Closure of an Oronasal Fistula in anIrradiated Palate by Tissue and BoneDistraction OsteogenesisPeter J. Taub, MD*James P. Bradley, MD*†

Henry K. Kawamoto, MD, DDS*

Los Angeles, CaliforniaPittsburgh, Pennsylvania

Uses for distraction osteogenesis in the craniofacialskeleton have expanded during the last decade. Ithas become an important rung in the reconstructiveladder for correction of difficult defects. Distrac-tion of irradiated bone has been successfully per-formed in an animal model but has not been re-ported in human subjects. We present a case ofdistraction osteogenesis in a patient with multiplefailed reconstructive attempts to close an irradiatedpalatal defect. An additional benefit included im-provement in support of the upper lip from bonetransported and the potential for placing dental im-plants.

Key Words: Oronasal fistula, palate, irradiation, dis-traction osteogenesis

Applications of distraction osteogenesis(DOG) in the craniofacial skeleton havegrown in the last decade.1 DOG has beenused 1) to lengthen the mandible in pa-

tients with craniofacial microsomia and microgna-thia, 2) to advance the maxilla after Le Fort I osteot-omy, 3) to advance the midface and/or foreheadafter Le Fort III or monobloc osteotomy, and 4) toexpand the orbits.2–5 The principles of DOG havealso been used to correct mandibular defects afterablative surgery or trauma with bone transport.6

DOG has been attempted to lengthen the hardpalate for velopharyngeal incompetence and has

been successfully performed in the canine palate.7–9

It has not been clinically used in irradiated bone,perhaps because of poor vascularity and poorwound healing. However, DOG of an irradiated ca-nine mandible showed stable bone healing after aconsolidation phase.10

In this article, we present a patient with a largeoronasal fistula after maxillary tumor resection andirradiation. Failure of several procedures using localmucosal turn-in flaps and pedicled temporalis myo-osseous flaps led to a persistent oronasal fistula.When the patient refused to undergo a radial fore-arm microvascular free flap, we offered DOG of theanterior segment of the remaining palate to close thedefect. The purpose of this article is to show thatDOG is a viable method of oronasal fistula closureand reconstruction of difficult defects in an irradi-ated palate.

CASE REPORT

A53-year-old man was seen with a large oronasalfistula in an irradiated palate after multiple

failed attempts at closure. He had previously under-gone an anterior maxillary resection and postopera-tive radiation (4,500 rad) for a squamous cell cancer5 years before. Immediate reconstruction consisted ofonly soft-tissue closure and resulted in deficiency ofthe alveolar ridge, bony hard palate, and caudalbony septum of the nose. Subsequently, he devel-oped a large oronasal fistula that also included theright maxillary sinus. Attempted closures of the fis-tula with local mucosal flaps and an iliac bone graftfailed. A temporalis myofascial osseous turnoverflap containing a “L”-shaped calvarial bone (outertable) was then tunneled to reach the intraoral defect.Postoperatively, this resulted in soft-tissue slough-ing, extrusion of bone, complete resorption of thebone, and a recurrent large fistula. The patient was

From the *Division of Plastic and Reconstructive Surgery, Uni-versity of California, Los Angeles, California; and the †Division ofPlastic and Reconstructive Surgery, University of Pittsburgh, Pitts-burgh, Pennsylvania.

Address correspondence to Dr Kawamoto, 1301 20th Street#460, Santa Monica, CA 90404.

495

then offered a radial forearm free flap to close thedefect.

When we initially examined the patient, he hadan appearance of midfacial collapse because of theabsence of the anterior alveolar ridge and a relativelyprominent lower lip (Fig 1). Intraorally, two rightmolars remained. A 1.5 × 2.0-cm oronasal fistula ex-isted where the left canine and premolars would nor-mally have been (Fig 2). The patient’s computed to-mography scan confirmed the lack of left maxillarybone and anterior half of the palate.

DOG was undertaken to transport the remain-ing palatal bone to the left and toward the anteriordefect (Fig 3). The procedure was begun with twoincisions: one just anterior to the remaining right mo-lars and the other at the posterior aspect of the oro-nasal fistula. A tunnel was created between theseincisions to accommodate a reciprocating saw thatwas used for the osteotomy. The vomer was de-tached from this segment to mobilize it, but the pala-tal mucosa was left intact to optimize blood supply.Next, the piriform rim was exposed to allow for anasal septal osteotomy and mobilization of the pre-maxilla. A distraction device (Howmedica-LeibingerInc., Dallas, TX) was then fixed with 14-mm screwsto the mobilized anterior palatal segment and to theremaining palatal bone posteriorly.

