The acid-etch technique has been widely used in

ORIGINAL ARTICLE Comparison of bond strength between orthodontic brackets bonded with halogen and plasma arc curing lights: An in-vitro and in-vivo study Michael D. Signorelli, a Elizabeth Kao, b Peter W. Ngan, c and Marcia A. Gladwin d Bitburg, Germany, and Morgantown, WVa Introduction: This study assessed in-vitro shear bond strength and in-vivo survival rate of orthodontic brackets bonded with either a halogen or a plasma arc light. Methods: Ninety extracted premolars were divided into 6 groups of 15. Stainless steel brackets were bonded to the teeth by using either a halogen light with a 20-second curing time or a plasma arc light with a 2-, 6-, or 10-second curing time. Brackets were debonded either within 30 minutes of bonding or after thermocycling for 24 hours. Bond strengths were tested on a testing machine at a crosshead speed of 1 mm/minute. The bracket failure interface was measured with a modified adhesive remnant index score. Data were analyzed by using ANOVA and Tukey-Kramer multiple comparison tests. For the in-vivo study, a split-arch design was used to determine the bracket-failure rate and distribution in 25 patients. The patients were followed for a mean period of 1.1 years (386 days). Survival analysis was carried out to compare the failure rates of the 2 techniques. Results: No significant differences in bond strengths were found 30 minutes after bonding between the halogen light (13.6 3.8 MPa) and the plasma arc light with 2-, 6-, or 10-second curing times (9.6 2.9, 14.2 4.6, 16.0 3.0 MPa, respectively). Similar bond strengths were also found between the halogen light with a 20-second (16.1 3.6 MPa) curing time and plasma arc light with 6 seconds (18.2 4.6 MPa) of curing time after 24 hours of thermocycling. For the in-vivo study, no significant difference was found in bracket failure rates between the 2 light sources (4.9% in both groups). No significant differences were found between ARI scores for the halogen light and the plasma arc light at either 30 minutes or 24 hours after debonding. Conclusions: These results indicate that the plasma arc light with a 6-second curing time can produce similar bond strength and bracket-failure rates as the halogen light that requires a longer curing time. (Am J Orthod Dentofacial Orthop 2006;129:277-82) The acid-etch technique has been widely used in orthodontics to directly bond attachments to enamel surfaces. For years, self-cured resin was the only type of adhesive available. This resin begins to polymerize upon mixing, and working time is limited. It has been shown that the bond strength of the resin is weakened when air is incorporated during mixing or when the pastes are incompletely mixed. 1 Buonocore 2 introduced the first photosensitive a US Air Force, Bitburg, Germany. b Professor, Department of Restorative Dentistry, School of Dentistry, West Virginia University, Morgantown. c Professor and chair, Department of Orthodontics, School of Dentistry, West Virginia University, Morgantown. d Professor, Division of Dental Hygiene, School of Dentistry, West Virginia University, Morgantown. Reprint requests to: Dr Peter Ngan, West Virginia University, School of Dentistry, 1076 Health Science Center North, PO Box 9480, Morgantown, WV 26506; e-mail, pngan@hsc.wvu.edu. Submitted, March 2004; revised and accepted, July 2004. 0889-5406/$32.00 Copyright 2006 by the American Association of Orthodontists. doi:10.1016/j.ajodo.2004.07.043 light-cured resin in 1970. The greatest advantage of a light-cured system is longer working time before polymerization begins. Visible light curing (VLC) was introduced about 1980. 3 Polymerization of VLC resins is based on the presence of a photo initiator, camphorquinone, which is sensitive to light in the 470-nm wavelength spectrum. 4,5 Most VLC units have a broad wavelength width between 400 and 520 nm, with a light intensity commonly about 400 mw/cm 2. The curing time for bonded metal brackets with conventional VLC units is variable. For example, the manufacturer of Transbond XT (3M Unitek, Monrovia, Calif) recommends a 20-second cure, but Oesterle et al 6 and Wang and Meng 7 independently found stronger bond strengths with a 40-second cure. Various methods have been introduced to enhance the polymerization of bonding agents. In the late 1980s, argon lasers were introduced with curing times of only 10 seconds for filled resins and 5 seconds for unfilled resins. The argon laser operates in a wavelength range 277

