Biomedical subjects

John F McCabe

Publications and source records attributed to John F McCabe.

At least 19 recordsLinked to original sources

PubMed2006

Hardness measured with traditional Vickers and Martens hardness methods.

OBJECTIVE: To determine the differences, if any, between hardness measured with traditional Vickers and Martens hardness test methods on denture teeth under 2, 10 and 50 N loads. METHOD: Hardness of acrylic resin (VIV), composite resin (ORT) and porcelain (POR) denture tooth materials was measured using a traditional Vickers hardness (HV) method and Martens hardness (HM) method at 2, 10 and 50N test loads. Vickers hardness was also calculated from the force-indentation depth curves (HVfid) that were recorded during Martens hardness. Indentation creep of the three test materials was also determined during Martens hardness testing. RESULTS: HM values were the same irrespective of the test force used. However, HV values were different for the three test forces. ANOVA using Tukey's test of the HM data showed that the hardness of POR was significantly higher than VIV or ORT (P<0.001). Moreover, ORT had a significantly higher hardness than VIV (P<0.001). The statistical analysis of HVfid data showed similar results. ANOVA of the HV data showed the hardness of VIV to be significantly higher than ORT (P<0.001) under 2, 10 and 50 N test load. The HV values for POR under 2 and 10 N test load could not be calculated because of inability to measure the indentation diagonals. Under the 50 N load, the hardness of POR was significantly higher than VIV and ORT. POR had a significantly lower creep value than any other material tested while VIV showed a statistically significantly higher creep than ORT. SIGNIFICANCE: This study confirms that the visco-elastic recovery of the materials has a very significant effect on the outcome of the hardness tests of denture teeth and the Martens hardness test method has obvious advantages when testing dental materials.

Acrylic Resins↗

PubMed2006

Response to thermal stimuli of glass ionomer cements.

OBJECTIVES: This study was designed to determine the dimensional changes of glass ionomers caused by thermal stimuli under both dry and wet conditions. METHODS: Eight cylindrical specimens (6 mm x 4 mm) were made (using a stainless steel mold) of each of the following materials: a conventional luting glass ionomer, two high viscosity restorative glass ionomers, a resin-modified glass ionomer and a resin composite which was used as a control. The thermal expansion characteristics were determined by a thermal mechanical analyzer (TMA) under wet and dry conditions by heating the samples from 25 to 70 degrees C at 10 degrees C min (-1). RESULTS: All materials showed contraction on heating in dry ambient conditions. In wet conditions, all glass ionomers maintained their original dimensions on heating, but the resin-modified glass ionomer expanded. The resin composite showed expansion in both wet and dry conditions. The results are explained in terms of the opposing effects of thermal expansion and desiccation on heating. SIGNIFICANCE: Under wet conditions glass ionomers maintain their original dimensions when heated. This kind of behavior may be considered as 'smart' behavior.

Composite Resins↗

PubMed2006

Micro-tensile bond strength to surface and subsurface enamel.

UNLABELLED: Testing of bond strength to enamel has been undertaken on both surface and subsurface enamel on the assumption that it is the same substrate. Over recent years, the micro-tensile method has become widely used for the evaluation of dental adhesives. However, there have been no micro-tensile bond strength studies comparing surface and subsurface enamel using orthodontic adhesives. It has been proposed that the bond strength to subsurface enamel is greater than that to surface enamel. OBJECTIVE: The micro-tensile bond strength (MTBS) of Transbond (light activated paste, 3 M Unitek, USA) to surface and subsurface enamel was investigated. METHODS: Ten sound freshly extracted lower human first molars were divided into two groups of five each. Surface enamel of the first group was removed with 500-grit silicone carbide paper, whilst in the second group it was left intact. Gelatin capsules filled with Transbond composite paste were bonded perpendicular to the enamel using light cure adhesive Primer (Transbond XT, 3 M Unitek, USA) in both groups, and stored in distilled water at 37 degrees C for 24 h. Crowns of all specimens were sectioned bucco-lingually into vertical slabs of approximately 0.6 mm thickness. Slabs in each group were prepared into an hourglass shape of 1mm width and 0.6 mm thickness. Micro-tensile bond strength was determined using a mini tensile testing machine (COMPACT Gauge, DILLON). RESULTS: Mean values of MTBS were: 31.2 MPa (SD14.5) for surface enamel and 47.9 (SD15.0) for subsurface enamel. SIGNIFICANCE: There was a significant difference between the two groups P = 0.001. It was concluded that subsurface enamel has significantly higher bond strength than surface enamel. This has implications for bonding in restorative dentistry and orthodontics.

