Introduction
It has been estimated that 73,800 new cases of cancers of the kidney and renal pelvis will be diagnosed in the United States in 2020 and that 14,800 people will die of them [
1]. Approximately 85% of all kidney tumors are renal cell carcinoma (RCC). RCC occurs in bilateral kidneys, including synchronous RCC (diagnosed concomitantly or within 3 months of the former tumor) and metachronous RCC (tumor diagnosed 3 months after former tumor detection) in approximately 5% of all RCC patients [
2‐
5]. Due to the relative rarity of bilateral presentation, even until now, there are limited data in the literature concerning patients treated with sequential bilateral kidney surgery. Few literatures have evaluated the functional impact of bilateral kidney surgery and how functional outcomes are influenced by tumor characteristics, modality selection, and patient-related risk factors [
6‐
9].
Nevertheless, several studies have elucidated renal functional outcomes in patients with bilateral synchronous tumors who have undergone sequential bilateral kidney surgery. Simmons et al. have demonstrated that bilateral partial nephrectomy (PN) is associated with significantly improved estimated glomerular filtration rate (eGFR) compared to radical nephrectomy (RN) [
6]. Packiam et al. have reported that patients for nonmetastatic bilateral synchronous tumors who have received simultaneous PN show lower mean postoperative 3 months (− 6% vs. -24%,
p = 0.015) and median postoperative 12 months (− 4% vs. -22%,
p < 0.001) reduction in eGFR compared to staged (within 6 months) PN, respectively [
7]. Singer et al. have observed that nephron-sparing surgery (NSS) enables dialysis to be avoided in more than 95% of patients [
8]. Another study reported average 28.9% eGFR decline after treatment of both kidneys with 608-day follow-up (59 → 41.9 ml/min/1.73 m
2) [
9].
However, few studies have compared long term functional outcomes according to procedure sequence. Current guidelines still lack an optimal surgical sequencing approach [
10‐
12]. Therefore, the objective of this study was to evaluate renal functional outcomes after sequential PN and RN in patients with bilateral RCC.
Results
Baseline characteristics between bilateral RCC group and unilateral RCC group are detailed in Table
1. Pre-propensity table for the cohort before matching are detailed in supplementary Table
1. There were no significant differences in baseline characteristics between groups after propensity score matching. Median time between surgeries in bilateral RCC group was 146.0 days (37.5 days for synchronous and 1899.0 days for metachronous subgroup). In the bilateral RCC cohort, subgroup comparative analysis for renal functional outcome (PN followed by PN vs. PN followed by RN vs. RN followed by PN group) was conducted (Table
2). Patients underwent sequential PN (
n = 48), PN followed by RN (
n = 8), or RN followed by PN (
n = 25). Patients with sequential RN were excluded in the subgroup analysis due to intuitively prominent renal function deterioration compared to other groups. The mean eGFR follow-up duration was 65.6 ± 47.6 months (range, 0–166 months). Final eGFR after bilateral surgery was 79.4 ± 33.9, 41.4 ± 28.3, and 61.2 ± 29.8 ml/minute/1.73 m
2 in these three groups, respectively (
p = 0.003). There were significant differences in eGFR decline from baseline and de novo CKD among groups, with PN followed by RN group showing the worst functional outcomes (all
p < 0.05). PN followed by PN group had significantly higher latest eGFR (
p = 0.003), less eGFR decline from baseline during the entire postoperative follow-up period (all
p < 0.05). They had also significantly less de novo CKD stage ≥3 occurred within a year after 2nd operation (
p = 0.036), less de novo CKD stage ≥3 occurred after the one-year minimal follow-up period after 2nd operation (median/mean follow-up period: 54.0/65.6 months,
p < 0.001), and less total number of patients with CKD at the last follow-up than other groups. RN followed by PN group had significantly less eGFR decline than PN followed by RN group at postoperative 1 week (
p = 0.031) and the latest follow-up period (
p = 0.001) from baseline.
