- Email: [email protected]
Evaluation of Fiber-Composite Laminate in the Restoration of Immature, Nonvital Maxillary Central Incisors
JOURNAL OF ENDODONTICS Copyright © 2001 by The American Association of Endodontists
Printed in U.S.A. VOL. 27, NO. 1, JANUARY 2001
Evaluation of Fiber-Composite Laminate in the Restoration of Immature, Nonvital Maxillary Central Incisors Jeffrey R. Pene, DDS, MSD, Jack I. Nicholls, PhD, and Gerald W. Harrington, DDS, MSD
According to Cvek (5), the prognosis for nonvital, immature incisors is dependent, in large part, on the stage of root development at the time of devitalization. The results he reported for the occurrence of cervical fractures 4 yr after filling the canal included 77% in stage 1 (wide, divergent apical opening, root length less than 1⁄2 final length); 53% in stage 2 (wide, divergent apical opening, root length 1⁄2 final length); 43% in stage 3 (wide, divergent apical opening, root length 2⁄3 of final length); and 28% in stage 4 (open apical foramen, nearly completed root length). He classified the stages of root development by radiographic estimation of the width of the apical foramen and the length of the root. Recent studies have proposed the use of acid-etched composites to strengthen endodontically treated immature teeth (6). Combined with various prefabricated posts this method has been advocated for usage during and after apexification (7). It has been postulated that one of the apparent shortcomings of composite material used in this manner is the propagation of microcracks resulting from inherent porosities and flexure in the composite material leading to subsequent failure of the restoration (8). Karna (8) suggested the use of fiber composite laminate to prevent propagation of microcracks and thereby reinforce the composite material, making it more resistant to fracture. The purpose of this study was to develop a model for immature maxillary central incisors and to evaluate whether a ribbon-reinforced composite can increase the fracture resistance of the model over composite alone.
The purpose of this study was to evaluate the use of fiber-composite laminate, a reinforcement ribbon embedded throughout a composite restoration, to reinforce immature maxillary central incisors. Twenty-six mature maxillary central incisors were prepared a minimum of 3 mm below the facial cementoenamel junction to simulate immature nonvital teeth. They were separated into 3 groups: group 1 was unfilled and served as a control; group 2 was filled to the depth of the preparation with composite; and group 3 was filled with composite and Connect Reinforcement Ribbon. The specimens were subjected to class I loading in an Instron Testing Machine until catastrophic failure occurred. The results indicate a highly significant difference between the groups (p < 0.003). Group 1 fractured at an average load of 31.08 kg, group 2 at 51.00 kg, and group 3 at 37.93 kg. These findings suggest that composite alone increases fracture resistance of the immature tooth model more than composite with Reinforcement Ribbon.
Previous research indicates that the most common site of dental impact injuries in the developing dentition is the maxillary anterior teeth (1, 2). Frequently these injuries can lead to necrosis of the pulp and subsequent arrested development of the tooth. The clinician is thus faced with the difficult task of treating these teeth endodontically in addition to restoring teeth with thin dentinal walls. Apexification treatment of these immature teeth has been shown to be predictable with a high degree of success (3–5). The thin dentinal walls at the cementoenamel junction, however, make them prone to fracture from secondary injuries (i.e. mastication or minor trauma), often leaving then nonrestorable (5). Studies have shown that, despite successful endodontic treatment, ⬃28 to 77% of these teeth— depending on the stage of root development—will fracture during or after treatment, leaving many clinicians to view the procedure as having a poor prognosis for teeth arrested in the early stages of development (5).
MATERIALS AND METHODS Twenty-six intact, extracted maxillary central incisors were used in this study. The teeth were stored in water except during restoration and experimental testing. Craze lines in enamel were noted using a dissecting microscope, and any teeth with fracture lines extending into dentin were not used. Additional teeth were used as pilot specimens to establish the experimental model. Immature Tooth Model Preparation Teeth were lingually accessed using a #4 round bur and a tapered diamond bur to a standardized dimension of 3 mm mesiodistally by 3 mm incisogingivally. For the purpose of establishing the experimental model the mesiodistal dimension of the access was ex18
Vol. 27, No. 1, January 2001
Restoration of Immature Nonvital Incisors
19
FIG 2. Enlargement of the canal in the drill press. FIG 1. Surveyor used to align the canal perpendicular to the base of the Duralay-lined acrylic block.
