| | Comparison of in vivo and in vitro shear bond strength☆☆☆★Received 1 September 2001; received in revised form 1 March 2002; accepted 1 March 2002. Abstract Much of the research into composite adhesives suggests that these materials will deteriorate in the oral environment, but most of these claims are made on the basis of extrapolation from in vitro experiments; relatively little in vivo research has been conducted into the mechanical properties of composite adhesives. For this study, we recruited 20 volunteers to wear removable appliances containing orthodontic brackets bonded to enamel slabs for 12 weeks. Each appliance carried 3 brackets bonded with Transbond (3M Unitek UK; Bradford, United Kingdom) and 3 brackets bonded with Heliosit (Ivoclar-Vivadent; Leicester, United Kingdom). The bond strengths were tested at intervals of 4, 8, and 12 weeks. Control specimens were stored in sterile water at 37°C and were debonded at the same time intervals. Transbond specimens debonded after 4 weeks in vivo had significantly (P < .05) lower bond strengths (9.78 megapascals [MPa]) than did the controls (14.34 MPa). In vivo, specimens bonded with Heliosit had significantly (P < .05) lower bond strengths after 4 weeks (8.16 MPa vs 10.96 MPa) and 8 weeks (9.96 MPa vs 13.61 MPa) than did the controls. These results indicate differences between bond strength testing in vitro and in vivo. Further research is required into the effects of the oral environment on bond strength. (Am J Orthod Dentofacial Orthop 2003;123:2-9)
Although its origins have recently been the cause of controversy,1, 2, 3 the use of brackets bonded directly to teeth as components in fixed orthodontic appliances has been described since 1965.4 Newman4 originally described the use of an epoxy adhesive, but other materials have become more commonly used, principally composite adhesives based on dimethacrylate resins with inorganic fillers.5 Development of these materials continues, but, as yet, none possesses all the properties that an operator might desire.6
A commonly encountered problem is the bond failure of a bracket during treatment. The frequency of this has been investigated by various authors, including Zachrisson,7 O'Brien et al,8 and Sunna and Rock,9 and has been found to vary between 0.5% and 16%. Various factors can contribute to the likelihood of a bond failure, including operator technique, patient behavior, variation in the enamel surface, and bracket properties.
The mechanical properties of composites have been extensively researched in the laboratory, but the clinical situation has not been accurately replicated.9 Some researchers have suggested that this is the result of changes that occur because of exposure to the oral environment.10, 11, 12, 13, 14
Biodegradation is the result of a combination of disintegration and dissolution in saliva, chemical and physical degradation, wear caused by chewing food, erosion by the food itself, and bacterial activity.15 Matasa11 proposed that biodegradation of the adhesive composite can contribute to the failure of the bond between the bracket and the tooth. Studies of composite biodegradation have led to various explanations. These are not mutually exclusive and are summarized below:
•Unreacted composite acrylic leaches from the composite when it is immersed in water.15, 16
•Water can hydrolyze the acrylic,17 particularly at the surface layers and at the interface of the acrylic and the filler particles. It can also hydrolyze the filler particles themselves.18, 19 This hydrolysis of composite is enhanced in artificial saliva.10
•Composite degrades in the presence of food simulants such as ethanol, probably because of reactions with the alcohol hydroxyl group or molecular oxygen.13 The exposure to food simulants has been shown to weaken the bond strength.20
•Nonspecific porcine liver esterases enhance degradation because of enzymatic hydrolysis.21
•Composite degrades in natural saliva in vitro, probably as a combination of hydrolysis and enzymatic hydrolysis.22
•Certain bacteria can consume composite, using it as a source of carbon.11
Øilo15 reviewed the degradation of composite materials in the mouth and stated that it is such a complex interaction of processes that it cannot be reproduced in vitro. He concluded that the correlation between tests in vitro and the clinical situation has not been established and that there is a need to develop standardized in vitro and in vivo tests. Similarly, Sunna and Rock9 concluded that bond strengths ex vivo did not correlate with clinical failure rates and questioned the applicability of these tests to the clinical situation.
