- Email: [email protected]
Novel Technique Prevents Lymphoceles After Transperitoneal Robotic-assisted Pelvic Lymph Node Dissection: Peritoneal Flap Interposition
Surgical Techniques in Urology Novel Technique Prevents Lymphoceles After Transperitoneal Robotic-assisted Pelvic Lymph Node Dissection: Peritoneal Flap Interposition Christopher Lebeis, David Canes, Andrea Sorcini, and Alireza Moinzadeh To determine the efficacy of our novel technique to prevent lymphocele formation after pelvic lymph node dissection (PLND) after robotic-assisted radical prostatectomy (RARP) using the existing peritoneum of the bladder. TECHNICAL We evaluated 155 consecutive patients undergoing RARP with PLND over 24 months. Group A CONSIDERATIONS included the first 77 patients with PLND using standard technique (no peritoneal flap). Group B included the subsequent 78 patients (1 patient excluded) with PLND and peritoneal interposition flap. The peritoneal interposition flap is created by rotating and advancing the peritoneum around the lateral surface of the ipsilateral bladder to the dependent portion of the pelvis and fixing it to the bladder itself. A cystogram was performed in 91% of the patients 7-14 days after the surgery. Lymphocele formation rates were compared (based on symptoms, cystogram findings, and radiographic confirmation). RESULTS The 2 groups were statistically equivalent in terms of prostate-specific antigen, age, blood loss, body mass index, Gleason score, prostate size, pathology, or heparin use. Lymphocele formation occurred in 9 of 77 (11.6%) group A patients and in 0 of 77 group B patients (P ¼ .003). Mean time to lymphocele detection in group A was 30.4 days. Mean follow-up in groups A and B were 383.97 and 379 days, respectively (P ¼ .91). CONCLUSION Strategic rotation and fixation of a peritoneal flap around the lateral aspect of the bladder during transperitoneal RARP with PLND is a novel technique to prevent lymphocele formation. Given the sample size and single institutional study, a prospective, randomized, multi-institutional trial is planned. UROLOGY 85: 1505e1509, 2015. 2015 Elsevier Inc.
INTRODUCTION
A
s urologists have gained more experience with robotic-assisted radical prostatectomy (RARP), the number and complexity of prostatectomy procedures have increased. The 2 most common indications for pelvic lymph node dissection (PLND) are intermediate risk stratification based on the American Urological Association’s prostate cancer risk classification (prostate-specific antigen level, 10-20 ng/mL or a Gleason score of 7 or clinical stage T2b)1 and/or a 2% probability of lymph node invasion based on nomograms developed at Memorial Sloan Kettering Cancer Center (MSKCC).2,3 Robotic PLND is an established technique to determine nodal status of prostate cancer patients.4 Although robotic PLND safety has been determined, complications may occur. Lymphocele formation is the most common Financial Disclosure: The authors declare that they have no relevant financial interests. From the Institute of Urology, Lahey Hospital and Medical Center, Burlington, MA Address correspondence to: Alireza Moinzadeh, M.D., Institute of Urology, Lahey Hospital and Medical Center, 41 Mall Road, Burlington, MA 01805. E-mail: Alireza. [email protected] Submitted: September 5, 2014, accepted (with revisions): February 27, 2015
ª 2015 Elsevier Inc. All Rights Reserved
postoperative complication of PLND.5,6 Most are clinically insignificant and resolve spontaneously; however, 2%-15% will require percutaneous or surgical drainage.4,7-9 In cases of symptomatic lymphoceles, patients typically present with lower abdominal pain, lower urinary tract symptoms, fevers, lower extremity swelling, or deep vein thrombosis. These symptoms and the treatment of the lymphocele can cause a significant amount of morbidity to the patient. A postoperative cystogram, commonly used to detect anastomotic leak, may demonstrate the presence of a lymphocele as it compresses the bladder and causes distortion of the bladder contour (Fig. 1A). Most patients with a symptomatic lymphocele will require either percutaneous drainage10 or laparoscopic marsupialization.11 Based on the computed tomography (CT) appearance of lymphoceles (Fig. 1B) and the scarcity of lymphoceles after cystectomy, we hypothesize that the edges of the peritoneum incised just lateral to the medial obliterated ligament reapproximate, thus sealing off the PLND bed from the peritoneal cavity. Additionally, the bladder http://dx.doi.org/10.1016/j.urology.2015.02.034 0090-4295/15
1505
Figure 1. (A) Postoperative cystogram in a patient who had bilateral lymphoceles. Note the hourglass shape of the bladder from the extrinsic compression. (B) Borders of a lymphocele. The yellow line outlines the lymphocele cavity. Medial lymphocele wall abuts perivesical adipose tissue. The lateral wall is the pelvic sidewall. Superiorly is the overlying peritoneum, indicated by the green line. B, bladder; L, lymphocele.
forms the medial wall of a lymphocele cavity. Therefore, we proposed that sliding a peritoneal surface between the lateral wall of the bladder and the lymphadenectomy bed might prevent lymphoceles from forming in this location. Instead of allowing the bladder to scar over the lymphadenectomy bed, thereby necessitating fenestration of the peritoneum when the lymphocele occurs, we developed a novel technique at the time of PLND to preemptively create a peritoneal window using the existing peritoneum overlying the bladder. Our hypothesis was that our technique should prevent the bladder from excluding the lymphadenectomy bed, allowing a pathway for continuous drainage of lymphatic fluid into the peritoneal cavity and thus prevent lymphocele formation.
