Performance of third-trimester ultrasound for prediction of small-for-gestational-age neonates and evaluation of contingency screening policies
To assess the performance of third-trimester fetal biometry and fetal Doppler studies for the prediction of small-for-gestational-age (SGA) neonates, and to explore contingency strategies using a first-trimester prediction model based on maternal and fetal parameters and third-trimester ultrasound.
This was an observational cross-sectional study of uncomplicated singleton pregnancies. Risk assessment for chromosomal abnormality was carried out in 4702 pregnancies using a combination of ultrasound markers (fetal nuchal translucency thickness (NT) and nasal bone assessment) and biochemistry (free beta-human chorionic gonadotropin (β-hCG) and pregnancy-associated plasma protein-A (PAPP-A)) at 11 to 13 + 6 weeks. Maternal demographic characteristics and method of conception were recorded. Third-trimester (30–34 weeks) fetal biometry (biparietal diameter (BPD), head circumference (HC), abdominal circumference (AC) and femur length (FL)) and umbilical artery (UA) and middle cerebral artery Doppler studies were performed routinely in a subgroup (n = 2310). Reference ranges for birth weight were constructed using the cohort of 4702 women, and neonates were classified as small (SGA, ≤ 5th centile) or appropriate (AGA) for gestational age. First-trimester, third-trimester and integrated first- and third-trimester prediction models for SGA were constructed using regression analysis and three different contingency strategies of rescanning in the third trimester were investigated.
According to the areas under the receiver–operating characteristics curves (AUCs), AC (AUC = 0.85) and ultrasound-estimated fetal weight (EFW, AUC = 0.87) were equally good predictors of SGA. The model was marginally improved by the addition of UA Doppler, smoking status and first-trimester indices (free β-hCG and PAPP-A multiples of the median) (combined model, AUC = 0.88), but the difference was not statistically significant. A contingency strategy of rescanning 50% of the population in the third trimester according to the risk estimated by a first-trimester prediction model yielded a detection rate of 79% for a 25% screen-positive rate.
Third-trimester ultrasound is effective in screening for SGA in uncomplicated pregnancies. The use of a contingency screening policy can reduce the need for unnecessary examinations. Copyright © 2012 ISUOG. Published by John Wiley & Sons, Ltd.
Intrauterine growth restriction is associated with increased perinatal mortality and morbidity and suboptimal long-term outcome1–4. Growth-restricted fetuses, once identified as such, benefit from intensive antenatal surveillance5. However, there is controversy as to the effectiveness of screening for growth restriction in low-risk pregnancies and the impact of antenatal detection of smallness on perinatal complications6–8.
Symphysiofundal height measurement is the most common method of antenatal screening for reduced fetal growth, although it has been reported to have low sensitivity and specificity9. The incorporation of maternal demographic characteristics in customized growth charts improves the detection rate of small-for-gestational-age (SGA) fetuses10. Fetal biometry by ultrasound achieves good prediction for SGA but its use is usually restricted to high-risk pregnancies6, 11; few studies have assessed the effectiveness of third-trimester ultrasound biometry in screening for SGA fetuses in low-risk obstetric populations12–14. Recently, the combination of first-trimester fetal parameters, biochemical indices and maternal demographic characteristics has been shown to be predictive for SGA15.
Our study examined the screening performance of a single ultrasound examination at 30–34 gestational weeks in low-risk pregnant women. We also report on the value of using maternal demographic characteristics and first-trimester fetal and biochemical indices to identify the pregnancies most likely to benefit from a third-trimester examination, in order to reduce the need for unnecessary visits.
This was an observational cross-sectional study. The data and pregnancy outcomes were collected from Leto Maternity Hospital and from Attikon University Hospital between January 2009 and July 2010.
