Abstract
Objectives
The use of all-ceramic restorations has been steadily increasing in the decades since the introduction of high-strength core ceramics as an alternative to metal cores. Even with high-strength ceramic cores, the crowns are susceptible to fractures during clinical function, probably the major concern associated with all-ceramic restorations. The aim of this study was to analyze fracture features of clinically fractured crowns in order to determine crack initiation sites and crack propagation paths.
Methods
Fractographic methods were used to analyze 27 all-ceramic full coverage crowns with alumina cores and feldspatic veneering ceramic. The fractured crowns were supplied by dentists in private and public practices with information of type of cement and time in function if available.
Results
The fractured crowns comprised 13 incisors, 3 premolars and 11 molars. The results revealed that all fractures initiated in the cervical margin of the crowns and usually from the approximal area close to the most coronally placed curvature of the margin. There was no statistically significant association between the cement used and time in function.
Significance
Fractography of clinically failed all-ceramic crowns can provide information on the fracture modes and design considerations. The findings are in contrast to fracture modes from in vitro tests. The results suggest that more clinically relevant in vitro test methods to study the durability of ceramic crowns should be developed.
1
Introduction
All-ceramic restorations are considered to be the most esthetically pleasing alternative when restoring broken or decayed teeth. The use has been steadily increasing in the decades since the introduction of high-strength core ceramics as an alternative to metal cores . Ceramic materials are, however, stiff and brittle and therefore susceptible to fracture in particular when exposed to tensile stresses. Several clinical studies of all-ceramic restorations reveal that fracture of the restorations is the main reason for failure . The clinical trials available do unfortunately not explain why or how the restorations fracture. Several laboratory tests and some finite element analyses have been performed to understand the fracture mechanisms of all-ceramic crowns, but it has not yet been possible to fully explain the mechanisms of clinical fractures or to simulate clinical fractures in vitro . Systematic fractographic analyses of ceramic restorations fractured in clinical use may reveal the fracture origin, the fracture path, the stress at fracture and maybe the reason for fracture . If we understand why some restorations fail early, while others remain functional much longer, we may reduce the number of fractures or increase the time in function before fracture occurs. Additionally, the information may be used to develop laboratory test that are more clinically relevant than the ones we use today.
The aim of this study was to analyze fracture features of clinically fractured crowns in order to determine crack initiation and crack propagation in all-ceramic, single crowns with alumina cores.
2
Materials and methods
27 fractured all-ceramic crowns with alumina cores (Procera Alumina, Nobel Biocare, Stockholm, Sweden) were collected and analyzed. The crowns were obtained from various practicing dentists in Norway and Sweden. The parts that had fallen out of the patients’ mouth were collected, if not lost. Parts that remained in the mouth, were carefully removed by the dentist, intact or by splitting with a bur if necessary. The damage done by this removal was distinguished during the analyses. Metal instruments leave gray marks on the ceramic surface and drilling gives a rough surface with drilling grooves. The available pieces were packed to avoid damage in the post and sent by mail. The dentists also supplied available information of time in function, type of cement used and special events occurring at the time of fracture. The crowns were inspected visually to determine how many parts they consisted of and whether any parts were missing. The separate parts were loosely reassembled to visualize the shape and orientation of the crown, taking care not to destroy any surfaces ( Fig. 1 ). The crowns were cleansed with ethylenediaminetetraacetic acid (EDTA) and then placed in distilled water in an ultrasonic bath for 5 min to remove plaque, calculus and debris. Analyses were performed in optical light microscopies (Leica DM IRM, Wetzlar, Germany) with gradual increase in magnification. The fracture surfaces of each crown were mapped with photographs where important fractographic features were documented. The starting point and crack propagation were determined using standard fractographic methods . Each crown was thoroughly searched for wake hackle in the veneering ceramic and the crack propagation was determined according to the direction of these. The region of the starting point was more closely examined for defects, such as pores, cracks or flaws. If there were several fracture paths, contact damage and chipping, the direction of the hackle were used to determine the primary and secondary fractures according to standard fractographic methods. The analyses were performed twice by the same operator with at least two months between the two analyzing sessions to assess the repeatability of the method. The second analyses were performed blinded of the previous results. In the cases where the two results differed, a third analysis was performed to determine crack initiation and propagation.