METHODS Data were reviewed using our institutional review boarde approved prospectively collected radical prostatectomy database. The study consisted of 155 consecutive patients who underwent RARP with PLND during a 24-month period. We created the peritoneal interposition flap in 78 sequential patients and retrospectively compared it with the previous 77 patients. Patients were included if they had biopsy-proven prostate cancer with intermediate- or high-risk features according to the D’Amico risk stratification and were candidates for concurrent PLND.12 Exclusion criteria included prior prostate irradiation. Preoperative CT, magnetic resonance imaging, or bone scans were obtained on a case-by-case basis. All patients received 5000 IU of subcutaneous heparin before the surgery began, and intermittent leg compression devices were used during the entire operation. All operations were performed at a single institution by 2 surgeons. PLND was performed either bilaterally or unilaterally based on the surgeon’s discretion at the beginning of the case and patients were not excluded if only a unilateral dissection was preformed. At a minimum, an obturator PLND with defined boundaries (pubic bone, lateral side wall, obturator nerve, and external iliac vein) was completed. A combination of electrocautery and Hem-o-lok clips (Teleflex Medical) were used. All PLNDs were performed by senior staff. At the end of the procedure, a single Jackson-Pratt drain was placed for all patients. 1506
Of the 155 consecutive patients, 154 were included. In the first 77 patients (group A), PLND was performed and then the RARP was completed. No peritoneal flap was created. In the subsequent 77 patients (group B), PLND and RARP were completed using the exact same technique and materials as the previous group, but we created the peritoneal interposition flap at the end of the case. The peritoneal flap, developed by dropping the bladder from the abdominal wall, was used after completion of the vesicourethral anastomosis (Fig. 2A). This flap uses the redundant peritoneal surface that usually covers the preperitoneal space, and this is entered when dropping the bladder down. The flap (Fig. 2B point A) was brought to the most dependent portion of the pelvis (posterior and caudal) and fixed to the lateral aspect of the bladder (Fig. 2B point B) using interrupted VICRYL sutures (Ethicon). At least 2 sutures were placed, with additional sutures placed as needed, so that the peritoneal surface would face the adjacent lymph node bed. This flap covered the perivesical adipose tissue, thus preventing contact with the PLND bed (Fig. 2C; additional Supplementary Video). The flap is completed at this point if performing it unilaterally. Otherwise, it is repeated on the other side in the exact same fashion. Additionally, the apex of the flap is secured to the ventral perivesical adipose tissue when both sides have a flap created to anchor them in place. The tissue used for the flaps does not need to be split, and there is more than enough tissue to complete the flap on both sides. The pneumoperitoneum can be reduced to ensure that the perivesical fat does not come in contact with the lymphadenectomy bed once the sutures have been placed. Theoretically, this window prevented the bladder from scarring over the lymphadenectomy bed and allowed for continuous egress of lymphatic fluid into the peritoneal cavity to be reabsorbed if a lymph “leak” occurs. Postoperative management was not different between the 2 groups. All patients continued on 5000 IU subcutaneous heparin every 12 hours until discharged from the hospital. On postoperative day 0, patients were instructed to ambulate and were not given anything by mouth. On postoperative day 1, patients were given a clear liquid diet, the Jackson-Pratt drain was removed, and patients were transitioned to oral pain medication with home discharge, if criteria were met. Patients were seen for follow-up in 7-12 days after surgery, and the majority (140 of 154 patients, 91%) had a cystogram to assess for any signs of leak or evidence of a lymphocele. If no UROLOGY 85 (6), 2015
Figure 2. (A) In this illustration, the bladder has been dropped down, and the urethral anastomosis has been completed. The pelvic lymph node dissection has been carried out on the patient’s left side only. The peritoneal flap is under the bladder, lying in the abdomen (not seen). (B) The peritoneal flap is used by folding the lateral aspect of the peritoneum (point A) around the bladder laterally to the adipose tissue of the bladder in the dependent portion of the pelvis (point B). The arrow indicates the course of the flap. Letters C and D represent right-side markings not used in this case. (C) The tacking sutures have been placed and this brings the flap into place in the pelvis as indicated by point A of the flap being sutured to point B on the lateral surface of the bladder. The peritoneal surface is adjacent to the area of the pelvic lymph node dissection, and the perivesical adipose tissue is covered. Letter I represents middle of peritoneum. Letters C and D represent right-side markings not used in this case. leak was seen, the Foley catheter was removed. If there was evidence of a lymphocele, either an ultrasound or an abdominalpelvis CT was performed for confirmation. Lymphocele rates were compared based on patient symptoms, initial cystogram findings, and radiographic confirmation with a combination of abdominal-pelvic CT and/or pelvic ultrasound. Initial cystogram findings that are consistent with a lymphocele can be seen in Figure 1A, which shows the distortion of the bladder contour caused by the extrinsic compression of the lymphocele. If the patient experienced symptoms that were suspicious for a lymphocele (lower abdominal pain, lower urinary tract symptoms, fevers, lower extremity swelling, or deep vein thrombosis), they would undergo a confirmatory radiologic study to visualize the lymphocele. A representative radiograph of a lymphocele can be seen in Figure 1A. To detect a 10% difference between the 2 groups, at least 75 patients would be needed in each group assuming a power of 80 and an alpha value of .05. Continuous variables were reported as a mean and were compared using an unpaired t test. Categorical data were compared using a 2-tailed Fisher exact test. A P value of <.05 was considered significant.
RESULTS One patient, in group B, was excluded given prior brachytherapy. This patient did not develop a lymphocele or have any other perioperative complications. Patient demographics are summarized in Table 1. The 2 groups were statistically equivalent in terms of prostate-specific antigen, age, estimated blood loss, body mass index, Gleason score, prostate size, proportion that had a unilateral PLND, pathology, and heparin use. There was not a significant difference in the duration of follow-up (group A mean, 383.97 days; group B mean, 379 days; P ¼ .9106). Complications were comparable between the 2 groups with the exception of clinically significant lymphoceles (Table 2). Lymphocele formation occurred in 9 of 77 (11.6%) group A patients and 0 of 77 group B patients (P ¼ .0033, Fisher exact test). The average time UROLOGY 85 (6), 2015
to lymphocele detection in group A was 30.4 days (range, 6-72 days). Cystogram demonstrated a lymphocele in 6 of 9 patients (66.7%) who had a clinically significant lymphocele. The cystograms showed significant distortion of the bladder contour. In all these cases, patients had accompanying symptoms either at time of cystogram (4 patients) or at a later date in the remaining 2 patients. The remaining lymphoceles were identified by patients’ symptoms and were confirmed radiographically. All 9 patients with a clinically significant lymphocele underwent percutaneous drainage; 2 of these patients failed less-invasive drainage and needed laparoscopic fenestration. Post hoc power analysis yielded a value of 86.7% for this study, using the incidence of clinically significant lymphoceles (11.6%) and number of subjects in this study (77 per group) with an alpha value of 0.05.
