Complications of open primary and post-chemotherapy retroperitoneal lymph node dissection for testicular cancer

Urologic Oncology: Seminars and Original Investigations 28 (2010) 504 –509

Original article

Complications of open primary and post-chemotherapy retroperitoneal lymph node dissection for testicular cancer Vairavan S. Subramanian, M.D., Carvell T. Nguyen, M.D., Ph.D., Andrew J. Stephenson, M.D., Eric A. Klein, M.D.* Glickman Urological and Kidney Institute, Cleveland Clinic, Cleveland, OH 44195, USA Received 23 September 2008; received in revised form 28 October 2008; accepted 28 October 2008

Abstract Objective: Treatment decisions regarding the use of retroperitoneal lymph node dissection (RPLND) for low-stage and advanced testicular cancer may be influenced by the morbidity of the procedure. We sought to compare the complication profile of primary (P-) and post-chemotherapy (PC-) RPLND using a standardized complication grading scale. Materials and methods: A retrospective analysis was conducted of 112 and 96 patients who underwent P-RPLND and PC-RPLND, respectively, between 1982 and 2007 for perioperative outcomes and late complications. Postoperative complications were graded using a 5-tiered scale based on the severity and/or level of intervention required for resolution. Results: P-RPLND patients had rates of 5%, 24%, and 7% for intraoperative, postoperative, and late complications, respectively. For PC-RPLND, these rates were 12%, 32%, and 7%, respectively (P ⫽ 0.11, 0.19, and 1, respectively). Major postoperative complications (grades III–V) were observed in 3 (3%) P-RPLND and 8 (8%) PC-RPLND patients (P ⫽ 0.15), including 1 fatal pulmonary embolus in a PC-RPLND patient. Ileus accounted for 63% and 45% of postoperative complications of P-RPLND and PC-RPLND, respectively. PC-RPLND was associated with significantly greater operative times, blood loss, and transfusion rates (P ⬍ 0.001). Compared with PC-RPLND after first-line chemotherapy for advanced NSGCT, there were no significant differences in perioperative outcomes for PC-RPLND performed in other settings. Conclusions: P-RPLND and PC-RPLND are associated with low rates of serious short- and long-term complications and negligible mortality, without significant differences between the 2 procedures. The safe morbidity profile of RPLND performed by fellowship-trained urologic oncologists should be considered during treatment decision-making for low-stage and advanced testicular cancer. © 2010 Elsevier Inc. All rights reserved. Keywords: Testicular neoplasms; Retroperitoneum; Lymph node excision; Complications; Morbidity

1. Introduction Open retroperitoneal lymph node dissection (RPLND) is an established staging and therapeutic modality for patients with low-stage (clinical stage [CS] I, IIA, and IIB) non-seminomatous germ cell testicular cancer (NSGCT). However, surveillance and adjuvant chemotherapy with 2 cycles of bleomycinetoposide-cisplatin (BEPx2) are alternative management strategies for CS I and induction chemotherapy with BEPx3 or EPx4 is an acceptable alternative for CS IIA and IIB. Laparoscopic RPLND is also being employed increasingly at select centers in low-stage NSGCT patients. In advanced stage NS-

GCT and seminoma (CS IIC and III), post-chemotherapy (PC-) RPLND has an established role for the resection of residual masses given the risk of residual viable malignancy and/or teratoma. Treatment decisions regarding the use of RPLND for low-stage and advanced testicular cancer may be influenced by the morbidity profile of the procedure. We endeavored to identify the rate of perioperative and late complications of primary (P-) and PC-RPLND in a contemporary cohort of patients using a 5-tiered scale based on prior reports [1–5] for urologic procedures to grade complications.

2. Materials and methods * Corresponding author. Tel.: ⫹1-216-444-5591; fax: ⫹1-216-4453532. E-mail address: [email protected] (E.A. Klein). 1078-1439/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.urolonc.2008.10.026

Between 1982 and 2007, a total of 204 patients underwent 208 RPLND procedures for testicular cancer or ex-

V.S. Subramanian et al. / Urologic Oncology: Seminars and Original Investigations 28 (2010) 504 –509 Table 1 Clinical characteristics and intraoperative data P-RPLND No. pts (%) Mean pt age (range) No. primary tumor pathology (%) Mixed GCT Seminoma Rhabdomyosarcoma Sertoli cell tumor No. nodal pathology (%) No tumor or necrosis Mixed GCT Teratoma only Rhabdomyosarcoma Seminoma Adenocarcinoma Median operative time (min) (range) Median estimated blood loss (mL) (range) No. transfusion rate (%) Median length of stay (days) (range) Median follow up (months) (range) Antegrade ejaculation postoperative (%)

