Baseline participant characteristics
A total of 6,821 twin pregnancies were initially included in the study. After excluding cases with fetal defects, missing information, and preexisting chronic hypertension, a total of 6,307 twin pregnancies remained, including 4,750 full-term infants and 7,864 premature infants. Among these pregnancies, 1,013 cases (16.1%) were diagnosed with HDP. The selection process is illustrated in Fig. 1.
Mothers with HDP were typically older, often nulliparous, and frequently used assisted reproductive techniques. The group exhibited increased rates of both preexisting and gestational diabetes, cesarean deliveries, and placenta previa, while showing a lower incidence of oligohydramnios. Twin infants from HDP mothers had a greater difference in birth weight discordance, lower GA, lower BW, lower BW Z scores, and a higher incidence of fetal growth restriction (Table 1). After PSM and OW, all variable SMDs were < 0.1, indicating a balanced comparison between the two groups. This demonstrates effective control of confounding factors, with OW achieving better balance than PSM (Figure S1). Therefore, the following results will primarily be explained based on the OW.
Table 1
Baseline characteristics of twin pregnant mothers with or without HDP
Characteristic | Non-HDP (n = 5294) | HDP (n = 1013) | p |
Maternal characteristics | | | |
Advanced maternal age | 1137 (21.5%) | 276 (27.3%) | < 0.001 |
Nulliparous | 3112 (58.7%) | 646 (63.7%) | 0.003 |
Minority nationality | 228 (4.4%) | 44 (4.5%) | 0.865 |
Assisted reproduction (IVF-ET or artificial fertilization) | 3032 (57.5%) | 624 (62.0%) | 0.008 |
Dichorionic | 3915 (75.6%) | 744 (75.3%) | 0.872 |
Spontaneous abortion | 591 (11.2%) | 107 (10.6%) | 0.585 |
Artificial abortion | 1378 (26.1%) | 264 (26.1%) | 1 |
Gestational diabetes | 1112 (21.0%) | 256 (25.3%) | 0.003 |
Pre-gestational diabetes | 78 (1.5%) | 27 (2.7%) | 0.01 |
Infection in mid-to-late pregnancy | 315 (6.0%) | 65 (6.4%) | 0.565 |
Hypothyroidism | 460 (8.7%) | 99 (9.8%) | 0.277 |
Maternal anemia | 1541 (29.1%) | 297 (29.3%) | 0.91 |
Cesarean section | 4792 (90.8%) | 967 (95.8%) | < 0.001 |
Placental abnormality* | 439 (8.4%) | 91 (9.1%) | 0.458 |
Placenta previa | 126 (2.4%) | 23 (2.3%) | 0.91 |
Placenta accreta | 71 (1.4%) | 24 (2.4%) | 0.023 |
Velamentous placenta | 126 (2.4%) | 25 (2.5%) | 0.823 |
Battledore placenta | 94 (1.8%) | 15 (1.5%) | 0.599 |
Abnormal amniotic fluid* | 427 (8.1%) | 81 (8.0%) | 1 |
Amniotic fluid turbidity | 166 (3.1%) | 42 (4.2%) | 0.102 |
Bloody amniotic fluid | 37 (0.7%) | 5 (0.5%) | 0.672 |
Oligohydramnios | 212 (4.0%) | 23 (2.3%) | 0.006 |
Polyhydramnios | 41 (0.8%) | 14 (1.4%) | 0.064 |
Nuchal cord* | 834 (15.8%) | 161 (15.9%) | 0.925 |
Neonatal characteristics | Non-HDP (n = 10588) | HDP (n = 2026) | |
