Original Paper
Urologia Internationalis
Received: March 22, 2013 Accepted: June 3, 2013 Published online: November 23, 2013
Urol Int 2014;92:169–173 DOI: 10.1159/000353652
Predictive Factors for Impaired Renal Function following Nephroureterectomy in Upper Urinary Tract Urothelial Cell Carcinoma O. Rodríguez Faba a J. Palou a A. Breda a P. Maroto b J.M. Fernández Gómez c A. Wong a H. Villavicencio a a
Department of Urology, Universitat Autònoma de Barcelona, Fundació Puigvert, and b Department of Oncology, Universitat Autònoma de Barcelona, Sant Pau Hospital, Barcelona; c Department of Urology, Universidad de Oviedo, Oviedo, Spain
Key Words Hydronephrosis · Radical nephroureterectomy · Impaired renal function
Abstract Objectives: Despite the uncertain value of adjuvant chemotherapy after radical nephroureterectomy (RNU) it is clear that impaired renal function represents a contraindication to its administration. The objective of this study was to identify possible predictive clinical factors for impaired renal function following RNU in patients with upper urinary tract urothelial cell carcinoma (UUT-UCC). Patients and Methods: A retrospective analysis was conducted of 546 patients who underwent RNU between 1992 and 2008 at our institution. Data of interest for this study included estimated glomerular filtration rate (eGFR), age, pathological stage and preoperative hydronephrosis (HN). The predictive value of HN, age and pathological stage for impaired renal function after RNU was calculated by multivariate linear regression analysis. Results: In total, 138 patients met the criteria for inclusion, including 108 men (78%). Mean age at surgery was 67 ± 10 years. There was a significant correlation (p < 0.001) between pre- and postoperative eGFR (decrease of 21% after NU). Preoperative HN was present in 51 patients (37%). On linear regression analysis, preoperative eGFR ≤60 ml/min (p = 0.012; OR = 4.60) and HN (p = 0.027; OR = 10.34) were confirmed to
© 2013 S. Karger AG, Basel 0042–1138/13/0922–0169$38.00/0 E-Mail [email protected] www.karger.com/uin
be predictive factors for a postoperative eGFR ≤60 ml/min. When postoperative eGFR ≤45 ml/min was used as the criterion for impaired renal function, predictive factors proved to be preoperative eGFR ≤45 ml/min (p < 0.0001; OR = 18.53), HN (p = 0.038; OR = 0.380) and age ≥70 years (p < 0.0001; OR = 0.169). Conclusions: Preoperative HN, older age and preoperative eGFR 90%, however this rate decreases to 45 ml/min per 1.73 m2 must be achieved if the patient is to receive cisplatin- or carboplatin-based regimens, respectively [9]. Since renal insufficiency resulting from RNU represents one argument against adjuvant chemotherapy [10], the ability to predict which patients would develop renal insufficiency following RNU could be extremely useful. Two previously published series have reported the changes in renal function after RNU [10, 11]; in the present study, we retrospectively analysed our RNU database and identified clinical predictive factors for the development of renal insufficiency following RNU.
Table 1. Patient characteristics (n = 138)
Gender, n (%) Male Female Age at surgery, years Mean (range) Median Age ≥70 years, n (%) Hydronephrosis, n (%) Present Absent Pathological stage, n (%) pT0 pTa pTis pT1 pT2 pT3 pT4 pTx pN+
108 (78) 30 (22) 66 (35 – 89) 67 50 (36) 51 (37) 87 (63) 1 (0.72) 21 (15) 3 (2) 49 (35) 17 (12) 38 (27) 5 (4) 4 (3) 9 (6.5)
Methods After internal review board approval had been obtained for the study, our database of patients treated with RNU for UUT-UCC between 1992 and 2008 was retrospectively analysed. Clinical factors included in the analysis were hydronephrosis (HN) at surgery (present/absent), age and pathological stage. Although the CDKEPI equation has been proven to be more accurate, specifically at eGFR >60 ml/min [12], we followed the practice of other authors [11] in estimating GFRs using the equation developed from the Modification of Diet in Renal Disease (MDRD) study [13]. The eGFR cut-offs of 60 and 45 ml/min per 1.73 m2 were selected. HN was defined as present or absent according to CT scan findings. Two age cohorts were distinguished using 70 years as the cut-off, and pathological stage was classified as either localised disease (≤pT2) or locally advanced/advanced disease (pT3–4). Groups were compared using the predetermined cut-offs for eGFR, 60 and 45 ml/min per 1.73 m2, which represent the eligibility criteria for cisplatin and carboplatin [9, 14]. The χ2 test was used to evaluate the differences between age groups. Multivariate logistic regression analysis was used to determine the predictive factors for impaired renal function after RNU. All statistical calculations were two-tailed, and p < 0.05 was deemed significant. Statistical analysis was performed using SPSS v.19 (SPSS, Inc., Chicago, Ill., USA) for Microsoft Windows.