PC-RPLND

112 (54) 29 (16–53)

96 (46) 31 (18–60)

108 (96) 0 (0) 3 (3) 1 (1)

86 (90) 9 (9) 1 (1) 0 (0)

P value 0.069

63 (56) 41 (37) 7 (6) 1 (1) 0 (0) 0 (0) 270 (150–640)

34 (35) 10 (10) 47 (49) 0 (0) 2 (2) 3 (3) 305 (120–1020)

⬍0.001

450 (75–4500)

1000 (50–14000)

⬍0.001

7 (6) 6 (3–32)

40 (42) 6 (3–52)

63 (0–275)

51 (0–303)

80

41

⬍0.001 0.2 0.081

505

Postoperative complications occurred from the date of surgery to 30 days after and included complications prior to and after discharge from the initial hospitalization. All postoperative complications were graded on a 5-tiered severity scale (Table 2) as grade I– oral medication or bedside care; grade II–intravenous therapy including total parenteral nutrition (TPN), heparin, or antibiotics; grade III–intubation, interventional radiology, dialysis, or reoperation; grade IV– major organ resection or chronic disability; and grade V– death [5]. Ileus was defined as nothing by mouth (NPO) status maintained beyond postoperative day 5 or reinsertion of a nasogastric tube (NG) (grade I) but was considered a grade II complication if the patient was readmitted to the hospital or required parenteral nutrition. Acute renal failure was defined as a greater than 50% increase of serum creatinine from baseline. Late postoperative complications occurred from 30 days after RPLND to last follow-up. All available records were analyzed for late complications related to surgery and hospitalization such as small bowel obstructions and incisional hernias. Data on postoperative antegrade ejaculation was collected on patients at follow-up visits and by phone questionnaires. Mean follow-up was 63 months for P-RPLND patients and 51 months for PCRPLND patients (P ⫽ 0.081).

⬍0.001

P-RPLND ⫽ primary retroperitoneal lymph node dissection; PCRPLND ⫽ post-chemotherapy RPLND; GCT ⫽ germ cell tumor.

tragonadal germ cell tumor, with 112 undergoing P-RPLND and 96 PC-RPLND. Two surgeons performed 171 (82%) of the operations. P-RPLND was performed for NSGCT CS I, IIA, and IIB in 88%, 11%, and 1% patients, respectively, 3 patients with rhabdomyosarcoma, and 1 with Sertoli cell tumor (Table 1). PC-RPLND was performed after first-line and salvage chemotherapy in 79 and 7 NSGCT patients, respectively, after first- or second-line chemotherapy in 9 patients with advanced seminoma, and for rhabdomyosarcoma in 1 patient. A reoperative RPLND was performed in 7 patients. In the PC-RPLND cohort, 81 (84%) patients received bleomycin-containing chemotherapy regimens and 15 (16%) received chemotherapy without bleomycin (most commonly EPx4). Clinical information and follow-up data were obtained from a retrospective review of clinic records and direct patient correspondence for those followed outside our institution. The study was approved by the Institutional Review Board. 2.1. Complication grading Operative reports were reviewed for all intraoperative complications. A vascular injury was considered a complication if it was not necessitated by direct invasion of tumor and if repair required the assistance of a vascular surgeon or the patient received a blood transfusion intra-operatively.

2.2. Statistical analysis Complication and hospitalization data were compared between patients undergoing P-RPLND and PC-RPLND. Continuous variables were compared using the Student’s t-test for normally distributed data and the Mann-Whitney U test for non-normally distributed data. Categorical variables were compared using the Pearson’s ␹2 test. All analyses were performed using SPSS statistical software (SPSS, Inc., Chicago, IL) with P ⬍ 0.05 considered significant.

