Fetal MRI at 36 weeks predicts neonatal macrosomia significantly better than USG: PREMACRO study
Prenatal estimation of fetal weight (EFW) began during the very early days of obstetric ultrasound. The estimations were initially based solely on abdominal circumference with biparietal diameter being added some years later. Hadlock introduced an algorithm for prediction based on the measurement of biparietal diameter, head circumference, abdominal circumference and femur length. Unfortunately this algorithm is still relatively imprecise, identifying only 73% of LGA neonates > 95th centile for a fixed false positive rate of 10%.
The purpose of EFW is to follow serial growth and identify fetuses at the extremes, either small or large-for-gestational age (SGA or LGA). Both these groups have higher rates of morbidity and mortality; therefore once identified, managing clinicans have the possibility to modify the timing and/or mode of birth in order to optimize clinical outcomes and minimise perinatal risks.
In a blinded prospective study, PREMACRO, Caroline Kadji et al tested the hypothesis that MRI-EFW performs better than standard US-EFW based on Hadlock et al. does in the prediction for LGA ≥ 95th centile for gestation in a large, unselected population of women at 36 weeks of gestation. It also evaluated the performance of MRI-EFW and US-EFW in the prediction of LGA and SGA at various cut-offs.
From May 2016 through February 2019, women received counselling at the 36 weeks clinic. In this prospective, single-centre, blinded study, pregnant women with singleton pregnancies between 36+0 – 36+6 weeks gestation, underwent both standard evaluation of estimated fetal weight with US according to Hadlock et al. and MRI according to the formula developed by Baker et al., based on the measurement of the fetal body volume (FBV). Birth weight centile was considered as the gold standard for the US and MRI-derived centile. The primary outcome was the area under the receiver-operating characteristic curve (AUROC) for the prediction of ≥ 95th centile LGA neonates. Secondary outcomes included the comparative prediction of LGA neonates ≥ 90th, 97th, 99th centile, small-for-gestational age (SGA) ≤ 10th, 5th, 3th centile for gestational age and maternal/perinatal complications.
- Of 2,914 women who were initially approached, results from 2,378 were available for analysis. Total FBV measurements were possible for all fetuses and the time required to perform the planimetric measurements by MRI was 3.0 (range; 1.3-5.6) minutes. The AUROC curve for prediction of birthweight ≥ 95th centile by prenatal MRI was 0.985 and by US it was 0.900 (difference = 0.085, P<0.001).
- For a fixed false-positive rate of 5 %, MRI-EFW detected 80.0% whereas US-EFW detected 59.1% of birthweight ≥ 95th centile. The positive predictive value (PPV) was 42.6% for MRI-EFW and 35.4 % for US-EFW, and the negative predictive value (NPV) was 99.0 for MRI-EFW and 98.0 for US-EFW.
- For a fixed false-positive rate of 10 %, MRI-EFW detected 92.4% whereas US-EFW detected 76.2 % of birthweight ≥ 95th centile. The PPV was 29.9 % for MRIEFW and 26.2 % for US-EFW and the NPV was 99.6 for MRI-EFW and 98.8 for US-EFW.
- The AUROC curves for prediction of LGA ≥ 90th, 97th , 99th centile and SGA ≤ 10th, 5th , 3 th centile was significantly larger by prenatal MRI as compared to US (p<0.05 for all).
The findings of this study showed that in singleton pregnancies evaluated at 36 weeks, MRIEFW performs significantly better that US-EFW for the detection of LGA neonates ≥ 95th centile for gestational age. For a fixed false-positive rate of 5%, detection rates of birthweight ≥ 95th centile by MRI-EFW was 80.0% whereas it was only 59.1% using US-EFW. The degree of improvement for MRI-EFW as compared to US-EFW was even more pronounced for SGA neonates. Study further showed that there was a significant association between MRI-EFW ≥ 95th and certain perinatal outcomes.
Macrosomia is a risk factor for adverse perinatal outcomes and so far, traditional screening methods have been relatively poor in terms of performance. This explains the findings of a decision-analysis which suggested that the number of elective caesarean sections needed to avoid one permanent brachial plexus injury is quite high and the recommendation to limit planned caesarean for suspected macrosomia only when estimated fetal weight exceeds 4500 g for women with diabetes and 5000 g for those without diabetes.
Another approach to suspected macrosomia is to induce labour between 37+0 and 38+6 weeks, which effectively arrests the problem of continued fetal growth, thereby reducing the associated risks of fetal and maternal morbidity. However, these benefits should be balanced against the potential negative effects of early-term induction of labor, which includes both short-term complications, such as neonatal respiratory morbidity and long-term developmental and behaviourial concerns; therefore this proposed strategy remains the subject of much ongoing debate. Here, study suggests a new approach that includes MRI-EFW to identify fetuses at the extremes of growth.
One could argue about the practicality of using fetal MRI for making these predictions in clinical practice with regards to cost and limited availability of most facilities. First, with the simplification of the technique cost arguments may be easily outweighed by the benefit of using an accurate method. Second, given the time needed to undergo such an examination, the argument of accessibility to MRI is and will remain country dependent and therefore may in some places be a false one.
The authors concluded, "This prospective observational study showed that in routine singleton pregnancies evaluated at 36 weeks of gestation, MRI-EFW performs significantly better than US-EFW in identifying LGA and SGA neonates. Future interventional studies based on a more accurate method of fetal weight estimation need to be conducted in order to determine whether this could improve neonatal and maternal morbidity or alternatively, whether this could reduce unnecessary intervention such as caesarean sections and induction of labor, while maintaining stable neonatal and maternal related morbidity."
Source: : Kadji C, Cannie MM, Kang X et al, American Journal of Obstetrics and Gynecology (2021),