Effect
of bonded
amalgam
W. Stephan Eakle, DDSTa Michal Alton M. Lacy, DDS, PhDC
Staninec,
on the fracture DDS,b
resistance
of teeth
and
University of California,SanFrancisco,Calif. The purpose of this study was to determine whether amalgam bonded to tooth structure with an adhesive resin cement can increase the fracture resistance of restored teeth. Extracted paired upper premolars were prepared for G.V. Black type mesioocclusodistal amalgam restorations. In one tooth of each pair (the experimental group), the enamel walls were etched with phosphoric acid and were painted with an adhesive resin (Panavia), and amalgam was condensed and carved. For the other tooth in each pair (the control group), amalgam was placed in the same manner but was not etched and lined with resin. The teeth were thermocycled and mounted for testing and then were loaded until fracture. A significant difference (p < 0.05, the paired Student’s t-test) was found in the force needed to fracture the bonded amalgam group (70.5 i 21.6 kg) compared with that needed to fracture the conventional amalgam group (60.3 i- 16.8 kg). SEM examination of fractures at the interface occurred predominantly within the resin. (J PROSTHET DENT 1992;68:
257-60.)
emoval of tooth structure by cavity preparation has beenshown to weaken teeth and to increasetheir susceptibility to fracture.l-g Compositeresin restorations that have been bonded to etched enamelincreasethe fracture resistanceof the teeth.1°-r5The major shortcomingsof the compositeresinshave beenrapid wear and the sequelaeof polymerization shrinkage in functional areasof posterior teeth.rs-rgGold and porcelain inlays bonded to teeth have alsobeen shown to increasefracture resistance20,21; however, the cost is prohibitive for many patients. Conventionally placed amalgam restoresthe tooth contours and occlusionbut has no adhesionto tooth structure and thus doesnot strengthen the prepared tooth. A restoration that hasthe durability of amalgamand that increasesthe fracture resistanceof the tooth by bonding cuspsand restoration together is desirable. The purpose of this study was to compare the fracture resistanceof teeth restored with conventional amalgam with those restored with amalgambonded to tooth structure with a resin liner. MATERIAL
AND
METHODS
Matched pairs of caries-free undamagedmaxillary premolars that were extracted for orthod.ontic purposeswere usedin the study. The teeth were mounted in dental stone for convenienceof handling during cavity preparation and placement of amalgam.
Presentedat the meetingof the AmericanAssociationfor Dental Research, Cincinnati,Ohio. aAssociate Professor,Departmentof RestorativeDentistry. bAssociateClinical Professor,Departmentof RestorativeDentistry. CAssociate Professor,Departmentof RestorativeDentistry. 10/l/37726
THE
JOURNAL
OF PROSTHETIC
DENTISTRY
Fourteen pairs of teeth were prepared with mesioocclusodistal (MOD) amalgamtype cavities by meansof high speedwith water spray and new No. 56 burs. The isthmus width wasonethird the intercuspal distance.Proximal box forms were 4 mm wide faciolingually and were 1.5mm deep at the gingival floor. The gingival floor was located in enamel1mm coronalto the cementoenameljunction (CEJ) (Fig. 1). Metal gaugeswere used as guides to keep the preparations asidentical aspossible.The teeth werestored in 100% humidity after preparation. One tooth in each pair was randomly assignedto the conventional amalgamgroup and the other to the bonded amalgamgroup. In the conventional amalgamgroup, metal matrix bandswere placed on the teeth, and amalgam(Tytin, Kerr/Sybron, Romulus, Mich.) was condensedvertically and laterally into the preparations. After condensation, the amalgamwascarved to the marginsfollowing the contours of the tooth but wasnot finished and polished.In the bonded amalgamgroup, the enamelwas etched for 30 secondswith 37% phosphoric acid gel and wasthen rinsed for 30 secondsand dried with oil-free air. An autopolymerizing resin cement (Panavia EX, Kuraray Co., Ltd., Osaka, Japan) wasmixed to a creamy consistencyaccording to the manufacturer’s instructions and a thin layer was painted on the enameland dentin of the preparations with a fine brush. Amalgam was immediately condensed into the preparations and wascarved as in the conventional amalgamgroup, The cavosurfacemarginswere coated with the polyvinyl alcohol gel (Oxyguard, Kuraray Co., Ltd.) supplied with the resin cement to prevent oxygen inhibition of the set of the resin. The teeth were returned to 100% humidity after restoration. Twenty-four hours after restoration, the teeth were removed from the dental stone and were thermal cycled between 5OC and 55’ C water baths for 240cycles. The time 257
