Effect of cyclic loading on the vertical microgap of long-span zirconia frameworks supported by 4 or 6 implants Rodrigo Tiossi, DDS, PhD,a Érica Alves Gomes, DDS, PhD,b Adriana Cláudia Lapria Faria, DDS, PhD,c Renata Cristina Silveira Rodrigues, DDS, PhD,d and Ricardo Faria Ribeiro, DDS, PhDe Fluminense Federal University (UFF), School of Dentistry, Niterói, Rio de Janeiro, Brazil; University of São Paulo (USP), Ribeirão Preto School of Dentistry, Ribeirão Preto, São Paulo, Brazil Statement of problem. Few studies have investigated the microgap of long-span complete-arch fixed dental prosthesis zirconia frameworks. Purpose. The purpose of this study was to evaluate the effects of cyclic loading on the vertical microgap of maxillary 12-unit prostheses supported by 4 implants and on 14-unit prostheses supported by 6 implants. Material and methods. One-piece zirconia frameworks were fabricated with a computer-aided design/computer-aided manufacturing technique and divided into 2 groups (n¼5): a group of 12-unit prostheses and a group of 14-unit prostheses. The vertical microgap between the frameworks and prosthetic abutments was measured with an optical microscope (80) under 2 reading conditions. Condition 1 (1-screw test): 1A, the screw on implant 1 was tightened and readings were made on all implants; 1B, the screw was changed to implant 4 (implant 6 for the 14-unit group) and readings were made on all implants. Condition 2: the microgap was measured with all screws tightened before cyclic loading (2A). Specimens were submitted to 200 N underwater (37 C) cyclic loading at a 2-Hz frequency for 1106 cycles. Microgap reading condition 2 was repeated after cyclic loading (2B). The data were submitted to a linear mixed-effects model for statistical comparison (a¼.05). Results. A lower (P.05). Conclusions. The vertical microgap was significantly reduced after tightening all the screws in the framework, possibly leading to a nonpassive situation. Longer-span frameworks showed an increased microgap. Cyclic loading had no influence on the vertical microgap within each group. (J Prosthet Dent 2014;-:---)
Clinical Implications Some concern exists that long-span zirconia frameworks present higher vertical microgaps because of uncontrolled sintering shrinkage. A comparison of 12-unit and 14-unit-span zirconia frameworks found a higher misfit for the longer ones. The results also indicate that cyclic loading did not influence the microgap. Supported by Research Grant Nos. 2010/02218-5, 2011/01506-0, 2010/00124-3, and 2010/19221-9 from the Foundation for Research Support of the State of São Paulo (FAPESP). a
Assistant Professor, Department of Prosthodontics, School of Dentistry, Fluminense Federal University. Postdoctoral Research Associate, Department of Dental Materials and Prosthodontics, University of São Paulo Ribeirão Preto School of Dentistry. c Laboratory Specialist, Department of Dental Materials and Prosthodontics, University of São Paulo Ribeirão Preto School of Dentistry. d Assistant Professor, Department of Dental Materials and Prosthodontics, University of São Paulo Ribeirão Preto School of Dentistry. e Full Professor, Department of Dental Materials and Prosthodontics, University of São Paulo Ribeirão Preto School of Dentistry. b
Tiossi et al
2
Volume Dental implants have changed the treatment strategies for the oral rehabilitation of edentulous patients1,2 and provided predictable treatment alternatives.3 However, mechanical and technical difficulties are still present.3-5 Fabrication procedures such as impression making,6 definitive cast pouring,7 and framework casting8-10 can still be inaccurate. The nonpassive fitting of an implant-supported prosthesis with an increased vertical microgap may induce higher stresses to the supporting structures and contribute to mechanical and biologic complications.11-17 The computer-aided design/computer-aided manufacturing (CAD/ CAM) method for framework fabrication has greatly minimized the distortion inherent in conventional fabrication methods.11 The fabrication of titanium frameworks with a CAD/CAM technique provides precise and repeatable outcomes.18-21 However, the high cutting temperatures of titanium could lead to distortions22 and impair tool life and surface quality. This could lead to low cutting accuracy due to the thermal expansion of the tool and the workpiece.22 The high demand for esthetics has led to the increased use of zirconia frameworks.1,23 Various CAM methods are available for fabricating zirconia frameworks. Zirconium oxide (ZrO2) can be milled after complete sintering, and zirconia linear shrinkage of 20% to 25% will occur before the milling procedure.1 Most companies fabricate the framework from a presintered, softer ZrO2 block; this process is faster, and the diamond rotary instrument used in the milling process wears less.1,24 However, the linear shrinkage after the sintering process leads to further framework distortion, and a higher microgap is expected. Zirconia frameworks may also damage the titanium interface of implants or the prosthetic-abutment titanium interface during function.25 A previous study26 found no significant differences when the implant-abutment microgap of zirconia, alumina, and titanium abutments was compared after dynamic loading. Titanium implants
