Bacterial Adhesion on Smoothand Rough Titanium Surfaces AfterTreatment With Different InstrumentsPoliana Mendes Duarte,* Andre Figueiredo Reis,† Patrıcia Moreira de Freitas,‡and Claudia Ota-Tsuzuki*

Background: Newly formed biofilm after implant debride-ment may challenge the long-term stability of peri-implanttherapy. This in vitro study aimed to assess the roughnessand adherence of Streptococcus sanguinis after treatment ofsmooth and rough titanium surfaces with an erbium-doped:yttrium, aluminum, and garnet (Er:YAG) laser, metal andplastic curets, and an air-powder abrasive system.

Methods: Forty titanium disks with smooth-machined sur-faces and 40 with sand-blasted and acid-etched surfaceswere divided into the following treatment groups: Er:YAG la-ser; plastic curet; metal curet, and air-powder abrasive sys-tem. The surface roughness (roughness average [Ra]) beforeand after treatments was determined using a profilometer.S. sanguinis (American Type Culture Collection 10556) wasgrown on treated and untreated specimens, and the amountsof retained bacteria on the surfaces were measured by the cul-ture method. Rough and smooth surfaces with and withouta suspension of S. sanguinis were also analyzed using scan-ning electron microscopy (SEM).

Results: For smooth surfaces, the roughest surfaces wereproduced by metal curets (repeated-measures analysis ofvariance [ANOVA] and Tukey test; P <0.05). The rough-surface profile was not altered by any of the treatments(repeated-measures ANOVA; P >0.05). Rough surfacestreated with metal curets and air-powder abrasion showedthe lowest level of bacterial adhesion (two-way ANOVAand Tukey test; P <0.05). SEM analysis revealed distinctsurface profiles produced by all devices.

Conclusions: Metal curets are not recommended forsmooth titanium surface debridement due to severe texturealteration. Rough surfaces treated with a metal curet and theair-powder abrasive system were less susceptible to bacte-rial adhesion, probably due to texture modification and thepresence of abrasive deposits. J Periodontol 2009;80:1824-1832.

KEY WORDS

Air-powder abrasive system; Er:YAG laser; metal curets;mucositis; peri-implantitis; plastic curets; SEM; scaling.

Titanium dental implants have beenconsidered excellent alternativesto conventional prostheses in the

oral rehabilitation of partially and totallyedentulous subjects. Therefore, varioustypes of implant surfaces, ranging fromsmooth machined to rough surfaces, arecurrently present in human oral cavi-ties.1 Despite the efforts to improveosseointegration by the modification ofimplant surfaces, evidence has shownthat bacterial infection inducing muco-sitis or peri-implantitis can jeopardizethe long-term success of some implantrehabilitations.2,3 Both peri-implant dis-eases are infectious disorders associ-ated with pathogenic bacterial speciescommonly observed in periodontaldiseases.2,3 Therefore, similar to peri-odontal treatment, the removal of bac-terial biofilm and calculus depositsaround implants seems to be crucial inthe prevention and treatment of peri-implant infections.4-6

Various procedures and instrumentshave beenproposed to reduce the numberof pathogenic species and, consequently,to improve or preserve periodontal healtharound titanium implants.4,5 Besides themechanical removal of biofilm by plasticcurets, air-powder abrasive systems,and the application of chemical agentsand local antimicrobials, lasers havebeen introduced as a potential alternativein reducing pathogens on implant sur-faces.4,6-9 Among lasers used indentistry,

* Department of Periodontics, Dental Research Division, Guarulhos University, Guarulhos,SP, Brazil.

† Department of Operative Dentistry, Dental Research Division, Guarulhos University.‡ Department of Restorative Dentistry, School of Dentistry, University of Sao Paulo, Sao

Paulo, SP, Brazil. doi: 10.1902/jop.2009.090273

Volume 80 • Number 11

1824

the erbium-doped:yttrium, aluminum, and garnet(Er:YAG) laser has been considered a promising ther-apy for peri-implant disinfection.10 The Er:YAG laserhas demonstrated a high potential for bacterial re-duction on implant surfaces11 and the ability to re-move biofilm from both smooth and rough titaniumsurfaces, leading to significant improvement ofperi-implant clinical parameters and new bone-to-implant formation.12-14

Several studies15-20 investigated the effects of dif-ferent mechanical treatments, such as metal andplastic curets and air-powder abrasive systems, ontitanium surfaces with respect to texture changes,cleaning efficacy, and fibroblast attachment. In addi-tion, the effects of the Er:YAG laser on the morphologiccharacteristics of implant surfaces, effectiveness forremoving biofilm, and biocompatibility of osteoblasticand fibroblastic cells were investigated.7,11,21-23 How-ever, there is limited information about the influence ofthe titanium surface modification after treatment withdifferent instruments on bacterial adherence. The sur-face profile and roughness produced by different in-struments could have an important impact on thenewly formed biofilm and, consequently, can be an im-portant factor in peri-implant health maintenance.Therefore, the aim of this in vitro study was to assessthe surface roughness (roughness average [Ra]) andadherence of Streptococcus sanguinis (AmericanType Culture Collection [ATCC] 10556) after treat-ment of smooth and rough titanium surfaces with anEr:YAG laser, metal and plastic curets, and an air-powder abrasive system.

