- Email: [email protected]
Evaluation of two dentin adhesives in cervical lesions
Ev.aba%km
of two dtmtin
adhesives
in cervical
lesions
B. Van Meerbeek, DDQ, PhD,a M. Braem, DDS, PhD,b P. Lambrechts, DDS, PhD,C and G. Vanherle, MD, DDSc School of Dentistry, Catholic University of Leuven, Leuven, and Orofacial Morphology and Fun&ion, Rijksuniversitair Centrum Antwerpen, Antwerpen, Belgium The clinical effectiveness of two dentin adhesives, Clearfil New Bond and Scotchbond 2, was evaguated in two di8erent cavity designs. Group A was without enamel bevel or acid etch and with a butt-joint cavity; and group B had enamel bevel, acid etch, and feather-edged cavities. The retentive rate and marginal adaptation were monitored for 2 years. In the Clearfil system, 21% of group A restorations failed after 2 years, whereas virtually all the group B restorations (99%) were retained. In addition, after 2 years, the total of debonded group A restorations in the Scotchbond 2 system expanded to 13%, whereas no restorations from group B were lost. There was clearly marginal deterioration in time irrespective of the bonding system. Nevertheless, the marginal adaptation of cervical lesions restored with Clearfll New Bond adhesive in combination with Clearfil Ray composite resin revealed fewer defects compared with the Scotchbond 2 adhesive with Silux Plus composite resin restorations. SEM evaluation disclosed composite resin remnants on the dentin surface in cavities with lost fillings, whioh indicates partial cohesive failure ofthe adhesive joint, (J PROSTHETDENT 1993;70:308-14.)
T
he innocuous application of acid to the enamel surface was an important advancement in restorative dentistry, and the mechanical retention created by the acid etch is now the basis of reliable enamel b0nding.l Although some researchers initially suggested a similar approach with dentin, many others did not recommend acid etching.2 It is now commonly accepted that merely acid etching dentin does not create equivalent micromechanical retention.” Thus, research has focused on the chemical adhesion of polymers to the dentin surface. However, the abundance of dentinal fluid* and the creation of an irregular smear layer during the cavity preparation prevent the formation of a durable chemical link between the superficial dentinal molecules and the resinous material.5, 6 The present generation of dentin adhesives enhances the bonding potential of dentin by impregnation of the superficial dentinal layer and its smear layer with a specific primer solution. Despite the need for clinical studies on the effectiveness of dentin adhesives, most research is limited to in vitro studies on basic physiologic aspects of dentin.7 This study evaluated the in vivo effectiveness of Clearfil New Bond -_-“‘Aspirant,National Fund for Scientific Researchof Belgium, Department of Operative Dentistry and Dental Materials, Catholic University of Leuven. “Professor,Orofacial Morphology and Function, Rijksuniversitair Centrum Antwerpen. ’ Professor,Department of Operative Dentistry and Dental Materials, Catholic University of Leuven. Copyright c’ 1993by The Editorial Council of THE JOURNAL OF PROSTHETIC DENTISTRY. 0022~3913/93/$1.00 + .lO. 10/l/49047
308
and Scotchbond 2 adhesive systems. The retentive rate and marginal quality of these two systems were monitored for 2 years. MATERIAL
AND
METHODS
Class V cervical lesions caused by erosion, abrasion, or occlusal stress were selected for examination of the bond and marginal adaptation of dentin adhesives. This study involved 306 class V cervical restorations with two different adhesive systems in two cavity categories. In group A, there was no preparation, additional mechanical retention, or etching of the adjacent enamel performed. In these preparations, the dentin cavosurfaces were approximately go-degree butt-joint angles (Fig. 1). In group B, the incisal enamel of the cervical lesions was beveled and acid etched. The cavities were finished to a feather edge (Fig. 1). Clearfil New Bond adhesive (Kuraray, Osaka, Japan) was placed in 189 cervical preparations (group A = 41, group B = 148) followed by the application of Clearfil Ray composite resin (Kuraray). One hundred seventeen fillings (group A = 50, group B = 67) were placed by use of Scotchbond 2 dental adhesive system (3M, Dental Products Div., St. Paul, Minn.) in combination with Silux Plus composite resin (3M). The restorations were inserted randomly in groups A and B. The procedures for conditioning the dentin surface and the successive application of the different products are summarized in Table I. The manufacturer’s instructions were followed precisely, except for the omission of phosphoric acid etching for group A restorations. Clinical efficacy was determined by the percentage of lost restorations after 6 and 24 months of clinical service. The evaluation of the marginal adaptation for the two adhesive
VOLUME70
NUMBER4
VAN
MERRBEEK
THE
ET AL
DENTISTRY
Feather edge
Butt joint
Cleaning
Cleaning
Enamel: beveled + etched
Dentin conditioning
Dentin conditioning
Bonding agent
Bonding agent
Composite resin
Composite resin Fig.
1. A, and B, Two cavity designs.
systems was performed with a mirror and explorer. The results were recorded with an index system (Table II).8 The restorations with excellent marginal adaptation over their entire outline were ranked as Rr, and Rs indicated a slight enamel defect. The restorations with several enamel defects or partial loss of the filling were defined as Rs, and lost restorations were classified as Rd. Small and severe chip fractures at the enamel margin were classified as Rs and Rs, respectively. A small defect at the cervical border was reported as R7, and a severe cervical defect was termed Rs. Finally, the total number of clinically acceptable restorations was computed from a combination of four indexesRI, Rs, Rs, and RT,--recorded as Rizs7 versus Rsdes,which signifies failing restorations. For the evaluation on the scanning electron microscope (SEM) (PSEM 500, Philips, Eindhoven, The Netherlands), an impression of the cervical fillings or cavities with lost fillings was made with Xantopren Blue light body impression material (Bayer Dental AG, Leverkusen, Germany) and cast with Araldite epoxy resin (Araldite DRL and Hardener, Ciba Geigy, Dilbeek, Belgium). All replicas were mounted on metal stubs and coated by vaporization of a pure gold layer of approximately 30 nm in an argon gas en-
1993
OF PROSTHETIC
Group B
Group A
OCTOBER
JOURNAL
vironment for 4 minutes each (SEM coating unit E-1500, Polaron Equipment Ltd., Watford, U.K.). RESULTS Clinical performance
after
6 months
After a 6-month evaluation for both adhesive systems, a decline of excellent restorations was noted, especially those in group A, or without enamel beveling and bonding (Table III). In both adhesive systems, the marginal adaptation was seriously affected and restorations were lost. In group A, Clearfil New Bond adhesive system performed least favorably. Sixteen percent of the Clearfil New Bond restorations were lost during the initial stages of service, whereas only 4.5 % of the Scotchbond 2 fillings failed. In group B, 2% of the Clearfil New Bond restorations failed, but there was complete retention for Scotchbond 2 restorations (Table III, Rd). The group B restorations scored better in regard to the optimal marginal adaptation (Table III, RI) than the group A restorations, except for those with Clearfil New Bond adhesive, which had excellent marginal adaptation in 73 % of the fillings for both cavity designs. For the Scotchbond 2 system, only 30% of the restorations in group A and 56 %
THE
JOI.JRNAt.
