Antibacterial

properties
of currentorthodontic
bandcements
RobertP. Vokus,
DDS
George
J. Cisneros,
DMD,
MMSc
Michael
Levi,ScD

Abstract lease fluoride, it couldalso havean antibacterial ef-

Purpose: Manufacturerscommonlyprovide informa- fect in addition to its potential remineralizingprop-
tion on the physicalpropertiesof dentalmaterials,but in- erties, andshouldproveuseful in the clinical setting.
formationon their antibacterialpropertiesis often miss- A new dual-cure, resin-based, glass ionomer band
ing. This study determinedthe antibacterialpropertiesof cement, Band-Lok, has been introduced that can
four currentlyusedorthodonticbandcementsagainstthree bond to metal as well as enamel. Its manufacturer
different strains of Streptococcus mutans. purports that it releases fluoride, makingit useful in
Methods: The cements utilized were Durelon, TM the prevention of enameldecalcification.
Ketac, vMMizzyT~Zinc Phosphate, and Band-Lok,~ a The purposeof our study wasto comparethe anti-
recentlyintroduced,resin-based,dual-cureglass ionomer bacterial activity of four currentlyavailableorthodontic
cement.Diskdiffusion assaymethodology wasusedto test band cements: Durelon (ESPEPremier, Norristown,
for zones of bacterial inhibition aroundcementsamples. PA), Ketac (ESPEPremier, Norristown, PA), Mizzy
Zonesof inhibitionweremeasured in millimetersusingan Zinc Phosphate(Mizzy, Inc., Cherry Hill, NJ), and
electroniccaliper. In additionto curedcementplugsand Band-Lok (Reliance Orthodontics, Itasca, IL).
J~eshlymixedcementsamples,a newvariation, in the form In addition, a study was conducted to determine
of a cementplug surroundinga stainless-steel band,was the fluoride ion release and pH changes undergone
tested. Twelvecombinations resultedJ~om the four cement by cured samples of Band-Lokcement at 24-, 48-,
typesandthree forms. and 72-h intervals.
Results: Of the variables studied, the mix forms of
Durelon,Ketac, and MizzyZinc Phosphatecementshowed Methods
the greatestbacterialinhibition(Kruskal-
Wallis,P<O.05). Prior to study initiation, the primaryinvestigator
Amongthe cementstested, Mizzy Zinc Phosphateshowed received extensive training and supervisionin proper
the largest zones of inhibition, with DurelonandKetac
microbiological techniques and procedures. In addi-
havingcomparable zones of inhibition (Kruskal-Wallis,
tion, a pilot study was performedto verify the ap-
< O. 05). Band-Lokdid not exhibit an inhibitory effect
propriateness of the study design.
againstanyof the three strains orS. mutanstested.
Conclusion:A "containment effect" of no bacterialin- Cement sample preparation
hibition wasobservedin the cementsamplessurrounded The four cementswere tested in three forms gen-
by the stainless-steel bandmaterial.(PediatrDent20:1, erating 12 experimental samplegroups as follows:
43-48, 1998) 1. Cured Durelon cement plug
ixed orthodontic appliances predispose teeth 2. Cured Ketac Glass Ionomer cement plug

