Reconstructive Urology Urinary Tract Fistulas: Transrenal Reversible Ureteral Occlusion With Detachable Semicompliant Balloons Mareike Franke, Sabine Winand, De-Hua Chang, Sebastian Wille, David Maintz, and Christopher Bangard OBJECTIVE METHODS
RESULTS
CONCLUSION
To evaluate transrenal reversible ureteral occlusion with detachable balloons in patients with treatment-refractory urinary tract fistulas. From September 2009 to September 2013, reversible occlusion of 18 ureters (7 men, 3 women; 27-74 years) with treatment-refractory urinary leakage mostly because of tumor disease or iatrogenic surgical injury was performed. Nephrostomy was exchanged fluoroscopically into an 8F or 9F sheath, which was introduced into the ureter to the point of intended balloon inflation. The prepared semicompliant balloon on a special microcatheter was introduced into the sheath, inflated, and detached. A nephrostomy tube was placed in the pelvicalyceal system. After healing of the fistulas, urinary flow was restored by transureteral removal or computed tomographye guided percutaneous puncture of the balloons. Hundred percent successful placement of the balloons and initial urinary fistula occlusion was achieved (18 of 18 ureters). One patient was lost during follow-up. Six of the remaining 9 patients needed reocclusion because of balloon dislocation or deflation (secondary technical success 83%; 5 of 6 ureters). A z-shaped ureter made reocclusion unsuccessful. Mean duration of ureteral occlusion was 74 days (5-250 days). After healing of the fistulas, intentional ureteral recanalization by percutaneous puncture or transureteral balloon removal was feasible. Anterograde urination was achieved in 5 of 9 patients (clinical success rate 55%). Four fistulas did not heal. Two patients died from their underlying disease. Transrenal reversible off-label ureteral occlusion with semicompliant detachable balloons is feasible with a high technical success rate. Long-term ureteral occlusion can be achieved in nearly all patients with a moderate clinical success rate in heavily diseased patients. UROLOGY -: -e-, 2015. 2015 Elsevier Inc.
F
istulas of the urinary tract are serious complications of surgical interventions, traumatic injuries, pelvic malignancy, or radiation therapy. Pelvic malignancy and radiation therapy that damage pelvic tissue may be responsible for surgical failure to close the fistula. Surgery fails in up to 11% in patients who are not optimal surgical candidates.1 Treatment of fistulas of the urinary tract still remains a challenge.2 For these reasons, an image-guided, percutaneous, nonsurgical treatment approach is useful. Retrograde ureteral stent placement or nephrostomy are options. When fistulas persist, ureteral occlusion protects fistula Financial Disclosure: The authors declare that they have no relevant financial interests. From the Department of Radiology, University Hospital Cologne, Cologne, Germany; and the Department of Urology, University Hospital Cologne, Cologne, Germany Address correspondence to: Christopher Bangard, M.D., Department of Radiology, University Hospital Cologne, Kerpener Str. 62, Cologne 50937, Germany. E-mail: [email protected] Submitted: January 7, 2015, accepted (with revisions): March 3, 2015
ª 2015 Elsevier Inc. All Rights Reserved
from urinary flow. In combination with percutaneous nephrostomy for urine diversion, many approaches of irreversible ureteral occlusions have been described3: Several groups used isobutyl-2-cyanoacrylate to permanently clog the ureter in patients.4,5 Results remained inconsistent because material softened in the urine with the result of being expelled by ureteral peristalsis. There are approaches of permanent ureteral occlusions with coils and gelatin sponges in 5 patients.6 Other groups7 proved this method to be successful and safe for permanent ureteral occlusion with a low complication rate. Since 2009, several groups have used the Amplatzer vascular plugs for successful permanent ureteral occlusion.8-10 There are only few reports available about nonpermanent ureteral occlusion in patients. Ureteral occlusions with detachable latex balloons filled with silicone were performed in 7 patients in 1982.11 In 1984, a larger group of 20 patients was treated, but in 6 patients, the contralateral kidney was rendered nonfunctional to obviate the need for bilateral http://dx.doi.org/10.1016/j.urology.2015.03.042 0090-4295/15
1
nephrostomy.12 Nondetachable catheter balloons were used by Horenblas et al13 for reversible ureteral occlusion in patients. They proved long-term ureteral occlusion with percutaneous transrenal catheter balloons to be safe. Although some of the balloons were in place for 5 months, they did not cause ureteral pressure necrosis. The aim of our study was to evaluate the feasibility of reversible transrenal off-label ureteral occlusion with detachable semicompliant balloons as treatment of last resort for patients with urinary fistulas.