tended to 3 mm, allowing clearance for the subsequent engineering twist drill. This prevented drill contact with enamel and thereby reduced the risk of fracture or craze line formation. All teeth had a notch placed in the incisal edge to allow absolute straight-line access and were then instrumented to 3 mm below the facial cementoenamel junction with 1 to 6 Gates-Glidden drills. To simulate an immature tooth, a surveyor pin was placed in the canal to the prepared depth, and the entire tooth set into a Duralay (Reliance Dental Manufacturing Co., Worth, IL)-lined acrylic block. Thus the surveyor pin and the long axis of the canal were perpendicular to the top of the acrylic block in the surveyor (Fig. 1). After the Duralay had set the acrylic block was placed in a drill press and the canal enlarged to 3 mm in diameter using an engineering twist drill (Fig. 2). The preparation was extended a minimum of 3 mm below the facial cementoenamel junction. The 3 mm diameter was chosen to approximate stage 3 of development where the canal is roughly one-half the entire mesiodistal dimension (5) (Figs. 3 and 4). After preparation, the teeth were randomly assigned to 1 of 3 groups, with 6 teeth in the unrestored control group and 10 teeth in each of the two experimental groups. Group 1: Unrestored Positive Control The access cavity and canal preparation were unrestored. Group 2: Dentin-Bonded Composite Enamel was etched with 35% phosphoric acid for 30 s, then rinsed thoroughly with water. Tenure Conditioner (Den-Mat Corp., Santa Maria, CA) was applied uniformly with an applicator brush to the
FIG 3. Radiographs comparing an immature tooth to a prepared immature tooth model.
internal walls of the entire canal and access. After 15 s this layer was thoroughly washed and dried. A dual-cure dentin bonding agent (Tenure A & B, Den-Mat Corp.) was placed on the internal surfaces of the canal and chamber and smoothed to an even thin layer with the applicator brush. This layer was then light-cured for 30 s. Using a small composite syringe tip in a Centrix injection gun (Dentsply Caulk, Milford, DE), Prisma VLC Hybrid Composite (Dentsply Caulk, Milford, DE) was injected into the canal space, completely filling it to the level of the cavo-surface margin. Excess composite was removed with a plastic instrument and the composite light-cured for 2 min.
20
Pene et al.
Journal of Endodontics
FIG 4. Proximal radiographs comparing an immature tooth to a prepared immature tooth model.
Group 3: Ribbon-Reinforced Dentin-Bonded Composite The procedural steps for group 3 were identical to those used for group 2. After the composite was injected into the canal space, a 2-mm-wide piece of Reinforcement Ribbon (RR) (Connect, Kerr USA, Orange, CA) was cut to a length equal to twice the depth of each canal preparation. The RR was coated with the bonding agent and excess removed. The RR material was handled with cotton pliers only. Using a #9 Schilder plugger, the RR strip was shaped into a “V” and inserted to the depth of the canal with the width in a faciolingual direction. Some of the composite was displaced as the ends of the “V” were pushed even with the cavo-surface margin. A second piece of RR, which had been prepared in the same manner, was then shaped into a “V” and inserted into the canal to the same depth. The width of the second “V” was placed in a mesiodistal direction and thus perpendicular to the first strip. The RR was adapted close to the canal walls with the plugger each time a new piece was placed. The ends of the RR were then packed below the cavo-surface margin, and additional hybrid composite was used to fill the void resulting from RR placement. Excess composite was removed with a plastic instrument and the restoration light cured for 2 min.
Fracture Strength Evaluation An Instron Testing Machine (model TTMBL; Instron Corp., Canton, MA) was used to apply a controlled load to each specimen at a crosshead speed of 5.0 mm/min. To ensure a fixed position for this load application, composite was bonded to the lingual surface of each specimen 2 mm below the incisal edge. The composite was leveled to provide an even point of contact that would remain stable during the loading trial. Each specimen block was fixed in a jig that ensured a loading angle of 130 degrees to the long axis of the tooth. This angle was chosen because it simulates the average angle of contact between maxillary and mandibular incisors in a class I occlusion (9). The force was then applied in a linguolabial vector. The magnitude of the load was continuously recorded by the Instron machine until catastrophic failure occurred.
