Renal function change after radical cystectomy for urothelial carcinoma patients with a solitary kidney may be independent of urinary diversion type

Bladder cancer is the tenth most commonly diagnosed cancer according to the 2020 world statistics for men and women [1]. The national cancer registration statistics of South Korea in 2019 also reported bladder cancer as the tenth most prevalent cancer (5-year survival rate, 87.6%) in South Korea for both sexes, and 72.5% of affected male patients are diagnosed at localized stages [2]. Of the bladder cancers, muscle-invasive bladder cancer has a poor prognosis with high morbidity and mortality, and the standard treatment comprises radical cystectomy (RC) and urinary diversion (UD) [3, 4, 5]. Although RC with UD is the treatment of choice, the operation lowers the quality of life because of surgery-related complications and comorbidities [6, 7]. In particular, most of the late complications (3 months after RC) are related to UD; therefore, surgeons should consider preoperative factors before determining the UD type [8, 9]. Before deciding on the type of diversion, patient age, medical history (including neurological and psychiatric illnesses), performance status, life expectancy, and previous abdominal operation or radiation history are considered [10].

Based on previous studies, complex forms of UD for patients with impaired renal function or a single kidney are not recommended [8, 11, 12]. However, a previous retrospective study demonstrated that the likelihood of estimated glomerular filtration rate (eGFR) decline did not significantly differ according to UD type in patients with compromised baseline renal function [13]. Many surgeons do not consider a neobladder (NB) when performing RC for patients with a single kidney because of the presumption that NB may further decrease renal function than an ileal conduit (IC). However, there is a lack of evidence. Therefore, we aimed to compare the change in renal function after RC in patients with a single kidney, between the IC and NB groups.

In total (n=86), 62.8% (54/86) of patients underwent IC, whereas 37.2% (32/86) underwent NB. Baseline patient characteristics (Table 1) were similar between the two groups, except for age and body mass index (BMI). Patients in the IC group were older than those in the NB group: IC (70.35±9.05) vs. NB (65.63±7.98), p=0.017. BMI was statistically higher in the NB group than in the IC group: IC (22.36±2.47) vs. NB (23.61±3.10), p=0.043. Preoperative renal function, smoking history, operation type (open, laparoscopic, or robotic), previous transurethral resection of the bladder, nephrectomy history, cystectomy pathology, and neoadjuvant/adjuvant chemotherapy and radiation therapy did not differ between the two groups.

Table 1
Patient demographics of ileal conduit and neobladder

• Click for larger image
• Click for full table
• Download as Excel file

The renal function of the two groups showed no statistical difference in the postoperative creatinine in terms of the mean and SD at 1, 3, 6, 12, 24, 36, 48, and 60 months: p=0.122, 0.401, 0.601, 0.675, 0.102, 0.091, 0.083, and 0.482, respectively. Similar results were observed for the postoperative eGFR (CKD-EPI) in mean±SD at 1, 3, 6, 12, 24, 36, 48, and 60 months (Fig. 1): p=0.674, 0.903, 0.295, 0.873, 0.114, 0.201, 0.211, and 0.758, respectively. The eGFR difference values (mean±SD), compared between the IC and NB groups during the same period, were 0.267, 0.852, 0.946, 0.236, 0.645, 0.778, and 0.706, respectively. The eGFR difference proportion (eGFR difference value/preoperative eGFR; expressed as mean±SD) at 1, 3, 6, 12, 24, 36, 48, and 60 months showed similar results: p=0.144, 0.260, 0.939, 0.643, 0.076, 0.382, 0.434, and 0.895, respectively. No significant difference was observed between the IC and NB groups even after a two-way repeated ANOVA test (absolute eGFR [CKD-EPI], p=0.817; GFR difference, p=0.332; and GFR difference proportion p=0.453).

