White spot lesions are often seen on the teeth after orthodontic treatment, resulting in unpleasant esthetics. The aim of this in-vitro study was to compare subjectively and objectively the esthetic outcomes of white spot lesions treated with 3 commercially available products that have been reported to have a positive effect on the remineralization of enamel.
Forty extracted premolars were randomly allocated into 1 of 4 groups (n = 10). The teeth were exposed to a demineralization solution at 37°C for 14 days to produce white spot lesions that were about 100 μm deep. Each group was then randomly assigned to receive either control treatment with artificial saliva or treatment with 1 of the 3 commercially available products: Restore toothpaste (Dr. Collins Inc, Orange County, Calif), which contains NovaMin (Dr. Collins Inc); Prevident 5000 (Colgate, New York, NY); and MI Paste Plus (GC America, Alsip, Ill). All groups were evaluated 5 times at 6 time points during the study.
The subjective and objective results were mixed within groups and between groups for the products tested.
We found no conclusive evidence that any of these 3 materials produced more favorable esthetic white spot lesion remineralization results.
White spot lesions after orthodontic treatment are an anguishing matter for the orthodontist, the patient, and the parents. Much effort has been directed toward prevention of white spot lesions; yet, despite best efforts, they still occur. The reported incidence rates of white spot lesions are 2% to 96% of tooth surfaces. It has been reported that the greatest prevalence of white spot lesions is on the cervical and middle thirds of the crowns of the first molars, lateral incisors, and canines. White spot lesions can develop rapidly in the presence of plaque-produced acid by-products. Large variations in the rates of demineralization and remineralization might be due to large variations in enamel composition, including local concentration gradients of specific mineral ions, as well as endogenous organic material and organic acids. The fact that many white spot lesions occur in the anterior part of the mouth carries with it the issue of unpleasant esthetics.
Tooth color is certainly a part of pleasing tooth esthetics. Tooth color depends on both intrinsic and extrinsic colorations. Extrinsic color depends on the absorption of materials on the enamel surface. Intrinsic color depends on the enamel and dentin properties of light absorption and scattering. Hydroxyapatite crystals have been found to be a key factor in the light-scattering properties of enamel. Demineralization of enamel can increase the scattering coefficient of light by a factor of 3. Demineralization in white spot lesions occurs mostly in the subsurface region of enamel and can occur rapidly around bonded appliances up to a depth of 75 μm. This demineralized enamel causes a diffuse backscattering of light that is seen as a dull white chalky area.
How to treat white spot lesions after appliance removal to produce a sound and esthetically pleasing enamel surface is a question yet to be fully answered. There is not 1 published proven method whereby white spot lesions can be predictably and esthetically remineralized. Various methods and products have been proposed, such as remineralization through saliva, mouth rinses, and toothpastes with various components such as increased fluoride concentrations, casein phosphopeptide amorphous calcium phosphate, and calcium sodium phosphosilicate glass. More aggressive and invasive techniques, such as microabrasion and composite restorations, have also been used to resolve white spot lesions.
Research has shown a reduction in the size of white spot lesions over time without the use of any products directed at their resolution. Resolution is thought to occur via 2 mechanisms: (1) the concentration of various soluble ions in saliva, particularly calcium, phosphate, and fluoride, which promote remineralization of the lesions ; and (2) surface abrasion to remove the dissolved surface enamel of the lesion, exposing the underlying enamel crystals, which are tightly packed and thus provide proper light reflection.
The ability of the crystal lattice of enamel to redeposit dissolved hydroxyapatite as fluoroapatite is the basis for the use of fluoride as a remineralization agent. High concentrations of fluoride have been shown to remineralize white spot lesions via hypermineralization, resulting in a remineralized but usually unsightly tooth. This remineralization occurs on the enamel surface and inhibits ion movement into the subsurface of the lesion, thereby affecting the reflection of light contacting the surface.
Materials that deliver calcium and phosphate ions to enamel have been proposed to promote remineralization with a more esthetically pleasing result. Casein phosphopeptide amorphous calcium phosphate (Recaldent; Cadbury Enterprises Pte, Melbourne, Australia) is 1 such material. Casein phosphopeptide amorphous calcium phosphate acts by binding to the tooth surface and plaque. This binding maintains high concentration gradients of soluble calcium and phosphate ions as a result of the stabilization of the amorphous calcium phosphate, both of which play an integral role in remineralization. These calcium and phosphate ions deposit in the crystal voids in the enamel and promote crystal growth, which has a positive impact on the reflective properties of the enamel.
