- Email: [email protected]
All You Need to Know About Urethrovesical Anastomotic Urinary Leakage Following Radical Prostatectomy
All You Need to Know About Urethrovesical Anastomotic Urinary Leakage Following Radical Prostatectomy Stavros I. Tyritzis,* Ioannis Katafigiotis and Constantinos A. Constantinides From the Department of Urology, Athens University Medical School, Laiko Hospital, Athens, Greece
Purpose: Radical prostatectomy is a challenging operation demanding a high level of surgical expertise and experience. Urinary leakage at the urethrovesical anastomosis is one of the most common short-term complications of radical prostatectomy, reaching an incidence of 0.3% to 15.4%. In this review we investigate and discuss all matters directly related to urethrovesical anastomotic leak, specifically how to diagnose it properly, how to determine when it is clinically significant and when intervention is required, how to prevent or predict it and, finally, the possible long-term sequelae. Materials and Methods: We conducted a systematic analysis of the literature searching for English and nonEnglish language publications from a preidentified time frame (1985 to 2011) using primary search databases (PubMed®, Web of Science®). Manual selection was performed by 2 authors and the third reviewed the final common selection. We also created an algorithm for the diagnosis and management of urethrovesical anastomotic leak. Results: A total of 72 studies were finally selected, including 48 (67%) observational case series, 16 (22.2%) prospective trials, 1 letter to the editor, 1 review and 1 systematic review which was focused only on laparoscopic radical prostatectomy. We also found 2 experimental studies performed in animal models and 3 case reports. Of these studies 7 reported results from fewer than 20 patients. No consensus was recorded on a strict definition of urethrovesical anastomotic leak. The factors determining possible definitions included postoperative day of urethrovesical anastomotic leak, amount of extravasation on cystography and the need for intervention. Urethrovesical anastomotic leak should be classified according to the Clavien classification system, depending on severity and the need for intervention. To our knowledge the role of the open, laparoscopic or robotic approach in the incidence of urethrovesical anastomotic leak has not been systematically investigated. Risk factors for urethrovesical anastomotic leak include obesity, prostate size, previous prostatic surgery, type of anastomosis technique, suture number and type, eversion of the mucosa, a difficult anastomosis or an anastomosis under tension, reconstruction of the musculofascial plate, blood loss, intraoperative flush test result and postoperative urinary tract infection. Diagnosis can be determined primarily by establishing the nature of the drain output. Retrograde cystography, computerized tomography cystography, transrectal ultrasound, contrast enhanced ultrasound and excretory urography are the indicated imaging modalities, and are not always necessary. Finally, the development of anastomotic stricture and incontinence due to urethrovesical anastomotic leak are additional complications. Conclusions: We gathered all relevant critical information concerning urethrovesical anastomotic leak to encourage standardization in the diagnosis and management of this common complication. Systematic meta-analysis of each debatable issue is required to provide definite answers. 0022-5347/12/1882-0369/0 THE JOURNAL OF UROLOGY® © 2012 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION
AND
http://dx.doi.org/10.1016/j.juro.2012.03.126 Vol. 188, 369-376, August 2012 RESEARCH, INC. Printed in U.S.A
Abbreviations and Acronyms CT ⫽ computerized tomography LRP ⫽ laparoscopic radical prostatectomy POD ⫽ postoperative day RARP ⫽ robot-assisted radical prostatectomy RP ⫽ radical prostatectomy TRUS ⫽ transrectal ultrasound UVAL ⫽ urethrovesical anastomotic leak Submitted for publication October 16, 2011. Nothing to disclose. Supplementary material for this article can be obtained at http://urology.med.uoa.gr. * Correspondence: Department of Urology, Athens University Medical School, Laiko Hospital, 17 Agiou Thoma str., 11527, Athens, Greece (telephone: 0030 210 7456344; FAX: 0030 210 9327744; e-mail: [email protected]).
Editor’s Note: This article is the first of 5 published in this issue for which category 1 CME credits can be earned. Instructions for obtaining credits are given with the questions on pages 680 and 681.
www.jurology.com
369
370
URETHROVESICAL ANASTOMOTIC URINARY LEAK AFTER RADICAL PROSTATECTOMY
Key Words: anastomosis, surgical; anastomotic leak; prostatectomy RADICAL prostatectomy is a challenging operation that demands high levels of surgical expertise and experience. RP can be accomplished with open or with minimally invasive laparoscopic or robot-assisted approaches. Regardless of the approach, complications of RP remain an important issue. Complication rates of the open approaches are approximately 2% to 10% for radical retropubic prostatectomy and 4% to 22% for radical perineal prostatectomy.1 LRP (transperitoneal and extraperitoneal approaches) and RARP reach a complication rate of approximately 1.5% to 17.8% after the learning curve.2 One of the most common short-term complications of RP is urinary urethrovesical anastomotic leakage. In this review we gather all relevant information concerning UVAL to stimulate standardization in its diagnosis. We address the proper diagnosis of UVAL, its clinical significance, appropriate intervention, prevention or prediction and, finally, possible long-term sequelae.
MATERIALS AND METHODS We conducted a systematic review of the literature searching for English and nonEnglish language publications (articles or abstracts) for a preidentified time frame (1985 to 2011). The primary search databases used were PubMed and Web of Science. The literature search strategy and search terms are depicted in figure 1. We reviewed the literature and manually selected those publications with the greatest relevance. All relevant articles had to report the incidence of UVAL and include a definition of UVAL.
Anastomotic leakage and RP
(92/relevant 44)
Urinary leakage and RP
(218/relevant 55)
Eligible papers
72
Cystography and RP
(57/relevant 19)
Figure 1. Data extraction strategy
Numerous reports were excluded from study because the term urinary leakage was used to describe incontinence. Items were also excluded unless management of UVAL was reported or any parameter (clinical or surgical) related to UVAL was found. For each eligible publication we determined the type of study, and recorded the sample size, the level of evidence provided and the number of citations. The statistical significance of any variable related to UVAL was also noted. All existing definitions and classifications were recorded. A variety of diagnostic and therapeutic methods were presented, allowing us to create a summarizing algorithm.
RESULTS Literature Search Figure 1 shows the number of hits and eligible studies stratified by the respective combination of terms used. We summarized the major characteristics from each of the 72 studies we used for data extraction. The studies were published between 1990 and 2011, with the majority published in 2005 or thereafter. Of the 72 studies 48 (67%) were observational case series, 16 (22.2%) prospective trials, 1 a letter to the editor, 1 a review and 1 a systematic review focused only on laparoscopic RP. We also found 2 experimental studies performed in animal models and 3 case reports. There were 7 studies that reported results from less than 20 patients. However, a few reports described novel techniques. Incidence The incidence of UVAL is higher in the first 7 to 10 days postoperatively and decreases with time, suggesting that the healing of the anastomosis is time dependent. However, the incidence of UVAL varies significantly due to several parameters such as postoperative day, the amount of extravasation on cystography and the need for intervention. The effect of POD on the incidence of UVAL, at least for the open procedure, can be easily understood from the study by Dalton et al.3 In this study extravasation was found in 78% of patients on POD 8, in 38% on POD 11, in 20% on POD 14 and in 8% of patients 3 weeks after surgery. In the first 5 to 8 days after radical retropubic prostatectomy the incidence of UVAL ranged between 67% and 78%.3 The incidence of urine extravasation on cystograms obtained at PODs 14 and 21 was 0.1% and 20%, respectively.3 In more recent studies the incidence on PODs 4 to 5 and 7 was 28.5% and 25%, respectively. The incidence of UVAL 3 weeks postoperatively ranged from 8% to 14.1%.2,3 This important finding suggests that after 3 weeks the healing process has been completed, and it can
URETHROVESICAL ANASTOMOTIC URINARY LEAK AFTER RADICAL PROSTATECTOMY
be speculated that there is little to be gained by leaving the catheter in place any longer. The incidence rates with LRP and RALP usually follow the same pattern. The laparoscopic incidence rates of UVAL have been reported to be 10%, 15.1% and 17.2% on PODs 4, 5 and 7, respectively.2 Rassweiler et al stratified the incidence rates of UVAL by the need for intervention.2 The incidence rates for LRP requiring intervention were 0.9% to 2.5% vs 0.5% for the open procedure. In a review by Mochtar et al the mean incidence of UVAL for LRP was 9.7%, and no statistical difference between the transperitoneal and the extraperitoneal approach was found.4 Definition-Classification There is no consensus on the definition of UVAL. The 2 most common definitions are based on extravasation on cystography in relation to POD and the production of urine via the drain. Although cystography appears to be the imaging technique of choice for the definition of UVAL, there is a great variance in the POD on which it is performed. On the other hand, it has been suggested that catheter removal can be accomplished safely without cystography. The current literature defines UVAL as persistent contrast extravasation on cystography between PODs 3 and 14.2 An issue of the utmost importance is establishing that the liquid drained is urine, especially if lymphadenectomy has been performed. A high creatinine level in the drain output is the key finding. It is essential to distinguish the clinically significant from the nonsignificant event. It has been advocated that the production of fluid with an increased creatinine concentration from the wound drain or considerable contrast extravasation on a cystogram on POD 7 resulting in an extension of the planned catheterization period can be considered clinically significant leakage.4 Others define UVAL as clinically significant when it 1) extends intraperitoneally, 2) necessitates interventional drainage, 3) results in ileus necessitating readmission and intravenous fluids, and 4) is associated with fever and signs of sepsis necessitating antibiotic therapy or drainage.5 On the other hand, characterization as not clinically significant involves radiological leakage on a cystogram or production via the drain that does not result in extra catheterization time.4 It is also important to refer to the severity grades of UVAL. Usually leakage is classified as minor, moderate or severe, as extraperitoneal or intraperitoneal, and according to the extension of extravasation in relation to the anastomosis or the pelvis.6,7 Patil et al were the first to report quantifiable definitions of UVAL using conventional cystography.8 Extravasations confined immediately around the
