Diagnostic value of the routine scan during pregnancy

Given that this practice is routine in many developed countries of the world, there is considerable evidence regarding the value of this screening assessment during pregnancy. This evidence has been recently systematically reviewed.

A recent review by NICE and the National Collaborating Centre for Women’s and Children’s Health (NCC-WCH) of 17 studies, including randomized controlled trials (RCTs) and prospective and retrospective studies, reported that the sensitivity and specificity of detecting fetal structural anomalies before 24 weeks of gestation reported from the included studies were 24.1% (range 13.5–85.7%) and 99.92% (range 99.40–100.00%) respectively, while overall sensitivity and specificity were 35.4% (range 15.0–92.9%) and 99.86% (range 99.40–100.00%), respectively.¹ Meta-analysis of likelihood ratios showed positive and negative likelihood ratios before 24 weeks of 541.54 (95% CI 430.80–680.76) and 0.56 (95% CI 0.54–0.58), respectively. Meta-analysis of likelihood ratios showed overall positive and negative likelihood ratios were 242.89 (95% CI 218.35–270.18) and 0.65 (95% CI 0.63–0.66), respectively.

When these studies were assessed by criteria adopted by a Royal College of Obstetricians and Gynaecologists (RCOG) workgroup, ultrasound performed before 24 weeks gestation appeared to demonstrate detection rates for anomalies likely to be lethal of 83.6%, for anomalies associated with possible survival and long-term morbidity of 50.6%, for anomalies amenable to intrauterine therapy of 100% (but n = 3), and for anomalies associated with possible short-term immediate morbidity of 15.5%.²

Taken together, these studies, which were conducted across Europe and the United States, showed that second-trimester ultrasound demonstrates high specificity but poor sensitivity for identifying fetal structural anomalies, good summary value for positive likelihood ratio but poor negative likelihood ratio. However, the values reported in the assessed studies varied widely by centre and condition.¹

Few studies have examined routine anomaly screening in the first trimester in low-risk pregnancy populations, most data emanating from tertiary units managing high-risk pregnancies. One study published in 1999 was a prospective cross-sectional study at a university hospital in the UK, and included 6634 unselected women carrying 6443 fetuses. All women underwent either transabdominal or transvaginal sonography at 11–14 weeks. The study reported that incidence of anomalous fetuses was 1.4%, and sensitivity (detection rate) was 59.0% (37/63 (95% CI 46.5–72.4%). The specificity was 99.9%. Positive and negative likelihood ratios were 624.5 and 0.41 respectively. Although good-quality data are limited, this study suggested that the high specificity and positive likelihood ratio reported were moderated by modest sensitivity and negative likelihood ratio.³

In terms of clinical effectiveness, a systematic review has demonstrated that routine ultrasound screening for fetal anomalies is associated with improved pregnancy dating, detection of fetal abnormality, a threefold increase in termination of pregnancy for fetal abnormality (OR 3.19, 95% CI 1.54–6.60), and reductions in the number of undiagnosed twins and inductions for ‘post-term’ pregnancy.¹ No studies have as yet demonstrated improved long-term pregnancy outcomes, or any evidence of improved effectiveness of a routine first-trimester scan for detecting major fetal malformation compared with a routine second-trimester scan.

Antenatal detection of cardiac anomalies poses particular challenges and is probably the investigation that has been shown to be improved by coordinated training aimed at extending the standard views sought – from simply the four-chamber view to including the views of the cardiac outflow tracts as well as the drainage of the pulmonary veins. Employing such extended cardiac views when screening for cardiac anomalies also appears to be cost-beneficial when compared to screening approaches limited to demonstrating solely the four-chamber view. A few studies have shown that with conditions such as transposition of the great arteries (TGA), antenatal detection can improve neonatal survival and outcome by optimizing the birth and immediate neonatal care plans. It is therefore now recommended to aim to achieve a good four-chamber view of the fetal heart and outflow tracts during the conduct of the routine anomaly scan. While raised nuchal translucency is associated with a higher prevalence of cardiac anomaly in affected fetuses, it is not recommended as part of routine screening for cardiac anomalies as its predictive accuracy is not sufficient for widespread adoption.

The practice of screening for neural tube defects by maternal serum α-fetoprotein determination has now been largely replaced by direct ultrasound visualization of the neural tube, which has the capacity to detect more than 80% of significant lesions by a combination of direct visualization of the lesion and the visualization of the so-called ‘head signs’ of the lemon-shaped head and the banana-shaped cerebellum during the routine anomaly san at 18–21 weeks. Thus when routine ultrasound screening is performed to detect neural tube defects, α-fetoprotein testing is not required.