3. Results A total of 112 patients underwent P-RPLND and of these 63 (56%) had pathologic stage I (pN0) and 49 (44%) had pathologic stage II (30 pN1 and 19 pN2-3), with 7 of the latter having teratoma only. Among the 86 patients treated with primary or salvage chemotherapy for NSGCT, 34 Table 2 Severity scale for early complications Grade of complication

Intervention for complication

I II

Oral medication or bedside care Intravenous therapy (total peripheral nutrition, heparin, antibiotics) Intubation, interventional radiology, dialysis, re-operation Major organ resection or chronic disability Death

III IV V

506

V.S. Subramanian et al. / Urologic Oncology: Seminars and Original Investigations 28 (2010) 504 –509

Table 3 Summary of complications P-RPLND No. pts with intraoperative complications (%) No. pts with postoperative complications (%) No. postoperative complications No. pts with late complications (%) No. worst complication/pt (%) Grade I Grade II Grade III Grade IV Grade V

PC-RPLND

P value

6 (5)

11 (12)

0.11

27 (24)

31 (32)

0.19

32

44

8 (7)

7 (7)

19 (17) 5 (5) 3 (3) 0 (0) 0 (0)

15 (16) 8 (8) 6 (6) 1 (1) 1 (1)

1 0.15

P-RPLND ⫽ Primary retroperitoneal lymph node dissection; PCRPLND ⫽ Post-chemotherapy RPLND.

(35%) patients had necrosis/fibrosis, 10 (10%) had viable cancer, 47 (49%) had teratoma only, and 3 (3%) had adenocarcinoma. Two of the adenocarcinoma occurrences were found in 1 patient with psoas muscle-invasive adenocarcinoma who had recurrences 3 years apart. An additional 2 (2%) of the 9 seminoma patients had viable seminoma. Initial hospitalization following RPLND was similar in both cohorts with a median of 6 days (P ⫽ 0.2). 3.1. Intraoperative complications (Tables 3 and 4) In the P-RPLND cohort, there were 6 (5%) intraoperative complications including 5 vascular complications which either necessitated assistance of a vascular surgeon (n ⫽ 4) or resulted in significant blood loss requiring a transfusion (n ⫽ 1). There was 1 ureteral injury repaired by primary anastomosis intraoperatively. There were 12 intraoperative vascular complications in 11 patients (11%) in the PCRPLND group (P ⫽ 0.11) which met the above criteria. Of note, no major vascular complications occurred in PCRPLND done for a seminoma primary. Unilateral nephrectomy was performed in 2 P-RPLND (2%) and 2 PC-RPLND (2%) patients with the indication for 1 P-RPLND patient being a dysplastic kidney and the remainder removed for oncologic reasons. Median operative time, defined as the time from skin incision to closing, was significantly longer during PC-RPLND at 305 minutes compared with 270 minutes for P-RPLND (P ⬍ 0.001). Median estimated blood loss (EBL) was significantly greater at 1000 cc for PCRPLND relative to P-RPLND at 450 cc (P ⬍ 0.001) and resulted in a higher transfusion rate (42% vs. 6%) (P ⬍ 0.001). 3.2. Postoperative complications (Tables 3 and 5) A total of 32 postoperative complications occurred in 27 patients (24%) after P-RPLND, and there were 44 compli-

cations in 31 patients (32%) after PC-RPLND P⫽0.19. The majority of complications in both cohorts were minor (grades I and II), and ileus treated by conservative management accounted for 63% and 45% of all complications in P-RPLND and PC-RPLND patients, respectively. Major postoperative complications (grade III-V) were observed in 3 (3%) P-RPLND and 8 (8%) PC-RPLND patients (P ⫽ 0.15), including 1 fatal pulmonary embolus (PE) in a PC-RPLND patient. Among the P-RPLND group, there were 4 grade III complications in the 3 patients including 1 who required surgical exploration for early small bowel obstruction, 2 who developed chylous ascites requiring paracentesis and TPN, and 1 who had an inferior vena caval filter placed following a PE. In the PC-RPLND group, there were 6 grade III complications: 3 patients developed chylous ascites or lymphocele requiring paracentesis and/or drain placement, 1 patient who had received bleomycin required admission to the intensive care unit for pulmonary edema, and 1 patient developed deep vein thrombosis requiring vena cava filter placement. The only grade IV complication occurred in a PC-RPLND patient who developed lower extremity paraplegia and myonecrosis following infrarenal aortic ligation and aortobifemoral bypass due to aortic rupture from an adherent tumor. This patient also required temporary hemodialysis (grade III) for acute renal failure due to massive blood loss. The only grade V complication occurred in a PC-RPLND patient who developed multiple PEs, respiratory failure, and died on postoperative day 23. 3.3. Late complications (Tables 3 and 6) A total of 8 (7%) patients in the P-RPLND cohort developed complications more than 30 days after RPLND compared to 7 (7%) patients after PC-RPLND (P ⫽ 1). Four of the patients in each needed incisional hernia repairs. One in each group developed small bowel obstruction requiring exploratory laparotomy with bowel resection. One patient developed ureteral obstruction secondary to retroperitoneal fibrosis 22 months after P-RPLND and ultimately required a segmental resection with ureteroureterostomy. Table 4 Intraoperative complications Injured structure

P-RPLND

PC-RPLND

Aorta Common iliac artery Common iliac vein Inferior vena cava Lumbar artery Lumbar vein Renal artery or branch Renal vein Ureter

0 0 0 0 1 1 1 2 1

2 1 1 2 3 0 1 2 0

P-RPLND ⫽ Primary retroperitoneal lymph node dissection; PCRPLND ⫽ Post-chemotherapy RPLND.

