0099-2399/92/1805-0209/$03.00/0 JOURNAL OF ENDODONTICS Copyright © 1992 by The American Association of Endodontists
Printed in U.S.A.
VOL. 18, NO. 5, MAY 1992
Effects of Moisture Content and Endodontic Treatment on Some Mechanical Properties of Human Dentin Tzyy-Jou G. Huang, BMD, MSD, Herbert Schilder, DDS, and Dan Nathanson, DMD, MSD
The objective of this study was to determine whether significant differences exist between the mechanical properties of human dentin from treated pulpless teeth and dentin from normal vital teeth. Dentin specimens (n = 262) were obtained from 54 freshly extracted normal vital human teeth and 24 treated human pulpless teeth. These specimens were subjected to different experimental conditions (wet, air dried, desiccated, and rehydrated). Compression, indirect tensile, and impact tests were conducted to measure the mechanical properties of those specimens. All data obtained were analyzed with t tests. The results showed that the dehydration of dentin increases the Young's modulus, proportional limit (in compression), and especially the ultimate strength (in both compression and tension). Substantial dehydration changes the fracture characteristics of dentin specimens under static compressive and indirect tensile Ioadings. The measurements of impact-breaking energies of desiccated dentin were not found to be significantly decreased. The compressive and tensile strengths of dentin from treated pulpless teeth obtained in this study do not appear to be significantly different from those of normal dentin (p > 0.05), while the mean values of Young's modulus and proportional limit in compression tests appear to be lower. Fifty percent of the dentin specimens from treated pulpless teeth exhibit greater plastic deformation than normal dentin in compression. The results of this study do not support the theory that dehydration after endodontic treatment per se weakens dentin structure in terms of compressive and tensile strengths. Other mechanical properties of treated pulpless teeth, however, may not be the same as those of normal vital teeth.
teeth. Certain mechanical terminologies have been frequently applied by many authors to describe the physical condition of teeth that have had root canal treatment, such as increased "brittleness," "friability," and "fragility;" reduced "resiliency," "elasticity," and "strength" (1-3). The reduction in tooth structure and the effect of dehydration on the dentinal tubules are widely considered to be the main reasons associated with increased weakness and brittleness in pulpless teeth (3). The recent study by Reeh et al. (4), however, revealed that endodontic procedures reduce the tooth stiffness by only 5%. In addition, the role that water plays in the biomechanical behavior of teeth has not been well understood nor fully investigated. As early as 1895, Black (5) concluded that the dentin of human pulpless teeth had less crushing strength than normal teeth. Later investigations (6-8) did not support Black's finding until Carter et al. (9) in 1983 showed the reduction of shear strength and shear toughness of endodontically treated teeth. In this study, an effort was made to reaffirm those studies, and, especially, to realize the effects of moisture content on the biomechanical properties in human teeth. The complexity of the clinical problems, the myriad of physical factors effecting clinical behavior of treated pulpless teeth, as well as the intricacies of analyzing and testing their interrelationships posed a dilemma. Accordingly, the decision was made to divide out several parameters of the problems and to limit the project to measurements of these factors alone. Extensive mechanical tests on human dentin of treated pulpless and vital teeth were performed. Moreover, the definitive mechanical terminologies of material testing science were applied to interpret the experimental data obtained. Potential difficulties due to variations and experimental errors that might occur in such a study were noted, evaluated, and avoided as much as possible. MATERIALS AND METHODS Teeth Collection
Two groups of freshly extracted human permanent teeth were collected. Group one consisted of 54 teeth with minimal caries that were vital at the time of extraction and group two
It is generally believed that endodontic treatment renders teeth weaker and more subject to fracture than normal vital 209
210
Huang et al.
consisted of 24 teeth that had been endodontically treated at least 1 yr before extraction. Both groups comprised mixtures of anterior and posterior teeth with donor patients' age varying from 20 to 88 yr. The teeth were stored in separate vials containing a special storage solution (composed of 4 mM Na2HPO4, 2 mr~ CaC12, 20 mM KC1, 18 mra NaCI, and 1% NAN3). This solution was used to prevent either dehydration or decalcification of the specimens during storage. Specimens preparation was performed as quickly as possible in each case, never more than 7 days after the tooth was extracted, with a steady flow of water in the cutting field.
Preparation of the Dentin Specimen For compression and indirect tensile tests, cylindrical specimens of dentin were cut out of the coronal and/or radicular portion of teeth from both groups (Fig. 1). This was accomplished by using an Isomet 77-1180 low-speed saw (Buehler Ltd., Evanston, IL) and trepan burs (Brasseler USA, Inc., Savannah, GA) attached to a Unimat-3 miniature lathe (Hobby Product Co., Columbus, OH). The sizes of the specimens for compression tests were from 0.13967 to 0.07819 inches (3.54 to 2.00 mm) in length and 0.06340 to 0.03992 inches (1.60 to 1.00 mm) in diameter (length to diameter ratio of 1.7:2.2). The sizes for indirect tensile tests were 0.04730 to 0.07680 inches (1.20 to 1.95 mm) in diameter and 0.03940 to 0.06450 inches (1.00 to 1.64 mm) in length (diameter to length ratio 1.5:1). The two ends of each cylindrical specimen were trimmed parallel and smooth. For impact tests, specimens were obtained only from coronal dentin of molar teeth from group one and made into a rectangular shape of 0.0630 inches x 0.0630 inches × 0.394 inches (1.60 x 1.60 x 10.00 mm). No notches were made on such small impact test specimens because it was not easy to prepare the notches consistently with the same geometry for all of them. Each prepared specimen was checked for defects under a dissecting microscope at 30 times magnification and any specimens with visible cracks or flaws were discarded. No
Journal of Endodontics
specimens that involve decay or structures such as cementum were used, i.e. only specimens with sound dentin were tested and these rules applied to both groups of teeth. Each specimen was then stored in a labeled plastic tube. Storage solution was placed only in the tubes for wet specimens. Specimens for compression and indirect tensile tests were divided into subgroups and received different treatments: remained wet, air dried at room temperature for 3 days, desiccated over CaSO4 for 2 days at room temperature, or rehydrated previously air-dried specimens in the storage solution for approximately 2 h. Specimens for impact tests were studied only wet and desiccated. In all tests, the wet and rehydrated specimens were tested wet, and the dried specimens were tested dry.
Compression Test An Instron 1331 testing machine (Instron Corp., Canton, MA) of 2200-1b capacity load cell was used to apply loads to specimens. Each specimen was tested between two superhardened steel platens. Due to the small size of the specimens, the strain gauge (1.969 inches) had to be fastened to the platens. Immediately before testing, the length and diameter of each specimen were measured by a micrometer caliper (L. S. Starrett Co., Athol, MA) accurate to 0.000039 inch (1 #m) and recorded. A thin layer of grease was painted in the center of the two platens' surfaces to reduce friction. Each specimen was compressed until it failed at the speed of 0.0015 to 0.0030 inches/min. A stress-strain curve was drawn by the X-Y chart recorder and the characteristics of deformation of each specimen were observed with a magnifying lens at x3 during the compression. After each specimen failed, the peak load (lb) and consumed time (s) were recorded. In addition, the mode of the fractured specimen was inspected under the dissecting microscope and recorded. The ultimate strength was calculated by the formula: Stress = maximum load/area The proportional limit was measured on the stress-strain curve. The modulus of elasticity (Young's modulus, or E) was calculated by the formula: E = stress/strain
"'" ~'1 Silo !lie...... .... • -4........ J f i I
FIG 1. Diagrams indicating the locations (crown and/or root) from which the dentin specimens were taken. The horizontal lines represent the slicing planes.
