ORIGINAL ARTICLE

Comparison of bond strength between aconventional resin adhesive and a resin-modified glass ionomer adhesive: An in vitroand in vivo studyAndrew Summers, DDS, MS,a Elizabeth Kao, DMD, MS,b Jeffrey Gilmore, DDS, MS,c Erdogan Gunel, PhD,d

and Peter Ngan, DMDe

Sioux Falls, Utah, Morgantown, WVa, and Marietta, Ohio

The objectives of this study were (1) to compare the in vivo survival rates of orthodontic brackets bonded with aresin-modified glass ionomer adhesive (Fuji Ortho LC; GC America, Alsip, Ill) after conditioning with 10%polyacrylic acid and a conventional resin adhesive (Light Bond; Reliance Orthodontic Products, Itasca, Ill) bondedwith 37% phosphoric acid, (2) to compare the in vitro bond shear/peel bond strength between the 2 adhesives,(3) to determine the mode of bracket failure in the in vivo and in vitro tests according to the adhesive remnant index(ARI), and (4) to compare the changes in surface morphology of enamel surface after etching or conditioning with10% polyacrylic acid, with scanning electron microscopy. In the in vitro study, 50 extracted premolars wererandomly divided into 4 groups: brackets bonded with Fuji Ortho LC or Light Bond adhesive that were debondedafter either 30 minutes or 24 hours. Bond strengths were determined with a testing machine at a crosshead speedof 1 mm/min. Data were analyzed with analysis of variance and a paired Student t test. The in vivo study consistedof 398 teeth that were randomly bonded with Fuji Ortho LC or Light Bond adhesive in 22 subjects with thesplit-mouth technique. Bracket survival rates and distribution were followed for 1.3 years. Data were analyzed withKaplan-Meier product-limit estimates of survivorship function. The in vitro study results showed significantdifferences (P � .05) among the adhesives and the debond times. Light Bond had significantly greater bondstrengths than Fuji Ortho LC at 24 hours (18.46 � 2.95 MPa vs 9.56 � 1.85 MPa) and 30 minutes (16.19 � 2.04MPa vs 6.93 � 1.93 MPa). Mean ARI scores showed that Fuji Ortho LC had significantly greater incidences ofenamel/adhesive failure than Light Bond adhesive (4.9 vs 4.1). For the in vivo study, no significant differences infailure rate, sex, or location in dental arch or ARI ratings were found between the 2 adhesives. These resultssuggest that, compared with conventional resin, brackets bonded with resin-modified glass ionomer adhesivehad significantly less shear bond strength in vitro. However, similar survival rates of the 2 materials studied after1.3 years indicate that resin-reinforced glass ionomers can provide adequate bond strengths clinically. Theweaker chemical bonding between the adhesive and the enamel might make it easier for clinicians to clean upadhesives on the enamel surface after debonding. (Am J Orthod Dentofacial Orthop 2004;126:200-6)

Etching enamel with 37% phosphoric acid isused routinely by orthodontists to bond ortho-dontic brackets to enamel.1 The disadvantages

of this procedure are the loss of enamel during etching,2

aPrivate practice, Sioux Falls, Utah.bProfessor, Department of Restorative Dentistry, West Virginia UniversitySchool of Dentistry.cPrivate practice, Marietta, Ohio.dProfessor, Department of Statistics, West Virginia University School ofDentistry.eProfessor and chair, Department of Orthodontics, West Virginia UniversitySchool of Dentistry.Reprints requests to: Dr Peter Ngan, West Virginia University, School ofDentistry, Department of Orthodontics, Health Science Center North, P.O. Box9480, Morgantown, WV 26506; e-mail, pngan@hsc.wvu.edu.Submitted, April 2003; revised and accepted, June 2003.0889-5406/$30.00Copyright © 2004 by the American Association of Orthodontists.doi:10.1016/j.ajodo.2003.06.013

