Oral Cavity Reconstruction Outcomes Using a Porcine Urinary Bladder Matrix: A Retrospective Case Series

Objective. The purpose of this study is to assess healing outcomes in full-thickness mucosal wounds following the use of a porcine urinary bladder matrix to augment mixed oral cavity repairs. Materials and Methods. A retrospective chart analysis was conducted over a 58-month timespan. Participants included individuals with osteoradionecrosis. Descriptive measures obtained in the postoperative setting were used to examine wound healing outcomes. Results. Thirty-nine encounters with 35 patients met inclusion criteria for assessment. The mean defect size repaired was 14 cm². Successful healing occurred in 64% of cases. Scarring was observed in 10 cases, and 3 cases demonstrated transient functional deficits. Reapplication of the xenograft was required in 4 cases. Only 1 acute event of hemorrhage and 1 infection were observed in the postoperative period. Conclusions. Use of porcine urinary bladder matrix grafts for oral cavity reconstruction was well tolerated in a diverse number of wound scenarios with a relatively low risk of postoperative complication. The use of porcine urinary bladder matrix was not observed to provide any noteworthy advantages for the healing of recalcitrant osteoradionecrosis wounds.

The oral cavity represents an area of critical tissue surface where depreciated function from poor healing can have drastic consequences on quality of life. Some of the main goals of reconstruction of oral cavity defects are to reestablish the mucosal lining, replace lost mucosa with tissue of similar function, preserve or improve function of the remaining structures, and achieve a pleasing aesthetic outcome.¹ Unique challenges to the repair of oral cavity wounds resulting in full-thickness mucosal defects include having limited local tissue available to utilize for covering a wound and low patient tolerance of surgical techniques that restrict movement of oral tissue through tension, graft bulk, or scarring. In addition, operations in the oral cavity are sometimes performed in previously irradiated tissue, which creates added challenges to achieving reconstructive optimization of functional and aesthetic outcomes by producing a poor vascular bed that is subject to scarring or, in some cases, the development of osteoradionecrosis (ORN). 

Depending on their size and location within the oral cavity, full-thickness mucosal wounds can be managed using a number of reconstructive techniques.^1-7 When possible, primary closure is preferred if it can be performed without generating tension.^4,7,8 For limited-size defects not amenable to primary closure, the use of a split-thickness skin graft (STSG) is preferred along with mucosal grafts when harvesting is possible.^4,6 More extensive defects with larger tissue loss may be reconstructed with the use of free, local, regional, and myomucosal flaps.^1-3,5-7 This graded approach to the management of full-thickness mucosal defects provides a gestalt to the techniques employed by surgeons. While some studies^6,7 have attempted to generate algorithms to approach reconstruction of certain oral cavity defects, in practice a fair amount of individual variation exists in modality selection for any restoration. Decision making is influenced by several factors, including defect size and location, prior therapy (eg, previous surgery, radiation, chemotherapy, etc), comorbidities, donor site morbidity, and surgeon preference.^4,7 

As an additional option, the application of extracellular matrix materials (also referred to as acellular dermal matrix [ADM]) as either primary or adjunctive therapy to wound care presents great potential for reconstructive procedures in sensitive regions. Decellularized biologic scaffold materials have been utilized in breast reconstruction, esophageal patch repairs, radiation damage, open pilonidal wounds, and hernia repairs.^9-12 In the chronic wound setting, they have been used in the treatment of diabetic and venous stasis ulcers.¹ In the head and neck, applications have encompassed the repairs of nasal septal, palatal, periorbital, periodontal, and tympanic membrane defects.^13-17 

Benefits include the material’s ability to promote a conductive microenvironment. Preclinical research exploring the polarized macrophage phenotype profiles recruited during incorporation of decellularized extracellular matrix products as part of the wound healing process have observed notably greater M2 cell-type responses.^9,18 This represents a repair pathway more associated with constructive tissue remodeling outcomes rather than the cicatricial, inflammatory outcomes characterizing an M1-type response.¹⁸ 

