Resorbable Distraction of the Mandible: Technical ... · used in orthopedic applications as pure polydiox-anone, pure polyglycolicacid (PGA),and pure poly ... on mandibular distraction

1

widespread clinicalapplication is relatively recent.l "Initially, it was recognized that polymers can befashioned into medical devices that would eventuallybe resorbed by the body; however, the exactmechan-ism was unclear. Initially, pure homopolymers wereused in orthopedic applications as pure polydiox-anone, pure polyglycolic acid (PGA), and pure polyt-lactic acid. In a few early reports, copolymers suchas PGA-pure poly i-lactic acid were used. Mostcomplications in these early studies were reportedwith homopolymers." These complications includedfibrous encapsulation, sterile sinus formation, and anintense inflammatory response. Fewer problems werenoted with the use of the copolymer materials. Theseearly resorbable polymers enjoyed some popularity,but because of the extreme length of time untilresorption, the amount of inflammation caused bythese polymers, and questions regarding theirmechanical stability, they were never widely used. Inthe late 1980s and early 1990s, interest was againstirred in resorbablematerials with the realization thatmixing 2 different types of polymers would result inmarkedly different resorption characteristics depend-ing on the proportions of each polymer," Thesepolymer macromolecules,when properly formulated,can serve a structural function and are then graduallydegraded and resorbed by the body.' A plethora ofresorbable polymers were subsequently formulatedfor use in medical devices. The most extensivelystudied is the Lacto Sorb (Biomet Corporation,Indianapolis, IN),which is a random linear copolymerconsisting of 82%poly t-lactide and 18%PGA. Thiscompound resorbs during 12 months and losesstrength during 3 to 4 months. It has intermediateresorption characteristics, which allow gradualresorption through hydrolysis and phagocytosis'

Until the introduction of these new copolymers,titanium plates and screws of various sizes had beenused even in infants and children. This led to concernby many practitioners about plates and screws beingcarried through the bone and into contact with thedura and brain through the process of absorptionand bony deposition. In 1992,the Lacto Sorb systemof plates and screws was introduced for clinical

Resorbable Distraction of the Mandible:Technical Evolution and Clinical ExperienceFernando D. Burstein MD, FACS, FAAP

Atlanta, Georgia, USA

The author has used various designs of single-stageresorb able distractors to lengthen the mandible in100patients aged 7 days to 16years (mean, 4.24Yi 49girls, 51boys) with predictable results and minimalmorbidity since 2002.The range of distraction was15 to 30 mm (mean, 25.4 mm). Indications forsurgery included mandibular hypoplasia associatedwith Pierre Robin sequence, craniofacial microso-mia, Treacher Collins, and Nagers syndrome. Thehistory of the development of resorbable polymersfor use in craniofacial surgery and the evolution ofdistraction osteogenesis are reviewed. The meldingof these 2 innovative technologies has led to thedevelopment of a new class of single-stage resorb-able devices. These devices are quite different fromthe titanium distraction devices that have been usedin clinical practice for more than 15 years. Throughcontinued clinical application, the surgical metho-dology for resorbable distraction has been refinedand simplified, although still-evolving continuedexperience with resorbable distraction has greatlydecreased operative time and improved results.

Key Words: Mandible distraction, resorbable, man-dibular hypoplasia

The parallel development of resorbable sur-gical materials and the application of dis-traction osteogenesis to the craniofacialskeleton have led to the melding of these

technologies in the development of single-stageresorbable distractors. Interest in bioabsorbable poly-mer materials has existed for more than 30 years, but

From the Emory University Plastic Surgery Division, Center forCraniofacial Disorders, Children's Healthcare of Atlanta, Atlanta,Georgia.

The author received support from Biomet Corporation inpreparing these studies.

Address correspondence and reprint requests to Fernando D.Burstein, MD, FACS, FAAP, Suite 500, 975, Johnson Ferry Rd.,Atlanta, GA 30342;E-mail: [email protected]

637

on mandibular distraction in neonates and infantswith Pierre Robin sequence in 2005.16 Since then, ourexperience has grown with application of variousdesigns of mandibular distractors in 100 patients over6 years. This experience has lead to the recognitionthat successful application of resorbable distractionrequires technical modifications from the techniquesused in the application ofmetallic distraction.

