J. Dent. 1992;
20: 55-58
55
In vitro evaluation of two microleakage detection tests* D. G. Charlton Department
and B. K. Moore
of Dental Materials,
Indiana University
School of Dentistry,
Indianapolis,
Indiana, USA
ABSTRACT Two microleakage
detection tests were evaluated to compare their capabilities for revealing the extent of leakage at the interface between amalgam restorations and tooth structure. Class V cavities were prepared in the buccal surfaces of extracted human canine and premolar teeth and were treated in one of the following ways: no liner or varnish, a copal varnish, or an adhesive resin liner. The cavities were restored with a highcopper single composition spherical amalgam alloy and the specimens thermally cycled. Following storage for 6 months, the teeth were exposed to separate solutions of 45Caand an ultraviolet fluorescing dye. The teeth were hemisected longitudinally and leakage was evaluated by determining the degree of penetration of the two
tracers at the occlusal and gingival margins. Statistical analysis revealed a significant difference between the two leakage techniques (P < 0.05). The radioisotope test generally indicated a greater degree of leakage than did the ultraviolet dye test. KEY WORDS: J. Dent. 1992;
Microleakage, Radioisotopes, Ultraviolet dye 20: 55-58
(Received 12 March 1991;
reviewed 11 June 1991;
accepted 14 July 1991)
Correspondence should be addressed to: Mr D. G. Charlton, Department of Dental Materials, Indiana University School of Dentistry, Indianapolis, Indiana, USA.
INTRODUCTION
METHODS
The ability of restorative material to seal the margins of a cavity preparation is critical to the success of a restoration. Many methods have been employed to evaluate leakage between restorative materials and tooth structure. Radioactive isotopes (Granath, 1967; Guzmanet al., 1969; McCurdy et al., 1974) dyes (Hirsch and Weinreb, 1958; Kakar and Subramanian, 1963) pressurized gas (Pickard and Gayford, 1965) and calcium hydroxide (Wright et al., 1988) are just a few of the materials that have been successfully used to reveal leakage. The selection of a specific technique for leakage detection is often based on a number of considerations, including availability of facilities and materials, and ease of use. An unspoken assumption underlying the selection process is that different techniques generally yield the same results. Little work however, has been done to test this assumption. The purpose of this study was to compare the ability of two techniques for revealing microleakage between amalgam restorations and tooth structure following three different cavity lining treatments.
A total of 18 extracted, non-carious human canine and premolar teeth were selected and stored in distilled water. Following cleaning, a 3 mm mesiodistal by 2 mm occlusogingival by 2 mm deep Class V cavity preparation was prepared in the buccal surface of each tooth using a 557 tungsten carbide bur in a high-speed handpiece with water coolant. All margins were kept entirely within enamel. No additional finishing of the margins was performed following cavity preparation. Three test groups of six teeth each were treated in the following manner: group 1, no treatment; group 2, two separate applications of Copalite varnish (Harry J. Bosworth Co., Skokie, IL, USA); group 3, application of Amalgambond (Parke11 Biomaterials, Farmingdale, NY, USA)-an adhesive resin system primarily recommended for the bonding of amalgam to tooth structure. The Copalite treatment involved two applications of the varnish using a small sponge and cotton forceps. Oil-free, compressed air was used to dry the first layer prior to application of the second layer. The second layer was allowed to dry undisturbed following its application. The Amalgambond application involved three separate steps. Amalgambond Activator, a solution of 10 per cent citric acid and 3 per cent ferric chloride, was applied to the
*The opinions expressed herein are those of the authors and do not necessarily reflect the opinions of the Department of Defense or the United States Air Force. D (1992) Butterworth-Heinemann 0300-5712/92/010055-04
Ltd.