Risk assessment for chromosomal abnormality by a combination of ultrasound markers (fetal nuchal translucency thickness (NT) and nasal bone assessment) and biochemical measurements (free beta-human chorionic gonadotropin (β-hCG) and pregnancy-associated plasma protein-A (PAPP-A)) was carried out at 11 to 13 + 6 weeks of gestation at both hospitals, using the same protocol. Maternal blood was drawn at the time of the ultrasound examination and analyzed either simultaneously or within 24 h. Maternal demographic characteristics (weight, height, parity and smoking status), method of conception (spontaneous or assisted, including ovulation induction and in-vitro fertilization) and ultrasound parameters (crown–rump length (CRL) and NT) were recorded in a computer database (Astraia software; Astraia GmbH, Munich, Germany). Gestational age in days was defined by last menstrual period (LMP). In cases with uncertain dates or if the difference between the LMP- and CRL-derived dates was 7 days or more, the gestational age was corrected according to CRL. Women were considered to be parous if they had had a previous delivery at or beyond 24 weeks.
Third-trimester ultrasound examination was performed routinely at 30–34 weeks and included measurements of biparietal diameter (BPD), head circumference (HC, calculated from BPD and occipitofrontal diameter measurements), abdominal circumference (AC, calculated from anteroposterior and transabdominal diameter measurements) and femur length (FL). Color Doppler was used to assess the umbilical artery (UA) and the fetal middle cerebral artery (MCA) and to measure their pulsatility indices (PI). Fetal Doppler studies were performed in the absence of fetal and breathing movements over three consecutive cardiac cycles.
The study sample consisted of viable, singleton pregnancies with known outcomes that were delivered beyond 24 weeks. Exclusion criteria were hypertensive disorders of pregnancy, gestational diabetes mellitus and pre-eclampsia. Women with either a history of a previous pregnancy complicated by these conditions or a medical history of hypertension and diabetes mellitus (Type I or Type II) were also excluded. Finally, pregnancies with chromosomal abnormalities and/or structural defects, pregnancies resulting in miscarriage or intrauterine death and pregnancies diagnosed with severe early-onset growth restriction prior to the routine third-trimester scan were not considered in the analysis.
All ultrasound scans were performed by sonographers certified by The Fetal Medicine Foundation (FMF). The blood samples for PAPP-A and free β-hCG were analyzed by the Kryptor system (Brahms, Berlin, Germany). Biochemical analyte values were adjusted for maternal weight, method of conception, gestational age, cigarette smoking and race according to FMF algorithms and the results were converted to multiples of the median (MoM). The ethics committees of the two institutes approved the use of these data for analysis.
Data management and statistical analysis
Descriptive exploratory data analysis was used and normality of the distributions was tested by the Kolmogorov–Smirnov test. The distributions of skewed parameters were made Gaussian by logarithmic transformation. Comparisons of normally distributed parameters between appropriate-for-gestational-age (AGA) and SGA groups were made by unpaired t-test, whereas Mann–Whitney U-test was applied for non-normally distributed continuous variables. Comparisons when the examined variable was normally distributed in one group and non-normally distributed in the other group were carried out by Mann–Whitney U-test. Dichotomous categorical variables were compared by χ2 test.
The cohort of 4702 women with known outcomes and no pregnancy complications was used to construct reference ranges for birth weight (BW) in relation to gestational age, using the mean and SD model developed by Royston and Wright16. Log-transformed BW was regressed against gestational age at delivery (GAD), with GAD expressed in days, as this is considered the optimal methodology with which to produce accurate reference ranges16. Scaled absolute residuals derived by this regression were regressed against gestational age to define the association between SD of BW (SDGA) and gestational age16. The scaled absolute residuals were computed from residuals by removing the sign and multiplying by 1.25. Furthermore, we added fetal gender as a regressor to construct gender-adjusted BW standards. First- and second-degree terms for GAD sufficiently described our data and fetal gender was entered as a dichotomous variable in the regression equation. Specifically:
where BWGA is BW in relation to gestational age.
