- Email: [email protected]
poly-l -lactide mesh tray with particulate cellular bone and marrow and platelet-rich plasma for a mandibular defect: Evaluation of tray fit and bone quality in a dog model
Journal of Cranio-Maxillo-Facial Surgery 40 (2012) e453ee460
Contents lists available at SciVerse ScienceDirect
Journal of Cranio-Maxillo-Facial Surgery journal homepage: www.jcmfs.com
Application of custom-made bioresorbable raw particulate hydroxyapatite/ poly-L-lactide mesh tray with particulate cellular bone and marrow and platelet-rich plasma for a mandibular defect: Evaluation of tray fit and bone quality in a dog modelq Akira Matsuo*, Hidetoshi Takahashi, Harutsugi Abukawa, Daichi Chikazu Department of Oral and Maxillofacial Surgery, Tokyo Medical University, 6-7-1 Nishishinjuku Shinjuku-ku, Tokyo 160-0023, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Paper received 21 April 2011 Accepted 6 March 2012
The purpose of this study was to evaluate tray fit and bone quality of particulate cancellous bone and marrow (PCBM) mandibular reconstruction using custom-made bioresorbable forged composites of a raw particulate hydroxyapatite/poly-L-lactide (HA/PLLA) tray in a dog model. Mesh sheets of HA/PLLA were formed in a tray shape according to the mandible stereolithographs of 14 beagle dogs. Platelet-rich plasma (PRP) was obtained from venous blood, and PCBM was harvested from the iliac crest. Bone defects were made bilaterally on the lower borders of the mandible. The PCBM and PRP were mixed and compressed into the defects and a custom-made HA/PLLA or a manually adopted titanium tray was fixed by screws. Tray fit and bone qualities were evaluated using computed tomography, microfocus computed tomography and confocal laser scanning microscopy. In buccal side, there is no significant difference with tray fit between the HA/PLLA and the titanium type, but in lingual side, it was better in the HA/PLLA type than that of the Ti type. Bone volume fraction (BV/TV) had markedly increased on the HA/PLLA side at 12 months. In conclusion, the custom-made HA/PLLA tray was easily and accurately adapted to the mandible, and had achieved sufficient bone quality by 12 months. Ó 2012 European Association for Cranio-Maxillo-Facial Surgery.
Keywords: Mandibular reconstruction Particulate cancellous bone and marrow Platelet-rich plasma Custom-made Bioresorbable tray
1. Introduction There are various reconstruction methods for mandibular defects, but no consensus has been reached on the optimal method (Goh et al., 2008). A titanium (Ti) tray with particulate cancellous bone and marrow (PCBM), has many advantages such as the invasiveness of the recipient bed (less than vascularized free bone grafts), and the ease of obtaining a suitable contour, which helps establish functional oral rehabilitation using implants (Carlson and Marx, 1996; Matsuo et al., 2011a,b). Although the tray is the key component in obtaining a good configuration in PCBM reconstruction, there is no agreed optimal system. Ready-made Ti trays are the most common material used (Dunback et al.,1994), but they also have some disadvantages, such as difficulty in making suitable contours for certain defects (Yamashita et al., 2008). They may also require removal, especially when dental implants are involved (Cheung et al., 1994). Recent advances in the field of bioresorbable materials could eliminate these disadvantages q No support in the form of grants was obtained for this article. * Corresponding author. Tel.: þ81 3 3342 6111x62725; fax: þ81 3 3342 1723. E-mail address: [email protected] (A. Matsuo).
(Kinoshita et al., 1997; Louis et al., 2004). Although block and granular forms of hydroxyapatite were used for mandibular reconstruction and alveolar augmentation in the 1980s, they have fallen into disuse because of the many disadvantages that result from their unresorbable characteristics, such as its high rate of infection, and failure to form a union (Bosker and Powers, 1995). A forged composite of raw particulate hydroxyapatite/poly-Llactide (HA/PLLA) is a bio-resorbable material containing hydroxyapatite. With this material, not only the PLLA, but also the HA is bio-resorbable because it is unsintered and uncalcined. The initial mechanical strength of HA/PLLA is greater than pure PLLA, because it contains HA (Shikinami and Okuno, 2001, 2005). HA/PLLA plates and screws have already been used for clinically (Kim et al., 2007, Macarini et al., 2008), but mesh sheets have been adopted for clinical use only in cardiac surgery (Ito et al., 2008). Custom-made HA/PLLA trays may provide a solution to the problems of the Ti tray in mandibular reconstruction. Recently, we reported clinical cases of custom-made HA/PLLA trays for mandibular reconstruction (Matsuo et al., 2010), but no comparative study of tray fitting and bone quality between a HA/PLLA and a Ti tray has been reported.
1010-5182/$ e see front matter Ó 2012 European Association for Cranio-Maxillo-Facial Surgery. doi:10.1016/j.jcms.2012.03.004
e454
A. Matsuo et al. / Journal of Cranio-Maxillo-Facial Surgery 40 (2012) e453ee460
To study tray fit and bone quality in PCBM reconstruction using custom-made HA/PLLA trays, we performed morphological and morphometrical evaluations using computed tomography (CT), microfocus CT (micro-CT) and confocal laser scanning microscopy (CLSM) of custom-made HA/PLLA and manually adapted Ti trays in dog models.
resorbable sutures. Antibiotics were mixed in food for 5 days after the operation. 2.4. Postoperative observations and specimen preparations
2. Materials and methods
Tetracycline (TC: 20 mg/kg) was injected in the first half and calcein (Cal: 8 mg/kg) was injected in the latter half of the observation period, as bone labelling agents, once per week.
2.1. Animals
2.5. CT evaluations
A total of 14 1.6e2-year-old healthy beagles, each weighing around 10 kg, were used for this study. The treatment of animals in this study conformed to the Association for Assessment and Accreditation of Laboratory Animal Care standards and was approved by the Tokyo Medical University Subcommittee on Research Animal Care.
CT scans were carried out on the live dogs at 1, 3, 6 and 12 months after surgery in the same conditions as presurgery. Morphological observation of the tray fit was performed first on a 3D reconstructed image. Then, 12 cases of CT images 1 month after surgery were used for morphometrical evaluation. Three trans-axial slice images, from the centre of the defect and from both mesial and distal areas 5 mm distal from the edge of the defect, were selected for measurement (Fig. 2A). Then, the distance between the inside of the tray and the outer surface of the native bone on both the buccal and lingual areas was measured on the slice, and an average of the 3 points was taken to represent data of the case (Fig. 2B). Progression of the new bone formations were sequentially evaluated in each animal on 2D transaxial sections and 3D reconstructed images. All observation and measurements were performed using the computer software Sim/Plant OMS (Materialize Dental, Leuven, Belgium). The dogs were sacrificed at 1, 3, 6 or 12 months under general anaesthesia with perfusion fixation using 10% buffered formalin. After the whole mandible was removed en bloc, reflex fixation was performed with 20% buffer formalin.