371
anastomosis are usually characterized as small or grade I, while those that extend to more distant areas are characterized as moderate or grade II. Finally the ones that are intraperitoneal, extend in the perivesical space or are confined in the pelvis are characterized as large, significant or grade III.3 Standardization is required for unbiased evaluation in the reporting of complications. Thus, UVAL should be classified according to the Clavien classification of surgical complications.9 It is apparent that UVAL grading depends on the severity and the need for intervention. Predictive Factors The predictive factors of UVAL can be related to patient characteristics, preoperative and postoperative outcomes, surgeon experience or the technical details of the procedure. The literature may present these factors as debatable since the number of studies on each factor is too limited to perform a metaanalysis. Thus, our assumptions can be considered observations rather than statements of certainty. The patient factors that might be implicated are ischemic heart disease7 and obesity.10,11 Obesity is thought to create difficulties in accessing the vesicourethral anastomosis or to decrease visibility, but only in the open approach.10 In cases of LRP or RARP UVAL occurs but it is not affected by obesity.10 However, other groups suggest the opposite.11 In addition, urinary tract infection appears to increase the possibility of UVAL.12 Excessive bleeding can impair visualization, rendering precise suture placement difficult.13 The role of hemorrhage as a putative risk factor is also suggested by the coexistence of pelvic hematoma and UVAL as depicted on postoperative cystography and changes in postoperative hematocrit. Again, the opposite opinion has been encountered, suggesting that blood loss has no predictive value in identifying leakage.12 Blood clot formation has also been implicated since a blood clot can obstruct the urethral catheter and interfere with urine drainage, compromising the healing of the anastomosis.14,15 Despite the fact that no trial has determined the actual learning curve for achieving a 0% incidence of UVAL, one could argue that surgical experience is related to UVAL. In LRP the learning curve is considered to be approximately 50 procedures but ranges from 40 to 180.4 The achievement of a watertight anastomosis demands expertise in laparoscopic suturing and this is apparent in the higher incidence of UVAL in initial series of LRP and its decrease in subsequent series.16 Several aspects of the surgery are also debatable. Eversion of the mucosa might be deleterious in the accomplishment of a watertight anastomosis.7,17,18 A comparison of anastomoses with and without
372
URETHROVESICAL ANASTOMOTIC URINARY LEAK AFTER RADICAL PROSTATECTOMY
eversion of the mucosa shows the slight superiority of the noneversion technique.7 Moreover prostate size14 and previous prostatic surgery7,8,12 have been recognized, but their roles are supported by both arguments. Specific technical details in the performance of the anastomosis are potential predictive factors. Interrupted or running sutures do not seem to have an effect.19 –21 In addition, the fewer the interrupted sutures used (2, 4, or more in various positions), the more they increase the chance of UVAL, but not in a statistically significant manner.22–24 Sammon et al reported that a double layer vs a single layer anastomosis could be advantageous.25 Others have reported on catheterless or even sutureless techniques, which are all risks for UVAL. However, this finding was not clinically significant because no further intervention was required and it did not add to patient morbidity.26 –28 The type of suture, which exhibits different characteristics, might also have a role. A comparison in the literature of MAXON™, DEXON™ and chromic catgut sutures revealed an advantage of chromic sutures over the others, but the data were based only on a small number of DEXON anastomoses. Barbed vs monofilament sutures were also compared without revealing significant differences.29 Yasui et al reported less UVAL when using LAPRA-TY® clips with a single knot technique.30 A difficult anastomosis is another possible risk factor for UVAL. The posterior or posterolateral segment of the anastomosis is more commonly recognized as the most difficult because it is deep inside the pelvis and requires great surgical skill for its construction, especially in LRP.12,15 Another difficulty lies in the length or thickness of the urethral stump. A comparison showed that patients with a difficult anastomosis had leakage significantly more often than those with no difficulties (32% vs 4.9%).12 The role of musculofascial posterior reconstruction is also debatable. Although the majority of studies agree that it is beneficial,31 2 recent randomized trials have shown that it has no protective role apart from the earlier recovery of continence.32 Thus, the benefit of posterior reconstruction can be considered controversial at best. A widely accepted method of checking the integrity of the anastomosis is the intraoperative flush test. It is a common practice after completion of the anastomosis to fill the bladder through the indwelling catheter with 60 to 250 ml sterile water and to check for a possible leak before the closure begins. An unsatisfactory test is usually defined as the leakage of the sterile water through the anastomosis. The value of an unsatisfactory flush test is often emphasized as it may indicate the need for additional sutures or even complete revision of the anas-
tomosis in the case of a major leak. Others recognize the importance of the test due to its ability to predict a leak on postoperative cystography.14 –16 More specifically, it has been reported that 97.1% of the cystograms performed on POD 7 can be avoided according to the results of the intraoperative test and that the test can be the only independent predictor of success for the early catheter removal (p ⫽ 0.0069).33 The comparison among patients with unsatisfactory test results showed that more patients leak (48.4%) than do not leak (15.8%) (p ⫽ 0.002).12 We should also mention that patients undergoing salvage RP after pelvic radiation are at a definite risk for UVAL compared to treatment naïve patients.34 Another cause of UVAL related to a wide bladder neck is the proximity of the ureteral orifices to the anastomosis. When this is the case, the balloon of the catheter might place the orifice outside the anastomosis, leading to leakage. This can be diagnosed with excretory urography and the surgeon should be aware of such an event. Lastly Parivar et al reported no correlation between positive surgical margins and healing of the urethrovesical anastomosis.35 Diagnosis The simplest way to diagnose UVAL is by assessing the clinical condition of the patient in conjunction with careful monitoring of the fluid emerging from the drain. The most urgent clinical condition that can imply significant UVAL is the complaint of abdominal distention and pain caused by ileus and peritonitis.36 It is important to mention that a high output of urine through the drain during the first hour can be encountered, and that this usually decreases in the first 2 or 3 days.4 A high urine output through the drain is not necessarily due to UVAL, but it can be the result of an iatrogenic ureteral injury which requires urgent management. Apart from the obvious result of the intraoperative flush test, retrograde cystography is often performed before catheter removal. The catheter is removed and retrograde cystography is performed between PODs 2 and 21.2 When the first cystography depicts moderate to large leakage in the anastomosis, prolonged catheterization is applied depending on surgeon preference. Before the second attempt, retrograde cystography is performed.2,15 Others suggest that when the initial cystography shows minimal leakage, then the catheter can be removed a week later, usually without performing a new cystography.7 Although cystography is considered a valuable diagnostic tool, the fact that it is an x-ray application and has a significant cost has led many authors to remove the catheter without cystography. This is evident in LRP and RARP series in which it has
URETHROVESICAL ANASTOMOTIC URINARY LEAK AFTER RADICAL PROSTATECTOMY
