bs_bs_banner
International Journal of Urology (2014)
doi: 10.1111/iju.12446
Original Article
Cytoreductive nephrectomy for metastatic renal cell carcinoma: A population-based analysis of perioperative outcomes according to clinical stage Toshio Takagi,1 Toru Sugihara,2,3 Hideo Yasunaga,4 Hiromasa Horiguchi,5 Kiyohide Fushimi,6 Tsunenori Kondo,1 Yukio Homma3 and Kazunari Tanabe1 1
Department of Urology, Tokyo Women’s Medical University, Tokyo, Japan; 2Department of Quantitative Health Sciences, Cleveland Clinic Foundation, Cleveland, Ohio, USA; 3Department of Urology, 4Department of Health Economics and Epidemiology Research, School of Public Health, The University of Tokyo, 5Department of Clinical Data Management and Research, Clinical Research Center, National Hospital Organization Headquarters, and 6Department of Health Care Informatics, Tokyo Medical and Dental University, Tokyo, Japan
Abbreviations & Acronyms CCI = Charlson Comorbidity Index CN = cytoreductive nephrectomy DIC = disseminated intravascular coagulation DPC = Diagnosis Procedure Combination ICD-10 = International Classification of Diseases and Related Health Problems, Tenth Revision IQR = interquartile range MIS = minimally-invasive surgery mRCC = metastatic renal cell carcinoma PLOS = postoperative length of stay Correspondence: Toshio Takagi M.D., Department of Urology, Tokyo Women’s Medical University, 8-1 Kawada-cho, Shinjuku-ku, Tokyo 162-8666, Japan. Email: [email protected] Received 1 November 2013; accepted 18 February 2014.
© 2014 The Japanese Urological Association
Objectives: To investigate the perioperative outcomes of cytoreductive nephrectomy according to clinical T stage, and to analyze factors affecting these outcomes. Methods: The Japanese Diagnosis Procedure Combination database from 2007 to 2012 was used to evaluate perioperative outcomes including in-hospital mortality, complications, blood transfusion, anesthesia time, postoperative length of stay and total cost in patients who underwent cytoreductive nephrectomy for metastatic renal cell carcinoma, according to clinical T stage. Multivariable regression analyses including sex, age, clinical N stage, hospital volume, type of hospital, Charlson Comorbidity Index and clinical T stage were carried out to identify outcome predictors. Results: The present study enrolled 1074 patients including 270 with T1, 215 with T2, 479 with T3 and 110 with T4. Age, sex and Charlson Comorbidity Index were not significantly different among the four stages. A low clinical T stage was associated with minimally-invasive surgery (P < 0.001). The blood transfusion rate, anesthesia time, postoperative length of stay and total cost increased significantly with increasing clinical T stage (all P < 0.001). Multivariable regression analyses showed that increasing clinical T stage was significantly associated with unfavorable perioperative outcomes except in-hospital mortality (T4/T1: postoperative complications OR 2.34; blood transfusion OR 5.27; anesthesia time +14%; postoperative length of stay +13.2%; total cost +13.4%; all P < 0.05). Clinical N stage was the only significant predictive factor for in-hospital mortality (N1/N0: OR 3.34, P = 0.004; N2/N0: OR 3.48, P = 0.008). Conclusions: Clinical T stage is significantly associated with perioperative outcomes, other than in-hospital mortality, in patients with metastatic renal call carcinoma undergoing cytoreductive nephrectomy. Clinical N stage is significantly associated with in-hospital mortality.
Key words: kidney neoplasm, metastasis, nephrectomy, predictor.
Introduction The incidence of renal cell carcinoma is increasing in the USA.1–3 According to the National Cancer Institute’s Surveillance, Epidemiology and End Results registry, the overall age-adjusted incidence of kidney cancer in the USA increased from 7.1 to 10.8 cases per 100 000 from 1983 to 2002.4 Although many tumors are detected at an early stage, often as an incidental finding during imaging examinations, 20% of patients with renal cell carcinoma have distant metastasis at the time of diagnosis.2 The results of randomized studies during the era of interferon therapy shown that CN was beneficial in patients with mRCC.5,6 After targeted therapy for mRCC became standard, there was a paradigm shift in the management strategies for mRCC. Based on the results of a Surveillance, Epidemiology and End Results analysis, Tsao et al. reported that treatment with CN has decreased in the USA during the era of targeted therapy, from 50% in 2005 to 38% in 2008.7 Although ongoing prospective trials including the Clinical Trial to Assess the Importance of Nephrectomy will provide information regarding the true benefit of CN in the era of targeted therapy, these prospective randomized trials exclude patients with high surgical 1
T TAKAGI ET AL.