2.2. Tray preparations The HA/PLLA tray was composed of 0.8 mm-thick u-HA/PLLA mesh sheet (OsteotransÔ, Takiron Co. Ltd, Hyogo, Japan). This bioresorbable material was prepared by machining a plate, produced by a unique compression moulding process of unsintered and uncalcined hydroxyapatite (u-HA) in PLLA. The composite contained 40 wt % HA. The initial bending strength was 270 MPa, and the bending module was 7.5e9.5 GPa (Shikinami et al., 2005). Tray customization is described as follows: First, CT images of the patient’s mandible were obtained at a slice thickness of 1.25 mm (Light Speed Ultra: General Electric Company, Fairfield, CT) before surgery. A Stereolithographic model of the mandible was made from CT data after deleting artifacts (Fig. 1A). An HA/PLLA mesh sheet (Fig. 1B) was roughly trimmed, and formed while being dipped in hot water (85 C) and further adjusted, by hand, to the shape of the 3D model. After cooling for shape fixation, the tray was removed from the model, then precisely trimmed and smoothed (Fig. 1C). Finally, the customized HA/PLLA mesh tray was sterilized by ethylene oxide gas. As controls, 1.3 mm titanium mesh sheets were manually adapted to the other side of the mandible (Timesh screen 1.3: Synthes, West Chester, PA). 2.3. Surgery General anaesthesia was induced with intravenous pentobarbiturate (0.5 mg/kg). Prior to surgery, 20 ml of venous blood was collected to produce platelet-rich plasma (PRP) and autonomous thrombin. PRP was separated using a centrifuge, (Smart Prep System, Harvest Technologies Corp., Norwell, MA). It was programmed for an initial spin of 2500 rpm for 10 min, followed by a 1-min interval, and then an additional spin of 2300 rpm for 3 min. A 10 20-mm bone block was harvested from the right iliac crest. The block was fragmented into particulated cortical bone and cancellous marrow. The mandible was then exposed from the lower border of the mandible using an extraoral approach and the periosteum was preserved on both the buccal and lingual sides. Then, 10 10 mm-bone defects were formed bilaterally on the lower border of the mandible in the premolar area (Fig. 1D) and PCBM, PRP and the autonomous thrombin were mixed and transferred to the tray. A custom-made bioresorbable tray was adapted to fit the defect on the right side. This process was easily done, simply by cutting tray the redundant tray area using scissors. The tray was then fixed by 8 HA/PLLA screws 2.0 mm in diameter from the buccal side (Fig. 1E). On the other side, the titanium tray was manually adapted to the mandible. A square form of the ready-made titanium mesh sheet was cut and bent into the site. Then, PCBM and PRP were compressed into the tray, and fixed by 8 1.5-mm Ti screws. The wound was closed using 3-0
2.6. Micro-CT evaluations The reconstructed area of the mandible was excised and the Ti tray was removed from the mandible, avoiding the artefact on the microCT image and the need to prepare the specimen. The micro-CT was performed using a SkyScan system (Desktop Micro, CT 1072 Kontch, Belgium). This is based on fan-beam tomography and has a multislice mode. Scanning was conducted at 100 kVp and 98 mA, acquiring consecutive tomographic slices with a slice thickness of 20 mm. Digital data were then reconstructed to obtain CT images with a pixel size of 18.2 mm in a 512 512 matrix. Bone microstructures were measured using computer software (Tri/3D Bon, Ratoc Co. Ltd., Tokyo, Japan). Using 3D reconstruction images by volume rendering method, 2 volume of interest (VOI) areas were selected for morphometrical analysis. The VOIs were independently based on the new bone formation areas, and the native bone areas between the root apex and the border of the defect (Fig. 2B). The threshold was automatically set by the software. Bone volume (BV) was calculated using tetrahedrons corresponding to the enclosed volume of the triangulated surface. Total tissue volume (TV) was analysed. The bone volume fraction (BV/ TV) was determined by the formula 100 BV/TV. 2.7. CLSM evaluations A specimen was cut from the centre of the defects into 2 segments, and non-decalcified ground sections were made from the halves using the Exakt Micro Grinding System (Exakt Apparatebau, Norderstedt, Germany), similar to our previous study (Takeuchi et al., 2010). We scanned the ground sections at 20 magnification using CLSM (Zeiss LSM 510, Carl Zeiss AG, Oberkochen, Germany). A whole transaxial composite image was integrated from several images. In the CLSM images, the TC labelling area representing the first half of the observation period was coloured by red fluorescence, and the Cal labelling area representing
A. Matsuo et al. / Journal of Cranio-Maxillo-Facial Surgery 40 (2012) e453ee460
e455
Fig. 1. Customized method of the HA/PLLA mesh tray and surgery. (A) Stereolithograph made from CT data. An HA/PLLA mesh sheet (B) was formed into the shape of the tray according to the mandible model (C). (D) A bone defect on the lower border of the mandible. (E) The particulate cancellous bone and marrow (PCBM) and Platelet-rich plasma (PRP) were mixed and compressed into the try, and a HA/PLLA tray was fixed with screws ( ).
Fig. 2. Morphometrical measurements. (A) Sagittal view of the mandible. M: 5-mm mesial line from the edge of the defect. C: centre of the defect. D: 5-mm distal line from the edge ) Tray fit was measured between the native bone and the tray. New bone (NB) was excluded from the of the defect. ( ) Defect. (B) transaxial view of the mandible. ( measurements. ( ) VOI was set independently in the new bone (NB) and native bone (B) for 3D microstructural measurement.
the latter half of the observation period was coloured by green fluorescence. 2.8. Histology Residual halves were decalcified by 10% EDTA and embedded in paraffin. Histological sections were stained with haematoxylin and eosin.
2.9. Statistical analysis All data were presented as means standard deviation (SD). Student’s t-test was used for all statistical evaluations. All analyses were performed using statistical software (SPSS version 11.5 J; SAS Institute; Cary, NC). P values less than 0.05 were considered to represent statistically significant differences.
e456
A. Matsuo et al. / Journal of Cranio-Maxillo-Facial Surgery 40 (2012) e453ee460
3. Results
3.2. Progression of bone formation in dog models
3.1. Evaluations of the tray fit
At 1 month after surgery, new bone had formed near the native bone on both sides, but had not completely filled the defect. At 3 months, new bone formation had progressed and the shape of the outer surface was becoming defined, but there were still large gaps between the outer surface and the tray on both sides. Bone remodelling gradually progressed towards the tray until month 12 (Fig. 4).
On 3D CT image observation, tray fit was better in the custommade HA/PLLA trays than in the Ti trays. Visible gaps between trays and the surrounding bone were seen only in the Ti tray cases. On CT morphometrical measurements, the average gap between the native bone and the HA/PLLA tray was 0.12 0.13 mm on the buccal side and 0.78 0.78 mm on the lingual side. The average gap was 0.13 0.18 mm on the buccal side and 1.97 0.76 mm on the lingual side between the native bone and the Ti tray. Comparing the HA/PLLA and the Ti tray, the gaps were statistically significantly larger with the Ti tray than with the HA/PLLA tray on the lingual side (P < 0.05), but there was no significant difference on the buccal side between these two groups (Fig. 3).
3.3. Micro CT and CLSM images At 1 month after surgery, new bone formations were progressing from the native bone, and had adhered to the crushed bone on micro-CT images. On CLSM images, strong fluorescent findings corresponded to the new bone formation areas. Weak fluorescent
Fig. 3. (A) Tray fit. 3D reconstructed CT images 6 months after surgery, showing gaps between the tray and native bone. ( ) (B) Comparisons of the tray fit between the HA/PLLA and the Ti tray. *P < 0.05.
Fig. 4. Progression of bone formation. Images show progression of bone formation in an animal on transaxial sectional CT images. New bone formation ( ) began close to the new bone (1 month), and progressed towards the tray side (3e12 months).
A. Matsuo et al. / Journal of Cranio-Maxillo-Facial Surgery 40 (2012) e453ee460
e457
findings were seen around the bone chips away from the new bone formation area (Fig. 5, arrow). At 3e6 months, isolated crushed bone segments were incorporated into the new bone, and the outer surface of the new bone formation area was defined, but a large gap was seen between the tray and the bone surface on both sides. Both TC and Cal were irregularly labelled in the new bone formation area. Cortical and cancellous bone were clearly defined by 12 months, Cal labelling in lamellar structures was piled up at the outer surface, especially on the HA/PLLA side, and the gap between the tray and the bone surface had decreased on both sides (Fig. 5, triangle). There were no marked differences in bone formation or structures between the lingual and buccal sides in either tray.
was 54.4 19.4% and 74.2 6.4% (N ¼ 4). At 12 months, the average BV/TV of the new bone on the HA/PLLA and Ti sides was 74.7 5.4% and 77.4 3.4% respectively, and the values of the native bone on both sides were 53.0 8.1% and 47.4 8.2%, respectively (N ¼ 3). At 1 month and 12 months, no statistically significant difference in BV/TV between the HA/PLLA and Ti side was seen, but it was significantly higher on the Ti side at 3 and 6 months (P < 0.05: Fig. 6). At 12 months, BV/TV of the new bone in both sides was significantly higher than that of native bone sites (P < 0.05).
