Kropp et al. suggested using SIS as an acellular urethroplasty matrix [60]. In one of the control groups, full thickness preputial skin from the host rabbit was used. The success rates were the same in both groups (100%), yet preputial graft procedures resulted in the formation of urethral diverticulum in all eight animals.

Chen et al. utilized BAMG (obtained and processed from porcine bladder submucosa) for ventral onlay urethroplasties in 10 rabbits [27]. Complete graft epithelialization was achieved after 2 months; the migration of organized muscle bundles was detected 6 months after implantation. No strictures or complications were observed.

In a study conducted by Nuininga et al., rabbits were divided into four groups, with six animals in each, according to the type of biomatrix used: in group 1, the rabbits underwent partial urethral replacement, which was replaced with one layer of SIS; in group 2, four-layer SIS patch grafts were utilized; group 3 was treated with collagen-based matrices produced from bovine tendons; and in group 4, the animals underwent a sham control operation (urethras were incised ventrally and then sutured) [61]. In groups 1 and 3, complete epithelialization was observed at 1 month after implantation, whereas in group 2, it was achieved at 3 months. One rabbit from group 3 developed a stricture and one rabbit from group 2 developed a fistula near the operation site.

Yang et al. suggested using an acellular corpus spongiosum matrix substitute as a patch [62]. These kinds of matrices were obtained from rabbit urethras and transplanted to repair urethral defects of 10–15 mm in length. Complete epithelialization of the extracellular matrix was achieved at 3 weeks post operation. Well-formed smooth muscle cells were observed after 6 weeks. No strictures or complications were reported.

Huang et al. explored the potential of SIS for urethral reconstruction in rabbits [63]. Tubular SIS grafts were applied in group 1 (n = 6), while a ventral onlay graft was used as a patch in group 2 (n = 6). The epithelium covered the grafts fully after 6 weeks.

An interesting study was published by Kanatani et al., wherein two types of tubularized graft based on type I collagen sponge, reinforced with co-poly(Lactide/ε-caprolactone) (CLLC) fabric, were created [51]. In group 1, the P(LA/CL) fibers were knitted into a vascular stent style; in group 2, the P(LA/CL) fibers were weaved into a mesh style. Each of the grafts was applied to a 15-mm urethral defect. Numerous complications (stenoses, fistulae, or stone formation) were observed in group 1, whereas no complications were seen in group 2. The authors concluded that creation of a scaffold suitable for urethral tissue regeneration depends not only on the biomaterial composition, but also on the fabrication technique.

In a very important study, Dorin et al. determined the maximum potential distance of normal native tissue regeneration when using tubularized unseeded matrices [64]. Twelve rabbits were divided into four groups. In each group, urethral defects of varying length (0.5, 1.0, 2.0, and 3.0 cm) were created. At week 4, only the 0.5 cm group had a normal layer of epithelium surrounded by a layer of smooth muscle within the urethral lumen. In all groups with longer defects, strictures developed by 4 weeks. Therefore, this study proved that acellular matrices in tubularized grafts may be used successfully only to repair defects limited in size.

Villoldo et al. assessed the efficacy of onlay urethroplasty with SIS in rabbits [65]. A total of 30 animals included in the study had 15 mm of the ventral wall of the penile urethra excised. One month later, the rabbits were divided into two groups: group 1 was the control group; in group 2, where the created defect was patched with an SIS onlay graft, epithelialization was achieved after 15 days and smooth muscle bundles were detected at 6 months follow-up. The fact that urethral strictures were created 1 month before urethroplasties were performed brings this study a step closer to clinical reality.

Chung et al. studied the potential of using acellular matrices derived from silk fibroin for urethral repair [66]. The results of urethroplasty with silk fibroin matrices (SFMs) were compared to those with SIS. The follow-up period was 3 months, with no intermediate time points. The success rates of SFM and SIS were found to be the same. No strictures or complications were reported. However, the silk fibroin scaffolds showed reduced immunogenicity.