Calcium sodium phosphosilicate glass (NovaMin; Dr. Collins Inc, Orange County, Calif) is marketed as a treatment for dentin sensitivity. Its success and mechanism of action as a sensitivity-reduction agent promoted interest as a remineralization agent. It is relatively new and less studied as a remineralization product that is thought to maintain high concentrations of calcium and phosphate on the enamel surface. NovaMin is composed of inorganic calcium phosphate minerals plus silica. The theorized method of action is sodium ions released through interaction with oral fluids, resulting in an increase in pH. This increase in pH releases calcium and phosphate to form a layer of calcium phosphate. This layer then crystallizes, producing a crystalline hydroxycarbonate apatite layer filling the crystal voids in the enamel.
In this in-vitro study, we sought to compare subjectively and objectively the esthetic outcome of white spot lesions treated with 3 commercially available products: Prevident 5000 (Colgate, New York, NY), MI Paste Plus (GC America, Alsip, Ill), and Restore toothpaste (Dr. Collins Inc). All 3 claim to have a positive esthetic effect on the remineralization of enamel. The methods and results of this study were to be the basis for a clinical trial to evaluate the effectiveness of the 3 products in vivo.
Material and methods
Forty extracted premolars without buccal restorations, obvious decalcifications, or obvious caries were selected as the sample teeth. Universal premolar brackets (Ormco, Orange, Calif) were bonded to the facial surfaces of the sample teeth with Transbond (3M Unitek, Monrovia, Calif). The sample teeth were randomly allocated into 1 of 4 groups (n = 10). Teeth from each group were mounted into an epoxy resin block with the buccal surfaces exposed. Clear nail polish was used to seal the exposed buccal enamel, except for the area gingival to the bracket, allowing for the production of a clinically appearing white spot lesion. Each tooth was assigned a unique identification number and stored in an artificial saliva solution (2.200 g/L of gastric mucin, 0.381g/L of sodium chloride, 0.213 g/L of calcium chloride dihydrate, 0.738 g/L of dipotassium phosphate, and 1.114 g/L of potassium chloride; pH was adjusted to 7.00 with 85% lactic acid). All samples were exposed to a demineralization solution (3 mmol/L of monopotassium phosphate, 3 mmol/L of calcium chloride dihydrate, and 0.1 M of lactic acid 85% solution with the pH adjusted to 4.5 with about 5 g of potassium hydroxide) at 37°C for 14 days to produce a white spot lesion. This white spot lesion production method was verified by exposing 3 additional premolars to the demineralization solution for 14 days, followed by sectioning of the 3 premolars and measurement of the lesion depth with polarized light microscopy. Lesion depth was found to be near 100 μm, which is in line with previous studies evaluating remineralization. After 14 days and visual verification of white spot lesion appearance on all sample teeth, all brackets were debonded, and any residual adhesive was removed with a high-speed hand piece and a number 12 fluted carbide bur, with care taken not to remove enamel of the white spot lesions.
Each group was then randomly assigned to either a control of artificial saliva (group 1) or to 1 of 3 commercially available products: Restore toothpaste, which contains NovaMin (group 2); Prevident 5000 (group 3), or MI Paste Plus (group 4). The 4 sample blocks were placed in 4 separate containers of artificial saliva and placed in an incubator at 37°C. Twice a day for 28 days, each block was removed from the artificial saliva solution, treated with its designated product applied with a toothbrush according to the manufacturers’ directions, and returned to its container without rinsing. The time frame chosen corresponds with previously published studies evaluating remineralization. The solution of artificial saliva in each container was changed after each treatment. Separate toothbrushes were used for each block to prevent cross-contamination with a different product.
All groups were evaluated 5 times at 6 time points during the study. The time points were pretreatment, immediately after demineralization, day 7 of treatment, day 14 of treatment, day 21 of treatment, and day 28 of treatment. All teeth were photographed in a light-controlled environment using a standardized camera (Nikon, Melville, NY) and settings with a 60mm/AF micro lens.
Since the purpose of this study was to objectively measure changes in the esthetic appearance of white spot lesions and since appearance depends on the interaction of light with the enamel surface, the groups were evaluated according to 4 parameters involving the enamel surface.
The shade of the gingival-buccal area was measured with a Vita Easyshade meter (Vident, Brea, Calif). The area was measured 3 times at each time point, and the value recorded was the shade reported by the meter at least twice.
The reflective aspects of the same gingival-buccal area were measured with a glossmeter (Novo-Curve; Rhopoint Instruments, East Sussex, United Kingdom). The area was measured 3 times at each time point for gloss of the enamel. The mean of the 3 glossmeter readings at each time point was used for the analysis of the value.
The Cielab (Brisbane, Queensland, Australia) color values of the gingival-buccal area were measured with Photoshop (Adobe Systems, San Jose, Calif).
Surface profilometry (Talyscan 150; Ametek, Paoli, Pa) was used to measure the surface smoothness of 3 0.8 × 0.8-mm areas in the gingival-buccal area of 1 teeth from each group.
