COMPLICATIONS OF LAPAROSCOPIC PARTIAL NEPHRECTOMY IN 200 CASES

0022-5347/05/1731-0042/0 Vol. 173, 42 47, January 2005 THE JOURNAL OF UROLOGY Printed in U.S.A. Copyright 2005 by AMERICAN UROLOGICAL ASSOCIATION DOI: 10.1097/01.ju.0000147177.20458.73 COMPLICATIONS OF LAPAROSCOPIC PARTIAL NEPHRECTOMY IN 200 CASES ANUP P. RAMANI, MIHIR M. DESAI, ANDREW P. STEINBERG, CHRISTOPHER S. NG, SIDNEY C. ABREU, JIHAD H. KAOUK, ANTONIO FINELLI, ANDREW C. NOVICK AND INDERBIR S. GILL* From the Section of Laparoscopic and Robotic Surgery, Glickman Urological Institute, Cleveland Clinic Foundation, Cleveland, Ohio ABSTRACT Purpose: We analyzed complications of the initial 200 cases treated with laparoscopic partial nephrectomy for a suspected renal tumor. Materials and Methods: Since August 1999, 200 consecutive patients have undergone laparoscopic partial nephrectomy. Mean patient age was 61.6 years, mean body mass index was 29.9 and mean tumor size was 2.9 cm (range 1 to 10). There were 51 central tumors (25%) and 15 solitary kidneys (7.5%). A central tumor was defined as any tumor infiltrating up to the collecting system or renal sinus, during the excision of which entry into and repair of the collecting system was necessary. Mean estimated blood loss was 247 cc and mean operative time was 3.3 hours. Data on complications were obtained from a prospectively maintained computerized database and via telephone calls to patients and/or local referring physicians. Results: A total of 66 patients (33%) had 1 or more complications, which were intraoperative in 11 (5.5%), postoperative in 24 (12%) and delayed in 31 (15.5%). Overall 30 patients (15%) had a nonurological complication and 36 (18%) had a urological complication, including hemorrhage in 19 (9.5%) and urine leakage in 9 (4.5%). Hemorrhage occurred intraoperatively in 7 cases (3.5%) and postoperatively in 4 (2%), while it was delayed in 8 (4%). Of patients with urine leakage none required reoperation, 6 (3%) required a Double-J stent (Medical Engineering Corp., New York, New York) only, 2 (1%) required a Double-J stent with computerized tomography guided drainage and 1 required no treatment. Open conversion was necessary in 2 patients (1%), reoperation was done in 4 (2%) and elective laparoscopic radical nephrectomy was performed in 1 (0.5%). Conclusions: Laparoscopic partial nephrectomy is an advanced procedure with potential for complications. It requires considerable experience with reconstructive laparoscopy. KEY WORDS: kidney, laparoscopy, nephrectomy, kidney neoplasms, complications Nephron sparing surgery is now a standard of care in the management of select, small renal tumors. 1 Laparoscopic partial nephrectomy is emerging as an attractive nephron sparing option at centers where there is expertise in minimally invasive surgery. 2 6 Laparoscopic partial nephrectomy was initially reserved for patients with a favorably located, small, peripheral, exophytic renal tumor. 3, 7 However, with growing laparoscopic experience and the refinement of technique in select patients with a compromised or threatened global nephron mass we are currently applying laparoscopic partial nephrectomy to more substantive tumors that are larger or infiltrating more deeply into the renal parenchyma. 