Fifty caries-free human third molars were selected for this study after obtaining Indiana University Institutional Review Board approval. The molars were stored at 4℃ in 0.1% thymol and used within six months after extraction. A 1.5 mm coronal dentin disc was cross-sectioned from each molar using a low-speed saw (IsoMet, Buehler, Lake Bluff, IL, USA) with a diamond disc (15LC Diamond Wafering Blade, Buehler) under deionized running water (Figure 1). The non-pulpal outer sides of the dentin discs were flattened to a uniform thickness using an automatic polishing machine (Struers Rotopol 31, Struers, Cleveland, OH, USA) with 500 grit silicon carbide grinding paper (Struers). The deep pulpal side of each dentin disc was polished with 1,200, 2,400 and 4,000 grit papers (Struers) and finally using a 1 µm diamond polishing suspension (Struers). As a final cleaning step, the polished specimens were sonicated in de-ionized water for 3 minutes.

To prepare the clinically used concentration of TAP (1 g/mL), 1 g of United States Pharmacopeia grade antibiotic powders compounded of equal portions of metronidazole, ciprofloxacin and minocycline (Champs Pharmacy, San Antonio, TX, USA) was mixed with 1 mL of sterile water.22 To prepare a diluted pasty consistency of methylcellulose-based triple antibiotic paste (DTAP), 100 mg of United States Pharmacopeia grade antibiotic powders compounded of equal portions of metronidazole, ciprofloxacin and minocycline was dissolved in 100 mL of sterile water. Then, 8 g of methyl cellulose powder (Methocel 60 HG, Sigma-Aldrich, St. Louis, MO, USA) was added to the 100 mL of 1 mg/mL solution of TAP and mixed for 30 minutes using a magnetic stir bar to obtain a homogenous DTAP with 1 mg/mL concentration of TAP. A commercial Ca(OH)₂ intracanal medicament (UltraCal XS, Ultradent, South Jordan, UT, USA) was also used in this study.

The dentin discs were randomized into three treatment groups and two control groups (Figure 1). For the first control group, the pulpal side of each dentin disc was irrigated with 20 mL of sterile water for 5 minutes using a 27 gauge needle and 5 mL syringes. Then, each sample was stored for four weeks at 37℃ in a sealed 2 mL conical sample cup (Fisher Scientific, Pittsburgh, PA, USA) containing a cotton pellet saturated with distilled water. After four weeks, the pulpal side of each dentin disc was irrigated with 20 mL of sterile water for 10 minutes. For the second control group (no medicament), the pulpal side of each dentin disc was slowly irrigated with 20 mL of 1.5% NaOCl for 5 minutes. Then, each sample was stored for four weeks at 37℃ in a sealed 2 mL conical sample cup containing a cotton pellet saturated with distilled water. After that, the pulpal side of each dentin disc was irrigated with 20 mL of 17% EDTA for 10 minutes. For each of the three treatment groups, the pulpal side of each dentin disc was slowly irrigated with 20 mL of 1.5% NaOCl for 5 minutes. Then, 0.1 mL of TAP, DTAP, or Ca(OH)₂ paste was applied on the pulpal side of each specimen using a disposable syringe (BD, Franklin Lakes, NJ, USA). The treated dentin disks were kept for four weeks at 37℃ in a sealed 2 mL conical sample cup containing a cotton pellet saturated with distilled water. After four weeks, the treated side of each dentin specimen in all treatment groups was irrigated with 20 mL of 17% EDTA for 10 minutes. The amount of intracanal medicaments applied in the three treatment groups (0.1 mL) was just enough to cover the pulpal surface of dentin discs with the medicament to avoid application of excessive amount of paste. The four weeks of intracanal medication and the time and volume of both EDTA and NaOCl irrigation used in this study were based on the current ER protocol recommended by AAE.10

The treatment side of each dentin disc was bisected by an imaginary line into two halves and labeled (Figure 2). One half of each treatment side was randomly selected and subjected to microhardness measurements using a Vickers microhardness tester (LM247, Leco, St. Joseph, MI, USA). Three indentations, spaced 200 µm apart, were made on each specimen using a 50 g load and a 10 second dwell time. The indentations were precisely measured using an optical microscope equipped with a digital camera and image analysis software. The representative hardness value for each specimen was obtained as the mean of the results from the three indentations.

The chemical integrity of treated dentin was evaluated using Fourier transform infrared (FTIR) spectrophotometer (Jasco Inc., Tokyo, Japan) with a diamond attenuated total reflectance accessory. Five dentin disc samples were randomly selected from each group for FTIR measurement. Three different areas were selected from the other half, not subjected to microhardness testing (Figure 2). FTIR spectra were then collected from the three selected areas in each sample between 800 and 2,000 cm^-1 at 4 cm^-1 resolution using 100 scans. Each obtained spectrum was processed by smoothing, baseline correction and normalization to the amide I peak using Spectra manager software (Jasco Inc.). The ratio of integrated areas of the phosphate v1 and v3 peaks to the amide I peak was quantified to calculate the relative contents of these inorganic and organic components. The representative final ratio allocated for each dentin sample was obtained as the mean of the ratios obtained from the three single scans. Larger phosphate/amide I ratios compared to untreated dentin samples corresponded to a higher collagen deproteinization, whereas smaller phosphate/amide I ratios compared to untreated dentin samples corresponded to a higher dentin demineralization.

Two dentin samples were randomly selected from each group for scanning electron microscopic (SEM) analysis to detect any morphological changes in the treated dentin side after experimental treatments. Each selected specimen was irrigated with 10 mL of de-ionized water, sonicated for 5 min in de-ionized water, and air dried for 48 hours using low vacuum pressure desiccator (Bel-Art, Wayne, NJ, USA) and silica gel crystals. Then, samples were sputter coated for 70 seconds with gold/palladium using a sputter coater (POLARON Sputter Coating System, Energy Beam Sciences, Agawam, MA, USA) and images were taken from the treated side of the dentin samples with an SEM (JEOL 7800F, JEOL, Peabody, MA, USA) in secondary electron imaging mode.

All data were checked for normality using the Kolmogorov-Smirnov test and the normality assumptions were satisfied. The differences in microhardness and phosphate/amide I ratios among experimental groups were examined using one-way ANOVA followed by Fisher's least significant difference. Pearson correlation coefficients overall and by group were calculated to determine the correlations between phosphate/amide I ratios and microhardness measurements. A 5% level of statistical significance was applied for all analyses.