AND MATERIALS
56
J. Dent. 1992;
20: No. 1
dentine cavity walls for 10 s and to the enamel walls for 30 s using a tine brush. The cavity was then rinsed with tapwater for 20 s and dried with compressed air for 20 s. Amalgambond Adhesive, an aqueous solution of hydroxyethyl methacrylate (HEMA), was then applied to the dentine walls with a brush and left undisturbed for 30 s. A gentle stream of compressed air was then used to thin the adhesive, however, attempts were made to avoid excessive thinning. Amalgambond Base and Catalyst were mixed according to the manufacturer’s instructions and applied to the dentine and enamel walls of the cavity. The base solution is a mixture of HEMA 4-methacryloxyethyl trimellitate anhydride and methyl methacrylate, while the catalyst is tri-n-butyl borane. The specimens were restored with Tytin (Sybron/Kerr, Romulus, MI, USA), a high-copper single composition spherical amalgam alloy. The amalgam was triturated for 5 s in a triturator (Vari-Mix II, Caulk/Dentsply, Milford, DE, USA) and condensed into the preparations using 2T and 3T condensers (GC American Dental, Scottsdale, AZ, USA). The restorations were carved flush to the external surface of the teeth using an interproximal instrument (IPC, GC American Dental). Carving was begun when the amalgam first offered resistance to the carver. Only one cavity was restored with each mix of the amalgam. Following carving, the teeth were immersed in deionized water and stored for 1 week at 37 “C. The specimens were thermocycled for 2500 cycles between a 5°C water-bath and a 45 “C water-bath. The dwell time in each bath was 30 s and the transfer time was 10 s. After thermocycling, the teeth were stored in deionized water at 37°C. At 6 months, microleakage testing was performed. The apices of the specimens were sealed with compound and the teeth coated with nail varnish up to 1 mm from the cavosurface margins. While the varnish was still tacky, tin foil was adapted over the teeth up to 1 mm from the cavity margins. Nail polish was then applied to the seams of the foil to ensure complete sealing. The specimens were immersed for 2 h in a room temperature aqueous solution (pH 5.5) of 45CaC1, having a radioactivity of 0.1 mCi/ml. The teeth were then removed, rinsed in running tapwater for 1 h, and scrubbed with a toothbrush and liquid soap. Following scrubbing, the specimens were immersed in a 2 per cent aqueous solution of a fluorescent dye (Zyglo Penetrant ZL-54, MagnaFlux Corp., Chicago, IL, USA) for 24 h at 37°C. The teeth were then removed from the solution and cleaned by scrubbing with a toothbrush and liquid soap. The protective tin foil was removed and the teeth were rescrubbed. Each tooth was sectioned buccolingually using a sectioning machine (Thin Sectioning Machine, Hamco Machines, Inc., Rochester, NY, USA) with a diamond disc under a constant stream of water. Each sectioned specimen was placed with its cut surface down directly onto an intraoral radiographic film (Eastman Kodak Co., Rochester, NY, USA) placed on a supporting plastic plate. The specimens were wrapped in tin foil and placed in a light-proof container for 17 h. The exposed films were then developed in an automatic film
\
I 4 \1
4
,
T
Fig. 1. Standard
used for evaluation of leakage with the radioisotope and ultraviolet dye techniques.
processor. The autoradiographs were examined by two independent evaluators who scored the degree of leakage according to a predetermined standard (Fig. I). The diagram corresponds to the following degrees of leakage: Category Category Category Category
1 ‘No leakage up to one-half of the distance 2 Penetration from the margin to the axial wall from the margin to the axial 3 Penetration wall along the axial wall 4 Penetration
Two measurements were made for both halves of each sectioned specimen: one measurement for the occlusal margin and one for the gingival margin. This produced 12 occlusal and 12 gingival scores for each treatment group. The evaluators scored the specimens three times with at least a 24 h lapse between evaluations. To prepare the specimens for evaluation using the fluorescent dye, the sectioned teeth were painted on the facial surface with a dark-coloured nail polish to reduce scattered illumination from residual dye on the exposed facial surface. The specimens were mounted with compound on cardstock with the cut surfaces facing upward and were illuminated by two ultraviolet lamps (Blak-Ray, Ultra-Violet Products, Inc., San Gabriel, CA, USA) with aluminium cones positioned over the lamps to direct the light toward the specimens. The specimens were scored for dye penetration according to the same standard and under the same conditions used for the radioisotope evaluation. The three sets of radioisotope test scores for each of the evaluators were compiled and the Pearson correlation method was used to evaluate for intraevaluator agreement. A majority score for each evaluator was then determined and the correlation test was performed to test for interevaluator agreement. The two evaluators’ scores were compared and a majority score obtained. This produced one set of data for the radioisotope test. The same procedure was followed to evaluate the ultraviolet dye test
Charlton
and Moore:
Table 1. Radioisotope test leakage scores
Treatment group No treatment Copalite Amalgambond n= 12. Vertical lines connect probability level.