278 Signorelli et al American Journal of Orthodontics and Dentofacial Orthopedics February 2006 of 454 to 496 nm of the visible light spectrum, with an intensity that approaches 800 mw/cm 2. 8 In-vitro studies have shown that the bond strengths were comparable to those of VLC units. 9-11 More recently, the plasma arc light has been used for rapid curing of restorative materials. 12 This system has a filter that narrows the spectrum of visible light to a band centered on the 470 nm wavelength for activation of camphorquinone while producing a high light intensity of 1200 mw/cm 2. This enables the curing time to be significantly shortened. 12 A high-energy, high-pressure ionized gas in the presence of an electric current is used to create a light source strong enough to increase the polymerization rate of composite resins and resin-modified glass ionomers. 13 The use of such light-curing units has been recently reported for orthodontic purposes with a recommended cure time of 2 seconds per bracket in 1 study 12 and 3 seconds per bracket in another. 14 The objectives of this study were to compare (1) the in-vitro shear bond strength and the in-vivo survival rate and distribution of brackets bonded with plasma arc light or halogen light, and (2) the in-vitro and in-vivo bracket-failure interface for the 2 curing-light systems. MATERIAL AND METHODS In-vitro bond strength study Two different light units for curing orthodontic bracket adhesive were compared: a plasma arc lightcuring system (OrthoLite, 3M Unitek, Monrovia, Calif) and a conventional halogen light-curing system (Ortholux XT, 3M Unitek). The OrthoLite produces a light intensity of approximately 2000 mw/cm 2, and the Ortholux XT produces approximately 400 mw/cm 2. 13 Both curing units produce wavelengths between 400 and 500 nm. The plasma arc light has an 8-mm diameter light guide, and the halogen light has a 7-mm light guide. Ninety freshly extracted human premolars were collected. The criteria for selection included noncarious and nonrestored buccal surfaces with no visible cracks from the extraction forceps. The teeth were not exposed to pretreatment chemical agents such as hydrogen peroxide or bleach. They were cleansed of tissue and debris, steam-autoclaved for 45 minutes in liquid cycle at 250 F, and stored in 0.1% thymol. Steam sterilization at a low temperature reduces bacteria contamination; this process has been shown not to affect the enamel surface or the bond strength. 15 The teeth were then embedded in epoxy resin (Buehler, Lake Bluff, Ill), and a dental surveyor was used to align the facial surface of each tooth so that it was perpendicular with the bottom of the stainless steel mounting ring. This oriented the bonding surface to be parallel to the force applied during the shear strength test. The mounted teeth were kept moist in a humidor while the epoxy resin polymerized and then stored in water until bonding. Before bonding, the teeth were randomly divided into 6 groups, each containing 15 teeth. Group I: brackets were cured with a halogen light for 20 seconds (10 seconds mesial and 10 seconds distal) and debonded within 30 minutes. Group II: bracketes were cured with a halogen light for 20 seconds, thermocycled for 24 hours (approximately 580 cycles with a 1-minute dwell time each in 5 C and 55 C), and debonded. Group III: brackets were cured with a plasma arc light for 2 seconds (1 second mesial and 1 second distal) and debonded within 30 minutes. Group IV: brackets were cured with a plasma arc light for 6 seconds (3 seconds mesial and 3 seconds distal) and debonded within 30 minutes. Group V: brackets were cured with a plasma arc light for 6 seconds, thermocycled for 24 hours, and debonded. Group VI: brackets were cured with a plasma arc light for 10 seconds and debonded within 30 minutes. The bonding surface of each tooth was pumiced for 10 seconds and rinsed for 10 seconds with distilled water. The enamel surface was conditioned with 37% phosphoric acid gel for 30 seconds and rinsed with distilled water. The surface was thoroughly dried, and a thin layer of Transbond XT sealant (3M Unitek) was applied. APC precoated maxillary premolar brackets (3M Unitek) with a bracket base area of 11.35 mm 2 were used. Each bracket was placed on the tooth, and an explorer was used to seat the brackets with a constant force by 1 operator (M.D.S.). Excess adhesive was removed, and the bracket adhesive was light-cured with the designated curing unit and the curing time indicated by group assignment. Each curing unit was tested and calibrated according to the manufacturer s instructions to ensure that maximum intensity output was obtained. 16 The plasma light was calibrated by inserting the curing tip fully into the calibration port and then depressing the hand switch. The halogen light was calibrated by placing the fiber-optic probe directly on top of the built-in sensor until the light indicated that the probe intensity was adequate. For groups, I, III, IV, and VI, testing was completed within 30 minutes of bonding. For groups II and V, testing was completed 24 hours after the thermocycling. between 5 C 2 C in a refrigerated circulating bath (Lindberg/Blue, Asheville, NC) and 55 C 2 C in a heated water bath for approximately 580 cycles. Debonding forces in Newtons were determined by using a testing machine (Instron, Canton, Mass) with a