4, 8 Laparoscopic partial nephrectomy is a technically challenging procedure that is associated with a steep learning curve and potential for complications. We analyzed complications in 200 consecutive patients undergoing laparoscopic partial nephrectomy for a renal tumor at our institution. MATERIALS AND METHODS Submitted for publication February 17, 2004. * Correspondence and requests for reprints: Section of Laparoscopic and Robotic Surgery, Urological Institute, Cleveland Clinic Foundation, A100, 9500 Euclid Ave., Cleveland, Ohio 44195 (telephone: 216-445-1534; FAX: 216-445-7031; e-mail: gilli@ccf.org). 42 The initial 200 patients who underwent laparoscopic partial nephrectomy for a solitary renal tumor since August 1999, as performed by a single surgeon (ISG), were evaluated. All patients underwent preoperative spiral computerized tomography (CT) with 3-dimensional volume reconstruction to evaluate tumor location, depth of invasion, proximity to the renal sinus or hilum and vascular anatomy. Our technique of laparoscopic partial nephrectomy, wherein we aim to replicate open surgical principles, has been described previously. 2 Briefly, a transperitoneal or retroperitoneal approach is used to perform en bloc Satinsky clamping of the renal hilum or individual clamping of the renal artery and vein using bulldog clamps. Subsequently intraoperative laparoscopic ultrasonography, tumor excision with an adequate margin, suture repair of the collecting system if needed, suture repair of transected vessels and hemostatic sutured renorrhaphy are performed with all suturing performed with pure laparoscopic free-hand intracorporeal techniques. Table 1 lists demographic data. Of the 200 patients 117 (59%) were male. Mean age was 61.6 years (range 17 to 87). Mean body mass index was 29.9 (range 18 to 56). The transperitoneal approach was used in 122 patients (62%) and the retroperitoneal approach was used in 76 (38%). Mean tumor size on preoperative CT was 2.9 cm (range 1 to 10). Mean depth of parenchymal invasion on intraoperative ultrasonography was 1.5 cm (range 0.2 to 5). Hemorrhage was defined as bleeding requiring transfusion or therapeutic intervention. Urine leakage was defined as urinary extravasation requiring therapeutic intervention or that identified radiographically on the 1-month postoperative excretory urogram. Demographic and intraoperative data were prospectively entered into a computerized database. Postoperative data were

COMPLICATIONS OF LAPAROSCOPIC NEPHRECTOMY 43 TABLE 1. Demographic data No. pts 200 Mean age SD (range) 61.6 13.1 (17 87) Mean body mass index SD (range) 29.9 7.6 (18 56) No. male 117 (58.5) No. lt side 88 (44) No. transperitoneal approach 122 (61) Mean mg/dl preop serum creatinine SD 1.09 0.4 (0.4 3.4) (range) Mean cm tumor size SD (range) 2.9 1.2 (1 10) Mean cm parenchymal invasion depth SD 1.5 0.8 (0.2 5) (range) Mean cc blood loss SD (range) 247 273 (25 1,500) No. pelvicaliceal repair 142 (71) Mean mins operative time SD (range) 199 56.1 (45 360) Mean mins ischemia time SD (range) 28.7 11.8 (15 58) No. open conversion 2 (1) No. reoperation 4 (2) collected by personal and telephone interviews with the patient and/or referring physician. RESULTS Of the procedures 198 (99%) were completed laparoscopically. Two cases (1%) were converted to open surgery. In 1 case failure to progress due to dense postoperative adhesions necessitated elective open conversion. In the second case persistent parenchymal hemorrhage despite individual clamping of the renal artery and vein, likely due to inadequate compression by the laparoscopic bulldog clamps, led to emergency open partial nephrectomy. Overall 66 patients (33%) had a perioperative complication. Hemorrhagic complications (tables 2 and 3). Mean estimated blood loss was 247 cc (range 25 to 1,500). Hemorrhage in 19 patients (9.5%) occurred