d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 1280–1286
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.intl.elsevierhealth.com/journals/dema
The influence of substance loss and ferrule height on the fracture resistance of endodontically treated premolars. An in vitro study Abdulaziz Samran a,b,c,∗ , Shadi El Bahra a,d , Matthias Kern a a
Department of Prosthodontics, Propaedeutics and Dental Materials, School of Dentistry, Christian-Albrechts University at Kiel, Kiel, Germany b Department of Fixed Prosthodontics, School of Dentistry, Ibb University, Ibb, Yemen c Al-Farabi Dental College, Riyadh, Saudi Arabia d Department of Removable Prosthodontics, School of Dentistry, Damascus University, Damascus, Syria
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objective. This study evaluated the effect of different ferrule heights and varying degrees of
Received 3 February 2013
substance loss on the fracture resistance of endodontically treated premolars.
Received in revised form
Methods. Eighty extracted and endodontically treated lower premolars were used and divided
14 August 2013
into 5 test groups (n = 16) depending on the ferrule height: A (0.0 mm), B (0.5 mm), C (1.0 mm),
Accepted 7 October 2013
D (1.5 mm) and E (2.0 mm) respectively. Teeth in subgroups were either with 1 or 2 residual coronal dentin walls which were 3 mm in height and 1 mm in thickness. Teeth were restored with glass fiber posts and cast crowns. All specimens were then subjected to dynamic load-
Keywords:
ing in a masticatory simulator for 1,200,000 loading cycles with a nominal load of 5 kg at
Residual coronal wall
1.2 Hz combined with thermal cycling (5–55 ◦ C, dwell time 30 s). Then specimens were quasi-
Ferrule height
statically loaded at 30◦ in a universal testing machine until fractured. Data were analyzed
Glass fiber post
with 2-way ANOVA, followed by multiple comparisons using Tukey HSD test (˛ = .05).
Resin cement
Results. Mean (SD) failure loads for groups ranged from 679.5 ± 164.9 N to 1084.5 ± 269.9 N. Two-way ANOVA revealed that both the ferrule height and the number of residual coronal walls had a significant influence on the fracture resistance (P < .001 and P = .006, respectively). Significant increases were produced in the final fracture resistance, when the ferrule height was increased, which was reduced to approximately 37% when teeth with 2 mm ferrule height were compared with teeth without a ferrule. Significance. Under the conditions of this in vitro study, increasing the number of residual coronal walls and ferrule height had a significant effect on the fracture resistance of endodontically treated premolars restored with prefabricated posts. © 2013 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
Endodontically treated teeth (ETT) have been problematic because of coronal destruction from dental caries, fractures,
previous restoration, and endodontic therapy. This results in an increase of the likelihood of fracture of the treated tooth during function [1]. The prognosis of ETT is influenced by different parameters such as amount of hard tissue loss [2], presence of a minimum of 1.5–2.0 mm ferrule height
∗ Corresponding author at: Department of Prosthodontics, Propaedeutics and Dental Materials School of Dentistry, Christian-Albrechts University at Kiel, Arnold-Heller Strasse 16, 24105 Kiel, Germany. Tel.: +49 431 5972877; fax: +49 431 597 2860. E-mail addresses: [email protected], [email protected] (A. Samran). 0109-5641/$ – see front matter © 2013 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.dental.2013.10.003
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 1280–1286
preparation [3], material and design of post and core material used [4]. The susceptibility to fracture of a restored tooth is increased with the loss of coronal dentin. Therefore, tooth longevity will depend on the amount of remaining tooth structure and the efficiency of the restorative procedures to replace lost structural integrity [5]. A treatment using posts should be used only for retention of a core within the remaining tooth structure but not for aiming to strengthen the tooth [6]. Several post system techniques are available for the restoration of ETT and can be divided into custom-made cast post and core systems and prefabricated post systems. There are many concerns when using custom-made metallic posts due to their inhomogeneous stress distribution, biological side effects due to microleakage and corrosion, and the dark color under all-ceramic restorations [7]. The popularity of the fiber posts is due to their favorable physical properties [8] and their bond strength is equivalent to adhesively or conventionally luted gold posts [9]. They were reported to reduce the risk of tooth fracture and display higher survival rates than teeth restored with rigid zirconia posts [10]. An important element for tooth preparation is the incorporation of a ferrule design. Sorensen and Engelman described the ferrule as the coronal dentinal extension of the tooth structure occlusal to the shoulder preparation [11]. Libman and Nicholls suggested that to achieve the full benefits of ferrule effect, it should be a minimum of 1.5 mm in height with parallel dentin walls, totally encircling the tooth and ending on sound tooth structure [12]. Many studies investigating the ferrule effect have used cast posts and cores [13–15], but there is little information as to whether the ferrule is of additional value in providing reinforcement in teeth restored with prefabricated glass-fiber posts and composite cores [16]. Mangold and Kern [17] described the influence of posts on fracture resistance of ETT with varying substance loss but they did not indicate the effect of the different ferrule height in their study. Therefore, the aim of this study was to evaluate the fracture resistance of endodontically treated premolars (ETPs) restored with glass-fiber posts when different ferrule heights and varying degrees of substance loss were incorporated. The null hypothesis of the study was that neither ferrule height nor the amount of residual coronal dentin would affect the fracture resistance of crowned premolars.
