Adverse Health Outcomes Associated with Surgical Management of the Small Renal Mass

Brian Shuch,*, Janet M. Hanley, Julie C. Lai, Srinivas Vourganti, Claude M. Setodji, Andrew W. Dick, Wong-Ho Chow, Chris S. Saigal and the Urologic Diseases in America Project
From the Department of Urology, Yale School of Medicine, New Haven, Connecticut (BS); Department of Urology, UCLA School of Medicine, Los Angeles (CSS), and RAND Corporation, Santa Monica (JMH, JCL, CMS, AWD), California; Department of Urology, SUNY Upstate Medical University, Syracuse, New York (SV); and Department of Epidemiology, MD Anderson Cancer Center, Houston, Texas (WHC)

Purpose: Partial and radical nephrectomy are treatments for the small renal mass. Partial nephrectomy is considered the gold standard as it may protect against renal dysfunction compared to radical nephrectomy. However, both treatments may cause adverse health outcomes. Materials and Methods: A matched cohort study was performed using the SEER (Surveillance, Epidemiology and End Results)-Medicare data set. Individuals treated with partial or radical nephrectomy for 4 cm or smaller nonmetastatic renal cell carcinoma were compared to 2 control groups (nonmuscle invasive bladder cancer and noncancer). A greedy algorithm matched surgical groups to controls. Medicare claims were examined for renal, cardiovascular and secondary cancer events. Results: Patients who underwent partial nephrectomy (1,471) and radical nephrectomy (4,299) were matched to controls. The time to event model demonstrated an increased risk of renal events for both treatments. Compared to the bladder cancer control and noncancer control groups, radical nephrectomy hazard ratios for renal events were 2.415 (p <0.0001) and 6.211 (p <0.0001), respectively, while partial nephrectomy hazard ratios were 1.513 (p <0.0001) and 4.926 (p <0.0001), respectively. Secondary cancers were increased for partial nephrectomy and radical nephrectomy compared to both control groups (p <0.0001). Cardiovascular events were increased for both treatments compared to noncancer controls (p <0.0001), but not compared to bladder cancer controls. Conclusions: Partial nephrectomy and radical nephrectomy may lead to adverse health outcomes. Compared to controls, partial nephrectomy and radical nephrectomy are associated with worsened renal outcomes. The increase in secondary cancers and cardiovascular events with both treatments is notable, and requires further investigation. Further research should investigate if active surveillance of the appropriately selected small renal mass limits adverse health outcomes. Key Words: watchful waiting, nephrectomy, kidney neoplasms SURGICAL management has been the mainstay of treatment for renal cell carcinoma. Despite an increased incidence of the disease, RCC mortality rates have remained fairly stable.1,2 These ndings led investigators
http://dx.doi.org/10.1016/j.juro.2013.08.074 Vol. 191, 301-309, February 2014 Printed in U.S.A.

Abbreviations and Acronyms AS active surveillance BCC bladder cancer control CCI Charlson comorbidity index CKD chronic kidney disease CSS cancer specific survival ESRD end stage renal disease HTN hypertension NCC noncancer control PN partial nephrectomy RCC renal cell carcinoma RN radical nephrectomy SRM small renal mass
Accepted for publication August 26, 2013. Supported by the National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases (HHSN276201200016C). * Correspondence: Department of Urology, Yale School of Medicine, 789 Howard Ave., FMP300, New Haven, Connecticut 06519 (e-mail: brian.shuch@yale.edu). Nothing to disclose. Financial interest and/or other relationship with WiserCare.

For other articles on related topics see pages 479 and 487. Editors Note: This article is the rst of 5 published in this issue for which category 1 CME credits can be earned. Instructions for obtaining credits are given with the questions on pages 564 and 565.