intraoperatively in 7 (3.5%) and postoperatively in 4 (2%), while it was delayed to after hospital discharge in 8 (4%). In these 19 patients mean tumor size was 2.6 cm, 10 tumors (53%) were central and the mean depth of tumor invasion was 2.2 cm. A total of 18 patients (9%) required blood transfusion. Intraoperative hemorrhage in 7 cases (3.5%) occurred due to laparoscopic bulldog malfunction (3), laparoscopic Satinsky clamp malfunction (1), systemic coagulopathy (1) and in the setting of multiple renal arteries (2). No patient in this series had significant hemorrhage upon renal revascularization following hilar unclamping. Hemorrhage was controlled laparoscopically with parenchymal suturing in 5 cases and by laparoscopic radical nephrectomy in the presence of a normal contralateral kidney in 1. One patient required emergency conversion to open partial nephrectomy to achieve adequate hemostasis. Postoperative hemorrhage prior to discharge home developed in 4 patients (2%), ostensibly from the partial nephrectomy bed. In each of these 4 patients hemorrhage occurred on day 2. Intraoperatively in all 4 patients complete hemostasis upon hilar unclamping had been confirmed to surgeon satisfaction. In 3 patients no obvious precipitating cause of postoperative bleeding could be found. In 1 patient bleeding developed presumably due to preexisting uremia and coagulopathy. All 4 patients responded to conservative treatment (table 3). Delayed hemorrhage after discharge home occurred in 8 patients (4%). Again, intraoperative parenchymal hemostasis to surgeon satisfaction had been documented in each case upon the completion of laparoscopic partial nephrectomy with a median initial hospital stay of 3 days. Hemorrhage developed at a mean of 16 days postoperatively (range 6 to 30). A possible precipitating cause of delayed hemorrhage could be identified in 5 patients, namely vigorous exercise on day 14 postoperatively, an accidental fall on day 30, coagulopathy, systemic heparinization and a splenic tear in 1 each TABLE 2. Intraoperative hemorrhagic complications during laparoscopic partial nephrectomy Pt 9 Pt 32 Pt 36 Pt 146 Pt 184 Pt 186 Pt 200 Mean 0.8 0.8 3.4 0.9 1.4 0.7 0.9 1.2 Preop serum creatinine (mg/dl) Tumor location Lower lat Mid posterior Mid anterior Lower medial Mid lat Lower lat Upper posterolat Tumor size (cm) 2 1.8 3.3 2.2 3.9 3.3 3 2.8 Invasion depth (cm) 1 1 1 1.6 2 2 2.5 1.7 Laparoscopic approach Retroperitoneal Retroperitoneal Transperitoneal Retroperitoneal Retroperitoneal Retroperitoneal Retroperitoneal Retroperitoneal in 6 pts, transperitoneal in 1 Ischemia time (mins) 17 32 29 45 Not available 30 46 35 Intraop blood loss (cc) 1,200 1,500 1,500 1,500 1,500 500 1,100 1,425 Bulldog clamp malfunction Multiple arteries Bulldog clamp malfunction Satinsky clamp malfunction Multiple arteries Uremia, thrombocytopenia, hepatic dysfunction Possible cause Bulldog clamp malfunction Transfusion 3.4 U Parenchymal suture repair, transfusion 4 U PRBCs Laparoscopic radical nephrectomy, no transfusion Open conversion, transfusion 5 U PRBCs Parenchymal suture repair, no transfusion Transfusion 7 U PRBCs platelets Parenchymal suture repair, transfusion 4 U PRBCs Treatment Parenchymal suture repair, transfusion 4 U packed red blood cells (PRBCs) Operative time (hrs) 3 2.5 3 2.7 6 4.5 3.8 3.6 % Kidney incised Not available Not available Not available 20 30 5 30 23 Hospital stay (days) 2 3 9 3 5 2 4 4 Followup (mos) 34 24 27 5 4 16 2 14.2 Hemorrhagic complications developed in 19 patients (9.5%), including intraoperatively in 7 (3.5%).

