Abstract
Introduction
The benefit of partial nephrectomy (PN) compared to radical nephrectomy (RN) for T1a renal cell carcinoma (RCC) remains uncertain, with observational studies conflicting with level 1 evidence. Therefore, the purpose of this population-based study was to compare long-term outcomes in patients undergoing PN or RN for T1a RCC.
Methods
We studied 5670 patients in Ontario, Canada undergoing PN or RN for T1a RCC. The primary outcome was overall survival (OS). Secondary outcomes were cancer-specific survival (CSS), chronic kidney disease (CKD), renal replacement therapy, and myocardial infarction (MI). We used multivariable Cox proportional hazard models to evaluate the association between PN or RN and these outcomes. A sensitivity analysis was performed in patients with a preoperative serum creatinine available.
Results
Median followup was 77 months. Compared to RN, PN was associated with significantly improved OS (hazard ratio [HR] 0.73, 95% confidence interval [CI] 0.63–0.84), reduced risk of CKD (HR 0.18, 95% CI 0.12–0.27), and improved CSS (HR 0.45, 95% CI 0.30–0.65). The risk of MI was not significantly different between groups (HR 0.91, 95% CI 0.62–1.34). Few patients (n=15) required renal replacement therapy. In the sensitivity analysis, the association between type of surgery and OS and CKD persisted, while the association with CSS did not.
Conclusions
Our study found that in patients undergoing surgery for T1a RCC, PN was associated with improved OS and reduced risk of CKD compared to RN. However, few patients in either group required renal replacement therapy.
Introduction
The incidence of kidney cancer is increasing in several countries.1 This increase is thought to be due to the rising prevalence of obesity and hypertension, both of which are established risk factors for kidney cancer,2 and the increased use of diagnostic imaging.3 The latter may explain the stage migration that has been observed over time, with the vast majority of tumors detected in the modern era being stage T1a tumors (tumors <4 cm).4
Several guidelines recommend that patients with clinical stage T1a tumors be managed preferentially with partial nephrectomy (PN) over radical nephrectomy (RN).5–7 This recommendation is based on several observational studies demonstrating that PN is associated with a reduced risk of renal dysfunction and improved overall survival (OS) compared to RN.8 However, many of these observational studies have been limited by sample size, followup, and inclusion of patients with heterogenous kidney cancer characteristics. Furthermore, few have been population-based, limiting their generalizability. The only randomized trial comparing these surgical approaches validated the increased risk of renal dysfunction with RN;9 however, OS in this trial was improved in the RN arm.10
Given that the therapeutic benefit of PN remains uncertain in patients presenting with clinical stage T1a kidney cancer,8 the most common stage of presentation in the present era, we performed a population-based study evaluating long-term survival and renal disease following PN compared to RN in these patients.
Methods
Setting and design
We performed a population-based cohort study of kidney cancer patients undergoing nephrectomy for stage T1a kidney cancer between 1995 and 2014 using linked administrative databases from Ontario, Canada. Our study was approved by the Research Ethics Board at the University Health Network, Toronto, Ontario, Canada.
Data sources
We used the Canadian Institute for Health Information Discharge Abstract Database, Same Day Surgery Database, National Ambulatory Care Reporting System, and Ontario Health Insurance Plan databases to obtain information on use of healthcare services and hospitalizations; the Ontario Cancer Registry to obtain information on cancer diagnosis date and cause of death, where applicable; and the Registered Person’s Database to obtain patient demographics, including date of birth, gender, place of residence via postal code, and date of death. We abstracted pathology records from Cancer Care Ontario and linked them with the administrative database records. Several of these databases have been validated and have been described in detail elsewhere.11
Study patients
To derive a cohort of patients undergoing a single nephrectomy for stage T1a renal cell carcinoma (RCC), we first identified hospitalizations containing a record for a PN or RN. We then linked these records with the Ontario Cancer Registry and only records with a kidney cancer diagnosis date within 14 days of nephrectomy date were kept. These records were linked to abstracted pathology reports, which contained information on histology and tumor size, and only records with histology consistent with RCC and maximal tumor size ≤4.0 cm were kept. To compare only a single PN vs. RN, we further excluded patients with any nephrectomy prior to or following the identified nephrectomy for kidney cancer.