Results
Of the 546 patients, 138 were found to be suitable for the study, including 108 men (78%). Pathological, surgical and demographic data are shown in table 1. Comparison of pre- and postoperative serum creatinine lev170
Urol Int 2014;92:169–173 DOI: 10.1159/000353652
els revealed a mean increase of 36% after RNU (p < 0.001). A similar analysis for eGFR demonstrated an overall mean decrease of 21% following RNU (p < 0.001). Analysis of the distribution of eGFR before and after surgery revealed a significant correlation (p < 0.001) between pre- and postoperative eGFR. Outcomes were further stratified by pathological stage. Of the 43 patients with pathological stage pT3–4, 44% (n = 19) had a preoperative eGFR >60 ml/min. In this subgroup of patients, only 9% maintained a postoperative eGFR >60 ml/min. We then stratified our study population by age at surgery, using 70 years as the cut-off. Patients ≥70 years of age were found to be more likely to have a preoperative mean eGFR ≤60 ml/min compared with younger patients (54.96 vs. 61.96; p = 0.032). Among the 51 patients (37%) who presented HN at diagnosis, overall mean pre- and postoperative eGFR were 50.90 and 41.73 ml/min, respectively (p < 0.001). Of the 16 patients with HN and a preoperative eGFR of >60 ml/min, only 1 (6%) maintained a postoperative eGFR >60 ml/min; in contrast, 33 of the 34 (97%) patients with HN and a preoperative eGFR of >45 ml/min maintained a postoperative eGFR >45 ml/min (table 2). On multivariate analysis, factors predictive for a postoperative eGFR ≤60 ml/ min were preoperative eGFR ≤60 ml/min and HN. Factors predictive for a postoperative eGFR ≤45 ml/min were a preoperative eGFR ≤45 ml/min, HN and age ≥70 years (table 3). Rodríguez Faba /Palou /Breda /Maroto / Fernández Gómez /Wong /Villavicencio
Table 2. Effect of surgery on renal function: data stratified by hydronephrosis, stage and age (n = 138 patients)
CKD is associated with an increased risk of cardiovascular events and death from any cause [15]. CKD may also render patients ineligible for cisplatin-based chemotherapy. In a cohort of 662 patients with a normal serum creatinine, Huang et al. [4] analysed the effect of RN or
partial nephrectomy (PN) for solitary cortical tumours ≤4 cm. Multivariate analyses showed that RN remained an independent risk factor for a new onset of eGFR. Barlow et al. [16] analysed a series of 276 patients who underwent RN or PN for cortical tumours. After surgery, 52% of patients had developed new-onset eGFR 60 ml/min, while following RNU only 1 (2%) retained an eGFR at this level. The multivariate analysis confirmed this finding at both eGFR cut-offs. These findings suggest that patients with ipsilateral HN and UUT-UCC may not be good candidates for adjuvant
Predictive Factors of Renal Insufficiency after Nephroureterectomy
Urol Int 2014;92:169–173 DOI: 10.1159/000353652
Preoperative Creatinine, mg/dl eGFR, ml/min/1.73 m2
Postoperative
p
110 (60 – 228) 149 (72 – 1,105)
Urologia Internationalis
Received: March 22, 2013 Accepted: June 3, 2013 Published online: November 23, 2013
Urol Int 2014;92:169–173 DOI: 10.1159/000353652
Predictive Factors for Impaired Renal Function following Nephroureterectomy in Upper Urinary Tract Urothelial Cell Carcinoma O. Rodríguez Faba a J. Palou a A. Breda a P. Maroto b J.M. Fernández Gómez c A. Wong a H. Villavicencio a a
Department of Urology, Universitat Autònoma de Barcelona, Fundació Puigvert, and b Department of Oncology, Universitat Autònoma de Barcelona, Sant Pau Hospital, Barcelona; c Department of Urology, Universidad de Oviedo, Oviedo, Spain
Key Words Hydronephrosis · Radical nephroureterectomy · Impaired renal function