OF PROSTHETIC
VAN
DENTISTRY
MEERBEEK
ET AL
I. Clinical procedures for the investigation of adhesives in the two types of cavities
Table
Adhesive
Category
Clearfil New Bond A Enamel Dentin
B
Enamel
Dentin
Cleaning
Pumice-water slurry Pumice-water slurry Pumice-water slurry (3)* Bevel + acid etching (4) Pumice-water slurry (1)
Application
Clearfil
New Bond
Clearfil
Ray -
Clearfil
New
(5) Clearfil Clearfil
Ray (7) New Bond
c&5) Clearfil
Scotchbond2 A Enamel Dentin
B
Enamel
Dentin
Fig. 2. A, SEM photomicrograph of dentin surface of lost Clearfil New Bond restoration reveals several remnants of composite material attached to dentin surface. B, Higher magnificat.ion of enclosed area in A (bar = 1 mm).
of the restoratmns in group B exhibited an optimal marginal adaptation. For the two dentin bonding agents, 98% of group B cervical rest,orations were clinically satisfactory. In group A, 93f’c of the Scotchbond 2 restorations were ranked as clinically acceptable, whereas Clearfil New Bond adhesive was slightly less favorable with a clinical acceptance rate of f&k’, (Table III, Rlz;l;). Clinical
performance
after
24 months
The result,s of marginal adaptation and retention, which demonstrate further decline after 24 months with the most dramatic changes evident in group A, can be seen in Table ITI. In group A: 21”; of the restorations were lost for the Clearfil New Bond system, whereas only 13 % were lost for the Scotchbond 2 system. The retentive rate was more positive for group B because no more restorations were lost after the 6-month control (Table III, RJ. Clearfil New Bond restorations recorded the greatest number of restorations with an optimal marginal adaptation (Table III, RI). An obvious decline in the excellent requ1t.swas not,ed for both adhesive systems if no additional mechanical ret.ention or acid etching was performed. Nev$10
Pumice-water slurry Pumice-water slurry Scotchprep Pumice-water slurry Bevel + acid etching Pumice-water slurry Scotchprep -
Bond
Ray (6) -
Scotchbond Silux Plus -
2
Scotchbond
2
Silux Plus
Scotchbond Silux Plus
2
*Sequence of application
ertheless, 50% of the Clearfil New Bond restorations with technique A had no marginal defects on the entire cavosurface outline of the filling. Conversely, Scotchbond 2 had an excellent margin in less than 10% of the restored lesions. No reduction in marginal quality was noted for group B fillings of the Clearfil New Bond system, but only 24% of the corresponding fillings in the Scotchbond 2 group scored the same ranking (Table III, RI). The percentage of clinically acceptable results was different in both groups (Table III, R1257).After 2 years, only 21 5%of the Clearfil New Bond restorations and 15% of the Scotchbond 2 restorations in group A were clinically acceptable. The restorations placed with additional enamel etching scored better; 94% and 98% of the group B restorations of Clearfil New Bond and Scotchbond 2, respectively, were classified as acceptable. SEM
evaluation
The SEM investigation confirmed cohesive bond failures on the dentin surface of lost group A restorations. For both bonding agents, composite resinous remnants could clear11 be detected on the dentin surface (Figs. 2 and 3). The butt-joint restorations often exhibited gaps and
VAN
MEERBEEK
Table
THE
ET AL
JOURNAL
OF PROSTHETIC
DENTISTRY
Index for marginal adaptation and retention
II. Clinically
acceptable
RI, No marginal defects Rz, Slight defect on the enamelside Rs, Small chip fracture Rr, Small cervical defect Ri257,Clinically acceptable
Clinically
unacceptable
Rs, Severaldefectson the enamelside or partial loss of the filling R4, Loss of restoration Rs, Severechip fracture Rs, Severecervical defect R34ss,Clinically unacceptable
marginal composite resinous fractures causing steep margins (Fig. 4). The feather-edge cavosurfaces displayed a gradual, smoother adaptation to the incisal enamel (Fig. 5). Clearfil New Bond adhesive in combination with Clearfil Ray composite resin exhibited better marginal adaptation than the combination of Scotchbond 2 adhesive with Silux Plus composite resin (Figs. 4 and 5). DISCUSSION The clinical effectiveness of two dentin adhesives is illustrated by its retentive rates in nonbeveled, nonetched cavities, as in group A. This situation is emphasized by the ADA guidelines, which state that a maximum loss of 5% after 6 months and 20% after 3 years is permissible. According to these guidelines, the Scotchbond 2 system would be a candidate for provisional acceptance based on this study’s results. These results were supported by several identical clinical investigations (Table IV).g-ll In these clinical trials for dentin bonding agents, Scotchbond 2 was the first adhesive system that recorded such a high retentive rate. Indeed, Scotchbond 1 (3M) restorations in group A exhibited a loss of 42% after 2 years, whereas the Gluma system (Bayer Dental AG) lost 67% (Table V, Rb).s 12,l3 The progressive improvement in clinical effectiveness of dentin bonding agents is illustrated by the Clearfil New Bond system; although not as recent as the Scotchbond 2 system, Clearfil New Bond adhesive recorded suitable retention, with a loss of 21% after 2 years. This gradual improvement of adhesion for the tested products has also been reported by Tyas.14 Use of a group B configuration resulted in better retention and improved the marginal adaptation. This suggests that the vulnerable dentin adhesion is improved with enamel bonding. Comparing these results with previous studies8912,l3 confirmed the superiority of the presently tested systems. This was especially true after 2 years where the failure rate was lower than in the earlier systems (Table V, R4, arrows). The group B restorations consistently performed better, which again became noteworthy after 2 years. The SEM photomicrographs demonstrated incomplete coverage of the dentin surface with composite resinous remnants (Figs. 2 and 3). The major surface was free of OCTOBER
1993
3. A, SEM photomicrograph of dentin surface of lost Scotchbond 2 restoration discloses several remnants of composite material attached to dentin surface. B, Higher magnification of large enclosed area in A (bar = 1 mm). C, Higher magnification of small enclosed area in A (bar = 0.1 mm). Fig.
