european urology 50 (2006)

Transcription

1 european urology 50 (2006) available at journal homepage: Kidney Cancer Surgical Care, Morbidity, Mortality and Follow-up after Nephrectomy for Renal Cancer with Extension of Tumor Thrombus into the Inferior Vena Cava: Retrospective Study Since 1990s Jérôme Rigaud *, Jean-François Hetet, Guillaume Braud, Simon Battisti, Loïc Le Normand, Pascal Glemain, Georges Karam, Olivier Bouchot Clinique Urologique, Hôtel Dieu, CHU Nantes, France Article info Article history: Accepted February 8, 2006 Published online ahead of print on March 24, 2006 Keywords: Renal cell carcinoma Vena cava Thrombus Cancer Kidney Morbidity Mortality Survival Abstract Objectives: The aim of our survey was to evaluate surgical care, morbidity, mortality and follow-up of patients who had undergone surgical exeresis of a renal cancer with extension of tumor thrombus into the inferior vena cava. Patients and methods: Between June 1991 and March 2003, 40 (5.4%) patients were operated on for an enlarged nephrectomy with thrombectomy. The upper limit of the tumor thrombus was below the sus-hepatic veins in 21 (52.5%) patients and above the sus-hepatic veins in 19 (47.5%) patients with six (15%) located in the right atrium. Results: Cardiopulmonary bypass (CPB) was used for 12 patients (30%). A per-operative embolism was noted for three (7.5%) patients: two cases of cruoric embolism and one case of gaseous embolism, systematically occurring in patients operated on without CPB. Early mortality was 7.5% (three patients) attributable to hemorrhagic complications. Overall survival at 2 and 5 years was 45.2% and 38.7%, respectively. Disease-free survival at 2 and 5 years was 28.3% and 8.9% respectively. Only the pn stage had a statistically significant prognosis value for overall survival but not for disease-free survival. At the end of the study, only one (2.5%) patient could be considered free of the disease with sufficient follow-up after the surgery. Conclusion: Patients with renal cancer and tumor extension in the inferior vena cava need multidisciplinary cooperation to adapt a good surgical strategy, particularly with the use of CPB. However, the rate of patients free of disease after such surgery was low. # 2006 European Association of Urology. Published by Elsevier B.V. All rights reserved. * Corresponding author. Cinique Urologique, CHU Hôtel Dieu, 1 Place Alexis Ricordeau, Nantes, France. Tel ; Fax: address: jrigaud@chu-nantes.fr (J. Rigaud) /$ see back matter # 2006 European Association of Urology. Published by Elsevier B.V. All rights reserved. doi: /j.eururo

2 european urology 50 (2006) Introduction The management of localized and metastatic renal cancer has changed in the last 20 years, predicated on major advancements in renal imaging, surgical techniques and development of effective adjuvant therapies like immunotherapies [1]. In 5 10% of cases of renal tumors, a venous tumor extension is found, which usually extends into the inferior vena cava up to the right atrium [2 4]. Before the 1970s, patients with such a tumor extension were considered inoperable because of the major risk involved with surgery. The evolution was then fatal, as no alternative therapeutics could be proposed for these patients. Advances in surgical technique, particularly the development of the cardiopulmonary bypass (CPB), permit removal of tumor thrombus without significant surgical risk [5]. Lymph node involvement or distant metastases were associated with a poor prognosis. However, venous extension is not associated infrequently with adverse prognostic factor. The problem is to evaluate whether this invasive surgery is beneficial for the patient, in particular for the oncological results. The aim of our study was to report the outcome of patients undergoing nephrectomy with exeresis of a tumor thrombus in the inferior vena cava in our department. We analysed the surgical procedure according to the level of thrombus, morbidity, mortality, pathological results and overall and disease-free survival. Also, we analysed the prognostic factors for this surgery. 