MATERIALS AND METHODS

SampleSmooth and rough disks made out of commerciallypure titanium (grade 4) measuring 5 mm in diameterand 3 mm in thickness were used.§ The smooth sur-faces were machined, whereas the rough surfaces weresand-blasted with aluminum oxide beads and acid-etched with nitric acid. The disks were removed fromthe originalpackaging and stored individually inmicro-tubes before experimental procedures. The disks werehandled by the lateral walls to avoid contact with thetreated surfaces that could alter the surface profile.

TreatmentsThe smooth and rough disks were randomly dividedinto one of the following groups.Er:YAG laser (n = 10 smooth; n = 10 rough). Sam-ples were irradiated with the Er:YAG laseri workingat 2,940 nm. The energy and repetition rate of thisequipment ranged from 60 to 500 mJ and 1 to15 Hz, respectively. A periodontal handpiece (#2056)was used with a prismatically cut glass tip (1.1 ·0.5 mm). The fluency and repetition rate used for

the laser irradiation were 8.4 J/cm2 and 10 Hz, re-spectively. Laser parameters were set at 120 mJ/pulse, and the energy delivered at the end of the tip,taking into account the transmitting factor (0.56)for the selected tip, was 67.2 mJ/pulse. The tip wasused at an incidence angle of 45! under continuouswater irrigation. The application tip was moved frombottom to top and maintained in slight contact withthe disk surface.

Plastic curet (n = 10 smooth; n = 10 rough). Sam-ples were scaled from bottom to top with the plasticcuret¶ in which the tip was placed at a contact angleof 70!.

The curet had a steel handle, and tips were pro-duced from a high-grade resin. Each side of the curetwas used for five specimens and then replaced bya new side.

Metal curet (n = 10 smooth; n = 10 rough). Sam-ples were scaled from bottom to top using a Graceycuret# in which the blade was placed at a contact an-gle of 70!. Each side of the curet was used for fivedisks and then replaced by a new side.

Air-powder abrasive system (n = 10 smooth; n =10 rough). Samples were treated with an air-powderabrasive system** using medium water and an air-power setting. The insert was perpendicularly appliedto the surfaces and worked by directing fine particlesof sodium bicarbonate that were propelled by com-pressed air in a water spray at a distance of 5 mm fromthe surface.

The entire top surface of the disks was treated for50 seconds by the same operator (PMD), resultingin ;30 curet strokes for each sample. No attemptwas made to standardize the application of the scalingforce because the operator applied the curets freely.This protocol was applied in an attempt to simulateapproximately five clinical visits for implant decon-tamination.

Surface RoughnessAll disks were rinsed with distilled water and allowed toair dry. The Ra (mm) of the titanium surfaces beforeand after treatments were determined using a sur-face profilometer.†† The surface roughness of eachspecimen was scanned with a diamond microneedle(10 mm diameter) using a cutoff of 0.8 mm (lc) anda speed of 0.1 mm/second (ISO 4228). A maskedoperator (AFR) made all measurements in two lon-gitudinal and two transversal directions, and thescanned area was limited to the size of the disks(5 mm in diameter).

§ AS Technology, Sao Jose dos Campos, SP, Brazil.i KaVo KEY II, Kavo, Biberach, Germany.¶ Implacare, Hu-Friedy, Chicago, IL.# 5-6, Hu-Friedy.** Jet Sonic, Gnatus, Ribeirao Preto, SP, Brazil.†† TR200, Time Group, Beijing, China.

J Periodontol • November 2009 Duarte, Reis, Freitas, Ota-Tsuzuki

1825

Adhesion AssayBesides the above described treated disks, 10 roughand 10 smooth disks did not receive any treatmentand were used as controls to the assay.Saliva coating of the specimens. Unstimulated salivawas collected from each one of four healthy male do-nors (age range, 18 to 24years) for1 hour per day for7days (in February 2009). All donors provided writteninformed consent. This study protocol was approvedby the Ethics Committee in Clinical Research ofGuarulhos University. The saliva samples were fro-zen at -20!C until a total of 500 ml was collected.Subsequently, the saliva sample was pooled andcentrifuged(27,000 · g for30 minutesat4!C). The su-pernatant was pasteurized (30 minutes at 60!C) to in-activate endogenous enzymes, recentrifuged (27,000· g for 30 minutes at 4!C) in sterile bottles, and storedat -20!C.Thepasteurizationefficacywasevaluatedbyplating 100 ml saliva onto brain-heart infusion (BHI)agar‡‡ and observing the absence of bacterial growthafter 72 hours. The disks were autoclaved (15 minutesat 127!C) and placed in a well of a sterile 24-wellpolystyrene cell-culture plate§§ containing 500 mlsaliva for 4 hours to allow salivary pellicle formation.

