Surg Endosc DOI 10.1007/s00464-014-3737-x
and Other Interventional Techniques
Transanal endoscopic microsurgery for surgical repair of rectovesical fistula following radical prostatectomy Eiji Kanehira • Takashi Tanida • Aya Kamei Masafumi Nakagi • Mitsuharu Iwasaki • Hirofumi Shimizu
•
Received: 3 March 2014 / Accepted: 8 July 2014 Ó Springer Science+Business Media New York 2014
Abstract Background Rectovesical fistula is a rare complication following prostatectomy, associated with significant symptoms such as urinary drainage from anus or faecaluria. While several surgical procedures have been described to treat this condition, none of them has been accepted as the universal standard. Transanal endoscopic microsurgery (TEM) is a well-established endoluminal procedure for local excision of rectal tumors. But its application to the repair of rectovesical fistula has been almost unknown. Methods We performed TEM as a surgical repair for refractory rectovesical fistula developing after radical prostatectomy in 10 patients. Under the magnified threedimensional view, through the stereoscope, the fistula and the surrounding rectal mucosa were precisely resected. The defect and the muscle layer of the rectum were closed by hand-sew technique in four layers. Results Fistula was completely closed in 7 patients, who eventually underwent enterostomy closure, while in the other 3 patients the fistula recurred. In the three recurrent cases, the fistula was associated with wide, tough scar tissue due to previous irradiation, HIFU, or repeated surgical repair attempts. Conclusions Rectovesical fistulas associated with wide, tough scar tissue due to multi-time attempt of surgical repair or any type of energy ablation should not be indicated for repair by TEM. However, for simple fistulas
without tough, fibrotic surroundings, TEM can be indicated as a minimally invasive surgical option with very low morbidity, without any incision in healthy tissue for approach. Keywords Rectovesical fistula Rectourethral fistula Transanal endoscopic microsurgery Endoluminal surgery Rectovesical fistula is a rare complication [1, 2] following radical prostatectomy and sometimes becomes refractory. Such symptoms as urinary drainage from anus or faecaluria deteriorate their quality of life significantly. And they often result in double diversion with enterostomy and suprapubic catheterization. A variety of surgical procedures has been attempted to treat the fistula. But none of them has been accepted as the universal standard. Transanal endoscopic microsurgery (TEM) [3, 4] was originally developed to perform complete en-bloc resection of rectal tumors and closure of the defect by suturing. But its application to fistula closure has been rarely reported. We performed TEM as surgical repair of the rectovesical fistula in 10 patients. Herein, we demonstrate our surgical technique and report the outcomes.
Materials and methods Patients
E. Kanehira (&) T. Tanida A. Kamei M. Nakagi Department of Surgery, Medical Topia Soka, 1-11-18 Yatsuka, Soka, Saitama 340 0028, Japan e-mail: [email protected] M. Iwasaki H. Shimizu Department of Urology, Medical Topia Soka, Soka, Japan
Since July 2004 through December 2013, we treated 10 patients (average age 68.3 years, range 62–76 years) with rectovesical fistula as a complication found after radical prostatectomy for prostatic cancer. The most common complaint was urinary drainage from the anus (7 patients), while pneumaturia (5) or faecaluria (5) was also noted as
123
Surg Endosc Table 1 Patient characteristics and outcomes Patient
Age
Previous surgical attempt (times)
Foley catheter
Stoma
Previous ablation
Months from onset to TEM
OP time (minutes)
Complication
Successful correction
1
66
1
Y
Y
N
6
95
N
Y
2
69
2
Y
Y
Y (irradiation & HIFU)
36
210
N
N
3 4
70 64
2 1
Y Y
Y Y
N N
14 8
101 93
N N
Y Y
5
68
3
Y
Y
N
6
108
N
N
6
71
1
Y
Y
N
12
116
N
Y
7
76
1
Y
Y
N
14
124
N
Y
8
62
0
Y
Y
N
9
120
N
Y
9
70
3
Y
Y
N
13
165
N
N
67
0
Y
Y
N
11
140
N
Y
68.3
80 %
100 %
100 %
10 %
12.9
127.2
0%
70 %
10
significant symptoms. All patients had an indwelling silicone rubber Foley catheter and had a diverting enterostomy (ileostomy in 7, colostomy in 3). Time between the detection of the fistula and the treatment in our hospital ranged 7–34 months (average 12.9 months). None of the patients had an evident sign of recurrence or metastasis of the prostatic cancer. In all cases, rectoscopy, cystoscopy, and the cystography were performed. In all cases, rectoscopy demonstrated the orifice of the fistula in the lower rectum near the dentate line. The diameter of the orifice ranged 5–20 mm. By cystography, the fistula tract was visualized in 5 patients while in the other 5 patients it was not visualized. 