ABSTRACT
OBJECTIVE: Audit the crown-rump length (CRL) measurements taken at 11 to 13 weeks scan, using operator-specific median multiples of the median (MoM) for pregnancy-associated plasma protein-A (PAPP-A) and free ß-human chorionic gonadotropin (ß-hCG) plots, to identify deviations potentially related to a systematic CRL bias. METHODS: Study population included consecutive singleton pregnancies undergoing first trimester combined screening, scanned by sonologists with at least 100 scans, during the 2011 to 2012 period. Previously described plots for PAPP-A and ß-hCG median MoM points, with their 95% confidence intervals circles, in relation with the expected deviation line were used. These plots have been modified to adjust the deviation line to the sonologist-specific expected MoM variation for each CRL millimetre bias according to each sonologist-specific median gestational ages at both blood sampling and ultrasound. RESULTS: Twenty-eight sonologists performing 9472 scans were included, accounting for 36% of the 77 sonologists and 70% of the 13 643 scans initially considered. Mean gestational age was 10 + 2 weeks at blood sampling and 12 + 4 weeks at ultrasound. Fifteen sonologists (53%) did not demonstrate any CRL bias, 10 (36%) present with a significant CRL underestimation, being above 2 mm in 6 (21%), and in 3 (11%) the observed deviation could not be explained by a systematic CRL bias. CONCLUSIONS: In sonologists with more than 100 NT measurements, operator-specific PAPP-A and ß-hCG median MoM plots are able to identify deviations potentially related to a systematic CRL bias. Systematic underestimation above 2 mm was observed in 1/5 of them. © 2016 John Wiley & Sons, Ltd.
