The aim of this study was to evaluate the Knoop microhardness (KHN) and chemical composition of high-viscous glass-ionomer cement (HVGIC) after 10 years of clinical service.
Six HVGIC samples were cut from 10-year ART restorations. The sections were embedded in acrylic moulds with their longitudinal profile exposed. KHN was determined by performing three sequences of five indentations at 10, 30, 50, 70 and 90 μm of HVGIC outer surface. For the control group ( n = 6), HVGIC specimens were stored in distilled water for 24 months. Hardness measurements were taken at days 7, 30, 60, 120, 180, 360, and 720. For chemical analysis using SEM-EDX, 10-year and control specimens were dehydrated and coated with carbon. Data were analysed using T -test and ANOVA/Tukey’s test ( p < 0.05).
A significant KHN increase was observed in the control group up to the 180-day period. From this point the values stabilized and no more significant differences were found between the 10-year and the control KHN values. No statistical differences were observed amongst the KHN from inner distances compared to the outer surface of the 10-year HVGIC specimens. In one 10-year specimen, SEM-images identified the transformation of HVGIC in an altered layer with no glass filler particles detectable, and raised Ca, K and P contents.
KHN values of ten-year HVGIC specimens were similar to the control group values at 180-day storage period. Except for one 10-year specimen in that an altered layer could be seen, chemical composition was similar amongst the depths evaluated.
High viscosity glass-ionomer cement (HVGIC) is the material of choice for atraumatic restorative treatment (ART) due to properties such as biocompatibility, chemical adhesion to tooth structures, and fluoride release and uptake. The effectiveness of HVGIC as sealants and single-surface ART restorations is already well supported. However, for multi-surface ART restorations, the survival rates are less satisfactory, and a variable performance has been reported. Therefore, it is essential to investigate the mechanical properties of HVGICs indicated for ART for a better understanding of the clinical behaviour of such material.
Surface hardness is regarded as an important property to predict the clinical performance of a restorative material, and to assess the interaction between the material and the medium in which it is found. Hardness refers to the plastic deformation of a solid material when a force is applied. In dentistry it is commonly used as a parameter for restorations under occlusal stress, and relates to functional parameters such as resistance and wear. Additionally, a strong explanatory power between HVGIC Knoop hardness and compressive and flexural strength has been reported.
HVGIC may show a variable performance under different storage conditions. Although water is the most employed storage medium, current studies attempt to simulate oral conditions (pH-cycling; demineralization and remineralization solutions) on GICs.
However, in vitro studies do not effectively reflect what actually occurs in the oral environment, and in situ studies often have short-term results. Van Duinen et al. reported that under oral conditions, glass-ionomer gradually changes into a harder and smoother enamel-like structure with increased calcium and phosphate content. However, in vivo long-term hardness data have not yet been reported in the scientific literature.
The aim of this study, therefore, was to evaluate the effect of the oral environment (long-term in clinical service) on the surface microhardness and chemical composition of HVGIC.
Materials and methods
The study protocol was approved by the Bauru School of Dentistry Institutional Review Board (# 000115/2008). Informed consent was granted.
In a randomized clinical trial conducted ten years ago, 274 single- and multiple-surface HVGIC posterior restorations were performed as part of the restorative strategy of a prenatal-oral-health-care programme directed at 43 high-caries active pregnant patients (decayed teeth = 9.8 ± 5.5). The cavities were restored with hand-mixed HVGIC Fuji IX (GC Dental Co., Tokyo, Japan) according to the ART approach and manufacturer’s recommended powder/liquid ratio. After 10 years, 129 posterior restorations could be evaluated showing a success rate of 58.1%. Twenty-eight restorations (21.7%) had been replaced by another restorative material, and 26 restorations (20.2%) were recorded as clinical failures. Considering the failed restorations, six HVGIC samples (biopsies) were cut from them before replacement, and prepared for hardness test and chemical analysis, thus providing HVGIC specimens that were under the oral conditions of six different individuals.
The related teeth were isolated under a rubber dam. The specimens were cut in bulk near the interface by using a diamond bur (# 3195 – KG Sorensen, São Paulo, SP, Brazil) with water coolant. Three of the six specimens showed enamel presence. The fragments were maintained in position with clinical tweezers until their dislodgment, and positioned in a wax matrix with identification of the surface exposed to the oral environment. Then, the sections were embedded in an acrylic resin according to the demarcated position with the profile (inner to outer material surface) exposed.
