- Email: [email protected]
An in vitro investigation of lingual bonding
An in vitro investigation of lingual bonding Leonard Chumak, DDS, MCID,* Khadry A. Galil, BDS, D. Oral Surg., PhD,** David C. Way, DDS, MS, FRCD (C),*** Leonard N. Johnson, BS, MS, PhD,**** and W. Stuart Hunter, DDS, MS, PhD***** London,
Ontario,
Canada
An in vitro investigatio? was undertaken to evaluate the bonding of orthodontic appliances onto lingual surfaces; 53 maxillary premolars, 37 mandibular premolars, and 37 mandibular incisors were used. Brackets were bonded onto the lingual and lgbial surfaces and fractured with an lnstron machine. Enamel damage associated with debonding also was assessed. Results indicated comparable bond strengths (t test) on lingual (Li) and labial (La) surfaces: maxillary premolars-Li-138.2 kg/cm’,’ La-l 27.7 kg/cm? mandibular premolars-Li-136.2 kg/cm2, La-l 21.6 kg/cm2; and mandibular incisors-Li-166.3 kg/cm2, La-161 .l kg/cm*. Adaptation of lingual bracket bases resulted in significantly higher lingual bond .s!rengths for maxillary premolars (166.9 kg/cm*) and mandibula! premolars (180.4 kg/cm*), but not for mandibular incisors (149.2 kg/cm2). On debonding, the percentages of lingual surfaces exhibiting horizontal “crescent-shaped” fracture lines and enamel fragment fractures were significantly higher (x2 test) than the corresponding percentages for labial surfaces: maxillary premolars--i-67.9%, La-5.7%; mandibular premolars-Lj-62.2%, La-13.5%; and mandibular incisbrs-Li-43.2%, La-18.9%. Furthermore, an increase in vertical enamel fracture lines (cracks) subsequent to debonding was seen labially and lingually. Bonding procedures for lingual surfaces should be identical to those advocated for labial surfaces. Care during debonding must be exercised to eliminate possible enamel damage. (AM J ORTHOD D~NTOFAC ORTHOP 1989;95:20-8.)
T
LA
ingual orthodqntics has added a new dimension to the bonding spectrum; however, questions have arisen with regard to lingual bonding procedures and how, if at all, they should differ from labial bonding procedures. Some clinicians believe that bonding techniques currently accepted for labial surfaces should be the same for lingual surfaces”*15; others believe that certain alterations, such as increased etching time and roughened lingual surfaces, are necessary to achieve satisfactory bond strengths. “Jo Such controversy exists because of paucity of documented research on the histology of lingual enamel, the effects of acid etching on lingual surfaces, and the retentiveness of bonded lingual brackets. Also the variable morphology of lingual
From the University of Western Ontario. Supported by MRC support grant. *Civilian Orthodontist, Canadian Forces Base, Lahr, West Germany. **Honorary Lecturer Anatomy; Oral and Maxillofacial Surgery and Hospital Dentistry. ***Clinical Professor, Department of Orthodontics. ***:Associate Professor. Department of Biomaterials Science and Medical Biophysics. *****Professor and Chairman. Department of Orthodontics.
20
enamel surfaces has led to the suggestion that bracket base adaptation is required if a proper bond is to, be established and for ease of bracket placement.” Strong adhesion of resin to enamel can result in undesirable enamel fractures during debonding. Such fractures have been shown to occur in two forms: either visible fracture lines on the enamel surface or complete separation of an enamel fragment.6.22 Numerous resin systems ranging from unfilled to highly filled have been marketed and ideally one would prefer a system that provides adequate bond strength while preserving enamel integrity. The objective of the present study was to provide an in vitro investigation of bonding orthodontic appliances t0 lingual surfaces as follows: 1. To compare labial and lingual bond strengths on maxillary premolars, mandibular premolars, and mandibular incisors 2. To determine the effect of lingual bracket base adaptation on bond strength 3. To compare the effect of debonding on labial and lingual enamel surfaces with respect to (a) site and nature of bond fracture and (b) the incidence of vertical fracture lines in the enamel
Volume 95 Number 1
In vitro
investigation
of lingual
bonding
21
Fig. 1. Tooth in Densite block mounted in surveyor base modified with metal insert.
MATERIALS AND METHODS
The in vitro samples comprised 53 maxillary premolars (group I), 37 mandibular premolars (group II), and 37 mandibular incisors (group III). The premolars were collected from adolescents who had undergone extraction of these teeth as part of orthodontic treatment, whereas the incisors were gathered from adults who lost these teeth as a result of periodontal involvement. Only noncarious and nonrestored teeth were selected. All teeth were stored in 70% alcohol until they were ready for testing. Then the teeth were mounted in blocks of Class II dental stone (Densite). After removal of any calculi present, the labial and lingual surfaces of each tooth received a 15-second prophylaxis with a rubber cup and a slurry of zirconium silicate,* followed by thorough rinsing and drying. To detect the presence of vertical fracture lines (cracks), each tooth was then examined under a stereo light microscope.? When a light was moved across the teeth at a distance of 5 cm, fracture lines appeared between reflected light and dark areas in the concentrated light beam. The fracture lines were recorded diagrammatically on enlarged drawings that represented the labial and lingual surfaces of each tooth. Labial and lingual surfaces of each tooth were then conditioned for 60 seconds with a 43% phosphoric acid ge1.S Each tooth was rinsed for 10 seconds, followed by final drying. Contoured Siamese orthodontic brackets§ were bonded to the labial surfaces by means of a heavily filled (67%), one-paste, chemically cured resin system.11Lingual brackets§ were bonded to all lingual surfaces with the same technique and resin. However, half the lingual bracket bases in each of the three groups were custom adapted with How pliers before bonding. *Zircate, L.D. Caulk Co., Milford, Del. t&l Zeiss, Oberkochen, West Germany. SConditioner gel, TP Laboratories, La Porte, Ind. 5G.A.C. International, Central Islip, N.Y. IjUnite, Unitek Canada Ltd.. Scarborough, Ontario,
Canada
I Fig. 2. A, Diagrammatic representation of crescent-shaped ture line (type B).
frac-
All bonds were allowed to bench cure for approximately 20 minutes, after which the teeth were stored in water at room temperature for 24 hours to ensure optimum polymerization. The labial and lingual bonds were fractured using an Instron universal testing machine.* The Densite blocks were secured in a surveyor base that was modified with a metal insert to ensure proper stabilization of the block (Fig. 1). Adjusting the universal joint in the surveyor base enabled proper orientation of the teeth so that the direction of force application, when applied at the top of the bracket tie wings, would be as nearly parallel to the bracket base as possible. The load on each bracket was applied at a crosshead speed of 0.02 in/min, and the force was recorded on a strip chart recorder until the breaking point. Shear force values were converted and reported in kilograms per square centimeter (kg/cm*) to account for the difference in bracket base dimension of labial and lingual brackets. Statistical analysis by means of a paired Student’s t test was performed on each of the three groups of teeth to determine whether a statistically significant difference in bond strength could be established between labial nonadapted and lingual nonadapted and adapted brackets. After bond fracture all labial and lingual surfaces and the corresponding bracket bases were examined with the stereo light microscope to determine the site and nature of the bond fracture. Each fracture was classified into one of the following categories: fInStmn
Corp.,
Canton, Mass
22
Am. J. Orthod.
