Lethal photosensitization and guided bone regeneration in treatment of peri-implantitis: an experimental study in dogs

Transcription

1 Jamil Awad Shibli Marilia Compagnoni Martins Fernando Salimon Ribeiro Valdir Gouveia Garcia FranciscoHumbertoNocitiJr Elcio Marcantonio Jr Lethal photosensitization and guided bone regeneration in treatment of peri-implantitis: an experimental study in dogs Authors affiliations: Jamil Awad Shibli, Department of Periodontology, Dental Research Division, Guarulhos University, Guarulhos, SP, Brazil Marilia Compagnoni Martins, Fernando Salimon Ribeiro, Elcio Marcantonio Jr, Department of Periodontology, Dental School of Araraquara, State University of Sao Paulo (UNESP), Araraquara, SP, Brazil Valdir Gouveia Garcia, Dental School of Marília, University of Marília (UNIMAR), Marília, SP, Brazil;DentalSchoolofAraçatuba, State University of Sao Paulo (UNESP), Araçatuba, SP, Brazil Francisco Humberto Nociti Jr, Department of Periodontology, Dental School of Piracicaba, University of Campinas, (UNICAMP) SP, Brazil Correspondence to: Elico Marcantonio Jr Departmento de Periodontia Faculdade de Odontologia de Araraquara-UNESP R. Humaitá, Araraquara, SP Brazil Tel.: þ Fax: þ Key words: guided bone regeneration, histology, peri-implantitis, photodynamic therapy/ photosensitizers, re-osseointegration Abstract: The purpose of this study was to evaluate the effect of lethal photosensitization and guided bone regeneration (GBR) on the treatment of ligature-induced peri-implantitis in different implant surfaces. The treatment outcome was evaluated by clinical and histometric methods. A total of 40 dental implants with four different surface coatings (10 commercially pure titanium surface (cpti); 10 titanium plasma-sprayed (TPS); 10 acid-etched surface; 10 surface-oxide sandblasted) were inserted into five mongrel dogs. After 3 months, the animals with ligature-induced peri-implantitis were subjected to surgical treatment using a split-mouth design. The controls were treated by debridment and GBR, while the test side received an additional therapy with photosensitization, using a GaAlAs diode laser, with a wavelength of 830 nm and a power output of 50 mw for 80 s (4 J/cm 2 ), and sensitized toluidine blue O (100 mg/ml). The animals were sacrificed 5 months after therapy. The control sites presented an earlier exposition of the membranes on all coating surfaces, while the test group presented a higher bone height gain. Re-osseointegration ranged between 41.9% for the cpti surface and 31.19% for the TPS surface in the test sites; however differences were not achieved between the surfaces. The lethal photosensitization associated with GBR allowed for better re-osseointegration at the area adjacent to the periimplant defect regardless of the implant surface. Date: Accepted 15 January 2005 To cite this article: Shibli JA, Martins MC, Ribeiro FS, Garcia VG, Nociti Jr FH, Marcantonio Jr E. Lethal photosensitization and guided bone regeneration in treatment of periimplantitis: an experimental study in dogs. Clin. Oral Impl. Res. 17, 2006; doi: /j x Copyright r Blackwell Munksgaard 2006 Animal studies have shown that bacterial biofilm accumulation around dental implants promoted by ligature placement may develop peri-implant tissue breakdown, also known, as peri-implantitis (Lang et al. 1993; Schou et al. 1993; Ericsson et al. 1996). Plaque-induced peri-implantitis has been indicated as one of the etiological factors associated with long-term failure of dental implants (Quirynen et al. 2002). Different implant surfaces have been proposed in order to improve clinical and histological outcome of dental implants and, therefore, the interaction between different implant surfaces and peri-implantitis therapy may be of interest. Recent studies (Persson et al. 2001b; Kolonidis et al. 2003; Schou et al. 2003a; Shibli et al. 2003b, 2003c) have investigated the importance of implant surface cleanliness after peri-implantitis treatment. It has been hypothesized that surface contaminants, released from contaminated implant surfaces, enhance and perpetuate the inflammatory response, thus altering the healing process, and possibly leading to the dissolution of titanium (Baier et al. 1984, 1988; Olefjord & Hansson 1993; Esposito et al. 1999; Shibli et al. 2004). In addition, the alterations of the titanium oxide layer may not allow re-osseointegration. In earlier studies, Shibli et al. (2003b, 273

