9Archivio Italiano di Urologia e Andrologia 2021; 93, 1 ORIGINAL PAPER No conflict of interest declared. DOI: 10.4081/aiua.2021.1.9 INTRODUCTION Urothelial carcinoma of the upper urinary tract (UTUC) is among the ten most common cancers, is more frequent in males and diagnosis is generally done in the sixth decade (1, 2). Established risk factors are exposure to tobacco, arsenic, and aristolochic acid, as well as alcohol con- sumption (3). Some genetic polymorphisms are also asso- ciated with an increased risk of UTUC or faster disease progression that introduces variability in the inter-indi- vidual susceptibility to the risk factors previously men- tioned (4). The disease has high mortality, with more than 150.000 patients dying each year worldwide (1). Extirpative surgery with removal of kidney, entire ureter and bladder cuff – radical nephroureterectomy (RNU) – is the treatment of choice for non-metastatic high-risk UTUC. Conversely, low-risk cases (unifocal, < 2 cm, low- grade and superficial cancers) are amenable of kidney- sparing treatments providing equal survival outcomes but better preservation of renal function (4). Despite this rec- ommendation, a relevant rate of low-risk cases still under- go RNU for several reasons, mainly concerns on clinical understaging or challenging anatomical locations with inherent risk of tumor spillage and complications (4). The issue of renal function impairment after RNU is gen- erally postponed to the need for radicality, but UTUC patients are at high risk of chronic kidney disease (CKD) because of patient’s age and comorbidities, smoking exposure, potential impairment of contralateral kidney due to diagnostic procedures or bilateral UTUC. Indeed, despite adjuvant chemotherapy might prolong survival (5) and reduce the risk of disease recurrence in locally- advanced UTUC (6), nearly only 50% of patients are still eligible for platinum-based protocols, due to post-oper- ative renal failure (7, 8). Finally, it should be noted that CKD might determine worse mortality due to non-can- cer but also cancer-related causes, as found in patients treated for renal cell carcinoma (RCC) (9-11). Thus, the identification of patients at risk of significant renal function decline may allow clinicians to better assess the opportunity of kidney-sparing rather than Objective: The aim of our study was to inves- tigate frequency and predictors both of post- operative acute kidney injury (AKI) and renal function decline in a population of consecutive upper tract urothelial carcinoma (UTUC) patients who underwent radical nephroureterectomy (RNU). Materials and methods: Between October 2014 and February 2020, 93 patients underwent RNU at our Institution. After con- sidered exclusion criteria, 89 patients were selected. Perioperative clinical factors were retrospectively collected. Estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD- EPI) equation. We defined AKI as an increase in serum creati- nine by ≥ 0.3 mg/dl or a 1.5-1.9-fold increase in serum creati- nine from baseline to I post-operative day (POD). A significant renal function reduction was defined as an eGFR reduction of 40% from baseline at discharge and at last clinical evaluation. Frequency of AKI and eGFR decline was investigated. Association between perioperative clinical factors and AKI and eGFR reduction at discharged and last follow-up was studied using univariate and multivariate models. Results: AKI was detected at I POD in 45 patients. On multivari- ate analysis, pre-operative eGFR was an independent predictor of AKI (OR 1.03; p = 0.042). Further, AKI was found to be a sig- nificant predictor of eGFR reduction ≥ 40% at discharge at uni- variate analysis (OR 19.42; p = 0.005) and at multivariate analysis (OR 12.49; p = 0.02). In a multivariate logistic regres- sion model post-operative AKI (OR 5.18; p = 0.033), lack of ipsi- lateral preoperative hydronephrosis (OR 0.17; p = 0.016), pre- operative eGFR (OR 1.04; p = 0.047) and antiplatelet therapy (OR 5.14; p = 0.018) were found to be independent predictors of an eGFR reduction higher than 40% at last clinical evaluation made at a median of 15 (IQR 5-30) months. Conclusions: In our cohort, AKI was present in almost 50% of patients after RNU and it was a strong predictor of renal func- tion decline after RNU. KEY WORDS: Upper tract urothelial