0022-5347/14/1912-0301/0 THE JOURNAL OF UROLOGY 2014 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.

www.jurology.com

301

302

POTENTIAL RISKS OF OVERTREATMENT OF SMALL RENAL MASS

to hypothesize that surgery for the small renal mass is often unnecessary. Hollingsworth et al suggested, the current paradigm for treating kidney cancer is not based on empiricism, and these ndings call to question the appropriateness of extirpative surgery in all patients.2 Over diagnosis is an increasingly prevalent phenomenon as many patients undergo treatment for cancers that may have remained clinically silent.3 For prostate cancer the treatment related side effects (impotence and incontinence) are easily recognizable and quantiable. Adverse health outcomes associated with the treatment of RCC involve long-term consequences from nephron loss as opposed to the immediate impact seen after the treatment of prostate cancer. The extent of chronic kidney disease has recently been linked to cardiac mortality, which is also believed to be related to loss of renal function after treatment of RCC.4,5 Because of these concerns, partial nephrectomy expanded to an elective setting.6 In addition to limiting the removal of normal parenchyma, PN offers an oncologic outcome similar to that of radical nephrectomy.7,8 Multiple series have demonstrated that compared to RN, PN may protect against CKD and proteinuria.912 CKD may also increase cancer risk, which is a new concern with RCC overtreatment.13,14 Recently PN has been labeled the standard of care for the SRM.15 However, studies comparing PN and RN have demonstrated renal events with both treatments.9,12 With data emerging on the safety of active surveillance, it is important to understand adverse health outcomes associated with current surgical paradigms.6 While studies compare RN to PN, to our knowledge none has assessed surgery vs no intervention or AS. Ideally a randomized clinical trial would improve our understanding of the risk/ benet ratio of SRM treatment since AS may also have risks, including anxiety and tumor progression, and those related to frequent imaging. To better understand the adverse health outcomes associated with surgical management, we conducted a matched cohort study using the SEERMedicare linkage.16 We measured cardiovascular, renal and secondary cancer events associated with PN and RN compared to controls.

METHODS
SEER-Medicare was used to identify individuals 66 years old or older diagnosed with RCC from 1992 to 2007 with 2 or more years of followup. As SEER-Medicare claims are available through 2009, this served as the end of followup. Patients with RCC were identied in SEER using ICD-O-3 (International Classication of Diseases for Oncology, 3rd Edition) codes.12 To exclude nonRCC renal

malignancies, only recognizable subtypes were selected. Inpatient Medicare les for ICD-9 (International Classication of Diseases, 9th Revision) claims identied patients treated with PN (55.4) or RN (55.5, 55.51). Cases of bilateral nephrectomy (55.54), or 2 or more surgical claims were excluded from study. We also excluded patients with nodal or metastatic disease at presentation. SRMs were dened as tumors measuring 4 cm or less. No exclusion was used for pathological up staging. Demographic variables including age, gender, race, region and treatment year were obtained. The inclusion of patients 66 years old or older allowed more than 85% of comorbidities to be captured.17 Inpatient and outpatient claims were used to calculate CCI.18 Patients with ESRD at diagnosis were excluded from study. Differences exist between patients undergoing PN and RN and, therefore, each category was separately matched to controls.19 Noncancer controls included individuals without a prior cancer chosen from a 5% random sample of Medicare beneciaries 66 years old or older. To control for nonmeasurable differences in health characteristics and followup practices between cancer and noncancer groups, we selected a separate urological cancer control. Nonmuscle invasive bladder cancer was chosen due to excellent cancer specic survival and similar oncologic surveillance. The bladder cancer controls were selected from SEER-Medicare, and included patients 66 years old or older with low grade (grade 1 or 2) nonmuscle invasive bladder cancer (Ta, T1 and Tis) who did not undergo cystectomy.20 As different cancers have unique biology, comparisons were performed only for adverse health outcomes assessment. A greedy algorithm was chosen to match RCC cases to controls. With the greedy algorithm, a randomly treated subject is matched to controls by the closest propensity score based on dened covariates.21 For surgical groups the controls were matched by age (3 years), year of diagnosis, race, gender, CCI and HTN. Hypertension was chosen since it is not represented in the CCI. Year of surgery was used to match NCCs using the rst month of Medicare eligibility. We evaluated claims for cardiovascular, renal and secondary cancer events more than 30 days postoperatively to exclude perioperative complications.12 We used CKD cardiovascular event coding similar to that used by Go et al.4 Renal events were coded as in the study by Miller et al, and included dialysis services, ESRD, transplantation, nephrology consultation and ESRD hospitalizations.12 Secondary cancers were identied using ICD-9 codes, excluding benign tumors. In the RCC and BCC groups the same cancer was excluded as this may represent recurrence. Multiple cancer events were consolidated into a solitary outcome. Claims were evaluated through the end of followup or death. We examined the CSS associated with PN, RN and the BCC group calculated from surgery until death or last followup. The chi-square test, ANOVA and generalized linear modeling were used to evaluate differences between groups. Kaplan-Meier estimates were performed for time to event and CSS. Differences were calculated using log rank tests. Cox proportional hazard models assessed the