Table 1
Baseline characteristics after propensity score matching
Age, mean (SD) | 54.0 ± 12.7 | 55.9 ± 12.9 | 54.2 ± 13.4 | 0.549 |
Sex, male, N (%) | 74 (84.1%) | 74 (82.2%) | 142 (79.8%) | 0.856 |
DM, yes, N (%) | 17 (19.3%) | 18 (20.0%) | 22 (16.7%) | 0.792 |
HTN, yes, N (%) | 47 (53.4%) | 42 (46.7%) | 89 (50.0%) | 0.564 |
BMI, mean (SD) | 25.0 ± 3.7 | 24.2 ± 3.1 | 24.7 ± 3.0 | 0.367 |
ECOG performance status, N (%) | | | | 0.415 |
≤ 1 | 86 (97.7%) | 86 (95.6%) | 167 (93.8%) | |
≥ 2 | 2 (2.3%) | 4 (4.4%) | 11 (6.2) | |
Surgical type, N (%) (Open / Laparoscopic / Robotic) | 37 (42.0%) / 13 (14.8%) / 38 (43.2%) | 44 (48.9%) / 9 (10.0%) / 37 (41.1%) | 94 (52.8%) / 11 (6.2%) / 73 (41.0%) | 0.183 |
EBL, ml, mean (SD) | 249.1 ± 107.5 | 259.0 ± 217.9 | 226.1 ± 261.5 | 0.878 |
Ischemic time, min, mean (SD)a | 21.5 ± 9.8 | 22.2 ± 10.7 | 21.4 ± 16.3 | 0.904 |
Pathologic tumor size, mm, mean (SD) | 33.5 ± 34.8 | 32.1 ± 27.6 | 32.8 ± 23.3 | 0.949 |
Baseline eGFR, ml/min/1.73 m2, mean (SD) | 71.6 ± 32.5 | 76.0 ± 25.4 | 78.6 ± 22.3 | 0.224 |
Baseline CKD, stage ≥3 | 28 (31.8%) | 16 (17.8%) | 39 (21.9%) | 0.068 |
Table 2
Functional outcome of bilateral RCC cohort regardless of time interval between tumors
Preoperative eGFR, ml/min/1.73 m2, mean (SD) (before 1st surgery) | 84.7 ± 25.4 | 83.1 ± 19.6 | 79.6 ± 33.4 | 0.759 |
Latest eGFR, ml/min/1.73 m2, mean (SD) | 79.4 ± 33.9 | 41.4 ± 28.3 | 61.2 ± 29.8 | 0.003 |
| p = 0.108 a | |
eGFR decline from baseline, mean (SD) |
Postoperative 1 week (after 2nd surgery) | 6.9 ± 18.5 | 28.5 ± 18.2 | 10.6 ± 16.5 | 0.009 |
| p = 0.031 a | |
Postoperative 1 month (after 2nd surgery) | 5.1 ± 15.0 | 28.7 ± 23.2 | 9.9 ± 13.2 | 0.001 |
| p = 0.058 a | |
Postoperative 1 year (after 2nd surgery) | 5.8 ± 17.5 | 29.4 ± 24.2 | 31.3 ± 6.3 | 0.018 |
| p = 0.035 a | |
Latest | −0.64 ± 22.5 | 30.5 ± 26.6 | −2.7 ± 21.3 | 0.001 |
| p = 0.001 a | |
De novo CKD, stage≥3 |
Baseline CKD, before 1st surgery | 8 (16.7%) | 1 (12.5%) | 5 (20.0%) | 0.874 |
Baseline CKD, before 2nd surgery | 6 (12.5%) | 1 (12.5%) | 9 (36.0%) | 0.049 |
De novo, ≤ 1 year after 2nd surgery | 9 (18.8%) | 4 (50%) | 5 (20%) | 0.036 |
De novo, > 1 year after 2nd surgery | 0 (0%) | 2 (25.0%) | 1 (4.0%) | < 0.001 |
Total number of CKD patients last follow-up | 15 (31.3%) | 7 (87.5%) | 15 (60%) | < 0.001 |
We also performed another subgroup analysis between bilateral synchronous RCC (PN followed by PN) subgroup and unilateral RCC (single PN) subgroup (Table
3). The bilateral synchronous RCC (PN followed by PN) subgroup had no significant difference in eGFR decline at the latest follow-up period from baseline compared to the unilateral RCC (single PN) subgroup (
p = 0.770), although it had higher de novo CKD rate until postoperative 1 year and during the entire follow-up period (13.8% vs. 6.9%,
p = 0.016). Additional subgroup analysis between bilateral synchronous and metachronous RCC (PN followed by PN) subgroups revealed no significant differences in variables among groups (Table
4).