2 2003c) demonstrated the potential use of photodynamic therapy associated with guided bone regeneration (GBR) in the treatment of experimental peri-implant lesions. The average re-osseointegration obtained in this study was 25.15%, similar to previous studies that have used mechanical treatment associated with systemic antibiotics (Persson et al. 1999; Wetzel et al. 1999; Nociti et al. 2001a; Schou et al. 2003a, 2003c). Photodynamic therapy uses a lowpower laser, following the application of a photosensitizing substance, such as toluidine blue O (TBO). The mechanism by which TBO kills microorganisms, such as Porphyromonas gingivalis, Prevotella intermedia, Actinobacillus actinomycetemcomitans, andfusobacterium nucleatum, is currently under research. However, it is believed that lethal photosensitization of these microorganisms may involve changes in their membranes, on the plasma membrane proteins, and DNA damage mediated by singlet oxygen (Dobson et al. 1992; Sarkar & Wilson 1993; Henry et al. 1995; Bhatti et al. 1997, 1998; Wainwright 1998; Bhatti et al. 2000). The objective of this study was to evaluate the efficacy of lethal photosensitization associated with GBR, for the treatment of ligature-induced peri-implantitis in dogs, using different dental implantcoated surfaces. Material and methods Extraction Implant Abutment +Lig. Lig. Animals and anesthesia The outline of the experiment is presented in Fig. 1. Briefly, five healthy male mongrel dogs, with an average age of 2 years and an average weight of 18 kg, were used in the study. Animal selection, management, and surgical protocol followed routines approved for this study by the local Institutional Animal Care and Ethics Committee. All surgical and clinical procedures, as well as laser irradiation, were performed under general anesthesia accomplished with 0.05 mg/kg of subcutaneous preanesthesia sedation (Atropine sulfate 0.5 mg, Ariston Inds. Química e Farms. LTDA, São Paulo, Brazil), intravenous injection of chlorpromazine (Amplictil 25 mg, Rhodia Farma LTDA, São Paulo, Brazil), and thiopental (Tiopental AB- BOTT Laboratórios do Brasil Ltda, São Paulo, Brazil). Tooth extraction An edentulous ridge was created by the extraction of all mandibular premolars and first molars. The alveoli were allowed to heal for a period of 3 months. The upper premolars were also extracted to prevent occlusion trauma interference during plaque-induced peri-implantitis. During the healing period, bacterial biofilm control was performed by scrubbing with 0.12% chlorhexidine daily, in addition to scaling and root planning once a month, until cotton ligatures were placed. Implant design, surface, and surgery Forty dental implants of four different surfaces were used: 10 commercially pure titanium implants (cpti; Sterngold, Implamed, Attleboro, MA, USA); 10 titanium Treatment % chlorhexidine Membrane removal Biopsy 10 months 0.12% chlorhexidine Fig. 1. Outline of the experiment. Animals n ¼ 5, dental implants n ¼ 40. Ligatures were placed at 0 month ( þ Lig.) and removed at 3 months ( Lig.). plasma-sprayed (TPS; Sterngold, Implamed); 10 hybrid surfaces machined titanium in the three first screws and acid etched in other screws acid (Osseotite s - 3i s Implants Innovations, Palm Beach Gardens, FL, USA), and 10 sandblasted with titaniun oxide oxide (Porous, Conexão Implants, São Paulo, SP, Brazil). All implants were 10 mm in length and 3.75 mm in diameter. The implants were randomly distributed among the dogs, so that each dental implant surface could be represented on each mandibular side. Dental implants were placed after full thickness flap surgery under aseptic conditions. The recipient sites were prepared according to the surgical techniques indicated by each implant manufacturer. The flaps were sutured with single interrupted sutures, submerging all implants. Potassic and sodic benzilpenicilin (Fort Dodge Saúde Animal LTDA, Campinas, SP, Brazil) was given once a week for 2 weeks, in order to prevent postsurgical infection, pain was controlled with paracetamol (ABBOTT Laboratórios do Brasil Ltda), and the sutures were removed after 10 days. Experimental peri-implantitis Three months after dental implant installation, healing abutments were installed, according to the instructions of each dental implant system. After 2 months of a plaque control program and the healing of the soft tissue, cotton floss ligatures were placed around the dental implants, as previously described (Shibli et al. 2003a). At 90 days, when approximately 40% of the initial bone support was lost, the ligatures were removed, and a 2-month plaque control program was initiated by daily scrubbing of the implants with 0.12% chlorhexidine and scaling of the abutment surface. Randomization of the mandibular quadrants and a split-mouth design was established as follows: control side mechanical debridment and GBR, and test side mechanical debridment, lethal photosensitization, and GBR to compare the effect of lethal photosensitization (Table 1). Clinical evaluation and treatment A crestal incision was made through the mucosa, and buccal and lingual full-thickness flaps surgery was performed to remove 274 Clin. Oral Impl. Res. 17, 2006 /

