EVALUATION OF WEIGHT LOSS AND SURFACE ROUGHNESS OF COMPOMERS AFTER SIMULATED TOOTHBRUSHING ABRASION TEST

1DDS, MSc, PhD, Assistant Professor, Department of Operative Dentistry, Endodontics and Dental Materials, Bauru Dental School, University of São Paulo, Bauru, Brazil. 2DDS, MSc, PhD, Assistant Professor, Department of Operative Dentistry, University North of Paraná (UNOPAR), Londrina, Brazil. 3DDS, MSc, PhD, Graduate student (Doctor degree), Department of Operative Dentistry, Endodontics and Dental Materials, Bauru Dental School, University of São Paulo, Bauru, Brazil. 4-DDS, MSc, PhD, Associate Professor, Department of Operative Dentistry, Endodontics and Dental Materials, Bauru Dental School, University of São Paulo, Bauru, Brazil.


INTRODUCTION
The association of composite resins and glass ionomer cement components determined a category of restorative materials called compomers 7,15 . Fluoride release, good biocompatibility, adhesion to enamel and dentin are some of their benefits. They are light cured, easy to handle and present superior strength when compared to conventional glass ionomer cements. Compomer is classified as a resin composite modified by polyacid, because it does not present acid-base reaction like the ionomer cements 20 .
Wear is defined as progressive loss of substance from the surface of a material caused by a mechanical action 10 . Clinically, wear of restorative materials can result in loss of contour, increase in surface roughness, staining and plaque retention 27 . Thus, evaluation of the mechanical properties 1,4,7,12,15,26,27,28,30 of restorative materials is necessary to ascertain their indications and limitations. Some variables of compomers' composition can create different wear rates, such as curing method, monomer system, particle size, filler treatment and the effect of aging 3 .
"In vitro" toothbrushing test is effective to achieve rapid comparative data and is reliable to evaluate the wear resistance of restorative materials under specific standardized conditions 29,32 . "In vivo" methods suffer from certain disadvantages, which include lack of reproducibility and time involved. The aim of this study was to evaluate the wear resistance of five commercial compomers, Dyract (Dentsply), Dyract AP (Dentsply), Compoglass F (Ivoclar/ Vivadent), Freedom (SDI), F2000 (3M ESPE), compared to the resin composites Z100 (3M ESPE) and Silux Plus (3M ESPE) in relation to their weight loss and surface roughness alteration after simulated toothbrushing abrasion test. The null hypotheses were that there would be no differences in weight loss and no significant changes in surface roughness after simulated toothbrushing test.

MATERIAL AND METHODS
The materials utilized in this study are presented in Table  1. Ten cylindrical specimens for each commercial brand were prepared with 5mm diameter and 3mm thickness. The specimens were obtained from a silicon mold, which was filled in a single increment with the materials and covered with a polyester matrix. They were polymerized through the matrix for 60 seconds on each side with a light intensity between 450 to 500 mW/cm 2 by the curing unit XL 3000 (3M ESPE). Afterwards, they were polished using Sof-Lex (3M ESPE) discs. Then, specimens were ultrasonically cleaned for 10 minutes and individually stored in deionized water at 37 o C for 24h until first initial weight measurement.
The initial weight measurement was done on an analytical balance with 0.0001g accuracy (Sartorius-Werke A.G., Germany). Each specimen was dried with absorbent paper to remove excess water. The samples were weighted every 24 hours during 2 weeks until they reached a constant weight on three consecutive days of measurements. The mean of the last three measurements was considered the baseline weight. The superficial roughness analysis was determined using a Hommel Tester T 1000 roughness meter (Hommel Tester T 1000 -Hommelwerke). They were expressed in Ra values (Roughness average-µm). For each tested specimen surface, tracings were performed for five randomized directions in different locations. Baseline roughness was obtained by the arithmetic mean of these readings. The toothbrush abrasion machine was adapted by Vieira 31 (1960), with the cycle's speed set at 374 strokes per minute. Soft nylon bristles toothbrushes heads (Colgate Classic TM , Colgate-Palmolive Co., Osasco, São Paulo, Brazil) were adjusted to the toothbrushing simulation machine. The amplitude of movement was 3.8cm with 200 grams of weight. The specimens were submitted to 100,000 strokes of toothbrushing, performing a total of four hours and fortyfive minutes to each group of material. This period is correspondent to 4.2 years of toothbrushing 6 . Slurry was prepared with 50 grams of Colgate MFP (Colgate-Palmolive Co., Osasco, São Paulo, Brazil) toothpaste and 100 grams of distilled water, according to the ISO specification 8 . During the test, slurry was constantly renewed for each 10,000 strokes. The slurry pH was checked during the test and presented the value of 8.6 without changes. After 50,000 strokes the specimens were moved from the right to the left side in the testing machine. Following the test, the specimens were removed and ultrasonically cleaned with water for 10 minutes. Final weight and roughness measurements were made according to initial measurement protocols. Weight loss was obtained by the percentage alteration between initial and final measurements. Roughness alteration was observed by the percentage difference between baseline and final reading means 23

