- Email: [email protected]
Hand-assisted laparoscopic nephroureterectomy for the treatment of transitional cell carcinoma of the upper urinary tract
ADULT UROLOGY
HAND-ASSISTED LAPAROSCOPIC NEPHROURETERECTOMY FOR THE TREATMENT OF TRANSITIONAL CELL CARCINOMA OF THE UPPER URINARY TRACT MICHAEL D. STIFELMAN, R. ERNEST SOSA, ADELINO ANDRADE, ARTHUR TARANTINO, AND STEVEN J. SHICHMAN
ABSTRACT Objectives. Nephroureterectomy with removal of the bladder cuff is the standard of care for patients with upper tract transitional cell carcinoma. Historically, it has been performed using two separate incisions or one large incision extending from the lateral flank to the symphysis pubis. We describe an alternative technique using endoscopic and hand-assisted laparoscopic techniques and present our experience. Methods. During the past 18 months, 22 patients at two institutions underwent hand-assisted laparoscopic nephroureterectomy. In 19 patients, the distal ureter and bladder cuff were managed endoscopically. In 3 patients, the distal ureter and the bladder cuff were removed by an extravesical, laparoscopic technique. The intraoperative parameters assessed included operative time, estimated blood loss, specimen weight, surgical margin status, pathologic grade and stage, and acute complications. Postoperative endpoints included the time to sustained fluid intake, parenteral narcotic requirement (milliequivalents of morphine sulfate), oral narcotic requirement (number of tablets), length of stay, time until return to normal activity, and rate of tumor recurrence. Results. The average age of our patient population was 65 years (range 42 to 86), 10 patients were men and 12 were women, and the average American Society of Anesthesiologists classification was 2.2. All but 2 patients had their specimens removed en bloc. No intraoperative complications occurred. The average operative time was 272 minutes (range 190 to 440), and the average blood loss was 180 mL (range 50 to 400); no patient required a transfusion. The mean specimen weight was 457 g (range 190 to 1420). All 22 patients had negative surgical margins. Postoperatively, the time to sustained fluid intake averaged 2.1 days (range 1 to 7), the mean parenteral narcotic requirement was 55 mEq (range 12 to 107.8) of morphine sulfate, the mean oral narcotic requirement was 5.8 tablets (range 1 to 14), and the average length of stay was 4.1 days (range 3 to 14). One patient developed thrombophlebitis of the right external jugular vein from a central line and required 2 weeks of intravenous antibiotics. The mean time to return to normal activity was 19 days; the mean follow-up was 13 months. Six patients had disease recurrence: four low-grade, low-stage bladder tumors and two metastatic tumors. All patients were alive at 18 months. Conclusions. Hand-assisted laparoscopic nephroureterectomy with endoscopic management of the bladder cuff is a viable and efficacious alternative to open nephroureterectomy. The technique allows the surgeon to perform an en bloc resection of the kidney, ureter, and bladder cuff without compromising oncologic principles. Patients benefit from a decrease in pain and hospital stay and quicker convalescence. Longer follow-up and comparative studies to standard open techniques are underway. UROLOGY 56: 741–747, 2000. © 2000, Elsevier Science Inc.
A
pproximately 5300 patients within the United States have been diagnosed with transitional cell carcinoma involving the upper urinary tract in
1999, and 1600 patients can expect to die of this disease.1 The traditional treatment for upper tract urothelial tumors is total nephroureterectomy
From the Department of Urology, James Buchanan Brady Foundation, New York Presbyterian Hospital, Weill Medical College, Cornell University, New York, New York; and Connecticut Surgical Group, PC, Hartford Hospital, Hartford, Connecticut
Reprint requests: Michael D. Stifelman, M.D., 540 First Avenue, Suite 1012, New York, NY 10016 Submitted: February 28, 2000, accepted (with revisions): June 20, 2000
© 2000, ELSEVIER SCIENCE INC. ALL RIGHTS RESERVED
0090-4295/00/$20.00 PII S0090-4295(00)00751-2 741
with excision of a cuff of bladder.2 Extirpation can be achieved using one large incision extending from the patient’s flank to the symphysis pubis. Alternatively, two separate incisions may be used at the flank and lower abdominal region. This latter approach allows the surgeon better exposure for excision of the bladder cuff. In either scenario, the kidney, ureter, and bladder are removed en bloc to prevent tumor spillage. Minimally invasive antegrade and retrograde endoscopic approaches have been reported for the treatment of this disease. The results have been varied, and disadvantages include a high ipsilateral recurrence rate, the need for multiple procedures, and the requirement for accurate preoperative confirmation that the tumor is truly low stage and low grade.3– 6 High-grade tumors have significantly increased recurrence rates when treated with local excision alone compared with nephroureterectomy.7 Thus, ureteroscopic and/or percutaneous treatments should be reserved for patients with a solitary kidney, bilateral disease, or decreased renal reserve, and low-grade and low-stage tumors. The advent of hand-assisted laparoscopy (HAL) has provided a new alternative for treating patients with upper tract urothelial tumors using techniques of minimal access. The potential benefits of HAL include a decreased hospital stay, lessened analgesic requirements, and a faster recovery. HAL facilitates en bloc removal of the total specimen, maintaining the oncologic principles used in open surgery. We review our technique and report our results of nephroureterectomy using HAL with endoscopic management of the bladder cuff. MATERIAL AND METHODS PREPARATION AND POSITION A clear liquid diet is administered the day before surgery along with a mechanical bowel preparation. The patient fasts for 8 hours before surgery. Preoperative antibiotics are infused in the operating room. General anesthesia with endotracheal intubation is used in all cases. A nasogastric tube and a urethral catheter are used to decompress the stomach and bladder, respectively. Venous insufflation stockings are used to prevent lower extremity venous stasis. The patient is initially placed in the dorsolithotomy position for excision of the bladder cuff. Once complete, the patient is prepared again and draped in the modified lateral decubitus position at a 45° angle. The operating table is slightly flexed. The patient is secured to the operating table with 3-in. tape to permit side-to-side rotation for intraoperative exposure.