2. Patients and methods 2.1. Population Between June 1991 and March 2003, 743 patients underwent an enlarged nephrectomy for a renal tumor in our department. Nephrectomy with tumor thrombectomy was done in 40 (5.4%) patients because of tumor extension into the inferior vena cava. The population was 25 men and 15 women (sex ratio of 1.6). The average age was years (range 39 80) Clinical data The renal tumor was located on the right kidney in 27 (67.5%) cases, on the left kidney in 10 (25%) cases, and on both kidneys in three (7.5%) cases. As for the cases of bilateral tumors, the patients underwent an enlarged exeresis on the side corresponding to the vena cava tumor thrombus, and a partial exeresis on the contra-lateral side during the same operative time. The vena cava tumor thrombus affected the right kidney in 29 (72.5%) cases and the left kidney in 11 (27.5%) cases. The upper limit of the vena cava tumor thrombus was infrahepatic (level 1) in six (15%) cases, intra-hepatic (level 2) in 15 (37.5%) cases, supra-hepatic (level 3) in 13 (32.5%) cases and intra-atrial (level 4) in six (15%) cases. Distant metastasis was found in 15 (37.5%) patients, essentially located in the lungs. The rationale for resection of these metastatic patients was for palliative indications attributable to haematuria or pain (eight cases) and for clinical trial with treatment by adjuvant immunotherapy (seven cases) Statistical analysis The data were entered in a database of the Excel 2000 type (Microsoft Corporation, Redmond, WA, USA) and was analysed statically with the use of the StatView software, version 5.0 (SAS, Cary, NC, USA). The overall survival and survival without recurrence were calculated by the Kaplan-Meier method. The different clinical and anatomic-pathological data was analysed as predictive factors for the overall survival or survival without recurrence, with a statistical comparison measured by the log-rank test. The qualitative variables were compared with the use of a chi-square test and the quantitative variables with the use of a Student t-test. Test results were considered statistically significant for p < Results 3.1. Surgical strategy The surgical procedure was established in function of the upper limit of the vena cava tumor thrombus (Table 1) Under the sus-hepatic veins (level 1 and 2) (21 patients) The vena cava tumor thrombus was approached exclusively by an abdominal incision in all cases. Control of the inferior vena cava always was achieved in an intra-abdominal approach. For two patients, clamping of the aorta and hepatic pedicle (Pringle maneuver) was performed to reduce the blood losses Intra-atrial (level 4) (six patients) All the patients underwent CPB as well as an incision of the right atrium to allow complete extraction of the tumor thrombus. Due to the fact that blood circulation had been interrupted, hepatic clamping did not prove necessary Supra-hepatic but sub-atrial (level 3) (13 patients) These patients underwent an abdominal incision exclusively (seven cases) or CPB (six cases) Patients operated on without CPB. Venous control above the tumor thrombus was performed five times by trans-abdominal intra-pericardial incision without opening the thorax. In the other two cases, upper

3 304 european urology 50 (2006) Table 1 Level of the thrombus, transfusion, morbidity and mortality according to the surgical technique Variables Abdominal (n = 28) CPB (n = 12) p value * Level of the thrombus Infra-hepatic 6 0 Intra-hepatic 15 0 Supra-hepatic 7 6 Intra-atrial 0 6 Transfusion (no. of packed red blood cells) Total Supra-hepatic (n = 13) Supra-hepatic and intra-atrial (n = 19) Embolism (tumoral or gaseous) Morbidity Mortality Hospitalisation time (d) Abdominal = abdominal incision exclusively; CPB = abdominal incision and cardiopulmonary bypass. * p values are taken from the chi-square distribution. control of the thrombus was possible by abdominal incision exclusively. For all the patients, the hepatic pedicle had to be clamped (Pringle maneuver) Patients operated on with CPB. The heart was opened in three cases of cardiac arrest to better control the upper part of the vena cava tumor thrombus, which extended up to the ostium of the right atrium, thus allowing as complete an exeresis as possible Exeresis of tumor thrombus Exeresis of the tumor thrombus was achieved in all cases through a longitudinal cavotomy incised along the tumor thrombus to the orifice of renal vein. In 23 (57.5%) cases, the tumor thrombus did not adhere to the venous wall. In case of adhesion between the tumor thrombus and wall of the inferior vena cava (42.5%), the maximum of tumor tissue was removed, and a portion of the inferior vena cava invaded by the tumor was resected with an appropriate margin. No prosthetic replacement of the