Impact of BMI on pregnancy outcomes in PCOS patients undergoing ultralong GnRH-a protocol with blastocyst transfer.

There were no significant differences in infertility type, infertile causes, basal progesterone or testosterone. Females in underweight group were younger than those in overweight group ( p  = 0.018) and obese group ( p  = 0.019). Males in underweight and normal weight group were younger than those in overweight and obese group, all p  < 0.05). The duration of infertility in the underweight and normal weight group were significantly shorter than that in overweight and obese group (all p  ≤ 0.001). AFC of patients in obese group were larger than that in underweight and normal weight group ( p  = 0.009 and p  = 0.012, respectively). Females in overweight group and obese group showed higher fasting plasma glucose, higher serum insulin and lower AMH level than those in underweight and normal weight (all p  < 0.001) (Table  1 ). Table 1 Comparison of baseline characteristics Underweight ( N  = 125) Normal weight ( N  = 845) Overweight ( N  = 517) Obese ( N  = 217) P BMI (Median, IQR) — (kg/m 2 ) 17.63 (16.90,18.22) abc 21.26 (20.03,22.50) de 25.78 (24.84,26.71) f 29.55 (28.84,30.81) 0.000 * Maternal age (Median, IQR) —(year) 28 (27,30.5) bc 29 (27,32) 29 (27,32) 30 (27,32) 0.013 * Male age (Median, IQR) —(year) 30 (28,32) bc 31 (29,34) de 32 (30,35) 32 (29,35) 0.000 * Infertility duration (Median, IQR) —(year) 3 (2,4) bc 3 (2,4) de 3 (2,5) 4 (2,6) 0.000 * Infertility type  Primary infertility— no. (%) 82 (65.6) 498 (58.9) 313 (60.5) 138 (63.6) 0.377  Secondary infertility— no. (%) 43 (34.4) 347 (41.1) 204 (39.5) 79 (36.4) Infertile causes  PCOS for single cause— no. (%) 32 (25.6) 205 (24.3) 146 (28.2) 65 (30.0) 0.220  PCOS combined with other female factors — no. (%) 19 (15.2) 167 (19.8) 91 (17.6) 40 (18.4)  PCOS combined with male factors— no. (%) 49 (39.2) 267 (17.6) 174 (33.7) 74 (34.1)  PCOS combined with both factors— no. (%) 25 (20.0) 206 (18.4) 106 (20.5) 38 (17.5)  AMH (Median, IQR) —(ng/ml) 7.30 (4.80,9.68) bc 6.38 (4.51,9.23) de 5.77 (4.11,8.43) 5.39 (4.01,7.62) 0.000 *  AFC (Median, IQR) 28 (24,30.5) c 28 (24,33) e 28 (24,35) 24 (24,37) 0.002 * Basal hormones  LH (Median, IQR)—(IU/L) 8.55 (6.19,13.16) abc 7.07 (4.93,11.09) de 6.34 (4.19,9.61) f 5.64 (3.79,7.95) 0.000 *  FSH (Median, IQR)—(IU/L) 7.01 (5.92,8.23) abc 6.30 (5.39,7.36) 6.34 (5.20,7.26) 6.01 (5.17,7.32) 0.000 *  PRL (Median, IQR)—(ng/ml) 12.25 (8.30,17.23) bc 11.21 (8.36,15.93) de 10.06 (7.32,13.88) 9.75 (7.41,13.43) 0.000 *  Progesterone (Median, IQR)—(ng/ml) 0.5 (0.34,0.65) 0.48 (0.36,0.67) 0.46 (0.36,0.63) 0.46 (0.34,0.61) 0.074  E 2 (Median, IQR)—(pg/ml) 47.68 (36.36,56.42) c 43.00 (34.32,53.30) 42.80 (34.24,52.76) 40.70 (34.25,49.97) 0.021 *  Testosterone (Median, IQR)—(ng/ml) 0.41 (0.27,0.56) 0.44 (0.30,0.59) 0.43 (0.30,0.62) 0.43 (0.31,0.61) 0.593  Fasting plasma glucose (Median, IQR)—(mmol/L) 4.82 (4.59,5.15) abc 4.97 (4.67,5.24) de 5.01 (4.76,5.44) 5.14 (4.87,5.42) 0.000 *  Fasting serum insulin (Median, IQR)—(mU/L) 6.29 (4.00,8.86) abc 8.68 (5.86,12.20) de 13.50 (10.10,18.90) 16.30 (11.38,23.95) 0.000 * “a” represents P value less than 0.05 between underweight group and normal weight group, “b” represents P value less than 0.05 between underweight group and overweight group, “c” represents P value less than 0.05 between underweight group and obese group, “d” represents P value less than 0.05 between normal weight group and overweight group, “e” represents P value less than 0.05 between normal weight group and obese group, “f” represents P value less than 0.05 between overweight group and obese group * P  < 0.05 was statistical significance Comparison of baseline characteristics “a” represents P value less than 0.05 between underweight group and normal weight group, “b” represents P value less than 0.05 between underweight group and overweight group, “c” represents P value less than 0.05 between underweight group and obese group, “d” represents P value less than 0.05 between normal weight group and overweight group, “e” represents P value less than 0.05 between normal weight group and obese group, “f” represents P value less than 0.05 between overweight group and obese group * P  < 0.05 was statistical significance Underweight and normal weight groups demonstrated significantly shorter ovarian stimulation duration and lower gonadotropin requirements than overweight and obese counterparts (all p  < 0.001). These leaner cohorts concurrently exhibited higher LH, E 2 , and progesterone levels on hCG trigger day. Embryologically, underweight and normal weight patients yielded more oocytes, 2PN zygotes, and day-3 embryos versus overweight and obese groups (all p  ≤ 0.001). Notably, the underweight group demonstrated significantly superior reproductive outcomes compared with other BMI categories, including: (1) the highest fertilization rate ( p  < 0.0083 vs. all groups), (2) superior day-3 high-quality embryo counts ( p  < 0.05 vs. all groups), and (3) enhanced blastocyst metrics—including formation rate (all p  < 0.0083) , available blastocysts (all  p  < 0.05), and high-quality blastocysts (all p  < 0.05)—exceeding normal weight and obese groups. Conversely, no intergroup differences existed in oocyte maturation