EAKLE,
l/3
intercuspal
STANINEC,
AND
LACY
distance
I
3. After fracture of tooth, amalgaminterface shows resin liner firmly attached. Fig.
1. Teeth prepared with sametype MOD cavities with parallel walls, an isthmus width one third the intercuspal distance, and cavosurfacemarginsin enamelwith no bevels. Fig.
I. Load to fracture teeth restored with conventional amalgamversus bonded amalgam Table
Mean Groups
n
Amalgam Bondedamalgam Matched
pairs
t-test:
fracture Load
14
(kg)
SD
60.3 70.5
14
rf:16.8 +21.6
p < 0.05
were examined by scanning electron microscopy (Model SX 40A, International Scientific Instruments, Inc., Milpitas, Calif.). RESULTS
2. Testing head contacted inclines of cuspsbut did not contact restoration. A continually increasingload was applied until fracture of tooth.
Fig.
in each bath was 30 seconds.Then the roots of the teeth were given a thin coating of vinyl plastic (PlastiDip, PDI, Inc., Blaine, Minn.) to simulatethe periodontal membrane. Teeth were mounted in acrylic resin blocks,and their roots were embeddedto within 2 mm of the CEJ and then were stored in tap water for 48 hours before testing. A universal testing machine (Model 1122,Instron Corp., Canton, Mass.) with a tapered plunger wasusedto deliver a continually increasingload at 5 mm/min to the occlusal surfacesof the restored teeth until the point of fracture. The plunger contacted only the occlusalinclines of the facial and lingual cuspsand not the restorations (Fig. 2). The forcesrequired to fracture the teeth wererecorded, and the results were analyzed with the matched pairs t-test. Fractured surfacesof the cavity preparations and restorations 258
Teeth restored with amalgambonded to the tooth fractured at a significantly greater load than did teeth restored with conventional amalgam(Table I). In the conventional amalgamgroup, all the specimens fractured through the cavity preparation. One cusp separated from the remainder of the tooth, or the tooth split longitudinally down the root. In the bonded amalgam group, most of the specimensfractured through the resin interface, which separateda cuspfrom the rest of the tooth as in the conventional amalgamgroup. The cusp that did not fracture remained bondedto the restoration. Scanning electron microscopicexamination of a buccolingual longitudinal section through the bonded amalgamat the interface of the amalgam,resin, and enamelshoweda thin, irregular layer of resin with extensionsand small islandsof embeddedamalgam.Examination of the bonded amalgam surfacethat had separatedfrom the cavity wall after fracture showedamalgamwith resin attached (Fig. 3). Magnification of amalgamsurrounded by resin showedthe intimate interlocking of the resin and amalgamthat resulted from the condensationof freshly mixed amalgamonto wet (unset) resin (Fig. 4). Fig. 5 showsthe adaptation of the
AUGUST
1992
VOLUME
68
NUMBER
2
EFFECT
OF BONDED
AMALGAM
ON TOOTH
STRENGTH
Fig. 4. Magnification of amalgam interface shows intimate interlocking of resin and amalgam that resulted from condensing amalgam onto unset resin.
Fig. 5. Mesiodistal section of restoration and tooth at gingival margin shows close adaptation of amalgam and resin liner to etched enamel margin. Resin is seen extending into amalgam, with which it interlocks.