200

the necessity of strict adherence to a dry field, themultiple steps required, and the remaining resin residuethat cannot be easily removed after debonding of thebracket. A resin-modified glass ionomer (RMGI) adhe-sive (Fuji Ortho LC; GC America, Alsip, Ill) has beenintroduced that can be used for bonding bracketswithout acid etching.5-9 However, previous studieshave shown that nonetched RMGI adhesives havelower bond strength and a higher failure rate whencompared with conventional acid-etch resin bondingagents. Conditioning the tooth surface with 10% poly-acrylic acid does not cause as much damage to theenamel surface as etching with 37% phosphoric acidbut enhances the bond strength of RMGI adhesives, atleast under in vitro condition.13-15 The purposes of thisstudy were (1) to determine the in vivo survival rate of

American Journal of Orthodontics and Dentofacial OrthopedicsVolume 126, Number 2

Summers et al 201

orthodontic brackets bonded with an RMGI adhesiveafter conditioning with 10% polyacrylic acid with aconventional resin adhesive bonded with 37% phos-phoric acid used as the control group, (2) to determinethe in vitro shear/peel bond strength of the 2 adhesives,(3) to determine the mode of failure in the in vivo andin vitro tests, and (4) to compare the changes in surfacemorphology of enamel surface after etching or condi-tioning with 10% polyacrylic acid, with scanning elec-tron microscopy (SEM).

MATERIAL AND METHODSIn vitro bond strength study

Two bonding materials were tested: Fuji Ortho LC,an RMGI, and Light Bond (Reliance OrthodonticProducts, Itasca, Ill), a light-cured resin. Fifty ex-tracted, noncarious premolars were randomly dividedinto 4 groups, as shown in Table I. The teeth werecleaned, steam autoclaved, and stored in distilled waterand 0.9% thymol. The teeth were mounted in epoxideresin (Buehler, Lake Bluff, Ill), with a surveyor used toensure that the bonding surface was parallel to thedebonding force, pumiced, and stored in distilled water.

For groups I and II, the bonding procedure con-sisted of pumicing the tooth surface for 10 seconds withflour pumice, followed by a rinse of 10 seconds withwater. The bonding surface was conditioned with 10%polyacrylic acid for 20 seconds and rinsed for 10seconds. Each tooth was then wiped with a moist cottonroll to ensure that the bonding surface was not desic-cated, and excess water was removed. Fuji Ortho LCRMGI capsule was triturated for 10 seconds and thenapplied to a GAC micro-arch universal orthodonticpremolar bracket (GAC International, Bohemia, NY)with a base dimension of 3.12 � 3.40 mm, covering theentire base of the bracket without bubbles or voids. Thebracket was applied to the tooth with a constant force,and the surrounding flash was carefully removed. Theadhesive was light-cured with the Ortholux XT visiblelight-curing unit (3M Unitek, Monrovia, Calif) for atotal of 40 seconds, with 20-second curing intervalsfrom the mesial and distal aspects of the bracket.

For Groups III and IV, the bonding procedure

Table I. Four test groups with number of samples, testmaterial, and time of testing

Group n Bonding material Time of testing

I 13 RMGI (Fuji Ortho LC) 30 minII 12 RMGI (Fuji Ortho LC) 24 hIII 13 Resin (Light Bond) 30 minIV 12 Resin (Light Bond) 24 h

consisted of pumicing the tooth surface for 10 seconds,followed by a rinse for 10 seconds with water. Theenamel was etched for 30 seconds with 37% phosphoricacid and then rinsed for 10 seconds. The tooth wasdried with a stream of air until a chalky white appear-ance was observed. A thin layer of Light Bond sealantwas applied with a brush and light-cured for 10seconds. A bracket was applied to the tooth with aconstant force with the Light Bond adhesive. Flash wascarefully removed, and the adhesive was light-cured fora total of 40 seconds, with 20-second curing on themesial and distal aspects.

The teeth were mounted in a testing ring with thefacial enamel surface perpendicular to the base of themounting ring, with a dental surveyor. Epoxide resinwas used to secure the tooth in the mounting ring. Teethfrom all groups were stored in an incubator with 100%humidity immediately after bonding. Debonding force(in newtons) was determined within 30 minutes afterbonding in Groups I and III and 24 hours after bondingin Groups II and IV, with a testing machine (Instron,Canton, Mass) with a cross-head speed of 1 mm/min.