3.4. Bone morphometrical evaluations
At 1 month, new bone was lined with osteoclasts and osteoblasts. However, non-vital crushed bone was also observed in a distant area of the new bone. At 12 months, no foreign bodies were detected around the tray, and all new bone areas were filled with the vital bone (Fig. 7).
At 1 month after surgery, the average BV/TV of the HA/PLLA and Ti sides was 46.4 16.9% and 32.2 7.2%, respectively (N ¼ 3). At 3 months, it was 44.5 13.2% and 62.4 11.6% (N ¼ 4). At 6 months, it
3.5. Histological findings
Fig. 5. Micro CT and CSLM images. Transaxial images of the HA/PLLA and titanium tray. The left side shows micro-CT and the right side shows CLSM images for each. In CLSM images, tetracycline is labelled by red and calcein is green. (B) Native bone. (NB) New bone. (CB) Crushed bone. Ti tray ( ). Borders between the native and new bone (---). Upper: 1 month after reconstruction. ( ) Weak labelling around the bone chips. Middle: 6 months after reconstruction. Lower: 12 months after reconstruction. ( ) Lamellar labelling at the outer surface.
e458
A. Matsuo et al. / Journal of Cranio-Maxillo-Facial Surgery 40 (2012) e453ee460
4. Discussion We set out to determine the tray fit and bone quality of PCBM mandibular reconstruction using custom-made bioresorbable HA/ PLLA tray. Surgeons find it difficult to manipulate a titanium mesh tray to achieve a suitable contour that will fit large defects. This technique is also time-consuming and decreases mechanical strength (Yamashita et al., 2008). A fracture after PCBM mandibular reconstruction has been reported (Aytak et al., 2005). Custommade Ti trays using rapid prototyping can solve this problem, and clinical trials have already been performed on orbital floor fractures (Metzger et al., 2007). In mandibular reconstruction, a case in which a Ti tray was pre-bent and reinforced by laser welding has also been reported (Yamashita et al., 2008). However, this is not a common technique in the clinical setting because of technical difficulties. In addition, Ti trays sometimes require removal, because they can hinder implant insertion (Cheung et al., 1994). Bioresorbable materials may be the best option to solve this problem. Initially PGA plates or screws were clinically used as the
Fig. 6. Changes of the bone volume fraction after reconstruction between the HA/PLLA and Ti sides. *P < 0.05.
bioabsorbable osteosynthesis material (Bostman et al., 1987). However, a high prevalence of foreign body reaction was reported (Bostman et al., 1990). Then, many experimental and clinical studies of PLLA plates or screws reported good results, with little foreign body reaction (Bessho et al., 1997; Eppley et al., 1997). Pure PLLA (Kinoshita et al., 1997) and DLA/PLLA (Louis et al., 2004) have been reported for use in mandibular reconstruction, but an important point for successful PCBM reconstruction is rigid fixation (Kruger and Krumholz, 1984), which is incompatible with these methods. We previously reported 2 PCBM reconstruction cases using custom-made HA/PLLA trays with rigid fixation (Matsuo et al., 2010), but the custom-made tray fit was not validated. In this study, there was no significant difference in tray fit between the HA/PLLA and the Ti on the buccal side, but on the lingual side fit was better in the HA/PLLA type than in the Ti type. On the buccal side, the difference in gap size was not significant between the Ti and HA/PLLA tray because the screws were inserted from the buccal side. However, an initially unseen gap formed between the tray and the mandible because the lingual side was not fixed with screws. These findings suggest that customized trays may be more efficient for fitting to the lingual side. In addition, bone formation was not affected by the large size of the gap. Therefore, we can expect the custom-made resorbable trays to yield better bone formation in the lingual area and reduce operating times, especially in clinically compromised cases with markedly curved or deep lingual areas, for which manual bending is difficult. Regarding bone quality of the HA/PLLA tray, both bone conduction and induction were more active in the PCBM graft than in the block bone graft (Lieberman and Friedlaender, 2005). In the present study, fluorescent labelling around the bone chips were seen early, even in the distant part from the new bone formation area. This suggests that bone induction is actively initiated early around the PCBM. From 3 to 6 months after surgery, the outer configuration of the new bone formation area was observed. This is
Fig. 7. Histological findings. Figure indicates the HeE staining of the HA/PLLA side. (A) New bone formation area at 1 month. ( ) Osteoblast, ( ) osteoclast. (B) Bone chips ( ) away from the bone formation area at 1 month. (C) Around the tray ( ) at 12 months. (D) Outer surface ( ) of the new bone areas at 12 months.
A. Matsuo et al. / Journal of Cranio-Maxillo-Facial Surgery 40 (2012) e453ee460
similar to previous pure PLLA tray (Kinoshita et al., 1997) and autogenous scaffold findings (Fennis et al., 2002), but no study has evaluated bone quality of the bone remodelling period 3 months after surgery. Although clinically sufficient amounts of bone formation formed in PCBM reconstruction using a Ti tray, its X-ray opacity is low in many cases compared with that of the surrounding native bone more than 1 year after reconstruction in clinical studies (Marx and Garg, 1998; Iwamoto et al., 2009; Matsuo et al., 2011a,b). On the other hand, a clinical study of mandibular reconstruction using HA/PLLA trays showed that the CT value of the reconstructed bone was higher with Ti trays, and was similar to native bone (Matsuo et al., 2010). This finding was obtained even in marginal resection cases, which had not reached critical defect size. Using HA/PLLA we expect rapid bone growth because it contains HA (Shikinami et al., 2005). To confirm this we did not need to make critical size defects in this study, and our aim was mainly evaluation during the remodelling period 3 months after surgery. In addition, we used histological, histomorphometrical and time labelling methods to evaluate new bone quality. In this study, the outer surface of the new bone formation area was defined by 6 months on both sides. However, the BV/TV of the HA/PLLA trays was significantly lower than in the Ti tray at 3 and 6 months. These findings indicate that bone remodelling was delayed on the HA/PLLA side until 6 months. However, after 6 months, new bone formation in the lamellar structure labelled by Cal progressed actively on the outer surface towards the tray, and BV/TV had quickly increased by 12 months, especially on the HA/PLLA tray in a morphometrical study. This suggests that the pattern of bone remodelling was initially relatively different between HA/PLLA and Ti trays, but similar bone qualities were obtained by 12 months, both of which were higher than that of native bone. We should note that these results may be affected by both tray materials and bending methods. It has been reported that HA/PLLA tray resorption becomes active after 1 year, and is resorbed approximately 3e5 years after surgery (Shikinami et al., 2005). In a pure PLLA tray, resorption of the tray had already begun 12 months after reconstruction (Kinoshita et al., 1997). In the present study, there was no foreign body reaction and no tray resorption at 12 months after reconstruction. In mandibular reconstruction, the reconstructed bone requires sufficient strength, especially in continuity defects, because a large occlusal force is loaded onto the mandible (Tie et al., 2006). However, the mechanical properties of the cancellous bone graft are lower than those of a block bone graft (Lieberman and Friedlaender, 2005). Clinically, the tray is removed at least 6 months after surgery, usually after 1 year, based on consideration of the mechanical properties. If implants are inserted early, the alveolar part of the HA/PLLA alone tray can be easily removed without additional invasive surgery. In general, bone strength is proportionate to bone microstructures (Seemann, 2002). In PCBM reconstruction using HA/PLLA trays, sufficient structure of new bones was obtained, and the HA/PLLA tray also maintained the sufficient strength by 12 months. We therefore think that delaying bone formation for 3e6 months in a HA/PLLA tray does not effect the clinical outcome. The results of this study regarding bone quality were limited to small defects using an extraoral approach. Further studies should evaluate bone strength, tray strength and infection resistibility of HA/PLLA trays in continuity defects using an intraoral approach. This study included several limitations. First, our study did not include critical-size continuity defects as the main purpose of this study was to evaluate tray fit. Therefore, the contact area between the tray and native bone had to be made as large as possible. Secondly, we observed the animals until 12 months after reconstruction, but the tray was mainly resorbed after 1 year (Shikinami
