Probability of survival of implant-supported metal ceramic and CAD/CAM resin nanoceramic crowns

Abstract

Objectives

To evaluate the probability of survival and failure modes of implant-supported resin nanoceramic relative to metal-ceramic crowns.

Methods

Resin nanoceramic molar crowns (LU) (Lava Ultimate, 3M ESPE, USA) were milled and metal-ceramic (MC) (Co–Cr alloy, Wirobond C+, Bego, USA) with identical anatomy were fabricated ( n = 21). The metal coping and a burnout-resin veneer were created by CAD/CAM, using an abutment (Stealth-abutment, Bicon LLC, USA) and a milled crown from the LU group as models for porcelain hot-pressing (GC-Initial IQ-Press, GC, USA). Crowns were cemented, the implants ( n = 42, Bicon) embedded in acrylic-resin for mechanical testing, and subjected to single-load to fracture (SLF, n = 3 each) for determination of step-stress profiles for accelerated-life testing in water ( n = 18 each). Weibull curves (50,000 cycles at 200N, 90% CI) were plotted. Weibull modulus ( m ) and characteristic strength ( η ) were calculated and a contour plot used ( m versus η ) for determining differences between groups. Fractography was performed in SEM and polarized-light microscopy.

Results

SLF mean values were 1871N (±54.03) for MC and 1748N (±50.71) for LU. Beta values were 0.11 for MC and 0.49 for LU. Weibull modulus was 9.56 and η = 1038.8N for LU, and m = 4.57 and η = 945.42N for MC ( p > 0.10). Probability of survival (50,000 and 100,000 cycles at 200 and 300N) was 100% for LU and 99% for MC. Failures were cohesive within LU. In MC crowns, porcelain veneer fractures frequently extended to the supporting metal coping.

Conclusion

Probability of survival was not different between crown materials, but failure modes differed.

Significance

In load bearing regions, similar reliability should be expected for metal ceramics, known as the gold standard, and resin nanoceramic crowns over implants. Failure modes involving porcelain veneer fracture and delamination in MC crowns are less likely to be successfully repaired compared to cohesive failures in resin nanoceramic material.

Introduction

Since osseointegration of dental implants is a predictable treatment modality, there is a demand for comprehensive understanding of the main complications of prosthetic designs eventually affecting success rates . Increasing efforts have been devoted in attempt to describe differences in prosthetic outcomes for implant-supported reconstructions, for instance, regarding external relative to internal connections, cemented versus screwed prostheses in systematic reviews and laboratory studies . However, when considering the final prostheses material for implant-supported reconstructions, very few studies have addressed the outcomes of all-ceramic materials which currently are increasingly demanded from both patients and dentists for improved esthetic results.

The use of high-strength ceramics such as stabilized zirconia has gained attention for use as a substructure material, especially for large span reconstructions due to its high mechanical properties . However, long-term clinical studies are virtually exclusive for tooth-supported zirconia-veneered fixed dental prostheses (FDP), and they include a very high variation in results among trials where the main reported complication is fracture of the veneering porcelain . Although clinical results for implant-supported zirconia-veneered single-crowns are sparse , the existing studies agree that the main failure is fracture of the veneering porcelain and with a unacceptable variation in failure rates varying from 3% to 24.5% with risks of chipping 3.8 times higher than metal-ceramics . A more recent 5-year prospective study has reported on 42.8% of zirconia-veneered porcelain fracture complications for implant-supported single crowns, which does raise concerns for its indication . Reasons for such high failure rates and ways to diminish them have been extensively presented in the literature and include an array of materials and laboratory handling precautions. In addition, patient-related factors such as the decreased proprioception and lower tactile sensitivity contributes to making implant-supported reconstructions more prone to failure .

While future trials reporting the successful use of zirconia-veneered implant-supported reconstructions are warranted, the development of metal free alternatives, such as resin nanoceramic has gained interest also in implant prosthodontics. Resin based materials are claimed to have more resiliency compared to ceramics resulting in improved dampening from occlusal forces to be more repair-friendly should chipping occur, and easier to grind during computer-assisted machine/computer-assisted manufacture (CAD/CAM) process or during occlusal adjustments.