METHODS Patients and Clinical Indications From September 2009 until September 2013, 10 patients (7 males; age range, 27-74 years; mean age, 55 years) with urine leakage were treated with reversible balloon occlusion. The institutional review board approved this retrospective study. Informed written consent including the off-label use was obtained from all patients included in the study. The reason for leakage was pelvic malignancy with or without previous surgery or radiation therapy resulting in complex fistulas (Table 1). Two patients had an iatrogenic damage of one ureter during surgical vascular prosthesis placement of iliac arteries with ureterocutaneous fistulas and one had traumatic rupture of the urethra with a vesico-perineal-cutaneous fistula (Table 1). Intravenous urography, cystography, and/or cystoscopy confirmed the diagnosis of urine leakage. Ureteral occlusion was indicated as treatment of last resort, when surgical treatment was not possible and/or fistula persisted despite urine diversion by balloon nephrostomy and ureteral stent placement or catheter drainage.
Technical Procedure of Balloon Occlusion All procedures were performed by a single interventional radiologist with 14 years of experience in all kinds of interventions including genitourinary ones. For off-label ureteral occlusion, Gold Balloons and Magic-MABDTE catheters (both from Balt Extrusion, Montmorency, France) were used. The Gold Balloon is a semicompliant latex balloon originally designed for the closure of arteriovenous fistulas and malformations and for vessel occlusion. Different sizes, from Gold Balloon size 1 (nominal 7 mm in diameter and 10 mm in length) to Gold Balloon size 6 (nominal 15 mm in diameter and 25 mm in length), are available. Its semicompliant material allows the surrounding tissue to determine the form of the inflated balloon. Restriction of the maximum balloon diameter can lead to an increase of the maximum length. The balloon contains a radiopaque marker (diameter 0.5 mm). The Gold Balloon has to be mounted on a special microcatheter to be inserted to the point of detachment. The fitting of the Gold Balloon onto the MABDTE microcatheter is described in detail in Figure 1A-D. The Magic-MABDTE catheter (MABDTE stands for magic balloon detachable Teflon microtube tip) is a flow-guided, tapered, 165-cm-long microcatheter with a proximal part of 2.7F and a distal part of 1.8F that ends with a Teflon microtube tip. It contains a 0.010-inch mandrel that is needed for fitting the balloon onto the catheter. All procedures were carried out fluoroscopically under local anesthesia without conscious sedation. Balloon nephrostomy had been introduced before ureteral occlusion. Nephrostomies were changed over a straight or j-shaped 0.035-inch heavy-duty wire (Safe-T-J curved, Cook Medical) and a 45-cm-long 9F or 2
8F sheath (Cook Medical) that was introduced into the ureter to the point of intended balloon release. The detachment point was chosen proximal to ureteral crossing of iliac vessels to guarantee easy access in case of necessary percutaneous puncture. If the guide wire did not penetrate the distal hole of the nephrostomy, for example, because of blood clots, a 4F straight catheter (Nylex; Cordis Corporation) and a 0.035-inch stiff hydrophilic wire (Radifocus Guide Wire M; Terumo Europe, Leuven, Belgium) were used to introduce the sheath. The prepared balloon on a Magic-MABDTE catheter was introduced into the sheath via an introducer made of steel. The sheath was jetted with physiological saline solution. At point of release, the balloon was inflated with a mixture of physiological saline solution and contrast medium (ratio approximately 4:1; maximum volume, 0.75 mL for size 4 and 2.5 mL for size 5 Gold Balloon) until significant rise of resistance during the inflation with a 1- or 3-mL Luer lock syringe. The balloon was released via slow pull back of the catheter against the sheath. The maximum nominal diameter and length of the size 4 and 5 Gold Balloons were 9 mm 16 mm and 12 mm 28 mm, respectively. The balloon size was chosen according to the size of the ureter. The semicompliant material led to different configurations of the same balloon size dependent on the ureteral environment. If a balloon dislocated during or immediately after detachment, a new balloon was detached until complete ureteral obstruction was verified by slow contrast injection via the sheath (Fig. 2). Afterward, a balloon nephrostomy was placed in the pelvicalyceal system before intervention. Primary and secondary technical success was defined as complete ureteral obstruction after the first or the second intervention. The latter was necessary in case of balloon deflation or departure before fistulas had healed. After healing of the fistulas, urinary flow was restored by transureteral removal or computed tomography (CT)-guided percutaneous puncture of the balloons with 17.522 gauge needles.
Follow-up Nephrostomy was exchanged every 6 weeks, and anterograde nephrostograms were routinely performed to assess occlusiveness of the balloons. Irrespective of inspection interval, any alteration of urine quantity via nephrostomy or newly occurring urination via the fistula immediately led to clinical visit and control of balloon occlusion and fistula. By this, the duration of ureteral occlusion was assessed. Clinical success was defined as restoration of normal or anterograde urination via bladder or pouch after healing of the fistula assessed by anterograde nephroureterogram or retrograde cystogram or conduitogram. Ureteral pressure necrosis with potential contrast leakage or ureteral stricture at the site of balloon inflation was ruled out in the same manner. After 3 months, ureteral stricture was ruled out sonographically.
Statistical Analysis Distributions of continuous variables were summarized by means standard deviations.