POTENTIAL RISKS OF OVERTREATMENT OF SMALL RENAL MASS

303

inuence of treatment. Forest plots evaluated adverse health outcomes for each RCC treatment compared to controls.

p Value

0.61 1

3,646 (85.85) 272 (6.4) 198 (4.66) 131 (3.08)

1 1,819 (42.83) 2,428 (57.17) 1,819 (42.83) 2,428 (57.17) 1 1,836 (42.71) 2,463 (57.29) 1,836 (42.71) 2,463 (57.29) 1 694 (48.33) 742 (51.67) 694 (48.33) 742 (51.67) 1 715 (48.61) 756 (51.39) 715 (48.61) 756 (51.39)

We identied 7,964 Medicare beneciaries, of which 5,770 met the inclusion criteria, including 1,471 treated with PN and 4,299 with RN. For the PN cohort 1,436 NCCs and 1,471 BCCs were identied. For the RN cohort 4,247 NCCs and 4,299 BCCs were identied. Patients were exactly matched for age, gender, race, HTN and CCI categories (table 1). Median followup for the kidney cancer, BCC and NCC groups was 4.1, 4.1 and 5.8 years, respectively. The 5 and 10-year CSS for the PN, RN and BCC groups was 98.4% and 97.9%, 97.4% and 95.7%, and 97.4% and 95.8%, respectively. In the PN group cardiovascular, renal and secondary cancer events occurred in 932 (63.8%), 245 (16.7%) and 633 (43.0%) patients, respectively. For the BCC group cardiovascular, renal and secondary cancer events occurred in 996 (67.7%), 175 (11.9%) and 517 (35.2%) individuals. The NCC group had cardiovascular, renal and secondary cancer events occur in 937 (65.3%), 107 (7.5%) and 421 (29.3%) patients, respectively. In the RN group cardiovascular, renal and secondary cancer events occurred in 2,903 (67.5%), 1,166 (27.1%) and 1,846 (42.9%) patients, respectively. For the BCC group cardiovascular, renal and secondary cancer events occurred in 3,033 (70.6%), 570 (13.3%) and 1,700 (39.5%) individuals. The NCC group had cardiovascular, renal and secondary cancer events in 2,928 (68.9%), 408 (9.1%) and 1,150 (27.1%) patients, respectively. Compared to their respective BCC groups, neither the PN group (p 0.069) nor the RN group (p 0.228) had increased cardiovascular events (g. 1). Compared to NCCs, the PN (p <0.0001) and RN (p <0.0001) groups experienced increased cardiovascular events. The Cox proportional hazard model demonstrated a hazard ratio of 2.70 (95% CI 2.38e3.09, p <0.0001) for PN and 2.63 (CI 2.44e2.83, p <0.0001) for RN compared to the respective NCC groups (table 2). Kaplan-Meier estimates of renal events are shown in gure 2. Compared to the respective BCC group, patients treated with PN (p <0.0001) and RN (p <0.0001) experienced increased renal events. Similarly, compared to the respective NCC groups, more renal events were observed with PN (p <0.0001) and RN (p <0.0001). The HR for renal events with PN compared to BCCs and NCCs was 1.51 (CI 1.25e1.84, p <0.0001) and 4.93 (CI 3.70e6.54, p <0.0001), respectively. The HR for renal events with RN compared to BCCs and NCCs