amalgam and resin liner to the enamel at the gingival margin. For the most part, the cohesive strength of the resin was the limiting factor in the reinforcement of these teeth against fracture. However, four of the bonded specimens fractured through the tooth and a portion of the amalgam, which left amalgam bonded to both buccal and lingual cusps (Fig. 6). There was no evidence, such as a mark or shiny spot, on the amalgam to indicate that the testing head had contacted the amalgam. DISCUSSION Teeth restored with amalgam bonded to etched enamel with a resin cement had a significantly greater fracture resistance than did teeth restored with conventional amalgam. The adhesive resin used in this study develops micromechanical retention with etched enamel in the same way that other restorative resins do. It may also develop some chemical bonds via phosphate esters that interact with calcium ions in the tooth. When amalgam is packed into the cavity over the wet resin, a mechanical interlocking of resin and amalgam occurs. This interlocking is probably a much more significant factor in retention of the amalgam than are the chemical bonds that occur between the resin and the components of the amalgam, as Lacy et a1.22have suggested. Staninecz3 found that in class II cavities, amalgam bonded to the tooth resulted in a restoration that was more resistant to displacement than was conventionally placed amalgam retained with proximal grooves or dovetails. If this finding holds true in clinical trials, then the bonding of amalgam to tooth structure will permit more conservative cavity preparations. Results of this study are consistent with results of similar studies of fracture resistance with restorations (gold inlays, composite resins, and porcelain inlays) bonded to tooth structure. Nearly all these studies reported an increased fracture resistance with bonded restorations.lOX,B, 21 Typical cuspal fractures with conventional MOD
THE
JOURNAL
OF PROSTHETIC
DENTISTRY
Fig. 6. Fractured tooth that also fractured through a portion of the amalgam and left amalgam fragments still bonded to cavity walls. There is no sign on amalgam that testing head contacted amalgam along the fracture site.
restorations occur at the base of the cusp and angle to the external surface so that they usually are supragingival or slightly subgingiva14 In the laboratory, teeth with conventional or bonded MOD restorations frequently fracture with a cusp spliting vertically into the root. We have observed this same type of fracture clinically with teeth restored with bonded MOD composite resins. Because bonded restorations reduce cusp flexure, there should be less long-term structural fatigue of the tooth at the base of the cusps and therefore fewer cuspal fractures. However, a load that exceeds the cohesive or shear bond strength of the resin and fractures the tooth might fracture into the root rather than at the base of the cusp, which makes restoration of the tooth more difficult. It is our opinion that bonded intracoronal restorations should not be substituted for circumferential reinforcement of teeth when the operator
EAXLE,
judges that the teeth are weak and susceptible to fracture. Staninec and Holtz4 found that the bonded amalgam had a mean tensile bond strength of 1404 psi to etched enamel and of 469 psi to dentin. This evidence indicates that the bond of amalgam to tooth via a resin liner is strong. Our results showed that the restoration remained bonded to the tooth in some of the fractures that occurred through the amalgam. A study of microleakage with bonded amalgam indicates a better seal to the tooth than that produced with composite resinz4 In another study, the progression of artificial caries was inhibited along cavity walls in bonded amalgam when compared with that progression in conventional amalgam without the resin liner.25 Although the bonded amalgam technique looks very promising as a result of laboratory studies, the limitations of these studies must be recognized. The continually increasing load applied to the teeth in this study is not typical of the type of loading that occurs clinically. Except for a single traumatic incidence, teeth that crack or fracture in the mouth usually do so as a result of repeated episodes of stress that fatigue the crystalline structure and produce microcracks that propagate until failure of the structure occurs. The effects of masticatory stresses on the durability of the resin bond also were not determined. It is not known whether microleakage, recurrent caries, or retention of the amalgam is a problem if the resin fatigues and breaks down. Long-t,erm clinical trials are needed to assess these factors and to determine the longevity of bonded amalgam restorations and their usefulness in reducing the incidence of tooth fracture. The laboratory studies and recent clinical applications have, in the short-term, demonstrated that amalgam can be bonded to tooth structure and can thereby produce cuspal reinforcement.
SUMMARY
3. 4. 5. 6.
7. 8. 9.
10.
11. 12. 13.
14. 15. 16.
17.
18.
19. 20.
21.
23. 24
25
We gratefully acknowledge the help of Mr. Larry Watanabe and Dr. Eric Clark with the technical procedures and that of Ms. Vickie Leow wit,h graphic illustrations.
REFERENCES 1. Snyder, DE. The cracked-tooth Oral Surg 1976;41:698-704.
260
syndrome
and fractured
posterior
cusp.
AND
LACY
2. Cameron
22.