remnants, which indicated that the dentin adhesives were incapable of providing a bond over the entire surface for these lost restorations (Figs. 2 and 3). These results are diametrically opposed to the SEM studies with previous bonding systems, in which there were no resinous remnants on the dentin surface.15 Marginal behavior did not improve at a rate similar to the retentive rate for Scotchbond 2 adhesive. The scores for 311
THE
JOURNAL
OF PROSTHETIC
DENTISTRY
VAN
Fig. 4. Incisal cavosurface of Scotchbond 2 restoration from group A characterized by fractured parts, partial debonding, and typically steep margin (bar = 0.1 mm).
Table
III.
following
Results of marginal adaptation the index system of Table II
6 Months C’learfil New Bond Scotchbond
2
24 Months Clearfil New Bond Scotchhond :!
R1257
73
84
30
93
50 8
21 15
IV. Summary of Scotchbond system clinical studies
Rq
RI
Rrm
R4
16 4.5
73 56
98 98
2 0
it,, no marginal det’ect; Rln;. clinically acceptable;
5. Combination of feather-edge, acid etching, Clearfil New Bond adhesive, and Clearfil Ray composite resin provided optimal marginal adaptation. Fig.
Table
B
21 13
ET AL
and retention
A Rl
MEERBEEK
76 24 Rq,
94 98
1 0
loss oi filling.
marginal accuracy for Clearfil New Bond adhesive were more encouraging (Table III, RI). However, the marginal adaptation al the cervical surface was better for the Scotchbond 2 restorations of group A than the marginal adaptation of its predecessor, Scotchbond 1 adhesive in a previous study. After 2 years, 26 I’; of the group A restorations for the Scotchbond 1 system recorded cervical defects versus only- 13 ‘: for Scotchbond 2 adhesive (criterion R7s according to Vanherle et al.“). ‘4 possible explanation for the different clinical performance between the Clearfil New Bond and Scotchbond 2 systems may be attributed to the selection of the composite resin in the final restorative procedure. Clearfil New Bond adhesive. in combination with Clearfil Ray composite resin, displayed a margin superior to the other systems tested in our trials after 2 years (Table V). Furthermore, this was the only combination when the R1 scores for group A restorations equaled or approached those for group B (Table III). The Clearfil Ray product is a hybrid composite resin with a higher volume of filler fraction than that of Silux Plus, which is a typical microfine composite.i6 This difference in composition is clearly reflected in the Young’s modulus, a
Retention
2 dental adhesive rate in percentage
6 Months
Bastos et aLg Duke et al.1° Jordan et al.”
2 Years
85
93 91
86
98
N/A*
*Not available. All restorations were placed by the same method (no enamel bevel, no enamel acid etch, butt-joint).
paramet.er for the resistance to deformation: 9446 MPa _+ 22 for Silux Plus composite resin versus 27384 MPa t 327 for Clearfil Ray composite resin.161 li The higher the modulus of elasticity, the greater the resistance to deformation.lfis 1T When structurally different materials are inserted in the cervical region, stress factors are evident. Extensive incisal or occlusal loads associated with erosive and abrasive lesions induced compressive and tensile stresses at the enamel-dentin junction in the cervical regionIs A bonded composite resin restoration placed cervically is subject to the same stresses as the tooth, certainly if adhesion is involved. The restoration is progressively dislodged at the cavosurfaces and eventually debonds. When a microfine composite resinous material such as Silux Plus with a relatively low modulus of elasticity is inserted (Fig. 6), plastic deformation may compensate for the induced strain. The forces created by compression of the restoration are localized at the bulk of the composite resin as compressive stress and less as shear stress at the adhesive interface. Therefore, the adhesive bond is preserved while the marginal adaptation is adversely affected, which is only valid when the ad-
VOLUME
70
NUMBER
4
VAN
MEERBEFX
ET AL
Fig. 6. Hypothesized mechanism that explains different clinical performance of microfine versus hybrid composite material. When microfine composite material such as Silux Plus with a relatively low modulus of elasticity is placed, plastic deformation may compensate for induced strain. Forces created by compression of filling are localized mainly at bulk of composite resin as compressive stress flarge arrows) with minimal shear stress at adhesive interface (small arrows).