44 COMPLICATIONS OF LAPAROSCOPIC NEPHRECTOMY TABLE 3. Postoperative and delayed hemorrhagic complications during laparoscopic partial nephrectomy Pt No. Preop Serum Creatinine (mg/dl) Tumor Location Tumor Size (cm) Invasion Depth (cm) Laparoscopic Approach Intraop Blood Loss (cc) Ischemia Time (mins) Operative Time (hrs) % Kidney Excised Surgeon Perception No. Sutures/Bolsters Bleeding Day No. Transfusions Hospital Stay (days) Possible Cause Treatment Followup (mos) Postop (4 pts or 2%): 158 1.8 Mid anterior 1.6 1.1 Transperitoneal 500 Not available 5 10 2/2 2 2 7 Unknown Bed rest 5 159 1.8 Mid anterior 1.6 0.7 Transperitoneal 500 0* 5 10 Not available 2 4 7 Azotemia Transfusion 8 161 0.8 Mid anterolat 3.4 1.7 Transperitoneal 400 0* 5.5 15 Not available 2 2 5 Unknown Transfusion 8 170 0.8 Posterolat 4.7 Not available Retroperitoneal 600 68 4 20 4/2 2 2 5 Unknown Transfusion 6 Mean 1.3 2.8 1.03 Transperitoneal in 3 pts, 500 4.8 13.75 2 2.5 6 6.7 retroperitoneal in 1 Delayed (8 pts or 4%): 35 1.2 Mid anterior 2.5 1 Transperitoneal 300 29 3 Not available 78 0.7 Mid lat 2.6 2 Retroperitoneal 200 33 2.5 Not available 4/2 6 10 2 Unknown Rehospitalization for delayed nephrectomy at 10 days 2/2 14 7 4 Unknown Bed rest 13 109 1.1 Lower lat 2.8 2 Transperitoneal 200 36 3.3 60 2/2 14 4 3 Exercise induced Bed rest continuous bladder irrigation 113 1 Mid anterior 3 1.5 Transperitoneal 50 45 2.5 15 4/2 14 0 3 Unknown Arterial embolization 130 0.9 Upper medial 4 1.8 Transperitoneal 200 30 2.5 30 2/2 30 4 5 Fall Bed rest 8 142 1.1 Mid lat 2.2 1.8 Transperitoneal 150 37 3.7 25 4/2 10 2 3 Uremia, coagulopathy Arterial embolization 169 1.3 Upper anterior 2.4 1.1 Transperitoneal 100 38 2.8 15 4/2 30 2 2 Splenic tear Transfusion 7 187 0.9 Upper lat 2.9 0.9 Transperitoneal 300 32 4 15 2/2 7 2 31 Heparin for pulmonary embolism Mean 1.02 2.8 1.5 Transperitoneal in 7 pts, retroperitoneal in 1 Exploratory laparotomy, splenectomy parenchymal suture repair 187.5 35 3.0 27 15.6 3.8 6.6 9.7 Patients 35, 78 and 130 had a body mass index of 30 or greater, patient 142 had preoperatively documented coagulopathy, caliceal entry necessitating suture repair occurred in 142 patients, fibrin glue was not used in any, and in patients 159 and 161 cold ischemia was achieved with laparoscopic ice slush. 23 10 9 7 1

COMPLICATIONS OF LAPAROSCOPIC NEPHRECTOMY 45 (tables 3 and 4). No obvious cause was evident in 3 patients. Treatment was conservative in 4 patients and percutaneous selective angioembolization was performed in 2. One patient underwent exploratory laparotomy for successful control of renal parenchymal oozing on postoperative day 7. One patient underwent exploratory laparotomy with delayed nephrectomy on day 10. Urinary leakage (table 5). Nine patients (4.5%) had urine leakage. Intraoperatively pelvicaliceal entry occurred in 8 of the 9 patients (89%), requiring suture repair. Treatment involved cystoscopic placement of a Double-J stent in 6 cases and a Double-J stent with CT guided percutaneous drainage in 2. In 1 patient asymptomatic urine leakage detected by routine excretory urography at 1 month resolved spontaneously with observation. No patient with urine leakage required operative reexploration. Other urological complications. Overall 36 patients (18%) had urological complications, including the mentioned hemorrhage in 19 (9.5%), urinary leakage in 9 (4.5%), transient renal insufficiency in 4 (2%), inferior epigastric artery injury in 1, epididymitis in 1, transient