3 Table 1. Distribution of dental implants with different surfaces in five dogs Animal Test side Control side PM2 PM3 PM4 M PM2 PM3 PM4 M 1 cpti TPS Oxide Acid Acid Oxide TPS cpti 2 TPS Oxide Acid cpti Oxide TPS cpti Acid 3 Oxide cpti TPS Acid TPS cpti Acid Oxide 4 cpti Acid Oxide TPS Acid TPS Oxide cpti 5 Acid TPS CpTi Oxide cpti Acid Oxide TPS PM2, PM3, PM4, and M, mandibular premolars and molar; cpti, commercially pure titanium; TPS, titanium plasma-sprayed; Oxide, surface sandblasted with titanium oxide; Acid, acid-etched surface. the abutments and the granulation tissue present in bone craters around the dental implants. A plastic scaller (Implacare IMPHDL6, Hu-friedy Mfg Co Inc., Chicago, IL, USA) was used to prevent any damage to exposed dental implant surface. Pre- and post-treatment bone defects were measured by a single trained examiner, i.e., the distance from the top of the cover screw to the bottom of the peri-implant defect, using a periodontal probe (PCPUNC 15 Hu-friedy Mfg Co Inc.), at four sites (mesial, buccal, distal, and lingual). The implant surfaces of the test group received 100 mg/ml TBO (Sigma LTDA, Poole, UK) injected into the peri-implant defect, as far as the bony edge, on the previously contaminated implant surface, using a thin needle as previously described (Shibli et al. 2003b, 2003c). TBO was placed for 1 min, and then carefully removed. The stained area was subsequently irradiated with a GaAlAs diode laser (Thera-Lase, DMC Equipamentos LTDA, São Carlos, SP, Brazil) with a measured power output of 50 mw, to emit radiation in collimated beams (1 cm 2 )withawavelength of 830 nm, for 80 s, and a total energy of 4 J (energy density of 4 J/cm 2 ). The diode laser was in contact with the mesial, distal, buccal, and lingual surfaces by a scanning method, for 20 s, on each surface. Multiple perforations of the bone surface were previously made to facilitate the filling of the defect with the coagulum. All the sites, including control sides, received a PTFE membrane (TefGen- PLUS, Lifecore Biomedical Inc., Chasca, MN, USA), allowing implant penetration through the site. The membrane was adjusted to extend circumferentially 3 5 mm over the adjacent alveolar bone, avoiding ingrowths of the soft connective tissue. The membranes were stabilized at the buccal and lingual sites not only by cpti tacks (INP Implantes Nacionais and Próteses Comércio Ltda, São Paulo, SP, Brazil) but also by cover screws. To allow flap apposition and closure after placement, incisions were made buccally and lingually after membrane placement. Primary wound closure was achieved with horizontal mattress sutures alternated with interrupted sutures. Anti-inflammatory medication (2 mg betamethasone Celestone, Schering-Plough S/A, Rio de Janeiro, RJ, Brazil) was administered twice a day and appropriate analgesia (paracetamol) for 3 days following surgery in order to reduce postoperative swelling and pain. Antibiotics were administered neither before nor after the surgical treatment of the peri-implant defect. Two weeks after local therapy, the sutures were removed and a fluorochrome (Oxytetracycline 25 mg/kg body weight; Pfizer do Brasil, Sao Paulo, SP, Brazil) was injected intravenously (i.v.). Four months after treatment, the PTFE membranes were surgically removed in both groups and the implants were allowed to heal for 1 month. Four days before the euthanasia, a second fluorochrome (18 mg/ kg red alizarine body weight; Sigma Chemical Co., St Louis, MO, USA) was injected i.v. for observation of the remodelation of peri-implant bone tissue at 5 months after treatment. Five months after treatment, the animals were sacrificed by induction of deep anesthesia followed by intravenous sodium pentobarbital euthanasia. Histological procedures The mandibles were removed and block biopsies of each implant site were dissected and fixed in 4% neutral formalin for 48 h. The biopsies were then prepared for ground sectioning according to the methods previously described by Donath & Breuner (1982). The specimens were cut in a mesio-distal plane using a cutting grinding unit (Exact s Cutting, System, Apparatebau Gmbh, Hamburg, Germany). One central section of each biopsy was prepared and reduced to a final thickness of mm, by micro-grinding and polishing using a micro-grinding unit (Exact s Cutting, System, Apparatebau Gmbh). Before staining, each section was evaluated with respect to the location of the fluorochrome marker by microscopy (Leitz DM-RBE microscopy, Leica, Bensheim, Germany) equipped with an image system (Qwin, Leica). In the unstained sections, fluorescence light and a filter cube compatible with the fluorochrome were used to assess the bone defect border, as well as the remodeling of the new bone. The sections were stained in toluidine blue in order to assess the histometric parameters: (1) Distance from the original bottom of the defect identified by the difference in coloration after staining and by the fluorochrome (a) to the most coronal point of the newly formed bone with intimate contact with the implant surface, (b) ( ¼ re-osseointegation); (2) area of (a) to the most apical border of the newly formed bone, (c) to implant shoulder, (d) ( ¼ bone fill); (3) percentage of osseointegration (mineralized bone contact with the implant surface); and (4) bone area within the limits of the implant threads at the portion of the implant, apical of the peri-implant defect, where peri-implantitis did not occur (Fig. 2). Data analysis Data were obtained in pixels and pixels 2 and transformed into percentage to prevent the possible influence of the different macrostructure of the different implant 275 Clin. Oral Impl. Res. 17, 2006 /