been advocated that the catheter can be safely removed at POD 4 to 10 without cystography.3 The same concept has been proposed for the open procedure, suggesting catheter removal on POD 8 to 9 in appropriate selected patients.3 Others support the same practice, but suggest that it is safe between POD 14 to 20. In recent RARP series there has been a general tendency not to perform cystography if the surgeon plans to remove the catheter after POD 8.37 Recently Han et al reported that CT cystography is superior to conventional cystography.38 Lee et al made a similar observation by evaluating multidetector CT cystography, which revealed a statistically significant difference in detection rates compared to conventional cystography (80.4% vs 54.3%).39 In both studies CT cystography upgraded UVAL (for example from mild to moderate), establishing the need for prolonged catheterization. Furthermore, CT eliminated the blind spots of conventional cystography where extravasation is located. Another modality is transrectal ultrasonography with or without the use of enhancing contrast medium. TRUS can be used to successfully assess the integrity of the UVAL while maintaining good specificity and sensitivity, with a much lower cost than cystography and with no emission of radiation.6 The drawbacks of TRUS include the depiction of paraurethral fluid which can mimic UVAL, reflections from the catheter which can obscure the ventral part of the anastomosis, patient discomfort and a hypothetical negative effect on the anastomosis due to the applied pressure. A modified version of the cystogram is the pericatheterogram, during which a small caliber catheter (10Fr) is introduced into the urethral meatus alongside the existing Foley. Contrast medium is instilled in the urethra under gravity until bladder is full and x-ray films are obtained.40 Most recently contrast enhanced ultrasound was reported to be a promising imaging modality, with a low cost, that seems to be equivalent to cystography for detecting large UVAL.41 In general CT cystography has a superior sensitivity and specificity with a greater cost and radiation exposure. Despite its limitations, TRUS has a higher accuracy for the close tissue patterns in the supposed perianastomotic lesion. A randomized trial comparing conventional cystography, CT, enhanced ultrasound and TRUS cystography would probably establish these assumptions. Lastly, an interesting point was reported by Schenck and Schneider, who suggested that uroscopy (if blood was cloudy or colored by old blood) could predict UVAL in 86% of cases.42 Management We composed an algorithm based on our experience with more than 200 RPs per year and the current
373
literature (fig. 2). In this algorithm we incorporated another algorithm by Do et al, who adjusted their maneuvers according to the amount of extravasation on cystography.43 The prevalence of UVAL requiring intervention has been estimated to be 0.9% to 2.5% after LRP and 0.5% after an open procedure.14 The most common practice is prolonged catheterization and, again, the timing of catheter removal varies. The consensus is that cystography must be repeated before catheter removal.7,12 The majority of the authors removed the catheter only if subsequent cystography showed no extravasation.7,39,40 Several conservative measures have been proposed to help the healing process, the most ordinary being the application of gentle traction to the catheter.14,15,36 However, some suggest that traction might induce ischemia on the anastomosis and impair healing, although this has not been verified scientifically. Correct replacement of the catheter has been proposed to help in the management of UVAL. If no fenestrated catheter has been used, radiological mobilization and adjustment of the catheter, deflation of the balloon and external fixation can contribute to proper urine drainage.4 Other conservative ways to manage UVAL concern the drain. The drain may act as a pipette when it touches the anastomosis and this is easily recognized during retrograde cystography. Switching from active to passive or gravity drainage, pulling back the drain and withdrawing the drain are proper maneuvers.14,15,36 In the presence of persistent UVAL, 2 specific maneuvers of the transurethral catheter have been reported. An underwater suction drainage system can be used as in extrapleural drainage, with connection to the transurethral catheter.2 The second option involves low intermittent wall suction and an 18 to 20 gauge angiocatheter needle in the tubing connecting the catheter to the wall suction device.14 The continuous needle vented catheter suction prevents the continuous suction of the bladder mucosa, especially when the musosa obstructs the catheter tip. Aspiration of a postoperative urinoma can be accomplished with interventional radiology, and CT guided drainage and placement of a pelvic drain.8,33,36 When these methods fail and before the decision for reoperation, ureteral stenting or even nephrostomy placement are the next steps.4 Of patients treated with LRP 1.6% to 2.2% require ureteral stenting or nephrostomy tube placement.4 In cases with a major leak the insertion of a mono-J catheter and a minimum of 2 weeks of catheterization have been recommended as an alternative.15 The insertion of a nephroureteral stent has been also combined with the connection of the stent to a TRU-CLOSE® suction drainage system.41 This combination effec-
374
URETHROVESICAL ANASTOMOTIC URINARY LEAK AFTER RADICAL PROSTATECTOMY
Lymphadenectomy performed?
YES NO (probably lymphorrhea but:) Check catheter function Flush test Blocked
OK
Remove clots or change catheter using guidewire
Creatinine determination
PROLONGED CATHETERIZATION
LOW
HIGH
<300cc >300cc Lymphorrhea=Prolonged drainage
Apply gentle traction
REDUCES
PERSISTS
REMOVE DRAIN IF OUTPUT <50cc
Perform cystography+ Rule out ureteral injury Order IVU
Check algorithm by Do et al
Deflate balloon and tack catheter Mobilize drain
REDUCES
PERSISTS
Suprapubic catheter Nephrostomy Endoscopic evaluation Consider revision Figure 2. Management algorithm. IVU, excretory urography.
tively diverts the urine output, effacing the output from the transurethral catheter and the pelvic drain. It should be emphasized that urinary diversion using stents might be efficient to stop UVAL and allow healing. However, we found no clear evidence of its efficacy in achieving a potent anastomosis. The last reserve is reoperation, and this depends on the amount of urine drainage and the use of the transperitoneal or extraperitoneal approach. When the transperitoneal route has been chosen, then
UVAL can lead to uroperitoneum and to more dramatic manifestations such as ileus.4 Ileus or peritonitis are not always manifested in cases of major leaks. Reoperation may vary from complete reconstruction of the anastomosis to a partial anastomotic repair.15,36 Data extracted from LRP series show that reoperation can been performed early in the postoperative period such as POD 2, or later such as PODs 6 to 12.15–36 The reoperation in the LRP series is performed with laparoscopy and the approach, transperitoneal or extraperitoneal, is the same as
URETHROVESICAL ANASTOMOTIC URINARY LEAK AFTER RADICAL PROSTATECTOMY
for the initial operation.15,36 While continuous suturing was used in the initial operation, in the majority of cases the reoperation is performed with interrupted sutures.15 Complications The most detrimental short-term complication of UVAL is uroperitoneum. Uroperitoneum is a result of the transperitoneal approach of the LRP, and it can lead to more serious clinical conditions such as peritonitis, ileus and renal function deterioration, and even necessitate laparotomy.4,36 Although no relationship has been established between anastomotic strictures or bladder neck contractures and UVAL, the data are contradictory. Several authors support the role of UVAL in inducing strictures10,13,17,44,45 among others who suggest that UVAL obstructs normal wound healing, leading to a persistent inflammatory reaction and finally to anastomotic stricture.8,22 However, some authors propose that strictures cannot be attributed to UVAL since different pathophysiological mechanisms and factors are responsible for the stricture.18,46 Incontinence is a complication that has been attributed to the radical prostatectomy itself. The consensus is that UVAL does not seem to affect longterm incontinence.8,18 However, early recovery of continence may be delayed due to UVAL and among patients with leakage there is a higher incidence of incontinence.8,18 Prevention The only additional measures that can be regarded as preventative with the exception of applying the basic surgical principles of anastomosis are the use of a fenestrated catheter, the application of fibrin sealant in the anastomosis and the placement of oxidized cellulose sponges around the anastomosis. A fenestrated catheter is used to prevent exudate composed of blood and urine from accumulating
375
in the urethral space, thereby preventing UVAL.21,47,48 The fenestration is placed below the balloon, helping the free drainage of the exudate from the important site of UVAL that is healing. The fenestration also aids the drainage when the catheter tip hole is located away from the accumulated fluid or when bladder spasm obstructs the catheter tip hole. The use of fibrin sealant reinforces the anastomosis, promoting healing without inflammation and the formation of fibrosis.49 Although the data in the literature are based on a small number of cases or on preliminary results, they exhibit a trend in decreasing leakage. The application of oxidized cellulose sponges around the anastomotic site in the pelvic fossa to block the escape of exudate has also been reported.21 Similar adhesives as well as laser welding of the anastomosis were proposed.50 Finally, we should emphasize the importance of informing the patient about drains and catheters as well as compliance. Accidental removal, blockage or displacement of catheters or drains is often seen during patient mobilization. Thus, it is important to anticipate such events by evaluating the patient’s understanding of the operation and postoperative course.
CONCLUSIONS UVAL is a troubling, short-term complication that occurs independent of the approach used to perform RP and may create repercussions in patient quality of life. The experience of the surgeon in using the technique with mastery is probably the most crucial parameter in avoiding UVAL. We were able to construct an algorithm for the diagnosis and management of this complication. It is crucial that the management of UVAL be tailored to each patient. The urologist should wait before each intervention, and should always evaluate the patient and drain signs.
REFERENCES 1. Lepor H and Kaci L: Contemporary evaluation of operative parameters and complications related to open radical retropubic prostatectomy. Urology 2003; 62: 702. 2. Hakimi AA, Faleck DM, Sobey S et al: Assessment of complication and functional outcome reporting in the minimally invasive prostatectomy literature from 2006 to the present. BJU Int 2012; 109: 26. 3. Dalton DP, Schaeffer AJ, Garnett JE et al: Radiographic assessment of the vesicourethral anastomosis directing early decatheterization following nerve-sparing radical retropubic prostatectomy. J Urol 1989; 141: 79.
4. Mochtar CA, Kauer PC, Laguna MP et al: Urinary leakage after laparoscopic radical prostatectomy: a systematic review. J Endourol 2007; 21: 1371.
7. Ramsden AR and Chodak GW: Can leakage at the vesico-urethral anastomosis be predicted after radical retropubic prostatectomy? BJU Int 2004; 93: 503.
5. Tan G, Srivastava A, Grover S et al: Optimizing vesicourethral anastomosis healing after robotassisted laparoscopic radical prostatectomy: lessons learned from three techniques in 1900 patients. J Endourol 2010; 24: 1975.