hematuria in 1 and ureteral injury in 1. Ureteral injury developed in 1 patient when it was mistaken for the aberrant lower pole artery, arising from the common iliac artery and coursing alongside the ureter, in a kidney with 3 arteries. The injury was recognized intraoperatively and laparoscopic ileal ureter was performed. Leakage from the ileo-ileal anastomosis necessitated delayed laparotomy and repair. In 1 patient intraoperative ultrasonography showed that the central infiltrating tumor penetrated more deeply than anticipated. Because of a normal contralateral kidney, laparoscopic radical nephrectomy was performed. Four patients (2%) required at least 1 hemodialysis session following surgery. A man undergoing 65% resection of a solitary kidney for a 6.5 cm tumor had acute renal failure postoperatively, requiring dialysis for 3 weeks. Serum creatinine 1 year later was 3.2 mg/dl (normal 0.8 to 1.4). Another man underwent staged bilateral partial nephrectomy for tumor. Although postoperative renal function was appropriate for bilateral resection, the long-term course was complicated by severe liver dysfunction due to heavy alcohol and cocaine addiction. This patient ultimately progressed to end stage renal disease 6 months after surgery, requiring maintenance dialysis. One patient with preexisting chronic renal insufficiency underwent 35% resection and presented in acute renal failure on postoperative day 30, which recovered with transient dialysis. The other patient required angioembolization for delayed hemorrhage on day 10 with subsequent acute renal failure. The administration of radio contrast medium during angioembolization was thought to have caused the episode of acute tubular necrosis. Serum creatinine normalized after 2 sessions of dialysis. Nonurological complications. A total of 30 patients (15%) had a nonurological complication (table 6). After the comple- TABLE 4. Likely causes of hemorrhage No. Pts Intraop: 7 (3.5) Bulldog clamp malfunction 3 (1.5) Multiple arteries 2 (1) Satinsky clamp malfunction 1 (0.5) Generalized bleeding due to uremia 1 (0.5) Postop: 4 (2) Cause unknown 3 (1.5) Coagulopathy/azotemia 1 (0.5) Delayed: 8 (4) Unknown 3 (1.5) Early vigorous exercise 1 (0.5) Violent fall 1 (0.5) Uremia/coagulopathy 1 (0.5) Heparinization 1 (0.5) Splenic tear 1 (0.5) TABLE 5. Urine leakage Pt 25 Pt 83 Pt 119 Pt 133 Pt 135 Pt 154 Pt 171 Pt 184 Pt 193 Mean Tumor location Upper medial Upper medial Mid medial Lower lat Mid lat Mid anterior Mid lat Mid lat Mid posterior Tumor size (cm) 2 3 2.8 3 3.5 1.5 2.7 3.9 3.6 2.8 Invasion depth (cm) 1 1.8 2 1.1 2.4 Not available 1.8 2 1.3 1.6 Caliceal repair No Yes Yes Yes Yes Yes Yes Yes Yes Yes in 8 pts, no in 1 Laparoscopic approach Retroperitoneal Transperitoneal Retroperitoneal Transperitoneal Transperitoneal Transperitoneal Transperitoneal Open Retroperitoneal Transperitoneal in 5 pts, retro- peritoneal in 3 Operative time (hrs) 1.5 2.8 2.8 2.5 3.8 3.2 2.5 6 3.7 3.2 % Kidney excised Not available Not available 10 5 25 15 25 30 20 16 No Yes Yes Yes Yes Yes Yes Yes Yes Yes in 8 pts, no in 1 Intraop watertightness testing Presentation day 30 60 60 15 15 4 30 30 7 28 Symptomatic No No No Yes Yes Yes Yes No Yes Yes in 5 pts, no in 4 Stent Stent Treatment Stent Stent Observation Stent CT guided drainage Stent Stent Stent CT guided drainage Percutaneous drainage 0 0 0 Not available 0 0 45 0 0 (days) Stent duration (days) 42 30 60 30 90 42 30 30 42 72 Urine leakage developed in 9 patients (4.5%).

