Two-sided Dorsal Plus Ventral Oral Graft Bulbar Urethroplasty: Long-term Results and Predictive Factors
Enzo Palminteri, Elisa Berdondini, et Al.
85: 942 e 947, 2015
Center for Urethral and Genitalia Reconstructive Surgery, Arezzo, Italy.
To evaluate long-term outcomes of the 2-sided dorsal plus ventral oral graft (DVOG) urethroplasty by preserving the narrow urethral plate in tight bulbar strictures and investigate which factors might influence long-term outcomes.
This is a single-center retrospective study of 166 patients who underwent DVOG urethroplasty for tight bulbar strictures by a single surgeon (E.P.) between 2002 and 2013. The strictured urethra was opened ventrally; the exposed urethral plate was incised in the midline and augmented dorsally and ventrally using 2 oral grafts. Outcome was considered a failure when any postoperative instrumentation was needed. According to stricture length, patients were classified in 3 groups as follows: 1.5 cm (group 1), >1.5 and 3.9 cm (group 2), and 4 cm (group 3). Time to failure was analyzed using Kaplan-Meier estimates and Cox regression.
Median follow-up was 47 months (interquartile range, 33-95.5 months). Of the 166 patients, 149 (89.8%) were successful and 17 (10.2%) were failures. Most of the failures (90%) were observed during the first 5 years of follow-up; afterward, the success rate remained stable. The stricture length was a significant predictor of surgical outcome (odds ratio, 1.743 per cm; confidence interval, 1.2-2.5; P<.001); patients with a urethral stricture 4 cmpresented a higher risk of late failure.Age, stricture etiology, and previous treatment were not significant predictors of surgical outcome.
With long-term follow-up, the treatment of tight bulbar strictures using a 2-sided DVOG urethroplasty showed a high success rate. The stricture length is an independent predictor of failure.
Bulbar urethral strictures are treated by various reconstructive techniques.1-4 Short or obliterative strictures may be treated with excision and anastomotic urethroplasty, whereas for longer strictures, patch urethroplasty preferably using an oral graft has been advocated.3 The graft may be placed dorsally or ventrally by using dorsal or ventral urethrotomy approaches.5-7 However, because transecting procedures might impair sexual function probably as a consequence of the vascular damage and urethral shortening,2,8,9 new nontransecting and urethra-sparing techniques have been promoted for short stenoses.10 On the other hand, procedures augmenting the preserved urethral plate by using 1-side grafts might be unable to provide an adequate lumen in very tight strictures.11,12 Therefore, the best approach to treat these strictures is still under debate.13 In this context, a new urethra-sparing technique was described in 2008 to treat tight bulbar strictures with a focally subobliterative segment, which are too severe for a simple 1-side grafting.14 Our technique combines a dorsal plus ventral oral graft (DVOG), without transecting and shortening the urethra; the urethral plate is preserved to avoid the post-transection sexual complications15 and augmented on both sides to obtain an adequate urethral augmentation. The aim of the present study was to extend our previously published short-term follow-up,14,15 by reporting a long-term follow-up in a wide series of patients. To our knowledge, this is the study on 2-sided DVOG bulbar urethroplasty with the longest follow-up. Furthermore, we have investigated which factors might influence long-term outcome and deterioration of the success rate.
We performed a retrospective study including 166 patients who underwent 2-sided DVOG urethroplasty between 2002 and 2013 for tight bulbar urethral strictures at our high-volume institution. Tight strictures were characterized by a very narrow residual urethral plate (width
Our technique of DVOG urethroplasty has been previously described in detail.14 A Y-inverted perineal incision is made, and the bulbocavernous muscles are divided.
Dorsal Graft. Using a ventral-sagittal urethrotomy, the strictured bulbar urethra is opened. The urethra is left open for 1 cm both proximally and distally in the healthy urethra. As suggested by Asopa et al,7 the exposed dorsal urethra is incised in the midline down to the tunica albuginea; the margins of the incised dorsal urethra are dissected from the tunica albuginea without lifting the 2 halves of the bisected urethra: an elliptical raw area is created where the first graft is placed as a dorsal inlay, quilted to the corpora cavernosa, and sutured to the urethral margins (Fig. 1; points 1-4).
