Abstract
Purpose:
We sought to provide a contemporary understanding of chronic kidney disease and its relevance to kidney cancer surgery. Another purpose was to resolve points of discrepancy regarding the survival benefits of partial vs radical nephrectomy by critically evaluating the results of prospective and retrospective studies in the urological literature.
Materials and Methods:
We performed a comprehensive literature search for relevant articles listed in MEDLINE® from 2002 to 2018 using the key words radical nephrectomy, partial nephrectomy, glomerular filtration rate, kidney function and chronic kidney disease. We also assessed select review articles and society guidelines about chronic kidney disease pertinent to urology and nephrology.
Results:
Complete evaluation of the potential consequences of chronic kidney disease involves assessment of the cause, the glomerular filtration rate level and the degree of albuminuria. Chronic kidney disease is commonly defined in the urological literature solely as a glomerular filtration rate less than 60 ml/minute/1.73 m2. This ignores the significance of the cause of chronic kidney disease, and the presence and degree of albuminuria. Although this glomerular filtration rate is relevant for preoperative assessment of patients who undergo surgery of kidney tumors, recent studies suggest that a glomerular filtration rate less than 45 ml/minute/1.73 m2 represents a more discerning postoperative prognostic threshold. Reported survival benefits of partial over radical nephrectomy in retrospective studies were likely influenced by selection bias. The lack of survival benefit in the partial nephrectomy cohort in the only randomized trial of partial vs radical nephrectomy was consistent with data demonstrating that patients in each study arm were at relatively low risk for mortality due to chronic kidney disease when accounting for the chronic kidney disease etiology and the postoperative glomerular filtration rate.
Conclusions:
The prognostic risk of chronic kidney disease in patients with kidney cancer is increased when the preoperative glomerular filtration rate is less than 60 ml/minute/1.73 m2 or the postoperative rate is less than 45 ml/minute/1.73 m2. Additional factors, including nonsurgical causes of chronic kidney disease and the degree of albuminuria, can also dramatically alter the consequences of chronic kidney disease after kidney cancer surgery. Urologists must have a comprehensive knowledge of chronic kidney disease to assess the risks and benefits of partial vs radical nephrectomy when managing tumors with increased complexity and/or oncologic aggressiveness.
| ACR |
albumin-to-creatinine ratio |
| CKD |
chronic kidney disease |
| CKD-M/S |
CKD due to medical etiologies in patient who required kidney cancer surgery |
| CKD-S |
CKD primarily due to surgical removal of nephrons |
| eGFR |
estimated GFR |
| EORTC |
European Organisation for Research and Treatment of Cancer |
| GFR |
glomerular filtration rate |
| KDIGO |
Kidney Disease Improving Global Outcomes® |
| KDOQI |
Kidney Disease Outcomes Quality Initiative |
| NKF |
National Kidney Foundation® |
| PN |
partial nephrectomy |
| RN |
radical nephrectomy |
In the last 2 decades PN has emerged as standard treatment for oncologic control of small renal tumors while maximizing kidney function. The paradigm shift from RN to PN has been driven by many factors, including the incidental presentation of most kidney tumors, advances in surgical technique, better understanding of the natural history and most importantly appreciation of the significance of CKD and the impact of kidney surgery on functional outcomes.1–3 Interest in preserving kidney function eventually evolved into a primary objective, if not the primary objective, for many surgeons, compelling them to preserve kidney function at all costs.4–7 This initiative was largely supported by an abundance of retrospective data suggesting equivalent oncologic outcomes and superior functional outcomes, fewer cardiovascular events and improved overall survival for PN compared to RN.8–10
However, in 2011 the results of the EORTC-30904, the only prospective randomized trial comparing elective PN to RN, revealed that patients who underwent PN for small (less than 5 cm) solitary, localized renal tumors did not have better overall survival despite equivalent oncologic outcomes and superior postoperative kidney function.11 In fact, on intent to treat analysis the RN cohort had significantly better overall survival and fewer cardiac events. This was particularly unexpected because the cohorts in this randomized study appeared to be similar to those in previously published retrospective studies demonstrating a survival advantage in patients who underwent PN.
Attempts to reconcile disparate conclusions drawn from retrospective and prospective data have raised questions not only about the validity of the trial results but also about the impact of postoperative kidney function on morbidity and mortality in patients with kidney cancer and even the potential overuse of elective PN.12 The objective of this study was to provide a contemporary understanding of CKD and its relevance to kidney cancer surgery and to address the conclusions of the retrospective and prospective studies by critically evaluating the results of the EORTC-30904 trial11 and analogous retrospective studies in the urological literature.