THE JOURNAL OF CRANIOFACIAL SURGERY I VOLUME 19, NUMBER 3 May 2008

use, followed later by systems from several manu-facturers, each having unique handling and resorp-tion characteristics. In comparison to their titanium,these plates were significantly thicker because ofstrength considerations. Initially, plates were fash-ioned in the same shapes and sizes as titaniumplates. As the unique properties of the resorbablematerials were discovered, designs that took advan-tage of these characteristicswere customized for spe-cific clinical applications.

Paralleling the application of resorbable plate/screw fixation was the emergence of distraction incraniofacial surgery. The pioneering work establish-ing the principles of distraction osteogenesis wasdone by Dr. Ilizarov" in Russia and was published in1954. In 1973, Snyder et al reported on their workusing gradual distraction for mandibular lengthen-ing? Karp et al" published their article on distractionosteogenesis in the craniofacial skeleton in 1990.McCarthy et a19

,10 further defined both the scientificbasis and clinical principles that set the stage fordistraction of the craniofacial skeleton. This impor-tant work stimulated many others to apply thisexciting technique. Initially, mandibular distractiondevices were all external, modeled after orthopedicdevices used for hand and wrist trauma.ll,12Although these devices offered excellent mechanicalstrength, there were also problems, includingunsightly pin site scars, complications from pin siteinfections, dislodgement of devices, and other diffi-culties, which led surgeons to design internal man-dibular devices.l+P These metallic devices wereinserted through either an oral or external approachand gave excellent mechanical strength to the dis-traction. The major drawback to these internaldevices was that they had to be removed at the endof the distraction. Often, bone would tend to over-grow the device, making removal quite difficult.

In 2000, Cohen et al13 published his initial workusing a maxillary distraction device made of aresorbable polymer. This device required a secondstage for stabilization. The author, building on Dr.Cohen's experience, as well as on his own work withdistraction,began working on a single-stageresorbabledevicein 2000. Thiswork was done in conjunctionwithengineersat BiometCorp.Combining the developmentof resorbable materials with internal distraction wasa new frontier. Our efforts led to the design anddevelopment of several resorbable devices for use incraniofacial distraction osteogenesis. This new classof devices allowed for I-stage distraction osteogenesisof the craniofacial skeleton. We reported our initialexperience with a I-stage resorbable distractor in2002.14,15 This was followed by our report focusing

638

MATERIALS AND METHODS

From 2000 to 2007, 100 patients have undergonesingle-stageresorbable distraction of the mandible.

Therewere 49 girls and 51 boys ranging from 7 days to16 years old (mean, 4.24 y). Thirty-two patients hadunilateral distraction, and 68 had bilateral distraction.Indications for mandibular distraction include man-dibular hypoplasia associated with Pierre Robinsequence (43 patients), craniofacial microsomia, (37patients), Treacher Collins (8 patients), and Naggerssyndrome (4 patients), as well as various otherindications (8 patients). Distraction distances rangefrom 15 to 30 mm (mean, 25.4 mm).

Todate, 10 patients have undergone more than 1course of distraction to keep up with their growth.

The current technique for mandibular dis-traction using a resorb able device is as follows(Figs 1-9). The distractor placement is accomplishedthrough an external approach (Fig 9). A small inci-sion is made approximately 2 cm below the man-dible, and the platysma is divided. This is followedby subperiosteal dissection of the mandible. Theproper vector is determined preoperatively, andthe distractor cable is passed in the direction ofthe vector deep to the soft tissues. Most fre-quently, the distractor flex cable is brought out just

Fig 1 A, Infant with Pierre Robin sequence and severeobstructive apnea. B, Same patient 2 years after 20-=bilateral mandibular distraction.