Ventral Graft. Following dorsal augmentation, the urethra is also graft-enlarged ventrally anastomosing the second graft to the mucosal margins of the urethral plate (Fig. 1; points 5-8). Finally, the spongiosum is closed over the ventral graft. A 16- to 18-Fr silicone Foley catheter was left in situ. The double patch was used to better enlarge the urethra in very tight strictures, in which a single patch appeared intraoperatively inadequate to make a sufficiently wide lumen. In this regard, the preoperative assessment by urethrography and urethroscopy (with failure to pass a 7-Fr rigid ureteroscope through the narrow part of the urethra) may help in assessing the degree of urethral narrowing and to define a tight stricture. During intraoperative endoscopy, a guidewire was inserted through the urethral lumen to facilitate correct stricturotomy and to avoid additional damage to the urethral plate. However, despite any preoperative evaluation, only the intraoperative urethral features will determine the choice of the surgical technique. We performed this technique in all the very tight strictures, including the short ones ( Postoperative Course and Follow-up Criteria
Patients were discharged from the hospital 3 days after surgery. All patients received 2 g per day of ceftriaxone from the day of operation to the day of discharge, and subsequently ciprofloxacin 500 mg orally twice a day until the catheter is removed. In all patients, the catheter was left in place for 3 weeks. At 3 weeks after surgery, the bladder was filled with contrast medium, the Foley catheter was removed, and voiding cystourethrography was obtained. Follow-up assessment included uroflowmetry and urine culture every 4 months in the first year and annually, thereafter. Urethrography and urethroscopy were performed in patients presenting any new or residual obstructive symptoms (such as slowed stream, strained stream, and hesitancy) or peak flow rate (Qmax)
Statistical analysis was performed using the SPSS 12.0 software (IBM Corp, Armonk, NY). Evaluation of data distribution showed a non-normal distribution of the study data set. Differences between groups in medians for quantitative variables and differences in distributions for categorical variables were tested with the Kruskal-Wallis 1-way analysis of variance and the chisquare test, respectively. Using multiple logistic regression with the enter method, the statistical significant variables as assessed in the univariate analysis were entered and investigated to predict the success of the procedure. The logistic regression analysis was carried out using date from patients for whom complete data were available. Age, etiology, urethral length, previous urethrotomy, and urethroplasty were used as predictors of success. Patients were also stratified according to the urethral length in the following 3 homogeneous groups with short, medium, or long strictures: stricture length was 1.5 cm in group 1, >1.5 and 3.9 cm in group 2, and 4 cm in group 3. Time to failure was analyzed using the Kaplan-Meier estimates and cox regression. Survival times were measured in months and were censored at the date of a patient being lost to follow-up or at the date of the last follow-up. An alpha value of 5% was considered as threshold for significance. Data are presented as mean standard deviation and median plus interquartile range (IQR). Odds ratios and 95% confidence intervals (CIs) were calculated for the parameters in each group using no complications as the reference group.
Descriptive data of the cohort are summarized in Table 1. Stricture length was 1.5 cm in 41 patients (24.7%; group 1), >1.5 and 3.9 cm in 99 patients (59.6%; group 2), and 4 cm in 26 patients (15.7%; group 3). Mean standard deviation stricture length was 2.6 1.3 cm (range, 1-10 cm). A total of 127 of 166 patients (76.6%) have undergone previous treatments before referral to our center: internal urethrotomy in 64 patients (38.6%), urethroplasty in 2 (1.2%), dilatation in 3 (1.8%), and multiple procedures in 58 (35%). In patients previously treated with urethrotomy, the number of urethrotomies ranged from 1 to 20 (mean, 3). There were no early postoperative complications such as wound infections, hematomas, or bleeding. At voiding urethrography after catheter removal at 3 weeks, in 9 patients (5.4%), a mild leakage at the graft anastomosis was observed. However, this resolved spontaneously with a 12-Fr catheter for 2 additional weeks. The median follow-up was 47 months (IQR, 33-95.5 months). No patient dropped out from the follow-up protocol.