risk, such as those with poor performance status. Large retrospective studies can sometimes provide important information, because they include such patients. Some studies have retrospectively reviewed mortality and morbidity after CN.8,9 Trinh et al. analyzed data from the Nationwide Inpatient Sample from 1998 to 2007, and found that older age, more preoperative comorbidities and treatment at a hospital with a smaller number of beds were negative prognostic factors for in-hospital mortality.9 Abdollah et al. analyzed data from the Florida Inpatients Databases from 1998 to 2008, and reported that more advanced age, more comorbidities and the cumulative number of secondary surgical procedures were independent predictors of increased in-hospital mortality, complications, and transfusions. However, these population-based analyses did not discuss associations between clinical stage and perioperative outcomes because of database limitations, and they were not specifically from the era of targeted therapy. Our study used the Japanese national database from 2007 to 2012 to examine perioperative outcomes (in-hospital mortality, postoperative complications, blood transfusion rate, anesthesia time, length of postoperative stay and total cost), and compared these outcomes according to clinical T stage. Predictive factors for perioperative outcomes were determined using multivariable analysis.
Complications were defined as follows: in-hospital mortality; sepsis (ICD10 codes: A32.7, A40.x, A41.x, A42.7, B37.7 and T81.4); disseminated intravascular coagulation (D65); pulmonary embolism (I26.x); cardiac events (ischemic heart disease [I20.x–24.x], heart failure [I11.0, I50.x]); other vascular complications (I71.x, I73.9, I74.x, I77.x, I80–I83.x, I89.x, K55.0, K55.9, T79.0, T79.1 and T81.7); stroke (I60.x–64); pneumonia or influenza (J10.x–18.x); other respiratory complications (J46, J69.0, J70.x, J80–J86.x, J93.x, J94.2, J95.x and J96.x); peritonitis or peritoneal abscess (K65.x, N73.3 and N73.5); ileus (K56.x and K91.3); acute renal failure (N17x, N28.0 and N99.0); genitourinary infection (N10, N30.x, N41.x, N45.x, N70.x–N73.x and N76); other genitourinary complications (N13.x, N32.x, N35.x, 36.x, N82.x and N99.x); disruption of operation wound (T81.3); and intraoperative complications (S34.x–S39.x, T81.2). Hospital volume was categorized into three groups according to the annual nephrectomy caseload of each institution (low 30), so that the number of patients in each group was almost equivalent. Other variables included sex, age, comorbidities at admission, oncological T and N categories, use of MIS, and hospital academic status (academic or non-academic). Comorbidities were converted to a CCI score based on Quan’s protocol.12
Methods
The threshold of significance was set at P < 0.05. Univariate comparisons were made using the χ2-test or the Kruskal–Wallis test, as appropriate. For each end-point, multivariable logistic and linear regression analysis was carried out including the aforementioned variables. To adjust for the institutional clustering effect, generalized estimating equations were applied.12 In the multivariable analyses, anesthesia time, PLOS and total cost were logarithmically transformed because of their skewed distribution. All statistical analyses were carried out using IBM SPSS version 20.0 (IBM SPSS, Armonk, NY, USA).
Data source The data for the present study were obtained from the DPC database, which is an inpatient administrative claims database in Japan. The database is highly representative of the overall population, and includes approximately 45% of all acute inpatient treatments in Japan.10 The data-collection term is discontinuous: July to December in 2007 to 2010, January to December in 2011 and January to March in 2012. The database includes main diagnoses, pre-existing comorbidities and complications that occurred after admission, coded according to the ICD-10 codes, as well as procedures coded according to the Japanese original procedure codes.11 Oncological status according to the TNM classification is also available. Given the anonymous nature of the data collection process, informed consent from patients for inclusion in the study was not required. Approval of the study protocol was obtained from the institutional review board and the ethics committee of The University of Tokyo.