EQUIPMENT The equipment for endoscopic excision of the bladder cuff includes a resectoscope with a 30° lens, Collings’ knife, 5-mm trocar, 5-mm laparoscopic grasping instrument, ureteral stent, and a 2-0 chromic Surgi-Tie (U.S. Surgical, Norwalk, Conn). A 24F resectoscope equipped with a Collings’ knife is used for incisional release of the distal ureter and bladder cuff. The equipment required for HAL nephroureterectomy in742
FIGURE 1. Endoscopic management of ureter and bladder cuff. (a) Cystoscopy and placement of trocar and ureteral stent, (b) outlining incision, (c) bladder cuff and ureteral dissection, and (d) securing of distal ureter with tie.
cludes a hand device (Pneumosleeve, Dexterity, Atlanta, Ga; Handport, Smith and Nephew, Andover, Mass; or Intromit, Applied Medical, Rancho Santa Margarita, Calif), two or three 10 to 12-mm trocars, a 30° laparoscope, Maryland dissector, right-angle dissectors, endoscopic shears (U.S. Surgical), Harmonic scalpel (Ethicon, Sommers, NJ), bipolar cautery, small and large clips, an in-line endoscopic stapling device, and a laparoscopic suction/irrigating device.
ENDOSCOPIC MANAGEMENT OF BLADDER CUFF The bladder is inspected cystoscopically to ensure that no urothelial tumors are present. A 5F open-ended ureteral stent is inserted into the ureter on the side of the tumor. With the bladder distended, a 5-mm trocar is percutaneously inserted into the bladder dome (Fig. 1a). The valve on the trocar is left open and placed to allow gravity drainage. This provides a continuous flow system and increased visibility. The Collings’ knife is used on pure cut to score a 2-cm cuff of bladder mucosa about the ureteral orifice (Fig. 1b). A laparoscopic grasper is introduced through the trocar to hold the intravesical ureter during its dissection (Fig. 1c). The ureter is partially mobilized in a circumferential manner. The 2-0 chromic Surgi-Tie is introduced through the trocar, the stent is removed, the distal ureter is engaged, and the snare is tightened, closing off the ureteral lumen. The excess suture is cut and the distal ureter grasped again to continue dissection about the ureter until the extravesical space is reached and the ureter is UROLOGY 56 (5), 2000
through the hand-device site. A closed system drain is placed in the vicinity of the cystotomy and brought out through a stab wound. The renal vascular stumps, adrenal gland, and operative site are inspected for bleeding at low insufflation pressures. The trocar sites are inspected and closed. The handdevice site is closed last. Postoperatively, the patient is left with two trocar scars and a 6.5-cm hand-device scar (Fig. 3). For right nephroureterectomy, the right colon and hepatic flexure are mobilized medially, and the right lobe of the liver is released from the body sidewall. The coronary ligaments are incised to expose the upper renal pole and the vena cava. The duodenum is mobilized medially by a Kocher maneuver to expose the anterior aspect of the renal hilum. In a similar fashion to the left side, the artery is divided posteriorly, an in-line stapler divides the renal vein, and the kidney is mobilized within Gerota’s fascia. The ipsilateral ureter is handled as described for the left side. Of note, in the first 3 patients, the ureter and bladder cuff were taken by an extravesical laparoscopic approach. For this, the ureter was mobilized to the bladder. A clip was placed on the ureter to prevent the spillage of urine before the bladder was open. With slight traction on the ureter, the detrusor muscle was opened with electrocautery and blunt dissection to expose the bladder mucosa. The bladder cuff was incised and removed together with the distal ureter. The opening in the bladder was closed with 2-0 chromic sutures tied intracorporeally with the nondominant hand.
INTRAOPERATIVE PARAMETERS FIGURE 2. (a) Left trocar configuration and (b) right trocar configuration.
totally freed (Fig. 1d). A 20F Foley catheter is placed to decompress the bladder.
NEPHROURETERECTOMY The technique of performing HAL nephrectomy has been previously described and can easily be expanded to include nephroureterectomy.8,9 In brief, after excision of the bladder cuff and freeing of the distal ureter, the patient is placed in a 60°, lateral decubitus position with slight operating room table flexion to hyperextend the ipsilateral flank. The skin is cleansed with a Betadine preparation from the nipples to the groin. The operative field is widely draped to allow room for the hand device. The placement of the hand device and trocars for left and right-side tumors is outlined in Figure 2. For left-side nephrectomy, the colon is reflected medially from the iliac vessels to the splenic flexure. The dissection is carried in a cephalad direction, freeing the lateral splenic attachments to the level of the gastric fundus. The colon, spleen, and pancreas are mobilized en bloc, leaving the anterior face of Gerota’s fascia exposed. The aorta and renal artery are palpated. The investing tissues over the hilum are divided, and the renal vein is exposed and dissected free anteriorly. The renal hilum is identified posteriorly, and the renal artery is dissected free, clipped, and divided. The renal vein is divided using an in-line stapler. The kidney is dissected free, keeping Gerota’s fascia on, and leaving the adrenal gland behind. To free the ureter and remain optically correct, the video monitors are moved to the foot of the operating table. The ureter is mobilized to the level of the bladder using a combination of blunt dissection, clips, and electrocautery. The final adventitial attachments to the bladder are identified by gentle traction on the ureter and divided between clips. The chromic tie identifies the distal ureter and bladder cuff. The kidney, ureter, and bladder cuff are retrieved en bloc and intact UROLOGY 56 (5), 2000
The charts of all patients undergoing nephroureterectomy within the past 18 months were reviewed at our two institutions. The demographic data included age, sex, and American Society of Anesthesiologists classification. The intraoperative time was measured from the start of the endoscopic bladder intervention to the completion of the skin closure after removal of the specimen. The intraoperative time included the interval required to prepare, drape, and position the patient. Other intraoperative parameters included the estimated blood loss, the need for transfusion, complications, the specimen weight in grams, and an assessment of the surgical margins.
POSTOPERATIVE PARAMETERS The endpoints included the time to the first oral intake, parenteral pain medication requirement, oral pain medication requirement, and length of hospital stay. The time until the first oral intake was measured from the time the patient exited the operating room to the resumption of sustained fluid intake. The amount of postoperative parenteral analgesia was measured in milliequivalents of morphine sulfate and was defined as (milligrams morphine sulfate) ⫹ (milligrams meperidine)/10 ⫹ (milligrams hydromorphine) ⫻ 10 ⫹ (milligrams ketorolac)/6. The number of oral narcotic tablets ingested during the patient’s hospital stay was recorded. Patients were discharged when they were afebrile, ambulating, tolerating a regular diet, and a return of bowel function had been confirmed. All patients went home with Foley catheter drainage. On postoperative day 10, a cystogram was performed before removing the catheter. Using a retrospective chart review method and telephone interviews, we determined the time to the resumption of normal activities.