RESORBABLE DISTRACTION OF THE MANDIBLE / Burstein

Fig 2 A, Infant with Pierre Robin sequence in process ofmandibular distraction. Note protective tubing over drivescrew extensions. B, Same patient 3 years after distraction.

under the sideburn or under the ear (Figs 2 and 3).The proximal and distal portions of the distract orare then reassembled on the distractor drive screw.The distractor is applied to the mandible with a 3- to5-mm gap between the proximal and distal por-tions to allow for the osteotomy. The author recom-mends using a self-tapping low-speed drill to avoidburning and thereby weakening the bone. Mono-cortical or bicortical screws can be used depend-ing on the thickness of the bone. Once the devicehas been secured to the mandible, the drive screwis backed out into the proximal housing, and theosteotomy is performed. The most frequently usedosteotomy technique is a monocortical osteotomyapproximately at the angle of the mandible, whichhas been previously described.14-16 This involvesusing a small reciprocating saw blade, with greatcare being taken to only penetrate the cortical bone.A bone spreader is then used to open the osteotomy,

Fig 3 Same patient as in Figure 1.

preserving the inferior alveolar nerve. Alternatively,an inverted-L osteotomy or stepped osteotomy hasbeen used depending on the size and shape of themandible. In applying resorbable distraction to themandible, several technicalpoints have been found tobe extremely helpful. These "pearls" are unique to re-sorbable technology and represent important techni-cal differences from the standard techniques used toapply titanium devices (Table1).These include the useof a combined drill and tap device that has quite abit of torque but is low speed to prevent burningthe bone. In addition, infant bone can be extremelythin and can be easily overdrilled using a high-speedbur. Application of the device before making theosteotomy allows for a more accurate prediction ofthe final vector and easier activation of the distractor.If available, a three-dimensional acrylic life-sizedmodel of the skull that can be sterilized is extremelyhelpful (Fig8). These models are manufactured fromthin-cut computed tomographic scan data by MedicalModeling (Denver, CO). The location of the inferioralveolar nerve and tooth follicles is also shownon the model. The design of the osteotomy can bemodified based on the shape of the mandible, loca-tion of the inferior alveolar nerve, or tooth follicles.The exact vectors can be planned on the model, andthe distractor can be contoured to fit the mandibleprecisely before being applied to the patient. Themodel of distractor can be chosen based on the three-dimensional anatomy of the mandible. This techniqueeliminates the frustration of trying to thermally con-tour the device in situ. Each device should be acti-vated 5 mm after application to confirm that theosteotomies are complete and that the segments aremobile. Once this has been done, the drive screws arebacked out until bone-to-bone contact is achieved.Partial or incomplete osteotomiesmay result in devicefailure, asymmetric distraction, or both.

A 48-hour latency period is allowed beforestarting distraction at a rate of 2 mm/ d until thedesired amount of distraction is achieved. Patients aregiven 5 days of antibiotics, usually a first-generationcephalosporin, and allowed a soft diet. After com-pleting the distraction, the drive screws are left inplace for an additional 4 to 6 weeks to allow the

Table 1. Key Technical Points (pearls) for ResorbableDistraction

Low-speed drillApply before osteotomyThree-dimensional modelCheck osteotomiesCheck device for security

639

THE JOURNAL OF CRANIOFACIAL SURGERY / VOLUME 19, NUMBER 3 May 2008

Fig 4 Patient with craniofacial microsomia. A, Beforeunilateral mandibular distraction (25mm). B,One year afterdistraction. Note restoration of facial symmetry.

regenerate to consolidate. The drive screw is backedout in the officewithout sedation.

RESULTS

One hundred patients underwent distraction ofthe mandible. Sixty-eight patients had bilateral

distraction, and 32 had unilateral distractions. Therange of mandibular distraction was 20 to 25 mm(mean, 25.4 mm; Figs 1-7). There were 6 complica-tions (6%). Two patients required debridement forexposed distractors, and 2 patients required removalfor infection. The patients requiring removal forinfection had completed the consolidation phaseand retained the new mandibular length. These

Fig 5 Patient with temporomandibular joint dysfunctionankylosis and mandibular hypoplasia secondary to perina-tal temporomandibular joint dysfunction infection. A,Before release of right ankylosis and simultaneous bilateralasymmetric distraction (25mm right, 15mm left). B,Patient3 years after distraction.