non-significant
0.57 f 0 0.51 f 0.14 0.33 + 0.25
differences
at the 0.05
Table III. Rank orders of the leakage scores
Treatment group No treatment Copalite Amalgambond
Occlusal RAD lJV 2 3 1
of microleakage
detection
tests
57
Table II. Ultraviolet dye test leakage scores
Ridit values (mean + s.d.) Occlusal Gingival 0.61 + 0.26 0.67 f 0.17 0.33 rt 0.30
Evaluation
Gingival RAD UV
2 3 1
n= 12. Vertical lines connect non-significant differences at the 0.05 probability level. 1, The least degree of leakage; 3, the greatest degree of leakage. RAD, radioisotope test; UV, ultraviolet dye test.
scores. The Wilcoxon signed-rank test was performed to determine if a significant difference existed between the scores for the two tests at the 0.05 probability level. Ridit Analysis was performed for both tests to obtain ridit means and standard deviations for the occlusal and gingival margins for the three treatment groups. Variances were tested for homogeneity using Bartlett’s test. When variances were homogeneous, one-way analysis of variance was used to test for significant differences. When variances were found to be non-homogeneous, Welch’s test was used. In both cases, testing was done at the 0.05 probability level and multiple comparisons were made using Newman-Keul’s test.
RESULTS For all evaluations, intraevaluator correlations were greater than 0.90. Interevaluator correlations for the radioisotope and dye tests were 0.92 and 0.79 respectively. Because significant differences were found between occlusal and gingival margin leakage scores, the scores could not be combined for simplified analysis. The Wilcoxon signed-rank test found that the distributions for the radioisotope and ultraviolet dye tests were not identical. Ridit means and standard deviations are presented for the radioisotope test (Table I) and the ultraviolet dye test (Table 11). Rank orders of the leakage scores are given in Table ZZZ.
DISCUSSION Although many methods can be used to evaluate microleakage at the interface of restorative materials and tooth structure, dye and radioisotope techniques appear to be the most popular. Each has its own advantages. An
Treatment group No treatment Copalite Amalgambond n= 12. Vertical lines connect probability level.
Ridit values (mean f s.d.) Occlusal Gingival 0.45 + 0.26 0.73 + 0.18 0.27 f 0 non-significant
0.55 + 0.32 0.66 * 0.21 0.33 * 0.15
differences
at the 0.05
ultraviolet fluorescing dye has been found to be particularly useful in that the dye fluoresces brightly when exposed to ultraviolet light (Christen and Mitchell, 1966). This makes photography of the specimens possible for the production of a permanent record. The dyes also tend to be non-toxic and inexpensive (Myer et al., 1974). Because the dyes are non-toxic they can be used intraorally to evaluate microleakage (Loiselle et al., 1969). The radioisotope method, on the other hand, employs tracers which penetrate deeper than dyes. The autoradiographs produced are capable of detecting minute amounts of the tracers which are not visible using other methods (Going, 1972). The freedom of researchers to choose among the various leakage detection tests is predicated upon the assumption that they all yield similar results. The finding of this study that the radioisotope and ultraviolet dye tests produced statistically different results seriously undermines that assumption. For identical specimens, the radioisotope technique tended to indicate a greater degree of leakage than did the ultraviolet dye test. These results are at odds with the findings of Crim et al. (1985), who compared the effectiveness of basic fuchsin dye and 4SCa as microleakage tracers. They found the two tracers to be equally effective in assessing degree of leakage. Their work, however, was primarily undertaken to evaluate the effect of different thermocycling techniques on leakage and only secondarily examined the correlation between tracers. In addition, the two techniques do appear to have produced dissimilar results for leakage at the gingival margin. Examination of the rank orders of the leakage scores for the two tests in this study (Table ZZZ)reveals that although the rank orders for occlusal margin leakage are the same, the ultraviolet dye test found a statistically significant difference between the ‘No treatment’ and Copalite groups which was not seen with the radioisotope test. The tests produced even greater differences for gingival margin leakage. Although the differences were not statistically significant, the rank orders were reversed for the ‘No treatment’ and Copalite groups. In addition, the radioisotope test indicated a significant difference in leakage between the ‘No treatment’ and Amalgambond groups which was not found by the ultraviolet dye test. Although the sample size in this study was relatively small, the differences produced by the two leakage detection techniques suggest that they are not equivalent in their abilities to reflect the degree of leakage at the
