Association between pre-pregnancy body mass index and neonatal outcomes in women undergoing assisted reproductive technology: a retrospective study Authors: Huili Qian, Yunyun Wang, Caiying Bai, Lina Gao, Xiaoling Liu, Jinrui Cui, Xiaohui Liu, Xiaohua Ding Published: 2026 ABSTRACT ObjectiveART and pre-pregnancy BMI are strongly associated with neonatal outcomes, but there are fewer relevant studies. Our objective was to analyze the association between pre-pregnancy body mass index (BMI) and neonatal outcomes in women planning assisted reproductive technology (ART).MethodsThis was a retrospective study that included 994 ART singleton mothers who delivered in hospitals from July 2020 to June 2024. Mothers were categorized into 4 groups based on BMI (kg/m2): underweight (< 18.5), normal weight (18.5–23.9, reference), overweight (24.0–27.9), and obesity (≥ 28.0). Associations between BMI classification and neonatal outcomes were analyzed using univariate and multivariate logistic regression.ResultsAmong all mothers, pre-pregnancy underweight increased the adjusted odds ratio (aOR) for small for gestational age (SGA) to 3.97-fold compared to normal-weight mothers; overweight/obesity significantly increased the risk of macrosomia and large for gestational age (LGA), with aORs of 5.047 and 2.935-fold, respectively. In gestational weight gain (GWG) subgroup analyses, when GWG was adequate, pre-pregnancy underweight mothers were more likely to develop SGA than normal-weight mothers (aOR 4.649, 95% CI: 1.316–16.426), and overweight/obese mothers were not associated with adverse neonatal outcomes. When GWG was excessive, pre-pregnancy underweight mothers were 3.986 times more likely to be at risk of developing SGA than normal-weight mothers; pre-pregnancy overweight/obese mothers were 4.466 and 3.010 times more likely to develop macrosomia and LGA, respectively, compared to normal-weight mothers.ConclusionMaternal preconception underweight is associated with SGA regardless of whether GWG is adequate or excessive. For pre-pregnancy overweight or obese mothers, maintaining GWG in the target range can reduce the risk of adverse neonatal outcomes; if GWG is excessive, it is associated with macrosomia and LGA. Background According to the latest statistics from the European Society of Human Reproduction and Embryology [1], around 48 million couples worldwide suffer from infertility, and more than 9 million babies have been born through assisted reproductive technology (ART). In 2021, 1.96% of all babies were born through ART in Australia [2], compared with 2.3% in the USA [3], and a wider range of between 2% and 7% in Europe [1, 4]. With increasing social pressures and delayed childbearing age, more and more couples will choose ART, which is growing by 5–10% per year in many developed countries [1]. Currently, several studies [5–7] have shown that babies born to women undergoing ART are at higher risk of adverse outcomes, including preterm birth (PTB), low birth weight (LBW), and small for gestational age (SGA), compared to natural conception. Mothers’ pre-pregnancy body mass index (BMI) is strongly associated with ART pregnancy outcomes. According to the WHO report, the global prevalence of overweight and obesity reached 43% (2.5 billion people) in 2022 [8]. In China, about 1/2 of adults are overweight or obese, and by 2030, it is likely that 65.3% of the population will have a BMI greater than 24.0 kg/m2 (nearly 800 million) [9]. Overweight and obesity are common in the general population, but they are even more prevalent among women seeking ART assistance. The results of several studies have shown that among women planning ART, pre-pregnancy overweight/obese women are at a higher risk of pregnancy comorbidities and adverse pregnancy outcomes than normal-weight women [10–12]. In addition to this, this population has lower ART cycle success rates and is more likely to experience miscarriage [13–15]. Therefore, exploring the relationship between pre-pregnancy BMI and ART pregnancy outcomes is essential to improving success rates and neonatal health. The aim of this study was to investigate the relationship between pre-pregnancy BMI and neonatal outcomes in women undergoing ART, with a view to providing evidence on weight health management for women planning ART. Materials and methods Study population Mothers receiving ART who delivered in the hospital from July 2020 to June 2024 (n = 1879) were included, and after excluding twin births (n = 796), gestational weeks < 28w (n = 61), stillbirths (n = 5), missing information on neonatal measurements at birth (n = 5), and missing information on mother’s height, gestational weight gain (GWG), and prenatal weight (n = 18), a total of 994 women were included in this study. Maternity data collection On the day the mother reports to the hospital