J. Dent. 1990; 18: 243-249
243
Clinical evaluation of occlusal glass ionomer, resin, and amalgam restorations R. J. Smales, D. C. Gerke and I. L. White* Department of Dentistry, The University Service, Adelaide, South Australia
of Adelaide,
Adelaide,
South Australia
and *South Australian
Dental
ABSTRACT The purpose of the study was to evaluate four materials (a glass ionomer (polyalkenoate) silver cermet, two composite resin restoratives and a high copper content dental amalgam) placed in either conventional Class I cavities or in modified odontotomy-enameloplasty-sealant (OES) fissure preparations. One experienced operator inserted 438 occlusal, Class I restorations in the posterior permanent teeth of 124 patients in a private dental practice. Restorations were assessed for bulk loss of material, surface voids and cracking, restoration margin fractures and staining, and surface staining and roughness, by using colour transparencies taken at baseline and at recalls for up to 3 years. The glass ionomer cermet was the most difficult material to handle and also gave the least satisfactory clinical result. Loss of material and surface voids were common in the cerrnet restorations with surface cracking or crazing being seen in 11.4 per cent of the restorations, especially in the larger, conventional Class I preparations. One posterior resin was more viscous and difftcult to handle than the other resin and exhibited more surface voids. The amalgam alloy was used in Class I preparations only and showed more restoration margin fractures and surface staining than did the other three materials. However, there were no unsatisfactory clinical assessments given for either restoration margin fracture and staining, or surface staining and roughness for any of the materials. Patient acceptance of the modified OES fissure preparation was extremely good. KEY WORDS: J. Dent. 1990; 1990)
Polyalkenoate (glass ionomer) cermets, Posterior composites, Amalgam, Clinical trial 18: 243-249
(Received 13 February 1990;
reviewed
11 April 1990;
accepted 29 June
Correspondence should be addressed too:Mr R. J. Smales, Department of Dentistry, The University of Adelaide, Adelaide, South Australia 5000.
INTRODUCTION In recent years there has been considerable interest shown in the use of alternative dental restorative materials to amalgam alloys for the restorative management of occlusal pit and fissure caries, and in minimizing the amount of sound tooth substance removed during cavity preparation. Depending on the dentist’s philosophy of treatment methods, and on each individual clinical situation, measures for treating occlusal pit and fissure caries may include: no direct operative intervention (Eames 1988); the placement of resin sealants (Elderton, 1985); preventive resin restorations (Simonsen, 1978); fissure widening (McLean and Wilson, 1974; Le Bell and Forsten, 1980); modified odontotomy-enameloplasty-sealant @ES) restorations (Gerke, 1987); and conventional Class I amalgam restorations. There have been very few reports published comparing the behaviour and performance of high-copper amalgam, 0 1990 Butterworth-Heinemann 0300-5712/90/050243-07
Ltd.
posterior composite resin, and glass ionomer cermet restorative materials as fissure fillings or larger, Class I restorations (Gerke and Smales, 1989). The purpose of the present study was to compare the handling properties, behaviour and performance of two posterior composite resins and one glass ionomer cermet used as occlusal fissure restoratives and to restore conventional Class I preparations in which a high copper content dental amalgam was used as a control.
MATERIALS
AND METHODS
Between 1985 and 1987, one operator (D.C.G.) placed 438 occlusal restorations in 20 premolar and 418 molar teeth in 124 patients treated at a private dental practice in Adelaide, South Australia. Some 50 per cent of the restorations were placed in patients under 21 years of age, with 92 per cent of the restorations being placed in patients under41 years of age. Details of the four materials used are shown in Table I.