Bisphenol A-Glycidyl Methacrylate↗

PubMed2006

Light exposure required for optimum conversion of light activated resin systems.

OBJECTIVE: The aim of this study was to evaluate the relationship between the degree of conversion (DC) of composites and the light intensity using LED-curing units and also to determine the amount of exposure required to achieve optimal curing. METHOD: The light outputs of light-curing units and the depths of cure of composites exposed to these units were determined using the methods outlined in modified ISO standards, ISO/TS10650 and ISO 4049, respectively. The distributions of DC in composites were investigated by IR spectra of microareas obtained at various depths from the irradiated surface of thin specimens cut out from the cured composites. IR spectra were measured using a Fourier transform infrared spectrometer equipped with a microscopic unit. DC was calculated from the changes in the amount of C=C double bonds in the IR spectra. RESULTS: The light intensity at various depths through the cured composite was calculated from the attenuation coefficient of each material, obtained from the linear relationship between the depth of cure and the logarithm of the amount of exposure, which is defined as the product of the irradiance and irradiation time. There was a third or fourth order regression relationship between DC and the logarithm of total light energy at a particular depth. SIGNIFICANCE: The minimum light energy required to produce a saturated DC was about 1000 s mW/cm2.

Composite Resins↗

PubMed2006

In vitro biofilm formation on the surface of resin-based dentine adhesives.

Prevention of root caries on exposed root surfaces in the aging population is a significant challenge. Bonding resins can be applied to exposed root surfaces as sealants; however, minimal data exists regarding biofilm formation on the surface of these resins. We hypothesized that an antibacterial dentine-bonding resin containing methacryloxydodecyl-pyridiniumbromide (MDPB) may reduce biofilm formation. Biofilms were produced in pooled stimulated natural whole saliva, supplemented with 1% sucrose, on the surface of 5 dentine-bonding resins (Clearfil SE, OptiBond Solo, Protect Bond, Protect Bond Primer, and Xeno III) using untreated root surfaces as controls. Biofilms were stained using the Live:Dead Baclight bacterial viability stain, viewed with confocal microscopy, and analyzed using ImageJ image-analysis software. Resin surfaces encouraged attachment of live bacteria compared with root surfaces. All resins showed similar bacterial colonization in sections adjacent to the resin surface, but in the central and outer portions of biofilms, Xeno III and Protect Bond Primer showed a viable bacterial load similar to that of the root surface. Fluoride-releasing resins (OptiBond Solo/Protect Bond) did not show reduced biofilm formation. Thus, antibacterial agents within the resins have a minimal effect on biofilm formation, particularly when directly adjacent to the root surface.

Anti-Bacterial Agents↗

PubMed2006

Film thickness and rheological properties of luting agents for crown cementation.

The film thickness and rheological properties [viscosity and tan sigma], of different resin based crown and bridge luting agents, including Panavia 21, Superbond, All Bond C&B Cement, Variolink were compared with zinc phosphate cement. A modification of the method in ISO 9917 (1991) and a controlled stress rheometer were used to monitor the changes in the film thickness and rheological properties respectively. For each material, the film thickness and rheological characteristics were determined at 60s after start of mixing and then made at regular intervals until immediately before the material was set. For the initial film thickness, the difference among the values for Varionk, Superbond and zinc phosphate cement was not significant (P>0.05), and less than 25microm. The values for All Bond C&B Cement and Panavia 21 were at least twice that of the other materials tested. All Bond C&B Cement produced the highest film thickness value at both the initial period and at 240s. Regarding viscosity, Superbond had the lowest initial value but the highest value at 240s. Variolink gave the highest initial value of viscosity, but the differences in the initial values among the other materials were small. During the whole period of testing, zinc phosphate cement gave the highest values of tan sigma, and Panavia 21 showed no significant change in any of the measured properties. The methods were able to distinguish the behaviour of different cement materials. Some resin cements produce larger than optimal values of film thickness and in some cases this is related to a rapid increase in viscosity after mixing. It is expected that differences in the ability of cements to form thin films will affect the seating of cemented restorations.

Analysis of Variance↗

PubMed2005

An alcohol-free tissue conditioner--a laboratory evaluation.