640

Fig 6 Fifteen year old boy with Pierre Robin sequence,severe micrognathia, and obstructive apnea. A, Beforedistraction. B,After bilateral30-mm mandibular distraction.

cases were early in the author's series, and thematerial that protruded beyond the last screw wasnot trimmed sufficiently at the time of surgery. Thisresulted in pressure against the skin by this protu-berant edge. In 1 early case, the mandibular distractorfailed when the distal plate became detached fromthe mandible and had to be reapplied to completethe distraction process. In 1 infant case, the proximalmandibular segment separated into 2 fragmentsduring distraction, and reoperation was necessary.The distractor was wired to the proximal segment,and distraction was successfully completed.

DISCUSSION

Resorbable I-stage mandibular distractors com-bine advances in resorbable materials and dis-

traction to allow for predictable distraction with asingle operation. Extensive studies of the safety andefficacyof resorbable fixation in children and infants

Fig 7 A and B, Same patient as in Figure 6, lateral view.

can be used to lengthen the mandible or advance themaxilla, orbits, and even cranium without having touse bone grafts?-ll This greatly shortens the timerequired for surgery and allows gradual stretchingof the nerves, cartilage, and blood vessels, resultingin decrease in surgical morbidity. In addition, thereis less relapse because distraction, done in a gradualmanner, allows the soft tissues to adapt and pre-vents structural inadequacies seen with conventionalmethods.19,2o

Engineering resorbable distraction devicesinvolves a whole new technology. Instead of conven-tional techniques such as milling, forging, andwelding used in metallic distraction devices, resorb-able distractors rely upon polymer science. Polymersof polylactic acid are formed in special molds or aremachined/milled from plate blanks of resorb ablematerial that are compression molded depending onthe manufacturer. Sterilization is achieved by either 'Yirradiation or by ethylene oxide.4,5 This process istime-consuming and currently quite costly due to theearly nature of the technology. Ounce for ounce, thepolymers that we are currently using is not as strongas an equivalent volume of titanium. Therefore,resorbable devices tend to be somewhat thicker andbulkier. Rather than bending, as is done with titaniumplates, the resorb able materials are thermally sensitiveand can be molded into three-dimensional shapesafter they are heated in a sterile water bath. It isimportant when thermally bending this class ofdevice that any threaded portions have the drivescrew or drive screw blank in place to avoid thermallydistorting the threaded surfaces. After heating andbending, the drive screw should be activated to makesure that proper turning occurs. To have a truly 1-stage device, the drive screw must be removable atthe end of the distraction but held in place longenough to allow the regenerate bone to provide

RESORBABLE DISTRACTION OF THE MANDIBLE / Burstein

Fig 8 Three-dimensional model of infant mandibledemonstrating location of inferior alveolar nerve.

have been published.l" This has become the norm forcraniofacial fixation for this age group. Resorptionbegins within 6 weeks and, in most cases, is completeby 9 months, leaving almost no perceptible trace ofthe material and allowing enough time for bonegrowth for stabilization of the construct.T'' Distrac-tion osteogenesis, when applied to the facial bones,

Fig 9 Artist rendition of device being applied to mandiblethrough submental approach. Proposed osteotomy isshown in red.

Fig 10 Photograph of resorbable distractors from infant,top to pediatric bottom.

641

screw and is much more comfortable for the patient.It is important to choose a drive screw and cablelength that will allow the desired distraction but notprotrude excessively. The flexible extension can beprotected from excessivebending using a cap, rubbertubing, or headband (Figs2 and 3).