BWDT# | 10.53 ± 8.70 | 12.76 ± 10.63 | < 0.001 |
Birth weight | 2351.08 ± 497.59 | 2278.96 ± 461.49 | < 0.001 |
Z-score | -0.90 ± 0.95 | -0.83 ± 0.95 | 0.003 |
Gestational age | 35.39 ± 2.39 | 35.19 ± 1.84 | < 0.001 |
Birthweight percentile | 0.31 ± 0.22 | 0.30 ± 0.22 | 0.020 |
Data are presented as mean ± SD or n (%). |
BWDT, discordant birth weight in twins; IVF-ET, in vitro fertilization-embryo transfer; HDP, hypertensive disorders of pregnancy; |
# The counting unit of BWDT is for twins; |
* Occurrence of any one fetus. |
Association between HDP and perinatal characteristics of twins
In investigating the relationship between HDP and BW and GA, we controlled for potential confounding factors such as placental abnormalities, amniotic fluid abnormalities, and nuchal cord abnormalities. As showed in Table 2, our study findings showed that HDP increased the likelihood of preterm birth (2.38 [2.13–2.66]crude, 2.67 [2.37–3.02]adjusted, 2.59 [2.23–3.01]PSM, and 2.68 [2.27–3.15]OW) in twin pregnancies, particularly for early preterm and late preterm infants, with OW-OR (95% CI) of 1.29 [1.00-1.67] and 2.31 [2.00-2.68], respectively. Interestingly, HDP reduced the probability of extremely preterm births (GA <28 weeks) (0.32 [0.12–0.77]OW). HDP also increased the occurrence of LBW (1500–2500 grams) (1.54 [1.34–1.78]OW) and the incidence of SGA (1.18 [1.00-1.42]OW). Furthermore, there is a clear correlation between the incidence of HDP and the degree of birth weight discordance in twins (Figure S2). HDP also increased the rates of BWDT (> 15%, > 20%, and > 25%) simultaneously (1.51 [1.20–1.88], 1.71 [1.31–2.23] and 2.05 [1.46–2.91], respectively) (Model 4).
Table 2
The relationship between gestational age and birth weight of newborns from mothers with and without HDP.
Characteristic | No. (%) | | OR (95% CIs) |
Non-HDP (n = 10588) | HDP (n = 2026) | | Model 1 Crude | Model 2 Adjusted | Model 3 PS-Matched | Model 4 Overlap weighting |
BWDT | | | | | | | |
BWDT > 15 | 2582 (24.4%) | 666 (32.9%) | | 1.52 [1.31–1.75] | 1.50 [1.29–1.75] | 1.50 [1.22–1.84] | 1.51 [1.20–1.88] |
BWDT > 20 | 1464 (13.8%) | 436 (21.5%) | | 1.71 [1.44–2.02] | 1.70 [1.42–2.03] | 1.69 [1.33–2.17] | 1.71 [1.31–2.23] |
BWDT > 25 | 742 (7.0%) | 273 (13.5%) | | 2.08 [1.68–2.55] | 2.06 [1.65–2.56] | 2.22 [1.62–3.09] | 2.05 [1.46–2.91] |
Gestational age | | | | | | | |
Preterm (< 37 weeks) | 5978 (56.5%) | 1335 (65.9%) | | 2.38 [2.13–2.66] | 2.67 [2.37–3.02] | 2.59 [2.23–3.01] | 2.68 [2.27–3.15] |
Late preterm (34+ 1~36+ 6 weeks) | 5311 (50.2%) | 1217 (60.1%) | | 2.31 [2.10–2.55] | 2.29 [2.07–2.55] | 2.19 [1.91–2.51] | 2.31[2.00-2.68] |
Early preterm (32+ 1~33+ 6 weeks) | 485 (4.6%) | 93 (4.6%) | | 1.12 [0.95–1.32] | 1.28 [1.08–1.52] | 1.36 [1.08–1.73] | 1.29 [1.00-1.67] |
Very preterm (28+ 1~31+ 6 weeks) | 182 (1.7%) | 25 (1.2%) | | 0.65 [0.51–0.82] | 0.74 [0.57–0.95] | 0.77 [0.56–1.06] | 0.74 [0.52–1.05] |