Table 3
Comparative analysis between bilateral synchronous tumor subgroup (PN followed by PN) and unilateral RCC (single PN)
Preoperative eGFR, ml/min/1.73 m2, mean (SD) | 78.2 ± 24.3 | 82.2 ± 22.9 | 0.106 |
aLatest eGFR, ml/min/1.73 m2, mean (SD) | 80.2 ± 22.7 | 83.1 ± 20.4 | 0.182 |
eGFR decline from baseline, mean (SD) |
Postoperative 1 week | 7.2 ± 16.1 | −3.5 ± 15.4 | < 0.001 |
Postoperative 1 month | 3.5 ± 14.6 | −1.7 ± 15.5 | 0.037 |
Postoperative 1 year | 5.3 ± 17.3 | 0.9 ± 16.3 | 0.099 |
bLatest | −2.0 ± 23.8 | −1.0 ± 18.5 | 0.770 |
De novo CKD, ≤ 1 year | 8 (13.8%) | 8 (7.5%) | 0.022 |
De novo CKD, total | 8 (13.8%) | 10 (6.9%) | 0.016 |
Table 4
Comparative analysis between PN followed by PN subgroups (synchronous vs. metachronous) [latter surgery]
Preoperative eGFR, ml/min/1.73 m2, mean (SD) | 76.5 ± 22.4 | 82.1 ± 25.1 | 0.434 |
aLatest eGFR, ml/min/1.73 m2, mean (SD) | 80.2 ± 24.0 | 78.1 ± 31.2 | 0.247 |
eGFR decline from baseline, mean (SD) |
Postoperative 1 week | 6.5 ± 18.0 | 7.4 ± 19.7 | 0.871 |
Postoperative 1 month | 3.0 ± 14.9 | 8.4 ± 15.1 | 0.230 |
Postoperative 1 year | 3.5 ± 17.5 | 9.3 ± 17.5 | 0.266 |
bLatest | −3.7 ± 24.1 | 4.0 ± 19.5 | 0.250 |
De novo CKD, stage ≥3 |
Baseline CKD, Before 1st surgery | 7 (24.1%) | 1 (5.3%) | 0.123 |
Baseline CKD, Before 2nd surgery | 5 (17.2%) | 1 (5.3%) | 0.381 |
De novo, ≤ 1 year after 2nd surgery | 5 (17.2%) | 4 (21.1%) | 1.000 |
De novo, > 1 year after 2nd surgery | 0 (0%) | 0 (0%) | 1.000 |
Total number of pts. with CKD at last follow-up | 10 (34.5%) | 5 (26.3%) | 0.751 |
Multivariate analysis for the prediction of de novo CKD until postoperative 1 year revealed that hypertension (Hazard ratio [HR]: 2.159, 95% Confidence Interval [CI]: 1.233–3.783,
p = 0.007), pathologic tumor size at 1st surgery (HR: 1.012, 95% CI: 1.006–1.024,
p = 0.010) and PN followed by RN sequence (HR: 1.837, 95% CI: 1.028–3.635,
p = 0.007) were significant factors (Table
5). A multivariate analysis for the prediction of de novo CKD during the entire period revealed that hypertension (HR: 1.905, 95% CI: 1.172–3.265,
p = 0.010), PN followed by RN sequence (HR: 1.888, 95% CI: 1.088–4.055,
p < 0.001), and metachronous RCC (HR: 2.682, 95% CI: 1.032–6.973,
p = 0.043) were significant predictive factors (Table
6). The difference in time interval between tumor occurrences in metachronous RCC was not significantly related to de novo CKD incidence (
p = 0.083 for de novo CKD within postoperative 1 year and
p = 0.056 for de novo CKD during the entire follow-up period).