Adhesion assay. Saliva was aspirated from eachwell and replaced with 500 ml BHI broth (double con-centrated) and 500 ml saliva. Inoculums were pre-pared by harvesting the standard reference strainS. sanguinis (ATCC 10556) cells from BHI agar platespreviously inoculated and incubated under micro-aerofilic conditions for 24 hours (candle jar; 37!C).The bacterial cells were suspended in sterile salinesolution, adjusting the turbidity to optical density(OD)630 0.15 (;106 colony forming units (CFUs)/ml), and each well was inoculated with 100 ml of thisinoculum suspension. Plates were incubated for 16hours under microaerofilic conditions. Afterwards,the specimens were washed in sterile saline solutionto remove unattached cells and inserted in microtubescontaining 1,000 ml sterile peptone water. The micro-tubes were vigorously vortexed for 2 minutes to freethe bacteria attached on the surface of each specimenand sonificated to disperse bacterial cells. Serial dilu-tions (10-4 to 10-8) of these suspensions were madeand inoculated in BHI agar plates for 48 hours. Testswere performed in triplicate, and the CFUs were de-termined using a stereomicroscope by an examiner(COT) who was masked to the experimental groups.

Scanning Electron Microscopy (SEM)Smooth and rough disks from each treatment groupwere observed using SEM. In addition, smooth andrough untreated disks were examined to observeany preexisting surface defects. Specimens were at-tached to stubs, placed into the vacuum chamber ofa scanning electron microscope,ii and the central

areas of the disks were photographed at magnifica-tions ·200 and ·1,000. The microscopic appearanceof S. sanguinis on treated and control titanium sur-faces was also observed by SEM. Samples were fixedin 2.5% glutaraldehyde in 0.05 mol/l cacodylatebuffer, pH 7.4. Subsequently, they were fixed, post-fixed, dehydrated in ascending ethanol concentra-tions up to 100%, sputter-coated with gold,¶¶ andobserved by SEM. Representative areas of treatedand untreated smooth and rough surfaces after bacte-rial incubation were photographed at a magnifica-tion ·10,000.

Statistical AnalysesStatistical analyses were performed using software.##

Comparisons were made among untreated andtreated smooth and rough surfaces and among thefour types of treatment modalities. The Ra was regis-tered for each scanned position and averaged foreach disk and, subsequently, for each group beforeand after treatments. Repeated-measures analysisof variance (ANOVA) was used to compare the sur-face roughness among groups before and after treat-ments. CFUs for disks inoculated in triplicate wereaveraged and submitted to logarithmic transforma-tion. Normal distributions of log10 CFU values perspecimen were confirmed by the Kolmogorov-Smirnov test. Two-way ANOVA was used for compar-isons among CFU formed on untreated (control) andtreated rough and smooth surfaces. When significantdifferences were detected by repeated-measuresANOVA or two-way ANOVA, a pairwise comparisonwas performed using the Tukey test. The significancelevel established for all analyses was 5%.

RESULTS

Surface RoughnessThe Ra for smooth and rough titanium surfaces beforeand after treatments are shown in Table 1. Statisticalanalysis revealed no significant differences in Rawithin smooth or rough surfaces before treatments(P >0.05), indicating a similar profile within each typeof surface. For smooth surfaces, there was an increasein Ra values after treatment with a metal curet(P <0.05), whereas no significant differences were ob-served after treatment with the Er:YAG laser, plasticcuret, and air-powder abrasive system (P >0.05).The smooth specimens treated by the metal curet pre-sented the roughest surface compared to the othertreatments (P <0.05). No significant changes in Ravalues were registered after treatment of rough sur-faces with any instrument (P >0.05).

‡‡ Difco, Sparks, NE.§§ Costar Corning, New York, NY.ii LEO 435 VP, LEO Electron Microscopy, Cambridge, U.K.¶¶ MED 010, BAL-TEC, Furstentum, Liechtenstein.## SANEST, Minas Gerais Agriculture Research Institute, Belo Horizonte,

MG, Brazil.

