Adhesion of composite resin to bone—a pilot study

C 1994 The British Association of Oral and Maxillolhcial Surgeons

Adhesion of composite resin to bone-a J. G. Meechan,

J. F. McCabe,

pilot study

A. D. Beynon

The Dentul School, University of Newcustle upon Tyne, Newcastle upon Tyne

SC’MMA R Y. This pilot study investigated the adhesion of composite resin to pig calvarium using All-bond 2 dentine bonding agent in an in vitro model. The bone was subjected to different methods of preparation. Acidetching decreased the bond strength between bone and composite. Roughening the bone with a surgical bur prior to application of the adhesive produced bond strengths in the range 11.1-16.1 MPa.

INTRODUCTION The use of miniplates has revolutionised the treatment of maxillofacial fractures, however the application of these devices falls short of the ideal. Perfect adaptation to bone can be time-consuming and difficult in some areas. In addition the drilling of screw-holes has the potential lo damage associated structures for example dental roots. It would be beneficial to USC a material that was highly malleable during the adaptation phase and rigid at the fixation stage. The elimination of screws would also be an advantage. Modern dentine adhesives used in combination with composite resins, which arc efyective in the presence of moisture may offer a suitable method of attaching rigid fvation devices to bone without the aid of screws. Furthermore, composite resins are initially soft and plastic until they set to a rigid condition. Theoretically these materials might be of value in the fixation of maxillofacial fractures when used in conjunction with a suitable adhesive to achieve composite-bone bonding. There have been rapid developments in the field of dentine bonding agents over recent years and it is possible that some of these materials may offer a potential means of bonding to bone. One such material (All-bond 2) has attracted particular ‘attention because it is able to bind to moist dcntine.’ It was the purpose of this in-vitro pilot study to investigate the bond strength between bone and composite resin when using All-bond 2 dentine adhesive (Bisco, Itasca, Illinois) and to determine the most optimal conditions for its use.



The bone used in this study was pig calvarium obtained from an abattoir. Samples of approximately 1 cm in diameter wcrc obtained and embedded in a polyester block with the cortical aspect at the surface. The bone surface was then prepared by one of the following methods:

1. lapping with PlOO grade Carborundum paper 2. lapping with PI200 grade Carborundum paper 3. lapping with PI200 grade Carborundum paper followed by a 15 s etch with 10% phosphoric acid as recommended for dentine with the Allbond 2 kit 4. roughening with a No. 8 round surgical bur. Each treatment was performed on six specimens. Following bone preparation All-bond 2 dentine adhesive was applied to an area of bone 5 mm in diameter which had been isolated using a PTFE splitring template. The method recommended by the manufacturers was used with the exception that the etching stage was omitted for samples prepared by methods 1, 2 and 4 above. Immediately following application and light-curing of the resin bonding agent 5 mm diameter cylinders of P50 composite (3M, Loughborough, England) were adapted to the bond using the split-ring template and light-cured. The prepared specimens were stored overnight in distilled water at 37’C and then the shear bond strength dctcrmincd using the method outlined in IS0 Draft Technical Report 11405 ( 1992). This consists of a two part guillotine-like device in which the test resin block is housed in one half of the device and the blade of the other half applies a shear force to remove the bonded composite or cement from the bone surface (Fig. 1). By carrying out the procedure on a Universal Testing Machine (Instron, High Wycombc, UK) the force to break the bond in shear can be recorded. The mode of failure was determined as adhesive, cohesive or mixed by viewing debonded surfaces under 2 x magnification. The results were subjected to ANOVA and the Student-Newman-Keuls Procedure.

RESULTS The results are shown in Table 1. Thcrc were significant differences in the bond strengths achieved between the composite and the different bone prep-


British Journal

of Oral and Maxillofacial


Fig. 1 -The device used to measure shear bond strength. Tests wcrc performed using this rig on the load ccl1 of an lnstron Universal Testing Machine.