RESULTS Since May 1998 we have performed 22 HAL nephroureterectomies using the described technique. All patients had a preoperative diagnosis of transitional cell carcinoma of the upper urinary tract confirmed by either ureteroscopic biopsy and/or pos743
FIGURE 3. Patient with two trocar scars (white arrows) and 6.5-cm hand-device scar (black arrow).
itive cytologic findings with a filling defect. The metastatic workup was negative for all patients. The average age of our patient population was 65 years (range 42 to 68). Ten patients were men and 12 women. In terms of overall health and patient comorbidity, the average American Society of Anesthesiologists classification was 2.2. No intraoperative complications occurred in this series. Two specimens were not removed en bloc. In one of the earliest patients, ureteral thinning occurred during dissection of the intravesical ureter. Rather than risk inadvertent disruption, the lower ureter was clipped and divided. At the end of the laparoscopic procedure, the patient was placed into the dorsolithotomy position and the most distal aspect of the ureter with the bladder cuff was removed cystoscopically. In the second patient, the distal ureter was taken extravesically. The mucosal margin was positive for atypia. The Handport incision was extended and a larger cuff of bladder resected to achieve a negative margin. The average operative time, including repositioning, was 272 minutes (range 190 to 440). The estimated blood loss was minimal, with an average blood loss of 180 mL (range 50 to 400). No patient required a transfusion. The mean specimen weight 744
was 457 g (range 144 to 1420), and the mean specimen size volume (length ⫻ height ⫻ width) was 780 mL (range 205 to 2112). Pathologically, 3 patients had low-grade tumors (grade I, I-II), 10 had medium-grade tumors (grade II), and 9 patients had high-grade tumors (grade III). All tumors were transitional cell carcinoma. Five lesions were Stage Ta, 8 were T1, 2 were T2, and 7 were T3 lesions; no T4 lesions were found. No patient had a positive margin on the final pathologic specimen. In the postoperative period, the patients were started on clear fluids the day after surgery. The time to sustained fluid intake without nausea or vomiting averaged 2.1 days (range 1 to 7). Patients were either given patient-controlled anesthesia or prescribed intravenous narcotics for the first 24 to 48 hours after surgery. The total parenteral narcotic requirement averaged 55 mEq (range 12 to 107.8) morphine sulfate. Once patients were tolerating clear liquids, they were converted to oral narcotics for pain management. The oral narcotic requirement averaged 5.8 tablets (range 1 to 14) during the patients’ hospital stay. The average length of stay was 4.1 days (range 3 to 14). One patient developed thrombophlebitis in the right external jugular vein at the site of a preoperative UROLOGY 56 (5), 2000
TABLE I. Results of laparoscopic and hand-assisted laparoscopic nephroureterectomy series
Investigators Salomon et al.14 Hsu et al.16 McDougall et al.13 Keeley and Tolley15 Wolf et al.11 Current series
Patients (n) 4 11 10 22 2 22
Approach Laparoscopic retroperitoneal Laparoscopic retroperitoneal Laparoscopic transabdominal Laparoscopic transabdominal HAL HAL
Mean Operating Time (min)
Blood Loss (mL)
Hospital Stay (days)
220
220
5.7
210
150
498
233
156 345 272
NA* 625 180
Pain Medication (mg MSO4)
Return to Normal Activity (days)
1
NR
2
14
14
5
16
19.6
5.5
NR
NR
3.1† 4.1
48.3† 55
14† 19
KEY: MSO4 ⫽ morphine sulfate; NR ⫽not reported; NA ⫽ not applicable (see footnote); HAL ⫽ hand-assisted laparoscopy. * Blood loss was quantified by transfusion requirement; the mean transfusion requirement was 0.3 U per patient (versus 0.9 U per patient in the open group). † Data for the complete series. It included 11 hand-assisted nephrectomies.
central line placement and required 2 weeks of intravenous antibiotics. The mean time to a return to normal activity was 19 days; the mean follow-up was 13 months. Six patients had disease recurrence. Four patients had low-grade, low-stage bladder tumors, not involving the resection site. Two patients with grade III, pT3 tumors manifested metastatic lesions. All patients were alive at 18 months. COMMENT Several published studies have addressed the merits of HAL nephrectomy for both benign and malignant disease.8 –11 Only two published reports are available of HAL nephroureterectomy.11,12 The total number of patients in both these studies was three. Our initial series of 22 patients who underwent HAL nephroureterectomy revealed very encouraging short-term results. The technique provides an effective and reproducible minimally invasive method for removing the kidney, ureter, and bladder cuff en bloc. Removing the specimen en bloc potentially decreases the risk of tumor spillage and provides for accurate staging, similar to an open operation. We reviewed and summarized the published data on laparoscopic and HAL nephroureterectomy to compare previous data with our experience (Table I). Our average operative time of 272 minutes was similar to other laparoscopic nephroureterectomy series. McDougall et al.13 reported on their initial 10 patients who underwent laparoscopic nephroureterectomy, with an average operative time of 8.3 hours. Salomon et al.14 published their experience on retroperitoneoscopic nephroureterectomy with iliac excision to manage the bladder cuff for renal pelvic tumors. Their opUROLOGY 56 (5), 2000
erative time averaged 3.7 hours. Keeley et al.15 reported on 22 laparoscopic nephroureterectomies, with a rapid mean operative time of 2.6 hours. In comparison, our operative time of 4.5 hours was comparable to the other series.14,16 The blood loss in our series was also similar to that in other laparoscopic series.14 –16 The HAL technique does not compromise the ability to remove an intact specimen. Intact tumor extirpation proved to be independent of specimen size. This is illustrated by specimen weights of up to 1200 g and a conversion rate of zero. In the laparoscopic series of Keeley and Tolley,15 the conversion rate for 22 patients was 13%. The postoperative narcotic requirements, time to oral intake, and length of hospital stay for HAL and laparoscopic nephroureterectomy are summarized in Table I. The advantages of HAL and laparoscopy compared with open surgery are realized with these benefits. The shorter time for convalescence associated with these minimally invasive techniques not only allows patients to resume their normal activities sooner, but also reflects the smaller amount of trauma sustained by the patient. Kloosterman et al.17 compared laparoscopic cholecystectomy with open cholecystectomy in relation to the amount of trauma induced and immunologic compromise sustained by both procedures. Using the white blood cell count, interleukin-6 assay, HLA-DR expression, and phytohemagglutinin skin testing, they quantified the acute phase response and T-cell mediated immunity postoperatively for these two procedures. They concluded that laparoscopy was associated with a decreased acute phase response (trauma) and preservation of the patient’s cell-mediated immunity compared with open surgery.17 It will be of interest to com745
