Analysis of survival for patients with chronic kidney disease primarily related to renal cancer surgery

Abstract

© 2017 The Authors BJU International © 2017 BJU International Published by John Wiley & Sons Ltd Objectives: To evaluate predictors of long-term survival for patients with chronic kidney disease primarily due to surgery (CKD-S). Patients with CKD-S have generally good survival that approximates patients who do not have CKD even after renal cancer surgery (RCS), yet there may be heterogeneity within this cohort. Patients and Methods: From 1997 to 2008, 4 246 patients underwent RCS at our centre. The median (interquartile range [IQR]) follow-up was 9.4 (7.3–11.0) years. New baseline glomerular filtration rate (GFR) was defined as highest GFR between nadir and 6 weeks after RCS. We retrospectively evaluated three cohorts: no-CKD (new baseline GFR of ≥60 mL/min/1.73 m 2 ); CKD-S (new baseline GFR of <60 mL/min/1.73 m 2 but preoperative GFR of ≥60 mL/min/1.73 m 2 ); and CKD due to medical aetiologies who then require RCS (CKD-M/S, preoperative and new baseline GFR both <60 mL/min/1.73 m 2 ). Analysis focused primarily on non-renal cancer-related survival (NRCRS) for the CKD-S cohort. Kaplan–Meier analysis assessed the longitudinal impact of new baseline GFR (45–60 mL/min/1.73 m 2 vs <45 mL/min/1.73 m 2 ) and Cox regression evaluated relative impact of preoperative GFR, new baseline GFR, and relevant demographics/comorbidities. Results: Of the 4 246 patients who underwent RCS, 931 had CKD-S and 1 113 had CKD-M/S, whilst 2 202 had no-CKD even after RCS. Partial/radical nephrectomy (PN/RN) was performed in 54%/46% of the patients, respectively. For CKD-S, 641 patients had a new baseline GFR of 45–60 mL/min/1.73 m 2 and 290 had a new baseline GFR of <45 mL/min/1.73 m 2 . Kaplan–Meier analysis showed significantly reduced NRCRS for patients with CKD-S with a GFR of <45 mL/min/1.73 m 2 compared to those with no-CKD or CKD-S with a GFR of 45–60 mL/min/1.73 m 2 (both P ≤ 0.004), and competing risk analysis confirmed this (P < 0.001). Age, gender, heart disease, and new baseline GFR were all associated independently with NRCRS for patients with CKD-S (all P ≤ 0.02). Conclusion: Our data suggest that CKD-S is heterogeneous, and patients with a reduced new baseline GFR have compromised survival, particularly if <45 mL/min/1.73 m 2 . Our findings may have implications regarding choice of PN/RN in patients at risk of developing CKD-S.