244
J. Dent
1990; 18: No. 5
Tab/e 1. Materials used in the studv Manufacturer
Composition
/&a/
Ketac-Silver (capsules)
ESPE GmbH, Seefeld, Oberbay, FRG
Polyalkenoate (glass ionomer), 40% silver cermet
August 1986
Visio-Molar (syringe)
ESPE GmbH
Quartz filler, 86% by weight
February 1986
P-30 (syringe) Dispersalloy (capsules)
3M, St Paul, MN, USA Johnson B Johnson, East Windsor, NJ, USA
Zinc glass, 84% by weight Admixture, 12% copper
May 1985
Material
The materials were placed sequentially as they became available, but early problems with the handling of VisioMolar for the smaller, modified OES fissure preparations, and later problems with the clinical performance of Ketac-Silver, led to a preference for using the less viscous P-30 resin. Because of the numerous clinical studies already published on the long-term behaviour of Dispersalloy, only a few amalgam restorations were included as Class I restoration controls. The choice between an OES preparation and a conventional Class I cavity was determined by the size of the carious lesion. Patients were selected solely on the basis of requiring treatment of Class I caries. Briefly, all narrow and sticky fissures were first widened minimally with a high speed tine flame-shaped diamond point. Any caries extending into dentine was then removed with small round steel burs at slow speed. No linings or bases were required as the depth of the preparations was minimal. For the composite resins, the enamel was etched with a 37 per cent phosphoric acid gel for 30 s, then washed and dried before placing the bonding resin and appropriate composite restorative. The bonding resin and restorative were light cured separately. For the cermet, the prepared fissure walls were first conditioned with a 10 per cent solution of polyacrylic acid for 10 s, then washed and dried before inserting the cermet. All materials were placed using metal instruments (Gerke, 1987) and in accordance with manufacturer’s instructions. For larger lesions, conventional Class I preparations were cut at high speed using small pear-shaped tungsten carbide burs. Any residual caries was removed as described previously. Occasionally, a glass ionomer cement base was placed in deeper preparations for resin and amalgam restorations. Insertion of the resins and glass-ionomer cermet was as for the fissure preparations. A cavity varnish was applied prior to placing the amalgam restorations. Apart from the amalgams, which were polished subsequently, all restorations were finished at the time of their insertion. Care was taken not to dehydrate the newly placed cermet restorations. Clinical photographs were taken at the time of finishing, and thereafter at 6-monthly intervals for periods of up to 3 years, using Ektachrome (Eastman Kodak Co., Rochester,
use
April 1986
NY, USA) 35 mm colour transparency film at 1 : 1 magnification. At each time point, bulk loss of material, and surface voids or exposed porosities were assessed as being either present or absent. Restoration margin fractures and discrepancies, marginal staining, surface staining or tarnishing, and surface roughness were assessed separately for all restorations by comparison of the photographs with four standard sets of enlarged 2~ colour transparencies, as has been described elsewhere (Smales and Creaven, 1985). Briefly, the four clinical factors assessed were rated by one examiner (R.J.S.) against separate linear scales, graded from 0 to 12,where 0 indicated no detection of the factor (staining, roughness, etc.) and 12 indicated that the factor was severe. A lack of sufficient colour contrast precluded rating the restorations of the two composite resin materials for margin fractures and discrepancies, and surface roughness. Each clinical factor was analysed separately using a mixed model analysis of variance, which included a random effects term for restorations, a covariate effect for restoration age, and a fixed effect for each material, the statistical significance of which was determined by a likelihood ratio test. Calculations were performed using the BMDP program 3V on a Vax 1l/785 computer (Dixon, 1988). Significant differences between the scored means over the study period were assumed to exist when there was an alpha probability of 0.05 or less. The failure rates of the restorations over the study period for the different materials were analysed using the BMDP life-table survival program 1L (Dixon, 1988). Repeat assessments were made (RJ.S.) of 30 restorations 2 months after the completion of the initial assessments. For bulk loss of material and for surface voids, weighted Kappa values were 0.61 and 0.64 respectively. For the four clinical factors, paired t tests showed no significant differences between the original and repeat assessments (P > 0.05).
RESULTS There were no significant differences in the distribution of the materials between the posterior teeth (P > 0.05). Standard Class I preparations accounted for 37 per cent of
et al.: Clinical evaluation
Smales
of occlusal
restorations
245
conjunction with Table ZZZ.Apart from surface roughness (P = 0.43) all the clinical factors were found to deteriorate with time over the study (P < 0.01). Although there were significant statistical differences found between the materials for the four clinical factors evaluated, no unsatisfactory clinical assessments were given. Table V shows the distribution of restoration observations for material retention. Losses of material usually occurred from over the cusp slopes and unprepared fissure extensions, especially with Ketac-Silver (14.5 per cent) (P < 0.001).