4,247 74.71 (75) 2,252/1,995 53.03

3,646 (85.85) 272 (6.4) 198 (4.66) 131 (3.08)

Table 1. Demographics and clinical features of selected patients treated with PN and RN, and their respective bladder cancer and noncancer controls

RN

p Value

0.45 1

3,751 (87.25) 244 (5.68) 185 (4.3) 119 (2.77)

4,299 74.91 (75) 2,361/1,938 54.92

BCC

4,299 74.82 (75) 2,361/1,938 54.92

3,751 (87.25) 244 (5.68) 185 (4.3) 119 (2.77)

RN

p Value

0.37 1

1,436 73.57 (73) 818/618 56.96

(84.68) (6.82) (5.01) (3.48)

(84.68) (6.82) (5.01) (3.48)

1,216 98 72 50

1,436 73.25 (73) 818/618 56.96

p Value

0.47 1

1,216 98 72 50

1,471 73.38 (73) 875/596 59.48

BCC

1,268 87 68 48

1,471 73.25 (73) 875/596 59.48

PN

No. pts Mean pt age (median) No. male/female % Male No. race (%): White Black Hispanic Other No. prior HTN (%): Yes No No. CCI (%): 0 1 2 3 or Greater

1,268 87 68 48

940 346 106 79

(20.77) (23.52) (7.21) (5.37)

(86.2) (5.91) (4.62) (3.26)

940 346 106 79

(20.77) (23.52) (7.21) (5.37)

(86.2) (5.91) (4.62) (3.26)

924 (64.35) 329 (22.91) 112 (7.8) 71 (4.94)

PN

924 (64.35) 329 (22.91) 112 (7.8) 71 (4.94)

NCC

2,787 968 341 203

(64.83) (22.52) (7.93) (4.72)

2,787 968 341 203

(64.83) (22.52) (7.93) (4.72)

2,773 947 337 190

(65.29) (22.3) (7.94) (4.47)

2,773 947 337 190

RESULTS

4,247 74.77 (74) 2,252/1,995 53.03

NCC

(65.29) (22.3) (7.94) (4.47)

304

POTENTIAL RISKS OF OVERTREATMENT OF SMALL RENAL MASS

Figure 1. Kaplan-Meier estimate of time to cardiovascular event. PN (black) compared to bladder cancer (A) and noncancer (B) controls (red), and RN (black) compared to bladder cancer (C ) and noncancer (D) controls (red).

was 2.42 (CI 2.19e2.67, p <0.0001) and 6.21 (CI 5.41e7.14, p <0.0001), respectively (table 2). Figure 3 shows the Kaplan-Meier estimates of secondary cancers, which were more common with PN (p <0.0001) and RN (p <0.0001) than in the BCC and NCC groups. The HR for cancer events in the PN group compared to the BCC and NCC groups was 1.36 (CI 1.21e1.53, p <0.0001) and 4.46 (CI 3.76e5.29, p <0.0001), respectively. The HR for cancer events with RN compared to BCCs and NCCs was 1.20 (CI 1.12e1.28, p <0.0001) and 4.43 (CI 4.00e4.88, p <0.0001), respectively (table 2). Figure 4 presents the magnitude of risk with PN and RN compared to controls. Figure 4, A

demonstrates that the increased cardiovascular risk appears similar between PN and RN. Figure 4, B shows that renal events are similar compared to NCCs. However, compared to BCCs, the magnitude of risk appears greater with RN, an observation consistently seen in other studies comparing RN to PN. Figure 4, C shows that PN and RN have a similarly increased cancer risk compared to controls.

