Effect of two dentin bonding different cavity designs. James
M. Mixson,
David
L. Moore,
University
DMD, DDS,
of Missouri,
MS,a MS,C
J. David
and
School of Dentistry,
Daniel
agents on microleakage
Eick,
in two
PhD,b
E. Tira,
PhDd
Kansas City, MO.
Studies have shown that cavity design and dentinal bonding agents can affect composite resin microleakage. This study compared the microleakage at cementurn-dentin margins of box- and Vshaped preparations restored with two different bonding agents. Twenty freshly extracted, human third molars were prepared with one box-shaped and one V-shaped preparation on the mesial or distal surface. Occlusal margins were terminated in etched enamel and gingival margins were in cementum-dentin. There was no statistically significant difference (p > 0.05) between Dual Cure Scotchbond and Scotchbond 2 samples restored in the boxshaped preparations, but Scotchbond 2 had statistically significantly less microleakage @ < 0.06) than Dual Cure adhesive in V-shaped preparations. The sealing of dentinal bonding agents varied in different cavity designs. (J PROSTHET DENT 1992;67:441-5.)
M*
rcroleakage, or the marginal permeability to bacterial, chemical and molecular invasion at the interface between the teeth and restorative materials remains a problem with composite resin restorations. Two factors that influence the marginal seal are the type of dentin bonding agent and the cavity design. Buonocorel showed that microleakage of composite resins is greatly reduced with acid-etched enamel margins. However, Gwinnett2 noted the difficulty in bonding to vital dentin that is 70% inorganic apatite by weight embedded in a collagen matrix that is 20% by weight. He also stated that preliminary studies had demonstrated that conservative removal of the smear layer before bonding increased adhesion to dentin. Eick et a1.3described the smear layer as a combination of inorganic hydroxyapatite and organic denatured collagen formed by the deposition of the dentinal protein matrix during cutting procedures. Dual Cure Scotchbond adhesive (3M Products Co., St. Paul, Minn.) bonds to an intact smear layer, whereas Scotchbond 2 (3M Products Co.) removes the smear layer before the adhesive is placed. Studies have demonstrated that Dual Cure Scotchbond adhesive significantly reduced microleakage.4p5 Although few studies using Scotchbond 2 have been completed, the preliminary results reveal comparable leakage with these two bonding agents.6,7 The effect of cavity design on a dentinal bonding agent to prevent leakage has been reviewed by numerous investigators. Retief et a1.8found no differences in microleakage between composite resins restored in box tooth prepara-
tions with butt or beveled cavosurface margins. However, Crim et ah9 demonstrated that beveled class V box composite resin restorations displayed significantly less microleakage than identical restorations with butt margins. The shape of the cavity has been recently identified as critical in determining the width of marginal gaps. Asmussen and Munksgaard’O revealed that the marginal gap increased from 4 pm to 13 pm as the cavity design changed from V-shaped to box shaped. This variation in gap formation was explained by the alteration in the ratio between the volume of filling and surface area of cavity walls. This volume:area ratio is elevated from 0.4 mm for V-shaped restorations to 1 mm for box-shaped restorations. The contraction forces increased as the volume increased, and as the surface area of the cavity walls diminished, so did the forces bonding the composite resin to the cavity walls. Although direct statistical comparisons in sectional microleakage studies between box- and V-shaped restorations are impossible because of cavity geometry and the different measurements associated with them, data in a former study6 suggested comparable microleakage for the two designs. The bonding to V-shaped erosion lesions is a challenge to dentists because of the lack of parallel walls or grooves. Studies have not compared the sealing of traditional bonding agents and Scotchbond 2 agent restored in different class V cavity preparations. This investigation compared the in vitro marginal sealing at the cementum-dentin margins of Dual Cure Scotchbond adhesive and Scotchbond 2 in box- and V-shaped tooth preparations.
‘Assistant Professor,Operative Dentistry.
MATERIAL
bCurator’s Professor and Acting Chairman, cProfessor, Chairman, Operative Dentistry. dAssociate Professor, Behavioral Sciences. lOl1134356
THE JOURNAL OFPROSTHETIC
DENTISTRY
Oral Biology.
AND
METHODS
A pilot study was performed to determine the sample size by restoring 10 teeth with composite resin and then sectioning occlusogingivally through the center of the res441
MIXSON
1. Schematic representation for scoring box-shaped restorations. (E) Enamel, (Co) composite resin, (Ce) Cementum. 0, No penetration; 1, up to ‘/z gingival wall; 2, up to gingivoaxial line angle; 3, up to ‘/z axial wall; 4, up to occlusoaxial line angle; 5, up to ‘/z occlusal wall; 6, up to occlusal margin.
Fig.