the restorations and the modified OES fissure preparations for 63 per cent (Table ZZ). The number of P-30 restorations placed reflected an increasing preference for this material as the study progressed. Ketac-Silver gave variable consistencies of mixes and working times. It was considered difficult to dispense this material into the small fissure preparations from the large straight nozzle of the capsules available at the time. Attempts to condense the cermet with metal instruments were also difficult as the material adhered to them and was pulled away from the cavity margins. The result was often an overfilled cavity with the material being smeared over the adjacent cusp slopes and any unprepared fissures. Visio-Molar was more viscous and difficult to handle than P-30, which was considered to be relatively easy to condense and get to flow into the narrow preparations. Table ZZZshows the mean scores over the study period of bulk loss of material, for surface voids or exposed porosities, and for the four clinical factors assessed in each case. Table ZV also shows the results of the asymptotic t-tests for the four clinical factors, to be read in
Table II. Number of restorations placed by size of occlusal preparations
Cavity size Material
Standard
Ketac-Silver Visio-Molar P-30 Dispersalloy
62 22 63 13
Modified
OES
:: 188 0
x2= 47.49, d.f.= 3, P < 0.01.
Table 111.Observation means + standard deviations over 3 years for the four materials placed Material Ketac-Silver Visio-Molar P-30 Dispersalloy P (from likelihood ratio test)
Bulk losses 0.22 0.01 0.02 0.03
f f f +
0.56 0.10 0.14 0.17
< 0.001*
Surface voids 0.23 0.11 0.04 0.00
~fr0.42 f 0.31 f 0.19 f 0.00
< 0.001*
Margin
fracture
0.64 f 0.76 1.35 + 1.25 0.002*
Margin stain
Surface stain
Surface rough
0.64 0.58 0.38 0.38
0.35 & 0.69 0.17 k 0.46 0.10+0.40 1.82 * 1.83
3.10 !I 1.31 1.47 f 1.26
< 0.001*
< 0.001*
k + + +
1.07 0.85 0.89 0.92
< 0.001*
*Significant difference between materials.
Table IV. Asvmptotic t-test scores of materials for predicted cell means* Material Bulk losses Visio-Molar P-30 Dispersalloy Surface voids Visio-Molar P-30 Dispersalloy Margin fracture Dispersalloy Margin staining Visio-Molar P-30 Dispersalloy Surface staining Visio-Molar P-30 Dispersalloy Surface roughness Dispersalloy
Ketac-Silver
-5.10 -8.95 -2.24
Visio-Molar
P-30
-0.58 - 1.29
- 1.87
1.37 0.68
1.70
1.87 -0.14
1.08
0.35 - 12.28
- 13.83
- 5.98 - 7.43 - 5.24 4.04
- 3.49
- 2.68 - 2.47 - 2.99 - 4.20 12.58 - 14.07
*Values of 2.0 or greater considered to be of statistical significance.
246
J. Dent
1990;
18: No. 5
Table VI shows the distribution of restoration observations for surface voids or exposed porosities. No voids were observed with Dispersalloy, but were seen frequently with Ketac-Silver (23 per cent) and to a lesser extent with the more viscous composite restorative Visio-Molar (10.9 per cent) (P < 0.001). Table VII shows the percentages of restoration observations with ideal rating assessments given for the various clinical characteristics examined over the study period. Minor restoration margin fractures and discrepancies, and surface staining or tarnish, typically limited, were seen most often with restorations of Dispersalloy, while some moderate surface roughness and limited margin staining was found with the restorations of Ketac-Silver. The findings for the restorations of Dispersalloy were similar to those reported previously @males and Gerke, 1984). Table VIII shows that 11.4 per cent of the Ketac-Silver restorations failed as a result of surface cracking or crazing, sometimes within 6 months of placement. In almost all instances, these problems were associated with the larger, Class I cavity preparations, rather than with the smaller, modified OES fissure preparations (P < 0.01).
Table V. Distribution of restorations for bulk of material retained, by number of observations made Bulk retention of material Intact Part lost
Material Ketac-Silver Visio-Molar P-30 Dispersalloy x2= 56.86, d.f.=
207
35 1 IO 1
5% 33 3, P < 0.001.