V.S. Subramanian et al. / Urologic Oncology: Seminars and Original Investigations 28 (2010) 504 –509

507

Table 5 Postoperative complications (within 30 days of RPLND) by grade Grade

P-RPLND

PC-RPLND

Complication

Number

Complication

Number

I

Ileus with NPO/NG Pulmonary (atelectasis, pneumonia) Infection (wound) C. difficile colitis

14 4 1 1

II

Ileus/SBO with TPN or readmission Deep venous thrombosis Intravenous line sepsis treated with IV antibiotics

6 1 1

10 3 3 2 2 10 2 1

III

Chylous ascites with percutaneous drainage SBO with re-operation

2 1

Pulmonary embolus with IVC filter placement

1

Ileus with NPO/NG Pulmonary (atelectasis, pneumonia) Infection (wound, epidural site) C. difficile colitis Acute renal failure (temporary) Ileus/SBO with TPN or readmission Pulmonary embolus Disseminated intravascular coagulation with post-operative transfusions Fever of unknown origin and readmission Chylous ascites with percutaneous drainage Pulmonary (pneumonia, edema) requiring intensive care unit DVT with IVC filter placement Abdominal abscess with percutaneous drainage Acute renal failure with temporary dialysis Lymphocele with percutaneous drainage Lower extremity paraplegia and calf muscle necrosis following intraoperative infrarenal aortic ligation with aorto-bifemoral bypass Multiple pulmonary emboli resulting in respiratory failure and multisystem organ failure, death

IV

V

1 2 2 1 1 1 1 1

1

P-RPLND ⫽ Primary retroperitoneal lymph node dissection; PC-RPLND ⫽ Post-chemotherapy RPLND; NPO ⫽ Nothing by mouth; NG ⫽ Nasogastric tube; SBO ⫽ Small bowel obstruction; TPN ⫽ Total peripheral nutrition; DVT ⫽ Deep venous thrombosis; IV ⫽ Intravenous; IVC ⫽ Inferior vena cava.

Data on post-RPLND ejaculatory function was available on 70 P-RPLND patients and 54 PC-RPLND patients. Antegrade ejaculation was preserved in 60 (80%) P-RPLND patients compared with 22 (41%) PC-RPLND cases (P ⬍ 0.001). Among patients who had prospective nerve-sparing templates, 42 (91%) P-RPLND and 5 (71%) PC-RPLND patients retained antegrade ejaculation. 3.4. Analysis of predictors of complications We hypothesized that patients undergoing PC-RPLND after first-line chemotherapy may have fewer intraoperative and postoperative complications compared with PCRPLND performed for other indications (post-salvage chemotherapy, advanced seminoma, late recurrence, and

Table 6 Late complications (after 30 days from RPLND) Complication

P-RPLND

PC-RPLND

Incisional hernia Chronic lower extremity edema Small bowel obstruction SBO with bowel resection Ureteral obstruction with ureterectomy/repair Wound dehiscence with chronic sinus

4 0 1 1 1 1

4 1 1 1 0 0

P-RPLND ⫽ Primary retroperitoneal lymph node dissection; PCRPLND ⫽ Post-chemotherapy RPLND; SBO ⫽ Small bowel obstruction.

reoperative RPLND). Though patients numbers in the latter group were small, we identified no significant increase in the risk of intraoperative vascular complications (13% vs. 6%, P ⫽ 0.3) or postoperative complications (34% vs. 26%, P ⫽ 0.5). We also hypothesized that patients receiving bleomycin may experience higher intraoperative complications due to perinodal fibrosis and postoperative complications due to pulmonary complications but no such differences were observed for intraoperative (11% vs. 13%, P ⫽ 0.8) or postoperative complications (35% vs. 20%, P ⫽ 0.3). Patients were also stratified by date of surgery (1982– 1997 vs. 1998 –2007) though no decrease in intraoperative or postoperative complications was noted with more recent cases (data not shown). 4. Discussion RPLND is an important tool in the staging of testis cancer and treatment of residual disease in the retroperitoneum. Previous reports have recognized that primary RPLND for low-stage NSGCT and RPLND after chemotherapy are 2 different entities and should be studied separately [6]. Other differences in PC-RPLND patients such as the primary tumor and types of chemotherapy agents used may also make this a heterogeneous group. With different options for the management of low-stage NSGCT and controversy over the routine application of PC-RPLND, it is critical to understand the various risks from this surgery and