COMMENT Lymphoceles are the most common complication for patients undergoing PLND.6 They may cause a significant amount of morbidity when they are symptomatic. Although our study focused on clinically significant lymphoceles, there is an even higher rate of occult lymphoceles. The incidence of asymptomatic lymphoceles has been reported to be from 30.4% to 51%.7,8,13 There are fewer studies that have looked at robotic PLND, and after reviewing these reports, symptomatic lymphoceles occur in 2%-7% of patients after a robotic PLND.4,5,8,14,15 We believe our higher rate of lymphoceles seen in group A is partially attributable to the heparin that all our patients received preoperative and twice a day postoperatively. However, both groups received the same perioperative care and deep vein thrombosis (DVT) prophylaxis. More patients may choose active surveillance for low-risk prostate cancer; however, patients undergoing RARP with PLND for 1507
Table 1. Patient demographics and perioperative data Demographics Age (y) PSA (ng/dL) EBL (cc) BMI (kg/m2) Final Gleason score sum Prostate size (cc) Number of lymph nodes resected, n Number of patients with positive lymph nodes, n pT2, n pT3, n pT4, n Follow-up (d) Length of stay (d) Bilateral PLND,y n Unilateral PLND,y n
P Value*
Group A (Mean)
Group B (Mean)
60.74 7.08 201.3 29.33 7.08 48.36 4.19
60.41 7.12 235.36 28.89 7.04 50.87 3.81
.77 .96 .12 .52 .70 .39 .42
2
2
1.00
48 28 1 383.97 1.25 45 32
46 31 1 379 1.14 39 38
.87 .74 1.00 .9106 .24 .42 .42
BMI, body mass index; EBL, estimated blood loss; PLND, pelvic lymph node dissection; PSA, prostate-specific antigen. * Compared by the unpaired t test, P <.05 was considered significant. The number of lymph nodes and pathologic stage were compared by the 2-tailed Fisher exact test. y Compared by the 2-tailed Fisher exact test, P <.05 was considered significant.
Table 2. Patient complications Complications
Group A, Group B, n P n (%) (%) Value*
Lymphocele Procedure for lymphocele Wound infection Cerebrovascular accident Scrotal pain Urinary tract infection Superficial bleed Urinary leak Deep vein thrombosis Ileus Urinary retention
9 9 2 1 0 2 0 5 3 0 1
(11.6) (11.6) (2.59) (1.29) (0) (2.59) (0) (6.49) (3.89) (0) (1.29)
0 0 4 0 1 1 1 3 0 1 0
(0) (0) (5.71) (0) (1.42) (1.42) (1.42) (4.28) (0) (1.42) (0)
.0033 .0033 .4246 1.0000 .4762 1.0000 .4762 .7215 .2466 .4762 1.0000
* Compared by the unpaired t test, P <.05 was considered significant. The number of lymph nodes and pathologic stage were compared by the 2-tailed Fisher exact test.
intermediate- and high-risk disease will continue to be at risk for this complication. Several techniques have been studied to reduce the incidence of lymphoceles. Scholz et al16 used fibrin glue for patients with gynecologic malignancies undergoing PLND, but this did not demonstrate a reduction in lymphocele formation. Kim et al17 used octreotide in an attempt to reduce the amount of lymphorrhea formed by the PLND, but there was no significant reduction in lymphocele formation. Several studies have shown a lower incidence of lymphoceles in patients who undergo transperitoneal PLssscompared with that of traditional open or extraperitoneal approaches.13 The initial peritoneotomy is likely the main reason for the decreased incidence of lymphocele formation during a transperitoneal PLND. 1508
The opening created during the transperitoneal approach allows the lymphatic fluid to drain away from the pelvis and into the abdomen. Stolzenburg et al18 demonstrated this theory by performing bilateral peritoneal fenestrations with extraperitoneal PLND and noting a significant decrease in the number of lymphoceles formed. This supports our hypothesis as it demonstrates that this window that is originally created performing the PLND transperitoneally may seal off and thus trap the lymphatic fluid in the pelvis. We surmise that after release of pneumoperitoneum during a transperitoneal RARP with PLND, the perivesical fat adheres over the PLND bed, thereby sealing it off. This creates a closed space where lymphatic fluid accumulates, with no access to the peritoneal space for reabsorption, ultimately resulting in lymphocele formation. As demonstrated on the CT scan (Fig. 1B), the perivesical fat of the bladder forms the medial aspect of the lymphocele and the peritoneum is the superior border, with the pelvic sidewall forming the lateral aspect. Although many symptomatic lymphoceles are managed by percutaneous drainage or with conservative management, some require laparoscopic fenestration to create a peritoneal window. Laparoscopic fenestration with creation of a window served as the impetus to preemptively rotate and secure a peritoneal flap in this location, decreasing the likelihood of closure of this space. Our goal is to create a pathway lined by peritoneum and direct the lymphatic fluid out of the pelvis and into the peritoneal cavity where it can be absorbed. The peritoneal flap is obtained by dropping the bladder from the abdominal wall at the beginning of the procedure, as there is more than sufficient redundant peritoneal surface that is used for one or both sides. The lateral aspect of the flap, lateral to the obliterated ligament, is brought into the dependent part of the pelvis and is fixed onto the lateral aspect of the bladder. If the patient undergoes bilateral PLND, this is then repeated on the contralateral side to create bilateral flaps, and there is no need to split the tissue. We also anchor this in place with a stitch that attaches it to the ventral perivesical adipose tissue. As this flap is created from redundant tissue, it does not pull the bladder over to the side of the flap. Two important goals are accomplished by this technique. First, the adipose tissue no longer opposes the lymphadenectomy bed because the peritoneal flap is interposed between the bladder fat and the pelvic sidewall. Second, this window allows drainage of the lymph fluid drain into the peritoneal cavity to be reabsorbed. Interposition of peritoneum is not technically demanding and typically requires w5-7 minutes using only 2 additional VICRYL sutures (Ethicon). Despite our findings, this study has some limitations. It is a nonrandomized and retrospective study. Although this is a retrospective study, there were no changes to surgical technique in terms of the PLND, and the groups had no significant difference in patient demographics. Additionally, the study was completed at a single UROLOGY 85 (6), 2015