Previous research has demonstrated that several composites directly bonded to titanium abutments as molar crowns (IAC ® – integrated abutment crowns) presented similar single load to fracture (SLF) values compared to metal ceramic crowns (MC) . Considering that fatigue plays a more relevant role in simulating clinical failures , recent studies have evaluated the use of hand-layered indirect composites for implant-supported molar crowns and found promising probability of survival (reliability) when tested under fatigue . Given that flaws, such as voids, are inherently introduced when an indirect composite is hand-layered, pressed blocks of resin nanoceramics have been developed for milling through the CAD/CAM process from several companies in attempt to further improve them from a fatigue resistance perspective.

Considering metal ceramic as the gold standard for comparisons, this study sought to investigate the reliability and failure mode of MC for implant-supported molar crowns compared to CAD/CAM fabricated resin nanoceramic. Our tested null hypothesis was that there would be no difference in probability of survival or failure modes when subjected to step-stress accelerated life testing in water.

Materials and methods

Crown fabrication

Forty-two Ti-6Al-4V abutments (Stealth abutments, shouldered, 3 mm well, 5 mm diameter, Bicon LLC, Boston, MA, USA) were selected for the study and divided in two groups ( n = 21 each) to support metal ceramic (MC) or Lava Ultimate (LU) (Lava Ultimate, 3M ESPE, St. Paul, MN, USA) maxillary first molar crowns. A waxed model maxillary first molar crown was replicated on an E4D Dentist CAD/CAM system (D4D Technologies, Richardson, TX). Crowns were milled from Lava Ultimate blocks ( n = 21, shade A3, 3M ESPE) in an E4D mill (3M ESPE), polished with diamond paste and bristle brush, then buffed to a high gloss with a cotton buff. The bonding surface of each Lava Ultimate crown was sandblasted with a 240 mesh alumina (Ney-Brasive™ J.M. Ney Co., Bloomfield, CT, USA). Crowns were cleaned by sonicating in ethanol, air dried, and their intaglio surface primed with RelyX Ceramic Primer (3M ESPE) applied with a fibertip brush, then dried with compressed air.

An anatomic metal coping ( Fig. 1 ) with a 360° collar and veneer top were designed and exported as .STL files for fabrication. Twenty-one metal copings were made via selective laser melting from non-precious cobalt–chromium dental alloy (Wirobond ® C+, Bego USA, Lincoln, RI, USA). Veneer tops for lost-wax investment and hot-pressing were made via wax printing (BeCe ® Wax-Up material, Bego USA). An opaque layer (GC™ Initial™ Opaque, GC America Inc. Alsip, IL, USA) was applied to the copings before investing them along with the wax veneer tops. After investment burnout, pressable ceramic (GC Initial™ IQ – One Body, Layering-over-Metal Pressable Ceramic) was hot-pressed directly onto the opaqued metal. Glaze (GC Initial IQ Glaze) was applied and fired on the crowns after hot-pressing by a commercial laboratory (Valley Dental Arts, Stillwater, MN). The intaglio surface of each MC crown was sandblasted with 240 mesh alumina, then cleaned by sonicating in ethanol and air dried.

Fig. 1
Stereomicroscopy micrographs of crowns embedded in epoxy resin and sectioned proximally. A) Metal ceramic crown shows a varying thickness of the supporting metal coping (between 0.5 and 1.5 mm) and an approximate thickness of 1.5 mm for the porcelain veneer. B) Resin nanoceramic crown bonded directly to the titanium abutment with a final thickness at the region underneath sliding contact fatigue of approximately 3.5 mm. a (abutment), m (metal coping), p (porcelain), RN (resin nanoceramic).

