Clin Plastic Surg 31 (2004) 377–385

Biomaterials in craniofacial reconstruction

Younghoon R. Cho, MD, PhD, Arun K. Gosain, MD*

Department of Plastic , Medical College of Wisconsin, 9200 West Wisconsin Avenue, Milwaukee, WI 53226, USA

Biomaterials have become an integral component quent infection, Wolfe [16] describes the safety and of craniofacial reconstruction. Their increasing ease efficacy of autogenous bone graft reconstruction. It of use, long ‘‘shelf-life,’’ and safety enables them to has been noted that resorption of autogenous bone be used effectively and play an important role in re- graft can be particularly problematic in certain situ- ducing operating times [1]. The ideal biomaterial is ations, particularly in reconstruction of the malar biocompatible with surrounding tissue, radiolucent, eminence in patients with Treacher Collins Syndrome, easily shaped or molded, strong enough to endure and when bone grafts are placed for augmentation of trauma, stable over time, able to maintain volume, and the chin [1]. However, difficulties may also be en- osteoactive [1–5]. There are various biomaterials cur- countered with the use of alloplastic materials in these rently available and specific usages have been charac- reconstructive situations, as they may result in erosion terized well in the literature. This article reviews of the underlying recipient bone. different biomaterials that can be used in craniofacial In experimental studies in which alloplastic reconstruction, including autogenous bone, methyl implants were removed, it was noted that the under- methacrylate and hard tissue replacement, hydroxy- lying recipient bone underwent transformation to a apatite, porous polyethylene, bioactive glass, and de- trabecular architecture with decreased bone density mineralized bone. [17]. This may be particularly troublesome if an alloplastic implant must be removed following infec- tion or dislodgment, which would result in greater Autogenous bone deformity than was initially present. Although some cite the morbidity involved in harvesting cranial bone, The first documented bone autograft was described it has been documented that with proper training, by Walther in 1821 where he replaced the bone plug plastic surgeons can harvest bone grafts easily and after trephination and noted partial healing [6,7]. Full comfortably with minimal morbidity [18]. healing of the wound was prevented by subsequent Cranial bone, iliac bone, ribs, and tibia are the wound suppuration. In 1885, Macewen [8] reported a most commonly used bone graft donor sites. In certain successful reimplantation of bone pieces into a cranial clinical circumstances, vascularized bone grafts, such defect. Other donor sites have been attempted for as free fibula or free iliac bone, provide distinct ad- repairing cranial defects, including the tibia (1889), vantages over nonvascularized bone grafts. The clini- cranium (1890), fascia (1906), rib (1911), scapula cal setting in which vascularized bone grafts have (1912), illium (1914), and sternum (1915) [6,9–15]. been used most frequently has been for reconstruction Today, many physicians believe that autogenous of large segments of the mandible. However, vascu- bone remains the biomaterial of choice for cranio- larized bone grafts require greater amounts of time facial reconstruction. In a series of 73 cranioplasties and skill and are not necessary in most clinical cir- reconstructed with autogenous bone with no subse- cumstances. Disadvantages of the autogenous bone grafts that are not seen with other biomaterials are the potential morbidity of the donor site and the * Corresponding author. additional time required to harvest the graft. In addi- E-mail address: [email protected] (A.K. Gosain). tion, autogenous bone graft often undergoes signi-

0094-1298/04/$ – see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.cps.2004.03.001 378 Y.R. Cho, A.K. Gosain / Clin Plastic Surg 31 (2004) 377–385 ficant resorption when used for augmentation of posite consisting of a polymethylmethacrylate sub- the facial skeleton, rendering it unreliable for long- strate sintered with polyhydroxyethyl and a calcium term augmentation. hydroxide coating shielding the poylmethylmetha- crylate from the external surface. It has a 20% to 30% material porosity (150 to 350 microns), holds Methyl methacrylate a negative surface charge (À8toÀ15 mV) and has substantial compressive strength (50,000 lb/in2 Methyl methacrylate is an acrylic-based resin that in particulate form and 5000 lb/ in2 in molded form) has found many uses in today’s craniofacial recon- [21]. Its porosity promotes vascular ingrowth. The struction. Its use in craniofacial reconstruction was hydrophilic surface diminishes bacterial adhesion fol- first described during the early stages of World War II lowing preimplantation soaking in antibiotic solution. when there was a growing interest in acrylic resins In addition, the negative surface charge further deters [6,19]. Its first reported human use was in 1940 by adhesion of bacteria to the implant. Zander [6,7]. Unlike autogenous bone, which can The implant can be prefabricated in custom shapes have variable resorption of between 25% and 40% for facial augmentation. When placed in a subperi- of the bone graft over time, methyl methacrylate is osteal position, the implants can be adequately held resistant to absorption [20]. A meta-analysis review- in place by closure of the overlying periosteum. Gu- ing 