Development and Validation of a Nomogram Predicting Intraoperative Adverse Events During Robot-assisted Partial...

Area of Science:

• Urology
• Surgical Oncology
• Medical Informatics

Background:

• Predicting intraoperative adverse events (IOAEs) in partial nephrectomy (PN) is clinically significant.
• Robot-assisted PN (RAPN) is a common procedure for kidney cancer.
• Accurate risk prediction can optimize patient selection and surgical planning.

Purpose of the Study:

• To develop and internally validate a preoperative nomogram for predicting IOAEs in patients undergoing RAPN.
• Identify key predictors of intraoperative complications in RAPN.
• Assess the clinical utility of the developed predictive model.

Main Methods:

• Observational study using data from the Vattikuti Collective Quality Initiative (VCQI) database.
• Backward stepwise logistic regression identified independent predictors of IOAEs.
• Nomogram validation included bootstrapping, AUC, and goodness-of-fit testing; Decision Curve Analysis (DCA) assessed clinical utility.

Main Results:

• A total of 2114 patients were analyzed, with 7.5% experiencing IOAEs.
• Five independent predictors of IOAEs were identified: RENAL nephrometry score, clinical tumor size, PN indication, Charlson comorbidity index, and multifocal tumors.
• The nomogram demonstrated good internal validity with an AUC of 0.76 and was clinically useful at threshold probabilities >5%.

Conclusions:

• A validated preoperative nomogram for predicting IOAEs in RAPN was successfully developed.
• The nomogram can aid in identifying patients at higher risk for intraoperative complications.
• This tool can support clinical decision-making regarding the suitability of RAPN for individual patients.