Fig. 2. Schematic representation for scoring V-shaped restorations. (E) Enamel, (Co) composite resin, (Ce) Cementum. 0, No penetration; 1, up to ‘/z gingival wall; 2, to depth of groove; 3, up to ‘/z occlusal wall; 4, up to occlusal margin.
toration. A mean microleakage value of 5.5 + 1.2 was recorded from surfaces scored by two calibrated examiners. Statistical methodology dictated that a sample size of 10 was necessary for each test condition to detect a statisti442
ET AL
cally significant difference of 20 % with parameters (Y= 0.05 and p = 0.05. Twenty extracted third molar teeth were selected and stored in normal saline at 5’ C for a maximum of 48 hours before preparation. Each tooth was prepared with a No. 169 carbide bur. One box-shaped and one V-shaped preparation were randomly made on either the mesial or distal surface for a total of 40 samples, or 20 box-shaped and 20 V-shaped. The class V box-shaped preparations had a 3 mm occlusogingival (OG) width, a 5 mm buccolingual (BL) length, and a 3 mm axial depth. The V-shaped preparations were designed to simulate an erosion lesion with a 3 mm OG width and a BL length of 6 to 7 mm that tapered at the cavity extremities. The depth at the center of the groove was 3 mm. The width and length for the preparations were checked with a Boley gauge caliber (Miltex Instrument Co., Lake Success, N.Y.) and depth was verified by use of a periodontal probe. Care was exercised to ensure that half of the preparation remained in enamel occlusal to the cementoenamel junction (CEJ), and half was apical to the CEJ in cementum. All enamel margins received a 0.5 mm bevel. Each tooth was randomly restored with either Dual Cure Scotchbond, P-30 or Scotchbond 2, P-50 materials. One material was used per tooth to prevent cross-contamination. The enamel bevels were gel-etched with 37% phosphoric acid for 30 seconds, rinsed 10 seconds, and dried 20 sec0nds.l’ Dentinal bonding agents were applied according to the manufacturer’s and recommendation, the composite resin was then incrementally placed along occlusal and gingival walls. Each increment was light-cured 40 seconds. The restored teeth were stored 7 days in normal saline with thymol, then finished with a No. 8884 diamond finishing bur, and polished with medium and fine Soflex disks (3M Products Co.). The restored teeth were thermocycled for 100 cycles with a 30 seconds dwell between 5’ C and 55” C. The apical foramen, roots, and crowns of the teeth were then covered with one coat of red wax to within 1 mm of the cementumdentin cavosurface margins. The samples were stored in 50 % by weight silver nitrate for 2 hours, rinsed, then placed in radiographic developer and exposed to fluorescent light for 2 hours.12 Teeth were sectioned buccolingually to separate mesial and distal halves with a low-speed diamond saw (Isomet Model 11-1180, Buehler Ltd., Evanston Ill.) and pulpal remnants removed. The halves were mounted in self-curing clear casting resin. Specimens were then sectioned through the restorations into four sections of 0.75 to 1 mm width. Sectioning for the box-shaped restorations began at the end of the buccal or lingual wall and extended to the opposite wall. Sectioning for the V-shaped restorations began 1 to 2 mm inside the restoration and extended to within 1 to 2 mm of the opposite end in order to maintain standard cavity dimensions for scoring. All eight surfaces of the four sections per restorations were scored.13 The degree of stain penetration was assessed at x30
APRIL
1992
VOLUME
67
NUMBER
4
TWO
DENTIN
Table
BONDING
AGENTS
Mann-Whitney
I.
U test of median scores for dentin bonding agents in box-shaped preparations* Median
1 2 3 4 5 6 7 8 9 10
Rank
Scotchbond
Dual core Scotchbond
Sample
Group
scores 2
3.5 10.0 10.0 10.0 10.0 10.0 10.5 17.5 19.5
1.0 1.0 1.5 2.0 2.0 2.0 2.0 2.0 2.5 3.0 iz
1.5 2.0 2.0 2.0 2.0 2.0 2.0 3.0 4.0
median
Dual core Scotchbond
Scotchbond
2
1.5 1.5 3.5 10.0 10.0 10.0 10.0 10.0 16.0
n, lo; u, 65. *Not significant (p > 0.05).
Table
II.
Mann-Whitney
U test of median scores for dentin bonding agents in V-shaped preparations* Median
Sample
Dual core Scotchbond
1 2 3 4 5 6 I 8 9 10
scores
Rank Scotchbond
1.0 1.0 1.0 1.5 1.5 2.0 2.0 2.0 2.0 2.5 iii
Group median
2
0.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.5 2.0 iz
Dual cure Scotchbond 6.5 6.5 6.5 13.0 13.0 17.0 17.0 17.0 17.0 20.0 z = 133.50
Scotchbond
2
1.0 6.5 6.5 6.5 6.5 6.5 6.5 6.5 13.0 17.0 Z = 76.50
n, IO; U, 78.5. *Significant (50.05).