Assessment of exposure
The type of nephrectomy, PN vs. RN, was based on the recorded procedure code during the relevant hospitalization. The date of nephrectomy was considered the index date.
Assessment of outcomes
The primary outcome was OS, defined as the time from the date of nephrectomy to death from any cause, or December 31, 2016, whichever came first. The secondary outcomes were time to diagnosis of chronic kidney disease (CKD), defined as time from nephrectomy to date of CKD diagnosis, or March 31, 2015, whichever came first; time to renal replacement therapy, defined as time from nephrectomy to first hospitalization code for kidney transplant or dialysis, or March 31, 2015, whichever came first; time to myocardial infarction (MI), defined as time from nephrectomy to date of MI diagnosis, or March 31, 2015, whichever came first; and cancer-specific mortality, defined as time from nephrectomy to death from kidney cancer, or December 31, 2011, whichever came first. Deaths within 30 days of nephrectomy were attributed to death from kidney cancer.12 The end dates were chosen based on the last update for the relevant databases at the time that the study cutoff date.
Statistical analysis
We compared baseline characteristics using standardized differences, whereby a threshold of >0.10 indicated a significant difference.13 We conducted time-to-event analysis using multivariable Cox proportional hazard regression to estimate the association of type of nephrectomy on the risk of the primary and secondary outcomes. The proportional hazards assumption was verified by evaluating Schoenfield residuals.14 For the secondary outcomes, we estimated the cause-specific hazard, as we were interested in understanding the potential etiology of kidney cancer survival outcomes related to type of nephrectomy.15 Covariates in the multivariable model were chosen a priori and included age, income quintile, Charlson score, year of surgery, tumor size, and histology. We confirmed the absence of significant collinearity based on the variance inflation factor.16 For the renal function outcomes, we excluded patients with any previous history of diabetes, hypertension, CKD, or renal replacement therapy to avoid potential bias of preferential use of PN based on specific comorbidities. All statistical analyses were performed using SAS (version 9.3; SAS Institute, Cary, NC, U.S.).
Sensitivity analysis
We repeated the analyses adjusting for serum creatinine within one year prior to nephrectomy in patients in whom this data was available.
Results
A total of 5670 patients met the inclusion criteria, of which 3167 (55.9%) underwent RN and 2503 (44.1%) underwent PN (Table 1). Other than income quintile, all other baseline characteristics were significantly different between the PN and RN groups. The analyses for the renal function outcomes included 2110 patients.
Table 1.
Cohort characteristics of 5670 patients undergoing nephrectomy for pT1a renal cell carcinoma between 1995 and 2015 in Ontario, Canada
| Partial nephrectomy (n=2503) | Radical nephrectomy (n=3167) | SD | |
|---|---|---|---|
| Gender, n (%) | 0.61 | ||
| Female | 959 (38.3) | 1354 (42.8) | |
| Male | 1544 (61.7) | 1813 (57.2) | |
| Age group, n (%) | 0.30 | ||
| 18–39 | 191 (7.6) | 148 (4.7) | |
| 40–44 | 153 (6.1) | 181 (5.7) | |
| 45–49 | 277 (11.1) | 249 (7.9) | |
| 50–54 | 348 (13.9) | 311 (9.8) | |
| 55–59 | 366 (14.6) | 428 (13.5) | |
| 60–64 | 353 (14.1) | 447 (14.1) | |
| 65–69 | 337 (13.5) | 507 (16.0) | |
| 70–74 | 248 (9.9) | 396 (12.5) | |
| 75–79 | 161 (6.4) | 322 (10.2) | |
| 80+ | 69 (2.8) | 178 (5.6) | |
| Income quintile, n (%) | 0.09 | ||
| 1 (lowest) | 442 (18) | 632 (20) | |
| 2 | 504 (20) | 679 (21) | |
| 3 | 500 (20) | 647 (20) | |
| 4 | 515 (21) | 613 (19) | |
| 5 (highest) | 542 (22) | 596 (19) | |
| Charlson score, median (interquartile range) | 2 (2–3) | 2 (2–3) | 0.15 |
| Year of surgery, n (%) | 0.78 | ||
| 1995–2000 | 144 (6) | 876 (28) | |
| 2001–2005 | 232 (9) | 520 (16) | |
| 2006–2010 | 790 (32) | 979 (31) | |
| 2011–2014 | 1337 (53) | 792 (25) | |
| Tumour size, median (interquartile range) | 2.5 (2.0–3.2) | 3.0 (2.5–3.6) | 0.61 |
| Histology, n (%) | 0.16 | ||
| Clear-cell | 1817 (73) | 2510 (79) | |
| Papillary | 527 (17) | 402 (13) | |
| Chromophobe | 157 (6) | 150 (5) | |
| Other | 102 (4) | 105 (3) |
SD: standardized difference.