58
J. Dent.
1992;
20:
No. 1
restoration/tooth interface. The radioisotope technique generally indicated a greater degree of leakage than did the ultraviolet dye technique. This should not be construed, however, to mean that the radioisotope test is necessarily more discerning than the ultraviolet dye test. For example, the radioisotope test did not detect a significant difference between the ‘No treatment’ and Copalite groups at the occlusal margin that was detected by the ultraviolet dye test. This outcome occurred, in large measure, because the differences in degrees of leakage detected by the two tests were not consistent for all groups. Future research should be directed toward attempting to determine if these inconsistencies extend to other microleakage detection tests. The possibility that testing order may affect the outcome should also be examined. In this study, the specimens were first exposed to the radioisotope and then, in a separate step, were exposed to the ultraviolet dye. This was intentionally done to prevent dye contamination of the radioisotope solution. Ideally specimens should be exposed to both media simultaneously to prevent testing order from having an effect on the outcome. Microleakage detection tests have played an important role in dental research and have enabled investigators to assess the ability of restorative materials to seal the restoration/tooth interface. The results of this study suggest, however, that individual tests do not yield equivalent results and that they should not be considered to be interchangeable.
International
References Christen A G. and Mitchell D. F. (1966) A fluorescent dye method for demonstrating leakage around dental restorations. J. Dent. Res. 45, 1485-1492. Crim G. A., Swartz M. L. and Phillips R. W. (1985) Comparison of four thermocycling techniques. J. Prosthet Dent. 53, 50-53. Going R. E. (1972) Microleakage around dental restorations: a summarizing review. J. Am. Dent Assoc. 84, 1349-1357. Granath L. (1967) Studies on microleakage with restorative materials. Part I. Introductory experiments on amalgam. J. Dent Res. 46, 1331-1336. Guzman H. J., Swartz M. L. and Phillips R. W. (1969) Marginal leakage of dental restorations subjected to thermal stress. J. Prosthet Dent. 21, 166-175. Hirsch L. and Weinreb M. M. (1958) Marginal fit of direct acrylic restorations. J. Am. Dent. Assoc. 56, 13-21. Kakar R. C. and Subramanian V. (1963) Sealing qualities of various restorative materials. J. Prosthet Dent. 13, 156-165. Loiselle R. J., Goldberg A. F., Gross R. L. et al. (1969) Marginal leakage-an in vivo assessment. J. Am. Dent. Assoc. 78, 758-760. McCurdy C. R., Swartz M. L., Phillips R. W. et al. (1974) A comparison of in vivo and in vitro microleakage of dental restorations. J. Am. Dent. Assoc. 88, 592-602. Myer J. M., Dennison J. B. and Craig R. G. (1974) Improved method of neutron activation analysis for microleakage studies. J. Dent. Res. 53, 356-363. Pickard H. M. and Gayford J. J. (1965) Leakage at the margins of amalgam restorations. Br. Dent. J. 119, 69-77. Wright W., Mazer R. M., Leinfelder K. F. ef al. (1988) Effect of cavity varnish on the clinical microleakage of amalgam. J. Dent. Res. 67, (abstr. 308), 139.
