Using CT-Based Pelvimetry and Visceral Obesity Measurements to Predict Total Mesorectal Excision Quality for Patients Undergoing Rectal Cancer Surgery

BACKGROUND: A complete total mesorectal excision is the criterion standard in curative rectal cancer surgery. Ensuring quality is challenging in a narrow pelvis, and obesity amplifies technical difficulties. Pelvimetry is the measurement of pelvic dimensions, but its role in gauging preoperatively the difficulty of proctectomy is largely unexplored. OBJECTIVE: To determine pelvic structural factors associated with incomplete total mesorectal excision after curative proctectomy and build a predictive model for total mesorectal excision quality. DESIGN: Retrospective cohort study. SETTING: A quaternary referral center database of patients diagnosed with rectal adenocarcinoma (2009-2017). PATIENTS: Curative-intent proctectomy for rectal adenocarcinoma. INTERVENTIONS: All radiological measurements were obtained from preoperative CT images using validated imaging processing software tools. Completeness of total mesorectal excision was obtained from histology reports. MAIN OUTCOME MEASURES: Ability of radiological pelvimetry and obesity measurements to predict total mesorectal excision quality. RESULTS: Of the 410 cases meeting inclusion criteria, 362 underwent a complete total mesorectal excision (88%). Multivariable regression identified a deeper sacral curve (per 100 mm2 [OR: 1.14; 95% CI, 1.06-1.23; p < 0.001]) and a greater transverse distance of the pelvic outlet (per 10 mm [OR:1.41, 95% CI, 1.08-1.84; p = 0.012]) to be independently associated with incomplete total mesorectal excision. An increased area of the pelvic inlet (per 10 cm2 [OR: 0.85; 95% CI, 0.75-0.97; p = 0.02) was associated with a higher rate of complete mesorectal excision. No difference in visceral obesity ratio and visceral obesity (ratio >0.4 vs <0.4) between BMI (<30 vs >= 30) and sex was identified. A model was built to predict mesorectal quality using the following variables: depth of sacral curve, area of pelvic inlet, and transverse distance of the pelvic outlet. LIMITATIONS: Retrospective analysis is not controlled for the choice of surgical approach. CONCLUSIONS: Pelvimetry predicts total mesorectal excision quality in rectal cancer surgery and can alert surgeons preoperatively to cases of unusual difficulty. This predictive model may contribute to treatment strategy and aid in the comparison of outcomes between traditional and novel techniques of total mesorectal excision.