In this laboratory study, teeth restored with bonded amalgam were significantly more resistant to fracture than were teeth restored with conventional amalgam. The mode of failure of bonded amalgam restorations was mixed, but cohesive resin failure predominated.
STANIA-EC,
CE. The cracked tooth syndrome: additional findings. J Am Dent Assoc 1976;93:971-5. Cave1 WT, K&q WP: Blankenau RJ. An in viva study of cuspal fracture J PROSTHET DENT 1985;53:38-42. Eakle WS, Maxwell EH, Braly BV. Fractures of posterior adult teeth. J Am Dent Assoc 1986;112:215-8. Vale WA. Cavity preparation and further thoughts on high speed. Br Dent J 1959;107:333-6. Mondelli J, Steagall L, Ishikiriama A, Nawrro M, Soares F. Fracture strength of human teeth with cavity preparations. J PROSTHET DENT 1980;43:419-22. Larson TD, Douglas WH, Geistfeld RE. Effect of prepared cavities on the strength of teeth. Oper Dent 1981;6:2-5. Bell JG, Smith MC, dePont JJ. Cuspal failures of MOD restored teeth. Aust Dent J 1982;27:283-7. Blaser PK, Lund MR, Cochran MA, Potter RH. Effects of design of class II preparations on resistance of teeth to fracture, Oper Dent 1983;8:610. Share J, Mishell Y, Nathanson D. Effect of restorative mat,erial on fracture resistance of tooth structure in vitro [Abstract]. J Dent Res 1982;61:247. Simonsen RJ, Barouch E: Gelb M. Cusp fracture resistance from composite resin in class II restorations [Abstract]. J Dent Res 1983;62:254. Landy NA, Simonsen RJ. Cusp fracture strength in class II composite resin restorations [Abstract]. J Dent Res 1984:63:175. M&hell Y, Share J, Nathanson D. Fracture resistance of Cl. II amalgam vs. light activated composite restorations in vitro [Abstract]. J Dent Res 1984;63:293. Newman SM, Pisko-Dubienshy R. Effects of composite restorations on strength of posterior teeth [Abstract]. J Dent Res 1984;63:523. Eakle WS. Fracture resistance of teeth with class II bonded composite restorations. J Dent Res 1986;65:149-53. Phillips RW, Avery DR, Mehra RJ, Swarm ML, McCune RJ. Observat,ions on a composite resin for class II restorations: three year report. J PROSTHF,T DENT 1973;30:891-7. Leinfelder KF: Sluder TB, Santos JF, Wall JT. Five-year clinical evaluation of anterior and posterior restorations of composite resin. Oper Dent 1980;5:57-65. Jensen ME, Redford DA, Williams BT, Gardner F. Posterior etchedporcelain restorations: an in vitro study. Compend Contin Educ Dent 1987;8:615-22. Lutz F, Moermann W. In viva wear of posterior tooth composites [Abstract], J Dent Res 1982;61:215. Moffa JP, Jenkins WA, Hamilton JC. The longevity of composite resins for the restoration of posterior teeth [Abstract]. J Dent Res 1984;63:199. Eakle WS, Staninec M, Clark EJ. Effect of bonded inlays on fracture resistance of teeth [Abstract]. J Dent Res 1989;68:303. Lacy A, Rupprecht R, Watanabe L, Hiramatsu D. Amalgam-amalgam and amalgam-composite resin bond strengths [Abstract]. J Dent Res 1989;68:189. Staninec M. Retention of amalgam restorations: undercuts versus bonding. Quintessence Int 1989;20:347-51. Staninec M, Halt M. Bonding of amalgam to tooth structure: tensile adhesion and microleakage tests. J PROSTHET D~XT 1988:59:397402. Torii Y, Staninec M, Kawakami M, Imazato S, Torri M, Tsuchitani Y. Inhibition in vitro of caries around amalgam restorations by bonding amalgam to tooth structure. Oper Dent 1989;14:142-8.