pare HAL and/or laparoscopic nephroureterectomy with open nephroureterectomy in this light. A similar decrease in the acute phase response and preservation of cell-mediated immunity would lead one to ponder whether laparoscopy could attenuate the biologic potential of a tumor. More research and longer follow-up are necessary to look at these issues. Endoscopic management of the bladder cuff has been described by other investigators.13,15,18 Our technique has some slight modifications. However, the general principles remain the same. These include excision of the bladder cuff with mobilization of the ureteral meatus and intramural ureter using a resectoscope with a Collings’ knife. The ureter is tied to complete occlusion before full mobilization to prevent spillage of ipsilateral urine into the retroperitoneum. A Foley catheter is left in place for 10 to 14 days to allow the defect to heal. A cystogram is performed to confirm a lack of extravasation before removing the catheter. One concern raised by using this technique is the risk of tumor seeding and implantation. Two cases have been reported of presumed tumor seeding after endoscopic incision of the ureteral meatus and bladder cuff.19,20 Whether the recurrence was due to seeding during the excision of the distal ureter remains unclear. Palou et al.21 reported on 31 patients who underwent endoscopic excision of the distal ureter before nephroureterectomy. Most of the patients had high-grade tumors with invasion, and none had tumor relapse.21 Certainly, careful follow-up and further investigation is warranted. Several potential advantages and disadvantages to the endoscopic approach for managing the distal ureter and bladder cuff exist. The transurethral technique allows the intraoperative hand to remain useful and assist in the dissection of the distal ureter. Less laparoscopic dissection of the distal ureter and bladder wall is necessary. For an open approach to the distal ureter, an incision larger than 7.5 cm and significant retraction are required to gain adequate exposure. The larger incision and added retraction may increase the patient’s postoperative pain and discomfort. However, we acknowledge that the selection of techniques is largely a matter of surgeon preference. Moreover, the open approach to the bladder cuff and distal ureter should be specifically considered in cases in which the patient has had a recent history of bladder tumors. Moreover, because of the discovery of bladder diverticula on follow-up cystoscopy and the long catheterization time after endoscopic management, we have begun to evaluate alternate ways of dealing with the distal ureter and bladder cuff. Finally, why did we choose to perform this pro746
cedure using HAL versus pure laparoscopy? One of the major objectives in performing a nephroureterectomy for transitional cell carcinoma is to remove the specimen intact. Therefore, morcellation of the tumor is not an option. Whatever technique is selected, an incision must be made large enough to remove the specimen en bloc. Our incision in the lower midline or right lower quadrant is equal in length to those created to remove the specimen after laparoscopy.13,14 Having one’s hand in the operative field allows for tactile sensation, a threedimensional spatial orientation, tissue palpation, blunt dissection, and retraction. The hand in the wound plays the role of the nondominant hand in open surgery. It works in concert with the instrument-bearing dominant hand for effective and safe tissue dissection. Our preliminary data support the view that this hybrid technique allows for efficient and safe en bloc removal of a tumor-bearing kidney with ureter and bladder cuff. REFERENCES 1. Cancer Statistics 1999. CA Cancer J Clin 49: 8, 1999. 2. Anderstrom C, Johansson SL, Pettersson S, et al: Carcinoma of the ureter: a clinicopathological study of 49 cases. J Urol 142(2 Pt 1): 280 –283, 1989. 3. Blute ML, Segura JW, Patterson DE, et al: Impact of endourology on diagnosis and management of upper urinary tract urothelial cancer. J Urol 141: 1298 –1301, 1989. 4. Grossman HB, Schwartz SL, and Konnak JW: Ureteroscopic treatment of urothelial carcinoma of the ureter and renal pelvis. J Urol 148(2 Pt 1): 275–277, 1992. 5. Smith AD, Orihuela E, and Crowley AR: Percutaneous management of renal pelvic tumors: a treatment option in selected cases. J Urol 137: 852– 856, 1987. 6. Tasca A, and Zattoni F: The case for percutaneous approach to transitional cell carcinoma of the renal pelvis. J Urol 143: 902–904, 1990. 7. Zincke H, and Neves RJ: Feasibility of conservative surgery for transitional cell cancer of the upper urinary tract. Urol Clin North Am 11: 717–724, 1984. 8. Nakada SY: Techniques in endourology: hand assisted laparoscopic nephrectomy. J Endourol 13: 9 –14, 1999. 9. Stifelman MD, Sosa RE, and Shichman SJ: Hand assisted laparoscopy (HAL). Current Surg Tech Urol 12: 1– 4, 1999. 10. Slakey DP, Wood JC, Hender D, et al: Laparoscopic living donor nephrectomy: advantages of the hand assisted method. Transplantation 68: 581–583, 1999. 11. Wolf JS Jr, Moon TD, and Nakada SY: Hand assisted laparoscopic nephrectomy, comparison to standard laparoscopic nephrectomy. J Urol 160: 22–27, 1998. 12. Keeley FX, Sharma NK, and Tolley DA: Hand-assisted laparoscopic nephroureterectomy. BJU Int 83: 504 –505, 1999. 13. McDougall EM, Clayman RV, and Elashry O: Laparoscopic nephroureterectomy for upper tract transitional cell carcinoma: the Washington University experience. J Urol 154: 975–980, 1995. 14. Salomon L, Hoznek A, Cicco A, et al: Retroperitoneoscopic nephroureterectomy for renal pelvic tumors with a single iliac incision. Urology 161: 541–544, 1999. 15. Keeley FX Jr, and Tolley DA: Laparoscopic nephroureterectomy: making management of upper urinary-tract transitional-cell carcinoma entirely minimally invasive. J Endourol 12: 139 –141, 1998. UROLOGY 56 (5), 2000
16. Hsu TH, Gill IS, Fazeli-Matin S, et al: Radical nephrectomy and nephroureterectomy in the octogenarian and nonagenarian: comparison of laparoscopic and open approaches. Urology 53: 1121–1125, 1999. 17. Kloosterman T, von Blomberg BM, Borgstein P, et al: Unimpaired immune function after laparoscopic cholecystectomy. Surgery 115: 424 – 428, 1994. 18. Gill IS, Soble JJ, Miller SD, et al: A novel technique for management of the en bloc bladder cuff removal and distal ureter during laparoscopic nephroureterectomy. J Urol 161: 430 – 434, 1999.
UROLOGY 56 (5), 2000
19. Arango O, Bielsa O, Carles J, et al: Massive tumor implantation in the endoscopic resected area in modified nephroureterectomy. J Urol 157: 1839, 1997. 20. Fernandez Gomez JM, Barmadah SE, Perez Garcia J, et al: Risk of tumor seeding after nephroureterectomy combined with endoscopic resection of the ureteral meatus. Arch Esp Urol 51: 829 – 831, 1998. 21. Palou J, Caparros J, Orsola A, et al: Transurethral resection of intramural ureter as the first step of nephroureterectomy. J Urol 154: 43– 44, 1995.