Complications from using biologic ADM materials are typically minimal. Drawbacks experienced when using xenograft or allograft matrices include early resorption of the matrix postoperatively, seroma/hematoma formation in defects in which large grafts were used, and a lack of randomized controlled trials evaluating ADM products providing level I evidence.⁵ Other possible risks include those that come with the incorporation of a foreign body, such as the potential for infectious agent (viral) inoculation into the host or the potential for an immunologic reaction to the implanted material.¹⁹ The manufacturing process of producing any acellular matrix also is important.^19,20 While cellular stripping of the source tissue helps minimize the implant’s risk of stimulating host immune response, many matrix preparation techniques are used. Certain types can result in premature matrix degradation with loss of growth factors that can cause an accelerated rate of product reabsorption and a greater risk of scar formation.^19,20 

The purpose of this research is to assess outcomes of wound healing in oral cavity repairs using a porcine urinary bladder matrix (UBM). Postoperative outcomes investigated to assess wound healing using UBM performance included determining the frequency of infection, hemorrhage, and scar development, as well as inspecting for functional return of speech and swallowing.

Retrospective analysis was conducted through the surgical department at Sanford Medical Center, a tertiary facility with 545 beds and the largest teaching hospital in Sioux Falls, South Dakota. Inclusion was limited to patients who underwent MatriStem UBM (ACell Inc, Columbia, MD) reconstruction between September 2011 and July 2016. Participants were selected via medical record query for surgical cases that used pig bladder products for head and neck wound repairs. Exclusion criteria included procedures for the surgical repair of an oral cavity defect not using pig bladder products. An initial query for head and neck repairs using UBM identified 189 cases, from which 39 cases involving 35 patients who presented specifically for oral cavity wound repair with UBM were selected for analysis. 

From the cohort, 7 subgroups were developed based on tissue site placement of the graft for reconstruction. These included reconstructions on the mandible, maxilla, tongue, palate, pharynx, lip, and mixed surfaces. Mixed wound reconstruction sites included those with graft coverage over more than 1 structural area. Areas of combined coverage included the tongue, oral floor, buccal wall, retromolar region, and palate surfaces. 

No comparison was made with a control group because the primary purpose was to evaluate the frequency of postoperative infection, hemorrhage, and scarring events in healed cases. Each case was inspected through the time required for complete healing, defined as the point when complete remucosalization of the wound was achieved, which was identified during postoperative follow-up appointments or at the point of graft treatment interruption. Failed healing was defined as premature total graft loss by detachment or intentional removal prior to the wound achieving complete remucosalization or persistence of a wound defect after resorption of the UBM graft. Therefore, a secondary objective was to identify the quality of wound healing outcomes as complete or incomplete. Only descriptive measures obtained in the postoperative setting were used to examine wound healing outcome results with the UBM product. 

Technique

The application of UBM was performed using a 3-layer sheet cut to about the size of the defect in all cases. This product is available in 2 forms: a single or multilayer sheet or a powder, particulate form.^21,22 

In each case, UBM powder was first applied with sufficient density to lightly cover the wound surface without creating product mounding and then followed  with overlay of the rehydrated UBM trilaminar sheet. Quilting sutures using either 5-0 Chromic or 4-0 Vicryl (Ethicon, Bridgewater, NJ) secured the graft to its position over the wound site (Figure).

A summary of patient demographic information is provided in Table 1. Malignancy resection composed the majority of inciting pathology necessitating reconstruction of the oral wound repairs reviewed. Others included benign lesions, bone pathology, and secondary surgical revisions. An analysis of wound source pathologies is provided in Table 2.

Successful healing was confirmed in 64% (n = 25) of all encounters; the remaining 36% (n = 14) had failed healing. Within the 25 successfully healed cases, 52% (n = 13) had outcomes with the regeneration of normal appearing mucosa and excellent tissue function, while 48% (n = 12) developed scarring and functional impairment. 

For the 14 cases of failed healing, the associated causes of failure included ORN (n = 7; 50%), osteomyelitis (n = 1; 7.1%), premature graft removal due to loosening (n = 1; 7.1%), hemorrhage (n = 1; 7.1%), or malignancy recurrence requiring additional resection (n = 4; 28.6%). Causes of failure along with the site of grafting appear in Table 3. Notably, no cases repaired for ORN achieved definitive tissue healing. In 2 cases, ORN wounds did obtain complete remucosalization but were found to redevelop mucosal ulcerations and bone exposure during a period of postoperative observation over 6 months. For the 4 cases of incomplete wound healing classified as having malignancy recurrence, 2 required early graft removal postoperatively due to pathology report findings of positive margins, which prompted further resection. For the remaining 2 cases, persistent ulceration was noted at the wound site after UBM graft disintegration. One case went on to develop leukoplakia requiring CO₂ laser ablation. The other underwent biopsy after gradual ulcer expansion was observed. The results of the biopsy showed squamous cell carcinoma, and additional resection was performed. Finally, in 1 case, premature graft detachment occurred as a result of hemorrhaging beneath the graft on postoperative day 7, which required removal of the graft to stop the bleeding. 