Sampling and end-points The DPC database was searched for patients with a main diagnosis at admission of renal malignancy (ICD-10 code: C64), with an oncological M status of M1, who underwent nephrectomy (Japanese codes for open nephrectomy: K773 and K773-2, and for laparoscopic nephrectomy: K773-1) from 2007 to 2012. Laparoscopic surgery was defined as MIS. The end-points evaluated were in-hospital mortality, perioperative complications, blood transfusion rate, anesthesia time, PLOS and total cost. The exchange rate was set at JP¥100 to US$1. 2
Statistical analysis
Results The clinical T stage was T1 in 270 patients, T2 in 215 patients, T3 in 479 patients and T4 in 110 patients. Table 1 shows the patient and tumor characteristics according to clinical T stage. Age, sex, CCI and type of hospital were not significantly different among the four stages. A higher clinical T stage was significantly associated with a higher clinical N stage (P < 0.001). Patients with a low clinical T stage were more likely to undergo MIS (T1: 44.1%, T4: 9.1%, P < 0.001). Table 2 shows comparisons of perioperative outcomes among the four stages. The blood transfusion rate (P < 0.001), anesthesia time (P < 0.001), PLOS (P < 0.001) and total cost (P < 0.001) increased significantly with increasing clinical T stage, but in-hospital death did not (P = 0.103). Postoperative complications tended to increase with increasing clinical T stage (7.8% in T1, 11.6% in T2, 12.5% in T3, and 16.4% in T4), but this increase was not statistically significant (P = 0.082). Tables 3 and 4 show the results of multivariable regression analyses for postoperative outcomes. A higher clinical T stage was significantly associated with a greater postoperative complication rate (T4/T1: OR 2.34, P = 0.024), blood transfusion © 2014 The Japanese Urological Association
CN for metastatic kidney cancer
Table 1
Comparisons of 1074 patients who underwent cytoreductive nephrectomy, according to clinical T stage
Variable
T1
T2
T3
T4
P-value
270
215
479
110
123 (45.6) 94 (34.8) 53 (19.6)
100 (46.5) 61 (28.4) 54 (25.1)
215 (44.9) 177 (37.0) 87 (18.2)
57 (51.8) 37 (33.6) 16 (14.5)
0.149
198 (73.3) 72 (26.7)
161 (74.9) 54 (25.1)
365 (76.2) 114 (23.8)
83 (75.5) 27 (24.5)
0.856
165 (61.1) 66 (24.4) 39 (14.4)
139 (64.7) 55 (25.6) 21 (9.8)
284 (59.3) 133 (27.8) 62 (12.9)
76 (69.1) 21 (19.1) 13 (11.8)
0.342
230 (85.2) 26 (9.6) 14 (5.2)
165 (76.7) 32 (14.9) 18 (8.4)
311 (64.9) 105 (21.9) 63 (13.2)
45 (40.9) 39 (35.5) 26 (23.6)
International Journal of Urology (2014)
doi: 10.1111/iju.12446
Original Article
Cytoreductive nephrectomy for metastatic renal cell carcinoma: A population-based analysis of perioperative outcomes according to clinical stage Toshio Takagi,1 Toru Sugihara,2,3 Hideo Yasunaga,4 Hiromasa Horiguchi,5 Kiyohide Fushimi,6 Tsunenori Kondo,1 Yukio Homma3 and Kazunari Tanabe1 1
Department of Urology, Tokyo Women’s Medical University, Tokyo, Japan; 2Department of Quantitative Health Sciences, Cleveland Clinic Foundation, Cleveland, Ohio, USA; 3Department of Urology, 4Department of Health Economics and Epidemiology Research, School of Public Health, The University of Tokyo, 5Department of Clinical Data Management and Research, Clinical Research Center, National Hospital Organization Headquarters, and 6Department of Health Care Informatics, Tokyo Medical and Dental University, Tokyo, Japan
Abbreviations & Acronyms CCI = Charlson Comorbidity Index CN = cytoreductive nephrectomy DIC = disseminated intravascular coagulation DPC = Diagnosis Procedure Combination ICD-10 = International Classification of Diseases and Related Health Problems, Tenth Revision IQR = interquartile range MIS = minimally-invasive surgery mRCC = metastatic renal cell carcinoma PLOS = postoperative length of stay Correspondence: Toshio Takagi M.D., Department of Urology, Tokyo Women’s Medical University, 8-1 Kawada-cho, Shinjuku-ku, Tokyo 162-8666, Japan. Email: [email protected] Received 1 November 2013; accepted 18 February 2014.