Chumak et al.
Fig.
2 (Cont’d).
B, Scanning electron micrograph of crescent-shaped
Dentofac. Orthop. January 1989
fracture line. (Original magnifi-
cation x 35.)
In addition, the labial and lingual surfaces of all teeth were examined further under the stereo light microscope to determine whether any additional vertical enamel fracture lines occurred as a result of debonding. RESULTS
Fig. 3. A, Diagrammatic representation loss of an ename! fragment (type C).
of enamel fracture with
‘I&e A. Bond fracture occurring at the resin bracket, resinresin, or resin enamel interface. Debonding resulted in no apparent damage to the enamel. Type B. Bond fracture as stated for type A. Debonding resulted in a horizontal “crescent-shaped” fracture line (variable length) visible on the enamel surface (Fig. 2). Type C. Bond fracture involving more extensive damage to the enamel. Debonding resulted in fracture within the enamel and loss of an enamel fragment (Fig. 3).
For maxillary premolars, the shear bond strength mean values of the labial and lingual (nonadapted) bonded brackets were 127.7 kg/cm’ and 138.2 kg/cm2, respectively. Although this suggested a slightly stronger bond strength on the lingual brackets, the difference between the two means was not statistically significant at the 5% level. Comparison of labial with lingual (adapted) bonded brackets showed mean shear strength values of 123.1 kg/cm2 and 166.9 kg/cm2, respectively. This difference was statistically significant, indicating that adaptation resulted in a true increased lingual bond strength as related to labial bond strength (p < 0.01) (Table I). For mandibular premolars, the mean values of the labial and the lingual (nonadapted) shear strengths were 121.6 kg/cm2 and 136.2 kg/cm2, respectively. Similar to maxillary premolars, the mean values were not significantly different. Adaptation of lingual bracket bases resulted in a
Volume 95 Number 1
In vitro
Fig. 3 (Cont’d). B, Scanning electron (Original magnification x 35.)
micrograph
of enamel
fracture
investigation
with loss of an enamel
Table I. Comparison of shear strengths of labial and lingual bonded brackets-Maxillary Mean
Group Group A Labial Lingual Group B Labial Lingual
NAB
= Nonadapted
(kg I cm’)
127.7 138.2 Means
(NAB)
123.1 166.9 Means
(AB)
brackets;
AB
= adapted
SD (kg I cm’)
not significantly
significantly
13.3 21.4 different, p > 0.05 10.3 31.5 different, p < 0.01
of lingual
bonding
23
fragment.
premolars Range
(kglcmz)
103.3-155.9 91.2-182.3
108.7-145.0 123.6-224.8
brackets.
significantly higher mean shear strength value of 180.4 kg/cm* when compared with the labial mean shear strength value of 128.3 kg/cm* (p < O.OOl), which again was similar to the findings for maxillary premolars (Table II). The mandibular incisor results did not follow the same trend observed for the maxillary and mandibular premolars. Comparing the shear bond strengths of the labial with the lingual (nonadapted) brackets, the mean values were 161.1 kg/cm* and 166.3 kg/cm’, respectively. Adaptation of the lingual bracket bases resulted in a mean shear strength value of 149.2 kg/cm*; when
compared to the labial mean shear strength value of 162.9 kg/cm*, the difference was not found to be statistically significant at the 5% level (Table III). With the bond fracture classification (types A, B, and C) previously described, the following results were noted. On labial surfaces of maxillary premolars, a resin fracture with no enamel damage (type A) was observed in most instances (94.3%). However, on lingual surfaces the percentage of fractures classified as types B and C was 76% for nonadapted and 58% for adapted brackets. A chi-square statistical analysis con-
24
Am. J. Orthod.
Chumak et al.
Table II. Comparison of shear strengths of labial and lingual bonded brackets-Mandibular Group Group A Labial Lingual Group B Labial Lingual
NAB
N
14 14
(NAB)
brackets;
AB
= adapted
(kg I cm=)
121.6 136.2 Means
14 14
(AB)
= Nonadapted
Mean
not significantly
128.3 180.4 Means
Range
p > 0.05
15.3 23.6 different, p < 0.0001
significantly
101.5-163.1 151.9-222.8
brackets.
Group
Group B Labial Lingual
NAB
= Nonadapted
Mean
(kglcm’)
161.1 166.3 Means
(NAB)
162.9 149.2 Means
(AB)
brackets;
AB
= adapted
(kglcm2)
94.2-155.9 87.1-172.2
Table Ill. Comparison of shear strengths of labial and lingual bonded brackets-Mandibular
Group A Labial Lingual
premolars
SD (kg I cm’)
20.2 25.3 different,
Dentofac. Orthop. January 1989
incisors
SD (kglcm2)
Range
(kg/cm=)
122.2-190.0 136.4-213.1
not significantly
18.2 21.7 different, p > 0.05
131.9-188.1 110.3-213.1
not significantly
13.9 30.9 different, p > 0.05
brackets.
Table IV. Fracture site on debonding maxillary premolars Group Fracture A-Resin fracture B-Resin fracture with on enamel surface C-Enamel fracture
NAB
= Nonadapted
brackets;
type
crescent-shaped
fracture
line
= adapted
Group
B
Labial
Lingual (NAB)
Labial
Lingual W)
28 (97%) 1 (3%)
7 (24%) 11 (38%)
22 (92%) 1 (4%)
10 (42%) 6 (25%)
0 (0%) x2 = 28.80; AB
A
11 (38%) 1 df; p < 0.001
1 (4%) x2 = 17.64;
1 dfip
8 (33%) < 0.001
brackets.
firmed this to be a significant difference (p < 0.001) (Table IV). The crescent fracture lines (type B) were situated horizontally on the lingual surface in the area in which the superior edge of the bracket base was originally placed. These crescent fracture lines varied in length from a few millimeters to larger ones occupying two thirds the width of the lingual surface (Fig. 2, A). Minute crescents were difficult to identify visually without the use of a stereo light microscope. The enamel surface layer in the area of these small crescent fracture lines remained intact. In larger crescents surface integrity became disrupted (Fig. 2, B).