Primary outcome
Median followup for OS was 77 months, during which there were 1187 deaths, 260 in the PN group and 927 in the RN group. Compared to RN, PN was associated with significantly improved OS (hazard ratio [HR] 0.73, 95% confidence interval [CI] 0.63–0.84) (Supplementary Table 1).
Secondary outcomes
The results for the secondary outcomes are summarized in Table 2 and detailed in Supplementary Table 1. Multivariable Cox proportional hazard models found that compared to RN, PN was associated with significantly reduced risk of CKD (HR 0.18, 95% CI 0.11–0.27) and significantly improved cancer-specific mortality (HR 0.45, 95% CI 0.30–0.65). The risk of MI was not significantly different between groups (HR 0.91, 95% CI 0.62–1.34). The number of patients (n=15) requiring renal replacement therapy was limited, precluding multivariable analysis; univariable analysis found that type of surgery was not associated with renal replacement therapy (HR 0.25, 95% CI 0.06–1.12). Serum creatinine within one year prior to surgery was available in 2411 (43%) patients. A multivariable sensitivity analysis further adjusting for preoperative serum creatinine demonstrated that OS remained significantly improved in the PN group. Due to the limited number of cancer-specific deaths (n=31) and patients diagnosed with CKD (n=46), these models included only type of surgery and serum creatinine as covariates; PN was associated with a reduced risk of CKD (HR 0.09, 95% CI 0.04–0.20) but not improved cancer-specific survival (HR 0.84, 95% CI 0.42–1.72). In this sensitivity analysis, only two patients were diagnosed with MI, both in the PN group, and two patients required renal replacement therapy, one in each group.
Table 2.
Long-term outcomes in 5670 patients undergoing partial vs. radical nephrectomy for pT1a renal cell carcinoma
| Outcome | Partial nephrectomy (n=2503) | Radical nephrectomy (n=3167) |
|---|---|---|
| Chronic kidney disease | ||
| # of events | 26 | 223 |
| Hazard ratio* (95% CI) | 0.18 (0.11–0.27) | Reference |
| Renal replacement therapy | ||
| # of events | 2 | 13 |
| Hazard ratio† (95% CI) | 0.25 (0.06–1.12) | Reference |
| Myocardial infarction | ||
| # of events | 42 | 131 |
| Hazard ratio* (95% CI) | 0.91 (0.62–1.34) | Reference |
| Death from kidney cancer | ||
| # of events | 36 | 184 |
| Hazard ratio* (95% CI) | 0.45 (0.30–0.65) | Reference |
Adjusted for gender, age group, income quintile, Charlson score, year of surgery, tumor size, and histology.
Univariate model due to low number of events.
CI: confidence interval.
Discussion
This population-based study spanning nearly 20 years found that compared to RN, PN for T1a kidney cancer was associated with significantly improved OS and reduced risk of CKD. The association with cancer-specific mortality was inconsistent; type of surgery was not associated with risk of MI, and the requirement for renal replacement therapy occurred infrequently in either group. These findings reaffirm the preferred use of PN for stage T1a kidney cancer, when feasible.
