Screening for fetal growth restriction with universal third trimester ultrasonography in nulliparous women in the Pregnancy Outcome Prediction (POP) study: a prospective cohort study

Summary Background Fetal growth restriction is a major determinant of adverse perinatal outcome. Screening procedures for fetal growth restriction need to identify small babies and then differentiate between those that are healthy and those that are pathologically small. We sought to determine the diagnostic effectiveness of universal ultrasonic fetal biometry in the third trimester as a screening test for small-for-gestational-age (SGA) infants, and whether the risk of morbidity associated with being small differed in the presence or absence of ultrasonic markers of fetal growth restriction. Methods The Pregnancy Outcome Prediction (POP) study was a prospective cohort study of nulliparous women with a viable singleton pregnancy at the time of the dating ultrasound scan. Women participating had clinically indicated ultrasonography in the third trimester as per routine clinical care and these results were reported as usual (selective ultrasonography). Additionally, all participants had research ultrasonography, including fetal biometry at 28 and 36 weeks' gestational age. These results were not made available to participants or treating clinicians (universal ultrasonography). We regarded SGA as a birthweight of less than the 10th percentile for gestational age and screen positive for SGA an ultrasonographic estimated fetal weight of less than the 10th percentile for gestational age. Markers of fetal growth restriction included biometric ratios, utero-placental Doppler, and fetal growth velocity. We assessed outcomes for consenting participants who attended research scans and had a livebirth at the Rosie Hospital (Cambridge, UK) after the 28 weeks' research scan. Findings Between Jan 14, 2008, and July 31, 2012, 4512 women provided written informed consent of whom 3977 (88%) were eligible for analysis. Sensitivity for detection of SGA infants was 20% (95% CI 15–24; 69 of 352 fetuses) for selective ultrasonography and 57% (51–62; 199 of 352 fetuses) for universal ultrasonography (relative sensitivity 2·9, 95% CI 2·4–3·5, p<0·0001). Of the 3977 fetuses, 562 (14·1%) were identified by universal ultrasonography with an estimated fetal weight of less than the 10th percentile and were at an increased risk of neonatal morbidity (relative risk [RR] 1·60, 95% CI 1·22–2·09, p=0·0012). However, estimated fetal weight of less than the 10th percentile was only associated with the risk of neonatal morbidity (pinteraction=0·005) if the fetal abdominal circumference growth velocity was in the lowest decile (RR 3·9, 95% CI 1·9–8·1, p=0·0001). 172 (4%) of 3977 pregnancies had both an estimated fetal weight of less than the 10th percentile and abdominal circumference growth velocity in the lowest decile, and had a relative risk of delivering an SGA infant with neonatal morbidity of 17·6 (9·2–34·0, p<0·0001). Interpretation Screening of nulliparous women with universal third trimester fetal biometry roughly tripled detection of SGA infants. Combined analysis of fetal biometry and fetal growth velocity identified a subset of SGA fetuses that were at increased risk of neonatal morbidity. Funding National Institute for Health Research, Medical Research Council, Sands, and GE Healthcare.


Supplementary References
Plot of measurements of abdominal circumference (AC) measured in millimeters (mm) at the time of the clinical (reported) 20 week measurement and the research measurements at ~28 and ~36 weeks. The green line is the 50 th percentile and the black lines above and below are the 5 th and 95 th percentiles of a previously published reference range.

A. B.
The receiver operating characteristic (ROC) curve for A. Small for gestational age (SGA, <10 th percentile), and B. severe SGA (<3 rd percentile), using the estimated fetal weight (EFW) percentile (i.e. the percentile from the last scan performed prior to birth). Solid lines represent universal ultrasonography and dashed lines represent selective ultrasonography. When the results of selective sonography were analysed, 58% (2,311/3,977) women did not have a clinically indicated scan at or after 26 weeks gestational age. In this group, EFW was imputed using a sex-specific population median. Areas under the ROC curves (95% confidence interval) are 0.

A. B.
The receiver operating characteristic (ROC) curve for A. Small for gestational age (<10 th percentile), and B. severe small for gestational age (<3 rd percentile), using the estimated fetal weight percentile from the 36 week research scan. The outcome was, necessarily, confined to births that occurred following the 36 week research scan. Areas under the ROC curves (95% confidence interval) are 0.86 (0.85-0.88) and 0.91 (0.89-0.94), respectively. Regression models were fitted between each measurement and GA within each GA interval (18-22, 26-30 and 34-38 completed weeks), i.e. excluding GAs without data points using published methodology. 5 Doppler PIs were log-transformed prior to model fitting. The equations of the mean and SD models give the expected mean and SD at each GA within the respective GA range. GA-specific z scores were calculated as (observed value -fitted mean) / fitted SD. Lowest and highest 10% (deciles) were determined from the z score distributions. For the AC growth velocity, change in the z score between 20 week scan and the last scan was calculated and the lowest decile of the difference was determined. For 97% of women (n=3850), the 36 week scan was the last scan. If the 36 week scan measurement was missing (delivery occurred before the 36 week scan or data were missing at the 36 week scan), the decile from the 28 week scan was used instead. Similarly for AC growth velocity, the lowest decile from the change between 20 and 28 week scans was used if the 36 week scan result was missing. For all z scores, mean=0.0 and SD=1.0. For AC growth velocity (change in z score), mean=0.0 and SD=1. Data are mean (SD) for biometric measurements and geometric mean (geometric SD) for Doppler measurements. Biometric and Doppler measurements were performed as previously described. 2;6-8 AC and HC were measured using the ellipse function of the machine. BPD was measured from the outer surface (nearside to the probe) to the inner surface (far side to the probe). Umbilical Doppler was assessed in a free loop of cord in the middle of its length (i.e. outside the regions of the umbilical and placental insertions), and uterine Doppler was assessed where the vessels cross the external iliac artery and vein.