The cut surfaces were ground flat with water-cooled-silicon-carbide papers (320-, 600-, and 1200-grade paper; Buehler, Lake Bluff, IL, USA), and polished with felt paper, wet with diamond spray (1 μm; Buehler, Lake Bluff, IL, USA), in order to obtain a flat surface necessary for the hardness measurement. Prior to the test, the specimens were sonically cleaned for 5 min.
The hardness was measured by using a Knoop diamond (KHN), under a load of 25 g, applied for 10 s (HMV-2 microhardness tester; Shimadzu Corporation, Tokyo, Japan). Three sequences of five indentations were performed. The first sequence of indentations was made in the central region of the specimen, and the other two at a 100-μm-distance to both sides of the central row of indentations. The indentations were made at 10, 30, 50, 70, and 90 μm from the outer surface, totalizing 90 indentations (15 in each specimen).
For chemical analysis, the 10-year specimens were dehydrated in a graded series of ethanol (25%, 50%, 75%, 90%, 99.8%), dried in hexamethyldisilazane (HMDS) for 15 min, and vacuum-desiccated for 24 h. The specimens were mounted on aluminium stubs, and sputter-coated with carbon. The coated specimens were analysed with a Scanning Electron Microscope/energy dispersive X-ray analysis (SEM/EDX) – (a LEO stereo 440 scan, with an energy-dispersive X-ray detector with Ge crystal and INCA software, 20 kV). EDX data were collected from several cement regions. The measurements were made either in the cement matrix or in the glass particle. A set of three lines perpendicular to the outer surface was performed, with measurements at 10 and 50 μm of distance from the outer surface of HVGIC, totalizing thus 6 measurements at fixed distances in each specimen. In the specimen whose surface was altered (altered layer), 6 additional measurements were made in different areas of this altered layer (AL).
For the control group ( n = 6), the same commercial HVGIC (Fuji IX) was hand-mixed according to the manufacturer’s instructions using the recommended powder/liquid ratio, and inserted in metal moulds (3 mm thick, and 6-mm in diameter). The moulds were mounted on top of a glass plate, then covered with acetate strips, compressed with a 0.5 kg weight, and allowed to set for 15 min. Subsequently, cement excesses were removed, the surfaces protected with petroleum jelly, and the specimens were stored in distilled water at 37 °C. After 7 days, the HVGIC specimens were embedded in acrylic moulds, and polished following the same protocol as that described for in vivo specimens. Hardness was determined by performing fifteen indentations in each specimen surface, totalizing 90 indentations. A distance of 20 μm was maintained between each one (KHN, 25 g, 10 s). Hardness measurements were taken at 7, 30, 60, 120, 180, 360, and 720 days after the beginning of the storage. The indentations were not made in depth.
Six additional HVGIC control specimens were prepared for chemical analysis after 7 days of water storage. EDX data were collected from several cement regions (punctual beam). The measurements were made either in the cement matrix or in the glass particle, totalizing 6 measurements in each specimen. Additionally, the powder of the glass ionomer was analysed using the X-ray fluorescence method (FRX) (Axios Advanced Spectrophotometer) with a wave-length detector and Super-Q software.
Variance equality and normal distribution of the data were tested for all the variables using the Bartlett and Kolmogorov-Smirnov tests, respectively (Statistica v. 9.1, StatSoft Inc., Tulsa, USA). All data showed homoscedasticity and normal distribution. KHN values were analysed using ANOVA/Tukey’s test for water storage time (control) and for the distance of the outer surface in contact with the oral environment (10-year specimens). A difference was regarded as statistically significant if p -value ≤ 0.05. Differences between 10-year × control KHN values were analysed by the t -test using the Bonferroni correction. In this analysis, a difference was considered significant if p -value ≤ 0.007. The chemical composition was expressed in a weight percentage, and analysed by ANOVA/Tukey’s test (significant if p ≤ 0.05).