To date, EORTC 30904 is the only randomized trial that has compared RN vs. PN.10 This multicenter trial randomized 541 patients with tumors <5 cm suspicious for RCC to RN vs. PN. Median followup was 9.3 years for OS. In the intention-to-treat analysis, PN was associated with significantly worse OS but there was no significant difference in cancer-specific mortality (only 2% of patients died of cancer). In the subgroup analysis of patients with confirmed RCC histology, the association for OS was not statistically significant. This trial also found that cardiovascular deaths were less common in the RN group.10 RN was favorable in terms of lower perioperative morbidity,17 while PN provided better renal function outcomes.9 The results of this trial have been controversial and several criticisms, including premature study closure, crossover between groups, the design of a non-inferiority trial with but the OS benefit with RN being based on a test of superiority, among others, have made it difficult to interpret the results.
Despite the only level 1 evidence on this topic supporting the use of RN over PN for small renal masses, several guidelines recommend the preferential use of PN, when feasible.5–7 This is based on the biological rationale and several observational studies demonstrating the benefit of PN over RN. A systematic review and meta-analysis published in 2012 included 36 studies evaluating 31 729 RN and 9281 PN patients undergoing surgery for localized kidney cancer.18 This study found that PN was associated with significantly improved OS, improved cancer-specific mortality, and reduced risk of CKD. A more recent meta-analysis also found a reduced risk of CKD associated with PN, but no difference in cardiovascular outcomes between PN and RN.19
The reduced risk of CKD related to PN is biologically plausible given the nephron-sparing concept behind PN. In our study, the reduced risk of CKD in patients undergoing PN may have contributed to the observed benefit in OS. Indeed, the landmark study by Go et al found that increasing glomerular filtration rate was inversely associated with risk of death.20 Their study also found that increasing glomerular filtration rate was inversely associated with cardiovascular outcomes;20 despite the associated observed benefit in risk of CKD provided by PN in our study, there was no association with MI, consistent with the previously described metaanalysis.19 This can be explained by the proposed theory that the biology of CKD associated with surgery may be different than that of CKD associated with medical conditions,21 as those medical conditions, such as diabetes and hypertension, continue to contribute to the risk of cardiovascular outcomes. It is worth noting that the requirement for renal replacement therapy occurred infrequently. The results of our large, population-based study support a previous analysis of 514 patients from EORTC 30904, which found that only four patients in each group developed end-stage renal disease. Put together, these results suggest that although RN likely increases the risk of CKD compared to PN, the risk of end-stage renal disease requiring renal replacement therapy after RN may be low. Therefore, in select cases, such as those with complex tumors, the risk of PN may outweigh the benefit, and RN remains an option in these select cases for the management of stage T1a kidney cancer.
In the overall cohort, cancer-specific mortality was significantly improved in the PN arm, consistent with the findings of a meta-analysis.18 This likely represents selection bias, as there is no biological rationale supporting PN as a more oncological effective procedure. We attempted to reduce selection bias by restricting our cohort to patients with pathological tumor size <4 cm and further adjusting for histology and tumor size; however, the possibility of residual confounding remains. In our sensitivity analysis restricting to patients with preoperative serum creatinine, there was no association between type of surgery and cancer-specific mortality, though this should be interpreted with caution given the limited number of events.
While several observational studies have compared survival outcomes following RN or PN, our study has several strengths. Our population-based study design in a universal healthcare setting improves generalizability. Furthermore, we used regularly updated administrative databases, several of which have been validated, allowing us to accurately capture various outcomes over a prolonged period. We also had detailed information on various baseline characteristics, allowing us to restrict our study to a homogenous population of patients undergoing a single nephrectomy for RCC tumors <4 cm, and to further adjust for differences in characteristics between patients undergoing RN or PN. Finally, we did a sensitivity analysis adjusting for preoperative creatinine, albeit this information was available in a reduced cohort.
This study is not without limitations. Although we attempted to reduce confounding by adjusting for known prognostic factors, the possibility of residual confounding remains. Additionally, our cohort may have included patients with solitary kidney prior to nephrectomy, though this is expected to be infrequent. We were unable to evaluate additional cancer-related outcomes, such recurrence or metastasis, as these are difficult to define using administrative databases. Finally, there may be subsets of patients for whom RN is preferred over PN; this was not an objective of this study but is an area of future research given the potential increased risk of perioperative morbidity associated with PN.17,22,23
Conclusions
Our population-based study of patients undergoing nephrectomy for T1a kidney cancer found that compared to RN, PN was associated with significantly improved OS and reduced the risk of CKD. However, few patients in either group required renal replacement therapy. These findings reaffirm the preferred use of PN for these patients.