8 patients had undergone the previous attempt of fistula closure in another hospital before they were referred to us. In all cases, the procedure of previous surgical attempts was only direct closure of the fistula orifice by suturing through conventional transanal approach (Table 1). Operative technique Before the operation, cystoscopy is performed and both ureters were catheterized for protection during the operation. Operation was performed under general anesthesia. The patient is placed in jack-knife position with legs apart. The original TEM (Richard-Wolf GmbH., Germany) system [3] is utilized. The short version of the operating rectoscope is inserted gently into the lower rectum. Then the handle of the rectoscope is attached and fixed at the upper end of the supporting arm, which is firmly connected to the operating table. The distal end of the rectoscope is covered with the metal working insert with five channels, through which the binocular microscope and other operating instruments are introduced into the rectal cavity. The magnified view is obtained by three dimensions through
123
the eyepieces of the microscope for the operator, while a monitor also shows the operation field for the other OR staffs. By adjusting the position of the rectoscope, the orifice of the fistula is well visualized at the center of the endoscopic view field. On the rectal mucosa around the fistula, orifice marking dots are drawn by the tip of the needle-shaped electrode with high-frequency coagulation so that the circle will be approximately 3 cm in diameter (Fig. 1A). Normal saline is injected into the submucosal layer underneath this circle. Then the mucosa is dissected by the electrocautery so that the proper muscle layer is completely exposed. The mucosa covering the fistula cavity is also resected. The resection is extended to the mucosa of the urinary bladder until a part of the catheter tube to the bladder is recognized (Fig. 1B). When the deepest part of the fistula is not sufficiently exposed, it is ablated with high-frequency coagulation as much as possible so that viable epitheria would not remain in the fistula tract. After the resection of the mucosal layer and the fistula, the defect is closed by suturing. In endoluminal suturing, instead of the needle driver specially designed for TEM, we mainly use a pair of laparoscopic needle drivers (KarlStorz GmbH, Germany) called ‘‘flamingo and parrot’’. Firstly, the bottom of the fistula, which is regarded as the urinary bladder layer, is sutured with 4–0 absorbable monofilament thread. Knot-tying is performed in the rectal cavity, facilitated by the fine tips of the needle drivers. Then the deeper muscle layer of the rectal wall, which is the central part of the round mucosal defect, is approximated by the same manner with 3–4 interrupted sutures with 4–0 absorbable monofilament thread (Fig. 1C). The shallower muscle layer is then closed in a same manner with 3–4 stitches. Lastly, the mucosal layer is approximated by running suture with 3–0 absorbable monofilament
Surg Endosc Fig. 1 A Operative view via the rectoscope. The fistula orifice is visualized on the anterior wall of the lower rectum. Coagulation dots are drawn along the planned resection circle around the fistula orifice. B The mucosa around the fistula and the fistula epithelia are resected with highfrequency needle electrode. C The muscle layer is closed by hand-sew suture technique with 4–0 monofilament absorbable thread. D The mucosal layer is closed to finish the repair
thread. For the last part of the suturing, we use the original TEM needle driver. And instead of knot-tying at the mucosal closure, the end of the thread is secured with a suture clip (LapraTyÒ, Ethicon Endosurgery, USA) fixed at the end of the thread by specially designed clip applier (Fig. 1D). Patients start walking on POD 1, when water intake is also allowed. Soft diet is started on POD 2. Broad spectrum antibiotics are administered intravenously for 3 days postoperatively. Patients are discharged on POD 5–7. Between 5th and 8th week, after the operation, cystography and rectoscopy are performed to observe the repaired part. When the recurrence of the fistula is not recognized, the catheter to the urinary bladder was removed. After removal of the urinary bladder catheter, when any symptoms related to possible fistula recurrence are not noted for approximately 1 month, the enterostomy is reversed.
history of repeated irradiation and HIFU as adjuvant treatments for prostatic cancer prior to our surgical repair, and the size of the fistula orifice was 2 cm in diameter. In addition, this case underwent the previous repair attempts twice. Both of the other 2 cases had previous surgical attempts three times. In those 3 cases, during the operation, we found the tissue around the fistula was extremely tough due to severe fibrosis, and dissection and closure were difficult. One of the recurrent cases underwent a further repair by perineal approach with gracilis muscle transposition, and the fistula was successfully closed eventually. The other 2 patients refused any additional surgical treatment. In 7 cases (70 %), the fistula was judged completely closed. In all 7 patients, the enterostomy was reversed. After the enterostomy closure, none of them complained of symptoms suggesting recurrent fistula.