for delivery, the doctor collects her basic information and enters it into the hospital’s HIS system. Basic information: maternal age at delivery (years), gestational week (weeks), previous pregnancies (gravidity, parity, spontaneous abortions, embryonic arrest, cesarean sections, vaginal deliveries), ethnicity (Han-Chinese, ethnic minority), higher education (academic degree of college and above), place of residence (rural, urban), and reasons for ART. Height and early pregnancy weight are available in the outpatient medical record for women who have been documented in our hospital (required to be documented before 12 weeks of pregnancy). For women who are documented in other hospitals, self-reporting of height and pre-pregnancy weight is required. Nurses asked hospitalized mothers to use electronic scales in the unit to obtain prenatal weight. GWG was the difference between prenatal weight (as recorded upon admission for delivery) and early pregnancy weight (as documented in outpatient medical records prior to 12 weeks of gestation). The doctor asked the mother in detail about the presence of pregnancy complications and asked for the diagnostic certificates. If the woman had a routine antenatal visit at our hospital, the doctor checked the outpatient information. Comorbidities in pregnancy were collected: hypertensive disorders in pregnancy (including gestational hypertension, pre-eclampsia-eclampsia, chronic hypertension, and chronic hypertension with superimposed pre-eclampsia) [16], gestational diabetes mellitus (GDM), subclinical hypothyroidism, polyhydramnios, and oligohydramnios. It should be noted that although we collected data on pregestational diabetes mellitus (PGDM), due to its extremely low incidence (12 cases, 1.21%), it was not included as an independent covariate in the multivariate regression analysis to avoid model instability. Pre-pregnancy BMI and GWG In July 2022, the National Health Commission of the People’s Republic of China issued the Standard of Recommendation for Weight Gain During the Pregnancy Period [17], which applies to the range of weight gain of Chinese women with a single pregnancy. Based on the Chinese adult body mass index, four categories were classified: underweight (< 18.5 kg/m2), normal weight (18.5–23.9 kg/m2), overweight (24.0–27.9 kg/m2), and obesity (≥ 28.0 kg/m2). Obesity (n = 57, 5.73%) was grouped with overweight in the statistical analysis. For GWG classification, a weight gain of 11.0–16.0 kg was considered adequate when underweight was gained; when normal weight was gained, a weight gain of 8.0–14.0 kg was the adequate range; when overweight was gained, a weight gain of 7.0–11.0 kg was considered adequate GWG; and when obesity was gained, a weight gain of 5.0–9.0 kg was adequate GWG. Weight gain during pregnancy below adequate GWG was considered insufficient weight, and more than adequate GWG is considered excessive weight. Basic information and adverse outcomes of newborns at birth Immediately after the mother delivered the fetus, the midwife reported the time of birth and sex of the newborn, and then the umbilical cord was disconnected. The newborn is wiped clean and weighed without any clothing to obtain the birth weight (to the nearest 10 g), and the length, head circumference, shoulder width, and abdominal circumference (to the nearest 1 cm) are obtained using a soft tape measure. The above procedures will be completed and entered into the HIS system within 5 min after the delivery of the fetus. Adverse neonatal outcomes in this study were defined using the INTERGROWTH-21st international standard [18]: PTB (born at < 37 weeks gestational age), LBW (birth weight < 2,500 g), macrosomia (birth weight > 4000 g), SGA (birth weight below the 10th percentile of the average weight for the same gestational age), and large for gestational age (LGA, birth weight above the 90th percentile of the average weight for the same gestational age). The gestational age range for SGA and LGA in the INTERGROWTH-21st international standard definition is 33–42 weeks (with 969 cases [97.48%] falling within this range in this study). The respective numbers of SGA and LGA cases were 35 (3.61%) and 125 (12.90%). Statistical analysis Continuous variables reported means and standard deviations and were analyzed using ANOVA for between-group comparisons. Categorical variables reported frequencies and percentages, and between-group comparisons were made using chi-square tests or Fisher’s exact tests. Logistic regression analysis was used to explore the association between adverse neonatal outcomes and preconception BMI. In the regression analysis, pre-pregnancy normal weight was used as a control group to compare with underweight and overweight/obesity. The results were adjusted according to maternal age at delivery, parity, ethnicity, higher