Abstracts
Hydrodynamisches Verhalten der Dentin-TubulusFliissigkeit unter okklusaler Belastung. (Hydrodynamic behaviour of the tubular fluid in dentine under occlusal load.) Lutz F., Krejci I., lmfeld Th. et al. (1991) Schweiz. Monatsschr. Zahnmed. 101, 24-30. In an in vitro study, the fluid movements were quantitatively recorded in natural teeth incorporated rigidly in a closed measuring system and exposed to an occlusal load of 122.6 N. Fluid movements under load are the result of elastic deformations of the tooth. Occlusal loads which were repeated within a short period of time caused a reversible plastic deformation. In comparison to the unfilled tooth the placement of an
MOD restoration combined with a glass ionomer cement base caused the following alterations in fluid movements: composite inlay, - 3.1 per cent; the placement of the base per se induced a reduction of - 24.7 per cent; ceramic inlay, + 34.7 per cent; MOD amalgam, - 12.7 per cent. In this in vitro study, the physical properties of the restorative materials governed the magnitude of the fluid movements under load. The results are in contradiction to clinical findings and favour the pulpitisrelated genesis of postoperative sensitivity. The intratubular and intrapulpal fluid movements under occlusal load seem to trigger the pain sensation. (24 references) Author’s abstract selected by W. Kullmann
20: 55-58
55
In vitro evaluation of two microleakage detection tests* D. G. Charlton Department
and B. K. Moore
of Dental Materials,
Indiana University
School of Dentistry,
Indianapolis,
Indiana, USA
ABSTRACT Two microleakage
detection tests were evaluated to compare their capabilities for revealing the extent of leakage at the interface between amalgam restorations and tooth structure. Class V cavities were prepared in the buccal surfaces of extracted human canine and premolar teeth and were treated in one of the following ways: no liner or varnish, a copal varnish, or an adhesive resin liner. The cavities were restored with a highcopper single composition spherical amalgam alloy and the specimens thermally cycled. Following storage for 6 months, the teeth were exposed to separate solutions of 45Caand an ultraviolet fluorescing dye. The teeth were hemisected longitudinally and leakage was evaluated by determining the degree of penetration of the two
tracers at the occlusal and gingival margins. Statistical analysis revealed a significant difference between the two leakage techniques (P < 0.05). The radioisotope test generally indicated a greater degree of leakage than did the ultraviolet dye test. KEY WORDS: J. Dent. 1992;
Microleakage, Radioisotopes, Ultraviolet dye 20: 55-58
(Received 12 March 1991;
reviewed 11 June 1991;
accepted 14 July 1991)
Correspondence should be addressed to: Mr D. G. Charlton, Department of Dental Materials, Indiana University School of Dentistry, Indianapolis, Indiana, USA.
INTRODUCTION
METHODS
The ability of restorative material to seal the margins of a cavity preparation is critical to the success of a restoration. Many methods have been employed to evaluate leakage between restorative materials and tooth structure. Radioactive isotopes (Granath, 1967; Guzmanet al., 1969; McCurdy et al., 1974) dyes (Hirsch and Weinreb, 1958; Kakar and Subramanian, 1963) pressurized gas (Pickard and Gayford, 1965) and calcium hydroxide (Wright et al., 1988) are just a few of the materials that have been successfully used to reveal leakage. The selection of a specific technique for leakage detection is often based on a number of considerations, including availability of facilities and materials, and ease of use. An unspoken assumption underlying the selection process is that different techniques generally yield the same results. Little work however, has been done to test this assumption. The purpose of this study was to compare the ability of two techniques for revealing microleakage between amalgam restorations and tooth structure following three different cavity lining treatments.
A total of 18 extracted, non-carious human canine and premolar teeth were selected and stored in distilled water. Following cleaning, a 3 mm mesiodistal by 2 mm occlusogingival by 2 mm deep Class V cavity preparation was prepared in the buccal surface of each tooth using a 557 tungsten carbide bur in a high-speed handpiece with water coolant. All margins were kept entirely within enamel. No additional finishing of the margins was performed following cavity preparation. Three test groups of six teeth each were treated in the following manner: group 1, no treatment; group 2, two separate applications of Copalite varnish (Harry J. Bosworth Co., Skokie, IL, USA); group 3, application of Amalgambond (Parke11 Biomaterials, Farmingdale, NY, USA)-an adhesive resin system primarily recommended for the bonding of amalgam to tooth structure. The Copalite treatment involved two applications of the varnish using a small sponge and cotton forceps. Oil-free, compressed air was used to dry the first layer prior to application of the second layer. The second layer was allowed to dry undisturbed following its application. The Amalgambond application involved three separate steps. Amalgambond Activator, a solution of 10 per cent citric acid and 3 per cent ferric chloride, was applied to the
*The opinions expressed herein are those of the authors and do not necessarily reflect the opinions of the Department of Defense or the United States Air Force. D (1992) Butterworth-Heinemann 0300-5712/92/010055-04
Ltd.