institution by 2 surgeons. As such, extrapolation to other institutions may be limited. However, the technique was easily transferred between one surgeon to the next after its development by the senior author (A.M.). It is not likely that a “learning curve” played a role in reduction of the lymphocele as each surgeon had performed >400 RARP before the study date. Overall, there was a low lymph node yield compared with that of other studies, but we routinely do not perform an extended PLND and 45% were unilateral PLND (group A, 42%; group B, 49%; P ¼ .42). This technique is not applicable to extraperitoneal cases. We do not know the long-term effects of the flap on continence or other unforeseen consequences, but the short-term complication rate was comparable with a standard RARP (group A). Postoperative lymphocele evaluation was limited by clinical diagnosis. The cystogram served as a surrogate for detection and found 66.7% of the lymphoceles. A cystogram is not the best imaging study to evaluate the presence of a lymphocele, and a CT or ultrasound would more accurately detect an asymptomatic lymphocele. Our goal in this study, however, was to detect symptomatic lymphoceles. However, this limitation in detection with the cystogram and not a CT or ultrasound existed in both groups and is therefore not likely to influence data interpretation. We are currently planning a multiinstitutional, randomized, controlled study. In our subsequent study, all patients will receive a postoperative ultrasound to better evaluate for a lymphocele and this will improve the limitations of our present work.
CONCLUSION Peritonealization of the lateral aspect of the bladder with a peritoneal interposition flap is a novel technique to preemptively create a permanent pathway for lymph drainage into the peritoneal space after RARP or PLND. In this preliminary report, no patients with peritoneal interposition flaps developed symptomatic lymphoceles after transperitoneal RARP or PLND. Additional studies are warranted, as this technique may be applicable to all patients undergoing transperitoneal RARP with PLND. Acknowledgment. Vinald Francis, Lahey Hospital and Medical Center, for illustration. References 1. Prostate Cancer Clinical Guideline Update Panel Prostate cancer Guideline for the Management of Clinically Localized Prostate Cancer: 2007 update. American Urological Association Education and Research, Inc. Available at: http://www.auanet.org/content/ guidelines-and-quality-care/clinical-guidelines/main-reports/proscan07/ content.pdf. Accessed December 9, 2014.
UROLOGY 85 (6), 2015
2. Cagiannos I, Karakiewicz P, Eastham JA, et al. A preoperative nomogram identifying decreased risk of positive pelvic lymph nodes in patients with prostate cancer. J Urol. 2003;170:1798-1803. 3. Touijer KA, Ahallal Y, Guillonneau BD. Indications for and anatomical extent of pelvic lymph node dissection for prostate cancer: practice patterns of uro-oncologists in North America. Urol Oncol. 2013;31:1517-1521; e1-e2. 4. Zorn KC, Katz MH, Bernstein A, et al. Pelvic lymphadenectomy during robot-assisted radical prostatectomy: assessing nodal yield, perioperative outcomes, and complications. Urology. 2009;74: 296-302. 5. Feicke A, Baumgartner M, Talimi S, et al. Robotic-assisted laparoscopic extended pelvic lymph node dissection for prostate cancer: surgical technique and experience with the first 99 cases. Eur Urol. 2009;55:876-883. 6. Musch M, Klevecka V, Roggenbuck U, et al. Complications of pelvic lymphadenectomy in 1,380 patients undergoing radical retropubic prostatectomy between 1993 and 2006. J Urol. 2008;179:923-929. 7. Freid RM, Siegel D, Smith AD, et al. Lymphoceles after laparoscopic pelvic node dissection. URL. 1998;51:131-134. 8. Orvieto MA, Coelho RF, Chauhan S, et al. Incidence of lymphoceles after robot-assisted pelvic lymph node dissection. BJU Int. 2011;108:1185-1190. 9. Sogani PC, Watson RC, Whitmore WF. Lymphocele after pelvic lymphadenectomy for urologic cancer. URL. 1981;17:39-43. 10. Kim JK, Jeong YY, Kim YH, et al. Postoperative pelvic lymphocele: treatment with simple percutaneous catheter drainage. Radiology. 1999;212:390-394. 11. Fallick ML, Long JP. Laparoscopic marsupialization of lymphocele after laparoscopic lymph node dissection. J Endourol. 1996;10: 533-534. 12. D’Amico AV, Whittington R, Malkowicz SB, et al. Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA. 1998;280:969-974. 13. Solberg A, Angelsen A, Bergan U, et al. Frequency of lymphoceles after open and laparoscopic pelvic lymph node dissection in patients with prostate cancer. Scand J Urol Nephrol. 2003;37: 218-221. 14. Raheem OA, Bazzi WM, Parsons JK, et al. Management of pelvic lymphoceles following robot-assisted laparoscopic radical prostatectomy. Urol Ann. 2012;4:111-114. 15. Liss MA, Palazzi K, Stroup SP, et al. Outcomes and complications of pelvic lymph node dissection during robotic-assisted radical prostatectomy. World J Urol. 2013;31:481-488. 16. Scholz HS, Petru E, Benedicic C, et al. Fibrin application for preventing lymphocysts after retroperitoneal lymphadenectomy in patients with gynecologic malignancies. Gynecol Oncol. 2002;84: 43-46. 17. Kim WT, Ham WS, Koo KC, et al. Efficacy of octreotide for management of lymphorrhea after pelvic lymph node dissection in radical prostatectomy. Urology. 2010;76:398-401. 18. Stolzenburg JU, Wasserscheid J, Rabenalt R, et al. Reduction in incidence of lymphocele following extraperitoneal radical prostatectomy and pelvic lymph node dissection by bilateral peritoneal fenestration. World J Urol. 2008;26:581-586.