Extremely preterm (< 28 weeks) | 5978 (56.5%) | 1335 (65.9%) | | 0.25 [0.11–0.47] | 0.33 [0.15–0.65] | 0.27 [0.10–0.60] | 0.32 [0.12–0.77] |
Birth weight | | | | | | | |
< 2500 g | 5978 (56.5%) | 1335 (65.9%) | | 1.49 [1.35–1.65] | 1.63 [1.46–1.81] | 1.80 [1.57–2.06] | 1.62 [1.40–1.89] |
LBW (1500 ~ 2500 g) | 5311 (50.2%) | 1217 (60.1%) | | 1.49 [1.36–1.65] | 1.55 [1.39–1.71] | 1.66 [1.45–1.89] | 1.54 [1.34–1.78] |
VLBW (1000 ~ 1500 g) | 485 (4.6%) | 93 (4.6%) | | 1.00 [0.79–1.25] | 1.16 [0.91–1.47] | 1.28 [0.92–1.78] | 1.16 [0.82–1.65] |
ELBW (< 1000 g) | 182 (1.7%) | 25 (1.2%) | | 0.71 [0.46–1.07] | 1.00 [0.62–1.53] | 1.26 [0.68–2.37] | 0.98 [0.51–1.85] |
Birthweight percentile | | | | | | | |
SGA (< 10th percentile) | 1980 (18.7%) | 431 (21.3%) | | 1.17 [1.04–1.32] | 1.18 [1.04–1.34] | 1.40 [1.18–1.65] | 1.18 [1.00-1.42] |
< 5th percentile | 880 (8.3%) | 221 (10.9%) | | 1.21 [1.18–1.24] | 1.22 [1.19–1.25] | 0.96 [0.78–1.18] | 1.00 [0.79–1.26] |
BWDT, discordant birth weight in twins; CIs, confidence intervals; ELBW, extremely low birth weight; HDP, hypertensive disorders of pregnancy; LBW, low birth weight; OR, odds ratio; SGA, small for gestational age; VLBW, very low birth weight. |
In stratified analysis, we found that both gestational hypertension or preeclampsia increased the risk of late preterm birth and BWDT > 15%, while reducing the risk of extremely preterm birth (Figure S3). Sensitivity analysis based on chorionicity or sex revealed that HDP increased the rates of late preterm birth and BWDT > 15% in all groups. Moreover, HDP increased the occurrence of fetal growth restriction at the 5th percentile in male infants or in dichorionic infants. Additionally, the occurrence of BWDT > 20% or 25% in dichorionic twin pregnancies is higher than that in monochorionic twin pregnancies (Figure S4-5).
Association between HDP and neonatal complications
After further controlling for confounding factors, including Z-score, gestational age, and SGA, we found that HDP can increase the crude or adjusted risks of 1-minute Apgar score <7, NICU admission, oxygen requirement, and NEC in twins, with crude OR (95%CI) of 1.22 [1.00-1.48], 1.75 [1.59–1.93], 1.28 [1.16–1.42], and 2.31 [2.00-2.68], respectively. After using OW method (Model 4), twins born to mothers with HDP still had a higher risk of oxygen requirement, 1-minute Apgar score <7, and NICU admission, with OW-OR (95%CI) of 1.29 [1.07–1.56], 1.43 [1.04–1.99], and 1.58 [1.32–1.90], respectively. Additionally, HDP was found to decrease the incidence of hypoalbuminemia (0.69 [0.53–0.91] OW). However, there were no significant changes in other complications, such as respiratory diseases, ROP, IVH, PVL, feeding intolerance, neonatal anemia, and neonatal hypoalbuminemia (Table 3).