747
HAND-ASSISTED LAPAROSCOPIC NEPHROURETERECTOMY FOR THE TREATMENT OF TRANSITIONAL CELL CARCINOMA OF THE UPPER URINARY TRACT MICHAEL D. STIFELMAN, R. ERNEST SOSA, ADELINO ANDRADE, ARTHUR TARANTINO, AND STEVEN J. SHICHMAN
ABSTRACT Objectives. Nephroureterectomy with removal of the bladder cuff is the standard of care for patients with upper tract transitional cell carcinoma. Historically, it has been performed using two separate incisions or one large incision extending from the lateral flank to the symphysis pubis. We describe an alternative technique using endoscopic and hand-assisted laparoscopic techniques and present our experience. Methods. During the past 18 months, 22 patients at two institutions underwent hand-assisted laparoscopic nephroureterectomy. In 19 patients, the distal ureter and bladder cuff were managed endoscopically. In 3 patients, the distal ureter and the bladder cuff were removed by an extravesical, laparoscopic technique. The intraoperative parameters assessed included operative time, estimated blood loss, specimen weight, surgical margin status, pathologic grade and stage, and acute complications. Postoperative endpoints included the time to sustained fluid intake, parenteral narcotic requirement (milliequivalents of morphine sulfate), oral narcotic requirement (number of tablets), length of stay, time until return to normal activity, and rate of tumor recurrence. Results. The average age of our patient population was 65 years (range 42 to 86), 10 patients were men and 12 were women, and the average American Society of Anesthesiologists classification was 2.2. All but 2 patients had their specimens removed en bloc. No intraoperative complications occurred. The average operative time was 272 minutes (range 190 to 440), and the average blood loss was 180 mL (range 50 to 400); no patient required a transfusion. The mean specimen weight was 457 g (range 190 to 1420). All 22 patients had negative surgical margins. Postoperatively, the time to sustained fluid intake averaged 2.1 days (range 1 to 7), the mean parenteral narcotic requirement was 55 mEq (range 12 to 107.8) of morphine sulfate, the mean oral narcotic requirement was 5.8 tablets (range 1 to 14), and the average length of stay was 4.1 days (range 3 to 14). One patient developed thrombophlebitis of the right external jugular vein from a central line and required 2 weeks of intravenous antibiotics. The mean time to return to normal activity was 19 days; the mean follow-up was 13 months. Six patients had disease recurrence: four low-grade, low-stage bladder tumors and two metastatic tumors. All patients were alive at 18 months. Conclusions. Hand-assisted laparoscopic nephroureterectomy with endoscopic management of the bladder cuff is a viable and efficacious alternative to open nephroureterectomy. The technique allows the surgeon to perform an en bloc resection of the kidney, ureter, and bladder cuff without compromising oncologic principles. Patients benefit from a decrease in pain and hospital stay and quicker convalescence. Longer follow-up and comparative studies to standard open techniques are underway. UROLOGY 56: 741–747, 2000. © 2000, Elsevier Science Inc.
A
pproximately 5300 patients within the United States have been diagnosed with transitional cell carcinoma involving the upper urinary tract in
1999, and 1600 patients can expect to die of this disease.1 The traditional treatment for upper tract urothelial tumors is total nephroureterectomy
From the Department of Urology, James Buchanan Brady Foundation, New York Presbyterian Hospital, Weill Medical College, Cornell University, New York, New York; and Connecticut Surgical Group, PC, Hartford Hospital, Hartford, Connecticut
Reprint requests: Michael D. Stifelman, M.D., 540 First Avenue, Suite 1012, New York, NY 10016 Submitted: February 28, 2000, accepted (with revisions): June 20, 2000
© 2000, ELSEVIER SCIENCE INC. ALL RIGHTS RESERVED
0090-4295/00/$20.00 PII S0090-4295(00)00751-2 741
with excision of a cuff of bladder.2 Extirpation can be achieved using one large incision extending from the patient’s flank to the symphysis pubis. Alternatively, two separate incisions may be used at the flank and lower abdominal region. This latter approach allows the surgeon better exposure for excision of the bladder cuff. In either scenario, the kidney, ureter, and bladder are removed en bloc to prevent tumor spillage. Minimally invasive antegrade and retrograde endoscopic approaches have been reported for the treatment of this disease. The results have been varied, and disadvantages include a high ipsilateral recurrence rate, the need for multiple procedures, and the requirement for accurate preoperative confirmation that the tumor is truly low stage and low grade.3– 6 High-grade tumors have significantly increased recurrence rates when treated with local excision alone compared with nephroureterectomy.7 Thus, ureteroscopic and/or percutaneous treatments should be reserved for patients with a solitary kidney, bilateral disease, or decreased renal reserve, and low-grade and low-stage tumors. The advent of hand-assisted laparoscopy (HAL) has provided a new alternative for treating patients with upper tract urothelial tumors using techniques of minimal access. The potential benefits of HAL include a decreased hospital stay, lessened analgesic requirements, and a faster recovery. HAL facilitates en bloc removal of the total specimen, maintaining the oncologic principles used in open surgery. We review our technique and report our results of nephroureterectomy using HAL with endoscopic management of the bladder cuff. MATERIAL AND METHODS PREPARATION AND POSITION A clear liquid diet is administered the day before surgery along with a mechanical bowel preparation. The patient fasts for 8 hours before surgery. Preoperative antibiotics are infused in the operating room. General anesthesia with endotracheal intubation is used in all cases. A nasogastric tube and a urethral catheter are used to decompress the stomach and bladder, respectively. Venous insufflation stockings are used to prevent lower extremity venous stasis. The patient is initially placed in the dorsolithotomy position for excision of the bladder cuff. Once complete, the patient is prepared again and draped in the modified lateral decubitus position at a 45° angle. The operating table is slightly flexed. The patient is secured to the operating table with 3-in. tape to permit side-to-side rotation for intraoperative exposure.
EQUIPMENT The equipment for endoscopic excision of the bladder cuff includes a resectoscope with a 30° lens, Collings’ knife, 5-mm trocar, 5-mm laparoscopic grasping instrument, ureteral stent, and a 2-0 chromic Surgi-Tie (U.S. Surgical, Norwalk, Conn). A 24F resectoscope equipped with a Collings’ knife is used for incisional release of the distal ureter and bladder cuff. The equipment required for HAL nephroureterectomy in742
FIGURE 1. Endoscopic management of ureter and bladder cuff. (a) Cystoscopy and placement of trocar and ureteral stent, (b) outlining incision, (c) bladder cuff and ureteral dissection, and (d) securing of distal ureter with tie.
cludes a hand device (Pneumosleeve, Dexterity, Atlanta, Ga; Handport, Smith and Nephew, Andover, Mass; or Intromit, Applied Medical, Rancho Santa Margarita, Calif), two or three 10 to 12-mm trocars, a 30° laparoscope, Maryland dissector, right-angle dissectors, endoscopic shears (U.S. Surgical), Harmonic scalpel (Ethicon, Sommers, NJ), bipolar cautery, small and large clips, an in-line endoscopic stapling device, and a laparoscopic suction/irrigating device.