rate, high-quality blastocyst rate, cycles without available blastocysts, or moderate-severe OHSS incidence. Paradoxically, overall ovarian hyperstimulation syndrome (OHSS) occurrence was higher in underweight and normal weight patients than overweight and obese cohorts (Table  2 ). Table 2 Ovarian stimulation results and embryology outcomes Underweight Normal weight Overweight Obese P Duration of stimulation (Median, IQR)—(day) 11 (10,12.5) bc 11 (10,13) de 12 (11,15) 13 (11,16) 0.000 * Total gonadotropin dose (Median, IQR)—(IU) 1500 (1125,1800) bc 1500 (1200,1950) de 2100 (1650,2712.5) f 2625 (2025,3225) 0.000 * Hormones on hCG trigger day  LH (Median, IQR)—(IU/L) 0.62 (0.43,0.91) ab 0.52 (0.35,0.80) de 0.38 (0.26,0.60) 0.38 (0.23,0.60) 0.000 *  E 2 (Median, IQR)—(pg/ml) 3098 (2321,4289) abc 2699 (1847,3660) de 2037 (1389,2884) 1980 (1284,2834) 0.000 *  Progesterone (Median, IQR)—(ng/ml) 0.64 (0.45,0.92) bc 0.57 (0.39,0.82) de 0.50 (0.34,0.70) 0.51 (0.35,0.71) 0.000 *  Endometrial thickness on hCG trigger day (Median, IQR)—(mm) 10.7 (9.7,12.4) 11.0 (9.7,12.5) 11.2 (10.0,12.7) 10.9 (10.0,12.6) 0.153  ICSI— no. (%) 19 (15.2) 146 (17.3) 90 (17.4) 41 (18.9) 0.857 Total No. of oocytes retrieved 2203 14,360 7857 3200 No. of oocytes retrieved (Median, IQR) 16 (12,22) bc 16 (12,21) de 14 (10,20) 14 (9,19) 0.000 * No. of mature oocytes (Median, IQR) 15 (11,19) bc 14 (10,19) de 12 (8.5,18) 12 (8,17) 0.000 * Oocytes maturation rate—no./total no. (%) 1940/2203 (88.1) 12,561/14360 (87.5) 6875/7857 (87.5) 2800/3200 (87.5) 0.891 No. of 2 pronuclear zygotes (Median, IQR) 12 (9,16) bc 11 (7,15) de 10 (6,14) 9 (5.5,14) 0.000 * Fertilization rate—no./total no. (%) 1595/1940 (82.2) abc 9971/12561 (79.4) 5424/6875 (78.9) 2205/2800 (78.8) 0.011 * Total No. of embryos on Day3 1544 9725 5287 2137 No. of embryos on Day3 (Median, IQR) 12 (9,16) bc 11 (7,15) de 9 (6,14) 9 (5,13) 0.000 * No. of high-quality embryos on Day3 (Median, IQR) 5(3,8) abc 4 (2,7) d 4 (2,6) 3 (2,5) 0.000 * High-quality embryo on Day3 rate—no./total no. (%) 713/1544 (46.2) ab 4015/9725 (41.3) 2203/5287 (41.7) 905/2137 (42.3) 0.004 * Total No. of available blastocysts 868 5073 2792 1065 No. of available blastocysts (Median, IQR) 6 (4,9) abc 5(3,8) de 5 (3,7) 5 (2,7) 0.000 * Blastocyst formation rate—no./total no. (%) 868/1544 (56.2) ac 5073/9725 (52.2) 2792/5287 (52.8) 1065/2137 (49.8) 0.002 * No. of high-quality blastocysts (Median, IQR) 3 (1,5) abc 2 (1,4) 2 (1,3) 2 (1,3) 0.000 * High-quality blastocyst rate—no./total no. (%) 429/868 (49.4) 2341/5073 (46.1) 1274/2792 (45.6) 478/1065 (44.9) 0.192 OHSS—no. (%) 46 (36.8) bc 246 (29.1) de 109 (21.1) 29 (13.4) 0.000 * Medium or severe OHSS—no. (%) 3 (2.4) 37 (4.4) 21 (4.1) 6 (2.8) 0.566 No. of cycles without available blastocysts—no. (%) 2 (1.6) 27 (3.2) 19 (3.7) 8 (3.7) 0.683 “a” represents P value less than 0.05 between underweight group and normal weight group, “b” represents P value less than 0.05 between underweight group and overweight group, “c” represents P value less than 0.05 between underweight group and obese group, “d” represents P value less than 0.05 between normal weight group and overweight group, “e” represents P value less than 0.05 between normal weight group and obese group, “f” represents P value less than 0.05 between overweight group and obese group * P  < 0.05 was statistical significance Ovarian stimulation results and embryology outcomes “a” represents P value less than 0.05 between underweight group and normal weight group, “b” represents P value less than 0.05 between underweight group and overweight group, “c” represents P value less than 0.05 between underweight group and obese group, “d” represents P value less than 0.05 between normal weight group and overweight group, “e” represents P value less than 0.05 between normal weight group and obese group, “f” represents P value less than 0.05 between overweight group and obese group * P  < 0.05 was statistical significance 1082 patients underwent their fresh blastocyst transfer. Patients in overweight group (67.3%) and obese group (72.8%) had higher fresh blastocyst transfer rate than that normal weight group (59.3%, all p  < 0.0083). The causes for fresh blastocyst transfer cancellation were different among groups. Cancellation for OHSS in obese group (42.4%) was lower than that in underweight group (70%, p  = 0.004) and for embryonic factors was higher in obese group (28.8%) and in overweight group (27.8%) than in underweight group (8%, all p  < 0.0083) (Table  3 ). Table 3 Main causes for fresh blastocyst transfer cancellation Underweight (50) Normal weight (344) Overweight (169) Obese (59) P Main causes for fresh blastocyst transfer cancellation  Embryonic factors—no. (%) 4 (8.0) bc 63 (18.3) 47/169 (27.8) 17/59 (28.8) 0.004 *  OHSS—no. (%) 35 (70.0) c 200 (58.2) 99 (58.6) 25 (42.4) 0.031 *  More than one causes—no. (%) 9 (18.0) b 40 (11.6) 8 (4.7) 4 (6.8) 0.021 *  Others—no. (%) 2 (4.0) c 41 (11.9) 15 (8.9) f 13 (22.0) 0.015 * “a” represents P value less than 0.05 between underweight group and normal weight group, “b” represents P value less than 0.05 between underweight group and overweight group, “c” represents P value less than 0.05 between underweight group and obese group, “d” represents P value less than 0.05 between normal