8. Patil N, Krane L, Javed K et al: Evaluating and grading cystographic leakage: correlation with clinical outcomes in patients undergoing robotic prostatectomy. BJU Int 2009; 103: 1108.
6. Eggert T, Palisaar J, Metz P et al: Assessing the vesico-urethral anastomosis after radical retropubic prostatectomy: transrectal ultrasonography can replace cystography. BJU Int 2007; 100: 1268.
9. Constantinides CA, Tyritzis SI, Skolarikos A et al: Short- and long-term complications of open radical prostatectomy according to the Clavien classification system. BJU Int 2009; 103: 336. 10. Sandhu JS, Gotto GT, Herran LA et al: Age, obesity, medical comorbidities and surgical tech-
376
URETHROVESICAL ANASTOMOTIC URINARY LEAK AFTER RADICAL PROSTATECTOMY
nique are predictive of symptomatic anastomotic strictures after contemporary radical prostatectomy. J Urol 2011; 185: 2148. 11. Herman MP, Raman JD, Dong S et al: Increasing body mass index negatively impacts outcomes following robotic radical prostatectomy. JSLS 2007; 11: 438. 12. Gnanapragasam VJ, Baker P, Naisby GP et al: Identification and validation of risk factors for vesicourethral leaks following radical retropubic prostatectomy. Int J Urol 2005; 12: 948. 13. Surya BV, Provet J, Johanson KE et al: Anastomotic strictures following radical prostatectomy: risk factors and management. J Urol 1990; 143: 755. 14. Fenig DM, Slova D and Lepor H: Postoperative blood loss predicts the development of urinary extravasation on cystogram following radical retropubic prostatectomy. J Urol 2006; 175: 146. 15. Moinzadeh A, Abouassaly R, Gill IS et al: Continuous needle vented Foley catheter suction for urinary leak after radical prostatectomy. J Urol 2004; 171: 2366. 16. Liatsikos E, Rabenalt R, Burchardt M et al: Prevention and management of perioperative complications in laparoscopic and endoscopic radical prostatectomy. World J Urol 2008; 26: 571. 17. Stolzenburg JU, Ho KM, Do M et al: Impact of previous surgery on endoscopic extraperitoneal radical prostatectomy. Urology 2005; 65: 325. 18. Webb DR, Sethi K and Gee K: An analysis of the causes of bladder neck contracture after open and robot-assisted laparoscopic radical prostatectomy. BJU Int 2009; 103: 957.
thral anastomosis in radical retropubic prostatectomy. Int Braz J Urol 2009; 35: 158. 25. Sammon JD, Muhletaler F, Peabody JO et al: Long-term functional urinary outcomes comparing single- vs double-layer urethrovesical anastomosis: two-year follow-up of a two-group parallel randomized controlled trial. Urology 2010; 76: 1102. 26. Tewari A, Rao S and Mandhani A: Catheter-less robotic radical prostatectomy using a custommade synchronous anastomotic splint and vesical urinary diversion device: report of the initial series and perioperative outcomes. BJU Int 2008; 102: 1000.
28. Simforoosh N, Javaherforooshzadeh A, Aminsharifi A et al: Early continence after open and laparoscopic radical prostatectomy with sutureless vesicourethral alignment: an alternative technique, 8 years’ experience. Urol J 2009; 6: 163.
41. Schoeppler GM, Buchner A, Zaak D et al: Detection of urinary leakage after radical retropubic prostatectomy by contrast enhanced ultrasound - do we still need conventional retrograde cystography? BJU Int 2010; 106: 1632.
29. Sammon J, Kim TK, Trinh QD et al: Anastomosis during robot-assisted radical prostatectomy: randomized controlled trial comparing barbed and standard monofilament suture. Urology 2011; 78: 572. 30. Yasui T, Itoh Y, Maruyama T et al: The singleknot method with Lapra-Ty clips is useful for training surgeons in vesicourethral anastomosis during laparoscopic radical prostatectomy. Int Urol Nephrol 2009; 41: 281.
20. Poulakis V, Skriapas K, de Vries R et al: Vesicourethral anastomosis during endoscopic extraperitoneal radical prostatectomy: a prospective comparison between the single-knot running and interrupted technique. Urology 2006; 68: 1284.
32. Sutherland DE, Linder B, Guzman AM et al: Posterior rhabdosphincter reconstruction during robotic assisted radical prostatectomy: results from a phase II randomized clinical trial. J Urol 2011; 185: 1262.
21. Ball AJ, Bordeau KP, Davis JW et al: Modified running vesicourethral anastomosis after robotically assisted laparoscopic radical prostatectomy: use of solitary Lapra-Ty to secure posterior approximation. Urology 2005; 66: 16.
33. Ischia JJ and Lindsay S: Is a cystogram necessary after radical prostatectomy? ANZ J Surg 2005; 75: 825.
24. Mazaris EM, Chatzidarellis E, Varkarakis IM et al: Reducing the number of sutures for vesicoure-
39. Lee HJ, Shin CI, Hwang SI et al: MDCT cystography for detection of vesicourethral leak after prostatectomy. AJR Am J Roentgenol 2008; 191: 1847. 40. Castillo OA, Alston C and Sanchez-Salas R: Persistent vesicourethral anastomotic leak after laparoscopic radical prostatectomy: laparoscopic solution. Urology 2009; 73: 124.
31. Coelho RF, Chauhan S, Orvieto MA et al: Influence of modified posterior reconstruction of the rhabdosphincter on early recovery of continence and anastomotic leakage rates after robot-assisted radical prostatectomy. Eur Urol 2011; 59: 72.
23. Gallo L, Perdonà S, Autorino R et al: Vesicourethral anastomosis during radical retropubic prostatectomy: does the number of sutures matter? Urology 2007; 69: 547.
38. Han KS, Choi HJ, Jung DC et al: A prospective evaluation of conventional cystography for detection of urine leakage at the vesicourethral anastomosis site after radical prostatectomy based on computed tomography. Clin Radiol 2011; 66: 251.
27. Krane LS, Bhandari M, Peabody JO et al: Impact of percutaneous suprapubic tube drainage on patient discomfort after radical prostatectomy. Eur Urol 2009; 56: 325.
19. Srougi M, Paranhos M, Leite KM et al: The influence of bladder neck mucosal eversion and early urinary extravasation on patient outcome after radical retropubic prostatectomy: a prospective controlled trial. BJU Int 2005; 95: 757.
22. Teber D, Erdogru T, Cresswell J et al: Analysis of three different vesicourethral anastomotic techniques in laparoscopic radical prostatectomy. World J Urol 2008; 26: 617.
37. Williams TR, Longoria OJ, Asselmeier S et al: Incidence and imaging appearance of urethrovesical anastomotic urinary leaks following da Vinci robotic prostatectomy. Abdom Imaging 2008; 33: 367.
34. Eandi JA, Link BA, Nelson RA et al: Robotic assisted laparoscopic salvage prostatectomy for radiation resistant prostate cancer. J Urol 2010; 183: 133.
42. Schenck M and Schneider T: Easy and effective assessment of the anastomosis after radical retropubic prostatectomy. Uroscopy before imaging? Urologe A 2010; 49: 364. 43. Do HM, Franz T and Stolzenburg JU: Algorithm for the treatment of anastomotic failure after laparoscopic prostatectomy. Urologe A 2011; 50: 1426. 44. Msezane LP, Reynolds WS, Gofrit ON et al: Bladder neck contracture after robot-assisted laparoscopic radical prostatectomy: evaluation of incidence and risk factors and impact on urinary function. J Endourol 2008; 22: 97. 45. Rebuck DA, Haywood S, McDermott K et al: What is the long-term relevance of clinically detected postoperative anastomotic urine leakage after robotic-assisted laparoscopic prostatectomy? BJU Int 2011; 108: 733. 46. Hanson GR, Odom E, Borden LS Jr et al: Postoperative drain output as a predictor of bladder neck contracture following radical prostatectomy. Int Urol Nephrol 2008; 40: 351. 47. Lawrentschuk N, Bolton DM and Angus D: Fenestrated urethral catheter to aid anastomotic drainage after radical prostatectomy. Urology 2005; 65: 160. 48. Kylmälä T, Kaipia A and Matikainen M: Management of prolonged urinary leakage at the urethrovesical anastomosis. Urol Int 2005; 74: 298.
35. Parivar F, Fournier GR Jr and Narayan P: Urethral anastomotic healing after radical retropubic prostatectomy: impact of positive urethral margin. Urology 1994; 44: 705.
49. Diner EK, Patel SV and Kwart AM: Does fibrin sealant decrease immediate urinary leakage following radical retropubic prostatectomy? J Urol 2005; 173: 1147.
36. Shah G, Vogel F and Moinzadeh A: Nephroureteral stent on suction for urethrovesical anastomotic leak after robot-assisted laparoscopic radical prostatectomy. Urology 2009; 73: 1375.