Fig. 1
Post-radical cystectomy eGFR at 1, 3, 6, 12, 24, 36, 48, and 60 months. Renal function change between the ileal conduit and neobladder groups was evaluated by post-operative eGFR using the CKD-EPI equation at 1, 3, 6, 12, 24, 36, 48, and 60 months. There was no significant difference between the two groups in each follow-up time by the t-test and two-way repeated ANOVA test. eGFR, estimated glomerular filtration rate; CKD-EPI equation, Chronic Kidney Disease Epidemiology equation.

• Click for larger image
• Download as PowerPoint slide

Fig. 2 shows the time-to-event analysis for eGFR decline-free survival, determined postoperatively as 10 points decline in eGFR, compared to the preoperative value. The log-rank p-value between the IC and NB groups showed no statistical difference at 0.108. Other definitions of eGFR decline-free survival, using postoperative 10% or 20% decline in eGFR compared to preoperative values, are presented in Fig. 2. No significant differences were observed in the 10% or 20% decline in eGFR decline-free survival definitions: 10% decline (p=0.067) and 20% decline (p=0.151). Similar results were observed in the eGFR of MDRD; there were no significantly different results between the IC and NB groups when compared with the previously mentioned eGFR decline-free survival definitions.

Fig. 2
Post-radical cystectomy eGFR decline-free survival. Multiple definitions of eGFR decline were used: 10 points, 10%, and 20% decline in eGFR using the CKD-EPI equation. (A) eGFR decline free survival, which is defined as GFR difference value (GFR DV=preoperative value–follow-up value) >10 points, showed no significant difference between the ileal conduit and neobladder groups. Estimated GFR decline-free survival that used GFR difference proportion (GFR DP=eGFR difference value/preoperative eGFR) >10% (B) and >20% (C) also presented no statistical difference. eGFR, estimated glomerular filtration rate; CKD-EPI equation, Chronic Kidney Disease Epidemiology equation; DV, difference value; DP, difference proportion; IC, ileal conduit.

• Click for larger image
• Download as PowerPoint slide

Endpoints other than renal function considered in our study were recurrence-free survival, overall survival, and cancer-specific survival. The median follow-up until urothelial cell carcinoma recurrence after RC was 23.5 months (interquartile range 7.0–48.0 mo), and the median follow-up until survival after RC was 35.8 months (interquartile range 16.5–70.2 mo). Recurrence-free and overall survival were not affected by the UD type: the log-rank p-values were 0.202 and 0.086, respectively (Fig. 3). However, cancer-specific survival was statistically different according to UD type (Fig. 3); NB had better cancer-specific survival than IC (log-rank p-value 0.036). Multivariable analysis was also performed for cancer-specific survival. However, the hazard ratio of the UD type was not significant (hazard ratio 0.087, 95% confidence interval 0.009–1.383, p=0.109) (Table 2).

Fig. 3
Recurrence-free survival, overall survival, cancer-specific survival. Recurrence-free survival (A) and overall survival (B) showed no significant difference according to urinary diversion type. Cancer-specific survival (C) showed a significant difference (p-value=0.036). However, urinary diversion type was not a significant factor in multivariate analysis. IC, ileal conduit.

• Click for larger image
• Download as PowerPoint slide

Table 2
Univariate and multivariate analysis for cancer-specific survival

• Click for larger image
• Click for full table
• Download as Excel file

In this single-center study, we aimed to determine the effect of UD on renal function in patients with a single kidney who underwent RC for urothelial cell carcinoma. We cautiously suggest that the diversion type has no impact on renal function, survival, and recurrence.