45 studies of routine cranioplasty with methyl yuron has extensive experience in the use of these methacrylate showed an infection rate of approxi- implants for malar augmentation, chin augmentation, mately 5%. In a series of 42 cranioplasties by Manson and augmentation of the temporal regions for correc- and colleagues [20], isolated cranioplasties showed no tion of the ‘‘hourglass’’ facial deformity [22]. The infection. However, patients who had undergone si- hourglass facial deformity may occur following tem- multaneous reconstruction of the cranial vault, the poral atrophy following elevation of the temporalis orbital walls, and nose had an infection rate of 23%. muscle and as the result of bilateral irradiation of the All patients who had infection with methyl meth- orbits and anterior cranial base during infancy. Cus- acrylate had experienced a previous infection, indicat- tom implants of HTR polymer can be placed in the ing that a history of infection in the region is a temporal region for augmentation. A recent report of significant risk factor for subsequent infection. Methyl four such cases demonstrates the successful correction methacrylate was found to be stronger than the adja- of the hourglass deformity using this technique [22]. cent bone to compression and torsion testing. In Eppley et al [23] described the successful use addition, they found that cranial orbital reconstruction of computer-generated HTR for cranial reconstruc- adjacent to previously infected ethmoidal sinuses were tion in fourteen patients who had large (greater than more directly related to infection than was the material 150 cm2) preexisting defects of the cranium or cranio- used for reconstruction. The authors cited an additional orbital region. They used a preoperative high-reso- advantage of methyl methacrylate to be low cost, pre- lution 3D CT scan to reconstruct the defect. The dictable resultant shape, ready availability, and suit- manufacturer then used this reconstruction to fabricate ability for complex defects.. The authors concluded the HTR implant with less than 1 mm accuracy. At the that methyl methacrylate is the cranioplasty material of time of surgery the implant was secured using metal or choice in adults with good soft tissue quality who have resorbable fixation. To minimize the risk of infection not had previous infection. However, criticism of in cases where the frontal sinus was in proximity to methyl methacrylate is that it is an inert and fixed the implant, the sinus was cranialized, covered with a substance that will not adapt to the changing cranio- pericranial flap, or obliterated with hydroxyapatite facial skeleton. This is particularly important if one cement paste. The authors reported no postoperative is considering skeletal reconstruction in a growing complications or infections and good reconstructive child. An additional disadvantage of methyl metha- results. This technique simplifies the reconstruction crylate is that there is no bone incorporation or and reduces operative time by eliminating the need ingrowth, making it susceptible to infection or dis- to harvest bone graft and to shape the graft intra- lodgement through the duration of the reconstruction. operatively [23].

Hard Tissue Replacement Hydroxyapatite

Hard Tissue Replacement (HTR; Walter Lorenz Hydroxyapatite is the primary mineral component Surgical, Jacksonville, FL) is another polymeric com- of teeth and bone and comprises up to 70% of the Y.R. Cho, A.K. Gosain / Clin Plastic Surg 31 (2004) 377–385 379 calcified skeleton. It is a calcium phosphate com- forms of hydroxyapatite. The authors found that pound arranged in a hexagonal structure and can be cranial bone grafts were not reliable for long-term produced synthetically as a ceramic by a process augmentation, with complete bone graft resorption called sintering. It is one of the more common forms observed in each of the facial recipient sites. The of calcium phosphate in clinical use. It has excellent volume maintenance of the hydroxyapatite compos- tissue compatibility and the advantage of being osteo- ites was much more predictable, with no significant active, radiolucent, and readily available. Osteoactiv- resorption noted in either the cement paste or ceramic ity is the ability of the biomaterial to be replaced with forms of hydroxyapatite over 1 year. Bone replace- bone formation either through osteoinduction or ment was greater within the ceramic forms than within osteoconduction. Hydroxyapatite is a porous material the cement paste forms of hydroxyapatite. Increased that promotes osteointegration, or the formation of a bone replacement within the center of the ceramic bond between the implant and adjacent bone. In implants was attributed to the porosity of the implants. addition, it is biocompatible and does not produce a The cement paste implants used for facial augmenta- chronic inflammatory response [4,24–26]. tion demonstrated a lip of bone, which surrounded the Hydroxyapatites have traditionally been available implants and blended into the adjacent facial bone in ceramic forms, which are nonresorbable. Hydroxy- cortex. However, there was little or no bone within the apatite has also been made available for clinical use in center of the implants. A similar result was observed cement paste form since 1992 [3] and in granular form in a clinical example of a 3-year follow-up on the use since 1993 [27]. Hydroxyapatite cement paste is of