education, place of residence, hypertensive disorders in pregnancy, GDM, polyhydramnios, and oligohydramnios. GWG is strongly associated with neonatal outcomes [19–21]. Therefore, subgroup analyses were performed. Insufficient weight accounted for 6.44% (n = 64) (descriptive statistics), so we analyzed only adequate weight and excessive weight. And the results were also adjusted for maternal age at delivery, parity, ethnicity, higher education, place of residence, hypertensive disorders in pregnancy, GDM, polyhydramnios, and oligohydramnios. Statistical analyses were performed using IBM SPSS Statistics (V.26, IBM) and R (V.4.2.3), with P < 0.05 indicating statistically significant differences. Results Maternal and neonatal characteristics Overweight/obese mothers had smaller gestational weeks, lower education, and a higher incidence of gestational comorbidities (hypertensive disorders of pregnancy, GDM, and polyhydramnios) compared to pre-pregnancy underweight and normal weight mothers (Table 1). Among the GWG, the percentage of adequate weight was 38.13%, with the corresponding percentage of pre-pregnancy underweight mothers being 53.33%; 55.43% of the mothers were categorized as having excessive weight, with the corresponding percentage of both pre-pregnancy overweight and obese mothers exceeding 60%. None of the differences in neonatal anthropometrics were statistically significant. Among the adverse neonatal outcomes, the risk of PTB and LGA increased with an increasing pre-pregnancy BMI. Among them, the children of pre-pregnancy underweight mothers had a high proportion of LBW and SGA (11.11% and 11.24%, respectively), while pre-pregnancy overweight and obese mothers had a high proportion of macrosomia (7.05% and 10.53%) and LGA (21.55% and 23.64%). Table 1Maternal and neonatal characteristicsTotal(n = 994)Pre-pregnancy BMIF/χ2P-value <18.5(n = 90)18.5–23.9(n = 606)24.0-27.9(n = 241)≥ 28.0(n = 57)Mean (SD) /n (%)Maternal characteristicsAge at delivery, years31.97(3.62)30.88(3.76)32.00(3.47)32.13(3.80)32.75(3.92)3.8260.010aGestational age, weeks38.45(1.97)38.81(1.59)38.54(1.85)38.23(2.27)37.89(2.27)3.9570.008aGravidity, times1.72(1.04)1.68(1.10)1.69(1.00)1.85(1.15)1.63(0.94)1.5150.209aParity, times0.18(0.47)0.13(0.37)0.15(0.42)0.25(0.57)0.16(0.49)2.7360.042aSpontaneous abortions, n (%)0.8090.847bNo890(89.54)80(88.89)541(89.27)216(89.63)53(92.98)Yes104(10.46)10(11.11)65(10.73)25(10.37)4(7.02)Embryonic arrest,n (%)5.9610.114bNo917(92.25)83(92.22)568(93.73)214(88.80)52(91.23)Yes77(7.75)7(7.78)38(6.27)27(11.20)5(8.77)Cesarean sections,n (%)0.820cNo956(96.18)88(97.78)581(95.87)231(95.85)56(98.25)Yes38(3.82)2(2.22)25(4.13)10(4.15)1(1.75)Vaginal deliveries, n (%)5.1990.158bNo885(89.03)81(90.00)548(90.43)205(85.06)51(89.47)Yes109(10.97)9(10.00)58(9.57)36(14.94)6(10.53)Ethnicity, n (%)10.7980.013bHan-Chinese897(90.24)73(81.11)557(91.91)215(89.21)52(91.23)ethnic minority97(9.76)17(18.89)49(8.09)26(10.79)5(8.77)Higher education,n (%)14.7070.002bNo287(28.87)28(31.11)151(24.92)83(34.44)25(43.86)Yes707(71.13)62(68.89)455(75.08)158(65.56)32(56.14)Place of residence, n (%)20.517< 0.001brural341(34.31)40(44.44)175(28.88)101(41.91)25(43.86)urban653(65.69)50(55.56)431(71.12)140(58.09)32(56.14)Height, m1.62(0.05)1.63(0.05)1.62(0.05)1.61(0.05)1.62(0.04)4.0880.007aPre-pregnancy weight, kg58.74(9.04)45.52(4.48)55.99(4.98)66(5.24)78.08(7.72)670.237< 0.001aGWG, n (%)34.199< 0.001bSufficient379(38.13)48(53.33)242(39.93)77(31.95)12(21.05)Insufficient64(6.44)12(13.33)30(4.95)16(6.64)6(10.53)Excessive551(55.43)30(33.33)334(55.12)148(61.41)39(68.42)Reason for ART, n (%)36.4000.050bDiminished ovarian reserve23(2.31)1(1.11)13(2.15)6(2.49)3(5.26)PCOS48(4.83)2(2.22)23(3.80)17(7.05)6(10.53)Tubal obstruction199(20.02)23(25.56)121(19.97)47(19.50)8(14.04)Tubal factor159(16.00)9(10.00)91(15.02)48(19.92)11(19.30)Primary infertility53(5.33)4(4.44)31(5.12)16(6.64)2(3.51)Male factor103(10.36)16(17.78)63(10.40)20(8.30)4(7.02)Unexplained40(4.02)1(1.11)28(4.62)10(4.15)1(1.75)Other70(7.04)9(10.00)48(7.92)11(4.56)2(3.51)Missing299(30.08)25(27.78)188(31.02)66(27.39)20(35.09)Hypertensive disorders in pregnancy, n (%)40.703< 0.001bNo861(86.62)84(93.33)544(89.77)197(81.74)36(63.16)Yes133(13.38)6(6.67)62(10.23)44(18.26)21(36.84)GDM, n (%)28.160< 0.001 bNo890(89.54)85(94.44)560(92.41)202(83.82)43(75.44)Yes104(10.46)5(5.56)46(7.59)39(16.18)14(24.56)Subclinical hypothyroidism,n (%)0.951cNo939(94.47)86(95.56)571(94.22)227(94.19)55(96.49)Yes55(5.53)4(4.44)35(5.78)14(5.81)2(3.51)Polyhydramnios, n (%)< 0.001cNo970(97.59)90(100.00)595(98.18)236(97.93)49(85.96)Yes24(2.41)0(0.00)11(1.82)5(2.07)8(14.04)Oligohydramnios, n (%)0.023cNo978(98.39)87(96.67)601(99.17)233(96.68)57(100.00)Yes16(1.61)3(3.33)5(0.83)8(3.32)0(0.00)Neonatal characteristicsSex, n (%)1.2950.730bMale504(50.70)49(54.44)304(50.17)125(51.87)26(45.61)Female490(49.30)41(45.56)302(49.83)116(48.13)31(54.39)Newbirth