The bracket failure interface was examined underlight microscopy to determine whether the failureoccurred at the enamel-adhesive or the bracket-adhe-sive interface. The brackets were then assessed with amodified adhesive remnant index (ARI) and scored forthe amount of resin material adhering to the bracket.14

The criteria for the modified ARI scale are shown inTable II. Data were analyzed by analysis of variance(ANOVA) and paired Student t test.

The facial surfaces of 3 noncarious premolars wereobserved with SEM to compare the effects of 37%phosphoric acid and 10% polyacrylic acid on dentalenamel. The first tooth was the control, with no enameltreatment. The second tooth had the facial surfaceconditioned with 10% phosphoric acid for 20 seconds,followed with a rinse for 10 seconds. The third toothhad the facial surface etched with 37% phosphoric acidfor 30 seconds and then rinsed for 10 seconds. Theteeth were dehydrated in a series of alcohol concentra-tions ranging from 40% to 95% and prepared for

Table II. Modified ARI scale and correspondingdefinitions

Score Definition

5 All of adhesive remained on bracket4 More than 90% of adhesive remained on bracket3 More than 10% but less than 90% of adhesive remained

on bracket2 Less than 10% of adhesive remained on bracket1 No adhesive remained on bracket

American Journal of Orthodontics and Dentofacial OrthopedicsAugust 2004

202 Summers et al

observation with SEM by gold sputtering of the enamelsurfaces.

In vivo survival distribution study

The in vivo portion of the experiment included 22patients who received comprehensive orthodontic treat-ment in the Department of Orthodontics, West VirginiaUniversity School of Dentistry. The selection criteriaincluded no decalcification on teeth, good oral hygiene,and permanent dentition. A split-arch technique wasused, in which the maxillary right quadrant and themandibular left quadrant were bonded with either FujiOrtho LC RMGI or Light Bond resin adhesive. Thiswas determined randomly by a coin toss. The remain-ing 2 quadrants were bonded with the material that wasnot chosen for the other 2 quadrants. Each tooth waspumiced for 10 seconds and rinsed for 10 seconds. Thequadrants bonded with Fuji Ortho LC had the bondingsurfaces conditioned with 10% polyacrylic acid for 20seconds and rinsed for 10 seconds. The teeth werewiped with a moist cotton roll to ensure that thebonding surface was not desiccated, and excess waterwas removed. Fuji Ortho LC RMGI capsule wastriturated for 10 seconds and applied to the base of anorthodontic bracket covering the entire base of thebracket, without bubbles or voids. The bracket wasapplied to the tooth with a constant force, and any flashwas carefully removed. The adhesive was light-curedwith the Ortholux XT visible light-curing unit for atotal of 40 seconds, with 20-second curing intervalsfrom the mesial and distal aspects of the bracket.

The quadrants bonded with Light Bond resin adhe-sive had the bonding surfaces prepared with acidetching of the enamel for 30 seconds with 37% phos-phoric acid and then rinsed for 10 seconds. Each toothwas dried with a stream of air until a chalky whiteappearance was observed. A thin layer of Light Bondsealant was applied with a brush and light-cured for 10seconds. Light Bond adhesive was placed on thebracket and applied to the tooth with a constant force.Flash was carefully removed, and the adhesive waslight-cured with the Ortholux XT visible light-curingunit for a total of 40 seconds, with a 20-second cureinterval from the mesial and distal aspects.

Bracket failures were noted during the study. Anyfailed bracket was saved, and the tooth was no longerfollowed in the study. The bracket failure interface wasobserved with light microscopy to determine the failureinterface. The failed brackets were then assessed withthe modified ARI and scored with respect to the amountof resin material adhering to the bracket.16

Data analysis

Data in the in vitro study were analyzed byANOVA and paired Student t test. For the in vivo data,significant differences in the bracket survival rateamong the 2 materials, patient sex, location in the oralcavity, and ARI scores were determined with theKaplan-Meier product limit survival estimates and thelog-rank test at P � .05.