F to increased plaque accumulation. Favored

sites for this accumulationinclude cervical
margins and where cement has been lost beneath
bands. 1 Enamel demineralization in the form
of "white spot" lesions is a negative sequela of
treatment in 50%of patients undergoingorthodon-
2tic treatment.
Previous studies have demonstrated that small
carious lesions can be remineralized in the presence
of fluoride ions. 3-5 Ifa cementcould be shownto re-
Pediatric Dentistry- 20:1, 1998
3. Cured Mizzy Zinc Phosphate cement plug
4. Cured Band-Lok cement plug
5. CuredDureloncementplug, stainless-steel band
6. Cured Ketac Glass Ionomer cement plug,
stainless-steel band
7. Cured Mizzy Zinc Phosphate cement plug,
stainless-steel band
8. CuredBand-Lok cementplug, stainless-steel band
AmericanAcademyof Pediatric Dentistry 43
 9. Durelon cement, fleshly mixed three strains of S. mutans.Thezone of bacterial in-
10. Ketac Glass Ionomer cement, fleshly mixed hibition, if any, whichoccurred around the samples
was measured from the edge of the cement sample
11. Mizzy Zinc Phosphate cement, fleshly mixed
to the edge of the zoneof inhibition, to the nearest
12. Band-Lokcement, fleshly mixed tenth of a millimeter, with an electronic digital cali-
A custom, 5-mmdiameter by 5-mmhigh poly- per. For all samplesin the study, measurements were
ethylene mold was fabricated to produce uniformly performedthree times, and averaged.
sized cement plug samples. Cement samples were The inoculated Petri dishes and cement samples
prepared using fleshly mixedcementaccording to the were incubated at 35°Cand the zones of bacterial
manufacturer’sinstructions, and cured shortly before inhibition were measuredby a single investigator at
placementin the preparedwells in the Petri dishes. 24-, 48-, and 72-h intervals. The electronic caliper
Samplesof the cementand stainless-steel band ma- was zeroed prior to each measurementto ensure ac-
terials were prepared using a custom, 5-mmdiameter curacy of the measurements.
moldcreated by stock stainless-steel band material In eachPetri dish, five samplesof onetype of cement,
(0.125 in. x 0.004in.), the cementoccupyingthe cen- in eachof the three forms,wasplacedwithinthe prepared
ter of the samples. wells, andinoculatedwith oneof three strains ofS. mu-
Each cement in the mixed sample group was pre- tans;, e.g. 5 Durelonplugs, 5 Durelonplugswith band,
pared as above.Usingsterile syringes, the uncuredce- and 5 Dureloncementsamples, freshly mixed, in one
ment was placed into the 5-mmwells, whichhad been Petri dish inoculatedwith S. mutans#25175(Fig 1).
prepared within the agar and allowedto cure.
pHandFluoride release
Bacterial sample preparation
A study of the pHchangesand fluoride ion release
S. mutansbacteria wasusedbecauseof its association of Band-Lokcement was madeat 24-, 48-, and 72-
with dental caries.6 Additionally,previousinvestigators h intervals. Five uniform samples of freshly mixed
haveusedS. mutansin similar experiments. 7’ 8 Thefol-
and self-cured Band-Lokcement were prepared ac-
lowingstrains orS. mutanswereuseddue to their clas- cording to the manufacturer’s directions. The
sification as bacteriacapableof dentalcaries production: samples were placed in 40 mLof de-ionized water
S mutans ATCC#25175 in covered polyethylene tubes, then inverted and
S rnutans ATCC#33402 gently agitated three times prior to pHtesting with
S rnutans ATCC#35668 a pH electrode (Orion Research, Cambridge, MA),
Mueller Hinton agar with 5%sheep blood was uti- and fluoride release with a fluoride electrode (Orion
lized as a suitable growthmedium for S. mutans. Uni- Research, Cambridge, MA). The samples were de-
form 5-ram wells were cut within the agar to allow pleted by 6 mL(2 mLfor fluoride, 4 mLfor pH)
cementsamplesto touch the agar. Eachplate had5 wells
madein equalthirds of the plate for a total of 15 wells Plug and Band Samples
per plate (Figure1).
ThePetri dishes wereuniformlyinoculatedwith one
strain ofS. mutansusing the standard methodfor disk
diffusionsusceptibility to providefor a uniformbacte-
rial lawnof growth. 9 Streakingof the plates wasaccom-
plished using a sterile, cotton-tipped applicator which
had beensaturated with onestrain of bacteria. Theap-
plicator wasdrawnalong the surface of the Petri dish
in a zig-zag pattern to cover the entire plate surface.
The Petri dish was then rotated 90° and the applica-
tor wasagain drawnacross the entire surface in a zig-
zag fashion to ensure completecoverageof the plate.
This streaking of the plates was completedfollow-
ing preparation of the wells.
Experimental technique
Each sample was placed in prepared wells within Fig1. Experimental
PetriDish
Set-Up.
5 plug,5 plug
and
agar in Petri dishes that were inoculated with one of band,5 mixed
samples.