Video clips cited in this article can be found on the internet at: http://dx.doi.org/10.1016/ j.urology.2015.02.034.
1509
INTRODUCTION
A
s urologists have gained more experience with robotic-assisted radical prostatectomy (RARP), the number and complexity of prostatectomy procedures have increased. The 2 most common indications for pelvic lymph node dissection (PLND) are intermediate risk stratification based on the American Urological Association’s prostate cancer risk classification (prostate-specific antigen level, 10-20 ng/mL or a Gleason score of 7 or clinical stage T2b)1 and/or a 2% probability of lymph node invasion based on nomograms developed at Memorial Sloan Kettering Cancer Center (MSKCC).2,3 Robotic PLND is an established technique to determine nodal status of prostate cancer patients.4 Although robotic PLND safety has been determined, complications may occur. Lymphocele formation is the most common Financial Disclosure: The authors declare that they have no relevant financial interests. From the Institute of Urology, Lahey Hospital and Medical Center, Burlington, MA Address correspondence to: Alireza Moinzadeh, M.D., Institute of Urology, Lahey Hospital and Medical Center, 41 Mall Road, Burlington, MA 01805. E-mail: Alireza. [email protected] Submitted: September 5, 2014, accepted (with revisions): February 27, 2015
ª 2015 Elsevier Inc. All Rights Reserved
postoperative complication of PLND.5,6 Most are clinically insignificant and resolve spontaneously; however, 2%-15% will require percutaneous or surgical drainage.4,7-9 In cases of symptomatic lymphoceles, patients typically present with lower abdominal pain, lower urinary tract symptoms, fevers, lower extremity swelling, or deep vein thrombosis. These symptoms and the treatment of the lymphocele can cause a significant amount of morbidity to the patient. A postoperative cystogram, commonly used to detect anastomotic leak, may demonstrate the presence of a lymphocele as it compresses the bladder and causes distortion of the bladder contour (Fig. 1A). Most patients with a symptomatic lymphocele will require either percutaneous drainage10 or laparoscopic marsupialization.11 Based on the computed tomography (CT) appearance of lymphoceles (Fig. 1B) and the scarcity of lymphoceles after cystectomy, we hypothesize that the edges of the peritoneum incised just lateral to the medial obliterated ligament reapproximate, thus sealing off the PLND bed from the peritoneal cavity. Additionally, the bladder http://dx.doi.org/10.1016/j.urology.2015.02.034 0090-4295/15
1505
Figure 1. (A) Postoperative cystogram in a patient who had bilateral lymphoceles. Note the hourglass shape of the bladder from the extrinsic compression. (B) Borders of a lymphocele. The yellow line outlines the lymphocele cavity. Medial lymphocele wall abuts perivesical adipose tissue. The lateral wall is the pelvic sidewall. Superiorly is the overlying peritoneum, indicated by the green line. B, bladder; L, lymphocele.
forms the medial wall of a lymphocele cavity. Therefore, we proposed that sliding a peritoneal surface between the lateral wall of the bladder and the lymphadenectomy bed might prevent lymphoceles from forming in this location. Instead of allowing the bladder to scar over the lymphadenectomy bed, thereby necessitating fenestration of the peritoneum when the lymphocele occurs, we developed a novel technique at the time of PLND to preemptively create a peritoneal window using the existing peritoneum overlying the bladder. Our hypothesis was that our technique should prevent the bladder from excluding the lymphadenectomy bed, allowing a pathway for continuous drainage of lymphatic fluid into the peritoneal cavity and thus prevent lymphocele formation.