Kajbafzadeh et al. explored the role of preputial acellular matrix (PAM) for urethral reconstruction [67]. Prepuces were obtained from circumcised boys. The prepuces were decellularized and the produced matrices were used as patches in ventral onlay urethroplasty procedures: PAM in group 1; and PAM plus fibrin sealant in group 2. The effectiveness of repair was evaluated in both groups at a 9 month follow-up. The authors noted that satisfactory vascularity and smooth muscle layer formation were more significant in group 2 (PAM + fibrin sealant).

In all the above-mentioned studies, only male rabbits were used as animal models. In most studies, a longitudinal urethral defect was created, varying in length (10–20 mm); in some studies it was a tubular defect 5–30 mm long. SIS was used as a urethroplasty matrix in four of ten studies, with BAMG being the second most frequently utilized type of scaffold (applied in three of ten studies). One should note that no episodes of graft rejection were reported. Acellular matrices were found to be highly effective in animal models, yet their application range was limited by the size of urethral defect. Complete formation of the urothelial layer on the inner surface of the graft was achieved 4–12 weeks after implantation; either single smooth muscle cells or a regenerated smooth muscle layer was observed 2–12 months after implantation.

The feasibility of applying a bladder submucosal, collagen-based inert matrix as a free graft substitute for urethral repair in patients with hypospadias was explored [34]. All four patients had had a history of hypospadia repair procedures and required yet another repair. The neourethras were created from BAMGs in the size range 5–15 cm. Postoperatively, only one patient developed a fistula. Histological evaluation showed typical urethral stratified epithelium at the site of surgery.

Mantovani et al. were the first to utilize an SIS graft for substitution urethroplasty in a 72 year-old patient with a long stricture of the penile and bulbar urethra [68]. The dorsal onlay technique was used. At a 16 month follow-up, the maximum urine flow rate was 14 ml/s. No complications were observed intraoperatively or postoperatively. In the same year, El-Kassaby et al. used a BAMG matrix for urethral stricture repair [35]. A total of 28 patients with urethral strictures of varied length (1.5–16 cm) underwent reconstructive surgery wherein the ventral onlay technique was used. The success rate was 86%, as one patient developed a urethrocutaneous fistula that closed spontaneously 1 year after repair.

Lin et al. suggested using an acellular dermal matrix (ADM) graft for urethral reconstruction [42]. Homologous ADM was applied as a tubularized graft in 16 patients with urethral strictures and hypospadias. No episodes of rejection were observed. During the 46 month follow-up period, only four patients needed periodical urethral dilatation.

Le Roux et al. evaluated SIS as a substitute in endoscopic urethroplasty [38]. Nine patients were enrolled. Optical urethrotomy was performed prior to the SIS graft implantation. Subsequently, a tubularized SIS graft was implanted into the stricture site. Only two patients had patent lumen at 1 and 2 years follow-up, respectively; all the other patients developed recurring strictures. The authors concluded that endoscopic urethroplasty with unseeded SIS grafts is not to be recommended.

Donkov et al. assessed the success rates of SIS grafts used for dorsal onlay substitution urethroplasty [69]. The graft was fixed using a modified Barbagli technique. The success rate was 89% (eight of nine patients) at 18 months follow-up.

While Hauser et al. also chose SIS for dorsal onlay substitution urethroplasty, the success rate was much lower: only one of five patients did not have stricture recurrence at a 12 month follow-up [70]. The complications observed postoperatively were extravasation, severe urethritis, and urinary tract infection.

Palminteri et al. evaluated the role of SIS grafts for substitution urethroplasties [39]. Three techniques were employed: a dorsal inlay graft was performed in 14 patients, ventral onlay in one patient, and dorsal onlay plus ventral onlay in five patients; 21 month follow-up data demonstrated an average success rate of 85%. The average stricture length was 3 cm. No complications were noted. The three failures were in penile and penile–bulbar urethral strictures with an average length of 5.7 cm.