Reprint requests to: DR. W. STEPHAN EAKLE DEPARTMENT OF RESTORATIVE SCHOOL OF DENTISTRY UNIVERSITY OF CALIFORNIA SAN FRANCISCO: CA 94143
DENTISTRY
AUGUST
1992
VOLUME
68
NUMBER
2
of bonded
amalgam
W. Stephan Eakle, DDSTa Michal Alton M. Lacy, DDS, PhDC
Staninec,
on the fracture DDS,b
resistance
of teeth
and
University of California,SanFrancisco,Calif. The purpose of this study was to determine whether amalgam bonded to tooth structure with an adhesive resin cement can increase the fracture resistance of restored teeth. Extracted paired upper premolars were prepared for G.V. Black type mesioocclusodistal amalgam restorations. In one tooth of each pair (the experimental group), the enamel walls were etched with phosphoric acid and were painted with an adhesive resin (Panavia), and amalgam was condensed and carved. For the other tooth in each pair (the control group), amalgam was placed in the same manner but was not etched and lined with resin. The teeth were thermocycled and mounted for testing and then were loaded until fracture. A significant difference (p < 0.05, the paired Student’s t-test) was found in the force needed to fracture the bonded amalgam group (70.5 i 21.6 kg) compared with that needed to fracture the conventional amalgam group (60.3 i- 16.8 kg). SEM examination of fractures at the interface occurred predominantly within the resin. (J PROSTHET DENT 1992;68:
257-60.)
emoval of tooth structure by cavity preparation has beenshown to weaken teeth and to increasetheir susceptibility to fracture.l-g Compositeresin restorations that have been bonded to etched enamelincreasethe fracture resistanceof the teeth.1°-r5The major shortcomingsof the compositeresinshave beenrapid wear and the sequelaeof polymerization shrinkage in functional areasof posterior teeth.rs-rgGold and porcelain inlays bonded to teeth have alsobeen shown to increasefracture resistance20,21; however, the cost is prohibitive for many patients. Conventionally placed amalgam restoresthe tooth contours and occlusionbut has no adhesionto tooth structure and thus doesnot strengthen the prepared tooth. A restoration that hasthe durability of amalgamand that increasesthe fracture resistanceof the tooth by bonding cuspsand restoration together is desirable. The purpose of this study was to compare the fracture resistanceof teeth restored with conventional amalgam with those restored with amalgambonded to tooth structure with a resin liner. MATERIAL
AND
METHODS
Matched pairs of caries-free undamagedmaxillary premolars that were extracted for orthod.ontic purposeswere usedin the study. The teeth were mounted in dental stone for convenienceof handling during cavity preparation and placement of amalgam.
Presentedat the meetingof the AmericanAssociationfor Dental Research, Cincinnati,Ohio. aAssociate Professor,Departmentof RestorativeDentistry. bAssociateClinical Professor,Departmentof RestorativeDentistry. CAssociate Professor,Departmentof RestorativeDentistry. 10/l/37726
THE
JOURNAL
OF PROSTHETIC
DENTISTRY
Fourteen pairs of teeth were prepared with mesioocclusodistal (MOD) amalgamtype cavities by meansof high speedwith water spray and new No. 56 burs. The isthmus width wasonethird the intercuspal distance.Proximal box forms were 4 mm wide faciolingually and were 1.5mm deep at the gingival floor. The gingival floor was located in enamel1mm coronalto the cementoenameljunction (CEJ) (Fig. 1). Metal gaugeswere used as guides to keep the preparations asidentical aspossible.The teeth werestored in 100% humidity after preparation. One tooth in each pair was randomly assignedto the conventional amalgamgroup and the other to the bonded amalgamgroup. In the conventional amalgamgroup, metal matrix bandswere placed on the teeth, and amalgam(Tytin, Kerr/Sybron, Romulus, Mich.) was condensedvertically and laterally into the preparations. After condensation, the amalgamwascarved to the marginsfollowing the contours of the tooth but wasnot finished and polished.In the bonded amalgamgroup, the enamelwas etched for 30 secondswith 37% phosphoric acid gel and wasthen rinsed for 30 secondsand dried with oil-free air. An autopolymerizing resin cement (Panavia EX, Kuraray Co., Ltd., Osaka, Japan) wasmixed to a creamy consistencyaccording to the manufacturer’s instructions and a thin layer was painted on the enameland dentin of the preparations with a fine brush. Amalgam was immediately condensed into the preparations and wascarved as in the conventional amalgamgroup, The cavosurfacemarginswere coated with the polyvinyl alcohol gel (Oxyguard, Kuraray Co., Ltd.) supplied with the resin cement to prevent oxygen inhibition of the set of the resin. The teeth were returned to 100% humidity after restoration. Twenty-four hours after restoration, the teeth were removed from the dental stone and were thermal cycled between 5OC and 55’ C water baths for 240cycles. The time 257