RESULTS Technical Feasibility In 8 patients, a bilateral ureteral occlusion was performed; in 2 patients with 1-sided ureteral leakage, 1-sided ureteral occlusion was sufficient (Table 1). The average UROLOGY
-
(-), 2015
UROLOGY -
(-), 2015 Table 1. Patient characteristics and technical and clinical outcomes Number of Balloons: Total (1st/2nd) Occlusion Patient
Age (y)
Sex
1 2
27 45
M M
3 4
53 56
M M
5
74
F
6 7 8
53 45 59
M M F
9 10
68 66
M F
R
L
R
L
Enterovesical Vesico-perineal-cutaneous
2 (1/1) 1
3 (1/2) 1
79 (5/154) 95
59 (5/113) 60
Rectovesical Ureterocutaneous near prosthetic vascular graft Ureterocutaneous
1 2 (1/1)
1 —
18 51 (39/63)
21 —
1
—
131
—
Indiana pouch Uretero-pouch Recto-vesico-vaginal
1 3 (1/2) 2 (1/1)
1 1 1
204 24 (10/38) 5 (5/t.f.)
250 48 116
Ileal conduit Vesicovaginal
2 (1/1) 1
1 1
32 (26/38) n.a.
104 n.a.
Diagnosis
Site of Leakage
CML Traumatic rupture of bladder and urethra Colorectal cancer Iatrogenic (vascular prosthesis) Iatrogenic (vascular prosthesis) End-stage incontinence Urothelial carcinoma Carcinoma of cervix uteri
Prostate cancer Carcinoma of corpus uteri
Duration of Occlusion: Mean (1st/2nd) Occlusion (d)
CML, chronic myelogenous leukemia; F, female; L, left; M, male; n.a., not applicable; R, right; t.f., technical failure.
Clinical Outcome Cancer-related death Alive, no leakage Postoperative death According to wish of patient: surgical fistula reconstruction Alive, no leakage Alive, no leakage Alive, no leakage Alive, large fistula with poor healing tendency; improvement of general condition due to reduced frequency of urinary tract infections Alive, no leakage Lost during follow-up
3
Figure 1. (A) Fitting the Gold Balloon onto the microcatheter (units of the ruler in mm): The Gold Balloon has to be mounted on the (B) distal transparent tube of the microcatheter. The microcatheter carries a 0.010-inch mandrel that overshoots the distal tube for several centimeters (arrow). The Gold Balloon has to be fitted onto the blunt wet needle included in the package and has to be inflated up to half of the maximum volume with sterile saline solution to control its integrity. (C) The partially inflated balloon has to be detached from the blunt needle by retraction. To fit it onto the distal transparent tube of the microcatheter, it is essential to pull the mandrel back into the catheter until the mandrel protrudes above the tube only by 2-3 mm (arrow). (D) Both the Gold Balloon and the distal transparent tube with the slightly protruding mandrel have to be fixed tight with the fingers to avoid kinking. The collar of the mounted Gold Balloon must not be >1 mm away from the end of the tube. By pulling the mandrel back by 1-2 cm, the Gold Balloon deflates and is ready for introduction. The microcatheter has to be straight during the whole procedure.
fluoroscopy time for the both-sided ureteral occlusion was 31.6 8.2 minutes and for the 1-sided occlusion, 22.2 17 minutes. Primary technical success rate was 100%. Occlusion of ureter was complete in all the 18 ureters in the 10 patients. Complete external diversion of the urinary stream in all of the 8 patients who received bilateral ureteral occlusion was accomplished. In the latter, all urine evacuated via the nephrostomy and the bladder or pouch urine output suspended completely. During the intervention, 5 balloons dislocated during or immediately after detachment into the bladder or the renal pelvis without consequence. Three of these balloons dislocated during the first intervention in the first patient. One patient had a z-shaped proximal ureter that prevented introduction of the long sheath to the intended distal detachment point of the balloon. During the attempt to pass the z-shaped ureter part, a guide-wire perforation of the ureter occurred. The balloon was detached proximal to the z-shaped part and proximal to the perforation with complete ureteral occlusion. No other method-related complication occurred. One patient was lost during follow-up. Five of the remaining 9 patients (55%) or 6 of the remaining 16 ureters (38%) needed a reocclusion because of balloon departure or balloon deflation after 15 14 days. Five of these 6 ureters were occluded successfully (secondary 4
technical success rate, 83%). Complete reocclusion of the z-shaped ureter was not possible because of balloon dislocation into the renal pelvis and attributed for the only technical failure. In total, 26 balloons were implanted in the ureters of 10 patients. Average length of time of ureteral occlusion was 74 days with a wide range (5-250 days; Table 1). No ureteric pressure necrosis occurred and no ureteral stricture was observed in the 3month follow-up sonography. Two major complications were nephrostomy associated: One ureteral pressure necrosis occurred because of a pre-existing balloon nephrostomy in malposition and 1 parenchymal pseudoaneurysm led to severe hematuria after nephrostomy that prompted successful transarterial coil embolization. Six occlusion balloons in 4 patients were punctured under CT guidance with 17.5-22 gauge needles (6 balloons) and 3 balloons were removed by ureterorenoscopy. In 1 patient, percutaneous transgression of small bowel with a 22-gauge needle was necessary to reach the balloon without any complications. The balloons punctured under CT guidance were all voided as documented by missing of the radiopaque balloon marker on plain film roentgenograms or abdominal CTs, which were performed in the following months and years independent of the performed procedure. Thirteen of the remaining 15 balloons went off UROLOGY
-
(-), 2015
occlusion but died 3 days after an operation of his preexisting nephrostomy-induced ureteral necrosis due to reduced general condition. Patient number 4 wished a surgical reconstruction of the ureter, which was successful but caused a ureteral stricture (at first, urologists opted against surgical reconstruction because of several preceding operations and wound infection in the right groin and abdominal wound with lymphatic fistula next to a prosthetic vascular graft). In patient number 8 with a large tissue defect and a big recto-vesico-vaginal fistula, ureteral occlusion of the left ureter (technical failure of occlusion of the right one) led to a reduced frequency of urinary tract infections and a better general condition, but the fistula persisted. The larger a fistula the less likely it seemed to heal.