magnification by two calibrated examiners. Microleakage was scored for box-shaped restorations according to the criteria described in Fig. 1 and for V-shaped restorations according to the criteria in Fig. 2. Rater reliability was greater when a 4-point scale was used to score V-shaped preparations. Because each preparation type had a different measurement scale, the effect of cavity design on microleakage of different dentin bonding agents could not be tested. Leakage was scored beginning at gingival cavosurface margins. Intrarater and interrater reliability was calculated by having each examiner independently score the same 40 randomly presented samples, 20 box-shaped and 20 V-shaped on two different occasions 2 weeks apart. During the study a consensus score was reached by all examiners. Examiners did not prepare or restore tooth samples and were uninformed as to the materials or surface scored. The median microleakage score from the eight surfaces per restoration was recorded. The data were subjected to the Mann-Whitney U testi to determine statistically sig-
TiiE
JOURNAL
OF PROSTHETIC
DENTISTRY
nificant differences @ < 0.05) between Dual Cure Scotchbond, P-30 and Scotchbond 2, P-50 materials for each preparation. Intrarater and interrater reliability was calculated by using the Spearman rank correlation14 for boxand V-shaped restorations and resulted in values of >0.74 and >0.84, respectively. Representative photomicrographs were taken at x30 magnification.
RESULTS Results of the Mann-Whitney U test between respective treatment conditions are shown in Tables I and II. There was no statistically significant difference in microleakage between Dual Cure Scotchbond adhesive and Scotchbond 2 adhesive materials placed in box-shaped preparations (Table I). However, Scotchbond 2 adhesive had statistically significantly less microleakage (p < 0.05) than Dual Cure Scotchbond adhesive in V-shaped preparations (Table II), The overall median scores ‘for Dual Care Scotchbond and Scotchbond 2 adhesives in box-shaped prepara443
MIXSON
ET AL
Fig. 3. Silver nitrate penetration observed at tooth-restoration interface of box-shaped restoration,
Fig. 4. Silver nitrate penetration observed at tooth-restoration interface of V-shaped restoration.
tions were both 2, and in V-shaped preparations the medians were 2 and 1, respectively. Leakage patterns were commonly greater at the cavosurface angles of the restorations.13 Typical stain penetration is evident in Figs. 3 and 4, but all of the restorations exhibited a certain degree of microleakage at the toothrestoration interface.
Scotchbond adhesive in class V box preparations. Chappell et a1.16demonstrated earlier that the Scotchbond 2 adhesive had a significantly greater bond strength than other adhesives. Scanning electron microscopic studies revealed that the Scotchbond 2 adhesive completely covered the dentin surface to a thickness of 8 pm and fracture patterns commonly disclosed resin tag penetration into dentinal tubules. This elevated bond strength may decrease microleakage, particularly in gingival V-shaped restorations that rely principally on dentinal bonding instead of mechanical retention. Microleakage may reflect the bond strength at margins primarily against the forces of polymerization shrinkage15 and these forces may be more difficult to overcome in box tooth preparations when the volume of composite resin is greater. The ultimate bond strength at the tooth-restoration interface is probably related to the bond between the smear layer and underlying dentin surface with bonding agents such as Dual Cure Scotchbond adhesive that leave the smear layer intact. The strength of this bond may be less than the bonding agents such as Scotchbond 2 adhesive that remove the smear layer and bond directly to the exposed’dentin. These results suggested that Scotchbond 2 adhesive was more effective in providing a marginal seal in erosive-type lesions and that microleakage may vary depending on the design of the cavity. However, microleakage is a complex phenomenon and, in this study, neither bonding system prevented microleakage at cementurn-dentin margins as effectively as at the etched enamel margins.
DISCUSSION This study demonstrated that the sealing of two bonding agents can differ depending upon cavity shape. It may be more difficult for both bonding agents to compensate for the polymerization shrinkage generated in box-shaped preparations with a greater volume of composite resin.1° This may be more evident with highly filled posterior composites such as P-30 and P-50 that are more viscous and less capable of diminishing polymerization shrinkage stresses.15In V-shaped preparations, the volume of composite resin and polymerization shrinkage are probably less. Scotchbond 2 adhesive exhibited less microleakage because of its ability to better withstand stress. However, box- and V-shaped gingival preparations alike displayed leakage when restored with Dual Cure Scotchbond, P-30 or Scotchbond 2, P-50 materials. This incomplete bonding could be caused by the expansion-contraction of thermally stressed composite resin, failure of initial bonding to tooth structure, or polymerization shrinkage during the curing of the composite resin. Although not statistically significant, Scotchbond 2 adhesive generally demonstrated less leakage than Dual Cure 444
APRIL
1992
VOLUME
67
NUMBER
4
TWO
DENTIN
BONDING
AGENTS
Crim GA, Swarm ML, Phillips RW. An evaluation of cavosurface design and microleakage. Gen Dent 19&1;32:56-8. 10. Asmussen E, Munksgaard EC. Bonding of restorative resins to dentine: status of dentine adhesives and impact on cavity design and filling techniques. Int Dent J 1988,38:97-104. 11. Mixson JM, Eick JD, Tira DE, Moore DL. The effects of rinse volumes and air and water pressures on enamel-composite resin bond strength.