Fiala et al. used SIS grafts for urethroplasties in 50 patients [71]. The success rate averaged 80%, with a mean follow-up time of 31 months. Recurring strictures developed in one of 10 bulbar, five of 31 bulbopenile, and four of nine penile strictures; these occurred during the first 6 months postoperatively. Farahat et al. placed SIS grafts endoscopically [72]. Only two cases of 10 exhibited stricture recurrence at a mean follow-up time of 14 months; there were no complications.

El-Kassaby et al. conducted a comparative study between BAMGs and buccal mucosal grafts in anterior urethral strictures [36]. The patients were followed for 25 months on average. The success rate for BAMG was 53%, and that of the buccal mucosal graft was 100%. The authors divided patients from the two groups into subgroups: (a) those with a healthy urethral bed (less than two prior surgeries), where BAMG surgeries were successful in 89% of cases and buccal mucosa surgeries in 100%; and (b) those with an unhealthy urethral bed (more than two prior surgeries), where the success rates were 33% (BAMG) and 100% (buccal mucosa).

Palminteri et al. reported the longest (as of today) mean follow-up period (71 months), having worked with patients who had bulbar urethral strictures and underwent urethroplasties using SIS [40]. The success rate was 76%. It should be noted that the failure rate was 14% for strictures <4 cm and 100% for strictures >4 cm.

In the studies mentioned above, the most frequently used type of matrix was SIS, which was, in most cases, used as a material for patch grafts. Tubularized grafts were applied twice: in endoscopic urethroplasties (with a success rate of 22%) [38] and in conventional urethroplasties (with a success rate of 88%); however, when describing the cases of conventional urethroplastic surgery, the authors did not mention the lengths of the urethral stricture [42]. No episodes of graft rejection were reported. Biopsies taken in five studies out of 13 showed normal urethral tissue characteristics at the site of surgery. Almost all studies using acellular matrices reported success rates of 75% or more. Failures were more common in patients with long urethral strictures (>4 cm), in those who had had previous urethroplasties (unhealthy urethral bed, unsatisfactory vascularity), and in those with penile or penile–bulbar strictures. The two studies that used endoscopic approaches exhibited different success rates: 22% vs 80% [38, 72]. The key differences among the groups were in the way the SIS grafts were secured: Le Roux et al. used tubularized grafts [38], while Farahat et al. chose patches [72]. Additionally, the average stricture length in the former study was 2.2 cm, as opposed to 1.47 cm in the latter study. Apparently, the difference in the success rates can be attributed to these two factors.

In addition, Donkov et al. and Hauser et al. both applied SIS in a similar technique (dorsal onlay substitution), but the success rates were completely different (89% vs 20%, respectively) [69, 70]. The difference can be explained by the difference in the way SIS was secured: in the study by Donkov et al. [69], the patch was spread-fixed onto the tunica albuginea, while the mucosa was sutured to the graft margins; in the study by Hauser et al. [70], SIS was anastomosed only to the incised urethra, without being fixed onto the tunica albuginea. Moreover, the strictures in the former study were shorter than those in the latter (4–6 cm vs 3.5–10 cm).

De Filippo et al. were the first to use a tubularized tissue-engineered construct derived from bladder submucosa and seeded with autologous urothelial bladder cells for urethroplasty in rabbits [56]. The cells were obtained via an open biopsy. Urethroplasties were performed to repair 10 mm-long defects. In the control group, similar defects were repaired with unseeded tubularized BAMGs. After 6 months of follow-up, there were no strictures in the former group, whereas the control group exhibited strictures at the site of surgery.

A similar study was conducted by Li et al., although instead of urothelial bladder cells they used OKs [57]. The ventral onlay graft procedure was used to repair 20 mm-long defects. Rabbits in the experimental group (BAMG plus OKs) developed no strictures or complications; in the control group, two rabbits died of infection, two had fistulae, and the remaining eight developed strictures.

Fu et al. used tubularized BAMGs seeded with autologous preputial keratinocytes to repair 15 mm-long defects in three rabbits; there was no control group in the study [58]. At a 2 month follow-up, urethrography showed a wide urethral caliber in all three animals, and histological evaluation showed complete epithelialization at the site of surgery.