2.
Materials and methods
2.1.
Test groups
Eighty recently extracted caries-free lower premolars, which were removed for orthodontics or periodontal reasons, were selected and then stored in 0.1% thymol solution (Caelo, Hilden, Germany). The teeth were cleaned with a hand scaler and stored then at room temperature during the study. Endodontic access cavities were prepared using a water cooled air turbine handpiece. During root canal preparation the working length was set at 1 mm short of the apical foramen. The teeth were endodontically prepared using the step-back technique to an ISO size 50 (K-files; Dentsply De Trey, Constance, Germany), irrigated with 3% sodium hypochlorite solution (Hedinger, Stuttgart, Germany) and dried with
1281
paper points (Coltene/Whaledent Inc., Langenau, Germany). Each canal was obturated using the lateral condensation method with gutta-percha points (Coltene/Whaledent Inc.) and sealed with an eugenol-free epoxyamine resin sealer (AH Plus; Dentsply De Trey). After 24-h water storage at 37 ◦ C, gutta-percha was removed using no. 2,3,4 Gates-Glidden burs (Maillefer, Ballaigues, Switzerland). The teeth roots were embedded into brass tubes, using an auto-polymerizing resin (Technovit 4000; Heraeus Kulzer, Wehrheim, Germany) up to 2 mm apical to the cemento-enamel junction (CEJ) and oriented their long axes perpendicular to horizontal using a custom-made surveyor. The ETPs received 0.8 mm shoulder finish lines which were mesial and distal 1 mm more coronal than the facial and lingual surfaces and which were cervical to the (CEJ). Burs were replaced after 8 preparations, in order to ensure high cutting efficacy. For teeth preparations, diamond rotary cutting instruments (Brasseler, Lemgo, Germany) under copious air-water cooling were used (with a 2◦ taper to achieve a 4◦ convergence angle) in a high-speed handpiece mounted on a custom-made parallelometer to standardize the preparation for all specimens. The teeth were assigned randomly to 5 groups of 16 teeth each according to the ferrule height. The properties of the specimens included in each group were as follows: group A: specimens without circumferential ferrule; group B: circumferential ferrule 0.5 mm above the finish line; group C: circumferential ferrule 1 mm above the finish line; group D: circumferential ferrule 1.5 mm above the finish line; group E: circumferential ferrule 2 mm above the finish line. Teeth in subgroups had either 1 residual facial wall (A1, B1, C1, D1 and E1) or 2 residual facial and lingual walls (A2, B2, C2, D2 and E2). The walls were 3 mm high and 1 mm thick. Post spaces were accomplished with a tapered drill (ER-post kit; Brasseler) of ISO size 90 to achieve an intraradicular post length of 7.5 mm for all teeth. The coronal opening of the post space was enlarged in a facio-lingual direction to a 3 mm in width and 2 mm in depth to resist rotation and to standardize the coronal openings and the thickness of residual coronal walls. The walls of the post preparation were roughened using a diamond-coated hand instrument 3 times (ER Post Systems; Brasseler) [18]. The glass-fiber posts (Komet ER DentinPost; ISO size 90, Brasseler) were airborne-particle abraded for 5 s at a distance of 30-mm with 50 m alumina particles (Heraeus Kulzer) at 0.25 MPa and ultrasonically cleaned in 96% isopropanol (German Federal Monopoly Administration for Spirits, Hamburg, Germany) for 3 min. The post spaces were then irrigated with a 3% sodium hypochlorite solution and dried with paper points, followed by irrigating with 70% ethanol (German Federal Monopoly Administration for Spirits) and drying with paper points. The posts were luted with adhesive composite-resin cement using a microbrush (Panavia 21 TC; Kuraray Medical, Osaka, Japan) after conditioning the dentin with the system’s autopolymerizing primer (ED-Primer; Kuraray) for 60 s. The resin cements were mixed and applied according to the manufacturer’s instructions. Excess luting resin was used to coat the coronal portion of the post. An auto-polymerizing composite resin (Clearfil Core; Kuraray Medical, Osaka, Japan) was applied as the core material according to the manufacturer’s
1282
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 1280–1286
Fig. 1 – Dimensions of preparation, restoration and ferrule height (in mm).