When total enamel fractures (type C) occurred on lingual surfaces, the incisal portion of the enamel fracture followed the configuration of a crescent (Fig. 3). In a buccolingual longitudinal section, enamel fragment fractures appeared wedge-shaped and invariably extended to the dentinoenamel junction. Scanning electron microscopy showed that the superior fracture plane primarily followed longitudinal sections of enamel prisms, and only infrequently crossed prisms transversely. The fracture site observations for mandibular premolars were similar to maxillary premolars. On lingual surfaces the percentage of fractures classified as types
Volume 95
Number
In vitro investigation of lingual bonding 25
1
Table V. Fracture site on debonding mandibular premolars GroupA Fracture
type
A-Resin fracture B-Resin fracture with crescent-shaped on enamel surface C-Enamel fracture
fracture
line
= Nonadapted
brackets;
AB
= adapted
B
Labial
Lingual (NAB)
Labial
Lingual (a)
17 (89%) 2(11%)
7(370/c) 7(370/c)
15 (83%) 3(17%)
7 (39%) 7 (39%)
0 (0%) x' =
NAB
Group
5 (26%) 9.16:l
df;p
0 (0%) x' = 5.72;
< 0.01
4(22%)
I df; p < 0.02
brackets.
B and C was significantly higher than on respective labial surfaces for both nonadapted (p < 0.01) and adapted groups ( p < 0.02) (Table V). Fracture site examination was performed on the incisor group as a whole since nonadapted and adapted shear strength mean values failed to show a significant difference. Similar to premolars, the majority of fractures on the labial surfaces were resin fractures (type A) with no enamel damage (8 1%). Lingually the percentage of such fractures was less, with types B and C present in 43% of the sample. The chi-square statistical test showed this to be a significant difference (p < 0.05) (Table VI). All teeth exhibited an increase in the number of vertical fracture lines (cracks) labially and lingually after debonding. The numerical data collected were insufficient to quantify such results. DlSCUSSlON
It has been suggested’2.‘9 that surface grinding is necessary for clinical success in lingual bonding. However, the need to roughen enamel to increase bond strength has never been documented conc1usively,4.‘3 even though grinding possibly may improve the etch pattern.’ The results of this study indicate that lingual bond strengths are comparable to labial strengths; if this also is true in vivo, it would eliminate the need for such a procedure. Furthermore, favorable lingual bond strengths for all three groups of teeth investigated also would suggest that increased etching time on lingual surfaces is not necessary. However, comparison of labial and lingual etching patterns has never been documented and merits investigation if support is to be given to the above statement. Since the anatomic configuration of lingual surfaces is highly variable, especially for mandibular premolars that may have one or two lingual cusps,’ custom adaptation of bracket bases before bonding has been recommended.”
Table VI. Fracture site on debonding mandibular incisors (nonadapted and adapted groups combined) Fracture
type
A-Resin fracture B-Resin fracture with crescent-shaped fracture line on enamel surface C-Enamel fracture
I
Labial
Lingual
30(81%)
21(57%)
4 (11%)
3(8%) x2 = 4.04;
3 (8%)
13(35%)
1 df; p < 0.05
Adaptation of the lingual bracket bases in half of each sample group yielded significantly higher lingual mean shear strengths for the premolars as compared with their respective labial shear strengths, whereas the incisors failed to exhibit such a relationship. These results may be explained in the following manner. First, lingual premolar brackets are designed in such a way that their bases can be adapted easier than incisor bracket bases. Second, the lingual surfaces of both maxillary and mandibular premolars are more uniformly convex as compared to lingual incisor contour, which can vary from concave to convex. This variable lingual incisor contour may result in microscopic differences in surface enamel rod concentrations between the concave and convex parts. However, the relationship between such differences in surface enamel rod concentrations and bond strength has not been investigated and remains unclear. Buonocore’ stated that a thin, continuous adhesive layer between the bracket base and the tooth surface would result in greater bond strength. Correspondingly, Evans and Powers8 showed that tensile bond strength decreased as resin thickness increased. Therefore,‘adaptation of lingual premolar brackets in which the curvature of the bracket base closely approximated.that of the tooth surface produced a thin, uniform layer of
26
A O/ LINGUAL
0B
Am. .I. Orthod. Dentofac. Orthop. January 1989
Chumak et al.
P
A
SIDE VIEW
LINGUAL
---
TOP
VIEW
failed to conform with the contour of the tooth. In such situations the bracket would be forced into contact at the highest point of curvature of the enamel surface, thereby displacing most of the adhesive from this area. The lack of adhesive in this area would result in a starved joint.5 Simultaneously, on either side of the contact area, large areas of resin would be present to fill the space discrepancy between the curvature of tooth and the bracket (Fig. 4, C). Since a one-paste chemically cured resin system was used, insufficient contact between the resin primer and such large areas of resin would result in incomplete polymerization. Similar results have been document&d by Evans and Powers.’ Clinically this lack of polymerization has possible implications: unpolymerized resin could potentially become replaced with plaque and a decrease in bond strength could result. Zachrisson, Skogan, and Hoymyhl-22 studied enamel fracture lines on labial surfaces of debonded teeth and found these teeth to have significantly more vertical fracture lines than those of an untreated control group. The results of the present study tie in agreement with Zachrisson and associates and similar trends also were observed’on lingual surfaces. Furthermore, on lingual surfaces, debonding resulted in a significantly. higher prevalence of horizontal crescent-shaped fracture lines (type B) and enamel fractures with loss of an enamel fragment (type C) in all three groups of teeth examined. Crescent fracture lines were initiated internally in stressed enamel in the area in which the superior edge of the bracket base was originally placed. They appeared to be the initial stage sf enamel damage leading to possible enamel fragment loss. As the force on the bracket was increased, either the resin would fracture, resulting in a crescent-shaped fracture line on the enamel surface, or enamel fragment fracture would occur. Studies by Eick and associates’ are in agreement. Scanning electron microscopy showed that the superior fracture plane of an enamel fracture primarily followed longitudinal sections of enamel prisms. Other authors14,16are in agreement, indicating this to be the most likely line of fracture. Possibly lingual enamel rod orientation is more oblique to the surface, resulting in greater susceptibility to enamel fracture. A limited investigation performed in this study suggested that labial enamel thickness was greater than lingual enamel thickness for both maxillary and mandibular premolars; Wakuri” confirmed these findings. Furthermore, Gillings and BuonocoTe” demonstrated that enamel was very thin on lingual surfaces of incisors. Because it is a brittle material, enamel will resist fracture only if it is present in sufficient bulk;
--b A
LINGUAL
I
SIDE VIEW
LINGUAL
TOP
VIEW
Flg. 4. A, Premolar. Shaded area represents a uniform adhesive layer between a properly contoured bracket base and corresponding tooth surface. B, Incisor. Shaded area represents a nonuniform adhesive layer between an improperly contoured bracket base and corresponding tooth surface. C, Shaded area represents a nonunifprm and discontinuous adhesive layer with the presence of a starved joini (A, minimal r&in) and areas of excess resin (B, unpolymerized resin) at the interface between an improperly contoured bracket base and corresponding tooth surface.