American Journal of Orthodontics and Dentofacial Orthopedics Volume 129, Number 2 Signorelli et al 279 crosshead speed of 1 mm/minute. The stainless steel rings were mounted on a base jig that could be adjusted on X and Y axes to ensure that the applied force was parallel to the long axis of the tooth. The alignment was further checked against a vertically mounted ruler. The teeth were immersed in water throughout the debonding procedure with the force applied at the bracket-tooth interface. The force was applied by placing the looped end of an 0.018-in stainless steel wire, attached superiorly to the load cell of the testing machine, under the gingival bracket wings. The force was then applied at a crosshead speed of 1 mm/minute. The bond strengths in MPa were calculated based on the bracket base area. After debonding, adhesive left on the bracket base was examined with an optical microscope by 1 investigator (M.D.S.) at 10 times magnification to determine the bracket-failure interface. A modified adhesive remnant index (ARI) was used to evaluate the adhesive remaining on a bracket after debonding (Table 1). 17,18 To determine intrarater reliability, 10 specimens were included in an error study to measure the ARI 2 weeks apart. The coefficient of reliability was found to be 0.92. In-vivo bracket-survival study The in-vivo study included 25 patients (11 male, 14 female) from the School of Dentistry Orthodontic Clinic at West Virginia University. Criteria for patient selection included intact permanent dentition, no decalcification on teeth, and treatment requiring comprehensive orthodontic treatment with fixed appliances. No preference was placed on type of malocclusion or whether extractions were indicated. Precoated APC brackets were bonded to 445 teeth, and 222 were cured with the plasma arc light and 223 with the halogen light. The adhesive applied to the precoated brackets contained the same ingredients as found in the Transbond XT adhesive. A split-arch technique was used to equate any variability between the right and left quadrants. Force of mastication and hygiene effectiveness can vary from the left side to the right side, and these variables could affect bracket failure. The patients were alternately assigned to 1 of 2 groups. In group I, the brackets for the maxillary left and mandibular right quadrants were bonded with the plasma arc light curing unit. The contralateral sides were bonded with a conventional halogen light curing unit. In group II, the pattern was reversed. The same bonding procedures were used as in the in-vitro study. For plasma arc curing, a 6-second exposure time (3 seconds mesial and 3 seconds distal) was used. For halogen curing, a 20-second exposure time (10 seconds mesial 10 seconds distal) was used. The date and quadrant of bracket Table I. Modified ARI scores Score Definition 0 No adhesive left on bracket 1 Less than 25% of adhesive left on bracket 2 25% of adhesive left on bracket 3 50% of adhesive left on bracket 4 75% of adhesive left on bracket 5 100% of adhesive left on bracket failures were recorded. The failed brackets were collected to determine the bracket-failure interface. Data analysis For the in-vitro study, significant differences in shear bond strength (MPa) and ARI scores between test groups were determined by using ANOVA and Tukey- Kramer multiple comparison tests. Level of significance was preset at P.05. The bracket-survival rate was computed with the Kaplan-Meier product-limit survival estimates. RESULTS In vitro bond strength study The shear bond strengths for brackets bonded with the 2 types of curing lights and varied curing times are shown in Table II. Significant differences in the shear bond strength were found among the 6 test groups (P.0001). The Tukey-Kramer test showed that, among the four 30-minute test groups, no significant differences were found between brackets bonded with the plasma arc light with curing times of 2, 6, or 10 seconds (9.6 2.9, 14.2 4.6, 16.0 3.0 MPa, respectively) and the halogen light control group with a curing time of 20 seconds (13.6 3.8 MPa). For the samples with 24 hours of thermocycling, no significant differences were found between the brackets bonded with the plasma arc light for 6 seconds (18.2 4.6 MPa) and the halogen light control group with a curing time of 20 seconds (16.1 3.6 MPa). Among the plasma arc groups tested 30 minutes after bonding, those with curing times of 6 and 10 seconds (14.2 4.6 MPa, 16.0 3.0 MPa, respectively) were found to have significantly higher mean bond strengths than the 2-second curing group (9.6 2.9 MPa). The plasma light with 6 seconds curing time had the highest mean shear bond strength of 18.2 4.6 MPa when debonded after 24 hours. This was significantly higher than groups I and III. In-vitro bracket-failure interface The ARI scores of adhesive remaining on the bracket after debonding for the 6 groups are shown in Table III.