50. Grummet JP, Costello AJ, Swanson DA et al: Laser welded vesicourethral anastomosis in an in vivo canine model: a pilot study. J Urol 2002; 168: 281.
Purpose: Radical prostatectomy is a challenging operation demanding a high level of surgical expertise and experience. Urinary leakage at the urethrovesical anastomosis is one of the most common short-term complications of radical prostatectomy, reaching an incidence of 0.3% to 15.4%. In this review we investigate and discuss all matters directly related to urethrovesical anastomotic leak, specifically how to diagnose it properly, how to determine when it is clinically significant and when intervention is required, how to prevent or predict it and, finally, the possible long-term sequelae. Materials and Methods: We conducted a systematic analysis of the literature searching for English and nonEnglish language publications from a preidentified time frame (1985 to 2011) using primary search databases (PubMed®, Web of Science®). Manual selection was performed by 2 authors and the third reviewed the final common selection. We also created an algorithm for the diagnosis and management of urethrovesical anastomotic leak. Results: A total of 72 studies were finally selected, including 48 (67%) observational case series, 16 (22.2%) prospective trials, 1 letter to the editor, 1 review and 1 systematic review which was focused only on laparoscopic radical prostatectomy. We also found 2 experimental studies performed in animal models and 3 case reports. Of these studies 7 reported results from fewer than 20 patients. No consensus was recorded on a strict definition of urethrovesical anastomotic leak. The factors determining possible definitions included postoperative day of urethrovesical anastomotic leak, amount of extravasation on cystography and the need for intervention. Urethrovesical anastomotic leak should be classified according to the Clavien classification system, depending on severity and the need for intervention. To our knowledge the role of the open, laparoscopic or robotic approach in the incidence of urethrovesical anastomotic leak has not been systematically investigated. Risk factors for urethrovesical anastomotic leak include obesity, prostate size, previous prostatic surgery, type of anastomosis technique, suture number and type, eversion of the mucosa, a difficult anastomosis or an anastomosis under tension, reconstruction of the musculofascial plate, blood loss, intraoperative flush test result and postoperative urinary tract infection. Diagnosis can be determined primarily by establishing the nature of the drain output. Retrograde cystography, computerized tomography cystography, transrectal ultrasound, contrast enhanced ultrasound and excretory urography are the indicated imaging modalities, and are not always necessary. Finally, the development of anastomotic stricture and incontinence due to urethrovesical anastomotic leak are additional complications. Conclusions: We gathered all relevant critical information concerning urethrovesical anastomotic leak to encourage standardization in the diagnosis and management of this common complication. Systematic meta-analysis of each debatable issue is required to provide definite answers. 0022-5347/12/1882-0369/0 THE JOURNAL OF UROLOGY® © 2012 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION
AND
http://dx.doi.org/10.1016/j.juro.2012.03.126 Vol. 188, 369-376, August 2012 RESEARCH, INC. Printed in U.S.A
Abbreviations and Acronyms CT ⫽ computerized tomography LRP ⫽ laparoscopic radical prostatectomy POD ⫽ postoperative day RARP ⫽ robot-assisted radical prostatectomy RP ⫽ radical prostatectomy TRUS ⫽ transrectal ultrasound UVAL ⫽ urethrovesical anastomotic leak Submitted for publication October 16, 2011. Nothing to disclose. Supplementary material for this article can be obtained at http://urology.med.uoa.gr. * Correspondence: Department of Urology, Athens University Medical School, Laiko Hospital, 17 Agiou Thoma str., 11527, Athens, Greece (telephone: 0030 210 7456344; FAX: 0030 210 9327744; e-mail: [email protected]).
Editor’s Note: This article is the first of 5 published in this issue for which category 1 CME credits can be earned. Instructions for obtaining credits are given with the questions on pages 680 and 681.
www.jurology.com
369
370
URETHROVESICAL ANASTOMOTIC URINARY LEAK AFTER RADICAL PROSTATECTOMY
Key Words: anastomosis, surgical; anastomotic leak; prostatectomy RADICAL prostatectomy is a challenging operation that demands high levels of surgical expertise and experience. RP can be accomplished with open or with minimally invasive laparoscopic or robot-assisted approaches. Regardless of the approach, complications of RP remain an important issue. Complication rates of the open approaches are approximately 2% to 10% for radical retropubic prostatectomy and 4% to 22% for radical perineal prostatectomy.1 LRP (transperitoneal and extraperitoneal approaches) and RARP reach a complication rate of approximately 1.5% to 17.8% after the learning curve.2 One of the most common short-term complications of RP is urinary urethrovesical anastomotic leakage. In this review we gather all relevant information concerning UVAL to stimulate standardization in its diagnosis. We address the proper diagnosis of UVAL, its clinical significance, appropriate intervention, prevention or prediction and, finally, possible long-term sequelae.
MATERIALS AND METHODS We conducted a systematic review of the literature searching for English and nonEnglish language publications (articles or abstracts) for a preidentified time frame (1985 to 2011). The primary search databases used were PubMed and Web of Science. The literature search strategy and search terms are depicted in figure 1. We reviewed the literature and manually selected those publications with the greatest relevance. All relevant articles had to report the incidence of UVAL and include a definition of UVAL.
Anastomotic leakage and RP
(92/relevant 44)
Urinary leakage and RP
(218/relevant 55)
Eligible papers
72
Cystography and RP
(57/relevant 19)
Figure 1. Data extraction strategy
Numerous reports were excluded from study because the term urinary leakage was used to describe incontinence. Items were also excluded unless management of UVAL was reported or any parameter (clinical or surgical) related to UVAL was found. For each eligible publication we determined the type of study, and recorded the sample size, the level of evidence provided and the number of citations. The statistical significance of any variable related to UVAL was also noted. All existing definitions and classifications were recorded. A variety of diagnostic and therapeutic methods were presented, allowing us to create a summarizing algorithm.
RESULTS Literature Search Figure 1 shows the number of hits and eligible studies stratified by the respective combination of terms used. We summarized the major characteristics from each of the 72 studies we used for data extraction. The studies were published between 1990 and 2011, with the majority published in 2005 or thereafter. Of the 72 studies 48 (67%) were observational case series, 16 (22.2%) prospective trials, 1 a letter to the editor, 1 a review and 1 a systematic review focused only on laparoscopic RP. We also found 2 experimental studies performed in animal models and 3 case reports. There were 7 studies that reported results from less than 20 patients. However, a few reports described novel techniques. Incidence The incidence of UVAL is higher in the first 7 to 10 days postoperatively and decreases with time, suggesting that the healing of the anastomosis is time dependent. However, the incidence of UVAL varies significantly due to several parameters such as postoperative day, the amount of extravasation on cystography and the need for intervention. The effect of POD on the incidence of UVAL, at least for the open procedure, can be easily understood from the study by Dalton et al.3 In this study extravasation was found in 78% of patients on POD 8, in 38% on POD 11, in 20% on POD 14 and in 8% of patients 3 weeks after surgery. In the first 5 to 8 days after radical retropubic prostatectomy the incidence of UVAL ranged between 67% and 78%.3 The incidence of urine extravasation on cystograms obtained at PODs 14 and 21 was 0.1% and 20%, respectively.3 In more recent studies the incidence on PODs 4 to 5 and 7 was 28.5% and 25%, respectively. The incidence of UVAL 3 weeks postoperatively ranged from 8% to 14.1%.2,3 This important finding suggests that after 3 weeks the healing process has been completed, and it can
URETHROVESICAL ANASTOMOTIC URINARY LEAK AFTER RADICAL PROSTATECTOMY
be speculated that there is little to be gained by leaving the catheter in place any longer. The incidence rates with LRP and RALP usually follow the same pattern. The laparoscopic incidence rates of UVAL have been reported to be 10%, 15.1% and 17.2% on PODs 4, 5 and 7, respectively.2 Rassweiler et al stratified the incidence rates of UVAL by the need for intervention.2 The incidence rates for LRP requiring intervention were 0.9% to 2.5% vs 0.5% for the open procedure. In a review by Mochtar et al the mean incidence of UVAL for LRP was 9.7%, and no statistical difference between the transperitoneal and the extraperitoneal approach was found.4 Definition-Classification There is no consensus on the definition of UVAL. The 2 most common definitions are based on extravasation on cystography in relation to POD and the production of urine via the drain. Although cystography appears to be the imaging technique of choice for the definition of UVAL, there is a great variance in the POD on which it is performed. On the other hand, it has been suggested that catheter removal can be accomplished safely without cystography. The current literature defines UVAL as persistent contrast extravasation on cystography between PODs 3 and 14.2 An issue of the utmost importance is establishing that the liquid drained is urine, especially if lymphadenectomy has been performed. A high creatinine level in the drain output is the key finding. It is essential to distinguish the clinically significant from the nonsignificant event. It has been advocated that the production of fluid with an increased creatinine concentration from the wound drain or considerable contrast extravasation on a cystogram on POD 7 resulting in an extension of the planned catheterization period can be considered clinically significant leakage.4 Others define UVAL as clinically significant when it 1) extends intraperitoneally, 2) necessitates interventional drainage, 3) results in ileus necessitating readmission and intravenous fluids, and 4) is associated with fever and signs of sepsis necessitating antibiotic therapy or drainage.5 On the other hand, characterization as not clinically significant involves radiological leakage on a cystogram or production via the drain that does not result in extra catheterization time.4 It is also important to refer to the severity grades of UVAL. Usually leakage is classified as minor, moderate or severe, as extraperitoneal or intraperitoneal, and according to the extension of extravasation in relation to the anastomosis or the pelvis.6,7 Patil et al were the first to report quantifiable definitions of UVAL using conventional cystography.8 Extravasations confined immediately around the