COMMENT
Figure 2. Anteroposterior roentgenogram after occlusion of the right distal ureter with a Gold Balloon size number 4. The radiopaque marker can be seen within the balloon that completely occludes the ureter so that the contrast material injected via the sheath cannot pass the balloon.
without need of intervention and 2 balloons remained within the ureters until the patients died. Clinical Outcome Normal or anterograde urination via bladder or pouch was achieved in 5 of 9 patients with a clinical success rate of healed fistulas of 55%. Four fistulas did not heal despite ureteral occlusion: 2 of these patients died during the observation period from their underlying disease, 1 of chronic myelogenous leukemia and immunodeficiency, and the other patient had a successful ureteral balloon UROLOGY
-
(-), 2015
Many ureteral occlusion methods are irreversible. To our knowledge, there is no dedicated approved balloon catheter for reversible obstruction of the ureter to dry urinary fistulas. In case of treatment-refractory urinary tract fistulas despite retrograde ureteral stent placement or catheter drainage and balloon nephrostomy, we suggest a new reversible off-label method with detachable neuroradiologic balloons designed for arteriovenous fistula occlusion14-16 to induce ureteral occlusion. The technical success rate was high. The only technical failure and complication occurred because of a z-shaped ureter. In this case, the long sheath could not be introduced safely and the method should not have been applied. Duration of ureter occlusion was highly variable between 5 and 250 days. This can be attributed to ureteral peristalsis and to spontaneous volume loss due to insufficiency of the balloon valve over time. Follow-up is essential. As done for 55% of patients or 38% of the ureters, reocclusion can be performed easily without the need for a new kidney puncture. Compared with the reversible methods described by G€unther11,12 and Horenblas,13 there are some advantages of the presented method. G€unther used detachable latex balloons that had to be filled via a 3F catheter with a mixture of silicone elastomer and silicone fluid that took >15 minutes for vulcanization. Attachment of the balloon to the catheter needed several knots and was inconvenient. The procedure was performed via 14F and 11F sheaths instead of 8F or 9F with the presented method. In 2 cases, the balloons were dislodged into the renal pelvis and extracted percutaneously via a Dormia stone basket. Despite follow-up between 1 and 20 months, clinical outcome of the patients and whereabouts of most of the balloons were not reported for the majority of patients. Late failures of ureteral occlusion in 3 of 21 ureters were lower than with our technique. In the largest reported series17 (52 ureters occluded with detachable balloons), the need for reocclusion in 31% was comparable with our method. In contrast to the off-label technique by Horenblas, only the nephrostomy, 5
but not a second catheter, protrudes from the patients’ body with the presented method. Prevention of a second percutaneous transrenal puncture to insert the diagnostic or Foley balloon in addition to the nephrostomy reduces the risk of complications.13,18 A nondetachable balloon catheter is more prone to dislodge because its extracorporeal part can accidently be pulled on. This is underlined by more frequent reocclusions due to balloon loss in the series described by Horenblas (15 reocclusions of 13 ureters described by Horenblas vs 6 reocclusions of 16 ureters with the presented method). Dislodgement of the balloon catheter has the potential to harm the ureter or the kidney. The semicompliant material of the Gold Balloon minimizes the risk of a ureteral pressure necrosis and makes complete occlusion of the ureter as atraumatic as possible. Even after 250 days of balloon occlusion, no sign of ureteral damage or stricture was observed in contrast to the surgical reconstruction in one of our patients. Horenblas reported a maximum of 169 days of ureteral occlusion. We could demonstrate that even longer occlusion periods are possible, which may improve the prognosis for patients with large tissue damage and longer periods of healing. Limitation of the presented study is the low number of patients treated. The method can only be applied if the ureter is
RESULTS
CONCLUSION