weight, g3196.01(534.17)3120.56(474.53)3189.01(488.92)3234.07(620.77)3228.60(671.40)1.1120.343aLength, cm50.02(2.91)50.04(2.10)50.00(2.69)50.09(3.47)49.86(3.59)0.1130.952aHead circumference, cm34.01(2.06)34.02(1.83)33.92(1.97)34.21(2.26)34.16(2.48)1.2860.278aAbdominal circumference, cm31.38(2.47)31.21(2.26)31.33(2.29)31.55(2.82)31.46(3.07)0.6160.605aShoulder width, cm35.4(2.69)35.27(2.43)35.31(2.46)35.64(3.12)35.56(3.43)1.0120.387aPTB, n (%)8.9240.030bNo899(90.44)84(93.33)557(91.91)211(87.55)47(82.46)Yes95(9.56)6(6.67)49(8.09)30(12.45)10(17.54)LBW, n (%)4.9690.174bNo923(92.86)80(88.89)571(94.22)220(91.29)52(91.23)Yes71(7.14)10(11.11)35(5.78)21(8.71)5(8.77)Macrosomia,n (%)< 0.001cNo957(96.28)88(97.78)594(98.02)224(92.95)51(89.47)Yes37(3.72)2(2.22)12(1.98)17(7.05)6(10.53)SGAd, n (%)0.004cNo934(96.39)79(88.76)575(96.96)227(97.84)53(96.36)Yes35(3.61)10(11.24)18(3.04)5(2.16)2(3.64)LGAd, n (%)31.830< 0.001bNo844(87.10)85(95.51)535(90.22)182(78.45)42(76.36)Yes125(12.90)4(4.49)58(9.78)50(21.55)13(23.64)Abbreviations BMI body mass index, GWG gestational weight gain, ART assisted reproductive technology, GDM gestational diabetes mellitus, PTB preterm birth, LBW low birth weight, SGA small for gestational age, LGA large for gestational agea ANOVAb Chi-square testc Fisher’s exact testd n = 969 Table 1Maternal and neonatal characteristicsTotal(n = 994)Pre-pregnancy BMIF/χ2P-value <18.5(n = 90)18.5–23.9(n = 606)24.0-27.9(n = 241)≥ 28.0(n = 57)Mean (SD) /n (%)Maternal characteristicsAge at delivery, years31.97(3.62)30.88(3.76)32.00(3.47)32.13(3.80)32.75(3.92)3.8260.010aGestational age, weeks38.45(1.97)38.81(1.59)38.54(1.85)38.23(2.27)37.89(2.27)3.9570.008aGravidity, times1.72(1.04)1.68(1.10)1.69(1.00)1.85(1.15)1.63(0.94)1.5150.209aParity, times0.18(0.47)0.13(0.37)0.15(0.42)0.25(0.57)0.16(0.49)2.7360.042aSpontaneous abortions, n (%)0.8090.847bNo890(89.54)80(88.89)541(89.27)216(89.63)53(92.98)Yes104(10.46)10(11.11)65(10.73)25(10.37)4(7.02)Embryonic arrest,n (%)5.9610.114bNo917(92.25)83(92.22)568(93.73)214(88.80)52(91.23)Yes77(7.75)7(7.78)38(6.27)27(11.20)5(8.77)Cesarean sections,n (%)0.820cNo956(96.18)88(97.78)581(95.87)231(95.85)56(98.25)Yes38(3.82)2(2.22)25(4.13)10(4.15)1(1.75)Vaginal deliveries, n (%)5.1990.158bNo885(89.03)81(90.00)548(90.43)205(85.06)51(89.47)Yes109(10.97)9(10.00)58(9.57)36(14.94)6(10.53)Ethnicity, n (%)10.7980.013bHan-Chinese897(90.24)73(81.11)557(91.91)215(89.21)52(91.23)ethnic minority97(9.76)17(18.89)49(8.09)26(10.79)5(8.77)Higher education,n (%)14.7070.002bNo287(28.87)28(31.11)151(24.92)83(34.44)25(43.86)Yes707(71.13)62(68.89)455(75.08)158(65.56)32(56.14)Place of residence, n (%)20.517< 0.001brural341(34.31)40(44.44)175(28.88)101(41.91)25(43.86)urban653(65.69)50(55.56)431(71.12)140(58.09)32(56.14)Height, m1.62(0.05)1.63(0.05)1.62(0.05)1.61(0.05)1.62(0.04)4.0880.007aPre-pregnancy weight, kg58.74(9.04)45.52(4.48)55.99(4.98)66(5.24)78.08(7.72)670.237< 0.001aGWG, n (%)34.199< 0.001bSufficient379(38.13)48(53.33)242(39.93)77(31.95)12(21.05)Insufficient64(6.44)12(13.33)30(4.95)16(6.64)6(10.53)Excessive551(55.43)30(33.33)334(55.12)148(61.41)39(68.42)Reason for ART, n (%)36.4000.050bDiminished ovarian reserve23(2.31)1(1.11)13(2.15)6(2.49)3(5.26)PCOS48(4.83)2(2.22)23(3.80)17(7.05)6(10.53)Tubal obstruction199(20.02)23(25.56)121(19.97)47(19.50)8(14.04)Tubal factor159(16.00)9(10.00)91(15.02)48(19.92)11(19.30)Primary infertility53(5.33)4(4.44)31(5.12)16(6.64)2(3.51)Male factor103(10.36)16(17.78)63(10.40)20(8.30)4(7.02)Unexplained40(4.02)1(1.11)28(4.62)10(4.15)1(1.75)Other70(7.04)9(10.00)48(7.92)11(4.56)2(3.51)Missing299(30.08)25(27.78)188(31.02)66(27.39)20(35.09)Hypertensive disorders in pregnancy, n (%)40.703< 0.001bNo861(86.62)84(93.33)544(89.77)197(81.74)36(63.16)Yes133(13.38)6(6.67)62(10.23)44(18.26)21(36.84)GDM, n (%)28.160< 0.001 bNo890(89.54)85(94.44)560(92.41)202(83.82)43(75.44)Yes104(10.46)5(5.56)46(7.59)39(16.18)14(24.56)Subclinical hypothyroidism,n (%)0.951cNo939(94.47)86(95.56)571(94.22)227(94.19)55(96.49)Yes55(5.53)4(4.44)35(5.78)14(5.81)2(3.51)Polyhydramnios, n (%)< 0.001cNo970(97.59)90(100.00)595(98.18)236(97.93)49(85.96)Yes24(2.41)0(0.00)11(1.82)5(2.07)8(14.04)Oligohydramnios, n (%)0.023cNo978(98.39)87(96.67)601(99.17)233(96.68)57(100.00)Yes16(1.61)3(3.33)5(0.83)8(3.32)0(0.00)Neonatal characteristicsSex, n (%)1.2950.730bMale504(50.70)49(54.44)304(50.17)125(51.87)26(45.61)Female490(49.30)41(45.56)302(49.83)116(48.13)31(54.39)Newbirth weight, g3196.01(534.17)3120.56(474.53)3189.01(488.92)3234.07(620.77)3228.60(671.40)1.1120.343aLength, cm50.02(2.91)50.04(2.10)50.00(2.69)50.09(3.47)49.86(3.59)0.1130.952aHead circumference, cm34.01(2.06)34.02(1.83)33.92(1.97)34.21(2.26)34.16(2.48)1.2860.278aAbdominal circumference, cm31.38(2.47)31.21(2.26)31.33(2.29)31.55(2.82)31.46(3.07)0.6160.605aShoulder