Bacterial Adhesion After Implant Treatment Volume 80 • Number 11

1826

Adhesion AssayFigure 1 presents the adherence ofS. sanguinis CFU (log10) to controland treated rough and smooth sur-faces. There were no differences inthe levels of S. sanguinis adhesionamong control and treated smoothsurfaces. With regard to rough sur-faces, control and Er:YAG laser–treated disks showed the highestlevels of S. sanguinis adhesion,followed by those treated withplastic curets, metal curets, andthe air-powder abrasive system, re-spectively (P <0.05). Control, laser-treated and plastic curet-treatedrough surfaces presented a highernumber of bacterial adhesion thanthe corresponding smooth ones(P <0.05). Differences betweenrough and smooth surfaces treatedwithmetalcuretsand theair-powderabrasive system were not significant(P >0.05).

SEMFigure 2 presents the SEM photomicrographs of thecontrol and treated smooth and rough titanium sur-faces. The smooth control surfaces exhibited pro-nounced circumferential machining marks, whereasthe rough control specimens showed irregular topog-raphy due to surface porosities. Surface imperfectionsthat appeared as metal scratches or tags were com-monly observed in the smooth control disks and wereirregular in size, shape, and distribution. The Er:YAGlaser produced slight alterations on smooth and roughtitanium surfaces. The machining marks were still ob-served on smooth surfaces treated with the Er:YAG la-ser; however, these marks were less evident than thecontrol ones. Some thin scratch lines over the originalsurfaces, probably produced by the tip of the laser de-

vice, were also observed on smooth and rough spec-imens treated with the laser. Plastic curets did notappear to significantly affect either of the titanium sur-faces, retaining surface characteristics similar to thecontrol ones. Only minor grooves produced by thetips of the curet and slight flattening of the ridges ofthe porosities were noted for smooth and rough sur-faces treated with plastic curets, respectively. Thesurfaces treated with metal curets displayed an evi-dent modification on smooth and rough surfaces.Many scratches and disruptions of the original ma-chined surfaces were observed in the smooth speci-mens treated with metal curets. In addition, theirregularities of the rough specimens appeared tohave smoothed out by flattening of the ridges aftertreatment with metal curets. The air-powder abrasive

Table 1.

Ra (mean – SD; mm) of Smooth and Rough Surfaces Before and After Treatments

Smooth (n = 10) Rough (n = 10)

Treatment Pretreatment Post-Treatment Pretreatment Post-Treatment

Er :YAG laser 0.18 – 0.02 0.23 – 0.06 0.70 – 0.07 0.68 – 0.06

Plastic curets 0.19 – 0.02 0.24 – 0.02 0.70 – 0.03 0.70 – 0.08

Metal curets 0.20 – 0.02 0.38 – 0.08*† 0.71 – 0.03 0.73 – 0.27

Air-powder system 0.18 – 0.02 0.20 – 0.06 0.70 – 0.07 0.69 – 0.08

* Differences between pre- and post-treatment (repeated-measures ANOVA and Tukey test; P <0.05).† Differences among groups after treatments (repeated-measures ANOVA and Tukey test; P <0.05).

Figure 1.Adherence of S. sanguinis (mean CFU/log10) to untreated and treated smooth and rough surfaces.*Differences between smooth and rough surfaces within each treatment group (two-way ANOVA andTukey test; P <0.05); †‡§differences among treated and control rough surfaces (two-way ANOVA andTukey test; P <0.05).

J Periodontol • November 2009 Duarte, Reis, Freitas, Ota-Tsuzuki

1827

system caused some sand-blastingaspect and irregular crater-like defectson the smooth surfaces. In addition,treatment with the air-powder abrasivesystem leveled down the edges of theelevations of the rough surfaces. Pow-der deposits were observed on roughsurfaces treated with the air-powderabrasive system at a higher magnifica-tion (·1,000) (Fig. 3).

Figure 4 presents the SEM photomi-crographs of the appearance of S. san-guinis adhered to control and treatedsmooth and rough titanium surfaces.In general, there was a trend towardthe formation of chains of bacterial cellson the titanium surfaces. A moderatecolonization was noted on the controland treated smooth surfaces, with theattached bacteria predominantly form-ing a monolayer compared to rough sur-faces. A more intense colonization ofS. sanguinis was observed on the roughsurfaces, which exhibited multilayers ofbacterial cells. A more sparse distribu-tion of cells was detected on the roughsurfaces treated with metal curets andthe air-powder abrasive system.

DISCUSSION

Several methods including chemical(i.e., chlorhexidine or metronidazoleapplication) and mechanical cleaning(i.e., air-powered abrasive, plastic cu-rets, and laser therapies) have beenproposed for implant debridement. Itwas demonstrated that the treatmentof implant surfaces with chlorhexidinerinse, for example, did not cause dam-age on different types of implant sur-faces, but it also did not removealready existing biofilm from such sur-faces.20 Therefore, chemical methodshave been proposed to be used in asso-ciation with mechanical cleaningmethods.7,13 Due to differences in ma-terial composition and applicationmethods, the available mechanicalcleaning instruments may differently af-fect the surface of titanium abutmentsand implants and, consequently, mayhave a direct effect on de novo biofilmformation after supportive implant ther-apy and/or treatment of peri-implant in-fections. Therefore, the first aim of thisstudy was to examine the effects of

Figure 2.Scanning electron photomicrographs of untreated (control) and treated smooth and roughsurfaces (original magnification ·200).