46 COMPLICATIONS OF LAPAROSCOPIC NEPHRECTOMY TABLE 6. Overall complications No. Intraop No. Postop No. Delayed Total No. Urological 9 (4.5) 9 (4.5) 18 (9) 36 (18) Ureteral injury 1 1 (0.5) Renal insufficiency 3 1 4 (2) Urine leakage 2 7 9 (4.5) Hematuria 1 1 (0.5) Epididymitis 1 1 (0.5) Inferior epigastric artery injury 1 1 (0.5) Hemorrhage 7 4 8 19 (9.5) Nonurological 2 (1) 15 (7.5) 13 (6.5) 30 (15) Pulmonary: 1 (0.5) 7 (3.5) 2 (1) 10 (5) Pleural injury 1 1 (0.5) Atelectasis 3 3 (1.5) Pneumonia 2 1 3 (1.5) Pulmonary embolism 1 1 (0.5) Pleural effusion 1 1 2 (1) Cardiovascular: 0 5 (2.5) 4 (2) 9 (4.5) Atrial fibrillation 3 3 (1.5) Deep vein thrombosis 2 2 4 (2) Congestive cardiac failure 2 2 (1) Gastrointestinal: 1 (0.5) 2 (1) 1 (0.5) 4 (2) Colonic segmental ischemia 1 1 (0.5) Bowel injury 1 1 (0.5) Prolonged ileus 1 1 (0.5) Delayed splenic hemorrhage 1 1 (0.5) Sepsis 0 0 2 (1) 2 (1) Musculoskeletal: 0 1 (0.5) 4 (2) 5 (2.5) Gluteal fasciotomy 1 1 (0.5) Wound infection 4 4 (2) Totals 11 (5.5) 24 (12) 31 (15.5) 66 (33) tion of laparoscopic partial nephrectomy and during port site closure a small (less than 1 cm) superficial, serosal bowel abrasion due to the Carter-Thomson port site closure needle occurred in 1 patient. To give the benefit of the doubt a single figure-of-8 serosal stitch (3-zero polyglactin) was placed laparoscopically without any sequelae. An elderly patient with generalized atherosclerotic disease had delayed segmental ischemia of the colon for unclear reasons, probably thromboembolic, requiring secondary exploratory laparotomy and colon resection. In 1 patient with a small superficial splenic capsular tear managed satisfactorily intraoperatively by argon beam coagulation postoperative bleeding was managed conservatively with blood transfusion and bed rest. One patient had a pleural entry, which was recognized intraoperatively and suture repaired laparoscopically without sequelae. DISCUSSION Hemorrhage from the partial nephrectomy bed is a vital concern intraoperatively and postoperatively. Intraoperative parenchymal hemorrhage can occur during 2 distinct steps of the operation, each of which has a distinct etiology, that is 1) during parenchymal resection due to inadequate hilar clamping and 2) upon renal revascularization due to inadequate suture repair of the partial nephrectomy bed. Inadequate hilar clamping can occur due to clamp malfunction or a missed renal artery. Of our 200 patients intraoperative hemorrhage occurred in 7 (3.5%) due to inadequate hilar control, including bulldog clamp malfunction in 3, Satinsky clamp malfunction in 1, a missed accessory vessel in 2 and slow, generalized oozing from the partial nephrectomy bed likely due to uremic coagulopathy in 1. In no case in this series did significant hemorrhage occur following hilar unclamping, attesting to the efficacy of laparoscopic parenchymal suture repair. In 2 patients mild parenchymal oozing was noted following hilar unclamping, necessitating 1 additional parenchymal hemostatic stitch. The laparoscopic Satinsky clamp is inherently more reliable for en bloc renal artery and vein clamping compared to the currently available laparoscopic bulldog clamps, which have somewhat suboptimal occlusive strength, especially in the case of an atherosclerotic renal artery. The development of more reliable bulldog clamps, similar to those available for open surgery, would overcome this problem. Recently we modified our hemostatic technique of laparoscopic partial nephrectomy to incorporate routinely the bioadhesive FloSeal (Baxter Healthcare Corp., Deerfield, Illinois) to layer the partial nephrectomy bed during parenchymal suture repair. This modification, which has been reported in a separate communication, significantly decreased our hemorrhage rate to 3% in the most recent 63 patients compared to the 9.5% rate in these initial 200 cases. 