British Journal of Oral and Maxillofacial Surgery 48 (2010) 131–135

Available online at www.sciencedirect.com

Repair of oronasal fistulas with human amnioticmembrane in minipigsMarco Rainer Kesting a,c,1, Denys John Loeffelbein a,c,!,1, Mariella Classen a,Julia Slotta-Huspenina b, Rafael Johannes Hasler a,c, Frank Jacobsen c, Kilian Kreutzer a,Sammy Al-Benna c, Klaus-Dietrich Wolff a, Lars Steinstraesser c

a Department of Oral and Maxillofacial Surgery, Technische Universität München, Ismaninger Str. 22, D-81675 München, Germanyb Department of Pathology, Technische Universität München, Ismaninger Str. 22, D-81675 München, Germanyc Department of Plastic Surgery, Burn Unit, BG-Universitätsklinikum Bergmannsheil, Ruhr-Universität Bochum, Bürkle-de-la-Camp-Platz 1, D-44789Bochum, Germany

Accepted 23 April 2009Available online 23 May 2009

Abstract

We evaluated the use of multilayer human amniotic membrane (HAM) as a grafting material for the repair of mid-palate oronasal fistulas inseven Berlin minipigs. After two weeks, three animals had the fistulas repaired with multilayered HAM grafts, three had them repaired witha collagen-based dermal substitute (INTEGRA®, Integra Life Sciences, Plainsboro, NJ, USA), and one fistula was left untreated to serve as acontrol. Grafts were interposed between the oral and nasal mucosa, traversing the fistulas. After healing for 40 days, the pigs were killed forclinical, histological, and immunohistochemical examination.

Two of the three fistulas closed with HAM were successful, the diameter of the third was reduced in size, and there was no change in thediameter of the fistula in the control.

This study shows successful closure of oronasal fistulas in minipigs using interposed grafts of cryopreserved HAM, and offers promise asa simple and effective technique for tension-free closure of such fistulas.© 2009 The British Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.

Keywords: Cleft lip and palate; Oronasal fistula repair; Human amniotic membrane

Introduction

The formation of an oronasal fistula is a serious compli-cation of the primary repair of cleft palates. Its incidencevaries widely, ranging from 0% to 68% in published reports.1

Many causes have been proposed for the formation of a fis-tula after repair of a cleft palate, including paucity of localtissue for closure, and excessive scarring.1–5 Various tech-

! Corresponding author at: Department of Oral and Maxillofacial Surgery,Technische Universität München, Ismaninger Str. 22, D-81675 München,Germany. Tel.: +49 89 4140 2929; fax: +49 89 4140 4993.

E-mail address: loeffelbein@mkg.med.tum.de (D.J. Loeffelbein).1 Both authors contributed equally to this work.

niques for repair have been suggested including local2 andfree flaps,3 tissue expansion,4 and the use of allogenic tissuesand biomaterials.1,5

Fresh human amniotic membrane (HAM) has been usedfor nearly a century in reconstructive surgery,6 but its usein developed countries has been limited by the risk ofcross-infection. The introduction of a method for glycerol-cryopreservation allowed donors and tissue to be retestedafter the “infection window”, and HAM transplantationbecame a regular procedure in ophthalmological operations.7

Ophthalmological studies advocated the use of the tissue forits low immunogenicity, minimal inflammation and scarring,and enhancement of epithelialisation.8 Its success in ophthal-mology as an adjunct in wound healing encouraged its usein extraocular, immunogenic tissue. Oral and maxillofacial

0266-4356/$ – see front matter © 2009 The British Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.doi:10.1016/j.bjoms.2009.04.025

132 M.R. Kesting et al. / British Journal of Oral and Maxillofacial Surgery 48 (2010) 131–135

surgeons have described its use for gingival wound healing,9

intraoral lining in vestibuloplasty,10,11 and in the prefabri-cation of flaps.12 To overcome shortcomings in mechanicalstability, Kruse et al. proposed a multilayered applicationfor deep corneal defects.13 In vivo studies have shown thatdefects in the abdominal wall can be closed successfully withcryopreserved multilayered HAM.14 We evaluated the use ofmultilayered HAM as a grafting material for the repair oforonasal fistulas in minipigs.