Of particular interest is the paper published by Gu et al. [73] They implanted silastic tubes into the peritoneal cavities of nine rabbits. After 2 weeks, the authors harvested the tubes covered by tissue, histological analysis of which revealed myofibroblasts embedded in collagen bundles covered by an outer layer of mesothelium. The tissue was everted and used as a tubularized graft to repair 15 mm-long urethral defects. At 1 month post operation, histological and immunohistochemical analyses showed normal urethral architecture.

In their study, Micol et al. used high-density collagen gel tubes as grafts that were seeded with autologous bladder smooth muscle cells [29]. It took the authors approximately 24 h to produce each tubularized graft. They did not place catheters postoperatively. Seven of sixteen rabbits developed fistulae. In both groups (with eight animals in each), equal numbers of smooth muscle cells were observed after 1 month.

Gu et al. isolated mesothelial cells via omentum biopsy and seeded them onto BAMGs [74]. Fifteen mm-long defects were substituted by tubularized matrices seeded with cells (in nine rabbits) and by those without cells (in nine rabbits from the control group). In the latter, all animals developed strictures, while no stricture formation was observed in the experimental group. At 6 months after implantation, the neo-urethras could not be histologically distinguished from the native urethras.

Xie et al. prepared electrospun silk fibroin matrices and seeded them with urothelial bladder cells that had been isolated and expanded [18]. Dorsal urethral defects 30 mm long were created in female dogs. At 6 months after implantation, the neo-urethras could not be histologically distinguished from the native urethras. It should be noted that the artificial defects involved only the mucosa, not the smooth muscle layer.

Li et al. seeded BAMGs with either epithelial differentiated or undifferentiated rabbit adipose-derived stem cells [75]. BAMGs were labelled with 5-bromo-2′-deoxyuridine. Thirty-six rabbits were divided into three groups (12 animals/group); 20 mm ventral urethral defects were created. In group A, defects were substituted only with BAMGs; in group B with BAMGs plus TGFβ1 siRNA-transfected fibroblasts (Und-rASCs); and in group C with BAMGs plus epithelial-differentiated rabbit adipose-derived stem cells (Epith-rASCs). Only one stricture was observed in group C, whereas almost all animals developed strictures in the first two groups. Complete epithelialization was achieved only in group C.

Wang et al. used a denuded human amniotic scaffold (dHAS); 5 × 10 mm defects were created on the ventral wall of the urethra and were repaired with either dHAS alone (group 1, n = 6) or dHAS + urethral urothelial cells (group 2, n = 6). In group 1, one animal developed an infection and one had a fistula; group 2 exhibited no complications or strictures (the efficacy rate was 100%). Mild inflammatory infiltration was observed in cell-seeded dHAS grafts, as revealed by less pronounced accumulation of CD4+ and CD8+ cells (or neutrophils or other immune cells). Histopathological analysis identified a mild immune response in the cell-seeded dHAS group (p < 0.05). Urethral defects in group 2 were completely resolved in 1 month. At 3 months after surgery, the formation of a smooth muscular layer and rich blood vessels was apparent [76].