Results
Discussion
In all cases, the operation was not converted to any other procedure. Operation time ranged from 93 to 210 min with an average of 127.2 min. During the operation, no complication was encountered. Bleeding was less than 20 ml in all cases. Postoperative complication was nil either. In 3 patients, fistula recurred within 8 weeks. They complained of symptoms such as pneumaturia or faecaluria again. Recurrence was confirmed by colonoscopy and/or cystography. Of these 3 cases with recurrence, one had a
Rectovesical fistula is a rare complication after radical prostatectomy. It has been reported that the incidence was 0.5–9 % [1, 2]. There seems to be a chance to close the fistula spontaneously when intestinal diversion and/or urinary diversion are constructed. However, majority of those patients eventually need surgical repair [1, 2]. A variety of surgical procedures has been tried to challenge this problem, although none of them has been universally accepted so far. In a systematic review with
123
Surg Endosc
416 cases, perineal approach with gracilis muscle interposition was selected in the majority of cases [5]. This approach gained the highest rate of successful closure. Other studies [5–8] also support the fact that perineal approach with gracilis muscle transposition must be the most reliable treatment with 83–100 % success rate, even when the fistula was more complicated as larger than 2 cm, surrounded by fibrotic scar tissue, severely affected by irradiation or ablation. Drawback of this approach might be a possibility of infection in the perineum, or numbness in the thigh. Trans-sphincter approach such as York-Mason procedure has been reported with preferable outcomes [9]. Hadely et al. [10] have reported the largest series of YorkMason repair for 40 years with excellent outcomes. They concluded that York-Mason procedure should be considered as the first choice, as it was less invasive associated with very few incidence of morbidity. They also suggest that those who underwent irradiation and/or ablation in the fistula site prior to the repair had a tendency to fail and might not be a good candidate for their procedure. Wilbert et al. [11] reported 2 cases with a rectourethral fistula which was successfully corrected by TEM. The most important advantage in using TEM, we think, is the fact that visualization of the fistula is excellent without any incision in healthy tissue such as in York-Mason procedure or perineal approach. As described in the past literatures [3, 4, 12, 13] and the authors know from our experiences, TEM has a lot of technical advantages compared with conventional transanal approach. The magnified threedimensional image obtained via the binocular endoscope enables surgeons to identify, for example, tiny tissue structures, difference between healthy tissues and ischemic tissues, and so on. This ability plays an important role in stitching correct layer of the proper muscle in multiple layer suturing. Wide operation field in TEM enhances maneuverability of curved instruments. In the original TEM procedure, the needle driver specially designed for TEM is supposed to be used [3]. This needle driver is not designed for knot-tying, and as in ordinary TEM, the suture is secured with specially designed suture clips made of pure silver. During the repair of the fistula, we use mainly a pair of laparoscopic needle drivers instead of the original needle driver of TEM, because in the repair of the fistula, the closure is performed by multiple layers. And we believe suture in the deeper layers should be secured not with silver clips, but with knot-tying, as we do not want to remain the silver clips inside the rectal wall. Although TEM has less simplicity and convenience, once the institute is equipped with it and surgeons are familiar with the unique dexterity required, there should be a chance to apply it to this challenging treatment.
123
Moreover, the fact that TEM instruments are reusable may encourage the institutes to expand its indications. Another advantage is that TEM is associated with extremely low incidence of morbidity [12, 13]. We presume that conventional transanal surgical approach fails with considerably high possibility, although there have been very few reports in the past and it is difficult to predicate. In the present series, for example, 8 patients had histories of previous surgical repair attempts via conventional transanal approach, and all failed. Moreover in a recent systematic review, conventional transanal surgical repair was revealed to be an unfavorable procedure associated with the worst failure rate of 41 %, compared with a variety of other surgical treatments [8]. We think that in the conventional transanal approach, the technical precision is not as high as in trans-sphincter approach or TEM, as it has narrower operative field and visibility, and maneuverability should be deteriorated. In our series, fistula repair failed in 3 patients (30 %). In one patient, the fistula orifice was as large as 2 cm, surrounded by tough scar tissue due to repeated irradiation and HIFU ablation prior to the reference. During the operation after fistulectomy, ideal tissue approximation was not feasible. The other 2 cases of recurrence had undergone repeated surgical repair via conventional transanal approach before they were referred to us. At the reference, the rectal wall around the fistula was very tough with scars and looked ischemic. So far we are discouraged to indicate the current surgical treatment for patients with history of irradiation and/or HIFU [5, 7, 10]. In addition, we think the cases with previous surgical attempts are associated with more risk of failure than fresh cases [10, 14]. Recently, some favorable results by transanal approach using flexible colonoscopy with a newly developed clip [15, 16] have been reported [17, 18]. When we know to which type of fistula this kind of endoscopic treatment is appropriately indicated, this procedure should be regarded as minimally invasive choice of treatment and beneficial to the indicated patients. However, this challenge has been started only recently and it should wait for larger series of experience. Throughout our experience of 10 patients, we think that fistulas with history of irradiation and/or ablation are not ideal candidates for repair by TEM. In addition, fistulas associated with wide, tough scar tissue due to past attempt of surgical repair should not be indicated for TEM, either. Those complicated fistulas must be considered to be corrected by perineal approach with gracilis muscle interposition. However, we think TEM is a very beneficial option in the treatment of simple fistulas without tough, fibrotic surroundings. It provides a chance of complete closure of fistula with very low morbidity, without any incision in healthy tissue for approach.