instructions. After completing polymerization of the resin, cores were prepared to the required dimensions (Fig. 1).
2.2.
Cast crown fabrication
Impressions of the prepared specimens were made with a polyether impression material (Permadyne Penta H; 3M/Espe, Seefeld, Germany). After 30 min, the impressions were poured in type IV stone (GC Fujirock EP, Leuven, Belgium). To obtain identical crown dimensions in all specimens, a stylized reference crown with a 30◦ angulation of the buccal cusp to the vertical tooth axis was created in wax (Crowax; Renfert GmbH, Hilzingen, Germany). Then, the crowns were duplicated onto the other dies by inserting heated liquid wax into a silicone mold (Deguform, Degudent, Hanau-Wolfgang, Germany). The crown wax patterns were measured using a wax caliper to make sure that all patterns have the same dimensions. Then, the wax patterns were invested and casted in cobalt–chromium alloy (Wirobond C; Bego, Bremen, Germany) following the instructions of the manufacturer. The internal surfaces of the crowns were airborne-particle abraded with 50 m alumina (Aluminum Oxide Abrasive; Heraeus Kulzer) at 0.25 MPa pressure and then ultrasonically cleaned in 96% isopropanol (German Federal Monopoly Administration for Spirits). The tooth preparations were cleaned with a rotary brush (Omnident; Rodgau, Germany) and pumice (Sterilbimspaste; Ernst Hinrichs GmbH, Goslar, Germany). Then, the crowns were cemented using glassionomer cement (Ketac Cem Maxicap; 3M/Espe) which was mixed according to the manufacturer’s instructions. During the cementing procedures, each crown was held in place for 7 min under a 5-kg load using custom made devise. Materials used in the restorative procedures are listed in Table 1.
2.3.
Loading of the specimens
After storing the specimens in deionized water at 37 ◦ C for 3 days, all specimens underwent combined masticatory loading simulation in a dual-axis masticatory simulator (Willytec, Munich, Germany) with a nominal load of 5 kg for 1.2 million cycles and thermocycling at 5–55 ◦ C for 6499 cycles. All specimens which survived the dynamic loading were quasi-statically loaded with a crosshead speed of 1 mm/min at an angle of 30◦ to the longitudinal axis of the tooth in a universal testing machine (Zwick Z010/TN2A; Zwick, Ulm, Germany) until they were fractured. Loading was on the lingual incline of the buccal cusp at a distance of 2 mm from the central fossa of the crown. The failure load of the specimen was determined when the force-versus-time graph showed an abrupt change in load, indicating a sudden decrease in the specimen’s resistance to compressive loading. Specimens were visually examined for the type and location of failure, as well as the direction of failure.
2.4.
Statistical analysis
Data were explored for normality using Kolmogorov–Smirnov and Shapiro–Wilk tests, which showed that data were normally distributed. Levene test for homogeneity of variance indicated homogeneity of variances between groups. Two-way analysis of variance (ANOVA) was used to compare fracture resistance means among the five groups followed by multiple comparisons using Tukey HSD test (˛ = .05). The confidence level was 95%. Statistical analysis was performed with SPSS 18.0 (SPSS 18.0 for Windows; SPSS, Inc., Chicago, IL). According to the significance level (˛ = .05) and the sample size (n = 8), the test of choice had adequate power to detect
1283
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 1280–1286
Table 1 – Materials used for restorative procedures. Material ER Dentin Post Clearfil Core Permadyne Penta H Panavia 21 Ketac Cem Maxicap Cobalt-chromium alloy
Company Brasseler, Lemgo, Germany Kuraray, Osaka, Japan 3M/Espe, Seefeld, Germany Kuraray, Osaka, Japan 3M/Espe, Seefeld, Germany Wirobond C, Bego, Bremen, Germany
Batch number 676303 041523 Heavy 434544Light 422524 041344 347837 3533
Table 2 – Fracture loads in N [means (SD)]. Group A B C D E
1 residual coronal wall
2 residual coronal walls
A1: 679.5 (164.9) B1: 742.6 (166.6) C1: 824.7 (194.3) D1: 854.0 (232.1) E1: 932.2 (206.4)
A2: 754.9 (193.4) B2: 824.0 (157.7) C2: 933.9 (145.5) D2: 1052.3 (187.0) E2: 1084.5 (269.9)
A, no ferrule; B, 0.5 mm ferrule; C, 1 mm ferrule; D, 1.5 mm ferrule; E, 2 mm ferrule height; 1, 1 residual wall; 2, 2 residual walls.