It is likely that composites suffer more degradation in vivo than in vitro, because these processes interact and frequently enhance one another, and the composite will also be subjected to the mechanical forces involved in mastication and orthodontic treatment. Despite this, the majority of research into dental composites continues in vitro because it is difficult to expose the materials to (and retrieve them from) the oral environment without interfering with the environment itself or taxing the compliance of subjects.
One approach to this problem is to investigate materials with teeth that have already been designated for removal, such as premolars in an orthodontic patient. A material can be applied to a tooth and left in situ for a predetermined time; the tooth can then be extracted, ideally with as little disruption to the material as possible, allowing experiments to be carried out on the material. Technically, the experiments are ex vivo, but the tooth experiences all the in vivo conditions until removal.
This approach has been used to test glass ionomer cements.23 The method exposes material to the oral environment but has practical difficulties because it must be planned around the patient's treatment. A large and regular turnover of patients would be required to recruit a sample of patients large enough to be statistically useful.
Roulet and Walti24 described an alternative approach. They investigated the influence of oral fluids on composite and glass ionomer restorative materials with a hollow bridge in a bounded edentulous area that could accommodate specimens of restorative materials that could be removed and replaced. Roulet and Walti24 demonstrated some degradation in the materials without reference to extraoral controls. The device provided the ability to authentically analyze the effects of the oral environment on the properties of a dental material. However, the device had several shortcomings:
•It had to be made to very precise specifications and was expensive in terms of materials and clinical and laboratory time.
•It accommodated only small samples of material.
•It required a subject who was partially edentulous but who had remaining teeth that were suitable for a bridge capable of accommodating the material specimens.
Such a platform is not ideally suited to the testing of orthodontic bonding materials. In addition, the typical orthodontic patient is likely to be considerably younger, with a different diet and physiologic environment than those in the study of Roulet and Walti.24
Clearly, a method to test the effect of the oral environment on orthodontic composite is needed. Such a system should fulfill the following criteria:
•It should focus on clinically relevant properties of the composite (usually bond strength).
•It should be reproducible and capable of comparing materials.
•It should expose the materials fully to the oral environment.
•It should be well tolerated by clinical subjects for long periods of time.
•It should be suitable for use by a wide range of clinical subjects without requiring selection on grounds of specific dentition or occlusion.
•It should allow the investigator to introduce materials to the oral environment, retrieve them, test them, and then replace them with minimal disruption to the subject and the operator.
•It should be relatively inexpensive.
This study describes an appliance that was designed to meet these criteria. It provides a platform for introducing dental materials into the oral environment and exposing them for intermediate periods of time (4, 8, and 12 weeks); it also allows them to be retrieved and subsequently tested for bond strength.
Material and methods  Appliance design An initial pilot study was undertaken to determine a design that would achieve the previously listed goals and be well tolerated. The following basic design elements were used (Fig 1): lower arch removable appliance, Adams clasps to mandibular first molars, and acrylic base lingual to mandibular teeth.