ENDOSCOPIC MANAGEMENT OF BLADDER CUFF The bladder is inspected cystoscopically to ensure that no urothelial tumors are present. A 5F open-ended ureteral stent is inserted into the ureter on the side of the tumor. With the bladder distended, a 5-mm trocar is percutaneously inserted into the bladder dome (Fig. 1a). The valve on the trocar is left open and placed to allow gravity drainage. This provides a continuous flow system and increased visibility. The Collings’ knife is used on pure cut to score a 2-cm cuff of bladder mucosa about the ureteral orifice (Fig. 1b). A laparoscopic grasper is introduced through the trocar to hold the intravesical ureter during its dissection (Fig. 1c). The ureter is partially mobilized in a circumferential manner. The 2-0 chromic Surgi-Tie is introduced through the trocar, the stent is removed, the distal ureter is engaged, and the snare is tightened, closing off the ureteral lumen. The excess suture is cut and the distal ureter grasped again to continue dissection about the ureter until the extravesical space is reached and the ureter is UROLOGY 56 (5), 2000
through the hand-device site. A closed system drain is placed in the vicinity of the cystotomy and brought out through a stab wound. The renal vascular stumps, adrenal gland, and operative site are inspected for bleeding at low insufflation pressures. The trocar sites are inspected and closed. The handdevice site is closed last. Postoperatively, the patient is left with two trocar scars and a 6.5-cm hand-device scar (Fig. 3). For right nephroureterectomy, the right colon and hepatic flexure are mobilized medially, and the right lobe of the liver is released from the body sidewall. The coronary ligaments are incised to expose the upper renal pole and the vena cava. The duodenum is mobilized medially by a Kocher maneuver to expose the anterior aspect of the renal hilum. In a similar fashion to the left side, the artery is divided posteriorly, an in-line stapler divides the renal vein, and the kidney is mobilized within Gerota’s fascia. The ipsilateral ureter is handled as described for the left side. Of note, in the first 3 patients, the ureter and bladder cuff were taken by an extravesical laparoscopic approach. For this, the ureter was mobilized to the bladder. A clip was placed on the ureter to prevent the spillage of urine before the bladder was open. With slight traction on the ureter, the detrusor muscle was opened with electrocautery and blunt dissection to expose the bladder mucosa. The bladder cuff was incised and removed together with the distal ureter. The opening in the bladder was closed with 2-0 chromic sutures tied intracorporeally with the nondominant hand.
INTRAOPERATIVE PARAMETERS FIGURE 2. (a) Left trocar configuration and (b) right trocar configuration.
totally freed (Fig. 1d). A 20F Foley catheter is placed to decompress the bladder.
NEPHROURETERECTOMY The technique of performing HAL nephrectomy has been previously described and can easily be expanded to include nephroureterectomy.8,9 In brief, after excision of the bladder cuff and freeing of the distal ureter, the patient is placed in a 60°, lateral decubitus position with slight operating room table flexion to hyperextend the ipsilateral flank. The skin is cleansed with a Betadine preparation from the nipples to the groin. The operative field is widely draped to allow room for the hand device. The placement of the hand device and trocars for left and right-side tumors is outlined in Figure 2. For left-side nephrectomy, the colon is reflected medially from the iliac vessels to the splenic flexure. The dissection is carried in a cephalad direction, freeing the lateral splenic attachments to the level of the gastric fundus. The colon, spleen, and pancreas are mobilized en bloc, leaving the anterior face of Gerota’s fascia exposed. The aorta and renal artery are palpated. The investing tissues over the hilum are divided, and the renal vein is exposed and dissected free anteriorly. The renal hilum is identified posteriorly, and the renal artery is dissected free, clipped, and divided. The renal vein is divided using an in-line stapler. The kidney is dissected free, keeping Gerota’s fascia on, and leaving the adrenal gland behind. To free the ureter and remain optically correct, the video monitors are moved to the foot of the operating table. The ureter is mobilized to the level of the bladder using a combination of blunt dissection, clips, and electrocautery. The final adventitial attachments to the bladder are identified by gentle traction on the ureter and divided between clips. The chromic tie identifies the distal ureter and bladder cuff. The kidney, ureter, and bladder cuff are retrieved en bloc and intact UROLOGY 56 (5), 2000
The charts of all patients undergoing nephroureterectomy within the past 18 months were reviewed at our two institutions. The demographic data included age, sex, and American Society of Anesthesiologists classification. The intraoperative time was measured from the start of the endoscopic bladder intervention to the completion of the skin closure after removal of the specimen. The intraoperative time included the interval required to prepare, drape, and position the patient. Other intraoperative parameters included the estimated blood loss, the need for transfusion, complications, the specimen weight in grams, and an assessment of the surgical margins.
POSTOPERATIVE PARAMETERS The endpoints included the time to the first oral intake, parenteral pain medication requirement, oral pain medication requirement, and length of hospital stay. The time until the first oral intake was measured from the time the patient exited the operating room to the resumption of sustained fluid intake. The amount of postoperative parenteral analgesia was measured in milliequivalents of morphine sulfate and was defined as (milligrams morphine sulfate) ⫹ (milligrams meperidine)/10 ⫹ (milligrams hydromorphine) ⫻ 10 ⫹ (milligrams ketorolac)/6. The number of oral narcotic tablets ingested during the patient’s hospital stay was recorded. Patients were discharged when they were afebrile, ambulating, tolerating a regular diet, and a return of bowel function had been confirmed. All patients went home with Foley catheter drainage. On postoperative day 10, a cystogram was performed before removing the catheter. Using a retrospective chart review method and telephone interviews, we determined the time to the resumption of normal activities.