Mean preoperative Qmax was 6.5 mL/s (range, 1-13 mL/s). Overall, 149 patients (89.8%) were considered successful and 17 as failures (10.2%). In the 149 successful cases, mean postoperative Qmax was 29.5 mL/s (range, 17-46 mL/s), mean stricture length was 2.48 cm (range, 1-8 cm), mean dorsal graft length was 2.15 cm, and mean ventral graft length was 4.1 cm. In the 17 failed cases, Qmax was 7.6 mL/s (range, 2-13 mL/s), mean stricture length was 3.53 cm (range, 1.5-10 cm), mean dorsal graft length was 2.3 cm, and mean ventral graft length was 4.3 cm. In the 21 cases with obstructive symptoms or Qmax The 17 failures were treated with urethrotomy in 10 patients (58.8%) with stenotic rings (located at the distal graft anastomosis in 3 cases, at the proximal graft anastomosis in 4, at the proximal and distal graft anastomosis in 3, respectively), and redo surgery in 7 patients (41.2%) with long recurrences: perineostomy in 3, ventral oral graft in 1, dorsal end-to-end plus ventral oral graft in 1, dorsolateral preputial graft in 1, and dorsolateral preputial plus oral graft in 1, respectively. The decision making on which kind of redo urethroplasty to perform was based on the local urethral conditions and the preferences of the patients. Median time to failure was 24 months (IQR, 12-36 months). Of the 17 failures, 15 (90%) were observed during the first 5 years of follow-up and 2 (10%) thereafter. Both patients with a late failure presented a urethral stricture >4 cm. No significant differences were recorded for age, stenosis etiology, and previous treatment (both urethrotomy and urethroplasty) between patients with success or failure.
On multivariate analysis (Table 2), urethral stricture length (odds ratio ¼ 1.743; 95% CI, 1.199-2.534; P <.001) was identified as an independent risk factor for failure. The risk for failure is increased by 70% per cm of urethral stricture length. When patients were stratified according to the urethral stricture length, patients with a longer urethral stricture (group 3) presented a higher rate of failure (6 of 26 patients; 23.1%) compared with those in group 2 (9 of 99 patients; 9.1%) and group 1 (2 of 41 patients; 4.9%). This difference corresponds to a significant difference in failure-free survival (P ¼ .006) visualized in Figure 2. The corresponding hazard ratio for patients with a >4 cm long urethral stricture was 1.390 (95% CI, 1.098-1.759). We did not observe any complication regarding the oral harvesting.
Surgical treatment of urethral stricture diseases is a continuous evolving process. Short and/or subobliterative bulbar strictures are traditionally treated by excision and primary anastomotic urethroplasty, whereas longer strictures are usually repaired by patch graft urethroplasty preferably using oral mucosa.3 Anyway, the shortening of the urethra and the vascular injury after urethral transection may lead to an increased risk of sexual complications, explaining the new trend to shift to the patch grafting without transecting in all nontraumatic and nonobliterative bulbar strictures, even the short ones.7-10,12 However, the graft technique is mainly suggested in cases which require simple augmentation of the urethral plate without excision of the scarred urethra, although the question remains if this procedure is also suitable to treat tight bulbar urethral strictures including a particularly narrow area. The best approach to treat this kind of strictures, reducing at the same time the risk of complications related to traditional end-to-end anastomosis, is still an open problem with several proposed solutions.
Guralnick et al suggested the graft-augmented anastomotic repair with the aim of reducing the urethral chordee. The technique consists in the excision of the narrow portion and anastomosis on one side of the urethra in conjunction with a patch graft on the opposite urethral side.16,17 A 90%-93% success rate was reported by using this procedure.18,19 Recently, Andrich and Mundy10 described a new anastomotic repair of bulbar urethra strictures without transecting the urethra, whereas Elliott et al and Barbagli et al favored the ventral grafting stating that excision of the narrow urethral plate was unnecessary because this grafting procedure provides a new sufficiently wide urethral plate.11,13 After the current trends of urethral-sparing surgery, we introduced the concept of a 2-sided dorsal plus ventral grafting for very narrow strictures where a single graft would not be sufficient to obtain a lumen of adequate width and the transection may compromise the urethral vascularization and length.14 The easy and quick ventral urethral access allows achieving a good visualization of both strictured tract and urethral plate.6,11,16 As there is no mobilization-rotation of the urethra, the ventral approach is less aggressive than the dorsal approach.