Stanford et at. (6) and Craig and Peyton (10) previously described potential experimental errors that should be considered in small size specimen testing of these types. One factor that would create errors (especially in the Young's modulus) in our test might be the gap effect due to nonparallelism of the two ends of the specimens. Stanford (6) stated that there is no known equation for the correction of this factor, since it involves probability of how the platens contact the ends of the specimens. Craig and Peyton (10) found that the length to diameter ratio appeared to lower the Young's modulus for the ratio between 1 and 0.5. In our study, two materials, aluminum alloy 2024-T4 and acrylic cast, were also made into similar shapes and dimensions as the dentin specimens to receive compression tests. Comparison was then made to the previously published data of those materials to verify the amount of error that may occur in our tests.
Dentin Moisture Content
Vol. 18, No. 5, May 1992
211
TABLE 1. Compressive properties of human dentin in different conditions*
Condition of Specimen Normal vital Wet Air-dried Desiccated
Rehydrated Treated pulpless Wet Air-dried Rehydrated
Location of Specimen
No. of Specimens
Crown Root Crown Root Crown Root Root Crown Root Root Root
Young's Modulus (E) (106 psi)
Proportional Limit (psi)
Ultimate Strength (psi)
Mean
SD
Mean
SD
Mean
SD
11 17 7 6 4 6 6
1.82 (2.17)t 1.86 (2.21)t 2.00 (2.39)t 2.27 (2.72)t 2.17 (2.60)t 2.47 (2.96)t 1.82 (2.17)t
0.30 0.27 0.29 0.27 0.46 0.38 0.29
22,500 22,300
4,700 3,400
42,800 36,400
3,000 4,400 I
22,900
3,000
56,700
5,000 t
29,000
4,200
59,700
3,40011
29,500 34,400 24,600
5,700 7,700 6,000
70,100 69,700 39,900
7,800 7,700 3,300
8 34 4 3
1.27 (1.49)t 1.49 (1.76)t 2.79 (3.36)t --
0.50 0.49 1.05
17,600 19,600 24,800 18,100
6,400 5,400 5,200 10,200
45,100 37,600 50,900 28,300
7,200 5,600 5,300 7,200
* The difference between any two means joined by the same thick vertical lines is not significant (t test, p > 0.05), and the difference between those joined by the same thin vertical lines is significant (t test, p < 0.05). t Mean values calibrated using the linear interpolation method based on the calibration data in Table 2. The acrylic cast accepted Young's modulus (E), 0.40 x 10e psi (midpoint of the range 0.35 to 0.45), was chosen to calibrate.
Indirect Tensile Test An Instron model 4202 testing machine with a 200-1b capacity load cell was used for this test. A layer of aluminum foil with a thickness of 0.00059 inches was taped on the two plat~tls as padding material. Each specimen was placed and oriented between the two platens and the load was applied at 0.003 inches/min cross-head speed until failure occurred. All of the specimens were oriented between the platens in such a way that the applied loads were approximately parallel to the direction of dentinal tubules. This ensured that the tensile stress created was approximately at right angles to the tubules even though the tubules do not run in regular directions. Continuously during the testing, recordings were made of the applied load versus time and recorded on load-time charts. A magnifying lens was used to observe changes of the specimens during load application. After failure occurred, the maximum load was recorded and each fractured specimen was examined under the dissecting microscope. The tensile stress was calculated by the equation: Stress = 2 x maximum load/Tr × diameter x thickness
Impact Test Since specimens of 0.0630 inches x 0.0630 inches x 0.394 inches were the largest geometrically regular dentin blocks that could be obtained from human teeth, an impact testing machine with small capacity had to be used. A Charpy impact machine model CIM-1 (Manlab Inc. (Physmet), Cambridge, MA) with 12-inch-lb maximum energy capacity was the smallest one that could be found, but the anvil was still too large to accommodate the small rectangular dentin specimen. A specially built anvil was made according to the specification of ASTM E23 (American Society for Testing and Materials, group E). This converted the Charpy impact machine model into an Izod-type impact machine. Both Charpy and Izodtype have equal standing in the scientific community for this type of test. The specimens were positioned and fixed to the
anvil and the pendulum was activated. The pendulum fractured the specimens into two parts and the energy consumed was recorded. Some of them, which were struck into several pieces, showed much higher energy readings and were not counted. RESULTS
Compressive Properties of Dentin The results and the mean values of the compressive tests appear in Table 1. The statistical analysis (t test) indicates that dehydration of dentin tends to increase the Young's modulus, proportional limit, and especially the ultimate strength. Figure 5 depicts the stress-strain relationships of those specimens under different degrees of dehydration. The slopes of the linear portion of the curves (Young's modulus in graphic form) become steeper, and the end points of the curves (ultimate stresses when failures occur) become much higher as the result of progressive dehydration. The mean values of ultimate strength of wet dentin specimens show no significant difference between treated pulpless teeth and normal vital teeth (p > 0.05). However, the Young's modulus and the proportional limit of wet dentin specimens from pulpless teeth are significantly lower than those from normal vital teeth (p < 0.05). Fifty percent of the specimens from pulpless teeth demonstrated greater plastic deformation than did those from vital teeth. These observations are well characterized in Figs. 6 and 7. Wet conditions only are compared between two groups of tooth samples, since specimens treated with other conditions may have multiple variables. The obtained values in this study, Young's modulus of aluminum alloy 2024-T4 and standard cast acrylic blocks, were slightly lower than the published values as may be observed in Table 2. The Young's moduli in Table 1, accordingly, were corrected through this calibration. For dentin specimens from vital teeth, the fracture patterns of most of the wet and air-dried specimens are shown in Fig.
Joumal of Endodontics
Huang et al.
212
TABLE 2. Calibration data for determination of Young's modulus (E)
Material
Observed E (108 p s i )
Accepted E* (10s psi)
8.65
10.60
0.39
0.35-0.45
Aluminum alloy 2024-T4 Acrylic cast
• Source: Measurement group, Raleigh, NC,
K\\\\\",3
22
(a)
(b)
(c)
FIG 3. Selected fracture patterns of dentin specimens under indirect tensile Ioadings. a and b, Wet and/or air-dried dentin specimens, c, A few dentin specimens from treated pulpless teeth demonstrating excessive deformation. TABLE 4. Impact test of human dentin (normal vital)
r~
(a)
(b)
(c)
(d)
FIG 2. Selected fracture patterns of dentin specimens under compressive Ioadings. a and b, Wet and/or air-dried dentin specimens from normal vital teeth, c, A desiccated dentin specimen from normal vital teeth. The sma/I arrows indicate that dentin fragments ejected in all directions when it fractured, d, The bulging deformation under pressure of some wet dentin specimens from treated pulpless teeth.
Condition of Specimen
No. of Specimens
Wet Desiccated
13 14
Energy (inch-lb) Mean SD 0.046 0.035
0.019 0.013
TABLE 3. Indirect tensile strength of human root dentin in different conditions*
Condition of Specimen Normal vital Wet Air-dried Desiccated Treated pulpless Wet Air-dried Rehydrated
No. of Specimens
Ultimate Strength (psi) Mean
SD
30 17 11
8,650 12,290 16,900
1,230 1,5301 2,660
FIG 4. Drawings indicating a dentin specimen for impact test and a few ruptured specimens after impact Ioadings.
42 7 5
8,640 10,270 8,170
1,420 I 1,5401 1,250
dentin specimens between treated pulpless teeth and normal vital teeth (p > 0.05). For dentin specimens from vital teeth, most wet specimens failed by diametral cleft formation with or without secondary cracks. The fracture modes of air-dried specimens were mainly triple-cleft formation, and all except one of the desiccated specimens appeared as triple-cleft fractures (Fig. 3, a and b). Specimens from treated pulpless teeth have similar fracture characteristics as do samples from vital teeth. A few specimens, however, demonstrated excessive ductility which rendered them inappropriate to receive this test. Their data were not included (Fig. 3c).