EAKLE,
l/3
intercuspal
STANINEC,
AND
LACY
distance
I
3. After fracture of tooth, amalgaminterface shows resin liner firmly attached. Fig.
1. Teeth prepared with sametype MOD cavities with parallel walls, an isthmus width one third the intercuspal distance, and cavosurfacemarginsin enamelwith no bevels. Fig.
I. Load to fracture teeth restored with conventional amalgamversus bonded amalgam Table
Mean Groups
n
Amalgam Bondedamalgam Matched
pairs
t-test:
fracture Load
14
(kg)
SD
60.3 70.5
14
rf:16.8 +21.6
p < 0.05
were examined by scanning electron microscopy (Model SX 40A, International Scientific Instruments, Inc., Milpitas, Calif.). RESULTS
2. Testing head contacted inclines of cuspsbut did not contact restoration. A continually increasingload was applied until fracture of tooth.
Fig.
in each bath was 30 seconds.Then the roots of the teeth were given a thin coating of vinyl plastic (PlastiDip, PDI, Inc., Blaine, Minn.) to simulatethe periodontal membrane. Teeth were mounted in acrylic resin blocks,and their roots were embeddedto within 2 mm of the CEJ and then were stored in tap water for 48 hours before testing. A universal testing machine (Model 1122,Instron Corp., Canton, Mass.) with a tapered plunger wasusedto deliver a continually increasingload at 5 mm/min to the occlusal surfacesof the restored teeth until the point of fracture. The plunger contacted only the occlusalinclines of the facial and lingual cuspsand not the restorations (Fig. 2). The forcesrequired to fracture the teeth wererecorded, and the results were analyzed with the matched pairs t-test. Fractured surfacesof the cavity preparations and restorations 258
Teeth restored with amalgambonded to the tooth fractured at a significantly greater load than did teeth restored with conventional amalgam(Table I). In the conventional amalgamgroup, all the specimens fractured through the cavity preparation. One cusp separated from the remainder of the tooth, or the tooth split longitudinally down the root. In the bonded amalgam group, most of the specimensfractured through the resin interface, which separateda cuspfrom the rest of the tooth as in the conventional amalgamgroup. The cusp that did not fracture remained bondedto the restoration. Scanning electron microscopicexamination of a buccolingual longitudinal section through the bonded amalgamat the interface of the amalgam,resin, and enamelshoweda thin, irregular layer of resin with extensionsand small islandsof embeddedamalgam.Examination of the bonded amalgam surfacethat had separatedfrom the cavity wall after fracture showedamalgamwith resin attached (Fig. 3). Magnification of amalgamsurrounded by resin showedthe intimate interlocking of the resin and amalgamthat resulted from the condensationof freshly mixed amalgamonto wet (unset) resin (Fig. 4). Fig. 5 showsthe adaptation of the
AUGUST
1992
VOLUME
68
NUMBER
2
EFFECT
OF BONDED
AMALGAM
ON TOOTH
STRENGTH
Fig. 4. Magnification of amalgam interface shows intimate interlocking of resin and amalgam that resulted from condensing amalgam onto unset resin.
Fig. 5. Mesiodistal section of restoration and tooth at gingival margin shows close adaptation of amalgam and resin liner to etched enamel margin. Resin is seen extending into amalgam, with which it interlocks.