DISCUSSION
Our study has several signicant ndings. We demonstrated that RN and PN increase renal

Table 2. Cox proportional hazard model of adverse health outcomes for PN and RN compared to noncancer and bladder cancer controls HR (95% CI) Cardiovascular Events PN vs: NCC BCC RN vs: NCC BCC * All values p <0.0001. 2.70 (2.375e3.086) 0.92 (0.842e1.007) 2.63 (2.439e2.825) 0.97 (0.921e1.019) p Value <0.0001 0.0696 <0.0001 0.2287 HR (95% CI) Renal Events* 4.93 (3.704e6.536) 1.51 (1.248e1.835) 6.21 (5.405e7.143) 2.42 (2.188e2.674) HR (95% CI) Secondary Ca Events* 4.46 (3.759e5.291) 1.36 (1.209e1.527) 4.42 (4.000e4.878) 1.20 (1.120e1.279)

POTENTIAL RISKS OF OVERTREATMENT OF SMALL RENAL MASS

305

Figure 2. Kaplan-Meier estimate of time to renal event. PN (black) compared to bladder cancer (A) and noncancer (B) controls (red), and RN (black) compared to bladder cancer (C ) and noncancer (D) controls (red).

Figure 3. Kaplan-Meier estimate of time to cancer event. PN (black) compared to bladder cancer (A) and noncancer (B) controls (red), and RN (black) compared to bladder cancer (C ) and noncancer (D) controls (red).

306

POTENTIAL RISKS OF OVERTREATMENT OF SMALL RENAL MASS

Figure 4. Forest plots comparing attributable risk of cardiovascular events (A), renal events (B) and cancer events (C ) associated with each surgical treatment compared to respective controls using linear scale. (Note: PN and RN groups are compared to separately matched patients.)

events. PN, the current gold standard, signicantly increased renal events with a HR of approximately 1.5 and 5 compared to BCCs and NCCs, respectively. In addition, we noted a similar 2.5-fold increased risk of cardiovascular events with PN and RN compared to NCC. If there is a difference in the risk of cardiovascular events associated with one type of treatment for RCC, it may be small and difcult to quantify. Finally, we found that PN and RN were associated with increased cancer events. Even when compared to BCCs, which may have a higher smoking prevalence, PN and RN increased the risk of cancer approximately 1.4 and 1.2-fold, respectively. The question of whether PN offers a survival advantage compared to RN for the small renal mass is still controversial.8,22 Recently we questioned whether the survival advantage in claims based

data is a result of selection bias.23 Without denitive data, many clinicians favor PN due to the increased renal events associated with RN observed in single institutional and claims based comparisons.9,12 With these concerns in mind, in 2009 the AUA (American Urological Association) issued guidelines considering PN the gold standard for the T1a tumor when technically feasible.6 Despite the perceived health risks associated with RN, there were similar cardiovascular deaths in EORTC (European Organization for the Research and Treatment of Cancer) 30904.8 Claims based comparisons have also been unable to demonstrate differences in cardiovascular events between PN and RN.12,19 However, while this may suggest no difference in cardiovascular risk, it is also possible that both procedures cause equivalent harm. In fact, both types of surgery are associated with a high

POTENTIAL RISKS OF OVERTREATMENT OF SMALL RENAL MASS

307

incidence (60%) of cardiovascular events.12 CKD (33% stage 3 or greater at 5 years) and proteinuria (34.5%) may also develop in patients after undergoing PN.12,13 Perhaps the mitigating effect of PN on the development of CKD should be viewed as less harmful rather than as protective. While some patients with a SRM require treatment, many may not. Therefore, an understanding of adverse health outcomes with treatment is critical. Ideally a clinical study evaluating treatment vs AS in patients with a SRM would improve our understanding of the risk/benet ratio of either approach. As such data are not available, we used SEER-Medicare to perform a matched cohort study to better understand the risks associated with surgical treatment. Matching patients with PN and those with RN to controls allowed us to assess adverse health outcomes. We chose 2 control groups as both may help control unmeasured cofounders. In addition to differences in health, the NCC group may have fewer followup visits than the cancer control group, hindering the capture of health events. BCCs were chosen not to study disease behavior but because they undergo periodic cancer surveillance in a manner similar to that seen in patients with RCC. However, one could argue these patients are less healthy due to a higher incidence of smoking,24 which is poorly characterized using SEER-Medicare.16 Due to these concerns we believed both controls were informative, with true risk estimates potentially lying between both controls. Our data reveal long-term adverse events associated with PN and RN. While PN is believed to cause less renal harm, decreased renal function can occur with removal of adjacent normal parenchyma, damage to segmental arteries and long ischemic intervals. While increased cardiovascular events were observed for both surgeries compared to the NCC group, this was not observed compared to the BCC group. This nding may be explained, in part, by a greater number of cardiovascular events in patients with bladder cancer given that tobacco is a cardiovascular risk factor. Increased secondary cancers were associated with PN and RN compared to both control groups. This unexpected nding should be considered hypothesis generating. The increased risk of secondary cancers in the RCC groups requires validation and further investigation to better understand this association. Possible mechanisms could include worsened postoperative renal function increasing cancer risk, radiation exposure from frequent computerized tomography surveillance or complex immunologic factors that put all patients with RCC (even those treated