statistical differences which could be used to provide clinical recommendations (F = .11). Failure modes were recorded and statistically analyzed with Chi-square (X2 ) testing for significant correlation between design and failure modes.
3.
Results
None of the specimens failed during masticatory simulation. The mean values of the fracture strength and standard deviations are displayed in Table 2. They ranged from 679.5 ± 164.9 N to 1084.5 ± 269.9 N. The fracture resistance of each group increased when the ferrule height increased and a second residual coronal wall existed. Two-way ANOVA (Table 3) indicated that both the ferrule height and the number of residual walls had a significant influence on the fracture resistance (P ≤ .001 and P = .006, respectively). There was no statistically significant interaction between the factors ferrule height and residual coronal walls (P = .889). Chi-square (X2 ) test revealed that there were no significant differences in fracture modes among the 10 groups (Table 4). The mode of failure was determined by visual inspection of all specimens. The type of fracture behavior and the frequency are illustrated in Fig. 2. There were 2 typical root fracture modes: cervical third fracture (favorable mode), middle and apical third (catastrophic mode). All groups had almost complete favorable fracture mode. The fracture behavior in A1, B1, and C1 subgroups with 1 residual coronal wall differed slightly from that in subgroups with 2 residual coronal walls, where the fracture line crossed into the dental substance which began further facially. Nearly all the teeth had a facial fracture by 2–4 mm below the crown margin and lingual along the crown margin.
4.
Discussion
The present study investigated the influence of five ferrule heights on the fracture resistance of crowned lower premolars. Teeth in subgroups were either with 1 or 2 residual coronal dentin walls. Eight specimens per group were exposed to thermal cycling and mechanical loading and loaded until they
fractured. Eight specimens per subgroup were chosen because 8 specimens can be loaded at a time in the masticatory simulator. Teeth are generally prepared; however, with their finish lines following the coronal extension of the gingival tissue level interproximally. To mimic this clinical condition, the finish lines in this study were mesial and distal 1 mm more coronal than the facial and lingual surfaces and which were cervical to the CEJ. A custom-made parallelometer was used to standardize the preparation for all specimens and the required dimensions were obtained prior to core fabrication by reducing the tooth structure in a stepwise manner using a digital sliding caliper to control dimensions. After core fabrication only a low speed handpiece with a fine grain diamond was used to finish the preparation and only a minimal additional amount of dentin was removed by that procedure. It must be admitted that this resulted in a slight overestimation of the remaining coronal tooth structure. However, as this was done in the same manner in each group it is assumed that this did not affect the results considerably. The first hypothesis that the ferrule height would not affect the fracture resistance of crowned premolars had to be rejected. The ferrule height had a significant influence on the final fracture resistance (P ≤ .001), which was reduced by approximately 37% when teeth with 2 mm ferrule height were compared with teeth without ferrule. In addition, the amount of residual coronal dentin had a significant influence on the final fracture resistance of the restored teeth (P = .006). Therefore, the second hypothesis that the amount of residual coronal dentin would not affect the fracture resistance of crowned premolars was also rejected. Unfortunately, the authors identified no other studies that evaluated the effect of the ferrule height and the number of residual walls on the fracture strength of the crowned premolars. None of the specimens failed during masticatory simulation. For that reason, the fracture resistance of the aged specimens to quasi-static loading could be determined in all groups. The fracture resistance of the restored premolars ranged from 679.5 ± 164.9 N (group A1) to 1084.5 ± 269.9 N (group E2) and can be compared well to previous in vitro studies
1284
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 1280–1286
Table 3 – Summary of 2-way ANOVA of main factors. Sum of squares
dfa
Mean square
F
P
Ferrule Wall Ferrule × wall Error
912,096.450 304,057.800 42,919.200 2,663,122.500
4 1 4 70
228024.113 304,057.800 10,729.800 38,044.607
5.994 7.992 .282
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.intl.elsevierhealth.com/journals/dema
The influence of substance loss and ferrule height on the fracture resistance of endodontically treated premolars. An in vitro study Abdulaziz Samran a,b,c,∗ , Shadi El Bahra a,d , Matthias Kern a a
Department of Prosthodontics, Propaedeutics and Dental Materials, School of Dentistry, Christian-Albrechts University at Kiel, Kiel, Germany b Department of Fixed Prosthodontics, School of Dentistry, Ibb University, Ibb, Yemen c Al-Farabi Dental College, Riyadh, Saudi Arabia d Department of Removable Prosthodontics, School of Dentistry, Damascus University, Damascus, Syria
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objective. This study evaluated the effect of different ferrule heights and varying degrees of
Received 3 February 2013
substance loss on the fracture resistance of endodontically treated premolars.