adhesive, thereby achieving superior bond strength (Fig. 4, A). The variable surface contour of the incisors combined with the inability to properly adapt bracket bases resulted in varying thicknesses of resin between the bonding pad and tooth surface, and consequently varying bond strengths (Fig. 4, B). Unpolymerized resin, detected by scratching with an explorer, was occasionally present in the nonadapted sample in which the bracket bonding pad curvature
Volume 95 i’hnber 1
thinning lingual enamel possibly is more prone to fracture. Longitudinal sections of types B and C fractures showed that crescent fracture lines and enamel fragment fractures routinely extended to and involved the dentinoenamel junction. Enamel fracture is likely to occur at the junction because it is an area of transformation between two dental tissues with different prope~ies.3,‘6‘18
In vivo the small crystals of dentin merge imperceptibly with the larger crystals of ename1,2 therefore ‘. structurally reinforcing the enamel cap. However, in vitro dentin desiccation may occur.21 This may reduce enamel cap support significantly, leading to enamel fracture at the dentinoenamel junction. It is therefore proposed that crescent fracture lines and enamel fragment fractures occur with increased frequency on lingual surfaces as compared with labial surfaces because (1) the enamel rod orientation may render the lingual enamel more prone to fracture and (2) lingual enamel is thinner. These two factors in an in vitro environment in which enamel may have lost structural support from the underlying desiccated dentin yielded the results documented. It is evident from the preceding discussion that undesirable enamel damage may occur during debonding in vitro, especially on lingual surfaces. Clinically the occurrence of crescent fracture lines and enamel fractures has never been reported on lingual surfaces. However, as stated in the results section, they are difficult to see and if present may easily be overlooked. It also is important to note that in clinical situations accidental debonding of orthodontic brackets by occlusal forces is more analogous to the experimental debonding design presented in this article than to conventional debonding procedures. In accidental debonding and in this study, the force to dislodge the brackets is directed gingivally. Conventional debonding does not necessarily produce such a force direction. Hence teeth that have lost brackets as a result of occlusal forces possibly should undergo careful scrutinization for enamel damage. Furthermore, the possibility of dentin desiccation as a contributing factor in types @ and C fractures has been proposed. If tiue, great attention must be focused on devitalized teeth in vivo because their susceptibility to lingual surface enamel fractures may be quite high. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS
1. In vitro lingual bond strengths are comparable to labial bond strengths, which would suggest that the labial bonding technique need not be altered for lingual
In vitro investigation
of lingual bonding
27
bonding. Therefore it is not necessary to roughen lingual surfaces or to increase etching time for lingual bonding. 2. Lingual bracket base adaptation produces a uniform thin layer of resin and a strong bond, and also minimizes unpolymerized resin when a one-paste resin system is used. This is accomplished more easily for premolars than for mandibular incisors. 3. Adaptation of lingual bracket bases is recommended for the following reasons: (a) ease of bracket placement, (b) minimizing of unpolymerized resin beneath bracket bases, and (c) an increase in bond strength. 4.’ There is an increase in vertical enamel fracture lines both labially and lingually when these surfaces are debonded with a vertical shear force in vitro. 5. Lingual surfaces in vitro are more susceptible than labial surfaces to horizontal crescent fracture line formation and enamel fractures with loss of a fragment when brackets are removed with a vertical shear force. Adaptation of lingual bracket bases does not appear to increase the incidence of such fractures. 6. Care must be exercised when debonding to minimize enamel stress. REFERENCES 1. Ash MM. Wheeler’s Atlas of tooth form. 5th ed. Philadelphia: WB Saunders, 1984. 2. Bhaskar SN. Oral histology and embryology. 9th ed. St. Louis: The CV Mosby Company, 1980:69-71. MS. Tensile strength and modulus of 3. Bowen RL, Rodriguez elasticity of tooth structure and several restorative materials. J Am Dent Assoc 1962;64:378-87. 4. Bozalis WG, Marshall GW, Cooley RO. Mechanical pretreatment and etching of primary-tooth enamel. ASDC J Dent Child 1979;44:43-9. 5. Buonocore MG. The use of adhesives in dentistry. 1 st ed. Springfield, Illinois: Charles C Thomas, 1975:25-30. 6. Diedrich P. Enamel alterations from bracket bonding and debonding: a study with the scanning electron microscope. AM J ORTHOD 1981;79:500-22. AW. I. Eick JD, Johnson LN, Fromer JR, Good RJ, Neumann Surface topography: its influence on wetting and adhesion in a dental adhesive system. J Dent Res 1972:.51:780-8. 8. Evans LB, Powers JM. Factors affecting in vitro bond strength of no-mix orthodontic cements. AM J ORTHOD 1985;87:508-12. 9. Galil KA, Wright GZ. Acid etch patterns on buccal surfaces of permanent teeth. Pediatr Dent 1979;1:230-4. 10. Gillings B, Buonocore M. An investigation of enamel thickness in human lower incisor teeth. J Dent Res 1961;40:105-18. (personal communication). Commack, New 11. GAC International York: 1985. 12. Kelly VM. JCO interviews. J Clin Orthod 1982;16:461-76. 13. Knoll M, Gwinett AJ, Wolff M. Should primary enamel be ground prior to bonding. J Clin Orthod 1985;19: 137-8. 14. Munechika T, Suzuki K, Nishiyama M, Ohashi M, Horie K. A comparison of the tensile bond strengths of composite resins to
28
15. 16. 17. 18. 19. 20.
Chunmk
Am. J. Orthod.
et al.
longitudinal and transverse sections of enamel prisms in human teeth. J Dent Res 1984;63:1079-82. Ormco, Division of Sybron Corporation. Lingual orthodontics, bonding instructions. Glendora, California: 1985. Rasmussen ST, Patchin RE, Scott DB, Heuer AM. Fracture properties of human enamel and dentine. J Dent Res 1976; 5.5:154-64. Rasmussen ST. Fracture properties of human teeth in proximity to the dentinoenamel junction. J Dent Res 1984;63: 1279-83. Stanford JW, Weigel KV, Paffenbarger GC, Sweeney WT. Compressive properties of hard tooth tissues and some restorative materials. J Am Dent Assoc 1960;60:746-56. Unitek Corporation. Lingual orthodontics, bonding instructions. Monrovia, California: 1985. Wakuri H. The experimental studies of enamel cap. Bull Tokyo Dent Co11 1971;12:117-47.