with surveillance) at higher risk for secondary cancers. Alternatively, biological risk factors for RCC may increase the risk of specic secondary cancers. While our research methodology was not designed to investigate differences in specic cancers, it was noted that both kidney groups had an increase in breast, gastrointestinal, genitourinary, hematological and skin cancers compared to the BCC group (p <0.05) (data not shown). There were fewer thoracic cancers, possibly related to differences in smoking (p <0.01). For further insight our ndings should be evaluated in a cohort of patients treated only with surveillance for RCC. An interesting nding is that in a similar cohort used in a different analysis overall survival improved in those with PN, despite an increase in adverse health outcomes.23 This nding was believed to be due to nonmeasured variables that could not be controlled for using claims based data. One explanation is that the coding of events many vary in severity and many are not necessarily life threatening. Thus, while events may occur more often in a particular group, they may not greatly inuence survival. Similar ndings can be observed in recent data from the EORTC 30904 trial where, despite an increased number of renal events (estimated glomerular ltration rate less than 60 ml/minute/1.73 m2) with RN, there was no difference in overall survival.25 In 2009 the AUA guidelines did not consider AS the standard for the T1a mass. However, the panel recommended this as an acceptable approach for high risk patients. Since then, additional series have demonstrated a low incidence of progression with AS.2630 With AS delayed treatment rarely causes unfavorable consequences such as up staging and progression.30 With the safety of AS becoming increasingly established and our ndings showing that both RCC surgical modalities share signicant adverse health outcomes, it may be time to reevaluate whether more patients should be considered for AS. While we are not arguing to change the current surgical paradigm for RCC, the potential risks associated with treatment must be weighed against the potential benets of surgical extirpation, especially in those with advanced age and comorbidity. Our study has some limitations, many of which are related to the use of SEER-Medicare data. Our analysis was restricted to the elderly. However, the median age of patients with RCC is approximately 64 years, which is younger than this population. As age is known to inuence the use of PN, fewer patients may have been offered this treatment approach. However, due to the methodology of matching cases to controls, this would not affect

308

POTENTIAL RISKS OF OVERTREATMENT OF SMALL RENAL MASS

our observations. SRMs are often incidentally discovered with imaging. Using claims based data, the reasons for such imaging are impossible to ascertain and could result in unmeasured confounders. Matching cases and controls was performed using several important variables. However, other factors such as obesity or smoking status may differ between groups. In addition, followup imaging frequency differs among cancer groups. Finally, the secondary cancers associated with RCC might be related to hereditary syndromes. However, these are rare, many are associated with benign tumors and most have an early onset, generally excluded by our methodology.

CONCLUSIONS
While the comparative effectiveness of partial and radical nephrectomy for the SRM is frequently debated, both options may increase the risk of renal and cardiovascular events compared to controls. We observed an increased risk of secondary cancers after both treatments. Whether this is related to decreasing kidney function requires further study. Signicant adverse health outcomes are associated with both surgical interventions in the Medicare population. Further investigation is needed to learn if AS could mitigate these risks and afford similar survival. If this is demonstrated, reassessment of the current surgical treatment paradigm for the SRM may be in order.

