Partial versus radical nephrectomy in very elderly patients: a propensity score analysis of surgical, functional and...

To compare the outcomes of PN to those of RN in very elderly patients treated for clinically localized renal tumor.

A purpose-built multi-institutional international database (RESURGE project) was used for this retrospective analysis. Patients over 75 years old and surgically treated for a suspicious of localized renal with either PN or RN were included in this database. Surgical, renal function and oncological outcomes were analyzed. Propensity scores for the predicted probability to receive PN in each patient were estimated by logistic regression models. Cox proportional hazard models were estimated to determine the relative change in hazard associated with PN vs RN on overall mortality (OM), cancer-specific mortality (CSM) and other-cause mortality (OCM).

A total of 613 patients who underwent RN were successfully matched with 613 controls who underwent PN. Higher overall complication rate was recorded in the PN group (33% vs 25%; p = 0.01). Median follow-up for the entire cohort was 35 months (interquartile range [IQR] 13-63 months). There was a significant difference between RN and PN in median decline of eGFR (39% vs 17%; p < 0.01). PN was not correlated with OM (HR = 0.71; p = 0.56), OCM (HR = 0.74; p = 0.5), and showed a protective trend for CSM (HR = 0.19; p = 0.05). PN was found to be a protective factor for surgical CKD (HR = 0.28; p < 0.01) and worsening of eGFR in patients with baseline CKD. Retrospective design represents a limitation of this analysis.

Adoption of PN in very elderly patients with localized renal tumor does not compromise oncological outcomes, and it allows better functional preservation at mid-term (3-year) follow-up, relative to RN. Whether this functional benefit translates into a survival benefit remains to be determined.