OBJECTIVES: An alcohol-free tissue conditioner based on a n-butyl methacrylate/i-butyl methacrylate copolymer has recently been developed. The purpose of the present study was to compare some key properties of the new tissue conditioner with those of poly(ethyl methacrylate)-based conventional materials containing ethyl alcohol. The effect of a coating, which consisted of poly(ethyl methacrylate) and methyl methacrylate, was also evaluated. METHODS: The new alcohol-free tissue conditioner (Fictioner) and three tissue conditioners containing ethyl alcohol (FITT, Hydro-Cast, SR-Ivoseal) were evaluated. The coated alcohol-free material was also used. Gelation characteristics, dynamic viscoelastic properties and compatibility with dental stones were measured using a displacement rheometer, dynamic viscoelastometer and profilometer, respectively. In addition, weight changes during immersion in water were determined. RESULTS: The working time and gelation time of the alcohol-free tissue conditioner were similar to those of the conventional materials. This alcohol-free material had significantly lower shear storage modulus and shear loss modulus, and higher loss tangent (P<0.05) than FITT and SR-Ivoseal at 0.01 and 1 Hz. The alcohol-free material maintained its inherent viscoelastic properties and exhibited only a slight change in weight during 14 days of water immersion when compared to the conventional materials. The application of the coating significantly reduced the loss of the initial viscoelastic properties and surface quality during the test periods. CONCLUSIONS: The coated alcohol-free tissue conditioner would be superior to the conventional materials containing ethyl alcohol in view of viscoelastic properties after gelation, compatibility with dental stones and durability.

Acrylic Resins↗

PubMed2005

Fluoride release from aged resin composites containing fluoridated glass filler.

OBJECTIVES: The aim of this study was to evaluate the fluoride release from aged resin composites containing different types of fluoridated glass filler into both deionized distilled water and lactic acid solution. METHODS: Three resin composites, UniFil S (containing fluoro-alumino-silicate glass filler), Reactmer (containing pre-reacted glass-ionomer filler) and Beautifil (containing both types of fillers) were used. A conventional glass-ionomer cement, Ketac-Fil, was used as a control. Five disk specimens of each material were prepared and aged in water for 10 weeks. After aging, specimens were immersed in deionized distilled water for a further 6 days and then in aqueous lactic acid (pH 4.0) for 2 days. This process was repeated twice more and the specimens were subsequently immersed in water for a further 12 days. Fluoride release was measured every 2 days throughout the post-aging period. RESULTS: The amount of fluoride release for aged UniFil S and Beautifil markedly increased in acid solution compared with water storage. The difference was not so great for aged Reactmer and Ketac-Fil. UniFil S and Beautifil gave significantly greater fluoride release in water following immersion in acid solution (p<0.05, two-way ANOVA and Scheffe's test), but Reactmer and Ketac-Fil showed no such increase in fluoride release after acid immersion. SIGNIFICANCE: These results suggested that the nature of the fluoridated glass filler within a resin composite and the way in which the material interacts with an acidic environment affected the amount of fluoride released.

Analysis of Variance↗

PubMed2004

Effect of two-step adhesive systems on inhibition of secondary caries around fluoride-releasing resin composite restorations in root dentine.

OBJECTIVE: The purpose of this study was to investigate the effect of two-step adhesive systems on secondary caries inhibition around fluoride-releasing materials in vitro. METHODS: Two self-etching primer systems, Clearfil SE Bond (SE) and UniFil Bond (UB), and two one-bottle systems with a total-etch wet-bonding technique, Single Bond (SB) and One-Step (OS), were used prior to placement of resin composites either with (Reactmer) or without (Z100) fluoride release. Class V cavities prepared in extracted human premolars were restored with various combinations of materials: Reactmer/SE, Reactmer/UB, Reactmer/SB, Reactmer/OS, Z100/SE, Z100/UB, Z100/SB and Z100/OS. After storage for 14 days, the restored teeth were incubated in bacterial medium containing sucrose with Streptoccus mutans for 2 weeks. Water sorption and desorption of the adhesives and fluoride release from the resins either coated with adhesive or uncoated were also determined. RESULTS: The one-bottle groups showed higher water sorption and desorption than the self-etching primer groups. Although fluoride release from the Reactmer specimens was suppressed by the adhesive coating, the one-bottle groups allowed significantly higher fluoride release than the self-etching primer groups. On microradiographs, the radio-opaque layers adjacent to the Reactmer restorations were thick and clear, while the layers adjacent to the Z100 restorations were thin and unclear. For the Reactmer restorations, the radio-opaque layers associated with the one-bottle groups were significantly thicker than for the self-etching primer groups. CONCLUSIONS: These results indicated that the use of one-bottle wet-bonding systems for bonding of fluoride-releasing resin composites to dentine may contribute to inhibit secondary caries compared to self-etching primer systems.

