HEISLER
2. Nemetz H, Tjan AHL. Reversible hydrocolloid: the standard of excellence. J Oral Health 1988;78:45-50. 3. Kornfield M. Month rehabilitation. 2nd ed. St. Louis: CV Mosby, 1974:943-6. 4. Appleby DS. The combination hydocolloid/alginate impression. J Am Dent Assoc 1983;106:194-5. 5. Lin C, Ziebert GJ, Donegan SJ, Dhuru WB. Accuracy of impression materials for complete arch fixed partial dentures. J PROSTHET DENT 1988;59:288-91.
6. Herring HW, Tames MA, Zardiackas LD. Accuracy of reversible/ irreverible hydrocolloid compared with other impression materials. J PRO~THET DENT 1984;52:275-9.
‘7. Johnson GA, Craig RG. Accuracy and bond strength of combination agar/alginate hydrocolloid impression materials. J PROSTHET DENT 1986;55:1-6.
3. Hellie CM, Charbeneau GT, Craig RG, Brandon HE. Quantitative evaluation of proximal tooth movement effected by wedging: a pilot study. J PROSTHET DENT 1985;53:335-41. 9. Tjan AHL, Li T. Effects of reheating on the accuracy of addition silicone putty-wash impressions. J PROSTHET DENT 1991:65:743-B.
atigue life of three conditions
core materials
AND
TJAN
10. Tjan AHL. Effect of contaminants on the adhesion of light-bodied silicones to putty silicones in putty-wash impression technique. J PROSTHET DENT 1988;59:562-7. 11. Fusayama T, Kurasaki N, Node I-I. A laminated hydrocolloid impression for indirect inlays. J PROSTHET DENT 1982;47:171-6. 12. Skinner EW, Hoblit NE. Study of the accuracy of hydrocolloid impressions. J PROSTHET DENT 1956;6:80-6. 13. Appleby DC, Pameijer CH, Boffa, J. The combined reversible hydrocollaid/irreversible hydrocolloid impression system. J PROSTHET DENT 1980;44:27-35. Reprint requests to: DR. WILLIAM H. HEISLER SCHOOL OF DENTISTRY LOMA LINDA UNIVERSITY LOMA LINDA, CA 92350
We thank Gingipac and Columbus Dental for supplying materials and equipment for this study.
under
simulated
chewing
obert E. Kovarik, DMD, MS,a Larry C. Breeding, DMD, MS,b and W. Franklin Caughman, DDS, MEdC University of Kentucky College of Dentistry, Lexington, Ky., and Medical College of Georgia, Augusta, Ga. There has been an increase in the use of prefabricated post systems to restore endodontically treated teeth. Various restorative materials are being used as core buildups on these posts. The purpose of this study was to compare three core materials that are used with prefabricated stainless steel posts. Two types of prefabricated posts were placed in extracted teeth, followed by core buildups in amalgam, composite resin, or glass ionomer. The teeth were prepared for full cast crowns with the margins of the crown preparation extending 0.5 to 1.0 mm below tbe margins of the core buildup. Crowns were fabricated and cemented with zinc phosphate cement. A custom-designed chewing machine was used to cyclically load the teeth with vertical and horizontal forces for one million cycles or until failure occurred. Results indicated highly significant differences in the survival of the post-core-crown restorations depending on which core buildup material was used. Amalgam cores had the lowest failure rate, followed by composite resin cores. All teeth restored with crowns over glass-ionomer core buildup failed. The type of prefabricated post used had no effect on the survival of the post-core-crown restorations regardless of the core buildup used. (J PROSTHET DENT 1992;68:584-90.)
I
n the past 30 years there has been a dramatic increase in both the number of endodontic procedures being performed and the effectiveness and predictability of these procedures. Dentists are now faced with restoring
aAssistant Professor, Department of Oral Health Practice, University
of Kentucky
College of Dentistry.
bAssociate Professor, Department of Oral Health Practice, University
of Kentucky
College of Dentistry. of Restorative ical College of Georgia. 10/l/39549
CAssociateProfessor, Department
584
Dentistry,
Med-
more endodontically treated teeth than ever before.la 2 The restoration of these teeth offers many challenges for the restorative dentist because of the large percentage of failures. This high incidence of failures has led to the development of a multitude of restorative alternatives for endodontically treated teeth. The restorative dentist may choose from cast post and cores, coronal-radicular core buildups, pin-retained core buildups, and prefabricated posts. Each can be used with a variety of post materials and core materials. In some situations radicular support for the restoration may not be necessary.3 The use of prefabricated posts with amalgam, composite OCTOBER
1992
VOLUME
68
NUMBER
4
FATIGUE
LIFE
OF CORE
MATERIALS
resin, or glass-ionomer core build-ups has become widespread.lp 4 Reduced chair time and easeof manipulation of prefabricated posts, as compared with cast post and cores, make the procedure appealing to private practitioners. This usually results in reduced cost to the patient. Most of the research on prefabricated systems focuses on post retention and stress distribution from the post to the surrounding root. These studies have looked at the design of prefabricated posts with respect to size, shape, and surface characteristics.5, 6Prefabricated parallel posts of equal length and diameter have been shown to be more retentive than cast post and cores7 This increased retention has been attributed to its serrated surface and parallel design. It is generally concluded that a threaded or serrated post with parallel sides offers the greatest retention while minimizing stress concentrations around the post. Amalgam, composite resin, and reinforced glass-ionomer (glass ionomer containing metal alloy powders) restorative materials have been suggested as core materials to be placed in conjunction with prefabricated postssr2 Standlee et al6 showed that amalgam and composite resin cores adapt well to all kinds of tooth-retained pins, and Kao et a1.13showed that glass ionomers can actually be strengthened by the incorporation of pins. Core buildups in amalgam offer high compressive strength and ease of manipulation. Fast-setting amalgam alloys allow crown preparation within 20 to 30 minutes after condensationi Composite resins have also gained acceptance as a choice of core materials.15sl6 They offer an advantage over amalgam in that they can be prepared immediately and impressed for restoration with a full casting. With dentin bonding agents, composite resins also offer the advantage of being adherent to tooth structure. The disadvantages of composite resins are problems with microleakage and dimensional stability, which may affect marginal adaptation of castings. I7 When a composite resin absorbs moisture, it expands significantly, even to the extent of affecting the ability to seat a crown. It has been suggested that when composite resin cores are used, some time be allowed for the resin to come to equilibrium with the moist oral environment before a final impression is made. The most recent addition to the materials used as core buildups are the reinforced glass-ionomer materials. The advantages of these materials are color difference with tooth structure, fluoride release to adjacent tooth structure, chemical bonding to tooth structure, and rapid setting.8 However, all glass ionomers including reinforced materials are inherently weak and should generally not be used for highstress-bearing applications. l8 A recent review of the literature by Engelmanlg concluded that composite resin and glass-ionomer materials should be used cautiously as core buildup materials. The purpose of this study was to investigate the fatigue life and modes of failure of three core materials with two preformed post systems by means of a custom-designed chewing machine that applied subcritical loads over many cycles. THE
JOURNAL
OF PROSTHETIC
DENTISTRY
The use of the fatigue-testing apparatus as a means of stressing samples offers some advantages. The majority of previous studies used tensile testing to measure a failure limit of post and cores. Clinically the amount of tensile stresses on the post are minimal.1,20 Other studies have attempted to mimic the shear and compressive forces that are found clinically but have done so by applying one catastrophic blow to failure. l, 21This mode of failure would occur only for traumatic injuries. Subcritical loads applied to the core over many cycles more closely model the actual service requirements of a post-core-crown restoration.