hesive bond is sufficiently strong. A relatively greater retentive rate compared with an inferior marginal adaptation score was noted for the combination of the Scotchbond 2 adhesive with Silux Plus composite. When a more rigid composite resin such as Clearfil Ray is placed (Fig. 7), the shear stress at the adhesive interface can exceed the compression stress and act primarily on the dentinal bond. The rate of loss for Clearfil New Bond adhesive was higher than for Scotchbond 2 adhesive, but the margins of Clearfil New Bond adhesive remained clinically better. This hypothesis was supported with an SEM study by van Dijken et al.lg They discovered that the greater polymerization shrinkage of the microfilled composite resins, in combination with a higher coefficient of thermal expansion, resulted in more severe marginal defects than when hybrid composite resins were used. Lambrechts et a1.20reported cohesive and adhesive chip fractures three to four times more often at the enamel cavosurfaces with microfilled composite resins than with conventional composite resins. The high coefficient of thermal expansion, elevated water sorption, greater polymerization shrinkage, and low tensile strength that are typical of microfilled composite resins were proposed as explanations. CONCLUSION The adhesion of current dental materials to dentin is more promising. Although the clinical effectiveness of present dentin bonding agents has not achieved the effi-
OCTOBER
1993
THE
JOURNAL
OF PROSTHETIC
DENTISTRY
Fig. 7. Hypothesized explanation for different clinical performance of micro&e versus hybrid composite material. When hybrid composite material such as Clearfil Ray is inserted, shear stresses (large arrows) at adhesive interface may exceed the compressive stress (small arrows), acting primarily on dentin bond. Table V. Percentage of lost fillings and marginal
adaptation performance of four different dentin adhesives after 6 and 24 months8* I3 24 Months
6 Months
Group A Scotchbond 1 Gluma Dentin Bond Clearfil New Bond Scotchbond 2 Group B Scotchbond 1 Gluma Dentin Bond Clearfil New Bond Scotchbond 2
Rl
R4
Rl
R4
30.3 10.9 73.0 29.5
15.1
3.9
43.6 16.2 4.5
6.1 50.0 7.5
41.866.7-21.4 -13.2
52.2 58.0 73.2 56.2
2.2 0.0 2.4 0.0
25.0 35.2 75.6 23.9
14.57.4-1.1 -0.0
Arrows highlight the improvement in clinical effectiveness of the two adhesive systems presently investigated over the two systems investigated earlier.
cacy of enamel bonding agents, a trend of improvement is apparent with specific products. Nevertheless, more longitudinal in vivo studies are necessary to support their effectiveness. Clinical marginal adaptation is directly influenced by the type of composite resin used. REFERENCES* 1. Buonocore DH. A simple method of increasing the adhesion of acrylic filling materials to enamel surfaces. J Dent Res 1955;34:849-53. 2. Fusayama T. The problems preventing progress in adhesive restorative dentistry. Adv Dent Res 1988;2:158-61. 3. Asmussen E. Clinical relevance of physical, chemical, and bonding properties of composite resins. Oper Dent 1985;10:61-73.
313
THE
JOURNAL
OF PROSTHETIC
DENTISTRY
VAN
,1-. Psshley DH. Interactions of dental materials with dentin. Proceedings of conference on enamel-dentin-pulp-bone-periodontal tissue interaction with dental materials. Tram Acad Dent Mater 1990;3:55-73. 5. Bowen RL. Bonding agents and adhesives: reactor response. Adv Dent Res 1988;2:155-7. 6. White GJ, Beech DR, Tyas MJ. Dentin smear layer: an asset or a liability for bonding? Dent Mater 1989;5:379-83. 7. Finger WJ. Dentin bonding agents. Relevance of in vitro investigations. Am d Dent 1988;1:184-8. 8. Vanherle G, Verschueren M, Lambrechts P, Braem M. Clinical investigation of dental adhesive systems. Part I: an in-viva study. J PROSTHET DENT 1986;55:157-63, 9. Bastes PA, Leintelder KF, Mazer, RB. CIinical evaluation of Exp-71/72 as a dentin adhesive-two year report. St Paul, Minn: Scotchbond 2 Dental Adhesive System. 3M Clinical Update, 1989. 10. Duke ES, Robbins JW, Snyder DE. Dentin adhesive evaluation for restoring cervical abrasions and root car&. [Abstract] J Dent Res 1989;68:206. il. Jordan RE, Suzuki M, Maclean DF. Early evaluation of Tenure and Scot,chbond 2 for conservative restoration of cervical erosion lesions. J Esthet Dent 1989;1:10-3. 12. Braem M, Lambrechts P, Van Meerbeek B, Vanherle G. Clinical evaluation of two adhesion promotors in cervical abrasive and/or erosive lesions after 6 months. [Abstract]. J Dent Res 1989;68:621. 13. Vanherle G, Lambrechts P, Braem M. An evaluation of different adhesive restorations in cervical lesions. J PROSTHET DENT 1991;65:341-7.
Availability
of
JOURNAL
MEERBEEK
ET AL
14. Tyas MJ. Clinical evaluation of three dentine bonding agents. Aust Dent J 1989;34:559-62. 15. Braem M, Lambrechts P, Vanherle G. Clinical evaluation of dental adhesive systems. Part II: a scanning electron microscopy study. J PROSTHET DENT 1986;55:551-60. 16. Willems G, Lambrechts P, Braem M, Celis JP, Vanherle G. A classification of dental composites according to their morphological and mechanical characteristics. Dent Mater 1992;8:310-9. 17. Braem M, Lambrechts P, Van Doren V, Vanherle G. The impact of composite structure on its elastic response. J Dent Res 1986;65:648-53. 18. Lee WC, Eakle WS. Possible role of tensile stress in the etiology of cevical erosive lesions of teeth. J PROSTHET DENT 1984;62:374-80. 19. Van Dijken J, Hijrstedt P. Marginal adaptation of composite resin restorations placed with or without intermediate low-viscous resin. An SEM investigation. Acta Odontol Stand 1987;45:115-23. 20. Lambrechts P, Ameye C, Vanherle G. Conventional and microfilled composite resins. II. Chip fractures. J PROSTHET DENT 1982;48:527-38. Reprint requests to: DR. B. VAN MEERBEEK DEPARTMENT OF OPERATIVE DENTISTRY AND DENTAL MATERIALS KATHOLIEKE UNIVERSITEIT TE LEUVEN uz ST. RAFAHL KAPUCIJNENVOER 7 B-3000 LEUVEN BELGIUM
back
issues,
1987-1992
Back issues of THE JOURNAL OF PROSTHETIC DENTISTRY are available for purchase from the publisher, Mosby, at a cost of $7.50 per issue. (Foreign postage is not included.) The following quantity discounts are available: 25 % off on quantities of 12 to 23, and one third off on quantities of 24 or more. Please write to Mosby, Subscription Services, 11830 Westline Industrial Drive, St. Louis, MO 63146-3318, or call (800)325-4177, ext. 4351, or (314)453-4351 for information on availability of particular issues for that period from 1987 to 1992. If unavailable from the publisher, photocopies of complete issues are available from University Microforms International, 300 N. Zeeb Rd., Ann Arbor, MI 48106, (313)7614700.