Postoperatively, after 48 hours, the patient be-gan 1 mm per day distraction for 14 days. The palataldefect was obliterated in 9 days (Fig 4). Additionaldistraction added bone to the alveolar ridge and thusprovided maxillary projection and improved lip sup-port. The distraction device was removed after 4

Fig 1. Photograph of the patient with a large oronasalfistula in an irradiated palate after multiple failed attemptsat closure. Note the appearance of midface hypoplasia andthe prominent lower lip from the loss of bone and softtissue in the anterior maxilla.

Fig 2. Intraoral photograph of the pa-tient demonstrating a large oronasalfistula in the anterior portion of his ir-radiated palate.

THE JOURNAL OF CRANIOFACIAL SURGERY / VOLUME 12, NUMBER 5 September 2001

496

months. At the time of this procedure, mucosal flapswere elevated in the site of the fistula to definitivelyseparate the oral and nasal mucosa and obliterate thecommunication. Posttreatment radiographs revealednew bone formation. The patient is being consideredfor anterior alveolar implants.

DISCUSSION

The technique of osseous distraction was first re-ported at the turn of the century when lengthening ofthe long bones of the lower extremity was describedby Codivilla.11 These initial attempts, however, hadan unacceptable rate of complication, including in-fection at the pin sites and skin necrosis beneath thedistraction device. These problems were addressedby Ilizarov who contributed tremendously to betterunderstanding of the principles of DOG. He showedthat corticotomy alone with preservation of the peri-osteum and intramedullary elements yielded signifi-cantly improved results.12

Early distraction in the craniofacial skeleton wasreported by Snyder et al. who, in 1973, showed thatmandibular length could be successfully increased ina canine model using an extraoral device.13 A modelof severe crossbite was created by removing a 1.5-cmsegment of mandible and allowing the bone to healwith collapse of the ipsilateral mandible. A completeosteotomy was then performed, and the mandiblewas successfully distracted during a period of 14days.

McCarthy et al. opened a new era by document-ing the successful distraction of the human mandiblein four subjects using a similar device.2 They showedthat the membranous bones of the craniofacial skel-eton were much more suited to surgical distractionthan Illizarov’s endochondral bones of extremities.This was attributed to the smaller size, better bloodsupply, and ease of surgical manipulation. Theirlaboratory also elucidated the process of repair in thedistracted bone. A longitudinal scaffold of collagenwas produced and subsequent ossification pro-

Fig 3. Intraoral illustrations including the operative technique of distraction osteogenesis (DOG) device placement andthe result of full distraction. (A) A midpalatal osteotomy is performed through two small lateral mucosal incisions witha reciprocating saw. The DOG device is secured in a collapsed, undistracted position. (B) After full distraction of 14 mm,the DOG device is seen in distracted position with intervening regenerate bone and an anteriorly displaced maxilla.

CLOSURE OF ORONASAL FISTULA / Taub et al

497

ceeded from the free edges of the mandible.14 Also,four zones of membranous reossification in the hu-man mandible could be recognized: 1) a fibrous, cen-tral zone, 2) a transition zone, 3) a remodeling zone,and 4) a peripheral, mature bone zone.

Carls et al. introduced DOG of the palate in acanine cleft palate model.7 DOG was used to

lengthen the palate 7 to 10 mm caudally with in-traoral devices attached to the dentition anteriorlyand palate posteriorly. Ascherman et al. performedpalatal distraction on five dogs fitted with a central,submucosal jackscrew apparatus and two anteriorand two posterior footplates.9 They believe that dis-traction of the hard palate might be a viable option in

Fig 4. Intraoral photo-graph of the patient afterdistraction osteogenesis(DOG) was complete. (A)Distraction device shownafter 14 mm of palatallengthening. Although thefistula appeared closed at 9days , addi t ional DOGadded bone to the alveolarridge region for improvedmaxillary projection. (B)The oronasal fistula hasbeen obliterated.