CONCLUSION
9.
A dentin bonding agent that removed the smear layer recorded less microleakage than an adhesive that bonded to the smear layer in V-shaped preparations. REFERENCES 1. Buonocore MG. A simple method of increasing the adhesion of acrylic filling materials to enamel surfaces. J Dent Res 1955;34:849-53. 2. Gwinnett AJ. Dentin bonding: present status. NY State Dent J 1985; 51635-B. 3. Eick JD, Wilko RA, Anderson CH, Sorensen SE. Scanning electron microscopy of cut tooth surfaces and identification of devices by use of the electron microprobe. J Dent Res 1970;49:1359-68. 4. Neo J, Chalkley Y, Jensen M. Composite resin microleakage: effect of bonding agents and polishing time [Abstract]. J Dent Res 1985;63:179. 5. Hansen EK. Effect of Scotchbond dependent on cavity cleaning, cavity diameter, and cavosurface angle. Stand J Dent Res 1984;92:141-7. 6. Mixson J, Eick D, McMillen S, et al. In vitro microleakage of four bonding agents [Abstract]. J Dent Res 1989$9:966. 7. Tsai Y, Swartz ML, Phillips RW, Moore B. A comparative study: bond strength and microleakage with dentin bond systems. Oper Dent 1990;15:53-60. 8. Retief DH, Woods E, Jamison HC. Effect of cavosurface treatment on marginal leakage in class V composite resin restorations. J PROSTHET
J PROSTHET DENT 1989;62:522-6. 12. Wu U, Cobb E, Derman K, Rupp NW. Detecting marginal leakage of
dental composite restorations. J Biomed Mater Res 1983;17:37-43. 13. Mixson JM, Eick JD, Chappell RP, Tira DE, Moore DL. Comparison
of two-surface and multiple-surface scoring methodologies for in vitro microleakage studies. Dent Mater 1991;7:191-6. 14. Schefler WC. Statistics for the biological sciences. 2nd ed. Reading, Mass: Addison-Wesley Pub1 Co, 1979:174-224. 15. Erickson RL. Dentin bonding agents: a perspective on research and clinical use. J Biomater Appl 1987;1:336-72. 16. Chappell RP, Eick JD, Mixson JM, Theisen FC. Shear bond strength and scanning electron microscopic observation of four dentin adhesives. Quintessence Int 1990;21:303-10. Reprint requests to: DR. JAMES M. MIXSON SCHOOL OF DENTISTRY UNIVERSITY OF MISSOURI 650 EAST 25~~ STREET KANSAS CITY, MO 64108-2795
DENT 1982;47:496-501.
Adhesion T. Ishijima,
of resin to casting
alloys
DDS, PhD,a A. A. Caputo, PhD,b and R. Mito, DDSC
Higashi-Nippon-Gakuen University, School of Dentistry, California, School of Dentistry, Los Angeles, Calif.
Hokkaido,
Japan, and University
of
The bond strength of a composite resin bonded to various dental casting alloys with three adhesive systems-Silicoater, Panavia, and Superbond C&B-was investigated. The metal surfaces were treated solely with aluminum oxide blasting before application of the adhesive. Thermal cycling caused a reduction in bond strength for all combinations of the adhesive systems and alloys, but the Silicoater system recorded the greatest bond strength. The 4-META system was equivalent to Panavia system in bond strengths to most metals and exhibited greater strength DENT 1992;67:445-9.) with others. (J PROSTHET
T
he combination of casting alloys and resin is common in the fabrication of dental prostheses. Classically, the attachment of resins to casting alloys has been accomplished by mechanical retention. During intraoral function, dental prostheses are subjected to an inordinate load and thermal cycling. These factors, individually or in com-
aAssistant Professor, Department of Removable Prosthodontics, Higashi-Nippon-Gakuen University, School of Dentistry. bProfessor and Chairman, Section of Biomaterials Science, UCLA School of Dentistry. CAdjunct Associate Professor and Chairman, Section of Hospital Dentistry, UCLA School of Dentistry.
10/l/32693
THE
JOURNAL
OF PROSTHETIC
DENTISTRY
bination, can exceed the mechanical retention, causing separation at the resin-metal interface. A chemical attachment of the resin to metal improves resistance to separation and ensures greater longevity of the prosthesis. Three new adhesive systems have been developed for creating chemical retention of composite resin to dental casting alloys. These commercial systems are the Silicoater system (Kulzer Inc., Irvine, Calif.), Superbond C&B (Sun MedicalCo.,Ltd.,Kyoto, Japan),andPanaviaEX (Kuraray Co., Ltd., Okayama, Japan). The Silicoater system was introduced by Musil and Tiller.i, 2in 1984 and is based on the adhesion of composite resin to silane bonding agents. It also involves coating the metal with an SiOx-C intermediate layer that bonds to the metal and also supplies the -OH
445
M. Mixson,
David
L. Moore,
University
DMD, DDS,
of Missouri,
MS,a MS,C
J. David
and
School of Dentistry,
Daniel
agents on microleakage
Eick,
in two
PhD,b
E. Tira,
PhDd
Kansas City, MO.