Bacterial Adhesion After Implant Treatment Volume 80 • Number 11

1828

different instruments used for implant and abutmentdecontamination on the morphologic characteristicsand roughness of smooth and rough titanium sur-faces. The second aim was to evaluate if the adhesionof S. sanguinis on such surfaces could be changed af-ter treatment with these different methods. The lengthof treatment period (50 seconds) was meant to simu-late multiple patient recalls, because some damage isfrequently not observed until after various applica-tions. The standardization of the force of scaling didnot seem practical because the treatments proposedvary widely in application methods. Thus, this studyproposed to simulate how the instruments would per-form on the titanium surfaces when a human hand ap-plies them freely.

Under the protocol used in the present study, theEr:YAG laser did not significantly alter the roughness(Table 1) and morphology (Fig. 2) of smooth andrough titanium surfaces, except for minor mechanicaldamage caused by the contact of the tip. The clinicalimplication of these findings is that the Er:YAG lasermay be used for the removal of biofilm from implantswithout significantly injuring their surfaces. To date,there are only a few studies11,21-24 describing theeffects of the Er:YAG laser on rough and smooth tita-nium surfaces. Differences on irradiation parameters(contact mode, water irrigation, and angle and timeof irradiation) make it difficult to compare the out-comes of the different studies.11,21-24 In agreementwithour results, Schwarz et al.22 didnot findalterationsin machine-polished and sand-blasted and acid-etched implant surfaces after treatment with anEr:YAG laser (85 mJ/pulse and 12.7 J/ cm2). Also,Kreisler et al.11,23 observed that laser treatment at 60or 120 mJ/pulse of sand-blasted and acid-etchedsurfaces did not result in any visible microscopicchanges, even without water cooling. Visible alterations

in different types of implant surfaces were reported byKreisler et al.24 using Er:YAG laser irradiation at pulseenergies over 120 mJ and by Matsuyama et al.21 whentitanium surfaces were irradiated with an Er:YAG laserat 100 and 200 mJ/pulse, corresponding to an outputenergy density of 35.4 and 70.8 J/cm2, respectively.

Similar to Er:YAG laser treatment, plastic curet in-strumentation caused no or minimal changes on thesmooth and rough titanium surfaces. These data arein accordance with previous studies18,20,25,26 thatalso showed no marked alterations after treatmentof smooth and rough surfaces with plastic curets. Ifthe preservation of surface integrity is the primary ob-jective, the plastic curet may be one of the instrumentsof choice for implant biofilm debridement; neverthe-less, the ability of this tool to effectively remove calcu-lus and biofilm from smooth and rough surfaces hasbeen widely questioned.20

The results of the present study demonstrate thatthe instrumentation with metal curets producedrougher textures compared to the other instruments(Table 1) and yielded definite damage on machinedtitanium surfaces (Fig. 2). The scratching and im-proper modification of the smooth surfaces of abut-ments and implants by metal instruments were alsoobserved in previous studies.16,25-27 Concerningrough surfaces, the roughness analysis did not revealany difference between pre- and post-treatment withmetal curets (Table 1). However, SEM picturesshowed an evident difference between metal curet-treated and control surfaces (Fig. 2). Metal curets flat-tened out the rough surfaces, removed the edges ofthe irregularities, and formed surfaces with muchlower visible levels of porosity, as previously demon-strated by Ruhling et al.18

Although the SEM photomicrographs demon-strated some titanium surface changes after treatment

Figure 3.Scanning electronphotomicrographs of anuntreated rough surface (control) (A) anda rough surface treatedwith the air-powder abrasive system (B) (originalmagnification ·1,000). Note the presence of sodium carbonate deposits in the treated rough surface compared to the control.

J Periodontol • November 2009 Duarte, Reis, Freitas, Ota-Tsuzuki

1829

with the air-powder system, a quantitative roughnessalteration was not detected by the profilometer forsmooth and rough surfaces. In agreement with ourSEM results, Meschenmoser et al.27 also observedsome small craters caused by salt crystals hitting

the smooth titanium surface aftertreatment with an air-powder abrasivesystem. In addition, Parham et al.15

demonstrated that the use of an air-powder abrasive system resulted inrounding of the angles and edges ofrough titanium surfaces and occasionalsurface pitting. More recently, Kreisleret al.23 noted that an air-powder abra-sive system led to changes of roughimplant surfaces, consisting in thereduction of the edges of the elevations.Considering that this device has beenwidely recommended for biofilm re-moval around dental implants, it isimportant to note that the surfacechanges may be affected by the hard-ness of the titanium, time exposure,air pressure, size and hardness of theabrasive particles, and distance and an-gulation of the tip.28 In addition, im-proper use of an air-powder abrasivesystem may result in subcutaneous em-physema due to the presence of air inthe interstices of connective tissue.29,30