width, cm35.4(2.69)35.27(2.43)35.31(2.46)35.64(3.12)35.56(3.43)1.0120.387aPTB, n (%)8.9240.030bNo899(90.44)84(93.33)557(91.91)211(87.55)47(82.46)Yes95(9.56)6(6.67)49(8.09)30(12.45)10(17.54)LBW, n (%)4.9690.174bNo923(92.86)80(88.89)571(94.22)220(91.29)52(91.23)Yes71(7.14)10(11.11)35(5.78)21(8.71)5(8.77)Macrosomia,n (%)< 0.001cNo957(96.28)88(97.78)594(98.02)224(92.95)51(89.47)Yes37(3.72)2(2.22)12(1.98)17(7.05)6(10.53)SGAd, n (%)0.004cNo934(96.39)79(88.76)575(96.96)227(97.84)53(96.36)Yes35(3.61)10(11.24)18(3.04)5(2.16)2(3.64)LGAd, n (%)31.830< 0.001bNo844(87.10)85(95.51)535(90.22)182(78.45)42(76.36)Yes125(12.90)4(4.49)58(9.78)50(21.55)13(23.64)Abbreviations BMI body mass index, GWG gestational weight gain, ART assisted reproductive technology, GDM gestational diabetes mellitus, PTB preterm birth, LBW low birth weight, SGA small for gestational age, LGA large for gestational agea ANOVAb Chi-square testc Fisher’s exact testd n = 969 Maternal and neonatal characteristics Embryonic arrest, n (%) Cesarean sections, n (%) Higher education, n (%) Subclinical hypothyroidism, n (%) Macrosomia, n (%) Abbreviations BMI body mass index, GWG gestational weight gain, ART assisted reproductive technology, GDM gestational diabetes mellitus, PTB preterm birth, LBW low birth weight, SGA small for gestational age, LGA large for gestational age a ANOVA b Chi-square test c Fisher’s exact test d n = 969 Association between pre-pregnancy BMI and neonatal adverse outcomes Among all mothers (Fig. 1), pre-pregnancy underweight mothers were 3.97 times more likely to have SGA than normal-weight mothers. Overweight/obese mothers were more likely to have macrosomia (aOR 5.047; 95% CI: 2.402–10.607) and LGA (aOR 2.935; 95% CI: 1.949–4.421), compared to pre-pregnancy normal-weight mothers. After adjusting for variables, the association between pre-pregnancy overweight/obesity and PTB was not statistically significant. Fig. 1Association between pre-pregnancy BMI and neonatal adverse outcomes. The cOR, aOR values, and corresponding p-values for neonatal adverse outcomes in pre-pregnancy BMI classifications (normal weight, underweight, overweight/obesity). Pre-pregnancy normal weight was used as the reference group. SGA and LGA included 969 mothers. Adjusted for maternal age at delivery, parity, ethnicity, higher education, place of residence, hypertensive disorders of pregnancy, GDM, polyhydramnios, and oligohydramnios. cOR, crude odds ratio; aOR adjusted odds ratio; PTB, preterm birth; LBW, low birth weight; SGA, small for gestational age; LGA, large for gestational age Association between pre-pregnancy BMI and neonatal adverse outcomes. The cOR, aOR values, and corresponding p-values for neonatal adverse outcomes in pre-pregnancy BMI classifications (normal weight, underweight, overweight/obesity). Pre-pregnancy normal weight was used as the reference group. SGA and LGA included 969 mothers. Adjusted for maternal age at delivery, parity, ethnicity, higher education, place of residence, hypertensive disorders of pregnancy, GDM, polyhydramnios, and oligohydramnios. cOR, crude odds ratio; aOR adjusted odds ratio; PTB, preterm birth; LBW, low birth weight; SGA, small for gestational age; LGA, large for gestational age Association between pre-pregnancy BMI and neonatal adverse outcomes in GWG subgroups 3.3.1 Insufficient weight The proportion of mothers with insufficient weight was 6.44% (n = 64) (Table 2). Among them, 30 (46.88%) had normal weight. Intergroup comparisons revealed significant differences in LGA incidence (P = 0.025), with the highest rate observed in the overweight group (31.25%) and no LGA cases in the underweight group. No statistically significant differences were found in PTB, LBW, SGA, or macrosomia rates across groups (P > 0.05). Table 2Information on women with insufficient weight and neonatal outcomesTotal(n = 64)Pre-pregnancy BMIF/χ2P-value<18.5(n = 12)18.5–23.924.0-27.9≥ 28.0(n = 6)(n = 30)(n = 16)Mean (SD) /n (%)PTB0.620 bNo9(75.00%)21(70.00%)14(87.50%)4(66.67%)Yes3(25.00%)9(30.00%)2(12.50%)2(33.33%)LBW0.614 bNo9(75.00%)24(80.00%)14(87.50%)4(66.67%)Yes3(25.00%)6(20.00%)2(12.50%)2(33.33%)MacrosomiaNo0.531 bYes12(100.00%)30(100.00%)15(93.75%)6(100.00%)0(0.00%)0(0.00%)1(6.25%)0(0.00%)SGA aNo0.651 bYes10(90.91%)25(96.15%)14(87.50%)4(100.00%)1(9.09%)1(3.85%)2(12.50%)0(0.00%)LGA a0.025 bNo11(100.00%)25(96.15%)11(68.75%)3(75.00%)Yes0(0.00%)1(3.85%)5(31.25%)1(25.00%)Abbreviations BMI body mass index, PTB preterm birth, LBW low birth weight, SGA small for gestational age, LGA large for gestational agea n = 57b Fisher’s exact test Table 2Information on women with insufficient weight and neonatal outcomesTotal(n = 64)Pre-pregnancy BMIF/χ2P-value<18.5(n = 12)18.5–23.924.0-27.9≥ 28.0(n = 6)(n = 30)(n = 16)Mean (SD) /n (%)PTB0.620 bNo9(75.00%)21(70.00%)14(87.50%)4(66.67%)Yes3(25.00%)9(30.00%)2(12.50%)2(33.33%)LBW0.614 