Surg Endosc Disclosures Eiji Kanehira, Takashi Tanida, Aya Kamei, Masafumi Nakagi, Mitsuharu Iwasaki, and Hirofumi Shimizu have no conflicts of interest or financial ties to disclose.
References 1. Thomas C, Jones J, Ja¨ger W, Hampel C, Thu¨roff JW, Gillitzer RJ (2009) Incidence, clinical symptoms and management of rectourethral fistulas after radical prostatectomy. J Urol 183:608–612 2. Buckley JC (2011) Complications after radical prostatectomy: anastomotic stricture and rectourethral fistula. Curr Opin Urol 21:461–464 3. Buess G, Kipfmu¨ller K, Hack D, Gru¨ssner R, Heintz A, Junginger T (1988) Technique of transanal endoscopic microsurgery. Surg Endosc 2:71–75 4. Lirici MM, Kanehira E, Melzer A, Schurr MO (2014) The outburst age: how TEM ignited the MIS revolution. Minim Invasive Ther Allied Technol 23:1–4 5. Zmora O, Tulchinsky H, Gur E, Goldman G, Klausner JM, Rabau M (2006) Gracilis muscle transposition for fistulas between the rectum and urethra or vagina. Dis Colon Rectum 49:1316–1321 6. Zmora O, Potenti FM, Wexner SD, Pikarsky AJ, Efron JE, Nogueras JJ, Pricolo VE, Weiss EG (2003) Gracilis muscle transposition for iatrogenic rectourethral fistula. Ann Surg 237:483–487 7. Voelzke BB, McAninch JW, Breyer BN, Glass AS, Garcia-Aguilar J (2013) Transperineal management for postoperative and radiation rectourethral fistulas. J Urol 189:966–971 8. Hechenbleikner EM, Buckley JC, Wick EC (2013) Acquired rectourethral fistulas in adults: a systematic review of surgical repair techniques and outcomes. Dis Colon Rectum 56:374–383 9. Renschler TD, Middleton RG (2003) 30 years of experience with York-Mason repair of recto-urinary fistulas. J Urol 170:1222–1225
10. Hadley DA, Southwick A, Middleton RG (2012) York-Mason procedure for repair of recto-urinary fistulae: a 40-year experience. BJU Int 109:1095–1098 11. Wilbert DM, Buess G, Bichler KH (1996) Combined endoscopic closure of rectourethral fistula. J Urol 155:256–258 12. Kanehira E, Tanida T, Kamei A, Nakagi M, Hideshima A (2014) A single surgeon’s experience with transanal endoscopic microsurgery over 20 years with 153 early cancer cases. Minim Invasive Ther Allied Technol 23:5–9 13. Arezzo A, Arolfo S, Cravero F, Migliore M, Allaix ME, Morino M (2014) Which treatment for large rectal adenoma? Preoperative assessment and therapeutic strategy. Minim Invasive Ther Allied Technol 23:21–27 14. Nfonsam VN, Mateka JJ, Prather AD, Marcet JE (2013) Shortterm outcomes of the surgical management of acquired rectourethral fistulas: does technique matter? Res Rep Urol 30:47–51 15. Schurr MO, Arezzo A, Ho CN, Anhoeck G, Buess G, Di Lorenzo N (2008) The OTSC clip for endoscopic organ closure in NOTES: device and technique. Minim Invasive Ther Allied Technol 17:262–266 16. Weiland T, Fehlker M, Gottwald T, Schurr MO (2012) Performance of the OTSC System in the endoscopic closure of gastrointestinal fistulae–a meta-analysis. Minim Invasive Ther Allied Technol 21:249–258 17. Mangiavillano B, Pisani A, Viaggi P, Arena M, Opocher E, Mangano M, Santoro T, Masci E (2010) Endoscopic sealing of a rectovesical fistula with a combination of an over the scope clip and cyano-acrylate injection. J Gastrointest Oncol 1:122–124 18. Mori H, Kobara H, Fujihara S, Nishiyama N, Kobayashi M, Masaki T, Izuishi K, Suzuki Y (2012) Rectal perforations and fistulae secondary to a glycerin enema: closure by over-thescope-clip. World J Gastroenterol 28:3177–3180
123
and Other Interventional Techniques
Transanal endoscopic microsurgery for surgical repair of rectovesical fistula following radical prostatectomy Eiji Kanehira • Takashi Tanida • Aya Kamei Masafumi Nakagi • Mitsuharu Iwasaki • Hirofumi Shimizu
•
Received: 3 March 2014 / Accepted: 8 July 2014 Ó Springer Science+Business Media New York 2014
Abstract Background Rectovesical fistula is a rare complication following prostatectomy, associated with significant symptoms such as urinary drainage from anus or faecaluria. While several surgical procedures have been described to treat this condition, none of them has been accepted as the universal standard. Transanal endoscopic microsurgery (TEM) is a well-established endoluminal procedure for local excision of rectal tumors. But its application to the repair of rectovesical fistula has been almost unknown. Methods We performed TEM as a surgical repair for refractory rectovesical fistula developing after radical prostatectomy in 10 patients. Under the magnified threedimensional view, through the stereoscope, the