© 2014 The Japanese Urological Association
Objectives: To investigate the perioperative outcomes of cytoreductive nephrectomy according to clinical T stage, and to analyze factors affecting these outcomes. Methods: The Japanese Diagnosis Procedure Combination database from 2007 to 2012 was used to evaluate perioperative outcomes including in-hospital mortality, complications, blood transfusion, anesthesia time, postoperative length of stay and total cost in patients who underwent cytoreductive nephrectomy for metastatic renal cell carcinoma, according to clinical T stage. Multivariable regression analyses including sex, age, clinical N stage, hospital volume, type of hospital, Charlson Comorbidity Index and clinical T stage were carried out to identify outcome predictors. Results: The present study enrolled 1074 patients including 270 with T1, 215 with T2, 479 with T3 and 110 with T4. Age, sex and Charlson Comorbidity Index were not significantly different among the four stages. A low clinical T stage was associated with minimally-invasive surgery (P < 0.001). The blood transfusion rate, anesthesia time, postoperative length of stay and total cost increased significantly with increasing clinical T stage (all P < 0.001). Multivariable regression analyses showed that increasing clinical T stage was significantly associated with unfavorable perioperative outcomes except in-hospital mortality (T4/T1: postoperative complications OR 2.34; blood transfusion OR 5.27; anesthesia time +14%; postoperative length of stay +13.2%; total cost +13.4%; all P < 0.05). Clinical N stage was the only significant predictive factor for in-hospital mortality (N1/N0: OR 3.34, P = 0.004; N2/N0: OR 3.48, P = 0.008). Conclusions: Clinical T stage is significantly associated with perioperative outcomes, other than in-hospital mortality, in patients with metastatic renal call carcinoma undergoing cytoreductive nephrectomy. Clinical N stage is significantly associated with in-hospital mortality.
Key words: kidney neoplasm, metastasis, nephrectomy, predictor.
Introduction The incidence of renal cell carcinoma is increasing in the USA.1–3 According to the National Cancer Institute’s Surveillance, Epidemiology and End Results registry, the overall age-adjusted incidence of kidney cancer in the USA increased from 7.1 to 10.8 cases per 100 000 from 1983 to 2002.4 Although many tumors are detected at an early stage, often as an incidental finding during imaging examinations, 20% of patients with renal cell carcinoma have distant metastasis at the time of diagnosis.2 The results of randomized studies during the era of interferon therapy shown that CN was beneficial in patients with mRCC.5,6 After targeted therapy for mRCC became standard, there was a paradigm shift in the management strategies for mRCC. Based on the results of a Surveillance, Epidemiology and End Results analysis, Tsao et al. reported that treatment with CN has decreased in the USA during the era of targeted therapy, from 50% in 2005 to 38% in 2008.7 Although ongoing prospective trials including the Clinical Trial to Assess the Importance of Nephrectomy will provide information regarding the true benefit of CN in the era of targeted therapy, these prospective randomized trials exclude patients with high surgical 1
T TAKAGI ET AL.
risk, such as those with poor performance status. Large retrospective studies can sometimes provide important information, because they include such patients. Some studies have retrospectively reviewed mortality and morbidity after CN.8,9 Trinh et al. analyzed data from the Nationwide Inpatient Sample from 1998 to 2007, and found that older age, more preoperative comorbidities and treatment at a hospital with a smaller number of beds were negative prognostic factors for in-hospital mortality.9 Abdollah et al. analyzed data from the Florida Inpatients Databases from 1998 to 2008, and reported that more advanced age, more comorbidities and the cumulative number of secondary surgical procedures were independent predictors of increased in-hospital mortality, complications, and transfusions. However, these population-based analyses did not discuss associations between clinical stage and perioperative outcomes because of database limitations, and they were not specifically from the era of targeted therapy. Our study used the Japanese national database from 2007 to 2012 to examine perioperative outcomes (in-hospital mortality, postoperative complications, blood transfusion rate, anesthesia time, length of postoperative stay and total cost), and compared these outcomes according to clinical T stage. Predictive factors for perioperative outcomes were determined using multivariable analysis.
Complications were defined as follows: in-hospital mortality; sepsis (ICD10 codes: A32.7, A40.x, A41.x, A42.7, B37.7 and T81.4); disseminated intravascular coagulation (D65); pulmonary embolism (I26.x); cardiac events (ischemic heart disease [I20.x–24.x], heart failure [I11.0, I50.x]); other vascular complications (I71.x, I73.9, I74.x, I77.x, I80–I83.x, I89.x, K55.0, K55.9, T79.0, T79.1 and T81.7); stroke (I60.x–64); pneumonia or influenza (J10.x–18.x); other respiratory complications (J46, J69.0, J70.x, J80–J86.x, J93.x, J94.2, J95.x and J96.x); peritonitis or peritoneal abscess (K65.x, N73.3 and N73.5); ileus (K56.x and K91.3); acute renal failure (N17x, N28.0 and N99.0); genitourinary infection (N10, N30.x, N41.x, N45.x, N70.x–N73.x and N76); other genitourinary complications (N13.x, N32.x, N35.x, 36.x, N82.x and N99.x); disruption of operation wound (T81.3); and intraoperative complications (S34.x–S39.x, T81.2). Hospital volume was categorized into three groups according to the annual nephrectomy caseload of each institution (low 30), so that the number of patients in each group was almost equivalent. Other variables included sex, age, comorbidities at admission, oncological T and N categories, use of MIS, and hospital academic status (academic or non-academic). Comorbidities were converted to a CCI score based on Quan’s protocol.12
Methods
The threshold of significance was set at P < 0.05. Univariate comparisons were made using the χ2-test or the Kruskal–Wallis test, as appropriate. For each end-point, multivariable logistic and linear regression analysis was carried out including the aforementioned variables. To adjust for the institutional clustering effect, generalized estimating equations were applied.12 In the multivariable analyses, anesthesia time, PLOS and total cost were logarithmically transformed because of their skewed distribution. All statistical analyses were carried out using IBM SPSS version 20.0 (IBM SPSS, Armonk, NY, USA).