hydroxyapatite cement as an onlay to an area of formed when tetracalcium phosphate and dicalcium frontal bone depression in a 4-year-old girl. Biopsy phosphate react in the presence of water to form demonstrated that the implant had minimal bone hydroxyapatite [28]. The two calcium salts undergo ingrowth, with significant bone noted only at the an isothermic reaction to form a dense paste that has periphery of the implant [5]. The hydroxyapatite been claimed to be resorbable over time. Mixing the cement paste remained largely nonresorbed, with powder in a sodium phosphate buffer solution can minimal bone entering the center of the implant accelerate the setting time of the cement [29]. Initial material. Clinical experience to date indicates that studies by Constatino [3] and associates in feline although hydroxyapatite cement paste appears to cranial defects showed that new bone comprised provide an excellent biomaterial for cranial vault 77% of the tissue replacing the hydroxyapatite ce- reconstruction, to date there is no histologic evidence ment 1 year after implantation. The authors have also of significant bone ingrowth or resorption of this bio- used this material for obliteration and reconstruction material in humans. of the cat frontal sinus [30] and for reconstructing In a series of 56 pediatric and five adult patients, a fronto-orbital craniotomy in 12-week-old kittens Burstein et al [38] showed excellent results over a [31]. Whereas there were some subtle growth differ- mean follow-up of 20 months using hydroxyapatite ences in kittens reconstructed with the cement paste, cement paste, without any adverse affect on orbito- excellent contour reconstruction was achieved in all cranial growth with. They reported seven (11%) com- animals [31]. plications, including seroma formation (n = 4), visible A major advantage of the cement paste over the irregularities requiring reoperation (n = 2), and a case ceramic form of hydroxyapatite is that it can be easily in which a drain placed in contact with the hydroxy- shaped during surgery [32]. The cement paste form of apatite became fixed within the cement and required hydroxyapatite has been used extensively for adult operative removal. The authors reported no visible craniofacial reconstruction [33,34]. It has also been loss of implant volume or gross resorption of the used in growing children for cranioplasty and for implant, no visible thinning of the overlying skin or treatment of temporal hollowing following cranial visible edges at the implant-bony interface, and no vault remodeling [35]. Friedman and associates [36] implant migration. Intraoperative curing time was have reviewed a number of clinical applications in reported to be 20 minutes, compared with 5 days which they report excellent results using hydroxy- following reconstruction with hydroxyapatite gran- apatite cement paste for craniofacial reconstruction. ules. Curing time for the cement paste was further A large animal sheep model, however, has not reduced to 10 minutes when sterile water with mono- demonstrated significant bone ingrowth into hydroxy- sodium phosphate solution was mixed with dry apatite cement paste cranioplasty when studied over a cement [38]. 1-year period. In this study, Gosain et al [37] inves- Byrd and colleagues [39] reported their use of tigated craniofacial augmentation using autogenous porous granular hydroxyapatite for craniofacial aug- bone graft versus nonceramic (cement) and ceramic mentation (Interpore 200, Cross International, Irvine, 380 Y.R. Cho, A.K. Gosain / Clin Plastic Surg 31 (2004) 377–385

CA). The granules are mixed intraoperatively with that within untreated implants was predominantly blood to give them an adhesive consistency and held woven. Thereafter, increasing amounts of lamellar in place with clot formation. The authors originally bone gradually appeared in the untreated implants. reported the results for facial augmentation in In another study, Holmes and colleagues [44] found 43 patients. Twenty-six patients were followed for that demineralized bone matrix filled within coralline over a year, and all achieved excellent results. Areas hydroxyapatite resulted in enhanced new bone forma- of the craniofacial skeleton that benefited from aug- tion and an increase in the rate of healing of cranial mentation with hydroxyapatite included the skull, defects within rabbits compared with defects recon- zygomatic-maxillary, lateral mandible, perialar, peri- structed with coralline hydroxyapatite alone [44]. The orbital, and temporal regions. The authors felt that authors also emphasized the role of backscatter elec- their technique achieved a predictable result with tron microscopy for the analysis of bone ingrowth minimal migration of the granules. No cases of in- within hydroxyapatite implants, stressing improved ac- fection were reported, and only two patients required curacy of the technique over conventional histology. minor revisions. Interpore 200 is a nonabsorbable form of hydroxyapatite, and no clinical evidence of resorption was noted. Byrd and Hobar [39] updated Porous polyethylene (Medpor) this experience in 1996, reporting on more than 200 patients operated over an 8-year period. Porous polyethylene (Medpor; Porex Surgical, Holmes [40] reviewed his experience in experi- College Park, GA) is commonly used for facial aug- mental and clinical applications of ceramic forms of mentation and to restore continuity to craniofacial hydroxyapatite. He