Dentofac. .huary
Orthop. 1989
21 Whittaker DK. The enamel-dentine junction of human and Macaca irus teeth: a light and electron microscope study. J Anat 1978;125:323-35. 22. Zachrisson BU, Skogan 0, Hoymyhr S. Enamel cracks in debonded, debanded, and orthodontically untreated teeth. AM J ORTHOD1980;77:307-19. Reprint
requests
to:
Dr. Leonard Chumak Civilian Dental Clinic Canadian Forces Europe C.F.P.O. 5000 Bell~ille, Ontario, Canada KOK 3R0
Ontario,
Canada
An in vitro investigatio? was undertaken to evaluate the bonding of orthodontic appliances onto lingual surfaces; 53 maxillary premolars, 37 mandibular premolars, and 37 mandibular incisors were used. Brackets were bonded onto the lingual and lgbial surfaces and fractured with an lnstron machine. Enamel damage associated with debonding also was assessed. Results indicated comparable bond strengths (t test) on lingual (Li) and labial (La) surfaces: maxillary premolars-Li-138.2 kg/cm’,’ La-l 27.7 kg/cm? mandibular premolars-Li-136.2 kg/cm2, La-l 21.6 kg/cm2; and mandibular incisors-Li-166.3 kg/cm2, La-161 .l kg/cm*. Adaptation of lingual bracket bases resulted in significantly higher lingual bond .s!rengths for maxillary premolars (166.9 kg/cm*) and mandibula! premolars (180.4 kg/cm*), but not for mandibular incisors (149.2 kg/cm2). On debonding, the percentages of lingual surfaces exhibiting horizontal “crescent-shaped” fracture lines and enamel fragment fractures were significantly higher (x2 test) than the corresponding percentages for labial surfaces: maxillary premolars--i-67.9%, La-5.7%; mandibular premolars-Lj-62.2%, La-13.5%; and mandibular incisbrs-Li-43.2%, La-18.9%. Furthermore, an increase in vertical enamel fracture lines (cracks) subsequent to debonding was seen labially and lingually. Bonding procedures for lingual surfaces should be identical to those advocated for labial surfaces. Care during debonding must be exercised to eliminate possible enamel damage. (AM J ORTHOD D~NTOFAC ORTHOP 1989;95:20-8.)
T
LA
ingual orthodqntics has added a new dimension to the bonding spectrum; however, questions have arisen with regard to lingual bonding procedures and how, if at all, they should differ from labial bonding procedures. Some clinicians believe that bonding techniques currently accepted for labial surfaces should be the same for lingual surfaces”*15; others believe that certain alterations, such as increased etching time and roughened lingual surfaces, are necessary to achieve satisfactory bond strengths. “Jo Such controversy exists because of paucity of documented research on the histology of lingual enamel, the effects of acid etching on lingual surfaces, and the retentiveness of bonded lingual brackets. Also the variable morphology of lingual
From the University of Western Ontario. Supported by MRC support grant. *Civilian Orthodontist, Canadian Forces Base, Lahr, West Germany. **Honorary Lecturer Anatomy; Oral and Maxillofacial Surgery and Hospital Dentistry. ***Clinical Professor, Department of Orthodontics. ***:Associate Professor. Department of Biomaterials Science and Medical Biophysics. *****Professor and Chairman. Department of Orthodontics.
20
enamel surfaces has led to the suggestion that bracket base adaptation is required if a proper bond is to, be established and for ease of bracket placement.” Strong adhesion of resin to enamel can result in undesirable enamel fractures during debonding. Such fractures have been shown to occur in two forms: either visible fracture lines on the enamel surface or complete separation of an enamel fragment.6.22 Numerous resin systems ranging from unfilled to highly filled have been marketed and ideally one would prefer a system that provides adequate bond strength while preserving enamel integrity. The objective of the present study was to provide an in vitro investigation of bonding orthodontic appliances t0 lingual surfaces as follows: 1. To compare labial and lingual bond strengths on maxillary premolars, mandibular premolars, and mandibular incisors 2. To determine the effect of lingual bracket base adaptation on bond strength 3. To compare the effect of debonding on labial and lingual enamel surfaces with respect to (a) site and nature of bond fracture and (b) the incidence of vertical fracture lines in the enamel
Volume 95 Number 1
In vitro
investigation
of lingual
bonding
21
Fig. 1. Tooth in Densite block mounted in surveyor base modified with metal insert.
MATERIALS AND METHODS
The in vitro samples comprised 53 maxillary premolars (group I), 37 mandibular premolars (group II), and 37 mandibular incisors (group III). The premolars were collected from adolescents who had undergone extraction of these teeth as part of orthodontic treatment, whereas the incisors were gathered from adults who lost these teeth as a result of periodontal involvement. Only noncarious and nonrestored teeth were selected. All teeth were stored in 70% alcohol until they were ready for testing. Then the teeth were mounted in blocks of Class II dental stone (Densite). After removal of any calculi present, the labial and lingual surfaces of each tooth received a 15-second prophylaxis with a rubber cup and a slurry of zirconium silicate,* followed by thorough rinsing and drying. To detect the presence of vertical fracture lines (cracks), each tooth was then examined under a stereo light microscope.? When a light was moved across the teeth at a distance of 5 cm, fracture lines appeared between reflected light and dark areas in the concentrated light beam. The fracture lines were recorded diagrammatically on enlarged drawings that represented the labial and lingual surfaces of each tooth. Labial and lingual surfaces of each tooth were then conditioned for 60 seconds with a 43% phosphoric acid ge1.S Each tooth was rinsed for 10 seconds, followed by final drying. Contoured Siamese orthodontic brackets§ were bonded to the labial surfaces by means of a heavily filled (67%), one-paste, chemically cured resin system.11Lingual brackets§ were bonded to all lingual surfaces with the same technique and resin. However, half the lingual bracket bases in each of the three groups were custom adapted with How pliers before bonding. *Zircate, L.D. Caulk Co., Milford, Del. t&l Zeiss, Oberkochen, West Germany. SConditioner gel, TP Laboratories, La Porte, Ind. 5G.A.C. International, Central Islip, N.Y. IjUnite, Unitek Canada Ltd.. Scarborough, Ontario,
Canada
I Fig. 2. A, Diagrammatic representation of crescent-shaped ture line (type B).
frac-
All bonds were allowed to bench cure for approximately 20 minutes, after which the teeth were stored in water at room temperature for 24 hours to ensure optimum polymerization. The labial and lingual bonds were fractured using an Instron universal testing machine.* The Densite blocks were secured in a surveyor base that was modified with a metal insert to ensure proper stabilization of the block (Fig. 1). Adjusting the universal joint in the surveyor base enabled proper orientation of the teeth so that the direction of force application, when applied at the top of the bracket tie wings, would be as nearly parallel to the bracket base as possible. The load on each bracket was applied at a crosshead speed of 0.02 in/min, and the force was recorded on a strip chart recorder until the breaking point. Shear force values were converted and reported in kilograms per square centimeter (kg/cm*) to account for the difference in bracket base dimension of labial and lingual brackets. Statistical analysis by means of a paired Student’s t test was performed on each of the three groups of teeth to determine whether a statistically significant difference in bond strength could be established between labial nonadapted and lingual nonadapted and adapted brackets. After bond fracture all labial and lingual surfaces and the corresponding bracket bases were examined with the stereo light microscope to determine the site and nature of the bond fracture. Each fracture was classified into one of the following categories: fInStmn
Corp.,
Canton, Mass
22
Am. J. Orthod.