e459
et al., 2005). We should further investigate long-term follow-up and tray evaluation. Finally, we used PRP according to our clinical PCBM reconstruction protocol (Iwamoto et al., 2009; Matsuo et al., 2011a). Previously, PRP was said to promote bone formation in PCBM reconstruction (Marx et al., 1998), however recent animal studies have indicated contradictory results. Choi et al. (2004) reported inhibition, but Fennis et al. reported promotion (Fennis et al., 2002). We believe that this did not affect our results, because PRP was used on both sides and sufficient quality of bone formation was seen on both sides in the present study. 5. Conclusion The customized HA/PLLA tray was easily and accurately adapted to the mandible, and had achieved sufficient bone quality by 12 months after the reconstruction. Disclosure The authors have no financial conflicts of interest with regard to this study. Acknowledgements The authors are indebted to Mr. Roderick J. Turner and Professor J. Patrick Barron, Chairman of the Department of the International Medical Communications of Tokyo Medical University, for their editorial review of this manuscript. References Aytak C, Ozbec S, Kahveci R, Ozgenel Y, Akin S, Ozcan M: Titanium mesh fracture in mandibular reconstruction. J Craniofac Surg 16: 1120e1122, 2005 Bessho K, Iizuka T, Murakami K: A bioabsorbable poly-L-lactide miniplate and screw system for osteosynthesis in oral and maxillofacial surgery. J Oral Maxillofac Surg 55: 941e946, 1997 Bosker H, Powers MP: TMI reconstruction system. In: Fonseca RJ, Davis WH (eds), Reconstructive preprosthetic oral and maxillofacial surgery, 2nd edn. Philadelphia, PA: Saunders, 1995 Bostman O, Vainionpaa S, Hirvensalo E, Makela A, Vihtonen K, Tormala P, et al: A prospective randomized trial. J Bone Joint Surg [Br] 69-B: 615e619, 1987 Bostman O, Hirvensalo E, Makinen J, Rokkanen P: Foreign-body reactions to fracture fixation implants of biodegradable synthetic polymers. J Bone Joint Surg [Br] 72-B: 592e596, 1990 Carlson ER, Marx RE: Mandibular reconstruction using cancellous cellular bone grafts. J Oral Maxillofac Surg 54: 889e897, 1996 Cheung LK, Samman N, Tong ACK, Tiedman AH: Mandibular reconstruction with the Dacron ulethane tray. A radiological assessment of bone remodeling. J Oral Maxillofac Surg 52: 373e385, 1994 Choi BH, Huh JY, Suh, Lee SH: Effect of platelet-rich plasma on bone regeneration in autogenous bone graft. Int J Oral Maxillofac Surg 33: 56e59, 2004 Dunback J, Rodemer H, Spitzer WJ, Steinhauser EW: Mandibular reconstruction with cancellous bone, hydroxyapatite and titanium mesh. J Cranio-Maxillofac Surg 22: 151e155, 1994 Eppley BL, Sadove AM, Havlik RJ: Resorbable plate fixation in pediatric craniofacial surgery. Plast Reconstr Surg 100: 1e7, 1997 Fennis JP, Stoelinga PJ, Jansen JA: Mandibular reconstruction: a clinical and radiographic animal study on the use of autogenous scaffolds and platelet-rich plasma. Int J Oral Maxillofac Surg 31: 281e286, 2002 Goh BT, Lee S, Tiedman H, Stoelinga PJW: Mandibular reconstruction in adults: a review. Int J Oral Maxillofac Surg 37: 597e605, 2008 Ito T, Kudo M, Yozu R: Usefulness of osteosynthesis device made of hydroxyapatitepoly-L-Lactide composites in Port-Access cardiac surgery. Ann Thorac Surg 86: 1905e1908, 2008 Iwamoto I, Matsuo A, Kato N, Takeuchi S, Takahashi H, Hojo S, et al: Computed tomographic evaluation of bone quality of the mandible reconstructed by particular cellular bone and marrow combined with platelet rich plasma. Oral Science International 6: 63e72, 2009 Kim K, Isu T, Sugawara A, Matsumoto R, Isobe M: Utility of new bioabsorptive screws in cervical anterior fusion. Surg Neurol 68: 264e268, 2007 Kinoshita Y, Kobayashi M, Hidaka T, Ikeda Y: Reconstruction of mandibular continuity defects in dogs using Poly (L-lactide) mesh and autogenic particulate cancellous bone and marrow: preliminary report. J Oral Maxillofac Surg 55: 718e723, 1997
e460
A. Matsuo et al. / Journal of Cranio-Maxillo-Facial Surgery 40 (2012) e453ee460
Kruger E, Krumholz K: Results of bone grafting after rigid fixation. J Oral Maxillofac Surg 42: 491e496, 1984 Lieberman JR, Friedlaender GE: Bone regeneration and repair. Biology and clinical applications. Totowa, New Jersey: Humana Press, pp. 57 65, 2005 Marx RE, Garg AK: Bone structure, metabolism, and physiology: its impact on dental implantology. Implant Dentistry 7: 267e276, 1998 Louis P, Holmes J, Fernandes R: Resorbable mesh as a containment system in reconstruction of the atrophic mandible fracture. J Oral Maxillofac Surg 62: 719e723, 2004 Macarini L, Milillo P, Mocci A, Vinci R, Ettorre GC: Poly lactic acid-hydroxyapatite (HA/PLLA) bioabsorbable interference screws for tibial graft fixation in anterior cruciate ligament (ACL) reconstruction surgery: MR evaluation of osseointegration and degradation features. Radiol Med 113: 1185e1197, 2008 Marx RE, Carlson ER, Eichstaedt RM, Schimmele SR, Georgeff KR: Platelet-rich plasma. Growth factor enhancement for bone graft. Oral Surg Oral Med Oral Pathol Oral Rad Enodl 85: 638e646, 1998 Matsuo A, Chiba H, Takahashi H, Toyoda J, Abukawa H: Clinical application of a custom-made bioresorbable raw particulate hydroxyapatite/poly-L-lactide mesh tray for mandibular reconstruction. Odontology 98: 85e88, 2010 Matsuo A, Chiba H, Toyoda J, Abukawa H, Fujikawa K, Tsuzuki M: Mandibular reconstruction using a tray with particulate cancellous bone and marrow and platelet rich plasma by an intraoral approach. J Oral Maxillofac Surg 69: 1807e1814, 2011a
Matsuo A, Chiba H, Takahashi H, Toyoda J, Hasegawa O, Hojo S: Bone quality of mandibles reconstructed with particulate cellular bone and marrow, and platelet-rich plasma. J Cranio-Maxillofac Surg 39: 628e632, 2011b Metzger MC, Schon R, Schmelzeisen R: Preformed titanium meshes: a new standard? Skull Base 17: 269e272, 2007 Seemann E: Pathogenesis of bone fragility in women and men. Lancet 359: 1841e1850, 2002 Shikinami Y, Okuno M: Bioresorbable devices made of forged composites of hydroxyapatite(HA) particles and poly lactide (PLLA). Part II: practical properties of miniscrews and miniplates. Biomaterials 22: 3197e3211, 2001 Shikinami Y, Matsusue Y, Nakamura T: The complete process of bioresorption and bone replacement using devices made of forged composites of raw hydroxyapatite particles/poly lactide (F-u-HA/PLLA). Biomaterials 26: 5542e5559, 2005 Takeuchi S, Matsuo A, Chiba H: Beneficial role of periosteum in distraction ontogenesis of mandible: its preservation prevents the external bone resorption. Tohoku J Exp Med 220: 67e75, 2010 Tie Y, Wang DM, Ji T, Wang CT, Zhang CP: Three-dimensional finite-element analysis investigating the biomechanical effects of human mandibular reconstruction with autogenous bone grafts. J Cranio-Maxillofac Surg 34: 290e298, 2006 Yamashita Y, Yamaguchi Y, Tsuji M, Shigematsu M, Goto M: Mandibular reconstruction using autologous iliac bone and Titanium mesh reinforced by laser welding for implant placement. Int J Oral Maxillofac Implants 23: 1143e1146, 2008
Contents lists available at SciVerse ScienceDirect
Journal of Cranio-Maxillo-Facial Surgery journal homepage: www.jcmfs.com
Application of custom-made bioresorbable raw particulate hydroxyapatite/ poly-L-lactide mesh tray with particulate cellular bone and marrow and platelet-rich plasma for a mandibular defect: Evaluation of tray fit and bone quality in a dog modelq Akira Matsuo*, Hidetoshi Takahashi, Harutsugi Abukawa, Daichi Chikazu Department of Oral and Maxillofacial Surgery, Tokyo Medical University, 6-7-1 Nishishinjuku Shinjuku-ku, Tokyo 160-0023, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Paper received 21 April 2011 Accepted 6 March 2012
The purpose of this study was to evaluate tray fit and bone quality of particulate cancellous bone and marrow (PCBM) mandibular reconstruction using custom-made bioresorbable forged composites of a raw particulate hydroxyapatite/poly-L-lactide (HA/PLLA) tray in a dog model. Mesh sheets of HA/PLLA were formed in a tray shape according to the mandible stereolithographs of 14 beagle dogs. Platelet-rich plasma (PRP) was obtained from venous blood, and PCBM was harvested from the iliac crest. Bone defects were made bilaterally on the lower borders of the mandible. The PCBM and PRP were mixed and compressed into the defects and a custom-made HA/PLLA or a manually adopted titanium tray was fixed by screws. Tray fit and bone qualities were evaluated using computed tomography, microfocus computed tomography and confocal laser scanning microscopy. In buccal side, there is no significant difference with tray fit between the HA/PLLA and the titanium type, but in lingual side, it was better in the HA/PLLA type than that of the Ti type. Bone volume fraction (BV/TV) had markedly increased on the HA/PLLA side at 12 months. In conclusion, the custom-made HA/PLLA tray was easily and accurately adapted to the mandible, and had achieved sufficient bone quality by 12 months. Ó 2012 European Association for Cranio-Maxillo-Facial Surgery.