Materials and methods

We used seven male six-month-old Berlin minipigs (Fa.Schlesier, Großerkmannsdorf, Germany), weighing 22–31 kg(mean 25.8) each. They were kept according to the interna-tional principles of laboratory animal care. Water and foodwere given freely. The study was approved by the local ethicscommittee and the local government.

Human placentas were obtained under sterile conditionsfrom planned, uneventful Caesarean sections. All women hadgiven written informed consent. A 5 cm " 5 cm piece wasprepared from each placenta and stored according to a fixedprotocol.14 Thirty minutes before operation the HAM wasthawed, washed with sterile phosphate buffered saline andsecured in place with polyglactin 910 (Vicryl® 6-0, Ethicon,Norderstedt, Germany) into five layers before application(Fig. 1).14

Induction of anaesthesia, intubation, perioperative moni-toring, and analgesia were done as described previously.15

Pigs were killed 40 days after closure of the oronasal fis-tulas by T61® (Bayer, Leverkusen, Germany) 1 ml/5 kg bodyweight intravenously.

We made a full-thickness defect in the mid-hard palate(15 mm in diameter) with a 15 mm stainless steel biopsypunch, and subsequent osteotomy of the hard palate. Thenasal septum and edges of the bony palate were then

Fig. 1. Multilayered human amniotic membrane graft before insertion(haematoxylin and eosin, original magnification "60, scale bar 500 !m;detailed view original magnification "300, scale bar 50 !m).

Fig. 2. Oronasal fistula 14 days after creation.

rongeured 2 mm beyond the mucosal edges to allow for re-epithelialisation at the periphery of the defects. Fistulas wereallowed to mature for two weeks (Fig. 2).

Three of the seven animals were selected randomly to havethe fistulas repaired with grafts of multilayered HAM or acollagen-based dermal substitute (INTEGRA® RegenerationTemplate, INTEGRA Life Sciences, Plainsboro, NJ, USA).Multilayered HAM was prepared as previously described.14

INTEGRA® dermal substitute is a bilaminar membranesystem consisting of a “dermal” component of porouscoprecipitate of bovine tendon type I collagen and shark gly-cosaminoglycan (chondroitin-6-sulfate), and an “epidermal”silicone layer.

Grafts were interposed between the oral and nasal mucosatraversing the palatal fistulas, and sutured with polyglactinsuture (Vicryl® 4-0, Ethicon, Norderstedt, Germany) to thelateral edges of the mucoperiostal bilateral flap createdthrough “von Langenbeck” incisions. Mucosal edges of thefistula were not closed. One pig was not treated after creationof the defect to serve as a control.

Alginate dental casts of the pigs’ palates were taken at thetime the fistulas were made. Protective plates were manufac-tured individually and fixed with three sinus-lift screws afterreconstruction of the fistula.

All wounds were monitored morphologically by photo-graph on postoperative days 3 and 6 under short sedation.They were cleaned of blood clots and food pellets afterremoval of the palatal appliance, which was fixed again untilit was removed completely at day 9 when the width of the fis-tulas were measured. On day 40 the pigs were killed and thepalates examined. The complete hard palate was processedfor microscopic tissue analysis.

Paraffin-embedded specimens were cut in 4 !m sectionsand stained with haematoxylin and eosin, and Elastica vanGieson to visualise and characterise the tissue architecture.

M.R. Kesting et al. / British Journal of Oral and Maxillofacial Surgery 48 (2010) 131–135 133

Immunohistochemical analysis was done on sections usingan automated immunostainer (Ventana Medical systems Inc.,Tucson, Arizona, USA) according to the manufacturer’sinstructions, with minor modifications.16 The antibodies usedincluded factor VIII (Dako, Glostrup, Denmark), "-humanchorionic gonadotropin (Dako) and pan-cytokeratin (Dako).Positive controls for all antibodies investigated were used toconfirm the adequacy of the staining.