Methods
We performed a comprehensive literature search for relevant articles listed in MEDLINE® from 2002 to 2018 using the key words kidney cancer, radical nephrectomy, partial nephrectomy, glomerular filtration rate, kidney function and chronic kidney disease. Selected review articles and society guidelines on CKD pertinent to the fields of urology and nephrology were reviewed.
Results
Chronic Kidney Disease, Morbidity and Mortality
CKD is a significant public health concern in the United States and around the world. Currently CKD affects more than 30 million Americans or approximately 14.8% of the adult population of the United States with a higher prevalence of early CKD stages rather than end stage kidney failure. Almost 50% of Americans older than 70 years old have CKD.13,14 The high prevalence of conditions contributing to CKD, such as diabetes, obesity and hypertension, means that CKD will remain a significant public health issue.
Kidney failure is the end stage of CKD. We have known for decades that the risk of cardiovascular events and death are dramatically increased in patients with kidney failure who receive kidney replacement therapy.15 Adjusted mortality rates in these patients are more than 5 times that of their peers without kidney failure with more than 50% of deaths attributable to cardiovascular disease.14 However, until 20 years ago little was known about the risk of morbidity and death associated with earlier stages of CKD.
In 1999 the NKF launched the KDOQI, which aimed to increase the detection of early stages of CKD, improve the treatment of kidney disease, retard the progression of CKD and prevent progression to end stage renal disease. The NKF-KDOQI guidelines, which were released in 2002, defined CKD as certain findings for more than 3 months, including 1) GFR less than 60 ml/minute/1.73 m2 or 2) markers of kidney damage including but not limited to albuminuria, urine sediment abnormalities or imaging abnormalities.2 In the ensuing decade numerous groups assessed the impact of CKD on morbidity and mortality, finding a significant graded increase in the risk of cardiovascular events and death for GFR less than 60 ml/minute/1.73 m2.16–19
Much was learned during this time about the complexities of CKD. In 2012 the KDIGO Clinical Practice guidelines were released.3 These guidelines, which were based on a meta-analysis of 45 patient cohorts, revised the CKD classification, reflecting increased recognition that GFR alone does not adequately predict the risk of adverse outcomes. Rather, accurate prediction of disease progression, and the morbidity and mortality of CKD also require considering the cause of the disease and the degree of albuminuria. The guidelines recommended classifying CKD based on the CGA (cause, GFR category and albuminuria) staging system.3 Each CGA factor can individually and collectively affect prognosis in patients with CKD.
Chronic Kidney Disease
Cause
The 2012 KDIGO guidelines provide a simple classification of the cause of disease based on the presence or absence of systemic diseases affecting the kidney (eg diabetes, hypertensive or autoimmune disease) vs diseases limited to the kidney and urinary tract, and the location of anatomical and pathological kidney abnormalities (eg glomerular, tubulointerstitial and vascular diseases).3 CKD etiology can have a dramatic effect on CKD progression and clinical outcomes.20 For example, patients with autosomal dominant polycystic kidney disease can experience more rapid eGFR decline and progression to end stage renal disease than in many other kidney diseases. Despite this the annual mortality rate in patients with autosomal dominant polycystic kidney disease is significantly lower than in those with CKD due to other diseases, likely because of the absence of other impactful systemic manifestations.
Although it is not specifically mentioned in this classification system, CKD due to surgical resection of nephrons as in kidney donation, PN or RN would be considered a primary kidney disease without systemic manifestations affecting mortality and typically without a progressive decline in kidney function after stabilizing from the original event. Notably, in some situations CKD etiology may be ambiguous or there may be multiple contributing factors. Thus, the cause may be a less reliable predictor of prognosis than other factors, such as the GFR and the degree of albuminuria.21
Glomerular Filtration Rate
Of the various measures of kidney function GFR is still considered the primary indicator. A normal GFR in young adults is greater than 90 ml/minute/1.73 m2. The average annual rate of GFR decline with aging is about 0.75 ml/minute/1.73 m2 but there is a wide variation and the cause of decline is not well understood. GFR less than 15 ml/minute/1.73 m2 is defined as kidney failure.
