urethroplasty and reported a series of 38 patients where 65.8% were considered successful over a mean follow-up of 111 months.40 It is of interest, looking at this series, that the majority of the recurrences occurred within the first year and certainly this has been our first experience11 in the assessment of patients undergoing substitution urethroplasty who undergo objective assessment using urethrography or endoscopy and where there is no underlying progressive disease condition such as BXO. Certainly, in the longer term it is clear that the recurrence rate after a substitution procedure is far worse than many would otherwise have considered likely based on the existing literature, with a reported recurrence rate of 42% at 15 years for substitution procedures and 14% for anastomotic procedures.41 However, it needs to be borne in mind that this study reported on a mixed population of cases at a tertiary center and represents a worst case scenario. The complexity of the underlying stricture and whether it is a first time or redo procedure, will obviously dictate the subsequent success rate. We have previously demonstrated that if one utilizes flexible cystoscopy then the majority of recurrences for straightforward strictures, (not complex BXO which has to be considered a potentially progressive disease process), are present within the first 3–6 months following surgery.11

Oral Mucosal Grafts

The majority of patients now undergoing substitution urethroplasty, and in particular patients with BXO, are optimally managed by an oral mucosa patch substitution.Oral mucosa was first described in 1941 by Humby42 and was reintroduced into pediatric urology in 1992 by Berger.43 It is simple to harvest, tough, resilient, and easy to handle. It is taken as a full thickness graft and there is usually enough of it. It picks up a blood supply very effectively and has a thick epithelium with a thin dermis with a dense subdermal vascular plexus which allows early inosculation. In the majority of cases there is acceptable morbidity at the donor site. This mucosa is used to being wet and is resilient to skin diseases such as BXO, has a privileged immunology, and is also known from preclinical work (in our own laboratory) to have fibroblast behavior which results

in less fibrosis – quite different to that seen with skin. Oral mucosa can be harvested from the cheek (buccal mucosa), from the lip (labial mucosa), or from the tongue (lingual mucosa). If an extensive amount of oral mucosa is being harvested, or the patient has a small mouth then a nasal tube should be utilized for anesthesia, but for most cases this is not essential.

In harvesting buccal mucosa, the landmark to identify is the parotid duct opposite the upper second molar tooth. The length of the oral buccal mucosa that is necessary is then identified using a skin marker; we find infiltration with 1 in 200,000 lignocaine with adrenaline helpful and the buccal mucosa can then be excised with appropriate traction, sticking close to the submucosal layer, in the plane superficial to the underlying muscle. Labial mucosa can be managed in a similar fashion, but in our experience is much thinner and more difficult to handle and associated with greater morbidity.

Clearly, with any surgical procedure, there are potential complications (intraoperative hemorrhage, postoperative infection, pain, swelling, and damage to salivary ducts have all been reported). In some cases, patients do note initial limitation of oral opening, although this is usually transient. Occasionally, there can be loss or altered sensation within the cheek. A permanent palpable scar due to formation of a fibrous band may be noticed by the patient. Particularly, when harvesting tissue from the lower lip, both numbness and deformity have been reported. A recent review of their experience by Barbagli et al. in a survey of 295 patients reported that 98.4% would undergo the surgery again and these authors concluded that harvesting from a single cheek with closure of the donor site was a safe procedure with high patient satisfaction.44

Recently, in patients where a greater amount of oral mucosa needs to be harvested, it has been reported that it can be harvested from the tongue (lingual mucosa).45 The landmarks to be identified are the lingual duct and the lingual nerve. Having done this and having marked the extent of the mucosa necessary, it can be removed in a similar fashion to buccal mucosa. A comparative study of buccal and lingual mucosa is reported by Simmonato and anecdotally the resultant grafts from oral and lingual mucosa appear to be very similar in

though that this is a viable option only if there is a healthy well-vascularized urethral bed with limited residual ischemic spongiofibrosis and healthy urethral mucosa at both ends.53 Regrettably,where there is a lengthy stricture requiring substitution this is not often the case.Positive results had been reported by Fialo and colleagues using small intestinal submucosa (SIS) matrix54 although a recent update on this suggests that with longer term follow-up the success rate might not have been as good as previously reported.55 The use of SIS was reported by another group (Hauser et al) as being associated with a poor success rate.56 It is our perception that any technique which requires the ingrowth of endogenous epithelial and fibroblast cells is unlikely to be applicable to lengthy strictures or strictures where there is extensive spongiofibrosis or ischemia, or a poorly vascularized graft bed,and that the direction for future research should be toward the use of cell seeded matrices.