weight group and overweight group, “e” represents P value less than 0.05 between normal weight group and obese group, “f” represents P value less than 0.05 between overweight group and obese group * P  < 0.05 was statistical significance Main causes for fresh blastocyst transfer cancellation “a” represents P value less than 0.05 between underweight group and normal weight group, “b” represents P value less than 0.05 between underweight group and overweight group, “c” represents P value less than 0.05 between underweight group and obese group, “d” represents P value less than 0.05 between normal weight group and overweight group, “e” represents P value less than 0.05 between normal weight group and obese group, “f” represents P value less than 0.05 between overweight group and obese group * P  < 0.05 was statistical significance The number of blastocysts and high-quality blastocysts transferred among four groups were not different. Respecting the clinical outcomes, there was no significant difference in the positive hCG rate, biochemical pregnancy rate, clinical pregnancy rate, live birth rate, preterm labor rate, gestational age on delivery, neonatal birth weight, or incidence of gestational hypertension among the four groups (all p  > 0.05). The incidence of gestational diabetes mellitus and macrosomia in obese group were higher than that in normal weight group ( p  < 0.001) (Table  4 ). Table 4 Comparison of pregnancy outcomes and neonatal outcomes in fresh blastocyst transfer Underweight Normal weight Overweight Obese P Number of transferred cycles (fresh blastocyst)—no./total no. (%) 75/125 (60.0) 501/845 (59.3) de 348/517 (67.3) 158/217 (72.8) 0.000 *  No. of blastocysts transferred  1—no. (%) 62/75 (82.7) 404/501 (80.6) 272/348 (78.2) 129/158 (81.6) 0.692  2—no. (%) 13/75 (17.3) 97/501 (19.4) 76/348 (12.8) 29/158 (18.4) No. of high-quality blastocysts transferred  0—no. (%) 11/75 (14.7) 54/501 (10.8) 59/348 (16.9) 17/158 (10.8) 0.232  1—no. (%) 61/75 (81.3) 427/501 (85.2) 278/348 (79.9) 135/158 (85.4)  2—no. (%) 3/75 (4.0) 20/501 (4.0) 11/348 (3.2) 6/158 (3.8) Positive β-hcg—no./total no. (%) 55/75 (73.3) 396/501 (79.0) 269/348 (77.3) 121/158 (76.6) 0.684 Biochemical pregnancy—no./total no. (%) 10/55 (18.2) 50/396 (12.6) 29/269 (10.8) 15/121 (12.4) 0.500 Clinical pregnancy—no./total no. (%) 45/75 (60.0) 346/501 (69.1) 240/348 (69.0) 106/158 (67.1) 0.445 Twin pregnancy rate—no./total no. (%) 4/45 (8.9) 31/346 (9.0) 26/240 (10.8) 8/106 (7.5) 0.777 Miscarriage—no./total no. (%) 4/45 (8.9) 52/346 (15.0) 36/240 (15.0) 27/106 (25.5) 0.028 * Pregnancy complication—no./total no. (%)  Gestational hypertension—no./total no. (%) 2/45 (4.4) 2/346 (0.6) 2/240 (0.8) 1/106 (0.9) 0.120  Gestational diabetes mellitus—no./total no. (%) 2/45 (4.4) 13/346 (3.8) e 21/240 (8.8) 17/106 (16.0) 0.000 *  Premature rupture of membranes—no./total no. (%) 1/45 (2.2) 9/346 (2.6) 10/240 (4.2) 3/106 (2.8) 0.770  Live birth—no./total no. (%) 41/75 (54.70) 293/501 (58.5) 204/348 (58.6) 79/158 (50.0) 0.246  Stillbirth—no./total no. (%) 0/41 (0.0) 1/294 (0.3) 0/204 (0.0) 0/79 (0.0) 1.000  Preterm labor—no./total no. (%) 6/41 (14.6) 38/293 (13.0) 40/204 (19.6) 13/79 (16.5) 0.254  Birth of twins —no./total no. (%) 4/41 (9.8) 30/293 (10.2) 23/204 (11.3) 6/79 (7.6) 0.837  Gestational age on delivery Median (IQR)—weeks 39.43 (38.14,40.21) 39.00 (37.93,39.71) 39.00 (37.57,39.86) 39.00 (37.43,39.71) 0.237  Number of neonates (alive) 45 323 227 85 –  Birth weight Median (IQR)—(g) 3040 (2575,3375) 3150 (2800,3500) 3200 (2650,3580) 3350 (2825.3775) 0.074  Macrosomia—no./total no. (%) 1/45 (2.2) 17/323 (5.3) e 14/227 (6.2) 12/85 (14.1) 0.016 *  Low birth weight infants—no./total no. (%) 9/45 (20.0) 43/323 (13.3) 44/227 (19.4) 12/85 (14.1) 0.217 Neonatal sex ratio—no./total no. (%)  Male 29/45 (64.4) 189/323 (58.5) 132/227 (58.1) 47/85 (55.3) 0.796  Female 16/45 (35.6) 134/323 (41.5) 95/227 (41.9) 38/85 (44.7) “a” represents P value less than 0.05 between underweight group and normal weight group, “b” represents P value less than 0.05 between underweight group and overweight group, “c” represents P value less than 0.05 between underweight group and obese group, “d” represents P value less than 0.05 between normal weight group and overweight group, “e” represents P value less than 0.05 between normal weight group and obese group, “f” represents P value less than 0.05 between overweight group and obese group * P  < 0.05 was statistical significance Comparison of pregnancy outcomes and neonatal outcomes in fresh blastocyst transfer “a” represents P value less than 0.05 between underweight group and normal weight group, “b” represents P value less than 0.05 between underweight group and overweight group, “c” represents P value less than 0.05 between underweight group and obese group, “d” represents P value less than 0.05 between normal weight group and overweight group, “e” represents P value less than 0.05 between normal weight group and obese group, “f” represents P value less than 0.05 between overweight group and obese group * P  < 0.05 was statistical significance The 1082 fresh blastocyst transfer cycles were divided into a live birth group ( n  = 617) and a nonlive birth group ( n  = 465) according to whether the live birth outcome was