To evaluate transrenal reversible ureteral occlusion with detachable balloons in patients with treatment-refractory urinary tract fistulas. From September 2009 to September 2013, reversible occlusion of 18 ureters (7 men, 3 women; 27-74 years) with treatment-refractory urinary leakage mostly because of tumor disease or iatrogenic surgical injury was performed. Nephrostomy was exchanged fluoroscopically into an 8F or 9F sheath, which was introduced into the ureter to the point of intended balloon inflation. The prepared semicompliant balloon on a special microcatheter was introduced into the sheath, inflated, and detached. A nephrostomy tube was placed in the pelvicalyceal system. After healing of the fistulas, urinary flow was restored by transureteral removal or computed tomographye guided percutaneous puncture of the balloons. Hundred percent successful placement of the balloons and initial urinary fistula occlusion was achieved (18 of 18 ureters). One patient was lost during follow-up. Six of the remaining 9 patients needed reocclusion because of balloon dislocation or deflation (secondary technical success 83%; 5 of 6 ureters). A z-shaped ureter made reocclusion unsuccessful. Mean duration of ureteral occlusion was 74 days (5-250 days). After healing of the fistulas, intentional ureteral recanalization by percutaneous puncture or transureteral balloon removal was feasible. Anterograde urination was achieved in 5 of 9 patients (clinical success rate 55%). Four fistulas did not heal. Two patients died from their underlying disease. Transrenal reversible off-label ureteral occlusion with semicompliant detachable balloons is feasible with a high technical success rate. Long-term ureteral occlusion can be achieved in nearly all patients with a moderate clinical success rate in heavily diseased patients. UROLOGY -: -e-, 2015. 2015 Elsevier Inc.
F
istulas of the urinary tract are serious complications of surgical interventions, traumatic injuries, pelvic malignancy, or radiation therapy. Pelvic malignancy and radiation therapy that damage pelvic tissue may be responsible for surgical failure to close the fistula. Surgery fails in up to 11% in patients who are not optimal surgical candidates.1 Treatment of fistulas of the urinary tract still remains a challenge.2 For these reasons, an image-guided, percutaneous, nonsurgical treatment approach is useful. Retrograde ureteral stent placement or nephrostomy are options. When fistulas persist, ureteral occlusion protects fistula Financial Disclosure: The authors declare that they have no relevant financial interests. From the Department of Radiology, University Hospital Cologne, Cologne, Germany; and the Department of Urology, University Hospital Cologne, Cologne, Germany Address correspondence to: Christopher Bangard, M.D., Department of Radiology, University Hospital Cologne, Kerpener Str. 62, Cologne 50937, Germany. E-mail: [email protected] Submitted: January 7, 2015, accepted (with revisions): March 3, 2015
ª 2015 Elsevier Inc. All Rights Reserved
from urinary flow. In combination with percutaneous nephrostomy for urine diversion, many approaches of irreversible ureteral occlusions have been described3: Several groups used isobutyl-2-cyanoacrylate to permanently clog the ureter in patients.4,5 Results remained inconsistent because material softened in the urine with the result of being expelled by ureteral peristalsis. There are approaches of permanent ureteral occlusions with coils and gelatin sponges in 5 patients.6 Other groups7 proved this method to be successful and safe for permanent ureteral occlusion with a low complication rate. Since 2009, several groups have used the Amplatzer vascular plugs for successful permanent ureteral occlusion.8-10 There are only few reports available about nonpermanent ureteral occlusion in patients. Ureteral occlusions with detachable latex balloons filled with silicone were performed in 7 patients in 1982.11 In 1984, a larger group of 20 patients was treated, but in 6 patients, the contralateral kidney was rendered nonfunctional to obviate the need for bilateral http://dx.doi.org/10.1016/j.urology.2015.03.042 0090-4295/15
1
nephrostomy.12 Nondetachable catheter balloons were used by Horenblas et al13 for reversible ureteral occlusion in patients. They proved long-term ureteral occlusion with percutaneous transrenal catheter balloons to be safe. Although some of the balloons were in place for 5 months, they did not cause ureteral pressure necrosis. The aim of our study was to evaluate the feasibility of reversible transrenal off-label ureteral occlusion with detachable semicompliant balloons as treatment of last resort for patients with urinary fistulas.