AND MATERIALS
56
J. Dent. 1992;
20: No. 1
dentine cavity walls for 10 s and to the enamel walls for 30 s using a tine brush. The cavity was then rinsed with tapwater for 20 s and dried with compressed air for 20 s. Amalgambond Adhesive, an aqueous solution of hydroxyethyl methacrylate (HEMA), was then applied to the dentine walls with a brush and left undisturbed for 30 s. A gentle stream of compressed air was then used to thin the adhesive, however, attempts were made to avoid excessive thinning. Amalgambond Base and Catalyst were mixed according to the manufacturer’s instructions and applied to the dentine and enamel walls of the cavity. The base solution is a mixture of HEMA 4-methacryloxyethyl trimellitate anhydride and methyl methacrylate, while the catalyst is tri-n-butyl borane. The specimens were restored with Tytin (Sybron/Kerr, Romulus, MI, USA), a high-copper single composition spherical amalgam alloy. The amalgam was triturated for 5 s in a triturator (Vari-Mix II, Caulk/Dentsply, Milford, DE, USA) and condensed into the preparations using 2T and 3T condensers (GC American Dental, Scottsdale, AZ, USA). The restorations were carved flush to the external surface of the teeth using an interproximal instrument (IPC, GC American Dental). Carving was begun when the amalgam first offered resistance to the carver. Only one cavity was restored with each mix of the amalgam. Following carving, the teeth were immersed in deionized water and stored for 1 week at 37 “C. The specimens were thermocycled for 2500 cycles between a 5°C water-bath and a 45 “C water-bath. The dwell time in each bath was 30 s and the transfer time was 10 s. After thermocycling, the teeth were stored in deionized water at 37°C. At 6 months, microleakage testing was performed. The apices of the specimens were sealed with compound and the teeth coated with nail varnish up to 1 mm from the cavosurface margins. While the varnish was still tacky, tin foil was adapted over the teeth up to 1 mm from the cavity margins. Nail polish was then applied to the seams of the foil to ensure complete sealing. The specimens were immersed for 2 h in a room temperature aqueous solution (pH 5.5) of 45CaC1, having a radioactivity of 0.1 mCi/ml. The teeth were then removed, rinsed in running tapwater for 1 h, and scrubbed with a toothbrush and liquid soap. Following scrubbing, the specimens were immersed in a 2 per cent aqueous solution of a fluorescent dye (Zyglo Penetrant ZL-54, MagnaFlux Corp., Chicago, IL, USA) for 24 h at 37°C. The teeth were then removed from the solution and cleaned by scrubbing with a toothbrush and liquid soap. The protective tin foil was removed and the teeth were rescrubbed. Each tooth was sectioned buccolingually using a sectioning machine (Thin Sectioning Machine, Hamco Machines, Inc., Rochester, NY, USA) with a diamond disc under a constant stream of water. Each sectioned specimen was placed with its cut surface down directly onto an intraoral radiographic film (Eastman Kodak Co., Rochester, NY, USA) placed on a supporting plastic plate. The specimens were wrapped in tin foil and placed in a light-proof container for 17 h. The exposed films were then developed in an automatic film
\
I 4 \1
4
,
T
Fig. 1. Standard
used for evaluation of leakage with the radioisotope and ultraviolet dye techniques.
processor. The autoradiographs were examined by two independent evaluators who scored the degree of leakage according to a predetermined standard (Fig. I). The diagram corresponds to the following degrees of leakage: Category Category Category Category
1 ‘No leakage up to one-half of the distance 2 Penetration from the margin to the axial wall from the margin to the axial 3 Penetration wall along the axial wall 4 Penetration
Two measurements were made for both halves of each sectioned specimen: one measurement for the occlusal margin and one for the gingival margin. This produced 12 occlusal and 12 gingival scores for each treatment group. The evaluators scored the specimens three times with at least a 24 h lapse between evaluations. To prepare the specimens for evaluation using the fluorescent dye, the sectioned teeth were painted on the facial surface with a dark-coloured nail polish to reduce scattered illumination from residual dye on the exposed facial surface. The specimens were mounted with compound on cardstock with the cut surfaces facing upward and were illuminated by two ultraviolet lamps (Blak-Ray, Ultra-Violet Products, Inc., San Gabriel, CA, USA) with aluminium cones positioned over the lamps to direct the light toward the specimens. The specimens were scored for dye penetration according to the same standard and under the same conditions used for the radioisotope evaluation. The three sets of radioisotope test scores for each of the evaluators were compiled and the Pearson correlation method was used to evaluate for intraevaluator agreement. A majority score for each evaluator was then determined and the correlation test was performed to test for interevaluator agreement. The two evaluators’ scores were compared and a majority score obtained. This produced one set of data for the radioisotope test. The same procedure was followed to evaluate the ultraviolet dye test
Charlton
and Moore:
Table 1. Radioisotope test leakage scores
Treatment group No treatment Copalite Amalgambond n= 12. Vertical lines connect probability level.