Graft Position

A question which is often considered is in which position to place the graft. Barbagli and colleagues57 reported the dorsal free graft urethroplasty initially using skin, but subsequently buccal mucosa (a modification of the Monseur technique).58 Initially, this was applied in the context of an augmented anastomotic repair. Recently, there has been debate about whether to transect the urethra or not, because of concern over whether this would further damage the blood supply. In our view If there is a severely ischemic area of corpus spongiosum it is unlikely to be important whether the urethra is transected or not as the residual blood flow through that ischemic area is not likely to be significant. In our experience, the dorsal onlay approach with an anastomosis (Fig. 31.6), Mundy59,60 provides a very effective technique which is easy to use and durable, and the majority of authors in the field61-63 report success rates in the order of 90%, although a large retrospective analysis by Barbagli27 suggested a success rate of 70%. Ultimately, as demonstrated previously by Andrich,41 it is likely that the success rate will depend on the complexity of the cases operated on and the type/duration of follow-up, but a reasonable success rate to quote to patients would be in the order of 85% at 5 years.

Figure 31.6. dorsal onlay approach to anastomotic urethroplasty (Mundy, BJUi surgical atlas).

When dealing with an onlay substitution, the options are a ventral, lateral, or dorsal approach (Fig. 31.7a–c) and our preference is for a dorsal or lateral approach, which is certainly supported by Barbagli et al.64 as the thickness of the corpus spongiosum both dorsally and laterally is far less than ventrally. Therefore, there is likely to be less bleeding from an incision in this plane and potentially less interference with blood supply as one extends into the proximal and distal “more normal” urethra. Barbagli et al. certainly reported comparable success rates of the order of 82–85% using ventral, dorsal, or lateral grafts in a small series of 50 patients. Recently, Kulkarni et al. reported a one-sided anterior dorsal approach, preserving the bulbospongiosus muscle and lateral vascular and nerve supply to the urethra, as having a success rate of 92%.65 This was from a small series of 24 patients with a short mean followup of 22 months.

A review of dorsal or ventral onlay grafting has suggested comparable success rates of

84–88% at 3 years,whichever approach is utilized for the onlay.29 Asopa in 2001 described a ventral sagittal urethrotomy approach66 (Fig.31.7a),with placement of a dorsal inlay graft. This has been reported to produce good success rates,in a study where 58 men underwent treatment with a mean follow up of 42 months with a success rate of 87%.67 More recently, Palminteri and colleagues have suggested that in addition to placement of a dorsal inlay graft via a ventral sagittal approach, a ventral onlay could be applied as well (Fig. 31.8a–e) reporting a success rate of 89% with a mean follow-up of 22 months, in 48 cases.68

As previously noted, one-stage tube repairs should not be used routinely, and it is clear from a review of the literature looking at two-stage procedures that the revision rate for a two-stage

procedure is of the order of 20–25%, which equates well with Greenwell’s findings of a 30% failure rate with a tube urethroplasty.30

An important indication for a two-stage procedure is in complex retrieval situations after failed previous surgery, for example, to retrieve a situation where there has been inappropriate use of a stent in the penile urethra (Fig. 31.9).

Two-stage reconstruction should certainly be considered to be most appropriate whenever there is concern about the success of any reconstructive procedure in the penile urethra, particularly following failed hypospadic repair or in the presence of severe BXO. The major factors to consider are whether there is an adequate residual roof strip to allow a one-stage reconstruction since a full tube reconstruction

Figure 31.9. inappropriate placement of a stent resulting in the requirement for a two-stage procedure.

has such a high failure rate and whether there is adequate subcutaneous tissue cover to prevent fistula formation. The two-stage skin reconstruction (Fig 31.10a–d) is very appropriate for the failed hypospadic patient, but clearly for BXO this is not possible and buccal mucosa must be used. There is only a small literature base relating to this29 and in particular it must be emphasized that there is a 22.5% revision rate in the literature for a first-stage urethroplasty.69 For one-stage closure, this would have automatically led to a failure. Clearly, with this in mind it is clear why tube substitution procedures have such a high failure rate. Furthermore, when talking about a two-stage procedure, bear in mind that the patient must be warned that the second stage can only be completed when the first stage is adequate for closure. Therefore, if there are

one or more revision procedures, it may in fact be a threeor more-stage procedure.

In carrying out penile surgery, the important factor to bear in mind is the tendency for cordee and use of an artificial erection is advised during the reconstruction. For a twostage procedure using either skin or buccal mucosa, this is easily accomplished using a standard technique. It must be borne in mind, however, that there are complications following first-stage urethroplasty with 10–39% of patients showing contractual scarring of the initial graft and this requires new grafting techniques.70 After the first-stage procedure and application of the graft, the wound is smeared with chloramphenicol jelly, dressed with a non-adherent dressing followed by gauze (Fig. 31.11a–c).

Second-stage closure requires tubularization of the first-stage procedure (Fig. 31.12b–d) and it is our practice to try and achieve a roof of 25–30 mm to provide a satisfactory substitution of the urethral lumen allowing for the inevitable contraction that occurs during subsequent healing. It is very important to achieve adequate tissue overclosure of the subsequent anastomosis, with careful attention to the underlying subcutaneous tissue to avoid linear overclosure of the skin subcutaneous tissues and urethra by offsetting the anastomotic line from the overlying skin and subcutaneous closures. If there is inadequate penile tissue then a useful technique is that of mobilizing a flap of scrotum to provide sufficient tissue cover for the urethral recon struction.