Received in revised form
Methods. Eighty extracted and endodontically treated lower premolars were used and divided
14 August 2013
into 5 test groups (n = 16) depending on the ferrule height: A (0.0 mm), B (0.5 mm), C (1.0 mm),
Accepted 7 October 2013
D (1.5 mm) and E (2.0 mm) respectively. Teeth in subgroups were either with 1 or 2 residual coronal dentin walls which were 3 mm in height and 1 mm in thickness. Teeth were restored with glass fiber posts and cast crowns. All specimens were then subjected to dynamic load-
Keywords:
ing in a masticatory simulator for 1,200,000 loading cycles with a nominal load of 5 kg at
Residual coronal wall
1.2 Hz combined with thermal cycling (5–55 ◦ C, dwell time 30 s). Then specimens were quasi-
Ferrule height
statically loaded at 30◦ in a universal testing machine until fractured. Data were analyzed
Glass fiber post
with 2-way ANOVA, followed by multiple comparisons using Tukey HSD test (˛ = .05).
Resin cement
Results. Mean (SD) failure loads for groups ranged from 679.5 ± 164.9 N to 1084.5 ± 269.9 N. Two-way ANOVA revealed that both the ferrule height and the number of residual coronal walls had a significant influence on the fracture resistance (P < .001 and P = .006, respectively). Significant increases were produced in the final fracture resistance, when the ferrule height was increased, which was reduced to approximately 37% when teeth with 2 mm ferrule height were compared with teeth without a ferrule. Significance. Under the conditions of this in vitro study, increasing the number of residual coronal walls and ferrule height had a significant effect on the fracture resistance of endodontically treated premolars restored with prefabricated posts. © 2013 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
Endodontically treated teeth (ETT) have been problematic because of coronal destruction from dental caries, fractures,
previous restoration, and endodontic therapy. This results in an increase of the likelihood of fracture of the treated tooth during function [1]. The prognosis of ETT is influenced by different parameters such as amount of hard tissue loss [2], presence of a minimum of 1.5–2.0 mm ferrule height
∗ Corresponding author at: Department of Prosthodontics, Propaedeutics and Dental Materials School of Dentistry, Christian-Albrechts University at Kiel, Arnold-Heller Strasse 16, 24105 Kiel, Germany. Tel.: +49 431 5972877; fax: +49 431 597 2860. E-mail addresses: [email protected], [email protected] (A. Samran). 0109-5641/$ – see front matter © 2013 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.dental.2013.10.003
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 1280–1286
preparation [3], material and design of post and core material used [4]. The susceptibility to fracture of a restored tooth is increased with the loss of coronal dentin. Therefore, tooth longevity will depend on the amount of remaining tooth structure and the efficiency of the restorative procedures to replace lost structural integrity [5]. A treatment using posts should be used only for retention of a core within the remaining tooth structure but not for aiming to strengthen the tooth [6]. Several post system techniques are available for the restoration of ETT and can be divided into custom-made cast post and core systems and prefabricated post systems. There are many concerns when using custom-made metallic posts due to their inhomogeneous stress distribution, biological side effects due to microleakage and corrosion, and the dark color under all-ceramic restorations [7]. The popularity of the fiber posts is due to their favorable physical properties [8] and their bond strength is equivalent to adhesively or conventionally luted gold posts [9]. They were reported to reduce the risk of tooth fracture and display higher survival rates than teeth restored with rigid zirconia posts [10]. An important element for tooth preparation is the incorporation of a ferrule design. Sorensen and Engelman described the ferrule as the coronal dentinal extension of the tooth structure occlusal to the shoulder preparation [11]. Libman and Nicholls suggested that to achieve the full benefits of ferrule effect, it should be a minimum of 1.5 mm in height with parallel dentin walls, totally encircling the tooth and ending on sound tooth structure [12]. Many studies investigating the ferrule effect have used cast posts and cores [13–15], but there is little information as to whether the ferrule is of additional value in providing reinforcement in teeth restored with prefabricated glass-fiber posts and composite cores [16]. Mangold and Kern [17] described the influence of posts on fracture resistance of ETT with varying substance loss but they did not indicate the effect of the different ferrule height in their study. Therefore, the aim of this study was to evaluate the fracture resistance of endodontically treated premolars (ETPs) restored with glass-fiber posts when different ferrule heights and varying degrees of substance loss were incorporated. The null hypothesis of the study was that neither ferrule height nor the amount of residual coronal dentin would affect the fracture resistance of crowned premolars.