Data source The data for the present study were obtained from the DPC database, which is an inpatient administrative claims database in Japan. The database is highly representative of the overall population, and includes approximately 45% of all acute inpatient treatments in Japan.10 The data-collection term is discontinuous: July to December in 2007 to 2010, January to December in 2011 and January to March in 2012. The database includes main diagnoses, pre-existing comorbidities and complications that occurred after admission, coded according to the ICD-10 codes, as well as procedures coded according to the Japanese original procedure codes.11 Oncological status according to the TNM classification is also available. Given the anonymous nature of the data collection process, informed consent from patients for inclusion in the study was not required. Approval of the study protocol was obtained from the institutional review board and the ethics committee of The University of Tokyo.
Sampling and end-points The DPC database was searched for patients with a main diagnosis at admission of renal malignancy (ICD-10 code: C64), with an oncological M status of M1, who underwent nephrectomy (Japanese codes for open nephrectomy: K773 and K773-2, and for laparoscopic nephrectomy: K773-1) from 2007 to 2012. Laparoscopic surgery was defined as MIS. The end-points evaluated were in-hospital mortality, perioperative complications, blood transfusion rate, anesthesia time, PLOS and total cost. The exchange rate was set at JP¥100 to US$1. 2
Statistical analysis
Results The clinical T stage was T1 in 270 patients, T2 in 215 patients, T3 in 479 patients and T4 in 110 patients. Table 1 shows the patient and tumor characteristics according to clinical T stage. Age, sex, CCI and type of hospital were not significantly different among the four stages. A higher clinical T stage was significantly associated with a higher clinical N stage (P < 0.001). Patients with a low clinical T stage were more likely to undergo MIS (T1: 44.1%, T4: 9.1%, P < 0.001). Table 2 shows comparisons of perioperative outcomes among the four stages. The blood transfusion rate (P < 0.001), anesthesia time (P < 0.001), PLOS (P < 0.001) and total cost (P < 0.001) increased significantly with increasing clinical T stage, but in-hospital death did not (P = 0.103). Postoperative complications tended to increase with increasing clinical T stage (7.8% in T1, 11.6% in T2, 12.5% in T3, and 16.4% in T4), but this increase was not statistically significant (P = 0.082). Tables 3 and 4 show the results of multivariable regression analyses for postoperative outcomes. A higher clinical T stage was significantly associated with a greater postoperative complication rate (T4/T1: OR 2.34, P = 0.024), blood transfusion © 2014 The Japanese Urological Association
CN for metastatic kidney cancer
Table 1
Comparisons of 1074 patients who underwent cytoreductive nephrectomy, according to clinical T stage
Variable
T1
T2
T3
T4
P-value
270
215
479
110
123 (45.6) 94 (34.8) 53 (19.6)
100 (46.5) 61 (28.4) 54 (25.1)
215 (44.9) 177 (37.0) 87 (18.2)
57 (51.8) 37 (33.6) 16 (14.5)
0.149
198 (73.3) 72 (26.7)
161 (74.9) 54 (25.1)
365 (76.2) 114 (23.8)
83 (75.5) 27 (24.5)
0.856
165 (61.1) 66 (24.4) 39 (14.4)
139 (64.7) 55 (25.6) 21 (9.8)
284 (59.3) 133 (27.8) 62 (12.9)
76 (69.1) 21 (19.1) 13 (11.8)
0.342
230 (85.2) 26 (9.6) 14 (5.2)
165 (76.7) 32 (14.9) 18 (8.4)
311 (64.9) 105 (21.9) 63 (13.2)
45 (40.9) 39 (35.5) 26 (23.6)