Many challenges remain in applying resorbabledistraction of the mandible. These include making thedevices thinner, smaller, and easier to use. Currently,resorbable distraction devices are only approved forpatients younger than 2 years of age. We are stillexploring the limitations and applications of resorb-able distractors. Particularly gratifying results havebeen obtained in mandibular distraction for PierreRobin sequence that saved the infants from requiringa tracheotomy.l" The results for mandibular distrac-tion using I-stage resorbable devices are comparableto those of Denny et al and others using metallicdevices.18-23 Most of the complications were due totechnical errors, including incomplete osteotomies,cortical fractures in the bone, and application of thedistractor too close to the skin surface. Infections alloccurred during the breakdown of the distractorswhen granules of the material extruded from thewound, causing an inflammatory reaction. Interest-ingly,as the surgeon's experiencehas grown with thisnew technology, the number of complications hasdramatically decreased. We are currently investigat-ing combining distraction and morphogenic proteinin patients who have very poor bone stock, requireredistraction, have poor healing, or need distractionof greater than 25 mm.24,25 Preliminary results in ahalf-dozen patients reveal that this combination ofbone morphogenic protein with distraction mayresult in earlier bone healing and perhaps allow usto decrease the time for bone consolidation. In the last30 patients, a torque-measuring device has been usedto monitor the course of distraction. The torquemeasurement device is useful in assuring thatexcessive forces are not being applied during thedistraction process as well as measuring the forcesbeing applied from side to side (see related article inthis issue). We are using a torque-limiting distractordriver that disengages the turning handle if torquemore than 30 in. oz is applied, avoiding damage to thedistractor. In addition, this driver is clearly markedfor ease of use and comeswith a combined instructionsheet and log table to record the progress ofdistraction. We have found that measuring andrecording distraction torque can be helpful in deter-mining the status of the distractors. For example, anextremely low reading on 1 side in a bilateraldistraction case may indicate that the distractor hasbecome detached from the bone.

THE JOURNAL OF CRANIOFACIAL SURGERY / VOLUME 19, NUMBER 3 May 2008

Fig 11 Photograph of "L" device being applied tomandible. The neck between the drive screw and fixationholes allows for thermal contouring to nonlinear mandib-ular shapes.

structural support for the distracted segments. Sev-eral designs have been tested over the years (Figs 10and 11). The various shapes and sizes have beendiscussed in other publications.14-16,20 Most recently,an L device has been introduced.l" The unique featureof this design is that the distractor plates can bemolded to fit a nonlinear segment of the mandible.Initially, the drive screw was housed in a threadedhousing proximally and allowed to rotate freely,withno threads in the distal housing. Although appealingin its simplicity,this can occasionallylead to potentialproblems. If the distal segment of the distracted boneis abruptly pulled forward, it can lead the drive screwto be disassociated from its distal housing, thuscausing distraction failure. This problem can easilybe prevented by simply checking the distal housingbefore closure. To eliminate this potential source ofdevice failure, a new drive screw that is threadeddistally so that it screws into the distal housing hasbeen developed. There is a nonthreaded section nearthe distal tip of drive screw that allows the drivescrew to spin freely in distal housing, which preventsaccidental disengagement. Toremove the drive screw,the screw is simply backed out in the office. In theinitial designs, a fairly rigid cable with a hex headfor turning the drive screw was welded to the drivescrew. This junction occasionally failed when repeat-edly traumatized and was uncomfortable in somecases for the patient. The look of an antenna stickingout in the preauricular or postauricular sulcuswas notparticularly welcomed by patients and their families.In addition, rolling over 1 of the devices can dislodgethe fixation. To prevent this, we have developed aflexible proximal drive screw flex extension thatallows it to lie against the head when not beingactivated. This avoids undue pressure on the drive

642

RESORBABLE DISTRACTION OF THE MANDIBLE / Burstein

CONCLUSIONS

This has been a very exciting time in the develop-ment of resorbable distraction. With growing

experience, we have been able to develop morerefined techniques specifically suited to this class ofdevices. Just as titanium plate and screw fixation ininfants and children have largely been replaced byresorbable plates and screws, we feel that gradually,single-stage resorbable distractors will surpass theirtitanium predecessors. Although many challengesremain, we hope ongoing clinical and basic researchwill allow us to extend these applications to helpchildren with complex craniofacial disorders.