non-significant
0.57 f 0 0.51 f 0.14 0.33 + 0.25
differences
at the 0.05
Table III. Rank orders of the leakage scores
Treatment group No treatment Copalite Amalgambond
Occlusal RAD lJV 2 3 1
of microleakage
detection
tests
57
Table II. Ultraviolet dye test leakage scores
Ridit values (mean + s.d.) Occlusal Gingival 0.61 + 0.26 0.67 f 0.17 0.33 rt 0.30
Evaluation
Gingival RAD UV
2 3 1
n= 12. Vertical lines connect non-significant differences at the 0.05 probability level. 1, The least degree of leakage; 3, the greatest degree of leakage. RAD, radioisotope test; UV, ultraviolet dye test.
scores. The Wilcoxon signed-rank test was performed to determine if a significant difference existed between the scores for the two tests at the 0.05 probability level. Ridit Analysis was performed for both tests to obtain ridit means and standard deviations for the occlusal and gingival margins for the three treatment groups. Variances were tested for homogeneity using Bartlett’s test. When variances were homogeneous, one-way analysis of variance was used to test for significant differences. When variances were found to be non-homogeneous, Welch’s test was used. In both cases, testing was done at the 0.05 probability level and multiple comparisons were made using Newman-Keul’s test.
RESULTS For all evaluations, intraevaluator correlations were greater than 0.90. Interevaluator correlations for the radioisotope and dye tests were 0.92 and 0.79 respectively. Because significant differences were found between occlusal and gingival margin leakage scores, the scores could not be combined for simplified analysis. The Wilcoxon signed-rank test found that the distributions for the radioisotope and ultraviolet dye tests were not identical. Ridit means and standard deviations are presented for the radioisotope test (Table I) and the ultraviolet dye test (Table 11). Rank orders of the leakage scores are given in Table ZZZ.
DISCUSSION Although many methods can be used to evaluate microleakage at the interface of restorative materials and tooth structure, dye and radioisotope techniques appear to be the most popular. Each has its own advantages. An
Treatment group No treatment Copalite Amalgambond n= 12. Vertical lines connect probability level.
Ridit values (mean f s.d.) Occlusal Gingival 0.45 + 0.26 0.73 + 0.18 0.27 f 0 non-significant
0.55 + 0.32 0.66 * 0.21 0.33 * 0.15
differences
at the 0.05
ultraviolet fluorescing dye has been found to be particularly useful in that the dye fluoresces brightly when exposed to ultraviolet light (Christen and Mitchell, 1966). This makes photography of the specimens possible for the production of a permanent record. The dyes also tend to be non-toxic and inexpensive (Myer et al., 1974). Because the dyes are non-toxic they can be used intraorally to evaluate microleakage (Loiselle et al., 1969). The radioisotope method, on the other hand, employs tracers which penetrate deeper than dyes. The autoradiographs produced are capable of detecting minute amounts of the tracers which are not visible using other methods (Going, 1972). The freedom of researchers to choose among the various leakage detection tests is predicated upon the assumption that they all yield similar results. The finding of this study that the radioisotope and ultraviolet dye tests produced statistically different results seriously undermines that assumption. For identical specimens, the radioisotope technique tended to indicate a greater degree of leakage than did the ultraviolet dye test. These results are at odds with the findings of Crim et al. (1985), who compared the effectiveness of basic fuchsin dye and 4SCa as microleakage tracers. They found the two tracers to be equally effective in assessing degree of leakage. Their work, however, was primarily undertaken to evaluate the effect of different thermocycling techniques on leakage and only secondarily examined the correlation between tracers. In addition, the two techniques do appear to have produced dissimilar results for leakage at the gingival margin. Examination of the rank orders of the leakage scores for the two tests in this study (Table ZZZ)reveals that although the rank orders for occlusal margin leakage are the same, the ultraviolet dye test found a statistically significant difference between the ‘No treatment’ and Copalite groups which was not seen with the radioisotope test. The tests produced even greater differences for gingival margin leakage. Although the differences were not statistically significant, the rank orders were reversed for the ‘No treatment’ and Copalite groups. In addition, the radioisotope test indicated a significant difference in leakage between the ‘No treatment’ and Amalgambond groups which was not found by the ultraviolet dye test. Although the sample size in this study was relatively small, the differences produced by the two leakage detection techniques suggest that they are not equivalent in their abilities to reflect the degree of leakage at the