The 2002 NKF-KDOQI guidelines outlined 5 GFR stages of CKD severity based on the GFR.2 Because it is difficult to directly measure GFR, it is most commonly estimated using equations based on serum creatinine, such as the MDRD (Modification of Diet in Renal Disease) or the CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) equations. Unfortunately the urological community generally chose to adopt the 60 ml/minute/1.73 m2 threshold as the primary criterion for CKD and the related risk. In reality the risk of mortality and cardiovascular events increases progressively in the general population below an eGFR threshold of approximately 75 ml/minute/1.73 m2 and the risk is particularly high when eGFR is less than 45 ml/minute/1.73 m2 (fig. 1, A and C).3
Figure 1. HR and 95% CI (shaded areas) of all cause (A and B) and cardiovascular (C and D) mortality by eGFR (A and C) and ACR (B and D) adjusted for age, gender, ethnic origin, cardiovascular disease history, systolic blood pressure, diabetes, smoking and total cholesterol. Referents (diamond) were eGFR 95 ml/minute/1.73 m2 and ACR 5 mg/gm (mg/g) (0.6 mg/mmol). To convert ACR in mg/gm to mg/mmol multiply by 0.113. Values in parentheses indicate approximate conversions to mg/mmol. Circles indicate statistically significant. Triangles indicate not significant. Reprinted with permission from Matshushita K, van de Velde M, Astor BC et al: The Lancet 2010; 375: 2073. Copyright 2010 by Elsevier.
It is now appreciated that there can be greatly varying risk in these eGFR categories. This is particularly true when considering patients in whom eGFR ranges from 30 to 59 ml/minute/1.73 m2 (fig. 2).19 Acknowledging the meaningful differences in outcomes in this group, the 2012 KDIGO guidelines formally subdivided GFR category G3 into G3a (45 to 59 ml/minute/1.73 m2) and G3b (30 to 44 ml/minute/1.73 m2).3 Risk can be further stratified when the presence or absence of albuminuria is considered.
Figure 2. CKD prognosis based on eGFR (G) and ACR (A) in KDIGO 2012. Grid shows risk of progressive kidney function decrease and mortality, including low, moderately increased, high and very high. Reprinted with permission.3
Albuminuria
It is well recognized in the nephrology literature that proteinuria (albuminuria) is a powerful and independent predictor of adverse outcomes in CKD cases (fig. 1, B and D). Thus, the 2012 KDIGO clinical practice guideline also categorizes CKD by degree of albuminuria (A1 to A3).3 Albuminuria is most easily measured by the urine ACR. The reagent strip (dipstick) is substantially less accurate, although it is often used for screening. Urinalysis is done as part of preoperative testing and/or the initial office evaluation in virtually all patients who undergo kidney cancer surgery and those who show more than trace albuminuria likely have a moderate to severe elevation in ACR. Unfortunately proteinuria has long been overlooked as a critical component of the evaluation of CKD by urologists. This is a significant oversight because albuminuria can have a dramatic influence on all cause and cardiovascular mortality independently of eGFR (fig. 1, B and D).3 For example, patients with a GFR of 45 to 59 ml/minute/1.73 m2 can be classified as at moderate, high or even very high risk for current and/or future complications based on the albuminuria category (fig. 2).
Risk Prediction
The KDIGO guidelines recommend risk prediction in CKD cases based on cause, GFR, albuminuria and other measures (risk factors and comorbid conditions) specific to the outcome.3 In patients with GFR less than 60 ml/minute/1.73 m2 the Kidney Failure Risk Equation, which is calculated using patient age, gender, GFR, albuminuria and geographic region, provides quantitative 2 and 5-year estimates of the risk of kidney failure development requiring kidney replacement therapy.22 Notably, in patients older than 60 years without severely elevated albuminuria an eGFR of 45 to 59 ml/minute/1.73 m2 is associated with 2 and 5-year kidney failure risks of less than 1.7% and less than 5.2%, respectively.
Kidney Cancer Surgery and Chronic Kidney Disease
Managing renal tumors with RN can lead to a permanent decline in GFR, which often manifests as an increase in serum creatinine.23,24 However, historically it was thought that sacrificing normal uninvolved renal parenchyma for the sake of cancer control did not result in serious long-term morbidity unless the patient were to require kidney replacement therapy. This assumption was primarily rooted in data from the kidney transplant literature, which demonstrated that kidney donors did not have a substantially increased rate of kidney failure requiring dialysis or leading to death.25–27 However, key differences exist between donors and patients with cancer, making some of these assumptions invalid. While kidney donors tend to be younger, carefully selected and screened for medical comorbidity, patients with renal tumors are frequently older and they often have comorbid conditions known to cause kidney disease, such as hypertension, diabetes, vascular disease or metabolic syndrome.8,28
In 2006 Huang et al reported a study evaluating the impact of PN vs RN on postoperative kidney function in 662 patients with normal preoperative serum creatinine, a solitary and less than 4 cm renal mass and a normal contralateral kidney on imaging.8 Rather than using the serum creatinine concentration to estimate kidney function, the MDRD equation was used to estimate GFR. Several novel observations were made. Although all patients had normal serum creatinine preoperatively, 26% had baseline eGFR less than 60 ml/minute/1.73 m2. Also, after surgery the 3-year probability of freedom from new onset eGFR less than 60 ml/minute/1.73 m2 was 80% after PN but only 35% after RN (p <0.0001). Although eGFR category 3 had not yet been subdivided at the time of publication, the clinical difference between the eGFR declines of less than 60 vs less than 45 ml/minute/1.73 m2 was recognized and freedom from a decline below those 2 eGFR thresholds was reported. Multivariate analysis showed that RN remained an independent risk factor in patients with new onset eGFR less than 60 and less than 45 ml/minute/1.73 m2 (HR 3.82 and 1.8, respectively, each p <0.0001).