Abstract Objectives: Despite the uncertain value of adjuvant chemotherapy after radical nephroureterectomy (RNU) it is clear that impaired renal function represents a contraindication to its administration. The objective of this study was to identify possible predictive clinical factors for impaired renal function following RNU in patients with upper urinary tract urothelial cell carcinoma (UUT-UCC). Patients and Methods: A retrospective analysis was conducted of 546 patients who underwent RNU between 1992 and 2008 at our institution. Data of interest for this study included estimated glomerular filtration rate (eGFR), age, pathological stage and preoperative hydronephrosis (HN). The predictive value of HN, age and pathological stage for impaired renal function after RNU was calculated by multivariate linear regression analysis. Results: In total, 138 patients met the criteria for inclusion, including 108 men (78%). Mean age at surgery was 67 ± 10 years. There was a significant correlation (p < 0.001) between pre- and postoperative eGFR (decrease of 21% after NU). Preoperative HN was present in 51 patients (37%). On linear regression analysis, preoperative eGFR ≤60 ml/min (p = 0.012; OR = 4.60) and HN (p = 0.027; OR = 10.34) were confirmed to
© 2013 S. Karger AG, Basel 0042–1138/13/0922–0169$38.00/0 E-Mail [email protected] www.karger.com/uin
be predictive factors for a postoperative eGFR ≤60 ml/min. When postoperative eGFR ≤45 ml/min was used as the criterion for impaired renal function, predictive factors proved to be preoperative eGFR ≤45 ml/min (p < 0.0001; OR = 18.53), HN (p = 0.038; OR = 0.380) and age ≥70 years (p < 0.0001; OR = 0.169). Conclusions: Preoperative HN, older age and preoperative eGFR 90%, however this rate decreases to 45 ml/min per 1.73 m2 must be achieved if the patient is to receive cisplatin- or carboplatin-based regimens, respectively [9]. Since renal insufficiency resulting from RNU represents one argument against adjuvant chemotherapy [10], the ability to predict which patients would develop renal insufficiency following RNU could be extremely useful. Two previously published series have reported the changes in renal function after RNU [10, 11]; in the present study, we retrospectively analysed our RNU database and identified clinical predictive factors for the development of renal insufficiency following RNU.
Table 1. Patient characteristics (n = 138)
Gender, n (%) Male Female Age at surgery, years Mean (range) Median Age ≥70 years, n (%) Hydronephrosis, n (%) Present Absent Pathological stage, n (%) pT0 pTa pTis pT1 pT2 pT3 pT4 pTx pN+
108 (78) 30 (22) 66 (35 – 89) 67 50 (36) 51 (37) 87 (63) 1 (0.72) 21 (15) 3 (2) 49 (35) 17 (12) 38 (27) 5 (4) 4 (3) 9 (6.5)
Methods After internal review board approval had been obtained for the study, our database of patients treated with RNU for UUT-UCC between 1992 and 2008 was retrospectively analysed. Clinical factors included in the analysis were hydronephrosis (HN) at surgery (present/absent), age and pathological stage. Although the CDKEPI equation has been proven to be more accurate, specifically at eGFR >60 ml/min [12], we followed the practice of other authors [11] in estimating GFRs using the equation developed from the Modification of Diet in Renal Disease (MDRD) study [13]. The eGFR cut-offs of 60 and 45 ml/min per 1.73 m2 were selected. HN was defined as present or absent according to CT scan findings. Two age cohorts were distinguished using 70 years as the cut-off, and pathological stage was classified as either localised disease (≤pT2) or locally advanced/advanced disease (pT3–4). Groups were compared using the predetermined cut-offs for eGFR, 60 and 45 ml/min per 1.73 m2, which represent the eligibility criteria for cisplatin and carboplatin [9, 14]. The χ2 test was used to evaluate the differences between age groups. Multivariate logistic regression analysis was used to determine the predictive factors for impaired renal function after RNU. All statistical calculations were two-tailed, and p < 0.05 was deemed significant. Statistical analysis was performed using SPSS v.19 (SPSS, Inc., Chicago, Ill., USA) for Microsoft Windows.