Table 1


of bone preparation Bone preparation










I’100 Carborundum paper I’1200 carborundum paper PI200 Carborundum paper-acid etch Surgical bur

Shear stress for failure XlMPa(mean k sd.) 8.9 k4.9 8.4k3.7 4.1 k2.0 13.1 k2.1

DISCUSSION This pilot study was designed to test the feasibility of bonding composite resm to bone as a means of stabilising maxillofacial fractures. At this stage only mechanical properties were considered. All-bond 2 is an enamel and dcntine bonding agent that can bond to wet surfaces.’ The manufacturers of All-bond 2 recommend etching of enamel and dentine prior to use. The system is used by applying a mixture of two priming agents (NTG-GMA) [an adduct of N p-tolyl glycine and glycidly methacrylatc] in acetone and BPDM [biphcnyl dimethacrylate] in acetone, drying and then applying the unfilled BisGMA,/hydroxyethyl methacrylate resin which is light-cured. Etching of enamel, which is a well established

means of achieving bonding produces differential relief as a consequence of differences in susceptibility of prism cores and boundaries to dissolution. Enamel crystallites arc preferentially removed at differing rates depending upon the crystal orientation and availability to the etching solution. Enamel proteins (prinicpally enamelins) which arc rclcascd tend to diffuse away from the etched surface. The attachment of resins accordingly is to a fresh irregular surface, comprised almost entirely of newly exposed biological apatite crystallitcs. In contrast, in mineralised connective tissues: including dentinc and bone, the acid etching agent solubilises mineral, but leaves the bulk of the organic matrix intact. For dcntinc the partially demineralized surface layer may be impregnated by the resin bonding agent to form a hybrid layer, as described by Nakabayashi et al.’ It is clear from the results of the present study that bone behaves differently. Mature periosteal bone consists of circumferential lamellae laid down by subperiosteal osteoblasts, with successive lamellae containing collagen fibrils which are parallel to the surface but which differ in orientation from one lamella to another. The biological apatite crystals lie within, and to a lesser extent between the collagen fibres orientated in the direction of the libres. Etching of periosteal bone causes surface demineralisation, removing crystallites within the superficial lamellae, leaving a surface zone of dcmineralised collagen Gbrils. Bone acts as a composite material, with collagen fibrcs providing tensile and mineral providing compressive strength. Loss of mineral causes the bone to become deformable and this probably accounts for the lower bond strength of etched bone. The avoidance of acid solutions which are potentially toxic also reduces potential damage to bone cells, whose vitality should be maintained. The death of bone cells should be minimised whenever possible, since this will act as a stimulus for bone which could result in premature resorption, detachment. The bond strengths obtained in this investigation approach those reported between All-bond 2 and dentine. Kanca’ reported a mean shear bond strength of 16.1 MPa with All-bond 2 and acid-etched dry human dentine, greater strengths have been recorded with moist dentine.‘.j A value of 24 MPa was obtained in this laboratory using All-bond 2 according to the manufactures recommended procedure.4 The forces required to break the bonds obtained in the present study (222-3 l7N following treatment with a bur) were greater than those required for failure of the 2.6 mm diameter 20 mm long poly L lactide biorcsorbable screws described by Wittenberg et al$ these authors noted failure below the screw head at forces of 178 +2N. The upper end of the range of forces reported in the present study approach the lower end of the range of pull-out forces for the metal screws employed in commonly used miniplatc systems described by Wittenberg et a1.5 These last authors noted metal screws pulled out at forces in the range 318-431 N however failure here was due to


bone splintering thus failure below such forces is perhaps advantageous. It is worth noting that the All-bond system also adheres to both precious and non-precious metals with bond strengths in excess of 20 MPa.‘.3.4 This feature might be of use when considering alternative methods of miniplate fixation. It is necessary to emphasize the preliminary nature of these studies as a way of justifying the relatively small number of specimens used for bond testing. As outlined in the IS0 document on adhesion testing,” six specimens are adequate for initial screening purposes only. Greater specimen numbers will be used in future studies and the results subjected to more rigorous scrutiny using Weibull analysis.6 The use of a highly malleable material to allow good adaptation to the bone surface would offer benefits in fracture reduction. Metal miniplates are bendable using instrumentation howcvcr perfect adaptation may not be readily achieved. Bioabsorbable plates suffer from a degree of stiffness’ although Bos et al8 increased the malleability of poly L lactide plates by heating before adaptation in an animal model. The great malleability of composite resins prior to setting offers an easy method of adaptation to bone surfaces and thus the use of composite based miniplates is conceptually appealing. The physical properties of composite resin can be modified by altering the filter content or by incorporating reinforcing materials such as fibres. The resin used in this study was P50 a light-cured ‘posterior composite’. The biocompatibility of composite resins varies between light and chemicallycured systems” however there is evidence that posterior composites are more biocompatible than some other restorative dental materials. lo Having established the most favourable bone prcparation a further five bone specimens from the same animal were prepared by surface roughening with the surgical bur. The glass ionomer Ketac Bond (Espe, Seefeld, West Germany) was bonded in the manner described earlier using the method recommended by the manufacturer for bonding to dcntine. The mean (+sd) bond strength for Ketac Bond was 3.6+ 1.3 MPa. The bond strength of composite to bone was significantly greater than that achieved by the glass-ionomer (p < 0.001) t-test). This comparison was made as glass-ionomers have shown promise as potential bone substitutes in recent investigations.1’,‘2 The bond strength between bone and glass-ionomer was the weakest of any of the combinations tested in this study, the greatest stress required to produce failure being 5.7 MPa. In all cases failure was adhesive in nature. In conclusion the data presented in the present study show that simple preparation of the bone surface and the USCof a dentine bonding agent can produce bond strengths between bone and composite