371
anastomosis are usually characterized as small or grade I, while those that extend to more distant areas are characterized as moderate or grade II. Finally the ones that are intraperitoneal, extend in the perivesical space or are confined in the pelvis are characterized as large, significant or grade III.3 Standardization is required for unbiased evaluation in the reporting of complications. Thus, UVAL should be classified according to the Clavien classification of surgical complications.9 It is apparent that UVAL grading depends on the severity and the need for intervention. Predictive Factors The predictive factors of UVAL can be related to patient characteristics, preoperative and postoperative outcomes, surgeon experience or the technical details of the procedure. The literature may present these factors as debatable since the number of studies on each factor is too limited to perform a metaanalysis. Thus, our assumptions can be considered observations rather than statements of certainty. The patient factors that might be implicated are ischemic heart disease7 and obesity.10,11 Obesity is thought to create difficulties in accessing the vesicourethral anastomosis or to decrease visibility, but only in the open approach.10 In cases of LRP or RARP UVAL occurs but it is not affected by obesity.10 However, other groups suggest the opposite.11 In addition, urinary tract infection appears to increase the possibility of UVAL.12 Excessive bleeding can impair visualization, rendering precise suture placement difficult.13 The role of hemorrhage as a putative risk factor is also suggested by the coexistence of pelvic hematoma and UVAL as depicted on postoperative cystography and changes in postoperative hematocrit. Again, the opposite opinion has been encountered, suggesting that blood loss has no predictive value in identifying leakage.12 Blood clot formation has also been implicated since a blood clot can obstruct the urethral catheter and interfere with urine drainage, compromising the healing of the anastomosis.14,15 Despite the fact that no trial has determined the actual learning curve for achieving a 0% incidence of UVAL, one could argue that surgical experience is related to UVAL. In LRP the learning curve is considered to be approximately 50 procedures but ranges from 40 to 180.4 The achievement of a watertight anastomosis demands expertise in laparoscopic suturing and this is apparent in the higher incidence of UVAL in initial series of LRP and its decrease in subsequent series.16 Several aspects of the surgery are also debatable. Eversion of the mucosa might be deleterious in the accomplishment of a watertight anastomosis.7,17,18 A comparison of anastomoses with and without
372
URETHROVESICAL ANASTOMOTIC URINARY LEAK AFTER RADICAL PROSTATECTOMY
eversion of the mucosa shows the slight superiority of the noneversion technique.7 Moreover prostate size14 and previous prostatic surgery7,8,12 have been recognized, but their roles are supported by both arguments. Specific technical details in the performance of the anastomosis are potential predictive factors. Interrupted or running sutures do not seem to have an effect.19 –21 In addition, the fewer the interrupted sutures used (2, 4, or more in various positions), the more they increase the chance of UVAL, but not in a statistically significant manner.22–24 Sammon et al reported that a double layer vs a single layer anastomosis could be advantageous.25 Others have reported on catheterless or even sutureless techniques, which are all risks for UVAL. However, this finding was not clinically significant because no further intervention was required and it did not add to patient morbidity.26 –28 The type of suture, which exhibits different characteristics, might also have a role. A comparison in the literature of MAXON™, DEXON™ and chromic catgut sutures revealed an advantage of chromic sutures over the others, but the data were based only on a small number of DEXON anastomoses. Barbed vs monofilament sutures were also compared without revealing significant differences.29 Yasui et al reported less UVAL when using LAPRA-TY® clips with a single knot technique.30 A difficult anastomosis is another possible risk factor for UVAL. The posterior or posterolateral segment of the anastomosis is more commonly recognized as the most difficult because it is deep inside the pelvis and requires great surgical skill for its construction, especially in LRP.12,15 Another difficulty lies in the length or thickness of the urethral stump. A comparison showed that patients with a difficult anastomosis had leakage significantly more often than those with no difficulties (32% vs 4.9%).12 The role of musculofascial posterior reconstruction is also debatable. Although the majority of studies agree that it is beneficial,31 2 recent randomized trials have shown that it has no protective role apart from the earlier recovery of continence.32 Thus, the benefit of posterior reconstruction can be considered controversial at best. A widely accepted method of checking the integrity of the anastomosis is the intraoperative flush test. It is a common practice after completion of the anastomosis to fill the bladder through the indwelling catheter with 60 to 250 ml sterile water and to check for a possible leak before the closure begins. An unsatisfactory test is usually defined as the leakage of the sterile water through the anastomosis. The value of an unsatisfactory flush test is often emphasized as it may indicate the need for additional sutures or even complete revision of the anas-
tomosis in the case of a major leak. Others recognize the importance of the test due to its ability to predict a leak on postoperative cystography.14 –16 More specifically, it has been reported that 97.1% of the cystograms performed on POD 7 can be avoided according to the results of the intraoperative test and that the test can be the only independent predictor of success for the early catheter removal (p ⫽ 0.0069).33 The comparison among patients with unsatisfactory test results showed that more patients leak (48.4%) than do not leak (15.8%) (p ⫽ 0.002).12 We should also mention that patients undergoing salvage RP after pelvic radiation are at a definite risk for UVAL compared to treatment naïve patients.34 Another cause of UVAL related to a wide bladder neck is the proximity of the ureteral orifices to the anastomosis. When this is the case, the balloon of the catheter might place the orifice outside the anastomosis, leading to leakage. This can be diagnosed with excretory urography and the surgeon should be aware of such an event. Lastly Parivar et al reported no correlation between positive surgical margins and healing of the urethrovesical anastomosis.35 Diagnosis The simplest way to diagnose UVAL is by assessing the clinical condition of the patient in conjunction with careful monitoring of the fluid emerging from the drain. The most urgent clinical condition that can imply significant UVAL is the complaint of abdominal distention and pain caused by ileus and peritonitis.36 It is important to mention that a high output of urine through the drain during the first hour can be encountered, and that this usually decreases in the first 2 or 3 days.4 A high urine output through the drain is not necessarily due to UVAL, but it can be the result of an iatrogenic ureteral injury which requires urgent management. Apart from the obvious result of the intraoperative flush test, retrograde cystography is often performed before catheter removal. The catheter is removed and retrograde cystography is performed between PODs 2 and 21.2 When the first cystography depicts moderate to large leakage in the anastomosis, prolonged catheterization is applied depending on surgeon preference. Before the second attempt, retrograde cystography is performed.2,15 Others suggest that when the initial cystography shows minimal leakage, then the catheter can be removed a week later, usually without performing a new cystography.7 Although cystography is considered a valuable diagnostic tool, the fact that it is an x-ray application and has a significant cost has led many authors to remove the catheter without cystography. This is evident in LRP and RARP series in which it has
URETHROVESICAL ANASTOMOTIC URINARY LEAK AFTER RADICAL PROSTATECTOMY
been advocated that the catheter can be safely removed at POD 4 to 10 without cystography.3 The same concept has been proposed for the open procedure, suggesting catheter removal on POD 8 to 9 in appropriate selected patients.3 Others support the same practice, but suggest that it is safe between POD 14 to 20. In recent RARP series there has been a general tendency not to perform cystography if the surgeon plans to remove the catheter after POD 8.37 Recently Han et al reported that CT cystography is superior to conventional cystography.38 Lee et al made a similar observation by evaluating multidetector CT cystography, which revealed a statistically significant difference in detection rates compared to conventional cystography (80.4% vs 54.3%).39 In both studies CT cystography upgraded UVAL (for example from mild to moderate), establishing the need for prolonged catheterization. Furthermore, CT eliminated the blind spots of conventional cystography where extravasation is located. Another modality is transrectal ultrasonography with or without the use of enhancing contrast medium. TRUS can be used to successfully assess the integrity of the UVAL while maintaining good specificity and sensitivity, with a much lower cost than cystography and with no emission of radiation.6 The drawbacks of TRUS include the depiction of paraurethral fluid which can mimic UVAL, reflections from the catheter which can obscure the ventral part of the anastomosis, patient discomfort and a hypothetical negative effect on the anastomosis due to the applied pressure. A modified version of the cystogram is the pericatheterogram, during which a small caliber catheter (10Fr) is introduced into the urethral meatus alongside the existing Foley. Contrast medium is instilled in the urethra under gravity until bladder is full and x-ray films are obtained.40 Most recently contrast enhanced ultrasound was reported to be a promising imaging modality, with a low cost, that seems to be equivalent to cystography for detecting large UVAL.41 In general CT cystography has a superior sensitivity and specificity with a greater cost and radiation exposure. Despite its limitations, TRUS has a higher accuracy for the close tissue patterns in the supposed perianastomotic lesion. A randomized trial comparing conventional cystography, CT, enhanced ultrasound and TRUS cystography would probably establish these assumptions. Lastly, an interesting point was reported by Schenck and Schneider, who suggested that uroscopy (if blood was cloudy or colored by old blood) could predict UVAL in 86% of cases.42 Management We composed an algorithm based on our experience with more than 200 RPs per year and the current