Supplementary
In the research scans at 28 and 36 weeks, the screen display of gestational age (GA) equivalence of measurements in the machine was disabled, to prevent ad hoc assessment of the appropriateness of growth measurements. EFW was calculated using published formulae. 3

Summary of results
Inter-observer reliability and agreement statistics suggest a very small measurement error for fetal biometric measurements in both 20 week and 36 week scans. Differences in measurements between two sonographers were slightly larger for FL than for BPD, HC and AC. These differences had a cumulative effect on EFW which was calculated from the four measurements, resulting in a slightly higher mean coefficient of variation (CV) than was observed for individual measurements. There was more variation in Doppler measurements in both scans (mean CV range: 5.58-8.22%) than in biometric measurements (mean CV range: 0.46-3.15%). There was no clear indication that the difference in measurements between sonographers varied according to the mean of the two measurements, except for uterine artery Doppler PI measurements, where largest differences tended to occur at the top end of the distribution (Supplementary Figure 13: Bland-Altman plots). ROC denotes receiver operating characteristic, GA denotes gestational age and SGA denotes small for gestational age. SGA is defined as birth weight <10 th percentile and severe SGA is defined as birth weight <3 rd percentile. The area under the ROC curve is calculated using the estimated fetal weight percentile. Sensitivity Analysis 1: Including women who defaulted from one or more research scans and defining their missing scan(s) as screen negative. Sensitivity Analysis 2: Including women who defaulted from one or more research scans and, if they had a clinically indicated scan at 26-30 weeks or 34-38 weeks, the last EFW within the respective time window was used as the result of the research scan. If no clinically indicated scan was performed within that time window, the record was treated as screen negative. Sensitivity Analysis 3: Excluding all records where the research scan result was revealed for any reason. Sensitivity Analysis 4: Re-classifying research scan result as screen positive if the last research scan was negative but a subsequent last clinically indicated scan was screen positive. Abbreviations: SGA denotes small for gestational age, EFW denotes estimated fetal weight, ACGV denotes abdominal circumference growth velocity, RR denotes relative risk and CI denotes confidence interval. All EFW are based on population based percentiles. SGA is defined as birth weight <10 th percentile, screen positive is defined as EFW<10 th percentile and screen negative is defined as EFW≥10 th . ACGV is based on the change in the gestational age adjusted z score between the 20 week scan and the last scan before birth. Z score at each scan was calculated using growth charts generated by the Fetal Growth Longitudinal Study component of the INTERGROWTH-21st Project, an international consortium which constructed fetal growth standards using methods recommended by the WHO. 1 The lowest decile of the change in z score between the 20 week scan and the last scan was defined within the study cohort. For 97% of women (n=3850), the 36 week scan was the last scan. If the 36 week scan measurement was missing (delivery occurred before the 36 week scan or data were missing at the 36 week scan), the decile from the 28 week scan was used instead. The change in z score cut-off point of the lowest decile was -1.594 from 20 week scan to 36 week scan (-1.2255 from 20 week to 28 week scan). Neonatal morbidity is a composite outcome, i.e. ≥1 of the three outcomes specified: metabolic acidosis (defined as pH<7.1 and a base deficit of more than 10mmol/L), 5 minute Apgar <7, neonatal unit admission. Neonatal unit admission was defined as admission to the Neonatal Intensive Care Unit, the High dependency Unit, or the Special Care Baby Unit. Severe adverse perinatal outcome is a composite outcome, i.e. ≥1 of the following outcomes specified: stillbirth (not due to congenital anomaly), neonatal death at term (not due to congenital anomaly), hypoxic ischaemic encephalopathy at term, use of inotropes at term, mechanical ventilation at term, severe metabolic acidosis at term (defined as pH<7.0 and a base deficit of more than 12mmol/L). P-values are from 2-sided Fisher's exact test. In the analysis stratified by ACGV (defined by the INTERGROWTH-21 st growth standard), an EFW<10 th percentile was associated with the risk of any neonatal morbidity when the fetal ACGV was in the lowest decile (RR=4.33, 95% CI 1.96 to 9.57) but there was no association in the normal ACGV group (RR=1.12, 95% CI 0.76 to 1.66), P for interaction = 0.003. Abbreviations: FGR denotes fetal growth restriction, SGA denotes small for gestational age, AC denotes abdominal circumference, ACGV denotes abdominal circumference growth velocity, FL denotes femur length, HC denotes head circumference, RR denotes relative risk, CI denotes confidence interval and N/A denotes not applicable. The five previously described indicators of FGR were classified as the extreme decile associated with FGR (highest or lowest, as appropriate) compared with the other 9 deciles in the cohort. Neonatal morbidity is a composite outcome, i.e. ≥1 of the three outcomes specified: metabolic acidosis (defined as pH<7.1 and a base deficit of more than 10mmol/L), 5 minute Apgar <7, neonatal unit admission. Neonatal unit admission was defined as admission to the Neonatal Intensive Care Unit, the High dependency Unit, or the Special Care Baby Unit. Severe adverse perinatal outcome is a composite outcome, i.e. ≥1 of the following outcomes specified: stillbirth (not due to congenital anomaly), neonatal death at term (not due to congenital anomaly), hypoxic ischaemic encephalopathy at term, use of inotropes at term, mechanical ventilation at term, severe metabolic acidosis at term (defined as pH<7.0 and a base deficit of more than 12mmol/L). P-values are from 2-sided Fisher's exact test.