PARENCHYMAL SPARING SURGERY FOR CENTRAL RENAL TUMORS IN PATIENTS WITH HEREDITARY RENAL CANCERS

0022-5347/04/1721-0049/0 THE JOURNAL OF UROLOGY® Copyright © 2004 by AMERICAN UROLOGICAL ASSOCIATION

Vol. 172, 49 –53, July 2004 Printed in U.S.A.

DOI: 10.1097/01.ju.0000130930.70356.28

PARENCHYMAL SPARING SURGERY FOR CENTRAL RENAL TUMORS IN PATIENTS WITH HEREDITARY RENAL CANCERS DARREL E. DRACHENBERG,* OTHON J. MENA,* PETER L. CHOYKE,* W. MARSTON LINEHAN* , AND McCLELLAN M. WALTHER* † From the Urologic Oncology Branch and Diagnostic Radiology Department, Clinical Center, National Institutes of Health, Bethesda, Maryland

ABSTRACT

Purpose: Nephron sparing surgery has become accepted surgical practice for removing of renal tumors. The resection of central lesions has been thought to be more surgically challenging than that of peripheral tumors. We analyzed our experience with renal preservation surgery in patients with small hereditary central renal tumors. Materials and Methods: From 1992 to 2000 we performed 116 partial nephrectomies with 44 kidneys (38%) demonstrating central renal masses. Central renal tumors were defined radiologically as those completely encircled by parenchyma or transgressing the interpapillary line on computerized tomography. We compared this group to a similar series of 67 patients with hereditary renal cancer with only peripheral based tumors. Results: Mean tumor size was 3.2 cm (range 1.5 to 7.5). Mean operative time was 352 minutes (range 70 to 830). Renal hypothermia and vascular clamping were used in 19 of 44 procedures (41%). Mean ischemic time was 55 minutes (range 16 to 143). Mean blood loss was 4.6 l (range 0.1 to 23). The complication rate was 23% (10 of 44 cases) and with 18% (8 of 44) directly related to surgical technique. The mean transfusion requirement was 6.7 U (range 0 to 32) and 12 of 44 procedures (27%) required no blood products. Mean preoperative and postoperative serum creatinine was 1.05 (range 0.6 to 1.8) and 1.08 mg/dl (range 0.6 to 2.1), respectively. Mean followup was 33.7 months. No metastasis developed during followup. Conclusions: Central renal tumors are a common manifestation of hereditary renal cell carcinoma. There was no statistical difference found between common operative parameters when central and peripheral nephron sparing surgeries were compared. However, mean operative blood loss and transfusion requirements were increased in the central tumor group. KEY WORDS: kidney; kidney neoplasms; carcinoma, renal cell; nephrectomy; Hippel-Lindau disease

Partial nephrectomy has become a well established accepted procedure for the management of renal masses.1, 2 Nephron sparing surgery assumes special importance in the treatment of patients with solitary kidney, hereditary syndromes and underlying renal insufficiency or the potential to have it because of associated comorbidity, such as diabetes mellitus or atherosclerosis. However, the usefulness of nephron sparing surgery in patients with sporadic clinical T1 disease and normal renal function with normal functioning contralateral renal moiety remains controversial.3–5 Patient populations presenting with renal masses at our institution have a host of hereditary syndromes, including von Hippel-Lindau disease (VHL), hereditary papillary renal cell carcinoma (HPRC), Birt-Hogg-Dube´ syndrome (BHD) and familial renal oncocytoma (FRO). Nephron sparing surgery is imperative in these patients given the genetic predisposition toward multifocal and bilateral disease. With this in mind management strategies involve conservative observation with close periodic followup until tumors attain a maximum diameter of 3 cm, at which time patients undergo resection of the index tumor, and all smaller ipsilateral tumors and cysts. We did not see tumors smaller than 3 cm become metastatic in our VHL patient population at a median followup of 5 years.6