The pretreatment and day-28 photographs for each tooth were subjectively evaluated by a 5-member panel of nondental personnel and a 5-member panel of dental personnel. The raters were blinded to the treatment group of the teeth they evaluated. Each member privately viewed a PowerPoint (Microsoft Corp, Redmond, Wash) presentation of each tooth’s photo at pretreatment and day 28 side by side. All members viewed the presentation on the same laptop computer. They were asked to evaluate the degree of white spot lesions at day 28 compared with pretreatment on a 5-point Likert scale (Appendix) . The results were then evaluated for interrater agreement. Intraclass correlation coefficients (ICC) were used to assess interrater reliability across the 45 pairs of photographs of teeth of the examiners’ assessments comparing the white spot lesions. Five pairs of photos were repeated to allow evaluation of rater reliability. The ICC that was used to assess interrater agreement in this study treats both the raters and the teeth as randomly sampled from a larger population and quantifies the within-photograph variability relative to the between-photograph variability of the examiners’ assessments. Under these assumptions, the ICC was calculated by using variance components from a 2-way analysis of variance (ANOVA) random-effects model without interactions. This model allows for generalization of the results to the larger populations of dentists and nondentists. For intrarater agreement, the version of the ICC that treats examiners as fixed effects was used, since we were specifically interested in assessing the agreement between the same examiner on different occasions.
For the Vita shade measurements, when comparing the pretreatment with the day-28 chroma, an increase in the numeric portion of the score corresponded to a darker chroma, whereas a decrease indicated a lighter chroma. For the statistical analysis, a darker chroma was assigned a change score of +1, a lighter chroma was assigned a change score of –1, and no change in chroma was assigned a score of 0. Similar scoring criteria were applied to the hue; eg, going from “A” at pretreatment to “B” at day 28 was assigned a hue change score of +1, whereas going from “B” at pretreatment to “A” at day 28 was assigned a hue change score of –1, and having “A” at both pretreatment and day-28 resulted in a hue change score of 0. The Wilcoxon signed rank test was used to determine whether there were significant changes from pretreatment to day 28 in hue and chroma values. The Kruskal-Wallis test was used to compare the groups’ pretreatment and day-28 changes in hue and chroma.
For the glossmeter, the average of the 3 measurements taken on each sample at each time point was used. A paired t test was used to determine whether there was a significant change in the readings at day 28 compared with pretreatment.
For the Cielab color model investigation, the 3 measurements taken on each tooth were reduced to a consensus measure by taking an average of the 3 readings for each of the 3 parameters (L*, a*, and b*) for each tooth. Since the purpose of the study was to determine which material produces posttreatment enamel that is most esthetically similar to the predecalcified enamel, a difference score was created for the time points of pretreatment and day 28 by subtracting the pretreatment measurement from the consensus measurement at that time point. This difference score was calculated for each of the 3 parameters for each tooth. ANOVA was used to determine whether there were significant differences between the 4 treatments. The Tukey HSD test was used for the post-hoc analysis.
For surface smoothness, the average of the 3 profilometer measurements taken on each sample at each time point was used and compared with ANOVA and the Tukey HSD used for post-hoc analysis.
The ICC function from the psych library implemented in R version 2.11.1 (2010) was used for reliability analysis. Software (version 9.1; SAS, Cary, NC) was used for all other data analysis.
The subjective evaluation of the pretreatment photographs compared with the day-28 photographs proved to be unreliable for both the dental personnel and nondental personnel groups. For dental personnel, the ICC (95% confidence interval [CI]) was 0.083 (0.004, 0.200). Although it was low, the ICC for the dental personnel was at least significantly higher than zero ( F 44,176 = 1.6; P = 0.015) statistically speaking. For the nondental personnel, the ICC (95% CI) was 0.061 (–0.025, 0.19), which was not significantly higher than zero ( F 44,176 = 1.3; P = 0.092). The intrarater agreement for the nondental personnel was somewhat higher than that for the dental personnel ( Table I ).
Group 1 (control) samples were significantly darker ( P = 0.031) at day 28 compared with pretreatment, but no other group showed a significant change in the value measurement. There was a marginal (but not statistically significant) difference ( P = 0.062) between the groups’ value changes. Group 1 had no samples that became lighter, and group 4 (MI Plus) only had 1 sample that became lighter. Group 2 (Restore) had equal numbers of samples that became lighter and darker. Group 3 (Prevident) was the only group that had more samples become lighter than darker. Half of the samples (5 of 10) in both groups 1 and 3 had changes in hue ( P = 0.063, testing for a significant change in hue for both groups 1 and 3) compared with 30% (3 of 10) in group 2 and 10% (1 of 10) in group 4 ( P = 1.000 and P = 0.125, testing for significant changes in hue for groups 2 and 4, respectively). There was no significant difference between the 4 groups’ change in hue ( P = 0.194).
Group 3 tended to show the greatest change in glossmeter readings ( Table II ), but the groups’ differences were not statistically significant ( R 2 = 0.461; F 3,12 = 3.4; P = 0.053).