The objectives were to evaluate the antibacterial effect of self-etching adhesives on carious dentin and the clinical results of restorations in a randomized controlled clinical trial. 104 carious deciduous molars (n=92) were randomly allocated to receive one of the self-etching adhesives: Clearfil SE Bond - control (CB - n =53) or Clearfil SE Protect containing MDPB (CP - n=51) after caries removal selective. Clinical, radiographic, photographic and plaster model assessments (n=53 deciduous molars, CB=29, CP=24) of resin restorations were performed at baseline and at 6, 12, and 24 months. Dentin samples (n=51 deciduous teeth; CB=24 and CP=27) were collected immediately after excavation of carious dentin and 3 months later, and cultured for mutans streptococci - (MS), lactobacilli (LB) and total microorganisms (TM). Data were statistically analyzed (p<0.05).

One hundred and four teeth from 40 children aged 6 to 9 years with one or two superficial deciduous molar cavities (occlusal and/or proximal-occlusal cavities), presenting high caries activity, were selected for this study. Fifty-three primary molars were selected for clinical/ radiographical evaluation and 51 primary molars were selected for microbiological evaluation in two different moments.

Restorative treatments For this part of the study, 53 primary molars were randomly divided into Clearfil SE Bond (CB, n = 29) and Clearfil SE Protect (CP, n = 24). All restorative treatments were performed by the same examiner at the Graduate Dental Clinic of the Faculty of Dentistry of Piracicaba / State University of Campinas. A local anesthetic agent was used for all procedures. In the first consultation, the selected tooth was isolated with a rubber dam and dental prophylaxis performed with pumice paste. Next, 0.2% chlorhexidine gluconate was applied to the operative area. Class I/II cavities were prepared using a high-speed sterile carbide bur (#245, KG Sorensen, Barueri, São Paulo, Brazil) to remove frayed enamel, if necessary, and a size-compatible sterile round steel bur, in a low-speed handpiece, was used to clean decayed tissue (infected dentin) from the enamel-dentin junction and side walls of the cavity, leaving an affected dentin layer on the cavity floor to prevent pulp exposure. Soft and stainable carious dentin was detected using Caries Detector (Kuraray) and then one of the adhesive systems CB or CP was applied on the pulp floor. Following the manufacturer's instructions, the primer was quickly applied with a microbrush (KG Sorensen, Barueri, SP, Brazil) to enamel and dentin and left undisturbed for 20s (seconds). Next, the adhesive was applied to the entire cavity surface with a microbrush (KG Sorensen, Barueri, SP, Brazil). A gentle stream of oil-free air was applied, and light activation was performed for 10s with an Elipar Tri-light unit (3M ESPE, Seefeld 82229, Germany) with a power density of 800 mW/cm2. Teeth were restored with Filtek Z250 composite resin (3M ESPE, St Paul, MN, USA) in 2-3 mm increments and activated by light for 40s using the same light curing unit and power density. The occlusion was checked with carbon paper and adjusted. The restorations were completed with diamond burs (KG Sorensen, São Paulo, SP, Brazil) and Flex Discs (3M ESPE, St Paul, Minnesota, USA).

Direct clinical evaluation Each restoration was evaluated for color match, marginal fit, marginal discoloration, anatomical shape, and secondary caries using the USPHS criteria. The Alpha score indicates clinically optimal restoration, the Bravo score is a clinically acceptable situation, except for secondary caries, and the Charlie score indicates clinically unacceptable restorations that must be replaced. Before the evaluation, prophylaxis was performed with pumice and aqueous paste. All evaluations were carried out under normal clinical conditions with a dental surgical light, a mouth mirror, and a dental explorer. Evaluation of all restorations was carried out at 6, 12, and 24 months by two independent examiners who were blinded to the adhesive system group. If there was disagreement between the two examiners, a consensus was reached on the side of the chair.