bNo9(75.00%)24(80.00%)14(87.50%)4(66.67%)Yes3(25.00%)6(20.00%)2(12.50%)2(33.33%)MacrosomiaNo0.531 bYes12(100.00%)30(100.00%)15(93.75%)6(100.00%)0(0.00%)0(0.00%)1(6.25%)0(0.00%)SGA aNo0.651 bYes10(90.91%)25(96.15%)14(87.50%)4(100.00%)1(9.09%)1(3.85%)2(12.50%)0(0.00%)LGA a0.025 bNo11(100.00%)25(96.15%)11(68.75%)3(75.00%)Yes0(0.00%)1(3.85%)5(31.25%)1(25.00%)Abbreviations BMI body mass index, PTB preterm birth, LBW low birth weight, SGA small for gestational age, LGA large for gestational agea n = 57b Fisher’s exact test Information on women with insufficient weight and neonatal outcomes Abbreviations BMI body mass index, PTB preterm birth, LBW low birth weight, SGA small for gestational age, LGA large for gestational age a n = 57 b Fisher’s exact test 3.3.2 Adequate weight Among women with adequate weight (n = 379) (Table 1), SGA was more likely to occur in underweight women compared to normal-weight women before pregnancy (P = 0.017) (Fig. 2). After adjusting for variables, the risk was slightly higher. Apart from this, for overweight/obese mothers whose GWG was maintained within the recommended range, the data showed that they were not associated with the risk of developing neonatal adverse outcomes. Fig. 2Association of pre-pregnancy BMI with neonatal adverse outcomes in the adequate weight subgroup. In women with adequate weight, the cOR values, aOR values, and corresponding p-values for neonatal adverse outcomes were examined across pre-pregnancy BMI classifications (normal weight, underweight, overweight/obesity). Pre-pregnancy normal weight was used as the reference group. SGA and LGA included 370 mothers. Adjusted for maternal age at delivery, parity, ethnicity, higher education, place of residence, hypertensive disorders of pregnancy, GDM, polyhydramnios, and oligohydramnios. cOR, crude odds ratio; aOR adjusted odds ratio; PTB, preterm birth; LBW, low birth weight; SGA, small for gestational age; LGA, large for gestational age Association of pre-pregnancy BMI with neonatal adverse outcomes in the adequate weight subgroup. In women with adequate weight, the cOR values, aOR values, and corresponding p-values for neonatal adverse outcomes were examined across pre-pregnancy BMI classifications (normal weight, underweight, overweight/obesity). Pre-pregnancy normal weight was used as the reference group. SGA and LGA included 370 mothers. Adjusted for maternal age at delivery, parity, ethnicity, higher education, place of residence, hypertensive disorders of pregnancy, GDM, polyhydramnios, and oligohydramnios. cOR, crude odds ratio; aOR adjusted odds ratio; PTB, preterm birth; LBW, low birth weight; SGA, small for gestational age; LGA, large for gestational age 3.3.3 Excessive weight Among the women who gained excessive weight (n = 551) (Table 1), pre-pregnancy underweight was associated with an increased risk of SGA, and pre-pregnancy overweight/obesity increased the risk of PTB, macrosomia, and LGA (Fig. 3). After adjusting for variables, pre-pregnancy overweight/obesity was not statistically associated with PTB (P = 0.057), while the others remained significant. Fig. 3Association of pre-pregnancy BMI with neonatal adverse outcomes in the excessive weight subgroup. In women with excessive weight, the cOR values, aOR values, and corresponding p-values for neonatal adverse outcomes were examined across pre-pregnancy BMI classifications (normal weight, underweight, overweight/obesity). Pre-pregnancy normal weight was used as the reference group. SGA and LGA included 542 mothers. Adjusted for maternal age at delivery, parity, ethnicity, higher education, place of residence, hypertensive disorders of pregnancy, GDM, polyhydramnios, and oligohydramnios. cOR, crude odds ratio; aOR adjusted odds ratio; PTB, preterm birth; LBW, low birth weight; SGA, small for gestational age; LGA, large for gestational age Association of pre-pregnancy BMI with neonatal adverse outcomes in the excessive weight subgroup. In women with excessive weight, the cOR values, aOR values, and corresponding p-values for neonatal adverse outcomes were examined across pre-pregnancy BMI classifications (normal weight, underweight, overweight/obesity). Pre-pregnancy normal weight was used as the reference group. SGA and LGA included 542 mothers. Adjusted for maternal age at delivery, parity, ethnicity, higher education, place of residence, hypertensive disorders of pregnancy, GDM, polyhydramnios, and oligohydramnios. cOR, crude odds ratio; aOR adjusted odds ratio; PTB, preterm birth; LBW, low birth weight; SGA, small for gestational age; LGA, large for gestational age Discussion This study retrospectively analyzed ART singleton mothers delivering at our hospital to investigate the association between different pre-pregnancy BMIs and adverse neonatal outcomes. The first finding