fistula and the surrounding rectal mucosa were precisely resected. The defect and the muscle layer of the rectum were closed by hand-sew technique in four layers. Results Fistula was completely closed in 7 patients, who eventually underwent enterostomy closure, while in the other 3 patients the fistula recurred. In the three recurrent cases, the fistula was associated with wide, tough scar tissue due to previous irradiation, HIFU, or repeated surgical repair attempts. Conclusions Rectovesical fistulas associated with wide, tough scar tissue due to multi-time attempt of surgical repair or any type of energy ablation should not be indicated for repair by TEM. However, for simple fistulas
without tough, fibrotic surroundings, TEM can be indicated as a minimally invasive surgical option with very low morbidity, without any incision in healthy tissue for approach. Keywords Rectovesical fistula Rectourethral fistula Transanal endoscopic microsurgery Endoluminal surgery Rectovesical fistula is a rare complication [1, 2] following radical prostatectomy and sometimes becomes refractory. Such symptoms as urinary drainage from anus or faecaluria deteriorate their quality of life significantly. And they often result in double diversion with enterostomy and suprapubic catheterization. A variety of surgical procedures has been attempted to treat the fistula. But none of them has been accepted as the universal standard. Transanal endoscopic microsurgery (TEM) [3, 4] was originally developed to perform complete en-bloc resection of rectal tumors and closure of the defect by suturing. But its application to fistula closure has been rarely reported. We performed TEM as surgical repair of the rectovesical fistula in 10 patients. Herein, we demonstrate our surgical technique and report the outcomes.
Materials and methods Patients
E. Kanehira (&) T. Tanida A. Kamei M. Nakagi Department of Surgery, Medical Topia Soka, 1-11-18 Yatsuka, Soka, Saitama 340 0028, Japan e-mail: [email protected] M. Iwasaki H. Shimizu Department of Urology, Medical Topia Soka, Soka, Japan
Since July 2004 through December 2013, we treated 10 patients (average age 68.3 years, range 62–76 years) with rectovesical fistula as a complication found after radical prostatectomy for prostatic cancer. The most common complaint was urinary drainage from the anus (7 patients), while pneumaturia (5) or faecaluria (5) was also noted as
123
Surg Endosc Table 1 Patient characteristics and outcomes Patient
Age
Previous surgical attempt (times)
Foley catheter
Stoma
Previous ablation
Months from onset to TEM
OP time (minutes)
Complication
Successful correction
1
66
1
Y
Y
N
6
95
N
Y
2
69
2
Y
Y
Y (irradiation & HIFU)
36
210
N
N
3 4
70 64
2 1
Y Y
Y Y
N N
14 8
101 93
N N
Y Y
5
68
3
Y
Y
N
6
108
N
N
6
71
1
Y
Y
N
12
116
N
Y
7
76
1
Y
Y
N
14
124
N
Y
8
62
0
Y
Y
N
9
120
N
Y
9
70
3
Y
Y
N
13
165
N
N
67
0
Y
Y
N
11
140
N
Y
68.3
80 %
100 %
100 %
10 %
12.9
127.2
0%
70 %
10
significant symptoms. All patients had an indwelling silicone rubber Foley catheter and had a diverting enterostomy (ileostomy in 7, colostomy in 3). Time between the detection of the fistula and the treatment in our hospital ranged 7–34 months (average 12.9 months). None of the patients had an evident sign of recurrence or metastasis of the prostatic cancer. In all cases, rectoscopy, cystoscopy, and the cystography were performed. In all cases, rectoscopy demonstrated the orifice of the fistula in the lower rectum near the dentate line. The diameter of the orifice ranged 5–20 mm. By cystography, the fistula tract was visualized in 5 patients while in the other 5 patients it was not visualized. 8 patients had undergone the previous attempt of fistula closure in another hospital before they were referred to us. In all cases, the procedure of previous surgical attempts was only direct closure of the fistula orifice by suturing through conventional transanal approach (Table 1). Operative technique Before the operation, cystoscopy is performed and both ureters were catheterized for protection during the operation. Operation was performed under general anesthesia. The patient is placed in jack-knife position with legs apart. The original TEM (Richard-Wolf GmbH., Germany) system [3] is utilized. The short version of the operating rectoscope is inserted gently into the lower rectum. Then the handle of the rectoscope is attached and fixed at the upper end of the supporting arm, which is firmly connected to the operating table. The distal end of the rectoscope is covered with the metal working insert with five channels, through which the binocular microscope and other operating instruments are introduced into the rectal cavity. The magnified view is obtained by three dimensions through