BW centiles were constructed as follows:
where K is the z-score for the corresponding centile and assuming that SDGA was the same for male and female fetuses. Finally, we obtained the anti-log values for BWGA.
Fetuses with BW at or below the 5th centile according to this formula were classified as SGA. Our study cohort of 4702 women consisted of 4441 non-SGA and 261 SGA neonates, according to our reference ranges.
Regression analysis with first-, second- and third-degree equation terms was used in order to identify the best-fit polynomial equation that described the associations between log10 CRL and gestational age (calculated in days according to LMP) and log10 NT and CRL. The regression equations were fitted in the non-SGA fetuses. Delta values for NT and CRL for both non-SGA and SGA neonates were calculated as the difference between the observed and expected values, the latter being estimated from the regression equations.
Regression analysis was also applied in the subgroup of 2310 women who had undergone both first-trimester screening and a third-trimester growth ultrasound scan with Doppler assessment to develop nomograms for BPD, HC, AC, FL and EFW and to calculate the respective delta values. The EFW was extracted from the database and was calculated by the Hadlock formula17. The 2189 non-SGA fetuses of this subgroup were investigated in order to define UA-PI and MCA-PI nomograms. MoMs for UA-PI and MCA-PI were computed by dividing observed by expected values for each biophysical parameter. The potential relation of UA-PI and MCA-PI with maternal and pregnancy characteristics was assessed.
Regression diagnostics in terms of residual analysis were used to confirm the regression assumptions (linearity, homoscedasticity). Identification of outliers by Mahalanobi's and Cook's distances was performed and outliers were removed to improve our regression models, but they are included in our presentation of descriptive statistics.
A first-trimester prediction model for SGA was developed by multiple logistic regression analysis. Receiver–operating characteristics (ROC) curves were plotted based on the probabilities predicted by the model in order to evaluate the predictive value of the regression model. The utilized predictors were: delta NT, delta CRL, PAPP-A MoM, free β-hCG MoM, maternal age, maternal weight, maternal height, parity, smoking status and method of conception. Racial origin was not used as a predictor because 99.97% of the studied population was Caucasian.
Third-trimester prediction models for SGA were constructed separately for dBPD, dHC, dAC, dFL, z log10 EFW (z-score of log transformed EFW), UA-PI MoM and MCA-PI MoM. dBPD, dHC, dAC, dFL and z log10 EFW were not used simultaneously in the regression model due to significant collinearity among these variables. The individual ROC curves were plotted and compared. We combined third-trimester EFW (the best predictor of SGA in third trimester) with maternal/pregnancy characteristics and Doppler studies to develop a third-trimester combined model. Subsequently, first- and third-trimester parameters were combined by logistic regression modeling to develop an integrated model and evaluate its performance in identifying SGA fetuses.
Finally, contingency policies were evaluated. Baye's theorem was applied to recalculate the risk in the 2310 pregnancies that had a third-trimester scan. This methodology has been used extensively in screening strategies18. According to Baye's theorem: posterior odds = prior odds × likelihood ratio. In our study, posterior odds (Oposterior) were the odds for SGA after reevaluation in the third trimester, while prior odds (Oprior) were those according to the first-trimester prediction model. The first-trimester prediction model for SGA was applied to calculate case-specific Oprior. Predicted probabilities (PP) obtained by the first-trimester prediction model were converted to Oprior by the equation:
We assumed that z log10 EFW (the best predictor of SGA in the third trimester) had a Gaussian distribution in SGA and AGA neonates, with respective means of MSGA and MAGA and standard deviations of SDSGA and SDAGA. Assuming that YSGA and YAGA are the distributions of z log10 EFW for SGA and AGA, respectively, with MSGA≠ MAGA and SDSGA ≠ SDAGA, the likelihood ratio (LR) for each individual would be:
where x is the estimated z log10 EFW by third-trimester scan for the individual.