It is common for us to see patients for second opinions regarding management after nephrectomy is proposed to manage central occurring lesions. It is generally thought that these centrally arising tumors (those completely encircled by renal parenchyma, such that they would be undetectable on visual inspection, or those involving the renal sinus) are more surgically demanding, have greater operative time, incur greater blood loss and may result in a poorer postoperative renal functional outcome. Two recent reports addressed the location and outcome of the nephron sparing surgical approach but to our knowledge a report of surgery for hereditary renal cancer syndromes with a central location is not available.7, 8 Since nephron sparing surgeries are a necessity in the majority of our patients and tumor multiplicity is frequently noted, we reviewed our single institution experience with nephron sparing surgery in this highly specific patient population. MATERIALS AND METHODS

Patients. Between 1992 and 2000 after obtaining informed consent we performed 116 partial nephrectomies with 44 (38%) demonstrating central renal masses in 32 patients harboring a total of 46 central tumors. A total of 67 consecutive patients/surgeries that involved only peripheral lesions and were performed in the same period were selected and used for comparison. We defined central renal tumors radiologically as those completely encircling parenchyma or transgressing the interpapillary line. Tumors meeting these criteria during the study period were included (figs. 1 and 2).

Accepted for publication February 27, 2004. * Financial interest and/or other relationship with the National Cancer Institute. † Correspondence: Urologic Oncology Branch, National Cancer Institute, National Institutes of Health, Building 10, 2B47, 10 Center Dr., Bethesda, Maryland 20892 (FAX: 301-402-0922). 49

50

PARENCHYMAL SPARING SURGERY FOR CENTRAL RENAL TUMORS

FIG. 1. CT demonstrates hereditary renal cancer syndrome, central renal tumor of left kidney completely encircled by renal parenchyma and multiple bilateral renal cystic disease.

FIG. 2. CT shows large left central tumor transgressing interpapillary line. Patient underwent nephron sparing surgery.

Tumor location was determined through a retrospective review of computerized tomography (CT) in all patients undergoing nephron sparing surgery at our institution during the study period. CT was reviewed by 3 independent investigators to ensure that the definition was met in all indexed cases. Tumor size was determined by caliper measurement of preoperative CT and it is reported as the maximal diameter of the lesion since tumor size prompted initiation surgical management. Tumor staging used the 1997 American Joint Committee on Cancer/UICC TNM classification system9 and histopathological classification followed the guidelines of Thoenes et al.10 Pathological grading was performed with the Fuhrman grading classification system.11 Operative data and postoperative parameters were obtained from a review of the operative notes, patient charts and our patient database. Surgery. The surgical techniques used in this patient cohort were well described in earlier reports and they are briefly summarized.12 The choice of incision was based on the history of surgery and associated procedures to be performed, such as contralateral renal and adrenal or pancreatic surgery. After initial dissection Gerota’s fascia was opened. Perinephric fat was mobilized but an effort was made to maintain its vascular supply to allow renal defect closure. If visualized, tumors were removed via sharp capsular incision and circumscription with enucleation by Penfield neurotomes, ensuring a margin of healthy parenchyma removal with the pseudo-encapsulated tumors. Hemostasis was ensured through a combination of hemostatic figure-of-8 poly-