2.
Materials and methods
2.1.
Test groups
Eighty recently extracted caries-free lower premolars, which were removed for orthodontics or periodontal reasons, were selected and then stored in 0.1% thymol solution (Caelo, Hilden, Germany). The teeth were cleaned with a hand scaler and stored then at room temperature during the study. Endodontic access cavities were prepared using a water cooled air turbine handpiece. During root canal preparation the working length was set at 1 mm short of the apical foramen. The teeth were endodontically prepared using the step-back technique to an ISO size 50 (K-files; Dentsply De Trey, Constance, Germany), irrigated with 3% sodium hypochlorite solution (Hedinger, Stuttgart, Germany) and dried with
1281
paper points (Coltene/Whaledent Inc., Langenau, Germany). Each canal was obturated using the lateral condensation method with gutta-percha points (Coltene/Whaledent Inc.) and sealed with an eugenol-free epoxyamine resin sealer (AH Plus; Dentsply De Trey). After 24-h water storage at 37 ◦ C, gutta-percha was removed using no. 2,3,4 Gates-Glidden burs (Maillefer, Ballaigues, Switzerland). The teeth roots were embedded into brass tubes, using an auto-polymerizing resin (Technovit 4000; Heraeus Kulzer, Wehrheim, Germany) up to 2 mm apical to the cemento-enamel junction (CEJ) and oriented their long axes perpendicular to horizontal using a custom-made surveyor. The ETPs received 0.8 mm shoulder finish lines which were mesial and distal 1 mm more coronal than the facial and lingual surfaces and which were cervical to the (CEJ). Burs were replaced after 8 preparations, in order to ensure high cutting efficacy. For teeth preparations, diamond rotary cutting instruments (Brasseler, Lemgo, Germany) under copious air-water cooling were used (with a 2◦ taper to achieve a 4◦ convergence angle) in a high-speed handpiece mounted on a custom-made parallelometer to standardize the preparation for all specimens. The teeth were assigned randomly to 5 groups of 16 teeth each according to the ferrule height. The properties of the specimens included in each group were as follows: group A: specimens without circumferential ferrule; group B: circumferential ferrule 0.5 mm above the finish line; group C: circumferential ferrule 1 mm above the finish line; group D: circumferential ferrule 1.5 mm above the finish line; group E: circumferential ferrule 2 mm above the finish line. Teeth in subgroups had either 1 residual facial wall (A1, B1, C1, D1 and E1) or 2 residual facial and lingual walls (A2, B2, C2, D2 and E2). The walls were 3 mm high and 1 mm thick. Post spaces were accomplished with a tapered drill (ER-post kit; Brasseler) of ISO size 90 to achieve an intraradicular post length of 7.5 mm for all teeth. The coronal opening of the post space was enlarged in a facio-lingual direction to a 3 mm in width and 2 mm in depth to resist rotation and to standardize the coronal openings and the thickness of residual coronal walls. The walls of the post preparation were roughened using a diamond-coated hand instrument 3 times (ER Post Systems; Brasseler) [18]. The glass-fiber posts (Komet ER DentinPost; ISO size 90, Brasseler) were airborne-particle abraded for 5 s at a distance of 30-mm with 50 m alumina particles (Heraeus Kulzer) at 0.25 MPa and ultrasonically cleaned in 96% isopropanol (German Federal Monopoly Administration for Spirits, Hamburg, Germany) for 3 min. The post spaces were then irrigated with a 3% sodium hypochlorite solution and dried with paper points, followed by irrigating with 70% ethanol (German Federal Monopoly Administration for Spirits) and drying with paper points. The posts were luted with adhesive composite-resin cement using a microbrush (Panavia 21 TC; Kuraray Medical, Osaka, Japan) after conditioning the dentin with the system’s autopolymerizing primer (ED-Primer; Kuraray) for 60 s. The resin cements were mixed and applied according to the manufacturer’s instructions. Excess luting resin was used to coat the coronal portion of the post. An auto-polymerizing composite resin (Clearfil Core; Kuraray Medical, Osaka, Japan) was applied as the core material according to the manufacturer’s
1282
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 1280–1286
Fig. 1 – Dimensions of preparation, restoration and ferrule height (in mm).
instructions. After completing polymerization of the resin, cores were prepared to the required dimensions (Fig. 1).