RESULTSIn vitro bond strength study

The shear force recorded in newtons on the testingmachine was converted to megapascals by dividing theforce by the area of the bracket base (3.12 mm � 3.40mm � 10.608 mm2). The shear bond strengths of alltest groups are shown in Table III. The control group(Light Bond debonded at 24 hours) was found to havethe highest mean shear bond strength (18.46 � 2.95MPa). This was followed by the Light Bond groupdebonded at 30 minutes (16.19 � 2.04 MPa). The FujiOrtho LC group debonded at 24 hours had a mean shearbond strength of 9.56 � 1.85 MPa, and the Fuji OrthoLC group debonded at 30 minutes had the lowest meanshear bond strength (6.93 � 1.93 MPa).

Analysis of variance showed significant differencesin the shear bond strengths among the 4 groups (P �.05). Paired Student t tests showed a significant differ-ence between the control group (Light Bond at 24hours) and the 3 experimental groups: Fuji Ortho LC,24 hours (P � .0001), Fuji Ortho LC, 30 minutes (P �.0001), and Light Bond, 30 minutes (P � .03). Signif-icant differences were found between the Light Bond,30 minutes, and the Fuji Ortho LC groups (P � .0001).Significant differences were also found between theFuji Ortho LC, 24 hours, and Fuji Ortho LC, 30minutes, groups (P � .001).

Bracket failure interface study

The average ARI scores for the 4 test groups areshown in Table IV. Fuji Ortho LC, 30 minutes, and FujiOrtho LC, 24 hours, had the highest mean ARI scores

Table III. In vitro shear bond strength for the 4 testgroups

GroupBondingmaterial

Debondtime n

Mean(MPa)

SD(MPa)

Minimum(MPa)

Maximum(MPa)

I Fuji Ortho LC 30 min 13 6.93 1.93 3.43 9.83II Fuji Ortho LC 24 h 12 9.56 1.85 7.10 12.42III Light Bond 30 min 13 16.19 2.04 12.65 19.14IV Light Bond 24 h 12 18.46 2.95 15.44 23.47

SD, standard deviation.

American Journal of Orthodontics and Dentofacial OrthopedicsVolume 126, Number 2

Summers et al 203

(4.92 � 0.28 and 4.67 � 0.65, respectively). LightBond, 30 minutes, and Light Bond, 24 hours, had ARIscores of 4.31 � 0.75 and 4.08 � 0.67, respectively.Significant differences in ARI scores were found be-tween Light Bond, 24 hours, and Fuji Ortho LC, 30minutes (P � .05), and also between Light Bond, 24hours, and Fuji Ortho LC, 24 hours (P � .04). Nosignificant differences were found between Light Bond,24 hours, and Light Bond, 30 minutes (P � .44).Significant differences were found between Fuji OrthoLC, 30 minutes, Fuji Ortho LC, 24 hours, and LightBond, 30 minutes (P � .01). No differences were foundbetween Fuji Ortho LC, 30 minutes, and Fuji Ortho LC,24 hours (P � .21).

The distribution of ARI scores for the test bracketsis shown in Table V. Most of the brackets bonded withFuji Ortho LC had an ARI score of 5 (Table V). Mostof the brackets bonded with Light Bond had an ARIscore of 4. Light Bond had 4 brackets with a score of 3,whereas Fuji Ortho LC had 1 bracket in this category.No sample tested received a score of less than 3.

SEM study of enamel morphology

Three extracted premolars were examined withSEM at �1000 magnification. The untreated enamelshowed a smooth surface (Fig 1, A). After applicationof 37% phosphoric acid for 30 seconds, the enamelshowed surface irregularities typical of a Type I enameletching pattern; etching of prism cores was predomi-nant17 (Fig 1, B). Figure 1, C, shows the enamel surfaceafter application of 10% polyacrylic acid for 20 sec-onds. The enamel surface exhibits minimal surfaceirregularities and smooth precipitate in some areas.Comparison of the enamel surfaces shows that enamelconditioned with 37% phosphoric acid produced aqualitatively rougher enamel surface than the enamelconditioned with 10% polyacrylic acid, indicating agreater loss of enamel from conditioning with 37%phosphoric acid.