RESULTS Since May 1998 we have performed 22 HAL nephroureterectomies using the described technique. All patients had a preoperative diagnosis of transitional cell carcinoma of the upper urinary tract confirmed by either ureteroscopic biopsy and/or pos743
FIGURE 3. Patient with two trocar scars (white arrows) and 6.5-cm hand-device scar (black arrow).
itive cytologic findings with a filling defect. The metastatic workup was negative for all patients. The average age of our patient population was 65 years (range 42 to 68). Ten patients were men and 12 women. In terms of overall health and patient comorbidity, the average American Society of Anesthesiologists classification was 2.2. No intraoperative complications occurred in this series. Two specimens were not removed en bloc. In one of the earliest patients, ureteral thinning occurred during dissection of the intravesical ureter. Rather than risk inadvertent disruption, the lower ureter was clipped and divided. At the end of the laparoscopic procedure, the patient was placed into the dorsolithotomy position and the most distal aspect of the ureter with the bladder cuff was removed cystoscopically. In the second patient, the distal ureter was taken extravesically. The mucosal margin was positive for atypia. The Handport incision was extended and a larger cuff of bladder resected to achieve a negative margin. The average operative time, including repositioning, was 272 minutes (range 190 to 440). The estimated blood loss was minimal, with an average blood loss of 180 mL (range 50 to 400). No patient required a transfusion. The mean specimen weight 744
was 457 g (range 144 to 1420), and the mean specimen size volume (length ⫻ height ⫻ width) was 780 mL (range 205 to 2112). Pathologically, 3 patients had low-grade tumors (grade I, I-II), 10 had medium-grade tumors (grade II), and 9 patients had high-grade tumors (grade III). All tumors were transitional cell carcinoma. Five lesions were Stage Ta, 8 were T1, 2 were T2, and 7 were T3 lesions; no T4 lesions were found. No patient had a positive margin on the final pathologic specimen. In the postoperative period, the patients were started on clear fluids the day after surgery. The time to sustained fluid intake without nausea or vomiting averaged 2.1 days (range 1 to 7). Patients were either given patient-controlled anesthesia or prescribed intravenous narcotics for the first 24 to 48 hours after surgery. The total parenteral narcotic requirement averaged 55 mEq (range 12 to 107.8) morphine sulfate. Once patients were tolerating clear liquids, they were converted to oral narcotics for pain management. The oral narcotic requirement averaged 5.8 tablets (range 1 to 14) during the patients’ hospital stay. The average length of stay was 4.1 days (range 3 to 14). One patient developed thrombophlebitis in the right external jugular vein at the site of a preoperative UROLOGY 56 (5), 2000
TABLE I. Results of laparoscopic and hand-assisted laparoscopic nephroureterectomy series
Investigators Salomon et al.14 Hsu et al.16 McDougall et al.13 Keeley and Tolley15 Wolf et al.11 Current series
Patients (n) 4 11 10 22 2 22
Approach Laparoscopic retroperitoneal Laparoscopic retroperitoneal Laparoscopic transabdominal Laparoscopic transabdominal HAL HAL
Mean Operating Time (min)
Blood Loss (mL)
Hospital Stay (days)
220
220
5.7
210
150
498
233
156 345 272
NA* 625 180
Pain Medication (mg MSO4)
Return to Normal Activity (days)
1
NR
2
14
14
5
16
19.6
5.5
NR
NR
3.1† 4.1
48.3† 55
14† 19
KEY: MSO4 ⫽ morphine sulfate; NR ⫽not reported; NA ⫽ not applicable (see footnote); HAL ⫽ hand-assisted laparoscopy. * Blood loss was quantified by transfusion requirement; the mean transfusion requirement was 0.3 U per patient (versus 0.9 U per patient in the open group). † Data for the complete series. It included 11 hand-assisted nephrectomies.
central line placement and required 2 weeks of intravenous antibiotics. The mean time to a return to normal activity was 19 days; the mean follow-up was 13 months. Six patients had disease recurrence. Four patients had low-grade, low-stage bladder tumors, not involving the resection site. Two patients with grade III, pT3 tumors manifested metastatic lesions. All patients were alive at 18 months. COMMENT Several published studies have addressed the merits of HAL nephrectomy for both benign and malignant disease.8 –11 Only two published reports are available of HAL nephroureterectomy.11,12 The total number of patients in both these studies was three. Our initial series of 22 patients who underwent HAL nephroureterectomy revealed very encouraging short-term results. The technique provides an effective and reproducible minimally invasive method for removing the kidney, ureter, and bladder cuff en bloc. Removing the specimen en bloc potentially decreases the risk of tumor spillage and provides for accurate staging, similar to an open operation. We reviewed and summarized the published data on laparoscopic and HAL nephroureterectomy to compare previous data with our experience (Table I). Our average operative time of 272 minutes was similar to other laparoscopic nephroureterectomy series. McDougall et al.13 reported on their initial 10 patients who underwent laparoscopic nephroureterectomy, with an average operative time of 8.3 hours. Salomon et al.14 published their experience on retroperitoneoscopic nephroureterectomy with iliac excision to manage the bladder cuff for renal pelvic tumors. Their opUROLOGY 56 (5), 2000
erative time averaged 3.7 hours. Keeley et al.15 reported on 22 laparoscopic nephroureterectomies, with a rapid mean operative time of 2.6 hours. In comparison, our operative time of 4.5 hours was comparable to the other series.14,16 The blood loss in our series was also similar to that in other laparoscopic series.14 –16 The HAL technique does not compromise the ability to remove an intact specimen. Intact tumor extirpation proved to be independent of specimen size. This is illustrated by specimen weights of up to 1200 g and a conversion rate of zero. In the laparoscopic series of Keeley and Tolley,15 the conversion rate for 22 patients was 13%. The postoperative narcotic requirements, time to oral intake, and length of hospital stay for HAL and laparoscopic nephroureterectomy are summarized in Table I. The advantages of HAL and laparoscopy compared with open surgery are realized with these benefits. The shorter time for convalescence associated with these minimally invasive techniques not only allows patients to resume their normal activities sooner, but also reflects the smaller amount of trauma sustained by the patient. Kloosterman et al.17 compared laparoscopic cholecystectomy with open cholecystectomy in relation to the amount of trauma induced and immunologic compromise sustained by both procedures. Using the white blood cell count, interleukin-6 assay, HLA-DR expression, and phytohemagglutinin skin testing, they quantified the acute phase response and T-cell mediated immunity postoperatively for these two procedures. They concluded that laparoscopy was associated with a decreased acute phase response (trauma) and preservation of the patient’s cell-mediated immunity compared with open surgery.17 It will be of interest to com745