Supplementary Information
Supplementary Table 1.
Multivariable Cox proportional hazard models evaluating the association between partial or radical nephrectomy and outcomes in patients with pT1a renal cell carcinoma
| Outcome (hazard ratio [95% confidence interval]) | ||||
|---|---|---|---|---|
|
| ||||
| Predictor | OS | CKD | MI | CSS |
| Partial nephrectomy | 0.73 (0.63–0.84) | 0.18 (0.11– 0.27) | 0.92 (0.62–1.35) | 0.45 (0.30–0.65) |
| Female gender | 0.81 (0.72–0.92) | 0.72 (0.55–0.94) | 0.58 (0.42–0.80) | 0.65 (0.48–0.85) |
| Age | ||||
| 18–39 | Reference | Reference | Reference | Reference |
| 40–44 | 1.8 (0.75–4.6) | 0.79 (0.36–1.74) | 0.95 (0.24–3.80) | 1.12 (0.25–5.03) |
| 45–49 | 2.7 (1.2 –6.2) | 1.35 (0.70–2.62) | 1.46 (0.45–4.75) | 2.45 (0.69–8.71) |
| 50–54 | 4.8 (2.2–10.4) | 2.17 (1.16–4.05) | 1.43 (0.45–4.57) | 3.13 (0.92–10.6) |
| 55–59 | 7.8 (3.6–16.8) | 2.29 (1.24–4.25) | 1.88 (0.63–5.65) | 3.19 (0.96–10.6 |
| 60–64 | 10.7 (5.0–22.9) | 2.84 (1.54–5.24) | 2.21 (0.75–6.52) | 4.88 (1.50–15.9) |
| 65–69 | 13.8 (6.5–29.4) | 2.68 (1.44–5.00) | 3.77 (1.34–10.6) | 3.70 (1.13–12.1) |
| 70–74 | 20.4 (9.6–43.4) | 2.55 (1.30–5.00) | 4.28 (1.51–12.2) | 4.42 (1.35–24.5) |
| 75–79 | 33.8 (15.9–71.8) | 3.32 (1.64–6.73) | 5.27 (1.85–15.0) | 6.06 (1.85–19.9) |
| 80+ | 53.1 (24.7–114.1) | 3.73 (1.59–8.74) | 3.27 (1.00–10.68) | 8.82 (2.59–30.1) |
| Income quintile | ||||
| 1 (lowest) | Reference | Reference | Reference | Reference |
| 2 | 0.84 (0.71–1.00) | 1.16 (0.78–1.71) | 0.72 (0.46–1.13) | 1.27 (0.83–1.96) |
| 3 | 0.83 (0.69–0.98) | 1.03 (0.68–1.55) | 0.79 (0.50–1.23) | 1.42 (0.93–2.15) |
| 4 | 0.82 (0.68–0.98) | 1.06 (0.70–1.62) | 0.70 (0.43–1.11) | 1.31 (0.84–2.04) |
| 5 (highest) | 0.75 (0.62–0.89) | 0.88 (0.58–1.33) | 0.59 (0.37–0.96) | 1.10 (0.70–1.74) |
| Charlson score | 1.34 (1.30–1.37) | 0.84 (0.72–0.98) | 1.05 (0.96–1.16) | 1.37 (1.30–1.45) |
| Year of surgery | ||||
| 1995–2000 | Reference | Reference | Reference | Reference |
| 2001–2005 | 0.95 (0.81–1.11) | 1.04 (0.72–1.51) | 0.86 (0.58 –1.27) | 1.07 (0.75 –1.54) |
| 2006–2010 | 0.71 (0.60–0.83) | 1.66 (1.18–2.34) | 0.55 (0.36–0.85) | 0.68 (0.46–1.00) |
| 2011–2014 | 0.55 (0.44–0.69) | 2.14 (1.35–3.42) | 0.59 (0.32–1.09) | 6.42 (3.88–10.61) |
| Tumor size | 1.09 (1.01–1.17) | 1.05 (0.89–1.24) | 1.05 (0.86–1.28) | 1.18 (0.99–1.41) |
| Histology | ||||
| Clear-cell | Reference | Reference | Reference | Reference |
| Papillary | 0.96 (0.81–1.14) | 1.16 (0.82–1.65) | 0.85 (0.55–1.31) | 0.96 (0.65–1.41) |
| Chromophobe | 0.80 (0.57–1.12) | 0.89 (0.47–1.70) | 1.34 (0.65–2.77) | 0.38 (0.12–1.19) |
| Other | 1.22 (0.95–1.58) | 1.02 (0.52–1.99) | 0.32 (0.10–1.00) | 1.95 (1.16–3.27) |
CKD: chronic kidney disease; CSS: cancer-specific survival; MI: myocardial infarction; OS: overall survival.