The mean KHN values and standard deviations (SD) are presented in Table 1 .
|7 Days water||30 Days water||60 Days water||120 Days water||180 Days water||360 Days water||720 Days water||10 Years oral environment|
|91.7 a (±2.6)||102.8 a (±4.0)||117.3 b (±13.9)||116.4 b (±8.4)||135.7 c (±4.6)||128.6 c (±6.9)||139.3 c (±5.3)||146.4 c (±7.2)|
In the control group, the time of water storage had a significant impact on HVGIC surface microhardness. Fuji IX HVGIC showed a significant increase in the KHN values up to the 180th day and from this time point on, the values showed no significant variations until the final measurements were taken at day 720. The 10-year specimens were significantly harder than the control specimens at 7-, 30-, 60-, and 120-day measurements. At 180 days, the mean KHN values observed in the control group were not significantly different from those obtained for the 10-year specimens. In the 10-year specimens, no statistical differences were observed amongst the distances from the outer surface in contact with the oral environment ( Table 2 ).
|Distance from outer surface||Mean|
|10 μm||30 μm||50 μm||70 μm||90 μm|
|148.0 a (±4.9)||151.4 a (±16.0)||146.0 a (±9.7)||142.8 a (±5.5)||143.9 a (±4.4)||146.4 a (±7.2)|
The main SEM-EDX spectrum detected is shown in Table 3 . The measurements were made either in the cement matrix or in the glass particle, and an average value was expressed in Table 3 . According to the manufacturer, Fuji IX contains stroncium (Sr) glass rather than the conventional calcium (Ca) glass, which was confirmed by the chemical analysis of the control specimens by EDX and FRX methods that showed only traces of Ca, Tables 3 and 4 . The mineral profile observed in the glass particle of the 10-year specimens is shown in Fig. 1 a , with no presence of Ca. However, the presence of Ca was observed in the cement matrix of the 10-year specimens, as shown in Fig. 1 b.
|Control HVGIC ( n = 36)||4.6 a (±1.9)||17.5 a (±2.0)||15.6 a (±2.6)||2.7 a (±0.6)||0.0 a||0.1 a (±0.3)||13.2 a (±3.8)|
|10-year HVGIC ( n = 36)||5.1 a (±2.6)||16.9 a (±2.0)||15.8 a (±2.1)||2.7 a (±0.9)||1.1 b (±0.5)||3.7 b (±2.9)||11.1 b (±4.4)|
|(AL) ( n = 6)||3.0 a (±2.7)||11.8 b (±3.4)||9.2 b (±4.3)||7.5 b (±1.1)||2.5 c (±0.9)||15.9 c (±6.2)||9.3 b (±2.3)|
|Al 2 O 3||27.77|
|P 2 O 5||4.99|
|Na 2 O||2.10|
|Fe 2 O 3||0.06|
|K 2 O||0.02|
In one 10-year sample with enamel presence, SEM-images identified physical and chemical alterations of the HVGIC adjacent to the enamel walls. This altered layer (AL) was 5–15 μm thick ( Fig. 2 ). In this layer, no glass-filler particles were detectable, and a distinct mineral profile was observed ( Fig. 1 c). The altered layer (AL) disclosed raised calcium (Ca), phosphorus (P), and potassium (K) contents, and a lower aluminium (Al), silica (Si), and strontium (Sr) percentage in comparison to the unaltered cement, Table 3 .
It has been demonstrated that glass-ionomer stored in human saliva has an improved surface hardness as compared to samples stored in water. The reasoning behind this may be that water storage causes only extraction of components, thus resulting in a reduction of strength, whilst storage in saliva can increase the mineral content of glass-ionomer. In the present study, inferior hardness values were observed in the control group stored in water up to the 120-day measurement in comparison with 10-year values. At the 180-day measurement, the control group values stabilized, and no more significant differences were found between the 10-year and control KHN values. This could be explained by the cement maturation process, and maybe by the fact that the water medium was not replaced during the 2 years of the experiment. Thus, homeostasis could have been reached during cement maturation, impairing therefore a continuing extraction of HVGIC components. Another interesting point to be investigated is at which moment the KHN values stabilized in vivo .
In the study by Okada et al., only a 40-day period was evaluated, and until then, the authors observed that from 7 to 40 days of storage in distilled water, the Fuji IX hardness values increased in 4%. In the present study, we observed an increase in KHN of 12% from 7 to 30 days, and of 40% from 7 to 360 days. Ellakuria et al. observed an increase in hardness values of 20% and 46% for Ketac-Molar at 30 and 365-day periods, respectively. However, it should be highlight that the type of GIC and the manipulation process might have influenced the changes in the hardness values overtime. Therefore, the studies should be compared with caution.
In the 10-year specimens we observed no differences in KHN values amongst the distances from the outer surface in contact with the oral environment, showing that the maturation process is homogenous after 10 years. For our knowledge this is the first study evaluating hardness at different depths of GIC restorations after 10 years of clinical service.