FIG 5. Radiograph of representative specimen showing fracture at the cervical area of the tooth.
RESULTS Of the 26 specimens tested, all teeth showed either horizontal or oblique fractures that went through the cervical area of the root (Fig. 5). At the point where the fracture extended through the prepared canal space, the fracture was also through the restorative materials in groups 2 and 3 (Fig. 6). The results of the strength testing trials are summarized by group in Table 1. The results of a one-way analysis of variance test (ANOVA), shown in Table 2, indicate a significant difference exists between the group means (p ⱕ 0.003). The results of the Student-NewmanKeuls test, (shown in Table 3) indicate three significant subsets (p ⱕ 0.003). In this table each test group defines a separate statistical subset, with the control group having the lowest fracture values, and the composite-alone group having the highest fracture values. These results suggest that composite alone increased fracture resistance of immature teeth greater than ribbon-reinforced composite. DISCUSSION In the present study, stage 3 root development was selected for the model because it is the stage at which the root-to-canal ratio, in a mesiodistal dimension at the cementoenamel junction, is roughly 1:1. Because the mature maxillary central incisor model does not allow for the degree of apical closure or the fraction of final root length to be used as determinants, the root-to-canal ratio provided a consistent parameter for preparation. An effort was made to standardize the preparation of each specimen so that parameters such as absence of craze or fracture lines, canal preparation centering, placement of the applied load, embedment of the
Vol. 27, No. 1, January 2001
Restoration of Immature Nonvital Incisors
21
TABLE 3. Student-Newman-Keuls results
FIG 6. Fracture of tooth concomitant with fracture of the restorative material. TABLE 1. Fracture values in kilograms for the three test groups Test Groups Specimen No.
Control (kg)
Composite Only (kg)
Composite ⫹ RR (kg)
1 2 3 4 5 6 7 8 9 10 Mean Standard Deviation
29.5 35.7 28.1 34.8 27.8 30.6 — — — — 31.08 3.3926
52.3 55.1 51.2 56.9 50.8 51.4 51.8 48.2 48.5 43.8 51.00 3.6533
45.7 38.2 39.5 42.7 36.7 38.1 41.2 30.0 35.9 31.3 37.93 4.8173
TABLE 2. ANOVA Sum of Squares Between groups Within groups Total
df
Mean Square
F
1680.597 2 840.298 50.001 386.529 23 16.806
Significance .000
2067.126 25
teeth in acrylic to the facial cementoenamel junction, and dimensions of the canal preparation were similar, leaving tooth restoration as the only variable. Preparation of the immature canal space was done with a 3 mm diameter engineering twist drill in a drill press. Thus the naturally occurring ovoid shape of the immature
Group
n
Control Composite only Composite ⫹ RR
6 10 10
Statistical Subsets (p ⱕ 0.003) 1
2
3
31.0833 51.0000 37.9300
canal was sacrificed in favor of a standardized, repeatable preparation. What effect this difference would have clinically is unknown, and such a variable would be difficult to control without having stage 3 immature maxillary central incisors as specimens. The experimental model was quite successful in providing a repeatable system for each trial. All fractures passed through the cervical area of the root. Each tooth in experimental groups 2 and 3 also fractured through the respective restorative materials at the point where the fracture extended through the prepared canal space. This result is important for two reasons. First, according to Cvek (5), all teeth included in his clinical study fractured in the cervical area of the root. Teeth that fractured in the middle or apical part of the root were associated with a second, severe injury and were not included in his study, leaving cervically fractured teeth to account for his reported prognosis percentages. Second, failure of the restorative materials occurred concomitantly with fracture of the teeth. Thus the load at the point of fracture directly reflected the strength of the tooth and the restorative material at the fracture site. Other important factors leading to the observed consistent results include centering of the canal preparation that was made possible by absolute straight-line access and setting the load application point 2 mm below the incisal edge. Placement of a notch in the incisal edge to obtain absolute straight line access for a 3-mm diameter drill did not adversely affect the model, since no fractures occurred through the incisal portion of the tooth. With the load application point established at 2 mm below the incisal edge, all fractures were in the cervical area. This indicates no adverse effects by placing the point of force at what would be expected to be a weaker (more incisal) place on the tooth. The depth of the preparation to a minimum of 3 mm below the facial cementoenamel junction was adequate, because all of the fractures in the experimental groups extended through the filling materials during strength testing. The following observations may be made relative to the statistical analysis of the results. First there was a highly significant difference between both experimental groups and the control teeth (Table 3, p ⱕ 0.003). This indicates that placement of bonded composite, with or without the RR, strengthens the immature tooth model beyond an unfilled model. Second there was also a highly significant difference (Table 3, p ⱕ 0.003) between groups 2 and 3 with regard to load to fracture. This result suggests that placement of bonded composite alone strengthens the immature tooth model greater than placement of bonded composite in conjunction with the RR. Therefore the null hypothesis that no difference exists between composite alone and composite with RR can be rejected. This difference is statistically in favor of composite alone over composite with RR. Thus the theoretical benefits of the RR when used in a composite restoration of this type are not supported by the findings of this study. The significant difference between the composite-alone and the RR ⫹ composite groups can be explained as follows. The RR fabric must be bonded to the composite either mechanically or chemically and in this way the two work together to resist the
22
Pene et al.