* The difference between any two means joined by the same thick vertical lines is not significant (t test, p > 0.05), and the difference between those joined by the same thin vertical lines is significant (t test, p < 0.05).
2, a and b. They separated or cracked into two to three fragments. The desiccated specimens shattered into many pieces at the moment they failed (Fig. 2c). The average dentin specimens from treated pulpless teeth had similar fracture characteristics as did those of vital teeth. An exception was that those demonstrating more bulging during stress developed more minor cracks (Fig. 2d). Indirect Tensile Test The results of the indirect tensile tests are presented in Table 3. Statistical analysis indicates that dehydration increases dentin tensile strength dramatically. There is, however, no significant difference in ultimate tensile strength of wet
Impact Test In impact tests, the fracture modes showed no difference between desiccated and wet dentin. Table 4 shows that there is no significant difference between the two means (p > 0.05), although the mean value of the desiccated dentin appears lower than the wet one. The ruptured ends of the specimens demonstrate various styles of fractured surfaces (Fig. 4).
Vol. 18, No. 5, May 1992
Dentin Moisture Content
DISCUSSION
80 70
Effects of Moisture Content on Dentin
% The loss of moisture content, 9% by weight, of pulpless teeth was quantified by Helfer et al. (11). Many authors have implied that dehydration increases brittleness and makes teeth more susceptible to fracture. Studies corresponding to the findings of Helfer et al. (11) were reported as early as 1937 and 1943. Bodecker and his colleague (12, 13) used a dye penetration technique on both animal and human material to study water permeability of pulpless teeth. Their observation was that teeth which have been devitalized for a long period of time, at least 1 yr, demonstrate little permeability. They explained this by the existence of gas in the dentinal tubules which developed subsequent to the drying out of dental lymph some time after the removal of pulps. This gas, they suggested, made pulpless teeth highly impermeable to fluids. However, considering that pulpless teeth continue to exist in a water-based environment, saliva in the mouth and blood and tissue fluid surrounding the root, it has never been fully explained why dentin should dry out. Cementum is permeable without tubules and considerable evidence indicates enamel permeability to water even though enamel is a much denser calcified tissue. There is little reason to consider treated pulpless teeth in situ as "dried." In addition, the presence of the gas in dentinal tubules has never been verified. Regardless of whether the conclusions of Heifer et al. (11) and Bodecker (12, 13) were true or not, the effect of the dryness per se on the mechanical properties of tooth dentin was determined to be of sufficient interest to be investigated. It should be noted that it is not possible to utilize a dentin specimen from either a vital or pulpless tooth and maintain its original moisture content. During the specimen preparation, constant water flow had to be applied to avoid overheating the cutting field. Also the storage of the specimen in the storage solution would certainly change the water content. Specimens from teeth presumed to be dehydrated by reasons of pulplessness were required to be redehydrated in the laboratory after specimen preparation. It was observed in this study that different degress of dehydration of dentin were established after the preparation of the specimens. The results indicate that the dehydration of dentin tends to increase the Young's modulus, proportional limit, and especially the ultimate strength (both compressive and tensile). According to the stress-strain diagrams, the dried dentin specimens were stiffer than those of wet ones (Fig. 5). The fracture patterns of dried dentin specimens demonstrated different characteristics from the wet ones. Some air-dried dentin specimens were rehydrated in the storage solution for approximately 2 h before the compression and indirect tensile tests. They demonstrated similar characteristics to normal wet dentin. In terms of toughness, area under stress-strain curve (Fig. 5), there is no indication that the toughness of dried dentin is less than that of normal wet dentin. The energy absorbed by a material before rupture, i.e. toughness, is sometimes more accurately obtained from the stress-strain curve of the uniaxial tension test which was impractical to perform with the small dentin sample in this study. Many reports have shown that drying of bone increases its compressive and tensile strength (14-16). It can be assumed that the uniaxial tensile test which has been applied to larger bone specimens would have yielded
213
C
60
x
50
,
B
E= Z
40
'
10
0
I
0
1
2
3
Strain
4
Number
5 x 1 0 .2
6
7
In/in
FiG 5. Stress-strain diagrams in compression showing the differences between subgroups from vital teeth. A, typical curves of wet dentin
specimens; B, typical curves of air-6ried sp~imens; and C, typical curves of desiccated specimens.
a similar result for dentin. This study indicates that dehydration has the same influence on compressive and tensile properties of dentin as on those of bone. The impact tests were conducted to verify the difference in resistance to rupture between normal wet dentin and dried dentin. In order to magnify the differences if any, only desiccated and wet specimens were used in these tests and the results of the two groups were compared. The mean value of energy absorbed by the desiccated dentin specimens appears to be lower than that of wet ones, but statistical analysis indicates no significant differences between the two subgroups. Accordingly, significant differences would not be expected to exist between normal wet dentin and dentin that has only about 9% water loss (11). Effects of Endodontic Treatment on Dentin
Another objective of this study was to conduct compression and indirect tensile tests on dentin from treated pulpless teeth. As mentioned previously, there is no possible way to maintain the original water content of the dentin from presumably drier pulpless teeth. All of the values obtained in tests termed "wet specimen" should be regarded as values of rehydrated pulpless dentin because the presumable drier dentin may regain its moisture during preparation and storage. The data obtained using rehydrated dentin of normal teeth demonstrate the influence of rehydation on strength. The results show that the ultimate strength, both compressive and tensile, of wet dentin from pulpless teeth has no significant difference from that of wet dentin from normal vital teeth. The Young's modulus and proportional limit in compression appeared to be lower, however, for pulpless dentin samples. An interesting phenomenon observed in this study is that the stress-strain curves in compression of up to 50% of the wet dentin specimens from treated pulpless teeth exhibited greater plastic deformation than did those from normal vital teeth. These wet dentin specimens could be compressed in length by 15 to 25% before failure (Fig. 7). This phenomenon was also demonstrated in
214
Huang et al.
indirect tensile tests. There seems to be no correlation with the patient's age, tooth type, nor with the time periods between the completion of endodontic treatment and the extraction. Except in these parameters from which plasticitiy can be derived, the air-dried and rehydrated pulpless dentin specimens demonstrated similar characteristics to those airdried and rehydrated normal dentin specimens in all of the tests performed. The values of compressive strength of wet dentin from normal teeth in this study range from 37,200 to 46,700 psi for crown portions and 28,500 to 49,000 psi for root portions. These data confirm those from Stanford et al. (6), Peyton et al. (17), and Craig and Peyton (10). The values of compressive strength of dentin from treated pulpless teeth ranged from 31,600 to 54,000 psi for crown portions and 25,200 to 48,900 psi for root portions. These ranges are wider than the ranges reported by Stanford et al. (6), i.e. 32,600 _-!-4,400 psi (root portion) (6) and the data of Black (5), i.e. ranging from 23,438 to 31,250 psi. There was no significant difference between the treated pulpless teeth and normal vital teeth based upon the evaluation of the ranges and the statistical tests. This corresponds to the findings of Stanford et al. (6). Black (5), however, considered the strength of pulpless teeth to be significantly lower than that of normal teeth, i.e. 37,188 psi. Black's (5) interpretation of his own data may have been arbitrary when one considers the wide range of variation of the values. Carter et al. (9) claimed that they had quantified the amount of reduction in shear strength and shear toughness of treated pulpless teeth, although their conclusion is firmer than their data would allow. Perhaps more samples should be tested to verify this finding, or perhaps, punch shear testing is more sensitive in discriminating differences in strength than are the testings utilized in the present study. It may also be hypothesized mathematically that there are relationships or linkages between the reduced shear strength found by Carter et al. (9) and the lower Young's modulus and the greater plasticity found by our study from treated pulpless teeth. Half of the dentin specimens from pulpless teeth demonstrate more obvious plastic deformation than do the samples from teeth with vital pulps (Figs. 6 and 7). This indicates that the values of those compressive strengths must be considered to be different from the true ultimate strengths, which can only be derived by subtle correction required by the inherent changes of the diameter of the specimens while under compressive loading. As a result, many of the dentin specimens from treated pulpless teeth could have lower ultimate strength than obtained in this conventional type of calculation of ultimate strength, it was impossible to do the correction for the size of the specimen applied. This phenomenon may lead to the assumption that, after the endodontic treatment, some of the teeth may become more plastic and hence have lower strength. Whether the putative dehydration process that occurred in pulpless teeth could compensate for the aforementioned increase in plastic deformation cannot be evaluated or justified from this study. Nevertheless, this phenomenon may indicate that changes occur in dentin not only in water content but also in other components after the removal of pulps. These changes, potential altered mineral or organic content subsequent to the removal of pulp or endodontic tzeatment procedures, would also affect the mechanical properties. The values of indirect tensile strength of root dentin from normal vital teeth range from 6,440 to 10,650 psi. The meas-
Journal of Endodontics 7O 6O
%
so
x
A
== +o
B
Z
20
o
' I 'I 0
2
' I t I t i r ( + f P I + ~,l 4
6
8
10
12
Strain
14
16
18
I ~ I J I ' I ' I I 20
22
24
26
28
30
Number x 10 "2 in/in
FIG 6. Stress-strain diagrams in compression showing the average curve of wet dentin specimens from normal vital teeth (A), and the average curve of wet dentin specimens from treated pulpless teeth (B).