In cases that successfully healed, scarring developed in 40% (n = 10). In 12% (n = 3) of the cases, there was persistent functional deficit that did not completely resolve with tissue healing. In addition, 1 case (4%) had an infection documented in the postoperative period. The distribution of observed wound healing outcomes by tissue is provided in Table 4. 

The majority of postoperative scarring, functional impairment, and infection events were identified in reconstructions that involved either a large area impacting multiple tissue surfaces or tissues with a higher level of inherent mobility, such as the tongue. While scarring was the most common event encountered, there was not a correlated decrease in tissue function for successfully healed cases. 

Assessment of comorbidities known to be detrimental to healing included evaluation for the presence of hypertension, diabetes, vascular disease, chronic obstructive pulmonary disease, and smoking history. Out of 39 encounters reviewed, 20 encounters comprising 19 patients presented with such comorbidities (Table 5). Of those 20 encounters, 45% (n = 9) comprising 9 patients had multiple comorbidities. 

Overall, hypertension represented the most common comorbidity, followed by current smoking. For cases (12 cases total) with postoperative scarring, bleeding, infection, or functional impairment, only 33% (n = 4) had 1 or more comorbidities versus 77% (n = 10 [13 cases total]) of those with excellent outcomes having 1 or more comorbidities.

The use of extracellular matrix scaffold material in wound reconstructive surgery has received increased attention over the last decade.¹¹ The versatility of using UBM is demonstrated in the variety of cases in the present study. A minimal number of events affecting the safety of UBM graft use were observed, with wound healing occurring in the majority of cases.

In this study, 64% of cases achieved complete tissue healing with use of a UBM graft. This was lower than the healing outcomes achieved in studies (83%–100%) using other ADM graft types in the repair of full-thickness oral mucosal wounds.^4,8,23-25 The healing success outcomes of other modalities in oral cavity wound reconstruction also are favorable, including STSG (90%–100%),^4,24 buccal fat pad (86%),²⁶ and free tissue flap repairs (> 95%).^6,27

Previous studies assessing ADM in the repair of full-thickness mucosal defects have commented on its usefulness and recommended it as a modality to be considered in oral cavity repair options. Girod et al⁴ prospectively assessed 34 patients and compared the outcomes of oral cavity wounds reconstructed using ADM and STSG. The authors identified successful healing rates of 86.4% and 100% in the ADM and STSG groups, respectively.⁴ Both study groups were similar in average wound size repaired (ADM: 17.6 cm²; STSG: 16 cm²) and had predominantly the same repair distribution within the oral cavity, with most performed on the tongue and floor of the mouth.⁴ Key differences between the compared groups included having a separate primary surgeon for each group and a higher number of ADM subjects with preoperative radiation therapy (45%) compared with the STSG group (17%). The authors⁴ felt ADMs of greater thickness and the more frequent use of ADMs in patients with prior radiation exposure contributed to the higher graft failure observed in the ADM group. Noted advantages of the ADM graft over STSG were a lack of donor site morbidity, reduced procedure cost, and shorter operating times, as well as noninferior functional, cosmetic, and patient satisfaction outcomes.⁴

In another prospective study, Jamal et al⁸ compared repairs of buccal mucosa graft harvest sites by primary closure with ADM grafting on 20 healthy, nonchemotherapy patients (10/group). Mucosal graft harvest size was reported to be between 10 cm² and 14 cm². All patients had successful healing of buccal wound sites. The authors⁸ found that ADM was an effective means of wound repair but did not note any clear advantages over primary closure. There were no significant differences in postoperative pain scores between groups.⁸ More ADM group members experienced postoperative cheek swelling at 3 weeks than in primary closure groups (80% vs. 30%, respectively), but this did not affect functional outcomes and resolved in both groups by 3 months.⁸ 