Table 5
Multivariate Cox regression analyses of variables associated with de novo CKD within postoperative one year
Age (years) | 1.032 | 1.009–1.056 | 0.001 | 1.036 | 1.011–1.062 | 0.005 |
BMI | 1.011 | 0.924–1.106 | 0.810 | | | |
Sex, male | 0.711 | 0.348–1.454 | 0.350 | | | |
DM | 1.295 | 0.646–2.595 | 0.466 | | | |
HTN | 2.413 | 1.353–4.304 | 0.003 | 2.159 | 1.233–3.783 | 0.007 |
Baseline eGFR | 0.992 | 0.982–1.003 | 0.150 | | | |
Baseline Hb | 0.986 | 0.875–1.112 | 0.822 | | | |
EBL (1st surgery) | 1.000 | 0.999–1.001 | 0.799 | | | |
EBL (2nd surgery) | 1.000 | 0.999–1.001 | 0.663 | | | |
Ischemic time (1st surgery) | 1.005 | 0.974–1.037 | 0.763 | | | |
Ischemic time (2nd surgery) | 0.987 | 0.957–1.019 | 0.431 | | | |
Pathologic tumor size (1st surgery) | 1.015 | 1.006–1.024 | 0.001 | 1.012 | 1.003–1.021 | 0.010 |
Pathologic tumor size (2nd surgery) | 1.004 | 0.996–1.011 | 0.326 | | | |
Time between operations | 1.000 | 1.000–1.001 | 0.083 | | | |
Surgery sequence |
PN ➔ PN | Reference | | | Reference | | |
PN ➔ RN | 2.249 | 1.088–4.786 | 0.006 | 1.837 | 1.028–3.635 | 0.007 |
RN ➔ PN | 1.510 | 1.096–5.750 | 0.030 | 1.235 | 0.947–5.276 | 0.066 |
PN (unilateral case) | 0.663 | 0.285–1.543 | 0.341 | 0.315 | 0.119–0.832 | 0.261 |
Tumor chronology |
Synchronous | Reference | | | | | |
Metachronous | 1.370 | 0.660–2.840 | 0.398 | | | |
Tumor multiplicity |
Unilateral | Reference | | | | | |
Bilateral | 1.244 | 0.698–2.215 | 0.459 | | | |
Table 6
Multivariate Cox regression analyses of variables associated with de novo CKD during the entire follow-up period
Age (years) | 1.031 | 1.009–1.053 | 0.005 | 1.005 | 0.971–1.040 | 0.780 |
BMI | 1.029 | 0.946–1.120 | 0.500 | | | |
Sex, male | 0.607 | 0.306–1.205 | 0.154 | | | |
DM | 1.530 | 0.806–2.905 | 0.194 | | | |
HTN | 1.956 | 1.147–3.335 | 0.014 | 1.905 | 1.172–3.265 | 0.010 |
Baseline eGFR | 0.992 | 0.982–1.002 | 0.106 | | | |
Baseline Hb | 1.002 | 0.902–1.114 | 0.969 | | | |
EBL (1st surgery) | 1.000 | 0.999–1.000 | 0.784 | | | |
EBL (2nd surgery) | 1.000 | 0.999–1.001 | 0.680 | | | |
Ischemic time (1st surgery) | 1.001 | 0.972–1.032 | 0.921 | | | |
Ischemic time (2nd surgery) | 0.985 | 0.956–1.014 | 0.301 | | | |
Pathologic tumor size (1st surgery) | 1.016 | 1.007–1.025 | 0.001 | 1.013 | 1.000–1.027 | 0.051 |
Pathologic tumor size (2nd surgery) | 1.004 | 0.997–1.011 | 0.278 | | | |
Time between operations | 1.000 | 1.000–1.001 | 0.056 | | | |
Surgery sequence |
PN ➔ PN | Reference | | | Reference | | |
PN ➔ RN | 2.919 | 1.386–6.146 | 0.001 | 1.888 | 1.088–4.055 | < 0.001 |
RN ➔ PN | 2.190 | 1.165–4.114 | 0.015 | 1.041 | 1.003–3.258 | 0.103 |
PN (unilateral case) | 0.782 | 0.366–1.669 | 0.525 | 0.233 | 0.088–0.614 | 0.718 |
Tumor chronology |
Synchronous | Reference | | | Reference | | |
Metachronous | 2.250 | 1.121–4.516 | 0.023 | 2.682 | 1.032–6.973 | 0.043 |
Tumor multiplicity |
Unilateral | Reference | | | | | |
Bilateral | 1.421 | 0.824–2.449 | 0.206 | | | |
Discussion
Surgical management of RCC ultimately aims to balance numerous considerations with attempts to minimize operative morbidity and decline of renal function. To this end, when technically feasible, NSS has been advocated to maximize preservation of renal parenchyma and avoid the incidence and sequelae of renal function deterioration [
19‐
22]. In the setting of bilateral renal masses, guidelines favor the performance of bilateral PN when it is technically feasible [
10‐
12]. However, there is only a short reference to this simple guideline statement without specific details on how to do it in real clinical practice. For example, in case of the situation that both RN and PN are inevitably needed, it is a dilemma to decide which procedure has to be done first to achieve satisfactory renal function preservation. Nevertheless, using our prospectively collected database, we proved that doing RN first followed by PN sequence would be better in terms of preserving renal function than proceeding in reverse order. According to our multivariate analysis, hypertension and PN followed by RN sequence are commonly independent risk factors for de novo CKD within postoperative 1 year and the entire follow-up period. We also identified evidence in other literature supporting our findings. Krohn et al. have reported that 1 year after donation, the remained kidney manages to compensate up to 70% of renal function before surgery [