RESULTS
The Student paired t-test indicated that, after the abrasion test, all materials suffered significant weight loss ( Table 2) and roughness increase ( Table 3) (p<0.05). Table 2 presents the results of materials weight loss and Figure 1 graphically depicts the weight loss percentage for each experimental group. For all compomers, weight loss was higher than resin composites. Freedom and Dyract AP presented the lowest wear among compomers. There were no differences between Z100, Silux Plus and Freedom. F2000 presented the worst abrasion resistance results, without statistical differences with Dyract (p<0.05). The results of surface roughness alteration after toothbrushing can be observed in Table 3 and Figure 2. Dyract, Dyract AP, Z100, Compoglass F and Silux Plus showed the lowest surface roughness alteration, in increasing order, without differences between them. F2000 and Freedom showed the statistically highest alteration (p<0.05) when compared to other materials, presenting statistical differences between them. Table 2 and Figure 1 present the weight changes of tested materials. All materials presented significant weight loss after the abrasion test. Although Z100 was the most resistant material (1.27%), it was not statistically different from Silux Plus (2.20%) and the compomer Freedom (2.86%). Neither there was any statistically significant difference between the compomers Freedom, Dyract AP, Compoglass F and Dyract. The weight loss of F2000 was statistically higher than the weight loss of other compomers, showing no statistical differences with Dyract. The weight loss of Z100 composite was expected to be lower than the compomers 3,11,12,14,16,17,18,21,22 as they present 71% of filler weight, protecting the organic matrix from abrasion 6 . Clinically, this material also presents a better performance (compared to compomers) and is recommended to be utilized in areas with high masticatory load. In this study, 100,000 brushing cycles were done, corresponding to approximately 4.2 years of "in vivo" brushing 5 . Taking into consideration that Dyract AP and Freedom presents 73% and 77% of filler weight, the good performance of Silux Plus must also be observed as this material presents only 52% of inorganic filler by weight.

DISCUSSION
Wear of the organic matrix, exposure of inorganic particles and loss of the filler particles probably explain the mechanism of abrasion in composites 25 . The toothbrushing abrasion test provides contact between toothbrush bristles and dentifrices 2,3,6,9,13 but does not include all mechanisms of wear which restorative materials are subjected to in the oral cavity. Other types of wear are related to abrasion with foodstuff, load in occlusal areas of teeth and erosion 24 . They can act on organic matrix, on the filler particles or on the matrix/filler interface 24 . Additionally, in high cariogenic challenge conditions, acidic pH is established, leading to a decrease in wear resistance due to the abrasion test. This is in accordance with the findings of Attin, et al. 1 (1998), in which the compomers Dyract and Compoglass F presented higher levels of wear when submitted to brushing with acidic solution when compared to neutral pH solution.
For the aforementioned reasons, the present test is not  able to simulate the wear performance that happens "in vivo", but it is effective to provide rapid comparative data regarding abrasion wear resistance of restorative materials under specific standardized conditions 19 . For instance, in the present study Z100 presented 1.27% of weight loss and Silux Plus 2.20%, which did not differ statistically. However, when the two materials are clinically compared, the microfiller composites (Silux Plus) do not present a superior performance as the hybrid composites in posterior teeth.
On the other hand, they present low roughness and are largely used for anterior teeth and subgingival restorations. Beyond tested compomers, although F2000 presents 84% of filler weight, it showed the highest weight loss (5.57%). Freedom and Dyract AP had the lowest weight alteration and present 77% and 73% of filler weight respectively; these rates demonstrate that not always compomers with high percentage of inorganic filler in weight presents high abrasion resistance. Other composition characteristics must be taken into account, such as filler type/size and the silane treatment. The speculation that the deficient treatment of compomer filler particles, which connect them to the organic matrix, is one of the reasons to decrease the wear resistance of these materials must be considered. All compomers presented higher values of weight loss than composites.
The one-year clinical evaluation of the compomer Dyract in class I and II restorations in primary molars demonstrated that this material presented mean wear of 100mm in the first 6-month evaluation and 90mm in the following 6 months, totalizing a mean wear of 190mm in one year 26 . The authors observed loss of occlusal anatomy and proximal contacts. Hse and Wei 7 (1997) observed similar results. This study also evaluated Dyract's clinical behavior in primary molars after one-year control. The authors showed that the restorations presented marginal discoloration and occlusal wear higher than hybrid resin composite restorations.
The roughness alteration results of the studied materials are shown in Table 3 and Figure 2. All materials were subjected to the same initial polishing treatment and presented a significant increase in surface roughness mean, but in different rates. The materials roughness alteration values were: Dyract (30%), Dyract AP (35%), Z100 (41%), Compoglass F (63%) and Silux Plus (77%). They were not statistically different. Comparable results to roughness changes after toothbrushing abrasion test were also observed by Gladys, et al. 4 (1997), with similar roughness alteration for Dyract and the resin composites Silux Plus and Z100, being statistically different from a conventional ionomer cement or resin-modified ionomer.
A compomer (Dyract), a resin-modified ionomer cement (Fuji II LC) and a hybrid resin composite (Pekafill) were clinically compared and evaluated for three years, in terms of marginal integrity, color alteration and surface roughness by Van Dikjen 30 (1996). After this period, the author observed that the resin composite presented the best marginal adaptation and the smallest roughness alteration. Dyract presented low color alteration and intermediate roughness alteration among materials. Fuji II LC presented the highest color and roughness changes.
In this present study, Z100 did not present great differences regarding the roughness changes when compared to Silux Plus and some of the tested compomers (Dyract, Dyract AP and Compoglass F). These materials can be indicated for class V cavities, non-carious cervical lesions and areas adjacent to periodontal tissues. Regarding Freedom, it yielded statistically higher alterations than all tested materials. This can lead to more plaque accumulation, irritating periodontal tissues if inserted near them. According to the manufactuer, F2000 presents 84% of filler particles by weight and also showed high weight loss (5.57%). The  indication of this material must be carefully revised, as the reason for this increased roughness may be the higher dislodgement of inorganic particles, determining low abrasion resistance. Long-term clinical evaluations are necessary to better indicate this category of materials that were introduced in the market to combine the advantages of resin composites and ionomer cements.