Comparing Diagnosis of Fetal Growth Restriction and the Potential Impact on Management and Outcomes Using Different Growth Curves

Objectives The diagnosis of fetal growth restriction (FGR) is managed with close fetal surveillance and often requires iatrogenic delivery, as there is an associated increased risk of fetal demise. However, there is no standard reference for fetal growth. We sought to compare the intrauterine growth curve of Hadlock et al (Radiology 1991; 181:129–133) to other known growth curves to determine which one best identifies fetuses at risk without overburdening the patient and health care system with unnecessary intervention. Methods We retrospectively reviewed charts of singleton euploid pregnancies with a diagnosis of FGR (per Hadlock) at a tertiary care center from June 2014 to May 2015. We applied the estimated fetal weights from ultrasound at diagnosis of FGR to 4 population‐based growth curves by Brenner et al (Am J Obstet Gynecol 1976; 126:555–564), Williams et al (Obstet Gynecol 1982; 59:624–632), Alexander et al (Obstet Gynecol 1996; 87:163–168), and Duryea et al (Obstet Gynecol 2014; 124:16–22) and reassessed the incidence of FGR using each curve. We reviewed pregnancy demographics, risk factors, pregnancy management, and outcomes of FGR cohorts on each curve to evaluate whether poor outcomes may be missed or interventions may be avoided using the population‐based curves. A sensitivity analysis was also done to see how well each curve predicted small‐for‐gestational‐age birth weights. Results Applying any of the population‐based growth curves decreased the number of FGR diagnoses, iatrogenic deliveries, and primary cesarean deliveries. Brenner's growth curve identified the least number of FGR diagnoses at 22 of the 107 identified by Hadlock. Williams’ growth curve performed best in the sensitivity analysis with sensitivity of 99% and specificity of 97%. A small number of patients with absent/reversed end‐diastolic flow would have been missed by applying the population curves. Conclusions Applying the population‐based growth curves instead of Hadlock's for diagnosis of FGR decreases its incidence, therefore decreasing the number of visits for ultrasound and fetal surveillance and the number of iatrogenic deliveries. However, using these curves could miss a few fetuses with increased risk of fetal demise.