58
J. Dent.
1992;
20:
No. 1
restoration/tooth interface. The radioisotope technique generally indicated a greater degree of leakage than did the ultraviolet dye technique. This should not be construed, however, to mean that the radioisotope test is necessarily more discerning than the ultraviolet dye test. For example, the radioisotope test did not detect a significant difference between the ‘No treatment’ and Copalite groups at the occlusal margin that was detected by the ultraviolet dye test. This outcome occurred, in large measure, because the differences in degrees of leakage detected by the two tests were not consistent for all groups. Future research should be directed toward attempting to determine if these inconsistencies extend to other microleakage detection tests. The possibility that testing order may affect the outcome should also be examined. In this study, the specimens were first exposed to the radioisotope and then, in a separate step, were exposed to the ultraviolet dye. This was intentionally done to prevent dye contamination of the radioisotope solution. Ideally specimens should be exposed to both media simultaneously to prevent testing order from having an effect on the outcome. Microleakage detection tests have played an important role in dental research and have enabled investigators to assess the ability of restorative materials to seal the restoration/tooth interface. The results of this study suggest, however, that individual tests do not yield equivalent results and that they should not be considered to be interchangeable.
International
References Christen A G. and Mitchell D. F. (1966) A fluorescent dye method for demonstrating leakage around dental restorations. J. Dent. Res. 45, 1485-1492. Crim G. A., Swartz M. L. and Phillips R. W. (1985) Comparison of four thermocycling techniques. J. Prosthet Dent. 53, 50-53. Going R. E. (1972) Microleakage around dental restorations: a summarizing review. J. Am. Dent Assoc. 84, 1349-1357. Granath L. (1967) Studies on microleakage with restorative materials. Part I. Introductory experiments on amalgam. J. Dent Res. 46, 1331-1336. Guzman H. J., Swartz M. L. and Phillips R. W. (1969) Marginal leakage of dental restorations subjected to thermal stress. J. Prosthet Dent. 21, 166-175. Hirsch L. and Weinreb M. M. (1958) Marginal fit of direct acrylic restorations. J. Am. Dent. Assoc. 56, 13-21. Kakar R. C. and Subramanian V. (1963) Sealing qualities of various restorative materials. J. Prosthet Dent. 13, 156-165. Loiselle R. J., Goldberg A. F., Gross R. L. et al. (1969) Marginal leakage-an in vivo assessment. J. Am. Dent. Assoc. 78, 758-760. McCurdy C. R., Swartz M. L., Phillips R. W. et al. (1974) A comparison of in vivo and in vitro microleakage of dental restorations. J. Am. Dent. Assoc. 88, 592-602. Myer J. M., Dennison J. B. and Craig R. G. (1974) Improved method of neutron activation analysis for microleakage studies. J. Dent. Res. 53, 356-363. Pickard H. M. and Gayford J. J. (1965) Leakage at the margins of amalgam restorations. Br. Dent. J. 119, 69-77. Wright W., Mazer R. M., Leinfelder K. F. ef al. (1988) Effect of cavity varnish on the clinical microleakage of amalgam. J. Dent. Res. 67, (abstr. 308), 139.
Abstracts
Hydrodynamisches Verhalten der Dentin-TubulusFliissigkeit unter okklusaler Belastung. (Hydrodynamic behaviour of the tubular fluid in dentine under occlusal load.) Lutz F., Krejci I., lmfeld Th. et al. (1991) Schweiz. Monatsschr. Zahnmed. 101, 24-30. In an in vitro study, the fluid movements were quantitatively recorded in natural teeth incorporated rigidly in a closed measuring system and exposed to an occlusal load of 122.6 N. Fluid movements under load are the result of elastic deformations of the tooth. Occlusal loads which were repeated within a short period of time caused a reversible plastic deformation. In comparison to the unfilled tooth the placement of an
MOD restoration combined with a glass ionomer cement base caused the following alterations in fluid movements: composite inlay, - 3.1 per cent; the placement of the base per se induced a reduction of - 24.7 per cent; ceramic inlay, + 34.7 per cent; MOD amalgam, - 12.7 per cent. In this in vitro study, the physical properties of the restorative materials governed the magnitude of the fluid movements under load. The results are in contradiction to clinical findings and favour the pulpitisrelated genesis of postoperative sensitivity. The intratubular and intrapulpal fluid movements under occlusal load seem to trigger the pain sensation. (24 references) Author’s abstract selected by W. Kullmann