applied load. If this bonding is not provided the RR in the composite acts alone and stretches readily. This means that there is no load resistance from the RR during initial loading and therefore no reinforcement. The fractured surfaces were examined under a stereomicroscope, and no voids were found with the exception of the fiber extending through the composite. Therefore the volume occupied by the RR acts as a void in the composite. In turn, this void reduces the volume of composite resisting the applied load. Hence there is a reduction in the fracture strength of the RR ⫹ composite versus the composite alone. In conclusion a model for developmental stage 3 immature maxillary central incisors is presented that allows repeatable cervical fractures to test restorative materials. Statistical analysis of the results suggests that bonded composite with or without RR strengthens the immature tooth model beyond no fill, and bonded composite alone strengthens the immature tooth model to a higher degree than bonded composite used in conjunction with RR. This study was partially supported by the Graduate Endodontic Fund, University of Washington School of Dentistry, Seattle, WA. The authors thank Dr. Tracy Pene, Tom Wade, and Leng Tuach for their assistance; and Ann Mattiello and Gloria Pascual for their assistance with preparing this manuscript. Dr. Pene is a former graduate student in Endodontics, and is currently practicing in Newport Beach and Mission Viejo, CA. Dr. Nicholls is a professor,
Journal of Endodontics Department of Restorative Dentistry, and Dr. Harrington is a Professor and chairman, Department of Endodontics, University of Washington School of Dentistry, Seattle, WA. Address requests for reprints to Dr. Gerald W. Harrington, Department of Endodontics, University of Washington, Box 357448, Seattle, WA 98195-7448.
References 1. Ravn JJ. Dental injuries in Copenhagen school children, school years 1967–1972. Comm Dent Oral Epidemiol 1974;2:231– 45. 2. O’Mullane DM. Injured permanent incisor teeth: an epidemiological study. J Irish Dent Assoc 1972;18:160 –73. 3. Ghose LJ, Baghdady VS, Hikmat BYM. Apexification of immature apices of pulpless permanent anterior teeth with calcium hydroxide. J Endodon 1987;13:285–90. 4. Kerekes K, Heide S, Jacobsen I. Follow-up examination of endodontic treatment in traumatized juvenile incisors. J Endodon 1980;6:744 – 8. 5. Cvek M. Prognosis of luxated non-vital maxillary incisors treated with calcium hydroxide and filled with gutta percha. A retrospective clinical study. Endod Dent Traumatol 1992;8:45–55. 6. Rabie G, Trope M, Garcia C, Tronstad L. Strengthening and restoration of immature teeth with an acid etch resin technique. Endod Dent Traumatol 1985;1:246 –56. 7. Katebzadeh N, Dalton C, Trope M. Strengthening immature teeth during and after apexification. J Endodon 1998;24:256 –9. 8. Karna J. A fiber composite laminate endodontic post and core. Am J Dent 1996;9:230 –2. 9. Reitz PV, Aoki H, Yoshioka M, Uehara J, Kubota Y. A cephalometric study of tooth position as related to facial structure in profiles of human beings: a comparison of Japanese and American adults. J Prosthet Dent 1973;29:157– 66.
The Journal of Endodontics’ “Ask a Friend” feature provides a forum for subscribers to submit a clinical case about which they have questions and have it published in the Journal, along with a review by one of a panel of “Friends.” The person submitting the case may remain anonymous upon request. To learn of the procedures for submitting an article, check the January 1998 issue of the Journal.