2.2.
Cast crown fabrication
Impressions of the prepared specimens were made with a polyether impression material (Permadyne Penta H; 3M/Espe, Seefeld, Germany). After 30 min, the impressions were poured in type IV stone (GC Fujirock EP, Leuven, Belgium). To obtain identical crown dimensions in all specimens, a stylized reference crown with a 30◦ angulation of the buccal cusp to the vertical tooth axis was created in wax (Crowax; Renfert GmbH, Hilzingen, Germany). Then, the crowns were duplicated onto the other dies by inserting heated liquid wax into a silicone mold (Deguform, Degudent, Hanau-Wolfgang, Germany). The crown wax patterns were measured using a wax caliper to make sure that all patterns have the same dimensions. Then, the wax patterns were invested and casted in cobalt–chromium alloy (Wirobond C; Bego, Bremen, Germany) following the instructions of the manufacturer. The internal surfaces of the crowns were airborne-particle abraded with 50 m alumina (Aluminum Oxide Abrasive; Heraeus Kulzer) at 0.25 MPa pressure and then ultrasonically cleaned in 96% isopropanol (German Federal Monopoly Administration for Spirits). The tooth preparations were cleaned with a rotary brush (Omnident; Rodgau, Germany) and pumice (Sterilbimspaste; Ernst Hinrichs GmbH, Goslar, Germany). Then, the crowns were cemented using glassionomer cement (Ketac Cem Maxicap; 3M/Espe) which was mixed according to the manufacturer’s instructions. During the cementing procedures, each crown was held in place for 7 min under a 5-kg load using custom made devise. Materials used in the restorative procedures are listed in Table 1.
2.3.
Loading of the specimens
After storing the specimens in deionized water at 37 ◦ C for 3 days, all specimens underwent combined masticatory loading simulation in a dual-axis masticatory simulator (Willytec, Munich, Germany) with a nominal load of 5 kg for 1.2 million cycles and thermocycling at 5–55 ◦ C for 6499 cycles. All specimens which survived the dynamic loading were quasi-statically loaded with a crosshead speed of 1 mm/min at an angle of 30◦ to the longitudinal axis of the tooth in a universal testing machine (Zwick Z010/TN2A; Zwick, Ulm, Germany) until they were fractured. Loading was on the lingual incline of the buccal cusp at a distance of 2 mm from the central fossa of the crown. The failure load of the specimen was determined when the force-versus-time graph showed an abrupt change in load, indicating a sudden decrease in the specimen’s resistance to compressive loading. Specimens were visually examined for the type and location of failure, as well as the direction of failure.
2.4.
Statistical analysis
Data were explored for normality using Kolmogorov–Smirnov and Shapiro–Wilk tests, which showed that data were normally distributed. Levene test for homogeneity of variance indicated homogeneity of variances between groups. Two-way analysis of variance (ANOVA) was used to compare fracture resistance means among the five groups followed by multiple comparisons using Tukey HSD test (˛ = .05). The confidence level was 95%. Statistical analysis was performed with SPSS 18.0 (SPSS 18.0 for Windows; SPSS, Inc., Chicago, IL). According to the significance level (˛ = .05) and the sample size (n = 8), the test of choice had adequate power to detect
1283
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 1280–1286
Table 1 – Materials used for restorative procedures. Material ER Dentin Post Clearfil Core Permadyne Penta H Panavia 21 Ketac Cem Maxicap Cobalt-chromium alloy
Company Brasseler, Lemgo, Germany Kuraray, Osaka, Japan 3M/Espe, Seefeld, Germany Kuraray, Osaka, Japan 3M/Espe, Seefeld, Germany Wirobond C, Bego, Bremen, Germany
Batch number 676303 041523 Heavy 434544Light 422524 041344 347837 3533
Table 2 – Fracture loads in N [means (SD)]. Group A B C D E
1 residual coronal wall
2 residual coronal walls
A1: 679.5 (164.9) B1: 742.6 (166.6) C1: 824.7 (194.3) D1: 854.0 (232.1) E1: 932.2 (206.4)
A2: 754.9 (193.4) B2: 824.0 (157.7) C2: 933.9 (145.5) D2: 1052.3 (187.0) E2: 1084.5 (269.9)
A, no ferrule; B, 0.5 mm ferrule; C, 1 mm ferrule; D, 1.5 mm ferrule; E, 2 mm ferrule height; 1, 1 residual wall; 2, 2 residual walls.