We investigated three different strategies, with third-trimester scans being performed in 20%, 30% or 50% of pregnancies with the highest predicted probability for SGA according to the first-trimester model. Oposterior was calculated supposing that we rescanned 20%, 30% and 50% of fetuses and the remaining population maintained their first-trimester assigned risk. The contingency strategies were evaluated by ROC-curve analysis.
The level of significance was set at P < 0.05 and P-values for all tests were two-sided for all analyses.
A total of 4702 women had the first-trimester ultrasound assessment, of whom 2310 women also had results from the third-trimester examination available for analysis. Demographic characteristics, descriptive statistics of first- and third-trimester ultrasound parameters and distributions of biochemical indices in these 2310 women are presented in Table 1. There were no statistically significant differences in demographic data, distributions of first-trimester screening markers, percentage of preterm delivery, gestational age at delivery or percentage of SGA occurrence between women with and those without third-trimester ultrasound data.
Table 1. Demographic characteristics and distribution of ultrasound and biochemical indices in the population with data from both first- and third-trimester ultrasound examinations (n = 2310)
|Maternal/pregnancy characteristics|| || || |
| Maternal age (years, mean (range))||31.7 (16–52)||32.2 (21–43)||0.158|
| Prepregnancy maternal weight (kg, median (range))||63 (41–138)||58 (43–93.5)||< 0.001|
| Maternal height (cm, median (range))||165 (142–188)||163 (141–176)||< 0.001|
| Nulliparous (n (%))||1034 (47.2)||59 (48.7)||0.07|
| Smoker (n (%))||323 (14.8)||33 (27.3)||< 0.001|
| IVF pregnancy (n (%))||73 (3.3)||8 (6.6)||0.057|
| GA at delivery (days, median (range))||269 (211–296)||269 (228–286)||0.864|
| Birth weight (g, median (range))||3170 (1300–4890)||2500 (1400–2950)||< 0.001|
| Male fetal gender (n (%))||1139 (52.0)||61 (50.4)||0.721|
| Preterm delivery < 34 weeks (n (%))||39 (1.78)||2 (1.65)||0.956|
|First-trimester ultrasound parameters|| || || |
| CRL (mm, median (range))||63.5 (45–84)||61.1 (46–82.7)||0.01|
| NT (mm, median (range))||1.7 (0.8–8.9)||1.55 (0.8–2.7)||< 0.001|
| dCRL (mm, mean (range))||0.21 (−15.45 to 13.26)||− 1.96 (−12.21 to 9.61)||< 0.001|
| dNT (mm, mean (range))||0.04 (−0.88 to 6.1)||− 0.4 (−0.97 to 0.96)||0.007|
|Biochemical parameters|| || || |
| PAPP-A (MoM, median (range))||0.915 (0.15–6)||0.846 (0.212–2.52)||0.023|
| Free β-hCG (MoM, median (range))||1.026 (0.2–6)||1.03 (0.08–6)||0.03|
|Third-trimester ultrasound parameters|| || || |
| BPD (mm, mean (range))||83.8 (70.1–100)||81.1 (70.8–90)||< 0.001|
| HC (mm, mean (range))||301.6 (258.1–364.9)||292.4 (248.5–322.7)||< 0.001|
| AC (mm, mean (range))||282.6 (233–342.9)||267.1 (218–304)||< 0.001|
| FL (mm, mean (range))||61.6 (52–72)||59.6 (52.5–68)||< 0.001|
| EFW (g, median (range))||1975 (1160–3217)||1714 (1027–2374)||< 0.001|
| dBPD (mm, mean (range))||0.2 (−13.9 to 12.7)||− 3 (−11.1 to 5.2)||< 0.001|
| dHC (mm, mean (range))||0.6 (−30.4 to 50.5)||− 9.54 (−39.4 to 17.51)||< 0.001|
| dAC (mm, mean (range))||0.9 (−34.59 to 47.6)||− 15.8 (−49.6 to 11.12)||< 0.001|
| dFL (mm, mean (range))||0.1 (−9.2 to 10.2)||− 2.3 (−8.3 to 3.53)||< 0.001|
| z log10 EFW (mean (range))||0.06 (−2.01 to 2.3)||− 0.99 (−3.06 to 0.37)||< 0.001|
| GA at scan (days, median (range))||226 (210–244)||227 (210–244)||0.098|
| UA-PI (mean (range))||0.94 (0.57–1.4)||1.01 (0.65–2.06)||< 0.001|
| MCA-PI (mean (range))||2.05 (1.19–3.61)||2.01 (1.32–2.8)||0.282|
| UA-PI MoM (mean (range))||0.99 (0.61–1.51)||1.07 (0.7–2.26)||< 0.001|
| MCA-PI MoM (mean (range))||1 (0.58–1.73)||1 (0.65–1.34)||0.577|