Provision and application After the local authority (Newcastle Central Ethics Committee) gave its ethical approval, volunteers were recruited from the staff and students of Newcastle Dental Hospital, United Kingdom, under the following criteria: good general health, full dentition, low caries experience with no active caries, good oral hygiene, and age 18 to 40 years. The appliance was constructed in conventional orthodontic cold-cured acrylic and worn by the volunteer for 1 to 2 weeks to ensure it was well tolerated before the specimens were added. Enamel slabs were prepared by sectioning extracted human premolars that had been stored in a 0.5% aqueous solution of chloramine-T after extraction. The teeth were thoroughly washed in sterile water, then the enamel surface layer (approximately 0.5 mm) was removed from the buccal surface of each tooth with a water-cooled, high-speed diamond bur. Removal of the aprismatic enamel improved the reproducibility of the acid etching before bracket bonding. Each tooth was sectioned with a Cutangrind machine (Agate and General Stonecutters Ltd, London, United Kingdom) to isolate the buccal surface (the 7×8-mm enamel slab used in the experiment) from the rest of the tooth, which was discarded. The enamel slabs were washed in sterile water again before being placed into the appliance by cutting 3 indentations into the lingual aspect of the appliance. An enamel slab was placed in each indentation and incorporated into the appliance with cold-cure acrylic so that the buccal surface of the enamel was exposed on the lingual aspect of the appliance. After the enamel slabs were incorporated into the appliance, the appliance was returned to the volunteer to wear for another week. This allowed the enamel to equilibrate to the oral environment. The appliance was then removed from the volunteer, and brackets were bonded to the enamel according to the manufacturer's instructions. A single appliance could carry 6 brackets. The appliance was then returned to the volunteer to be retrieved at a predetermined time for testing. Ex vivo control appliances were constructed in a similar way and stored in sterile water at 37°C (Fig 2).
This allowed comparison of the in vivo data with previous research that used water as the control. Brackets (0.022 Andrews prescription, premolar brackets, Ormco-“A” Company, Orange, Calif) were bonded to the enamel slabs on the appliances and the controls. Two adhesive composites were used: Transbond (3M Unitek UK; Bradford, United Kingdom) and Heliosit (Ivoclar-Vivadent; Leicester, United Kingdom). Both are light-cured composites. Each appliance had 3 brackets bonded with Transbond and 3 bonded with Heliosit. Adhesive positions were randomly chosen. Each control appliance had 10 brackets bonded with a single adhesive. After bonding, an elastic module was placed on each bracket to more closely approximate clinical conditions. The appliances were worn by the volunteers, and the controls were stored in sterile water at 37°C. Volunteers were instructed to wear the appliance 24 hours a day, including mealtime, and to remove it twice a day only to clean it and their teeth with a toothbrush and fluoride toothpaste. Experimental debonding of brackets was carried out at intervals of 4, 8, and 12 weeks, as described by Hobson et al.20 Brackets were debonded with an Instron 5567 machine (Instron Instruments; High Wycombe, United Kingdom) that was fitted with a custom jig to allow the brackets to be debonded with a shear force parallel to the enamel surface as described by Fox et al.25 The jig could firmly hold an experimental or control appliance and orient it in all 3 planes to allow precise positioning of the appliance being tested. A ball-and-socket joint allowed further orientation of the bracket so that debonding could be carried out with a force applied parallel to the buccal surface of the enamel: a true shear force. The experimental appliance was firmly held in place in the jig by wiring it to the dental cast of the volunteer who wore it. The lingual portion of the plaster cast had been removed, and the cast was inverted so that when the bracket was debonded, the bracket bond was exposed to a shear force, but the appliance was compressed onto the cast and held firmly. The control appliances were formed so that they could be directly held in the same jig (Fig 3).
The Instron machine applied the force to the bracket via a wire loop (0.5-mm diameter, round cross section) at a crosshead speed of 1.0 mm/min. The load to cause failure was recorded in newtons, and the bond strength was calculated in megapascals (MPa) after the bracket surface area was measured. Statistical analysis was undertaken with Weibull, generalized linear analysis of variance (ANOVA), and post hoc Tukey tests; a spreadsheet and statistical software were used. Results Twenty volunteers were recruited (12 women, 8 men; mean age, 22 years). Two left the study because they could not tolerate the appliance despite several adjustments. One volunteer mislaid the appliance, and 1 left at the end of the university term. Two appliances each lost a single bracket within 4 weeks; otherwise, no accidental bond failures occurred. The mean bond strengths for Transbond and Heliosit, in vivo and in vitro, are shown in Table I.