glactin sutures to stop visibly bleeding vessels and the judicious use of thrombin saturated gelatin sponge packing for managing venous oozing. Cystic disease was excised by enucleation or unroofing with fulguration of the bases. Cyst walls were sent for pathological examination. Deeper lesions were sought by intraoperative color Doppler ultrasonography. An effort was made to remove surgically all potentially malignant solid and cystic lesions during the procedure with the realization that further disease was likely to occur in this patient population as the result of residual microscopic foci of disease rather than as true disease recurrence from positive margins. Therefore, margin status was not ascertained by frozen section. Deep tumors and cysts were ultrasonographically identified and surrounding vascular structures were noted, so that renal capsulotomy and incision were done in vessel sparing fashion. The decision to gain vascular control was based on the size and position of the lesion in the renal hilum. Ice slush was applied to the kidney and the artery was clamped with small bulldog vascular clamps, allowing the kidney 10 minutes of cooling time. If entrance to the collecting system was visualized or anticipated, the patient received methylene blue through intrapelvic injection. Any defects discovered were closed with 5-zero chromic catgut or polyglactin. Local papaverine and intravenous mannitol were given to prevent renal arterial spasm and minimize ischemic tubular injury. Realizing that further surgeries are anticipated in this patient cohort, renal hilar dissection was minimized to prevent future scarring and perinephric fat was laid over all defects to improve potentially future tissue planes as well as prevent adhesions and scarring to the abdominal wall and viscera. Urinary fistula was defined as persistent output of urine from flank drainage, necessitating further management with stenting, or persisting for greater than 2 weeks after surgery. Serum creatinine was determined preoperatively 1 day before the operation with the patient in a well hydrated state. Postoperative creatinine was determined serially at patient visits and the most recent creatinine value available was considered to reflect long-term postoperative renal function. Patients returned for serial postoperative visits, including repeat imaging of the abdomen and chest to evaluate recurrence, metastasis and the development of new tumors, a followup history and physical examination, and routine laboratory tests and differential renal scan as needed. Statistical analysis. A statistical comparison between our patients with a central lesion and a series of patients with hereditary renal cell cancer with only peripheral tumors was performed with standard statistical software (Instat, version 3.01, GraphPad Software, Inc., San Diego, California). Tests included Fisher’s exact test to compare complication rates and number of patients requiring renal clamping/ischemia, and the 2-tailed t test with the Welch correction to compare mean operative times, mean ischemic times, mean blood loss and mean transfusion requirements as well as mean preoperative and postoperative serum creatinine. RESULTS

Nephron sparing surgery for central renal tumors was performed in 17 men and 15 women with an average age of 32.9 years at surgery (range 18 to 66) (table 1). Mean followup was 33.7 months (range 1.5 to 101). All except 2 patients had multicentric ipsilateral renal tumors (central and peripheral) associated with the larger 3 cm or greater index tumor (indication for surgery). Two patients had synchronous ipsilateral central lesions found at surgery with the lesions transgressing the interpapillary line and completely encased by renal parenchyma. No patient had metastatic disease at surgery. Ten of 32 patients (31%) underwent 2 separate surgeries on the index kidney, 1 of 32 (3%) underwent 3 surgeries on the index kidney and the remaining 21 of

PARENCHYMAL SPARING SURGERY FOR CENTRAL RENAL TUMORS TABLE 1. Patient statistics No. pts No. men/women No. procedures No. multiple procedures No. disease: VHL HPRC Carcinoma FRO BHD Sporadic Mean mos followup (range)

Central

Peripheral

32 17/15 44 11

67 38/29 67 Not applicable

26 3 1 1 1 33.7 (1.5–101)

60 3 1 1 2 60.3 (3–139)

