Test characteristics of placental location screening by transabdominal ultrasound at 18–20 weeks




To determine the test characteristics of a second-trimester transabdominal fetal anomaly scan in screening for placenta previa.

Patients and methods

This was a case-control study carried out at a tertiary obstetric hospital involving 54 women (0.5% of the total number booked for antenatal care and delivery at the center) with clinically significant placenta previa at the time of delivery, and 168 randomly selected controls who attended for antenatal care and delivery in 2000–2003. Sensitivity, specificity and false positive rate of the placenta–os measurement in the second-trimester transabdominal fetal anomaly scan for identification of clinically significant placenta previa were assessed.


Women with placenta previa at delivery had significantly different second-trimester placenta–os measurements from those of controls (P < 0.0001). For women with placenta previa, the mean (standard deviation) placenta–os measurement was −0.96 cm (1.95) vs. 4.15 cm (1.77) for controls (P < 0.0001). Using the anomaly scan as a screening test for placenta previa with a placenta–os measurement cut-off point of ≤ 2 cm, all cases of placenta previa should be identified, but 11.1% (95% CI, 7.1–14.8) of women without placenta previa would have false positive results. At cut-off points of 3 cm and 1.5 cm the false positive rates were 25.7% (95% CI, 18.8–30.2%) and 6.6% (95% CI, 3.8–9.6%), respectively, but at 1.5 cm 3.8% of cases would be missed.


The second-trimester transabdominal fetal anomaly scan is a useful screening test for placenta previa. However, because of false positives among the much larger population of women without placenta previa, we estimate that with a cut-off point of ≤2 cm, 23 women would have to be followed up to confirm one true placenta previa. Copyright © 2006 ISUOG. Published by John Wiley & Sons, Ltd.