Evaluation of the RCOG guideline for the prediction of neonates that are small for gestational age and comparison with the competing risks model.

Abstract

To examine the predictive performance of the relevant guideline by the Royal College of Obstetricians and Gynaecologists (RCOG) for neonates that are small for gestational age (SGA), and to compare the performance of the RCOG guideline with that of our competing risks model for SGA. Prospective observational study. Obstetric ultrasound departments in two UK maternity hospitals. A total of 96 678 women with singleton pregnancies attending for routine ultrasound examination at 19-24 weeks of gestation. Risks for SGA for different thresholds were computed, according to the competing risks model using maternal history, second-trimester estimated fetal weight, uterine artery pulsatility index and mean arterial pressure. The detection rates by the RCOG guideline scoring system and the competing risks model for SGA were compared, at the screen positive rate (SPR) derived from the RCOG guideline. Small for gestational age (SGA), <10th or <3rd percentile, for different gestational age thresholds. At an SPR of 22.5%, as defined by the RCOG guideline, the competing risks model predicted 56, 72 and 81% of cases of neonates that are SGA, with birthweights of <10th percentile, delivered at ≥37, <37 and <32 weeks of gestation, respectively, which were significantly higher than the respective figures of 36, 44 and 45% achieved by the application of the RCOG guideline. The respective figures for neonates that were SGA with birthweights of <3rd percentile were 66, 79, 85 and 41, 45, 44%. The detection rate for neonates that were SGA with the competing risk approach is almost double than that obtained with the RCOG guideline. The competing risks approach for the prediction of SGA performs better than the existing RCOG guideline.