As earlier demonstrated by invivo31,32 and in vitro studies,33 in thepresent study, untreated rough surfacespresented higher levels of S. sanguinisadhesion than untreated smooth ones.The microbial adhesion to biomaterialshas been related to surface-free energy,chemical composition, and surfaceroughness,31-34 which enhance themicrobial retention within surface ir-regularities. An Ra ;0.2 mm has beensuggested as a threshold surface rough-ness value below which no further sig-nificant changes in the total amount ofadhering bacteria can be observeddue to the larger size of most bacte-ria.31,35 Therefore, instruments usedto decontaminate implants and abut-ments should not change the surfaceto make it more biofilm retentive, butthey should attempt to minimize denovo biofilm formation. In the presentstudy, smooth surfaces treated with allof the tested instruments demonstratedthe same level of S. sanguinis adhesionas untreated control surfaces, besides

the presence of different surface profiles at the SEMlevel (Fig. 2) and increased roughness after treatmentwith metal curets (Table 1). In agreement with theseresults, a previous in vivo study17 demonstrated thatthe plaque accumulation on abutments was very

Figure 4.Scanning electron photomicrographs of untreated (control) and treated smooth and roughsurfaces after incubation of a suspension of S. sanguinis (original magnification ·10,000).

Bacterial Adhesion After Implant Treatment Volume 80 • Number 11

1830

similar after treatment with plastic curets and the air-powder abrasive system, despite some differences inthe surface texture between groups. Based on our re-sults, one could speculate that, clinically, the type ofinstrument used does not play an important role onde novo biofilm formation on smooth surfaces. How-ever, at this stage, caution should be used becausethe comparison of the level of biofilm formation onsmooth titanium surfaces modified by these mechan-ical treatments has not been clinically evaluated.

Even though roughness was not different among allof the treated rough surfaces, the levels of S. sanguinisadhesion were lower on rough surfaces treated withmetal curets and the air-powder abrasive system,achieving a level similar to those presented by thesmooth surfaces treated with these instruments.One possible explanation for the reduced bacterial ad-hesion on the rough surfaces treated with a metal cu-ret may be the texture produced by this instrument,characterized by flattening of the edges of the surfaceelevations (Fig. 2). The unexpected finding regardingthe low bacterial adhesion rate on the surfaces treatedwith the air-powder abrasive system could be ex-plained by the presence of deposits of sodium car-bonate in the irregularities as observed at a highermagnification (Fig. 3). Therefore, it appears that thelevel of bacterial adhesion in the present study wasnot only an effect of the surface roughness but alsothe presence and nature of surface contaminants aftertreatment. However, caution must be used in the inter-pretation of this finding as an advantage because thepresence of sodium carbonate deposits in the surfaceirregularities might have other unknown biologic im-plications, such as cytotoxicity and impairment of fi-broblast and osteoblast adhesion.

One can argue that this study only observed the ef-fect of titanium surface treatments on the adhesionof one bacterial species, which is associated withhealthy implant sites. However, Streptococci, espe-cially S. sanguinis, are considered significant earlycolonizers that facilitate the attachment of organismsnormally incapable of binding to host surfaces andcan ultimately lead to the development of a biofilmcommunity.36-38 Many secondary colonizers, whichadhere to bacteria already in the biofilm mass, are wellrecognized to be involved in peri-implant diseases(e.g., Fusobacterium, Capnocytophaga, Porphyro-monas, and Prevotella species).8

CONCLUSIONS

Metal curets produced higher levels of damage onsmooth titanium surfaces, whereas all instrumentsproduced the same quantitative level of roughnesson rough surfaces. Therefore, considering surfaceroughness, metal curets are not recommended forthe treatment of smooth surfaces, whereas all tested

instruments are suitable for the debridement of roughsurfaces. S. sanguinis adhesion was lower on roughsurfaces treated with metal curets and the air-powderabrasive system, probably because of texture modifi-cations and the presence of abrasive deposits. How-ever, before the clinical recommendation of thesemodalities, further studies are necessary. This inves-tigation was performed strictly in vitro, and the effectof surface changes on biofilm accumulation in vivoneeds to be further studied. In addition, treatmentsthat produce little surface damage also need to betested according to their cleaning efficacy and impacton attachment of fibroblastic and osteoblast cellsduring the tissue repairing process.