inferior vena cava resection was done. In only one (4.7%) case, the inferior vena cava was resected totally because of massive extension of the tumor thrombus into the wall, but it was not replaced by a vascular prosthesis Transfusions The total rate of patients transfused during the hospitalisation was 87.5% (35 patients). The rate of patients transfused who were operated on with CPB or without CPB was 100% and 82.1%, respectively ( p = ). The transfused patients received an average of packed red blood cells (range, 2 23); 11 of them also received fresh deepfrozen plasmas (range, 1 18). The number of packed red blood cells transfused according to the upper limit of the vena cava tumor thrombus is shown in Table 1. Briefly, there were more transfusions for patients operated on without CPB when the tumor thrombus was supra-hepatic Mortality and morbidity There was no per-operative mortality. However, three (7.5%) patients died during the hospitalisation after the nephrectomy with thrombectomy: one patient underwent CPB and two did not ( p = ; Table 1). Two patients died of hemorrhagic complications on day 2 and day 11. The other patient died of medical complications on day 33 after a tumor thrombus pulmonary embolism occurred during the intervention. A per-operative embolism of the tumor thrombus was noted for two patients during dissection of the inferior vena cava. One patient died of medical complications after this tumor thrombus embolism on day 33. The other patient had no metastasis at the diagnosis, but we observed lung metastasis at 6 months and he died 9 months after the intervention. Similarly, a gaseous embolism was observed for 1 patient during venous unclamping but without consequence for the post-operative care. These three per-operative embolic complications (7.5%) were reported for only patients operated on without CPB but are not statistically significant ( p = ). Early morbidity (within 30 days after the operation) was 47.5% (19 patients). The patients operated on with CPB or without CPB had an early morbidity in 50% and 46.4% of the cases, respectively ( p = ); Table 1). This morbidity was characterized by urinal and pulmonary infections (seven cases), haematomas of the renal fossa (five cases), per-operative tumor or gaseous embolism (three

4 european urology 50 (2006) Table 2 Pathological characteristics Variables No. of patients (%) Histology Clear cell tumors 36 (90%) Collecting duct 2 (5%) Sarcomatoid carcinoma 2 (5%) Nuclear grading Grade 2 4 (10%) Grade 3 28 (70%) Grade 4 8 (20%) Pathological stage pt pt3b 30 (75%) pt3c 10 (25%) Infiltration of perirenal fat Yes 19 (47.5%) No 21 (52.5%) Fig. 1 Overall survival of the population (n = 40). Adhesion of the thrombus Yes 17 (42.5%) No 23 (57.5%) Pathological stage pn pn0 35 (87.5%) pn1 3 (7.5%) pn2 2 (5%) Clinical stage M M0 25 (62.5%) M1 15 (37.5%) cases), haemostasis splenectomy attributable to decapsulation of the spleen (two cases), renal failure (one case) and respiratory failure (one case). Another surgical intervention was necessary for four (10%) patients: three for a post-operative haematoma and one for draining a pleural effusion. The patients operated on with CPB or without CPB had a second surgical intervention in 9% and 10% of the cases, respectively ( p = ). No specific complications linked to the CPB were observed Pathological stage The pathological data are shown in Table 2 in accordance with the TNM 2002 classification. Distant visceral metastases were observed during radiological staging of the extension for 15 (37.5%) patients. The metastatic sites were lung (11 cases), Fig. 2 Disease-free survival of the population (n = 40). liver (four cases), homolateral adrenal gland (four cases), bone (two cases) and brain (one case) Survival and prognosis factors The mean follow-up of the population was months (range, months; median, 15.9 months). Only seven patients received adjuvant immunotherapy after the surgery because of the presence of metastasis at diagnosis. The other 33 patients had a wait-and-see procedure until recurrence. Table 3 Overall and disease-free survival according to the stage of tumor Overall survival Disease-free survival 2-year 5-year 2-year 5-year All patients (40) 45.2% 38.7% 28.3% 8.9% M0 (25 patients) 51.8% 44.4% 30.4% 8.7% N0M0 (22 patients) 59.5% 51% 35% 10% M0 = patients without distant metastasis; N0M0 = patients without lymph node and distant metastasis (i.e., localised tumor).