initially demonstrated bone regen- skeletal defects. Polyethylene resins are straight-chain erated within an implanted hydroxyapatite replica aliphatic hydrocarbons that are inert and promote of a coral skeletal structure placed within mandibular little tissue reactivity. Unlike bone grafts, porous defects in dogs. By 6 months postimplantation, polyethylene shows little evidence of implant degra- 88% of the implant areas were filled with the regen- dation. Like HTR, this is a porous (100 to 250 mm) erated bone. This regenerated bone was a woven biomaterial that permits bone and soft tissue ingrowth type at 2 months and changed to a lamellar type by when placed on facial skeletal defects. There are 6 months. The coralline hydroxyapatite implants were reports of sufficient soft tissue ingrowth with suffi- biodegradable, and 29% resorption of the implants cient vascularity to allow skin grafts to be placed was noted in two implanted defects examined at directly over the implant [45]. 12 months. Holmes and colleagues subsequently Dougherty and Wellisz [46] compared Medpor to studied porous hydroxyapatite ceramics as bone graft silicone implants for reconstruction of orbital floor substitutes. Dog models were established to use this defects using rabbits. The authors created 8-mm de- material in alveolar ridge augmentation [41] and in fects bilaterally in the maxillary sinuses that included cranial reconstruction [42]. In both models bone in- bone and mucosa. They reconstructed one side with growth extended across the entire extent of the im- Medpor and the other with silicone. One surface of the plant. Bone growth was noted throughout the pores of implant was exposed to the open maxillary sinus. the implant, and the bone appeared mature and well They studied the implant sites for 5 months post- vascularized. In contrast, little bone ingrowth was implantation through serial sacrifice of animals. The seen in reconstruction of similar defects using autog- porous polyethylene implants showed vascular and enous bone grafts. One-and-a-half years postopera- soft tissue ingrowth in the pores in the first week and tively the implant specimens consisted of 45% bone bone ingrowth by 3 weeks. The use of Medpor im- in the alveolar ridge, and 40% bone in the cranial plants resulted in more rapid closure of the obturated defects. These studies confirmed the applicability of defects with soft tissue and bone fixation, whereas the nonresorbable porous form of hydroxyapatite as a silicone implants developed a fibrous tissue capsule bone graft substitute material for facial reconstruction. without fixation to the adjacent skeleton. Holmes and colleagues [43] also found that bone In a recent review, Yaremchuk [47] described his induction within porous hydroxyapatite could be 11-year clinical experience in which 370 porous accelerated through the addition of an osteogenic polyethylene implants were placed in 162 patients. protein. The authors found that there was significantly The implants were used in various clinical scenarios, more bone ingrowth within implants treated with including acquired tumor-related and congenital osteoinductive protein. By 3 months postimplantation, defects, aesthetic procedures, secondary posttrau- the new bone within the implants treated with osteo- matic reconstruction, and orbital wall reconstruction inductive protein was predominantly lamellar, whereas following acute trauma. The distribution of implants Y.R. Cho, A.K. Gosain / Clin Plastic Surg 31 (2004) 377–385 381 included 21 frontal, 30 temporal, 145 internal orbit, Most biomaterials, including bioactive glasses, are 28 infraorbital rim, 58 malar, 29 paranasal, 13 nasal, osteoconductive, serving as a biocompatible interface 24 mandible, and 22 genioplasty. All implants were along which bone cells migrate. In addition, bioactive placed in a subperiosteal plane and fixed with tita- glasses are osteoproductive, which is defined as the nium screws. There were no reported cases of extru- process whereby a bioactive surface is colonized by sion, migration, or capsule formation. The author osteogenic stem cells from the defect environment as a reported a 10% reoperation rate, consisting of im- result of surgical intervention [52]. Bioglass particles plant removal for infection (3%) or displeasing con- range in size from 90 to 710 microns. Resorption of tour (2%), and implant revision/replacement for bioglass particles of 150 microns or less occurs as improvement of contour (6%). Yaremchuk [47] de- silica is released within the apatite gel layer. Larger scribed three key maneuvers for accurate skeletal bioglass particles are incorporated in the growing bone augmentation: (1) obtain wide subperiosteal exposure matrix and eventually broken down by osteoclasts. As of the area to be augmented; (2) screw fixation of the a result of their bioactive properties, the interfacial implant to the underlying skeleton to restrict move- bonding strength of most bioactive materials is equiv- ment; (3) ‘‘in-place’’ implant contouring to allow a alent to or greater than that of bone [53].Unlike smooth transition between the implant edge and the nonbioactive alloplasts, failure under mechanical facial skeleton. stress does not occur at the bone interface, but rather occurs in the host bone or within the biomaterial. This absence of failure at the bone interface is a unique and Bioactive glass particulate (Nova Bone) defining feature of bioactive materials [54]. We have recently reviewed the role of bioactive Bioactive materials are defined as those that elicit glass in craniomaxillofacial reconstruction [55]. Most a specific biological response at the interface of the clinical experience with bioactive glass has focused material that results in the formation of a bond on the repair of periodontal and alveolar ridge defects between the tissue and the material [48]. This mini- [56–60].However,therearereporteduseswith mizes the formation of a fibrous capsule around the varying degrees of success for reconstruction of other implant. Bioactive glasses have been shown to form areas of the head and . Bioactive glass has been a surface apatite layer in vivo that enhances the for- used for the repair of orbital floor fractures with mation and attachment of bone. Nova Bone (Porex maintenance of globe position in studies that have Surgical, College Park, GA) is a synthetic bioactive followed patients for up to 1 year [61,62]. A com- glass particulate consisting of 45% silica dioxide, posite of bioactive glass (80% to 90%) and autoge- 45% sodium oxide, 5% calcium, and 5% phosphate, nous iliac bone (10% to 20%) resulted in accelerated which is believed to be bioactive toward the produc- bone regeneration healing time compared with bone tion of new bone within the biomaterial. It was first graft alone for elevation of the floor of the maxillary introduced in 1971 by Hench and colleagues [49]. sinus floor [63]. Cordioli et al [64] have used these The bioactivity in glass particulates begins when they principles in 27 patients who would otherwise have are mixed with saline or blood [50]. The silicon- insufficient maxillary bone for implant placement. oxygen bonds are broken to release silicic acid, which They achieved simultaneous bone augmentation of condenses to form a negatively charged gel at the the maxillary sinus floor and placement of titanium surface of the particles. This gel serves to hold the implants for dental restoration. glass particles in a cohesive mass. Within several The authors have reported on bioactive glass par- hours, calcium phosphate is produced within the gel ticles (Nova Bone, Porex Surgical) mixed with autog- to crystallize into a new surface apatite layer. Bioac- enous bone particles harvested from cranial burr holes tivity is initiated within this surface layer when as an adjunct to cranial vault reconstruction [5,55]. collagen, mucopolysaccharides, and glycoproteins This was done to reconstruct full-thickness defects from surrounding bone are incorporated into the apa- in two patients 5 years and older, when minimal tite layer to mediate a direct chemical bond with the spontaneous bone regeneration was expected. On host bone, facilitating early bone formation at the follow-up CT scan, these patients demonstrated con- biomaterial-bone interface. The growing apatite layer version of most of the reconstructed defect to bone further serves to stimulate osteoprogenitor cells to density within 6 months. In a 4-year follow-up, both produce TGF-b by release of silicon from the glass patients have had stable reconstruction with no need surface. TGF-b serves as an osteogenic cytokine, for reoperation or biopsy of the biomaterial. leading to a rapid proliferation of bone in contact Some complications of bioactive glass have been with the glass particles [51]. reported when used as a ceramic implant for contour 382 Y.R. Cho, A.K. Gosain / Clin Plastic Surg 31 (2004) 377–385 restoration of the facial skeleton. Duskova et al [65] facial defect. The authors stress that this material af- have found extrusion rates in 20% of cases, requir- fords the advantage of nearly limitless supply without ing reoperation with implant size reduction or soft the risk of donor harvesting, which is particularly tissue coverage. useful in the pediatric population. Recently, demineralized bone paste has become commercially available (Synthes Maxillofacial, Paoli, Demineralized bone PA). An advantage of this biomaterial is that it is po- rous and is easily molded to fit the defect. The material Demineralized bone is another alternative for re- does not harden but remains as a paste intended to construction of the craniofacial skeleton. Salyer and serve as a matrix for bone ingrowth. It is therefore associates have studied the use of cortical deminer- well-suited to reconstruct defects in the craniofacial alized perforated bone for reconstructions (produced skeleton. However, demineralized bone paste is not by Pacific Coast Tissue Bank) in animal models and suited to reconstruct load-bearing regions in the skele- in the clinical setting [66–70]. One study evaluated ton because it will not harden to bear significant load the use of demineralized bone for reconstruction until significant bone ingrowth is complete. of calvarial defects in beagles [68]. The authors veri- fied that the demineralized bone implants were well- accepted within calvarial defects with little tissue Summary reaction and remarkably little osteoclastic activity. There was evidence of new bone growth by 8 and This review highlights some of the recent develop- 12 weeks following implantation of the demineral- ments in biomaterials that are suited to reconstruction ized bone. Bone growth tended to occur more fre- of the craniofacial skeleton. Although there is no ideal quently on the dural aspect of the calvaria and formed biomaterial, numerous alternatives are available to a continuum with the