carcinoma; Radical nephroureterectomy; Acute kidney injury; Renal function reduction; Chronic kidney disease. Submitted 26 September 2020; Accepted 15 October 2020 Acute kidney injury strongly influences renal function after radical nephroureterectomy for upper tract urothelial carcinoma: A single-centre experience Summary Alessandro Tafuri 1, 2, Katia Odorizzi 1, Giacomo Di Filippo 1, 3, Clara Cerrato 1, Giulia Fassio 1, Emanuele Serafin 1, Alessandro Princiotta 1, Damiano D’Aietti 1, Alessandra Gozzo 1, Antonio B. Porcaro 1, Matteo Brunelli 4, Maria Angela Cerruto 1, Alessandro Antonelli 1 1 Department of Urology, University of Verona, Azienda Ospedaliera Universitaria Integrata Verona, Verona, Italy; 2 Department of Neuroscience, Imaging and Clinical Science, Physiology and Physiopathology division, “G. D’Annunzio” University, Chieti, Italy; 3 Department of General and Hepatobiliary Surgery, University of Verona, Azienda Ospedaliera Universitaria Integrata Verona, Verona, Italy; 4 Department of Pathology, University of Verona, Azienda Ospedaliera Universitaria Integrata Verona, Verona, Italy. Archivio Italiano di Urologia e Andrologia 2021; 93, 1 A. Tafuri, K. Odorizzi, G. Di Filippo, et al. 10 extirpative surgery when feasible, to adopt appropriate protective strategies during the peri-operative period and to adequately schedule controls during the follow up. The aim of our study is to investigate the prevalence of AKI and CKD, as well as the degree of renal functional impairment, in a population of patients submitted to RNU, in order to identify the most significant predictors of these events. MATERIALS AND METHODS Data were retrospectively collected in our Institutional Review Board (IRB) - approved UTUC dataset prospec- tively compiled since October 2014. At admission, each patient provided written informed consent for data col- lection and analysis. Between October 2014 and February 2020, 93 consecu- tive patients underwent RNU at our Institution as pri- mary treatment for UTUC. Patients who underwent pri- mary neo-adjuvant chemotherapy were not considered. For the purposes of the present study 4 patients with solitary kidney were excluded, leaving 89 patients, 10 of which previously submitted to radical cystectomy. In total 87 patients were submitted to open RNU and 2 to robot-assisted RNU. The following data were considered: gender, age at sur- gery, body mass index (BMI), performance status [American Society of Anesthesiologists (ASA) classification and Eastern Cooperative Oncology Group (ECOG) score], comorbidities (presence of coronary artery disease (CAD), pulmonary dis- ease, hypertension, hyperlipidemia, diabetes mellitus), smoking exposure, hydronephrosis, operative time, blood loss, intraoperative transfusions, pathological TNM stage (8th edition) (12), grade, presence of tumor necrosis, sur- gical margins status, length of hospital stay. According to an internal protocol regulating postopera- tive management, estimated glomerular filtration rate [eGFR, calculated by the CKD-EPI equation (13)] and blood chemistry were collected in all cases on post-oper- ative day (POD) 1, 3 and at discharge. Acute kidney injury was defined as an increase in serum creatinine with respect to baseline by ≥ 0.3 mg/dl or a 1.5-1.9-fold at I POD, according to the Acute Kidney Injury Network (AKIN) classification (14). According to previous reports, a renal function decline was considered significative when eGFR reduction got over 40% with respect to the baseline (15). Based on our internal protocol for low risk patients, fol- low-up controls were scheduled after 3 months from sur- gery performing cystoscopy and blood samples. If nega- tive, subsequent cystoscopy and urinary cytology, abdominal ultrasound, and blood samples were sched- uled 9 months later and then yearly, for 5 years. For high-risk patients, cystoscopy and urinary cytology at 3 months were performed. If negative, cystoscopy and cytology every 3 months for a period of 2 years, and every 6 months thereafter until five years, and then year- ly were considered. Additionally, yearly computed tomog- raphy (CT) urography and chest CT was scheduled. However, many patients traveled to our tertiary center from far away only for surgery, and the follow-up con- trols were often performed elsewhere. Table 1. Study population's