9 Delayed bleeding occurred in 8 patients (4%). Potential factors that may have precipitated bleeding were identified in 5 patients, namely early vigorous physical exercise within the first 2 weeks postoperatively, a violent fall on postoperative day 30, heparin induced coagulopathy, a splenic tear and uremic coagulopathy in 1 each. In 3 patients no obvious precipitating cause of hemorrhage could be found. Van Poppel et al reported hemorrhage in 7.9% of 76 patients undergoing open partial nephrectomy. 10 They suggested that larger tumor size and a central location correlated with the risk of postoperative hemorrhage. Of our 12 patients who presented with postoperative (4) or delayed (8) bleeding 8 (67%) had a central tumor infiltrating to a depth of 2.2 cm and abutting the renal sinus. This is in contrast to our overall experience with laparoscopic partial nephrectomy in 200 patients, in which only 29% of patients had a central tumor with a mean infiltration depth of 1.5 cm. In patients undergoing laparoscopic partial nephrectomy we advocate complete bed rest for 24 hours postoperatively, followed by 2 weeks of restricted physical activity in an attempt to minimize physical jarring of the freshly operated renal remnant. The cumulative incidence of urinary fistulas following open partial nephrectomy is reportedly 6.5% (range 1.4% to 17.4%) (table 7). 1 In the current series urinary leakage developed in 4.5% of patients, of whom none required reoperation. These data attest to the efficacy of laparoscopic watertight suture repair of the collecting system by continuous intracorporeal

COMPLICATIONS OF LAPAROSCOPIC NEPHRECTOMY 47 TABLE 7. Open partial nephrectomy complications reported in literature No. Campbell et al 12 No. Lerner et al 13 No. Moll et al 14 No. Belldegrun et al 15 Steinbach et al 16 No. pts 259 169 164 146 140 878 Bleeding 6 (2.3) 6 (3.7) 3 (2.1) 2 (1.4) (2.3) Reoperation 8 (3.1) 3 (1.8) 1 (0.6) 3 (2.1) 2 (1.4) (1.9) Prolonged acute tubular necrosis 19 (7.3) 1 (0.7) (5) Dialysis 14 (5.4) (5.4) Urinary fistula 45 (17.4) 3 (1.8) 11 (6.7) 2 (1.4) 3 (2.1) (7.2) Death 4 (1.5) 1 (0.6) 3 (2.1) 2 (1.4) (2.4) Totals 96 (30) 7 (4.1) 18 (9.8) 11 (5.5) 10 (8.6) Mean No. suturing techniques. Routine placement of a ureteral catheter preoperatively for retrograde injection of methylene blue allows accurate intraoperative identification and the precise suture repair of pelvicaliceal entry. Suture repair watertightness can then be tested by repeat retrograde injection. Polascik et al compared the incidence of urinary fistulas with and without using methylene blue retrograde injection during open partial nephrectomy. 11 A urinary fistula occurred in 21% of patients in whom retrograde methylene blue was not used compared to 0% of those in whom retrograde injection of methylene blue was done routinely. All 9 patients in the current series were successfully treated without operative re-intervention, including a Double-J stent in 6, CT guided aspiration plus a Double-J stent in 2 and observation in 1 (table 5). Prolonged acute tubular necrosis with or without clinically overt renal failure may occur after nephron sparing surgery in 6.3% of patients (range 0.7% to 7.3%) (table 7). 1, 12 16 In our series 4 patients (2%) had acute renal failure necessitating dialysis. Intraoperatively it is important to ensure adequate intravenous hydration with mannitol and furosemide administration to promote diuresis and minimize ischemic sequelae. A European multi-institutional series of 53 patients undergoing laparoscopic partial nephrectomy showed an overall urological complication rate of 23%. 5 Hemorrhage developed in 5 patients (10%), that is intraoperatively in 4 (8%) and postoperatively in 1 (2%). Hemostasis required emergency open conversion in 2 cases (4%) and secondary radical nephrectomy in 1. Urine leakage occurred in 5 cases (10%), requiring a Double-J stent in 3, percutaneous nephrostomy in 1 and nephrectomy in 1. Overall 2 kidneys (4%) were lost. In a recent review Kim et al compared complications during 35 laparoscopic radical nephrectomies and 79 laparoscopic partial nephrectomies. 