Apical root strain as a function composite resin core
of post extension
Joanne N. Walton, DDS,a N. Dorin Ruse, PhD,b and Ned University of British Columbia, Vancouver, B. C., Canada
Glick,
into a
PhDC
This study tested the hypothesis that there is no difference in the amount of strain measured in the root of an extracted tooth at the apical end of a prefabricated post, whether a composite resin core completely covers the head of the post or the post is exposed at the occlusal surface of the core. Twenty extracted teeth were measured (1) after endodontic treatment (baseline), (2) after a composite resin core was fabricated over a prefabricated post, and (3) after a cast crown was cemented. The teeth were randomly divided into two groups after baseline testing. In group 1, the post was buried within the core, and in group 2, the head of the post was exposed on the occlusal surface of the core. Analysis of results indicated a statistically significant decrease in strain from the baseline to the crown when 1 mm of composite resin covered the head of the post. However, based on the small strain values measured, the difference may not be clinically significant. (J F~OSTHET DENT 1996;75:499-505.)
D
espite concerns about the physical properties of composite resins, le3 they have shown success as core buildup materials in clinical applications4 and are the most commonly used direct core materials for the restoration of endodontically treated teeth.5 Because of the ease and efficiency of use, composite resins have been recommended as the material of choice for this purpose.6 When additional retention is required for core material, the judicious use of passively fitting prefabricated posts is advised.7,s However, conflicting opinions have been expressed about the appropriate extension of the post into a composite resin core (Fig. 1). Nathanson suggested that approximately 1 mm of composite resin should overlie the head of a prefabricated post after the crown preparation has been completed to prevent the transmission of stresses on the restoration through the post into the root. On the other hand, Christenson9 advised that the head of the post should extend through the composite resin core to the surface of the preparation to prevent thin, unsupported areas of resin from fracturing during provisional restoration fabrication or on removal of the final impression. To address these divergent clinical opinions, this study tested the hypothesis that there is no difference in the amount of strain measured in the root of a tooth whether the post is exposed or buried in a composite resin core.
MATERIAL
AND METHODS
A list of the materials and instrumentation used in this study is presented in Table I. From a pool of 54 unrestored, freshly extracted mandib-
aAssistant Professor, Division of Prosthodontics, Faculty of Dentistry. bAssistant Professor, Division of Biomaterials, Faculty of Dentistry. cProfessor Emeritus, Department of Statistics and Department of Health Care and Epidemiology.
MAY 1996
ular first premolars that were measured and carefully checked for fractures, 20 intact teeth that were similar in length and root diameter were selected. Each tooth was radiographed from the facial and the proximal views to confirm that the dentin thicknesses were comparable. Throughout the testing procedures, the teeth were kept moist by immersing them in sterile, distilled water or by covering them with a moist gauze. The plan was to simulate a common clinical occurrence by performing endodontic treatment on each tooth and then restoring the teeth with prefabricated posts and composite resin cores over which cast gold alloy crowns would be cemented. Each tooth was tested at three separate stages: stage 1, after endodontic treatment but before post and core build-up (baseline); stage 2, at the post and core stage (post/core); and stage 3, after the cast crown had been cemented (crowned). Although the teeth were selected to be as similar to each other as possible, each tooth was also able to serve as its own control by use of this staged testing, which started with intact teeth. Endodontic treatment was completed for each tooth, and canal preparations were made to allow passive seating of a 15 mm long Unity size 3 prefabricated titanium alloy post to a level 5 mm above the apex of the root. lo The teeth were radiographed a second time to ensure that there was at least 1 mm of dentin lateral to the apical end of the post space. Strain gauges were then bonded to the surface of each root by use of a protocol similar to that described by Ross et al.ll Each tooth was lightly disked on the proximal surface 6 mm above the root apex to facilitate gauge bonding. The teeth were then cleansed and the gauges were prepared and bonded according to manufacturer’s recommendations (Fig. 2). The quality of the gauge bond was verified with a x10 microscope; no air bubbles were found between the root surface and the strain gauge. Even though the strain gauges were encapsulated in a polya-