MATERIAL
AND METHODS
Extracted human canines, which were unrestored and noncarious and had been stored in a solution of equal parts glycerin, hydrogen peroxide, and water, were used for this study. The dimensions of the teeth were measured mesiodistally and faciolingually at the cementoenamel junction (CEJ), as well as for root length from the CEJ. Only teeth that fell in the size range 6 to 7 mm mesiodistally, 7.5 to 8.5 mm faciolingually, and 18 to 21 mm in root length were chosen. The teeth were kept in a moist environment throughout the experiment except during placement of post and cores. Three types of core materials and two types of prefabricated posts were chosen. The core materials consisted of a fast-setting dental amalgam (Tytin, Kerr Division of Sybron, Romulus, Mich.), a posterior composite resin (Adaptic II, Johnson & Johnson, East Windsor, N.J.), and a reinforced glass ionomer (Ketac Silver, ESPEPremier, Norristown, Pa.). Two types of prefabricated stainless steel posts (post A, Flexipost, Essential Dental Systems, New York, N.Y.; post B, Para-Post, Whaledent International, New York, N.Y.) were chosen as the prefabricated posts on which the cores were to be built. These two posts were chosen because they vary greatly in the design of the head to which the cores are attached. At least five samples of each core material were placed over each of the two designs of prefabricated posts.
Post insertion The crowns of the teeth were removed 1 mm coronally to the CEJ with a No. 56 bur in a high-speed handpiece with copious water spray. The surface of the roots was lightly notched and mounted in tray acrylic resin to within 2 to 3 mm of the CEJ. The mounted teeth were then randomly divided into two groups to receive either post A or post B. The post spaces were prepared according to the manufacturer’s instructions. The posts were first trial inserted without cement and then cemented with zinc phosphate cement. Post A (Flexi-post, size No. 2) was embedded 10.5 mm, and the head was reduced to 4 mm in height to allow the core materials to completely cover the post. Post B (Para-Post, size No. 6) was reduced in length by 4.5 mm from the apical end and embedded 10.5 mm into the root of the tooth, which left 4 mm of post exposed. The groups were then randomly subdivided into three groups to receive amalgam, composite-resin, or glass-ionomer cores. 585
KOVARIK,
BREEDING,
AND CAUGHMAN
Vertical load Transducer tip,
J
1
Fig.
1. Diagram
of working portion
Median fatigue life (cycles)
Failures (%)
A B
73.3 68.4
377,841 392,583
Initial
flexure km)
5.70 -f- 0.77 5.86 i_ 0.66
XI. Comparison of failure rates, median fatigue life, and amount of initial flexure of post-core-crown restoration for three core materials
Reinforced
Table
Failures (%)
Core type Amalgam Composite resin Glass ionomer
Amalgam
Median fatigue life (cycles)
33.3 83.3 100
> 1,000,000 385,212 120,631
Initial
flexure bm)
2.52 t 0.27 8.71 rt 0.46 6.22 t 0.22
cores
After excess zinc phosphate cement was removed from around the post, a matrix band was placed and fast-setting Tytin amalgam was condensed around the post head to completely fill the band. These samples were returned to a moist environment and allowed to set for 24 hours.
Composite-resin
cores
After excess cement was removed, the enamel and dentin were etched with 37% phosphoric acid and rinsed copiously with water. Scotchbond II bonding agent (Johnson
586
apparatus.
& Johnson, East Windsor, N.J.) was applied according to the manufacturer’s instructions. A matrix band was placed and Adaptic II composite resin was inserted into the band in three increments, light curing for 30 seconds after each increment. The samples were returned to a moist environment for 24 hours.
Table I. Comparison of failure rates, median fatigue life, and amount of initial flexure of post-core-crown restoration for two types of posts Post type
of fatigue-testing
$ i *
glass-ionomer
cores
After excess cement was removed, the floor of the preparation, the dentin, and the enamel were treated with Duralon liquid for 8 seconds to partially remove the smear layer and rinsed copiously with water. A matrix band was placed, and Ketac silver was injected to completely fill the band. The surface of the glass ionomer was coated with varnish according to the manufacturer’s instructions. The samples were returned to a moist environment and allowed to set for 24 hours.
Crown preparation Twenty-four hours after placement of the appropriate core, the teeth were oriented along the long axis by means of a surveyor. A high-speed handpiece was mounted on the surveyor with a custom-designed attachment to the surveyor. Full cast gold crown preparations were completed with 5 mm of axial wall height, six degrees of taper, and a flat occlusal surface. The margins of the preparation were placed an average of 0.75 mm gingival to the margins of the core buildup and varied randomly from 0.5 to 1.0 mm for all preparations. A 40-degree bevel was placed at the axioocclusal walls of the preparation, leaving 4 mm of actual axial wall height. The tooth preparation was accomplished by means of medium grit diamond burs with a B-degree taper and copious water spray.
OCTOBER
1992
VOLUME
68
NUMBER
4
FATIGUE
LIFE
OF CORE
MATERIALS
Survival -
Post A (all cores)
Curve -----
Post
B
(all cores)
Millions
of Cycles
Fig. 2. Survival curves for post A and post B show percentage of surviving post-corecrown restorations as a function of the number of cycles of applied load on fatigue-testing apparatus.
