Objective: To estimate and compare the risk of morbid perinatal outcomes in pregnancies identified as small for gestational age (SGA) with customized compared with conventional standards of fetal growth.
Methods: Ultrasound-derived estimates of fetal weight were used to generate a fetal growth trajectory (N=7,510). The gestational age at delivery and pathologic and physiologic variables from 5,072 pregnancies were used to calculate a customized threshold for SGA. In a separate analysis of 32,070 pregnancies, rates of morbid outcomes were compared in participants classified as SGA according to a population-based birth weight standard only (SGApop only), a customized standard only (SGAcust only), and both methods (SGAboth).
Results: Eight-hundred seventy-five (2.7%) participants were SGApop only, 1,970 (6.1%) participants were SGAboth, and 609 (1.9%) participants were SGAcust only. The odds ratios of neonatal death in SGApop only and SGAcust only pregnancies were 1.78 (95% confidence interval [CI] 0.2-13.1) and 54.6 (95% CI 29.0-102.8), respectively. Rates of prematurity in the SGApop only and SGAcust only cohorts were 4.8% and 64.5%, respectively. After adjustment for the effect of prematurity, odds ratios of neonatal death in the SGApop only and SGAcust only cohorts were 4.8 (95% CI 0.6-37.0) and 2.9 (95% CI 1.4-6.1), respectively.
Conclusion: After adjustment for confounding stemming from premature delivery, there is little difference in the risk of adverse outcomes between SGAcust only and SGApop only participants. Adoption of customized fetal growth standards into clinical practice may not improve the ability to identify pregnancies with increased risk of perinatal morbidity.