Table VI. Distribution of surface voids Surface voids (no. of observations) Material Ketac-Silver Visio-Molar P-30 Dispersalloy x2= 71.65,
d.f.=
3, P
243
Clinical evaluation of occlusal glass ionomer, resin, and amalgam restorations R. J. Smales, D. C. Gerke and I. L. White* Department of Dentistry, The University Service, Adelaide, South Australia
of Adelaide,
Adelaide,
South Australia
and *South Australian
Dental
ABSTRACT The purpose of the study was to evaluate four materials (a glass ionomer (polyalkenoate) silver cermet, two composite resin restoratives and a high copper content dental amalgam) placed in either conventional Class I cavities or in modified odontotomy-enameloplasty-sealant (OES) fissure preparations. One experienced operator inserted 438 occlusal, Class I restorations in the posterior permanent teeth of 124 patients in a private dental practice. Restorations were assessed for bulk loss of material, surface voids and cracking, restoration margin fractures and staining, and surface staining and roughness, by using colour transparencies taken at baseline and at recalls for up to 3 years. The glass ionomer cermet was the most difficult material to handle and also gave the least satisfactory clinical result. Loss of material and surface voids were common in the cerrnet restorations with surface cracking or crazing being seen in 11.4 per cent of the restorations, especially in the larger, conventional Class I preparations. One posterior resin was more viscous and difftcult to handle than the other resin and exhibited more surface voids. The amalgam alloy was used in Class I preparations only and showed more restoration margin fractures and surface staining than did the other three materials. However, there were no unsatisfactory clinical assessments given for either restoration margin fracture and staining, or surface staining and roughness for any of the materials. Patient acceptance of the modified OES fissure preparation was extremely good. KEY WORDS: J. Dent. 1990; 1990)
Polyalkenoate (glass ionomer) cermets, Posterior composites, Amalgam, Clinical trial 18: 243-249
(Received 13 February 1990;
reviewed
11 April 1990;
accepted 29 June
Correspondence should be addressed too:Mr R. J. Smales, Department of Dentistry, The University of Adelaide, Adelaide, South Australia 5000.
INTRODUCTION In recent years there has been considerable interest shown in the use of alternative dental restorative materials to amalgam alloys for the restorative management of occlusal pit and fissure caries, and in minimizing the amount of sound tooth substance removed during cavity preparation. Depending on the dentist’s philosophy of treatment methods, and on each individual clinical situation, measures for treating occlusal pit and fissure caries may include: no direct operative intervention (Eames 1988); the placement of resin sealants (Elderton, 1985); preventive resin restorations (Simonsen, 1978); fissure widening (McLean and Wilson, 1974; Le Bell and Forsten, 1980); modified odontotomy-enameloplasty-sealant @ES) restorations (Gerke, 1987); and conventional Class I amalgam restorations. There have been very few reports published comparing the behaviour and performance of high-copper amalgam, 0 1990 Butterworth-Heinemann 0300-5712/90/050243-07
Ltd.
posterior composite resin, and glass ionomer cermet restorative materials as fissure fillings or larger, Class I restorations (Gerke and Smales, 1989). The purpose of the present study was to compare the handling properties, behaviour and performance of two posterior composite resins and one glass ionomer cermet used as occlusal fissure restoratives and to restore conventional Class I preparations in which a high copper content dental amalgam was used as a control.
MATERIALS
AND METHODS
Between 1985 and 1987, one operator (D.C.G.) placed 438 occlusal restorations in 20 premolar and 418 molar teeth in 124 patients treated at a private dental practice in Adelaide, South Australia. Some 50 per cent of the restorations were placed in patients under 21 years of age, with 92 per cent of the restorations being placed in patients under41 years of age. Details of the four materials used are shown in Table I.