70
._
g
60
e
50 x
E
/D
,/,';'
40 30
Z
==
2O
¢D 10
0 0
2
4
6
8
Strain
10
12
14
16
18
20
22
24
26
28
30
Number x 10 "z in/In
FIG 7. S t r e s s - s t r a i n d i a g r a m s o f h u m a n w e t dentin specimens f r o m
treated pulpless teeth in compression showing different styles of response to Ioadings. A, a style that is similar to those of vital teeth and to B-D, specimens demonstrate relatively more plastic deformation than those of vital teeth.
urements of stress occurred during loading were perpendicular to tubular directions. They correspond to the reports of Gau (t8) and Hannah (19). The values of indirect tensile strength of root dentin from treated pulpless teeth in this study range from 5,830 to 11,530 psi. In the present study, the statistical analysis reveals no significant difference between the two groups. This finding correlates with the work of Rivera and Yamauchi (20). They discussed the role of the collagen covalent intermolecular cross-link in providing dentin matrix with stability and tensile strength. They found no significant differences in cross-link content between normal and treated pulpless teeth. The data from the present study shed light on the role of water in the biomechanical behavior of human dentin. Dehydration of dentin per se does not weaken the tooth in terms of compressive and tensile strengths. In addition, if dehydration does occur in dentin in vivo after endodontic treatment, the amount of water loss can never be greater than it is in air or in a desiccant as under the conditions of this study. As a result, a significant increase of brittleness is unlikely to occur from water loss alone. Most of the strength values of dentin
Vol. 18, No. 5, May 1992
specimens from treated pulpless teeth appear to be within the strength range of normal dentin. Accordingly, the following conclusions can be drawn.
Dentin Moisture Content
215
Graduate Dentistry, Boston, MA. Dr. Schilder, is professor and chairman, Department of Endodontics, Boston University School of Graduate Dentistry. Dr. Nathanson, is professor and chairman, Department of Biomaterials, Boston University School of Graduate Dentistry. Address requests for reprints to Dr. Herbert Schilder, Department of Endodontics, Boston University School of Graduate Dentistry, 100 East Newton St., Boston, MA 02118.
CONCLUSIONS 1. Dehydration of human dentin tends to increase its Young's modulus, its proportional limit, and especially its ultimate strength under compression and tension tests. In other words, dehydration increases the stiffness and decreases the flexibility of dentin. This applies both to normal vital tooth samples and to treated pulpless tooth samples. 2. Substantial water loss of human dentin changes the fracture patterns under static compression and indirect tensile loadings, while not significantly decreasing its impact-breaking energy. 3. Wet dentin specimens from treated pulpless teeth generally show lower Young's modulus and proportional limit in compression than do those from normal vital teeth. 4. The stress-strain curves from compressive tests demonstrated that as much as 50% of the dentin specimens from treated pulpless teeth exhibit greater plastic deformation than do those from normal vital teeth. However, more sophisticated tests may be required to reexamine this observation. 5. The average values of ultimate strength, both compression and tension, for wet dentin specimens obtained in this study show no significant differences between treated pulpless teeth and normal vital teeth. 6. The results of this study indicate that dehydration does not appear to weaken dentin structure in terms of strength and toughness. 7. The data obtained in this study record only the mechanical properties of isolated dentin specimens. Additional variables should be considered in any attempt to explain the behavior of whole tooth in function. This study was part of Dr. Huang's thesis submitted in partial fulfillment of the requirements for the degree of MSD in endodontics at the Boston University School of Graduate Dentistry, Boston, MA. Dr. Huang is a part-time clinical instructor, Department of Endodontics, and DSc candidate, Department of Oral Biology, Boston University School of
References 1. Cohen S, Burns RC (eds.). Pathways of the pulp. 5th ed. St. Louis: CV Mosby, 1991 ;640-81. 2. Charbeneau GT. Planning for the restoration of non-vital teeth. J Mich Stat Dent Assoc 1960;24:75-8. 3. Tidmarsh BG. Restoration of endodontically treated posterior teeth. J Endodon 1976;2:374-5. 4. Reeh ES, Messer HH, Douglas WH. Reduction in tooth stiffness as a result of endodontic and restorative procedures. J Endodon 1989;15:512-6. 5. Black GV. An investigation of the physical characters of the human teeth in relation to their diseases, and to practical dental operations, together with the physical characters of filling materials. Dent Cosmos 1895;37:353, 469, 553,637,737. 6. Stanford JW, Weigel KV, Paffenbarger GC, Sweeney WT. Compressive properties of hard tooth tissues and some restorative rnaterials. J Am Dent Assoc 1960;60:746-56. 7. Fusayama T, Maeda T. Effect of pulpectomy on dentin hardness. J Dent Res 1969;48:452-60. 8. Lewinstein I, Grajower R. Root dentin hardness of endodontically treated teeth. J Endodon 1981 ;7:421-2. 9. Caner JM, Sorensen SE, Johnson RL, Teitelbaum RL, Levine MS. Punch shear testing of extracted vital and endodontically treated teeth. J Biomech 1983;16:841-8. 10. Craig RG, Peyton FA. Elastic and mechanical properties of human dentin. J Dent Res 1958;37:710-8. 11. Heifer AR, Melnick S, Schilder H. Determination of the moisture content of vital and pulpless teeth. Oral Surg 1972;34:661-70. 12. Bodecker CF, Lefkowitz W. Concerning the "vitality" of the calcified dental tissues. J Dent Res 1937;16:463-78. 13. Bodecker CF. Tooth condition, a factor in experimental isotope absorption. J Dent Res 1943;22:281-5. 14. Dempster WT, Liddicoat RT.Compact bone as anon-isotropicmaterial. Am J Mater 1952;91:331-62. 15. Evans FG, Lebow M. The strength of human compact bone as revealed by engineering technics. Am J Surg 1952;83:326-31. 16. Ascenz~ A, Bonucci E. The ultimate tensile strength of single osteons. Acta Anat 1964;58:160-83. 17. Peyton FA, Mahler DB, Hershenov B. Physical properties of dentin. J Dent Res 1952;31:366-70. 18. Gau DJ. The indirect tensile (diametral)strength of dentin [MSc Thesis] Indiana University, 1970. 19. Hannah CM. Tensile properties of human enamel and dentin [Abstract 113]. J Dent Res 1971 ;50:690-1. 20. Rivera E, Yamauchi M. Dentin collagen cross-links of root-filled and normal teeth [Abstract 2]. J Endodon 1990;16:190.