32 (66%) underwent a single procedure during the followup period for a total of 44 surgeries performed. The mean number of lesions (tumors and cysts) removed at surgery was 6.7 (range 1 to 25) with an average size of 1.7 cm (range 0.2 to 7.5) (table 2). Mean operative time was 352 minutes (range 70 to 830) with renal hypothermia and vascular clamping used in 19 of 44 procedures (41%). Mean ischemic time was 55 minutes (range 16 to 143) with an average blood loss of 4.6 (range 0.1 to 23). The complication rate was 23% (10 of 44 cases) with 18% (8 of 44) directly attributable to surgical technique, namely urinary fistula in 2 (4.5%), renal artery injuries in 2 (4.5%), and partial ureteral transection, postoperative bleeding, a perinephric abscess and an atrophic kidney in 1 each (2.25%). One renal artery injury involved an intimal tear with prolonged renal ischemia, resulting in nephrectomy. This patient had undergone previous contralateral nephrectomy for renal tumors and was rendered anephric. The remaining complications were not directly attributable to operative technique. The mean transfusion requirement was 6.7 U (range 0 to 32) and 12 of 44 procedures (27%) required no blood products. There was no extracorporeal renal surgery performed in any patient (table 3). Mean hospital stay was 7 days (range 5 days to 12 weeks). Clear cell pathology was present in all except 5 patients. It was illustrative of the syndromic pathology anticipated with VHL, HPRC, BHD and FRO (table 4). A total of 40 tumors were hilar in location and 6 were encased by renal parenchyma but did not transgress the interpapillary line (table 2). Except for a few large lesions that transgressed the interpapillary line and extended distal into a cortical location the majority of these tumors would not have been visible without suspicion on CT and intraoperative ultrasonographic localization. All tumors were pathological stage T1 lesions except in 1 patient with HPRC, in whom lesions were stage T2 (table 4). Two patients with VHL (4.5%) had stage T1 size lesions but a segmental renal vein tumor thrombus was present at operation. The tumor thrombus finding in 1 case was determined incidentally at pathological evaluation since it was removed en bloc with the 2.3 ⫻ 2.0 ⫻ 2.0 cm specimen, whereas the thrombus in the other case was discovered at surgery. This lesion was 5.0 ⫻ 4.0 ⫻ 3.0 cm. Formal segmental venotomy and thrombus extraction were performed after renal vascular control and hypothermia. Each lesion with tumor thrombi was Fuhrman nuclear grade II. Overall Fuhrman nuclear grade was I to III in 2 (4%), 40 (87%) and 4 (9%) tumors, respectively (table 4). Mean preoperative and postoperative serum creatinine TABLE 2. Tumor characteristics No. tumors: Transgressing interpapillary line Encircled by parenchyma Transgressing ⫹ encircled Av cm central tumor size (range) Av cm tumor size (range) Av No. tumors/pt (range)

46 34 6 6 3.2 (1.5–7.5) 1.7 (0.2–7.5) 6.7 (1–16)

51

was 1.05 (range 0.6 to 1.8) and 1.08 mg/dl (range 0.6 to 2.1), respectively (table 3). One patient with a solitary kidney required postoperative hemodialysis after a renal artery intimal tear and nephrectomy. No patient had metastatic disease and all were alive at the end of followup. No statement can be made regarding local recurrence rates in this population due to the multicenticity and bilaterality of disease as well as known microscopic lesions in the majority of patients with hereditary forms of renal cancer.12–14 A comparison of our central tumor cohort to a cohort of 67 patients with hereditary renal cancer who had only peripheral renal tumors demonstrated no statistical difference in mean operative time (352 vs 335 minutes), the percent of patients requiring renal ischemia (43% vs 40%), mean ischemic time (55 vs 55 minutes) and the complication rate (10 of 44 or 23% vs 16 of 67 or 24%). However, the mean operative blood loss (2.11 vs 4.61) and mean transfusion requirements (2.5 vs 6.7 U) were greater and statistically significant (p ⫽ 0.02 and 0.008, respectively). DISCUSSION

