External validation of the preoperative risk evaluation for partial nephrectomy (PREP) score.

Area of Science:

• Urology
• Surgical Oncology
• Robotic Surgery

Background:

• The preoperative risk evaluation for partial nephrectomy (PREP) score was developed to predict major complications after partial nephrectomy.
• External validation is crucial to assess the generalizability of risk prediction models in diverse patient populations and surgical settings.
• Robot-assisted partial nephrectomy (RAPN) is increasingly utilized, necessitating validated risk assessment tools specific to this approach.

Purpose of the Study:

• To externally validate the PREP score's performance in a large, multi-institutional cohort of patients undergoing robot-assisted partial nephrectomy (RAPN).
• To assess the discrimination and calibration of the PREP score in predicting major complications (Clavien-Dindo grade III-V) within 30 days post-RAPN.
• To identify specific patient characteristics associated with major complications in the RAPN cohort.

Main Methods:

• Retrospective analysis of 8,615 patients who underwent RAPN from the United States Kidney Cancer Data Network (US-KIDNET).
• Evaluation of PREP score's discrimination using the area under the receiver operating characteristic curve (AUC).
• Assessment of calibration using the Hosmer-Lemeshow (HL) goodness-of-fit test and logistic regression for individual component associations.

Main Results:

• The PREP score demonstrated poor discrimination (AUC = 0.547) and poor calibration (HL χ² = 150.2, p < 0.001) in the RAPN cohort.
• Performance was consistent across different robotic platforms and surgical approaches.
• Coronary artery disease, chronic obstructive pulmonary disease, chronic kidney disease, and obesity were independently associated with increased risk of major complications.

Conclusions:

• The PREP score, as currently formulated, does not perform well for predicting major complications in a large, multi-institutional cohort undergoing RAPN.
• Simple recalibration may be insufficient; coefficient re-estimation of the PREP score might be necessary for improved accuracy in robotic surgery settings.
• Specific comorbidities like CAD, COPD, CKD, and obesity are significant predictors of complications in RAPN patients.