44 AmericanAcademyof Pediatric Dentistry PediatricDentistry - 20:1, 1998

after each time point (incubation solution = 40 mLat ferences amongbacteria weretested for significance at
~°
24 h, 34 mLat 48 h, 28 mLat 72 h). each of the time intervals. For each of the forms,dif-
Statistical
analysis ferences amongcementtypes werealso tested for sig-
nificance at each of the time periods. For every
A Type I error of .05 and 80%power would de-
combinationof bacteria, form, and time point, differ-
tect a difference of 50%betweenthe radii of the mea-
ences were examined.
sured zonesof bacterial inhibition.
For purposesof illustration, meansof the zonesof
Asthe initial data werenot normallydistributed, a
bacterial inhibition were computedfor cementtypes
Box CoxTransformation was completedto transform
and cementforms at 24-, 48-, and 72-h time inter-
the data to a morenormaldistribution, resulting in
vals and presented in tabular form.
raw data exponentiatedto the -1.5 power.
A repeated-measures three-way ANOVA was per- Results
formedwith time utilized as the repeat factor (24, 48, There were significant differences amongthe ce-
and 72 h). Thethree factors utilized in the modelwere ment types, with Mizzy Zinc Phosphate having the
bacteria (ATCC#25175, #33402,and #35668), form greatest bacterial inhibition (P < 0.05, Kruskal-Wallis),
(mix, plug, and plug with band), and cement type followed by Durelon and Ketac with comparablein-
(Durelon, Ketac, MizzyZinc Phosphate, and Band- hibition (P < 0.05, Kruskal-Wallis), and Band-Lok
Lok). havingthe least inhibition at each of the time points
For each bacteria type and each hour of observa- (Tables1-3). There weresignificant differences among
tion, descriptive statistics were generatedincluding formsat all time points, such that the greatest bacteo
means,standard deviations, medians, and ranges for rial inhibition occurredfor the mixformand the least
the transformeddata. Becauseit wasdifficult to dis- inhibition occurred in the plug with band form
cern whether assumptions of the three-way ANOVA (Tables 1-3).
weremet, and as three-wayinteractions weresignifi-
Theeffectof differentmicroorganisms
cant, differences between factors--holding one or
moreof the other factors constant--were tested for There were no significant differences amongbac-
significance using nonparametricmethods. teria types for the plug and band forms of cement(P
AKruskal-Wallis test wasusedto test for differences < 0.05). Mixedcements demonstrated significantly
amongthe bacteria types for each of the time points. greater inhibition against S. mutans#33402 (P < 0.05)
Similar analyses weredoneto assess differences among (Table 2). MizzyZinc Phosphate, in the mix form,
forms and cementtypes. For each cementtype, dif- displayedsignificant differencesat all time points como

Plug Form 24 h 48 h 72 h
Durelon 0 0 0
Ketac 0 0 0
Zinc Phosphate 1.33" -+ 0.23 2.18" +0.16 2.50" +0.28
Band-Lok 0 0 0
Plug_ and Band Form 24 h 48h 72h
Durelon 0 0 0
Ketac 0 0 0
0.18" + 0.15 0.31" _+0.21 0.33" _+ 0.22
Band-Lok 0 0 0
Mixed Form 24 h 48 h 72 h
Durelon 2.15" + 0.18 2.18" +_ 0.19 2.32" + 0.21
Ketac 2.77" + 0.17 3.60" + 0.10 3.64" -+ 0.10
Zinc Phosphate 3.14 .+
__ 0.10 3.70" + 0.17 3.72’ + 0.17
Band-Lok 0 0 0
"P < O.05-Kruskal-Wallis

Pediatric Dentistry- 20:1, 1998 AmericanAcademyof Pediatric Dentistry 45

 Plug Form 24 h 48 h 72 h
Durelon 0.62" _+0.08 0.64 + 0.08 0.70 + 0.09
Ketac 0 0 0
Zinc Phosphate 2.40" + 0.57 3.28" + 0.61 3.74" + 0.66
Band-Lok 0 0 0
Plug~ and Band Form 24 h 48 h 72 h
Durelon 0.08" + 0.06 0.09" + 0.08 0.09" + 0.08
Ketac 0 0 0
Zinc Phosphate 0 0 0
Band-Lok 0 0 0
Mixed Form 24 h 48 h 72 h
Durelon 2.84" + 0.25 3.16" + 0.24 3.47" +0.19
Ketac 1.56" + 0.16 2.57" + 0.23 2.75" + 0.21
Zinc Phosphate 4.58" + 0.30 4.52" + 0.30 5.01" + 0.22
Band-Lok 0 0 0
"P < O.05-Kruskal-Wallis