implant. Fragmentation of the practicing surgeons that provide attractive alternatives implanted material was observed in the presence of to autogenous bone graft in the appropriate clinical new bone formation by 12 weeks postoperatively. settings. Biomaterials are a particularly well-suited for This fragmentation occurred in the absence of multi- skeletal augmentation, as autogenous bone can often nucleated cells, and the authors hypothesized that a undergo unpredictable resorption in these applica- hydrolytic enzyme was responsible for the degrada- tions. Although all of the biomaterials discussed in tion. No marked differences were found between this article seem to maintain their volume over time, demineralized bone grafts of calvarial origin and porosity of the biomaterial may be a significant fac- those of tibial origin. tor in determining bone ingrowth into the implant. Salyer and colleagues [67] published their first Methyl methacrylate is nonporous, and no bone in- clinical report on the use of demineralized bone in growth is expected. Cement paste implants tend to 1992. They have since reported on a number of contain micropores, and experimental and clinical innovative clinical applications of demineralized evidence indicates that there is less long-term bone bone, including that of skull reconstruction in Sia- ingrowth into these biomaterials than in implants with mese twins who had previously been separated from macroporous architecture. Biomaterials presently re- union at the skull vertex [69]. The implants were viewed that have a macroporous architecture and processed with microperforations, which are believed have demonstrated bone ingrowth in clinical or ex- to be centers of new bone formation. The demineral- perimental studies include ceramic and granular forms ized bone has been biopsied 4 years following initial of hydroxyapatite, Hard Tissue Replacement (HTR) reconstruction in one of the Siamese twins. Most of polymer, porous polyethylene (Medpor), bioactive the specimen represented large areas of nonvital glasses (Nova Bone), and demineralized bone paste. bone, lacking living bone cells. In several areas there Prefabricated biomaterials and those that set as a ce- appeared to be fragmentation of the autogenous bone ment are not designed to change dimension over time matrix. However, active resorption was not observed, and are therefore best-suited for cranial vault recon- osteoclasts were not seen, and there was no inflam- struction after completion of skull growth [71]. matory or fibrotic reaction in the adjacent soft tissues. Rubin and Yaremchuk [72] conducted an exhaus- Remodeling was seen in several areas contiguous tive review of the complications and toxicities of with nonvital bone. The authors also stressed the implantable biomaterials used in facial surgery. They technical advantages in using cortical demineralized reviewed nearly 200 clinical studies reporting series perforated bone—the material is pliable and can be of patients with implantable biomaterials in the . shaped intraoperatively to suit the specific cranio- Polymer and ceramic materials in the face had an Y.R. Cho, A.K. Gosain / Clin Plastic Surg 31 (2004) 377–385 383 overall infection rate of 3% and an exposure/extru- an Stelle der Trepanation [in German]. Zentralbl Chir sion rate of 1.2%; 4.6% of implants were removed 1890;4:65–6. because of implant-related complications. The au- [11] Beck C. Ueber eine neue Methode der Deckung von thors concluded that it is difficult to attribute many Scha¨deldefekten [in German]. Arch Klin Chir 1906; 80:266–71. of the complications solely to the implant material [12] Dobrotworski WJ. Die Rippen als Material zur Kno- itself, and that there is much overlap between surgical chenautoplastik [in German]. Zentralbl Chir 1911;32: technique, host response, and potential toxicity of the 1081–3. implant. The authors also noted that the biocompat- [13] Ro¨pke W. Zur Frage der Deckung von Scha¨deldefekten ibility of a material can vary depending on the [in German]. Zentralbl Chir 1912;35:1192–4. conditions under which the implant is placed. Pro- [14] Mauclaire P. Autogreffe craˆnienne emprunte´ea` la tu- plast had been used for implants in the malar, chin, be´rosite´ iliaque, et homogreffe se´euse interme´ningo- nasal, and orbital floor regions with acceptable com- ence´phalique [in French]. Soc Chir Bull Mem 1914; plication rates. However, when it was used as an 40:113–5. interpositional disk implant in the temporomandibu- [15] Mu¨ller P. Deckung von Scha¨deldefekten aus dem Ster- num [in German]. Zentralbl Chir 1915;23:409–10. lar joint, complication rates were significantly higher. [16] Wolfe SA. Frontal cranioplasty: risk factors and choice Nearly all of the Proplast implants fractured over time of cranial vault reconstructive material [discussion]. under the load of the temporomandibular joint, and Plast Reconstr Surg 1986;77:901–4. particulate fragments of Proplast would elicit a vig- [17] Gosain AK, McCarthy JG, Staffenberg D, Glat PM, orous foreign body reaction and contribute to erosion Simmons DJ. The histomorphologic changes in vascu- of the joint. Therefore, alloplastic implants used for larized bone transfers and their inter-relationship with facial augmentation may have a different outcome the recipient sites: a one year study. Plast Reconstr Surg when placed in positions subject to stress loading. 