Table IV. Average ARI scores for the 4 in vitro testgroups

GroupBondingmaterial

Debondtime n

MeanARI

score SD

MinimumARI

score

MaximumARI

score

I Fuji Ortho LC 30 min 13 4.92 0.28 4 5II Fuji Ortho LC 24 h 12 4.67 0.65 3 5III Light Bond 30 min 13 4.31 0.75 3 5IV Light Bond 24 h 12 4.08 0.67 3 5

In vivo bracket survival distribution

Twenty-two patients participated in this study (9male, 13 female). The mean observation time at finaldata collection was 481.4 days (1.32 years), with amaximum of 664 days (1.82 years) and a minimum of217 days (0.59 years). A total of 199 teeth were bondedwith Light Bond, and 199 teeth were bonded with FujiOrtho LC. During the observation period, 10 bracketsbonded with Light Bond adhesive failed, resulting in afailure rate of 5%. Thirteen brackets bonded with FujiOrtho LC failed, giving a slightly higher failure rate of6.5%. However, the Kaplan- Meier survival distribu-tion test showed no statistically significant correlationbetween the type of bonding material and bracketfailure rates (P � .41) (Fig 2). There was no statisti-cally significant correlation between sex and bracketfailure rates. The male subjects had a 5.2% bracketfailure rate, and the female subjects had a failure rate of6.2%.

There was no statistically significant correlationbetween the quadrant in which the teeth were bondedand the bracket failure rate (P � .99). The failure rateswere 6.0% for the maxillary right quadrant, 6.0% forthe maxillary left quadrant, 5.0% for the mandibularright quadrant, and 5.0% for the mandibular left quad-rant.

The ARI scores for all failed brackets (Table VI)showed that Light Bond had a slightly higher mean ARIvalue (3.7) than Fuji Ortho LC (3.5). There were nostatistically significant differences in the ARI scoresbetween the 2 bonding materials (P � .52). Mostbrackets that were bonded with Light Bond had an ARIscore of 5 (Table VII). There were more brackets withARI scores of 3 and 4 in the Fuji Ortho LC groups thanin the Light Bond groups.

DISCUSSIONIn vitro study

The shear/peel bond strengths of the 2 bondingmaterials were measured at 30 minutes and at 24 hours.This was designed to more fully simulate a clinicalsituation, because archwires are typically placed at thebonding appointment, when light-cured resins or ce-ments might not have been completely polymerized.2

The bond strengths of both the composite resin and theRMGI were significantly higher at 24 hours than at 30minutes. The reduced shear/peel bond strength imme-diately after bonding agrees with the findings ofBishara et al15 and is probably related to incompletepolymerization of light-cured materials.18,19

The bond strengths of the composite resin with 37%phosphoric acid at both 24 hours and 30 minutes were

American Journal of Orthodontics and Dentofacial OrthopedicsAugust 2004

204 Summers et al

significantly greater than those of the RMGI with 10%polyacrylic acid. These findings agree with those ofJobalia et al,6 Meehan et al,20 Bishara et al,15 andOwens et al.21 Bonding of glass ionomers is enhanced

Table V. Number of the in vitro brackets in each score

Group Material Debonded ARI � 1

I Fuji Ortho LC 30 min 0II Fuji Ortho LC 24 h 0III Light Bond 30 min 0IV Light Bond 24 h 0

Fig 1. A, SEM of untreated enamel surface. Bacid for 30 seconds. C, SEM of enamel conditall panels, original magnification �1000.

Fig 2. Bracket survival distribution over time with FujiOrtho LC and Light Bond.