5 306 european urology 50 (2006) Table 4 Predictive factors for local or metastatic recurrence Variables Local recurrence Metastatic recurrence n/total % p value * n/total % p value * Level of tumor thrombus Infra-hepatic 2/6 33% 4/6 66% Intra-hepatic 5/15 33% 7/15 46% Supra-hepatic 5/13 38% 9/13 69% Intra-atrial 2/6 33% 3/6 50% Side of the tumor thrombus Right 7/29 24% 17/29 58% Left 7/11 63% 6/11 54% Surgical technique Abdominal 9/28 32% 15/28 53% CPB 5/12 41% 8/12 66% Histological type Clear cell tumors 12/36 33% 21/36 58% Collecting duct 1/2 50% 1/2 50% Sarcomatoid carcinoma 1/2 50% 1/2 50% Nuclear grading Grade 2 0/4 0% 2/4 50% Grade 3 10/28 35% 17/28 60% Grade 4 4/8 50% 4/8 50% Stage pt pt3b 11/30 36% 18/30 60% pt3c 3/10 30% 5/10 50% Infiltration of perirenal fat Yes 9/19 47% 11/19 57% No 5/21 23% 12/21 57% Adhesion of the thrombus Yes 7/17 41% 11/17 64% No 7/23 30% 12/23 52% Stage pn pn0 13/35 37% 20/35 57% pn+ 1/5 20% 3/5 60% Stage M M0 10/25 40% 18/25 72% M+ 4/15 26% 5/15 33% Transfusion Yes 14/35 40% 21/35 60% No 0/5 0% 2/5 40% Abdominal = abdominal incision exclusively; CPB = abdominal incision and cardiopulmonary bypass. * p values are taken from the chi-square distribution. Twenty-two (55%) patients died of progression of their disease with a mean time of months after the surgical intervention (range, months; median, 9.2 months). Overall survival at 2 and 5 years was 45.2% and 38.7%, respectively (Fig. 1). Local and/or metastatic recurrence during followup of all the patients was noted for 28 (70%) patients with a mean time of months (range, months; median, 9 months) for the first recurrence. Disease-free survival at 2 and 5 years was 28.3% and 8.9%, respectively (Fig. 2). The overall survival and disease-free survival of the sub-group of 25 patients without distant metastasis at diagnosis (M0) and of the 23 patients without lymph node and distant metastasis (N0M0) are shown in Table 3. A local recurrence of the tumor during the followup period was observed for 14 (35%) patients with a mean time of months (range, 1 37 months; median, 9 months). The only predictive factor for local recurrence was the side of the kidney that was affected by the tumor thrombus. Patients with a tumor thrombus that started from the left kidney had more local recurrence (Table 4). The sites of the local recurrences were the cave thrombus (six cases), the lymph node alongside the aorta (four

6 european urology 50 (2006) Table 5 Overall survival and disease-free survival according to clinicopathological variables Variables 3-year overall survival p value * 3-year disease-free survival p value * Level of tumor thrombus Infra 38.9% 30.4% Supra 52.4% 25.1% Surgical technique Abdominal 37.6% 28.0% CPB 62.5% 30.3% Pathological stage pt pt3b 39.1% 26.2% pt3c 64.0% 31.7% Nuclear grading Grade % 100% Grade % 25.2% Grade % 15.0% Infiltration of perirenal fat Yes 44.7% 25.2% No 46.8% 33.4% Adhesion of the thrombus Yes 37.8% 16.4% No 48.4% 36.8% Pathological stage Pn pn0 49.0% 28.5% pn+ 20.0% 25.0% Clinical stage M M0 51.8% 30.4% M1 32.7% 22.2% Abdominal = abdominal incision exclusively; CPB = abdominal incision and cardiopulmonary bypass. * p values are based on the log-rank test. cases) and the renal fossa (four cases). Treatment of the local recurrence was re-operation for tumor exeresis (two cases), immunotherapy (five cases) or abstention (seven cases). A distant metastatic extension during the followup period was observed for 23 (57.5%) patients who did not feature these initial metastatic locations, after a mean time of months (range, months; median, 15 months). The metastatic recurrence was not preceded systematically by a Fig. 3 Survival outcomes of the population. local recurrence. Also, not all the patients suffering from a local recurrence subsequently had a metastatic recurrence. The predictive factor for metastatic recurrence was the metastatic status at diagnosis: Patients with no metastatic at diagnosis had more risk of developing metastasis after surgery (Table 4). The sites of metastasis were the lung (12 cases), bone (eight cases), brain (five cases), liver (five cases), a peritoneum carcinosis (three cases), the adrenal gland (two cases) and/or an inguinal adenopathy (one case). Therapeutic care of the metastatic evolution was undertaken for 14 patients through immunotherapy (eight cases), external radiotherapy on the bone or brain (four cases) and/or surgical exeresis of the metastasis (three cases). Abstention with a simple survey was adopted for the remaining nine patients. Table 5 shows the prognosis factor for overall survival or disease-free survival. Only the pn stage had a statistically significant prognosis value in terms of overall survival. At the end of the study, 22 patients died of progression of their disease. Of the 18 living patients, 12 had progression of the disease by the development of new metastatic sites (11) or by a