Bisphenol A-Glycidyl Methacrylate↗

PubMed2004

Effect of impression technique on bond strength.

If the effects of surface preparation (eg, acid etching, laser preparation, crystal growth) are to be investigated on the same tooth from which the bond strength is recorded, a method of surface replication is required that does not affect the subsequent bond. This study investigated the effect of 2 different methods of taking impressions on bond strength. Three groups of 11 mandibular incisors were used. The labial enamel was etched with 37% phosphoric acid for 30 seconds. Group A (control) had no impression taken; in group B (silicone), impressions were taken with silicone impression material before bonding; in group C (polyether), an impression was taken with polyether before bonding. After the impressions were taken, GAC brackets (A Company, San Diego, Calif) were bonded to the labial surfaces of the etched enamel with Transbond XT light-cured composite (3M Unitek, Monrovia, Calif). Teeth with bonded brackets were stored in water at 37 degrees C for 24 hours, and then bond strength was measured on a testing machine. The adhesive remnant index (ARI) was also recorded. The lowest bond strength was found after silicone replication (mean [standard deviation]: 8.6 [1.7] MPa) and the highest in the control group (21.2 [4.0] MPa). There was no significant difference between the control group and the polyether replication group (19.1 [4.7] MPa). The surface detail replications of polyether and silicone were found to be identical. It was concluded that polyether had no significant effect on bond strength and was suitable for surface replication before bonding. Polyether allows replication of the enamel surface without a significant effect on bond strength, and this technique could be used to examine the relationship between enamel preparation techniques and subsequent bond strength between composite and enamel.

Adolescent↗

PubMed2004

Water absorption, dimensional change and radial pressure in resin matrix dental restorative materials.

The purpose of this work was to study the relationship between water absorption, dimensional change (swelling) under cavity constraint and radial stress generation in resin matrix dental restorative materials. Water absorption was determined on disc specimens whilst swelling was determined on samples of materials restrained within cavities cut in cast polymethylmethacrylate and pressure generated was determined using a 'push-out' test. Four commercially available resin matrix materials were used. A giomer material gave significantly greater water absorption than two compomers and a fluoride releasing composite (p<0.05). The giomer material was the only material which produced a significant degree of swelling (p<0.05) when restrained within a cavity. The giomer product produced the greatest radial pressure (over 20 MPa in 1 month) following water storage, however a significant pressure generation was also observed for other materials despite their much lower water absorption values. The mechanism of water absorption and the amount of water absorbed determine the dimensional changes and radial pressure generated by resin matrix materials in a moist environment.

Absorption↗

PubMed2004

Temperature mediated coefficient of dimensional change of dental tooth-colored restorative materials.

OBJECTIVES: Restorative materials are constantly subjected to thermal challenges in the oral environment. Such challenges, if significant, can have unfavorable effects on the margins of restorations in terms of the seal between the material and the tooth structure. This study aimed to assess the Coefficient of Dimensional Change (CDC) of tooth-colored restorative materials. METHODS: Five cylindrical specimens (6 mm x 4 mm) were made (using a stainless steel mold) of each of the following: the compomers Dyract AP (Dentsply), or F2000 Compomer (3M); a resin composite, Z100 MP (3M); a resin-modified glass-ionomer, Fuji II LC Capsule (GC); the conventional glass-ionomers Fuji IX GP Fast (GC) or Ketac Fil Aplicap (ESPE). The light-cured materials were cured for 40 s at each end and also around the 'waist' after removal from the mold. All specimens were stored in distilled water at 37 degrees C for 24 h, before testing. The CDC for each specimen was determined using a thermal mechanical analyser, by heating the sample from 25 to 70 degrees C at 10 degrees C min(-1). RESULTS: All materials except the glass-ionomers showed expansion on heating. The temperature response was non-linear in each case and values of CDC were therefore calculated between 25 and 50 degrees C and between 50 and 70 degrees C. The mean values of CDC (x 10(-6) degrees C(-1)) between 25 and 50 degrees C were Dyract: 83.4, F2000: 66.1, Z100: 64.5. The glass-ionomer materials showed contraction, which was non-linear in nature and was associated with a loss of water on heating. ANOVA and Tukey's pairwise comparisons of mean CDC values for the other materials indicated significant differences between all pairs of materials. SIGNIFICANCE: The compomers and resin composite tested had similar values of CDC. The conventional and resin-modified glass-ionomers contracted on heating. For one glass-ionomer the dimensional change on heating was minimal as thermal expansion appeared to be compensated by water loss.

