Comparison between trans-vaginal and trans-abdominal ultrasound examination of the cervix in the second trimester of pregnancy: a prospective study.

Abstract

To compare transvaginal (TV) and trans-abdominal (TA) ultrasound assessment of cervical length (CL) at the time of the second-trimester scan for fetal anomalies. This was a prospective study including consecutive pregnant women attending the low-risk ultrasound clinic of two fetal medicine centres in Italy. The inclusion criteria were women between 19+0 and 22+0weeks of gestation, attending the prenatal ultrasound clinic for the routine second trimester screening for fetal anomalies. The primary outcome was to compare the CL measurement obtained at TV compared to TA ultrasound; the secondary outcome was to report the inter and intra-observer variability of CL measured with the two different approaches. All women underwent TV and TA assessment of the cervix performed by two experienced certified operators, blinded to each other. Intra-class correlation coefficients (ICC) and Bland-Altman analyses were used to analyse the data. Two hundred and fifty women were included in the analysis. All women had anteverted uterus. The mean gestational age at ultrasound was 20.7±0.7weeks; 1.2 % (3/250) scans were performed at 19weeks, 49.2 % (123/250) at 20weeks, 44.8 % (112/250) at 21weeks and 4.8 % (12/250) at 22weeks of gestations. Identification of the major landmarks of CL at TA ultrasound was achieved in all the included cases. There was good reliability between CL measured at TA (ICC 0.95, 95 % CI 0.93-0.97 for observer 1 and 0.92 %, 95 % CI0.89-0.94 for observer 2) and TV ultrasound 0.97, 95 % CI0.96-0.98 for observer 1 and 0.96, 95 % CI 0.95-0.97 for observer 2). There was also good reliability between the two observers for both the TA and TV assessment of the CL. Mean TA CL was 41.4±5.5 for observer 1 and 40.5±4.8 for observer 2 with no significant differences between the two measurements (mean difference 0.92 mm, 95 % CI-9.7 to 11.2). Likewise, there was no difference between the CL measured at TV ultrasound between the two observers (mean difference-0.83 mm, 95 % CI-5.97 to 4.30). Finally, there was no difference in the mean CL measured at TA compared to TV, either considering the overall population of women (mean difference:-0.43, 955 CI-8.65 to 7.79), or when stratifying the analysis according to the parity status and the operator. Among experienced operators, there was no difference between TV and TA ultrasound assessment of the CL at the time of the routine anomaly scan for fetal anomaly.