The bond strengths for Transbond in vitro decreased significantly between 4 weeks (14.34 MPa) and 8 weeks (12.24 MPa) ( P < .05). However, there was no significant decrease or increase with time in vivo ( P > .05). At all time intervals, the mean bond strengths for Transbond in vivo were lower than those for Transbond in vitro, although this difference was significant only for specimens debonded at 4 weeks ( P < .01). The bond strengths for Heliosit in vivo showed no significant differences with time. However, specimens in vitro showed a significant (P < .05) increase in bond strength between 4 and 8 weeks and a significant (P < .001) decrease between 8 and 12 weeks. At 4 and 8 weeks, the bond strength of Heliosit was significantly lower (P < .05) in vivo than in vitro. There was no difference between the mean bond strengths of the specimens in vivo and in vitro at 12 weeks (P > .05). Transbond produced significantly higher mean bond strengths than did Heliosit (P < .005). Weibull curves were plotted for both materials (Figs 4 through 7).
In addition, Weibull analysis of the bond strengths was used to calculate the stress required for 5% and 10% probability of bond failure (Table II) and 95% confidence intervals.
There were trends for in vivo and in vitro Heliosit specimens to fail at lower stresses over time, and Transbond specimens tended to require more stress to cause bond failure, but neither trend was significant. Discussion Only 2 volunteers could not tolerate the experimental appliance, and only 2 brackets were unavailable for bond strength testing. The brackets and their adhesives were fully exposed to the oral environment. Although the volunteers were not constantly monitored for compliance, we found that the appliances were being worn when they were collected for testing. Testing took place at different times on different days, so it was unlikely that the volunteers could predict collection times. Furthermore, the volunteers were motivated dental students, and we often challenged them when we met them in the dental school. It is reasonable to assume that the appliances were being worn correctly. Provision of the appliances was only slightly more complicated than provision of a conventional removable appliance, and management of the volunteers and experimental procedure was relatively simple. The appliance design could be improved by placing the enamel slabs and brackets on the buccal aspect of the mandible; this would more closely reproduce the usual placement of orthodontic brackets. Modification of the Instron machine to allow very precise standardization of bond strength testing and provide a force that was parallel to the buccal surface of the tooth was not especially complicated. This modification was recommended by Fox et al25 to achieve a true shear force, but previously it has been difficult to achieve in conventional bond strength tests. The appliance system could be modified for testing other aspects of orthodontic or restorative materials in vivo, such as water sorption, microleakage, microcrack propagation, or bacterial colonization. The enamel surface underwent minimal enamel preparation similar to that used for adhesive bridges. The preparation provided greater consistency of enamel etching to ensure that any variations in bond strength would more likely result from variations in the composite rather than from variations in the enamel surfaces after etching. Enamel surface variations have been reported to be a problem with premolars.26, 27 Apart from the shape of the appliances and the distribution of the composites on them, the only difference between the control and experimental appliances was the environments to which they were exposed. Some concern could be expressed because dental tissue from extracted teeth was placed in the mouths of the volunteers. The slabs predominantly consisted of enamel, the least vascular and least cellular tissue, and were consequently the least likely to be a cross-infection risk. In accordance with the International Standards Organization protocol 11405,31 the teeth were stored in chloramine-T (chloroparatoluene sulfonamide salt), a disinfectant with an inorganic chlorine base; it is active against bacteria, viruses, fungi, and spores. If cross-infection were a concern in the future and further sterilization were required, the enamel could be autoclaved without significantly altering the bond strengths.32, 33 Because the analysis of the specimens required multiple comparisons of data, such as bond strengths changing over time, ANOVA and post hoc Tukey tests were used to examine the results. In keeping with the recommendations of Fox et al25 and McCabe and Carrick,28 Weibull analysis was used to indicate bond reliability. In practice, a material with a given mean fracture stress but a high Weibull modulus is more dependable than a second material with a higher mean fracture stress but a low Weibull modulus. The first material will have its samples