Table VI. Descriptive statistics of all in vivo ARIscores

Adhesive n Mean Minimum Maximum

Fuji Ortho LC 13 3.5 1 5Light Bond 10 3.7 1 5

by surface conditioning with 10% polyacrylic acid.22

The acid removes contaminants and pellicles from thesubstrate surface and serves as a cleaning and wettingagent to improve the bonding of cement to enamel.Shammaa et al7 reported adequate bond strength whenglass ionomer (Fry Ortho LC) was bonded to enamelclinically without acid etching. However, the bondstrength was less compared with the values reported inthis study with 10% polyacrylic acid. Etching ofenamel with 37% phosphoric acid produces resin tagsto a depth of 80 �m that greatly increase the mechan-ical retention of composite resin to the enamel.23

Tavas and Watt24 recommended that adhesive bondstrength greater than 58 N is necessary for clinical use.According to this recommendation, Fuji Ortho LCadhesive, with a shear bond strength of 73.5 N (6.9MPa) at 30 minutes, has the potential to resist forcesduring orthodontic treatment. However, the averageforce transmitted to a bracket during mastication hasbeen reported to be between 40 and 120 N.25,26 The useof Fuji Ortho LC in areas of traumatic or heavyocclusion might be guarded.

ory of ARI

ARI � 2 ARI � 3 ARI � 4 ARI � 5

0 0 1 120 1 2 90 2 5 60 2 7 3

of enamel conditioned with 37% phosphoricwith 10% polyacrylic acid for 20 seconds. For

Table VII. Number of in vivo brackets in each ARIscore category

Group ARI � 1 ARI � 2 ARI � 3 ARI � 4 ARI � 5

Fuji Ortho LC 1 1 4 4 3Light Bond 0 1 2 0 7

categ

, SEMioned

American Journal of Orthodontics and Dentofacial OrthopedicsVolume 126, Number 2

Summers et al 205

The SEM results showed that the enamel surfaceconditioned with 37% phosphoric acid was qualita-tively rougher than when 10% polyacrylic acid wasused. The preferential etching of prism cones is typicalof a type I enamel etching pattern.17 Polyacrylic acidhas a larger molecular size than phosphoric acid. Thesmaller molecule of 37% phosphoric acid was able topenetrate to a greater depth; thus, potentially moreenamel can be lost during conditioning and debonding.SEM studies show that enamel surfaces after debondingof brackets are more porous with acid etching com-pared with clean and smooth enamel surface withoutetching.3 Repumicing of the bonded surface couldrestore the tooth surface to its original appearance.27

In vivo investigation

Bracket survival distribution was used to analyzethe in vivo data. This gives the clinician a betterperspective on when failure occurred compared withstudying the failure rates only at the final data collec-tion.

Our study showed no statistically significant differ-ences between Light Bond and Fuji Ortho LC asbonding materials. The failure rates of Light Bond andFuji Ortho LC were 5% and 6.5%, respectively. Inseveral studies, the failure rates for brackets bondedwith glass ionomer cements varied from 3.2% to50%.28-33 The lowest failure rates for Fuji Ortho LCwere reported by Fricker29 in 1998 (3.2%) and Silver-man et al30 in 1995 (3.3%). In addition, several clini-cians, after 3 years of using Fuji Ortho LC, experiencedbonding success rates comparable to those of conven-tional composite resins.31,32 The findings in the presentstudy confirm the observations that, overall, the clinicalperformance of Fuji Ortho LC was similar to that ofcomposite resin.

In the present study, the predominant mode ofbracket failure for the Fuji Ortho LC adhesive was atthe enamel-cement interface, in both in vitro and invivo conditions. The in vitro results suggest that chem-ical and mechanical bonding of glass ionomer cementto the bracket is stronger than the chemical bond ofglass ionomer to the enamel, even in the presence of theresin component. This agrees with the results reportedby McSherry,26 that glass ionomer cement bonds betterto the metal band of the bracket than to the enamel.

In contrast, when acid etching was used with LightBond materials, a higher percentage of failure occurredat the resin-bracket interface. This is probably becauseof the incomplete polymerization of the resin below thebracket base.19 Air entrapment behind the mesh of abracket can also affect polymerization, because ofoxygen inhibition of free radical polymerizing in light-

cured composite materials.35 Clinically, it is the au-thors’ experience that, during debonding, the resinadhesive remaining on the tooth is more difficult toremove than the RMGI adhesive.

