Introduction

Despite the association between nephrometry scores and the risk of complications following nephron sparing surgery [NSS] is established, data regarding predictive accuracy [PA] and clinical net-benefit [NB] of nephrometry scores-based prognostication are scarce. The aim of the study was to perform a head-to-head comparison of PADUA score-, RENAL score- and clinical tumour size-based prognostication of complications following NSS.

Methods

214 patients diagnosed with a cT1-2 cN0 cM0 renal mass elected for NSS and collected into a prospective database were assessed. The outcomes of the study were overall and Clavien-Dindo [CD] grade ≥2 complication rates. Logistic regression analyses predicting overall and CD grade ≥2 complications were fitted. PA was estimated with C index. NB was evaluated with decision curve analysis.

Results

The overall complications rate was 36%. The CD grade ≥2 complications rate was 20%. The most common specific complication was anemia/haematoma. At logistic regression analysis, PADUA score resulted associated with higher risk of overall complications rate (Odds ratio [OR] 1.24; 95% Confidence intervals [CI] 1.03-1.50; p=0.03). Conversely, RENAL score (OR 1.16; CI 0.95-1.41; p=0.1) and clinical tumour size were not associated with the risk of overall complications (OR 1.21; CI 0.99-1.47; p=0.058). Moreover, PADUA score, RENAL score and clinical tumour size were not associated with the risk of CD grade ≥2 complications (p>0.05). Harrel&[prime]s C index resulted 0.59 for PADUA score-, 0.57 for RENAL score- and 0.59 for clinical tumour size-based prediction. At decision curve analysis, the recorded NB was consistently low regardless of the prediction strategy (Figure 1).

Conclusions

PADUA and RENAL score were not associated with the risk of clinically relevant complications after NSS. Moreover, their ability to predict complications is diagnostically inadequate. Neither PADUA nor RENAL score were superior to the use of clinical tumour size only with respect to PA or NB. These findings suggest that, despite PADUA and RENAL scores represent accurate classification systems for the anatomic complexity of renal tumours, none of them is an optimal predictor of clinical outcomes.

Funding

none