These postoperative functional trends were subsequently corroborated in multiple retrospective series, population based cohorts and even in the randomized EORTC-30904 trial,11 confirming that loss of normal, functioning nephrons due to RN has a measurable effect on postoperative eGFR.29–32 Although the clinical outcomes of the resulting CKD in this patient population were unknown, the study suggested that preventing CKD by PN could potentially reduce nononcologic morbidity and improve overall survival in patients who were otherwise cured of kidney cancer from surgery.8
Consequently many investigators set out to determine whether nephron sparing surgery could improve nononcologic outcomes such as cardiovascular events and overall survival. The first of these studies included data on 648 patients who underwent RN or PN at Mayo Clinic from 1989 to 2003.33 Overall the investigators found no significant association of surgery type (RN or PN) with overall mortality. However, in patients younger than 65 years RN was associated with an increased risk of death from any cause compared with PN (RR 2.16, p=0.022).
This initial report33 was followed soon thereafter by an analysis of data from the SEER (Surveillance, Epidemiology, and End Results) cancer registry, which demonstrated that RN was associated with an increased risk of mortality (HR 1.38, p <0.01) and a 1.4 times greater incidence of cardiovascular events after surgery (p <0.05).34 In 2012 Tan et al used an updated SEER-Medicare data set and found that for cT1a lesions PN instead of RN resulted in a predicted survival increase of 5.6% (95% CI 1.9–9.3), 11.8% (95% CI 3.9–19.7) and 15.5% (95% CI 5.0–26.0) at 2, 5 and 8 years after treatment, respectively (each p <0.001).9 This corresponded to a number needed to treat of 7 patients at the 8-year time point. Treating 7 patients with PN rather than RN would result in 1 life saved during 8 years of followup. This represents a remarkable result with strong implications for patient treatment if it was due to the functional benefits of PN rather than to residual confounders, which are common in observational studies.
Despite the lack of substantiation in randomized trials, the urological community widely accepted the notion that PN reduced the risk of nononcologic morbidity and mortality by preserving nephrons and preventing CKD development. This resulted in strong endorsement of PN by society guidelines and a dramatic shift from RN to PN in patients with cT1a tumors, representing an important improvement in patient care.1,35–37 However, it also prompted surgeons to expand the indications for elective PN to more complex and/or aggressive tumors in an effort to optimize nephron preservation and minimize the risk of CKD.
However, the validity of this tenet was abruptly called into question in 2011 when the results of EORTC-30904 were published, which demonstrated an overall survival advantage in patients who underwent RN.11 To our knowledge EORTC-30904 remains the only randomized prospective trial to date to compare overall survival following RN vs PN as treatment of localized kidney cancer, targeting the exact population which should derive benefit from PN. The study was initially designed as a noninferiority trial intended to detect a 10% difference in overall survival. However, after 5 years it was redesigned to detect a more modest 3% difference. Ultimately the trial closed early due to suboptimal accrual after randomizing 541 patients. Nonetheless, the 2 groups were well balanced in regard to preoperative characteristics.
The primary outcome of the study was overall survival.11 On intent to treat analysis the investigators unexpectedly found that RN was associated with higher 10-year overall survival than PN (81.1% vs 75.7%). With a HR of 1.50 (95% CI 1.03–2.16) the statistical test of noninferiority for overall survival was not significant while the test for superiority in favor of RN was significant (p=0.77 vs 0.03). When considering only clinically and pathologically eligible renal cell carcinoma cases, the HRs were more modest and the findings were no longer significant but the results trended similarly in that there was no overall survival benefit to PN. There were only 12 deaths from renal cell carcinoma with no significant difference between PN and RN. There was also no difference in cancer progression, thus supporting oncologic equivalency, at least for the small renal tumors under study. Interestingly the rate of cardiovascular death also trended higher for PN vs RN (9.3% vs 7.3%). Despite these findings the study investigators ultimately concluded that PN remained the treatment of choice for pT1a tumors.