showed higher wear at the implant interface after cyclic loading when connected to zirconia abutments compared with those connected to titanium abutments.25 The vertical microgap of short-span zirconia fixed dental prosthesis frameworks has been considered sufficiently precise.27 Few studies have investigated the microgap of long-span completearch fixed dental prosthesis (CAFDP) zirconia frameworks.1,28 Four implants are considered the minimum number needed to restore a completely edentulous patient with a CAFDP replacing 10 to 12 units.29 This study evaluated the effects of cyclic loading on the vertical microgap of maxillary 12-unit CAFDPs supported by 4 implants and 14-unit CAFDPs supported by 6 implants. The tested null hypotheses were that the vertical microgaps of the 12unit and 14-unit CAFDP frameworks would present no significant differences and that cyclic loading would have no effect on the vertical microgap of the tested specimens.
MATERIAL AND METHODS A rapid prototyping method was used to fabricate a maxilla prototype from a computed tomography (CT) scan of a patient previously treated with implant-supported restorations (University of São Paulo, Ethics Commission approval No. 2011.1.608.58.3). Informed consent was obtained from the patient before using the CT scan
The Journal of Prosthetic Dentistry
-
Issue
-
and the patient records. Two prototypes of the same maxilla were fabricated. One prototype had 6 implants placed according to the “All-on-Six”30 treatment concept for rehabilitation of the maxilla, and the other had 4 implants placed according to the “All-onFour”31 treatment concept. The location of the dental implants was virtually planned in a 3-dimensional implant planning software (DentalSlice; BioParts Prototipagem Biomédica) for computer-assisted surgery. A maxillary stereolithographic surgical template (NeoGuide Cirurgia Guiada; Neodent) was fabricated (BioParts Prototipagem Biomédica) from the virtual planning to allow the correct placement of the implants in the maxilla prototypes. Prosthetic abutments (Mini conical abutments; Neodent) were attached to the implants and tightened to 20 Ncm. Impression copings were connected to each other with acrylic resin (DuraLay; Reliance Dental Mfg Co) and were used to transfer the implant positions correctly to the 10 polyurethane casts poured (F-16 FastCast Polyurethane; Axson). To minimize distortion, the acrylic resin was sectioned after polymerization and reconnected. Implant replicas were embedded in the polyurethane casts, and the prosthetic abutments were screwed to the implant replicas in the same position as in the initial maxilla prototype (Fig. 1). The casts of the patient’s maxilla and mandible were mounted on a semiadjustable articulator (A7Plus; Bioart) in
1 Maxilla replicas restored with 6 and 4 implants.
Tiossi et al
-
2014
3
maximum intercuspal position. The mandibular arch had been previously restored with an implant-fixed dental prosthesis. Burnout prosthetic cylinders (Burnout cylinder for mini conical abutments; Neodent) were screwed to the abutments, and a 12-unit and a 14unit implant-supported prosthesis were waxed. The wax patterns were reduced to the dimensions of the prosthetic frameworks, and digital impressions (3Shape D700; 3Shape A/S) were made. The frameworks were individually fabricated for each polyurethane cast by using a CAD/CAM method (Neoshape; Neodent). Standardized crowns were fabricated for the frameworks with pressable porcelain (Luminesse Low-Fusing & Pressable Porcelain; Talladium Inc). The
maxilla with 4 implants was restored with a 12-unit CAFDP (all-on-four) (group AO4), and the maxilla with 6 implants was restored with a 14-unit CAFDP (all-on-six) (group AO6) (n¼5). Interim luting agent (RelyX Temp; 3M ESPE Dental Products) was used for the cementation of the pressed porcelain crowns (Fig. 2). The vertical microgap between the CAFDPs with the prosthetic abutments was evaluated under 2 reading conditions (Sheffield test)32: Condition 1 (1-screw test): 1A, the screw located on implant 1 was tightened, and the readings were made on all the implants, including implant 1; 1B, the screw on implant 4 (implant 6 for AO6) was tightened and the readings were again made on all the implants,