REFERENCES

1. Goldstein JA, Quereshy FA, Cohen AR. Early experience withbiodegradable fixation for congenital pediatric craniofacialsurgery. J Craniofac Surg 1997;8:110-115

2. Tharnanon W, Sinn DP, Hobar pc, et al. Surgical outcomesusing bioabsorbable plating systems in pediatric craniofacialsurgery. Craniofac Surg 1998;9:441--444

3. Eppley BL, Reilly M. Degradation characteristics PLLA-PGAbone fixation devices. J Craniofac Surg 1997;8:116-120

4. Pietrzak WS, Sarver DR, Vestynen ML. Bioabsorbable polymerscience for the practicing surgeon. JCraniofac Surg 1997;8:87-91

5. Shalaby SW, Burg KL. Bioabsorbable polymers update:degradation mechanism, safety and application. Appl Biomater1995;6:219-221

6. llizarov GA. A new principle of osteosynthesis with the use ofcrossing pins and rings. In: Collection of Scientific Works ofthe Kurgan Regional Scientific Medical Society.Kurgan, USSR:Kurgan Regional ScientificMedical Society, 1954:145-160

7. Snyder C, Levine G, Swanson H, et al. Mandibular lengtheningby gradual distraction. Plast Reconstr Surg 1973;51:506-508

8. Karp N, Thome CH, McCarthy JG, et al. Bone lengthening ofthe craniofacial skeleton. Ann Plast Surg 1990;24:231-237

9. McCarthy JG, Schriber J, Karp N, et al. Lengthening of thehuman mandible by gradual distraction. Plast Reconstr Surg1992;89:1-8

10. McCarthy JG, Stelnicki EJ,Mehrara BJ.Distraction osteogenesis

of the craniofacial skeleton. Plast Reconst Surg 2001;107:1812-1827

11. Mofid MM, Manson PN, Robertson BC, et al. Craniofacialdistraction osteogenesis: review of 3278 cases. Plast ReconstrSurg 2001;108:1103-1114

12. Molina F. Mandibular elongation and remodeling by gradualdistraction. An experience of 277 cases. Ann Chir Plast Esthet2001;46:507-515

13. Cohen SR, Holmes RE, Amis P, et al. Internal craniofacialdistraction with biodegradable devices: early stabilization andprotected bone regeneration. Craniofac Surg 2000;11:354-366

14. Burstein FD, Williams JK, Hudgins R. Single stage distractionusing resorbable devices. Craniofac Surg 2002;13:776-782

15. Burstein FD, Wiliams JK. Resorbable bone distraction: currentstatus and future directions. Clin Plast Surg 2004;31:407-414

16. Burstein FD, Williams JK.Mandibular distraction osteogenesisin Pierre Robin sequence: application of a new single-stageresorbable device. Plast Reconstr Surg 2005;115:61-67

17. Eppley BL, Morales L, Wood R, et al. Resorbable PLLA-PGAplate and screw fixation pediatric craniofacial surgery:clinical experience in 1883 patients. Plast Reconstr Surg 2004;114:850-856

18. Denny AD, Talisman R, Hanson PR, et al. Mandibulardistraction osteogenesis in very young patients to correctairway obstruction. Plast Reconstr Surg 2001;108:302-311

19. Denny A, Amm C. New technique for airway correction inneonates with severe Pierre Robin sequence. J Pediatr 2005;147:97-101

20. Margulis A, Patel PK, Daw JL. Distraction osteogenesis of themandible with an internal bioresorbable device. J CraniofacSurg 2003;14:791-796

21. Huisinga-Fisher CE, Vaandrager JM, Prahl-Andersen B. Long-itudinal results of mandibular distraction osteogenesis inhemifacial microsomia. J Craniofac Surg 2003;14:924-933

22. Shetye PR, Grayson BH, Mackool RJ, et al. Long-term stabilityand growth following unilateral mandibular distraction ingrowing children with craniofacial microsomia. Plast ReconstrSurg 2006;15:985-995

23. Lin SJ, Patel PK. Distraction osteogenesis in the pediatricpopulation. Otolaryngol Head Neck Surg 2007;137:233-238

24. Farhadieh RD, Gianoutsos MP, Walsh WR. The role of bonemorphogenic proteins BMP2 and BMP4 and their related postreceptor signaling system (Smads) in distraction osteogenesis ofthe mandible. J Craniofac Surg 2004;15:714-718

25. Chen AW,Wei FC, Wang KT,et al. Effect of bone morphogenicproteins-2 on mandibular distraction osteogenesis in rabbits.Zhonghua Kou Qiang YiXue Za Zhi 2007;42:15-17

643