To investigate the occurrence of live-born twins with birth weight-discordance and its relationship to adverse birth outcomes.

A retrospective analysis was performed on 4 011 pairs of live-born twins in the Third Affiliated Hospital of Guangzhou Medical University from January 2011 to December 2020. Based on the birth-weight discordance (∆BW, ∆BW=(birth-weight_big-birth-weight_small)/birth-weight_big×100%)), twins were divided into 4 groups, including ∆BW≤15%,>15%-20%,>20%-25%, and>25%. The differences in maternal and neonatal outcomes among 4 groups were explored. Then the correlation between ∆BW and neonatal adverse outcomes were explored.

The ΔBW was 9 (4, 16)% and males were accounted for 53.8% (4 315 cases) of 4 011 pairs of twins. The gestational age was (35.3±2.7) weeks at birth. There were 2 908 pairs (72.5%) of twins with ΔBW≤15%, 481 pairs (12.0%) with ΔBW>15%-20%, 281 pairs (7.0%) with ΔBW 20%-≤25%, and 341 twin pairs (8.5%) with ΔBW>25%. With ∆BW of 20% as the diagnostic cutoff, the incidence of birth weight discordance was 15.5% (622/4 011). The proportion of natural births in the ∆BW≤15% group was higher than that in the ∆BW>15%-20% group (10.5% (288/2 740) vs. 6.3% (29/463), P<0.008 3). The ∆BW>25% group had a significantly higher prevalence of maternal hypertensive disorders during pregnancy than that of the other 3 groups (25.5% (87/341) vs. 16.7% (47/281) vs.17.3% (83/480) vs. 13.8% (400/2 899), all P<0.008 3). Univariate analysis found that the ΔBW>25% group had a lower gestational age and a higher rate of preterm birth than the other groups. The rate of extremely low birth weight (ELBW) or very low birth weight (VLBW), small for gestational age (SGA), and transferring to the department of neonatology in the smaller twins were significantly different among the 4 groups (all P<0.05). Multivariate analysis showed that higher degree of birth weight discordance was all positively associated with the rate of ELBW, SGA, and transferring to the department of neonatology in smaller twin, even after adjusting maternal age and gestational hypertension, year of birth, mode of delivery, gender, and gestational age (all P<0.05). Moreover, the Mantel-Haenszel test also indicated that there were significantly low to moderate correlations between ΔBW and the unfavorable outcomes (r=0.22, 0.53, 0.21, all P<0.001, respectively). The receiver operating characteristic (ROC) curve found that adverse birth outcomes would be well predicted by birth weight-discordant when the diagnostic cut-off of ΔBW was 12%-17%, with an acceptable sensitivity (0.53-0.78) and a high specificity (0.72-0.79).

Birth weight discordant is not uncommon in live-born twins, and is associated with adverse outcomes including ELBW, SGA, and transferring to the department of neonatology in the small twins. Besides, the risk is linearly related to the increase of ΔBW. In the future, more researches are needed to explore the underline mechanism and long-term impact of birth weight discordance, to guide the prevention and management.

Disease in twins; Birth weigh-discordance; Occurrence; Birth outcomes