Footnotes
An abstract of this study was presented as a podium presentation at the 2019 American Urological Association Meeting in Chicago, IL, U.S., and as a moderated poster at the 2019 Canadian Urological Association Meeting in Quebec City, QC, Canada
Competing interests: Dr. Finelli has been an advisory board member for AbbVie, Astellas, Bayer, Ipsen, Janssen, Sanofi, and TerSera; and has participated in clinical trials supported by Astellas, Bayer, and Janssen. The remaining authors report no competing personal or financial interests related to this work.
This paper has been peer-reviewed.
References
- 1.Znaor A, Lortet-Tieulent J, Laversanne M, et al. International variations and trends in renal cell carcinoma incidence and mortality. Eur Urol. 2015;67:519–30. doi: 10.1016/j.eururo.2014.10.002. [DOI] [PubMed] [Google Scholar]
- 2.Chow W-H, Dong LM, Devesa SS. Epidemiology and risk factors for kidney cancer. Nat Rev Urol. 2010;7:245. doi: 10.1038/nrurol.2010.46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Lightfoot N, Conlon M, Kreiger N, et al. Impact of non-invasive imaging on increased incidental detection of renal cell carcinoma. Eur Urol. 2000;37:521–7. doi: 10.1159/000020188. [DOI] [PubMed] [Google Scholar]
- 4.Kane CJ, Mallin K, Ritchey J, et al. Renal cell cancer stage migration: Analysis of the national cancer data base. Cancer. 2008;113:78–83. doi: 10.1002/cncr.23518. [DOI] [PubMed] [Google Scholar]
- 5.Ljungberg B, Albiges L, Abu-Ghanem Y, et al. European association of urology guidelines on renal cell carcinoma: The 2019 update. Eur Urol. 2019;75:799–810. doi: 10.1016/j.eururo.2019.02.011. [DOI] [PubMed] [Google Scholar]
- 6.Motzer RJ, Jonasch E, Agarwal N, et al. Kidney cancer, version 2.2017, NCCN clinical practice guidelines in oncology. J Natl Compr Canc Netw. 2017;15:804–34. doi: 10.6004/jnccn.2017.0100. [DOI] [PubMed] [Google Scholar]
- 7.Campbell S, Uzzo RG, Allaf ME, et al. Renal mass and localized renal cancer: AUA guideline. J Urol. 2017;198:520–9. doi: 10.1016/j.juro.2017.04.100. [DOI] [PubMed] [Google Scholar]
- 8.Pierorazio PM, Johnson MH, Patel HD, et al. Management of renal masses and localized renal cancer: Systematic review and meta-analysis. J Urol. 2016;196:989–99. doi: 10.1016/j.juro.2016.04.081. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Scosyrev E, Messing EM, Sylvester R, et al. Renal function after nephron-sparing surgery vs. radical nephrectomy: Results from EORTC randomized trial 30904. Eur Urol. 2014;65:372–7. doi: 10.1016/j.eururo.2013.06.044. [DOI] [PubMed] [Google Scholar]
- 10.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;59:543–52. doi: 10.1016/j.eururo.2010.12.013. [DOI] [PubMed] [Google Scholar]