Keywords: Mandibular reconstruction Particulate cancellous bone and marrow Platelet-rich plasma Custom-made Bioresorbable tray
1. Introduction There are various reconstruction methods for mandibular defects, but no consensus has been reached on the optimal method (Goh et al., 2008). A titanium (Ti) tray with particulate cancellous bone and marrow (PCBM), has many advantages such as the invasiveness of the recipient bed (less than vascularized free bone grafts), and the ease of obtaining a suitable contour, which helps establish functional oral rehabilitation using implants (Carlson and Marx, 1996; Matsuo et al., 2011a,b). Although the tray is the key component in obtaining a good configuration in PCBM reconstruction, there is no agreed optimal system. Ready-made Ti trays are the most common material used (Dunback et al.,1994), but they also have some disadvantages, such as difficulty in making suitable contours for certain defects (Yamashita et al., 2008). They may also require removal, especially when dental implants are involved (Cheung et al., 1994). Recent advances in the field of bioresorbable materials could eliminate these disadvantages q No support in the form of grants was obtained for this article. * Corresponding author. Tel.: þ81 3 3342 6111x62725; fax: þ81 3 3342 1723. E-mail address: [email protected] (A. Matsuo).
(Kinoshita et al., 1997; Louis et al., 2004). Although block and granular forms of hydroxyapatite were used for mandibular reconstruction and alveolar augmentation in the 1980s, they have fallen into disuse because of the many disadvantages that result from their unresorbable characteristics, such as its high rate of infection, and failure to form a union (Bosker and Powers, 1995). A forged composite of raw particulate hydroxyapatite/poly-Llactide (HA/PLLA) is a bio-resorbable material containing hydroxyapatite. With this material, not only the PLLA, but also the HA is bio-resorbable because it is unsintered and uncalcined. The initial mechanical strength of HA/PLLA is greater than pure PLLA, because it contains HA (Shikinami and Okuno, 2001, 2005). HA/PLLA plates and screws have already been used for clinically (Kim et al., 2007, Macarini et al., 2008), but mesh sheets have been adopted for clinical use only in cardiac surgery (Ito et al., 2008). Custom-made HA/PLLA trays may provide a solution to the problems of the Ti tray in mandibular reconstruction. Recently, we reported clinical cases of custom-made HA/PLLA trays for mandibular reconstruction (Matsuo et al., 2010), but no comparative study of tray fitting and bone quality between a HA/PLLA and a Ti tray has been reported.
1010-5182/$ e see front matter Ó 2012 European Association for Cranio-Maxillo-Facial Surgery. doi:10.1016/j.jcms.2012.03.004
e454
A. Matsuo et al. / Journal of Cranio-Maxillo-Facial Surgery 40 (2012) e453ee460
To study tray fit and bone quality in PCBM reconstruction using custom-made HA/PLLA trays, we performed morphological and morphometrical evaluations using computed tomography (CT), microfocus CT (micro-CT) and confocal laser scanning microscopy (CLSM) of custom-made HA/PLLA and manually adapted Ti trays in dog models.
resorbable sutures. Antibiotics were mixed in food for 5 days after the operation. 2.4. Postoperative observations and specimen preparations
2. Materials and methods
Tetracycline (TC: 20 mg/kg) was injected in the first half and calcein (Cal: 8 mg/kg) was injected in the latter half of the observation period, as bone labelling agents, once per week.
2.1. Animals
2.5. CT evaluations
A total of 14 1.6e2-year-old healthy beagles, each weighing around 10 kg, were used for this study. The treatment of animals in this study conformed to the Association for Assessment and Accreditation of Laboratory Animal Care standards and was approved by the Tokyo Medical University Subcommittee on Research Animal Care.
CT scans were carried out on the live dogs at 1, 3, 6 and 12 months after surgery in the same conditions as presurgery. Morphological observation of the tray fit was performed first on a 3D reconstructed image. Then, 12 cases of CT images 1 month after surgery were used for morphometrical evaluation. Three trans-axial slice images, from the centre of the defect and from both mesial and distal areas 5 mm distal from the edge of the defect, were selected for measurement (Fig. 2A). Then, the distance between the inside of the tray and the outer surface of the native bone on both the buccal and lingual areas was measured on the slice, and an average of the 3 points was taken to represent data of the case (Fig. 2B). Progression of the new bone formations were sequentially evaluated in each animal on 2D transaxial sections and 3D reconstructed images. All observation and measurements were performed using the computer software Sim/Plant OMS (Materialize Dental, Leuven, Belgium). The dogs were sacrificed at 1, 3, 6 or 12 months under general anaesthesia with perfusion fixation using 10% buffered formalin. After the whole mandible was removed en bloc, reflex fixation was performed with 20% buffer formalin.
