EAU 2018: Surgical Nomogram for Predicting the Likelihood of Postoperative Surgical Complications in Patients Treated with Partial Nephrectomy - The RECORd 2 Project

Copenhagen, Denmark (UroToday.com) The aim of this study was to analyse the predictive factors of surgical complications after partial nephrectomy (PN), and develop a surgical nomogram for predicting the likelihood of these complications.

The authors prospectively evaluated 4,308 patients who were surgically treated for renal tumors between January 2013 and December 2016 at 26 urological Italian Centers (RECORd 2 project). Postoperative complications occurring within 30 days of surgery were graded using the modified Clavien-Dindo scale. A multivariable logistic regression analysis for surgical complication was performed. The area under the receiving operator characteristic (ROC) curve (AUC) was used to quantify predictive discrimination. A nomogram was created from the multivariable model. Internal validation processes were performed using bootstrapping with 200 repetitions.

Overall, 2584 patients who underwent PN were evaluated for the final analyses. The median American Society of Anesthesiologists physical status (ASA PS) score was 2 (IQR 2-3). Clinical T1b and T2 were 24.3% and 3.3% of the cases. Median PADUA score was 7 (IQR 6-8). Overall, 34.3%, 27.7%, 38% of patients underwent open, laparoscopic and robotic PN, respectively. Enucleation was performed in 36.1% of patients. Postoperative surgical complications were recorded in 264 (10.2%) of patients: 5.9% were Clavien 2, 2.4% Clavien 3 and 0.2% Clavien 4a. 

At multivariable analysis age (p=0.01), ASA PS score (p=0.01), preoperative anaemia (p<0.0001), relative (p=0.67) and imperative (p=0.56) vs elective surgical indication, clinical T1b (p=0.93) and cT2 (p=0.01), preoperative creatinine (p=0.03), PADUA score (p=0.005), laparoscopic (p=0.01) and open (p<0.0001) vs robotic surgical approach and standard PN (p<0.65) vs enucleation, were evaluated to assess the risk of postoperative surgical complication. The AUC of the ROC curve of the model was 0.731. After internal validation using bootstrapping, it was 0.702. The final nomogram included age, ASA PS score, preoperative anaemia, surgical indication, Clinical T stage, PADUA score, surgical approach and surgical technique of resection (Figure 1). 

In conclusion, the authors developed and validated a nomogram for predicting postoperative surgical complications after PN using preoperative and surgical variables from a wide multicentre prospectively-recorded Italian dataset.

Figure 1 – Nomogram for predicting postoperative surgical complications after partial nephrectomy:


Presented by: Mari A., Careggi University Hospital, Dept. of Urology, Florence, Italy

Written by: Hanan Goldberg, MD, Urologic Oncology Fellow (SUO), University of Toronto, Princess Margaret Cancer Centre, twitter: @GoldbergHanan at the 2018 European Association of Urology Meeting EAU18, 16-20 March, 2018 Copenhagen, Denmark