characteristics (n = 89). N, % / Median [IQR] Age at surgery, years Follow up, months Gender Male Female Side Right Left BMI Smoking status No Yes Ex ASA score ≤ 2 > 2 ECOG score 0 1 2 Preoperative Hydronephrosis No Yes CAD No Yes Hypertension No Yes Preoperative Hb, mg/dL Preoperative Albumin, mg/dL Preoperative Creatinine, mg/dL Preoperative eGFR, mL/min/1.73 m2 Preoperative CKD stage 1 2 3 4 5 Hb I POD, mg/dL Hb III POD, mg/dL eGFR I POD, mL/min/1.73 m2 eGFR III POD, mL/min/1.73 m2 Hyperlipidemia No Yes Diabetes No Yes Antiplatelet therapy No Yes Anticoagulant therapy No Yes Antidiabetic therapy No Yes Operation time, min Blood loss, mL Transfusions No Yes Intraoperative complications No Yes pT stage ≤ 1 > 1 Grade Low High Tumour necrosis No Yes Lymph node status pN0 pN+ pNx Margin Status R0 R1 Length of stay, days eGFR at discharge, mL/min/1.73 m2 eGFR at last follow up, mL/min/1.73 m2 eGFR reduction I POD (%) eGFR reduction at III POD (%) eGFR reduction at discharge (%) eGFR reduction at last follow up (%) AKI I POD No Yes eGFR reduction at discharge, from baseline ≤ 39,9% > 40% eGFR reduction at last follow up, from baseline ≤ 39,9% > 40% AKI: Acute Kidney Injury; ASA: American Society of Anesthesiology; BMI: Body Mass Index; CAD: Coronary Artery Disease; CKD: Chronic Kidney Disease; ECOG: Eastern Cooperative Oncology Group; eGFR: estimated Glomerular Filtration Rate; Hb: Haemoglobin; POD: postoperative day. 71 [66-76] 15 [5-30] 61 (68.5%) 28 (31.5%) 43 (48.3%) 46 (51.7%) 25.6 [23.14-28] 24 (27%) 20 (22.5%) 45 (50.6%) 64 (71.9%) 25 (28.1%) 38 (42.7%) 39 (43.8%) 12 (13.5%) 38 (42.7%) 51 (57.3%) 77 (86.5%) 12 (13.5%) 39 (43.8%) 50 (56.2%) 13.4 [11.8-14.36] 38 [35.9-40.4] 1.1 [0.96-1.37] 58.39 [45.32-76.17] 9 (10.1%) 35 (39.3%) 38 (42.7%) 6 (6.7%) 1 (1.1%) 11.8 [10.5-12.9] 11.2 [10.3-12.6] 43.18 [33.49-52.99] 48.5 [36.63-57.23] 74 (83.1%) 15 (16.9%) 68 (76.4%) 21 (23.6%) 58 (65.2%) 31 (34.8%) 82 (92.1%) 7 (7.9%) 68 (76.4%) 21 (23.6%) 200 [178-252] 380 [250-600] 78 (87.6%) 11 (12.4%) 85 (95.5%) 4 (4.5%) 41 (46.1%) 48 (53.9%) 16 (18%) 73 (82%) 72 (83.7%) 14 (16.3%) 32 (36%) 16 (18%) 41 (46.1%) 83 (93.3%) 6 (6.7%) 10 [8-12] 51.16 [41.8-61.09] 48.45 [38.36-55.68] 25.14 [5.5-41.9] 17.7 [ -2.22-34.88] 8.57 [-7.42-32.53] 16.83 [-4.14-34.88] 44 (49.4%) 45 (50.6%) 74 (83.1%) 15 (16.9%) 69 (77.5%) 20 (22.5%) Categorical variables were expressed as frequencies and relative percentages. Continuous variables were expressed as median and interquartile range. Patients were divided into two groups according to the presence of I POD AKI. Categorical variables were compared between groups using the !2 test. Continuous variables were tested between subgroups with the independent samples t-test or Mann- Whitney test after testing for normality of distributions using Shapiro-Wilk test, as appropriate. Univariate logistic regression was used to identify outcomes’ predictors, and Odds Ratios and 95% confidence intervals were calculated for each signifi- cant variable. Significant variables at uni- variate analysis were entered into a multi- variate regression model to identify inde- pendent predictors. A p value < 0.05 was considered statistically significant. The analysis was carried out using SPSS soft- ware version 25.0 (SPSS Inc, Chicago, IL). RESULTS Demographics, clinical, operative and pathological data of the entire population are summarized in Table 1. The median age of the cohort was 71 years (IQR 66-76), 61 were males, 28 females; median follow-up time was 15 months (IQR 5-30). Median preoperative Hb was 13.4 g/dL (IQR 11.8-14.36) and median baseline eGFR was 58.39 ml/min/1.73 m2 (IQR 45.32-76.17). 51 patients (57.3%) had preoperative ipsilateral hydronephro- sis. Histology found pT1 or less in 41 patients (46.1%) and pT2 or higher in the other 48 (53.9%). Lymph node invasion in the pathological specimen was found in 16 (18%) cases. Post-operative median eGFR at 1, 3 POD and at discharge were 43.18 ml/min/1.73 m2 (IQR 33.49-52.99), 48.5 ml/min/1.73 m2 (IQR 36.63-57.23) and 51.16 ml/min/1.73 m2 (IQR 41.8-61.09) respectively. At the same timepoints, medi- an relative eGFR reduction was equal to 25.14%, 17.7%, and 8.57%. At discharge and last available follow up eGFR reduction was > 40% from the baseline in 15 (16.9%) and 20 (22.5%) patients, respectively. Overall, 45 patients (50.6%) developed AKI after surgery. The clinicopathological characteristics of the two groups (AKI com- pared to non-AKI) are reported in Table 2. On univariate analysis AKI was found sig- nificantly related to preoperative Hb value (OR 1.44; p = 0.003), preoperative eGFR (OR 1.04; p = 0.002), preoperative creati- nine serum level (OR 0.23, p = 0.018) and CKD stage < 2 (OR 0.29, p = 0.005). 