Analysis of Variance↗

PubMed2004

Fluoride release and recharge in giomer, compomer and resin composite.

OBJECTIVES: The aim of this study was to examine the fluoride recharging and releasing abilities of resin-based materials containing fluoridated glass filler to determine whether the extent of the glass-ionomer matrix of the material affects these properties. METHODS: Three materials having a different proportion of the hydrogel matrix surrounding the glass filler, namely: Reactmer paste, Dyract AP and Xeno CF, were used for this study. Five disk specimens of each material were placed into distilled/deionized water and the fluoride release measured during 38 days. For fluoride recharge the disks were exposed to 250 ppm F solution for 1 h and the pre- and post-recharge fluoride release were determined using an ion-selective electrode (total fluoride ions) and ion chromatography (free fluoride ions). RESULTS: The amounts of total and free fluoride release from each material at the initial period in descending order were Reactmer paste > Dyract AP > Xeno CF (p < 0.05 ANOVA and Scheffe's test). After fluoride recharge, Reactmer paste showed a greater amount of fluoride release than the other materials. Dyract AP and Xeno CF showed a similar total level of fluoride release after recharging. For Xeno CF the amount of total fluoride released after recharging was significantly greater than that of free fluoride, but there was no significant difference between total and free fluoride released after recharging for Dyract AP (p = 0.05 paired t-test). SIGNIFICANCE: These results suggested that the extent of the glass-ionomer matrix of the glass filler played an important role for fluoride-releasing and recharging abilities of the resin-based materials.

Analysis of Variance↗

PubMed2004

Effect of aperture size on irradiance of LED curing units.

OBJECTIVE: The aim of this study was to evaluate the light output from LED-curing units measured according to the current ISO standards. The effect of aperture size on irradiance was also evaluated. METHOD: The irradiance of light-curing units and the depths of cure of composites exposed to these units were determined using the methods outlined in ISO standards, ISO/TS10650 and ISO 4049, respectively. The irradiance measured through two different sized apertures was also measured. RESULTS: The irradiance measured with an aperture was greater than that without an aperture. For each light-curing unit and material, there was a linear relationship between the depth of cure and the logarithm of the amount of exposure, which is defined as the product of the irradiance and irradiation time. The correlation coefficients of these linear relationships, using data obtained from the different units was only moderate when the irradiance was measured without an aperture. The correlations improved with decreasing aperture size. SIGNIFICANCE: Since the irradiance is affected by the angular aperture of the light guide and the size of the mold for measuring depth of cure of resin in ISO 4049 is 4-mm diameter, the irradiance through a 4-mm aperture should be used in the determination of the relationship between depth of cure and light exposure.

Composite Resins↗

PubMed2004

Effect of the applied load on surface contact fatigue of dental filling materials.

OBJECTIVE: The purpose of this study was to investigate the effect of the applied load in a rolling-ball device on the surface contact fatigue of two dental filling materials: a microfilled composite (FA) and a glass ionomer (GFII). METHODS: Disk specimens 10 mm diameter by 1.5 mm thick were set into cavities cut in plates of PMMA. After setting, the specimens were ground and polished using wet-carborundum paper followed by 1 microm alumina and then stored for 24 h in water at 23 degrees C. The surface fatigue test was carried out using loads ranging from 100 to 500 gf through a ruby ball 2 mm diameter using a rolling-ball device. The ball was set to rotate at 720 rpm and a surface profile was determined on impressions recorded at regular intervals. Fatigue life was defined as the number of cycles to produce a track 5 microm deep and was determined by interpolation. Data were statistically analyzed by ANOVA and t-tests at a significance level of 0.05. RESULTS: At loads ranging from 200 to 500 gf the fatigue life was 0.7-10.6 x 10(5) cycles for FA, and 0.7-4.8 x 10(3) cycles for GFII. For both materials the fatigue life was reduced significantly by increasing the test load (p<0.05). The surface contact fatigue life for FA was at least 100 times greater than that for GFII at all loads. SIGNIFICANCE: The rolling ball device was confirmed as a simple and meaningful device for the study of surface contact fatigue. A microfilled composite is very much more resistant to contact fatigue than a glass ionomer.