6 In the partial nephrectomy group complications included intraoperative hemorrhage in 6 cases (7.5%), urine leakage in 2 (2.4%), ureteral injury in 1, postoperative atelectasis in 1, acute renal failure in 1 and clot retention in 1. In each group 1 case was converted to open surgery to achieve hemostasis. In our 200 cases open conversion was necessary in 2 (1%). In 1 patient there was failure to progress due to adhesions secondary to previous abdominal surgery. In the other patient bulldog clamps used for vascular control failed to occlude the renal pedicle securely, leading to hemorrhage upon parenchymal incision. Open partial nephrectomy was performed. CONCLUSIONS The specific reconstructive technical challenge during laparoscopic partial nephrectomy is to achieve parenchymal hemostasis and watertight pelvicaliceal repair in a time sensitive manner. Laparoscopic partial nephrectomy in the initial 200 patients was associated with hemorrhage and urine leakage in 9.5% and 4.5%, respectively. Most complications were managed conservatively, resulting in a 1% open conversion and a 2% reoperation rate. Laparoscopic partial nephrectomy is a technically complex operation requiring advanced laparoscopic skills. REFERENCES 1. Uzzo, R. G. and Novick, A. C.: Nephron sparing surgery for renal tumors: indications, techniques and outcomes. J Urol, 166: 6, 2001 2. Gill, I. S., Desai, M. M., Kaouk, J. H., Meraney, A. M., Murphy, D. P., Sung, G. T. et al: Laparoscopic partial nephrectomy for renal tumor: duplicating open surgical techniques. J Urol, 167: 469, 2002 3. Hoznek, A., Salomon, L., Antiphon, P., Radier, C., Hafiani, M., Chopin, D. K. et al: Partial nephrectomy with retroperitoneal laparoscopy. J Urol, 162: 1922, 1999 4. Gill, I. S., Matin, S. F., Desai, M. M., Kaouk, J. H., Steinberg, A., Mascha, E. et al: Comparative analysis of laparoscopic versus open partial nephrectomy for renal tumors in 200 patients. J Urol, 170: 64, 2003 5. Rassweiler, J. J., Abbou, C., Janetschek, G. and Jeschke, K.: Laparoscopic partial nephrectomy. The European experience. Urol Clin North Am, 27: 721, 2000 6. Kim, F. J., Rha, K. H., Hernandez, F., Jarrett, T. W., Pinto, P. A. and Kavoussi, L. R.: Laparoscopic radical versus partial nephrectomy: assessment of complications. J Urol, 170: 408, 2003 7. Gill, I. S., Delworth, M. G. and Munch, L. C.: Laparoscopic retroperitoneal partial nephrectomy. J Urol, 152: 1539, 1994 8. Desai, M. M., Gill, I. S., Kaouk, J. H., Matin, S. F. and Novick, A. C.: Laparoscopic partial nephrectomy with suture repair of the pelvicaliceal system. Urology, 61: 99, 2003 9. Gill, I. S., Ramani, A. P., Spaliviero, M., Xu, M., Finelli, A., Kaouk, J. H. et al: Improved hemostasis during laparoscopic partial nephrectomy using gelatin matrix thrombin sealant. Unpublished data 10. Van Poppel, H., Bamelis, B., Oyen, R. and Baert, L.: Partial nephrectomy for renal cell carcinoma can achieve long-term tumor control. J Urol, 160: 674, 1998 11. Polascik, T. J., Pound, C. R., Meng, M. V., Partin, A. W. and Marshall, F. F.: Partial nephrectomy: technique, complications and pathological findings. J Urol, 154: 1312, 1995 12. Campbell, S. C., Novick, A. C., Streem, S. B., Klein, E. and Licht, M.: Complications of nephron sparing surgery for renal tumors. J Urol, 151: 1177, 1994 13. Lerner, S. E., Hawkins, C. A., Blute, M. L., Grabner, A., Wollan, P. C., Eickholt, J. et al: Disease outcome in patients with low stage renal cell carcinoma treated with nephron sparing or radical surgery. J Urol, 155: 1868, 1996 14. Moll, V., Becht, E. and Zeigler, M.: Kidney preserving surgery in renal cell tumors: indications, techniques and results in 152 patients. J Urol, 150: 319, 1993 15. Belldegrun, A., Tsui, K. H., dekernion, J. B. and Smith, R. B.: Efficacy of nephron-sparing surgery for renal cell carcinoma: analysis based on the new 1997 tumor-node-metastasis staging system. J Clin Oncol, 17: 2868, 1999 16. Steinbach, F., Stockle, M., Muller, S. C., Thuroff, 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