7 308 european urology 50 (2006) local recurrence (one); four patients had metastasis at diagnosis and were not cured; and two patients had no progression of the disease but one was lost to follow-up after surgery and the other had a followup of 52 months. Finally, only one (2.5%) patient could be considered free of the disease with a sufficient follow-up after the surgery (Fig. 3). 4. Discussion Renal tumors with extension into the inferior vena cava require multidisciplinary actions and heavy surgical interventions. Our retrospective survey highlights higher mortality and morbidity rates (7.5% and 47.5%, respectively) than those for conventional enlarged nephrectomies without thrombus in the inferior vena cava. The 5-year overall survival and disease-free survival rates were 38.7% and 8.9%, respectively. However at the end of the study, only one (2.5%) patient could be considered free of the disease with a sufficient follow-up after the surgery. Methodological weakness could be made about this survey. Indeed, this study was retrospective with all the bias of this kind of study. During this period, the upper limit of the vena cava tumor thrombus was not evaluated pre-operatively for all patients by computed tomography, magnetic resonance imaging or transesophageal echocardiography. Furthermore, all metastatic patients at diagnosis did not receive adjuvant immunotherapy after the surgery during the study. Also the followup was not done systematically by regular computed tomography, and recurrence was defined only on this examination without biopsy. Evaluation of the upper limit of the inferior vena cava tumor thrombus is necessary to adapt the surgical strategy. Patients in whom the upper limit of the tumor thrombus is situated under the subhepatic veins conventionally undergo abdominal incision exclusively with control of the inferior vena cava on both sides of the thrombus. This control could be achieved with a different technique, particularly with careful mobilization of the retrohepatic portion of the inferior vena cava, similar to that performed during liver harvesting [6]. In contrast, practically all the patients suffering from a thrombus extending into the right atrium are operated on with CPB [5,7 11]. However, when the tumor thrombus is above the sus-hepatic veins but under the right atrium, the operative strategy is variable. Indeed, for some authors, the affection of the sus-hepatic veins requires a CPB for the extraction, whereas other authors perform the exeresis by the abdominal method exclusively. In our series, 13 patients had an extension of the tumor thrombus above the sus-hepatic veins but under the right atrium: Seven were operated on by the abdominal method without CPB but with transabdominal intra-pericardial clamping of the inferior vena cava in three fourths of the cases, whereas the other six were operated on with CPB. When comparing the two groups of patients, the transfusion rate and, therefore, the indirect blood losses are greater for the patients operated on without CPB. Furthermore, CPB brings a certain security during surgical interventions. In the sub-group of 19 patients with a thrombus extending above the sus-hepatic veins, patients operated on without CPB required more transfusions. Similarly, we noted three per-operative embolism complications (gaseous and tumorous) in only patients operated on without CPB. In contrast, no significant differences were found in terms of morbidity, mortality and hospitalisation time in the groups of patients operated on with or without CPB. It would appear that, for patients suffering from a tumor thrombus extending above the sus-hepatic veins, the CPB does not increase morbidity and mortality, but it increases operative security through better control of the thrombus, thus reducing the risks of embolism and haemorrhage [9,11 12]. In our survey, we performed a cavectomy without substitution in one case only. The rate of local recurrence was not correlated statistically with the infiltration of the wall. Also, the infiltration of the venous wall was not a prognostic factor for the overall or disease-free survival. However, the only predictive factor for local recurrence was the side of the thrombus: Patients with a tumor thrombus that started from the left kidney had more local recurrence, probably because dissection between the aorta and mesenteric artery could not be as extensive, whereas extensive