Table III, materials Core
Number and types of failure of post-core-crown restorations with the use of three different core build-up
material
Failure did not occur
Amalgam Compositeresin Glassionomer
Core fracture
8
0
1
2
0
4
0
8
0
Crown fabrication Rubber Sep material (George Taub Products and Fusion Co., Inc., Jersey City, N.J.) was applied to the tooth as a die spacer to within 1 mm of the margins and allowed to dry. Die Lube (Slaycris Products, Portland, Ore.) was applied and a wax pattern fabricated by means of a dip technique to obtain an even axial wall thickness of the patterns of 0.5 mm and an overall height of 8 mm. The occlusal surface of the pattern was fabricated with a slope of 40 degrees so as to contact the chewing machine on its buccal and lingual inclines. Contact along these surfaces in the chewing applied a combination of shear and compressive forces to the tooth. The wax patterns were sprued, invested, and cast in Albabond E (Heraeus, Dental Products Division, Queens Village, N.Y.). After each casting was successfully seated on the teeth, the castings and the teeth were cleaned and dried. The castings were then luted to the teeth with zinc phosphate cement (Flecks Mizzy, Mizzy Inc., Cherry Hill, N.J.) mixed according to manufacturers specifications. A 4 kg load of uniform seating pressure was applied for 8 minutes to all cast restorations. The teeth with cemented
THE
JOURNAL
OF PROSTHETIC
DENTISTRY
Post fracture
Tooth fracture
Cement failure
2 1 2
1 5 0
crowns were then returned to a moist environment for 24 hours before testing. The fatigue-testing machine used in this study has been described previously. 22The working portion of this device is illustrated in Fig. 1. The force bar, which applies the simulated chewing force on the tooth, is motor driven and moves horizontally in a reciprocating action. A vertical load was applied to the force bar by a compressed spring above the restoration. The horizontal motion of the force bar alternately contacted the buccal and lingual inclines of the tooth, yielding a combination of compressive and shear forces that mimics the forces of mastication. For this study a vertical load of 75 pounds was placed on the force bar. This was considered to be on the high end of the range of typical forces that a patient could generate during parafunctional activity.23 A complex arrangement of strain gauges was used to detect movement of the cast crown in relation to the tooth itself. The transducer consisted of a strip of beryllium-copper alloy formed into a box with the ends welded to form the sensing tip. Four contact strain gauges were attached to sides A and B. The
587
KOVARIK,
Survival -
Amalgam
BREEDING,
AND
CAUGHMAN
Curve
-----Composite
--.-
Glass
lonomer
100
0 0.0
0.5 Millions
1 .o
of Cycles
Fig. 3. Survival curves for amalgam, composite resin, and glass ionomer core buildups show percentage of surviving post-core-crown restorations as a function of the number of cycles of applied load on fatigue-testing apparatus. transducer -measured both horizontal and vertical components of motion of the crown with respect to the tooth with an accuracy of ~1 pm. The vertical and horizontal displacement was recorded continuously on a strip chart. A digital counter in the control box kept a record of the total number of cycles applied to each specimen. The machine speed was 60 cycles/min. A horizontal movement of 30 pm or more was arbitrarily chosen as the failure point of a casting. With this much motion the cement seal of the restoration would be broken. The horizontal movement (flexure of the post-core-crown restoration) during applied masticatory forces was measured, initially and throughout the test. One million cycles was chosen as the cutoff point of the test if the 30 pm end point had not been reached. One million cycles on this machine had previously been estimated to be equivalent to 5 years of heavy wear.22 After testing, all teeth were sectioned and examined to determine the cause of failure.
RESULTS Table I compares the two types of posts, regardless of the core material used, with regard to total percentage of failures, median fatigue life, and amount of initial flexure of the crown in relation to the tooth when cyclically loaded. The survival curves for the two types of posts are shown in Fig. 2. When the two types of posts were compared, there was no significant difference in the fatigue life of the postcore-crown restorations as tested by a Wilcoxon-MannWhitney test (p 0.765). Likewise there was no significant difference in the amount of flexure of the restoration as tested by a t test (p 0.8777). Table II compares the three core materials regardless of
588
post type, which was determined to have no effect, for total percentage of failures, median fatigue life, and initial flexure of the crown. The survival curves for the three core materials are shown in Fig. 3. When the three core types were compared, there was a significant difference in the fatigue life of the post-core-crown restorations as tested by Kruskal-Wallis analysis of variance by ranks (JJ < 0.0001). A multiple-range test indicates that the glass ionomer cores had a significantly lower fatigue life than the amalgam or composite-resin cores at a 95 % confidence level. There was a significant difference in the means of the amount of flexure of the crowns in relation to the tooth for the different core materials as tested by analysis of variance (p < 0.0001). Tukey’s HSD test (99 % confidence level) indicates that the amalgam cores permitted significantly less Aexure than glass ionomer cores, which permitted significantly less flexure than composite-resin cores. Sectioning of untested specimens showed that ail three core materials were well adapted to both types of posts before testing. Figs. 4 and 5 show representative samples and failure modes. Table III shows the modes of failure that were determined for each post-core-crown restoration tested.
DISCUSSION The intent of this study was to compare the durability of three core materials. Harsh clinical conditions were imposed on the materials. Specifically the margins of the crowns were placed just below the margins of the core (30 pm with respect to the tooth. In this sample the core material remained well adapted to the post. However, cement breakdown occurred at the post-root interface and at the crown-tooth interface. B, Failure of a post-core-crown restoration with a composite-resin core by post fracture.