244
J. Dent
1990; 18: No. 5
Tab/e 1. Materials used in the studv Manufacturer
Composition
/&a/
Ketac-Silver (capsules)
ESPE GmbH, Seefeld, Oberbay, FRG
Polyalkenoate (glass ionomer), 40% silver cermet
August 1986
Visio-Molar (syringe)
ESPE GmbH
Quartz filler, 86% by weight
February 1986
P-30 (syringe) Dispersalloy (capsules)
3M, St Paul, MN, USA Johnson B Johnson, East Windsor, NJ, USA
Zinc glass, 84% by weight Admixture, 12% copper
May 1985
Material
The materials were placed sequentially as they became available, but early problems with the handling of VisioMolar for the smaller, modified OES fissure preparations, and later problems with the clinical performance of Ketac-Silver, led to a preference for using the less viscous P-30 resin. Because of the numerous clinical studies already published on the long-term behaviour of Dispersalloy, only a few amalgam restorations were included as Class I restoration controls. The choice between an OES preparation and a conventional Class I cavity was determined by the size of the carious lesion. Patients were selected solely on the basis of requiring treatment of Class I caries. Briefly, all narrow and sticky fissures were first widened minimally with a high speed tine flame-shaped diamond point. Any caries extending into dentine was then removed with small round steel burs at slow speed. No linings or bases were required as the depth of the preparations was minimal. For the composite resins, the enamel was etched with a 37 per cent phosphoric acid gel for 30 s, then washed and dried before placing the bonding resin and appropriate composite restorative. The bonding resin and restorative were light cured separately. For the cermet, the prepared fissure walls were first conditioned with a 10 per cent solution of polyacrylic acid for 10 s, then washed and dried before inserting the cermet. All materials were placed using metal instruments (Gerke, 1987) and in accordance with manufacturer’s instructions. For larger lesions, conventional Class I preparations were cut at high speed using small pear-shaped tungsten carbide burs. Any residual caries was removed as described previously. Occasionally, a glass ionomer cement base was placed in deeper preparations for resin and amalgam restorations. Insertion of the resins and glass-ionomer cermet was as for the fissure preparations. A cavity varnish was applied prior to placing the amalgam restorations. Apart from the amalgams, which were polished subsequently, all restorations were finished at the time of their insertion. Care was taken not to dehydrate the newly placed cermet restorations. Clinical photographs were taken at the time of finishing, and thereafter at 6-monthly intervals for periods of up to 3 years, using Ektachrome (Eastman Kodak Co., Rochester,
use
April 1986
NY, USA) 35 mm colour transparency film at 1 : 1 magnification. At each time point, bulk loss of material, and surface voids or exposed porosities were assessed as being either present or absent. Restoration margin fractures and discrepancies, marginal staining, surface staining or tarnishing, and surface roughness were assessed separately for all restorations by comparison of the photographs with four standard sets of enlarged 2~ colour transparencies, as has been described elsewhere (Smales and Creaven, 1985). Briefly, the four clinical factors assessed were rated by one examiner (R.J.S.) against separate linear scales, graded from 0 to 12,where 0 indicated no detection of the factor (staining, roughness, etc.) and 12 indicated that the factor was severe. A lack of sufficient colour contrast precluded rating the restorations of the two composite resin materials for margin fractures and discrepancies, and surface roughness. Each clinical factor was analysed separately using a mixed model analysis of variance, which included a random effects term for restorations, a covariate effect for restoration age, and a fixed effect for each material, the statistical significance of which was determined by a likelihood ratio test. Calculations were performed using the BMDP program 3V on a Vax 1l/785 computer (Dixon, 1988). Significant differences between the scored means over the study period were assumed to exist when there was an alpha probability of 0.05 or less. The failure rates of the restorations over the study period for the different materials were analysed using the BMDP life-table survival program 1L (Dixon, 1988). Repeat assessments were made (RJ.S.) of 30 restorations 2 months after the completion of the initial assessments. For bulk loss of material and for surface voids, weighted Kappa values were 0.61 and 0.64 respectively. For the four clinical factors, paired t tests showed no significant differences between the original and repeat assessments (P > 0.05).
RESULTS There were no significant differences in the distribution of the materials between the posterior teeth (P > 0.05). Standard Class I preparations accounted for 37 per cent of
et al.: Clinical evaluation
Smales
of occlusal
restorations
245
conjunction with Table ZZZ.Apart from surface roughness (P = 0.43) all the clinical factors were found to deteriorate with time over the study (P < 0.01). Although there were significant statistical differences found between the materials for the four clinical factors evaluated, no unsatisfactory clinical assessments were given. Table V shows the distribution of restoration observations for material retention. Losses of material usually occurred from over the cusp slopes and unprepared fissure extensions, especially with Ketac-Silver (14.5 per cent) (P < 0.001).