280 Signorelli et al American Journal of Orthodontics and Dentofacial Orthopedics February 2006 Table II. In-vitro mean shear/peel bond strength for test groups Group Light source Total curing time Debond time Mean MPa SD Minimum MPa Maximum MPa I Ortholux XT 20 s 30 min 13.6 a,b 3.8 8.7 21.5 II Ortholux XT 20 s TC 24 h 16.1 b,c 3.6 11.9 23.2 III OrthoLite 2 s 30 min 9.6 a 2.9 5.0 13.5 IV OrthoLite 6 s 30 min 14.2 b,c 4.6 7.8 23.1 V OrthoLite 6 s TC 24 h 18.2 c 4.6 9.2 25.9 VI OrthoLite 10 s 30 min 16.0 b,c 3.0 12.7 23.1 Superscript of same letter indicates groups that are not significantly different from each other. TC, Thermocycled. Table III. In vitro mean ARI score for test groups Group Light source Total curing time Debond time Mean ARI SD Minimum ARI Maximum ARI I Ortholux XT 20 s 30 min 3.5 a 1.2 0 5 II Ortholux XT 20 s TC 24 h 3.9 b 1.3 0 5 III OrthoLite 2 s 30 min 2.5 a 1.7 0 5 IV OrthoLite 6 s 30 min 3.2 a 1.3 1 5 V OrthoLite 6 s TC 24 h 4.6 b 0.6 1 5 VI OrthoLite 10 s 30 min 3.3 a 1.5 0 5 Superscript of same letter indicates groups that are not significantly different from each other. TC, Thermocycled. ANOVA showed significant differences among the 6 test groups (P.0016). No significant differences were found between the halogen light control group (3.5 1.2 MPa) and the plasma arc light groups tested 30 minutes after bonding (2.5 1.7 MPa). No significant difference was found between the halogen light control group (3.9 1.3 MPa) and the plasma arc light group tested 24 hours after bonding (4.6 0.6 MPa). The Tukey-Kramer analysis showed significantly higher ARI scores for group V compared with group III (P.05) and for group II compared with group III (P.05). In-vivo bracket-survival distribution The mean observation time at final data collection was 386 days, or 1.1 years. Of 445 bonded teeth, 22 brackets failed, for a failure rate of 4.9%. Of the 22 failed brackets, 11 were cured with each light. The chi-square analysis showed no significant difference between the light source and the number of bracket failures. The bracket-survival distribution for the type of curing unit was determined by the Kaplan-Meier product-limit survival estimates analysis. It showed that the mean durations of bracket failure were 189 days for the plasma arc light group and 187 days for the halogen light group. The ranges for the 11 failed brackets in each group were 44 to 432 days for the plasma arc light and 51 to 365 days for the halogen light group. No pattern of bracket failure frequency with time was noted among the 2 groups. A t test indicated no significant differences between the 2 groups (P.005). Eight of the 11 failed brackets were collected from the halogen light group, and 4 of the 11 were collected from the plasma arc light group. The halogen arc light group had a mean ARI score of 3.2, and the plasma arc light group had a mean score of 3.0. The sample size of the collected failed brackets was too small to justify statistical analysis. DISCUSSION When evaluating the mean shear bond strengths of the 4 groups tested 30 minutes after bonding, no statistically significant differences were found between the control group cured with the halogen light for 20 seconds and brackets bonded with the plasma arc light with curing times of 2, 6, or 10 seconds. This agrees with the findings of Oesterle et al, 19 who found no significant difference in bond strength among brackets bonded with a plasma arc light with curing times of 3, 6, or 9 seconds and brackets bonded with a halogen light with a 40-second curing time and all groups tested at 30 minutes. Cacciafesta et al 20 reported similar findings with a light-cured glass ionomer 15 minutes after bonding. When evaluating the mean shear bond strengths

American Journal of Orthodontics and Dentofacial Orthopedics Volume 129, Number 2 Signorelli et al 281 after 24 hours of thermocycling, no significant difference was found between brackets bonded with the halogen light for 20 seconds and brackets bonded with the plasma arc light for six seconds. This also agrees with findings by Oesterle et al, 19 Sfondrini et al, 13 and James et al. 11 These results suggest that the plasma arc light can be used to bond orthodontic brackets to the enamel surface with a shorter curing time without a significant decrease in shear bond strength. The faster curing time saves chair time and might decrease the risk of bond failure from moisture contamination. 