4 systems evaluated in this study. The data relative to defect depth, percentage of osseointegration, and bone area within the Fig. 2. Schematic drawing illustrating the landmarks used for the histometric measurements. limits of the threads were obtained from all implants, independent of group. Paired t-test and parametric analysis of variance were performed to assess the differences between test and control groups and among the several implant surfaces, respectively. The unit of analysis was the animal (n ¼ 5), and the level of significance was Results Clinical observations One cpti (test group) and one oxide surface (control group) were lost during the experimental peri-implantitis phase in two animals. All the other remaining implants were included in the evaluation. The dental implants of the test group successfully integrated to the bone and survived the subsequent periods of treatment. However, 89.47% of the control sites depicted earlier exposition of the membranes when compared with the 15.79% of the test sites (Tables 2 and 3). Implants from the control group showed several exposed threads, while some dental implants from the test group appeared to be covered by new bone (Figs 3a c). When the effect of the treatment was compared between the test and the control group independent of the implant surface, the test group presented a higher bone gain (Table 6). Histological examinations and measurements The peri-implant soft and hard tissues from all treatment groups generally appeared healthy, with the junctional epithelium Table 2. Overview of clinical measurements (mm) for the test and the control group pre- and post-treatment as well as membrane exposure per animal Animal cpti TPS Acid Oxide Test Control Test Control Test Control Test Control Pre, post Pre, post Pre, post Pre, post Pre, post Pre, post Pre, post Pre, post 1 1, 0 5.5, 5w 3, , 3.75z 2.25, 0 5, 5z 2.25, 0 2.5, 3.25w 2 Failed implant 3, , , 4z 1.25, , 0.5y 2.5, , 3.2 n , , 2.5w 2.75, , 4.75 n 2.5, , 3.75w 3.25, 3.25w Failed implant 4 3, 0 3, 3.1w 4.5, 4 3, , , 2.25w 3, , 2.25w 5 3, , 4 n 7, 3y 1, 3 n 3.25, 2z 3-3w 3, , 6w Mean 2.43, , , , , , , , 3.79 SD 0.96, , , , , , , , 1.42 P NS NS 0.07 NS 0.96 NS S 0.37 NS S NS Calculated values are based on the mean value of four sites (mesial, buccal, distal, and lingual). n Exposure of membrane before the first post-operative month. wexposure of membrane between the first and second post-operative month. zexposure of membrane between the second and third post-operative month. yexposure of membrane after the third post-operative month. s Statistically significant. NS, not significant; cpti, commercially pure titanium; TPS, titanium plasma-sprayed; Oxide, surface sandblasted with titanium oxide; Acid, acid etched surface. Table 3. Clinical height measurements (mm SD) of the peri-implant bone defects for test and control groups Test side Control side cpti n TPS Oxide Acid cpti TPS Oxide n Acid Pretreatment Posttreatment Bone height w,z 2 0.5w,z y,z y,z w w w w gain (mm)k Bone gain (%)k z z z z Calculated values are based on the mean value for four sites (mesial, buccal, distal, and lingual). n One implant was lost at ligature phase (n ¼ 4 implants). wnot significant between pre- and post-treatment (P40.05). znot significant between groups test and control (P40.05). ystatistically significant between pre- and post-treatment (P 0.03). zstatistically significant between groups test and control (P 0.02). knot significant among the different implant surfaces (P 0.711). cpti, commercially pure titanium; TPS, titanium plasma-sprayed; Oxide, surface sandblasted with titanium oxide; Acid, acid etched surface. 276 Clin. Oral Impl. Res. 17, 2006 /

5 Fig. 3. (a) Peri-implant defect adjacent to commercially pure titanium (left) and acid surface (right) after mechanical debridment; (b) clinical view of the same defect 4 months after lethal photosensitization and guided bone regeneration. Note the contour of the newly formed bone (arrows) on the peri-implant defect; (c) clinical view of peri-implant defects from the control group covered by fibrous tissue after membrane removal. The membrane exposure occurred after 1 month post-treatment. extending a short distance apical to the implant shoulder. The connective tissue seemed to be compromised primarily by dense collagenous fibers, which ran parallel to the previously contaminated implant surface. The old bone was mostly lamellar and compact, and numerous osteocytes were presented in their lacunae, while the newly formed bone exhibited different stages of maturation and remodelation, mainly in the test group. The biopsies from the sites where the membrane had been exposed showed very few spicules of bone at the base of the connective tissue cap. In some cases, osteoblasts were connected to newly formed bone, indicating ongoing bone formation, and minor apposition of new bone could be found, specifically at the apical regions of the originally induced defects; however, thebonetissueappearedimmature.the bone marker projected in a lateral coronal direction and separated a triangular-shaped portion of newly formed regenerated bone from the old bone (Figs 4a c). Only the TPS implants presented surface debris or particle inclusions in the surrounding tissue at the bone area, within the limits of the implant threads at the apical portion of the peri-implant defect, where peri-implantitis did not occur (Figs 5a and b). The defect depth, the percentage of mineralized