the restorations and the modified OES fissure preparations for 63 per cent (Table ZZ). The number of P-30 restorations placed reflected an increasing preference for this material as the study progressed. Ketac-Silver gave variable consistencies of mixes and working times. It was considered difficult to dispense this material into the small fissure preparations from the large straight nozzle of the capsules available at the time. Attempts to condense the cermet with metal instruments were also difficult as the material adhered to them and was pulled away from the cavity margins. The result was often an overfilled cavity with the material being smeared over the adjacent cusp slopes and any unprepared fissures. Visio-Molar was more viscous and difficult to handle than P-30, which was considered to be relatively easy to condense and get to flow into the narrow preparations. Table ZZZshows the mean scores over the study period of bulk loss of material, for surface voids or exposed porosities, and for the four clinical factors assessed in each case. Table ZV also shows the results of the asymptotic t-tests for the four clinical factors, to be read in
Table II. Number of restorations placed by size of occlusal preparations
Cavity size Material
Standard
Ketac-Silver Visio-Molar P-30 Dispersalloy
62 22 63 13
Modified
OES
:: 188 0
x2= 47.49, d.f.= 3, P < 0.01.
Table 111.Observation means + standard deviations over 3 years for the four materials placed Material Ketac-Silver Visio-Molar P-30 Dispersalloy P (from likelihood ratio test)
Bulk losses 0.22 0.01 0.02 0.03
f f f +
0.56 0.10 0.14 0.17
< 0.001*
Surface voids 0.23 0.11 0.04 0.00
~fr0.42 f 0.31 f 0.19 f 0.00
< 0.001*
Margin
fracture
0.64 f 0.76 1.35 + 1.25 0.002*
Margin stain
Surface stain
Surface rough
0.64 0.58 0.38 0.38
0.35 & 0.69 0.17 k 0.46 0.10+0.40 1.82 * 1.83
3.10 !I 1.31 1.47 f 1.26
< 0.001*
< 0.001*
k + + +
1.07 0.85 0.89 0.92
< 0.001*
*Significant difference between materials.
Table IV. Asvmptotic t-test scores of materials for predicted cell means* Material Bulk losses Visio-Molar P-30 Dispersalloy Surface voids Visio-Molar P-30 Dispersalloy Margin fracture Dispersalloy Margin staining Visio-Molar P-30 Dispersalloy Surface staining Visio-Molar P-30 Dispersalloy Surface roughness Dispersalloy
Ketac-Silver
-5.10 -8.95 -2.24
Visio-Molar
P-30
-0.58 - 1.29
- 1.87
1.37 0.68
1.70
1.87 -0.14
1.08
0.35 - 12.28
- 13.83
- 5.98 - 7.43 - 5.24 4.04
- 3.49
- 2.68 - 2.47 - 2.99 - 4.20 12.58 - 14.07
*Values of 2.0 or greater considered to be of statistical significance.
246
J. Dent
1990;
18: No. 5
Table VI shows the distribution of restoration observations for surface voids or exposed porosities. No voids were observed with Dispersalloy, but were seen frequently with Ketac-Silver (23 per cent) and to a lesser extent with the more viscous composite restorative Visio-Molar (10.9 per cent) (P < 0.001). Table VII shows the percentages of restoration observations with ideal rating assessments given for the various clinical characteristics examined over the study period. Minor restoration margin fractures and discrepancies, and surface staining or tarnish, typically limited, were seen most often with restorations of Dispersalloy, while some moderate surface roughness and limited margin staining was found with the restorations of Ketac-Silver. The findings for the restorations of Dispersalloy were similar to those reported previously @males and Gerke, 1984). Table VIII shows that 11.4 per cent of the Ketac-Silver restorations failed as a result of surface cracking or crazing, sometimes within 6 months of placement. In almost all instances, these problems were associated with the larger, Class I cavity preparations, rather than with the smaller, modified OES fissure preparations (P < 0.01).
Table V. Distribution of restorations for bulk of material retained, by number of observations made Bulk retention of material Intact Part lost
Material Ketac-Silver Visio-Molar P-30 Dispersalloy x2= 56.86, d.f.=
207
35 1 IO 1
5% 33 3, P < 0.001.
Table VI. Distribution of surface voids Surface voids (no. of observations) Material Ketac-Silver Visio-Molar P-30 Dispersalloy x2= 71.65,
d.f.=
3, P