Crowns from both groups were bonded to the abutments using self-adhesive resin cement (RelyX™ Unicem™ 2 Automix Self-Adhesive Cement, 3M ESPE, St. Paul, MN, USA) where excess cement was wiped away with a microbrush. Due to their translucency, Lava Ultimate crowns were light cured 20 s each on buccal, lingual, and occusal sides with an LED curing light (Elipar S10 LED Curing Light, 3M ESPE). The cement was allowed to cure (dark cure) by resting undisturbed for at least 1 h before handling. Excess cement (oxygen inhibited) was wiped away with ethanol-soaked wipe. One additional crown of each group was fabricated and embedded in epoxy resin (Redelease, São Paulo, SP, Brazil), and polished from buccal to lingual (SiC papers #600, #1200, #2000, and #2500) to illustrate differences in veneering material thickness ( Fig. 1 ).

Crowns fixed to the implants were vertically attached to the analyzing rod of a surveyor by means of a silicone matrix which secured the crown to allow standardized positioning of samples into PVC tubes for pouring with self-curing acrylic resin (Orthodontic resin, Dentsply Caulk, Milford, DE, USA). The implant was exposed to allow its axial positioning during embedding. The sectioned PVC tube was positioned under the silicone key containing the implant/abutment/crown assembly in the center. Then, self-curing acrylic resin was poured and the analyzing rod of the surveyor lowered with the implant/abutment interface remaining 1 mm below the potting surface.

Step-stress accelerated life testing (SSALT) and reliability analysis

Three crowns from each group were subjected to single load to fracture testing with a spherical indenter (6 mm diameter, D-2 Steel) positioned at the mesio-lingual cusp at a crosshead speed of 1 mm/min (Model 800R, Test Resources, Inc., Shakopee, MN, USA) . Failure was defined as veneer material fracture. The mean load to failure values were used to determine step-stress profiles for accelerated life fatigue testing, undertaken in the remaining samples ( n = 18, each group) for probability of survival calculation. Fig. 2 shows sample distribution for mechanical testing.

Fig. 2
Schematic showing group division and number of samples distributed across SLF and step-stress accelerated life testing.

Step-stress accelerated life testing was performed in an axial direction at constant frequency of 2 Hz with the spherical indenter positioned 1 mm from the cusp tip, at the slope of the mesio-lingual cusp incline as done for the SLF test, which allowed for sliding of the indenter (0.3–0.5 mm) . Crowns were submerged in water at room temperature throughout mechanical testing. Although details of SSALT method used in this study has been reported elsewhere , in brief, three profiles are designed as mild, moderate, and aggressive, with the number of specimens assigned to each group being distributed in the ratios of 3:2:1, respectively (i.e. n = 9 in the mild, n = 6 in the moderate, and n = 3 in the aggressive load profile). These profiles are named based on the step-wise load increase that the specimen will be fatigued throughout the cycles until a certain level of load, meaning that specimens assigned to a mild profile will be cycled longer to reach the same load level of a specimen assigned to the aggressive profile. Fatigue loads throughout SSALT ranged from 200 N up to a maximum of 1150 N with a steady increase in load as a function of elapsed cycles. Samples that survived the maximum fatigue load (no fracture) were deemed suspended and accounted for in Weibull distribution.

Based upon the step-stress distribution of the failures, use level probability Weibull curves (probability of failure versus cycles) with use stress of 200 N and 90% two-sided confidence intervals were calculated and plotted (Alta Pro 9, ReliaSoft, Tucson, AZ, USA) using a power law relationship for damage accumulation. Reliability (the probability of an item functioning for a given amount of time without failure) for completion of a mission of 50,000 and 100,000 cycles at 200 N and 300 N (90% two-sided confidence interval) was calculated from the Weibull curves for group comparisons. If the Weibull use level probability calculated Beta were <1 for any group, then a Probability Weibull Contour plot (Weibull modulus versus characteristic strength) was calculated using final load magnitude to failure or survival of all groups. The Weibull modulus (90% two-sided confidence intervals) was calculated (Weibull 9++, Reliasoft, Tucson, AZ, USA) using the Fisher Matrix method. Weibull modulus ( m ) and characteristic strength Eta ( η ) (63.2% of the specimens would fail up to the calculated “ η ”) were identified for examining differences between groups.