123
the eyepieces of the microscope for the operator, while a monitor also shows the operation field for the other OR staffs. By adjusting the position of the rectoscope, the orifice of the fistula is well visualized at the center of the endoscopic view field. On the rectal mucosa around the fistula, orifice marking dots are drawn by the tip of the needle-shaped electrode with high-frequency coagulation so that the circle will be approximately 3 cm in diameter (Fig. 1A). Normal saline is injected into the submucosal layer underneath this circle. Then the mucosa is dissected by the electrocautery so that the proper muscle layer is completely exposed. The mucosa covering the fistula cavity is also resected. The resection is extended to the mucosa of the urinary bladder until a part of the catheter tube to the bladder is recognized (Fig. 1B). When the deepest part of the fistula is not sufficiently exposed, it is ablated with high-frequency coagulation as much as possible so that viable epitheria would not remain in the fistula tract. After the resection of the mucosal layer and the fistula, the defect is closed by suturing. In endoluminal suturing, instead of the needle driver specially designed for TEM, we mainly use a pair of laparoscopic needle drivers (KarlStorz GmbH, Germany) called ‘‘flamingo and parrot’’. Firstly, the bottom of the fistula, which is regarded as the urinary bladder layer, is sutured with 4–0 absorbable monofilament thread. Knot-tying is performed in the rectal cavity, facilitated by the fine tips of the needle drivers. Then the deeper muscle layer of the rectal wall, which is the central part of the round mucosal defect, is approximated by the same manner with 3–4 interrupted sutures with 4–0 absorbable monofilament thread (Fig. 1C). The shallower muscle layer is then closed in a same manner with 3–4 stitches. Lastly, the mucosal layer is approximated by running suture with 3–0 absorbable monofilament
Surg Endosc Fig. 1 A Operative view via the rectoscope. The fistula orifice is visualized on the anterior wall of the lower rectum. Coagulation dots are drawn along the planned resection circle around the fistula orifice. B The mucosa around the fistula and the fistula epithelia are resected with highfrequency needle electrode. C The muscle layer is closed by hand-sew suture technique with 4–0 monofilament absorbable thread. D The mucosal layer is closed to finish the repair
thread. For the last part of the suturing, we use the original TEM needle driver. And instead of knot-tying at the mucosal closure, the end of the thread is secured with a suture clip (LapraTyÒ, Ethicon Endosurgery, USA) fixed at the end of the thread by specially designed clip applier (Fig. 1D). Patients start walking on POD 1, when water intake is also allowed. Soft diet is started on POD 2. Broad spectrum antibiotics are administered intravenously for 3 days postoperatively. Patients are discharged on POD 5–7. Between 5th and 8th week, after the operation, cystography and rectoscopy are performed to observe the repaired part. When the recurrence of the fistula is not recognized, the catheter to the urinary bladder was removed. After removal of the urinary bladder catheter, when any symptoms related to possible fistula recurrence are not noted for approximately 1 month, the enterostomy is reversed.
history of repeated irradiation and HIFU as adjuvant treatments for prostatic cancer prior to our surgical repair, and the size of the fistula orifice was 2 cm in diameter. In addition, this case underwent the previous repair attempts twice. Both of the other 2 cases had previous surgical attempts three times. In those 3 cases, during the operation, we found the tissue around the fistula was extremely tough due to severe fibrosis, and dissection and closure were difficult. One of the recurrent cases underwent a further repair by perineal approach with gracilis muscle transposition, and the fistula was successfully closed eventually. The other 2 patients refused any additional surgical treatment. In 7 cases (70 %), the fistula was judged completely closed. In all 7 patients, the enterostomy was reversed. After the enterostomy closure, none of them complained of symptoms suggesting recurrent fistula.