statistical differences which could be used to provide clinical recommendations (F = .11). Failure modes were recorded and statistically analyzed with Chi-square (X2 ) testing for significant correlation between design and failure modes.
3.
Results
None of the specimens failed during masticatory simulation. The mean values of the fracture strength and standard deviations are displayed in Table 2. They ranged from 679.5 ± 164.9 N to 1084.5 ± 269.9 N. The fracture resistance of each group increased when the ferrule height increased and a second residual coronal wall existed. Two-way ANOVA (Table 3) indicated that both the ferrule height and the number of residual walls had a significant influence on the fracture resistance (P ≤ .001 and P = .006, respectively). There was no statistically significant interaction between the factors ferrule height and residual coronal walls (P = .889). Chi-square (X2 ) test revealed that there were no significant differences in fracture modes among the 10 groups (Table 4). The mode of failure was determined by visual inspection of all specimens. The type of fracture behavior and the frequency are illustrated in Fig. 2. There were 2 typical root fracture modes: cervical third fracture (favorable mode), middle and apical third (catastrophic mode). All groups had almost complete favorable fracture mode. The fracture behavior in A1, B1, and C1 subgroups with 1 residual coronal wall differed slightly from that in subgroups with 2 residual coronal walls, where the fracture line crossed into the dental substance which began further facially. Nearly all the teeth had a facial fracture by 2–4 mm below the crown margin and lingual along the crown margin.
4.
Discussion
The present study investigated the influence of five ferrule heights on the fracture resistance of crowned lower premolars. Teeth in subgroups were either with 1 or 2 residual coronal dentin walls. Eight specimens per group were exposed to thermal cycling and mechanical loading and loaded until they
fractured. Eight specimens per subgroup were chosen because 8 specimens can be loaded at a time in the masticatory simulator. Teeth are generally prepared; however, with their finish lines following the coronal extension of the gingival tissue level interproximally. To mimic this clinical condition, the finish lines in this study were mesial and distal 1 mm more coronal than the facial and lingual surfaces and which were cervical to the CEJ. A custom-made parallelometer was used to standardize the preparation for all specimens and the required dimensions were obtained prior to core fabrication by reducing the tooth structure in a stepwise manner using a digital sliding caliper to control dimensions. After core fabrication only a low speed handpiece with a fine grain diamond was used to finish the preparation and only a minimal additional amount of dentin was removed by that procedure. It must be admitted that this resulted in a slight overestimation of the remaining coronal tooth structure. However, as this was done in the same manner in each group it is assumed that this did not affect the results considerably. The first hypothesis that the ferrule height would not affect the fracture resistance of crowned premolars had to be rejected. The ferrule height had a significant influence on the final fracture resistance (P ≤ .001), which was reduced by approximately 37% when teeth with 2 mm ferrule height were compared with teeth without ferrule. In addition, the amount of residual coronal dentin had a significant influence on the final fracture resistance of the restored teeth (P = .006). Therefore, the second hypothesis that the amount of residual coronal dentin would not affect the fracture resistance of crowned premolars was also rejected. Unfortunately, the authors identified no other studies that evaluated the effect of the ferrule height and the number of residual walls on the fracture strength of the crowned premolars. None of the specimens failed during masticatory simulation. For that reason, the fracture resistance of the aged specimens to quasi-static loading could be determined in all groups. The fracture resistance of the restored premolars ranged from 679.5 ± 164.9 N (group A1) to 1084.5 ± 269.9 N (group E2) and can be compared well to previous in vitro studies
1284
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 1280–1286
Table 3 – Summary of 2-way ANOVA of main factors. Sum of squares
dfa
Mean square
F
P
Ferrule Wall Ferrule × wall Error
912,096.450 304,057.800 42,919.200 2,663,122.500
4 1 4 70
228024.113 304,057.800 10,729.800 38,044.607
5.994 7.992 .282