revealed that pre-pregnancy underweight mothers were associated with a higher risk of SGA. Interestingly, this association remained statistically significant in the subgroups with adequate and excessive weight (although the 25% SGA incidence in the insufficient weight subgroup with insufficient weight was not statistically significant). Furthermore, underweight mothers with excessive weight showed a slightly lower risk of SGA compared to those with adequate weight. This result suggests that for pre-pregnancy underweight mothers, the association between pre-pregnancy BMI and adverse neonatal outcomes is greater than that of GWG [19]. Additionally, the risk of SGA decreases as GWG increases. The second finding indicates that pre-pregnancy overweight/obesity increases the risk of macrosomia and LGA. Analysis within the insufficient weight subgroup also showed a statistically significant association between pre-pregnancy overweight/obesity and LGA. However, maintaining pre-pregnancy BMI within the recommended range may prevent adverse neonatal outcomes. Maternal preconception nutritional status is a major determinant of embryonic development [22]. Between conception and implantation, the embryo is highly sensitive to nutritive, metabolic, and inflammatory substances in the tubal fluid, and these change in response to maternal nutritional changes [23]. Our study is similar to existing findings [24–26] that preconception underweight is strongly associated with SGA. A large meta-analysis [19], which included data from 196,670 women in 25 cohort studies from Europe and North America, noted that when pre-pregnancy BMI was < 18.0 kg/m2, the risk of adverse pregnancy outcomes was lowest for mothers with an increase in GWG of 26.0–27.9 kg. Although this GWG value is well above the recommended range, it suggests that pre-pregnancy underweight mothers may need a higher energy intake during pregnancy to support fetal health. This is consistent with the physiological mechanism of prioritizing the supply of nutrients to the fetus by lowering the mother’s basal metabolic rate during early pregnancy [27]. This is supported by our findings that excessive maternal GWG in pre-pregnancy underweight mothers is still associated with the risk of developing SGA. And the risk of developing SGA decreased as GWG increased. A prospective study of the association between pregnancy weight trajectory and SGA from rural Gambia [24] demonstrated that, among pre-pregnancy underweight mothers, a GWG of more than 4.5 kg in the third to seventh month of pregnancy was able to reduce the risk of developing SGA. In addition, the results of a meta-analysis [20] showed a 21% reduction in the risk of SGA in pre-pregnancy underweight women with high gestational weight gain (≥ 1.0 SD), compared to normal weight moderate weight gain (-1.0-0.9 SD) women. Although LBW among pre-pregnancy underweight women was not statistically significant, this group did have a higher observed incidence of LBW (11.11%) than the other three groups. The results of a retrospective cohort study [10] with data from 494,097 ART cases from the United States showed that pre-pregnancy underweight increased the risk of LBW by 39% in singleton pregnancies. Therefore, for pre-pregnancy underweight women, it is recommended to actively adjust their nutritional status before undergoing ART and to appropriately relax the target range of GWG during pregnancy. It is logical that our study found that pre-pregnancy overweight/obesity in women undergoing ART was strongly associated with macrosomia and LGA. Overweight/obese mothers are often insulin-resistant, resulting in high levels of maternal blood glucose, which can easily pass through the placenta to support fetal growth. In addition, the increased accumulation of lipids in the placenta of obese mothers compared to normal-weight mothers may affect the normal placental lipid supply to the fetus, causing macrosomia and LGA [28, 29]. A study by Kluge et al. [11], which included 58,187 singleton deliveries by in-vitro fertilization (IVF) from 2002 to 2020, showed that as the BMI of overweight/obese women increased, the incidence of PTB and the risk of SGA and macrosomia increased with it. The results of a study by Perumal et al. [21] were similar. However, in our study, we found that the risk of PTB in pre-pregnancy overweight/obese women was 1.355 times higher than that in normal-weight women, but there was no statistical difference (P = 0.215). The possible reason for this is that nearly half of the pregnant women with hypertensive disorders of pregnancy and GDM in our study were overweight/obese mothers, so the association between BMI and PTB was not significant after adjusting for these two variables. Meanwhile, weight interventions for overweight/obese mothers prior to IVF have been proposed with a