bone contact with the dental implant surface, and the bone area within the limits of the threads are presented in Fig. 6. The highest percentage of osseointegration was 73.58% of the TPS surface and the lowest was 61.48% of the cpti surface (P ¼ 0.029). Bone filling was statistically significant between the test and control groups (P0.036) (Tables 4 and 6). In some specimens, the lateral aspect of the coronal part of the dental implant was covered by a dense connective tissue capsule separating the newly formed bone from the dental implant surface (Fig. 5b). The mean percentage of re-osseointegration was 31 41% for the test group and 0 14% for the control group (Tables 5 and 6). In some specimens in the control group, there was no contact between new bone and the previously contaminated implant surface (Figs 7a and b). Discussion The difficulties in obtaining re-osseointegration after treatment of peri-implantitis have been documented in several animal studies (Grunder et al. 1993; Ericsson et al. 1996; Persson et al. 1996; Hanisch et al. 1997; Wetzel et al. 1999; Shibli et al. 2003b). Recently, Kolonidis et al. (2003) showed that re-osseointegration occurred on a surface previously contaminated with bacterial biofilm. However, the animal model used by Kolonidis et al. (2003) is not comparable, either with the animal model used in our study, or with the previously cited studies (Grunder et al. 1993; Ericsson et al. 1996; Persson et al. 1996, 1999; Wetzel et al. 1999; Nociti et al. 2001a, 2001b; Persson et al. 2001a; Shibli et al. 2003b) because of lack of important factors such as peri-implant soft and hard tissue breakdown induced by ligatures, the presence of bone defect, and presence of bone debris at the implant surface. The amount of bone regeneration was significantly influenced by the photosensitization treatment. The earlier exposition of the membranes, as well as the maintenance of these barriers can jeopardize tissue regeneration (Lekholm et al. 1993; Machado et al. 1999; Haas et al. 2000). The complications because of membrane exposition in the control group may support the findings that the presence of periodontal pathogens at the dental implant fixture altered the cell responses, together with enhanced protease activities and foreign body reactions, decreasing the efficacy of GBR (Wakabayashi et al. 1996; Slots et al. 1999). The control group presented 5.6 times more earlier membrane exposition when compared with the test group. In the present investigation, the clinical evaluation pre- and post-treatment revealed a variable degree of appreciable hard-tissue fill of the peri-implant defects, mainly for the test group. The histometric analysis depicts a statistically higher amount of new bone formation for the test group. Although the percentages of bone fill observed in studies such as those by Wetzel et al. (1999) and Persson et al. (1999) were lower, our results ranged between 41% for the TPS surface and 60.87% for the acid surface, in agreement with Persson et al. (1996, 1999), although the last two studies cited utilized smooth surfaces. Re-osseointegration was achieved on all dental implant surfaces of the test group, principally at the base of the angular bony defect, in agreement with previous studies (Jovanovic et al. 1993; Singh et al. 1993; Persson et al. 1996; Shibli et al. 2003b). The higher percentage of re-osseointegration observed in our investigation was 41.90%, of the cpti surface, and the lowest was on the TPS surface (31.19%), which is in agreement with the results described by Hanisch et al. (1997) and higher when compared with the results presented by Wetzel et al. (1999) and Shibli et al. (2003b). On the other hand, Persson et al. (2001a) found % of re-osseointegration 277 Clin. Oral Impl. Res. 17, 2006 /

6 Shibli et al. Lethal photosensitization in treatment of peri-implantitis Fig. 4. (a) Ground section of commercially pure titanium implant from the test group. Fluorescence light (original magnification 200). The fluorochrome marks show the border of the bone defect (yellow) and the new bone formation (red); (b) mesio-distal ground section of the same area (toluidine blue staining, original magnification 200) showing the re-osseointegration with dental implant surface contamined previously; (c) the fluorochrome marks show the border of the bone defect (yellow) and the new bone formation (red). on the oxide surface (original magnification 200). Fig. 5. (a) Surface debris and particle inclusions of titanium plasma-sprayed (TPS) surface in the surrounding tissue at the bone area within the limits of the implant threads apical of the peri-implant defect (toluidine blue staining, original magnification 100); (b) ground section of one TPS surface from the test group (toluidine blue staining, original magnification 100) showing the borderline between the old original bone and the newly formed bone (arrows) in direct contact with the dental implant surface contamined previously (reosseointegration) (arrows heads). Note a dense connective tissue (CT) between the implant surface and newly formed bone. in sandblasted large grit acid-etched surfaces (SLA) and % for turned surfaces. The authors speculated that