Chumak et al.
Fig.
2 (Cont’d).
B, Scanning electron micrograph of crescent-shaped
Dentofac. Orthop. January 1989
fracture line. (Original magnifi-
cation x 35.)
In addition, the labial and lingual surfaces of all teeth were examined further under the stereo light microscope to determine whether any additional vertical enamel fracture lines occurred as a result of debonding. RESULTS
Fig. 3. A, Diagrammatic representation loss of an ename! fragment (type C).
of enamel fracture with
‘I&e A. Bond fracture occurring at the resin bracket, resinresin, or resin enamel interface. Debonding resulted in no apparent damage to the enamel. Type B. Bond fracture as stated for type A. Debonding resulted in a horizontal “crescent-shaped” fracture line (variable length) visible on the enamel surface (Fig. 2). Type C. Bond fracture involving more extensive damage to the enamel. Debonding resulted in fracture within the enamel and loss of an enamel fragment (Fig. 3).
For maxillary premolars, the shear bond strength mean values of the labial and lingual (nonadapted) bonded brackets were 127.7 kg/cm’ and 138.2 kg/cm2, respectively. Although this suggested a slightly stronger bond strength on the lingual brackets, the difference between the two means was not statistically significant at the 5% level. Comparison of labial with lingual (adapted) bonded brackets showed mean shear strength values of 123.1 kg/cm2 and 166.9 kg/cm2, respectively. This difference was statistically significant, indicating that adaptation resulted in a true increased lingual bond strength as related to labial bond strength (p < 0.01) (Table I). For mandibular premolars, the mean values of the labial and the lingual (nonadapted) shear strengths were 121.6 kg/cm2 and 136.2 kg/cm2, respectively. Similar to maxillary premolars, the mean values were not significantly different. Adaptation of lingual bracket bases resulted in a
Volume 95 Number 1
In vitro
Fig. 3 (Cont’d). B, Scanning electron (Original magnification x 35.)
micrograph
of enamel
fracture
investigation
with loss of an enamel
Table I. Comparison of shear strengths of labial and lingual bonded brackets-Maxillary Mean
Group Group A Labial Lingual Group B Labial Lingual
NAB
= Nonadapted
(kg I cm’)
127.7 138.2 Means
(NAB)
123.1 166.9 Means
(AB)
brackets;
AB
= adapted
SD (kg I cm’)
not significantly
significantly
13.3 21.4 different, p > 0.05 10.3 31.5 different, p < 0.01
of lingual
bonding
23
fragment.
premolars Range
(kglcmz)
103.3-155.9 91.2-182.3
108.7-145.0 123.6-224.8
brackets.
significantly higher mean shear strength value of 180.4 kg/cm* when compared with the labial mean shear strength value of 128.3 kg/cm* (p < O.OOl), which again was similar to the findings for maxillary premolars (Table II). The mandibular incisor results did not follow the same trend observed for the maxillary and mandibular premolars. Comparing the shear bond strengths of the labial with the lingual (nonadapted) brackets, the mean values were 161.1 kg/cm* and 166.3 kg/cm’, respectively. Adaptation of the lingual bracket bases resulted in a mean shear strength value of 149.2 kg/cm*; when
compared to the labial mean shear strength value of 162.9 kg/cm*, the difference was not found to be statistically significant at the 5% level (Table III). With the bond fracture classification (types A, B, and C) previously described, the following results were noted. On labial surfaces of maxillary premolars, a resin fracture with no enamel damage (type A) was observed in most instances (94.3%). However, on lingual surfaces the percentage of fractures classified as types B and C was 76% for nonadapted and 58% for adapted brackets. A chi-square statistical analysis con-
24
Am. J. Orthod.
Chumak et al.
Table II. Comparison of shear strengths of labial and lingual bonded brackets-Mandibular Group Group A Labial Lingual Group B Labial Lingual
NAB
N
14 14
(NAB)
brackets;
AB
= adapted
(kg I cm=)
121.6 136.2 Means
14 14
(AB)
= Nonadapted
Mean
not significantly
128.3 180.4 Means
Range
p > 0.05
15.3 23.6 different, p < 0.0001
significantly
101.5-163.1 151.9-222.8
brackets.
Group
Group B Labial Lingual
NAB
= Nonadapted
Mean
(kglcm’)
161.1 166.3 Means
(NAB)
162.9 149.2 Means
(AB)
brackets;
AB
= adapted
(kglcm2)
94.2-155.9 87.1-172.2
Table Ill. Comparison of shear strengths of labial and lingual bonded brackets-Mandibular
Group A Labial Lingual
premolars
SD (kg I cm’)
20.2 25.3 different,
Dentofac. Orthop. January 1989
incisors
SD (kglcm2)
Range
(kg/cm=)
122.2-190.0 136.4-213.1
not significantly
18.2 21.7 different, p > 0.05
131.9-188.1 110.3-213.1
not significantly
13.9 30.9 different, p > 0.05
brackets.
Table IV. Fracture site on debonding maxillary premolars Group Fracture A-Resin fracture B-Resin fracture with on enamel surface C-Enamel fracture
NAB
= Nonadapted
brackets;
type
crescent-shaped
fracture
line
= adapted
Group
B
Labial
Lingual (NAB)
Labial
Lingual W)
28 (97%) 1 (3%)
7 (24%) 11 (38%)
22 (92%) 1 (4%)
10 (42%) 6 (25%)
0 (0%) x2 = 28.80; AB
A
11 (38%) 1 df; p < 0.001
1 (4%) x2 = 17.64;
1 dfip
8 (33%) < 0.001
brackets.
firmed this to be a significant difference (p < 0.001) (Table IV). The crescent fracture lines (type B) were situated horizontally on the lingual surface in the area in which the superior edge of the bracket base was originally placed. These crescent fracture lines varied in length from a few millimeters to larger ones occupying two thirds the width of the lingual surface (Fig. 2, A). Minute crescents were difficult to identify visually without the use of a stereo light microscope. The enamel surface layer in the area of these small crescent fracture lines remained intact. In larger crescents surface integrity became disrupted (Fig. 2, B).