Critics of the study11 contested the results, raising a number of concerns. 1) The study was closed prematurely due to poor accrual and, thus, it was statistically underpowered to detect small differences. 2) While it was designed as a noninferiority trial, the finding of an overall survival benefit for RN over PN on intent to treat analysis was based on a test of superiority. 3) There was no standardization of surgical technique since surgeries were done at a total of more than 60 centers and there was unequal crossover between arms. However, concerns about external validity were evaluated and at least 1 group concluded that the results were generalizable to real world patients.38
Many found it difficult to ignore the randomized prospective data of EORTC-30904.11 The magnitude of benefits from elective PN began to be questioned, particularly as others in the urological community progressively pushed the limits and boundaries of elective nephron sparing surgery in the name of a purported survival benefit. Doubts about the benefits of elective PN were further bolstered when the kidney function outcomes from EORTC-30904 were published, which confirmed superior functional outcomes for PN while again demonstrating no benefit in overall survival.29 Subsequently investigators began critically reappraising the retrospective data as well as the EORTC results11 to reconcile these seemingly conflicting findings. In the course of this reappraisal several important observations have come to light.
Observation 1 is that CKD defined solely by eGFR less than 60 ml/minute/1.73 m2 is not an adequate description of CKD and postoperative baseline GFR less than 45 ml/minute/1.73 m2 is a more discerning threshold after PN or RN. By applying the 60 ml/minute/1.73 m2 threshold many retrospective studies failed to acknowledge the nonlinear increase in the risk of complications due to declining eGFR as well as the fact that the cause of CKD and the presence of albuminuria have a significant impact on CKD progression and prognosis (figs. 1 and 2).3
Investigators from the Cleveland Clinic were among the first to report that differing causes of CKD produce divergent clinical outcomes, which is particularly relevant in patients with kidney cancer.39 In a large cohort of more than 4,000 patients who underwent PN or RN for renal masses at Cleveland Clinic the authors compared those with postoperative eGFR greater than 60 ml/minute/1.73 m2 (referred to as no CKD) to those with preoperative eGFR greater than 60 ml/minute/1.73 m2 but postoperative eGFR less than 60 ml/minute/1.73 m2 (referred to as CKD-S) and to those with preoperative and postoperative eGFR less than 60 ml/minute/1.73 m2 (referred to as CKD due to medical etiologies, also needing surgery or CKD-M/S).32,39
The Cleveland Clinic investigators made several interesting observations.39 They found no difference in the rate of eGFR decline or the need for dialysis between patients with no CKD and those with CKD-S with each declining at about 0.7% per year, consistent with the aging process. In contrast, patients with CKD-M/S experienced an average GFR decline of 4.7% per year. All cause and nonkidney cancer related mortality was highest in those with CKD-M/S, followed by those with CKD-S, which approximated the survival characteristics of patients with no CKD even after surgery. The authors concluded that cause of CKD has a significant role in predicting the risk of mortality after PN or RN.32,39,40
As mentioned, while eGFR less than 60 ml/minute/1.73 m2 is generally diagnostic for CKD and confers increased risk, it is clear that the risks of cardiovascular events and mortality are significantly higher when eGFR has fallen to lower thresholds. Demirjian et al noted that the probability of various adverse outcomes, including a 50% decline in GFR, the need for dialysis and 5-year all cause mortality, were generally not increased unless postoperative GFR was less than 45 ml/minute/1.73 m2.32 Several other groups who evaluated this lower cutoff subsequently confirmed that postoperative eGFR less than 45 ml/minute/1.73 m2 is more strongly associated with mortality in patients who undergo kidney cancer surgery regardless of cause (fig. 3).41,42
Figure 3. Importance of postoperative GFR 45 ml/minute/1.73 m2 as prognostic threshold after kidney cancer surgery. A, probability of all cause mortality 5 years after kidney cancer surgery based on postoperative GFR in patients with preoperative eGFR greater than 60 ml/minute/1.73 m2 but postoperative eGFR less than 60 ml/minute/1.73 m2 (CKD-S) and those with preoperative and postoperative eGFR less than 60 ml/minute/1.73 m2 (CKD-M/S). Reprinted with permission from Lane B, Demirjian S, Derweesh I et al: Eur Urol 2015; 68: 996. Copyright 2015 by Elsevier. B1, nonkidney cancer related survival in 2,202 patients with postoperative eGFR greater than 60 ml/minute/1.73 m2 (No CKD) after kidney cancer surgery, CKD-S in 931 and CKD M/S in 1,113. B2, CKD-S further divided into cohorts with postoperative GFR 45 to 60 and less than 45 ml/minute/1.73 m2. Note nonrenal cancer related deaths in each 2-year interval (No. of events) below each chart. Reprinted with permission.42
The contemporary urological literature has also begun to reflect increased appreciation of the importance of proteinuria when evaluating patients with kidney cancer.27,43,44 Recent studies have noted that approximately 20% of patients who undergo kidney cancer surgery have preexisting proteinuria.43–45 Using the KDIGO guideline nomenclature, up to 25% of patients with kidney cancer who have baseline CKD 3a are at high or very high risk for CKD progression.3,43,46 In a recent study Zhang et al noted that patients with kidney cancer and proteinuria had compromised 5-year overall survival compared with those without proteinuria (negative or trace on dipstick) (65% vs 77%) and a lower rate of stable eGFR at 5 years (72% vs 86%, each p <0.001).46 On multivariable analysis proteinuria was an independent prognostic factor for overall survival and kidney function stability (each p <0.05).