recorded. Multivariate logistic regression analysis showed that the number of high-quality blastocysts and male age were main factors associated with live birth. Dominant favorable factors for live birth outcomes included the number of high-quality blastocysts transferred (OR = 1.737, 95% CI 1.271–2.375) and male age < 35 years (OR = 1.393, 95% CI 1.038–1.870) (Table  5 ). Table 5 Factors associated with live birth rate in fresh blastocyst transfer Factors Unadjusted Adjusted OR (95%CI) P OR (95%CI) P BMI—(kg/m 2)  18.5 ≤ BMI < 24 Ref Ref  BMI ≥ 28 0.710 (0.496–1.016) 0.061 0.749 (0.520–1.079) 0.121  24 ≤ BMI < 28 1.006 (0.762–1.328) 0.968 1.103 (0.830–1.465) 0.500  BMI < 18.5 0.856 (0.525–1.395) 0.533 0.833 (0.509–1.365) 0.469 Number of blastocysts transferred  1 Ref  2 1.223 (0.901–1.659) 0.197  Number of high-quality blastocysts transferred 1.763 (1.295–2.401) 0.000 * 1.737 (1.271–2.375) 0.000 * Infertile causes with male factors  No Ref  Yes 1.217 (0.955–1.550) 0.112 Type of infertility  Primary infertility Ref  Secondary infertility 1.080 (0.846–1.381) 0.536 Duration of infertility—(years) 0.936 (0.887–0.988) 0.016 * 0.954 (0.902–1.009) 0.102 Female age—(years)  < 35 Ref   ≥ 35 0.777 (0.521–1.159) 0.217 Male age —(years)   ≥ 35 Ref Ref   < 35 1.494 (1.125–1.984) 0.006 * 1.393 (1.038–1.870) 0.027 * Type of fertilization  IVF Ref  ICSI 0.966 (0.704–1.324) 0.828  AFC 1.007 (0.990–1.024) 0.420  AMH —(ng/ml) 1.004 (0.971–1.039) 0.798 Basal hormones  LH—(IU/L) 1.008 (0.985–1.031) 0.495  FSH—(IU/L) 1.008 (0.941–1.079) 0.827  PRL—(ng/ml) 0.993 (0.979–1.008) 0.364  Progesterone—(ng/ml) 1.234 (0.787–1.934) 0.360  E 2 —(pg/ml) 1.005 (0.998–1.012) 0.139  Testosterone—(ng/ml) 1.599 (0.896–2.853) 0.112  Fasting plasma glucose (Median, IQR)—(mmol/L) 0.955 (0.775–1.178) 0.670  Fasting serum insulin (Median, IQR)—(mU/L) 0.998 (0.989–1.006) 0.613  Duration of Gn 0.999 (0.962–1.039) 0.974  Dose of Gn 1.000 (1.000–1.000) 0.157 Hormones on hCG trigger day  LH—(IU/L) 0.868 (0.687–1.097) 0.235  E 2 —(pg/ml) 1.000 (1.000–1.000) 0.217  Progesterone—(ng/ml) 0.758 (0.484–1.187) 0.226  Endometrium thickness 1.021 (0.962–1.084) 0.498  No. of oocytes retrieved 0.990 (0.970–1.011) 0.359 * P  < 0.05 was statistical significance Factors associated with live birth rate in fresh blastocyst transfer * P  < 0.05 was statistical significance 1356 cycles of FBT were performed. No difference was found in the number of blastocysts transferred among groups while the number of high-quality blastocysts transferred in underweight group differed from other 3 groups (all p  < 0.0083). Obese group showed lower hCG positive rate than that in normal weight group and lowest clinical pregnancy rate than that in other 3 groups (all p  < 0.0083). The miscarriage rate increased in obese group when compared with underweight group and normal weight group. The live birth rate in obese group were lowest in 4 groups (all p  < 0.0083). Meanwhile overweight group showed lower live birth rate than underweight group ( p   0.05) (Table  6 ). Table 6 Comparison of pregnancy outcomes and neonatal outcomes in frozen-thawed blastocyst transfer Underweight Normal weight Overweight Obese P Number of transferred cycles —no 100 714 378 164 Endometrial thickness (Mean, SD)—(mm) 9.0 (8.3,10.0) 9.0 (8.2–10.0) 9.5 (8.5–10.4) 9.0 (8.2–10.3) 0.105 No. of blastocysts transferred  1—no. (%) 36/100 (36.0) 208/714 (29.1) 125/378 (33.1) 51/164 (31.1) 0.378  2—no. (%) 64/100 (64.0) 506/714 (70.9) 253/378 (66.9) 113/164 (68.9) No. of high-quality blastocysts transferred  0—no. (%) 18/100 (18.0) abc 212/714 (29.7) 123/378 (32.5) 59/164 (36.0) 0.001 *  1—no. (%) 52/100 (52.0) 389/714 (54.5) 203/378 (53.7) 87/164 (53.0)  2—no. (%) 30/100 (30.0) 113/714 (15.8) 52/378 (13.8) 18/164 (11.0) Positive β-hcg—no./total no. (%) 75/100 (75%) 525/714 (73.5) e 269/378 (71.2) 101/164 (61.6) 0.019 * Biochemical pregnancy—no./total no. (%) 8/75 (10.7) c 95/525 (18.1) 49/269 (18.2) 29/101 (28.7) 0.018 * Clinical pregnancy—no./total no. (%) 67/100 (67.0) c 430/714 (60.2) e 220/378 (58.2) f 72/164 (43.9) 0.000 * Twin pregnancy rate—no./total no. (%) 14/67 93/430 43/220 9/72 – Miscarriage—no./total no. (%) 6/67 (9.0) bc 82/430 (19.1) e 54/220 (24.5) 27/72 (37.5) 0.000 * Pregnancy complication—no./total no. (%)  Hypertensive disorders—no./total no. (%) 1/67 (1.5) 4/430 (0.9) 7/220 (3.2) 2/72 (2.8) 0.131  Gestational diabetes mellitus—no./total no. (%) 2/67 (3.0) 15/430 (3.5) 18/220 (8.2) 7/72 (9.7) 0.019 *  Premature rupture of membranes—no./total no. (%) 1/67 (1.5) 10/430 (2.3) 8/220 (3.6) 3/72 (4.2) 0.589  Live birth—no./total no. (%) 61/100 (61.0) bc 348/714 (48.7) e 166/378 (43.9) f 45/164 (27.4) 0.000 *  Preterm labor—no./total no. (%) 11/100 (11.0) 70/714 (9.8) 42/378 (11.1) 13/164 (7.9) 0.699  Birth of twins —no./total no. (%) 13/61 (21.3) 88/348 (25.3) 38/166 (22.9) 7/45 (17.8) 0.658  Gestational age on delivery Median (IQR)—weeks 39.14 (37.14,40.29) 38.86 (37.29,39.86) 38.57 (36.86,39.71) 37.86 (35.64,39.71) 0.125  Number of neonates (alive) 74 436 204 52  Birth weight Median (IQR)—(g) 2900 (2350,3450) 3050 (2505,3450) 3045 (2500,3442.5) 2950 (2062.5,3487.5) 0.642  Macrosomia—no./total no. (%) 0/74 (0.0) 20/436 (4.6) 13/204 (6.4) 5/52 (9.6) 0.061  Low birth weight infants—no./total no. (%) 23/74 (31.1) 