Indirect clinical evaluation

1. Photographic evaluation At each time-point, photos of the restorations were taken with a semi-professional digital camera (Sony Cyber-shot DSC-H7, Sony Brasil Ltda, São Paulo, São Paulo), for comparative evaluation of possible clinical changes that occurred during the experimental period. The assessment was performed using parameters similar to those of the clinical direct assessment: color combination, marginal discoloration, anatomical shape, and secondary caries, and excluding marginal integrity. Images with a 1.4X zoom were analyzed on the same computer by two examiners at different times. Individual scores were recorded for further analysis.
2. Plaster models A polyvinyl siloxane impression (Elite HD + Putty Soft, Zhermack, Italy) was taken of each restored tooth and placed on special type IV plaster. The models were inspected by an examiner (AMS) using a stereomicroscope with 16x magnification. The selected sets did not present structural changes (positive or negative bubbles) or other deformations that could interfere with the assessment. The marginal integrity and anatomical structures of the restorations were assessed. The criteria were compared with the assessment recorded in the baseline model. The presence of wear was classified as a change in the initial anatomical shape, and fractures were considered as a change in the marginal integrity of the restorations.

Dentin collections and microbiological procedures For this second part of the study, 51 primary molars were randomly divided into Clearfil SE Bond (CB, n = 24) and Clearfil SE Protect (CP, n = 27). After cavity preparation and selective caries removal, as described above, the initial collection (baseline) of dentin affected by caries was performed with a sterile spoon excavator and transferred to a modified sterile instrument to standardize the amount of dentin (ca. 0.6 mg) removed from each tooth. The collected dentin was placed in tubes containing 1 mL sterile 0.9% NaCl solution for subsequent laboratory procedures. Subsequently, the pulp wall was completely covered with one of the adhesive systems, CB or CP. The cavities were provisionally restored with glass ionomer cement (Maxxion R, FGM, Joinville, SC, Brazil). After 3 months, the individuals returned, and their teeth were evaluated clinically and radiographically. After local anesthesia and isolation with a rubber dam, the provisional restoration was removed and the final collection (reentry) of the decayed dentin was carried out following the same procedures described in the initial collection. The teeth were definitively restored with Filtek Z 250 composite resin (3M ESPE, St Paul, MN, USA), maintaining the same adhesive system as used previously.

The dentin samples were ultrasonically dispersed for 15s (Ultrasonic, LC30, Germany) and diluted in decimal series in 0.9% NaCl. Subsequently, triplicate aliquots obtained from the dilutions were spread on Mitis Salivarius agar (Difco-BD, Sparks, MD, USA) supplemented with sucrose and bacitracin (0.2 U/mL) for isolation of mutans streptococci (MS), Rogosa Agar (Merck, Darmstadt, Germany) for lactobacilli (LB) and Brain Heart Infusion agar (Difco-BD, Sparks, MD, USA) with 5% defibrinated sheep blood to assess the total microorganism (TM). These plates were incubated for 24-48h at 37o C in an atmosphere of 5% of CO2. After incubation, the plates were photographed, and Image J software were used to count the total number of colony-forming units per milliliter (CFU/mL).

Criteria for clinical evaluation of the dentin substrate After removal of the infected dentin, the remaining carious dentin was evaluated according to the following criteria: dentin consistency: 0 = hard (similar to normal dentin), 1 = quite hard (dentin spoon removes tissue decayed when forced), and 2 = soft (tissue easily removed by the dentin spoon); dentin color: 0 = dark brown, 1 = light brown, and 2 = yellow; and dentin moisture: 0 = dry and 1 = wet.

Inclusion Criteria:

• deciduous molars with one or two surfaces with caries lesions;
• children without systemic problems and who were not using any medication;
• absence of spontaneous pain, edema or fistula, and dental mobility;
• radiographically, caries lesions in the inner half of the dentin thickness, absence of pulp contact with the lesion, absence of internal or external root resorption, and other changes suggestive of degenerative pulp conditions, such as radiolucencies in the furcation or periapical regions or widening of periodontal spaces.

Exclusion Criteria:

• Caries lesions with involvement of the pulp or periodontal tissues;
• Children with systemic problems.