1996;97:1001–13. Although current implant materials have favorable [18] Kline RM, Wolfe SA. Complications associated with complication rates in most craniofacial applications, a the harvesting of cranial bone grafts. Plast Reconstr Surg 1995;95:5–13. biomaterial that fairs well in one clinical circum- [19] Woodhall B, Spurling RG. Tantalum cranioplasty for stance may not be ideal for all applications in facial war wounds of the skull. Ann Surg 1945;121:649–71. reconstructive surgery. [20] Manson PN, Crawley WA, Hoopes JE. Frontal cra- nioplasty: risk factors and choice of cranial vault re- constructive material. Plast Reconstr Surg 1986;77: 888–900. References [21] Eppley BL, Sadove AM, German RZ. Evaluation of HTR polymer as a craniomaxillofacial graft material. [1] Gosain AK, Persing JA. Biomaterials in the face: bene- Plast Reconstr Surg 1990;86:1085–92. fits and risks. J Craniofac Surg 1999;10:404–14. [22] Guyuron B. The hourglass facial deformity. J Cranio- [2] Damien JD, Parsons JR. Bone graft and bone graft maxillofac Surg 1990;18:187–91. substitutes: a review of current technology and appli- [23] Eppley BL, Kilgo M, Coleman JJ. Cranial reconstruc- cations. J Appl Biomater 1991;2:187–208. tion with computer-generated hard-tissue replacement [3] Costantino PD, Friedman CD, Jones K, Chow LC, patient-matched implants: indications, surgical tech- Sisson GA. Experimental hydroxyapatite cement cra- niques, and long-term follow-up. Plast Reconstr Surg nioplasty. Plast Reconstr Surg 1992;90:174–91. 2002;109:864–71. [4] Jackson IT, Yavuzer R. Hydroxyapatite cement: an [24] Dujovny M, Aviles A, Agner C. An innovative ap- alternative for craniofacial skeletal contour refine- proach for cranioplasty using hydroxyapatite cement. ments. Br J Plast Surg 2000;53:24–9. Surg Neurol 1997;48:294–7. [5] Gosain AK. Biomaterials in facial reconstruction. [25] Ross MH, Romwell LJ, Kaye GI. Histology: a text and Operative Techniques in Plastic and Reconstructive atlas. Third edition. Baltimore: Williams & Wilkins; Surgery 2003;9:23–30. 1995. p. 50–187. [6] Sanan A, Haines SJ. Repairing holes in the head: [26] Friedman CD, Costantino PD, Sajjandian A. Alloplas- a history of cranioplasty. Neurosurgery 1997;40(3): tic materials for facial skeletal augmentation. Facial 588–603. Plast Surg Clin North Am 1999;7:95–105. [7] Woolf JI, Walker AE. Cranioplasty: collective review. [27] Byrd HS, Hobar PC, Shewmake K. Augmentation of Int Abstr Surg 1945;81:1–23. the craniofacial skeleton with porous hydroxyapatite [8] Macewen W. Illustrative cases of cerebral surgery. granules. Plast Reconstr Surg 1993;91:15–22. Lancet 1885;1:881–3, 934–6. [28] Chow LC, Takigi S, Constantino PD, Friedman CD. [9] Seydel. Kleinere Mittheilungen [in German]. Zentralbl Self-setting calcium phosphate cements. Mat Res Soc Chir 1889;16:209–11. Symp Proc 1991;179:3–24. [10] Mu¨ller W. Zur Frage der tempora¨ren Scha¨delresektion [29] Costantino PD, Friedman CD, Jones K, Chow LC, 384 Y.R. Cho, A.K. Gosain / Clin Plastic Surg 31 (2004) 377–385

Pelzer HJ, Sisson Sr GA. Hydroxyapatite cement I. [44] Damien CJ, Parsons JR, Prewett AB, Huismans F, Basic chemistry and histologic properties. Arch Oto- Shors EC, Holmes RE. Effect of demineralized bone laryngol Head Neck Surg 1991;117:379–84. matrix on bone growth within a porous HA material: [30] Friedman CD, Costantino PD, Jones K, Chow LC, a histologic and histometric study. J Biomater Appl Pelzer HJ, Sisson Sr GA. Hydroxyapatite cement II. 1995;9:275–88. Obliteration and reconstruction of the cat frontal sinus. [45] Wellisz T, Kanel G, Anooshian RV. Characteristics of Arch Otolaryngol Head Neck Surg 1991;117:385–9. the tissue response to Medpor porous polyethylene [31] Lykins CL, Friedman CD, Costantino PD, Horioglu R. implants in the human facial skeleton. J Long Term Hydroxyapatite cement in craniofacial skeletal recon- Eff Med Implants 1993;3:223–35. struction and its effects on the developing craniofacial [46] Dougherty WR, Wellisz T. The natural history of allo- skeleton. Arch Otolaryngol Head Neck Surg 1998;124: plastic implants in orbital floor reconstruction: an ani- 153–9. mal model. J Craniofac Surg 1994;5:26–32. [32] Gosain AK, Song L, Riordan P, Amarante MT, Nagy [47] Yaremchuk MJ. Facial skeletal reconstruction using PG, Wilson CR, et al. A 1-year study of osteoinduc- porous polyethylene implants. Plast Reconstr Surg tion in hydroxyapatite-derived biomaterials in an adult 2003;111:1818–27. sheep model. Part I. Plast Reconstr Surg 2002;109: [48] Hench LL. Bioceramics: materials characteristics ver- 619–30. sus in vivo behavior. In: Ducheyne P, Lemons J, edi- [33] Gosain AK. Part I: the use of biomaterials in facial tors. Ann N Y Acad Sci 1998;523:54. reconstruction. Maxillofacial News (Newsletter of the [49] Hench LL, Splinter RJ, Allen WC, Greenlee TK. American Society of Maxillofacial Surgeons) Winter. Bonding mechanisms at the interface of ceramic pros- 2001. p. 6–11. thetic materials. J Biomed Mater Res 1972;2:117–41. [34] Gosain AK. Part II: the use of biomaterials in facial [50] Blaydon S, Amato MM, Neuhaus R, Shore JW. The reconstruction. Maxillofacial News (Newsletter of the orbito-facial uses of Novabone C/M, a bioactive American Society of Maxillofacial Surgeons) Winter. glass synthetic bone graft particulate for craniofacial 2002. p. 17–22. and maxillofacial surgery. Presented at the American [35] Gosain AK. Hydroxyapatite cement paste cranioplasty Society of Oculoplastic Plastic Reconstructive Sur- for the treatment of temporal hollowing after cranial gery Scientific Symposium. Dallas, Texas, October vault remodeling in a growing child. J Craniofac Surg 20, 2000. 