Impact of customized growth curves on screening for small for gestational age twins

The choice of a growth curve determines the screening for small-for-gestational-age (SGA) fetuses and little data is available on SGA twins. Our aim was to evaluate small-for-gestational-age (SGA) detection rate in twin pregnancies and assess whether the use of a customized curve allowed better identification of SGA fetuses.

Retrospective study including all twins between 2010 and 2013. Two groups were formed: the SGA and the non-SGA group. Four curves were compared: Hadlock's curve, a customized curve, EPOPé M0 and EPOPé M1. We defined a composite neonatal complication criterion (transfer to intensive care unit, respiratory distress and death).

472 fetuses were included with a 34.3% prevalence of SGA. Hadlock's curve showed better sensitivity for the detection of SGA <10th percentile (67.3% vs. 63%, 59.9% and 57.4% respectively). Diagnostic Odd Ratio were comparable for the detection of SGA. For the composite variable, there was a significant difference between the 2 groups using a customized curve adjusted for fetal sex (EPOPé M1).

The EPOPé (M0 and M1) and customized curves do not improve screening for SGA infants below the 10th percentile. The reduced effectiveness of customized curves can be related to the greater impact of placentation or cord insertion on the potential for fetal growth.