Results
Discussion
In all cases, the operation was not converted to any other procedure. Operation time ranged from 93 to 210 min with an average of 127.2 min. During the operation, no complication was encountered. Bleeding was less than 20 ml in all cases. Postoperative complication was nil either. In 3 patients, fistula recurred within 8 weeks. They complained of symptoms such as pneumaturia or faecaluria again. Recurrence was confirmed by colonoscopy and/or cystography. Of these 3 cases with recurrence, one had a
Rectovesical fistula is a rare complication after radical prostatectomy. It has been reported that the incidence was 0.5–9 % [1, 2]. There seems to be a chance to close the fistula spontaneously when intestinal diversion and/or urinary diversion are constructed. However, majority of those patients eventually need surgical repair [1, 2]. A variety of surgical procedures has been tried to challenge this problem, although none of them has been universally accepted so far. In a systematic review with
123
Surg Endosc
416 cases, perineal approach with gracilis muscle interposition was selected in the majority of cases [5]. This approach gained the highest rate of successful closure. Other studies [5–8] also support the fact that perineal approach with gracilis muscle transposition must be the most reliable treatment with 83–100 % success rate, even when the fistula was more complicated as larger than 2 cm, surrounded by fibrotic scar tissue, severely affected by irradiation or ablation. Drawback of this approach might be a possibility of infection in the perineum, or numbness in the thigh. Trans-sphincter approach such as York-Mason procedure has been reported with preferable outcomes [9]. Hadely et al. [10] have reported the largest series of YorkMason repair for 40 years with excellent outcomes. They concluded that York-Mason procedure should be considered as the first choice, as it was less invasive associated with very few incidence of morbidity. They also suggest that those who underwent irradiation and/or ablation in the fistula site prior to the repair had a tendency to fail and might not be a good candidate for their procedure. Wilbert et al. [11] reported 2 cases with a rectourethral fistula which was successfully corrected by TEM. The most important advantage in using TEM, we think, is the fact that visualization of the fistula is excellent without any incision in healthy tissue such as in York-Mason procedure or perineal approach. As described in the past literatures [3, 4, 12, 13] and the authors know from our experiences, TEM has a lot of technical advantages compared with conventional transanal approach. The magnified threedimensional image obtained via the binocular endoscope enables surgeons to identify, for example, tiny tissue structures, difference between healthy tissues and ischemic tissues, and so on. This ability plays an important role in stitching correct layer of the proper muscle in multiple layer suturing. Wide operation field in TEM enhances maneuverability of curved instruments. In the original TEM procedure, the needle driver specially designed for TEM is supposed to be used [3]. This needle driver is not designed for knot-tying, and as in ordinary TEM, the suture is secured with specially designed suture clips made of pure silver. During the repair of the fistula, we use mainly a pair of laparoscopic needle drivers instead of the original needle driver of TEM, because in the repair of the fistula, the closure is performed by multiple layers. And we believe suture in the deeper layers should be secured not with silver clips, but with knot-tying, as we do not want to remain the silver clips inside the rectal wall. Although TEM has less simplicity and convenience, once the institute is equipped with it and surgeons are familiar with the unique dexterity required, there should be a chance to apply it to this challenging treatment.
123
Moreover, the fact that TEM instruments are reusable may encourage the institutes to expand its indications. Another advantage is that TEM is associated with extremely low incidence of morbidity [12, 13]. We presume that conventional transanal surgical approach fails with considerably high possibility, although there have been very few reports in the past and it is difficult to predicate. In the present series, for example, 8 patients had histories of previous surgical repair attempts via conventional transanal approach, and all failed. Moreover in a recent systematic review, conventional transanal surgical repair was revealed to be an unfavorable procedure associated with the worst failure rate of 41 %, compared with a variety of other surgical treatments [8]. We think that in the conventional transanal approach, the technical precision is not as high as in trans-sphincter approach or TEM, as it has narrower operative field and visibility, and maneuverability should be deteriorated. In our series, fistula repair failed in 3 patients (30 %). In one patient, the fistula orifice was as large as 2 cm, surrounded by tough scar tissue due to repeated irradiation and HIFU ablation prior to the reference. During the operation after fistulectomy, ideal tissue approximation was not feasible. The other 2 cases of recurrence had undergone repeated surgical repair via conventional transanal approach before they were referred to us. At the reference, the rectal wall around the fistula was very tough with scars and looked ischemic. So far we are discouraged to indicate the current surgical treatment for patients with history of irradiation and/or HIFU [5, 7, 10]. In addition, we think the cases with previous surgical attempts are associated with more risk of failure than fresh cases [10, 14]. Recently, some favorable results by transanal approach using flexible colonoscopy with a newly developed clip [15, 16] have been reported [17, 18]. When we know to which type of fistula this kind of endoscopic treatment is appropriately indicated, this procedure should be regarded as minimally invasive choice of treatment and beneficial to the indicated patients. However, this challenge has been started only recently and it should wait for larger series of experience. Throughout our experience of 10 patients, we think that fistulas with history of irradiation and/or ablation are not ideal candidates for repair by TEM. In addition, fistulas associated with wide, tough scar tissue due to past attempt of surgical repair should not be indicated for TEM, either. Those complicated fistulas must be considered to be corrected by perineal approach with gracilis muscle interposition. However, we think TEM is a very beneficial option in the treatment of simple fistulas without tough, fibrotic surroundings. It provides a chance of complete closure of fistula with very low morbidity, without any incision in healthy tissue for approach.