the SLA surface could provide a better condi- 278 Clin. Oral Impl. Res. 17, 2006 / tion for coagulum stability, facilitating the bone regeneration process. However, the different coated surfaces evaluated in our study showed lower means of re-osseointe- gration than for the cpti surface. One may speculate about the differences between the present study and the study performed by Persson et al. (2001a): (1) the treatment using systemic antibiotics may decrease the number of periodontal pathogens present in the hard and soft peri-implant tissue, and (2) the SLA surface may present specific chemical properties that facilitate re-osseointegration. The healing of the dental implant is initiated immediately after implant insertion by initial blood clot formation in the peri-implant gaps and the development of a biofilm (D Hoedt 1985; Meyer et al. 1988; Zechner et al. 2003). The roughnesses of the implant surface are an important factor in this process. In our study, there were no statistical differences for the amount of reosseointegration in different dental implant surfaces. The adsorption of microorganisms as well as the application of the TBO on the implant surface may influence the direct coagulum implant contacts and, consequently, re-osseointegration. In addition, a series of co-ordinated events, including protein adsorption, proliferation, and deposition of bone tissue, are probably affected by experimental peri-implantitis. In addition, a dense connective tissue capsule that separated the newly formed bone from the dental implant surface was

7 Table 4. Percentage of bone fill measured from the base of the defect to the implant shoulder Implant Test Control P 95% CI surface n Mean SD Range Mean SD Range cpti TPS Oxide Acid n Not significant among the surfaces P ¼ cpti, commercially pure titanium; TPS, titanium plasma-sprayed; Oxide, surface sandblasted with titanium oxide; Acid, acid etched surface. Table 5. Percentage of implant bone contact in the previously contamined implant surface (re-osseointegration) measured from the base of the defect to the implant shoulder Implant Test Control P 95% CI surface n Mean SD Range Mean SD Range cpti to TPS Oxide Acid n Not significant among the surfaces P ¼ cpti, commercially pure titanium; TPS, titanium plasma-sprayed; Oxide, surface sandblasted with titanium oxide; Acid, acid etched surface. Fig. 6. Box-plots of histometric measurements of (a) defect depth (mm) in an apical direction from the implant shoulder to the level of the old bone, (b) percentage of mineralized bone contact with the implant surface, and (c) bone area within the limits of the implant threads. Table 6. Clinic and histometric effect of the treatment by comparing the test and the control group independent of the implant surface (n ¼ 5 animals) Variable Test Control P 95% CI Mean SD Range Mean SD Range Bone height gain (mm) n Re-osseo integration (%)w to Bone fill (%)w to n Clinical variable. whistometrical variables. observed in some specimens, as previously reported (Persson et al. 1996; Wetzel et al. 1999). Despite controversy regarding the amount of re-osseointegration (Persson et al. 1996; Nociti et al. 2001a, 2001b; Schou et al. 2003a, 2003c), these different results may be attributed to different experimental designs and variables such as the ligatureinduced peri-implantitis period, material used for ligatures (cotton, silk, orthodontic elastics), microstructure utilized, cleaning methods of the contaminated implant surface and their efficiency, bony defect shape, and combination of graft materials and GBR. In addition, the use of photodynamic therapy to kill periodontal pathogens offers some advantages over the use of conventional antimicrobials: it prevents the development of resistance among target organisms to the photochemically generated free radicals thought to be responsible for bacterial killing and, unlike antiseptics and antibiotics, there would be no need to maintain high concentrations of the TBO in the peri-implant defects for long periods. The results of the present investigation support the use of GBR associated with antimicrobial local therapy for the treatment of peri-implant defects. In addition, the use of photodynamic therapy as local treatment may be a useful alternative to antibiotics for the treatment of local infection by eradication of target cells using oxygen species produced by interaction between a photosensitizing agent and light of an appropriate wavelength (Dougherty et al. 1998; Kömerik & Wilson 2002). In conclusion, data from the present study showed that the treatment of ligature-induced peri-implantitis using lethal photosensitization, associated with GBR, may achieve significant bone fill associated with re-osseointegration. The different coating surfaces evaluated in our study presented the same healing responses to peri-implantitis treatment. However, these results should be considered with caution and further investigations must be conducted. 279 Clin. Oral Impl. Res. 17, 2006 /