VOLUME
70
NUMBER
4
of two dtmtin
adhesives
in cervical
lesions
B. Van Meerbeek, DDQ, PhD,a M. Braem, DDS, PhD,b P. Lambrechts, DDS, PhD,C and G. Vanherle, MD, DDSc School of Dentistry, Catholic University of Leuven, Leuven, and Orofacial Morphology and Fun&ion, Rijksuniversitair Centrum Antwerpen, Antwerpen, Belgium The clinical effectiveness of two dentin adhesives, Clearfil New Bond and Scotchbond 2, was evaguated in two di8erent cavity designs. Group A was without enamel bevel or acid etch and with a butt-joint cavity; and group B had enamel bevel, acid etch, and feather-edged cavities. The retentive rate and marginal adaptation were monitored for 2 years. In the Clearfil system, 21% of group A restorations failed after 2 years, whereas virtually all the group B restorations (99%) were retained. In addition, after 2 years, the total of debonded group A restorations in the Scotchbond 2 system expanded to 13%, whereas no restorations from group B were lost. There was clearly marginal deterioration in time irrespective of the bonding system. Nevertheless, the marginal adaptation of cervical lesions restored with Clearfll New Bond adhesive in combination with Clearfil Ray composite resin revealed fewer defects compared with the Scotchbond 2 adhesive with Silux Plus composite resin restorations. SEM evaluation disclosed composite resin remnants on the dentin surface in cavities with lost fillings, whioh indicates partial cohesive failure ofthe adhesive joint, (J PROSTHETDENT 1993;70:308-14.)
T
he innocuous application of acid to the enamel surface was an important advancement in restorative dentistry, and the mechanical retention created by the acid etch is now the basis of reliable enamel b0nding.l Although some researchers initially suggested a similar approach with dentin, many others did not recommend acid etching.2 It is now commonly accepted that merely acid etching dentin does not create equivalent micromechanical retention.” Thus, research has focused on the chemical adhesion of polymers to the dentin surface. However, the abundance of dentinal fluid* and the creation of an irregular smear layer during the cavity preparation prevent the formation of a durable chemical link between the superficial dentinal molecules and the resinous material.5, 6 The present generation of dentin adhesives enhances the bonding potential of dentin by impregnation of the superficial dentinal layer and its smear layer with a specific primer solution. Despite the need for clinical studies on the effectiveness of dentin adhesives, most research is limited to in vitro studies on basic physiologic aspects of dentin.7 This study evaluated the in vivo effectiveness of Clearfil New Bond -_-“‘Aspirant,National Fund for Scientific Researchof Belgium, Department of Operative Dentistry and Dental Materials, Catholic University of Leuven. “Professor,Orofacial Morphology and Function, Rijksuniversitair Centrum Antwerpen. ’ Professor,Department of Operative Dentistry and Dental Materials, Catholic University of Leuven. Copyright c’ 1993by The Editorial Council of THE JOURNAL OF PROSTHETIC DENTISTRY. 0022~3913/93/$1.00 + .lO. 10/l/49047
308
and Scotchbond 2 adhesive systems. The retentive rate and marginal quality of these two systems were monitored for 2 years. MATERIAL
AND
METHODS
Class V cervical lesions caused by erosion, abrasion, or occlusal stress were selected for examination of the bond and marginal adaptation of dentin adhesives. This study involved 306 class V cervical restorations with two different adhesive systems in two cavity categories. In group A, there was no preparation, additional mechanical retention, or etching of the adjacent enamel performed. In these preparations, the dentin cavosurfaces were approximately go-degree butt-joint angles (Fig. 1). In group B, the incisal enamel of the cervical lesions was beveled and acid etched. The cavities were finished to a feather edge (Fig. 1). Clearfil New Bond adhesive (Kuraray, Osaka, Japan) was placed in 189 cervical preparations (group A = 41, group B = 148) followed by the application of Clearfil Ray composite resin (Kuraray). One hundred seventeen fillings (group A = 50, group B = 67) were placed by use of Scotchbond 2 dental adhesive system (3M, Dental Products Div., St. Paul, Minn.) in combination with Silux Plus composite resin (3M). The restorations were inserted randomly in groups A and B. The procedures for conditioning the dentin surface and the successive application of the different products are summarized in Table I. The manufacturer’s instructions were followed precisely, except for the omission of phosphoric acid etching for group A restorations. Clinical efficacy was determined by the percentage of lost restorations after 6 and 24 months of clinical service. The evaluation of the marginal adaptation for the two adhesive
VOLUME70
NUMBER4
VAN
MERRBEEK
THE
ET AL
DENTISTRY
Feather edge
Butt joint
Cleaning
Cleaning
Enamel: beveled + etched
Dentin conditioning
Dentin conditioning
Bonding agent
Bonding agent
Composite resin
Composite resin Fig.
1. A, and B, Two cavity designs.
systems was performed with a mirror and explorer. The results were recorded with an index system (Table II).8 The restorations with excellent marginal adaptation over their entire outline were ranked as Rr, and Rs indicated a slight enamel defect. The restorations with several enamel defects or partial loss of the filling were defined as Rs, and lost restorations were classified as Rd. Small and severe chip fractures at the enamel margin were classified as Rs and Rs, respectively. A small defect at the cervical border was reported as R7, and a severe cervical defect was termed Rs. Finally, the total number of clinically acceptable restorations was computed from a combination of four indexesRI, Rs, Rs, and RT,--recorded as Rizs7 versus Rsdes,which signifies failing restorations. For the evaluation on the scanning electron microscope (SEM) (PSEM 500, Philips, Eindhoven, The Netherlands), an impression of the cervical fillings or cavities with lost fillings was made with Xantopren Blue light body impression material (Bayer Dental AG, Leverkusen, Germany) and cast with Araldite epoxy resin (Araldite DRL and Hardener, Ciba Geigy, Dilbeek, Belgium). All replicas were mounted on metal stubs and coated by vaporization of a pure gold layer of approximately 30 nm in an argon gas en-
1993
OF PROSTHETIC
Group B
Group A
OCTOBER
JOURNAL
vironment for 4 minutes each (SEM coating unit E-1500, Polaron Equipment Ltd., Watford, U.K.). RESULTS Clinical performance
after
6 months
After a 6-month evaluation for both adhesive systems, a decline of excellent restorations was noted, especially those in group A, or without enamel beveling and bonding (Table III). In both adhesive systems, the marginal adaptation was seriously affected and restorations were lost. In group A, Clearfil New Bond adhesive system performed least favorably. Sixteen percent of the Clearfil New Bond restorations were lost during the initial stages of service, whereas only 4.5 % of the Scotchbond 2 fillings failed. In group B, 2% of the Clearfil New Bond restorations failed, but there was complete retention for Scotchbond 2 restorations (Table III, Rd). The group B restorations scored better in regard to the optimal marginal adaptation (Table III, RI) than the group A restorations, except for those with Clearfil New Bond adhesive, which had excellent marginal adaptation in 73 % of the fillings for both cavity designs. For the Scotchbond 2 system, only 30% of the restorations in group A and 56 %
THE
JOI.JRNAt.