95/436 (21.8) 49/204 (24.0) 18/52 (34.6) 0.095 Neonatal sex ratio—no./total no. (%)  Male 39/74 (52.7) 224/436 (51.4) 86/204 (42.2) 32/52 (61.5) 0.040 *  Female 35/74 (47.3) 212/436 (48.6) 118/204 (57.8) 20/52 (38.5) a represents P value less than 0.05 between underweight group and normal weight group, b represents P value less than 0.05 between underweight group and overweight group, c represents P value less than 0.05 between underweight group and obese group, d represents P value less than 0.05 between normal weight group and overweight group, e represents P value less than 0.05 between normal weight group and obese group, f represents P value less than 0.05 between overweight group and obese group. * P P < 0.05 was statistical significance Comparison of pregnancy outcomes and neonatal outcomes in frozen-thawed blastocyst transfer a represents P value less than 0.05 between underweight group and normal weight group, b represents P value less than 0.05 between underweight group and overweight group, c represents P value less than 0.05 between underweight group and obese group, d represents P value less than 0.05 between normal weight group and overweight group, e represents P value less than 0.05 between normal weight group and obese group, f represents P value less than 0.05 between overweight group and obese group. * P P < 0.05 was statistical significance 1356 FBT cycles were divided into a live birth group ( n  = 620) and a nonlive birth group ( n  = 736) according to whether the live birth outcome was recorded. Multivariate logistic regression analysis showed that obesity, the number of high-quality blastocysts and endometrium thickness were main factors associated with live birth. Dominant favorable factors for live birth outcomes included the number of high-quality blastocysts transfer (OR = 1.480, 95% CI 1.251–1.751) and endometrium thickness (OR = 1.101, 95% CI 1.022–1.186). Risk factor for nonlive birth was obesity (OR = 0.437, 95% CI 0.298–0.641) (Table  7 ). Table 7 Factors associated with live birth rate in frozen-thawed blastocyst transfer Factors Unadjusted Adjusted OR (95%CI) P OR (95%CI) P BMI—(kg/m 2)  18.5 ≤ BMI < 24 Ref  BMI ≥ 28 0.398 (0.274–0.578) 0.000 * 0.437 (0.298–0.641) 0.000 *  24 ≤ BMI < 28 0.824 (0.641–1.058) 0.129 * 0.866 (0.669–1.120) 0.273  BMI < 18.5 1.645 (1.072–2.523) 0.023 * 1.493 (0.967–2.307) 0.071 Number of blastocysts transferred  1 Ref  2 1.166 (0.925–1.471) 0.193  Number of high-quality blastocysts transferred 1.540 (1.306–1.815) 0.000 * 1.480 (1.251–1.751) 0.000 *  Endometrium thickness 1.088 (1.012–1.170) 0.022 * 1.101 (1.022–1.186) 0.011 * Type of infertility  Primary infertility Ref  Secondary infertility 1.007 (0.808–1.255) 0.953  Duration of infertility—(years) 0.934 (0.893–0.976) 0.002 * 0.955 (0.911–1.002) 0.060 Female age—(years)   < 35 Ref   ≥ 35 0.844 (0.581–1.227) 0.375 Male age —(years)   < 35 Ref Ref   ≥ 35 0.781 (0.606–1.007) 0.057 0.895 (0.683–1.173) 0.422 Number of transplant cycles 0.882 (0.775–1.003) 0.055 0.932 (0.817–1.063) 0.295 * P  < 0.05 was statistical significance Factors associated with live birth rate in frozen-thawed blastocyst transfer * P  < 0.05 was statistical significance 1704 patients were divided into a live birth group ( n  = 1237) and a nonlive birth group ( n  = 467) according to whether the live birth outcome was recorded. Multivariate logistic regression analysis showed factors for cumulative live birth outcomes included obesity (OR = 0.438, 95% CI 0.312–0.615), the number of high-quality blastocysts transfer (OR = 1.269, 95% CI 1.132–1.423), duration of infertility (OR = 0.927, 95% CI 0.885–0.972) and serum LH level on hCG trigger day (OR = 0.758, 95% CI 0.619–0.930) (Table  8 ). Table 8 Factors associated with cumulative live birth rate Factors Unadjusted Adjusted OR (95% CI) P OR (95% CI) P BMI—(kg/m 2)  18.5 ≤ BMI < 24 Ref Ref  BMI ≥ 28 0.424 (0.311–0.579) 0.000 * 0.438 (0.312–0.615) 0.000 *  24 ≤ BMI < 28 0.801 (0.625–1.026) 0.079 0.844 (0.647–1.100) 0.209  BMI < 18.5 1.411 (0.874–2.279) 0.159 1.361 (0.839–2.208) 0.212 Number of high-quality blastocysts transferred 1.270 (1.134–1.422) 0.000 * 1.269 (1.132–1.423) 0.000 * PCOS for Infertile causes only  No Ref  Yes 0.992 (0.799–1.232) 0.943 Infertile causes with male factors  No Ref  Yes 1.175 (0.949–1.454) 0.139 Type of infertility  Primary infertility Ref  Secondary infertility 1.057 (0.849–1.314) 0.622 Duration of infertility—(years) 0.905 (0.866–0.946) 0.000 * 0.927 (0.885–0.972) 0.002 * Female age—(years)   < 35 Ref   ≥ 35 0.724 (0.512–1.024) 0.068 0.939 (0.633–1.393) 0.754 Male age —(years)   < 35 Ref   ≥ 35 0.680 (0.532–0.869) 0.002 * 0.781 (0.588–1.037) 0.087 Type of fertilization  IVF Ref  ICSI 0.821 (0.624–1.079) 0.157  AFC 1.001 (0.988–1.014) 0.931  AMH —(ng/ml) 1.004 (0.976–1.032) 0.780 Basal hormones  LH—(IU/L) 1.010 (0.990–1.030) 0.317  FSH—(IU/L) 1.004 (0.946–1.066) 0.890  PRL—(ng/ml) 1.033 (0.990–1.017) 0.633  Progesterone—(ng/ml) 0.923 (0.620–1.372) 0.691  E 2 —(pg/ml) 1.005 (0.999–1.011) 0.126  Testosterone—(ng/ml) 1.304 (0.790–2.153) 0.299  Fasting plasma glucose (Median, IQR)—(mmol/L) 0.929 (0.776–1.112) 0.424  Fasting serum insulin (Median, IQR)—(mU/L) 0.993 (0.985–1.001) 0.099 1.000 (0.991–1.009) 0.967  Duration of Gn 0.971 (0.939–1.004) 0.083 1.000 (0.964–1.036) 0.982  Dose of Gn 1.000 (1.000–1.000) 0.000 * Hormones on hCG trigger day  LH—(IU/L) 0.835 (0.687–1.015) 0.070 0.758 (0.619–0.930) 0.008 *  E 2 —(pg/ml) 1.000 (1.000–1.000) 0.000 *  Progesterone—(ng/ml) 1.229 (0.892–1.694) 0.206 * P  < 0.05 was statistical significance Factors associated with cumulative live birth rate * P  < 0.05 was statistical significance