THE JOURNAL
OF PROSTHETIC
DENTISTRY
499
THE JOURNAL
OF PROSTHETIC
DENTISTRY
WALTON,
RUSE, AND GLICK
Fig. 1. Diagrammaticrepresentationofoneprefabricated post extending to point 1 mm below occlusal surface of composite resin core &ft) and second post extending to surface of core (right). Fig. 2. Encapsulated strain gauge bonded 6 mm above apex. mide coating for moisture resistance, a layer of tray adhesive was applied over each gauge and the surrounding tooth structure to ensure that there would be no moisture contamination of the bond. To allow flexion of the root at the level of the apical third of the post under testing conditions, a layer of flexible impression material was placed over the strain gauges and around the circumference of each root (Fig. 31. Each tooth was suspended parallel to its long axis (Fig. 4)and embedded in an acrylic resin jig to stabilize the tooth during testing (Fig. 5). The occlusal surfaces of the teeth were slightly flattened with a 240~grit silicon carbide under-polisher to make the teeth pe~en~~l~ to the long axis of the tooth for subsequent compressive testing and to facilitate seating of the platen of the Universal testing machine (Instron Corp., Canton, Mass.). The strain gauge lead wires were then attached to a strain gauge readout device and the acrylic resin jig containing the tooth was placed in a water vessel on the universal testing machine (Fig. 6). Strain in the root at the apical portion of the post was measured in microstrain units (pe) while each tooth was subjected to a 200 N compressive load that was perpendicular to the long axis of the tooth. This magnitude of compressive load is consistent with average biting forces in the bicuspid regionI Each measurement was replicated 10 times to ensure
500
that the measurement error was negligible. Readings were made on the strain gauge readout device as soon as the maximum 200 N load was reached and the teeth were permitted to return to the initial passive state between load appli~tions as was measured by a return to zero on the readout device to ensure that the strains induced were not cumulative. These measurements were then averaged for each tooth and the standard deviation was calculated to establish baseline strain data. To ensure the consistency of readings over time, teeth were randomly retested up to 3 days after initial measurements were made. A matrix of polyvinyl siloxane putty material was placed over the crown of each tooth and then cut in half to allow measurement of subsequent post and core heights and to permit restoration of the teeth to baseline dimensions (Fig. 7). The crowns of the teeth were then removed to a level 2 mm above the facial eementoenamel junction and the teeth were randomly divided into two groups. For each tooth in group 1, one of the previously selected posts was fitted to finish 2 mm below the planned occlusal surface as measured against the putty matrix (Fig. 8). After cemen~tion of the post with a glass ionomer luting agent, an autopolyme~zing composite resin core was fabricated so that 1 mm of resin covered the top ofthe post and
VOLumE
75
IWMBER
5
WALTON,
THE JOURNAL
RUSE, AND GLICK
OF PROSTHETIC
Fig. 4. Device for suspending axis in preparation
Fig. 3. Layer of polyvinyl siloxane over strain gauge and around root.
impression
parallel
to its long
material
the occlusal surface of the core was, in turn, 1 mm below the planned occlusal surface of the crown (Fig. 9). This group is referred to in the Results section as “buried.” For group 2, teeth were treated in the same manner as those in group 1 with the exception that the head ofthe post was 1 mm below the planned occlusal surface when measured against the putty matrix. Thus, the head of the post was flush with the occlusal surface of the composite resin core (Fig. 10). This group is referred to as “exposed.” In both groups each tooth was subsequently prepared for a complete metal crown. The crown preparation was extended 1 mm apical to the tooth and core junction to provide a 1 mm collar of tooth structure.13, l4 The teeth were then tested on the Instron machine to measure root strain at the apical end of the post when a 200 N compressive force was applied to the occlusal surface of the core. Once again, a series of 10 readings was made and the mean and standard deviation were calculated for each tooth. At this point, the crowns of the teeth in both groups were waxed to baseline dimensions with the putty matrices (Fig. 11) and then invested and cast in a gold alloy. The crowns were examined for fit, the internal surfaces were abraded with 40 pm aluminum oxide, and they were then cemented with a glass ionomer luting agent (Fig. 12). Testing was then completed through 10 trials for each crowned tooth
MAY 1996
tooth for embedding.