of composite

resin to bone-a

pilot study


resin that are potentially useful clinically. However. considerable work is required, particularly concerning biocompatibility, if the proposed system is to be further dcvcloped. References 1. Kanca J. Dental adhesion and the All-bond system. Journal of Esthetic Dentistry I991 ; 3: 129. 2. Xakabayashi N, Nakamura MM,Yasuda N. Hybrid layer as dentin-bonding mechanism. Journal of Esthetic Dentistry 1991; 3: 133 138. 3. Barkmeier WW, Suh BI; Coolcy RL. Shear bond strength to dentin and Ni-Cr-Bc alloy with the All-bond universal adhesive system. Journal of Esthetic Dentistry 1991: 3: 148. 4. Nery S, McCabe JF, Wassell RW. Bond strengths of a multipurpose dental adhesive. British Society for Dental Research Abstract Xo. 157. 1993. 5. Wittcnberg JM, Wittenberg RI-I, Hipp JA. Biomechanical properties of resorbable poly-L-lactide plates and screws: a comparison with traditional systems. J Oral Maxillofac Surg 1991: 49: 512. 6. McCabe JF, Carrick TE. A statistical approach to the mechanical testing of dental materials. Dental Materials 1986; 2: 139-142. I. Shetty VR, Bcrtolami C. Biomcchanical properties of resorbable poly-L-lactide plates and screws: a comparison with traditional systems: Discussion. J Oral Maxillofac Surg 1991; 49: 517. 8. Bos RRM. Rozema FR, Boering G, Nijenhuis AJ, Pennings AJ, Jansen HWB. Bone-plates and screws of bioabsorbable poly (L-lactide) - an animal pilot study. Br J Oral Maxillofac Surg 1989: 27: 467. 9. Wennbcrg A. Mjor IA. Hcnstcn-Pettersen A. Biological evaluation of dental restorative materials-a comparison of different test methods. J Biomed Mater Res 1983; 17: 23. 10. Hctcm S. Jowett AK. Fcrguson MWJ. Biocompatibility testing of a posterior composite and dental cements using a new organ culture model. J Dent 1989; 17: 155. 11. Brook IM, Craig CT, Lamb DJ. Initial in-viva evaluation of glass-ionomer cements for use as alveolar bone substitutes. Clinical Materials 1991; 7: 295. 12. Brook IM, Craig CT, Hatton PV, Jonck LM. Bone cell interactions with a granular glass-ionomcr bone substitute mat&al in viva and in vitro culture models. Biomatcrials 1992; 13:72 1. 13. IS0 Draft Technical Report 11405. Dental Materials: Guidance on testing of adhesion to tooth structure, 1992.

The Authors .I. G. Meechan

BSc, BDS, PhD. FDS Lecturer in Oral Surgery .I. F. McCabe Rsc, FhD, DSc Reader in Dental Material Science A. D. Beynun BDS, PhD Senior Lecturer in Oral Anatomy The Dental School University of Newcastle upon Tyne Newcastle upon Tyne NE2 4BW

Correspondence and requests for olTprints to J. G. Meechan, Department of Oral Surgery: The Dental School, Framlington Place, Newcastle upon ‘fyne, NE2 4BW Paper received 14 April 1993 Acccptcd 2 September 1993