more closely grouped about its mean and is less likely to fail at low stress levels than is the second material. It would have been interesting to record initial bond strengths for the materials to compare them with the bond strengths as time passed, but there were practical problems with this. First, there were not enough premolars available to produce the extra enamel slabs required; second, there would be some debate as to when initial bond strength should be recorded. The aim of the experiment was to compare the bond strengths for 2 materials in 2 environments, and the experiment design was focused on this aim. The statistically significant decrease in bond strength between 4 and 8 weeks for Transbond in vivo was consistent with experiments in vitro by Prati et al29 and Meng et al.14 These showed an initially severe deterioration in the mechanical properties of composite immersed in water, followed by a stable period of several weeks when there was little or no deterioration, and then a resumption of deterioration but slower than the original. In the current clinical study, the reduction in bond strength between 8 and 12 weeks for Transbond in vivo was not significant. These results suggest that initial high bond strengths deteriorate within 8 weeks of bond placement and thus alleviate concerns that modern materials might have excessively high bond strengths. There was also a highly significant (P < .01) difference between the mean bond strengths of Transbond after 4 weeks in vivo and in vitro, with the mean bond strength in vivo lower than that of the control. Heliosit in vivo at 4 weeks and 8 weeks also had significantly (P < .05) lower mean bond strengths than did the controls. This was consistent with studies by Munksgaard et al,12 who proposed that composite was degraded by enzymes in saliva, and studies by Lee et al13, 30 and Hobson et al,20 who demonstrated that food simulants degraded composite and reduced bond strength ex vivo. The finding of a significant difference between bond strengths in vivo and the controls suggested that the accepted experimental environment of distilled water does not accurately represent the oral environment. This finding agreed with those of Söderholm et al,10 who questioned the clinical relevance of using sterile water for tests on composites, and Øilo,15 who proposed that research is needed to compare results of experiments in vitro and in vivo. For Transbond in vivo, the mean bond strength was found to increase from 4 to 8 weeks, then to decrease from 8 to 12 weeks. This variation was not statistically significant, but it does appear paradoxical compared with the controls. However, by using the Weibull analysis to plot a curve of probability of bond failure against stress (Fig. 5, Fig. 7), we found that the same leftward displacement is seen on the Transbond in vivo curve as on the curve for Transbond in vitro for stresses below 10 MPa or failure probability below 50%, which would encompass the clinical range of bond failure in recent studies.8, 9 This suggested that bonds become less dependable over time, in both the controls and the experimental group, and this was consistent with the general deterioration over time in the mechanical properties of composite found by previous investigators.12, 13, 20, 30 The variation in bond strengths for Heliosit could not readily be explained in the context of previous experimental research. The specimens in vivo showed an increase in mean bond strength and a rightward displacement on Weibull curves (Fig. 4, Fig. 6) as time passed; this suggested that the bond becomes more dependable over time. However, the variations in bond strength were not significant. The control specimens showed a significant (P < .001) increase in bond strength from 4 to 8 weeks, and then a significant (P < .001) decrease from 8 to 12 weeks. These apparently confusing results might be clarified by further research. Conclusions From this initial study, it can be demonstrated that this appliance offers a means of introducing and exposing materials to the oral environment while allowing routine laboratory tests, such as bond strength testing, to be conducted. The appliance has 2 main advantages over previous attempts at exposing materials to the oral environment for laboratory testing: good tolerance by clinical subjects and ease of provision and maintenance of the system. From the bond strength testing, the following conclusions are drawn:
•The mean bond strength of Transbond immersed in sterile water at 37°C declines significantly between 4 and 8 weeks.
•The mean bond strength of Transbond exposed to the oral environment for 4 weeks is significantly less than when exposed to a control environment of sterile water at 37°C for 4 weeks.
•The mean bond strengths of Heliosit exposed to the oral environment for 4 weeks and 8 weeks, respectively, are significantly less than when exposed to a control environment of sterile water at 37°C for 4 weeks and 8 weeks.