DENTISTRY
Fig. 5. Tooth embedded
in acrylic
resin testing
jig.
with the same compressive forces previously used. The strain readings recorded were averaged once again and the standard deviation was calculated.
RESULTS The strain measurements for each sample are summarized in Table II. Each measurement in Table II is the average of 10 replications. Replicate coefficients of variation are not shown and were not used in the analyses, but were less than 10% for every measurement without exception. One tooth in group 2 was excluded from analysis because its baseline strain value was an outlier, almost twice as high as any other strain value observed at baseline or otherwise. The effective sample sizes were therefore 10 in group 1 and nine in group 2.
F;cIl
THE JOURNAL
OF PROSTHETIC
DENTISTRY
WALTON,
Fig.
Fig.
6. Sample
compressed
Table I. Materials
and instruments
Item
gauge
Adhesive & catalyst Tray adhesive PVS Impression
materials
Acrylic resin Composite resin Universal testing
core material machine
Strain gauge readout device Gold
alloy
testing machine.
siloxane
putty
matrix.
1 post 2 mm below planned
occlusal
sur-
used in this study
#3-Ti alloy Metaserv Fuji I EA-13-062AQ-350P M Bond 200 Extrude Medium body President putty Formatray Ti-Core 4301 TN8C Transtrain Indicator Harmony Hard Type III
The hypothesis of null median difference between treatments was tested with a two-sample nonparametric rank procedure, Wilcoxon rank sum test (which is equivalent to the Mann-Whitney test) (Fig. 13). No significant difference was found (p = 0.4) between the mean strain in groups 1 and 2 at baseline. Even if the outlier in group 2 was included, there was no significant difference between the two groups (p = 0.25). Similarly, comparison of the two groups after a post and core was placed in each tooth but
502
Fig. 8. Group face.
Description
Unity post Grinder-polisher Glass ionomer luting cement Strain
on universal
7. Polyvinyl
RUSE, AND GLICK
Whaledent
Manufacturer
Location
Int’l.
New York, N. Y. Coventry, England Tokyo, Japan Raleigh, N. C.
Buehler, UK, Ltd. GC Corp. Micro-Measurements Group, Micro-Measurements Group, Kerr Mfg. Kerr Mfg. Coltene Mfg. Kerr Mfg. Essential Dental Systems Instron Canada, Inc. Micro-Measurements Group, Williams Gold
Inc. Inc
Inc.
Raleigh, N. C. Romulus, Mich. Romulus, Mich. Mahwah, N. Y. Romulus, Mich. Hackensack, N. J. Burlington, Ont. Raleigh, N. C. Buffalo, N. Y.
before crowning indicated no significant difference (JJ = 0.4). However, there was a significant reduction of strain 0, = 0.01) from the baseline measurement to the crowned condition for group 1 teeth in which the post was left buried within the core. The “crowned” strain value was less than the corresponding baseline stain value for every tooth in group 1 except the two that had the lowest initial measurements (both were below 50 pE, whereas most other baseline values exceeded 100 pE). When the two
VOLUME
75
NUMRER
5
WALTON,
RUSE, AND GLICK
THE JOtJRK&
Fig. Fig. 9. Composite occlusal surface.
resin core buildup
11. Wax-up
view of group 2 core buildup,
with post
treatment groups were compared with respect to strain values for teeth after crowning, the difference was statistically significant (p = 0.01). Again, even if the baseline outlier in group 2 had not been excluded, this difference would still have been significant (p < 0.01).
DISCUSSION Strain is described as the change in length per unit length when a stress is applied to an objectr’ and its measurement in teeth permits assessment of the dimensional changes that occur when functional stresses are applied. According to Standlee and Caputo,15 the application of occlusal function is the most severe stress that can be applied to a post and core or crown, and the most dangerous place for those stresses to cause strains are in the apical portion of the root, where fracture often renders a tooth nonrestorable.16 In addition, stress from occlusal forces on posts has
MAY 1996
of crown to baseline
DRESSY
dimensions
1 mm below planned
Fig.
Fig. 10. Occlusal “exposed.”