Plug Form
Durelon 0 0 0
Ketac 0 0 0
Zinc Phosphate 1.3 i" + 0.03 !,88’ + 0.27 2.20" + 0.40
Band-Lok 0 0 0
Plug and Band Form 24h 48 h 72h
Durelon 0
Ketac 0 0 0
Zinc Phosphate 0 0 0
Band-Lok 0 0 0
Mixed Form 24 h 48 h 72h
Durelon 1.57" + 0,05 1.63" + 0.05 1.80"_+ 0.09
Ketac 3.33" + 0.29 3,59" + 0.22 3.64" + 0.21
Zinc Phosphate 3.23" + 0.38 3.95" + 0.57 4.07" + 0.59
Band-Lok 0 0 0
"P < O.05-Kruskal-Wallis

pared to Band-Lok,Durelon,and Ketac (P < 0.05) Theeffectof different cements

(Tables1-3). All three strains ofS. mutansshowed
the
greatest inhibition with MizzyZincPhosphatein the
plug and mixform(P< 0.05) (Tables 1-3). The
organismsdemonstratedthe least inhibition with
Band-Lok(P < 0.05). Band-Lokshowedno inhibi-
tion in any form, regardlessof the organismtested
(Tables1-3).
46 AmericanAcademyof Pediatric Dentistry
andforms
Mixedcementsdemonstratedthe greatest inhibi-
tion of all organisms,with MizzyZinc Phosphate
showingthe greatest inhibition and Band-Lokthe
least (P < 0.05) (Tables 1-3). Band-Lok showed
bacterial inhibition in anyform(Tables1-3). In the
plug form, MizzyZinc Phosphateshowedsignificant
PediatricDentistry - 20:1, 1998
 though this containment effect maybe undesirable
TAeLE4, EVALUATION
OF PHCHANGES
ANn whenevaluating materials by the disk diffusion assay
methodology, it moreaccurately approximatesthe clini-
cal setting. Enameldemineralizationoccurs adjacent
pH changes over time to the band/tooth interface, not just beneath the
Sample 24 h 48 h 72 h band. Orthodontic bands maylimit the diffusion of
1 5.07 5.78 5.80 low-level fluoride ions from the bandedtooth and
2 5.06 5.33 5.68 therefore reduce the fluoride ion’s potential for
3 5.05 5.52 5.59 proximalorregional remineralization.
The finding that Band-Lokcementis not inhibi-
4 4.97 5.56 5.81
tory against S. mutansmaybe explainedby the evalu-
5 5.30 5.69 5.88 ation of fluoride release and pHchangesafter curing.
Fluoride release (ppm) In this aspect of the study, the lowestpHrecordedwas
Sample 24 h 48 h 72 h 4.97 with 5.09 the meanpHat 24 h. The pH contin-
1 0.91 1.48 1.48 ued to rise during the next 48 h with a meanpH of
2 1.30 1.71 1.85 5.75 at 72 h. ApHof at least 4.0 has beenreported as
8necessaryfor a cementto exhibit a bactericidal effect.
3 1.02 1.48 1.48
In DeSchepper and coworkers’study, 8 all antibacte-
4 1.50 1.71 1.80 rial activity of the cementsstudied waslost whenthe
5 0.91 1.45 1.65 pH of the liquid componentswas adjusted to 5.0.
The minimum range of fluoride release to inhibit
Samples wereinverted gently three times prior to testing for pH
and fluoride release. Samplesweredepleted by 6 mL(2 mLfor S. mutans ranges from20 to 300 ppm.13 In this study,
fluoride, 4 mLfor pH)after eachtime point, (i.e., incubation very low levels of fluoride release wererecordeddur-
solution = 40 mLat 24 h, 34 mLat 48 h, 28 mLat 72 h). ing the 72-h period of evaluationperiod followingthe
inhibition against all strains ofS. mutansand Durelon curing of samples.At 24 h, the highest recordedlevel
against S mutans#33402 (P < 0.05). All but a minor was only 1.50 ppm, with a meanof 1.13 ppm. At 72
inhibitory effect with Mizzy Zinc Phosphate and h, the highest level recordedwas1.85 ppm,with a mean
Durelon were blocked when cement plugs were of 1.65 ppm.
bandedwith stainless-steel rings (Tables 1-3). When evaluatingan antibacterial effect with the disk
Fluoride release andpHalteration of Band-tok diffusion assay methodology,diffusion of inhibitory
substancesis required. Apossible explanationfor the
The meanpH for the freshly mixed Band-Lokce- lack of an inhibitory effect maylie with the chemical