METHODS Data were reviewed using our institutional review boarde approved prospectively collected radical prostatectomy database. The study consisted of 155 consecutive patients who underwent RARP with PLND during a 24-month period. We created the peritoneal interposition flap in 78 sequential patients and retrospectively compared it with the previous 77 patients. Patients were included if they had biopsy-proven prostate cancer with intermediate- or high-risk features according to the D’Amico risk stratification and were candidates for concurrent PLND.12 Exclusion criteria included prior prostate irradiation. Preoperative CT, magnetic resonance imaging, or bone scans were obtained on a case-by-case basis. All patients received 5000 IU of subcutaneous heparin before the surgery began, and intermittent leg compression devices were used during the entire operation. All operations were performed at a single institution by 2 surgeons. PLND was performed either bilaterally or unilaterally based on the surgeon’s discretion at the beginning of the case and patients were not excluded if only a unilateral dissection was preformed. At a minimum, an obturator PLND with defined boundaries (pubic bone, lateral side wall, obturator nerve, and external iliac vein) was completed. A combination of electrocautery and Hem-o-lok clips (Teleflex Medical) were used. All PLNDs were performed by senior staff. At the end of the procedure, a single Jackson-Pratt drain was placed for all patients. 1506
Of the 155 consecutive patients, 154 were included. In the first 77 patients (group A), PLND was performed and then the RARP was completed. No peritoneal flap was created. In the subsequent 77 patients (group B), PLND and RARP were completed using the exact same technique and materials as the previous group, but we created the peritoneal interposition flap at the end of the case. The peritoneal flap, developed by dropping the bladder from the abdominal wall, was used after completion of the vesicourethral anastomosis (Fig. 2A). This flap uses the redundant peritoneal surface that usually covers the preperitoneal space, and this is entered when dropping the bladder down. The flap (Fig. 2B point A) was brought to the most dependent portion of the pelvis (posterior and caudal) and fixed to the lateral aspect of the bladder (Fig. 2B point B) using interrupted VICRYL sutures (Ethicon). At least 2 sutures were placed, with additional sutures placed as needed, so that the peritoneal surface would face the adjacent lymph node bed. This flap covered the perivesical adipose tissue, thus preventing contact with the PLND bed (Fig. 2C; additional Supplementary Video). The flap is completed at this point if performing it unilaterally. Otherwise, it is repeated on the other side in the exact same fashion. Additionally, the apex of the flap is secured to the ventral perivesical adipose tissue when both sides have a flap created to anchor them in place. The tissue used for the flaps does not need to be split, and there is more than enough tissue to complete the flap on both sides. The pneumoperitoneum can be reduced to ensure that the perivesical fat does not come in contact with the lymphadenectomy bed once the sutures have been placed. Theoretically, this window prevented the bladder from scarring over the lymphadenectomy bed and allowed for continuous egress of lymphatic fluid into the peritoneal cavity to be reabsorbed if a lymph “leak” occurs. Postoperative management was not different between the 2 groups. All patients continued on 5000 IU subcutaneous heparin every 12 hours until discharged from the hospital. On postoperative day 0, patients were instructed to ambulate and were not given anything by mouth. On postoperative day 1, patients were given a clear liquid diet, the Jackson-Pratt drain was removed, and patients were transitioned to oral pain medication with home discharge, if criteria were met. Patients were seen for follow-up in 7-12 days after surgery, and the majority (140 of 154 patients, 91%) had a cystogram to assess for any signs of leak or evidence of a lymphocele. If no UROLOGY 85 (6), 2015
Figure 2. (A) In this illustration, the bladder has been dropped down, and the urethral anastomosis has been completed. The pelvic lymph node dissection has been carried out on the patient’s left side only. The peritoneal flap is under the bladder, lying in the abdomen (not seen). (B) The peritoneal flap is used by folding the lateral aspect of the peritoneum (point A) around the bladder laterally to the adipose tissue of the bladder in the dependent portion of the pelvis (point B). The arrow indicates the course of the flap. Letters C and D represent right-side markings not used in this case. (C) The tacking sutures have been placed and this brings the flap into place in the pelvis as indicated by point A of the flap being sutured to point B on the lateral surface of the bladder. The peritoneal surface is adjacent to the area of the pelvic lymph node dissection, and the perivesical adipose tissue is covered. Letter I represents middle of peritoneum. Letters C and D represent right-side markings not used in this case. leak was seen, the Foley catheter was removed. If there was evidence of a lymphocele, either an ultrasound or an abdominalpelvis CT was performed for confirmation. Lymphocele rates were compared based on patient symptoms, initial cystogram findings, and radiographic confirmation with a combination of abdominal-pelvic CT and/or pelvic ultrasound. Initial cystogram findings that are consistent with a lymphocele can be seen in Figure 1A, which shows the distortion of the bladder contour caused by the extrinsic compression of the lymphocele. If the patient experienced symptoms that were suspicious for a lymphocele (lower abdominal pain, lower urinary tract symptoms, fevers, lower extremity swelling, or deep vein thrombosis), they would undergo a confirmatory radiologic study to visualize the lymphocele. A representative radiograph of a lymphocele can be seen in Figure 1A. To detect a 10% difference between the 2 groups, at least 75 patients would be needed in each group assuming a power of 80 and an alpha value of .05. Continuous variables were reported as a mean and were compared using an unpaired t test. Categorical data were compared using a 2-tailed Fisher exact test. A P value of <.05 was considered significant.
RESULTS One patient, in group B, was excluded given prior brachytherapy. This patient did not develop a lymphocele or have any other perioperative complications. Patient demographics are summarized in Table 1. The 2 groups were statistically equivalent in terms of prostate-specific antigen, age, estimated blood loss, body mass index, Gleason score, prostate size, proportion that had a unilateral PLND, pathology, and heparin use. There was not a significant difference in the duration of follow-up (group A mean, 383.97 days; group B mean, 379 days; P ¼ .9106). Complications were comparable between the 2 groups with the exception of clinically significant lymphoceles (Table 2). Lymphocele formation occurred in 9 of 77 (11.6%) group A patients and 0 of 77 group B patients (P ¼ .0033, Fisher exact test). The average time UROLOGY 85 (6), 2015
to lymphocele detection in group A was 30.4 days (range, 6-72 days). Cystogram demonstrated a lymphocele in 6 of 9 patients (66.7%) who had a clinically significant lymphocele. The cystograms showed significant distortion of the bladder contour. In all these cases, patients had accompanying symptoms either at time of cystogram (4 patients) or at a later date in the remaining 2 patients. The remaining lymphoceles were identified by patients’ symptoms and were confirmed radiographically. All 9 patients with a clinically significant lymphocele underwent percutaneous drainage; 2 of these patients failed less-invasive drainage and needed laparoscopic fenestration. Post hoc power analysis yielded a value of 86.7% for this study, using the incidence of clinically significant lymphoceles (11.6%) and number of subjects in this study (77 per group) with an alpha value of 0.05.