METHODS Patients and Clinical Indications From September 2009 until September 2013, 10 patients (7 males; age range, 27-74 years; mean age, 55 years) with urine leakage were treated with reversible balloon occlusion. The institutional review board approved this retrospective study. Informed written consent including the off-label use was obtained from all patients included in the study. The reason for leakage was pelvic malignancy with or without previous surgery or radiation therapy resulting in complex fistulas (Table 1). Two patients had an iatrogenic damage of one ureter during surgical vascular prosthesis placement of iliac arteries with ureterocutaneous fistulas and one had traumatic rupture of the urethra with a vesico-perineal-cutaneous fistula (Table 1). Intravenous urography, cystography, and/or cystoscopy confirmed the diagnosis of urine leakage. Ureteral occlusion was indicated as treatment of last resort, when surgical treatment was not possible and/or fistula persisted despite urine diversion by balloon nephrostomy and ureteral stent placement or catheter drainage.
Technical Procedure of Balloon Occlusion All procedures were performed by a single interventional radiologist with 14 years of experience in all kinds of interventions including genitourinary ones. For off-label ureteral occlusion, Gold Balloons and Magic-MABDTE catheters (both from Balt Extrusion, Montmorency, France) were used. The Gold Balloon is a semicompliant latex balloon originally designed for the closure of arteriovenous fistulas and malformations and for vessel occlusion. Different sizes, from Gold Balloon size 1 (nominal 7 mm in diameter and 10 mm in length) to Gold Balloon size 6 (nominal 15 mm in diameter and 25 mm in length), are available. Its semicompliant material allows the surrounding tissue to determine the form of the inflated balloon. Restriction of the maximum balloon diameter can lead to an increase of the maximum length. The balloon contains a radiopaque marker (diameter 0.5 mm). The Gold Balloon has to be mounted on a special microcatheter to be inserted to the point of detachment. The fitting of the Gold Balloon onto the MABDTE microcatheter is described in detail in Figure 1A-D. The Magic-MABDTE catheter (MABDTE stands for magic balloon detachable Teflon microtube tip) is a flow-guided, tapered, 165-cm-long microcatheter with a proximal part of 2.7F and a distal part of 1.8F that ends with a Teflon microtube tip. It contains a 0.010-inch mandrel that is needed for fitting the balloon onto the catheter. All procedures were carried out fluoroscopically under local anesthesia without conscious sedation. Balloon nephrostomy had been introduced before ureteral occlusion. Nephrostomies were changed over a straight or j-shaped 0.035-inch heavy-duty wire (Safe-T-J curved, Cook Medical) and a 45-cm-long 9F or 2
8F sheath (Cook Medical) that was introduced into the ureter to the point of intended balloon release. The detachment point was chosen proximal to ureteral crossing of iliac vessels to guarantee easy access in case of necessary percutaneous puncture. If the guide wire did not penetrate the distal hole of the nephrostomy, for example, because of blood clots, a 4F straight catheter (Nylex; Cordis Corporation) and a 0.035-inch stiff hydrophilic wire (Radifocus Guide Wire M; Terumo Europe, Leuven, Belgium) were used to introduce the sheath. The prepared balloon on a Magic-MABDTE catheter was introduced into the sheath via an introducer made of steel. The sheath was jetted with physiological saline solution. At point of release, the balloon was inflated with a mixture of physiological saline solution and contrast medium (ratio approximately 4:1; maximum volume, 0.75 mL for size 4 and 2.5 mL for size 5 Gold Balloon) until significant rise of resistance during the inflation with a 1- or 3-mL Luer lock syringe. The balloon was released via slow pull back of the catheter against the sheath. The maximum nominal diameter and length of the size 4 and 5 Gold Balloons were 9 mm 16 mm and 12 mm 28 mm, respectively. The balloon size was chosen according to the size of the ureter. The semicompliant material led to different configurations of the same balloon size dependent on the ureteral environment. If a balloon dislocated during or immediately after detachment, a new balloon was detached until complete ureteral obstruction was verified by slow contrast injection via the sheath (Fig. 2). Afterward, a balloon nephrostomy was placed in the pelvicalyceal system before intervention. Primary and secondary technical success was defined as complete ureteral obstruction after the first or the second intervention. The latter was necessary in case of balloon deflation or departure before fistulas had healed. After healing of the fistulas, urinary flow was restored by transureteral removal or computed tomography (CT)-guided percutaneous puncture of the balloons with 17.522 gauge needles.
Follow-up Nephrostomy was exchanged every 6 weeks, and anterograde nephrostograms were routinely performed to assess occlusiveness of the balloons. Irrespective of inspection interval, any alteration of urine quantity via nephrostomy or newly occurring urination via the fistula immediately led to clinical visit and control of balloon occlusion and fistula. By this, the duration of ureteral occlusion was assessed. Clinical success was defined as restoration of normal or anterograde urination via bladder or pouch after healing of the fistula assessed by anterograde nephroureterogram or retrograde cystogram or conduitogram. Ureteral pressure necrosis with potential contrast leakage or ureteral stricture at the site of balloon inflation was ruled out in the same manner. After 3 months, ureteral stricture was ruled out sonographically.
Statistical Analysis Distributions of continuous variables were summarized by means standard deviations.
