Introduction Renal cell carcinoma is the 14th most common cancer worldwide, accounting for nearly 2% of all malignancies, with the highest incidence in Western countries. The widespread use of imaging has increased incidental detection of small renal masses, allowing earlier diagnosis and promoting nephron-sparing strategies. Partial nephrectomy is now the preferred treatment for localized renal tumours when feasible, as it provides good oncological control while preserving renal function. Robotic-assisted surgery has further expanded its use by improving precision and visualization, even in complex cases. However, tumour anatomy remains a major determinant of surgical difficulty and perioperative outcomes. Several nephrometry scores, including PADUA and R.E.N.A.L., have been developed to standardize tumour complexity assessment. Although widely used, these systems may be relatively complex in routine practice. The RPN score (Radius, Position, iNvasion) was proposed as a simplified alternative based on three key anatomical variables. Study aim The aim of this study was to externally validate the RPN nephrometry score compared with PADUA and R.E.N.A.L. in predicting perioperative outcomes after robot-assisted partial nephrectomy (RAPN). Materials and methods Data from patients undergoing RAPN with the Da Vinci system between 2020 and 2025 at a single university hospital were retrospectively analyzed from a prospectively maintained database. Only patients with complete imaging and clinical data allowing calculation of all three scores were included. Outcomes assessed were estimated blood loss (EBL), warm ischemia time (WIT), prolonged operative time, major complications (Clavien-Dindo), and trifecta achievement, defined as no complications, WIT <20 minutes, and negative surgical margins. Predictive accuracy was evaluated using R² and RMSE, discrimination through ROC curves, and score comparisons with the DeLong test. Decision curve analysis (DCA) assessed clinical net benefit. Statistical significance was set at p<0.05. Results A total of 376 patients were included. The most common PADUA score was 7–8 (46%), while the most frequent R.E.N.A.L. score was 6–7 (45%). Median operative time was 130 minutes, median EBL was 100 mL, and median WIT was 13 minutes. Overall, 79% of patients had WIT <20 minutes. Postoperative complications occurred in 23% of patients, with major complications in 4.3%. Trifecta was achieved in 59% of cases. All three scores showed moderate discriminative ability for EBL, WIT, operative time, major complications, and trifecta, with no significant differences in predictive performance (p>0.05). DCA showed that RPN had a positive net benefit comparable to PADUA and R.E.N.A.L. only for EBL and operative time prediction. For WIT, PADUA performed best, while RPN had the lowest benefit. For major complications, R.E.N.A.L. showed the highest benefit and RPN the lowest. For trifecta, RPN and PADUA were similar, whereas R.E.N.A.L. performed better at threshold probabilities between 50% and 75%. Conclusions The simplified RPN score showed non-inferior clinical utility compared with PADUA and R.E.N.A.L. for predicting blood loss and operative time. Its three-variable design also demonstrated comparable discrimination across outcomes, making it a practical alternative to more complex systems. However, PADUA and R.E.N.A.L. remained superior for predicting prolonged WIT, major complications, and trifecta achievement.
Introduzione Il carcinoma a cellule renali rappresenta circa il 2% di tutte le neoplasie ed è più frequente nei Paesi occidentali. Negli ultimi anni, la diffusione delle tecniche di imaging ha aumentato il riscontro incidentale di piccole masse renali, favorendo diagnosi più precoci e approcci conservativi. La nefrectomia parziale è il trattamento di scelta per i tumori renali localizzati quando tecnicamente fattibile, poiché garantisce un buon controllo oncologico preservando la funzione renale. L’introduzione della chirurgia robot-assistita ha ampliato ulteriormente le indicazioni grazie a maggiore precisione, destrezza e qualità della visione, anche nei casi più complessi. Tuttavia, la difficoltà chirurgica varia in base alle caratteristiche anatomiche del tumore, che influenzano in maniera significativa gli esiti perioperatori. Per standardizzare la valutazione della complessità tumorale sono stati sviluppati diversi score nefrometrici, tra cui PADUA e R.E.N.A.L. Sebbene ampiamente utilizzati, questi sistemi possono risultare complessi e poco pratici nella routine clinica. Lo score RPN (Raggio, Posizione, iNvasione del seno renale) è stato proposto come alternativa semplificata basata su tre variabili anatomiche essenziali. Scopo dello studio Valutare la validità esterna dello score RPN rispetto ai sistemi PADUA e R.E.N.A.L. nella previsione degli esiti perioperatori della nefrectomia parziale robot-assistita (RAPN). Materiali e metodi Sono stati analizzati i dati di pazienti sottoposti a RAPN con sistema Da Vinci tra il 2020 e il 2025 presso un singolo ospedale universitario, raccolti in un database prospettico. Sono stati inclusi esclusivamente i casi con dati clinici e radiologici completi, necessari per il calcolo dei tre score. Gli endpoint perioperatori considerati comprendevano: perdita ematica stimata (EBL), tempo di ischemia calda (WIT), durata operatoria prolungata, complicanze maggiori secondo Clavien-Dindo e raggiungimento della trifecta, definita come assenza di complicanze, WIT <20 minuti e margini chirurgici negativi. L’accuratezza predittiva è stata valutata mediante R² e RMSE, la capacità discriminatoria con curve