Pictures were taken using the Axioskop 2 plus® (CarlZeiss Light Microscopy, Göttingen, Germany) connected toan AxioCam HRc® (Carl Zeiss MicroImaging, Göttingen,Germany) at magnifications of 5, 10, and 20 times.

Results

The fistula of the untreated control had epithelialised woundmargins, and it had decreased to 13 mm. Two of the threeHAM-covered fistulas remained closed with no visual evi-dence of a remaining fistular tract (Fig. 3). The diameter ofthe third HAM-covered fistula had been 3 mm on day 9 and6 mm on day 40. All the fistulas treated with INTEGRA®

dehisced and had a significantly larger diameter than thosetreated with HAM on day 40 (p = 0.043) (Table 1). Interest-ingly, all wounds treated with INTEGRA® that were infectedon day 9 had resolved by day 40. No wounds closed usingHAM were infected.

Staining with haematoxylin and eosin showed that theclosed fistulas had re-epithelialised totally in both the oraland nasal sites (Fig. 4). Staining for factor VIII showed thatendothelial cells and ingrowth of vessels were present in thesites grafted with HAM (Fig. 5). Numerous fibroblasts andbundles of matured collagen were seen in the scars whenstained by Elastica van Gieson (Fig. 6). "-human chorionic

Fig. 3. Postoperative view 40 days after closure with human amniotic mem-brane. Complete healing was seen.

Table 1Diameter of fistulas (mm).

Animal Graft material Size offistula

Fistula beforeclosure

After closure

Day 9 Day 40

1 Integra® 16 14 6 112 Integra® 15 15 8 123 HAM 15 13 0 04 Integra® 17 15 10 125 HAM 16 15 3 66 HAM 15 14 0 07 Control (no

closure)15 14 13 13

Fig. 4. Overview on the oronasal complex 40 days after reconstruction ofthe fistula with human amniotic membrane. Section with palatal and nasalsurfaces indicates complete closure (haematoxylin and eosin, original mag-nification "1.5, scale bar 4 cm).

Fig. 5. Staining with factor VIII antibody shows dispersed immunoreac-tions indicating endothelial structures (original magnification "5, scale bar1000 !m).

134 M.R. Kesting et al. / British Journal of Oral and Maxillofacial Surgery 48 (2010) 131–135

Fig. 6. Staining with Elastica van Gieson shows sufficient epithelial clo-sure of the oral surface with regularly arranged epithelial cells (originalmagnification "5, scale bar 1000 !m).

Fig. 7. No stain with "-human chorionic gonadotropin antibody suggestscomplete resorption of the fetal tissue graft (original magnification "5, scalebar 1000 !m).

gonadotropin did not stain, which suggested complete degra-dation and remodelling of the HAM grafts (Fig. 7).

The three fistulas that had been closed using INTEGRA®

and had dehisced were characterised by a pronounced inva-sion of lymphocytes.

Discussion

Formation of scars and a shortage of tissue complicate therepair of oronasal fistulas. Current methods for their clo-sure can be broadly divided into four groups that use localmucoperiostal flaps, pedicled flaps such as from the tongue,free flaps such as from the forearm, and biomaterials. Donorsite morbidity can be substantial so biomaterials seem ideal,but until now none have fulfilled the complex requirementsof biocompatibility and integration.