Regarding the chemical compositions, Okada et al. also investigated the Fuji IX HVGIC surface using X-ray photoelectron spectroscopy (XPS) and electron probe microanalysis (EPMA). The authors verified the presence of Ca, P, and K in the surface and subsurface of HVGIC only in the sample stored in human saliva. It was suggested that Ca and PO 4 ions from saliva may be adsorbed onto the surface of the cement, which is in accordance with our 10-year data. The glass-ionomer setting reaction was not completed immediately, and it is likely that saliva components react with the glass-ionomer overtime. The authors inferred from their data that Ca, K and P elements penetrated into the glass-ionomer matrix. However, the incorporation of Ca and P may not be clinically significant, since no difference was observed in our study amongst the KHN values of the evaluated depths. However, in further studies it would be interesting to investigate the hardness and chemical composition of the GIC restorations at short-term intervals until final cement maturation.
It should be pointed out that the preparation of the 10-year samples with a diamond bur (biopsy) and the correct positioning of the specimens in the wax matrix were extremely difficult to achieve. The SEM-EDX analysis was performed as a preliminary study, and disclosed interesting data regarding HVGIC alterations in a long-term oral environment. An altered layer could be observed in only one specimen of three that had enamel, suggesting that in spite of the presence of saliva, the transformation is possible only at an enamel interface. This finding is in accordance with the observations made by Van Duinen et al., who detected glass-ionomer alterations at the border with the enamel fissure. The authors speculated that the glass particles were dissolved, and a smoother cement surface comparable to natural enamel could be noticed after long-term exposure to an oral environment.
Under high magnification, a zone of interaction between HVGIC and enamel could be observed along the interface in all the three samples with enamel presence ( Fig. 3 ). This observation was similar to that reported by Ngo et al. and probably represents the ion-exchange or absorption layer. This ion-exchange layer is currently considered a significant feature in material classification. Additional information observed in the present study is the long-term maintenance of this layer confirming the findings observed in short-term studies. However, the altered layer expressed in Fig. 2 show distinct appearance, and is wider (5–15 μm).
Ngo et al. suggested that the ion-exchange layer between glass ionomer and enamel was more resistant to acid etching. Thus, the altered layer reported in this study could represent an additional protection against caries and cement erosion.
In the present study, each specimen was donated by a different voluntary, but the chemical analysis comparing the components of the specimens with enamel showed a similar profile, except for a superior fluoride (F) content in the sample with the altered layer. The analysis of the altered layer disclosed calcium and phosphate uptake, and loss of silica and aluminium therein. This finding may represent a further contribution to the “intermediate layer” theory postulated by Wilson et al., but its clinical significance is not conclusive so far. Van Duinen et al. observed more calcium and phosphate at the saliva-exposed side than in the deeper areas of a little material fragment. In the present study regarding the chemical profile and the KHN values, no difference could be observed between the saliva-exposed side (subsurface) and deeper HVGIC areas. For better understand of this process, the saliva and enamel content of the patients should be further investigated.
Wang et al. observed higher hardness values in HVGIC stored in acidic solutions (pH 3) with increased phosphate content. The authors suggested that high calcium and phosphate rates may have positive effects on the surface hardness of GICs. In the study of Wang et al. the specimens stored at pH 3 had lower hardness, but this could be overcome by an increase level of environmental phosphate. A microscopic surface reaction layer was observed in specimens stored at pH 3. This reaction layer is due to erosion of the cement and its thickness and composition varied depending on environmental phosphate levels. At higher phosphate levels a distinctive outermost layer dense structure rendered HVGIC less soluble under acidic conditions. The results suggest that the existence of this dense layer was critical for increase in the mechanical properties.
Thus, we can speculate that the reaction layer or altered layer observed in the current study could be seen in the specimen whose oral environment matched the ionic and pH ideal conditions.
Studies evaluating long-term surface hardness of HVGIC are rare, and data related to restorations that have been in clinical service for long periods have not been established in the literature. The long-term results obtained here could be regarded as representative, and can be useful in identifying appropriate mechanical characteristics of HVGICs, mainly in stress-bearing clinical situations such as multi-surface restorations.
For the conditions tested in this study, it was concluded that a period of 180-day water storage could simulate the maturation of the setting process of glass-ionomer according to the hardness measurements. Further studies evaluating long-term saliva and water storage with cariogenic challenge and fluoride therapy, and the correspondence of the setting process of glass-ionomer within the mouth and in vitro should be performed using different GICs.