2.2. Tray preparations The HA/PLLA tray was composed of 0.8 mm-thick u-HA/PLLA mesh sheet (OsteotransÔ, Takiron Co. Ltd, Hyogo, Japan). This bioresorbable material was prepared by machining a plate, produced by a unique compression moulding process of unsintered and uncalcined hydroxyapatite (u-HA) in PLLA. The composite contained 40 wt % HA. The initial bending strength was 270 MPa, and the bending module was 7.5e9.5 GPa (Shikinami et al., 2005). Tray customization is described as follows: First, CT images of the patient’s mandible were obtained at a slice thickness of 1.25 mm (Light Speed Ultra: General Electric Company, Fairfield, CT) before surgery. A Stereolithographic model of the mandible was made from CT data after deleting artifacts (Fig. 1A). An HA/PLLA mesh sheet (Fig. 1B) was roughly trimmed, and formed while being dipped in hot water (85 C) and further adjusted, by hand, to the shape of the 3D model. After cooling for shape fixation, the tray was removed from the model, then precisely trimmed and smoothed (Fig. 1C). Finally, the customized HA/PLLA mesh tray was sterilized by ethylene oxide gas. As controls, 1.3 mm titanium mesh sheets were manually adapted to the other side of the mandible (Timesh screen 1.3: Synthes, West Chester, PA). 2.3. Surgery General anaesthesia was induced with intravenous pentobarbiturate (0.5 mg/kg). Prior to surgery, 20 ml of venous blood was collected to produce platelet-rich plasma (PRP) and autonomous thrombin. PRP was separated using a centrifuge, (Smart Prep System, Harvest Technologies Corp., Norwell, MA). It was programmed for an initial spin of 2500 rpm for 10 min, followed by a 1-min interval, and then an additional spin of 2300 rpm for 3 min. A 10 20-mm bone block was harvested from the right iliac crest. The block was fragmented into particulated cortical bone and cancellous marrow. The mandible was then exposed from the lower border of the mandible using an extraoral approach and the periosteum was preserved on both the buccal and lingual sides. Then, 10 10 mm-bone defects were formed bilaterally on the lower border of the mandible in the premolar area (Fig. 1D) and PCBM, PRP and the autonomous thrombin were mixed and transferred to the tray. A custom-made bioresorbable tray was adapted to fit the defect on the right side. This process was easily done, simply by cutting tray the redundant tray area using scissors. The tray was then fixed by 8 HA/PLLA screws 2.0 mm in diameter from the buccal side (Fig. 1E). On the other side, the titanium tray was manually adapted to the mandible. A square form of the ready-made titanium mesh sheet was cut and bent into the site. Then, PCBM and PRP were compressed into the tray, and fixed by 8 1.5-mm Ti screws. The wound was closed using 3-0
2.6. Micro-CT evaluations The reconstructed area of the mandible was excised and the Ti tray was removed from the mandible, avoiding the artefact on the microCT image and the need to prepare the specimen. The micro-CT was performed using a SkyScan system (Desktop Micro, CT 1072 Kontch, Belgium). This is based on fan-beam tomography and has a multislice mode. Scanning was conducted at 100 kVp and 98 mA, acquiring consecutive tomographic slices with a slice thickness of 20 mm. Digital data were then reconstructed to obtain CT images with a pixel size of 18.2 mm in a 512 512 matrix. Bone microstructures were measured using computer software (Tri/3D Bon, Ratoc Co. Ltd., Tokyo, Japan). Using 3D reconstruction images by volume rendering method, 2 volume of interest (VOI) areas were selected for morphometrical analysis. The VOIs were independently based on the new bone formation areas, and the native bone areas between the root apex and the border of the defect (Fig. 2B). The threshold was automatically set by the software. Bone volume (BV) was calculated using tetrahedrons corresponding to the enclosed volume of the triangulated surface. Total tissue volume (TV) was analysed. The bone volume fraction (BV/ TV) was determined by the formula 100 BV/TV. 2.7. CLSM evaluations A specimen was cut from the centre of the defects into 2 segments, and non-decalcified ground sections were made from the halves using the Exakt Micro Grinding System (Exakt Apparatebau, Norderstedt, Germany), similar to our previous study (Takeuchi et al., 2010). We scanned the ground sections at 20 magnification using CLSM (Zeiss LSM 510, Carl Zeiss AG, Oberkochen, Germany). A whole transaxial composite image was integrated from several images. In the CLSM images, the TC labelling area representing the first half of the observation period was coloured by red fluorescence, and the Cal labelling area representing
A. Matsuo et al. / Journal of Cranio-Maxillo-Facial Surgery 40 (2012) e453ee460
e455
Fig. 1. Customized method of the HA/PLLA mesh tray and surgery. (A) Stereolithograph made from CT data. An HA/PLLA mesh sheet (B) was formed into the shape of the tray according to the mandible model (C). (D) A bone defect on the lower border of the mandible. (E) The particulate cancellous bone and marrow (PCBM) and Platelet-rich plasma (PRP) were mixed and compressed into the try, and a HA/PLLA tray was fixed with screws ( ).
Fig. 2. Morphometrical measurements. (A) Sagittal view of the mandible. M: 5-mm mesial line from the edge of the defect. C: centre of the defect. D: 5-mm distal line from the edge ) Tray fit was measured between the native bone and the tray. New bone (NB) was excluded from the of the defect. ( ) Defect. (B) transaxial view of the mandible. ( measurements. ( ) VOI was set independently in the new bone (NB) and native bone (B) for 3D microstructural measurement.
the latter half of the observation period was coloured by green fluorescence. 2.8. Histology Residual halves were decalcified by 10% EDTA and embedded in paraffin. Histological sections were stained with haematoxylin and eosin.
2.9. Statistical analysis All data were presented as means standard deviation (SD). Student’s t-test was used for all statistical evaluations. All analyses were performed using statistical software (SPSS version 11.5 J; SAS Institute; Cary, NC). P values less than 0.05 were considered to represent statistically significant differences.
e456
A. Matsuo et al. / Journal of Cranio-Maxillo-Facial Surgery 40 (2012) e453ee460
3. Results
3.2. Progression of bone formation in dog models
3.1. Evaluations of the tray fit
At 1 month after surgery, new bone had formed near the native bone on both sides, but had not completely filled the defect. At 3 months, new bone formation had progressed and the shape of the outer surface was becoming defined, but there were still large gaps between the outer surface and the tray on both sides. Bone remodelling gradually progressed towards the tray until month 12 (Fig. 4).
On 3D CT image observation, tray fit was better in the custommade HA/PLLA trays than in the Ti trays. Visible gaps between trays and the surrounding bone were seen only in the Ti tray cases. On CT morphometrical measurements, the average gap between the native bone and the HA/PLLA tray was 0.12 0.13 mm on the buccal side and 0.78 0.78 mm on the lingual side. The average gap was 0.13 0.18 mm on the buccal side and 1.97 0.76 mm on the lingual side between the native bone and the Ti tray. Comparing the HA/PLLA and the Ti tray, the gaps were statistically significantly larger with the Ti tray than with the HA/PLLA tray on the lingual side (P < 0.05), but there was no significant difference on the buccal side between these two groups (Fig. 3).
3.3. Micro CT and CLSM images At 1 month after surgery, new bone formations were progressing from the native bone, and had adhered to the crushed bone on micro-CT images. On CLSM images, strong fluorescent findings corresponded to the new bone formation areas. Weak fluorescent
Fig. 3. (A) Tray fit. 3D reconstructed CT images 6 months after surgery, showing gaps between the tray and native bone. ( ) (B) Comparisons of the tray fit between the HA/PLLA and the Ti tray. *P < 0.05.
Fig. 4. Progression of bone formation. Images show progression of bone formation in an animal on transaxial sectional CT images. New bone formation ( ) began close to the new bone (1 month), and progressed towards the tray side (3e12 months).
A. Matsuo et al. / Journal of Cranio-Maxillo-Facial Surgery 40 (2012) e453ee460
e457
findings were seen around the bone chips away from the new bone formation area (Fig. 5, arrow). At 3e6 months, isolated crushed bone segments were incorporated into the new bone, and the outer surface of the new bone formation area was defined, but a large gap was seen between the tray and the bone surface on both sides. Both TC and Cal were irregularly labelled in the new bone formation area. Cortical and cancellous bone were clearly defined by 12 months, Cal labelling in lamellar structures was piled up at the outer surface, especially on the HA/PLLA side, and the gap between the tray and the bone surface had decreased on both sides (Fig. 5, triangle). There were no marked differences in bone formation or structures between the lingual and buccal sides in either tray.
was 54.4 19.4% and 74.2 6.4% (N ¼ 4). At 12 months, the average BV/TV of the new bone on the HA/PLLA and Ti sides was 74.7 5.4% and 77.4 3.4% respectively, and the values of the native bone on both sides were 53.0 8.1% and 47.4 8.2%, respectively (N ¼ 3). At 1 month and 12 months, no statistically significant difference in BV/TV between the HA/PLLA and Ti side was seen, but it was significantly higher on the Ti side at 3 and 6 months (P < 0.05: Fig. 6). At 12 months, BV/TV of the new bone in both sides was significantly higher than that of native bone sites (P < 0.05).
