Predicting positive surgical margins in partial nephrectomy: A prospective multicentre observational study (the RECORd 2 project)

Abstract

Purpose: to evaluate clinical predictors of positive surgical margins (PSMs) in a large multicenter prospective observational study and to develop a clinic nomogram to predict the likelihood of PSMs after partial nephrectomy (PN). Materials and methods: We prospectively evaluated 4308 patients who had surgical treatment for renal tumors between January 2013 and December 2016 at 26 urological Italian Centers (RECORd 2 project). Two multivariable logistic models were evaluated to predict the likelihood of PSMs. Center caseload was dichotomized using a visual assessment adjusted for several predictors of PSMs. A nomogram predicting PSMs was developed. Results: Overall, 2076 patients treated with PN were evaluated. pT1a, pT1b, pT2 and pT3a were recorded in 68.7%, 22.6%, 2.1% and 6.6% of the patients, respectively. PSMs were recorded in 342 (16.5%) patients. From a null multivariable model against number of PN/year, 60 PN/year were identified as the best cut-off to define a high-volume centre. At multivariable analysis, clinical stage (cT1a vs. cT2 [OR 1.94]; p = 0.03), volume centre (≤60 PN/year) (OR 2.22; p < 0.0001), imperative vs elective indication (OR 2.10; p = 0.04), surgical technique (laparoscopic vs. open [OR 1.62; p = 0.002), lymphovascular invasion (OR 2.27; p = 0.01) and upstaging to pT3a (OR 2.81; p < 0.0001) were independent predictors of PSMs. The final nomogram included age, ASA score, Charlson score, clinical tumor stage, surgical indication, surgical approach, surgical technique, PADUA score, clamp procedure and volume centre. Conclusions: PSMs after PN were significantly more likely in patients with lower clinical stage, higher PADUA score, in individuals referred to laparoscopic PN and in those treated at lower volume centers. We used these data to develop a nomogram to predict such risk.