Observation 2 is that the similarity in survival between the PN and RN groups in EORTC-30904 may reflect that postoperative eGFR remained greater than 45 ml/minute/1.73 m2 in most patients in the 2 groups.11 In 2014 the EORTC investigators reported kidney function outcomes of the 2 trial groups, which was informative (fig. 4).29 At a median followup of 6.7 years postoperative eGFR less than 60 ml/minute/1.73 m2 was observed in 85.7% of patients who underwent RN vs 64.7% of those treated with PN (p <0.001). Postoperative eGFR less than 45 ml/minute/1.73 m2 was noted in 49% and 27.1% of RN and PN treated patients, respectively (p <0.001). Mean eGFR 1 year after surgery was 52.7 ml/minute/1.73 m2 (GFR category 3a) in the RN cohort and 66.8 ml/minute/1.73 m2 (GFR category 2) in the PN cohort. Also notable is that the difference in mean eGFR remained stable for several years, suggesting that a reduction in GFR due to surgery did not lead to progressive GFR decline even in the RN group.
Figure 4. Mean eGFR by time after surgery in patients assigned to RN or PN (NSS) in EORTC-30904 and percent with lowest recorded eGFR by treatment group specified binary cutoffs. Bars represent 95% CI. Reprinted with permission.29
Observation 3 is that the survival benefits of PN demonstrated in retrospective studies were likely influenced by selection bias. By 2011 abundant data from retrospective studies had been published suggesting a survival benefit for PN over RN. In fact, in a meta-analysis of 36 studies in a total of more than 41,000 patients Kim et al noted that PN correlated with a 19% reduction in all cause mortality and a 61% reduction in severe CKD.47 Interestingly, they also noted a 29% reduction in cancer specific mortality in favor of PN, which is best explained by selection bias. These results were directly contradicted by EORTC-30904.11 Kim et al cautioned that results should be interpreted in the context of "the low quality of existing evidence and the significant heterogeneity across studies."47
In the following year Shuch et al performed a matched cohort study comparing individuals treated with RN or PN of pT1a tumors to controls without cancer matched for age and comorbidity.48 Patients who underwent RN did not experience reduced survival compared to controls despite substantial loss of function. Patients treated with PN did better than controls, suggesting that they were generally healthier although they were matched for age and comorbid condition. These findings suggest that the survival advantage associated with PN in previous retrospective studies was likely strongly influenced by selection bias rather than by kidney function preservation.9,34,48
Further, a recent retrospective study in 3,133 patients with normal preoperative kidney function reported that those who underwent PN had better nonkidney cancer related survival than those treated with RN, although the functional dividend of PN, namely higher postoperative GFR, failed to be associated with survival.49 This again suggests that factors other than postoperative GFR are important and selection bias has an important role with respect to outcomes after PN or RN. Also, a 2018 analysis of PN and RN confirmed strong selection bias for clinical and pathological features in favor of PN. The authors stated that "the decision to perform PN vs RN is based on preoperative factors that are not captured in even very rich, granular institutional datasets," emphasizing the potential influence of unrecognized confounders in such studies.50
In regard to clinical implications, recent advances in our understanding of CKD and its relevance to kidney cancer surgery have direct implications on clinical management (supplementary Appendix, https://www.jurology.com).
Conclusions
It is evident that PN offers a significant kidney function advantage without compromising oncologic control of small renal masses. Observational and retrospective studies also demonstrate associations with reduced cardiovascular morbidity and nononcologic mortality. However, the results of EORTC-3090411 have prompted critical reevaluation of the impact of PN vs RN on noncancer related survival and our understanding of CKD as it pertains to kidney cancer surgery.