When total enamel fractures (type C) occurred on lingual surfaces, the incisal portion of the enamel fracture followed the configuration of a crescent (Fig. 3). In a buccolingual longitudinal section, enamel fragment fractures appeared wedge-shaped and invariably extended to the dentinoenamel junction. Scanning electron microscopy showed that the superior fracture plane primarily followed longitudinal sections of enamel prisms, and only infrequently crossed prisms transversely. The fracture site observations for mandibular premolars were similar to maxillary premolars. On lingual surfaces the percentage of fractures classified as types
Volume 95
Number
In vitro investigation of lingual bonding 25
1
Table V. Fracture site on debonding mandibular premolars GroupA Fracture
type
A-Resin fracture B-Resin fracture with crescent-shaped on enamel surface C-Enamel fracture
fracture
line
= Nonadapted
brackets;
AB
= adapted
B
Labial
Lingual (NAB)
Labial
Lingual (a)
17 (89%) 2(11%)
7(370/c) 7(370/c)
15 (83%) 3(17%)
7 (39%) 7 (39%)
0 (0%) x' =
NAB
Group
5 (26%) 9.16:l
df;p
0 (0%) x' = 5.72;
< 0.01
4(22%)
I df; p < 0.02
brackets.
B and C was significantly higher than on respective labial surfaces for both nonadapted (p < 0.01) and adapted groups ( p < 0.02) (Table V). Fracture site examination was performed on the incisor group as a whole since nonadapted and adapted shear strength mean values failed to show a significant difference. Similar to premolars, the majority of fractures on the labial surfaces were resin fractures (type A) with no enamel damage (8 1%). Lingually the percentage of such fractures was less, with types B and C present in 43% of the sample. The chi-square statistical test showed this to be a significant difference (p < 0.05) (Table VI). All teeth exhibited an increase in the number of vertical fracture lines (cracks) labially and lingually after debonding. The numerical data collected were insufficient to quantify such results. DlSCUSSlON
It has been suggested’2.‘9 that surface grinding is necessary for clinical success in lingual bonding. However, the need to roughen enamel to increase bond strength has never been documented conc1usively,4.‘3 even though grinding possibly may improve the etch pattern.’ The results of this study indicate that lingual bond strengths are comparable to labial strengths; if this also is true in vivo, it would eliminate the need for such a procedure. Furthermore, favorable lingual bond strengths for all three groups of teeth investigated also would suggest that increased etching time on lingual surfaces is not necessary. However, comparison of labial and lingual etching patterns has never been documented and merits investigation if support is to be given to the above statement. Since the anatomic configuration of lingual surfaces is highly variable, especially for mandibular premolars that may have one or two lingual cusps,’ custom adaptation of bracket bases before bonding has been recommended.”
Table VI. Fracture site on debonding mandibular incisors (nonadapted and adapted groups combined) Fracture
type
A-Resin fracture B-Resin fracture with crescent-shaped fracture line on enamel surface C-Enamel fracture
I
Labial
Lingual
30(81%)
21(57%)
4 (11%)
3(8%) x2 = 4.04;
3 (8%)
13(35%)
1 df; p < 0.05
Adaptation of the lingual bracket bases in half of each sample group yielded significantly higher lingual mean shear strengths for the premolars as compared with their respective labial shear strengths, whereas the incisors failed to exhibit such a relationship. These results may be explained in the following manner. First, lingual premolar brackets are designed in such a way that their bases can be adapted easier than incisor bracket bases. Second, the lingual surfaces of both maxillary and mandibular premolars are more uniformly convex as compared to lingual incisor contour, which can vary from concave to convex. This variable lingual incisor contour may result in microscopic differences in surface enamel rod concentrations between the concave and convex parts. However, the relationship between such differences in surface enamel rod concentrations and bond strength has not been investigated and remains unclear. Buonocore’ stated that a thin, continuous adhesive layer between the bracket base and the tooth surface would result in greater bond strength. Correspondingly, Evans and Powers8 showed that tensile bond strength decreased as resin thickness increased. Therefore,‘adaptation of lingual premolar brackets in which the curvature of the bracket base closely approximated.that of the tooth surface produced a thin, uniform layer of
26
A O/ LINGUAL
0B
Am. .I. Orthod. Dentofac. Orthop. January 1989
Chumak et al.
P
A
SIDE VIEW
LINGUAL
---
TOP
VIEW
failed to conform with the contour of the tooth. In such situations the bracket would be forced into contact at the highest point of curvature of the enamel surface, thereby displacing most of the adhesive from this area. The lack of adhesive in this area would result in a starved joint.5 Simultaneously, on either side of the contact area, large areas of resin would be present to fill the space discrepancy between the curvature of tooth and the bracket (Fig. 4, C). Since a one-paste chemically cured resin system was used, insufficient contact between the resin primer and such large areas of resin would result in incomplete polymerization. Similar results have been document&d by Evans and Powers.’ Clinically this lack of polymerization has possible implications: unpolymerized resin could potentially become replaced with plaque and a decrease in bond strength could result. Zachrisson, Skogan, and Hoymyhl-22 studied enamel fracture lines on labial surfaces of debonded teeth and found these teeth to have significantly more vertical fracture lines than those of an untreated control group. The results of the present study tie in agreement with Zachrisson and associates and similar trends also were observed’on lingual surfaces. Furthermore, on lingual surfaces, debonding resulted in a significantly. higher prevalence of horizontal crescent-shaped fracture lines (type B) and enamel fractures with loss of an enamel fragment (type C) in all three groups of teeth examined. Crescent fracture lines were initiated internally in stressed enamel in the area in which the superior edge of the bracket base was originally placed. They appeared to be the initial stage sf enamel damage leading to possible enamel fragment loss. As the force on the bracket was increased, either the resin would fracture, resulting in a crescent-shaped fracture line on the enamel surface, or enamel fragment fracture would occur. Studies by Eick and associates’ are in agreement. Scanning electron microscopy showed that the superior fracture plane of an enamel fracture primarily followed longitudinal sections of enamel prisms. Other authors14,16are in agreement, indicating this to be the most likely line of fracture. Possibly lingual enamel rod orientation is more oblique to the surface, resulting in greater susceptibility to enamel fracture. A limited investigation performed in this study suggested that labial enamel thickness was greater than lingual enamel thickness for both maxillary and mandibular premolars; Wakuri” confirmed these findings. Furthermore, Gillings and BuonocoTe” demonstrated that enamel was very thin on lingual surfaces of incisors. Because it is a brittle material, enamel will resist fracture only if it is present in sufficient bulk;
--b A
LINGUAL
I
SIDE VIEW
LINGUAL
TOP
VIEW
Flg. 4. A, Premolar. Shaded area represents a uniform adhesive layer between a properly contoured bracket base and corresponding tooth surface. B, Incisor. Shaded area represents a nonuniform adhesive layer between an improperly contoured bracket base and corresponding tooth surface. C, Shaded area represents a nonunifprm and discontinuous adhesive layer with the presence of a starved joini (A, minimal r&in) and areas of excess resin (B, unpolymerized resin) at the interface between an improperly contoured bracket base and corresponding tooth surface.