The recent development and increased use of renal imaging technologies such as CT and ultrasonography has led to a marked increase in the diagnosis of small, incidental asymptomatic renal masses and, as a result, the impetus for nephron sparing surgical strategy has grown.15 As mentioned, a central tumor location has been thought to be more surgically demanding, have greater operative time and incur greater blood loss, and it may result in a poorer postoperative renal functional outcome. In 2 recent studies groups reviewing their experience with central renal masses and partial nephrectomy used different definitions to describe what represents a central lesion.7, 8 These studies had an average central tumor size (2.5 and 3.45 cm, respectively) that was comparable to the average size in our group (3.2 cm). Hafez et al defined central as “extending centrally into the kidney beyond the renal medulla and into the renal sinus.”7 Black et al defined central as “lesions completely surrounded by renal parenchyma,” making no note of location and proximity to the renal sinus.8 We believe that the 2 definitions are important and should be used to give a more complete anatomical description of centrality. In the study of Hafez et al the 27 patients with central tumors tended to be selected for radical nephrectomy with higher frequency than those with peripheral tumors with only 54% of central tumors and a greater 61% of peripheral tumors undergoing a nephron sparing approach.7 However, because of the few patients in the central tumor group and, therefore, subsequent under powering, this only trended toward statistical significance (p ⫽ 0.1). It is important to note that this tendency is consistent with practice patterns that are in use for the management of central tumors presenting to a majority of urologists. More central lesions are de-selected from a nephron sparing procedure because of the widely believed concept that central tumors are not amenable to renal preserving surgical approaches due to technical challenges related to tumor location. Hafez et al found that renal parenchymal sparing surgery in the central tumor group was associated with increased ischemic time and an increased incidence of collecting system entry compared with more peripherally based lesions. They concluded that in the absence of other outcome differences nephron sparing surgery for central tumors is technically more demanding but equally efficacious when comparing complications, recurrence rates and disease-free survival. We believe that most tumors in patients with hereditary renal cancers are low grade and well encapsulated and, therefore, they may behave with lower malignant potential.12 In fact, none of our patients had metastatic disease during followup. Although some patients underwent more than 1

52

PARENCHYMAL SPARING SURGERY FOR CENTRAL RENAL TUMORS TABLE 3. Operative statistics Central No. tumors Mean mins operative time (range) % Renal hypothermia ⫹ vascular clamping (No./total No.) Mean mins ischemic time (range) Mean blood loss (range) No. complications: Urinary fistula Renal artery injury* Partial ureteral transection Postop bleeding Perinephric abscess Atrophic kidney Dilutional coagulopathy Air embolism

Total No. (%) Mean transfusion requirement (range) Mean mg/dl serum creatinine (range): Preop Postop * Resulted in nephrectomy in 1 patient.

(70–830) (19/44) (16–143) (0.1–23)

Peripheral

p Value

67 335 (106–690) 40 (25/67) 55 (24–120) 2.1 (0.15–22)

0.60 0.56 1 0.02

2 2 1 1 1 1 1 1 10 (23) 6.7 (0–32)

16 (24) 2.5 (0–34)

0.98 0.008

1.05 (0.6–1.8) 1.08 (0.6–2.1)

TABLE 4. Pathology No. Stage: T1 T2 T3b Cell type: Clear Other Grade: I II III IV

44 352 43 55 4.6

43 1 2 39 7 2 40 4 0

ipsilateral surgery for tumor enucleation, this most likely reflected disease multicentricity rather than true tumor bed recurrence. Interestingly in our cohort 2 patients (4.5%) harbored occult venous tumor thrombus, which was discovered incidentally at the time of the procedure. Each lesion was TNM pathological stage T1 in size and had no overt evidence of venous involvement on preoperative imaging. These patients currently have functioning kidneys and no evidence of recurrence or metastatic disease at a followup of 76 and 38 months, respectively. To our knowledge this is the first detailed description in the literature demonstrating tumor thrombus with parenchymal sparing surgery. Tumor thrombus associated with these small lesions that are thought to be amenable to a nephron sparing surgical approach may be more commonly found in centrally located small lesions. In the previous published review of our experience with parenchymal sparing surgery no peripheral lesion alone resulted in vascular extension and the 2 patients with T3b lesions harbored central tumors.16 The current TNM staging classification refers to T3b lesions as those that are found to invade the major veins or vena cava and it does not define what represents a major vein. Patient outcome with microscopic invasion of small venules or involvement in a segmental venous tributary has not been well defined. The recent analysis of Lang et al, who reviewed microscopic venous invasion (MVI) in 74 patients who underwent radical nephrectomy, revealed that MVI significantly increased the metastatic progression rate and decreased survival.17 However, MVI was not found to be an independent or significant prognostic factor apart from TNM stage (1992 classification) and Fuhrman nuclear grade. Previous studies also support this relationship.18 –20 These series showed that the MVI incidence in pT1 and pT2 tumors (1992 classification) was 7% to 25%. With the growing popularity and success of renal preservation surgery, and the

