Validation of a pelvic surgery difficulty risk model to predict difficult pelvic dissection and poor outcomes.

BACKGROUND: We previously developed the Pelvic Surgery Difficulty Index for predicting intraoperative events and postoperative outcomes associated with rectal mobilization with or without proctectomy ("deep pelvic dissection"). The aim of this study was to validate the scoring system as a prognostic tool for outcomes of pelvic dissection, regardless of the cause of dissection.

METHODS: Consecutive patients who underwent elective deep pelvic dissection at our institution from 2009 to 2016 were reviewed. Pelvic Surgery Difficulty Index score (0-3) was calculated from the following parameters: male sex (+1), prior pelvic radiotherapy (+1), and linear distance from sacral promontory to pelvic floor >13 cm (+1). Patient outcomes stratified by Pelvic Surgery Difficulty Index score were compared. The outcomes assessed included operative blood loss, operative time, length of hospital stay, cost, and postoperative complications.

RESULTS: A total of 347 patients were included. Higher Pelvic Surgery Difficulty Index scores were associated with significantly more blood loss, operative time, postoperative complications, hospital costs, and hospital stay. The model achieved good discrimination with area under the curve ≥0.7 for most outcomes.

CONCLUSION: Preoperative prediction of the morbidity associated with difficult pelvic dissection is possible with an objective, feasible, and validated model. Such a tool may facilitate preoperative preparation and allow for better risk stratification and uniform quality control across centers.