Studies have shown that cavity design and dentinal bonding agents can affect composite resin microleakage. This study compared the microleakage at cementurn-dentin margins of box- and Vshaped preparations restored with two different bonding agents. Twenty freshly extracted, human third molars were prepared with one box-shaped and one V-shaped preparation on the mesial or distal surface. Occlusal margins were terminated in etched enamel and gingival margins were in cementum-dentin. There was no statistically significant difference (p > 0.05) between Dual Cure Scotchbond and Scotchbond 2 samples restored in the boxshaped preparations, but Scotchbond 2 had statistically significantly less microleakage @ < 0.06) than Dual Cure adhesive in V-shaped preparations. The sealing of dentinal bonding agents varied in different cavity designs. (J PROSTHET DENT 1992;67:441-5.)
M*
rcroleakage, or the marginal permeability to bacterial, chemical and molecular invasion at the interface between the teeth and restorative materials remains a problem with composite resin restorations. Two factors that influence the marginal seal are the type of dentin bonding agent and the cavity design. Buonocorel showed that microleakage of composite resins is greatly reduced with acid-etched enamel margins. However, Gwinnett2 noted the difficulty in bonding to vital dentin that is 70% inorganic apatite by weight embedded in a collagen matrix that is 20% by weight. He also stated that preliminary studies had demonstrated that conservative removal of the smear layer before bonding increased adhesion to dentin. Eick et a1.3described the smear layer as a combination of inorganic hydroxyapatite and organic denatured collagen formed by the deposition of the dentinal protein matrix during cutting procedures. Dual Cure Scotchbond adhesive (3M Products Co., St. Paul, Minn.) bonds to an intact smear layer, whereas Scotchbond 2 (3M Products Co.) removes the smear layer before the adhesive is placed. Studies have demonstrated that Dual Cure Scotchbond adhesive significantly reduced microleakage.4p5 Although few studies using Scotchbond 2 have been completed, the preliminary results reveal comparable leakage with these two bonding agents.6,7 The effect of cavity design on a dentinal bonding agent to prevent leakage has been reviewed by numerous investigators. Retief et a1.8found no differences in microleakage between composite resins restored in box tooth prepara-
tions with butt or beveled cavosurface margins. However, Crim et ah9 demonstrated that beveled class V box composite resin restorations displayed significantly less microleakage than identical restorations with butt margins. The shape of the cavity has been recently identified as critical in determining the width of marginal gaps. Asmussen and Munksgaard’O revealed that the marginal gap increased from 4 pm to 13 pm as the cavity design changed from V-shaped to box shaped. This variation in gap formation was explained by the alteration in the ratio between the volume of filling and surface area of cavity walls. This volume:area ratio is elevated from 0.4 mm for V-shaped restorations to 1 mm for box-shaped restorations. The contraction forces increased as the volume increased, and as the surface area of the cavity walls diminished, so did the forces bonding the composite resin to the cavity walls. Although direct statistical comparisons in sectional microleakage studies between box- and V-shaped restorations are impossible because of cavity geometry and the different measurements associated with them, data in a former study6 suggested comparable microleakage for the two designs. The bonding to V-shaped erosion lesions is a challenge to dentists because of the lack of parallel walls or grooves. Studies have not compared the sealing of traditional bonding agents and Scotchbond 2 agent restored in different class V cavity preparations. This investigation compared the in vitro marginal sealing at the cementum-dentin margins of Dual Cure Scotchbond adhesive and Scotchbond 2 in box- and V-shaped tooth preparations.
‘Assistant Professor,Operative Dentistry.
MATERIAL
bCurator’s Professor and Acting Chairman, cProfessor, Chairman, Operative Dentistry. dAssociate Professor, Behavioral Sciences. lOl1134356
THE JOURNAL OFPROSTHETIC
DENTISTRY
Oral Biology.
AND
METHODS
A pilot study was performed to determine the sample size by restoring 10 teeth with composite resin and then sectioning occlusogingivally through the center of the res441
MIXSON
1. Schematic representation for scoring box-shaped restorations. (E) Enamel, (Co) composite resin, (Ce) Cementum. 0, No penetration; 1, up to ‘/z gingival wall; 2, up to gingivoaxial line angle; 3, up to ‘/z axial wall; 4, up to occlusoaxial line angle; 5, up to ‘/z occlusal wall; 6, up to occlusal margin.
Fig.