Composite Resins↗

PubMed2004

Effect of mixing method on the porosity of encapsulated glass ionomer cement.

OBJECTIVE: To observe cement specimens mixed by various methods non-destructively using microfocus computerized tomography (micro-CT) and to evaluate the effect of mixing method on porosity. METHOD: Five glass ionomer cements were used: two were hand-mixed and three were encapsulated. The latter were mixed either by shaking or rotating. Fifteen cylindrical specimens were prepared for each material by each mixing method and stored in distilled water at 37 degrees C until testing. The specimens were observed and analyzed using micro-CT. 201 sliced images were obtained horizontally along the length of the specimens. Three-dimensional reconstructions were made and the sizes, numbers and total volume ratio of bubbles in the specimens were calculated. RESULTS: Mixing method had a significant effect on porosity (P<0.05). For the luting/lining cement, mechanical mixing produced a significantly higher porosity (P<0.05). For the restorative cement, there were only small differences in porosity between specimens mixed by the two methods. More large bubbles were detected in specimens mixed mechanically than by hand for the fluid material. This suggested that for low-viscosity materials bubbles may combine until a certain viscosity is reached during setting. SIGNIFICANCE: The strength of glass ionomer cement is affected by incorporated porosity and this depends on the mixing method. For low-viscosity cements, hand-mixing is favored in order to reduce porosity and increase strength, but this is not generally applicable to high-viscosity cements.

Cementation↗

PubMed2004

Determination of fluoride ions released from resin-based dental materials using ion-selective electrode and ion chromatograph.

OBJECTIVES: The aim of this study was to investigate the quantities of total and free fluoride ions released from four resin-based composites containing fluoride (Heliomolar (HE), UniFil S (UF), Beautifil (BE) and Reactmer paste (RP)) and one glass ionomer cement (Ketac-Fil Plus Aplicap (KF)). METHODS: Five disk specimens of each material were prepared and the fluoride release was measured at 1, 2, 4, 7, 14, 21 and 28 days. The amounts of total and free fluoride ions released from materials were measured by using an ion-selective electrode (ISE) and ion chromatography (IC). Data were statistically analyzed using a t-test at a significant level of 0.05. RESULTS: For HE, fluoride release is undetectable using the ISE but was detected by the IC method. There were no significant differences in the free and total fluoride ions released from UF and BE (p>0.05). For RP, the free fluoride ions were significantly lower than the total fluoride ions after 4 days (p<0.05). KF showed significant difference between total and free fluoride ions at 7 and 28 days (p<0.05). CONCLUSIONS: The nature of the fluoride incorporated into dental materials affected the amounts of free and total fluoride ions released from materials.

Acrylic Resins↗

PubMed2004

Calbindin D-28k distribution in odontoblasts underneath tertiary dentine in human carious teeth.

OBJECTIVE: The aim of this study was to examine the calbindin D-28k immunoreactivity in carious teeth to know whether this protein may have a function in tertiary dentine formation. METHODS: Human extracted teeth with or without carious lesions were immersion-fixed with Zamboni fixative, demineralized in 4.13% EDTA solution (pH 7.4), frozen-sectioned, and processed for calbindin immunoreactivity and hematoxylin-eosin stain. The intensity of the immunostaining was evaluated by quantitative densitometry. RESULTS: In intact teeth, numerous odontoblasts were aligned underneath the secondary dentine and their cell bodies showed the immunoreactivity. In carious teeth, tertiary dentine had poor- or rich tubular patterns under the carious lesion. Underneath the tubule-poor tertiary dentine, distinct odontoblasts could not be seen at the central site. However, some cells with a flat appearance were located at this site and were immunonegative for calbindin D-28k. On the other hand, columnar odontoblasts were seen at the peripheral site, and their cell bodies and processes showed strong immunoreactivity. Underneath the tubule-rich tertiary dentine, columnar odontoblasts were abundantly distributed, and the strong immunoreactivity was observed in their cell bodies and processes. The immunoreactivity in odontoblasts underneath the tertiary dentine with poor or rich tubular pattern was more intense than that for the secondary dentine in intact teeth (P<0.05). On the other hand, the intensity of the immunoreactivity in odontoblasts was similar underneath the secondary dentine in intact and carious teeth. CONCLUSIONS: The present study demonstrated that calbidin D-28k was actively synthesised by odontoblasts under the carious lesion. These findings may suggest that this protein plays an important role in the tertiary dentine formation.

Adult↗