surgery to the right side is easier. Only the pathological lymph node status was a predictive factor for overall survival in our survey. This prognosis value was highlighted in other studies [4,7,13 17]. In contrast, the upper level of the thrombus, the parietal invasion and the operative technique (CPB versus absence of CPB) were not prognostic factors in terms of survival, as other studies have demonstrated [7,10,13 15,18 20].Even if the upper level of the thrombus was not a prognostic value in such patients, Lang et al. [21] demonstrated that the microscopic venous invasion was an independent prognostic factor for survival for patients with pt1 to pt3b N0 M0 renal cell carcinoma. It would appear that the prognosis for patients

8 european urology 50 (2006) operated on for a renal tumor with a thrombus in the vena cava is related more to the lymph node involvement than to the level of the tumor thrombus. In contrast, the prognosis value of the initial metastatic impairment is more disputed. In our survey, as in other studies, it did not have any prognosis value [7,10,15]. Nevertheless, some have reported improved survival for patients without distant metastasis at the time of diagnosis [13,14,17,22]. Probably, these patients required an adjuvant treatment to surgery which, alone, is not sufficient. However, in the field of renal neoplasia, few adjuvant or neo-adjuvant treatments have demonstrated real efficiency. It would nevertheless appear that adjuvant immunotherapy could improve survival for the patients having a good prognosis of the N0M0 level [23] or for patients having pulmonary metastasis at diagnosis [24]. Also, patients with pulmonary metastases will be considered for metastasectomy with a good survival outcome, particularly if the metastasis is solitary and metachronous, and the resection is completed (R0) [25]. 5. Conclusion The first important finding of this study is that patients with tumor thrombus in the inferior vena cava require a multidisciplinary approach, particularly if the tumor thrombus extends above the sushepatic veins. Indeed, utilisation of CPB increases operative security through better control of the thrombus, thus reducing the risks of embolism and haemorrhage without augmentation of morbidity or mortality. The second important point is that survival of such patients after nephrectomy was bad with only 2.5% of patient free of disease with sufficient followup. Probably these patients need an adjuvant treatment after surgery to extend survival. References [1] Lam JS, Shvarts O, Pantuck AJ. Changing concepts in the surgical management of renal cell carcinoma. Eur Urol 2004;45: [2] Bretheau D, Koutani A, Lechevallier E, Coulange C. A French national epidemiologic survey on renal cell carcinoma. Oncology Committee Association Francaise d Urologie. Cancer 1998;82: [3] Nesbitt JC, Soltero ER, Dinney CP, et al. Surgical management of renal cell carcinoma with inferior vena cava tumor thrombus. Ann Thorac Surg 1997;63: [4] Rabbani F, Hakimian P, Reuter VE, Simmons R, Russo P. Renal vein or inferior vena caval extension in patients with renal cortical tumors: impact of tumor histology. J Urol 2004;171: [5] Neves RJ, Zincke H. Surgical treatment of renal cancer with vena cava extension. Br J Urol 1987;59: [6] Gallucci M, Borzomati D, Flammia G, et al. Liver harvesting surgical technique for the treatment of retro-hepatic caval thrombosis concomitant to renal cell carcinoma: perioperative and long-term results in 15 patients without mortality. Eur Urol 2004;45: [7] Lebret T, Bohin D, Richard F, Botto H. Tumeur rénale avec thrombus s étendant à la totalité de la lumière de la veine cave inférieure: indication chirurgicale, techniques et résultats. Prog Urol 1998;8: [8] Reissigl A, Janetschek G, Eberle J, et al. Renal cell carcinoma extending into the vena cava: surgical approach, technique and results. Br J Urol 1995;75: [9] Montie JE, Jackson CL, Cosgrove DM, Streem SB, Novick AC, Pontes JE. Resection of large inferior vena caval thrombi from renal cell carcinoma with the use of circulatory arrest. J Urol 1988;139:25 8. [10] Sweeney P, Wood CG, Pisters LL, et al. Surgical management of renal cell carcinoma associated with complex inferior vena caval thrombi. Urol Oncol 2003;21: [11] Belis JA, Levinson ME, Pae Jr WE. Complete radical nephrectomy and vena caval thrombectomy during circulatory arrest. J Urol 2000;163: [12] Braud F, Soulie M, Soula P, Tollon C, Pontonnier F, Plante P. L adénocarcinome rénal avec thrombus cave rétro-hépatique: place de la circulation extra-corporelle sur une série rétrospective de 10 cas. Prog Urol 1999;9: [13] Bensalah K, Guille F, Vincendeau S, et