Within the limitations of this study, KHN of 10-year HVGIC was similar to the values obtained at 180-day of water storage. Except for one 10-year specimen in that an altered layer could be seen, chemical composition was similar amongst the depths evaluated.
The authors would like to thank the volunteers for their diligence in performing this study.
The authors would like to thank the Laboratory of Technologic Characterization of the University of São Paulo (LCT – POLI – USP) for assistance in the SEM/EDX analysis.
The authors would like to thank Angela Christine Charity for the English revision.
This investigation was supported by the Fundação de Apoio à Pesquisa do Estado de São Paulo – FAPESP (t he São Paulo State Foundation for Research Support , Grant 2007/04276-0.353 .
1. Frencken J.E., Pilot T., Songpaisan Y., Phantumvanit P.: Atraumatic restorative treatment (ART): rationale, technique, and development. Journal of Public Health Dentistry 1996; 56: pp. 135-140.
2. Frencken J.E, Wolke J.: Clinical and SEM assessment of ART high-viscosity glass-ionomer sealants after 8–13 years in 4 teeth. Journal of Dentistry 2010; 38: pp. 59-64.
3. Chen X., Cuijpers V., Fan M., Frencken J.E.: Marginal leakage of two newer glass-ionomer-based sealant materials assessed using micro-CT. Journal of Dentistry 2010; 38: pp. 731-735.
4. Frencken J.E., van’t Hof M.A., van Amerogen W.E., Holmgren C.J.: Effectiveness of single-surface ART restorations in the permanent dentition: a meta-analysis. Journal of Dental Research 2004; 83: pp. 120-123.
5. van’t Hof M.A., Frencken J.E., van Palenstein Helderman W.H., Holmgren C.J.: The atraumatic restorative treatment (ART) approach for managing dental caries: a meta-analysis. International Dental Journal 2006; 56: pp. 345-351.
6. Frencken J.E., van’t Hof M.A., Taifour D., Al-Zaher I.: Effectiveness of ART and traditional amalgam approach in restoring single-surface cavities in posterior teeth of permanent dentitions in school children after 6.3 years. Community Dentistry and Oral Epidemiology 2007; 35: pp. 207-214.
7. Mickenautsch S., Yengopal V., Banerjee A.: Atraumatic restorative treatment versus amalgam restoration longevity: a systematic review. Clinical Oral Investigations 2010; 14: pp. 233-240.
8. van Gemert-Schriks M.C.M., van Amerogen W.E., ten Cate J.M., Aartman I.H.A.: Three-year survival of single- and two-surface ART restorations in a high-caries child population. Clinical Oral Investigations 2007; 11: pp. 337-343.
9. Cefaly D.F.G., Barata T.J.E., Bresciani E., Fagundes T.C., Lauris J.R.P., Navarro M.F.L.: Clinical evaluation of multiple-surface ART restorations: 12-month follow-up. Journal of Dentistry for Children 2007; 74: pp. 203-208.
10. Farag A., van der Sanden W.J.M., Abdelwahab H., Mulder J., Frencken J.E: 5-Year survival of ART restorations with and without cavity disinfection. Journal of Dentistry 2009; 37: pp. 468-474.
11. Zanata R.L., Fagundes T.C., Freitas M.C.C.A., Lauris J.R.P., Navarro M.F.L.: Ten-year survival of ART restorations in permanent posterior teeth. Clinical Oral Investigations 2011; 15: pp. 265-271.
12. Bonifacio C.C., Kleverlaan C.J., Raggio D.P., Werner A., de Carvalho R.C.R., van Amerogen W.E.: Physical–mechanical properties of glass-ionomer cements indicated for atraumatic restorative treatment. Australian Dental Journal 2009; 54: pp. 233-237.
13. Ellakuria J., Triana R., Minguez N., Soler I., Ibaseta G., Maza J., et. al.: Effect of one-year water storage on the surface microhardness of resin-modified versus conventional glass-ionomer cements. Dental Materials 2003; 19: pp. 286-290.
14. Okada K., Tosaki S., Hirota K., Hume W.R.: Surface hardness change of restorative filling materials stored in saliva. Dental Materials 2001; 17: pp. 34-39.
15. Wang X.Y., Jin A.Y.U., Ngo H.C.: Influence of environmental calcium/phosphate and pH on glass-ionomers. European Journal of Oral Sciences 2007; 115: pp. 224-229.