•The changes in the bond strength of Heliosit in the control environment and the progressive increase in its bond strength in the oral environment require more research.
It would appear that further work is required to explore the differences between the bond strengths of Transbond and Heliosit in the oral environment compared with the standard laboratory control environment. Acknowledgements The authors would like to thank the volunteers who so diligently wore their appliances; John Askell and the staff in the orthodontic laboratory for their assistance in constructing the appliances; and Tom Carrick and Sandra Rusby in the Biomaterials laboratory for instructions and advice in the use of the Instron machine. Adhesives provided by 3M Unitek and Ivoclar-Vivodent. References 1.
1
Cueto HI.
A little bit of history: the first direct bonding in orthodontia.
Am J Orthod Dentofacial Orthop. 1990;98:276–277.
Full-Text PDF (329 KB)
|
CrossRef
2.
2
Mitchell DL.
The first direct bonding in orthodontia, revisited.
Am J Orthod Dentofacial Orthop. 1992;101:187–189.
Full-Text PDF (853 KB)
|
CrossRef
3.
3
Newman GV.
First direct bonding in orthodontia.
Am J Orthod Dentofacial Orthop. 1992;101:190–192.
Full-Text PDF (1241 KB)
|
CrossRef
4.
4
Newman GV.
Epoxy adhesives for orthodontic attachments: progress report.
Am J Orthod. 1965;51:900–912. 5.
5
Bowen RL.
Compatibility of various materials with oral tissues, I: the components in composite restorations.
J Dent Res. 1979;58:1493–1503. MEDLINE 6.
6
Matasa CG.
Adhesion and its ten commandments.
Am J Orthod Dentofacial Orthop. 1989;95:355–356. Abstract |
Full-Text PDF (167 KB)
|
CrossRef
7.
7
Zachrisson BU.
A posttreatment evaluation of direct bonding in orthodontics.
Am J Orthod. 1977;71:173–189. Abstract |
Full-Text PDF (6226 KB)
|
CrossRef
8.
8
O'Brien KD, Read MJF, Sandison RJ, Roberts CT.
A visible light-activated direct-bonding material: an in vivo comparative study.
Am J Orthod Dentofacial Orthop. 1989;95:348–351. Abstract |
Full-Text PDF (424 KB)
|
CrossRef
9.
9
Sunna S, Rock WP.
Clinical performance of orthodontic brackets and adhesive systems: a randomized clinical trial.
Br J Orthod. 1998;25:283–287. 10.
10
Söderholm K-JM, Mukherjee R, Longmate J.
Filler leachability of composites stored in distilled water or artificial saliva.
J Dent Res. 1996;75:1692–1699. MEDLINE |
CrossRef
11.
11
Matasa CG.
Microbial attack of orthodontic adhesives.
Am J Orthod Dentofacial Orthop. 1995;108:132–141. Abstract | Full Text |
Full-Text PDF (8238 KB)
|
CrossRef
12.
12
Munksgaard EC, Freund M.
Enzymatic hydrolysis of (di)methacrylates and their polymers.
Scand J Dent Res. 1990;98:261–267. 13.
13
Lee S-Y, Greener EH, Menis DL.
Detection of leached moieties from dental composites in fluids simulating food and saliva.
Dent Materials. 1995;11:348–353. 14.
14
Meng CL, Wang WN, Tarng TH, Luo YC, Lai JS, Arvystas MG.
Orthodontic resin under water immersion.
Angle Orthod. 1995;65:209–214. MEDLINE 15.
15
Øilo G.
Biodegradation of dental composites/glass ionomer cements.
Adv Dent Res. 1992;6:50–54. MEDLINE 16.
16
Hume WR, Gerzina TM.
Bioavailability of components of resin-based materials which are applied to teeth.
Crit Rev Oral Med Biol. 1996;7:172–179. 17.
17
Söderholm K-JM, Roberts MJ.
