[Predictive value of MRI pelvic measurements for "difficult pelvis" during total mesorectal excision].

ABSTRACT

Objective: Total mesorectal resection (TME) is difficult to perform for rectal cancer patients with anatomical confines of the pelvis or thick mesorectal fat. This study aimed to evaluate the ability of pelvic dimensions to predict the difficulty of TME, and establish a nomogram for predicting its difficulty. Methods: The inclusion criteria for this retrospective study were as follows (1) tumor within 15 cm of the anal verge; (2) rectal cancer confirmed by preoperative pathological examination; (3) adequate preoperative MRI data; (4) depth of tumor invasion T1-4a; and (5) grade of surgical difficulty available. Patients who had undergone non-TME surgery were excluded. A total of 88 patients with rectal cancer who underwent TME between March 2019 and November 2021 were eligible for this study. The system for scaling difficulty was as follows Grade I, easy procedure, no difficulties; Grade II, difficult procedure, but no impact on specimen quality (complete TME); Grade III, difficult procedure, with a slight impact on specimen quality (near-complete TME); Grade IV very difficult procedure, with remarkable impact on specimen quality (incomplete TME). We classified Grades I-II as no surgical difficulty and grades III-IV as surgical difficulty. Pelvic parameters included pelvic inlet length, anteroposterior length of the mid-pelvis, pelvic outlet length, pubic tubercle height, sacral length, sacral depth, distance from the pubis to the pelvic floor, anterior pelvic depth, interspinous distance, and inter-tuberosity distance. Univariate and multivariate logistic regression analyses were performed to identify the factors associated with the difficulty of TME, and a nomogram predicting the difficulty of the procedure was established. Results: The study cohort comprised 88 patients, 30 (34.1%) of whom were classified as having undergone difficult procedures and 58 (65.9%) non-difficult procedures. The median age was 64 years (56-70), 51 patients were male and 64 received neoadjuvant therapy. The median pelvic inlet length, anteroposterior length of the mid-pelvis, pelvic outlet length, pubic tubercle height, sacral length, sacral depth, distance from the pubis to the pelvic floor, anterior pelvic depth, interspinous distance, and inter-tuberosity distance were 12.0 cm, 11.0 cm, 8.6 cm, 4.9 cm, 12.6 cm, 3.7 cm, 3.0 cm, 13.3 cm, 10.2 cm, and 12.2 cm, respectively. Multivariable analyses showed that preoperative chemoradiotherapy (OR=4.97,95% CI 1.25-19.71, P=0.023), distance between the tumor and the anal verge (OR=1.31, 95% CI 1.02-1.67, P=0.035) and pubic tubercle height (OR=3.36, 95% CI 1.56-7.25, P=0.002) were associated with surgical difficulty. We then built and validated a predictive nomogram based on the above three variables (AUC = 0.795, 95%CI 0.696-0.895). Conclusion: Our research demonstrated that our system for scaling surgical difficulty of TME is useful and practical. Preoperative chemoradiotherapy, distance between tumor and anal verge, and pubic tubercle height are risk factors for surgical difficulty. These data may aid surgeons in planning appropriate surgical procedures.