RESULTS Technical Feasibility In 8 patients, a bilateral ureteral occlusion was performed; in 2 patients with 1-sided ureteral leakage, 1-sided ureteral occlusion was sufficient (Table 1). The average UROLOGY
-
(-), 2015
UROLOGY -
(-), 2015 Table 1. Patient characteristics and technical and clinical outcomes Number of Balloons: Total (1st/2nd) Occlusion Patient
Age (y)
Sex
1 2
27 45
M M
3 4
53 56
M M
5
74
F
6 7 8
53 45 59
M M F
9 10
68 66
M F
R
L
R
L
Enterovesical Vesico-perineal-cutaneous
2 (1/1) 1
3 (1/2) 1
79 (5/154) 95
59 (5/113) 60
Rectovesical Ureterocutaneous near prosthetic vascular graft Ureterocutaneous
1 2 (1/1)
1 —
18 51 (39/63)
21 —
1
—
131
—
Indiana pouch Uretero-pouch Recto-vesico-vaginal
1 3 (1/2) 2 (1/1)
1 1 1
204 24 (10/38) 5 (5/t.f.)
250 48 116
Ileal conduit Vesicovaginal
2 (1/1) 1
1 1
32 (26/38) n.a.
104 n.a.
Diagnosis
Site of Leakage
CML Traumatic rupture of bladder and urethra Colorectal cancer Iatrogenic (vascular prosthesis) Iatrogenic (vascular prosthesis) End-stage incontinence Urothelial carcinoma Carcinoma of cervix uteri
Prostate cancer Carcinoma of corpus uteri
Duration of Occlusion: Mean (1st/2nd) Occlusion (d)
CML, chronic myelogenous leukemia; F, female; L, left; M, male; n.a., not applicable; R, right; t.f., technical failure.
Clinical Outcome Cancer-related death Alive, no leakage Postoperative death According to wish of patient: surgical fistula reconstruction Alive, no leakage Alive, no leakage Alive, no leakage Alive, large fistula with poor healing tendency; improvement of general condition due to reduced frequency of urinary tract infections Alive, no leakage Lost during follow-up
3
Figure 1. (A) Fitting the Gold Balloon onto the microcatheter (units of the ruler in mm): The Gold Balloon has to be mounted on the (B) distal transparent tube of the microcatheter. The microcatheter carries a 0.010-inch mandrel that overshoots the distal tube for several centimeters (arrow). The Gold Balloon has to be fitted onto the blunt wet needle included in the package and has to be inflated up to half of the maximum volume with sterile saline solution to control its integrity. (C) The partially inflated balloon has to be detached from the blunt needle by retraction. To fit it onto the distal transparent tube of the microcatheter, it is essential to pull the mandrel back into the catheter until the mandrel protrudes above the tube only by 2-3 mm (arrow). (D) Both the Gold Balloon and the distal transparent tube with the slightly protruding mandrel have to be fixed tight with the fingers to avoid kinking. The collar of the mounted Gold Balloon must not be >1 mm away from the end of the tube. By pulling the mandrel back by 1-2 cm, the Gold Balloon deflates and is ready for introduction. The microcatheter has to be straight during the whole procedure.
fluoroscopy time for the both-sided ureteral occlusion was 31.6 8.2 minutes and for the 1-sided occlusion, 22.2 17 minutes. Primary technical success rate was 100%. Occlusion of ureter was complete in all the 18 ureters in the 10 patients. Complete external diversion of the urinary stream in all of the 8 patients who received bilateral ureteral occlusion was accomplished. In the latter, all urine evacuated via the nephrostomy and the bladder or pouch urine output suspended completely. During the intervention, 5 balloons dislocated during or immediately after detachment into the bladder or the renal pelvis without consequence. Three of these balloons dislocated during the first intervention in the first patient. One patient had a z-shaped proximal ureter that prevented introduction of the long sheath to the intended distal detachment point of the balloon. During the attempt to pass the z-shaped ureter part, a guide-wire perforation of the ureter occurred. The balloon was detached proximal to the z-shaped part and proximal to the perforation with complete ureteral occlusion. No other method-related complication occurred. One patient was lost during follow-up. Five of the remaining 9 patients (55%) or 6 of the remaining 16 ureters (38%) needed a reocclusion because of balloon departure or balloon deflation after 15 14 days. Five of these 6 ureters were occluded successfully (secondary 4
technical success rate, 83%). Complete reocclusion of the z-shaped ureter was not possible because of balloon dislocation into the renal pelvis and attributed for the only technical failure. In total, 26 balloons were implanted in the ureters of 10 patients. Average length of time of ureteral occlusion was 74 days with a wide range (5-250 days; Table 1). No ureteric pressure necrosis occurred and no ureteral stricture was observed in the 3month follow-up sonography. Two major complications were nephrostomy associated: One ureteral pressure necrosis occurred because of a pre-existing balloon nephrostomy in malposition and 1 parenchymal pseudoaneurysm led to severe hematuria after nephrostomy that prompted successful transarterial coil embolization. Six occlusion balloons in 4 patients were punctured under CT guidance with 17.5-22 gauge needles (6 balloons) and 3 balloons were removed by ureterorenoscopy. In 1 patient, percutaneous transgression of small bowel with a 22-gauge needle was necessary to reach the balloon without any complications. The balloons punctured under CT guidance were all voided as documented by missing of the radiopaque balloon marker on plain film roentgenograms or abdominal CTs, which were performed in the following months and years independent of the performed procedure. Thirteen of the remaining 15 balloons went off UROLOGY
-
(-), 2015
occlusion but died 3 days after an operation of his preexisting nephrostomy-induced ureteral necrosis due to reduced general condition. Patient number 4 wished a surgical reconstruction of the ureter, which was successful but caused a ureteral stricture (at first, urologists opted against surgical reconstruction because of several preceding operations and wound infection in the right groin and abdominal wound with lymphatic fistula next to a prosthetic vascular graft). In patient number 8 with a large tissue defect and a big recto-vesico-vaginal fistula, ureteral occlusion of the left ureter (technical failure of occlusion of the right one) led to a reduced frequency of urinary tract infections and a better general condition, but the fistula persisted. The larger a fistula the less likely it seemed to heal.