OF PROSTHETIC
VAN
DENTISTRY
MEERBEEK
ET AL
I. Clinical procedures for the investigation of adhesives in the two types of cavities
Table
Adhesive
Category
Clearfil New Bond A Enamel Dentin
B
Enamel
Dentin
Cleaning
Pumice-water slurry Pumice-water slurry Pumice-water slurry (3)* Bevel + acid etching (4) Pumice-water slurry (1)
Application
Clearfil
New Bond
Clearfil
Ray -
Clearfil
New
(5) Clearfil Clearfil
Ray (7) New Bond
c&5) Clearfil
Scotchbond2 A Enamel Dentin
B
Enamel
Dentin
Fig. 2. A, SEM photomicrograph of dentin surface of lost Clearfil New Bond restoration reveals several remnants of composite material attached to dentin surface. B, Higher magnificat.ion of enclosed area in A (bar = 1 mm).
of the restoratmns in group B exhibited an optimal marginal adaptation. For the two dentin bonding agents, 98% of group B cervical rest,orations were clinically satisfactory. In group A, 93f’c of the Scotchbond 2 restorations were ranked as clinically acceptable, whereas Clearfil New Bond adhesive was slightly less favorable with a clinical acceptance rate of f&k’, (Table III, Rlz;l;). Clinical
performance
after
24 months
The result,s of marginal adaptation and retention, which demonstrate further decline after 24 months with the most dramatic changes evident in group A, can be seen in Table ITI. In group A: 21”; of the restorations were lost for the Clearfil New Bond system, whereas only 13 % were lost for the Scotchbond 2 system. The retentive rate was more positive for group B because no more restorations were lost after the 6-month control (Table III, RJ. Clearfil New Bond restorations recorded the greatest number of restorations with an optimal marginal adaptation (Table III, RI). An obvious decline in the excellent requ1t.swas not,ed for both adhesive systems if no additional mechanical ret.ention or acid etching was performed. Nev$10
Pumice-water slurry Pumice-water slurry Scotchprep Pumice-water slurry Bevel + acid etching Pumice-water slurry Scotchprep -
Bond
Ray (6) -
Scotchbond Silux Plus -
2
Scotchbond
2
Silux Plus
Scotchbond Silux Plus
2
*Sequence of application
ertheless, 50% of the Clearfil New Bond restorations with technique A had no marginal defects on the entire cavosurface outline of the filling. Conversely, Scotchbond 2 had an excellent margin in less than 10% of the restored lesions. No reduction in marginal quality was noted for group B fillings of the Clearfil New Bond system, but only 24% of the corresponding fillings in the Scotchbond 2 group scored the same ranking (Table III, RI). The percentage of clinically acceptable results was different in both groups (Table III, R1257).After 2 years, only 21 5%of the Clearfil New Bond restorations and 15% of the Scotchbond 2 restorations in group A were clinically acceptable. The restorations placed with additional enamel etching scored better; 94% and 98% of the group B restorations of Clearfil New Bond and Scotchbond 2, respectively, were classified as acceptable. SEM
evaluation
The SEM investigation confirmed cohesive bond failures on the dentin surface of lost group A restorations. For both bonding agents, composite resinous remnants could clear11 be detected on the dentin surface (Figs. 2 and 3). The butt-joint restorations often exhibited gaps and
VAN
MEERBEEK
Table
THE
ET AL
JOURNAL
OF PROSTHETIC
DENTISTRY
Index for marginal adaptation and retention
II. Clinically
acceptable
RI, No marginal defects Rz, Slight defect on the enamelside Rs, Small chip fracture Rr, Small cervical defect Ri257,Clinically acceptable
Clinically
unacceptable
Rs, Severaldefectson the enamelside or partial loss of the filling R4, Loss of restoration Rs, Severechip fracture Rs, Severecervical defect R34ss,Clinically unacceptable
marginal composite resinous fractures causing steep margins (Fig. 4). The feather-edge cavosurfaces displayed a gradual, smoother adaptation to the incisal enamel (Fig. 5). Clearfil New Bond adhesive in combination with Clearfil Ray composite resin exhibited better marginal adaptation than the combination of Scotchbond 2 adhesive with Silux Plus composite resin (Figs. 4 and 5). DISCUSSION The clinical effectiveness of two dentin adhesives is illustrated by its retentive rates in nonbeveled, nonetched cavities, as in group A. This situation is emphasized by the ADA guidelines, which state that a maximum loss of 5% after 6 months and 20% after 3 years is permissible. According to these guidelines, the Scotchbond 2 system would be a candidate for provisional acceptance based on this study’s results. These results were supported by several identical clinical investigations (Table IV).g-ll In these clinical trials for dentin bonding agents, Scotchbond 2 was the first adhesive system that recorded such a high retentive rate. Indeed, Scotchbond 1 (3M) restorations in group A exhibited a loss of 42% after 2 years, whereas the Gluma system (Bayer Dental AG) lost 67% (Table V, Rb).s 12,l3 The progressive improvement in clinical effectiveness of dentin bonding agents is illustrated by the Clearfil New Bond system; although not as recent as the Scotchbond 2 system, Clearfil New Bond adhesive recorded suitable retention, with a loss of 21% after 2 years. This gradual improvement of adhesion for the tested products has also been reported by Tyas.14 Use of a group B configuration resulted in better retention and improved the marginal adaptation. This suggests that the vulnerable dentin adhesion is improved with enamel bonding. Comparing these results with previous studies8912,l3 confirmed the superiority of the presently tested systems. This was especially true after 2 years where the failure rate was lower than in the earlier systems (Table V, R4, arrows). The group B restorations consistently performed better, which again became noteworthy after 2 years. The SEM photomicrographs demonstrated incomplete coverage of the dentin surface with composite resinous remnants (Figs. 2 and 3). The major surface was free of OCTOBER
1993
3. A, SEM photomicrograph of dentin surface of lost Scotchbond 2 restoration discloses several remnants of composite material attached to dentin surface. B, Higher magnification of large enclosed area in A (bar = 1 mm). C, Higher magnification of small enclosed area in A (bar = 0.1 mm). Fig.