In a very recently published Korean study by Chun and Kim et al., they evaluated the combined effect of acellular bladder submucosa matrix (BSM) and autologous urethral tissue for the treatment of long segment urethral stricture in a rabbit model [20]. To prepare the BSM, porcine bladder submucosa was processed, decellularized, configured into a sheet-like shape, and sterilized (Fig. 9.2). Twenty rabbits were randomly divided into groups: normal control, urethral stricture, urethroplasty using BSM only, or BSM/autologous urethral tissue (n = 5 per group). Brief urethroplasty methods are represented in Fig. 9.3. The width of the penile urethra was measured by postoperative urethrograms at 4, 8, and 12 weeks. The control urethrogram showed a wide urethral caliber. Both graft groups showed a similar width to those of the normal group, and the stricture group revealed stenosis. The mean urethral width of the control, stricture, BSM, and BSM/autologous urethral tissue groups at week 12 was 10.3 ± 0.80, 3.8 ± 1.35, 8.8 ± 0.84, and 9.1 ± 1.14 mm, respectively. Although the difference in the width of the BSM and BSM/tissue graft urethroplasty groups was not statistically significant, the BSM/tissue graft group diameter was approximately 0.567 mm wider (Fig. 9.4). The histopathology study revealed that the BSM/autologous urethral tissue graft had a normal urethral lumen area, compact muscular layers, complete epithelialization, and progressive infiltration by vessels in the regenerated urethra. In contrast, the BSM grafts revealed keratinized epithelium, abundant collagenized fibrous connective tissue, and were devoid of bundles of circular smooth muscle (Fig. 9.5). According to these results, the BSM/tissue graft had well-organized integration with the recipient tissues, and the cells within embedded autologous tissue and BSM can synchronistically act as biological activators for incorporation, promoting ingrowth of surrounding urethral tissue into the scaffold. This indicates that autologous urethral tissue at the time of implantation is effective for urethral reconstruction. A BSM graft embedded with autologous urethral tissue can be applied for clinical treatment of strictures requiring partially resected urethra replacements.

Male rabbits were used in ten of the eleven studies reviewed above. Female dogs were used as animal models in one study. Despite the fact that preclinical and clinical studies demonstrated better success rates with SIS, six of the eleven reviewed studies utilized BAMGs. SIS grafts were not applied at all. In most studies, matrices were seeded with epithelial cells. Also, mesothelial cells, bladder smooth muscle cells, bone marrow mesenchymal stem cells, and rabbit adipose-derived stem cells were used. The authors did not mention any problems with expanding the cells or seeding them onto different matrices; 50% of the surgeries were performed with tubularized constructs; and 8 of 11 studies had control groups, where similar matrices without cells were used. All 11 studies showed significantly better results with cellularized matrices. Epithelial cells and smooth muscle cells at the site of surgery were observed at a mean follow-up time of 4 weeks.

Fossum et al. obtained urothelial cells via bladder washing [55]. The cells were cultured and seeded onto decellularized dermal matrices that were used to surgically treat six boys with scrotal or perineal hypospadias. The mean follow-up time was 87 months. Only one boy had to undergo an optical urethrotomy to improve voiding.

Feng et al. seeded two types of autologous cells (lingual keratinocytes and corporal smooth muscle cells) onto both sides of porcine acellular corpus spongiosum matrices (ACSMs) [59]. The best results were achieved in the group in which urethral defects were repaired with matrices containing both types of cells. Animals in the other two groups (ACSMs alone, and ACSMs plus OKs) developed strictures.

De Filippo et al. substituted 30 mm defects with either tubularized BAMGs alone, or BAMGs seeded with two types of cells (BAMGs plus BUCs plus BSMCs) [77]. The success rate in the group wherein cell-seeded matrices were utilized was 100%.

Mikami et al. formed tubularized tissue engineered urethras comprised of collagen mesh matrices and autologous cells of two types—keratinocytes and buccal mucosa smooth muscle cells [78]. These grafts were used to repair urethral defects in the experimental group (10 dogs) and were not applied to defects in the control group (10 dogs). The authors reported statistically significant differences in the number of complications between the two groups. The control dogs developed fistulae and strictures.

Orabi et al. used the same types of matrix in male dogs; 6 cm perineal urethra segments were removed and substituted with either tubularized matrices with two types of cells (BAMGs plus BUCs and BSMCs) or BAMGs alone (control group) [19]. In the control group, 100% of animals developed strictures and fistulae, whereas in the experimental group there were no strictures at the 12 month follow-up.

Li et al. hypothesized that TGFβ1 plays a very important role in the formation of fibrosis. Using OK- and TGFβ1 siRNA transfected fibroblast-seeded BAMGs, the authors were able to minimize the activity of TGFβ1 and avoid the formation of fibrosis [79]. In three groups of rabbits (nine animals/group) the urethral defects were repaired with: (a) BAMGs alone; (b) BAMGs plus OKs; and (c) BAMGs plus OKs plus TGF-β1-siRNA transfected fibroblasts. Rabbits in group (a) developed strictures, while those in groups (b) and (c) had none. However, in group (c) the formation of capillaries in the epithelial lower layer was observed at 6 months after implantation.