HAM and its extracellular matrix contain componentssuch as growth factors17 and antimicrobial peptides18 thatsuggest that it is an excellent candidate for use as a nativescaffold. Unlike other scaffolds it is cheap, and is easy toobtain and to process.19

This study shows that oronasal fistulas in pigs can beclosed successfully using interposed grafts of cryopreservedmultilayered HAM. The intraoral use of fresh monolayeredHAM has been reported for mandibular vestibuloplasty,10,11

but it is not resilient and was the source of leakage in recon-structed bladders in rats that led to a mortality rate of 42%.20

Mermet et al. described a limited benefit of monolayeredmembranes for the treatment of leg ulcers because of rapiddegradation.21 These problems were solved in our earlierstudy by the successful application of multilayered HAMto repair defects in the abdominal wall in rats.14

The clinical course of fistulas reconstructed using HAMwas better than that of the control that was closed usingINTEGRA®. Unlike INTEGRA®, which has been proposedas an implant for palatal wound healing,22,23 the pigs treatedwith HAM healed uneventfully. Two of three pigs had fistulasclosed successfully, and there was a considerable reduc-tion in the diameter of the fistula in the remaining pig40 days after operation. Histologically, the closed fistu-las consisted of matured tissue with collagen bundles andhad no remnants of amnion or an appreciable number oflymphocytes. All animals in the INTEGRA® group hadinflammation and larger fistulas. Ophof et al. reported com-plete re-epithelialisation and integration of tissue four weeksafter reconstruction of palatal full-thickness wounds in bea-gle dogs with INTEGRA®,22,23 but later studies from thisgroup showed that it did not improve the healing of palatalwounds as revascularisation was too slow to allow survivaland integration of INTEGRA® so they dispensed with its usefor palatal reconstruction.24

Kirschner et al. initiated this study’s pig model for theclosure of oronasal fistulas. They used an acellular dermalmatrix (AlloDerm®, LifeCell Corporation, New Jersey, USA)to repair fistulas, and achieved excellent results in piglets andpatients.1 Smoothness and pliability were similar to that oforal mucosa, but we used HAM as it is cheaper and widelyavailable, although it is more vulnerable than acellular dermalmatrices. Fabrication of a protective plate was essential toavoid mechanical alteration and inflammation that might becaused by food on the palatal folds.

HAM is currently used for reconstruction of the ocularsurface8 and abdominal wall,14 for bladder repair,20 treat-ment of leg ulcers,21 and burns.25 It is a useful reconstructiveoption for the repair of oronasal fistulas and as a surgicalpatch. Its use in the repair of cleft palates offers new opportu-nities for its application. With the implementation of prenataldiagnosis it will be possible to harvest and store the amnioticmembrane of a child with a cleft as an exclusive tissue sourcefor later reconstruction.

Acknowledgements

We thank Andrea Rittig for technical assistance. This workwas supported by the medical faculty of the Ruhr UniversityBochum (FoRUM, F321/01).

M.R. Kesting et al. / British Journal of Oral and Maxillofacial Surgery 48 (2010) 131–135 135

References

1. Kirschner RE, Cabiling DS, Slemp AE, Siddiqi F, LaRossa DD, LoseeJE. Repair of oronasal fistulae with acellular dermal matrices. PlastReconstr Surg 2006;118:1431–40.

2. Honnebier MB, Johnson DS, Parsa AA, Dorian A, Parsa FD. Closure ofpalatal fistula with a local mucoperiosteal flap lined with buccal mucosalgraft. Cleft Palate Craniofac J 2000;37:127–9.

3. Eufinger H, Machtens E. Microsurgical tissue transfer for rehabilita-tion of the patient with cleft lip and palate. Cleft Palate Craniofac J2002;39:560–7.

4. Kobus KF. Cleft palate repair with the use of osmotic expanders: apreliminary report. J Plast Reconstr Aesthet Surg 2007;60:414–21.

5. Clark JM, Saffold SH, Israel JM. Decellularized dermal grafting in cleftpalate repair. Arch Facial Plast Surg 2003;5:40–5.

6. Davis J. Skin transplantation with a review of 550 cases at the JohnsHopkins Hospital. Johns Hopkins Hospital Report 1910;15:310.

7. Lee SH, Tseng SC. Amniotic membrane transplantation for persis-tent epithelial defects with ulceration. Am J Ophthalmol 1997;123:303–12.