COMMENT Lymphoceles are the most common complication for patients undergoing PLND.6 They may cause a significant amount of morbidity when they are symptomatic. Although our study focused on clinically significant lymphoceles, there is an even higher rate of occult lymphoceles. The incidence of asymptomatic lymphoceles has been reported to be from 30.4% to 51%.7,8,13 There are fewer studies that have looked at robotic PLND, and after reviewing these reports, symptomatic lymphoceles occur in 2%-7% of patients after a robotic PLND.4,5,8,14,15 We believe our higher rate of lymphoceles seen in group A is partially attributable to the heparin that all our patients received preoperative and twice a day postoperatively. However, both groups received the same perioperative care and deep vein thrombosis (DVT) prophylaxis. More patients may choose active surveillance for low-risk prostate cancer; however, patients undergoing RARP with PLND for 1507
Table 1. Patient demographics and perioperative data Demographics Age (y) PSA (ng/dL) EBL (cc) BMI (kg/m2) Final Gleason score sum Prostate size (cc) Number of lymph nodes resected, n Number of patients with positive lymph nodes, n pT2, n pT3, n pT4, n Follow-up (d) Length of stay (d) Bilateral PLND,y n Unilateral PLND,y n
P Value*
Group A (Mean)
Group B (Mean)
60.74 7.08 201.3 29.33 7.08 48.36 4.19
60.41 7.12 235.36 28.89 7.04 50.87 3.81
.77 .96 .12 .52 .70 .39 .42
2
2
1.00
48 28 1 383.97 1.25 45 32
46 31 1 379 1.14 39 38
.87 .74 1.00 .9106 .24 .42 .42
BMI, body mass index; EBL, estimated blood loss; PLND, pelvic lymph node dissection; PSA, prostate-specific antigen. * Compared by the unpaired t test, P <.05 was considered significant. The number of lymph nodes and pathologic stage were compared by the 2-tailed Fisher exact test. y Compared by the 2-tailed Fisher exact test, P <.05 was considered significant.
Table 2. Patient complications Complications
Group A, Group B, n P n (%) (%) Value*
Lymphocele Procedure for lymphocele Wound infection Cerebrovascular accident Scrotal pain Urinary tract infection Superficial bleed Urinary leak Deep vein thrombosis Ileus Urinary retention
9 9 2 1 0 2 0 5 3 0 1
(11.6) (11.6) (2.59) (1.29) (0) (2.59) (0) (6.49) (3.89) (0) (1.29)
0 0 4 0 1 1 1 3 0 1 0
(0) (0) (5.71) (0) (1.42) (1.42) (1.42) (4.28) (0) (1.42) (0)
.0033 .0033 .4246 1.0000 .4762 1.0000 .4762 .7215 .2466 .4762 1.0000
* Compared by the unpaired t test, P <.05 was considered significant. The number of lymph nodes and pathologic stage were compared by the 2-tailed Fisher exact test.
intermediate- and high-risk disease will continue to be at risk for this complication. Several techniques have been studied to reduce the incidence of lymphoceles. Scholz et al16 used fibrin glue for patients with gynecologic malignancies undergoing PLND, but this did not demonstrate a reduction in lymphocele formation. Kim et al17 used octreotide in an attempt to reduce the amount of lymphorrhea formed by the PLND, but there was no significant reduction in lymphocele formation. Several studies have shown a lower incidence of lymphoceles in patients who undergo transperitoneal PLssscompared with that of traditional open or extraperitoneal approaches.13 The initial peritoneotomy is likely the main reason for the decreased incidence of lymphocele formation during a transperitoneal PLND. 1508
The opening created during the transperitoneal approach allows the lymphatic fluid to drain away from the pelvis and into the abdomen. Stolzenburg et al18 demonstrated this theory by performing bilateral peritoneal fenestrations with extraperitoneal PLND and noting a significant decrease in the number of lymphoceles formed. This supports our hypothesis as it demonstrates that this window that is originally created performing the PLND transperitoneally may seal off and thus trap the lymphatic fluid in the pelvis. We surmise that after release of pneumoperitoneum during a transperitoneal RARP with PLND, the perivesical fat adheres over the PLND bed, thereby sealing it off. This creates a closed space where lymphatic fluid accumulates, with no access to the peritoneal space for reabsorption, ultimately resulting in lymphocele formation. As demonstrated on the CT scan (Fig. 1B), the perivesical fat of the bladder forms the medial aspect of the lymphocele and the peritoneum is the superior border, with the pelvic sidewall forming the lateral aspect. Although many symptomatic lymphoceles are managed by percutaneous drainage or with conservative management, some require laparoscopic fenestration to create a peritoneal window. Laparoscopic fenestration with creation of a window served as the impetus to preemptively rotate and secure a peritoneal flap in this location, decreasing the likelihood of closure of this space. Our goal is to create a pathway lined by peritoneum and direct the lymphatic fluid out of the pelvis and into the peritoneal cavity where it can be absorbed. The peritoneal flap is obtained by dropping the bladder from the abdominal wall at the beginning of the procedure, as there is more than sufficient redundant peritoneal surface that is used for one or both sides. The lateral aspect of the flap, lateral to the obliterated ligament, is brought into the dependent part of the pelvis and is fixed onto the lateral aspect of the bladder. If the patient undergoes bilateral PLND, this is then repeated on the contralateral side to create bilateral flaps, and there is no need to split the tissue. We also anchor this in place with a stitch that attaches it to the ventral perivesical adipose tissue. As this flap is created from redundant tissue, it does not pull the bladder over to the side of the flap. Two important goals are accomplished by this technique. First, the adipose tissue no longer opposes the lymphadenectomy bed because the peritoneal flap is interposed between the bladder fat and the pelvic sidewall. Second, this window allows drainage of the lymph fluid drain into the peritoneal cavity to be reabsorbed. Interposition of peritoneum is not technically demanding and typically requires w5-7 minutes using only 2 additional VICRYL sutures (Ethicon). Despite our findings, this study has some limitations. It is a nonrandomized and retrospective study. Although this is a retrospective study, there were no changes to surgical technique in terms of the PLND, and the groups had no significant difference in patient demographics. Additionally, the study was completed at a single UROLOGY 85 (6), 2015
institution by 2 surgeons. As such, extrapolation to other institutions may be limited. However, the technique was easily transferred between one surgeon to the next after its development by the senior author (A.M.). It is not likely that a “learning curve” played a role in reduction of the lymphocele as each surgeon had performed >400 RARP before the study date. Overall, there was a low lymph node yield compared with that of other studies, but we routinely do not perform an extended PLND and 45% were unilateral PLND (group A, 42%; group B, 49%; P ¼ .42). This technique is not applicable to extraperitoneal cases. We do not know the long-term effects of the flap on continence or other unforeseen consequences, but the short-term complication rate was comparable with a standard RARP (group A). Postoperative lymphocele evaluation was limited by clinical diagnosis. The cystogram served as a surrogate for detection and found 66.7% of the lymphoceles. A cystogram is not the best imaging study to evaluate the presence of a lymphocele, and a CT or ultrasound would more accurately detect an asymptomatic lymphocele. Our goal in this study, however, was to detect symptomatic lymphoceles. However, this limitation in detection with the cystogram and not a CT or ultrasound existed in both groups and is therefore not likely to influence data interpretation. We are currently planning a multiinstitutional, randomized, controlled study. In our subsequent study, all patients will receive a postoperative ultrasound to better evaluate for a lymphocele and this will improve the limitations of our present work.