view to reducing adverse pregnancy outcomes, but the results have been unsatisfactory. In a prospective Nordic study [30], 317 infertile women were recruited, allocated covertly (ratio 1:1), and blinded to embryologists and statisticians. The control group underwent IVF only, and the intervention group underwent a weight loss intervention prior to IVF consisting of a low-calorie liquid formula diet for 12 weeks, followed by the introduction of a normal diet for 3–5 weeks. Their results showed that neonates in the intervention group had a 98.6 g decrease in birth weight compared to the control group (P = 0.46), and there was no statistically significant difference in PTB and SGA between the two groups. Similarly, several studies [13–15, 31] examined the effect of pre-IVF bariatric surgery or dietary exercise interventions on the rate of live births in overweight/obese mothers, but the results were equally unpromising. GWG control may be a more beneficial option for overweight/obese women. The results of subgroup analyses in this study showed no statistically significant difference in adverse neonatal outcomes among overweight/obese mothers in terms of GWG sufficiency. The results of a case-control study from Finland [32] showed that GWG adequacy was beneficial in lowering birth weight SD scores and preventing LGA, regardless of the presence of GDM and pre-pregnancy obesity. In summary, for pre-pregnancy overweight/obese mothers, we believe that the significance of weight control during pregnancy is enormous. Maintaining GWG within the recommended range can somewhat reduce the risk of adverse neonatal outcomes. This study employed BMI classification criteria and corresponding GWG values more suited to Chinese women, potentially reflecting the actual circumstances of Chinese women undergoing ART more accurately, thereby yielding more persuasive results. Moreover, our findings support existing research [11, 21, 24–26] and yielded intriguing results in subgroup analyses. Nevertheless, the transition from epidemiological association to clinical decision-making (such as termination of pregnancy) involves a complex evaluation process that must also account for the individual circumstances of the pregnant woman (underlying medical conditions, complications, fetal status, etc.). Our results merely expand the knowledge base for weight management during preconception and pregnancy among women undergoing ART. At the same time, there are some limitations to this study. First, some of the mothers who were not recorded in our hospital self-reported their height and pre-pregnancy weight. Although common in similar studies, this portion of the data may be informationally biased. Secondly, the sample size was small. In this case, the number of obese mothers was small (n = 57, 5.73%), which we considered and classified as overweight. Similarly, the number of underweight gainers was lower when stratified by GWG (n = 64, 6.44%), but we do not believe that underweight gainers could be merged with adequate weight gainers. Therefore, this section merely describes the data. Given the limited sample size, the results should be interpreted and extrapolated with caution. Thirdly, this study did not systematically collect data on neonatal admissions to the NICU, which constitutes a limitation of the present work. Although we observed a consistent association between pre-pregnancy underweight and SGA, we were unable to further assess the clinical severity or healthcare resource utilization among these SGA infants. Fourthly, although we adjusted for common pregnancy-related complications, PGDM was excluded from the correction as an independent covariate due to its low prevalence (1.21%). PGDM exerts a potent effect on fetal growth, and should its distribution be uneven across groups, it may introduce confounding bias to the results. Another limitation, due to the retrospective nature of the study, was the absence of data on IVF procedures (number of heterologous IVF cases, use of donor gametes), transferred embryos (status, transfer timing, etc.), as well as information on husbands (age, education, alcohol consumption, smoking, etc.), which may affect maternal GWG and neonatal outcomes, was missing at the time of data collection [33, 34]. Therefore, caution should be exercised when inferring the outcome of maternal preconception BMI and neonatal outcome when undergoing ART. Conclusion Among women on ART, the risk of SGA in pre-pregnancy underweight women decreased with increasing GWG. Maintaining pre-pregnancy overweight or obese maternal GWG in the target range is beneficial in reducing the risk of adverse neonatal outcomes. In the case of excess GWG, there is a strong association with macrosomia and LGA. Our findings provide evidence for prenatal and maternal health care for women planning ART.