21 The degree of polymerization is directly related to the amount of total energy that the resin absorbs, and the total light energy is the intensity of the light times the duration of the exposure. The 3M Unitek instruction book 16 lists the Ortholux XT as producing a light intensity of approximately 400 mw/cm 2 ; the OrthoLite produces 5 times as much, approximately 2000 mw/cm 2. Greater total light energy results in increased fracture toughness and greater flexural strength; these mean greater bond strength. 19 The manufacturer s current recommended curing time for the plasma arc light is 6 seconds: 3 seconds for the mesial surface and 3 seconds for the distal surface for metal brackets. A part of this investigation was to determine whether a shorter curing time of 2 seconds could produce the same bond strength. The results for the 30-minute test groups showed that the shear bond strengths produced by the plasma arc light with curing times of 6 or 10 seconds were significantly higher than with a curing time of 2 seconds. Although the 2-second curing time with the plasma arc light was significantly lower than the 6- and 10-second times, its mean shear bond strength of 9.6 MPa still exceeded the range of 6 to 8 MPa suggested by Reynolds 22 as clinically acceptable. Of the 15 teeth that were cured for 2 seconds with the plasma arc light and debonded within 30 minutes, 2 failed to reach 6 MPa. These findings suggest that more bond failures can occur with a 2-second curing time, because most orthodontists apply forces within 30 minutes of bonding. A longer exposure time will allow for more complete polymerization of the bracket adhesive and improve bond strength. Both 6 and 10 seconds of curing produced sufficient bond strengths for immediate loading of the bracket with orthodontic forces; clinicians would gain no additional advantage by curing longer. The results of this study indicate that bond strength continues to increase after the initial polymerization by the curing light. The ARI enables the clinician to determine the bracket-failure interface. A low score would be interpreted as a failure between the adhesive and bracket interface, and a high score would indicate a failure at the adhesive enamel interface. The highest mean score (4.6) was for the plasma arc light group with a 6-second curing time tested 24 hours after bonding. Because over 75% of the adhesive remained on the bracket base, this would make residual adhesive removal easier for the clinician. In the in-vivo investigation, the bracket survival rate distribution for the 2 curing systems was determined by using the Kaplan-Meier product limit survival estimates from the time of bracket placement to final data collection. This method enables the clinician to evaluate the performance of the bonded brackets over the entire test period instead of just at final data collection. Our results showed no significant differences in either failure rate or time of failure. The failure rate of 4.9% with both curing lights over the 1.1-year observation period is acceptable and concurs with the findings of other investigators. 14,23-25 These clinical findings further support the in-vitro findings that there is no significant decrease in bond strength between the plasma arc light and the halogen light groups. Sunna and Rock 23 found that 60% of the bond failures occurred within the first 6 months, or 180 days; this agrees with the mean duration of bracket failures with the halogen light (187 days) and the plasma arc light group (189 days). Because some patients did not reutrn failed brackets to clinicians, they were too few to justify a bracketfailure interface data analysis. Of the brackets collected and scored, the clinical impression was that there is no difference in the amount of adhesive remaining on the enamel surface after bracket failure or debonding. CONCLUSIONS Brackets bonded with the plasma arc light for 6 seconds were found to produce similar bond strength compared with the halogen light with a 20-second curing time. No significant differences were found in the site of bracket failure interface between the halogen and plasma arc curing lights. A longer than the recommended curing time (10 seconds) with the plasma arc light does not produce statistically significant higher mean bond strengths; the 2-second cure time showed significantly lower strength, although it was still within the clinically acceptable range. There is no statistically significant difference in the survival rate or time of bracket failure between the plasma arc light and the halogen light. Using the plasma arc light can save chair time without increasing the bracket-failure rate.