remnants, which indicated that the dentin adhesives were incapable of providing a bond over the entire surface for these lost restorations (Figs. 2 and 3). These results are diametrically opposed to the SEM studies with previous bonding systems, in which there were no resinous remnants on the dentin surface.15 Marginal behavior did not improve at a rate similar to the retentive rate for Scotchbond 2 adhesive. The scores for 311
THE
JOURNAL
OF PROSTHETIC
DENTISTRY
VAN
Fig. 4. Incisal cavosurface of Scotchbond 2 restoration from group A characterized by fractured parts, partial debonding, and typically steep margin (bar = 0.1 mm).
Table
III.
following
Results of marginal adaptation the index system of Table II
6 Months C’learfil New Bond Scotchbond
2
24 Months Clearfil New Bond Scotchhond :!
R1257
73
84
30
93
50 8
21 15
IV. Summary of Scotchbond system clinical studies
Rq
RI
Rrm
R4
16 4.5
73 56
98 98
2 0
it,, no marginal det’ect; Rln;. clinically acceptable;
5. Combination of feather-edge, acid etching, Clearfil New Bond adhesive, and Clearfil Ray composite resin provided optimal marginal adaptation. Fig.
Table
B
21 13
ET AL
and retention
A Rl
MEERBEEK
76 24 Rq,
94 98
1 0
loss oi filling.
marginal accuracy for Clearfil New Bond adhesive were more encouraging (Table III, RI). However, the marginal adaptation al the cervical surface was better for the Scotchbond 2 restorations of group A than the marginal adaptation of its predecessor, Scotchbond 1 adhesive in a previous study. After 2 years, 26 I’; of the group A restorations for the Scotchbond 1 system recorded cervical defects versus only- 13 ‘: for Scotchbond 2 adhesive (criterion R7s according to Vanherle et al.“). ‘4 possible explanation for the different clinical performance between the Clearfil New Bond and Scotchbond 2 systems may be attributed to the selection of the composite resin in the final restorative procedure. Clearfil New Bond adhesive. in combination with Clearfil Ray composite resin, displayed a margin superior to the other systems tested in our trials after 2 years (Table V). Furthermore, this was the only combination when the R1 scores for group A restorations equaled or approached those for group B (Table III). The Clearfil Ray product is a hybrid composite resin with a higher volume of filler fraction than that of Silux Plus, which is a typical microfine composite.i6 This difference in composition is clearly reflected in the Young’s modulus, a
Retention
2 dental adhesive rate in percentage
6 Months
Bastos et aLg Duke et al.1° Jordan et al.”
2 Years
85
93 91
86
98
N/A*
*Not available. All restorations were placed by the same method (no enamel bevel, no enamel acid etch, butt-joint).
paramet.er for the resistance to deformation: 9446 MPa _+ 22 for Silux Plus composite resin versus 27384 MPa t 327 for Clearfil Ray composite resin.161 li The higher the modulus of elasticity, the greater the resistance to deformation.lfis 1T When structurally different materials are inserted in the cervical region, stress factors are evident. Extensive incisal or occlusal loads associated with erosive and abrasive lesions induced compressive and tensile stresses at the enamel-dentin junction in the cervical regionIs A bonded composite resin restoration placed cervically is subject to the same stresses as the tooth, certainly if adhesion is involved. The restoration is progressively dislodged at the cavosurfaces and eventually debonds. When a microfine composite resinous material such as Silux Plus with a relatively low modulus of elasticity is inserted (Fig. 6), plastic deformation may compensate for the induced strain. The forces created by compression of the restoration are localized at the bulk of the composite resin as compressive stress and less as shear stress at the adhesive interface. Therefore, the adhesive bond is preserved while the marginal adaptation is adversely affected, which is only valid when the ad-
VOLUME
70
NUMBER
4
VAN
MEERBEFX
ET AL
Fig. 6. Hypothesized mechanism that explains different clinical performance of microfine versus hybrid composite material. When microfine composite material such as Silux Plus with a relatively low modulus of elasticity is placed, plastic deformation may compensate for induced strain. Forces created by compression of filling are localized mainly at bulk of composite resin as compressive stress flarge arrows) with minimal shear stress at adhesive interface (small arrows).