373
literature (fig. 2). In this algorithm we incorporated another algorithm by Do et al, who adjusted their maneuvers according to the amount of extravasation on cystography.43 The prevalence of UVAL requiring intervention has been estimated to be 0.9% to 2.5% after LRP and 0.5% after an open procedure.14 The most common practice is prolonged catheterization and, again, the timing of catheter removal varies. The consensus is that cystography must be repeated before catheter removal.7,12 The majority of the authors removed the catheter only if subsequent cystography showed no extravasation.7,39,40 Several conservative measures have been proposed to help the healing process, the most ordinary being the application of gentle traction to the catheter.14,15,36 However, some suggest that traction might induce ischemia on the anastomosis and impair healing, although this has not been verified scientifically. Correct replacement of the catheter has been proposed to help in the management of UVAL. If no fenestrated catheter has been used, radiological mobilization and adjustment of the catheter, deflation of the balloon and external fixation can contribute to proper urine drainage.4 Other conservative ways to manage UVAL concern the drain. The drain may act as a pipette when it touches the anastomosis and this is easily recognized during retrograde cystography. Switching from active to passive or gravity drainage, pulling back the drain and withdrawing the drain are proper maneuvers.14,15,36 In the presence of persistent UVAL, 2 specific maneuvers of the transurethral catheter have been reported. An underwater suction drainage system can be used as in extrapleural drainage, with connection to the transurethral catheter.2 The second option involves low intermittent wall suction and an 18 to 20 gauge angiocatheter needle in the tubing connecting the catheter to the wall suction device.14 The continuous needle vented catheter suction prevents the continuous suction of the bladder mucosa, especially when the musosa obstructs the catheter tip. Aspiration of a postoperative urinoma can be accomplished with interventional radiology, and CT guided drainage and placement of a pelvic drain.8,33,36 When these methods fail and before the decision for reoperation, ureteral stenting or even nephrostomy placement are the next steps.4 Of patients treated with LRP 1.6% to 2.2% require ureteral stenting or nephrostomy tube placement.4 In cases with a major leak the insertion of a mono-J catheter and a minimum of 2 weeks of catheterization have been recommended as an alternative.15 The insertion of a nephroureteral stent has been also combined with the connection of the stent to a TRU-CLOSE® suction drainage system.41 This combination effec-
374
URETHROVESICAL ANASTOMOTIC URINARY LEAK AFTER RADICAL PROSTATECTOMY
Lymphadenectomy performed?
YES NO (probably lymphorrhea but:) Check catheter function Flush test Blocked
OK
Remove clots or change catheter using guidewire
Creatinine determination
PROLONGED CATHETERIZATION
LOW
HIGH
<300cc >300cc Lymphorrhea=Prolonged drainage
Apply gentle traction
REDUCES
PERSISTS
REMOVE DRAIN IF OUTPUT <50cc
Perform cystography+ Rule out ureteral injury Order IVU
Check algorithm by Do et al
Deflate balloon and tack catheter Mobilize drain
REDUCES
PERSISTS
Suprapubic catheter Nephrostomy Endoscopic evaluation Consider revision Figure 2. Management algorithm. IVU, excretory urography.
tively diverts the urine output, effacing the output from the transurethral catheter and the pelvic drain. It should be emphasized that urinary diversion using stents might be efficient to stop UVAL and allow healing. However, we found no clear evidence of its efficacy in achieving a potent anastomosis. The last reserve is reoperation, and this depends on the amount of urine drainage and the use of the transperitoneal or extraperitoneal approach. When the transperitoneal route has been chosen, then
UVAL can lead to uroperitoneum and to more dramatic manifestations such as ileus.4 Ileus or peritonitis are not always manifested in cases of major leaks. Reoperation may vary from complete reconstruction of the anastomosis to a partial anastomotic repair.15,36 Data extracted from LRP series show that reoperation can been performed early in the postoperative period such as POD 2, or later such as PODs 6 to 12.15–36 The reoperation in the LRP series is performed with laparoscopy and the approach, transperitoneal or extraperitoneal, is the same as
URETHROVESICAL ANASTOMOTIC URINARY LEAK AFTER RADICAL PROSTATECTOMY
for the initial operation.15,36 While continuous suturing was used in the initial operation, in the majority of cases the reoperation is performed with interrupted sutures.15 Complications The most detrimental short-term complication of UVAL is uroperitoneum. Uroperitoneum is a result of the transperitoneal approach of the LRP, and it can lead to more serious clinical conditions such as peritonitis, ileus and renal function deterioration, and even necessitate laparotomy.4,36 Although no relationship has been established between anastomotic strictures or bladder neck contractures and UVAL, the data are contradictory. Several authors support the role of UVAL in inducing strictures10,13,17,44,45 among others who suggest that UVAL obstructs normal wound healing, leading to a persistent inflammatory reaction and finally to anastomotic stricture.8,22 However, some authors propose that strictures cannot be attributed to UVAL since different pathophysiological mechanisms and factors are responsible for the stricture.18,46 Incontinence is a complication that has been attributed to the radical prostatectomy itself. The consensus is that UVAL does not seem to affect longterm incontinence.8,18 However, early recovery of continence may be delayed due to UVAL and among patients with leakage there is a higher incidence of incontinence.8,18 Prevention The only additional measures that can be regarded as preventative with the exception of applying the basic surgical principles of anastomosis are the use of a fenestrated catheter, the application of fibrin sealant in the anastomosis and the placement of oxidized cellulose sponges around the anastomosis. A fenestrated catheter is used to prevent exudate composed of blood and urine from accumulating
375
in the urethral space, thereby preventing UVAL.21,47,48 The fenestration is placed below the balloon, helping the free drainage of the exudate from the important site of UVAL that is healing. The fenestration also aids the drainage when the catheter tip hole is located away from the accumulated fluid or when bladder spasm obstructs the catheter tip hole. The use of fibrin sealant reinforces the anastomosis, promoting healing without inflammation and the formation of fibrosis.49 Although the data in the literature are based on a small number of cases or on preliminary results, they exhibit a trend in decreasing leakage. The application of oxidized cellulose sponges around the anastomotic site in the pelvic fossa to block the escape of exudate has also been reported.21 Similar adhesives as well as laser welding of the anastomosis were proposed.50 Finally, we should emphasize the importance of informing the patient about drains and catheters as well as compliance. Accidental removal, blockage or displacement of catheters or drains is often seen during patient mobilization. Thus, it is important to anticipate such events by evaluating the patient’s understanding of the operation and postoperative course.
CONCLUSIONS UVAL is a troubling, short-term complication that occurs independent of the approach used to perform RP and may create repercussions in patient quality of life. The experience of the surgeon in using the technique with mastery is probably the most crucial parameter in avoiding UVAL. We were able to construct an algorithm for the diagnosis and management of this complication. It is crucial that the management of UVAL be tailored to each patient. The urologist should wait before each intervention, and should always evaluate the patient and drain signs.
REFERENCES 1. Lepor H and Kaci L: Contemporary evaluation of operative parameters and complications related to open radical retropubic prostatectomy. Urology 2003; 62: 702. 2. Hakimi AA, Faleck DM, Sobey S et al: Assessment of complication and functional outcome reporting in the minimally invasive prostatectomy literature from 2006 to the present. BJU Int 2012; 109: 26. 3. Dalton DP, Schaeffer AJ, Garnett JE et al: Radiographic assessment of the vesicourethral anastomosis directing early decatheterization following nerve-sparing radical retropubic prostatectomy. J Urol 1989; 141: 79.
4. Mochtar CA, Kauer PC, Laguna MP et al: Urinary leakage after laparoscopic radical prostatectomy: a systematic review. J Endourol 2007; 21: 1371.
7. Ramsden AR and Chodak GW: Can leakage at the vesico-urethral anastomosis be predicted after radical retropubic prostatectomy? BJU Int 2004; 93: 503.
5. Tan G, Srivastava A, Grover S et al: Optimizing vesicourethral anastomosis healing after robotassisted laparoscopic radical prostatectomy: lessons learned from three techniques in 1900 patients. J Endourol 2010; 24: 1975.
8. Patil N, Krane L, Javed K et al: Evaluating and grading cystographic leakage: correlation with clinical outcomes in patients undergoing robotic prostatectomy. BJU Int 2009; 103: 1108.