occlusal forces were greater than normal chewing forces but well within the range of forces that occur during parafunctional activity. Results of other studies have shown that when adequate tooth structure is present-at least a 2 mm ferrule below the margin of the core-any core material is acceptable. However, there are times clinically when remaining tooth structure is so scarce that the margins of the crown must be placed at or just below the core. It is under these conditions that the choice of core material may be important. All of the teeth restored with glass ionomer core buildups failed under the conditions of this study. In each case the core material showed fracture around the post head. This material simply did not have adequate strength to withstand occlusal loading when the majority of the forces
were borne by the core material (Fig. 4). The composite resin core buildups had adequate strength (no core fractures were observed); however, 83 % eventually failed. Fifty percent of the failures occurred as a result of failure of the zinc phosphate cement at the crown-core interface, the post-root interface, or both. Forty percent of the failures occurred as a result of fracture of the post (Fig. 5, B). It is interesting to note, however, that these two modes of failure were not seen frequently for amalgam cores, which were cycled significantly longer. The high number of cement failures and post fractures for composite resin cores is apparently related to the material’s low modulus of elasticit,y. Composite resin core build-ups were much more flexible (Table II) than the other core materials tested. This extra flexure of the core placed significantly more shearing
THE
JOURNAL
OF PROSTHETIC
DENTISTRY
KOVARIK,
stresses on the cement interface and the posts themselves. The glass ionomer cores, although less flexible than the composite resin cores, also allowed significantly more flexure than the amalgam cores. It would be expected that had the glass ionomer cores themselves not broken down so quickly, a high number of cement failures and post fractures may have also occurred. Five of 34 teeth fractured while being tested in this study. The propensity to fracture would be expected to be higher for this in vitro system than clinically. The teeth were mounted in a rigid block without a periodontal ligament to more evenly disperse stresses in the root. The roots were notched to increase retention in the mounting block, and these notches acted as stress raisers in the root. However, it is interesting to note that four of five tooth fractures occurred on teeth restored with post A. Under these test conditions, this implies that post A produced more internal stress on the root than post B. However, the clinical significance of this is questionable. Five of 34 teeth tested exhibited failure as a result of fracture of the post. All of these post fractures occurred at or just below the junction of the post with the core material, and all were with post B. The tendency of post B to fracture may be due to its serrated design, which acts as a stress raiser (defect) on the surface of the post. This could create greatly increased localized stress on the post and initiate fracture. The results of this study do not support the use of glass ionomer materials as core buildup materials when there is very little remaining natural tooth structure. Although statistical significance was not shown at the p < 0.05 level, the results of this study do not support the use of composite resin core buildups when very little tooth structure remains. CONCLUSIONS 1. Teeth restored with amalgam cores and prefabricated posts had a significantly lower failure rate than teeth restored with composite resin or reinforced glass ionomer cores and prefabricated posts. 2. Teeth restored with amalgam cores provided significantly more rigid abutments for crowns and fixed partial denture retainers than did glass ionomer or composite resin cores. 3. Under the conditions used in this study, reinforced glass ionomer cores did not have adequate strength to withstand simulated occlusal forces,
590
BREEDING,
AND
CAUGHMAN
REFERENCES 1.
2. 3.
I.
8. 9. 10. 11.
12.
13. 14.
15. 16.
17. 18. 19. 20. 21. 22.
23.
Plasmans PJJM, Visseren LGH, Vrijhoef MMA, Kayser AF. In vitro comparison of dowel and core techniques for endodontically treated molars. J Endodontol 1986;12:382-7. Hudis SI, Goldstein GR. Restoration of endodontically treated teeth: a review oftheliterature. J PROSTHET DENT 1986$5:33-S. Sorensen JA, Nayyar A, Nicholls JI, Nathanson D, Caputo A, Standlee J. Current clinical trends in restoring the endodontically treated tooth: highlights of a symposium. J Clin Dent 1988;2:39-47. Chapman KW, Worley JL, Von Fraunhofer JA. Retention of prefabricated posts by cements and resins. J PROSTHET DENT 1985;54:649-52. Ruemping DR, Lund MR: Schnell RJ. Retention of dowels subjected to tensile and torsional forces. J PROSTHET DENT 1979;41:159-62. Standlee JP, Caputo AA, Hanson EC. Retention of endodontic dowels: effect of cement, dowel length, diameter and design. J PROSTHET DENT 1978;39:401-5. Tjan AH, Whang SB. Retentive properties of some simplified dowelcore systems to cast gold dowel and core. J PROSTHET DENT 1983; 50:203-6. Taleghani M, Morgan RW. Reconstructive materials for endodontically treated teeth. J PROSTHET DENT 1987;57:446-9. Linde LA. The use of composites as core material in root filled teeth. Swed Dent J 1983;7:92-9. Hoag EP, Dwyer TG. A comparative evaluation of three post and core techniques.J PROSTHET DENT 1982;47:177-81. Marshak BL, Shemen BB, Cardash HS. Use of a special matrix system for constructing amalgam and composite cores. J PROSTHET DENT 1981;57:21-2. Brandal JL, Nicholls JI, Harrington GW. A comparison of three restorative techniques for endodontically treated anterior teeth. J PROSTHET DENT 1987;58:161-5. Kao EC, Hart S, Johnston WM. Fracture resistance of four core materials with incorporated pins. Int J Prosthodontol 1989;2:569-78. Nayyar A. Coronal-radicular buildup for endodontically treated teeth. In: Clark JW, ed. Clinical dentistry. ~014. Philadelphia: Harper&Row, 1983:1-28. Ghan KC, Fuller J, Khowassah M. The adaptation of new amalgam and composite resin to pins. J PROSTHET D~~~1978;38:392-5. Paige H, Hirsch SM, Gelb MN. Effects of temporary cements on crown-to-composite resin core bond strength. J PROSTHET DENT 1986;55:49-52. Oliva RA, Lowe JA. Dimensional stability of composite used as a core material. J PROSTHET DENT 1986;56:554-61. Wilson AD, McLean JW. Glass ionomer cements. Chicago: Quintessence Publishing, 1988. Engelman MJ. Core materials. J Calif Dent Assoc 1988;16:41-5. Caldwell RC. Adhesion of foods to teeth. J Dent Res 1962;41:821-32. Trabert KC, Caputo AA, Abou-Rass M. Tooth fracture-a comparison of endodontic and restorative treatments. J Endodontol 1978;4:341-5. Outhwaite WC, Twiggs SW, Fairhurst CW, King GE. Slots vs pins: a comparison of retention under simulated chewing stresses. 3 Dent Res 1982;61:400-2. Okeson JP. Management of temporomandibular disorders and occlusion. 2nd ed. St. Louis: CV Mosby, 1989.