ment samples was pH 5.09 at 24 h, whichrose to a composition of Band-Lok.As a resin-based, glass-
meanpH 5.75 at 72 h. The meanfluoride release ionomercement, substances whichmayhave provided
for the freshly mixed Band-Lokcementsamples was an inhibitory effect could havebeen chemicallybound
1.13 ppmat 24 h, 1.57 ppmat 48 h, and 1.65 ppm andunableto diffuse.
at 72 h (Table 4). Band-Lokcement appears inferior to other com-
Discussion monly used orthodontic band cements(i.e., Durelon,
Ketac, and MizzyZinc Phosphate)from the standpoint
of being inhibitory m S. mutans. However,Band-Lok
erties of orthodontic cementsusing the disk diffu-
cementcould prove to be clinically useful if it were
sion assay method.8’ 11 Cementplugs and freshly
shownto be less soluble in the oral environmentthan
mixed cement have been the most commonlyused
the other cements. If cementwash-outwascompletely
forms.8.12 An experimental design wasintroduced in
eliminated, then decalcification beneaththe bandmay
this study;i.e., stainless-steel bandmaterialin contact
not be an issue, due to the exdusionof acid-producing
with the cementsample. Thegoal of this combination
bacteria. Aninhibitory bacterial effect at the tooth/ce-
wasto moreclosely represent the clinical condition of
mentmarginwouldstill be highly desirable in the pre-
a cementedorthodonticband, and to evaluate any syn-
ventionof enameldecalcification, secondaryto bacterial
ergistic effect of the bandand cementcombination.The
colonization. Studies should be conductedto evalu-
bandactually inhibited the antibacterial effect of the
ate the relative solubility of Band-Lok cement.
cements, as shownin the plug or mix form. The band
In an attempt to increase the antibacterial effect
apparently contains those cement componentsthat
of Band-Lokcement, the manufacturercould consider
wouldordinarily diffuse and cause an inhibition. Al-
Pediatric Dentistry- 20:1, 1998 AmericanAcademyof Pediatric Dentistry 47
 modifications of its composition to increase the
amountof fluoride released by the cementfollowing
curing. Loweringthe pHof the cement, whichwould
also increaseits antibacterial effect, wouldmostprob-
ably be ill-advised, due to the potential for enamelde-
calcification shownwith cementsexhibiting a low pH
15
on setting.
MizzyZinc Phosphate, Durelon, and Ketac do have
significant antibacterial properties against the three
strains ofS. mutansshownto cause dental caries. These
results agree with previous studies that showedthe
greatest bacterial inhibition occurring with cements
containingzinc oxideor release fluoride. 11.16Thefind-
ing that the freshly mixedformof the cementsshowed
the most bacterial inhibition has also been shownby
previousstudies.7’ 8
Other researchers have used bacteria such as Lac-
tobacillus salivarius, Streptococcussobrinus, andActi-
nomycesviscosus, in addition to S. mutans, in an
attempt to more accurately duplicate the microbial
conditions of the mouth.7’ 11 S. mutanswas chosen
becauseof its proven role in the production of den-
tal caries. 6’ 17 Themouthis a complicatedmicrobio-
logical ecosystem that is virtually impossible to
duplicate in an experimental design, because of the
multitude of synergistic bacterial interactions in-
volved in the production of dental caries.
Since the completionof this study, Ultra Band-Lok
has been introduced by Reliance Orthodontics. Ultra
Band-Lok is similar to Band-Lok in that it is a resin-
based, glass-ionomercement,but differs in that it is
exclusively a light-cured cementwhereasBand-Lok is a
dual-cure cement. Accordingto the manufacturer, the
6fluoride release of the twoproductsare similar)
Conclusions