11Archivio Italiano di Urologia e Andrologia 2021; 93, 1 RNU and renal function decline Table 2. Clinicopathological characteristics stratified according I POD AKI. Group 1 (No, n = 44) Group 1 (yes, n = 45) p value N (%); Median [IQR] N (%); Median [IQR] Age at surgery, years Follow up, months Gender Male Female Side Right Left BMI Smoking status No Yes Ex ASA score ≤ 2 > 2 ECOG score 0 1 2 Preoperative Hydronephrosis No Yes CAD No Yes Hypertension No Yes Preoperative Hb, mg/dL Preoperative Albumin, mg/dL Preoperative Creatinine, mg/dL Preoperative eGFR, mL/min/1.73 m2 Preoperative CKD stage 1 2 3 4 5 Hb I POD, mg/dL Hb III POD, mg/dL eGFR I POD, mL/min/1.73 m2 eGFR III POD, mL/min/1.73 m2 Hyperlipidemia No Yes Diabetes No Yes Antiplatelet therapy No Yes Anticoagulant therapy No Yes Antidiabetic therapy No Yes Operation time, min Blood loss, mL Transfusions No Yes Intraoperative complications No Yes pT stage ≤ 1 > 1 Grade Low High Tumour necrosis No Yes Lymph node status pN0 pN+ pNx Margin Status R0 R1 Length of stay, days eGFR at discharge, mL/min/1.73 m2 eGFR at last follow up, mL/min/1.73 m2 eGFR reduction I POD (%) eGFR reduction at III POD (%) eGFR reduction at discharge (%) eGFR reduction at last follow up (%) eGFR reduction at discharge, from baseline ≤ 39,9% > 40% eGFR reduction at last follow up, from baseline ≤ 39,9% > 40% AKI: Acute Kidney Injury; ASA: American Society of Anesthesiology; BMI: Body Mass Index; CAD: Coronary Artery Disease; CKD: Chronic Kidney Disease; ECOG: Eastern Cooperative Oncology Group; eGFR: estimated Glomerular Filtration Rate; Hb: Haemoglobin; POD: postoperative day. 71 [66-79] 16 [5-35] 26 (59.1%) 18 (40.9%) 14 (31.8%) 30 (68.2%) 25.63 [22.99-27.65] 12 (27.3%) 8 (18.2%) 24 (54.5%) 29 (65.9%) 15 (34.1%) 18 (40.9%) 22 (50%) 4 (9.1%) 15 (34.1%) 29 (65.9%) 41 (93.2%) 3 (6.8%) 22 (50%) 22 (50%) 12.65 [11.05-13.9] 37.7 [35.55-40.35] 1.21 [0.98-1.68] 51.34 [40.97-65.89] 3 (6.8%) 12 (27.3%) 23 (52.3%) 5 (11.4%) 1 (2.3%) 11.6 [10.35-12.6] 11.45 [10.5-12.5] 51.13 [41.67-62] 51.01 [41.31-68.16] 36 (81.8%) 8 (18.2%) 32 (72.7%) 12 (27.3%) 33 (75%) 11 (25%) 40 (90.9%) 4 (9.1%) 32 (72.7%) 12 (27.3%) 200 [161-251] 400 [275-600] 39 (88.6%) 5 (11.4%) 44 (100%) 0 (0%) 15 (34.1%) 29 (65.9%) 7 (15.9%) 37 (84.1%) 35 (83.3%) 7 (16.7%) 17 (38.6%) 9 (20.5%) 18 (40.9%) 38 (86.4%) 6 (13.6%) 10 [8-11] 53.7 [44.49-64.32] 50.43 [39.41-59.84] 5.44 [-7.94-12.75] -0.35 [-18.41-13.48] -3.16 [-16.63-4.68] -1.31 [-10.62-17] 43 (97.7%) 1 (2.3%) 41 (93.2%) 3 (6.8%) 72 [66-76] 15 [5-29] 35 (77.8%) 10 (22.2%) 29 (64.4%) 16 (35.6%) 25.6 [23.4-28.8] 12 (26.7%) 12 (26.7%) 21 (46.7%) 35 (77.8%) 10 (22.2%) 20 (44.4%) 17 (37.8%) 8 (17.8%) 23 (51.1%) 22 (48.9%) 36 (80%) 9 (20%) 17 (37.8%) 28 (62.2%) 13.9 [13-15.2] 38 [36.3-40.9] 1.05 [0.91-1.24] 69.37 [53.46-80.37] 6 (13.3%) 23 (51.1%) 15 (33.3%) 1 (2.2%) 0 (0%) 12 [10.5-13.4] 10.9 [10.1-12.7] 38.44 [31.66-45.42] 45.42 [36.31-54.13] 38 (84.4%) 7 (15.6%) 36 (80%) 9 (20%) 25 (55.6%) 20 (44.4%) 42 (93.3%) 3 (6.7%) 36 (80%) 9 (20%) 200 [180-252] 365 [250-650] 39 (86.7%) 6 (13.3%) 41 (91.1%) 4 (8.9%) 26 (57.8%) 19 (42.2%) 9 (20%) 36 (80%) 37 (84.1%) 7 (15.9%) 15 (33.3%) 7 (15.6%) 23 (51.1%) 45 (100%) 0 (0%) 10 [9-13] 46.98 [41.35-54.67] 46.98 [38.32-54.67] 41.9 [35.50-48.89] 34.83 [22.90-42.15] 30.69 [10.66-43.09] 33.17 [13.36-44.92] 31 (68.9%) 14 (31.1%) 28 (62.2%) 17 (37.8%) 0.663 0.660 0.058 0.002 0.159 0.908 0.213 0.355 0.105 0.069 0.245 0.001 0.042 0.001 0.001 0.044 0.007 0.003 0.001 0.001 0.741 0.419 0.054 0.671 0.419 0.928 0.388 0.778 0.043 0.025 0.615 0.924 0.615 0.010 0.993 0.004 0.001 0.001 0.001 0.001 0.001 0.001 0.001 Archivio Italiano di Urologia e Andrologia 2021; 93, 1 A. Tafuri, K. Odorizzi, G. Di Filippo, et al. 12 On multivariate analysis, preoperative eGFR was the only independent predictor to the occurrence of AKI (OR 1.03; p = 0.042) (Table 3). On univariate analysis AKI (OR 19.42; p = 0.005), preop- erative eGFR (OR 1.05; p = 0.004), preoperative