Reprint requests to: DR.ROBERT E.KOVARIK DEPARTMENTOF ORALHEALTHPRACTICE UNIVERSITY OF KENTUCKY LEXINGTON,KY~O~~~-OO~~
OCTOBER
1992
VOLUME
68
NUMBER
4
2. Nemetz H, Tjan AHL. Reversible hydrocolloid: the standard of excellence. J Oral Health 1988;78:45-50. 3. Kornfield M. Month rehabilitation. 2nd ed. St. Louis: CV Mosby, 1974:943-6. 4. Appleby DS. The combination hydocolloid/alginate impression. J Am Dent Assoc 1983;106:194-5. 5. Lin C, Ziebert GJ, Donegan SJ, Dhuru WB. Accuracy of impression materials for complete arch fixed partial dentures. J PROSTHET DENT 1988;59:288-91.
6. Herring HW, Tames MA, Zardiackas LD. Accuracy of reversible/ irreverible hydrocolloid compared with other impression materials. J PRO~THET DENT 1984;52:275-9.
‘7. Johnson GA, Craig RG. Accuracy and bond strength of combination agar/alginate hydrocolloid impression materials. J PROSTHET DENT 1986;55:1-6.
3. Hellie CM, Charbeneau GT, Craig RG, Brandon HE. Quantitative evaluation of proximal tooth movement effected by wedging: a pilot study. J PROSTHET DENT 1985;53:335-41. 9. Tjan AHL, Li T. Effects of reheating on the accuracy of addition silicone putty-wash impressions. J PROSTHET DENT 1991:65:743-B.
atigue life of three conditions
core materials
AND
TJAN
10. Tjan AHL. Effect of contaminants on the adhesion of light-bodied silicones to putty silicones in putty-wash impression technique. J PROSTHET DENT 1988;59:562-7. 11. Fusayama T, Kurasaki N, Node I-I. A laminated hydrocolloid impression for indirect inlays. J PROSTHET DENT 1982;47:171-6. 12. Skinner EW, Hoblit NE. Study of the accuracy of hydrocolloid impressions. J PROSTHET DENT 1956;6:80-6. 13. Appleby DC, Pameijer CH, Boffa, J. The combined reversible hydrocollaid/irreversible hydrocolloid impression system. J PROSTHET DENT 1980;44:27-35. Reprint requests to: DR. WILLIAM H. HEISLER SCHOOL OF DENTISTRY LOMA LINDA UNIVERSITY LOMA LINDA, CA 92350
We thank Gingipac and Columbus Dental for supplying materials and equipment for this study.
under
simulated
chewing
obert E. Kovarik, DMD, MS,a Larry C. Breeding, DMD, MS,b and W. Franklin Caughman, DDS, MEdC University of Kentucky College of Dentistry, Lexington, Ky., and Medical College of Georgia, Augusta, Ga. There has been an increase in the use of prefabricated post systems to restore endodontically treated teeth. Various restorative materials are being used as core buildups on these posts. The purpose of this study was to compare three core materials that are used with prefabricated stainless steel posts. Two types of prefabricated posts were placed in extracted teeth, followed by core buildups in amalgam, composite resin, or glass ionomer. The teeth were prepared for full cast crowns with the margins of the crown preparation extending 0.5 to 1.0 mm below tbe margins of the core buildup. Crowns were fabricated and cemented with zinc phosphate cement. A custom-designed chewing machine was used to cyclically load the teeth with vertical and horizontal forces for one million cycles or until failure occurred. Results indicated highly significant differences in the survival of the post-core-crown restorations depending on which core buildup material was used. Amalgam cores had the lowest failure rate, followed by composite resin cores. All teeth restored with crowns over glass-ionomer core buildup failed. The type of prefabricated post used had no effect on the survival of the post-core-crown restorations regardless of the core buildup used. (J PROSTHET DENT 1992;68:584-90.)
I
n the past 30 years there has been a dramatic increase in both the number of endodontic procedures being performed and the effectiveness and predictability of these procedures. Dentists are now faced with restoring
aAssistant Professor, Department of Oral Health Practice, University
of Kentucky
College of Dentistry.
bAssociate Professor, Department of Oral Health Practice, University
of Kentucky
College of Dentistry. of Restorative ical College of Georgia. 10/l/39549
CAssociateProfessor, Department
584
Dentistry,
Med-
more endodontically treated teeth than ever before.la 2 The restoration of these teeth offers many challenges for the restorative dentist because of the large percentage of failures. This high incidence of failures has led to the development of a multitude of restorative alternatives for endodontically treated teeth. The restorative dentist may choose from cast post and cores, coronal-radicular core buildups, pin-retained core buildups, and prefabricated posts. Each can be used with a variety of post materials and core materials. In some situations radicular support for the restoration may not be necessary.3 The use of prefabricated posts with amalgam, composite OCTOBER
1992
VOLUME
68
NUMBER
4
FATIGUE
LIFE
OF CORE
MATERIALS
resin, or glass-ionomer core build-ups has become widespread.lp 4 Reduced chair time and easeof manipulation of prefabricated posts, as compared with cast post and cores, make the procedure appealing to private practitioners. This usually results in reduced cost to the patient. Most of the research on prefabricated systems focuses on post retention and stress distribution from the post to the surrounding root. These studies have looked at the design of prefabricated posts with respect to size, shape, and surface characteristics.5, 6Prefabricated parallel posts of equal length and diameter have been shown to be more retentive than cast post and cores7 This increased retention has been attributed to its serrated surface and parallel design. It is generally concluded that a threaded or serrated post with parallel sides offers the greatest retention while minimizing stress concentrations around the post. Amalgam, composite resin, and reinforced glass-ionomer (glass ionomer containing metal alloy powders) restorative materials have been suggested as core materials to be placed in conjunction with prefabricated postssr2 Standlee et al6 showed that amalgam and composite resin cores adapt well to all kinds of tooth-retained pins, and Kao et a1.13showed that glass ionomers can actually be strengthened by the incorporation of pins. Core buildups in amalgam offer high compressive strength and ease of manipulation. Fast-setting amalgam alloys allow crown preparation within 20 to 30 minutes after condensationi Composite resins have also gained acceptance as a choice of core materials.15sl6 They offer an advantage over amalgam in that they can be prepared immediately and impressed for restoration with a full casting. With dentin bonding agents, composite resins also offer the advantage of being adherent to tooth structure. The disadvantages of composite resins are problems with microleakage and dimensional stability, which may affect marginal adaptation of castings. I7 When a composite resin absorbs moisture, it expands significantly, even to the extent of affecting the ability to seat a crown. It has been suggested that when composite resin cores are used, some time be allowed for the resin to come to equilibrium with the moist oral environment before a final impression is made. The most recent addition to the materials used as core buildups are the reinforced glass-ionomer materials. The advantages of these materials are color difference with tooth structure, fluoride release to adjacent tooth structure, chemical bonding to tooth structure, and rapid setting.8 However, all glass ionomers including reinforced materials are inherently weak and should generally not be used for highstress-bearing applications. l8 A recent review of the literature by Engelmanlg concluded that composite resin and glass-ionomer materials should be used cautiously as core buildup materials. The purpose of this study was to investigate the fatigue life and modes of failure of three core materials with two preformed post systems by means of a custom-designed chewing machine that applied subcritical loads over many cycles. THE
JOURNAL
OF PROSTHETIC
DENTISTRY
The use of the fatigue-testing apparatus as a means of stressing samples offers some advantages. The majority of previous studies used tensile testing to measure a failure limit of post and cores. Clinically the amount of tensile stresses on the post are minimal.1,20 Other studies have attempted to mimic the shear and compressive forces that are found clinically but have done so by applying one catastrophic blow to failure. l, 21This mode of failure would occur only for traumatic injuries. Subcritical loads applied to the core over many cycles more closely model the actual service requirements of a post-core-crown restoration.