hesive bond is sufficiently strong. A relatively greater retentive rate compared with an inferior marginal adaptation score was noted for the combination of the Scotchbond 2 adhesive with Silux Plus composite. When a more rigid composite resin such as Clearfil Ray is placed (Fig. 7), the shear stress at the adhesive interface can exceed the compression stress and act primarily on the dentinal bond. The rate of loss for Clearfil New Bond adhesive was higher than for Scotchbond 2 adhesive, but the margins of Clearfil New Bond adhesive remained clinically better. This hypothesis was supported with an SEM study by van Dijken et al.lg They discovered that the greater polymerization shrinkage of the microfilled composite resins, in combination with a higher coefficient of thermal expansion, resulted in more severe marginal defects than when hybrid composite resins were used. Lambrechts et a1.20reported cohesive and adhesive chip fractures three to four times more often at the enamel cavosurfaces with microfilled composite resins than with conventional composite resins. The high coefficient of thermal expansion, elevated water sorption, greater polymerization shrinkage, and low tensile strength that are typical of microfilled composite resins were proposed as explanations. CONCLUSION The adhesion of current dental materials to dentin is more promising. Although the clinical effectiveness of present dentin bonding agents has not achieved the effi-
OCTOBER
1993
THE
JOURNAL
OF PROSTHETIC
DENTISTRY
Fig. 7. Hypothesized explanation for different clinical performance of micro&e versus hybrid composite material. When hybrid composite material such as Clearfil Ray is inserted, shear stresses (large arrows) at adhesive interface may exceed the compressive stress (small arrows), acting primarily on dentin bond. Table V. Percentage of lost fillings and marginal
adaptation performance of four different dentin adhesives after 6 and 24 months8* I3 24 Months
6 Months
Group A Scotchbond 1 Gluma Dentin Bond Clearfil New Bond Scotchbond 2 Group B Scotchbond 1 Gluma Dentin Bond Clearfil New Bond Scotchbond 2
Rl
R4
Rl
R4
30.3 10.9 73.0 29.5
15.1
3.9
43.6 16.2 4.5
6.1 50.0 7.5
41.866.7-21.4 -13.2
52.2 58.0 73.2 56.2
2.2 0.0 2.4 0.0
25.0 35.2 75.6 23.9
14.57.4-1.1 -0.0
Arrows highlight the improvement in clinical effectiveness of the two adhesive systems presently investigated over the two systems investigated earlier.
cacy of enamel bonding agents, a trend of improvement is apparent with specific products. Nevertheless, more longitudinal in vivo studies are necessary to support their effectiveness. Clinical marginal adaptation is directly influenced by the type of composite resin used. REFERENCES* 1. Buonocore DH. A simple method of increasing the adhesion of acrylic filling materials to enamel surfaces. J Dent Res 1955;34:849-53. 2. Fusayama T. The problems preventing progress in adhesive restorative dentistry. Adv Dent Res 1988;2:158-61. 3. Asmussen E. Clinical relevance of physical, chemical, and bonding properties of composite resins. Oper Dent 1985;10:61-73.
313
THE
JOURNAL
OF PROSTHETIC
DENTISTRY
VAN
,1-. Psshley DH. Interactions of dental materials with dentin. Proceedings of conference on enamel-dentin-pulp-bone-periodontal tissue interaction with dental materials. Tram Acad Dent Mater 1990;3:55-73. 5. Bowen RL. Bonding agents and adhesives: reactor response. Adv Dent Res 1988;2:155-7. 6. White GJ, Beech DR, Tyas MJ. Dentin smear layer: an asset or a liability for bonding? Dent Mater 1989;5:379-83. 7. Finger WJ. Dentin bonding agents. Relevance of in vitro investigations. Am d Dent 1988;1:184-8. 8. Vanherle G, Verschueren M, Lambrechts P, Braem M. Clinical investigation of dental adhesive systems. Part I: an in-viva study. J PROSTHET DENT 1986;55:157-63, 9. Bastes PA, Leintelder KF, Mazer, RB. CIinical evaluation of Exp-71/72 as a dentin adhesive-two year report. St Paul, Minn: Scotchbond 2 Dental Adhesive System. 3M Clinical Update, 1989. 10. Duke ES, Robbins JW, Snyder DE. Dentin adhesive evaluation for restoring cervical abrasions and root car&. [Abstract] J Dent Res 1989;68:206. il. Jordan RE, Suzuki M, Maclean DF. Early evaluation of Tenure and Scot,chbond 2 for conservative restoration of cervical erosion lesions. J Esthet Dent 1989;1:10-3. 12. Braem M, Lambrechts P, Van Meerbeek B, Vanherle G. Clinical evaluation of two adhesion promotors in cervical abrasive and/or erosive lesions after 6 months. [Abstract]. J Dent Res 1989;68:621. 13. Vanherle G, Lambrechts P, Braem M. An evaluation of different adhesive restorations in cervical lesions. J PROSTHET DENT 1991;65:341-7.
Availability
of
JOURNAL
MEERBEEK
ET AL
14. Tyas MJ. Clinical evaluation of three dentine bonding agents. Aust Dent J 1989;34:559-62. 15. Braem M, Lambrechts P, Vanherle G. Clinical evaluation of dental adhesive systems. Part II: a scanning electron microscopy study. J PROSTHET DENT 1986;55:551-60. 16. Willems G, Lambrechts P, Braem M, Celis JP, Vanherle G. A classification of dental composites according to their morphological and mechanical characteristics. Dent Mater 1992;8:310-9. 17. Braem M, Lambrechts P, Van Doren V, Vanherle G. The impact of composite structure on its elastic response. J Dent Res 1986;65:648-53. 18. Lee WC, Eakle WS. Possible role of tensile stress in the etiology of cevical erosive lesions of teeth. J PROSTHET DENT 1984;62:374-80. 19. Van Dijken J, Hijrstedt P. Marginal adaptation of composite resin restorations placed with or without intermediate low-viscous resin. An SEM investigation. Acta Odontol Stand 1987;45:115-23. 20. Lambrechts P, Ameye C, Vanherle G. Conventional and microfilled composite resins. II. Chip fractures. J PROSTHET DENT 1982;48:527-38. Reprint requests to: DR. B. VAN MEERBEEK DEPARTMENT OF OPERATIVE DENTISTRY AND DENTAL MATERIALS KATHOLIEKE UNIVERSITEIT TE LEUVEN uz ST. RAFAHL KAPUCIJNENVOER 7 B-3000 LEUVEN BELGIUM
back
issues,
1987-1992
Back issues of THE JOURNAL OF PROSTHETIC DENTISTRY are available for purchase from the publisher, Mosby, at a cost of $7.50 per issue. (Foreign postage is not included.) The following quantity discounts are available: 25 % off on quantities of 12 to 23, and one third off on quantities of 24 or more. Please write to Mosby, Subscription Services, 11830 Westline Industrial Drive, St. Louis, MO 63146-3318, or call (800)325-4177, ext. 4351, or (314)453-4351 for information on availability of particular issues for that period from 1987 to 1992. If unavailable from the publisher, photocopies of complete issues are available from University Microforms International, 300 N. Zeeb Rd., Ann Arbor, MI 48106, (313)7614700.
VOLUME
70
NUMBER
4