adhesive, thereby achieving superior bond strength (Fig. 4, A). The variable surface contour of the incisors combined with the inability to properly adapt bracket bases resulted in varying thicknesses of resin between the bonding pad and tooth surface, and consequently varying bond strengths (Fig. 4, B). Unpolymerized resin, detected by scratching with an explorer, was occasionally present in the nonadapted sample in which the bracket bonding pad curvature
Volume 95 i’hnber 1
thinning lingual enamel possibly is more prone to fracture. Longitudinal sections of types B and C fractures showed that crescent fracture lines and enamel fragment fractures routinely extended to and involved the dentinoenamel junction. Enamel fracture is likely to occur at the junction because it is an area of transformation between two dental tissues with different prope~ies.3,‘6‘18
In vivo the small crystals of dentin merge imperceptibly with the larger crystals of ename1,2 therefore ‘. structurally reinforcing the enamel cap. However, in vitro dentin desiccation may occur.21 This may reduce enamel cap support significantly, leading to enamel fracture at the dentinoenamel junction. It is therefore proposed that crescent fracture lines and enamel fragment fractures occur with increased frequency on lingual surfaces as compared with labial surfaces because (1) the enamel rod orientation may render the lingual enamel more prone to fracture and (2) lingual enamel is thinner. These two factors in an in vitro environment in which enamel may have lost structural support from the underlying desiccated dentin yielded the results documented. It is evident from the preceding discussion that undesirable enamel damage may occur during debonding in vitro, especially on lingual surfaces. Clinically the occurrence of crescent fracture lines and enamel fractures has never been reported on lingual surfaces. However, as stated in the results section, they are difficult to see and if present may easily be overlooked. It also is important to note that in clinical situations accidental debonding of orthodontic brackets by occlusal forces is more analogous to the experimental debonding design presented in this article than to conventional debonding procedures. In accidental debonding and in this study, the force to dislodge the brackets is directed gingivally. Conventional debonding does not necessarily produce such a force direction. Hence teeth that have lost brackets as a result of occlusal forces possibly should undergo careful scrutinization for enamel damage. Furthermore, the possibility of dentin desiccation as a contributing factor in types @ and C fractures has been proposed. If tiue, great attention must be focused on devitalized teeth in vivo because their susceptibility to lingual surface enamel fractures may be quite high. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS
1. In vitro lingual bond strengths are comparable to labial bond strengths, which would suggest that the labial bonding technique need not be altered for lingual
In vitro investigation
of lingual bonding
27
bonding. Therefore it is not necessary to roughen lingual surfaces or to increase etching time for lingual bonding. 2. Lingual bracket base adaptation produces a uniform thin layer of resin and a strong bond, and also minimizes unpolymerized resin when a one-paste resin system is used. This is accomplished more easily for premolars than for mandibular incisors. 3. Adaptation of lingual bracket bases is recommended for the following reasons: (a) ease of bracket placement, (b) minimizing of unpolymerized resin beneath bracket bases, and (c) an increase in bond strength. 4.’ There is an increase in vertical enamel fracture lines both labially and lingually when these surfaces are debonded with a vertical shear force in vitro. 5. Lingual surfaces in vitro are more susceptible than labial surfaces to horizontal crescent fracture line formation and enamel fractures with loss of a fragment when brackets are removed with a vertical shear force. Adaptation of lingual bracket bases does not appear to increase the incidence of such fractures. 6. Care must be exercised when debonding to minimize enamel stress. REFERENCES 1. Ash MM. Wheeler’s Atlas of tooth form. 5th ed. Philadelphia: WB Saunders, 1984. 2. Bhaskar SN. Oral histology and embryology. 9th ed. St. Louis: The CV Mosby Company, 1980:69-71. MS. Tensile strength and modulus of 3. Bowen RL, Rodriguez elasticity of tooth structure and several restorative materials. J Am Dent Assoc 1962;64:378-87. 4. Bozalis WG, Marshall GW, Cooley RO. Mechanical pretreatment and etching of primary-tooth enamel. ASDC J Dent Child 1979;44:43-9. 5. Buonocore MG. The use of adhesives in dentistry. 1 st ed. Springfield, Illinois: Charles C Thomas, 1975:25-30. 6. Diedrich P. Enamel alterations from bracket bonding and debonding: a study with the scanning electron microscope. AM J ORTHOD 1981;79:500-22. AW. I. Eick JD, Johnson LN, Fromer JR, Good RJ, Neumann Surface topography: its influence on wetting and adhesion in a dental adhesive system. J Dent Res 1972:.51:780-8. 8. Evans LB, Powers JM. Factors affecting in vitro bond strength of no-mix orthodontic cements. AM J ORTHOD 1985;87:508-12. 9. Galil KA, Wright GZ. Acid etch patterns on buccal surfaces of permanent teeth. Pediatr Dent 1979;1:230-4. 10. Gillings B, Buonocore M. An investigation of enamel thickness in human lower incisor teeth. J Dent Res 1961;40:105-18. (personal communication). Commack, New 11. GAC International York: 1985. 12. Kelly VM. JCO interviews. J Clin Orthod 1982;16:461-76. 13. Knoll M, Gwinett AJ, Wolff M. Should primary enamel be ground prior to bonding. J Clin Orthod 1985;19: 137-8. 14. Munechika T, Suzuki K, Nishiyama M, Ohashi M, Horie K. A comparison of the tensile bond strengths of composite resins to
28
15. 16. 17. 18. 19. 20.
Chunmk
Am. J. Orthod.
et al.
longitudinal and transverse sections of enamel prisms in human teeth. J Dent Res 1984;63:1079-82. Ormco, Division of Sybron Corporation. Lingual orthodontics, bonding instructions. Glendora, California: 1985. Rasmussen ST, Patchin RE, Scott DB, Heuer AM. Fracture properties of human enamel and dentine. J Dent Res 1976; 5.5:154-64. Rasmussen ST. Fracture properties of human teeth in proximity to the dentinoenamel junction. J Dent Res 1984;63: 1279-83. Stanford JW, Weigel KV, Paffenbarger GC, Sweeney WT. Compressive properties of hard tooth tissues and some restorative materials. J Am Dent Assoc 1960;60:746-56. Unitek Corporation. Lingual orthodontics, bonding instructions. Monrovia, California: 1985. Wakuri H. The experimental studies of enamel cap. Bull Tokyo Dent Co11 1971;12:117-47.
Dentofac. .huary
Orthop. 1989
21 Whittaker DK. The enamel-dentine junction of human and Macaca irus teeth: a light and electron microscope study. J Anat 1978;125:323-35. 22. Zachrisson BU, Skogan 0, Hoymyhr S. Enamel cracks in debonded, debanded, and orthodontically untreated teeth. AM J ORTHOD1980;77:307-19. Reprint
requests
to:
Dr. Leonard Chumak Civilian Dental Clinic Canadian Forces Europe C.F.P.O. 5000 Bell~ille, Ontario, Canada KOK 3R0