16. Wang X.Y., Yap A.U.J., Ngo H.C., Chung S.M.: Environmental degradation of glass-ionomer cements: a depth-sensing microindentation study. Journal of Biomedical Materials Research Part B: Applied Biomaterials 2007; 82B: pp. 1-6.
17. Silva K.G., Pedrini D., Delbem A.C.B., Cannon M.: Microhardness and fluoride release of restorative materials in different storage media. Brazilian Dental Journal 2007; 18: pp. 309-313.
18. Kirsten G.A., Takahashi M.K., Rached R.N., Giannini M., Souza E.M.: Microhardness of dentin underneath fluoride-releasing adhesive systems subjected to cariogenic challenge and fluorie therapy. Journal of Dentistry 2010; 38: pp. 460-468.
19. Basting R.T., Serra M.C., Rodrigues A.L.: In situ microhardness evaluation of glass-ionomer/composite resin hybrid materials at different post-irradiation times. Journal of Oral Rehabilitation 2002; 29: pp. 1187-1195.
20. Van Duinen R.N.B., Davidson C.L., De Gee A.J., Feilzer A.J.: In situ transformation of glass-ionomer into an enamel-like material. American Journal of Dentistry 2004; 17: pp. 223-227.
21. Zanata R.L., Navarro M.F.L., Pereira J.C., Franco E.B., Lauris J.R.P., Barbosa S.H.: Effect of caries preventive measures directed to expectant mothers on caries experience in their children. Brazilian Dental Journal 2003; 14: pp. 75-81.
22. Zanata R.L., Navarro M.F.L., Barbosa S.H., Lauris J.R.P., Franco E.B.: Clinical evaluation of three restorative materials applied in a minimal intervention caries treatment approach. Journal of Public Health Dentistry 2003; 63: pp. 221-226.
23. Dowling A.H., Fleming G.J.P.: Are encapsulated anterior glass-ionomer restoratives better than their equivalents?. Journal of Dentistry 2009; 37: pp. 133-140.
24. Brito C.R., Velasco L.G., Bonini G.A.V.C., Imparato J.C.P., Raggio D.P.: Glass ionomer cement hardness after different materials for surface protection. Journal of Biomedical Materials Research Part A 2010; 93A: pp. 243-246.
25. Luo J., Billington R.W., Pearson G.J.: Kinetics of fluoride release from glass components of glass ionomers. Journal of Dentistry 2009; 37: pp. 495-501.
26. Ngo H., Mount G.J., Peters M.C.R.B.: A study of glass-ionomer cement and its interface with enamel and dentin using a low-temperature, high-resolution scanning electron microscopic technique. Quintessence International 1997; 28: pp. 63-69.
27. Wilson A.D., Possner H.J., Powis D.R.: Mechanisms of adhesion of polyelectrolyte cements to hydroxyapatite. Journal of Dental Research 1983; 62: pp. 590-592.
28. Sennou H.E., Lebugle A.A., Grégoire G.L.: X-ray photoelectron spectroscopy study of the dentin-glass ionomer cement interface. Dental Materials 1999; 15: pp. 229-237.
29. Marquezan M., Fagundes T.C., Toledano M., Navarro M.F.L., Osorio R.: Differential bonds degradation of two resin-modified glass-ionomer cements in primary and permanent teeth. Journal of Dentistry 2009; 37: pp. 857-864.
30. Fagundes T.C., Toledano M., Navarro M.F.L., Osorio R.: Resistance to degradation of resin-modified glass-ionomer. Journal of Dentistry 2009; 37: pp. 342-347.
31. Cardoso M.V., Delmé K.I.M., Mine A., Neves A.A., Coutinho E., De Moor R.J.G., et. al.: Towards a better understanding of the adhesion mechanism of resin-modified glass-ionomer by bonding to differently prepared dentin. Journal of Dentistry 2010; 38: pp. 921-929.
32. Mount G.J., Tyas M.J., Ferracane J.L., Nicholson J.W., Berg J.H., Simonsen R.J., et. al.: A revised classification for direct tooth-colored restorative materials. Quintessence International 2009; 40: pp. 691-697.
33. Scholtanus J.D., Huysmans M.C.: Clinical failure of class-II restorations of a highly viscous glass-ionomer material over a 6-year period: a retrospective study. Journal of Dentistry 2007; 35: pp. 156-162.
34. Burke F.J.T., Lucarotti P.S.K.: Re-intervention in glass-ionomer restorations: what comes next?. Journal of Dentistry 2009; 37: pp. 39-43.