It has become evident that the presence and severity of CKD should not be defined solely as eGFR less than 60 ml/minute/1.73 m2. Additional factors, such as the cause of CKD, the level of GFR and the degree of albuminuria, can dramatically alter survival in patients after kidney cancer surgery. It has also become apparent that the survival advantage of PN in most retrospective studies was influenced by selection bias and the lack of an overall survival advantage for PN in EORTC-30904 was due to the fact that patients in the RN and PN cohorts were at relatively low risk for CKD mortality after surgery.11
Of course, these observations do not advocate for a return to RN as the reference standard for small renal masses. Rather, as the boundaries of elective PN continue to expand, particularly when considering patients at increased oncologic risk, the limitations of the evidence surrounding CKD after surgery are particularly relevant. A randomized trial of PN vs RN in such patients is necessary to improve our understanding of the relative merits of PN vs RN and the survival implications of postoperative kidney function.
References
- 1. : Management of small kidney cancers in the new millennium: contemporary trends and outcomes in a population-based cohort. JAMA Surg 2015; 150: 664. Google Scholar
- 2. : National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Intern Med 2003; 139: 137. Google Scholar
- 3. KDIGO 2012 Clinical Practice Guideline for the evaluation and management of chronic kidney disease. Kidney Int, suppl., 2013; 3: 1. Google Scholar
- 4. : Renal function after partial nephrectomy: effect of warm ischemia relative to quantity and quality of preserved kidney. Urology 2012; 79: 356. Google Scholar
- 5. : Every minute counts when the renal hilum is clamped during partial nephrectomy. Eur Urol 2010; 58: 340. Google Scholar
- 6. : Nonclamping partial nephrectomy: towards improved nephron sparing. Nat Rev Urol 2011; 8: 523. Google Scholar
- 7. : Impact of renal hilar control on outcomes of robotic partial nephrectomy: systematic review and cumulative meta-analysis. Eur Urol Focus 2019; 5: 619. Google Scholar
- 8. : Chronic kidney disease after nephrectomy in patients with renal cortical tumours: a retrospective cohort study. Lancet Oncol 2006; 7: 735. Google Scholar
- 9. : Long-term survival following partial vs radical nephrectomy among older patients with early-stage kidney cancer. JAMA 2012; 307: 1629. Google Scholar
- 10. : Elective partial nephrectomy in patients with clinical T1b renal tumors is associated with improved overall survival. Urology 2010; 76: 631. Google Scholar
- 11. : A prospective, randomised EORTC intergroup phase 3 study comparing the oncologic outcome of elective nephron-sparing surgery and radical nephrectomy for low-stage renal cell carcinoma. Eur Urol 2011; 59: 543. Google Scholar
- 12. : Do we know (or just believe) that partial nephrectomy leads to better survival than radical nephrectomy for renal cancer?World J Urol 2014; 32: 573. Google Scholar
- 13. : Trends in prevalence of chronic kidney disease in the United States. Ann Intern Med 2016; 165: 473. Google Scholar
- 14. : US Renal Data System 2018 annual data report: epidemiology of kidney disease in the United States. Am J Kidney Dis, suppl., 2019; 73: A7. Google Scholar
- 15. : Accelerated atherosclerosis in prolonged maintenance hemodialysis. N Engl J Med 1974; 290: 697. Google Scholar
- 16. : Kidney function as a predictor of noncardiovascular mortality. J Am Soc Nephrol 2005; 16: 3728. Google Scholar
- 17. : Rapid decline of kidney function increases cardiovascular risk in the elderly. J Am Soc Nephrol 2009; 20: 2625. Google Scholar
- 18. : Chronic kidney disease as a risk factor for cardiovascular disease and all-cause mortality: a pooled analysis of community-based studies. J Am Soc Nephrol 2004; 15: 1307. Google Scholar
- 19. : Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med 2004; 351: 1296. Google Scholar
- 20. : Evaluating the contribution of the cause of kidney disease to prognosis in CKD: results from the Study of Heart and Renal Protection (SHARP). Am J Kidney Dis 2014; 64: 40. Google Scholar
- 21. : KDOQI US commentary on the 2012 KDIGO clinical practice guideline for the evaluation and management of CKD.Am J Kidney Dis 2014; 63: 713. Google Scholar
- 22. : Multinational assessment of accuracy of equations for predicting risk of kidney failure: a meta-analysis. JAMA 2016; 315: 164. Google Scholar
- 23. : Matched comparison of radical nephrectomy vs nephron-sparing surgery in patients with unilateral renal cell carcinoma and a normal contralateral kidney. Mayo Clin Proc 2000; 75: 1236. Crossref, Medline, Google Scholar
- 24. : Natural history of chronic renal insufficiency after partial and radical nephrectomy. Urology 2002; 59: 816. Google Scholar