COMMENT
Figure 2. Anteroposterior roentgenogram after occlusion of the right distal ureter with a Gold Balloon size number 4. The radiopaque marker can be seen within the balloon that completely occludes the ureter so that the contrast material injected via the sheath cannot pass the balloon.
without need of intervention and 2 balloons remained within the ureters until the patients died. Clinical Outcome Normal or anterograde urination via bladder or pouch was achieved in 5 of 9 patients with a clinical success rate of healed fistulas of 55%. Four fistulas did not heal despite ureteral occlusion: 2 of these patients died during the observation period from their underlying disease, 1 of chronic myelogenous leukemia and immunodeficiency, and the other patient had a successful ureteral balloon UROLOGY
-
(-), 2015
Many ureteral occlusion methods are irreversible. To our knowledge, there is no dedicated approved balloon catheter for reversible obstruction of the ureter to dry urinary fistulas. In case of treatment-refractory urinary tract fistulas despite retrograde ureteral stent placement or catheter drainage and balloon nephrostomy, we suggest a new reversible off-label method with detachable neuroradiologic balloons designed for arteriovenous fistula occlusion14-16 to induce ureteral occlusion. The technical success rate was high. The only technical failure and complication occurred because of a z-shaped ureter. In this case, the long sheath could not be introduced safely and the method should not have been applied. Duration of ureter occlusion was highly variable between 5 and 250 days. This can be attributed to ureteral peristalsis and to spontaneous volume loss due to insufficiency of the balloon valve over time. Follow-up is essential. As done for 55% of patients or 38% of the ureters, reocclusion can be performed easily without the need for a new kidney puncture. Compared with the reversible methods described by G€unther11,12 and Horenblas,13 there are some advantages of the presented method. G€unther used detachable latex balloons that had to be filled via a 3F catheter with a mixture of silicone elastomer and silicone fluid that took >15 minutes for vulcanization. Attachment of the balloon to the catheter needed several knots and was inconvenient. The procedure was performed via 14F and 11F sheaths instead of 8F or 9F with the presented method. In 2 cases, the balloons were dislodged into the renal pelvis and extracted percutaneously via a Dormia stone basket. Despite follow-up between 1 and 20 months, clinical outcome of the patients and whereabouts of most of the balloons were not reported for the majority of patients. Late failures of ureteral occlusion in 3 of 21 ureters were lower than with our technique. In the largest reported series17 (52 ureters occluded with detachable balloons), the need for reocclusion in 31% was comparable with our method. In contrast to the off-label technique by Horenblas, only the nephrostomy, 5
but not a second catheter, protrudes from the patients’ body with the presented method. Prevention of a second percutaneous transrenal puncture to insert the diagnostic or Foley balloon in addition to the nephrostomy reduces the risk of complications.13,18 A nondetachable balloon catheter is more prone to dislodge because its extracorporeal part can accidently be pulled on. This is underlined by more frequent reocclusions due to balloon loss in the series described by Horenblas (15 reocclusions of 13 ureters described by Horenblas vs 6 reocclusions of 16 ureters with the presented method). Dislodgement of the balloon catheter has the potential to harm the ureter or the kidney. The semicompliant material of the Gold Balloon minimizes the risk of a ureteral pressure necrosis and makes complete occlusion of the ureter as atraumatic as possible. Even after 250 days of balloon occlusion, no sign of ureteral damage or stricture was observed in contrast to the surgical reconstruction in one of our patients. Horenblas reported a maximum of 169 days of ureteral occlusion. We could demonstrate that even longer occlusion periods are possible, which may improve the prognosis for patients with large tissue damage and longer periods of healing. Limitation of the presented study is the low number of patients treated. The method can only be applied if the ureter is