CONCLUSIONS

1. In vitro results showed significantly greater shearbond strengths when brackets were bonded with37% phosphoric acid and composite resin (LightBond) compared with RMGI (Fuji Ortho LC)bonded with 10% polyacrylic acid.

2. Significantly greater shear bond strengths can beobtained 24 hours after bonding brackets for bothmaterials.

3. The in vivo results showed no significant differencein bracket failure rates between Fuji Ortho LC andLight Bond after 1.3 years. Clinically, Fuji Ortho LCadhesive has adequate bond strength to withstandthe occlusal forces of chewing and biting.

4. The ARI study showed that the predominant bracketfailure interface for Fuji Ortho LC was at theenamel-adhesive interface. The weaker chemicalbonding between RMGI and the enamel might makeit easier for clinicians to clean up the adhesive on theenamel surface after debonding.

5. The SEM study showed that etching with 37%phosphoric acid on dental enamel for 30 secondsproduced a qualitatively rougher and more poroussurface than conditioning with 10% polyacrylicacid.4,10,11,12,34

REFERENCES

1. Newman GV. Epoxy adhesives for orthodontic attachments:progress report. Am J Orthod 1965;51:901-12.

2. Thompson RE, Way DC. Enamel loss due to prophylaxis andmultiple bonding/debonding of orthodontic attachments. Am JOrthod 1981;79:282-95.

3. Osorio R, Toledano M, Garcia-Godoy F. Enamel surface mor-phology after bracket debonding. J Dent Child 1998;65:313-7.

4. Joseph VP, Rossouw E. The shear bond strengths of stainlesssteel and ceramic brackets used with chemically and light-activated composite resins. Am J Orthod Dentofacial Orthop1990;97:121-5.

5. Komori A, Ishikawa H. Evaluation of a resin-reinforced glassionomer cement for use as an orthodontic bonding agent. AngleOrthod 1997;67:189-96.

6. Jobalia SB, Valente RM, Waldemar GR, BeGole EA, Evans CA.Bond strength of visible light-cured glass ionomer orthodonticcement. Am J Orthod Dentofacial Orthop 1997;112:205-8.

7. Shammaa I, Ngan P, Kim H, Kao E, Gladwin M, Gunel E, et al.Comparison of bracket debonding force between two conven-tional resin adhesives and a resin-reinforced glass ionomercement: An in vitro and in vivo study. Angle Orthod 1999;69:463-9.

8. Millett DT, Cattanach D, McFadzean R, Pattison J, McColl J.Laboratory evaluation of a compomer and a resin modified glass

American Journal of Orthodontics and Dentofacial OrthopedicsAugust 2004

206 Summers et al

ionomer cement for orthodontic bonding. Angle Orthod 1999;69:58-63.

9. Flores AR, Saez G, Federico B. Metallic bracket to enamelbonding with a photopolymerizable resin-reinforced glass iono-mer. Am J Orthod Dentofacial Orthop 1999;116:514-7.

10. Maskeroni AJ, Meyers CE, Lorton L. Ceramic bracket bonding:a comparison of bond strength with polyacrylic acid and phos-phoric acid enamel conditioning. Am J Orthod DentofacialOrthop 1990;97:168-75.

11. Årtun J, Bergland S. Clinical trials with crystal growth condi-tioning as an alternative to acid-etch enamel pretreatment. Am JOrthod 1984;85:333-40.

12. Farquhar RB. Direct bonding comparing a polyacrylic acid and aphosphoric acid technique. Am J Orthod Dentofacial Orthop1986;90:187-94.

13. Itoh T, Matsuo N, Fukushima T, Inoue Y, Oniki Y, MatsumotoM, et al. Effect of contamination and etching on enamel bondstrength of new light-cured glass ionomer cements. Angle Orthod1999;69:450-6.

14. Cacciafesta V, Jost-Brinkmann PG, Sussenberger U, MiethkeRR. Effects of saliva and water contamination on the enamelshear bond strength of a light-cured glass ionomer cement. Am JOrthod Dentofacial Orthop 1998;113:402-7.