Hb (OR 1.43; p = 0.036), the lack of ipsilateral hydronephro- sis (OR 0.30; p = 0.047), III POD creatinine serum level (OR 3.00; p = 0.018), I POD Hb (OR 1.57, p = 0.019) were predictors of eGFR reduction > 40% at discharge. On multivariate analysis only AKI retained its significance (OR 12.49; p = 0.02) (Table 4). Among several factors predicting eGFR reduction > 40% at last follow-up on univariate analysis, AKI (OR 5.18, p = 0.033), preoperative eGFR (OR 1.04, p = 0.047), the lack of ipsilateral hydronephrosis (OR 0.17; p = 0.016), and antiplatelet therapy (OR 5.14; p = 0.018) were found significantly associated to the outcome also on multivariate analysis (Table 5). DISCUSSION The present study shows that patients with UTUC candi- date to RNU have baseline poor renal function, with medi- an eGFR values close to 60 ml/min, and post-operatively suffer from further relevant decline given that an impair- ment exceeding 40% of baseline function was noted in 22.5% of cases at a median follow-up of 15 months. The factors associated with worse functional outcome were preoperative eGFR, lack of ipsilateral hydronephro- sis, antiaggregating therapy and the presence of post-oper- ative AKI, which represent the strongest predictor of CKD after RNU in the present cohort. Definitely these patients have major determinants of baseline impaired function and relevant risks to develop CKD, with inherent effects on non-cancer (11), but also cancer-related survival outcomes, firstly concerning the access to platinum-based chemotherapy regimens that showed significantly improved disease-free survival in locally advanced UTUC (6). Although renal function preservation in UTUC repre- sents a major issue it has been poorly investigated, with sparse reports in the literature. The main reason of this is that UTUC is often featured by aggressiveness and mul- tifocality, so that extirpative treatment is commonly priv- ileged, except for very selected low risk cases for whose kidney-sparing approaches might be preferred (16, 17). In 2006, Meyer et al. retrospectively analyzed 131 RNU patients reporting a 18% deterioration in eGFR after a median follow-up of 5 years. Such deterioration was found to be greater in patients with older age, and comor- bidities as diabetes mellitus, hypertension, pre-existing renal impairment and analgesic nephropathy (18). In a multicentric retrospective study evaluating 388 patient who underwent RNU for UTUC, Kaag MG et al. showed a mean 24% of eGFR decrease after surgery. They also reported that eligibility to platinum-based chemother- apy decreased from 49% before surgery to 19% post-sur- gery using a cut-off of 60 mL/min/1.73 m2, and from 80% to 55% using a cut-off of 45 mL/min/1.73 m2 (19). Kaag M. and his group identified age and preoperative eGFR as a predictors of renal function decline after RNU (20). They retrospectively enrolled 374 RNU patients and assessed early (1-5 months) and late (> 5 months) eGFR after surgery: multivariable analysis identified pre- operative eGFR lower than 60 mL/min/1.73 m2 and age > 70 years as preoperative predictors of clinically rele- vant eGFR loss after RNU, considering clinically relevant a loss of renal function compromising the possibility of chemotherapy recruitment. Shao et al. recently reported that among 242 RNU cases, 42.1% was eligible to cisplatin-based therapy prior to RNU whereas, following surgery, only 15.2% remained eligible, because of the worsening of renal function (8). In the present study, we also investigated the role of post- operative AKI, finding that a half of patients experienced this event. Interestingly, preoperative hemoglobin and preoperative eGFR predicted AKI, but only preoperative eGFR remained an independent predictor of AKI on multivariate analysis. The prevalence of AKI after RNU was previously unre- ported and resembles the data after radical nephrectomy for RCC (15). The most relevant finding of our analysis Table 3. Logistic regression analysis for I POD AKI predictors assessment. Univariate Multivariate OR 95% CI p value OR 95% CI p value Preoperative CKD stage 0.285 0.12-0.68 0.005 Preoperative creatinine 0.234 0.07-0.78 0.018 Preoperative eGFR 1.038 1.01-1.06 0.002 1.027 1.00-1.05 0.042 Preoperative Hb 1.441 1.13-1.83 0.003 1.295 0.99-1.68 0.051 AKI: Acute Kidney Injury; CKD: Chronic Kidney