Fig. 2. Schematic representation for scoring V-shaped restorations. (E) Enamel, (Co) composite resin, (Ce) Cementum. 0, No penetration; 1, up to ‘/z gingival wall; 2, to depth of groove; 3, up to ‘/z occlusal wall; 4, up to occlusal margin.
toration. A mean microleakage value of 5.5 + 1.2 was recorded from surfaces scored by two calibrated examiners. Statistical methodology dictated that a sample size of 10 was necessary for each test condition to detect a statisti442
ET AL
cally significant difference of 20 % with parameters (Y= 0.05 and p = 0.05. Twenty extracted third molar teeth were selected and stored in normal saline at 5’ C for a maximum of 48 hours before preparation. Each tooth was prepared with a No. 169 carbide bur. One box-shaped and one V-shaped preparation were randomly made on either the mesial or distal surface for a total of 40 samples, or 20 box-shaped and 20 V-shaped. The class V box-shaped preparations had a 3 mm occlusogingival (OG) width, a 5 mm buccolingual (BL) length, and a 3 mm axial depth. The V-shaped preparations were designed to simulate an erosion lesion with a 3 mm OG width and a BL length of 6 to 7 mm that tapered at the cavity extremities. The depth at the center of the groove was 3 mm. The width and length for the preparations were checked with a Boley gauge caliber (Miltex Instrument Co., Lake Success, N.Y.) and depth was verified by use of a periodontal probe. Care was exercised to ensure that half of the preparation remained in enamel occlusal to the cementoenamel junction (CEJ), and half was apical to the CEJ in cementum. All enamel margins received a 0.5 mm bevel. Each tooth was randomly restored with either Dual Cure Scotchbond, P-30 or Scotchbond 2, P-50 materials. One material was used per tooth to prevent cross-contamination. The enamel bevels were gel-etched with 37% phosphoric acid for 30 seconds, rinsed 10 seconds, and dried 20 sec0nds.l’ Dentinal bonding agents were applied according to the manufacturer’s and recommendation, the composite resin was then incrementally placed along occlusal and gingival walls. Each increment was light-cured 40 seconds. The restored teeth were stored 7 days in normal saline with thymol, then finished with a No. 8884 diamond finishing bur, and polished with medium and fine Soflex disks (3M Products Co.). The restored teeth were thermocycled for 100 cycles with a 30 seconds dwell between 5’ C and 55” C. The apical foramen, roots, and crowns of the teeth were then covered with one coat of red wax to within 1 mm of the cementumdentin cavosurface margins. The samples were stored in 50 % by weight silver nitrate for 2 hours, rinsed, then placed in radiographic developer and exposed to fluorescent light for 2 hours.12 Teeth were sectioned buccolingually to separate mesial and distal halves with a low-speed diamond saw (Isomet Model 11-1180, Buehler Ltd., Evanston Ill.) and pulpal remnants removed. The halves were mounted in self-curing clear casting resin. Specimens were then sectioned through the restorations into four sections of 0.75 to 1 mm width. Sectioning for the box-shaped restorations began at the end of the buccal or lingual wall and extended to the opposite wall. Sectioning for the V-shaped restorations began 1 to 2 mm inside the restoration and extended to within 1 to 2 mm of the opposite end in order to maintain standard cavity dimensions for scoring. All eight surfaces of the four sections per restorations were scored.13 The degree of stain penetration was assessed at x30
APRIL
1992
VOLUME
67
NUMBER
4
TWO
DENTIN
Table
BONDING
AGENTS
Mann-Whitney
I.
U test of median scores for dentin bonding agents in box-shaped preparations* Median
1 2 3 4 5 6 7 8 9 10
Rank
Scotchbond
Dual core Scotchbond
Sample
Group
scores 2
3.5 10.0 10.0 10.0 10.0 10.0 10.5 17.5 19.5
1.0 1.0 1.5 2.0 2.0 2.0 2.0 2.0 2.5 3.0 iz
1.5 2.0 2.0 2.0 2.0 2.0 2.0 3.0 4.0
median
Dual core Scotchbond
Scotchbond
2
1.5 1.5 3.5 10.0 10.0 10.0 10.0 10.0 16.0
n, lo; u, 65. *Not significant (p > 0.05).
Table
II.
Mann-Whitney
U test of median scores for dentin bonding agents in V-shaped preparations* Median
Sample
Dual core Scotchbond
1 2 3 4 5 6 I 8 9 10
scores
Rank Scotchbond
1.0 1.0 1.0 1.5 1.5 2.0 2.0 2.0 2.0 2.5 iii
Group median
2
0.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.5 2.0 iz
Dual cure Scotchbond 6.5 6.5 6.5 13.0 13.0 17.0 17.0 17.0 17.0 20.0 z = 133.50
Scotchbond
2
1.0 6.5 6.5 6.5 6.5 6.5 6.5 6.5 13.0 17.0 Z = 76.50
n, IO; U, 78.5. *Significant (50.05).
