Prediction of surgical difficulty in minimally invasive surgery for rectal cancer by use of MRI pelvimetry

Abstract

Background Technical difficulties in rectal surgery are often related to dissection in a limited surgical field. This study investigated the clinical value of MRI pelvimetry in the prediction of surgical difficulty associated with minimally invasive rectal surgery. Methods Patients with rectal cancer who underwent laparoscopic or robotic total mesorectal excision between 2005 and 2017 were reviewed retrospectively and categorized according to surgical difficulty on the basis of duration of surgery, conversion to an open procedure, use of the transanal approach, postoperative hospital stay, blood loss and postoperative complications. Preoperative clinical and MRI‐related parameters were examined to develop a prediction model to estimate the extent of surgical difficulty, and to compare anastomotic leakage rates in the low‐ and high‐grade surgical difficulty groups. Prognosis was investigated by calculating overall and relapse‐free survival, and cumulative local and distant recurrence rates. Results Of 121 patients analysed, 104 (86·0 per cent) were categorized into the low‐grade group and 17 (14·0 per cent) into the high‐grade group. Multivariable analysis indicated that high‐grade surgical difficulty was associated with a BMI above 25 kg/m2 (odds ratio (OR) 4·45, P = 0·033), tumour size 45 mm or more (OR 5·42, P = 0·042), anorectal angle 123° or more (OR 5·98, P = 0·028) and pelvic outlet less than 82·7 mm (OR 6·62, P = 0·048). All of these features were used to devise a four‐variable scoring model to predict surgical difficulty. In patients categorized as high grade for surgical difficulty, the anastomotic leakage rate was 53 per cent (9 of 17 patients), compared with 9·6 per cent (10 of 104) in the low‐grade group (P < 0·001). The high‐grade group had a significantly higher local recurrence rate than the low‐grade group (P = 0·002). Conclusion This study highlights the impact of clinical variables and MRI pelvimetry in the prediction of surgical difficulty in minimally invasive rectal surgery.