Printed in U.S.A.
VOL. 18, NO. 5, MAY 1992
Effects of Moisture Content and Endodontic Treatment on Some Mechanical Properties of Human Dentin Tzyy-Jou G. Huang, BMD, MSD, Herbert Schilder, DDS, and Dan Nathanson, DMD, MSD
The objective of this study was to determine whether significant differences exist between the mechanical properties of human dentin from treated pulpless teeth and dentin from normal vital teeth. Dentin specimens (n = 262) were obtained from 54 freshly extracted normal vital human teeth and 24 treated human pulpless teeth. These specimens were subjected to different experimental conditions (wet, air dried, desiccated, and rehydrated). Compression, indirect tensile, and impact tests were conducted to measure the mechanical properties of those specimens. All data obtained were analyzed with t tests. The results showed that the dehydration of dentin increases the Young's modulus, proportional limit (in compression), and especially the ultimate strength (in both compression and tension). Substantial dehydration changes the fracture characteristics of dentin specimens under static compressive and indirect tensile Ioadings. The measurements of impact-breaking energies of desiccated dentin were not found to be significantly decreased. The compressive and tensile strengths of dentin from treated pulpless teeth obtained in this study do not appear to be significantly different from those of normal dentin (p > 0.05), while the mean values of Young's modulus and proportional limit in compression tests appear to be lower. Fifty percent of the dentin specimens from treated pulpless teeth exhibit greater plastic deformation than normal dentin in compression. The results of this study do not support the theory that dehydration after endodontic treatment per se weakens dentin structure in terms of compressive and tensile strengths. Other mechanical properties of treated pulpless teeth, however, may not be the same as those of normal vital teeth.
teeth. Certain mechanical terminologies have been frequently applied by many authors to describe the physical condition of teeth that have had root canal treatment, such as increased "brittleness," "friability," and "fragility;" reduced "resiliency," "elasticity," and "strength" (1-3). The reduction in tooth structure and the effect of dehydration on the dentinal tubules are widely considered to be the main reasons associated with increased weakness and brittleness in pulpless teeth (3). The recent study by Reeh et al. (4), however, revealed that endodontic procedures reduce the tooth stiffness by only 5%. In addition, the role that water plays in the biomechanical behavior of teeth has not been well understood nor fully investigated. As early as 1895, Black (5) concluded that the dentin of human pulpless teeth had less crushing strength than normal teeth. Later investigations (6-8) did not support Black's finding until Carter et al. (9) in 1983 showed the reduction of shear strength and shear toughness of endodontically treated teeth. In this study, an effort was made to reaffirm those studies, and, especially, to realize the effects of moisture content on the biomechanical properties in human teeth. The complexity of the clinical problems, the myriad of physical factors effecting clinical behavior of treated pulpless teeth, as well as the intricacies of analyzing and testing their interrelationships posed a dilemma. Accordingly, the decision was made to divide out several parameters of the problems and to limit the project to measurements of these factors alone. Extensive mechanical tests on human dentin of treated pulpless and vital teeth were performed. Moreover, the definitive mechanical terminologies of material testing science were applied to interpret the experimental data obtained. Potential difficulties due to variations and experimental errors that might occur in such a study were noted, evaluated, and avoided as much as possible. MATERIALS AND METHODS Teeth Collection
Two groups of freshly extracted human permanent teeth were collected. Group one consisted of 54 teeth with minimal caries that were vital at the time of extraction and group two
It is generally believed that endodontic treatment renders teeth weaker and more subject to fracture than normal vital 209
210
Huang et al.
consisted of 24 teeth that had been endodontically treated at least 1 yr before extraction. Both groups comprised mixtures of anterior and posterior teeth with donor patients' age varying from 20 to 88 yr. The teeth were stored in separate vials containing a special storage solution (composed of 4 mM Na2HPO4, 2 mr~ CaC12, 20 mM KC1, 18 mra NaCI, and 1% NAN3). This solution was used to prevent either dehydration or decalcification of the specimens during storage. Specimens preparation was performed as quickly as possible in each case, never more than 7 days after the tooth was extracted, with a steady flow of water in the cutting field.
Preparation of the Dentin Specimen For compression and indirect tensile tests, cylindrical specimens of dentin were cut out of the coronal and/or radicular portion of teeth from both groups (Fig. 1). This was accomplished by using an Isomet 77-1180 low-speed saw (Buehler Ltd., Evanston, IL) and trepan burs (Brasseler USA, Inc., Savannah, GA) attached to a Unimat-3 miniature lathe (Hobby Product Co., Columbus, OH). The sizes of the specimens for compression tests were from 0.13967 to 0.07819 inches (3.54 to 2.00 mm) in length and 0.06340 to 0.03992 inches (1.60 to 1.00 mm) in diameter (length to diameter ratio of 1.7:2.2). The sizes for indirect tensile tests were 0.04730 to 0.07680 inches (1.20 to 1.95 mm) in diameter and 0.03940 to 0.06450 inches (1.00 to 1.64 mm) in length (diameter to length ratio 1.5:1). The two ends of each cylindrical specimen were trimmed parallel and smooth. For impact tests, specimens were obtained only from coronal dentin of molar teeth from group one and made into a rectangular shape of 0.0630 inches x 0.0630 inches × 0.394 inches (1.60 x 1.60 x 10.00 mm). No notches were made on such small impact test specimens because it was not easy to prepare the notches consistently with the same geometry for all of them. Each prepared specimen was checked for defects under a dissecting microscope at 30 times magnification and any specimens with visible cracks or flaws were discarded. No
Journal of Endodontics
specimens that involve decay or structures such as cementum were used, i.e. only specimens with sound dentin were tested and these rules applied to both groups of teeth. Each specimen was then stored in a labeled plastic tube. Storage solution was placed only in the tubes for wet specimens. Specimens for compression and indirect tensile tests were divided into subgroups and received different treatments: remained wet, air dried at room temperature for 3 days, desiccated over CaSO4 for 2 days at room temperature, or rehydrated previously air-dried specimens in the storage solution for approximately 2 h. Specimens for impact tests were studied only wet and desiccated. In all tests, the wet and rehydrated specimens were tested wet, and the dried specimens were tested dry.
Compression Test An Instron 1331 testing machine (Instron Corp., Canton, MA) of 2200-1b capacity load cell was used to apply loads to specimens. Each specimen was tested between two superhardened steel platens. Due to the small size of the specimens, the strain gauge (1.969 inches) had to be fastened to the platens. Immediately before testing, the length and diameter of each specimen were measured by a micrometer caliper (L. S. Starrett Co., Athol, MA) accurate to 0.000039 inch (1 #m) and recorded. A thin layer of grease was painted in the center of the two platens' surfaces to reduce friction. Each specimen was compressed until it failed at the speed of 0.0015 to 0.0030 inches/min. A stress-strain curve was drawn by the X-Y chart recorder and the characteristics of deformation of each specimen were observed with a magnifying lens at x3 during the compression. After each specimen failed, the peak load (lb) and consumed time (s) were recorded. In addition, the mode of the fractured specimen was inspected under the dissecting microscope and recorded. The ultimate strength was calculated by the formula: Stress = maximum load/area The proportional limit was measured on the stress-strain curve. The modulus of elasticity (Young's modulus, or E) was calculated by the formula: E = stress/strain
"'" ~'1 Silo !lie...... .... • -4........ J f i I
FIG 1. Diagrams indicating the locations (crown and/or root) from which the dentin specimens were taken. The horizontal lines represent the slicing planes.