magnification by two calibrated examiners. Microleakage was scored for box-shaped restorations according to the criteria described in Fig. 1 and for V-shaped restorations according to the criteria in Fig. 2. Rater reliability was greater when a 4-point scale was used to score V-shaped preparations. Because each preparation type had a different measurement scale, the effect of cavity design on microleakage of different dentin bonding agents could not be tested. Leakage was scored beginning at gingival cavosurface margins. Intrarater and interrater reliability was calculated by having each examiner independently score the same 40 randomly presented samples, 20 box-shaped and 20 V-shaped on two different occasions 2 weeks apart. During the study a consensus score was reached by all examiners. Examiners did not prepare or restore tooth samples and were uninformed as to the materials or surface scored. The median microleakage score from the eight surfaces per restoration was recorded. The data were subjected to the Mann-Whitney U testi to determine statistically sig-
TiiE
JOURNAL
OF PROSTHETIC
DENTISTRY
nificant differences @ < 0.05) between Dual Cure Scotchbond, P-30 and Scotchbond 2, P-50 materials for each preparation. Intrarater and interrater reliability was calculated by using the Spearman rank correlation14 for boxand V-shaped restorations and resulted in values of >0.74 and >0.84, respectively. Representative photomicrographs were taken at x30 magnification.
RESULTS Results of the Mann-Whitney U test between respective treatment conditions are shown in Tables I and II. There was no statistically significant difference in microleakage between Dual Cure Scotchbond adhesive and Scotchbond 2 adhesive materials placed in box-shaped preparations (Table I). However, Scotchbond 2 adhesive had statistically significantly less microleakage (p < 0.05) than Dual Cure Scotchbond adhesive in V-shaped preparations (Table II), The overall median scores ‘for Dual Care Scotchbond and Scotchbond 2 adhesives in box-shaped prepara443
MIXSON
ET AL
Fig. 3. Silver nitrate penetration observed at tooth-restoration interface of box-shaped restoration,
Fig. 4. Silver nitrate penetration observed at tooth-restoration interface of V-shaped restoration.
tions were both 2, and in V-shaped preparations the medians were 2 and 1, respectively. Leakage patterns were commonly greater at the cavosurface angles of the restorations.13 Typical stain penetration is evident in Figs. 3 and 4, but all of the restorations exhibited a certain degree of microleakage at the toothrestoration interface.
Scotchbond adhesive in class V box preparations. Chappell et a1.16demonstrated earlier that the Scotchbond 2 adhesive had a significantly greater bond strength than other adhesives. Scanning electron microscopic studies revealed that the Scotchbond 2 adhesive completely covered the dentin surface to a thickness of 8 pm and fracture patterns commonly disclosed resin tag penetration into dentinal tubules. This elevated bond strength may decrease microleakage, particularly in gingival V-shaped restorations that rely principally on dentinal bonding instead of mechanical retention. Microleakage may reflect the bond strength at margins primarily against the forces of polymerization shrinkage15 and these forces may be more difficult to overcome in box tooth preparations when the volume of composite resin is greater. The ultimate bond strength at the tooth-restoration interface is probably related to the bond between the smear layer and underlying dentin surface with bonding agents such as Dual Cure Scotchbond adhesive that leave the smear layer intact. The strength of this bond may be less than the bonding agents such as Scotchbond 2 adhesive that remove the smear layer and bond directly to the exposed’dentin. These results suggested that Scotchbond 2 adhesive was more effective in providing a marginal seal in erosive-type lesions and that microleakage may vary depending on the design of the cavity. However, microleakage is a complex phenomenon and, in this study, neither bonding system prevented microleakage at cementurn-dentin margins as effectively as at the etched enamel margins.
DISCUSSION This study demonstrated that the sealing of two bonding agents can differ depending upon cavity shape. It may be more difficult for both bonding agents to compensate for the polymerization shrinkage generated in box-shaped preparations with a greater volume of composite resin.1° This may be more evident with highly filled posterior composites such as P-30 and P-50 that are more viscous and less capable of diminishing polymerization shrinkage stresses.15In V-shaped preparations, the volume of composite resin and polymerization shrinkage are probably less. Scotchbond 2 adhesive exhibited less microleakage because of its ability to better withstand stress. However, box- and V-shaped gingival preparations alike displayed leakage when restored with Dual Cure Scotchbond, P-30 or Scotchbond 2, P-50 materials. This incomplete bonding could be caused by the expansion-contraction of thermally stressed composite resin, failure of initial bonding to tooth structure, or polymerization shrinkage during the curing of the composite resin. Although not statistically significant, Scotchbond 2 adhesive generally demonstrated less leakage than Dual Cure 444
APRIL
1992
VOLUME
67
NUMBER
4
TWO
DENTIN
BONDING
AGENTS
Crim GA, Swarm ML, Phillips RW. An evaluation of cavosurface design and microleakage. Gen Dent 19&1;32:56-8. 10. Asmussen E, Munksgaard EC. Bonding of restorative resins to dentine: status of dentine adhesives and impact on cavity design and filling techniques. Int Dent J 1988,38:97-104. 11. Mixson JM, Eick JD, Tira DE, Moore DL. The effects of rinse volumes and air and water pressures on enamel-composite resin bond strength.