Disease; eGFR: estimated Glomerular Filtration Rate; Hb: Haemoglobin; POD: postoperative day. Table 4. Logistic regression analysis for the assessment of eGFR reduction > 40% from baseline to discharge. Univariate Multivariate OR 95% CI p value OR 95% CI p value Preoperative Hb 1.43 1.02-2 0.036 1.091 0.76-1.57 0.636 Hb I POD 1.567 1.08-2.28 0.019 Preoperative eGFR 1.053 1.02-1.09 0.004 1.036 0.99-1.08 0.108 AKI I POD 19.419 2.42-155.54 0.005 12.491 1.48-105.40 0.02 Creatinine III POD 3.004 1.21-7.45 0.018 Preoperative hydronephrosis 0.304 0.09-0.98 0.047 0.519 0.13-2.02 0.344 AKI: Acute Kidney Injury; CKD: Chronic Kidney Disease; eGFR: estimated Glomerular Filtration Rate; Hb: Haemoglobin; POD: postoperative day. Table 5. Logistic regression analysis for the assessment of eGFR reduction > 40% from baseline to the last follow-up. Univariate Multivariate OR 95% CI p value OR 95% CI p value Preoperative Hb 1.359 1.02-1.81 0.037 1.085 0.75-1.57 0.663 Hb I POD 1.598 1.13-2.26 0.008 Hb III POD 1.768 1.21-2.58 0.003 Preoperative eGFR 1.055 1.02-1.09 0.001 1.043 1-1.09 0.047 AKI I POD 8.298 2.22-31 0.002 5.183 1.14-23.54 0.033 Creatinine III POD 67.524 8.45-539.64 0.0001 Creatinine at discharge 72.359 9.87-530.65 0.0001 Preoperative hydronephrosis 0.167 0.05-0.52 0.002 0.172 1.14-23.55 0.016 Antiplatelet therapy 3.947 1.4-11.16 0.01 5.139 1.32-19.93 0.018 AKI: Acute Kidney Injury; CKD: Chronic Kidney Disease; eGFR: estimated Glomerular Filtration Rate; Hb: Haemoglobin; POD: postoperative day. 13Archivio Italiano di Urologia e Andrologia 2021; 93, 1 RNU and renal function decline is that AKI affects long-term renal function impairment, indicating that any effort should be done to prevent AKI, especially in patients at risk. Identification of such patients would allow to optimize the perioperative management in order to reduce the inci- dence of AKI after surgery and its consequences. A dedi- cated pre-, intra- and postoperative management with avoidance of potentially nephrotoxic agents, close moni- toring of serum creatinine and urine output (remembering that urine output and serum creatinine are changing very late during the development of AKI), optimization of vol- ume status and hemodynamic parameters and use of alter- natives to radio contrast agents, represent the best AKI preventative measure to adopt in perioperative time (21- 23). Also, anesthesiologist may contribute to renal damage prevention avoiding the reduction of renal blood and renal hypoxia, and preventing hypotension during surgery (21). The median percentage eGFR reduction we sought was 8.6% at discharge, 16.8% at last follow-up after a median of 15 months from surgery. The rate of patients experienc- ing an eGFR decline ≥ 40% at last follow-up was 22.5%. The factors independently associated to this event were post-operative AKI, the lack of preoperative hydronephro- sis, preoperative hemoglobin, and antiplatelet therapy. The interpretation of these findings is that already estab- lished contralateral hypertrophy due to hydronephrosis of the affected urinary tract, as well as better post-operative course, with less blood loss and without AKI, facilitate the compensatory role of the remnant solitary kidney. Additionally, antiplatelet drugs represent a risk factors for renal function impairing, especially after RNU. As we already mentioned, these patients should be properly man- aged pre-operatively and accurately followed after surgery. After nephrectomy UTUC patients showed larger eGFR reduction than those with parenchymal tumors, reason- ably because the latter are generally younger and with less comorbidities. However, there are some intrinsic dif- ferences between these two conditions still to be investi- gated. Tae et al. indeed investigated by matched-pair comparison 554 patients who underwent nephrectomy for UTUC (n = 277) or parenchymal tumor (n = 277), balanced in terms of age, BMI, baseline eGFR and comorbidities. A significant larger decline in postopera- tive eGFR was found in UTUC cases (73.3% vs. 66.1%, p = 0.039) and multivariate analysis showed that the indication to nephrectomy due to UTUC (OR 1.84; p = 0.006) was an independent predictor of postoperative impaired renal function (24). Lee et al. showed similar findings in 616 patient who underwent nephrectomy for renal cancer (n = 319), or UTUC (n = 