Birth-weight reference ranges
BW was significantly related to GAD and fetal gender and the following equation describes their relation:
Scaled absolute residuals (SAR) were associated with GAD as follows:
Subsequently, we created gender-adjusted reference ranges for BW by combining Equations 1, 2 and 3 and we classified 121 (5.23%) newborns as SGA in the cohort of 2310 women.
First-trimester ultrasound indices
The regression equations that described the distribution of the first-trimester ultrasound parameters were:
These equations were utilized to compute delta values.
Third-trimester ultrasound indices
In the subgroup of 2310 women who had a third-trimester ultrasound scan, regression analysis was used to explore the relation between biometric fetal ultrasound variables and gestational age (GA, expressed in days), giving the following equations:
These equations were used to estimate delta values for BPD, HC, AC, FL and z-scores for log10 EFW.
Multiple regression analysis demonstrated that UA-PI and MCA-PI were significantly related to gestational age with significant independent contributions from maternal and pregnancy characteristics:
These equations were the basis for the computation of UA-PI MoM and MCA-PI MoM.
First-trimester prediction of SGA
Multiple logistic regression analysis identified the significant independent predictors for the first-trimester prediction of SGA. The regression equation was:
where P is the probability of SGA occurrence.
The ROC curve for the first-trimester prediction of SGA is depicted in Figures 1 and 2. The area under the curve (AUC) was 0.7224 (95% CI, 0.6742–0.7706) P < 0.001.
Third-trimester prediction of SGA
When examined alone, the following third-trimester parameters were significant predictors of SGA: dBPD (AUC = 0.745 (95% CI, 0.7015–0.7883) P < 0.001), dHC (AUC = 0.7509 (95% CI, 0.7074–0.7943) P < 0.001), dAC (AUC = 0.8470 (95% CI, 0.8136–0.8803) P < 0.001), dFL (AUC = 0.7371 (95% CI, 0.6904–0.7837) P < 0.001), z log10 EFW (AUC = 0.8657 (95% CI, 0.8352–0.8961) P < 0.001) and UA-PI MoM (AUC = 0.6350 (95% CI, 0.583–0.6869) P < 0.001). MCA-PI MoM was not predictive of SGA. Significant colinearity did not allow the combination of biometric parameters and EFW. EFW, which was, when used alone, the best predictor of SGA, when analyzed together with maternal and pregnancy characteristics and UA-PI MoM provided a combined model that was marginally better (R2 = 0.29, AUC = 0.8711 (95% CI, 0.8432–0.9018) P < 0.001) than was the model using EFW alone (R2 = 0.28, AUC = 0.8657 (95% CI, 0.8352–0.8961) P < 0.001). However, this improvement did not reach statistical significance. The equation for the third-trimester combined model was:
First- and third-trimester integrated prediction of SGA
We coevaluated maternal/pregnancy characteristics, first-trimester ultrasound indices, first-trimester biochemical analytes, third-trimester EFW and UA-PI MoM to develop an integrated prediction model for SGA. Multiple logistic regression analysis revealed z log10 EFW, PAPP-A MoM, free β-hCG MoM, UA-PI MoM and smoking status as the statistically significant independent predictors in the first- and third-trimester integrated model. The equation was:
This final integrated model that incorporated z log10 EFW, PAPP-A MoM, free β-hCG MoM, UA-PI MoM and smoking status resulted in further improvement of the prediction (R2 = 0.30, AUC = 0.8798 (95% CI, 0.8514–0.9082) P < 0.001), compared with the combined third-trimester model (R2 = 0.29, AUC = 0.8711 (95% CI, 0.8432–0.9018) P < 0.001), but this improvement did not reach statistical significance
The ROCs for the third-trimester prediction parameters AC and dEFW, the combined third-trimester model and the integrated first- and third-trimester model are presented in Figure 1.