3.4. Bone morphometrical evaluations
At 1 month, new bone was lined with osteoclasts and osteoblasts. However, non-vital crushed bone was also observed in a distant area of the new bone. At 12 months, no foreign bodies were detected around the tray, and all new bone areas were filled with the vital bone (Fig. 7).
At 1 month after surgery, the average BV/TV of the HA/PLLA and Ti sides was 46.4 16.9% and 32.2 7.2%, respectively (N ¼ 3). At 3 months, it was 44.5 13.2% and 62.4 11.6% (N ¼ 4). At 6 months, it
3.5. Histological findings
Fig. 5. Micro CT and CSLM images. Transaxial images of the HA/PLLA and titanium tray. The left side shows micro-CT and the right side shows CLSM images for each. In CLSM images, tetracycline is labelled by red and calcein is green. (B) Native bone. (NB) New bone. (CB) Crushed bone. Ti tray ( ). Borders between the native and new bone (---). Upper: 1 month after reconstruction. ( ) Weak labelling around the bone chips. Middle: 6 months after reconstruction. Lower: 12 months after reconstruction. ( ) Lamellar labelling at the outer surface.
e458
A. Matsuo et al. / Journal of Cranio-Maxillo-Facial Surgery 40 (2012) e453ee460
4. Discussion We set out to determine the tray fit and bone quality of PCBM mandibular reconstruction using custom-made bioresorbable HA/ PLLA tray. Surgeons find it difficult to manipulate a titanium mesh tray to achieve a suitable contour that will fit large defects. This technique is also time-consuming and decreases mechanical strength (Yamashita et al., 2008). A fracture after PCBM mandibular reconstruction has been reported (Aytak et al., 2005). Custommade Ti trays using rapid prototyping can solve this problem, and clinical trials have already been performed on orbital floor fractures (Metzger et al., 2007). In mandibular reconstruction, a case in which a Ti tray was pre-bent and reinforced by laser welding has also been reported (Yamashita et al., 2008). However, this is not a common technique in the clinical setting because of technical difficulties. In addition, Ti trays sometimes require removal, because they can hinder implant insertion (Cheung et al., 1994). Bioresorbable materials may be the best option to solve this problem. Initially PGA plates or screws were clinically used as the
Fig. 6. Changes of the bone volume fraction after reconstruction between the HA/PLLA and Ti sides. *P < 0.05.
bioabsorbable osteosynthesis material (Bostman et al., 1987). However, a high prevalence of foreign body reaction was reported (Bostman et al., 1990). Then, many experimental and clinical studies of PLLA plates or screws reported good results, with little foreign body reaction (Bessho et al., 1997; Eppley et al., 1997). Pure PLLA (Kinoshita et al., 1997) and DLA/PLLA (Louis et al., 2004) have been reported for use in mandibular reconstruction, but an important point for successful PCBM reconstruction is rigid fixation (Kruger and Krumholz, 1984), which is incompatible with these methods. We previously reported 2 PCBM reconstruction cases using custom-made HA/PLLA trays with rigid fixation (Matsuo et al., 2010), but the custom-made tray fit was not validated. In this study, there was no significant difference in tray fit between the HA/PLLA and the Ti on the buccal side, but on the lingual side fit was better in the HA/PLLA type than in the Ti type. On the buccal side, the difference in gap size was not significant between the Ti and HA/PLLA tray because the screws were inserted from the buccal side. However, an initially unseen gap formed between the tray and the mandible because the lingual side was not fixed with screws. These findings suggest that customized trays may be more efficient for fitting to the lingual side. In addition, bone formation was not affected by the large size of the gap. Therefore, we can expect the custom-made resorbable trays to yield better bone formation in the lingual area and reduce operating times, especially in clinically compromised cases with markedly curved or deep lingual areas, for which manual bending is difficult. Regarding bone quality of the HA/PLLA tray, both bone conduction and induction were more active in the PCBM graft than in the block bone graft (Lieberman and Friedlaender, 2005). In the present study, fluorescent labelling around the bone chips were seen early, even in the distant part from the new bone formation area. This suggests that bone induction is actively initiated early around the PCBM. From 3 to 6 months after surgery, the outer configuration of the new bone formation area was observed. This is
Fig. 7. Histological findings. Figure indicates the HeE staining of the HA/PLLA side. (A) New bone formation area at 1 month. ( ) Osteoblast, ( ) osteoclast. (B) Bone chips ( ) away from the bone formation area at 1 month. (C) Around the tray ( ) at 12 months. (D) Outer surface ( ) of the new bone areas at 12 months.
A. Matsuo et al. / Journal of Cranio-Maxillo-Facial Surgery 40 (2012) e453ee460
similar to previous pure PLLA tray (Kinoshita et al., 1997) and autogenous scaffold findings (Fennis et al., 2002), but no study has evaluated bone quality of the bone remodelling period 3 months after surgery. Although clinically sufficient amounts of bone formation formed in PCBM reconstruction using a Ti tray, its X-ray opacity is low in many cases compared with that of the surrounding native bone more than 1 year after reconstruction in clinical studies (Marx and Garg, 1998; Iwamoto et al., 2009; Matsuo et al., 2011a,b). On the other hand, a clinical study of mandibular reconstruction using HA/PLLA trays showed that the CT value of the reconstructed bone was higher with Ti trays, and was similar to native bone (Matsuo et al., 2010). This finding was obtained even in marginal resection cases, which had not reached critical defect size. Using HA/PLLA we expect rapid bone growth because it contains HA (Shikinami et al., 2005). To confirm this we did not need to make critical size defects in this study, and our aim was mainly evaluation during the remodelling period 3 months after surgery. In addition, we used histological, histomorphometrical and time labelling methods to evaluate new bone quality. In this study, the outer surface of the new bone formation area was defined by 6 months on both sides. However, the BV/TV of the HA/PLLA trays was significantly lower than in the Ti tray at 3 and 6 months. These findings indicate that bone remodelling was delayed on the HA/PLLA side until 6 months. However, after 6 months, new bone formation in the lamellar structure labelled by Cal progressed actively on the outer surface towards the tray, and BV/TV had quickly increased by 12 months, especially on the HA/PLLA tray in a morphometrical study. This suggests that the pattern of bone remodelling was initially relatively different between HA/PLLA and Ti trays, but similar bone qualities were obtained by 12 months, both of which were higher than that of native bone. We should note that these results may be affected by both tray materials and bending methods. It has been reported that HA/PLLA tray resorption becomes active after 1 year, and is resorbed approximately 3e5 years after surgery (Shikinami et al., 2005). In a pure PLLA tray, resorption of the tray had already begun 12 months after reconstruction (Kinoshita et al., 1997). In the present study, there was no foreign body reaction and no tray resorption at 12 months after reconstruction. In mandibular reconstruction, the reconstructed bone requires sufficient strength, especially in continuity defects, because a large occlusal force is loaded onto the mandible (Tie et al., 2006). However, the mechanical properties of the cancellous bone graft are lower than those of a block bone graft (Lieberman and Friedlaender, 2005). Clinically, the tray is removed at least 6 months after surgery, usually after 1 year, based on consideration of the mechanical properties. If implants are inserted early, the alveolar part of the HA/PLLA alone tray can be easily removed without additional invasive surgery. In general, bone strength is proportionate to bone microstructures (Seemann, 2002). In PCBM reconstruction using HA/PLLA trays, sufficient structure of new bones was obtained, and the HA/PLLA tray also maintained the sufficient strength by 12 months. We therefore think that delaying bone formation for 3e6 months in a HA/PLLA tray does not effect the clinical outcome. The results of this study regarding bone quality were limited to small defects using an extraoral approach. Further studies should evaluate bone strength, tray strength and infection resistibility of HA/PLLA trays in continuity defects using an intraoral approach. This study included several limitations. First, our study did not include critical-size continuity defects as the main purpose of this study was to evaluate tray fit. Therefore, the contact area between the tray and native bone had to be made as large as possible. Secondly, we observed the animals until 12 months after reconstruction, but the tray was mainly resorbed after 1 year (Shikinami