al. Facteurs pronostiques cliniques et anatomo-pathologiques des cancers du rein avec thrombus cave. Prog Urol 2004;14:160 6, discussion 165. [14] Kuczyk MA, Bokemeyer C, Kohn G, et al. Prognostic relevance of intracaval neoplastic extension for patients with renal cell cancer. Br J Urol 1997;80: [15] Tsuji Y, Goto A, Hara I, et al. Renal cell carcinoma with extension of tumor thrombus into the vena cava: surgical strategy and prognosis. J Vasc Surg 2001;33: [16] Libertino JA, Zinman L, Watkins Jr E. Long-term results of resection of renal cell cancer with extension into inferior vena cava. J Urol 1987;137:21 4. [17] Swierzewski DJ, Swierzewski MJ, Libertino JA. Radical nephrectomy in patients with renal cell carcinoma with venous, vena caval, and atrial extension. Am J Surg 1994;168: [18] Polascik TJ, Partin AW, Pound CR, Marshall FF. Frequent occurrence of metastatic disease in patients with renal cell carcinoma and intrahepatic or supradiaphragmatic intracaval extension treated with surgery: an outcome analysis. Urology 1998;52: [19] Kim HL, Zisman A, Han KR, Figlin RA, Belldegrun AS. Prognostic significance of venous thrombus in renal cell carcinoma. Are renal vein and inferior vena cava involvement different? J Urol 2004;171: [20] Moinzadeh A, Libertino JA. Prognostic significance of tumor thrombus level in patients with renal cell carci-

9 310 european urology 50 (2006) noma and venous tumor thrombus extension. Is all T3b the same? J Urol 2004;171: [21] Lang H, Lindner V, Letourneux H, Martin M, Saussine C, Jacqmin D. Prognostic value of microscopic venous invasion in renal cell carcinoma: long-term follow-up. Eur Urol 2004;46: [22] Bissada NK, Yakout HH, Babanouri A, et al. Long-term experience with management of renal cell carcinoma involving the inferior vena cava. Urology 2003;61: [23] Zisman A, Wieder JA, Pantuck AJ, et al. Renal cell carcinoma with tumor thrombus extension: biology, role of nephrectomy and response to immunotherapy. J Urol 2003;169: [24] Naitoh J, Kaplan A, Dorey F, Figlin R, Belldegrun A. Metastatic renal cell carcinoma with concurrent inferior vena caval invasion: long-term survival after combination therapy with radical nephrectomy, vena caval thrombectomy and postoperative immunotherapy. J Urol 1999;162: [25] Hofmann HS, Neef H, Krohe K, Andreev P, Silber RE. Prognostic factors and survival after pulmonary resection of metastatic renal cell carcinoma. Eur Urol 2005;48:77 81, discussion Editorial Comment Markus Kuczyk, Tübingen, Germany markus.kuczyk@med.uni-tuebingen.de The paper presented by Rigaud and coworkers deals with the clinical outcome of renal cell cancer patients subjected to the simultaneous removal of the primary tumour and an intracaval thrombosis. Although a multivariate statistical analysis that would have considered all available patient and tumour characteristics is not provided herein, exclusively the presence of nodal disease was demonstrated as in earlier studies to significantly predict the patients long-term survival after surgical treatment. Herewith, the present paper, as repeatedly done, addresses the question of whether major surgery is justified in the case of an extensively growing renal cell cancer presenting with an intracaval thrombosis. It has been demonstrated clearly before that the aforementioned surgical approach can induce an increased therapy-associated mortality and morbidity when compared with the simple resection of a tumour-bearing kidney. However, neither the presence nor the cranial extension of an intracaval thrombosis, but exclusively the presence of lymph node metastases, could be identified as having a significant impact on post-operative long-term survival. This observation is confirmed by Rigaud and co-workers. The clinical prognosis of patients with intracaval thrombosis in the absence of metastatic disease depends on the local extension of the primary tumour and, therefore, on the possibility of its complete resection with the achievement of a tumour-free margin status as well as the presence or absence of occult metastatic disease that cannot be visualized at the time of primary diagnosis. This selection bias affecting previously reported series should be recognized when the long-term survival rates diverging between different cohorts of patients are addressed. Consequently, the conclusion of the authors that the disease-free survival after treatment of renal cell cancer patients with intracaval thrombosis is poor should be handled with care because this observation might be true for the current series of patients, but it cannot, however, be transferred to the clinical efficacy of this surgical procedure in general.