We evaluated three different contingency strategies. The basis for this concept was to use the first-trimester risk assessment for SGA in order to confine third-trimester reassessment to 20% (contingency strategy S1), 30% (contingency strategy S2) and 50% (contingency strategy S3) of the study population. The ROC curves for strategy S1 (AUC = 0.7609 (95% CI, 0.7135–0.8082) P < 0.001), strategy S2 (AUC = 0.7876 (95% CI, 0.7429–0.8321) P < 0.001) and strategy S3 (AUC = 0.8429 (95% CI, 0.8038–0.8818) P < 0.001) are presented in Figure 2. The sensitivity and screen-positive rate for different screening strategies for the prediction of SGA ≤ 5th centile using first- and third-trimester parameters are presented in Table 2.
Table 2. Sensitivity for prediction of small-for-gestational age (SGA) ≤ 5th centile by third-trimester ultrasound and a combination of first- and third-trimester indices (n = 2310)
|Third trimester (BPD)||18.18||31.4||58.68||0.7450|
|Third trimester (HC)||21.49||33.06||63.64||0.7509|
|Third trimester (AC)||41.32||57.02||74.38||0.8470|
|Third trimester (FL)||26.45||41.32||60.33||0.7371|
|Third trimester (EFW)||47.11||60||81.82||0.8657|
|Third trimester (UA-PI)||18.49||25.21||39.5||0.6350|
|First- and third-trimester integrated||49.56||57.52||86.73||0.8798|
|Contingency strategy S1||34.21||44.74||61.4||0.7609|
|Contingency strategy S2||35.96||45.61||64.04||0.7876|
|Contingency strategy S3||42.98||58.77||78.95||0.8429|
This study evaluated fetal biometry and fetal Doppler studies (UA-PI and MCA-PI) in uncomplicated pregnancies in the prediction of the SGA neonate. We also examined combinations of maternal demographic characteristics, first-trimester fetal measurements and biochemical indices with third-trimester sonographic parameters. Finally, we investigated the function of contingency strategies using the first-trimester prediction model in order to stratify variable percentages of the population to undergo a third-trimester growth scan. We investigated the sensitivity of all models for screen-positive rates of 5%, 10% and 25%. The follow-up of a pregnancy at risk for SGA would consist of one or more ultrasound scans, fetal Doppler studies and cardiotocography. These investigations do not carry risks to the mother or the baby but may result in intervention, such as induction of labor or Cesarean section, in pregnancies with normal small fetuses. In implementing a policy of third-trimester screening for SGA, the selection of an appropriate screen-positive rate depends on the availability of resources and further studies will be needed to clarify whether such a policy can reduce the perinatal risks of the small fetus.
Amongst third-trimester fetal biometry parameters, AC and EFW were the best predictors of SGA, achieving sensitivities of 74.38% and 81.82%, respectively, for a 25% screen-positive rate (Table 2). Similar results have been reported previously by other investigators. Skovron et al.14 examined 768 singleton pregnancies and reported that AC (AUC = 0.785) and EFW (AUC = 0.793) were the best predictors of SGA at birth. David et al.12 found AC to have a sensitivity of 54% for a 20% screen-positive rate in the prediction of SGA. Finally, De Reu et al.13 used AC < 25th centile derived from parity- and sex-specific growth charts in 725 pregnancies and reported a sensitivity of 53% and a specificity of 81%.