15. Bishara SE, VonWald L, Laffoon JF, Jakobsen JR. Effect ofaltering the type of enamel conditioner on the shear bondstrength of a resin-reinforced glass ionomer adhesive. Am JOrthod Dentofacial Orthop 2000;118:288-94.

16. Oliver RG. The effect of different methods of bracket removal onthe amount of residual adhesive. Am J Orthod DentofacialOrthop 1988;93:196-200.

17. Summitt JB, Robbins JW, Schwartz RS, Santos J. Fundamentalsof operative dentistry: a contemporary approach. QuintessencePublishing; 2001.

18. Greenlaw R, Way DC, Galil KA. An in vitro evaluation of visiblelight-cured resins as an alternative to conventional resin bondingsystems. Am J Orthod Dentofacial Orthop 1989;96:214-20.

19. Swartz ML, Phillips RW, Rhodes B. Visible light-activatedresins-depth of cure. J Am Dent Assoc 1983;106:634-7.

20. Meehan MP, Foley TF, Mamandras AH. A comparison of theshear bond strengths of two glass ionomer cements. Am J OrthodDentofacial Orthop 1999;115:125-32.

21. Owens SE Jr, Miller BH. A comparison of shear bond strengths

of three visible light-cured orthodontic adhesives. Angle Orthod2000;70:352-6.

22. Cook PA, Luther F, Youngson CC. An in vitro study of the bondstrength of light-cured glass ionomer cement in the bonding oforthodontic brackets. Eur J Orthod 1996;18:199-204.

23. Diedrich P. Enamel alterations from bracket bonding anddebonding: a study with the scanning electron microscope. Am JOrthod 1981;79:500-22.

24. Tavis MA, Watts DC. A visible light activated direct bondingmaterial. An in vitro study. Br. J Ortho 1984;11:33-7.

25. Reynolds JR. A review of direct orthodontic bonding. Br JOrthod 1975;2:171-8.

26. McSherry PF. An in vitro evaluation of the tensile and shearstrengths of four adhesives used in orthodontics. Eur J Orthod1996;18:319-27.

27. Newman GV, Facq JM. The effects of adhesive systems on toothsurfaces. Am J Orthod 1971;59:67-75.

28. Millet DT, McCabe JF. Orthodontic bonding with glass ionomercement—a review. Eur J Orthod 1996;18:385-99.

29. Fricker JP. A new self-curing resin-modified glass-ionomercement for the direct bonding of orthodontic brackets in vivo.Am J Orthod Dentofacial Orthop 1998;113:384-6.

30. Silverman E, Cohen M, Demke RS, Silverman M. A newlight-cured glass ionomer cement that bonds brackets to teethwithout etching in the presence of saliva. Am J Orthod Dento-facial Orthop 1995;108:231-6.

31. Cook PA. Direct bonding with glass ionomer cement. J ClinOrthod 1990;24:509-11.

32. Hegarty DJ, Macfarlane TV. In vivo bracket retention compari-son of a resin-modified glass ionomer cement and a resin-basedbracket adhesive system after a year. Am J Orthod DentofacialOrthop 2002;121:496-501.

33. Gaworski M, Weinstein M, Borislow AJ, Braitman LE. Decal-cification and bond failure: a comparison of a glass ionomer anda composite resin bonding system in vivo. Am J OrthodDentofacial Orthop 1999;116:518-21.

34. Hitmi L, Muller C, Mujajic M, Attal JP. An 18-month clinicalstudy of bond failures with resin-modified glass ionomer cementin orthodontic practice. Am J Orthod Dentofacial Orthop 2001;120:406-15.

35. Maijer R, Smith DC. Variables influencing the bond strength ofmetal orthodontic bracket bases. Am J Orthod 1981;79:20-34.

RECEIVE THE JOURNAL’S TABLE OF CONTENTS EACH MONTH BY E-MAILTo 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 mailinglist. Note that TOC e-mails will be sent when a new issue is posted to the website.