REFERENCES
1. Chow WH, Devesa SS, Warren JL et al: Rising incidence of renal cell cancer in the United States. JAMA 1999; 281: 1628. 2. Hollingsworth JM, Miller DC, Daignault S et al: Rising incidence of small renal masses: a need to reassess treatment effect. J Natl Cancer Inst 2006; 98: 1331. 3. Welch HG and Black WC: Overdiagnosis in cancer. J Natl Cancer Inst 2010; 102: 605. 4. Go AS, Chertow GM, Fan D et al: Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med 2004; 351: 1296. 5. Matsushita K, van der Velde M, Astor BC et al: Association of estimated glomerular filtration rate and albuminuria with all-cause and cardiovascular mortality in general population cohorts: a collaborative meta-analysis. Lancet 2010; 375: 2073. 6. Campbell SC, Novick AC, Belldegrun A et al: Guideline for management of the clinical T1 renal mass. J Urol 2009; 182: 1271. 7. Belldegrun A, Tsui KH, deKernion JB et al: Efficacy of nephron-sparing surgery for renal cell carcinoma: analysis based on the new 1997 tumor-node-metastasis staging system. J Clin Oncol 1999; 17: 2868. 8. Van Poppel H, Da Pozzo L, Albrecht W et al: 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. 9. Huang WC, Levey AS, Serio AM et al: Chronic kidney disease after nephrectomy in patients with renal cortical tumours: a retrospective cohort study. Lancet Oncol 2006; 7: 735. 10. Lau WK, Blute ML, Weaver AL et al: Matched comparison of radical nephrectomy vs nephronsparing surgery in patients with unilateral renal cell carcinoma and a normal contralateral kidney. Mayo Clin Proc 2000; 75: 1236. 11. Weight CJ, Larson BT, Gao T et al: Elective partial nephrectomy in patients with clinical T1b renal tumors is associated with improved overall survival. Urology 2010; 76: 631. 12. Miller DC, Schonlau M, Litwin MS et al: Renal and cardiovascular morbidity after partial or radical nephrectomy. Cancer 2008; 112: 511. 13. Wong G, Hayen A, Chapman JR et al: Association of CKD and cancer risk in older people. J Am Soc Nephrol 2009; 20: 1341. 14. Stengel B: Chronic kidney disease and cancer: a troubling connection. J Nephrol 2010; 23: 253. 15. Huang WC and Donin NM: Partial nephrectomy is the standard of care for T1a kidney tumors. Urol Oncol 2013; 31: 140. 16. Engels EA, Pfeiffer RM, Ricker W et al: Use of Surveillance, Epidemiology, and End ResultsMedicare data to conduct case-control studies of cancer among the US elderly. Am J Epidemiol 2011; 174: 860. 17. Klabunde CN, Harlan LC and Warren JL: Data sources for measuring comorbidity: a comparison of hospital records and Medicare claims for cancer patients. Med Care 2006; 44: 921. 18. Klabunde CN, Potosky AL, Legler JM et al: Development of a comorbidity index using physician claims data. J Clin Epidemiol 2000; 53: 1258. 19. Huang WC, Elkin EB, Levey AS et al: Partial nephrectomy versus radical nephrectomy in patients with small renal tumorseis there a difference in mortality and cardiovascular outcomes? J Urol 2009; 181: 55. 20. Chamie K, Saigal CS, Lai J et al: Quality of care in patients with bladder cancer: a case report? Cancer 2012; 118: 1412. 21. Gu XS and Rosenbaum PR: Comparison of multivariate matching methods: structures, distances, and algorithms. J Comput Graph Stat 1993; 2: 405. 22. Tan HJ, Norton EC, Ye Z et al: Long-term survival following partial vs radical nephrectomy among older patients with early-stage kidney cancer. JAMA 2012; 307: 1629. 23. Shuch B, Hanley J, Lai J et al: Overall survival advantage with partial nephrectomy: a bias of observational data? Cancer 2013; 119: 2981. 24. Yun YH, Jung KW, Bae JM et al: Cigarette smoking and cancer incidence risk in adult men: National Health Insurance Corporation Study. Cancer Detect Prev 2005; 29: 15. 25. Scosyrev E, Messing EM, Sylvester R et al: Renal function after nephron-sparing surgery versus radical nephrectomy: results from EORTC randomized trial 30904. Eur Urol 2013; Epub ahead of print. 26. Crispen PL, Viterbo R, Boorjian SA et al: Natural history, growth kinetics, and outcomes of untreated clinically localized renal tumors under active surveillance. Cancer 2009; 115: 2844. 27. Jewett MA, Mattar K, Basiuk J et al: Active surveillance of small renal masses: progression patterns of early stage kidney cancer. Eur Urol 2011; 60: 39. 28. Kunkle DA, Crispen PL, Chen DY et al: Enhancing renal masses with zero net growth during active surveillance. J Urol 2007; 177: 849. 29. Mason RJ, Abdolell M, Trottier G et al: Growth kinetics of renal masses: analysis of a prospective cohort of patients undergoing active surveillance. Eur Urol 2011; 59: 863. 30. Crispen PL, Viterbo R, Fox EB et al: Delayed intervention of sporadic renal masses undergoing active surveillance. Cancer 2008; 112: 1051.