23]. A plausible explanation is that in the remained kidney, vasodilation and increased renal plasma flow can occur immediately after surgery. These changes, combined with the process of glomerular hypertrophy, can increase glomerular filtration of the remnant kidney by approximately 40% without occurring a concomitant increase in glomerular capillary pressure [
24‐
26]. This adaptive hyperfiltration also occurs in older kidney donors, although more modestly than in younger donors [
27]. Taner et al. have reported that compensatory hypertrophy and GFR increase can occur in the remaining kidney of medically complex living donors at a comparable rate to those of standard donors [
28]. These findings confirmed reassurance for delicately selected medically complex living donors. One study has concluded that RCC is not an independent risk factor for renal function decrease after nephrectomy. RCC patients with few morbidities could have the same deterioration of meanly 30% of kidney function compared with living donors. However, their lower baseline function can result in an increased risk for CKD [
29]. These findings imply that patients with RN who have sufficient period to compensate for their renal function until PN can preserve favorable renal function without risk for CKD.
PN is generally related to a lower risk of developing clinically significant CKD than RN. Postoperative impairment in kidney function occurs most commonly in the first year after nephrectomy and appears stable over time. Age, Tumor stage, and preoperative kidney function are predictors of incident CKD after kidney cancer operation [
30]. Bilateral PN needs more careful consideration compared to ipsilateral PN due to potential additional loss of renal function secondary to bilateral renal ischemia from hemorrhage, hypotension, and prolonged operative time [
19]. These challenges have made some surgeons support staged bilateral PN as opposed to simultaneous bilateral PN in a single setting. PN, if performed well, is also a possible choice for larger renal tumors as it generates tolerable surgical morbidity, better renal function preservation, and equivalent cancer control with potential for better long-term survival than RN [
31]. Bercz et al. showed acceptable oncological and functional outcomes from 65 patients. They reported significant postoperative renal function deterioration (44.8% eGFR decrease for synchronous (mostly RN → PN) and 30.4% decrease for metachronous tumor (all RN → PN) after the second operation, respectively), but hemodialysis was rarely required. Compared to their results, we showed more eGFR preservation if RN → PN (23.1% decrease) is performed respectively even though their cohort characteristics are slightly different [
32].
This study has some limitations. First, even with two large tertiary centers’ cohort, the study population was still small due to the rarity of bilateral RCCs with its retrospective nature. In addition, there was no analysis of differences in renal functions that might occur due to the heterogeneity of surgical techniques and different time intervals of operation for bilateral metachronous RCC. Thirdly, our result from eight patients that PN followed by RN can negatively affect renal function preservation is necessary to be verified in further studies with larger numbers of patients. Finally, it was difficult to identify genetic differences between these patients. These issues can be elucidated if multicenter prospective randomized clinical trials are performed in the near future. Nevertheless, to the best of our knowledge, the current study is the first of its kind that assesses functional aspects of both kidney cancer surgery performed in large institutions over a relatively long period of time. This is also the first study to find that RN followed by PN is superior in terms of preserving renal functions in the opposite order by evaluating cohorts through propensity score matching analysis.
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit
http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.