amalgam and resin liner to the enamel at the gingival margin. For the most part, the cohesive strength of the resin was the limiting factor in the reinforcement of these teeth against fracture. However, four of the bonded specimens fractured through the tooth and a portion of the amalgam, which left amalgam bonded to both buccal and lingual cusps (Fig. 6). There was no evidence, such as a mark or shiny spot, on the amalgam to indicate that the testing head had contacted the amalgam. DISCUSSION Teeth restored with amalgam bonded to etched enamel with a resin cement had a significantly greater fracture resistance than did teeth restored with conventional amalgam. The adhesive resin used in this study develops micromechanical retention with etched enamel in the same way that other restorative resins do. It may also develop some chemical bonds via phosphate esters that interact with calcium ions in the tooth. When amalgam is packed into the cavity over the wet resin, a mechanical interlocking of resin and amalgam occurs. This interlocking is probably a much more significant factor in retention of the amalgam than are the chemical bonds that occur between the resin and the components of the amalgam, as Lacy et a1.22have suggested. Staninecz3 found that in class II cavities, amalgam bonded to the tooth resulted in a restoration that was more resistant to displacement than was conventionally placed amalgam retained with proximal grooves or dovetails. If this finding holds true in clinical trials, then the bonding of amalgam to tooth structure will permit more conservative cavity preparations. Results of this study are consistent with results of similar studies of fracture resistance with restorations (gold inlays, composite resins, and porcelain inlays) bonded to tooth structure. Nearly all these studies reported an increased fracture resistance with bonded restorations.lOX,B, 21 Typical cuspal fractures with conventional MOD
THE
JOURNAL
OF PROSTHETIC
DENTISTRY
Fig. 6. Fractured tooth that also fractured through a portion of the amalgam and left amalgam fragments still bonded to cavity walls. There is no sign on amalgam that testing head contacted amalgam along the fracture site.
restorations occur at the base of the cusp and angle to the external surface so that they usually are supragingival or slightly subgingiva14 In the laboratory, teeth with conventional or bonded MOD restorations frequently fracture with a cusp spliting vertically into the root. We have observed this same type of fracture clinically with teeth restored with bonded MOD composite resins. Because bonded restorations reduce cusp flexure, there should be less long-term structural fatigue of the tooth at the base of the cusps and therefore fewer cuspal fractures. However, a load that exceeds the cohesive or shear bond strength of the resin and fractures the tooth might fracture into the root rather than at the base of the cusp, which makes restoration of the tooth more difficult. It is our opinion that bonded intracoronal restorations should not be substituted for circumferential reinforcement of teeth when the operator
EAXLE,
judges that the teeth are weak and susceptible to fracture. Staninec and Holtz4 found that the bonded amalgam had a mean tensile bond strength of 1404 psi to etched enamel and of 469 psi to dentin. This evidence indicates that the bond of amalgam to tooth via a resin liner is strong. Our results showed that the restoration remained bonded to the tooth in some of the fractures that occurred through the amalgam. A study of microleakage with bonded amalgam indicates a better seal to the tooth than that produced with composite resinz4 In another study, the progression of artificial caries was inhibited along cavity walls in bonded amalgam when compared with that progression in conventional amalgam without the resin liner.25 Although the bonded amalgam technique looks very promising as a result of laboratory studies, the limitations of these studies must be recognized. The continually increasing load applied to the teeth in this study is not typical of the type of loading that occurs clinically. Except for a single traumatic incidence, teeth that crack or fracture in the mouth usually do so as a result of repeated episodes of stress that fatigue the crystalline structure and produce microcracks that propagate until failure of the structure occurs. The effects of masticatory stresses on the durability of the resin bond also were not determined. It is not known whether microleakage, recurrent caries, or retention of the amalgam is a problem if the resin fatigues and breaks down. Long-t,erm clinical trials are needed to assess these factors and to determine the longevity of bonded amalgam restorations and their usefulness in reducing the incidence of tooth fracture. The laboratory studies and recent clinical applications have, in the short-term, demonstrated that amalgam can be bonded to tooth structure and can thereby produce cuspal reinforcement.
SUMMARY
3. 4. 5. 6.
7. 8. 9.
10.
11. 12. 13.
14. 15. 16.
17.
18.
19. 20.
21.
23. 24
25
We gratefully acknowledge the help of Mr. Larry Watanabe and Dr. Eric Clark with the technical procedures and that of Ms. Vickie Leow wit,h graphic illustrations.
REFERENCES 1. Snyder, DE. The cracked-tooth Oral Surg 1976;41:698-704.
260
syndrome
and fractured
posterior
cusp.
AND
LACY
2. Cameron
22.
