The effect of urinary diversion on long-term kidney function after cystectomy

•Large cystectomy cohort comparing long-term kidney function•First study to evaluate rate of change in kidney function with a longitudinal, linear mixed model•Neobladder diversion associated with lower rates of kidney function deterioration Cystectomy with urinary diversion is associated with decreased long-term kidney function due to several factors. One factor that has been debated is the type of urinary diversion used: ileal conduit (IC) vs. neobladder (NB). We tested the hypothesis that long-term kidney function will not vary by type of urinary diversion. We retrospectively identified all patients who underwent cystectomy with urinary diversion at our institution from January 1, 2007, to January 1, 2018. Data were collected on patient demographics, comorbid conditions, perioperative radiotherapy, and complications. Creatinine values were measured at several time points up to 120 months after surgery. Glomerular filtration rate (GFR) (ml/min per 1.73 m2) was calculated using the Chronic Kidney Disease Epidemiology Collaboration equation. A linear mixed model with inverse probability of treatment weighting (IPTW) was used to compare GFR between the IC and NB cohorts over time. Multiple sensitivity analyses were performed based on 2 different calculations of GFR (Chronic Kidney Disease Epidemiology Collaboration equation vs. Modification of Diet in Renal Disease), with and without excluding patients with preoperative GFR less than 40 ml/min per 1.73 m2. Among 563 patients who underwent cystectomy with urinary diversion, a NB was used for 72 (12.8%) individuals. Patients who had a NB were significantly younger, had a lower American Society of Anesthesiologists score, greater baseline GFR, better Eastern Cooperative Oncology Group performance status, lower median Charlson comorbidity index, and were less likely to have received preoperative abdominal radiation (all P < 0.05). Both NB and IC patients had decreased kidney function over time, with mean GFR losses at 5 years of 17% and 14% of baseline values, respectively. The IPTW-adjusted linear mixed model revealed that IC patients had slightly more deterioration in kidney function over time, but this was not statistically significant (estimate, 0.12; P = 0.06). The sensitivity analyses yielded a similar trend, in that GFR decrease appeared to be greater in the IC cohort. This trend was statistically significant when using Modification of Diet in Renal Disease (P = 0.04). Among highly selected patients with