Printed in U.S.A. VOL. 27, NO. 1, JANUARY 2001
Evaluation of Fiber-Composite Laminate in the Restoration of Immature, Nonvital Maxillary Central Incisors Jeffrey R. Pene, DDS, MSD, Jack I. Nicholls, PhD, and Gerald W. Harrington, DDS, MSD
According to Cvek (5), the prognosis for nonvital, immature incisors is dependent, in large part, on the stage of root development at the time of devitalization. The results he reported for the occurrence of cervical fractures 4 yr after filling the canal included 77% in stage 1 (wide, divergent apical opening, root length less than 1⁄2 final length); 53% in stage 2 (wide, divergent apical opening, root length 1⁄2 final length); 43% in stage 3 (wide, divergent apical opening, root length 2⁄3 of final length); and 28% in stage 4 (open apical foramen, nearly completed root length). He classified the stages of root development by radiographic estimation of the width of the apical foramen and the length of the root. Recent studies have proposed the use of acid-etched composites to strengthen endodontically treated immature teeth (6). Combined with various prefabricated posts this method has been advocated for usage during and after apexification (7). It has been postulated that one of the apparent shortcomings of composite material used in this manner is the propagation of microcracks resulting from inherent porosities and flexure in the composite material leading to subsequent failure of the restoration (8). Karna (8) suggested the use of fiber composite laminate to prevent propagation of microcracks and thereby reinforce the composite material, making it more resistant to fracture. The purpose of this study was to develop a model for immature maxillary central incisors and to evaluate whether a ribbon-reinforced composite can increase the fracture resistance of the model over composite alone.
The purpose of this study was to evaluate the use of fiber-composite laminate, a reinforcement ribbon embedded throughout a composite restoration, to reinforce immature maxillary central incisors. Twenty-six mature maxillary central incisors were prepared a minimum of 3 mm below the facial cementoenamel junction to simulate immature nonvital teeth. They were separated into 3 groups: group 1 was unfilled and served as a control; group 2 was filled to the depth of the preparation with composite; and group 3 was filled with composite and Connect Reinforcement Ribbon. The specimens were subjected to class I loading in an Instron Testing Machine until catastrophic failure occurred. The results indicate a highly significant difference between the groups (p < 0.003). Group 1 fractured at an average load of 31.08 kg, group 2 at 51.00 kg, and group 3 at 37.93 kg. These findings suggest that composite alone increases fracture resistance of the immature tooth model more than composite with Reinforcement Ribbon.
Previous research indicates that the most common site of dental impact injuries in the developing dentition is the maxillary anterior teeth (1, 2). Frequently these injuries can lead to necrosis of the pulp and subsequent arrested development of the tooth. The clinician is thus faced with the difficult task of treating these teeth endodontically in addition to restoring teeth with thin dentinal walls. Apexification treatment of these immature teeth has been shown to be predictable with a high degree of success (3–5). The thin dentinal walls at the cementoenamel junction, however, make them prone to fracture from secondary injuries (i.e. mastication or minor trauma), often leaving then nonrestorable (5). Studies have shown that, despite successful endodontic treatment, ⬃28 to 77% of these teeth— depending on the stage of root development—will fracture during or after treatment, leaving many clinicians to view the procedure as having a poor prognosis for teeth arrested in the early stages of development (5).
MATERIALS AND METHODS Twenty-six intact, extracted maxillary central incisors were used in this study. The teeth were stored in water except during restoration and experimental testing. Craze lines in enamel were noted using a dissecting microscope, and any teeth with fracture lines extending into dentin were not used. Additional teeth were used as pilot specimens to establish the experimental model. Immature Tooth Model Preparation Teeth were lingually accessed using a #4 round bur and a tapered diamond bur to a standardized dimension of 3 mm mesiodistally by 3 mm incisogingivally. For the purpose of establishing the experimental model the mesiodistal dimension of the access was ex18
Vol. 27, No. 1, January 2001
Restoration of Immature Nonvital Incisors
19
FIG 2. Enlargement of the canal in the drill press. FIG 1. Surveyor used to align the canal perpendicular to the base of the Duralay-lined acrylic block.