2
Materials and methods
27 fractured all-ceramic crowns with alumina cores (Procera Alumina, Nobel Biocare, Stockholm, Sweden) were collected and analyzed. The crowns were obtained from various practicing dentists in Norway and Sweden. The parts that had fallen out of the patients’ mouth were collected, if not lost. Parts that remained in the mouth, were carefully removed by the dentist, intact or by splitting with a bur if necessary. The damage done by this removal was distinguished during the analyses. Metal instruments leave gray marks on the ceramic surface and drilling gives a rough surface with drilling grooves. The available pieces were packed to avoid damage in the post and sent by mail. The dentists also supplied available information of time in function, type of cement used and special events occurring at the time of fracture. The crowns were inspected visually to determine how many parts they consisted of and whether any parts were missing. The separate parts were loosely reassembled to visualize the shape and orientation of the crown, taking care not to destroy any surfaces ( Fig. 1 ). The crowns were cleansed with ethylenediaminetetraacetic acid (EDTA) and then placed in distilled water in an ultrasonic bath for 5 min to remove plaque, calculus and debris. Analyses were performed in optical light microscopies (Leica DM IRM, Wetzlar, Germany) with gradual increase in magnification. The fracture surfaces of each crown were mapped with photographs where important fractographic features were documented. The starting point and crack propagation were determined using standard fractographic methods . Each crown was thoroughly searched for wake hackle in the veneering ceramic and the crack propagation was determined according to the direction of these. The region of the starting point was more closely examined for defects, such as pores, cracks or flaws. If there were several fracture paths, contact damage and chipping, the direction of the hackle were used to determine the primary and secondary fractures according to standard fractographic methods. The analyses were performed twice by the same operator with at least two months between the two analyzing sessions to assess the repeatability of the method. The second analyses were performed blinded of the previous results. In the cases where the two results differed, a third analysis was performed to determine crack initiation and propagation.
3
Results
Wake hackle around pores in the veneering ceramic were detected in all crowns and it was therefore possible to determine the direction of the fracture path in all the crowns ( Fig. 2 ). There was a 93 per cent match in the assessment of initiating area between the two sessions. One crown missed the part of the crown containing the initiation point, but the remaining parts of the crown indicated a start in the cervical area. This crown was first misjudged as having fractured from the occlusal surface. One was difficult to analyze due to many secondary fractures of the veneering material and the crack path was determined to go in opposite directions in the two sessions. The third analysis mapped the secondary chip-offs first. It was then possible to determine the primary crack propagation. Both of these restorations had been marked as “difficult” or “uncertain” in both analyzing sessions. Two crowns had two initiation points ( Fig. 3 ). For these it was possible to distinguish the primary fracture path from the secondary due to the directions of the wake hackle. Initiation flaws were difficult to detect with this method and were thus not analyzed.
An overview of the crowns retrieved, time in function, type of cement used and fracture modes is presented in Table 1 . The mean time in function was 3.1 years (median: 2.75, range: 0.6–10), but for 7 crowns the time in function was not known. The crowns cemented with glass ionomer cements had the longest mean time in function with 4.8 years compared to 2.3 years for both zinc phosphate and resin-based cements. Molars had a mean time in function of 3.6 years and the incicors 3.1 years. There was, however, no statistically significant correlation between type of cement or type of tooth and time in function. This may, however, be due to the low number of teeth included in the analysis which gives a low power of the calculations. 19 of 27 fractures had a direction from one approximal surface to the other, 2 fractures from one approximal surface to the buccal surface, and 5 crowns fractured from the oral or buccal side ( Fig. 4 ). For one crown it was not possible to determine the orientation of the crown and thus impossible to know whether the fracture initiation was in the approximal area or not. The typical approximal fractures started close to the most coronally placed point of the cervical margin ( Fig. 4 ). Some of the crowns had a semilunar fracture of the palatinal flange only or as the primary fracture.
No | Restoration type | Cement used | Time in function | Initiation point | Direction of fracture | Additional information |
---|---|---|---|---|---|---|
1 | Premolar | * | 1.2 | Cerv | Appr-appr | |
2 | Upper incisor | * | 1 | Cerv | Appr-appr | |
3 | Molar | ZP | * | Cerv | Appr-bucc | |
4 | Molar | GI | * | Cerv | Palatinal | |
5 | Molar | ZP | * | Cerv | Appr-appr | |
6 | Upper central | * | * | Cerv | Appr-appr | |
7 | Molar | Resin | 0.6 | Cerv | Appr-appr | |
8 | Molar | Resin | 2.1 | Cerv | Appr-appr | |
9 | Molar | GI | 3.2 | Cerv | Pal-bucc | |
10 | Molar | GI | 5.5 | Cerv | Appr-appr | |
11 | Upper lateral | GI | 3.4 | Cerv | Appr-appr | Two starting points |
12 | Lower central | ZP | 3 | Cerv | Appr-appr | |
13 | Upper lateral | GI | 1 | Cerv | Appr-appr | |
14 | Upper central | Resin | 2.5 | Cerv | Palatinal | |
15 | Upper lateral | Resin | 2.5 | Cerv | Appr-bucc | |
16 | Molar | Resin | 4 | Cerv | Appr-appr | |
17 | Upper central | * | 3 | Cerv | Appr-appr | Epileptical fit |
18 | Premolar | ZP | 0 | Cerv | Appr-appr | |
19 | Upper central | ZP | 2 | Cerv | Palatinal | Metal Post |
20 | Premolar | * | * | Cerv | Palatinal | |
21 | Upper central | * | 3 | * | Appr-appr | Start missing |
22 | Molar | * | * | Cerv | ** | |
23 | Molar | * | * | Cerv | Appr-appr | Two starting points |
24 | Molar | GI | 6 | Cerv | Appr-appr | |
25 | Upper central | R | 4 | Cerv | Appr-appr | |
26 | Upper central | R | 1.5 | Cerv | Appr-appr. | |
27 | Upper central | GI | 10 | Cerv | Appr-appr |