OF PROS~C
12. Completed
gold alloy crown.
been cited as a cause of vertical root fractures in endodontically treated teeth.17 Up to this point, the effect on apical root strain of varying the coronal extension of a prefabricated post into a composite resin core was not known. At each stage in this study, the strain measurement for each tooth had little error relative to the variation among teeth. However, for either group at each stage, the variation or range or strain values among the teeth was substantial, compared with the group’s average or mean. Moreover, these variations may be quite different for the two groups. This is not unexpected given the sensitive nature of gauges that measure such small amounts of strain and the inherent variability among natural teeth. The strain gauges detected small differences between teeth, but-given the small magnitude of strain recorded-such variations contributed to substantial standard deviations between samples. Indeed, strain appears to be a variable that has a more-or-less constant coefficient of variation (standard deviation/mean ratio) instead of constant standard deviation. Under these conditions, especially with small samples sizes, a t-test is not appropriate to compare outcomes of two groups, and our two-sample comparisons therefore use a nonparametric rank procedure, the Wilcoxon rank sum test.
CLINICAL Although is apparent,
SIGNIFICANCE the statistical difference between treatments it is questionable whether the observed treat-
503
WALTON,RUSE,ANDGLICK
THEJOURNALOFPROSTHETICDENTISTRY
140120 --
n Grou 1 post &Wed
qGroup
2 post exposed
Post/Core (p = 0.4)
Baseline (p = 0.4) Fig.
Table
II.
13.
Crowned (p = 0.01)
Bar chart of statistically analyzed data.
Results of strain measurements (in PEP Baseline Group
1 2 3 4 5 6 7 8 9 10
Mean? SW
SD~ean ICVN
141.0 49.2 268.1 183.3 370.8 20.5 109.8 204.9 271.8 102.3 172.2 109.4 0.64
*Each measurement given is the average iExcluding one outlier [ ] in group 2.
1
Post Group 168.9 84.9 255.2 55.0 420.2 282.4 260.6 385.6 120.2 1849.01 225.9 128.5 0.57
of 10 replications
2
Post
buried
119.1 82.3 247.4 261.9 31.3 35.4 24.2 249.9 237.8 210.8 150.0 101.1 0.67
and core
Crowned
Post
exposed
202.3 242.3 150.6 278.3 167.7 283.4 95.9 191.4 161.7 L352.51 197.1 61.9 0.31
Post
buried
113.2 82.9 133.0 36.1 14.2 33.8 40.8 37.3 121.1 92.8 70.5 43.0 0.61
Post
exposed 44.1 292.3 92.8 172.7 102.0 303.6 108.6 448.2 162.7 L381.31 191.9 130.5 0.68
for each tooth.
ment effect has clinical significance, because the strain values for the crowned teeth in which the post was left “exposed” instead of “buried” seem similar to strain values for the teeth at the baseline measurement. Furthermore, in evaluating the range of strain values that may be clinically important in dentin, a simple calculation of strain as a function of ultimate tensile strength of dentin over its modulus of elasticity with figures given for each by Craigi provides an estimation of maximum strain before fracture of approximately 2787 ue for dentin. This is a factor of al-
most seven higher than the great strain value used for calculations in this study. In light of these considerations, the potential clinical advantages of extending a prefabricated post to the surface of an anterior core buildup to support the thin composite resin at the incisal edge may outweigh concerns about possible differences in the amount of stress that was generated at the apical end of the post. However in a posterior tooth, where there is less likelihood of unsupported resin at the oeclusal surface and therefore less clinical advantage to exposing the post, the clinician
WALTON,
RUSE, AND GLICK
THE JOURNAL
may choose to cover the head of the post with composite resin. Further study of strain in endodontically treated teeth restored with posts and cores may be facilitated by (1) establishing baseline apical strain values of unrestored teeth before endodontic treatment to provide more clinical direction in assessing “restored’ versus “intact” strain; (2) developing standardized models to enhance comparability; (3) conducting finite element analysis with computer modeling based on known Poisson’s ratios of the elements involved; and (4) conducting additional testing at an angle to the long axis of the tooth and fatigue testing to create a more complete picture of the strain generated in the dentin of teeth restored with posts and cores and crowns.