tended to 3 mm, allowing clearance for the subsequent engineering twist drill. This prevented drill contact with enamel and thereby reduced the risk of fracture or craze line formation. All teeth had a notch placed in the incisal edge to allow absolute straight-line access and were then instrumented to 3 mm below the facial cementoenamel junction with 1 to 6 Gates-Glidden drills. To simulate an immature tooth, a surveyor pin was placed in the canal to the prepared depth, and the entire tooth set into a Duralay (Reliance Dental Manufacturing Co., Worth, IL)-lined acrylic block. Thus the surveyor pin and the long axis of the canal were perpendicular to the top of the acrylic block in the surveyor (Fig. 1). After the Duralay had set the acrylic block was placed in a drill press and the canal enlarged to 3 mm in diameter using an engineering twist drill (Fig. 2). The preparation was extended a minimum of 3 mm below the facial cementoenamel junction. The 3 mm diameter was chosen to approximate stage 3 of development where the canal is roughly one-half the entire mesiodistal dimension (5) (Figs. 3 and 4). After preparation, the teeth were randomly assigned to 1 of 3 groups, with 6 teeth in the unrestored control group and 10 teeth in each of the two experimental groups. Group 1: Unrestored Positive Control The access cavity and canal preparation were unrestored. Group 2: Dentin-Bonded Composite Enamel was etched with 35% phosphoric acid for 30 s, then rinsed thoroughly with water. Tenure Conditioner (Den-Mat Corp., Santa Maria, CA) was applied uniformly with an applicator brush to the
FIG 3. Radiographs comparing an immature tooth to a prepared immature tooth model.
internal walls of the entire canal and access. After 15 s this layer was thoroughly washed and dried. A dual-cure dentin bonding agent (Tenure A & B, Den-Mat Corp.) was placed on the internal surfaces of the canal and chamber and smoothed to an even thin layer with the applicator brush. This layer was then light-cured for 30 s. Using a small composite syringe tip in a Centrix injection gun (Dentsply Caulk, Milford, DE), Prisma VLC Hybrid Composite (Dentsply Caulk, Milford, DE) was injected into the canal space, completely filling it to the level of the cavo-surface margin. Excess composite was removed with a plastic instrument and the composite light-cured for 2 min.
20
Pene et al.
Journal of Endodontics
FIG 4. Proximal radiographs comparing an immature tooth to a prepared immature tooth model.
Group 3: Ribbon-Reinforced Dentin-Bonded Composite The procedural steps for group 3 were identical to those used for group 2. After the composite was injected into the canal space, a 2-mm-wide piece of Reinforcement Ribbon (RR) (Connect, Kerr USA, Orange, CA) was cut to a length equal to twice the depth of each canal preparation. The RR was coated with the bonding agent and excess removed. The RR material was handled with cotton pliers only. Using a #9 Schilder plugger, the RR strip was shaped into a “V” and inserted to the depth of the canal with the width in a faciolingual direction. Some of the composite was displaced as the ends of the “V” were pushed even with the cavo-surface margin. A second piece of RR, which had been prepared in the same manner, was then shaped into a “V” and inserted into the canal to the same depth. The width of the second “V” was placed in a mesiodistal direction and thus perpendicular to the first strip. The RR was adapted close to the canal walls with the plugger each time a new piece was placed. The ends of the RR were then packed below the cavo-surface margin, and additional hybrid composite was used to fill the void resulting from RR placement. Excess composite was removed with a plastic instrument and the restoration light cured for 2 min.
Fracture Strength Evaluation An Instron Testing Machine (model TTMBL; Instron Corp., Canton, MA) was used to apply a controlled load to each specimen at a crosshead speed of 5.0 mm/min. To ensure a fixed position for this load application, composite was bonded to the lingual surface of each specimen 2 mm below the incisal edge. The composite was leveled to provide an even point of contact that would remain stable during the loading trial. Each specimen block was fixed in a jig that ensured a loading angle of 130 degrees to the long axis of the tooth. This angle was chosen because it simulates the average angle of contact between maxillary and mandibular incisors in a class I occlusion (9). The force was then applied in a linguolabial vector. The magnitude of the load was continuously recorded by the Instron machine until catastrophic failure occurred.
FIG 5. Radiograph of representative specimen showing fracture at the cervical area of the tooth.