508

V.S. Subramanian et al. / Urologic Oncology: Seminars and Original Investigations 28 (2010) 504 –509

the means to prevent complications. Our series comprises a 25-year experience detailing complications of 112 patients undergoing P-RPLND and 96 PC-RPLND patients with extended follow-up. In our experience, the overall rate of intraoperative complications, postoperative and late complications as well as the severity of postoperative complications was not significantly different between P-RPLND and PC-RPLND. This is important because PC-RPLND is often regarded as a more morbid operation due to the decreased pulmonary, renal, and hematologic reserve as well as higher volume of disease [7] in patients who have undergone chemotherapy. Hospitalization was similar after both P-RPLND and PCRPLND while PC-RPLND was associated with longer operative times, greater intraoperative blood loss, and a higher transfusion rate. As shown in Table 4, the caliber of blood vessels injured in our PC-RPLND patients was generally larger relative to the P-RPLND patients and may have contributed to the higher intraoperative blood loss. Additionally, the difference in intraoperative and postoperative complications may have become statistically significant with a larger sample size. All of these underscore the point that PC-RPLND is a more technically difficult operation but that in expert hands, morbidity can be similar to those who have not had chemotherapy. A 19.6% rate of intraoperative and early postoperative complications from primary nerve-sparing RPLND for 239 patients with CS I NSGCT between 1995 and 2000 was reported by the German Testicular Cancer Study Group with wound infection (5.4%) being most common [3]. Indiana University reported a 10.6% rate of overall complications of P-RPLND in 478 patients between 1982 and 1992 with minimum 1 year follow-up though the average duration was not specified [6]. They stratified complications by “major” resulting in an additional 2 days or more of hospitalization and “minor” with 70% of their complications being “major”. The majority of their complications (43%) were related to small bowel obstruction (SBO) or ileus as compared with 63% of our P-RPLND complications. The same authors also reported a 20.7% complication rate for PC-RPLND in 603 cases performed in the same duration with follow-up interval not specified [1]. The operative mortality of 0.8% (5 patients) was similar to our 1% incidence. Pulmonary complications (8% of all patients) were the most common followed by a 4.8% rate of wound infection. Our incidence of pulmonary complications in PCRPLND including atelectasis, pneumonia, and PE was 7% and our wound complication rate was similarly 4%. As with our P-RPLND patients, SBO and ileus were the most common complication at 45%. Using a similar scale, 26% of the complications were considered “minor” and 74% “major”. Of note, 2% of patients in both the P-RPLND and PCRPLND groups experienced chylous ascites requiring percutaneous drainage as well as TPN. Other series in the literature report rates of 0.2% to 2% for P-RPLND and 2%

with PC-RPLND with higher incidence after vena cava resection [1,3,6,8]. Due to tissue effects, PC-RPLND for a seminoma primary tumor, after secondary or tertiary chemotherapy, or in the setting of reoperative surgery, could be potentially more morbid. Previous studies showed a 24.7% rate of perioperative complications with PC-RPLND in patients with seminoma, which was not found to be significantly different compared with 20.3% in PC-RPLND for NSGCT (P ⫽ 0.29) [9]. Similarly, a series of reoperative retroperitoneal surgeries for NSGCT showed a 27% incidence of perioperative complications [10], not dramatically different than published series of overall PC-RPLND complication rates. In our series, we compared patients undergoing induction chemotherapy prior to RPLND with patients who had PC-RPLND after salvage chemotherapy, for seminoma, for “desperation”, for late recurrence, and for reoperative RPLNDs. We found only a 6% rate of intraoperative complications in the latter group along with a 26% incidence of postoperative complications, not significantly different than the induction chemotherapy cohort, though our sample size was limited. There were only 3 early postoperative complications that were grade III or higher, including chylous ascites and lymphocele drained percutaneously, and a DVT requiring Greenfield filter placement. In an effort to accurately assess the severity of complications in our experience, we used a previously published 5-tiered severity scale to grade postoperative complications [5]. This method uses the level of intervention needed for resolution of the complication and therefore more accurately assesses the impact on the patient. Dividing complications by major or minor as in the aforementioned series is based on the author’s assessment of additional hospitalization rather than objective criteria. As such, the rate of major complications for both P-RPLND and PC-RPLND as reported in the literature is higher than minor complications. Additionally, this criterion does not allow for grading complications manifested after hospital discharge or on re-admission. The majority of complications in our series were grade I (medication or bedside care) or grade II (intravenous therapy). Loss of ejaculation in performing a bilateral template RPLND was an accepted morbidity until the late 1980s when preservation of lumbar sympathetic fibers critical to antegrade propulsion of semen was described. These initial series by Donohue and Jewett noted antegrade ejaculation rates of 90 to 100% for patients undergoing nerve-sparing P-RPLND [11,12]. Later experience with nerve-sparing PCRPLND has yielded ejaculation rates of 76.5% to 89% [13,14]. In both patient populations, this technique can be used to decrease morbidity without increasing local recurrence rates. Overall, 80% of our P-RPLND patients for whom data was available retained normal ejaculation compared with 41% of PC-RPLND patients.