ACKNOWLEDGMENTS

The authors were supported by the Sao Paulo StateResearch Support Foundation (grants 97/10823-0,05/02561-3, and 04/01175-0). The authors greatlyappreciate the assistance of AS Technology, SaoJose dos Campos, SP, Brazil, for supplying thetitanium disks. We also thank Dr. Elliot W. Kitajima(Department of Phytopathology, University of SaoPaulo, Piracicaba, SP, Brazil) for technical SEM sup-port. The authors report no conflicts of interestrelated to this study.

REFERENCES1. Esposito M, Murray-Curtis L, Grusovin MG, Coulthard

P, Worthington HV. Interventions for replacing missingteeth: Different types of dental implants. CochraneDatabase Syst Rev 2007;(4):CD003815.

2. Esposito M, Hirsch JM, Lekholm U, Thomsen P.Biological factors contributing to failures of osseointe-grated oral implants. (II). Etiopathogenesis. Eur J OralSci 1998;106:721-764.

3. Berglundh T, Persson L, Klinge B. A systematic reviewof the incidence of biological and technical complica-tions in implant dentistry reported in prospectivelongitudinal studies of at least 5 years. J Clin Peri-odontol 2002;29(Suppl. 3):197-212.

4. Klinge B, Gustafsson A, Berglundh T. A systematicreview of the effect of anti-infective therapy in thetreatment of peri-implantitis. J Clin Periodontol 2002;29(Suppl. 3):213-225.

5. Renvert S, Roos-Jansaker AM, Claffey N. Non-surgicaltreatment of peri-implant mucositis and peri-imp-lantitis: A literature review. J Clin Periodontol 2008;35(Suppl. 8):305-315.

6. Roos-Jansaker AM, Renvert S, Egelberg J. Treatmentof peri-implant infections: A literature review. J ClinPeriodontol 2003;30:467-485.

7. Schwarz F, Sculean A, Romanos G, et al. Influence ofdifferent treatment approaches on the removal of earlyplaque biofilms and the viability of SAOS2 osteoblastsgrown on titanium implants. Clin Oral Investig 2005;9:111-117.

8. Maximo MB, de Mendoncxa AC, Renata Santos V,Figueiredo LC, Feres M, Duarte PM. Short-term clinicaland microbiological evaluations of peri-implant dis-eases before and after mechanical anti-infective ther-apies. Clin Oral Implants Res 2009;20:99-108.

J Periodontol • November 2009 Duarte, Reis, Freitas, Ota-Tsuzuki

1831

9. Claffey N, Clarke E, Polyzois I, Renvert S. Surgicaltreatment of peri-implantitis. J Clin Periodontol2008;35(Suppl. 8):316-332.

10. Sculean A, Schwarz F, Becker J. Anti-infective ther-apy with an Er:YAG laser: Influence on peri-implanthealing. Expert Rev Med Devices 2005;2:267-276.

11. Kreisler M, Kohnen W, Marinello C, et al. Bactericidaleffect of the Er:YAG laser on dental implant surfaces:An in vitro study. J Periodontol 2002;73:1292-1298.

12. Schwarz F, Bieling K, Bonsmann M, Latz T, Becker J.Nonsurgical treatment of moderate and advancedperiimplantitis lesions: A controlled clinical study. ClinOral Investig 2006;10:279-288.

13. Schwarz F, Jepsen S, Herten M, Sager M, Rothamel D,Becker J. Influence of different treatment approacheson non-submerged and submerged healing of ligatureinduced peri-implantitis lesions: An experimentalstudy in dogs. J Clin Periodontol 2006;33:584-595.

14. Takasaki AA, Aoki A, Mizutani K, Kikuchi S, Oda S,Ishikawa I. Er:YAG laser therapy for peri-implantinfection: A histological study. Lasers Med Sci 2007;22:143-157.

15. Parham PL Jr., Cobb CM, French AA, Love JW, DriskoCL, Killoy WJ. Effects of an air-powder abrasivesystem on plasma-sprayed titanium implant surfaces:An in vitro evaluation. J Oral Implantol 1989;15:78-86.

16. Homiak AW, Cook PA, DeBoer J. Effect of hygieneinstrumentation on titanium abutments: A scanningelectron microscopy study. J Prosthet Dent 1992;67:364-369.

17. McCollum J, O’Neal RB, Brennan WA, Van Dyke TE,Horner JA. The effect of titanium implant abutmentsurface irregularities on plaque accumulation in vivo.J Periodontol 1992;63:802-805.

18. Ruhling A, Kocher T, Kreusch J, Plagmann HC.Treatment of subgingival implant surfaces with Tef-lon-coated sonic and ultrasonic scaler tips and variousimplant curettes. An in vitro study. Clin Oral ImplantsRes 1994;5:19-29.

19. Dennison DK, Huerzeler MB, Quinones C, CaffesseRG. Contaminated implant surfaces: An in vitro com-parison of implant surface coating and treatmentmodalities for decontamination. J Periodontol 1994;65:942-948.