We considered various factors for evaluating renal function. Absolute and relative comparisons were performed using the serum creatinine level, MDRD eGFR, and CKD-EPI eGFR. A previous study reported that the CKD-EPI equation was better than the MDRD equation for a wide range of GFR, especially elevated GFR [14]. In addition, the CKD-EPI equation showed enhanced precision, better accuracy, and less bias than the MDRD equation regarding the measured and calculated GFR [14]. In a previous study, the CKD-EPI eGFR equation demonstrated the greatest precision and accuracy, and the least bias among four formulas for comparing renal function before and after nephrectomy: Cockcroft-Gault, Modification of Diet in Renal Disease Study, reexpressed Modification of Diet in Renal Disease Study, and Chronic Kidney Disease-Epidemiology Study equations [15]. Recently published studies comparing the renal function of patients who underwent cystectomy also used the CKD-EPI equation for assessing renal function decline [17, 20].

To determine renal function decline, our study referred to previous journals and combined various definitions using different values and ratios [13, 16, 17, 18, 19, 20, 21, 22, 23, 24]. The follow-up duration and interval were different between previous studies; we followed up the renal function after cystectomy for 5 years. Other studies defined the renal function decline by comparing the assessment and preoperative periods using only one episode [13, 18, 19, 22]. We questioned this definition and determined that renal function decline should be used when there were two or more consecutive episodes. A one-time renal function decline may be a temporary event or may improve later. Moreover, considering the short survival time of patients with a single kidney, advanced renal function values from serum creatinine and eGFR (CKD-EPI and MDRD equations) may result from over-hydration at the time of death or a cachexic state [25]. In such cases, serum cystatin C may be helpful because it is not affected by muscle mass and hydration [26]. In addition to the several definitions of renal function decline, there are several laboratory tests that depend on a patient’s medical condition.

Previous studies have reported a significant relationship between renal function decline and neoadjuvant chemotherapy [18], higher preoperative eGFR [20], and urinary tract obstruction (ureteroenteric or enterourethral stricture) [13, 19, 24]. Another study reported similar complications of NB and IC, which contrasts the popular view that IC is simple and safe [24]. If urinary tract obstruction is managed timeously after NB, renal function may not be considered a contraindication. Therefore, indications of NB should be expanded in the future because the surgical outcomes of NB and IC do not differ.

This study had several limitations. First, there was a limited sample size, and the study was retrospective. For more statistically significant results, a large-scale, multicenter, prospective study is required in the future. Second, all patients who underwent nephrectomy before, concurrently with, or after RC were included in our analysis. A previous study reported that nephrectomy was a significant risk factor for the progression of chronic kidney disease and renal function decline after nephrectomy [27]. In our study, 37 patients underwent nephrectomy before cystectomy, 4 underwent RC concurrently with nephrectomy, and 35 underwent nephrectomy after cystectomy. Although the different timings of nephrectomy did not influence renal function change in the IC and NB groups, a sophisticated study that unifies nephrectomy timing should be conducted in the future. Third, the causes of nephrectomy varied in our study. Although most cases were caused by upper tract urothelial cell carcinoma, some patients underwent nephrectomy due to kidney cancer or tuberculosis: 2 patients had kidney cancer, 4 had tuberculosis, and 4 had an infection or congenital diseases. Patients with upper tract urothelial cell carcinoma are more vulnerable to bladder carcinoma (22%–47%) or contralateral upper tract recurrence (5%) [28, 29, 30]. Although our study results showed that the cause of nephrectomy was not statistically related to our primary endpoint, further studies with a large number of participants are needed to determine the association with nephrectomy.

Our study presents patients data who underwent RC and simultaneously had a single kidney, a patient population difficult to recruit. Furthermore, the study findings are applicable to a variety of nephrectomy timings and causes, and can contribute to various indications of RC.

Among several factors for choosing UD type, a single kidney was thought to be a contraindication for NB formation. However, our study showed that UD type (IC vs. NB) when undergoing RC did not impact postoperative renal function change to a single kidney patients. Accordingly, patients with a single kidney might be considered to be an indication of NB. Our study has strength in that various renal function parameters and renal function decline definitions were used for comparing postoperative renal function decline between IC and NB groups. Our study is meaningful that it can be a starting point to expand NB indication even to a single kidney patients.