This retrospective analysis included patients diagnosed with polycystic ovary syndrome (PCOS) according to the revised Rotterdam criteria [ 23 ] who underwent the ultralong GnRH-a protocol at the Reproductive Center of the Second Affiliated Hospital of Wenzhou Medical University between June 2016 and June 2023. Inclusion criteria comprised completion of in vitro fertilization/intracytoplasmic sperm injection (IVF/ICSI) cycles with oocyte retrieval. The revised Rotterdam criteria required fulfillment of at least two of the following: (1) oligo/anovulation manifested as irregular menstrual cycles; (2) clinical and/or biochemical hyperandrogenism; (3) polycystic ovarian morphology confirmed by ultrasonography. Exclusion criteria were: (1) ultrasonographic uterine abnormalities (e.g., untreated endometrial polyps, leiomyomas, intrauterine adhesions, adenomyosis, or congenital malformations); (2) active malignancy or systemic chronic diseases (including autoimmune or hematologic disorders); (3) genetic disorders in either partner; (4) cycles involving preimplantation genetic testing. A cohort of 1704 women aged 20–42 years was stratified according to China-specific BMI criteria [ 2 ] into four groups: underweight (BMI 14.52–18.49 kg/m 2 ; n  = 125), normal weight (BMI 18.50–24.99 kg/m 2 ; n  = 845), overweight (BMI 25.00–27.99 kg/m 2 ; n  = 517), and obese (BMI 28.00–40.06 kg/m 2 ; n  = 217). Patients initiated pituitary down-regulation using a full dose (3.75 mg) of GnRH-a (Triptorelin, Ferring, Kiel, Germany) during the early follicular phase. When down-regulation (no ovarian cysts > 8 mm; estradiol(E 2 ) < 50 pg/L; endometrium thickness ≤ 5 mm) was confirmed 32–38 days later, follicle-stimulating hormone (FSH) administration started and continued with adjusted dosage of gonadotropin according to patients’ antral follicular count (AFC), BMI, age, Anti-Mullerian hormone (AMH), ovarian response and so on. Gonadotropin stimulation continued until follicles reached specific size by ultrasonography revealed. One dose of 4,000–10,000 IU Human Chorionic Gonadotropin (hCG, Livzon, Guangdong, China) was given between 20:00 and 20:30 on the day when Gonadotropin administration ceased according to follicular growth and E 2 level. Oocyte retrieval was performed 34-36 hours later using a transvaginal ultrasound approach [ 24 ]. The cleavage of embryonic development was assessed according to the Istanbul consensus [ 25 ]. All cleavage stage embryos were further cultured to the blastocyst stage. Additionally, Gardner’s grading criteria was used to score all resuscitated blastocysts [ 26 ]. Blastocysts rated ≥ 3AA, 3AB, 3BA, and 3BB were classified into high-quality blastocysts [ 27 ]. According to the embryo development, conditions and willingness of the patients, one or two available blastocysts were selected for transfer 5 days after oocyte retrieval or elective freezing of all blastocysts. Excess blastocysts were vitrified accordingly. Frozen-thawed blastocyst transfer (FBT) was performed in patients who had not achieved a live birth, provided at least one blastocyst was available. A total of 1638 patients underwent blastocyst transfer cycles: 1070 patients completed one cycle, 391 completed two cycles, 132 completed three cycles, 35 completed four cycles, and 10 completed five cycles. This resulted in 2438 blastocyst transfer procedures. However, 66 patients were excluded from transfer: 56 had no available blastocysts for transfer, and 10 declined transfer procedure due to personal reasons. The observation period concluded upon meeting either of the following endpoints: 1. At least one newborn was delivered; 2. Discontinuation of further blastocyst transfer attempts in a 2-year period after oocyte retrieval; 3. Absence of viable blastocysts for transfer (Fig.  1 ). Fig. 1 Flow chart. Patients initiated pituitary down-regulation during the early follicular phase. When down-regulation was confirmed 32–38 days later, controlled ovarian hyperstimulation (COH) was performed. Oocyte pick-up was performed until follicles reached specific size. Available blastocysts were selected for transfer 5 days after oocyte pick-up or elective freezing of all blastocysts. Frozen-thawed blastocyst transfer (FBT) was performed in patients without newborn when there was at least one blastocyst available. The observation period concluded upon meeting either of the following endpoints: 1. At least one newborn was delivered; 2. Discontinuation of further blastocyst transfer attempts in a 2-year period after oocyte retrieval; 3. Absence of viable blastocysts for transfer (Fig. 1) Flow chart. Patients initiated pituitary down-regulation during the early follicular phase. When down-regulation was confirmed 32–38 days later, controlled ovarian hyperstimulation (COH) was performed. Oocyte pick-up was performed until follicles reached specific size. Available blastocysts were