2 Twelve-unit complete-arch fixed dental prosthesis specimen supported by 4 implants.
including implant 4 (implant 6 for AO6). Condition 2 (all screws tightened): 2A, the vertical microgap was measured after tightening all screws in the framework. The screws were tightened to 10 Ncm for all reading conditions. An optical microscope at 80 and a precision of 1 mm (Leica DFC295 attached to a Leica S8 APO; Leica Microsystems) was used for the measurements. The vertical microgap was inspected around each implant on the lingual surface of the CAFDP frameworks for each reading condition (Fig. 3); 3 readings were made on each cylinder to ensure the precision of the microgap measurements. A calibrated 200 N cyclic loading was applied to all specimens at 2 Hz for 1106 cycles (40 months of simulated function)33 in an electromechanical mastication fatigue testing machine (MSFM; Elquip, Equipamentos para Pesquisa Odontológica). The opposing mandibular dental arch was fabricated with bis-acrylic composite resin (Luxatemp; DMG Dental-Material GmbH), copying the anatomy of the patient’s initial implant-supported mandibular complete-arch prosthesis. The opposing occlusion was adjusted for each specimen before cyclic loading. Both the replicas of the maxilla and mandible and the CAFDP specimens were immersed in distilled water under constant circulation at 37 2 C for the fatigue testing. Only microgap reading condition 2 was repeated after the cyclic loadings for all specimens in the study (2B).
3 A, Specimen under microscope for microgap measurement. B, View of microgap measurement showing readings made on each abutment.
Tiossi et al
4
Volume The statistical comparison between the groups was performed by using orthogonal contrasts in a linear mixedeffects model (a¼.05). The linear mixed-effects model was selected because the results found for 1 specimen were grouped and the assumption of independence between the measurements within the same group was inadequate.34 The residuals should present normal distribution with a mean equal to zero and constant variance for the linear mixed-effects model. Because this was not the case for the results found in the study, a logarithmic transformation of the data was applied to the results. This transformation succeeded in providing a normal distribution and equal variances. The model was adjusted with the PROC MIXED function in specialized statistical software (SAS 9.1; SAS Institute Inc).
RESULTS Three microgap measurements were made on each cylinder for this study. The intraclass correlation coefficient between measurements was calculated for each reading condition, and the results are as follows: 1A, 0.992; 1B, 0.996; 2A, 0.910; and 2B, 0.896. Table I presents the mean and standard deviation results found for the vertical microgap (mm) measurements before the cyclic loadings and under all reading conditions. The statistical comparison (a¼.05) between the groups tested in this study before the cyclic loadings and for all reading conditions is presented in Table II. The CAFDPs supported by 4 implants (AO4) presented a significantly lower mean microgap than the CAFDPs supported by 6 implants (AO6) under all reading conditions (P
Clinical Implications Some concern exists that long-span zirconia frameworks present higher vertical microgaps because of uncontrolled sintering shrinkage. A comparison of 12-unit and 14-unit-span zirconia frameworks found a higher misfit for the longer ones. The results also indicate that cyclic loading did not influence the microgap. Supported by Research Grant Nos. 2010/02218-5, 2011/01506-0, 2010/00124-3, and 2010/19221-9 from the Foundation for Research Support of the State of São Paulo (FAPESP). a
Assistant Professor, Department of Prosthodontics, School of Dentistry, Fluminense Federal University. Postdoctoral Research Associate, Department of Dental Materials and Prosthodontics, University of São Paulo Ribeirão Preto School of Dentistry. c Laboratory Specialist, Department of Dental Materials and Prosthodontics, University of São Paulo Ribeirão Preto School of Dentistry. d Assistant Professor, Department of Dental Materials and Prosthodontics, University of São Paulo Ribeirão Preto School of Dentistry. e Full Professor, Department of Dental Materials and Prosthodontics, University of São Paulo Ribeirão Preto School of Dentistry. b
Tiossi et al
2