pare HAL and/or laparoscopic nephroureterectomy with open nephroureterectomy in this light. A similar decrease in the acute phase response and preservation of cell-mediated immunity would lead one to ponder whether laparoscopy could attenuate the biologic potential of a tumor. More research and longer follow-up are necessary to look at these issues. Endoscopic management of the bladder cuff has been described by other investigators.13,15,18 Our technique has some slight modifications. However, the general principles remain the same. These include excision of the bladder cuff with mobilization of the ureteral meatus and intramural ureter using a resectoscope with a Collings’ knife. The ureter is tied to complete occlusion before full mobilization to prevent spillage of ipsilateral urine into the retroperitoneum. A Foley catheter is left in place for 10 to 14 days to allow the defect to heal. A cystogram is performed to confirm a lack of extravasation before removing the catheter. One concern raised by using this technique is the risk of tumor seeding and implantation. Two cases have been reported of presumed tumor seeding after endoscopic incision of the ureteral meatus and bladder cuff.19,20 Whether the recurrence was due to seeding during the excision of the distal ureter remains unclear. Palou et al.21 reported on 31 patients who underwent endoscopic excision of the distal ureter before nephroureterectomy. Most of the patients had high-grade tumors with invasion, and none had tumor relapse.21 Certainly, careful follow-up and further investigation is warranted. Several potential advantages and disadvantages to the endoscopic approach for managing the distal ureter and bladder cuff exist. The transurethral technique allows the intraoperative hand to remain useful and assist in the dissection of the distal ureter. Less laparoscopic dissection of the distal ureter and bladder wall is necessary. For an open approach to the distal ureter, an incision larger than 7.5 cm and significant retraction are required to gain adequate exposure. The larger incision and added retraction may increase the patient’s postoperative pain and discomfort. However, we acknowledge that the selection of techniques is largely a matter of surgeon preference. Moreover, the open approach to the bladder cuff and distal ureter should be specifically considered in cases in which the patient has had a recent history of bladder tumors. Moreover, because of the discovery of bladder diverticula on follow-up cystoscopy and the long catheterization time after endoscopic management, we have begun to evaluate alternate ways of dealing with the distal ureter and bladder cuff. Finally, why did we choose to perform this pro746
cedure using HAL versus pure laparoscopy? One of the major objectives in performing a nephroureterectomy for transitional cell carcinoma is to remove the specimen intact. Therefore, morcellation of the tumor is not an option. Whatever technique is selected, an incision must be made large enough to remove the specimen en bloc. Our incision in the lower midline or right lower quadrant is equal in length to those created to remove the specimen after laparoscopy.13,14 Having one’s hand in the operative field allows for tactile sensation, a threedimensional spatial orientation, tissue palpation, blunt dissection, and retraction. The hand in the wound plays the role of the nondominant hand in open surgery. It works in concert with the instrument-bearing dominant hand for effective and safe tissue dissection. Our preliminary data support the view that this hybrid technique allows for efficient and safe en bloc removal of a tumor-bearing kidney with ureter and bladder cuff. REFERENCES 1. Cancer Statistics 1999. CA Cancer J Clin 49: 8, 1999. 2. Anderstrom C, Johansson SL, Pettersson S, et al: Carcinoma of the ureter: a clinicopathological study of 49 cases. J Urol 142(2 Pt 1): 280 –283, 1989. 3. Blute ML, Segura JW, Patterson DE, et al: Impact of endourology on diagnosis and management of upper urinary tract urothelial cancer. J Urol 141: 1298 –1301, 1989. 4. Grossman HB, Schwartz SL, and Konnak JW: Ureteroscopic treatment of urothelial carcinoma of the ureter and renal pelvis. J Urol 148(2 Pt 1): 275–277, 1992. 5. Smith AD, Orihuela E, and Crowley AR: Percutaneous management of renal pelvic tumors: a treatment option in selected cases. J Urol 137: 852– 856, 1987. 6. Tasca A, and Zattoni F: The case for percutaneous approach to transitional cell carcinoma of the renal pelvis. J Urol 143: 902–904, 1990. 7. Zincke H, and Neves RJ: Feasibility of conservative surgery for transitional cell cancer of the upper urinary tract. Urol Clin North Am 11: 717–724, 1984. 8. Nakada SY: Techniques in endourology: hand assisted laparoscopic nephrectomy. J Endourol 13: 9 –14, 1999. 9. Stifelman MD, Sosa RE, and Shichman SJ: Hand assisted laparoscopy (HAL). Current Surg Tech Urol 12: 1– 4, 1999. 10. Slakey DP, Wood JC, Hender D, et al: Laparoscopic living donor nephrectomy: advantages of the hand assisted method. Transplantation 68: 581–583, 1999. 11. Wolf JS Jr, Moon TD, and Nakada SY: Hand assisted laparoscopic nephrectomy, comparison to standard laparoscopic nephrectomy. J Urol 160: 22–27, 1998. 12. Keeley FX, Sharma NK, and Tolley DA: Hand-assisted laparoscopic nephroureterectomy. BJU Int 83: 504 –505, 1999. 13. McDougall EM, Clayman RV, and Elashry O: Laparoscopic nephroureterectomy for upper tract transitional cell carcinoma: the Washington University experience. J Urol 154: 975–980, 1995. 14. Salomon L, Hoznek A, Cicco A, et al: Retroperitoneoscopic nephroureterectomy for renal pelvic tumors with a single iliac incision. Urology 161: 541–544, 1999. 15. Keeley FX Jr, and Tolley DA: Laparoscopic nephroureterectomy: making management of upper urinary-tract transitional-cell carcinoma entirely minimally invasive. J Endourol 12: 139 –141, 1998. UROLOGY 56 (5), 2000
16. Hsu TH, Gill IS, Fazeli-Matin S, et al: Radical nephrectomy and nephroureterectomy in the octogenarian and nonagenarian: comparison of laparoscopic and open approaches. Urology 53: 1121–1125, 1999. 17. Kloosterman T, von Blomberg BM, Borgstein P, et al: Unimpaired immune function after laparoscopic cholecystectomy. Surgery 115: 424 – 428, 1994. 18. Gill IS, Soble JJ, Miller SD, et al: A novel technique for management of the en bloc bladder cuff removal and distal ureter during laparoscopic nephroureterectomy. J Urol 161: 430 – 434, 1999.
UROLOGY 56 (5), 2000
19. Arango O, Bielsa O, Carles J, et al: Massive tumor implantation in the endoscopic resected area in modified nephroureterectomy. J Urol 157: 1839, 1997. 20. Fernandez Gomez JM, Barmadah SE, Perez Garcia J, et al: Risk of tumor seeding after nephroureterectomy combined with endoscopic resection of the ureteral meatus. Arch Esp Urol 51: 829 – 831, 1998. 21. Palou J, Caparros J, Orsola A, et al: Transurethral resection of intramural ureter as the first step of nephroureterectomy. J Urol 154: 43– 44, 1995.
747