Results
Of the 546 patients, 138 were found to be suitable for the study, including 108 men (78%). Pathological, surgical and demographic data are shown in table 1. Comparison of pre- and postoperative serum creatinine lev170
Urol Int 2014;92:169–173 DOI: 10.1159/000353652
els revealed a mean increase of 36% after RNU (p < 0.001). A similar analysis for eGFR demonstrated an overall mean decrease of 21% following RNU (p < 0.001). Analysis of the distribution of eGFR before and after surgery revealed a significant correlation (p < 0.001) between pre- and postoperative eGFR. Outcomes were further stratified by pathological stage. Of the 43 patients with pathological stage pT3–4, 44% (n = 19) had a preoperative eGFR >60 ml/min. In this subgroup of patients, only 9% maintained a postoperative eGFR >60 ml/min. We then stratified our study population by age at surgery, using 70 years as the cut-off. Patients ≥70 years of age were found to be more likely to have a preoperative mean eGFR ≤60 ml/min compared with younger patients (54.96 vs. 61.96; p = 0.032). Among the 51 patients (37%) who presented HN at diagnosis, overall mean pre- and postoperative eGFR were 50.90 and 41.73 ml/min, respectively (p < 0.001). Of the 16 patients with HN and a preoperative eGFR of >60 ml/min, only 1 (6%) maintained a postoperative eGFR >60 ml/min; in contrast, 33 of the 34 (97%) patients with HN and a preoperative eGFR of >45 ml/min maintained a postoperative eGFR >45 ml/min (table 2). On multivariate analysis, factors predictive for a postoperative eGFR ≤60 ml/ min were preoperative eGFR ≤60 ml/min and HN. Factors predictive for a postoperative eGFR ≤45 ml/min were a preoperative eGFR ≤45 ml/min, HN and age ≥70 years (table 3). Rodríguez Faba /Palou /Breda /Maroto / Fernández Gómez /Wong /Villavicencio
Table 2. Effect of surgery on renal function: data stratified by hydronephrosis, stage and age (n = 138 patients)
CKD is associated with an increased risk of cardiovascular events and death from any cause [15]. CKD may also render patients ineligible for cisplatin-based chemotherapy. In a cohort of 662 patients with a normal serum creatinine, Huang et al. [4] analysed the effect of RN or
partial nephrectomy (PN) for solitary cortical tumours ≤4 cm. Multivariate analyses showed that RN remained an independent risk factor for a new onset of eGFR. Barlow et al. [16] analysed a series of 276 patients who underwent RN or PN for cortical tumours. After surgery, 52% of patients had developed new-onset eGFR 60 ml/min, while following RNU only 1 (2%) retained an eGFR at this level. The multivariate analysis confirmed this finding at both eGFR cut-offs. These findings suggest that patients with ipsilateral HN and UUT-UCC may not be good candidates for adjuvant
Predictive Factors of Renal Insufficiency after Nephroureterectomy
Urol Int 2014;92:169–173 DOI: 10.1159/000353652
Preoperative Creatinine, mg/dl eGFR, ml/min/1.73 m2
Postoperative
p
110 (60 – 228) 149 (72 – 1,105)