6. Eggert T, Palisaar J, Metz P et al: Assessing the vesico-urethral anastomosis after radical retropubic prostatectomy: transrectal ultrasonography can replace cystography. BJU Int 2007; 100: 1268.
9. Constantinides CA, Tyritzis SI, Skolarikos A et al: Short- and long-term complications of open radical prostatectomy according to the Clavien classification system. BJU Int 2009; 103: 336. 10. Sandhu JS, Gotto GT, Herran LA et al: Age, obesity, medical comorbidities and surgical tech-
376
URETHROVESICAL ANASTOMOTIC URINARY LEAK AFTER RADICAL PROSTATECTOMY
nique are predictive of symptomatic anastomotic strictures after contemporary radical prostatectomy. J Urol 2011; 185: 2148. 11. Herman MP, Raman JD, Dong S et al: Increasing body mass index negatively impacts outcomes following robotic radical prostatectomy. JSLS 2007; 11: 438. 12. Gnanapragasam VJ, Baker P, Naisby GP et al: Identification and validation of risk factors for vesicourethral leaks following radical retropubic prostatectomy. Int J Urol 2005; 12: 948. 13. Surya BV, Provet J, Johanson KE et al: Anastomotic strictures following radical prostatectomy: risk factors and management. J Urol 1990; 143: 755. 14. Fenig DM, Slova D and Lepor H: Postoperative blood loss predicts the development of urinary extravasation on cystogram following radical retropubic prostatectomy. J Urol 2006; 175: 146. 15. Moinzadeh A, Abouassaly R, Gill IS et al: Continuous needle vented Foley catheter suction for urinary leak after radical prostatectomy. J Urol 2004; 171: 2366. 16. Liatsikos E, Rabenalt R, Burchardt M et al: Prevention and management of perioperative complications in laparoscopic and endoscopic radical prostatectomy. World J Urol 2008; 26: 571. 17. Stolzenburg JU, Ho KM, Do M et al: Impact of previous surgery on endoscopic extraperitoneal radical prostatectomy. Urology 2005; 65: 325. 18. Webb DR, Sethi K and Gee K: An analysis of the causes of bladder neck contracture after open and robot-assisted laparoscopic radical prostatectomy. BJU Int 2009; 103: 957.
thral anastomosis in radical retropubic prostatectomy. Int Braz J Urol 2009; 35: 158. 25. Sammon JD, Muhletaler F, Peabody JO et al: Long-term functional urinary outcomes comparing single- vs double-layer urethrovesical anastomosis: two-year follow-up of a two-group parallel randomized controlled trial. Urology 2010; 76: 1102. 26. Tewari A, Rao S and Mandhani A: Catheter-less robotic radical prostatectomy using a custommade synchronous anastomotic splint and vesical urinary diversion device: report of the initial series and perioperative outcomes. BJU Int 2008; 102: 1000.
28. Simforoosh N, Javaherforooshzadeh A, Aminsharifi A et al: Early continence after open and laparoscopic radical prostatectomy with sutureless vesicourethral alignment: an alternative technique, 8 years’ experience. Urol J 2009; 6: 163.
41. Schoeppler GM, Buchner A, Zaak D et al: Detection of urinary leakage after radical retropubic prostatectomy by contrast enhanced ultrasound - do we still need conventional retrograde cystography? BJU Int 2010; 106: 1632.
29. Sammon J, Kim TK, Trinh QD et al: Anastomosis during robot-assisted radical prostatectomy: randomized controlled trial comparing barbed and standard monofilament suture. Urology 2011; 78: 572. 30. Yasui T, Itoh Y, Maruyama T et al: The singleknot method with Lapra-Ty clips is useful for training surgeons in vesicourethral anastomosis during laparoscopic radical prostatectomy. Int Urol Nephrol 2009; 41: 281.
20. Poulakis V, Skriapas K, de Vries R et al: Vesicourethral anastomosis during endoscopic extraperitoneal radical prostatectomy: a prospective comparison between the single-knot running and interrupted technique. Urology 2006; 68: 1284.
32. Sutherland DE, Linder B, Guzman AM et al: Posterior rhabdosphincter reconstruction during robotic assisted radical prostatectomy: results from a phase II randomized clinical trial. J Urol 2011; 185: 1262.
21. Ball AJ, Bordeau KP, Davis JW et al: Modified running vesicourethral anastomosis after robotically assisted laparoscopic radical prostatectomy: use of solitary Lapra-Ty to secure posterior approximation. Urology 2005; 66: 16.
33. Ischia JJ and Lindsay S: Is a cystogram necessary after radical prostatectomy? ANZ J Surg 2005; 75: 825.
24. Mazaris EM, Chatzidarellis E, Varkarakis IM et al: Reducing the number of sutures for vesicoure-
39. Lee HJ, Shin CI, Hwang SI et al: MDCT cystography for detection of vesicourethral leak after prostatectomy. AJR Am J Roentgenol 2008; 191: 1847. 40. Castillo OA, Alston C and Sanchez-Salas R: Persistent vesicourethral anastomotic leak after laparoscopic radical prostatectomy: laparoscopic solution. Urology 2009; 73: 124.
31. Coelho RF, Chauhan S, Orvieto MA et al: Influence of modified posterior reconstruction of the rhabdosphincter on early recovery of continence and anastomotic leakage rates after robot-assisted radical prostatectomy. Eur Urol 2011; 59: 72.
23. Gallo L, Perdonà S, Autorino R et al: Vesicourethral anastomosis during radical retropubic prostatectomy: does the number of sutures matter? Urology 2007; 69: 547.
38. Han KS, Choi HJ, Jung DC et al: A prospective evaluation of conventional cystography for detection of urine leakage at the vesicourethral anastomosis site after radical prostatectomy based on computed tomography. Clin Radiol 2011; 66: 251.
27. Krane LS, Bhandari M, Peabody JO et al: Impact of percutaneous suprapubic tube drainage on patient discomfort after radical prostatectomy. Eur Urol 2009; 56: 325.
19. Srougi M, Paranhos M, Leite KM et al: The influence of bladder neck mucosal eversion and early urinary extravasation on patient outcome after radical retropubic prostatectomy: a prospective controlled trial. BJU Int 2005; 95: 757.
22. Teber D, Erdogru T, Cresswell J et al: Analysis of three different vesicourethral anastomotic techniques in laparoscopic radical prostatectomy. World J Urol 2008; 26: 617.
37. Williams TR, Longoria OJ, Asselmeier S et al: Incidence and imaging appearance of urethrovesical anastomotic urinary leaks following da Vinci robotic prostatectomy. Abdom Imaging 2008; 33: 367.
34. Eandi JA, Link BA, Nelson RA et al: Robotic assisted laparoscopic salvage prostatectomy for radiation resistant prostate cancer. J Urol 2010; 183: 133.
42. Schenck M and Schneider T: Easy and effective assessment of the anastomosis after radical retropubic prostatectomy. Uroscopy before imaging? Urologe A 2010; 49: 364. 43. Do HM, Franz T and Stolzenburg JU: Algorithm for the treatment of anastomotic failure after laparoscopic prostatectomy. Urologe A 2011; 50: 1426. 44. Msezane LP, Reynolds WS, Gofrit ON et al: Bladder neck contracture after robot-assisted laparoscopic radical prostatectomy: evaluation of incidence and risk factors and impact on urinary function. J Endourol 2008; 22: 97. 45. Rebuck DA, Haywood S, McDermott K et al: What is the long-term relevance of clinically detected postoperative anastomotic urine leakage after robotic-assisted laparoscopic prostatectomy? BJU Int 2011; 108: 733. 46. Hanson GR, Odom E, Borden LS Jr et al: Postoperative drain output as a predictor of bladder neck contracture following radical prostatectomy. Int Urol Nephrol 2008; 40: 351. 47. Lawrentschuk N, Bolton DM and Angus D: Fenestrated urethral catheter to aid anastomotic drainage after radical prostatectomy. Urology 2005; 65: 160. 48. Kylmälä T, Kaipia A and Matikainen M: Management of prolonged urinary leakage at the urethrovesical anastomosis. Urol Int 2005; 74: 298.
35. Parivar F, Fournier GR Jr and Narayan P: Urethral anastomotic healing after radical retropubic prostatectomy: impact of positive urethral margin. Urology 1994; 44: 705.
49. Diner EK, Patel SV and Kwart AM: Does fibrin sealant decrease immediate urinary leakage following radical retropubic prostatectomy? J Urol 2005; 173: 1147.
36. Shah G, Vogel F and Moinzadeh A: Nephroureteral stent on suction for urethrovesical anastomotic leak after robot-assisted laparoscopic radical prostatectomy. Urology 2009; 73: 1375.
50. Grummet JP, Costello AJ, Swanson DA et al: Laser welded vesicourethral anastomosis in an in vivo canine model: a pilot study. J Urol 2002; 168: 281.