CONCLUSION
9.
A dentin bonding agent that removed the smear layer recorded less microleakage than an adhesive that bonded to the smear layer in V-shaped preparations. REFERENCES 1. Buonocore MG. A simple method of increasing the adhesion of acrylic filling materials to enamel surfaces. J Dent Res 1955;34:849-53. 2. Gwinnett AJ. Dentin bonding: present status. NY State Dent J 1985; 51635-B. 3. Eick JD, Wilko RA, Anderson CH, Sorensen SE. Scanning electron microscopy of cut tooth surfaces and identification of devices by use of the electron microprobe. J Dent Res 1970;49:1359-68. 4. Neo J, Chalkley Y, Jensen M. Composite resin microleakage: effect of bonding agents and polishing time [Abstract]. J Dent Res 1985;63:179. 5. Hansen EK. Effect of Scotchbond dependent on cavity cleaning, cavity diameter, and cavosurface angle. Stand J Dent Res 1984;92:141-7. 6. Mixson J, Eick D, McMillen S, et al. In vitro microleakage of four bonding agents [Abstract]. J Dent Res 1989$9:966. 7. Tsai Y, Swartz ML, Phillips RW, Moore B. A comparative study: bond strength and microleakage with dentin bond systems. Oper Dent 1990;15:53-60. 8. Retief DH, Woods E, Jamison HC. Effect of cavosurface treatment on marginal leakage in class V composite resin restorations. J PROSTHET
J PROSTHET DENT 1989;62:522-6. 12. Wu U, Cobb E, Derman K, Rupp NW. Detecting marginal leakage of
dental composite restorations. J Biomed Mater Res 1983;17:37-43. 13. Mixson JM, Eick JD, Chappell RP, Tira DE, Moore DL. Comparison
of two-surface and multiple-surface scoring methodologies for in vitro microleakage studies. Dent Mater 1991;7:191-6. 14. Schefler WC. Statistics for the biological sciences. 2nd ed. Reading, Mass: Addison-Wesley Pub1 Co, 1979:174-224. 15. Erickson RL. Dentin bonding agents: a perspective on research and clinical use. J Biomater Appl 1987;1:336-72. 16. Chappell RP, Eick JD, Mixson JM, Theisen FC. Shear bond strength and scanning electron microscopic observation of four dentin adhesives. Quintessence Int 1990;21:303-10. Reprint requests to: DR. JAMES M. MIXSON SCHOOL OF DENTISTRY UNIVERSITY OF MISSOURI 650 EAST 25~~ STREET KANSAS CITY, MO 64108-2795
DENT 1982;47:496-501.
Adhesion T. Ishijima,
of resin to casting
alloys
DDS, PhD,a A. A. Caputo, PhD,b and R. Mito, DDSC
Higashi-Nippon-Gakuen University, School of Dentistry, California, School of Dentistry, Los Angeles, Calif.
Hokkaido,
Japan, and University
of
The bond strength of a composite resin bonded to various dental casting alloys with three adhesive systems-Silicoater, Panavia, and Superbond C&B-was investigated. The metal surfaces were treated solely with aluminum oxide blasting before application of the adhesive. Thermal cycling caused a reduction in bond strength for all combinations of the adhesive systems and alloys, but the Silicoater system recorded the greatest bond strength. The 4-META system was equivalent to Panavia system in bond strengths to most metals and exhibited greater strength DENT 1992;67:445-9.) with others. (J PROSTHET
T
he combination of casting alloys and resin is common in the fabrication of dental prostheses. Classically, the attachment of resins to casting alloys has been accomplished by mechanical retention. During intraoral function, dental prostheses are subjected to an inordinate load and thermal cycling. These factors, individually or in com-
aAssistant Professor, Department of Removable Prosthodontics, Higashi-Nippon-Gakuen University, School of Dentistry. bProfessor and Chairman, Section of Biomaterials Science, UCLA School of Dentistry. CAdjunct Associate Professor and Chairman, Section of Hospital Dentistry, UCLA School of Dentistry.
10/l/32693
THE
JOURNAL
OF PROSTHETIC
DENTISTRY
bination, can exceed the mechanical retention, causing separation at the resin-metal interface. A chemical attachment of the resin to metal improves resistance to separation and ensures greater longevity of the prosthesis. Three new adhesive systems have been developed for creating chemical retention of composite resin to dental casting alloys. These commercial systems are the Silicoater system (Kulzer Inc., Irvine, Calif.), Superbond C&B (Sun MedicalCo.,Ltd.,Kyoto, Japan),andPanaviaEX (Kuraray Co., Ltd., Okayama, Japan). The Silicoater system was introduced by Musil and Tiller.i, 2in 1984 and is based on the adhesion of composite resin to silane bonding agents. It also involves coating the metal with an SiOx-C intermediate layer that bonds to the metal and also supplies the -OH
445