Stanford et at. (6) and Craig and Peyton (10) previously described potential experimental errors that should be considered in small size specimen testing of these types. One factor that would create errors (especially in the Young's modulus) in our test might be the gap effect due to nonparallelism of the two ends of the specimens. Stanford (6) stated that there is no known equation for the correction of this factor, since it involves probability of how the platens contact the ends of the specimens. Craig and Peyton (10) found that the length to diameter ratio appeared to lower the Young's modulus for the ratio between 1 and 0.5. In our study, two materials, aluminum alloy 2024-T4 and acrylic cast, were also made into similar shapes and dimensions as the dentin specimens to receive compression tests. Comparison was then made to the previously published data of those materials to verify the amount of error that may occur in our tests.
Dentin Moisture Content
Vol. 18, No. 5, May 1992
211
TABLE 1. Compressive properties of human dentin in different conditions*
Condition of Specimen Normal vital Wet Air-dried Desiccated
Rehydrated Treated pulpless Wet Air-dried Rehydrated
Location of Specimen
No. of Specimens
Crown Root Crown Root Crown Root Root Crown Root Root Root
Young's Modulus (E) (106 psi)
Proportional Limit (psi)
Ultimate Strength (psi)
Mean
SD
Mean
SD
Mean
SD
11 17 7 6 4 6 6
1.82 (2.17)t 1.86 (2.21)t 2.00 (2.39)t 2.27 (2.72)t 2.17 (2.60)t 2.47 (2.96)t 1.82 (2.17)t
0.30 0.27 0.29 0.27 0.46 0.38 0.29
22,500 22,300
4,700 3,400
42,800 36,400
3,000 4,400 I
22,900
3,000
56,700
5,000 t
29,000
4,200
59,700
3,40011
29,500 34,400 24,600
5,700 7,700 6,000
70,100 69,700 39,900
7,800 7,700 3,300
8 34 4 3
1.27 (1.49)t 1.49 (1.76)t 2.79 (3.36)t --
0.50 0.49 1.05
17,600 19,600 24,800 18,100
6,400 5,400 5,200 10,200
45,100 37,600 50,900 28,300
7,200 5,600 5,300 7,200
* The difference between any two means joined by the same thick vertical lines is not significant (t test, p > 0.05), and the difference between those joined by the same thin vertical lines is significant (t test, p < 0.05). t Mean values calibrated using the linear interpolation method based on the calibration data in Table 2. The acrylic cast accepted Young's modulus (E), 0.40 x 10e psi (midpoint of the range 0.35 to 0.45), was chosen to calibrate.
Indirect Tensile Test An Instron model 4202 testing machine with a 200-1b capacity load cell was used for this test. A layer of aluminum foil with a thickness of 0.00059 inches was taped on the two plat~tls as padding material. Each specimen was placed and oriented between the two platens and the load was applied at 0.003 inches/min cross-head speed until failure occurred. All of the specimens were oriented between the platens in such a way that the applied loads were approximately parallel to the direction of dentinal tubules. This ensured that the tensile stress created was approximately at right angles to the tubules even though the tubules do not run in regular directions. Continuously during the testing, recordings were made of the applied load versus time and recorded on load-time charts. A magnifying lens was used to observe changes of the specimens during load application. After failure occurred, the maximum load was recorded and each fractured specimen was examined under the dissecting microscope. The tensile stress was calculated by the equation: Stress = 2 x maximum load/Tr × diameter x thickness
Impact Test Since specimens of 0.0630 inches x 0.0630 inches x 0.394 inches were the largest geometrically regular dentin blocks that could be obtained from human teeth, an impact testing machine with small capacity had to be used. A Charpy impact machine model CIM-1 (Manlab Inc. (Physmet), Cambridge, MA) with 12-inch-lb maximum energy capacity was the smallest one that could be found, but the anvil was still too large to accommodate the small rectangular dentin specimen. A specially built anvil was made according to the specification of ASTM E23 (American Society for Testing and Materials, group E). This converted the Charpy impact machine model into an Izod-type impact machine. Both Charpy and Izodtype have equal standing in the scientific community for this type of test. The specimens were positioned and fixed to the
anvil and the pendulum was activated. The pendulum fractured the specimens into two parts and the energy consumed was recorded. Some of them, which were struck into several pieces, showed much higher energy readings and were not counted. RESULTS
Compressive Properties of Dentin The results and the mean values of the compressive tests appear in Table 1. The statistical analysis (t test) indicates that dehydration of dentin tends to increase the Young's modulus, proportional limit, and especially the ultimate strength. Figure 5 depicts the stress-strain relationships of those specimens under different degrees of dehydration. The slopes of the linear portion of the curves (Young's modulus in graphic form) become steeper, and the end points of the curves (ultimate stresses when failures occur) become much higher as the result of progressive dehydration. The mean values of ultimate strength of wet dentin specimens show no significant difference between treated pulpless teeth and normal vital teeth (p > 0.05). However, the Young's modulus and the proportional limit of wet dentin specimens from pulpless teeth are significantly lower than those from normal vital teeth (p < 0.05). Fifty percent of the specimens from pulpless teeth demonstrated greater plastic deformation than did those from vital teeth. These observations are well characterized in Figs. 6 and 7. Wet conditions only are compared between two groups of tooth samples, since specimens treated with other conditions may have multiple variables. The obtained values in this study, Young's modulus of aluminum alloy 2024-T4 and standard cast acrylic blocks, were slightly lower than the published values as may be observed in Table 2. The Young's moduli in Table 1, accordingly, were corrected through this calibration. For dentin specimens from vital teeth, the fracture patterns of most of the wet and air-dried specimens are shown in Fig.
Joumal of Endodontics
Huang et al.
212
TABLE 2. Calibration data for determination of Young's modulus (E)
Material
Observed E (108 p s i )
Accepted E* (10s psi)
8.65
10.60
0.39
0.35-0.45
Aluminum alloy 2024-T4 Acrylic cast
• Source: Measurement group, Raleigh, NC,
K\\\\\",3
22
(a)
(b)
(c)
FIG 3. Selected fracture patterns of dentin specimens under indirect tensile Ioadings. a and b, Wet and/or air-dried dentin specimens, c, A few dentin specimens from treated pulpless teeth demonstrating excessive deformation. TABLE 4. Impact test of human dentin (normal vital)
r~
(a)
(b)
(c)
(d)
FIG 2. Selected fracture patterns of dentin specimens under compressive Ioadings. a and b, Wet and/or air-dried dentin specimens from normal vital teeth, c, A desiccated dentin specimen from normal vital teeth. The sma/I arrows indicate that dentin fragments ejected in all directions when it fractured, d, The bulging deformation under pressure of some wet dentin specimens from treated pulpless teeth.
Condition of Specimen
No. of Specimens
Wet Desiccated
13 14
Energy (inch-lb) Mean SD 0.046 0.035
0.019 0.013
TABLE 3. Indirect tensile strength of human root dentin in different conditions*
Condition of Specimen Normal vital Wet Air-dried Desiccated Treated pulpless Wet Air-dried Rehydrated
No. of Specimens
Ultimate Strength (psi) Mean
SD
30 17 11
8,650 12,290 16,900
1,230 1,5301 2,660
FIG 4. Drawings indicating a dentin specimen for impact test and a few ruptured specimens after impact Ioadings.
42 7 5
8,640 10,270 8,170
1,420 I 1,5401 1,250
dentin specimens between treated pulpless teeth and normal vital teeth (p > 0.05). For dentin specimens from vital teeth, most wet specimens failed by diametral cleft formation with or without secondary cracks. The fracture modes of air-dried specimens were mainly triple-cleft formation, and all except one of the desiccated specimens appeared as triple-cleft fractures (Fig. 3, a and b). Specimens from treated pulpless teeth have similar fracture characteristics as do samples from vital teeth. A few specimens, however, demonstrated excessive ductility which rendered them inappropriate to receive this test. Their data were not included (Fig. 3c).
* The difference between any two means joined by the same thick vertical lines is not significant (t test, p > 0.05), and the difference between those joined by the same thin vertical lines is significant (t test, p < 0.05).