- 25. : 20 Years or more of follow-up of living kidney donors. Lancet 1992; 340: 807. Google Scholar
- 26. : No evidence of accelerated loss of kidney function in living kidney donors: results from a cross-sectional follow-up. Transplantation 2001; 72: 444. Google Scholar
- 27. : Long-term consequences of kidney donation. N Engl J Med 2009; 360: 459. Google Scholar
- 28. : Prevalence of proteinuria and other abnormalities in urinalysis performed in the urology clinic. Urology 2017; 103: 34. Google Scholar
- 29. : Renal function after nephron-sparing surgery versus radical nephrectomy: results from EORTC randomized trial 30904. Eur Urol 2014; 65: 372. Google Scholar
- 30. : New chronic kidney disease and overall survival after nephrectomy for small renal cortical tumors. Urology 2015; 86: 1137. Google Scholar
- 31. : Long-term renal function recovery following radical nephrectomy for kidney cancer: results from a multicenter confirmatory study. J Urol 2018; 199: 921. Link, Google Scholar
- 32. : Chronic kidney disease due to surgical removal of nephrons: relative rates of progression and survival. J Urol 2014; 192: 1057. Link, Google Scholar
- 33. : Radical nephrectomy for pT1a renal masses may be associated with decreased overall survival compared with partial nephrectomy. J Urol 2008; 179: 468. Link, Google Scholar
- 34. : Partial nephrectomy versus radical nephrectomy in patients with small renal tumors—is there a difference in mortality and cardiovascular outcomes?J Urol 2009; 181: 55. Link, Google Scholar
- 35. : Renal mass and localized renal cancer: AUA guideline. J Urol 2017; 198: 520. Link, Google Scholar
- 36. : Management of small renal masses: American Society of Clinical Oncology clinical practice guideline. J Clin Oncol 2017; 35: 668. Google Scholar
- 37. : EAU guidelines on renal cell carcinoma: 2014 update. Eur Urol 2015; 67: 913. Google Scholar
- 38. : Testing the external validity of the EORTC randomized trial 30904 comparing overall survival after radical nephrectomy vs nephron-sparing surgery in contemporary North American patients with renal cell cancer. BJU Int 2018; 121: 345. Google Scholar
- 39. : Surgically induced chronic kidney disease may be associated with a lower risk of progression and mortality than medical chronic kidney disease. J Urol 2013; 189: 1649. Link, Google Scholar
- 40. : Predictors of long-term survival after renal cancer surgery. J Urol 2018; 199: 384. Link, Google Scholar
- 41. : Survival and functional stability in chronic kidney disease due to surgical removal of nephrons: importance of the new baseline glomerular filtration rate. Eur Urol 2015; 68: 996. Google Scholar
- 42. : Analysis of survival for patients with chronic kidney disease primarily related to renal cancer surgery. BJU Int 2018; 121: 93. Google Scholar
- 43. : Proteinuria is a predictor of renal functional decline in patients with kidney cancer. J Urol 2016; 196: 658. Link, Google Scholar
- 44. : Impact of reduced glomerular filtration rate and proteinuria on overall survival of patients with renal cancer. J Urol 2016; 195: 588. Link, Google Scholar
- 45. : Association of urine dipstick proteinuria and postoperative renal function following robotic partial nephrectomy. J Endourol 2016; 30: 532. Google Scholar
- 46. : Proteinuria in patients undergoing renal cancer surgery: impact on overall survival and stability of renal function. Eur Urol Focus 2016; 2: 616. Google Scholar
- 47. : Comparative effectiveness for survival and renal function of partial and radical nephrectomy for localized renal tumors: a systematic review and meta-analysis. J Urol 2012; 188: 51. Link, Google Scholar
- 48. : Overall survival advantage with partial nephrectomy: a bias of observational data?Cancer 2013; 119: 2981. Google Scholar
- 49. : Renal cancer surgery in patients without preexisting chronic kidney disease: is there a survival benefit for partial nephrectomy?J Urol 2019; 201: 1088. Link, Google Scholar
- 50. : Radical versus partial nephrectomy for cT1 renal cell carcinoma. Eur Urol 2018; 74: 825. Google Scholar
The corresponding author certifies that, when applicable, a statement(s) has been included in the manuscript documenting institutional review board, ethics committee or ethical review board study approval; principles of Helsinki Declaration were followed in lieu of formal ethics committee approval; institutional animal care and use committee approval; all human subjects provided written informed consent with guarantees of confidentiality; IRB approved protocol number; animal approved project number.
No direct or indirect commercial, personal, academic, political, religious or ethical incentive is associated with publishing this article.