V.S. Subramanian et al. / Urologic Oncology: Seminars and Original Investigations 28 (2010) 504 –509

5. Conclusions P-RPLND and PC-RPLND are both associated with acceptable rates of morbidity, with the majority being low grade, and with negligible risk of perioperative mortality. Overall, the risk and severity of intraoperative and postoperative complications are higher with PC-RPLND though not statistically significantly so in our limited series. Our subgroup analyses did not identify any specific patient population at higher risk for morbidity. There is a significantly longer operative time, EBL, and transfusion rate with PCRPLND. For future series, we propose that a standardized complication grading system be applied to critically assess perioperative outcomes of RPLND in future comparisons.

References [1] Baniel J, Foster RS, Rowland RG, et al. Complications of postchemotherapy retroperitoneal lymph node dissection. J Urol 1995; 153:976 – 80. [2] Donat SM. Standards for surgical complication reporting in urologic oncology: Time for a change. Urology 2007;69:221–5. [3] Heidenreich A, Albers P, Hartmann M, et al. Complications of primary nerve sparing retroperitoneal lymph node dissection for clinical stage I nonseminomatous germ cell tumors of the testis: Experience of the German Testicular Cancer Study Group. J Urol 2003;169: 1710 – 4.

509

[4] Martin RCG, Brennan MF, Jaques DP. Quality of complication reporting in the surgical literature. Ann Surg 2002;235:803–13. [5] Stephenson AJ, Hakimi A, Snyder ME, et al. Complications of radical and partial nephrectomy in a large contemporary cohort. J Urol 2004;171:130 – 4. [6] Baniel J, Foster RS, Rowland RG, et al. Complications of primary retroperitoneal lymph node dissection. J Urol 1994;152:424 –7. [7] Baniel J, Sella A. Complications of retroperitoneal lymph node dissection in testicular cancer: Primary and post-chemotherapy. Semin Surg Oncol 1999;17:263–7. [8] Jewett MA, Wesley-James T. Early and late complications of retroperitoneal lymphadenectomy in testis cancer. Can J Surg 1991;34: 368 –73. [9] Mosharafa AA, Foster RS, Leibovich BC, et al. Is post-chemotherapy resection of seminomatous elements associated with higher acute morbidity? J Urol 2003;169:2126 – 8. [10] McKiernan JM, Motzer RJ, Bajorin DF, et al. Reoperative retroperitoneal surgery for nonseminomatous germ cell tumor: Clinical presentation, patterns of recurrence, and outcome. Urology 2003;62: 732– 6. [11] Jewett MA, Kong YS, Goldberg SD, et al. Retroperitoneal lymphadenectomy for testis tumor with nerve sparing for ejaculation. J Urol 1988;139:1220 – 4. [12] Donohue JP, Foster RS, Rowland RG, et al. Nerve-sparing retroperitoneal lymphadenectomy with preservation of ejaculation. J Urol 1990;144:287–92. [13] Coogan CL, Hejase MJ, Wahle GR, et al. Nerve sparing post-chemotherapy retroperitoneal lymph node dissection for advanced testicular cancer. J Urol 1996;156:1656 – 8. [14] Jacobsen KD, Ous S, Waehre H, et al. Ejaculation in testicular cancer patients after post-chemotherapy retroperitoneal lymph node dissection. Br J Cancer 1999;80:249 –55.