2, a and b. They separated or cracked into two to three fragments. The desiccated specimens shattered into many pieces at the moment they failed (Fig. 2c). The average dentin specimens from treated pulpless teeth had similar fracture characteristics as did those of vital teeth. An exception was that those demonstrating more bulging during stress developed more minor cracks (Fig. 2d). Indirect Tensile Test The results of the indirect tensile tests are presented in Table 3. Statistical analysis indicates that dehydration increases dentin tensile strength dramatically. There is, however, no significant difference in ultimate tensile strength of wet
Impact Test In impact tests, the fracture modes showed no difference between desiccated and wet dentin. Table 4 shows that there is no significant difference between the two means (p > 0.05), although the mean value of the desiccated dentin appears lower than the wet one. The ruptured ends of the specimens demonstrate various styles of fractured surfaces (Fig. 4).
Vol. 18, No. 5, May 1992
Dentin Moisture Content
DISCUSSION
80 70
Effects of Moisture Content on Dentin
% The loss of moisture content, 9% by weight, of pulpless teeth was quantified by Helfer et al. (11). Many authors have implied that dehydration increases brittleness and makes teeth more susceptible to fracture. Studies corresponding to the findings of Helfer et al. (11) were reported as early as 1937 and 1943. Bodecker and his colleague (12, 13) used a dye penetration technique on both animal and human material to study water permeability of pulpless teeth. Their observation was that teeth which have been devitalized for a long period of time, at least 1 yr, demonstrate little permeability. They explained this by the existence of gas in the dentinal tubules which developed subsequent to the drying out of dental lymph some time after the removal of pulps. This gas, they suggested, made pulpless teeth highly impermeable to fluids. However, considering that pulpless teeth continue to exist in a water-based environment, saliva in the mouth and blood and tissue fluid surrounding the root, it has never been fully explained why dentin should dry out. Cementum is permeable without tubules and considerable evidence indicates enamel permeability to water even though enamel is a much denser calcified tissue. There is little reason to consider treated pulpless teeth in situ as "dried." In addition, the presence of the gas in dentinal tubules has never been verified. Regardless of whether the conclusions of Heifer et al. (11) and Bodecker (12, 13) were true or not, the effect of the dryness per se on the mechanical properties of tooth dentin was determined to be of sufficient interest to be investigated. It should be noted that it is not possible to utilize a dentin specimen from either a vital or pulpless tooth and maintain its original moisture content. During the specimen preparation, constant water flow had to be applied to avoid overheating the cutting field. Also the storage of the specimen in the storage solution would certainly change the water content. Specimens from teeth presumed to be dehydrated by reasons of pulplessness were required to be redehydrated in the laboratory after specimen preparation. It was observed in this study that different degress of dehydration of dentin were established after the preparation of the specimens. The results indicate that the dehydration of dentin tends to increase the Young's modulus, proportional limit, and especially the ultimate strength (both compressive and tensile). According to the stress-strain diagrams, the dried dentin specimens were stiffer than those of wet ones (Fig. 5). The fracture patterns of dried dentin specimens demonstrated different characteristics from the wet ones. Some air-dried dentin specimens were rehydrated in the storage solution for approximately 2 h before the compression and indirect tensile tests. They demonstrated similar characteristics to normal wet dentin. In terms of toughness, area under stress-strain curve (Fig. 5), there is no indication that the toughness of dried dentin is less than that of normal wet dentin. The energy absorbed by a material before rupture, i.e. toughness, is sometimes more accurately obtained from the stress-strain curve of the uniaxial tension test which was impractical to perform with the small dentin sample in this study. Many reports have shown that drying of bone increases its compressive and tensile strength (14-16). It can be assumed that the uniaxial tensile test which has been applied to larger bone specimens would have yielded
213
C
60
x
50
,
B
E= Z
40
'
10
0
I
0
1
2
3
Strain
4
Number
5 x 1 0 .2
6
7
In/in
FiG 5. Stress-strain diagrams in compression showing the differences between subgroups from vital teeth. A, typical curves of wet dentin
specimens; B, typical curves of air-6ried sp~imens; and C, typical curves of desiccated specimens.
a similar result for dentin. This study indicates that dehydration has the same influence on compressive and tensile properties of dentin as on those of bone. The impact tests were conducted to verify the difference in resistance to rupture between normal wet dentin and dried dentin. In order to magnify the differences if any, only desiccated and wet specimens were used in these tests and the results of the two groups were compared. The mean value of energy absorbed by the desiccated dentin specimens appears to be lower than that of wet ones, but statistical analysis indicates no significant differences between the two subgroups. Accordingly, significant differences would not be expected to exist between normal wet dentin and dentin that has only about 9% water loss (11). Effects of Endodontic Treatment on Dentin
Another objective of this study was to conduct compression and indirect tensile tests on dentin from treated pulpless teeth. As mentioned previously, there is no possible way to maintain the original water content of the dentin from presumably drier pulpless teeth. All of the values obtained in tests termed "wet specimen" should be regarded as values of rehydrated pulpless dentin because the presumable drier dentin may regain its moisture during preparation and storage. The data obtained using rehydrated dentin of normal teeth demonstrate the influence of rehydation on strength. The results show that the ultimate strength, both compressive and tensile, of wet dentin from pulpless teeth has no significant difference from that of wet dentin from normal vital teeth. The Young's modulus and proportional limit in compression appeared to be lower, however, for pulpless dentin samples. An interesting phenomenon observed in this study is that the stress-strain curves in compression of up to 50% of the wet dentin specimens from treated pulpless teeth exhibited greater plastic deformation than did those from normal vital teeth. These wet dentin specimens could be compressed in length by 15 to 25% before failure (Fig. 7). This phenomenon was also demonstrated in
214
Huang et al.
indirect tensile tests. There seems to be no correlation with the patient's age, tooth type, nor with the time periods between the completion of endodontic treatment and the extraction. Except in these parameters from which plasticitiy can be derived, the air-dried and rehydrated pulpless dentin specimens demonstrated similar characteristics to those airdried and rehydrated normal dentin specimens in all of the tests performed. The values of compressive strength of wet dentin from normal teeth in this study range from 37,200 to 46,700 psi for crown portions and 28,500 to 49,000 psi for root portions. These data confirm those from Stanford et al. (6), Peyton et al. (17), and Craig and Peyton (10). The values of compressive strength of dentin from treated pulpless teeth ranged from 31,600 to 54,000 psi for crown portions and 25,200 to 48,900 psi for root portions. These ranges are wider than the ranges reported by Stanford et al. (6), i.e. 32,600 _-!-4,400 psi (root portion) (6) and the data of Black (5), i.e. ranging from 23,438 to 31,250 psi. There was no significant difference between the treated pulpless teeth and normal vital teeth based upon the evaluation of the ranges and the statistical tests. This corresponds to the findings of Stanford et al. (6). Black (5), however, considered the strength of pulpless teeth to be significantly lower than that of normal teeth, i.e. 37,188 psi. Black's (5) interpretation of his own data may have been arbitrary when one considers the wide range of variation of the values. Carter et al. (9) claimed that they had quantified the amount of reduction in shear strength and shear toughness of treated pulpless teeth, although their conclusion is firmer than their data would allow. Perhaps more samples should be tested to verify this finding, or perhaps, punch shear testing is more sensitive in discriminating differences in strength than are the testings utilized in the present study. It may also be hypothesized mathematically that there are relationships or linkages between the reduced shear strength found by Carter et al. (9) and the lower Young's modulus and the greater plasticity found by our study from treated pulpless teeth. Half of the dentin specimens from pulpless teeth demonstrate more obvious plastic deformation than do the samples from teeth with vital pulps (Figs. 6 and 7). This indicates that the values of those compressive strengths must be considered to be different from the true ultimate strengths, which can only be derived by subtle correction required by the inherent changes of the diameter of the specimens while under compressive loading. As a result, many of the dentin specimens from treated pulpless teeth could have lower ultimate strength than obtained in this conventional type of calculation of ultimate strength, it was impossible to do the correction for the size of the specimen applied. This phenomenon may lead to the assumption that, after the endodontic treatment, some of the teeth may become more plastic and hence have lower strength. Whether the putative dehydration process that occurred in pulpless teeth could compensate for the aforementioned increase in plastic deformation cannot be evaluated or justified from this study. Nevertheless, this phenomenon may indicate that changes occur in dentin not only in water content but also in other components after the removal of pulps. These changes, potential altered mineral or organic content subsequent to the removal of pulp or endodontic tzeatment procedures, would also affect the mechanical properties. The values of indirect tensile strength of root dentin from normal vital teeth range from 6,440 to 10,650 psi. The meas-
Journal of Endodontics 7O 6O
%
so
x
A
== +o
B
Z
20
o
' I 'I 0
2
' I t I t i r ( + f P I + ~,l 4
6
8
10
12
Strain
14
16
18
I ~ I J I ' I ' I I 20
22
24
26
28
30
Number x 10 "2 in/in
FIG 6. Stress-strain diagrams in compression showing the average curve of wet dentin specimens from normal vital teeth (A), and the average curve of wet dentin specimens from treated pulpless teeth (B).