MATERIAL
AND METHODS
Extracted human canines, which were unrestored and noncarious and had been stored in a solution of equal parts glycerin, hydrogen peroxide, and water, were used for this study. The dimensions of the teeth were measured mesiodistally and faciolingually at the cementoenamel junction (CEJ), as well as for root length from the CEJ. Only teeth that fell in the size range 6 to 7 mm mesiodistally, 7.5 to 8.5 mm faciolingually, and 18 to 21 mm in root length were chosen. The teeth were kept in a moist environment throughout the experiment except during placement of post and cores. Three types of core materials and two types of prefabricated posts were chosen. The core materials consisted of a fast-setting dental amalgam (Tytin, Kerr Division of Sybron, Romulus, Mich.), a posterior composite resin (Adaptic II, Johnson & Johnson, East Windsor, N.J.), and a reinforced glass ionomer (Ketac Silver, ESPEPremier, Norristown, Pa.). Two types of prefabricated stainless steel posts (post A, Flexipost, Essential Dental Systems, New York, N.Y.; post B, Para-Post, Whaledent International, New York, N.Y.) were chosen as the prefabricated posts on which the cores were to be built. These two posts were chosen because they vary greatly in the design of the head to which the cores are attached. At least five samples of each core material were placed over each of the two designs of prefabricated posts.
Post insertion The crowns of the teeth were removed 1 mm coronally to the CEJ with a No. 56 bur in a high-speed handpiece with copious water spray. The surface of the roots was lightly notched and mounted in tray acrylic resin to within 2 to 3 mm of the CEJ. The mounted teeth were then randomly divided into two groups to receive either post A or post B. The post spaces were prepared according to the manufacturer’s instructions. The posts were first trial inserted without cement and then cemented with zinc phosphate cement. Post A (Flexi-post, size No. 2) was embedded 10.5 mm, and the head was reduced to 4 mm in height to allow the core materials to completely cover the post. Post B (Para-Post, size No. 6) was reduced in length by 4.5 mm from the apical end and embedded 10.5 mm into the root of the tooth, which left 4 mm of post exposed. The groups were then randomly subdivided into three groups to receive amalgam, composite-resin, or glass-ionomer cores. 585
KOVARIK,
BREEDING,
AND CAUGHMAN
Vertical load Transducer tip,
J
1
Fig.
1. Diagram
of working portion
Median fatigue life (cycles)
Failures (%)
A B
73.3 68.4
377,841 392,583
Initial
flexure km)
5.70 -f- 0.77 5.86 i_ 0.66
XI. Comparison of failure rates, median fatigue life, and amount of initial flexure of post-core-crown restoration for three core materials
Reinforced
Table
Failures (%)
Core type Amalgam Composite resin Glass ionomer
Amalgam
Median fatigue life (cycles)
33.3 83.3 100
> 1,000,000 385,212 120,631
Initial
flexure bm)
2.52 t 0.27 8.71 rt 0.46 6.22 t 0.22
cores
After excess zinc phosphate cement was removed from around the post, a matrix band was placed and fast-setting Tytin amalgam was condensed around the post head to completely fill the band. These samples were returned to a moist environment and allowed to set for 24 hours.
Composite-resin
cores
After excess cement was removed, the enamel and dentin were etched with 37% phosphoric acid and rinsed copiously with water. Scotchbond II bonding agent (Johnson
586
apparatus.
& Johnson, East Windsor, N.J.) was applied according to the manufacturer’s instructions. A matrix band was placed and Adaptic II composite resin was inserted into the band in three increments, light curing for 30 seconds after each increment. The samples were returned to a moist environment for 24 hours.
Table I. Comparison of failure rates, median fatigue life, and amount of initial flexure of post-core-crown restoration for two types of posts Post type
of fatigue-testing
$ i *
glass-ionomer
cores
After excess cement was removed, the floor of the preparation, the dentin, and the enamel were treated with Duralon liquid for 8 seconds to partially remove the smear layer and rinsed copiously with water. A matrix band was placed, and Ketac silver was injected to completely fill the band. The surface of the glass ionomer was coated with varnish according to the manufacturer’s instructions. The samples were returned to a moist environment and allowed to set for 24 hours.
Crown preparation Twenty-four hours after placement of the appropriate core, the teeth were oriented along the long axis by means of a surveyor. A high-speed handpiece was mounted on the surveyor with a custom-designed attachment to the surveyor. Full cast gold crown preparations were completed with 5 mm of axial wall height, six degrees of taper, and a flat occlusal surface. The margins of the preparation were placed an average of 0.75 mm gingival to the margins of the core buildup and varied randomly from 0.5 to 1.0 mm for all preparations. A 40-degree bevel was placed at the axioocclusal walls of the preparation, leaving 4 mm of actual axial wall height. The tooth preparation was accomplished by means of medium grit diamond burs with a B-degree taper and copious water spray.
OCTOBER
1992
VOLUME
68
NUMBER
4
FATIGUE
LIFE
OF CORE
MATERIALS
Survival -
Post A (all cores)
Curve -----
Post
B
(all cores)
Millions
of Cycles
Fig. 2. Survival curves for post A and post B show percentage of surviving post-corecrown restorations as a function of the number of cycles of applied load on fatigue-testing apparatus.