selected for transfer 5 days after oocyte pick-up or elective freezing of all blastocysts. Frozen-thawed blastocyst transfer (FBT) was performed in patients without newborn when there was at least one blastocyst available. The observation period concluded upon meeting either of the following endpoints: 1. At least one newborn was delivered; 2. Discontinuation of further blastocyst transfer attempts in a 2-year period after oocyte retrieval; 3. Absence of viable blastocysts for transfer (Fig. 1) The study was approved by the Ethic Committee (Institutional Review Board) of the Second Affiliated Hospital and Yuying Children’s Hospital of Wenzhou Medical University. All participants provided informed consent for the procedures, and all information was handled confidentially. SPSS (version 29.0; IBM, Chicago) statistical software was used for data analysis. The measured variables were expressed as medians and interquartile ranges (IQRs); the nonparametric Kruskal‒Wallis H test was used to compare differences among the four groups, and Kruskal‒Wallis one-way analysis of variance test was used for multiple comparisons between groups. The chi-square test and Fisher’s exact test were used for categorical variables. Multivariate logistic regression analysis was performed based on the results of the univariate analysis. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated for the independent variables, and p  < 0.05 was considered to indicate statistical significance (Fig.  2 ). Fig. 2 Flow chart. A total of 1704 PCOS patients aged were treated with an ultralong GnRH—a protocol during a single oocyte retrieval cycle, followed by blastocyst transfer. Participants were stratified according to BMI criteria into four groups: underweight ( n  = 125), normal weight ( n  = 845), overweight ( n  = 517) and obese ( n  = 217). Statistical analysis was used to compare patient data (Fig. 2) Flow chart. A total of 1704 PCOS patients aged were treated with an ultralong GnRH—a protocol during a single oocyte retrieval cycle, followed by blastocyst transfer. Participants were stratified according to BMI criteria into four groups: underweight ( n  = 125), normal weight ( n  = 845), overweight ( n  = 517) and obese ( n  = 217). Statistical analysis was used to compare patient data (Fig. 2)

Body mass index (BMI) serves as a critical anthropometric indicator for assessing nutritional status, adiposity, and physical development. Notably, Chinese populations exhibit increased susceptibility to type 2 diabetes at lower BMI thresholds compared with White and Black populations [ 1 ]. Reflecting this distinctive risk profile, China has established specific BMI classification criteria for adults [ 2 ]. Elevated pre-pregnancy BMI significantly increases risks for gestational diabetes mellitus, hypertensive disorders, preterm delivery, fetal macrosomia and large-for-gestational-age infants [ 3 , 4 ]. The pathophysiological mechanisms linking pre-pregnancy obesity to adverse outcomes include insulin resistance, lipotoxicity chronic low-grade inflammation and oxidative stress. These factors collectively contribute to complications such as birth defects, miscarriage, and preeclampsia through disrupted placental function [ 5 ]. Obesity is associated with heightened activity of the mechanistic target of rapamycin and insulin growth factor 1 signaling pathways, which presumably contribute to fetal overgrowth [ 6 ]. It also disrupts endocrine homeostasis, manifesting as dysregulation of insulin dysregulation of insulin [ 7 ], androgens [ 8 , 9 ], and growth hormone [ 10 ]. Critically, insulin resistance and hyperandrogenism are pathognomonic features of polycystic ovary syndrome (PCOS) [ 11 ], whereas growth hormone supplementation improves oocyte quality in PCOS patients [ 12 ]. Compared with obese non-PCOS women, obese PCOS women exhibit elevated levels of tumor necrosis factor-α and malondialdehyde-biomarkers of lipid peroxidation [ 13 , 14 ] indicating chronic ovarian inflammation and oxidative stress [ 15 ]. This demonstrates PCOS-mediated amplification of obesity-dependent ovarian inflammatory and oxidative pathways [ 16 ], both known to impair ovarian function and oocyte quality. Consequently, subfertility, obesity, and PCOS exhibit intricate bidirectional relationships. Although BMI’s impact on assisted reproductive technology outcomes has been debated for decades [ 17 , 18 ], few studies specifically address BMI-live birth associations in PCOS populations. Existing research was limited by heterogeneous ovarian stimulation protocols [ 19 , 20 ]. As PCOS represents a leading cause of female infertility, the ultralong gonadotrophin-releasing hormone agonist (GnRH-a) protocol has gained widespread adoption in China with established feasibility [ 21 ] and is increasingly regarded as an optimal strategy for PCOS management [ 22 ]. These attributes make such patients ideal study candidates. Notably, no previous study has investigated BMI stratification in PCOS patients who completed the ultralong GnRH-a protocol with blastocyst transfer within a single oocyte retrieval cycle. Therefore, we stratified PCOS patients according to Chinese-specific BMI classification to establish evidence-based management strategies for high-BMI PCOS populations.

