The association between maternal complications and small for gestational age in twin pregnancies using singleton and twin birth weight references

Abstract

Introduction In singleton pregnancies, maternal complications, such as preeclampsia and thyroid dysfunction are associated with small for gestational age (SGA). However, data on the association between SGA and maternal complications in twin pregnancies are limited and conflicting. Small sample size and the application of singleton birth weight reference (SBWR) to define SGA in twins may be reasons for the inconsistent conclusions. Purpose of this study was to define SGA in dichorionic diamniotic (DCDA) and monochorionic diamniotic (MCDA) twin pregnancies using both SBWR and twin birth weight reference (TBWR) and to determine whether certain maternal complications are associated with SGA in twin pregnancies. Materials and methods This retrospective cohort study included all twins delivered in a single tertiary center between 2013 and 2018. SGA was defined as a twin with birth weight <10th centile for gestational age using either SBWR or TBWR. The association between SGA and maternal complications was analyzed separately for DCDA and MCDA twin pregnancies, expressed as odds ratio (OR) and adjusted odds ratio (aOR) with 95% confidence interval (95%-CI). Results A total of 2005 DCDA and 467 MCDA twin pregnancies were enrolled. In DCDA pregnancies, SGA was significantly associated with PE according to TBWR (22.73 vs. 14.56%, aOR 1.823, 95%-CI 1.137–2.922). This association was even more pronounced between SGA and severe PE (9.09 vs. 4.54%, aOR 2.234, 95%-CI 1.115–4.479). In turn, PE was associated with higher risk of SGA defined according to TBWR (8.31 vs. 4.99%, aOR 1.825, 95%-CI 1.139–2.925). No association was detected between SGA and other maternal complications. Using SBWR, no association was found between preeclampsia and SGA. In MCDA pregnancies, according to TBWR, SGA was associated only with severe PE (12.5 vs. 4.06%, aOR 3.470, 95%-CI 1.256–9.587) and lower risk of PROM (aOR 0.067, 95%-CI 0.014–0.322). Conclusion PE was associated with SGA in DCDA pregnancies only when TBWR was used, suggesting that DCDA pregnancies complicated with PE should be closely monitored for signs of SGA and vice versa. In MCDA pregnancies, SGA defined according to TBWR was associated with only severe PE (but not with all PE) and lower risk of PROM. More basic experiments are needed to investigate the mechanisms underlying PE and SGA in DCDA and MCDA twin pregnancies respectively.