Table III, materials Core
Number and types of failure of post-core-crown restorations with the use of three different core build-up
material
Failure did not occur
Amalgam Compositeresin Glassionomer
Core fracture
8
0
1
2
0
4
0
8
0
Crown fabrication Rubber Sep material (George Taub Products and Fusion Co., Inc., Jersey City, N.J.) was applied to the tooth as a die spacer to within 1 mm of the margins and allowed to dry. Die Lube (Slaycris Products, Portland, Ore.) was applied and a wax pattern fabricated by means of a dip technique to obtain an even axial wall thickness of the patterns of 0.5 mm and an overall height of 8 mm. The occlusal surface of the pattern was fabricated with a slope of 40 degrees so as to contact the chewing machine on its buccal and lingual inclines. Contact along these surfaces in the chewing applied a combination of shear and compressive forces to the tooth. The wax patterns were sprued, invested, and cast in Albabond E (Heraeus, Dental Products Division, Queens Village, N.Y.). After each casting was successfully seated on the teeth, the castings and the teeth were cleaned and dried. The castings were then luted to the teeth with zinc phosphate cement (Flecks Mizzy, Mizzy Inc., Cherry Hill, N.J.) mixed according to manufacturers specifications. A 4 kg load of uniform seating pressure was applied for 8 minutes to all cast restorations. The teeth with cemented
THE
JOURNAL
OF PROSTHETIC
DENTISTRY
Post fracture
Tooth fracture
Cement failure
2 1 2
1 5 0
crowns were then returned to a moist environment for 24 hours before testing. The fatigue-testing machine used in this study has been described previously. 22The working portion of this device is illustrated in Fig. 1. The force bar, which applies the simulated chewing force on the tooth, is motor driven and moves horizontally in a reciprocating action. A vertical load was applied to the force bar by a compressed spring above the restoration. The horizontal motion of the force bar alternately contacted the buccal and lingual inclines of the tooth, yielding a combination of compressive and shear forces that mimics the forces of mastication. For this study a vertical load of 75 pounds was placed on the force bar. This was considered to be on the high end of the range of typical forces that a patient could generate during parafunctional activity.23 A complex arrangement of strain gauges was used to detect movement of the cast crown in relation to the tooth itself. The transducer consisted of a strip of beryllium-copper alloy formed into a box with the ends welded to form the sensing tip. Four contact strain gauges were attached to sides A and B. The
587
KOVARIK,
Survival -
Amalgam
BREEDING,
AND
CAUGHMAN
Curve
-----Composite
--.-
Glass
lonomer
100
0 0.0
0.5 Millions
1 .o
of Cycles
Fig. 3. Survival curves for amalgam, composite resin, and glass ionomer core buildups show percentage of surviving post-core-crown restorations as a function of the number of cycles of applied load on fatigue-testing apparatus. transducer -measured both horizontal and vertical components of motion of the crown with respect to the tooth with an accuracy of ~1 pm. The vertical and horizontal displacement was recorded continuously on a strip chart. A digital counter in the control box kept a record of the total number of cycles applied to each specimen. The machine speed was 60 cycles/min. A horizontal movement of 30 pm or more was arbitrarily chosen as the failure point of a casting. With this much motion the cement seal of the restoration would be broken. The horizontal movement (flexure of the post-core-crown restoration) during applied masticatory forces was measured, initially and throughout the test. One million cycles was chosen as the cutoff point of the test if the 30 pm end point had not been reached. One million cycles on this machine had previously been estimated to be equivalent to 5 years of heavy wear.22 After testing, all teeth were sectioned and examined to determine the cause of failure.
RESULTS Table I compares the two types of posts, regardless of the core material used, with regard to total percentage of failures, median fatigue life, and amount of initial flexure of the crown in relation to the tooth when cyclically loaded. The survival curves for the two types of posts are shown in Fig. 2. When the two types of posts were compared, there was no significant difference in the fatigue life of the postcore-crown restorations as tested by a Wilcoxon-MannWhitney test (p 0.765). Likewise there was no significant difference in the amount of flexure of the restoration as tested by a t test (p 0.8777). Table II compares the three core materials regardless of
588
post type, which was determined to have no effect, for total percentage of failures, median fatigue life, and initial flexure of the crown. The survival curves for the three core materials are shown in Fig. 3. When the three core types were compared, there was a significant difference in the fatigue life of the post-core-crown restorations as tested by Kruskal-Wallis analysis of variance by ranks (JJ < 0.0001). A multiple-range test indicates that the glass ionomer cores had a significantly lower fatigue life than the amalgam or composite-resin cores at a 95 % confidence level. There was a significant difference in the means of the amount of flexure of the crowns in relation to the tooth for the different core materials as tested by analysis of variance (p < 0.0001). Tukey’s HSD test (99 % confidence level) indicates that the amalgam cores permitted significantly less Aexure than glass ionomer cores, which permitted significantly less flexure than composite-resin cores. Sectioning of untested specimens showed that ail three core materials were well adapted to both types of posts before testing. Figs. 4 and 5 show representative samples and failure modes. Table III shows the modes of failure that were determined for each post-core-crown restoration tested.
DISCUSSION The intent of this study was to compare the durability of three core materials. Harsh clinical conditions were imposed on the materials. Specifically the margins of the crowns were placed just below the margins of the core (30 pm with respect to the tooth. In this sample the core material remained well adapted to the post. However, cement breakdown occurred at the post-root interface and at the crown-tooth interface. B, Failure of a post-core-crown restoration with a composite-resin core by post fracture.