282 Signorelli et al American Journal of Orthodontics and Dentofacial Orthopedics February 2006 REFERENCES 1. Pollack BF, Blitzer MH. The advantages of visible light curing resins. NY State Dent J 1982;48:228-30. 2. Buonocore M. Adhesive sealing of pits and fissures for caries prevention with use of ultraviolet light. J Am Dent Assoc 1970;80:324-8. 3. Strassler H. Light-curing renaissance makes materials popular. Technological upgrades boost quality in light-curing devices. Dentist 1980;68:23-7. 4. Zachrisson BJ. A posttreatment evaluation of direct bonding in orthodontics. Am J Orthod 1977;71:173-89. 5. Fan PL, Wozniak WT, Reyes WD, Stanford JW. Irradiance of visible light-curing units and voltage variation effects. J Am Dent Assoc 1987;115:442-5. 6. Oesterle LJ, Messersmith ML, Devine SM, Ness CF. Light and setting times of visible light-cured orthodontic adhesives. J Clin Orthod 1995;29:31-6. 7. Wang WN, Meng CL. A study of bond strength between lightand self-cured orthodontic resin. Am J Orthod Dentofacial Orthop 1992;101:350-4. 8. Cipolla AJ. Laser curing of photoactivated restorative materials. ILT Systems;1993. p. 1-3. 9. Kupiec KA, Swenson RR, Blankenau RJ, Bhatia SJ. Laser vs VLC systems for bonding orthodontic brackets [abstract 3205]. J Dent Res 1997;76:414. 10. Sedivy M, Ferguson D, Dhuru V, Kittleson R. Orthodontic resin adhesive cured with argon laser: tensile bond strength [abstract 582]. J Dent Res 1993;72:176. 11. James J, Miller B, English J, Tadlock L, Buschang P. Effects of high-speed curing devices on shear bond strength and microleakage of orthodontic brackets. Am J Orthod Dentofacial Orthop 2003,125:555-61. 12. Cacciafesta V, Sfondrini MF, Sfondrini G. A xenon arc light-curing unit for bonding and bleaching. J Clin Orthod 2000;34:94-6. 13. Sfondrini MF, Cacciafesta V, Pistorio A, Sfondrini G. Effects of conventional and high-intensity light-curing on enamel shear bond strength of composite resin and resin-modified glassionomer. Am J Orthod Dentofacial Orthop 2001;119:30-5. 14. Manzo B, Liistro G, De Clerck H. Clinical trial comparing plasma arc and conventional halogen curing lights for orthodontic bonding. Am J Orthod Dentofacial Orthop 2004;125:30-5. 15. Pashley EL, Tao L, Pashley DH. Sterilization of human teeth, its effects on permaeability and bond strength. Am J Dent 1993;6: 189-91. 16. 3M Unitek. Instruction book for OrthoLite curing unit; 2001. 17. Shammaa I, Ngan P, Kim H, Kao E, Gladwin M, Gunel E, et al. Comparison of bracket debonding force between two conventional resin adhesives and a resin-reinforced glass ionomer cement: an in vitro and in vivo study. Angle Orthod 1999;69: 463-9. 18. Årtun J, Bergland S. Clinical trials with crystal growth conditioning as an alternative to acid-etch enamel pretreatment. Am J Orthod 1984;85:333-40. 19. Oesterle LJ, Newman SM, Shelhart WC. Rapid curing of bonding composite with a xenon plasma arc light. Am J Orthod Dentofacial Orthop 2001;119:610-6. 20. Cacciafesta V, Sfondrini MF, Klersy C, Sfondrini G. Polymerization with a micro-xenon light of a resin-modified glass ionomer: a shear bond study 15 minutes after bonding. Eur J Orthod 2002;24:689-97. 21. Klocke A, Korbmacher HM, Huck LG, Kahl-Nieke B. Plasma arc curing lights for orthodontic bonding. Am J Orthod Dentofacial Orthop 2002;122:643-8. 22. Reynolds JR. A review of direct orthodontic bonding. Br J Orthod 1975;2:171-8. 23. Sunna S, Rock WP. Clinical performace of orthodontic brackets and adhesives: a randomized study. Br J Orthod 1998;25:283-7. 24. Millett DT, Hallgren A, Cattanach D, McFadzean R, Pattison J, Robertson M, et al. A 5-year clinical review of bond failure with a light-cured resin adhesive. Angle Orthod 1998;68:351-6. 25. Sfondrini MF, Cacciafesta V, Scribante A, Klersy C. Plasma arc versus halogen light curing of orthodontic brackets: a 12-month clinical study of bond failures. Am J Orthod Dentofacial Orthop 2004;125:342-7. RECEIVE THE JOURNAL S TABLE OF CONTENTS EACH MONTH BY E-MAIL To receive the tables of contents by e-mail, send an e-mail message to majordomo@mosby.com Leave the subject line blank and type the following as the body of your message: Subscribe ajodo_toc You may sign up through our website at http://www.mosby.com/ajodo. You will receive an e-mail message confirming that you have been added to the mailing list. Note that TOC e-mails will be sent when a new issue is posted to the website.