CONCLUSION Peritonealization of the lateral aspect of the bladder with a peritoneal interposition flap is a novel technique to preemptively create a permanent pathway for lymph drainage into the peritoneal space after RARP or PLND. In this preliminary report, no patients with peritoneal interposition flaps developed symptomatic lymphoceles after transperitoneal RARP or PLND. Additional studies are warranted, as this technique may be applicable to all patients undergoing transperitoneal RARP with PLND. Acknowledgment. Vinald Francis, Lahey Hospital and Medical Center, for illustration. References 1. Prostate Cancer Clinical Guideline Update Panel Prostate cancer Guideline for the Management of Clinically Localized Prostate Cancer: 2007 update. American Urological Association Education and Research, Inc. Available at: http://www.auanet.org/content/ guidelines-and-quality-care/clinical-guidelines/main-reports/proscan07/ content.pdf. Accessed December 9, 2014.
UROLOGY 85 (6), 2015
2. Cagiannos I, Karakiewicz P, Eastham JA, et al. A preoperative nomogram identifying decreased risk of positive pelvic lymph nodes in patients with prostate cancer. J Urol. 2003;170:1798-1803. 3. Touijer KA, Ahallal Y, Guillonneau BD. Indications for and anatomical extent of pelvic lymph node dissection for prostate cancer: practice patterns of uro-oncologists in North America. Urol Oncol. 2013;31:1517-1521; e1-e2. 4. Zorn KC, Katz MH, Bernstein A, et al. Pelvic lymphadenectomy during robot-assisted radical prostatectomy: assessing nodal yield, perioperative outcomes, and complications. Urology. 2009;74: 296-302. 5. Feicke A, Baumgartner M, Talimi S, et al. Robotic-assisted laparoscopic extended pelvic lymph node dissection for prostate cancer: surgical technique and experience with the first 99 cases. Eur Urol. 2009;55:876-883. 6. Musch M, Klevecka V, Roggenbuck U, et al. Complications of pelvic lymphadenectomy in 1,380 patients undergoing radical retropubic prostatectomy between 1993 and 2006. J Urol. 2008;179:923-929. 7. Freid RM, Siegel D, Smith AD, et al. Lymphoceles after laparoscopic pelvic node dissection. URL. 1998;51:131-134. 8. Orvieto MA, Coelho RF, Chauhan S, et al. Incidence of lymphoceles after robot-assisted pelvic lymph node dissection. BJU Int. 2011;108:1185-1190. 9. Sogani PC, Watson RC, Whitmore WF. Lymphocele after pelvic lymphadenectomy for urologic cancer. URL. 1981;17:39-43. 10. Kim JK, Jeong YY, Kim YH, et al. Postoperative pelvic lymphocele: treatment with simple percutaneous catheter drainage. Radiology. 1999;212:390-394. 11. Fallick ML, Long JP. Laparoscopic marsupialization of lymphocele after laparoscopic lymph node dissection. J Endourol. 1996;10: 533-534. 12. D’Amico AV, Whittington R, Malkowicz SB, et al. Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA. 1998;280:969-974. 13. Solberg A, Angelsen A, Bergan U, et al. Frequency of lymphoceles after open and laparoscopic pelvic lymph node dissection in patients with prostate cancer. Scand J Urol Nephrol. 2003;37: 218-221. 14. Raheem OA, Bazzi WM, Parsons JK, et al. Management of pelvic lymphoceles following robot-assisted laparoscopic radical prostatectomy. Urol Ann. 2012;4:111-114. 15. Liss MA, Palazzi K, Stroup SP, et al. Outcomes and complications of pelvic lymph node dissection during robotic-assisted radical prostatectomy. World J Urol. 2013;31:481-488. 16. Scholz HS, Petru E, Benedicic C, et al. Fibrin application for preventing lymphocysts after retroperitoneal lymphadenectomy in patients with gynecologic malignancies. Gynecol Oncol. 2002;84: 43-46. 17. Kim WT, Ham WS, Koo KC, et al. Efficacy of octreotide for management of lymphorrhea after pelvic lymph node dissection in radical prostatectomy. Urology. 2010;76:398-401. 18. Stolzenburg JU, Wasserscheid J, Rabenalt R, et al. Reduction in incidence of lymphocele following extraperitoneal radical prostatectomy and pelvic lymph node dissection by bilateral peritoneal fenestration. World J Urol. 2008;26:581-586.
Video clips cited in this article can be found on the internet at: http://dx.doi.org/10.1016/ j.urology.2015.02.034.
1509