A noncomparative, retrospective study by Richardson et al²⁵ examined the healing outcomes of ADM used to augment a multilayered repair of anterior palatal fistula defects as an alternative to regional and free flap repair or an obturator in 12 cases (average age, 18.5 years; range, 2-29 years). The authors²⁵ identified a successful healing rate of 83.3% (10/12), complete fistula closure (8/10), and reduced size (2/10). All cases had at least 1 prior failed fistula closure by alternative means and were evaluated after ADM placement for 6 months.²⁵ The 2 cases of graft loss occurred in the youngest patients who did not comply with oral care instructions. Most ADM graft sloughing and degeneration was observed between 3 and 4 weeks.²⁵

Similar to the current study, works by Shi et al²³ and Rhee et al²⁴ retrospectively investigated ADM outcomes in a mixed series of oral cavity wound cases. Shi et al²³ identified successful healing in 94.4% (34/36) of the cases. Inclusion criteria for wound size in the study were defects of at least 10 mm² for palate and gingiva sites (average graft size [AGS] used for defect coverage, 2.89 cm²; total procedures [TP], 27/36); 15 mm² for buccal, tongue dorsum, and lip sites (AGS, 6.03 cm²; TP, 7/36); and 20 mm² for tongue margin and vestibular groove sites (AGS, 6.5 cm²; TP, 2/36). The authors²³ noted contracture at the graft site during healing in 7 cases repaired by ADM occurring primarily in lip and buccal defect repair sites. The 2 unsuccessful grafts were both noted to have occurred in palatal repair sites. Rhee et al²⁴ found a 90% success rate (26/29) with ADM graft repair for mixed oral cavity wounds, with a total average wound size of 25 cm². Of the 3 unsuccessfully healed graft sites, 2 were on tongue repair sites (graft sizes, 1 cm x 3 cm; 4 cm x 5 cm) and 1 in a mixed retromolar trigone, tonsil, and pharynx repair site (graft size, 2 cm x 3 cm) in which postoperative hemorrhage occurred.²⁴ Graft contracture during healing was noted in 1 case where tumor recurrence was later documented. Key differences in these studies^23,24 from the one herein included the use of an alternative ADM product (AlloDerm extracellular matrix from human cadaver tissue;  LifeCell Corp, Branchburg, NJ), application of a bolster dressing for the first 37 days placed over the ADM for reinforcement postoperatively, and exclusion of patients who had undergone prior radiation therapy or had existing ORN. 

In this study, transient functional deficits were observed in 3 cases. In each instance, there was significant oncologic resection impacting either the tongue or pharynx, which made the outcomes unsurprising given already-known potential sequela associated with such procedures.^28,29 The lack of a screening tool with objective scoring for assessing speech and swallowing (eg, Water Swallow Test, Toronto Bedside Swallowing Screening Test) along with the absence of a control group prevents determination of any measurable difference of wound healing rate from recovery without UBM grafting. Yet the attainment of comprehensible speech and a return to normal diet in the setting of xenograft employment in each case remains noteworthy. 

In addition, this study had several wounds that were caused by radiation damage and managed with grafting. The use of UBM as a healing modality for recalcitrant radiation wounds involving nonmucosal surfaces of the sacrum and perineal region for soft tissue necrosis has been successfully demonstrated in a prior small retrospective case series performed by Rommer et al.¹¹ There does not appear to be a great deal of published work related to the use of extracellular matrix grafts for the treatment of radiation wounds in the present literature.¹¹ The limited data suggest UBM does not significantly influence long-term patient recovery from oral ORN despite some evidence of short-term healing/mucosal coverage; however, larger studies are needed to assess this finding. Work by Girod et al⁴ also identified ADM graft wound healing failure to be greater with previous exposure to radiation at more than 6 months prior to treatment; however, this study did not specify whether ORN also was present. In addition, only 20% (2/10) of ADM-treated patients previously exposed to radiation had wound healing failure.⁴ Poor wound healing is a known challenge for cases with prior exposure to radiation and not entirely unexpected.^30,31 The added presence of known ORN is likely to decrease the chances of successful repair, as observed in this study. 