8 all the surgical phases. The authors are also indebted to Conexão Implants, Brazil, for providing the oxide dental implants, INP Brazil for the titanium tacks, and Implamed Brazil for cpti and TPS implants. Source of support: FAPESP (00/ ) and FUNDUNESP. Fig. 7. (a) Ground section of acid surface implant (test group) shown in Figs 3a and b. (toluidine blue staining, original magnification 25) showing the borderline (arrows) between the old original bone and the newly formed bone; (b) mesio-distal ground section of one oxide surface from the control group shown in Fig. 5c (Toluidine blue staining, original magnification 16). Note the connective tissue covering the dental implant surface and the absence of newly regenerated bone. Acknowledgements: This research was supported by FAPESP (Fundaçãode Amparo à Pesquisa do Estado de São Paulo, SP, Brazil) grant no. 00/ and FUNDUNESP (Fundação para o Desenvolvimento da UNESP, São Paulo, Brazil). The authors appreciate the assistance of Dr Vagner Samy Lemo in References Baier, R.E. & Meyer, A.E. (1988) Implant surface preparation. International Journal of Oral & Maxillofacial Implants 3: Baier, R.E., Meyer, A.E., Natiella, J.R., Natiella, R.R. & Carter, J.M. (1984) Surface properties determine bioadhesive outcomes: methods and results. Journal of Biomedical Materials Research 18: Bhatti, M., MacRobert, A., Henderson, B., Shepherd, P., Cridland, J. & Wilson, M. (2000) Antibody-target lethal photosensitization of Porphyromonas gingivalis. Antimicrobians Agents Chemotherapics 44: Bhatti, M., MacRobert, A., Meghji, S., Henderson, B. & Wilson, M. (1997) Effect of dosimetric and physiological factors on the lethal photosensitization of Porphyromonas gingivalis in vitro. Photochemistry and Photobiology 65: Bhatti, M., MacRobert, A., Meghji, S., Henderson, B. & Wilson, M. (1998) A study of the uptake of toluidine blue O by Porphyromonas gingivalis and the mechanism of lethal photosensitization. Photochemistry and Photobiology 68: D Hoedt, B. (1985) The influence of implant materials and material surfaces on blood clotting: a scanning electron microscopy study. In: van Steenburgh, D., Albrektsson, T., Bränemark, P.I., Henry, P.J., Holt, R. & Linden, C. eds. Tissue integration in oral and maxillofacial reconstruction. Proceedings of an International Congress. Current Clinical Practice Series 29: Dobson, J. & Wilson, M. (1992) Sensitization of oral bacteria in biofilms to killing by light from a low power laser. Archives of Oral Biology 37: Donath, K. & Breuner, G.A. (1982) A method for undecalcified bones and teeth with attached soft tissues. The Sage Schliff technique. Journal of Oral Pathology 11: Dougherty, T.J., Gomer, C.J., Henderson, B.W., Jori, G., Kessel, D., Korbelik, M. & Moan, J. (1998) Photodynamic therapy. Journal of the National Cancer Institute 90: Ericsson, I., Persson, L.G., Berglundh, T., Edlund, T. & Lindhe, J. (1996) The effect of antimicrobial therapy on peri-implantitis lesion. An experimental study in dog. Clinical Oral Implants Research 7: Esposito, E., Lausmaa, J., Hirsch, J.M. & Thomsen, P. (1999) Surface analysis of failed oral titanium implants. Journal of Biomedical Materials Research (Applied Biomaterials) 48: Grunder, U., Hurzeler, M.B., Schupbach, P. & Strub, J.R. (1993) Treatment of ligature-induced peri-implantitis using guided tissue regeneration: a clinical and histological study in the beagle dog. International Journal of Oral & Maxillofacial Implants 8: Hass, R., Baron, M., Dortbudak, O. & Watzek, G. (2000) Lethal photosensitization, autogenous bone, and e-ptfe membrane for the treatment of peri-implantitis: preliminary results. International Journal of Oral & Maxillofacial Implants 15: Hanisch, O., Tatakis, D.N., Boskovic, M., Rohrer, M.D. & Wikesjo, U.M.E. (1997) Bone formation and re-osseointegration in peri-implantitis defects following surgical implantation of rhbmp-2. International Journal of Oral & Maxillofacial Implants 12: Henry, C.A., Judy, M., Dyer, B., Wagner, M. & Matthews, J. (1995) Sensitivity of Porphyromonas and Prevotella species in liquid media to argon laser. Photochemistry and Photobiology 61: Jovanovic, S.A., Keney, E.B., Carranza, F.A. & Donath, K. (1993) The regenerative potencial of plaque induced peri-implant bone defects treated 280 Clin. Oral Impl. Res. 17, 2006 /

9 by submerged membrane technique: an experimental study. International Journal of Oral & Maxillofacial Implants 8: Kolonidis, S., Renvert, S., Hämmerle, C.H.F., Lang, N.P., Harris, D. & Claffey, N. (2003) Osseointegration on implant surfaces previously contamined with plaque. An experimental study in the dog. Clinical Oral Implants Research 14: Kömerik, N. & Wilson, M. (2002) Factors influencing the susceptibility of Gram-negative bacteria to toluidine blue O-mediated lethal photosensitization. Journal of Applied Microbiology 92: Lang, N.P., Bragger, U., Walther, D., Beamer, B. & Kornman, K.S. (1993) Ligature-induced peri-implant infection in cynomolgus monkeys. I. Clinical and radiographic findings. Clinical Oral Implants Research 4: Lekholm, U., Becker, W., Dahlin, C., Becker, B., Donath, K. & Morrison, E. (1993) The role of early versus late removal of GTAM membranes on bone formation at oral implants placed into immediate extraction sockets. An experimental study in dogs. Clinical Oral Implants Research 4: Machado, M.A., Stefani, C.M., Sallum, E.A., Sallum, A.W., Tramontina, V.A., Nogueira-Filho, G.R. & Nociti