l. Theorthodontic band’s containmenteffect tends
to limit anyantibacterial propertiesof the cements
tested
2. Band-Lokcementdoes not exhibit an antibacte-
rial effect againstthree strains ofS. mutanstested
3. The pH changes and fluoride release of
Band-Lokcement appear insufficient to be
bacteriostatic
4. The mix form of MizzyZinc Phosphate,Durelon,
and Ketac cementsexhibited the greatest bacte-
riostatic properties comparedwith other forms
5. Mizzy Zinc Phosphate is the most bacterio-
static, with Durelon and Ketac cementhaving
comparablebacteriostatic properties.
48 American
Academy
of PediatricDentistry
RobertP. Vokusis in orthodontic private practice in Fort Myers,
Florida. GeorgeJ. Cisnerosis associate professor, Albert Einstein
College of Medicineand director, Postgraduate OrthodonticPro-
gram, MontefioreMedicalCenter. MichaelLevi is assistant pro-
fessor, Albert Einstein Collegeof Medicineand associate director
of Microbiologyand assistant professor of Pathology,Montefiore
Medical Center, Bronx, NewYork.
The authors thank Dr. Norman Tinanoff and Mr. David M.
O’Sullivan, University of Connecticut Health Center, School of
Dental Medicine, for technical assistance in determiningcement
pHvalues and fluoride release. Theauthors alsothank Dr. Kathy
Freeman,Director of Bostatistics at MontefioreMedicalCenter,
for her assistancewiththe statistics.
References
i. MizrahE: Enameldemineralization following orthodontic
treatment. AmJ Orthod 82:62-67, 1982.
2. Gorelick L, Geiger AM,GwinnettAJ: Incidence ofwhite spot
formation after banding and bonding. AmJ Orthod 81:93-
98, 1982.
3. Ogaard B, Rolla G, Arends J: Orthodontic appliances and
enameldemineralization. Part 1. Lesion development.AmJ
Orthod Dentofacial Orthop 94:68-73, 1988.
4. OgaardB, Rolla G, ArendsJ, ten Cate JM:Orthodontic ap-
pliances and demineralization.Part 2. Preventionand treat-
ment of lesions. AmJ Orthod Dentofacial Orthop 94:123-
28, 1988.
5. Silverstone LM:Remineralization phenomena.Caries Res
11:59-84, 1977.
6. Zickert I, EmilsonCG,Krasse B: Correlation of level and
duration of Streptococcusinfection with incidence of dental
caries. Infect Immun,39: 982-85, 1983.
7. FischmanSA, Tinanoff N: The effect of acid and fluoride
release on the antimicrobial properties of four glass ionomer
cements. Pediatr Dent 16:368-70, 1984.
8. DeSchepperEJ, White RR, vonder Lehr W: Antibacterial
effects of glass ionomersAmJ Dent 2: 51-56, 1989.
9. National Committeefor Clinical LaboratoryStandards, Per-
formancestandards for antimicrobialtesting, vol. 15, no. 4,
9-10, 1995.
10. Tinanoff N: Personal communicationon laboratory tech-
nique, 1995.
11. Scherer W, LippmanN, KaimJ: Antimicrobial properties
of glass-ionomer cementsand other restorative materials.
Oper Dent, 14:77-81, 1984.
12. Wiltshire WA,Janse van RensburgSD:Fluoride release from
four visible light-cured orthodontic adhesive resins. AmJ
Orthod Dentofacial Orthop 108:278-83, 1995.
13. Ferretti GA,Tanzer JM, TinanoffN:The effect of fluoride
and stannousions on Streptococcusmutans.Viability, growth,
acid, glucan production, and adherence. Caries Res 16:298-
3O7, 1982.
14. Wisth PJ: The role of zinc phosphatecementin enamelsur-
face changes on banded teeth. Angle Orthod 40:329-33,
1970.
15. Loyola-RodriguezJP, Garcia-GodoyF, Lindquist R: Growth
inhibition of glass ionomercementson mutansstreptococci.
Pediatr Dent 16:346-49, 1994.
16. Hamada S, Slade HD: Biology, immunology, and
cariogenicity of Streptococcus mutans. Microbiol Rev 44:
331-84, 1980.
17. GangeP: Personal communication on the properties of Band-
LokTM and Ultra Band-LokTM cements, 1996.
PediatricDentistry- 20:1,1998