finding of a 7% to 25% incidence of MVI in patients with localized renal cell carcinoma it is highly likely that microscopic and macroscopic small venous invasion will become an increasingly frequent occurrence in patients considered for nephron sparing approaches. Hopefully further recognition and experience with MVI and nephron sparing surgery will help define its importance as a prognosticator and help determine future management and postoperative followup schemas. When one compares hereditary renal cancers from our database with tumors of only peripheral location managed by nephron sparing surgery, there was no statistically significant difference between operative parameters except greater mean operative blood loss and mean transfusion requirements. This is likely the result of the proximity of central lesions to the hilar vascular supply, and its many large branches and tributaries. However, despite the increase in blood loss there was no increase in average operative and ischemic times. The overall complication rate was also comparable (table 4). Similar operative and ischemic times, and complication rates do not seem to indicate greater technical challenge in this patient population. It is likely that the advent of new technologies that may decrease operative blood loss and facilitate nephron sparing surgeries, such as the ultrasonic scalpel, bipolar coagulation and argon beam coagulator, would impact these complications in a positive manner. They should be currently considered when performing such technically demanding procedures. These devices were not used in our patient population and many were not available at the time of our study. There was no statistically significant difference between mean preoperative and postoperative serum creatinine. Overall the excellent renal functional outcome reinforces the plausibility of considering parenchymal sparing surgery for all T1 and T2 lesions despite a hilar location. The overall complication rate was 23% with 18% directly attributable to surgical technique. The complication rate in our study is comparable to that in the 2 previously published series involving central tumors with observed complication rates of 21% in that of Hafez7 et al and 15% in that of Black et al.8 However, notably the urinary fistula rate was 10% in the patients of Hafez et al despite lesions 4 cm or less in size7 and in those of Black et al,8 whereas we had only a 4.5% fistula occurrence rate with no such limit on surgical sizes to less than 4 cm. It is likely that as technique improves and further experience is gained with this procedure, the complication rate will decrease to that seen with more peripheral based tumors.

PARENCHYMAL SPARING SURGERY FOR CENTRAL RENAL TUMORS CONCLUSIONS

Our study demonstrates that in patients with hereditary renal cancer who are at risk for multiple bilateral metachronous tumors nephron sparing surgery is feasible and desirable to preserve renal function and prevent the significant morbidity and mortality of dialysis, and transplant related immunosuppression. In this patient population enucleation techniques seem exceedingly well suited to minimize the loss of normal, functioning parenchyma. Increased surgical difficulty was not demonstrated by an increase in operative or ischemic time and the complication rate is comparable to that in other series. This fact is emphasized when one compares our procedures involving central renal masses to similar procedures at our institution involving only peripheral tumors. There was no increase in operative time, ischemic time, the percent of procedures requiring renal ischemia and the complication rate. Therefore, we recommend nephron sparing approaches in these patients regardless of tumor location and find it technically feasible and efficacious. REFERENCES