297), with the lat- ter older and more comorbid. The authors reported that UTUC patients had an increased risk of serum creatinine doubling and need for dialysis after radical nephrectomy, and the predictor of these unfavorable outcomes were old age, diabetes, low baseline eGFR and indication to nephrectomy due to UTUC (25). Other tools have been proposed for identifying and mon- itoring patients at risk of renal failure. Brardi et al. inves- tigated the role of doppler ultrasound derived renal resis- tive index (RRI) in a CKD population in the monitoring of renal function after a therapeutic and dietetic inter- vention to ameliorate the renal impairment. The authors found that RRI was a key parameter in mon- itoring patients with CKD and a helpful tool to drive clinical efforts to contrast renal function decline (26). The same team recently found that variation in time of eGFR positively correlates to sonographic measurements of average right and left kidney diameters and percentage variations of right and left renal cortical thickness in a population of 80 adult patients with various degrees of chronic kidney disease after received of a therapeutic and dietetic intervention to improve renal function. Patients were not dialysis-dependent, they did not undergo renal surgery, nor they were affected by any of the pathological conditions that can increase kidney size (27). According to these pieces of evidence, renal ultrasound derived parameters together with clinical factors could represent a useful tool for evaluating patients before RNU and sub- sequent follow up. The present study is a retrospective evaluation of a small population. However, our results are innovative. We showed that AKI at I POD is a strong predictor of renal function decline in patients who underwent RNU for UTUC who might need an adjuvant platinum-based chemotherapy. Identifying patients at high-risk of renal function decline has a pivotal role to provide a correct peri-operative management. A tailored pre-operative management and surgical procedure should be provided to patients at high risk of developing AKI. Additionally, when UTUC patients are counseled before treatment, the risk of renal function decline should be extensively explained. Further higher-level studies are needed to confirm our results. CONCLUSIONS In our cohort, almost 50% of patients developed AKI after RNU. Acute kidney injury was a strong predictor of renal function decline after radical nephroureterectomy at discharged and a 15 months follow-up. Identifying patients at high-risk of renal function decline is essential to provide a correct peri-operative management. REFERENCES 1. Siegel RL, Miller KD, Jemal A. Cancer statistics. CA Cancer J Clin. 2019; 69:7-34. 2. Horstmann M, Witthuhn R, Falk M, Stenzl A. Gender-specific dif- ferences in bladder cancer: a retrospective analysis. Gend Med. 2008; 5:385-94. 3. Colin P, Koenig P, Ouzzane A, et al. Environmental factors involved in carcinogenesis of urothelial cell carcinomas of the upper urinary tract. BJU international, 2009; 104:1436-1440. 4. Rouprêt M, Babjuk M, Burger M, et al. European Association of Urology Guidelines on Upper Urinary Tract Urothelial Carcinoma: 2020 Update. Eur Urol. 2020:S0302-2838(20)30427-9. 5. Bamias A, Moulopoulos LA, Koutras A, et al. The combination of gemcitabine and carboplatin as first-line treatment in patients with advanced urothelial carcinoma. A Phase II study of the Hellenic Cooperative Oncology Group. Cancer. 2006; 106:297-303. 6. Birtle A, Johnson M, Chester J, et al. Adjuvant chemotherapy in upper tract urothelial carcinoma (the POUT trial): a phase 3, open- label, randomised controlled trial. Lancet. 2020; 395:1268-1277. Archivio Italiano di Urologia e Andrologia 2021; 93, 1 A. Tafuri, K. Odorizzi, G. Di Filippo, et al. 14 7. Galsky MD, Hahn NM, Rosenberg J, et al. Treatment of patients with metastatic urothelial cancer “unfit” for cisplatin-based chemotherapy. J Clin Oncol. 2011; 29:p 2432-2438. 8. Shao IH, Lin YH, Hou CP, et al. Risk factors associated with inel- igibility of adjuvant cisplatin-based chemotherapy after nephroureterectomy. Drug Des Devel Ther. 2014; 8:1985-90. 9. Antonelli A, Minervini A, Sandri M, et al. Below safety limits, every unit of glomerular filtration rate counts: assessing the rela- tionship between renal function and cancer-specific mortality in renal cell carcinoma. Eur Urol. 2018; 74:661-667. 