Objectives: To evaluate the inter- and intra-observer reproducibility of a technique for examining physical characteristics of the uterine cervix in the third trimester of pregnancy using transvaginal ultrasound scanning (TVs). Study design: Forty-six women of over 27-week gestation underwent TVs of the cervix using a Toshiba 140 scanner with a 6 MHz probe. Twenty were scanned once and 14 women twice by the same operator, and another 12 women by two different operators. Measurements of cervical length, including length and width and opening of the internal os were taken from hardcopy images. Of the 26 women scanned twice, sets of images were analysed by a blinded single observer. Of the 20 women scanned once, measurements from the hardcopy images were taken by two blinded observers. Results: The mean difference in measurement of the distance from the tip of the cervix to foetal head was 2.3 mm for two hardcopy image observers,-1.96 mm for two scans by the same operator and - 1.52 mm for two scanners. Corresponding differences in width of opening of the internal os were 1.8, 0.44 and 0.32 mm, respectively, Interpretation of two images taken on each occasion gave similar results to those taken from three or more images. Conclusions: This method of obtaining TVs measurements from the cervix from hardcopy images that has satisfactory intra- and inter-observer error. It remains to be established whether this method will be better than Bishop score in predicting the outcome of induced or spontaneous labour. The technique appears to be suited for use in multicentre trials of TVs of the cervix at term. (C) 2002 Elsevier Science Ireland Ltd. All rights reserved.
|Pages (from-to)||31 - 35|
|Number of pages||5|
|Journal||European Journal of Obstetrics Gynecology and Reproductive Biology|
|Publication status||Published - 10 Oct 2002|