70
._
g
60
e
50 x
E
/D
,/,';'
40 30
Z
==
2O
¢D 10
0 0
2
4
6
8
Strain
10
12
14
16
18
20
22
24
26
28
30
Number x 10 "z in/In
FIG 7. S t r e s s - s t r a i n d i a g r a m s o f h u m a n w e t dentin specimens f r o m
treated pulpless teeth in compression showing different styles of response to Ioadings. A, a style that is similar to those of vital teeth and to B-D, specimens demonstrate relatively more plastic deformation than those of vital teeth.
urements of stress occurred during loading were perpendicular to tubular directions. They correspond to the reports of Gau (t8) and Hannah (19). The values of indirect tensile strength of root dentin from treated pulpless teeth in this study range from 5,830 to 11,530 psi. In the present study, the statistical analysis reveals no significant difference between the two groups. This finding correlates with the work of Rivera and Yamauchi (20). They discussed the role of the collagen covalent intermolecular cross-link in providing dentin matrix with stability and tensile strength. They found no significant differences in cross-link content between normal and treated pulpless teeth. The data from the present study shed light on the role of water in the biomechanical behavior of human dentin. Dehydration of dentin per se does not weaken the tooth in terms of compressive and tensile strengths. In addition, if dehydration does occur in dentin in vivo after endodontic treatment, the amount of water loss can never be greater than it is in air or in a desiccant as under the conditions of this study. As a result, a significant increase of brittleness is unlikely to occur from water loss alone. Most of the strength values of dentin
Vol. 18, No. 5, May 1992
specimens from treated pulpless teeth appear to be within the strength range of normal dentin. Accordingly, the following conclusions can be drawn.
Dentin Moisture Content
215
Graduate Dentistry, Boston, MA. Dr. Schilder, is professor and chairman, Department of Endodontics, Boston University School of Graduate Dentistry. Dr. Nathanson, is professor and chairman, Department of Biomaterials, Boston University School of Graduate Dentistry. Address requests for reprints to Dr. Herbert Schilder, Department of Endodontics, Boston University School of Graduate Dentistry, 100 East Newton St., Boston, MA 02118.
CONCLUSIONS 1. Dehydration of human dentin tends to increase its Young's modulus, its proportional limit, and especially its ultimate strength under compression and tension tests. In other words, dehydration increases the stiffness and decreases the flexibility of dentin. This applies both to normal vital tooth samples and to treated pulpless tooth samples. 2. Substantial water loss of human dentin changes the fracture patterns under static compression and indirect tensile loadings, while not significantly decreasing its impact-breaking energy. 3. Wet dentin specimens from treated pulpless teeth generally show lower Young's modulus and proportional limit in compression than do those from normal vital teeth. 4. The stress-strain curves from compressive tests demonstrated that as much as 50% of the dentin specimens from treated pulpless teeth exhibit greater plastic deformation than do those from normal vital teeth. However, more sophisticated tests may be required to reexamine this observation. 5. The average values of ultimate strength, both compression and tension, for wet dentin specimens obtained in this study show no significant differences between treated pulpless teeth and normal vital teeth. 6. The results of this study indicate that dehydration does not appear to weaken dentin structure in terms of strength and toughness. 7. The data obtained in this study record only the mechanical properties of isolated dentin specimens. Additional variables should be considered in any attempt to explain the behavior of whole tooth in function. This study was part of Dr. Huang's thesis submitted in partial fulfillment of the requirements for the degree of MSD in endodontics at the Boston University School of Graduate Dentistry, Boston, MA. Dr. Huang is a part-time clinical instructor, Department of Endodontics, and DSc candidate, Department of Oral Biology, Boston University School of
References 1. Cohen S, Burns RC (eds.). Pathways of the pulp. 5th ed. St. Louis: CV Mosby, 1991 ;640-81. 2. Charbeneau GT. Planning for the restoration of non-vital teeth. J Mich Stat Dent Assoc 1960;24:75-8. 3. Tidmarsh BG. Restoration of endodontically treated posterior teeth. J Endodon 1976;2:374-5. 4. Reeh ES, Messer HH, Douglas WH. Reduction in tooth stiffness as a result of endodontic and restorative procedures. J Endodon 1989;15:512-6. 5. Black GV. An investigation of the physical characters of the human teeth in relation to their diseases, and to practical dental operations, together with the physical characters of filling materials. Dent Cosmos 1895;37:353, 469, 553,637,737. 6. Stanford JW, Weigel KV, Paffenbarger GC, Sweeney WT. Compressive properties of hard tooth tissues and some restorative rnaterials. J Am Dent Assoc 1960;60:746-56. 7. Fusayama T, Maeda T. Effect of pulpectomy on dentin hardness. J Dent Res 1969;48:452-60. 8. Lewinstein I, Grajower R. Root dentin hardness of endodontically treated teeth. J Endodon 1981 ;7:421-2. 9. Caner JM, Sorensen SE, Johnson RL, Teitelbaum RL, Levine MS. Punch shear testing of extracted vital and endodontically treated teeth. J Biomech 1983;16:841-8. 10. Craig RG, Peyton FA. Elastic and mechanical properties of human dentin. J Dent Res 1958;37:710-8. 11. Heifer AR, Melnick S, Schilder H. Determination of the moisture content of vital and pulpless teeth. Oral Surg 1972;34:661-70. 12. Bodecker CF, Lefkowitz W. Concerning the "vitality" of the calcified dental tissues. J Dent Res 1937;16:463-78. 13. Bodecker CF. Tooth condition, a factor in experimental isotope absorption. J Dent Res 1943;22:281-5. 14. Dempster WT, Liddicoat RT.Compact bone as anon-isotropicmaterial. Am J Mater 1952;91:331-62. 15. Evans FG, Lebow M. The strength of human compact bone as revealed by engineering technics. Am J Surg 1952;83:326-31. 16. Ascenz~ A, Bonucci E. The ultimate tensile strength of single osteons. Acta Anat 1964;58:160-83. 17. Peyton FA, Mahler DB, Hershenov B. Physical properties of dentin. J Dent Res 1952;31:366-70. 18. Gau DJ. The indirect tensile (diametral)strength of dentin [MSc Thesis] Indiana University, 1970. 19. Hannah CM. Tensile properties of human enamel and dentin [Abstract 113]. J Dent Res 1971 ;50:690-1. 20. Rivera E, Yamauchi M. Dentin collagen cross-links of root-filled and normal teeth [Abstract 2]. J Endodon 1990;16:190.