Furthermore, it is versatile because based on the condition of the urethral plate, it allows to choose different kinds of graft augmentations shifting, if necessary, from a single grafting (ventral or dorsal by Asopa) to a double dorsal plus ventral grafting. Compared with the dorsal approach, ventral opening might be associated with more blood loss and a risk of vascular damage to the spongious tissue, but no comparative study has ever been able to prove this. Our technique consists of a semicircumferential reconstruction, with the advantage of maintaining the urethral plate structure and avoiding the typical collapse of substitutive tissues in the tubular repairs. Another advantage is that the technique is suitable to treat very narrow strictures focally or for longer sections. As in all grafting procedures and opposite to the transecting techniques, the fibrotic urethral segment is not removed, and the strictured tract is only enlarged. Therefore, comparing with anastomotic procedures, our technique could lead to a lower success rate at long-term follow-up, but on the other hand, the preservation of the urethral plate avoids sexual complications related to transecting techniques as we have previously shown.15
However, during the experience with this new technique, our feeling was that success probability was increased when a good lateralization of the 2 urethral halves was possible after the dorsal incision of the urethral plate, creating a wide enough dorsal raw area where to allocate the graft. In contrast, in case of severe urethral spongiofibrosis, this lateralization was not feasible, leading us to convert necessarily in a transecting and anastomotic procedure. Furthermore, it should be emphasized that the procedure might be technically difficult or impossible to perform in very proximal bulbar strictures, leading in these cases to convert in a single ventral grafting. In our previous series with this technique, we reported 89.6% of success at short-term follow-up (mean, 22 months). In the present study, our procedure has confirmed to maintain over time a high success rate (89.8%) at long-term follow-up (mean, 59.3 months). These data agree with similar data reported in the literature about 1-sided augmentation using dorsal or ventral or lateral grafts.4,11,13 However, it has to be considered that our series was characterized by cases with urethral plate very narrow in which the traditional single-graft enlargement was inadequate or insufficient to guarantee a good chance of success. Our study shows that most of the failures (90%) were observed during the first 5 years postoperatively; afterward, the success rate remained stable. This claims the need for strict follow-up especially in the first 5 years. Similarly to other studies, age, stenosis etiology, and previous treatments (both urethrotomy and urethroplasty) were not significant predictors of surgical outcome.13 Conversely, stricture length was identified as an independent risk factor for failure. The risk of having a reintervention was increased by 70% per cm of urethral stricture length. Strictures 4 cm (group 3) presented a higher risk of failure compared with strictures 4 cm to be a risk factor for failure. We must acknowledge some limitations to our study. Even if our criteria of urinary outcomes and follow-up are comparable with those used by previous studies,4,9 we realize that a more strict follow-up methodology might detect unrecognized restrictures. On the other hand, it must be underlined that patients experiencing recurrent diseases over several years and undergoing multiple failed treatments do not accept invasive follow-up methods once they report no urinary symptoms at last and their quality of life is improved than ever before. Although all the patients were included in a prospective database, our study is a retrospective single-center experience. However, it represents the largest available series with the longest follow-up where all the procedures were performed by the same surgeon (E.P.) who originally described this technique. This could explain the low rate of long-term failures. Multicenter experiences from different centers and surgeons should confirm our data and possibly support our technique as one of the possible options for the management of patients with bulbar urethral strictures. If confirmed, our data could also highlight the importance of a long-term follow-up (up to 5 years) in patients with urethral strictures independently from the stricture length.
In patients with tight bulbar strictures, the 2-sided DVOG urethroplasty provides high long-term success rates, which may decrease primarily during the first 5 years; afterward, the success rate remains stable. The stricture length is an independent predictor of failure.