Failed samples were first inspected in polarized light stereomicroscope (MZ-APO, Carl Zeiss Micro Imaging, Thornwood, NY, USA) and then at a scanning electron microscope (SEM) (Model 3500S, Hitachi Ltd., Osaka, Japan) for fractographic analysis. Criteria used for failure were delamination (abutment exposure), cohesive fracture within the composite (chipping), crown debond from abutment, or abutment fracture for composite crowns. For metal ceramic crowns criteria for failure was fracture within the veneering porcelain (cohesive) or exposing the metal coping.

Materials and methods

Crown fabrication

Forty-two Ti-6Al-4V abutments (Stealth abutments, shouldered, 3 mm well, 5 mm diameter, Bicon LLC, Boston, MA, USA) were selected for the study and divided in two groups ( n = 21 each) to support metal ceramic (MC) or Lava Ultimate (LU) (Lava Ultimate, 3M ESPE, St. Paul, MN, USA) maxillary first molar crowns. A waxed model maxillary first molar crown was replicated on an E4D Dentist CAD/CAM system (D4D Technologies, Richardson, TX). Crowns were milled from Lava Ultimate blocks ( n = 21, shade A3, 3M ESPE) in an E4D mill (3M ESPE), polished with diamond paste and bristle brush, then buffed to a high gloss with a cotton buff. The bonding surface of each Lava Ultimate crown was sandblasted with a 240 mesh alumina (Ney-Brasive™ J.M. Ney Co., Bloomfield, CT, USA). Crowns were cleaned by sonicating in ethanol, air dried, and their intaglio surface primed with RelyX Ceramic Primer (3M ESPE) applied with a fibertip brush, then dried with compressed air.

An anatomic metal coping ( Fig. 1 ) with a 360° collar and veneer top were designed and exported as .STL files for fabrication. Twenty-one metal copings were made via selective laser melting from non-precious cobalt–chromium dental alloy (Wirobond ® C+, Bego USA, Lincoln, RI, USA). Veneer tops for lost-wax investment and hot-pressing were made via wax printing (BeCe ® Wax-Up material, Bego USA). An opaque layer (GC™ Initial™ Opaque, GC America Inc. Alsip, IL, USA) was applied to the copings before investing them along with the wax veneer tops. After investment burnout, pressable ceramic (GC Initial™ IQ – One Body, Layering-over-Metal Pressable Ceramic) was hot-pressed directly onto the opaqued metal. Glaze (GC Initial IQ Glaze) was applied and fired on the crowns after hot-pressing by a commercial laboratory (Valley Dental Arts, Stillwater, MN). The intaglio surface of each MC crown was sandblasted with 240 mesh alumina, then cleaned by sonicating in ethanol and air dried.

Fig. 1
Stereomicroscopy micrographs of crowns embedded in epoxy resin and sectioned proximally. A) Metal ceramic crown shows a varying thickness of the supporting metal coping (between 0.5 and 1.5 mm) and an approximate thickness of 1.5 mm for the porcelain veneer. B) Resin nanoceramic crown bonded directly to the titanium abutment with a final thickness at the region underneath sliding contact fatigue of approximately 3.5 mm. a (abutment), m (metal coping), p (porcelain), RN (resin nanoceramic).

Crowns from both groups were bonded to the abutments using self-adhesive resin cement (RelyX™ Unicem™ 2 Automix Self-Adhesive Cement, 3M ESPE, St. Paul, MN, USA) where excess cement was wiped away with a microbrush. Due to their translucency, Lava Ultimate crowns were light cured 20 s each on buccal, lingual, and occusal sides with an LED curing light (Elipar S10 LED Curing Light, 3M ESPE). The cement was allowed to cure (dark cure) by resting undisturbed for at least 1 h before handling. Excess cement (oxygen inhibited) was wiped away with ethanol-soaked wipe. One additional crown of each group was fabricated and embedded in epoxy resin (Redelease, São Paulo, SP, Brazil), and polished from buccal to lingual (SiC papers #600, #1200, #2000, and #2500) to illustrate differences in veneering material thickness ( Fig. 1 ).