20. Augthun M, Tinschert J, Huber A. In vitro studies onthe effect of cleaning methods on different implantsurfaces. J Periodontol 1998;69:857-864.

21. Matsuyama T, Aoki A, Oda S, Yoneyama T, IshikawaI. Effects of the Er:YAG laser irradiation on titaniumimplant materials and contaminated implant abutmentsurfaces. J Clin Laser Med Surg 2003;21:7-17.

22. SchwarzF,RothamelD,SculeanA,GeorgT,ScherbaumW, Becker J. Effects of an Er:YAG laser and the Vectorultrasonic system on the biocompatibility of titaniumimplants in cultures of human osteoblast-like cells. ClinOral Implants Res 2003;14:784-792.

23. Kreisler M, Kohnen W, Christoffers AB, et al. In vitroevaluation of the biocompatibility of contaminatedimplant surfaces treated with an Er:YAG laser and anair powder system. Clin Oral Implants Res 2005;16:36-43.

24. Kreisler M, Gotz H, Duschner H. Effect of Nd:YAG,Ho:YAG, Er:YAG, CO2, and GaAIAs laser irradiationon surface properties of endosseous dental implants.Int J Oral Maxillofac Implants 2002;17:202-211.

25. Hallmon WW, Waldrop TC, Meffert RM, Wade BW. Acomparative study of the effects of metallic, non-metallic, and sonic instrumentation on titanium abut-ment surfaces. Int J Oral Maxillofac Implants 1996;11:96-100.

26. Mengel R, Buns CE, Mengel C, Flores-de-Jacoby L. Anin vitro study of the treatment of implant surfaces withdifferent instruments. Int J Oral Maxillofac Implants1998;13:91-96.

27. Meschenmoser A, d’Hoedt B, Meyle J, et al. Effects ofvarious hygiene procedures on the surface character-istics of titanium abutments. J Periodontol 1996;67:229-235.

28. Chairay JP, Boulekbache H, Jean A, Soyer A, BouchardP. Scanning electron microscopic evaluation of theeffects of an air-abrasive system on dental implants: Acomparative in vitro study between machined andplasma-sprayed titanium surfaces. J Periodontol 1997;68:1215-1222.

29. Van de Velde E, Thielens P, Schautteet H, VancloosterR. Subcutaneous emphysema of the oral floor duringcleaning of a bridge fixed on an IMZ implant. Casereport. Rev Belge Med Dent 1991;46:64-71.

30. Uehara M, Okumura T, Asahina I. Subcutaneouscervical emphysema induced by a dental air syringe:A case report. Int Dent J 2007;57:286-288.

31. Quirynen M, van der Mei HC, Bollen CM, et al. An invivo study of the influence of the surface roughness ofimplants on the microbiology of supra- and subgin-gival plaque. J Dent Res 1993;72:1304-1309.

32. Rimondini L, Fare S, Brambilla E, et al. The effect ofsurface roughness on early in vivo plaque colonizationon titanium. J Periodontol 1997;68:556-562.

33. Wu-Yuan CD, Eganhouse KJ, Keller JC, Walters KS.Oral bacterial attachment to titanium surfaces: Ascanning electron microscopy study. J Oral Implantol1995;21:207-213.

34. Yoshinari M, Oda Y, Kato T, Okuda K, Hirayama A.Influence of surface modifications to titanium on oralbacterial adhesion in vitro. J Biomed Mater Res 2000;52:388-394.

35. Bollen CM, Papaioanno W, Van Eldere J, Schepers E,Quirynen M, van Steenberghe D. The influence ofabutment surface roughness on plaque accumulationand peri-implant mucositis. Clin Oral Implants Res1996;7:201-211.

36. Kolenbrander PE, London J. Adhere today, heretomorrow: Oral bacterial adherence. J Bacteriol1993;175:3247-3252.

37. Ruhl S, Sandberg AL, Cisar JO. Salivary receptors forthe proline-rich protein-binding and lectin-like adhe-sins of oral actinomyces and streptococci. J Dent Res2004;83:505-510.

38. KolenbranderPE,PalmerRJJr.,RickardAH,JakubovicsNS, Chalmers NI, Diaz PI. Bacterial interactions andsuccessions during plaque development. Periodontol2000 2006;42:47-79.

Correspondence: Dr. Poliana Mendes Duarte, DentalResearch Division, Guarulhos University, Rua Dr. NiloPecxanha, 81. Predio U. 6o Andar - Centro, Guarulhos07011-040, SP, Brazil. Fax: 55-11-2464-1758; e-mail:pduarte@ung.br.

Submitted May 13, 2009; accepted for publication June17, 2009.

Bacterial Adhesion After Implant Treatment Volume 80 • Number 11

1832