1. Novick, A. C., Zincke, H., Neves, R. J. and Topley, H. M.: Surgical enucleation for renal cell carcinoma. J Urol, 135: 235, 1986 2. Steinbach, F., Sto¨ ckle, M., Mu¨ ller, S. C., Thu¨ roff, J. W., Melchior, S. W., Stein, R. et al: Conservative surgery of renal cell tumors in 140 patients: 21 years of experience. J Urol, 148: 24, 1992 3. Wunderlich, H., Reichelt, O., Schumann, S., Schlichter, A., Kosmehl, H., Werner, W. et al: Nephron sparing surgery for renal cell carcinoma 4 cm. or less in diameter: indicated or under treated? J Urol, 159: 1465, 1998 4. Ljungberg, B., Alamdari, F. I., Holmberg, G., Granfors, T. and Duchek, M.: Radical nephrectomy is still preferable in the treatment of localized renal cell carcinoma. A long-term follow-up study. Eur Urol, 33: 79, 1998 5. Lee, C. T., Katz, J., Shi, W., Thaler, H. T., Reuter, V. E. and Russo, P.: Management of renal tumors 4 cm. or less in a contemporary cohort. J Urol, 163: 730, 2000 6. Walther, M. M., Choyke, P. L., Glenn, G., Lyne, J. C., Rayford, W., Venzon, D. et al: Renal cancer in families with hereditary renal cancer: prospective analysis of a tumor size threshold for renal parenchymal sparing surgery. J Urol, 161: 1475, 1999 7. Hafez, K. S., Novick, A. C. and Butler, B. P.: Management of small solitary unilateral renal cell carcinomas: impact of cen-

53

tral versus peripheral tumor location. J Urol, 159: 1156, 1998 8. Black, P., Filipas, D., Fichtner, J., Hohenfellner, R. and Thu¨ roff, J. W.: Nephron sparing surgery for central renal tumors: experience with 33 cases. J Urol, 163: 737, 2000 9. Sobin, L. H. and Wittekind, C. H.: TNM Classification of Malignant Tumors, 5th ed. New York: Wiley-Liss, 1997 10. Thoenes, W., Storkel, S. and Rumpelt, H. J.: Histopathology and classification of renal cell tumors (adenomas, oncocytomas and carcinomas). The basic cytological and histopathological elements and their use for diagnostics. Pathol Res Pract, 181: 125, 1986 11. Fuhrman, S. A., Lasky, L. C. and Limas, C.: Prognostic significance of morphologic parameters in renal cell carcinoma. Am J Surg Pathol, 6: 655, 1982 12. Walther, M. M., Thompson, N. and Linehan, W.: Enucleation procedures in patients with multiple hereditary renal tumors. World J Urol, 13: 248, 1995 13. Zbar, B., Glenn, G., Lubensky, I., Choyke, P., Walther, M. M., Magnusson, G. et al: Hereditary papillary renal cell carcinoma: clinical studies in 10 families. J Urol, 153: 907, 1995 14. Toro, J. R., Glenn, G., Duray, P., Darling, T., Weirich, G., Zbar, B. et al: Birt-Hogg-Dube syndrome: a novel marker of kidney neoplasia. Arch Dermatol, 135: 1195, 1999 15. Smith, S. J., Bosniak, M. A., Megibow, A. J., Hulnick, D. H., Horii, S. C. and Raghavendra, B. N.: Renal cell carcinoma: earlier discovery and increased detection. Radiology, 170: 699, 1989 16. Herring, J. C., Enquist, E. G., Chernoff, A., Linehan, W. M., Choyke, P. L. and Walther, M. M.: Parenchymal sparing surgery in patients with hereditary renal cell carcinoma: 10 experiences. J Urol, 165: 777, 2001 17. Lang, H., Lindner, V., Saussine, C., Havel, D., Faure, R. and Jacqmin, D.: Microscopic venous invasion: a prognostic factor in renal cell carcinoma. Eur Urol, 38: 600, 2000 18. Hermanek, P. and Schrott, K. M.: Evaluation of the new tumor, nodes and metastases classification of renal cell carcinoma. J Urol, 144: 238, 1990 19. Mrstik, Ch., Salamon, J., Weber, R. and Sto¨ germayer, F.: Microscopic venous infiltration as predictor of relapse in renal cell carcinoma. J Urol, 148: 271, 1992 20. Van Poppel, H., Vandendriessche, H., Boel, K., Merterns, V., Goethuys, H., Haustermans, K. et al: Microscopic vascular invasion is the most relevant prognosticator after radical nephrectomy for clinically nonmetastatic renal cell carcinoma. J Urol, 158: 45, 1997