Crowns fixed to the implants were vertically attached to the analyzing rod of a surveyor by means of a silicone matrix which secured the crown to allow standardized positioning of samples into PVC tubes for pouring with self-curing acrylic resin (Orthodontic resin, Dentsply Caulk, Milford, DE, USA). The implant was exposed to allow its axial positioning during embedding. The sectioned PVC tube was positioned under the silicone key containing the implant/abutment/crown assembly in the center. Then, self-curing acrylic resin was poured and the analyzing rod of the surveyor lowered with the implant/abutment interface remaining 1 mm below the potting surface.

Step-stress accelerated life testing (SSALT) and reliability analysis

Three crowns from each group were subjected to single load to fracture testing with a spherical indenter (6 mm diameter, D-2 Steel) positioned at the mesio-lingual cusp at a crosshead speed of 1 mm/min (Model 800R, Test Resources, Inc., Shakopee, MN, USA) . Failure was defined as veneer material fracture. The mean load to failure values were used to determine step-stress profiles for accelerated life fatigue testing, undertaken in the remaining samples ( n = 18, each group) for probability of survival calculation. Fig. 2 shows sample distribution for mechanical testing.

Fig. 2
Schematic showing group division and number of samples distributed across SLF and step-stress accelerated life testing.

Step-stress accelerated life testing was performed in an axial direction at constant frequency of 2 Hz with the spherical indenter positioned 1 mm from the cusp tip, at the slope of the mesio-lingual cusp incline as done for the SLF test, which allowed for sliding of the indenter (0.3–0.5 mm) . Crowns were submerged in water at room temperature throughout mechanical testing. Although details of SSALT method used in this study has been reported elsewhere , in brief, three profiles are designed as mild, moderate, and aggressive, with the number of specimens assigned to each group being distributed in the ratios of 3:2:1, respectively (i.e. n = 9 in the mild, n = 6 in the moderate, and n = 3 in the aggressive load profile). These profiles are named based on the step-wise load increase that the specimen will be fatigued throughout the cycles until a certain level of load, meaning that specimens assigned to a mild profile will be cycled longer to reach the same load level of a specimen assigned to the aggressive profile. Fatigue loads throughout SSALT ranged from 200 N up to a maximum of 1150 N with a steady increase in load as a function of elapsed cycles. Samples that survived the maximum fatigue load (no fracture) were deemed suspended and accounted for in Weibull distribution.

Based upon the step-stress distribution of the failures, use level probability Weibull curves (probability of failure versus cycles) with use stress of 200 N and 90% two-sided confidence intervals were calculated and plotted (Alta Pro 9, ReliaSoft, Tucson, AZ, USA) using a power law relationship for damage accumulation. Reliability (the probability of an item functioning for a given amount of time without failure) for completion of a mission of 50,000 and 100,000 cycles at 200 N and 300 N (90% two-sided confidence interval) was calculated from the Weibull curves for group comparisons. If the Weibull use level probability calculated Beta were <1 for any group, then a Probability Weibull Contour plot (Weibull modulus versus characteristic strength) was calculated using final load magnitude to failure or survival of all groups. The Weibull modulus (90% two-sided confidence intervals) was calculated (Weibull 9++, Reliasoft, Tucson, AZ, USA) using the Fisher Matrix method. Weibull modulus ( m ) and characteristic strength Eta ( η ) (63.2% of the specimens would fail up to the calculated “ η ”) were identified for examining differences between groups.

Failed samples were first inspected in polarized light stereomicroscope (MZ-APO, Carl Zeiss Micro Imaging, Thornwood, NY, USA) and then at a scanning electron microscope (SEM) (Model 3500S, Hitachi Ltd., Osaka, Japan) for fractographic analysis. Criteria used for failure were delamination (abutment exposure), cohesive fracture within the composite (chipping), crown debond from abutment, or abutment fracture for composite crowns. For metal ceramic crowns criteria for failure was fracture within the veneering porcelain (cohesive) or exposing the metal coping.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free dental videos. Join our Telegram channel

Nov 23, 2017 | Posted by in Dental Materials | Comments Off on Probability of survival of implant-supported metal ceramic and CAD/CAM resin nanoceramic crowns

VIDEdental - Online dental courses

Get VIDEdental app for watching clinical videos