Jr., F.H. (2000) Treatment of ligature-induced peri-implantitis defects by regenerative procedures. Part II: a histometric study in dogs. Journal of Oral Science 42: Meyer, A., Baier, R., Natiella, J. & Meenagha, M. (1998) Investigation of tissue/implant interactions during the first two hours of implantation. Journal of Oral Implantology 14: Nociti Jr, F.H., Machado, M.A.N., Stefani, C.M. & Sallum, E.A. (2001b) Absorbable versus nonaborbable membranes and bone grafts in the treatment of ligature-induce peri-implantitis defects in dogs: a histometric investigation. International Journal of Oral & Maxillofacial Implants 16: Nociti Jr, F.H., Machado, M.A.N., Stefani, C.M., Sallum, E.A. & Sallum, A.W. (2001a) Absorbable versus nonaborbable membranes and bone grafts in the treatment of ligature-induce periimplantitis defects in dogs. Part I. A clinical investigation. Clinical Oral Implants Research 12: Olefjord, I. & Hansson, S. (1993) Surface analysis of four dental implant systems. International Journal of Oral & Maxillofacial Implants 8: Persson, L.G., Araujo, M.G., Berglundh, T., Grondahl, K. & Lindhe, J. (1999) Resolution of periimplantitis following treatment. An experimental study in the dog. Clinical Oral Implants Research 10: Persson, L.G., Berglundh, T., Lindhe, J. & Sennerby, L. (2001a) Re-osseointegration after periimplantitis at different implant surfaces. An experimental study in the dog. Clinical Oral Implants Research 12: Persson, L.G., Ericsson, I., Berglundh, T. & Lindhe, J. (1996) Guided bone regeneration in the treatment of periimplantitis. Clinical Oral Implants Research 7: Persson, L.G., Ericsson, I., Berglundh, T. & Lindhe, J. (2001b) Osseointegration following treatment of peri-implantitis and replacement of implant components. An experimental study in the dog. Journal of Clinical Periodontology 28: Quirynen, M., De Soete, M. & van Steenberghe, D. (2002) Infectious risks for oral implants: a review of the literature. Clinical Oral Implants Research 12: Sarkar, S. & Wilson, M. (1993) Lethal photosensitization of bacteria in subgingival plaque samples from patients with chronic periodontitis. Journal of Periodontal Research 28: Schou, S., Holmstrup, P., Keiding, N. & Fiehn, N.E. (1993) Ligature induced marginal inflamation around osseointegrated implants and ankylosed teeth. Clinical and radiographic observations in cynomolgus monkeys (Macaca fascicularis). Clinical Oral Implants Research 4: Schou, S., Holmstrup, P., Jrgensen, T., Skovgaard, L.T., Stolze, K., Hjrting-Hansen, E. & Wenzel, A. (2003a) Implant surface preparation in the surgical treatment of experimental peri-implantitis with autogenous bone graft and eptfe membrane in cynomolgus monkeys. Clinical Oral Implants Research 14: Schou, S., Holmstrup, P., Jrgensen, T., Stolze, K., Hjrting-Hansen, E. & Wenzel, A. (2003b) Autogenous bone graft and eptfe membrane in the treatment of peri-implantitis. I. Clinical and radiographic observations in cynomolgus monkeys. Clinical Oral Implants Research 14: Schou, S., Holmstrup, P., Skovgaard, L.T., Stolze, K., Hjrting-Hansen, E. & Gundersen, H.J.G. (2003c) Autogenous bone graft and eptfe membrane in the treatment of peri-implantitis. II. Stereologic and histologic observations in cynomolgus monkeys. Clinical Oral Implants Research 14: Shibli, J.A., Martins, M.C., Lotufo, R.F. & Marcantonio Jr, E. (2003a) Microbiologic and radiographic analysis of ligature-induced peri-implantitis. International Journal of Oral & Maxillofacial Implants 18: Shibli, J.A., Martins, M.C., Nociti Jr, F.H., Garcia, V.G. & Marcantonio Jr, E. (2003b) Treatment of ligature-induced peri-implantitis by lethal photosensitization and guided bone regeneration: a histologic study in dogs. Journal of Periodontology 74: Shibli, J.A., Martins, M.C., Theodoro, L.H., Lotufo, R.F.M., Garcia, V.G. & Marcantonio Jr, E. (2003c) Lethal photosensitization in microbiological treatment of ligature-induced peri-implantitis: a preliminary study in dogs. Journal of Oral Science 45: Shibli, J.A., Theodoro, L.H., Haypek, P., Garcia, V.G. & Marcantonio Jr, E. (2004) The effect of CO 2 laser irradiation on failed implant surfaces. Implant Dentistry 13: Singh, G., O Neal, R.B., Brennam, W.A., Strong, S.L., Horner, J.A. & Van Dyke, T.E. (1993) Surgical treatment of induced periimplantitis in the micro pig: clinical and histological analysis. Journal of Periodontology 64: Slots, J., MacDonald, E.S. & Nowzary, H. (1999) Infectious aspects of periodontal regeneration. Periodontology : Wakabayashi, R.C., Wong, F., Richards, D.W. & Johnson, P.W. (1996) Protease repertories of cells adherent to membranes recovered after guided tissue regeneration. Journal of Periodontal Research 31: Wetzel, A.C., Vlassis, J., Hammerle, C.H.F. & Lang, N.P. (1999) Attempts to obtain re-osseontegration following experimental peri-implantitis in dogs. Clinical Oral Implants Research 10: Zechner, W., Tangl, S., Fürst, G., Tepper, G., Thams, U., Mailath, G. & Watzek, G. (2003) Osseous healing characteristics of three different implant types. A histological and histomorphometric study in mini-pigs. Clinical Oral Implants Research 14: Clin. Oral Impl. Res. 17, 2006 /