END STAGE RENAL DISEASE AFTER SURGERY IN PATIENTS WITH NORMAL PREOPERATIVE KIDNEY FUNCTION: THE EFFECT OF NEPHRON-SPARING SURGERY IN DELAYING THE ONSET OF THE DISEASE

NTRODUCTION AND OBJECTIVES: In a recent subanalysis of 514 patients included in the EORTC randomized trial 30904, the incidence of End Stage Renal Disease (ESRD) was nearly identical between patients treated with nephron sparing surgery (NSS) or radical nephrectomy (RN). In the current paper, we aimed to report the rate and the predictors of ESRD after NSS vs. RN after accounting for clinical characteristics, comorbidities and individual patients’ cardiovascular risk. METHODS: A multi-institutional collaboration among five Tertiary Care Centers allowed collecting 2029 patients with a clinical T1a- T1b renal mass. Patients underwent RN (n¼693, 34.2%) or NSS (n¼1336, 65.8%) and showed normal estimated glomerular filtration rates (eGFR) before surgery (defined as a pre-operative eGFR60ml/ min/1.73m2). Descriptive, univariable and multivariable Cox regression analyses were used to predict the risk of ESRD (defined as postoperative eGFR <15ml/min/1.73m2). To adjust for inherent baseline differences among patients, we included as covariates: age, pre-operative GFR, clinical tumor size, hypertension (none vs. yes vs. controlled by medical therapy), diabetes, baseline Charlson comorbidity index (CCI), body mass index and smoker status. RESULTS: Median preoperative GFR was 90ml/min/1.73m2 (interquartile range, IQR 79-99). Median age was 61 yrs (IQR 52-69)and median clinical tumor size 3.5 cm (IQR 2.5-5). Overall, 10.0% vs. 18.2% vs. 21.3% of the patients had diabetes, uncontrolled or hypertension controlled by medical therapy, respectively. The 5 yr, 10 yr and 15 yr ESRD rates after surgery were 1.6%, 2.6% and 2.6% for NSS vs. 2.1%, 2.7% and 5.1% for RN, respectively (p¼0.5). At multivariable analyses, after accounting for clinical characteristics, comorbidities and individual cardiovascular risk, patients with hypertension not controlled by medical therapy (HR 4.2, p¼0.02) and those with more than 2 comorbidities (HR 8.5, p¼0.009) showed significantly higher risk to develop ESRD. Patients who underwent NSS showed virtually the same risk to develop ESRD relative to their RN-treated counterparts (HR 1.04, p¼0.9). However, mean time to ESRD was significantly higher in patients treated with NSS relative to RN (85 vs. 43 months, bootstrapped pvalue ¼0.04). CONCLUSIONS: Roughly 2% of the patients with normal GFR before kidney surgery will develop ESRD, with virtually no difference between those treated with NSS or RN. Interestingly, NSS showed a statistically significant effect in delaying the onset of ESRD relative to RN.