Table 3
Complication incidence in newborns of mothers with or without HDP
Characteristic | No. (%) | | OR (95% CIs) |
Non-HDP (n = 10588) | HDP (n = 2026) | | Model 1 Crude | Model 2 Adjusted | Model 3 PS-Matched | Model 4 Overlap weighting |
NICU admission | 5040 (47.6) | 1243 (61.4) | | 1.75 [1.59–1.93] | 1.57 [1.38–1.78] | 1.38 [1.17–1.63] | 1.58 [1.32–1.90] |
1-min Apgar < 7 | 562 (5.3) | 130 (6.4) | | 1.22 [1.00-1.48] | 1.45 [1.16–1.80] | 1.20 [0.90–1.59] | 1.43 [1.04–1.99] |
5-min Apgar < 7 | 108 (1.0) | 26 (1.3) | | 1.26 [0.80–1.91] | 1.82 [1.11–2.91] | 1.33 [0.71–2.52] | 1.90 [0.93–4.04] |
Oxygen therapy | 2877 (27.2) | 654 (32.4) | | 1.28 [1.16–1.42] | 1.29 [1.13–1.46] | 1.27 [1.07–1.51] | 1.29 [1.07–1.56] |
Invasive ventilation | 528 (5.0) | 77 (3.8) | | 0.75 [0.59–0.96] | 1.20 [0.89–1.61] | 1.21 [0.82–1.78] | 1.17 [0.77–1.79] |
Noninvasive ventilation | 1999 (18.9) | 436 (21.6) | | 1.18 [1.05–1.33] | 1.29 [1.12–1.49] | 1.28 [1.05–1.56] | 1.28 [1.03–1.58] |
Conventional oxygen therapy | 1862 (17.6) | 449 (22.2) | | 1.33 [1.19–1.50] | 1.42 [1.24–1.61] | 1.36 [1.15–1.62] | 1.40 [1.16–1.69] |
Composite respiratory morbidity | 1509 (14.3) | 295 (14.6) | | 1.03 [0.90–1.17] | 1.13 [0.95–1.35] | 1.15 [0.91–1.45] | 1.11 [0.86–1.43] |
NRDS | 1456 (13.8) | 290 (14.3) | | 1.05 [0.91–1.20] | 1.18 [0.99–1.41] | 1.19 [0.95–1.51] | 1.15 [0.89–1.49] |
BPD | 278 (2.6) | 25 (1.2) | | 0.46 [0.30–0.69] | 0.64 [0.36–1.09] | 0.65 [0.32–1.30] | 0.57 [0.27–1.19] |
Pulmonary hemorrhage | 72 (0.7) | 11 (0.5) | | 0.80 [0.40–1.44] | 1.30 [0.58–2.65] | 1.45 [0.54-4.00] | 1.29 [0.43–3.86] |
Pneumothorax | 30 (0.4) | 5 (0.3) | | 0.80 [0.27–1.90] | 0.91 [0.31–2.19] | 1.96 [0.43–1.10] | 0.90 [0.22–3.55] |
PPHN | 40 (0.4) | 3 (0.1) | | 0.39 [0.09–1.08] | 0.42 [0.09–1.28] | 0.35 [0.06–1.52] | 0.43 [0.07–2.03] |
NEC (any stage) | 159 (1.5) | 43 (2.1) | | 1.42 [1.00-1.98] | 1.56 [1.07–2.25] | 1.44 [0.88–2.37] | 1.53 [0.88–2.71] |
NEC (II-III stage) | 75 (0.7) | 18 (0.9) | | 1.26 [0.73–2.06] | 1.53 [0.85–2.60] | 1.79 [0.81–4.17] | 1.39 [0.61–3.23] |
Treated PDA | 666 (13.3) | 161 (13.1) | | 0.98 [0.82–1.18] | 1.18 [0.95–1.45] | 1.13 [0.86–1.50] | 1.17 [0.86–1.60] |
ROP (any stage) | 365 (6.8) | 60 (6.0) | | 0.88 [0.66–1.16] | 1.13 [0.80–1.58] | 1.24 [0.79–1.96] | 1.14 [0.70–1.87] |
ROP (II-V stage) | 62 (1.1) | 4 (0.4) | | 0.35 [0.10–0.84] | 0.72 [0.17–2.30] | 0.41 [0.01–1.93] | 0.73 [0.12–3.83] |
PVL | 129 (3.1) | 25 (2.5) | | 0.79 [0.50–1.19] | 0.96 [0.60–1.49] | 1.09 [0.60-2.00] | 1.00 [0.52–1.92] |