THE JOURNAL OF CRANIOFACIAL SURGERY / VOLUME 12, NUMBER 5 September 2001

498

treating patients with velopharyngeal incompetence.Also, Molina and Jacobo have reported a small seriesof patients with improvements in velopharyngeal in-competence after palatal distraction.8

Although healthy bone has been shown to besuccessfully manipulated by distraction, this processhas not been recommended to reconstruct defects ofthe head and neck after ablative radiation therapy.However, there have been attempted cases (J. G. Mc-Carthy and S. U. Stucki-McCormick, personal com-munication, April 2000). Irradiated tissues have beenshown to be relatively hypovascular and hypoxiccompared with nonirradiated tissues.15 Various tech-niques have been used by those in the field of radia-tion oncology to minimize the effects of harmful ra-diation. One of these is fractionation, in which thetotal desired dose is split, or fractionated, over afixed number of sessions. Gantous et al. showed thatdistraction of the craniofacial skeleton followed bystable healing is possible in an irradiated caninemandible.10 They subjected five dogs to a full courseof radiation therapy designed to treat human squa-mous cell cancer of the oral cavity. Each animal re-ceived 50 Gy of radiation in 20 fractions during aperiod of 4 weeks. After 6 months, 20 mm of man-dibular bone was removed and a transport fragmentof mandibular bone was distracted anteriorly fromthe proximal mandibular segment. Four of the fiveanimals demonstrated complete bony regenerationof the mandible. Bone within the distraction site wasof comparable size and histologically similar to thatof the remaining bone.

In the patient that we presented, the palataldefect resisted several attempts at closure. To bringin vascularized tissue into an irradiated area, hewas offered a radial forearm flap. However, wethought that DOG was a simpler method to closethe fistula and rebuild the anterior alveolus. Benefitsof DOG include bringing vascularized soft tissueand bone to the defect without the distal donor sitemorbidity.

In summary, we presented a patient in whompalatal distraction was successfully used to close anoronasal fistula in an irradiated palate. The case

highlights the importance of distraction osteogenesisas a therapeutic option in 1) less common areas of thecraniofacial skeleton and 2) tissues damaged by theeffects of radiation.

REFERENCES

1. Tessier P. Foreword. In: Joseph G. McCarthy (ed). Distractionof the Craniofacial Skeleton. New York: Springer-Verlag, 1999

2. McCarthy JG, Schreiber J, Karp NS, et al. Lengthening of thehuman mandible by gradual distraction. Plast Reconstr Surg1992;89:1–8

3. Chin M, Toth BA. Le Fort III advancement with gradual dis-traction using internal devices. Plast Reconstr Surg 1997;100:819–830

4. Cohen SR, Boydston W, Burstein FD, et al. Monobloc distrac-tion osteogenesis during infancy. Report of a case and presen-tation of a new device. Plast Reconstr Surg 1998;101:1919–1924

5. Lo AK, Colcleugh RG, Allen L, et al. The role of tissue ex-panders in an anophthalmic animal model. Plast ReconstrSurg 1990;86:399–408

6. Stucki-McCormick SU. Reconstruction of the mandibularcondyle using transport distraction. J Craniofac Surg 1997;8:48–52

7. Carls FR, Jackson IT, Topf JS. Distraction osteogenesis forlengthening of the hard palate: Part I. A possible new treat-ment concept for velopharyngeal incompetence. Experimentalstudy in dogs. Plast Reconstr Surg 1997;100:1635–1647

8. Molina F, Jacobo F. Distraction of the bone palate and its ef-fects on velopharyngeal competence. Presented at the AnnualMeeting of the American Society of Plastic and ReconstructiveSurgeons, Boston, Massachusetts, October 1998

9. Ascherman JA, Marin VP, Rogers L, et al. Palatal distraction ina canine cleft palate model. Plast Reconstr Surg 2000;105:1687–1694

10. Gantous A, Phillips JH, Catton P, et al. Distraction osteogen-esis in the irradiated canine mandible. Plast Reconstr Surg1994;93:164–168

11. Codivilla A. On the means of lengthening in the lower limbs,the muscles, and tissues which are shortened through the de-formity. Am J Orthop Surg 1905;2:353–369

12. Illizarov GA, Devyatov AA, Kamerin VK. Plastic and recon-struction of longitudinal bone defects by means of compres-sion and subsequent distraction. Acta Chir Plast 1980;22:32–41

13. Snyder CC, Levine GA, Swanson HM, et al. Mandibularlengthening by gradual distraction: Preliminary report. PlastReconstr Surg 1973;51:506–508

14. Karp NS, McCarthy JG, Schreiber JS, et al. Mandibular bonelengthening: A serial histologic study. Ann Plast Surg 1992;29:2–7

15. Marx RE. Osteoradionecrosis: A new concept of its patho-physiology. J Oral Maxillofac Surg 1983;41:283–288

CLOSURE OF ORONASAL FISTULA / Taub et al

499