Treatment of ORN depends on severity and may utilize conservative measures or require surgical management.^32,33 Conservative treatment for ORN involves avoidance of risk factors (ie, alcohol, tobacco, ill-fitting dentures), good oral hygiene (ie, prophylactic dental extractions, antiseptic solution rising), and frequent physician visits for the removal of sequestra.³² Hyperbaric oxygen therapy (HBOT) is another time-intensive option, requiring a series of 20 to 30 dives for 90 to 120 minutes, which may be considered as either a part of conservative therapy or with surgical resection.^32-34 At the medical center where this study was conducted, HBOT is typically not a reasonable option because of the highly rural demographic of patients concerned with time to reach the facility and travel distance. For that reason, HBOT was not used on any of the patients with ORN in this study.

The mouth presents an inherent population of natural flora, and situations of ORN also tend to involve bacteria. The antimicrobial cryptic peptides released and natural reabsorption of extracellular matrix products like UBM provide characteristics that make it ideally indicated for applications in already-infected wounds or when the risk for developing such sequela is high.^9,10,18,35 Yet, based on the limited results of this study, it does not appear that use of UBM conveys any noteworthy advantages over traditional treatment methods in these difficult-to-heal wounds. 

Scarring is an observational outcome that helps characterize the quality of mucosal tissue regeneration obtained. Since the potential benefit of using an extracellular matrix material is to promote a wound environment more favorable to regrowth of native tissue rather than default to scar tissue, this was considered an appropriate outcome for comparison. Based on comparison of this study’s results with those of others,^4,23-25 the use of a bolster to increase graft support in the early postoperative period may be useful to reduce scar tissue formation and increase wound healing success. 

It was noted in this study that scar tissue formation occurred more frequently when grafts were placed on tissue surfaces with a higher mobility, such as around the tongue and pharynx. Possible explanations for this are a size limitation to healing without scarring (since defect size tended to be larger in these repair sites) and that increased tissue mobility at the repair site results in loosening, which decreases graft contact with the underlying wound and prevents optimal ingrowth of native tissue. It is important to recognize that the increased presence of scarring at these sites did not translate to an increase in functional deficit or graft failure rate. Attention to placing the graft in well-vascularized wound surfaces also is important and exemplified in this study by the high failure rate observed in cases of ORN where tissue vascularity was poor and likely contributed to failed wound healing. 

When inspecting for short-term postop complications, the results of this study appear to indicate a low risk for concerning events such as hemorrhage and infection. This is consistent with the results of other studies.^23,24 In the single instance of hemorrhage developing 7 days after graft placement, a simple application of pressure to the wound site resolved the issue after graft removal. Also, the instance of oral thrush was attributed to the patient’s use of an oral steroid for inflammation discomfort and resolved readily with nystatin rinse and oral fluconazole. 

This study has several limitations important to consider in the interpretation of results. A major limitation is the lack of a control group with similar wounds that did not receive extracellular matrix application. The noncomparative, small sample size, short follow-up period, and retrospective nature of this study also are important limitations to consider. 

This study investigated the application of UBM for oral cavity reconstruction in a small cohort and found limited complications with acceptable wound healing and function. Based on the results of this and previous studies,^4,8,23,24,25 the use of ADM, UBM, and other ADM graft types appear to offer acceptable outcomes in the repair of full-thickness mucosal defects of the oral cavity. The incorporation of a bolster in the initial postoperative period may improve healing success rates. Increased graft size needed to cover a defect may be associated with increased likelihood of scar formation during healing. Application of UBM/ADM does not appear to convey any distinct advantages for use in patients with known ORN, although the small sample size in this study limits this conclusion.

The most equivalent reconstructive modality used in oral cavity repairs for ADM grafts is STSG. Limited comparison has demonstrated that ADM offers certain advantages over STSG by avoiding donor site morbidity, providing reduced operating time, and lowering procedure cost with noninferior cosmetic, functional, and patient satisfaction outcomes.⁴ Larger studies are needed to evaluate the relative efficacy of oral cavity reconstruction with UBM/ADM versus other modalities. Furthermore, there are no known studies comparing outcome differences among ADM graft types.^5,19 

Affiliation: Sanford Research; Department of Surgery, University of South Dakota Sanford School of Medicine, Sioux Falls, SD

Correspondence: William C. Spanos, MD, Assistant Professor, Department of Surgery, University of South Dakota Sanford School of Medicine, Sanford Ear, Nose, and Throat Clinic, 1310 West 22nd St, Sioux Falls, SD 57105; William.Spanos@SanfordHealth.org

Disclosure: The authors disclose no financial or other conflicts of interest.