- 11.Nayan M, Macdonald EM, Juurlink DN, et al. Medication use and survival in diabetic patients with kidney cancer: A population-based cohort study. Pharmacol Res. 2016;113:468–74. doi: 10.1016/j.phrs.2016.09.027. [DOI] [PubMed] [Google Scholar]
- 12.Welch HG, Black WC. Are deaths within 1 month of cancer-directed surgery attributed to cancer? J Natl Canc Inst. 2002;94:1066–70. doi: 10.1093/jnci/94.14.1066. [DOI] [PubMed] [Google Scholar]
- 13.Austin PC. Using the standardized difference to compare the prevalence of a binary variable between two groups in observational research. Comm Stats Simul Comput. 2009;38:1228–34. doi: 10.1080/03610910902859574. [DOI] [Google Scholar]
- 14.Schoenfeld D. Partial residuals for the proportional hazards regression model. Biometrika. 1982;69:239–41. doi: 10.1093/biomet/69.1.239. [DOI] [Google Scholar]
- 15.Lau B, Cole SR, Gange SJ. Competing risk regression models for epidemiologic data. Am J Epidemiol. 2009;170:244–56. doi: 10.1093/aje/kwp107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Hair JF, Anderson RE, Tatham RL, et al. Multivariate data analysis new york. NY: Macmillan; 1995. [Google Scholar]
- 17.Van Poppel H, Da Pozzo L, Albrecht W, et al. A prospective randomized EORTC intergroup phase 3 study comparing the complications of elective nephron-sparing surgery and radical nephrectomy for low-stage renal cell carcinoma. Eur Urol. 2007;51:1606–15. doi: 10.1016/j.eururo.2006.11.013. [DOI] [PubMed] [Google Scholar]
- 18.Kim SP, Thompson RH, Boorjian SA, et al. 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;188:51–7. doi: 10.1016/j.juro.2012.03.006. [DOI] [PubMed] [Google Scholar]
- 19.Wang Z, Wang G, Xia Q, et al. Partial nephrectomy vs. Radical nephrectomy for renal tumors: A meta-analysis of renal function and cardiovascular outcomes. Urol Oncol. 2016;34:533.e11–9. doi: 10.1016/j.urolonc.2016.07.007. [DOI] [PubMed] [Google Scholar]
- 20.Go AS, Chertow GM, Fan D, et al. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Eng J of Med. 2004;351:1296–305. doi: 10.1056/NEJMoa041031. [DOI] [PubMed] [Google Scholar]
- 21.Lane BR, Demirjian S, Derweesh IH, et al. 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;68:996–1003. doi: 10.1016/j.eururo.2015.04.043. [DOI] [PubMed] [Google Scholar]
- 22.Hadjipavlou M, Khan F, Fowler S, et al. Partial vs radical nephrectomy for t1 renal tumors: An analysis from the British Association of Urological Surgeons nephrectomy audit. BJU Int. 2016;117:62–71. doi: 10.1111/bju.13114. [DOI] [PubMed] [Google Scholar]
- 23.Venkatramani V, Koru-Sengul T, Miao F, et al. A comparison of overall survival and perioperative outcomes between partial and radical nephrectomy for cT1b and cT2 renal cell carcinoma—analysis of a national cancer registry. Urol Oncol. 2018;36:90.e9–e14. doi: 10.1016/j.urolonc.2017.11.008. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Table 1.