POTENTIAL RISKS OF OVERTREATMENT OF SMALL RENAL MASS

309

EDITORIAL COMMENT
The cardiovascular and metabolic sequelae of extirpative renal surgery are well described.1,2 Nonetheless, the authors ndings regarding the increased oncologic risk of secondary malignancies are provocative, and should stimulate further investigation into the metabolic and oncologic sequelae of extirpative surgery for RCC. There are major methodological issues to consider. The authors used nonmuscle invasive bladder cancer as a control. However, T1a RCC has a more aggressive biological prole. The authors counterargument that they are analyzing overall health related and not oncologic outcomes does not entirely hold. The picture emerging with renal cortical oncogenesis is complex. A signicant proportion of sporadic RCCs are driven by a set of metabolic, immunologic and environmental risk factors that also contribute to development of secondary malignancies and other adverse effects.3 Therefore, association may be confounded with causality and the reader is cautioned not to overinterpret the authors conclusions. Discussion of these ndings, especially in young and healthy patients, should be balanced with the knowledge that RCC is the deadliest urological malignancy, in which salvage therapy for recurrent or locally advanced disease is far more problematic than most other malignancies.
Ithaar H. Derweesh
University of California, San Diego La Jolla, California

REFERENCES
1. Sun M, Bianchi M, Hansen J et al: Chronic kidney disease after nephrectomy in patients with small renal masses: a retrospective observational analysis. Eur Urol 2012; 62: 696. 2. Bagrodia A, Mehrazin R, Bazzi WM et al: Comparison of rates and risk factors for development of osteoporosis and fractures after radical or partial nephrectomy. Urology 2011; 78: 614. 3. Russo P: End stage and chronic kidney disease: associations with renal cancer. Front Oncol 2012; 2: 28.

REPLY BY AUTHORS
We appreciate the comments that the work is provocative and deserves further investigation. Derweesh mentions that sequelae of extirpative renal surgery have been described. However, this has only been done by comparing outcomes of partial and radical nephrectomy to each other (references 2 and 3 in Editorial Comment) and not by assessing the overall impact of both surgical therapies compared to a nontreatment group. Without an available active surveillance cohort, matching to a cancer control group allowed comparisons of long-term, adverse health outcomes in patients with similar excellent cancer specic survival. In his argument Derweesh provides further support for the use of a cancer control group, pointing out similarities between our control group and kidney cancer, as bladder cancer is also complex, and is similarly associated with metabolic, immunologic and environmental risk factors.1e3

REFERENCES
1. Liu X, Ji J, Forsti A, Sundquist K et al: Autoimmune disease and subsequent urological cancer. J Urol 2013; 189: 2262. 2. H aggstr om C, Stocks T, Rapp K et al: Metabolic syndrome and risk of bladder cancer: prospective cohort study in the Metabolic Syndrome and Cancer project (Me-Can). Int J Cancer 2011; 128: 1890. 3. Leta siov a S, Medveov a A,  Sov c kov a A et al: Bladder cancer, a review of the environmental risk factors. Environ Health, suppl., 2012; 11: S11.