The incorporation of maternal smoking status and UA-PI marginally improved the prediction of SGA (combined third-trimester model). Further improvement was observed by adding first-trimester biochemical indices (integrated third-trimester model). Nevertheless, these models did not result in a statistically significant improvement in the prediction of SGA in comparison to using EFW alone.
We observed a novel correlation of UA-PI and MCA-PI with maternal demographic characteristics and fetal sex. UA-PI was lower in parous women and MCA-PI was correlated with maternal weight. Male fetuses had lower UA-PI and MCA-PI.
Offering routine third-trimester ultrasound in all pregnant women achieved good prediction of SGA; fetal Doppler studies did not confer a significant advantage. Aiming to reduce the need for third-trimester ultrasound, we investigated the merit of firstly using a first-trimester model for prediction of SGA based on fetal and maternal parameters, in order to assess the risk and then rescan in the third trimester those women with the highest risk. Of the three contingency strategies investigated, strategy S3 would result in comparable, albeit slightly lower, detection rates compared with using EFW alone (78.95% and 81.82%, respectively). The benefits of using contingency strategy S3 is that we would rescan only 50% and follow up 12.5% of the population, whereas with EFW we would have to rescan all cases and follow up 25% of the population. Moreover, for screen-positive rates of 5% and 10%, strategy S3 yielded acceptable detection rates for SGA, although lower than those possible using EFW (Table 2).
Routine third-trimester ultrasound for fetal biometry in low-risk pregnancies is controversial. A recent meta-analysis concluded that it offers no improvement in maternal or neonatal outcomes6. A large Swedish randomized trial compared 56 000 pregnancies with third-trimester ultrasound versus 153 000 without third-trimester ultrasound and found no significant difference in perinatal mortality7. They did, however, comment that there is a trend favoring routine ultrasound in the group of neonates with very severe growth restriction. Unfortunately, the authors did not provide information on the detection rate of SGA by ultrasound.
In contrast, Lindqvist and Molin5 compared the perinatal outcomes of 1215 SGA (≤ 5th centile) fetuses, of which 681 were identified antenatally by routine third-trimester ultrasound and 573 were not identified, with the outcomes of 24 585 AGA fetuses. The detection rate of the third-trimester scan was 54%, rising from 44% for moderate SGA to 62% for severe and 76% for extreme SGA. In general, SGA fetuses had a four-fold increased risk for adverse fetal outcome (including hypoxic encephalopathy, intracranial haemorrhage, low Apgar score, neonatal convulsions, low umbilical pH, cerebral palsy, neurodevelopmental delay and perinatal mortality). Unrecognized SGA fetuses had a four-fold higher risk of adverse outcome compared with those that were recognized antenatally. Furthermore, research on the long-term outcome of children born small shows higher rates of cerebral palsy and suboptimal neurodevelopmental outcome4, 19.
In a recent review, Figueras and Gardosi20 concluded that the routine use of third-trimester ultrasound for detection of SGA is questionable and called for further research to elucidate the benefit of this intervention. It is indeed difficult to reconcile the abundant literature highlighting the risks for the SGA fetus and the beneficial effects of increased surveillance with the failure of large trials to detect improvement in the perinatal indices. Our study has shown that third-trimester ultrasound examination has good screening performance for the detection of SGA. It is of note that all examinations were performed by experienced sonographers and this may account for the high detection rates.
Last but not least, a significant concern in introducing routine third-trimester ultrasound examinations is the additional cost. A first-trimester prediction model utilizes data already available from the first-trimester ultrasound examination at 11–14 weeks and the use of contingency strategies makes it possible to limit the number of unnecessary examinations without major compromise in the detection rate21.