Multivariable Cox proportional hazard models evaluating the association between partial or radical nephrectomy and outcomes in patients with pT1a renal cell carcinoma
| Outcome (hazard ratio [95% confidence interval]) | ||||
|---|---|---|---|---|
|
| ||||
| Predictor | OS | CKD | MI | CSS |
| Partial nephrectomy | 0.73 (0.63–0.84) | 0.18 (0.11– 0.27) | 0.92 (0.62–1.35) | 0.45 (0.30–0.65) |
| Female gender | 0.81 (0.72–0.92) | 0.72 (0.55–0.94) | 0.58 (0.42–0.80) | 0.65 (0.48–0.85) |
| Age | ||||
| 18–39 | Reference | Reference | Reference | Reference |
| 40–44 | 1.8 (0.75–4.6) | 0.79 (0.36–1.74) | 0.95 (0.24–3.80) | 1.12 (0.25–5.03) |
| 45–49 | 2.7 (1.2 –6.2) | 1.35 (0.70–2.62) | 1.46 (0.45–4.75) | 2.45 (0.69–8.71) |
| 50–54 | 4.8 (2.2–10.4) | 2.17 (1.16–4.05) | 1.43 (0.45–4.57) | 3.13 (0.92–10.6) |
| 55–59 | 7.8 (3.6–16.8) | 2.29 (1.24–4.25) | 1.88 (0.63–5.65) | 3.19 (0.96–10.6 |
| 60–64 | 10.7 (5.0–22.9) | 2.84 (1.54–5.24) | 2.21 (0.75–6.52) | 4.88 (1.50–15.9) |
| 65–69 | 13.8 (6.5–29.4) | 2.68 (1.44–5.00) | 3.77 (1.34–10.6) | 3.70 (1.13–12.1) |
| 70–74 | 20.4 (9.6–43.4) | 2.55 (1.30–5.00) | 4.28 (1.51–12.2) | 4.42 (1.35–24.5) |
| 75–79 | 33.8 (15.9–71.8) | 3.32 (1.64–6.73) | 5.27 (1.85–15.0) | 6.06 (1.85–19.9) |
| 80+ | 53.1 (24.7–114.1) | 3.73 (1.59–8.74) | 3.27 (1.00–10.68) | 8.82 (2.59–30.1) |
| Income quintile | ||||
| 1 (lowest) | Reference | Reference | Reference | Reference |
| 2 | 0.84 (0.71–1.00) | 1.16 (0.78–1.71) | 0.72 (0.46–1.13) | 1.27 (0.83–1.96) |
| 3 | 0.83 (0.69–0.98) | 1.03 (0.68–1.55) | 0.79 (0.50–1.23) | 1.42 (0.93–2.15) |
| 4 | 0.82 (0.68–0.98) | 1.06 (0.70–1.62) | 0.70 (0.43–1.11) | 1.31 (0.84–2.04) |
| 5 (highest) | 0.75 (0.62–0.89) | 0.88 (0.58–1.33) | 0.59 (0.37–0.96) | 1.10 (0.70–1.74) |
| Charlson score | 1.34 (1.30–1.37) | 0.84 (0.72–0.98) | 1.05 (0.96–1.16) | 1.37 (1.30–1.45) |
| Year of surgery | ||||
| 1995–2000 | Reference | Reference | Reference | Reference |
| 2001–2005 | 0.95 (0.81–1.11) | 1.04 (0.72–1.51) | 0.86 (0.58 –1.27) | 1.07 (0.75 –1.54) |
| 2006–2010 | 0.71 (0.60–0.83) | 1.66 (1.18–2.34) | 0.55 (0.36–0.85) | 0.68 (0.46–1.00) |
| 2011–2014 | 0.55 (0.44–0.69) | 2.14 (1.35–3.42) | 0.59 (0.32–1.09) | 6.42 (3.88–10.61) |
| Tumor size | 1.09 (1.01–1.17) | 1.05 (0.89–1.24) | 1.05 (0.86–1.28) | 1.18 (0.99–1.41) |
| Histology | ||||
| Clear-cell | Reference | Reference | Reference | Reference |
| Papillary | 0.96 (0.81–1.14) | 1.16 (0.82–1.65) | 0.85 (0.55–1.31) | 0.96 (0.65–1.41) |
| Chromophobe | 0.80 (0.57–1.12) | 0.89 (0.47–1.70) | 1.34 (0.65–2.77) | 0.38 (0.12–1.19) |
| Other | 1.22 (0.95–1.58) | 1.02 (0.52–1.99) | 0.32 (0.10–1.00) | 1.95 (1.16–3.27) |
CKD: chronic kidney disease; CSS: cancer-specific survival; MI: myocardial infarction; OS: overall survival.