RESULTS Of the 26 specimens tested, all teeth showed either horizontal or oblique fractures that went through the cervical area of the root (Fig. 5). At the point where the fracture extended through the prepared canal space, the fracture was also through the restorative materials in groups 2 and 3 (Fig. 6). The results of the strength testing trials are summarized by group in Table 1. The results of a one-way analysis of variance test (ANOVA), shown in Table 2, indicate a significant difference exists between the group means (p ⱕ 0.003). The results of the Student-NewmanKeuls test, (shown in Table 3) indicate three significant subsets (p ⱕ 0.003). In this table each test group defines a separate statistical subset, with the control group having the lowest fracture values, and the composite-alone group having the highest fracture values. These results suggest that composite alone increased fracture resistance of immature teeth greater than ribbon-reinforced composite. DISCUSSION In the present study, stage 3 root development was selected for the model because it is the stage at which the root-to-canal ratio, in a mesiodistal dimension at the cementoenamel junction, is roughly 1:1. Because the mature maxillary central incisor model does not allow for the degree of apical closure or the fraction of final root length to be used as determinants, the root-to-canal ratio provided a consistent parameter for preparation. An effort was made to standardize the preparation of each specimen so that parameters such as absence of craze or fracture lines, canal preparation centering, placement of the applied load, embedment of the
Vol. 27, No. 1, January 2001
Restoration of Immature Nonvital Incisors
21
TABLE 3. Student-Newman-Keuls results
FIG 6. Fracture of tooth concomitant with fracture of the restorative material. TABLE 1. Fracture values in kilograms for the three test groups Test Groups Specimen No.
Control (kg)
Composite Only (kg)
Composite ⫹ RR (kg)
1 2 3 4 5 6 7 8 9 10 Mean Standard Deviation
29.5 35.7 28.1 34.8 27.8 30.6 — — — — 31.08 3.3926
52.3 55.1 51.2 56.9 50.8 51.4 51.8 48.2 48.5 43.8 51.00 3.6533
45.7 38.2 39.5 42.7 36.7 38.1 41.2 30.0 35.9 31.3 37.93 4.8173
TABLE 2. ANOVA Sum of Squares Between groups Within groups Total
df
Mean Square
F
1680.597 2 840.298 50.001 386.529 23 16.806
Significance .000
2067.126 25
teeth in acrylic to the facial cementoenamel junction, and dimensions of the canal preparation were similar, leaving tooth restoration as the only variable. Preparation of the immature canal space was done with a 3 mm diameter engineering twist drill in a drill press. Thus the naturally occurring ovoid shape of the immature
Group
n
Control Composite only Composite ⫹ RR
6 10 10
Statistical Subsets (p ⱕ 0.003) 1
2
3
31.0833 51.0000 37.9300
canal was sacrificed in favor of a standardized, repeatable preparation. What effect this difference would have clinically is unknown, and such a variable would be difficult to control without having stage 3 immature maxillary central incisors as specimens. The experimental model was quite successful in providing a repeatable system for each trial. All fractures passed through the cervical area of the root. Each tooth in experimental groups 2 and 3 also fractured through the respective restorative materials at the point where the fracture extended through the prepared canal space. This result is important for two reasons. First, according to Cvek (5), all teeth included in his clinical study fractured in the cervical area of the root. Teeth that fractured in the middle or apical part of the root were associated with a second, severe injury and were not included in his study, leaving cervically fractured teeth to account for his reported prognosis percentages. Second, failure of the restorative materials occurred concomitantly with fracture of the teeth. Thus the load at the point of fracture directly reflected the strength of the tooth and the restorative material at the fracture site. Other important factors leading to the observed consistent results include centering of the canal preparation that was made possible by absolute straight-line access and setting the load application point 2 mm below the incisal edge. Placement of a notch in the incisal edge to obtain absolute straight line access for a 3-mm diameter drill did not adversely affect the model, since no fractures occurred through the incisal portion of the tooth. With the load application point established at 2 mm below the incisal edge, all fractures were in the cervical area. This indicates no adverse effects by placing the point of force at what would be expected to be a weaker (more incisal) place on the tooth. The depth of the preparation to a minimum of 3 mm below the facial cementoenamel junction was adequate, because all of the fractures in the experimental groups extended through the filling materials during strength testing. The following observations may be made relative to the statistical analysis of the results. First there was a highly significant difference between both experimental groups and the control teeth (Table 3, p ⱕ 0.003). This indicates that placement of bonded composite, with or without the RR, strengthens the immature tooth model beyond an unfilled model. Second there was also a highly significant difference (Table 3, p ⱕ 0.003) between groups 2 and 3 with regard to load to fracture. This result suggests that placement of bonded composite alone strengthens the immature tooth model greater than placement of bonded composite in conjunction with the RR. Therefore the null hypothesis that no difference exists between composite alone and composite with RR can be rejected. This difference is statistically in favor of composite alone over composite with RR. Thus the theoretical benefits of the RR when used in a composite restoration of this type are not supported by the findings of this study. The significant difference between the composite-alone and the RR ⫹ composite groups can be explained as follows. The RR fabric must be bonded to the composite either mechanically or chemically and in this way the two work together to resist the
22
Pene et al.