Xie et al. seeded autologous OKs and fibroblasts onto silk fibroin matrices; 5 cm urethral mucosal defects were repaired using matrices alone and matrices with two cell types (in five female dogs in each case) [80]. No strictures or complications developed in the group with cell-seeded matrices.

Tissue-engineered constructs with different cell types in a co-culture system are structurally more similar to the native urethra than other constructs. In three of the six studies reviewed above, BAMGs were utilized. The inner surface of the matrix was seeded with OKs (in four of six studies) or bladder urothelial cells (in two of six studies), while the outer surface of the matrix was seeded with smooth muscle cells or fibroblasts (in two of six studies). Each study included a control group in which matrices without cells were used. The authors did not mention any problems regarding culturing autologous cells or creating tissue-engineered constructs with two types of cells. The animal models were male rabbits (in three of six studies), male dogs (in two of six studies), and female dogs (in one study). Tubularized constructs were applied in 50% of studies. The authors used tissue-engineered constructs containing different cell types with 100% success in five of six studies and with 70% success in one study.

Bhargava et al. were the first to use a tissue-engineered construct for urethral repair in clinical patients [54]. Sterilized donor de-epidermized dermis was utilized as a matrix, onto which buccal mucosa keratinocytes and fibroblasts were seeded. Five patients with urethral stricture secondary to lichen sclerosis participated in the study. Three patients underwent a two-stage procedure and two patients a one-stage procedure. Postoperatively, one patient required complete graft excision because of marked fibrosis, and one had partial graft excision due to hyperproliferation of tissue. Three patients required some form of instrumentation because of recurring strictures. At present, the mean follow-up time is 111.8 months [81]. All patients who underwent urethroplasties are able to void independently. All four patients with tissue-engineered buccal mucosa (TEBM) void spontaneously. Three of five patients used intermittent self-calibration.

Raya-Rivera et al. used a PGA matrix, of which the inner surface was seeded with expanded bladder epithelial cells and the outer surface with bladder smooth muscle cells [50]. All patients had post-traumatic obliteration of membranous urethra. They had scar tissues excised and tissue-engineered urethras implanted. The success rate was 100% at a 71 month follow-up. Biopsies showed that the grafts had developed a normal-appearing architecture by 3 months after implantation.

To sum up, the authors applied tissue-engineered constructs with two types of cells to manage very complicated cases of urethral stricture disease: lichen sclerosis in Bhargava et al. and post-traumatic membranous urethral obliteration in Raya-Rivera et al. [50, 54] In the former, additional interventions were needed, while in the latter, the use of tubularized synthetic matrices seeded with urothelial and smooth muscle cells obtained from bladder demonstrated 100% success at long-term follow-up. Histological evaluation showed that the engineered grafts had developed apparently normal architecture by 3 months after implantation. All adverse fibrotic reactions were observed within the first year after implantation. The results have been durable for three out of five patients for 9 years, although all of them required re-interventions [81].

Urethral tissue engineering provides fertile ground for further research, with many problems remaining unresolved. One of the questions to be answered is, “What are the potential indications for substitution urethroplasty procedures using tissue-engineered constructs?” At present, clinicians are successfully utilizing oral (buccal) mucosa to repair complex urethral strictures and hypospadias. Its use has yielded good results, not only in patients with primary urethral strictures but also in those with recurring strictures and lichen sclerosis, with most authors reporting low donor-site morbidity (sometimes extremely low, especially when mucosa is harvested from the cheek or the ventral surface of the tongue). Moreover, the procedure of harvesting is simple, reproducible, and well tolerated by the majority of patients. The view that oral mucosa may not provide enough tissue for urethral repair seems to be groundless. Mucosa from both cheeks as well as from the ventral surface of the tongue may provide enough material to create a 16 cm graft.