In this laboratory study, teeth restored with bonded amalgam were significantly more resistant to fracture than were teeth restored with conventional amalgam. The mode of failure of bonded amalgam restorations was mixed, but cohesive resin failure predominated.
STANIA-EC,
CE. The cracked tooth syndrome: additional findings. J Am Dent Assoc 1976;93:971-5. Cave1 WT, K&q WP: Blankenau RJ. An in viva study of cuspal fracture J PROSTHET DENT 1985;53:38-42. Eakle WS, Maxwell EH, Braly BV. Fractures of posterior adult teeth. J Am Dent Assoc 1986;112:215-8. Vale WA. Cavity preparation and further thoughts on high speed. Br Dent J 1959;107:333-6. Mondelli J, Steagall L, Ishikiriama A, Nawrro M, Soares F. Fracture strength of human teeth with cavity preparations. J PROSTHET DENT 1980;43:419-22. Larson TD, Douglas WH, Geistfeld RE. Effect of prepared cavities on the strength of teeth. Oper Dent 1981;6:2-5. Bell JG, Smith MC, dePont JJ. Cuspal failures of MOD restored teeth. Aust Dent J 1982;27:283-7. Blaser PK, Lund MR, Cochran MA, Potter RH. Effects of design of class II preparations on resistance of teeth to fracture, Oper Dent 1983;8:610. Share J, Mishell Y, Nathanson D. Effect of restorative mat,erial on fracture resistance of tooth structure in vitro [Abstract]. J Dent Res 1982;61:247. Simonsen RJ, Barouch E: Gelb M. Cusp fracture resistance from composite resin in class II restorations [Abstract]. J Dent Res 1983;62:254. Landy NA, Simonsen RJ. Cusp fracture strength in class II composite resin restorations [Abstract]. J Dent Res 1984:63:175. M&hell Y, Share J, Nathanson D. Fracture resistance of Cl. II amalgam vs. light activated composite restorations in vitro [Abstract]. J Dent Res 1984;63:293. Newman SM, Pisko-Dubienshy R. Effects of composite restorations on strength of posterior teeth [Abstract]. J Dent Res 1984;63:523. Eakle WS. Fracture resistance of teeth with class II bonded composite restorations. J Dent Res 1986;65:149-53. Phillips RW, Avery DR, Mehra RJ, Swarm ML, McCune RJ. Observat,ions on a composite resin for class II restorations: three year report. J PROSTHF,T DENT 1973;30:891-7. Leinfelder KF: Sluder TB, Santos JF, Wall JT. Five-year clinical evaluation of anterior and posterior restorations of composite resin. Oper Dent 1980;5:57-65. Jensen ME, Redford DA, Williams BT, Gardner F. Posterior etchedporcelain restorations: an in vitro study. Compend Contin Educ Dent 1987;8:615-22. Lutz F, Moermann W. In viva wear of posterior tooth composites [Abstract], J Dent Res 1982;61:215. Moffa JP, Jenkins WA, Hamilton JC. The longevity of composite resins for the restoration of posterior teeth [Abstract]. J Dent Res 1984;63:199. Eakle WS, Staninec M, Clark EJ. Effect of bonded inlays on fracture resistance of teeth [Abstract]. J Dent Res 1989;68:303. Lacy A, Rupprecht R, Watanabe L, Hiramatsu D. Amalgam-amalgam and amalgam-composite resin bond strengths [Abstract]. J Dent Res 1989;68:189. Staninec M. Retention of amalgam restorations: undercuts versus bonding. Quintessence Int 1989;20:347-51. Staninec M, Halt M. Bonding of amalgam to tooth structure: tensile adhesion and microleakage tests. J PROSTHET D~XT 1988:59:397402. Torii Y, Staninec M, Kawakami M, Imazato S, Torri M, Tsuchitani Y. Inhibition in vitro of caries around amalgam restorations by bonding amalgam to tooth structure. Oper Dent 1989;14:142-8.
Reprint requests to: DR. W. STEPHAN EAKLE DEPARTMENT OF RESTORATIVE SCHOOL OF DENTISTRY UNIVERSITY OF CALIFORNIA SAN FRANCISCO: CA 94143
DENTISTRY
AUGUST
1992
VOLUME
68
NUMBER
2