ROC e il confronto tra score con test di DeLong. L’analisi decisionale (DCA) ha stimato il beneficio clinico netto. La significatività statistica è stata fissata a p<0,05. Risultati Sono stati inclusi 376 pazienti. Il punteggio PADUA più frequente era 7–8 (46%), mentre il R.E.N.A.L. più comune era 6–7 (45%). La durata mediana dell’intervento era di 130 minuti, la perdita ematica mediana di 100 mL e il WIT mediano di 13 minuti. Il 79% dei pazienti presentava un WIT inferiore a 20 minuti.Le complicanze postoperatorie si sono verificate nel 23% dei casi, ma solo il 4,3% erano complicanze maggiori. La trifecta è stata raggiunta nel 59% dei pazienti. Tutti e tre gli score hanno mostrato una moderata capacità discriminatoria per EBL, WIT, durata operatoria, complicanze maggiori e trifecta, senza differenze statisticamente significative tra loro. L’analisi DCA ha evidenziato che RPN offriva un beneficio clinico comparabile a PADUA e R.E.N.A.L. solo nella previsione della perdita ematica e della durata operatoria prolungata. Per il WIT, PADUA mostrava il miglior beneficio netto, mentre RPN risultava il meno performante. Per le complicanze maggiori, R.E.N.A.L. era superiore e RPN inferiore. Per la trifecta, RPN e PADUA erano sovrapponibili, mentre R.E.N.A.L. risultava migliore tra probabilità soglia del 50–75%. Conclusioni Lo score RPN semplificato ha mostrato un’utilità clinica non inferiore rispetto a PADUA e R.E.N.A.L. nella previsione della EBL e della durata chirurgica. La struttura a tre variabili lo rende uno strumento pratico, con capacità discriminatoria complessivamente comparabile. Tuttavia, PADUA e R.E.N.A.L. restano superiori nella previsione di WIT prolungato, complicanze maggiori e raggiungimento della trifecta.
EXTERNAL VALIDATION OF THE RADIUS, POSITION OF TUMOR, INVASION OF RENAL SINUS (RPN) SCORE FOR SURGICAL OUTCOMES OF ROBOT-ASSISTED PARTIAL NEPHRECTOMY PERFORMED WITH THE DA VINCI ROBOTIC SYSTEM
PADOVESE, ELENA
2025/2026
Abstract
Introduction Renal cell carcinoma is the 14th most common cancer worldwide, accounting for nearly 2% of all malignancies, with the highest incidence in Western countries. The widespread use of imaging has increased incidental detection of small renal masses, allowing earlier diagnosis and promoting nephron-sparing strategies. Partial nephrectomy is now the preferred treatment for localized renal tumours when feasible, as it provides good oncological control while preserving renal function. Robotic-assisted surgery has further expanded its use by improving precision and visualization, even in complex cases. However, tumour anatomy remains a major determinant of surgical difficulty and perioperative outcomes. Several nephrometry scores, including PADUA and R.E.N.A.L., have been developed to standardize tumour complexity assessment. Although widely used, these systems may be relatively complex in routine practice. The RPN score (Radius, Position, iNvasion) was proposed as a simplified alternative based on three key anatomical variables. Study aim The aim of this study was to externally validate the RPN nephrometry score compared with PADUA and R.E.N.A.L. in predicting perioperative outcomes after robot-assisted partial nephrectomy (RAPN). Materials and methods Data from patients undergoing RAPN with the Da Vinci system between 2020 and 2025 at a single university hospital were retrospectively analyzed from a prospectively maintained database. Only patients with complete imaging and clinical data allowing calculation of all three scores were included. Outcomes assessed were estimated blood loss (EBL), warm ischemia time (WIT), prolonged operative time, major complications (Clavien-Dindo), and trifecta achievement, defined as no complications, WIT <20 minutes, and negative surgical margins. Predictive accuracy was evaluated using R² and RMSE, discrimination through ROC curves, and score comparisons with the DeLong test. Decision curve analysis (DCA) assessed clinical net benefit. Statistical significance was set at p<0.05. Results A total of 376 patients were included. The most common PADUA score was 7–8 (46%), while the most frequent R.E.N.A.L. score was 6–7 (45%). Median operative time was 130 minutes, median EBL was 100 mL, and median WIT was 13 minutes. Overall, 79% of patients had WIT <20 minutes. Postoperative complications occurred in 23% of patients, with major complications in 4.3%. Trifecta was achieved in 59% of cases. All three scores showed moderate discriminative ability for EBL, WIT, operative time, major complications, and trifecta, with no significant differences in predictive performance (p>0.05). DCA showed that RPN had a positive net benefit comparable to PADUA and R.E.N.A.L. only for EBL and operative time prediction. For WIT, PADUA performed best, while RPN had the lowest benefit. For major complications, R.E.N.A.L. showed the highest benefit and RPN the lowest. For trifecta, RPN and PADUA were similar, whereas R.E.N.A.L. performed better at threshold probabilities between 50% and 75%. Conclusions The simplified RPN score showed non-inferior clinical utility compared with PADUA and R.E.N.A.L. for predicting blood loss and operative time. Its three-variable design also demonstrated comparable discrimination across outcomes, making it a practical alternative to more complex systems. However, PADUA and R.E.N.A.L. remained superior for predicting prolonged WIT, major complications, and trifecta achievement.| File | Dimensione | Formato | |
|---|---|---|---|
|
Tesi Elena Padovese .pdf
accesso aperto
Dimensione
2.11 MB
Formato
Adobe PDF
|
2.11 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/109079