previously used in women with normal ovarian response (NOR) and polycystic ovarian syndrome (PCOS) patients, effectively improving follicular output rate (FORT) and pregnancy outcomes [ 19 , 20 ]. There are few studies on the early combination with LE in the PPOS protocol for POR patients. Therefore, we conducted a retrospective cohort study to evaluate the impact of the early addition of LE in PPOS protocol on IVF/ICSI outcomes among expected POR patients including POSEIDON group 3 and 4 patients, intending to provide new thoughts for improving POR patients’ outcomes.

Additional file 1 Additional file 1