applied load. If this bonding is not provided the RR in the composite acts alone and stretches readily. This means that there is no load resistance from the RR during initial loading and therefore no reinforcement. The fractured surfaces were examined under a stereomicroscope, and no voids were found with the exception of the fiber extending through the composite. Therefore the volume occupied by the RR acts as a void in the composite. In turn, this void reduces the volume of composite resisting the applied load. Hence there is a reduction in the fracture strength of the RR ⫹ composite versus the composite alone. In conclusion a model for developmental stage 3 immature maxillary central incisors is presented that allows repeatable cervical fractures to test restorative materials. Statistical analysis of the results suggests that bonded composite with or without RR strengthens the immature tooth model beyond no fill, and bonded composite alone strengthens the immature tooth model to a higher degree than bonded composite used in conjunction with RR. This study was partially supported by the Graduate Endodontic Fund, University of Washington School of Dentistry, Seattle, WA. The authors thank Dr. Tracy Pene, Tom Wade, and Leng Tuach for their assistance; and Ann Mattiello and Gloria Pascual for their assistance with preparing this manuscript. Dr. Pene is a former graduate student in Endodontics, and is currently practicing in Newport Beach and Mission Viejo, CA. Dr. Nicholls is a professor,
Journal of Endodontics Department of Restorative Dentistry, and Dr. Harrington is a Professor and chairman, Department of Endodontics, University of Washington School of Dentistry, Seattle, WA. Address requests for reprints to Dr. Gerald W. Harrington, Department of Endodontics, University of Washington, Box 357448, Seattle, WA 98195-7448.
References 1. Ravn JJ. Dental injuries in Copenhagen school children, school years 1967–1972. Comm Dent Oral Epidemiol 1974;2:231– 45. 2. O’Mullane DM. Injured permanent incisor teeth: an epidemiological study. J Irish Dent Assoc 1972;18:160 –73. 3. Ghose LJ, Baghdady VS, Hikmat BYM. Apexification of immature apices of pulpless permanent anterior teeth with calcium hydroxide. J Endodon 1987;13:285–90. 4. Kerekes K, Heide S, Jacobsen I. Follow-up examination of endodontic treatment in traumatized juvenile incisors. J Endodon 1980;6:744 – 8. 5. Cvek M. Prognosis of luxated non-vital maxillary incisors treated with calcium hydroxide and filled with gutta percha. A retrospective clinical study. Endod Dent Traumatol 1992;8:45–55. 6. Rabie G, Trope M, Garcia C, Tronstad L. Strengthening and restoration of immature teeth with an acid etch resin technique. Endod Dent Traumatol 1985;1:246 –56. 7. Katebzadeh N, Dalton C, Trope M. Strengthening immature teeth during and after apexification. J Endodon 1998;24:256 –9. 8. Karna J. A fiber composite laminate endodontic post and core. Am J Dent 1996;9:230 –2. 9. Reitz PV, Aoki H, Yoshioka M, Uehara J, Kubota Y. A cephalometric study of tooth position as related to facial structure in profiles of human beings: a comparison of Japanese and American adults. J Prosthet Dent 1973;29:157– 66.
The Journal of Endodontics’ “Ask a Friend” feature provides a forum for subscribers to submit a clinical case about which they have questions and have it published in the Journal, along with a review by one of a panel of “Friends.” The person submitting the case may remain anonymous upon request. To learn of the procedures for submitting an article, check the January 1998 issue of the Journal.