10. Antonelli A, Palumbo C, Sandri M, et al. Renal function impair- ment below safety limits correlates with cancer-specific mortality in localized renal cell carcinoma: results from a single-center study. Clin Genitourin Cancer. 2020; 18:e360-e367. 11. Go AS, Chertow GM, Fan D, et al. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. New Engl J Med. 2004; 351:1296-1305. 12. Brierley JD, Gospodarowicz MK, Wittekind C. TNM classifica- tion of malignant tumours. 2017: John Wiley & Sons. 13. Levey AS, Stevens LA, Schmid CH, et al., A new equation to esti- mate glomerular filtration rate. Ann Int Med. 2009; 150:604-612. 14. Bellomo R, Ronco C, Kellum JA, et al. Acute Dialysis Quality Initiative workgroup. Acute renal failure - definition, outcome meas- ures, animal models, fluid therapy and information technology needs: the Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care 8:R204-12. 15. Garofalo C, Liberti ME, Russo D, et al. Effect of post-nephrecto- my acute kidney injury on renal outcome: a retrospective long-term study. World J Urol. 2018; 36:59-63. 16. Yakoubi R, Colin P, Seisen T, et al. Radical nephroureterectomy versus endoscopic procedures for the treatment of localised upper tract urothelial carcinoma: a meta-analysis and a systematic review of current evidence from comparative studies. Eur J Surg Oncol. 2014; 40:1629-34 17. Fang D, Seisen T, Yang K, et al. A systematic review and meta- analysis of oncological and renal function outcomes obtained after segmental ureterectomy versus radical nephroureterectomy for upper tract urothelial carcinoma. Eur J Surg Oncol. 2016; 42:1625- 1635. 18. Meyer JP, Delves GH, Sullivan ME, et al. The effect of nephroureterectomy on glomerular filtration rate. BJU internation- al. 2006; 98: 845-848. 19. Kaag MG, O'Malley RL, O'Malley P, et al. Changes in renal function following nephroureterectomy may affect the use of periop- erative chemotherapy. Eur Urol. 2010; 58:581-587. 20. Kaag M,Trost L, Thompson RH, et al. Preoperative predictors of renal function decline after radical nephroureterectomy for upper tract urothelial carcinoma. BJU international. 2014; 114:674-679. 21. Zarbock A, Koyner JL, Hoste EAJ, et al. Update on periopera- tive acute kidney injury. Anesth Analg, 2018; 127:1236-1245. 22. KDIGO clinical practice guideline for acute kidney injury. Kidney Int Suppl. 2012; 2:1-138. 23. Bell S, Prowle J. Postoperative AKI-Prevention Is better than cure? J Am Soc Nephrol. 2019; 30:4-6. 24. Tae BS, Ku JH, Kwak C, et al. Comparison of renal function after radical surgery for upper tract urothelial carcinoma versus renal cell carcinoma: propensity score matching. Urologia Internationalis. 2018; 101:400-408. 25. Lee, KH, Chen YT, Chung HJ, et al. Kidney disease progression in patients of upper tract urothelial carcinoma following unilateral radical nephroureterectomy. Renal failure. 2016; 38:77-83. 26. Brardi S, Cevenini G, Giovannelli V, Romano G. Longitudinal prospective observational type study about determinants of renal resistive index variations in chronic renal failure patients treated with conventional medical and dietetic therapy. Arch Ital Urol Androl. 2017; 89:305-309. 27. Brardi S, Cevenini G. Time changes of renal dimensions and variations of glomerular filtration rate in chronic kidney disease patients. Arch Ital Urol Androl. 2020; 92:21-24. Correspondence Alessandro Tafuri, MD aletaf@hotmail.it Katia Odorizzi, MD katia.odorizzi@virgilio.it Giacomo Di Filippo, MD giacomo.difilippo90@gmail.com Clara Cerrato, MD clara.cerrato01@gmail.com Giulia Fassio, MD fassio.giulia@gmail.com Emanuele Serafin, MD serafin.mnl@gmail.com Alessandro Princiotta, MD alessandroprinciotta0@gmail.com Damiano D’Aietti, MD damiano.daietti@gmail.com Alessandra Gozzo, MD la.ale.gozzo@gmail.com Antonio B. Porcaro, MD antoniobenito.porcaro@aovr.veneto.it Maria Angela Cerruto, MD mariaangela.cerruto@univr.it Alessandro Antonelli, MD (Corresponding Author) alessandro.antonelli@univr.it Department of Urology, University of Verona, Azienda Ospedaliera Universitaria Integrata Verona, Piazzale Stefani 1, 37126, Verona (Italy) Matteo Brunelli, MD matteo.brunelli@univr.it Department of Pathology, University of Verona, Azienda Ospedaliera Universitaria Integrata Verona, Verona (Italy)