Influence of water exposure on the tensile strength of composites.
J Dent Res. 1990;69:1812–1816. MEDLINE 18.
18
Söderholm K-J.
Degradation of glass filler in experimental composites.
J Dent Res. 1981;60:1867–1875. MEDLINE 19.
19
Söderholm K-JM.
Leaking of fillers in dental composites.
J Dent Res. 1983;62:126–130. MEDLINE 20.
20
Hobson RS, McCabe JF, Hogg SD.
The effect of food simulants on the dental composite bond strength.
J Orthod. 2000;27:55–59. MEDLINE 21.
21
Freund M, Munksgaard EC.
Enzymatic degradation of BISGMA/TEGDMA polymers causing decreased microhardness and greater wear in vitro.
Scand J Dent Res. 1990;98:351–355. 22.
22
Larsen IB, Munksgaard EC.
Effect of human saliva on surface degradation of composite resins.
Scand J Dent Res. 1991;99:254–261. 23.
23
Rezk-Lega F, Øgaard B, Arends J.
An in vivo study on the merits of two glass ionomers for the cementation of orthodontic bands.
Am J Orthod Dentofacial Orthop. 1991;99:162–167. Abstract |
Full-Text PDF (397 KB)
|
CrossRef
24.
24
Roulet JF, Wälti C.
Influence of oral fluid on composite resin and glass-ionomer cement.
J Pros Dent. 1984;52:182–188. 25.
25
Fox NA, McCabe JF, Buckley JG.
A critique of bond strength testing in orthodontics.
Br J Orthod. 1994;21:33–43. 26.
26
Kanemura N, Sano H, Tagami J.
Tensile bond strength to and SEM evaluation of ground and intact enamel surfaces.
J Dent. 1999;27:523–530. Abstract | Full Text |
Full-Text PDF (2111 KB)
|
CrossRef
27.
27
Mattick CR, Hobson RS.
A comparative micro-topographic study of the buccal enamel of different tooth types.
J Orthod. 2000;27:143–148. MEDLINE 28.
28
McCabe JF, Carrick TE.
A statistical approach to the mechanical testing of dental materials.
Dent Materials. 1986;2:139–142. 29.
29
Prati C, Tao L, Simpson M, Pashley DH.
Permeability and microleakage of Class II resin composite restorations.
J Dent. 1994;22:49–56. MEDLINE |
CrossRef
30.
30
Lee S-Y, Greener EH, Mueller HJ.
Effect of food and oral simulating fluids on structure of adhesive composite systems.
J Dent. 1995;23:27–35. Abstract |
Full-Text PDF (995 KB)
|
CrossRef
31.
31
International standards organization .
Dental materials-guidance on testing of adhesion to tooth structure.
Geneva, Switzerland: ISO/TR 11405; 1994;. 32.
32
DeWald JP.
The use of extracted teeth for in vitro bonding studies: a review of infection control considerations.
Dent Materials. 1997;13:74–81. 33.
33
Shaffer SE, Barkmeier WW, Gwinnett AJ.
Effect of disinfection/sterilization on in vitro enamel bonding.
J Dent Educ. 1985;49:658–659. MEDLINE Sunderland and Newcastle, United Kingdom aSpecialist registrar, Department of Orthodontics, Sunderland Royal Hospital, Sunderland, United Kingdom ☆ bSenior lecturer, Department of Child Dental Health, Newcastle Dental Hospital, Newcastle, United Kingdom. ☆☆ Reprint requests to: Stephen Murray, Dept of Orthodontics, Merlin Park Regional Hospital, Galway, Republic of Ireland; e-mail, murnau@ireland.com. ★ 0889-5406/2003/$30.00 + 0 PII: S0889-5406(02)56969-9 doi:10.1067/mod.2003.49 © 2003 American Association of Orthodontists. Published by Elsevier Inc. All rights reserved. | |
|
FULL TEXT