CONCLUSIONS Under the conditions of this study, there was a statistically significant decrease in apical root strain measured in endodontically treated teeth that were restored with a prefabricated post, composite resin core, and crown when the post was 1 mm short of the occlusal surface of the composite resin compared with when the post was flush with the occlusal surface of the core. The null hypothesis that there is no difference in the amount of strain measured in the root of a tooth whether the post is exposed or buried in a composite resin core is therefore rejected. The clinical significance of this finding is less clear because the strain stayed relatively constant from baseline to crown when the post was exposed and the strains measured were small relative to the strain required to cause a fracture in dentin. Possibilities for further investigation have been suggested. We gratefully acknowledge the important contribution made by Mr. Ian Boyd, a second-year dental student, who worked as a research assistant on this study. RE~~NCES 1. Oliva
RA, Lowe JA. Dimensional stability of composite material. J PROSTHET DENT 1986;56:554-61.
I
OF PROST~TIC
DE~IS~Y
2. Oliva RA, Lowe JA. Dimensional stability of silver amalgam and composite resin used as core materials. J P~OSTHET DENT 1987;57: 554-S. 3. Engelman
MJ. Core materials. J Calif Dent Assoc 1988;16:41-5. 4. Linde L. The use of composites as core material in root-filled teeth. II. Clinical investigation. Swed Dent J 1984;8:209-16. 5. Morgan0 SM, Hashem AF, Fotoohi K, Rose L. A nationwide survey of contemporary philosophies and techniques of restoring endodontically treated teeth. J PROSTHET DENT 1994;72:259-67. 6. Nathanson D. Modern Methods of restoring endo-ly treated teeth. Clin Topics Dent, Iowa Great Plains Nat’1 Library; 1992:15-17. 7. Sorenson JA, Ma~noff JT. Intracoronal reinforcement and coronal coverage: a study of endodontically treated teeth. J PROSTHET DENT 1984;51:780-4.
8. Sorenson JA. Preservation of tooth structure. J Calif Dent Assoc 1988;16:15-22. 9. Christenson. GE. Provo: Clin Res Assoc Newsletter, 1990;14: 1-3. 10. Mattison GD, Delivanis PD, Thacker RW Jr, Hassell &I. Effect of post p~paration on the apical seal. J PROSTHET DENT 1984;51:785-9. 11. Ross RS, Nicholls JI, Hawington GW. A comparison of strains generated during placement of five endodontic posts. J Endodont 1991;17:450-6. 12. Craig R, ed. Restorative dental materials. 9th ed. St Louis: CV Mosby, 1993:64-S. 13. Hoag EP, Dwyer TG. A comparative evaluation of three post and core techniques. J PROSTHET DE&T 1982;47:177-81. 14. Gelfand M, Goldman M, Sunderman EJ. Effect of complete veneer wowns on the compressive strength of endodontically treated posterior teeth. J PROSTHET DENT 1984;52:635-8. 15. Standlee JP, Caputo AA. Biomechanics. J Calif Dent Assoc 1988;16:4958. 16. Sorenson
JA, Martinoff JT. Clinically significant factors design. J PROSTHET DENT 1984$X&28-35. 17. Peters MC, Poort HW, Farah JW, Craig RG. Stress analysis restored with a post and core. J Dent Res 1983;62;760-3.
in dowel ofa tooth
Reprint requests to: DR. JOANNE N. WALTON DEPARTMENT OF CLINICAL DENTAL SCIEKES FACULTY OF DEMBTRY 2199 WESBR~K MALL
VANCOWER V6T 123 CANADA
B.C.
Copyright
0 1996 by The Editorial
PROSTHETIC
used as a care
Council of
THE
JOURN~
OF
DENTISTRY.
0022-3913/96/$5.00
+ 0.
10/l/71249
l-800~55-MOSBY
I
This number links you to the full text of articles published in more than 25,000 Journals, in&ding all Mosby journals. Mosby Express@, a rapid response information retrieval service, provides quick turnaround, 24-hour availability, and speedy delivery methods. For inquiries and pricing information, call our toll-free, 24-hour order line: l-800~55MOSBY; outside the United States: 415-259-5046; fax: 415-259-5019; E-mail:
[email protected].
D~u~ut
MAY 1996
505