IVH (any grade) | 496 (11.6) | 120 (11.6) | | 0.99 [0.80–1.23] | 1.26 [0.99–1.60] | 1.15 [0.84–1.59] | 1.19 [0.83–1.70] |
IVH (III-IV grade) | 174 (4.1) | 34 (3.3) | | 0.80 [0.54–1.14] | 1.28 [0.83–1.92] | 1.14 [0.65–1.98] | 1.12 [0.61–2.06] |
Neonatal anemia | 2099 (41.7) | 433 (35.0) | | 0.75 [0.66–0.86] | 0.91 [0.78–1.06] | 1.12 [0.61–2.06] | 0.91 [0.80–1.05] |
Severe anemia | 1165 (23.1) | 224 (18.1) | | 0.73 [0.63–0.86] | 0.96 [0.79–1.16] | 0.86 [0.71–1.06] | 0.91 [0.73–1.13] |
Sepsis | 324 (6.5) | 66 (5.4) | | 0.82 [0.62–1.07] | 1.07 [0.80–1.43] | 0.84 [0.58–1.21] | 1.05 [0.69–1.61] |
Neonatal hypothyroidism | 131 (2.6) | 23 (1.9) | | 0.71 [0.44–1.08] | 1.03 [0.61–1.67] | 1.03 [0.52–2.08] | 1.10 [0.52–2.32] |
Hypoproteinemia | 936 (18.7) | 169 (13.7) | | 0.69 [0.58–0.82] | 0.70 [0.57–0.85] | 0.67 [0.52–0.85] | 0.69 [0.53–0.91] |
Feeding intolerance | 750 (7.3) | 158 (8.0) | | 1.11 [0.93–1.33] | 1.20 [0.97–1.46] | 1.13 [0.87–1.47] | 1.19 [0.89–1.60] |
BPD, bronchopulmonary dysplasia; CIs, confidence intervals; HDP, hypertensive disorders of pregnancy; IVH, intraventricular hemorrhage; NEC, necrotizing enterocolitis; NICU, neonatal intensive care unit; NRDS, neonatal respiratory distress syndrome; OR, odds ratio; PDA, patent ductus arteriosus; PPHN, persistent pulmonary hypertension of newborn; PVL, periventricular leukomalacia; ROP, retinopathy of prematurity. |
Stratified analysis by GA revealed that HDP increased the NICU admission rate among full-term infants (Table S2). HDP also increased the risk of preterm infants having low Apgar scores at 1 and 5 minutes, requiring NICU admission, and needing conventional oxygen therapy, while reducing the risk of hypoalbuminemia (Fig. 2). Among late preterm infants, HDP increased the NICU admission rate and decreased the risk of hypoalbuminemia. However, HDP did not have a significant impact on early preterm infants and very preterm infants (Table S2). Furthermore, gestational hypertension decreased the occurrence of hypoalbuminemia, while preeclampsia increased the risk of 1-minute Apgar score <7, NICU admission, composite respiratory morbidity, NRDS, and the need for oxygen therapy (Figure S6). In sensitivity analysis, regardless of the chorionicity, sex, birth order, HDP increase the rate of hospitalization and the need for conventional oxygen in twins (Figure S7-9). HDP also increases the risk of a 1-minute Apgar score <7 in male or dichorionic infants. Additionally, HDP reduces the occurrence of hypoalbuminemia in female, later-born or monochorionic infants.