Objectives
First, to investigate the accuracy of transvaginal sonography (TVS) for presurgical evaluation of the distance between the most caudal part of the endometriotic lesion and the anal verge (lesion‐to‐anal‐verge distance (LAVD)) in women with rectosigmoid deep endometriosis (DE), compared with intraoperative measurement (IOM). Second, to assess the agreement between anastomosis height and LAVD measured using TVS.
Methods
This was a prospective observational multicenter study of symptomatic women who were scheduled for surgical treatment of rectosigmoid DE, by either discoid or segmental resection, between April 2017 and September 2019. Presurgical TVS was performed to evaluate the LAVD in two ways, depending on the level of the lesion. Method 1: for lesions at the level of the rectovaginal septum (RVS), the caudal part of the lesion was identified on TVS and an index finger was placed on the TVS probe at the level of the anal verge. The probe was withdrawn and the distance from the tip of the TVS probe down to the index finger was measured using a ruler, representing the LAVD. Method 2: for lesions above the RVS, the distance between the caudal part of the lesion and the lower lip of the posterior cervix was measured in a frozen image (LAVD‐1), and the distance between the lower lip of the posterior cervix and the anal verge (LAVD‐2) was measured using Method 1. These two measurements (LAVD‐1 and LAVD‐2) were added together and the result represented the total LAVD. During surgery, a rectal probe was used to perform IOM of LAVD, which was considered as the gold standard test. Agreement between LAVD measured using TVS and the IOM was assessed using Bland–Altman analysis. The intraclass correlation coefficient (ICC) for absolute agreement and Spearman's correlation coefficient were also calculated. Systematic and proportional bias were tested for significance using the paired t‐test. Similar analysis was performed to assess agreement between LAVD measured using TVS and anastomosis height.
Results
A total of 147 consecutive women were considered eligible for inclusion. Fourteen women were excluded initially. Thirty‐four discoid resections and 102 segmental resections were performed; both procedures were performed in three women. Two more women were excluded from the final analysis because the measurements represented extreme outliers. The mean LAVD measured using TVS was 114.8 ± 36.5 mm and the mean IOM was 116.9 ± 42.3 mm. There was no statistically significant difference between LAVD measured using TVS and IOM (mean difference, –2.12 mm (95% CI, –6.33 to 2.05 mm); P = 0.32). Bland–Altman analysis showed that there was good agreement between the two methods. The ICC was 0.81 (95% CI, 0.74–0.86) and Spearman's correlation coefficient was 0.68 (95% CI, 0.56–0.77). The mean difference between LAVD measured using TVS and anastomosis height was statistically, but not clinically, significant (mean difference, 10.25 mm (95% CI, 5.94–14.32 mm); P = 0.0005), and the ICC was 0.78 (95% CI, 0.66–0.85).
Conclus...