occlusal forces were greater than normal chewing forces but well within the range of forces that occur during parafunctional activity. Results of other studies have shown that when adequate tooth structure is present-at least a 2 mm ferrule below the margin of the core-any core material is acceptable. However, there are times clinically when remaining tooth structure is so scarce that the margins of the crown must be placed at or just below the core. It is under these conditions that the choice of core material may be important. All of the teeth restored with glass ionomer core buildups failed under the conditions of this study. In each case the core material showed fracture around the post head. This material simply did not have adequate strength to withstand occlusal loading when the majority of the forces
were borne by the core material (Fig. 4). The composite resin core buildups had adequate strength (no core fractures were observed); however, 83 % eventually failed. Fifty percent of the failures occurred as a result of failure of the zinc phosphate cement at the crown-core interface, the post-root interface, or both. Forty percent of the failures occurred as a result of fracture of the post (Fig. 5, B). It is interesting to note, however, that these two modes of failure were not seen frequently for amalgam cores, which were cycled significantly longer. The high number of cement failures and post fractures for composite resin cores is apparently related to the material’s low modulus of elasticit,y. Composite resin core build-ups were much more flexible (Table II) than the other core materials tested. This extra flexure of the core placed significantly more shearing
THE
JOURNAL
OF PROSTHETIC
DENTISTRY
KOVARIK,
stresses on the cement interface and the posts themselves. The glass ionomer cores, although less flexible than the composite resin cores, also allowed significantly more flexure than the amalgam cores. It would be expected that had the glass ionomer cores themselves not broken down so quickly, a high number of cement failures and post fractures may have also occurred. Five of 34 teeth fractured while being tested in this study. The propensity to fracture would be expected to be higher for this in vitro system than clinically. The teeth were mounted in a rigid block without a periodontal ligament to more evenly disperse stresses in the root. The roots were notched to increase retention in the mounting block, and these notches acted as stress raisers in the root. However, it is interesting to note that four of five tooth fractures occurred on teeth restored with post A. Under these test conditions, this implies that post A produced more internal stress on the root than post B. However, the clinical significance of this is questionable. Five of 34 teeth tested exhibited failure as a result of fracture of the post. All of these post fractures occurred at or just below the junction of the post with the core material, and all were with post B. The tendency of post B to fracture may be due to its serrated design, which acts as a stress raiser (defect) on the surface of the post. This could create greatly increased localized stress on the post and initiate fracture. The results of this study do not support the use of glass ionomer materials as core buildup materials when there is very little remaining natural tooth structure. Although statistical significance was not shown at the p < 0.05 level, the results of this study do not support the use of composite resin core buildups when very little tooth structure remains. CONCLUSIONS 1. Teeth restored with amalgam cores and prefabricated posts had a significantly lower failure rate than teeth restored with composite resin or reinforced glass ionomer cores and prefabricated posts. 2. Teeth restored with amalgam cores provided significantly more rigid abutments for crowns and fixed partial denture retainers than did glass ionomer or composite resin cores. 3. Under the conditions used in this study, reinforced glass ionomer cores did not have adequate strength to withstand simulated occlusal forces,
590
BREEDING,
AND
CAUGHMAN
REFERENCES 1.
2. 3.
I.
8. 9. 10. 11.
12.
13. 14.
15. 16.
17. 18. 19. 20. 21. 22.
23.
Plasmans PJJM, Visseren LGH, Vrijhoef MMA, Kayser AF. In vitro comparison of dowel and core techniques for endodontically treated molars. J Endodontol 1986;12:382-7. Hudis SI, Goldstein GR. Restoration of endodontically treated teeth: a review oftheliterature. J PROSTHET DENT 1986$5:33-S. Sorensen JA, Nayyar A, Nicholls JI, Nathanson D, Caputo A, Standlee J. Current clinical trends in restoring the endodontically treated tooth: highlights of a symposium. J Clin Dent 1988;2:39-47. Chapman KW, Worley JL, Von Fraunhofer JA. Retention of prefabricated posts by cements and resins. J PROSTHET DENT 1985;54:649-52. Ruemping DR, Lund MR: Schnell RJ. Retention of dowels subjected to tensile and torsional forces. J PROSTHET DENT 1979;41:159-62. Standlee JP, Caputo AA, Hanson EC. Retention of endodontic dowels: effect of cement, dowel length, diameter and design. J PROSTHET DENT 1978;39:401-5. Tjan AH, Whang SB. Retentive properties of some simplified dowelcore systems to cast gold dowel and core. J PROSTHET DENT 1983; 50:203-6. Taleghani M, Morgan RW. Reconstructive materials for endodontically treated teeth. J PROSTHET DENT 1987;57:446-9. Linde LA. The use of composites as core material in root filled teeth. Swed Dent J 1983;7:92-9. Hoag EP, Dwyer TG. A comparative evaluation of three post and core techniques.J PROSTHET DENT 1982;47:177-81. Marshak BL, Shemen BB, Cardash HS. Use of a special matrix system for constructing amalgam and composite cores. J PROSTHET DENT 1981;57:21-2. Brandal JL, Nicholls JI, Harrington GW. A comparison of three restorative techniques for endodontically treated anterior teeth. J PROSTHET DENT 1987;58:161-5. Kao EC, Hart S, Johnston WM. Fracture resistance of four core materials with incorporated pins. Int J Prosthodontol 1989;2:569-78. Nayyar A. Coronal-radicular buildup for endodontically treated teeth. In: Clark JW, ed. Clinical dentistry. ~014. Philadelphia: Harper&Row, 1983:1-28. Ghan KC, Fuller J, Khowassah M. The adaptation of new amalgam and composite resin to pins. J PROSTHET D~~~1978;38:392-5. Paige H, Hirsch SM, Gelb MN. Effects of temporary cements on crown-to-composite resin core bond strength. J PROSTHET DENT 1986;55:49-52. Oliva RA, Lowe JA. Dimensional stability of composite used as a core material. J PROSTHET DENT 1986;56:554-61. Wilson AD, McLean JW. Glass ionomer cements. Chicago: Quintessence Publishing, 1988. Engelman MJ. Core materials. J Calif Dent Assoc 1988;16:41-5. Caldwell RC. Adhesion of foods to teeth. J Dent Res 1962;41:821-32. Trabert KC, Caputo AA, Abou-Rass M. Tooth fracture-a comparison of endodontic and restorative treatments. J Endodontol 1978;4:341-5. Outhwaite WC, Twiggs SW, Fairhurst CW, King GE. Slots vs pins: a comparison of retention under simulated chewing stresses. 3 Dent Res 1982;61:400-2. Okeson JP. Management of temporomandibular disorders and occlusion. 2nd ed. St. Louis: CV Mosby, 1989.
Reprint requests to: DR.ROBERT E.KOVARIK DEPARTMENTOF ORALHEALTHPRACTICE UNIVERSITY OF KENTUCKY LEXINGTON,KY~O~~~-OO~~
OCTOBER
1992
VOLUME
68
NUMBER
4