Surg Endosc Disclosures Eiji Kanehira, Takashi Tanida, Aya Kamei, Masafumi Nakagi, Mitsuharu Iwasaki, and Hirofumi Shimizu have no conflicts of interest or financial ties to disclose.
References 1. Thomas C, Jones J, Ja¨ger W, Hampel C, Thu¨roff JW, Gillitzer RJ (2009) Incidence, clinical symptoms and management of rectourethral fistulas after radical prostatectomy. J Urol 183:608–612 2. Buckley JC (2011) Complications after radical prostatectomy: anastomotic stricture and rectourethral fistula. Curr Opin Urol 21:461–464 3. Buess G, Kipfmu¨ller K, Hack D, Gru¨ssner R, Heintz A, Junginger T (1988) Technique of transanal endoscopic microsurgery. Surg Endosc 2:71–75 4. Lirici MM, Kanehira E, Melzer A, Schurr MO (2014) The outburst age: how TEM ignited the MIS revolution. Minim Invasive Ther Allied Technol 23:1–4 5. Zmora O, Tulchinsky H, Gur E, Goldman G, Klausner JM, Rabau M (2006) Gracilis muscle transposition for fistulas between the rectum and urethra or vagina. Dis Colon Rectum 49:1316–1321 6. Zmora O, Potenti FM, Wexner SD, Pikarsky AJ, Efron JE, Nogueras JJ, Pricolo VE, Weiss EG (2003) Gracilis muscle transposition for iatrogenic rectourethral fistula. Ann Surg 237:483–487 7. Voelzke BB, McAninch JW, Breyer BN, Glass AS, Garcia-Aguilar J (2013) Transperineal management for postoperative and radiation rectourethral fistulas. J Urol 189:966–971 8. Hechenbleikner EM, Buckley JC, Wick EC (2013) Acquired rectourethral fistulas in adults: a systematic review of surgical repair techniques and outcomes. Dis Colon Rectum 56:374–383 9. Renschler TD, Middleton RG (2003) 30 years of experience with York-Mason repair of recto-urinary fistulas. J Urol 170:1222–1225
10. Hadley DA, Southwick A, Middleton RG (2012) York-Mason procedure for repair of recto-urinary fistulae: a 40-year experience. BJU Int 109:1095–1098 11. Wilbert DM, Buess G, Bichler KH (1996) Combined endoscopic closure of rectourethral fistula. J Urol 155:256–258 12. Kanehira E, Tanida T, Kamei A, Nakagi M, Hideshima A (2014) A single surgeon’s experience with transanal endoscopic microsurgery over 20 years with 153 early cancer cases. Minim Invasive Ther Allied Technol 23:5–9 13. Arezzo A, Arolfo S, Cravero F, Migliore M, Allaix ME, Morino M (2014) Which treatment for large rectal adenoma? Preoperative assessment and therapeutic strategy. Minim Invasive Ther Allied Technol 23:21–27 14. Nfonsam VN, Mateka JJ, Prather AD, Marcet JE (2013) Shortterm outcomes of the surgical management of acquired rectourethral fistulas: does technique matter? Res Rep Urol 30:47–51 15. Schurr MO, Arezzo A, Ho CN, Anhoeck G, Buess G, Di Lorenzo N (2008) The OTSC clip for endoscopic organ closure in NOTES: device and technique. Minim Invasive Ther Allied Technol 17:262–266 16. Weiland T, Fehlker M, Gottwald T, Schurr MO (2012) Performance of the OTSC System in the endoscopic closure of gastrointestinal fistulae–a meta-analysis. Minim Invasive Ther Allied Technol 21:249–258 17. Mangiavillano B, Pisani A, Viaggi P, Arena M, Opocher E, Mangano M, Santoro T, Masci E (2010) Endoscopic sealing of a rectovesical fistula with a combination of an over the scope clip and cyano-acrylate injection. J Gastrointest Oncol 1:122–124 18. Mori H, Kobara H, Fujihara S, Nishiyama N, Kobayashi M, Masaki T, Izuishi K, Suzuki Y (2012) Rectal perforations and fistulae secondary to a glycerin enema: closure by over-thescope-clip. World J Gastroenterol 28:3177–3180
123