In this observational study, we investigated the embryological and clinical outcomes following blastocyst transfer in women with PCOS undergoing an ultralong GnRH-a protocol, stratified by BMI in four groups: underweight, normal weight, overweight and obese. Our results indicated that obesity was associated with a reduced number of retrieved oocytes, mature oocytes, 2-pronuclear zygotes and day-3 embryos. When considering fresh blastocyst transfer, overweight and obese status showed limited impact on live birth rate, whereas obesity was significantly associated with a decreased live birth rate in the subsequent FBT. Indeed, approximately 50% of the women with PCOS are overweight or obese [ 28 ], a condition that can precede PCOS onset [ 29 ] or present alongside PCOS in normal weight individuals. The symptoms of PCOS tend to be worsened by weight gain and improved by weight loss [ 29 ]. Given the strong association between PCOS with obesity and infertility, an increasing number of obese women are turning to assisted reproductive technology treatments like IVF [ 30 ]. However, IVF success rates remain lower in obese women compared with those with normal-weight or even underweight women. Contributing factors include poorer oocyte and embryo quality, impaired endometrial receptivity, and a higher risk of miscarriage [ 31 – 33 ]. Notably, it remains unclear whether the adverse IVF outcomes observed in obese women with PCOS stem primarily from obesity itself or from underlying PCOS-related pathologies. Therefore, investigating the specific impact of BMI on IVF outcomes in the PCOS population controlling for treatment protocols could provide valuable insights. Our results demonstrated that obese women with PCOS exhibited higher fasting plasma glucose and serum insulin levels, along with a reduced number of retrieved and mature oocytes. These findings align with those of Yuan et al., who also reported fewer retrieved oocytes, mature oocytes, and 2-pronuclear embryos in overweight and obese women compared with their normal-weight counterparts [ 34 ]. Furthermore, we observed a higher rate of fresh embryo transfer cancellations due to embryonic factors in obese patients, which may be linked to metabolic dysregulation associated with obesity and PCOS. Supporting this, Chang et al. identified significant enrichment in metabolic pathways related to both obesity and PCOS [ 35 ]. Previous studies have emphasized that balanced metabolism is critical for optimal oocyte development, as metabolic disturbances in the follicular microenvironment can impair follicular cell function and compromise oocyte competence [ 36 ]. Additionally, mitochondrial dysfunction has been proposed as a key contributor to poor oocyte quality in obese women [ 37 ]. Disrupted mitochondrial activity may adversely affect oocyte maturation, fertilization potential, and subsequent embryonic development [ 38 ]. The current consensus regarding the impact of BMI on pregnancy outcomes in PCOS patients remains inconclusive [ 39 ]. Our study demonstrated comparable live birth rates between obese and normal-weight PCOS patients undergoing fresh blastocyst transfer with GnRH-a protocols. However, we observed significantly reduced live birth rates in obese patients during frozen blastocyst transfer (FBT). Multivariate logistic regression further identified obesity as an independent negative predictor of cumulative live birth (OR = 0.438, 95% CI 0.312–0.615), consistent with previous studies reporting the detrimental effects of obesity on reproductive outcomes in PCOS [ 39 , 40 ]. Emerging evidence suggests bariatric surgery may serve as an effective intervention to improve fertility and pregnancy outcomes in obese women, particularly those with PCOS [ 41 ]. Recent work by José Bellver and colleagues has revealed that elevated BMI may compromise endometrial receptivity through significant transcriptomic alterations in the endometrium of obese women compared with their nonobese counterparts [ 42 ]. These molecular changes likely contribute to the observed reduction in implantation rates and increased miscarriage rates. Notably, in obese PCOS patients, researchers have identified dysregulation of endometrial genes involved in critical biological processes, some of which have been previously linked to implantation failure and unexplained infertility [ 43 ]. Future investigations should prioritize elucidating the mechanisms by which obesity-induced gene expression changes impair endometrial receptivity. Such research will be crucial for developing targeted strategies to improve IVF outcomes in obese women with PCOS. The primary strengths of our study include its large sample size and the standardization of blastocyst transfers, which helped partially control for embryo-related variables—an advantage over many previous investigations examining BMI’s impact on IVF outcomes in PCOS populations. Our use of BMI cutoffs of 25 and 28 kg/m 2 (rather than the WHO standards of 25 and 30 kg/m 2 ) better reflects anthropometric characteristics specific to Chinese populations. Furthermore, our separate analyses of fresh and frozen blastocyst transfer outcomes, combined with multivariate regression modeling, enabled identification of key prognostic factors. However, several limitations should be acknowledged. First, the retrospective study design carries inherent risks of selection and information biases. Second, we lacked data on potentially important confounding variables including paternal BMI, dietary habits, specific obesity phenotypes and genetic factors. The absence of these unmeasured variables in our analyses may have influenced the observed reproductive outcomes.

This study demonstrates the significant combined impact of PCOS and elevated BMI on IVF outcomes. Our findings strongly suggest that elevated BMI in PCOS patients adversely affects multiple aspects of reproductive performance, from initial ovarian response to final perinatal outcomes. These results highlight the clinical importance of implementing pre-IVF interventions, particularly weight management strategies, for obese PCOS patients. Such preparatory measures may enhance oocyte and embryo quality, improve live birth rates, and potentially reduce the risk of pregnancy and neonatal complications.