1997;8:506–11. [51] Price N, Bendall SP, Frondoza C, Jinnah RH, Hunger- [36] Friedman CD, Costantino PD, Takagi S, Chow LC. ford DS. Human osteoblast-like cells (MG63) prolifer- Bonesource hydroxyapatite cement: a novel biomate- ate on a bioactive glass surface. J Biomed Mater Res rial for craniofacial skeletal tissue engineering and re- 1997;37:394–400. construction. J Biomed Mater Res 1998;43:428–32. [52] Wilson J, Low SB. Bioactive ceramics for periodontal [37] Gosain AK, Riordan PA, Song L, Amarante MT, treatment: comparative studies in the Patus monkey. Kalantarian B, Nagy PG, et al. Part II: a one year study J Appl Biomat NY 1992;3:123–9. of hydroxyapatite derivatives in reconstruction of cra- [53] Kitsugi T, Yamamuro T, Kokubo T. Bonding behavior nial defects: bioengineering implants to optimize bone of a glass-ceramic containing apatite and wollastonite replacement. Plast Reconstr Surg, in press. in segmental replacement of the rabbit tibia under load- [38] Burstein FD, Cohen SR, Hudgins R, Boydston W, bearing conditions. J Bone Joint Surg [Am] 1989;71: Simms C. The use of hydroxyapatite cement in sec- 264–72. ondary craniofacial reconstruction. Plast Reconstr Surg [54] Hench LL, West JK. Biological applications of bio- 1999;104:1270–5. active glasses. Life Chemistry Reports 1996;13: [39] Byrd HS, Hobar PC. Alloplastic nasal and perialar 187–241. augmentation. Clin Plast Surg 1996;23:315–26. [55] Gosain AK. Bioactive glass for bone replacement [40] Holmes RE. Bone regeneration within a coralline hy- in craniomaxillofacial reconstruction. droxyapatite implant. Plast Reconstr Surg 1979;63: Education Foundation Device and Technique Assess- 626–33. ment Committee. Plast Reconstr Surg, in press. [41] Holmes RE, Roser SM. Porous hydroxyapatite as a [56] Quinones CR, Lovelace TB. Utilization of a bioactive bone graft substitute in alveolar ridge augmentation: synthetic particulate for periodontal therapy and bone a histometric study. Int J Oral Maxillofac Surg 1987; augmentation techniques. Pract Periodont Aesthet 16:718–28. Dent 1997;9:1–7. [42] Holmes RE, Hagler HK. Porous hydroxyapatite as a [57] Han J, Meng H, Xu L. Clinical evaluation of bioactive bone graft substitute in cranial reconstruction: a histo- glass in the treatment of periodontal intrabony defects. metric study. Plast Reconstr Surg 1988;81:662–71. Zonghua Kou Qiang Yi Xue Za Zhi 2002;37:225–7. [43] Miller TA, Ishida K, Kobayashi M, Wollman JS. The [58] Sy IP. Alveolar ridge preservation using a bioactive induction of bone by an osteogenic protein and the glass particulate graft in extraction site defects. Gen conduction of bone by porous hydroxyapatite: a labo- Dent 2002;50:66–8. ratory study in the rabbit. Plast Reconstr Surg 1991;87: [59] Throndson RR, Sexton SB. Grafting mandibular third 87–95. molar extraction sites: a comparison of bioactive glass Y.R. Cho, A.K. Gosain / Clin Plastic Surg 31 (2004) 377–385 385

to a nongrafted site. Oral Surg Oral Med Oral Pathol [66] Hubli EH, Salyer KE, Gendler E. Demineralized bone Oral Radiol Endod 2002;94:413–9. bandeau in a patient with Kleeblattschadel skull defor- [60] Norton MR, Wilson J. Dental implants placed in ex- mity. Ann Plast Surg 1998;41:81–5. traction sited implanted with bioactive glass: human [67] Salyer KE, Gendler E, Menendez JL, Simon TR, Kelly histology and clinical outcome. Int J Oral Maxillofac KM, Bardach J. Demineralized perforated bone im- Implants 2002;17:249–57. plants in craniofacial surgery. J Craniofac Surg 1992; [61] Aitasalo K, Kinnunen I, Palmgren J, Varpula M. Repair 3:55–62. of orbital floor fractures with bioactive glass implants. [68] Salyer KE, Bardach J, Squier CA, Gendler E, Kelly J Oral Maxillofac Surg 2001;59:1390–6. KM. Cranioplasty in the growing canine skull using [62] Kinnunen I, Aitasalo K, Pollonen M, Varpula M. Re- demineralized perforated bone. Plast Reconstr Surg construction of orbital floor fractures using bioactive 1995;96:770–9. glass. J Craniomaxiollofac Surg 2000;28:229–34. [69] Salyer KE, Gendler E, Squier CA. Long-term outcome [63] Tadjoedin ES, de Lange GL, Lyaruu DM, Kuiper L, of extensive skull reconstruction using demineralized Burger EH. High concentrations of bioactive glass perforated bone in Siamese twins joined at the skull material (BioGran) vs. autogenous bone for sinus floor vertex. Plast Reconstr Surg 1997;99:1721–6. elevation. Clin Oral Implants Res 2002;13:428–36. [70] Salyer KE. Personal contributions to craniofacial sur- [64] Cordioli G, Mazzocco C, Schepers E, Brugnolo E, gery. Scand J Plast Reconstr Hand Surg Suppl 1995; Majzoub Z. Maxillary sinus floor augmentation using 27:19–47. bioactive glass granules and autogenous bone with [71] Gosain AK. Biomaterials for reconstruction of the simultaneous implant placement. Clinical and his- cranial vault. Plastic Surgery Education Foundation tological findings. Clin Oral Implants Res 2001;12: Device and Technique Assessment Committee. Plast 270–8. Reconstr Surg, in press. [65] Duskova M, Smahel Z, Vohradnik M, Tvrdek M, [72] Rubin JP, Yaremchuk MJ. Complications and toxicities Mazanek J, Kozak J, et al. Bioactive glass-ceramics of implantable biomaterials used in facial reconstruc- in facial skeleton contouring. Aesthetic Plast Surg tive and aesthetic surgery: a comprehensive review of 2002;26:274–83. the literature. Plast Reconstr Surg 1997;100:1336–53.