8. Dua HS, Gomes JA, King AJ, Maharajan VS. The amniotic membranein ophthalmology. Surv Ophthalmol 2004;49:51–77.

9. Rinastiti M, Harijadi, Santoso AL, Sosroseno W. Histological evalua-tion of rabbit gingival wound healing transplanted with human amnioticmembrane. Int J Oral Maxillofac Surg 2006;35:247–51.

10. Samandari MH, Yaghmaei M, Ejlali M, Moshref M, Saffar AS.Use of amnion as a graft material in vestibuloplasty: a preliminaryreport. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2004;97:574–8.

11. Güler R, Ercan MT, Ulutuncel N, Devrim H, Uran N. Measurement ofblood flow by the 133Xe clearance technique to grafts of amnion usedin vestibuloplasty. Br J Oral Maxillofac Surg 1997;35:280–3.

12. Ahn KM, Lee JH, Hwang SJ, et al. Fabrication of myomucosalflap using tissue-engineered bioartificial mucosa constructed withoral keratinocytes cultured on amniotic membrane. Artif Organs2006;30:411–23.

13. Kruse FE, Rohrschneider K, Volcker HE. Multilayer amniotic mem-brane transplantation for reconstruction of deep corneal ulcers.Ophthalmology 1999;106:1504–11.

14. Kesting MR, Loeffelbein DJ, Steinstraesser L, et al. Cryopreservedhuman amniotic membrane for soft tissue repair in rats. Ann Plast Surg2008;60:684–91.

15. Druecke D, Lamme EN, Hermann S, et al. Modulation of scar tis-sue formation using different dermal regeneration templates in thetreatment of experimental full-thickness wounds. Wound Repair Regen2004;12:518–27.

16. Quintanilla-Martinez L, Kremer M, Specht K, et al. Analysis of signaltransducer and activator of transcription 3 (Stat 3) pathway in multiplemyeloma: Stat 3 activation and cyclin D1 dysregulation are mutuallyexclusive events. Am J Pathol 2003;162:1449–61.

17. Koizumi NJ, Inatomi TJ, Sotozono CJ, Fullwood NJ, Quantock AJ,Kinoshita S. Growth factor mRNA and protein in preserved humanamniotic membrane. Curr Eye Res 2000;20:173–7.

18. Buhimschi IA, Jabr M, Buhimschi CS, Petkova AP, Weiner CP, SaedGM. The novel antimicrobial peptide beta3-defensin is produced by theamnion: a possible role of the fetal membranes in innate immunity ofthe amniotic cavity. Am J Obstet Gynecol 2004;191:1678–87.

19. Niknejad H, Peirovi H, Jorjani M, Ahmadiani A, Ghanavi J, SeifalianAM. Properties of the amniotic membrane for potential use in tissueengineering. Eur Cell Mater 2008;15:88–99.

20. Iijima K, Igawa Y, Imamura T, et al. Transplantation of preservedhuman amniotic membrane for bladder augmentation in rats. Tissue Eng2007;13:513–24.

21. Mermet I, Pottier N, Sainthillier JM, et al. Use of amniotic membranetransplantation in the treatment of venous leg ulcers. Wound RepairRegen 2007;15:459–64.

22. Ophof R, Maltha JC, Kuijpers-Jagtman AM, Von Den Hoff JW. Eval-uation of a collagen-glycosaminoglycan dermal substitute in the dogpalate. Tissue Eng 2007;13:2689–98.

23. Ophof R, Maltha JC, Von den Hoff JW, Kuijpers-Jagtman AM. Histo-logic evaluation of skin-derived and collagen-based substrates implantedin palatal wounds. Wound Repair Regen 2004;12:528–38.

24. Ophof R, Maltha JC, Kuijpers-Jagtman AM, Von den Hoff JW. Implan-tation of tissue-engineered mucosal substitutes in the dog palate. Eur JOrthod 2008;30:1–9.

25. Kesting MR, Wolff KD, Hohlweg-Majert B, Steinstraesser L. The roleof allogenic amniotic membrane in burn treatment. J Burn Care Res2008;29:907–16.