e459
et al., 2005). We should further investigate long-term follow-up and tray evaluation. Finally, we used PRP according to our clinical PCBM reconstruction protocol (Iwamoto et al., 2009; Matsuo et al., 2011a). Previously, PRP was said to promote bone formation in PCBM reconstruction (Marx et al., 1998), however recent animal studies have indicated contradictory results. Choi et al. (2004) reported inhibition, but Fennis et al. reported promotion (Fennis et al., 2002). We believe that this did not affect our results, because PRP was used on both sides and sufficient quality of bone formation was seen on both sides in the present study. 5. Conclusion The customized HA/PLLA tray was easily and accurately adapted to the mandible, and had achieved sufficient bone quality by 12 months after the reconstruction. Disclosure The authors have no financial conflicts of interest with regard to this study. Acknowledgements The authors are indebted to Mr. Roderick J. Turner and Professor J. Patrick Barron, Chairman of the Department of the International Medical Communications of Tokyo Medical University, for their editorial review of this manuscript. References Aytak C, Ozbec S, Kahveci R, Ozgenel Y, Akin S, Ozcan M: Titanium mesh fracture in mandibular reconstruction. J Craniofac Surg 16: 1120e1122, 2005 Bessho K, Iizuka T, Murakami K: A bioabsorbable poly-L-lactide miniplate and screw system for osteosynthesis in oral and maxillofacial surgery. J Oral Maxillofac Surg 55: 941e946, 1997 Bosker H, Powers MP: TMI reconstruction system. In: Fonseca RJ, Davis WH (eds), Reconstructive preprosthetic oral and maxillofacial surgery, 2nd edn. Philadelphia, PA: Saunders, 1995 Bostman O, Vainionpaa S, Hirvensalo E, Makela A, Vihtonen K, Tormala P, et al: A prospective randomized trial. J Bone Joint Surg [Br] 69-B: 615e619, 1987 Bostman O, Hirvensalo E, Makinen J, Rokkanen P: Foreign-body reactions to fracture fixation implants of biodegradable synthetic polymers. J Bone Joint Surg [Br] 72-B: 592e596, 1990 Carlson ER, Marx RE: Mandibular reconstruction using cancellous cellular bone grafts. J Oral Maxillofac Surg 54: 889e897, 1996 Cheung LK, Samman N, Tong ACK, Tiedman AH: Mandibular reconstruction with the Dacron ulethane tray. A radiological assessment of bone remodeling. J Oral Maxillofac Surg 52: 373e385, 1994 Choi BH, Huh JY, Suh, Lee SH: Effect of platelet-rich plasma on bone regeneration in autogenous bone graft. Int J Oral Maxillofac Surg 33: 56e59, 2004 Dunback J, Rodemer H, Spitzer WJ, Steinhauser EW: Mandibular reconstruction with cancellous bone, hydroxyapatite and titanium mesh. J Cranio-Maxillofac Surg 22: 151e155, 1994 Eppley BL, Sadove AM, Havlik RJ: Resorbable plate fixation in pediatric craniofacial surgery. Plast Reconstr Surg 100: 1e7, 1997 Fennis JP, Stoelinga PJ, Jansen JA: Mandibular reconstruction: a clinical and radiographic animal study on the use of autogenous scaffolds and platelet-rich plasma. Int J Oral Maxillofac Surg 31: 281e286, 2002 Goh BT, Lee S, Tiedman H, Stoelinga PJW: Mandibular reconstruction in adults: a review. Int J Oral Maxillofac Surg 37: 597e605, 2008 Ito T, Kudo M, Yozu R: Usefulness of osteosynthesis device made of hydroxyapatitepoly-L-Lactide composites in Port-Access cardiac surgery. Ann Thorac Surg 86: 1905e1908, 2008 Iwamoto I, Matsuo A, Kato N, Takeuchi S, Takahashi H, Hojo S, et al: Computed tomographic evaluation of bone quality of the mandible reconstructed by particular cellular bone and marrow combined with platelet rich plasma. Oral Science International 6: 63e72, 2009 Kim K, Isu T, Sugawara A, Matsumoto R, Isobe M: Utility of new bioabsorptive screws in cervical anterior fusion. Surg Neurol 68: 264e268, 2007 Kinoshita Y, Kobayashi M, Hidaka T, Ikeda Y: Reconstruction of mandibular continuity defects in dogs using Poly (L-lactide) mesh and autogenic particulate cancellous bone and marrow: preliminary report. J Oral Maxillofac Surg 55: 718e723, 1997
e460
A. Matsuo et al. / Journal of Cranio-Maxillo-Facial Surgery 40 (2012) e453ee460
Kruger E, Krumholz K: Results of bone grafting after rigid fixation. J Oral Maxillofac Surg 42: 491e496, 1984 Lieberman JR, Friedlaender GE: Bone regeneration and repair. Biology and clinical applications. Totowa, New Jersey: Humana Press, pp. 57 65, 2005 Marx RE, Garg AK: Bone structure, metabolism, and physiology: its impact on dental implantology. Implant Dentistry 7: 267e276, 1998 Louis P, Holmes J, Fernandes R: Resorbable mesh as a containment system in reconstruction of the atrophic mandible fracture. J Oral Maxillofac Surg 62: 719e723, 2004 Macarini L, Milillo P, Mocci A, Vinci R, Ettorre GC: Poly lactic acid-hydroxyapatite (HA/PLLA) bioabsorbable interference screws for tibial graft fixation in anterior cruciate ligament (ACL) reconstruction surgery: MR evaluation of osseointegration and degradation features. Radiol Med 113: 1185e1197, 2008 Marx RE, Carlson ER, Eichstaedt RM, Schimmele SR, Georgeff KR: Platelet-rich plasma. Growth factor enhancement for bone graft. Oral Surg Oral Med Oral Pathol Oral Rad Enodl 85: 638e646, 1998 Matsuo A, Chiba H, Takahashi H, Toyoda J, Abukawa H: Clinical application of a custom-made bioresorbable raw particulate hydroxyapatite/poly-L-lactide mesh tray for mandibular reconstruction. Odontology 98: 85e88, 2010 Matsuo A, Chiba H, Toyoda J, Abukawa H, Fujikawa K, Tsuzuki M: Mandibular reconstruction using a tray with particulate cancellous bone and marrow and platelet rich plasma by an intraoral approach. J Oral Maxillofac Surg 69: 1807e1814, 2011a
Matsuo A, Chiba H, Takahashi H, Toyoda J, Hasegawa O, Hojo S: Bone quality of mandibles reconstructed with particulate cellular bone and marrow, and platelet-rich plasma. J Cranio-Maxillofac Surg 39: 628e632, 2011b Metzger MC, Schon R, Schmelzeisen R: Preformed titanium meshes: a new standard? Skull Base 17: 269e272, 2007 Seemann E: Pathogenesis of bone fragility in women and men. Lancet 359: 1841e1850, 2002 Shikinami Y, Okuno M: Bioresorbable devices made of forged composites of hydroxyapatite(HA) particles and poly lactide (PLLA). Part II: practical properties of miniscrews and miniplates. Biomaterials 22: 3197e3211, 2001 Shikinami Y, Matsusue Y, Nakamura T: The complete process of bioresorption and bone replacement using devices made of forged composites of raw hydroxyapatite particles/poly lactide (F-u-HA/PLLA). Biomaterials 26: 5542e5559, 2005 Takeuchi S, Matsuo A, Chiba H: Beneficial role of periosteum in distraction ontogenesis of mandible: its preservation prevents the external bone resorption. Tohoku J Exp Med 220: 67e75, 2010 Tie Y, Wang DM, Ji T, Wang CT, Zhang CP: Three-dimensional finite-element analysis investigating the biomechanical effects of human mandibular reconstruction with autogenous bone grafts. J Cranio-Maxillofac Surg 34: 290e298, 2006 Yamashita Y, Yamaguchi Y, Tsuji M, Shigematsu M, Goto M: Mandibular reconstruction using autologous iliac bone and Titanium mesh reinforced by laser welding for implant placement. Int J Oral Maxillofac Implants 23: 1143e1146, 2008