Volume Dental implants have changed the treatment strategies for the oral rehabilitation of edentulous patients1,2 and provided predictable treatment alternatives.3 However, mechanical and technical difficulties are still present.3-5 Fabrication procedures such as impression making,6 definitive cast pouring,7 and framework casting8-10 can still be inaccurate. The nonpassive fitting of an implant-supported prosthesis with an increased vertical microgap may induce higher stresses to the supporting structures and contribute to mechanical and biologic complications.11-17 The computer-aided design/computer-aided manufacturing (CAD/ CAM) method for framework fabrication has greatly minimized the distortion inherent in conventional fabrication methods.11 The fabrication of titanium frameworks with a CAD/CAM technique provides precise and repeatable outcomes.18-21 However, the high cutting temperatures of titanium could lead to distortions22 and impair tool life and surface quality. This could lead to low cutting accuracy due to the thermal expansion of the tool and the workpiece.22 The high demand for esthetics has led to the increased use of zirconia frameworks.1,23 Various CAM methods are available for fabricating zirconia frameworks. Zirconium oxide (ZrO2) can be milled after complete sintering, and zirconia linear shrinkage of 20% to 25% will occur before the milling procedure.1 Most companies fabricate the framework from a presintered, softer ZrO2 block; this process is faster, and the diamond rotary instrument used in the milling process wears less.1,24 However, the linear shrinkage after the sintering process leads to further framework distortion, and a higher microgap is expected. Zirconia frameworks may also damage the titanium interface of implants or the prosthetic-abutment titanium interface during function.25 A previous study26 found no significant differences when the implant-abutment microgap of zirconia, alumina, and titanium abutments was compared after dynamic loading. Titanium implants
showed higher wear at the implant interface after cyclic loading when connected to zirconia abutments compared with those connected to titanium abutments.25 The vertical microgap of short-span zirconia fixed dental prosthesis frameworks has been considered sufficiently precise.27 Few studies have investigated the microgap of long-span completearch fixed dental prosthesis (CAFDP) zirconia frameworks.1,28 Four implants are considered the minimum number needed to restore a completely edentulous patient with a CAFDP replacing 10 to 12 units.29 This study evaluated the effects of cyclic loading on the vertical microgap of maxillary 12-unit CAFDPs supported by 4 implants and 14-unit CAFDPs supported by 6 implants. The tested null hypotheses were that the vertical microgaps of the 12unit and 14-unit CAFDP frameworks would present no significant differences and that cyclic loading would have no effect on the vertical microgap of the tested specimens.
MATERIAL AND METHODS A rapid prototyping method was used to fabricate a maxilla prototype from a computed tomography (CT) scan of a patient previously treated with implant-supported restorations (University of São Paulo, Ethics Commission approval No. 2011.1.608.58.3). Informed consent was obtained from the patient before using the CT scan
The Journal of Prosthetic Dentistry
-
Issue
-
and the patient records. Two prototypes of the same maxilla were fabricated. One prototype had 6 implants placed according to the “All-on-Six”30 treatment concept for rehabilitation of the maxilla, and the other had 4 implants placed according to the “All-onFour”31 treatment concept. The location of the dental implants was virtually planned in a 3-dimensional implant planning software (DentalSlice; BioParts Prototipagem Biomédica) for computer-assisted surgery. A maxillary stereolithographic surgical template (NeoGuide Cirurgia Guiada; Neodent) was fabricated (BioParts Prototipagem Biomédica) from the virtual planning to allow the correct placement of the implants in the maxilla prototypes. Prosthetic abutments (Mini conical abutments; Neodent) were attached to the implants and tightened to 20 Ncm. Impression copings were connected to each other with acrylic resin (DuraLay; Reliance Dental Mfg Co) and were used to transfer the implant positions correctly to the 10 polyurethane casts poured (F-16 FastCast Polyurethane; Axson). To minimize distortion, the acrylic resin was sectioned after polymerization and reconnected. Implant replicas were embedded in the polyurethane casts, and the prosthetic abutments were screwed to the implant replicas in the same position as in the initial maxilla prototype (Fig. 1). The casts of the patient’s maxilla and mandible were mounted on a semiadjustable articulator (A7Plus; Bioart) in
1 Maxilla replicas restored with 6 and 4 implants.
Tiossi et al
-
2014
3
maximum intercuspal position. The mandibular arch had been previously restored with an implant-fixed dental prosthesis. Burnout prosthetic cylinders (Burnout cylinder for mini conical abutments; Neodent) were screwed to the abutments, and a 12-unit and a 14unit implant-supported prosthesis were waxed. The wax patterns were reduced to the dimensions of the prosthetic frameworks, and digital impressions (3Shape D700; 3Shape A/S) were made. The frameworks were individually fabricated for each polyurethane cast by using a CAD/CAM method (Neoshape; Neodent). Standardized crowns were fabricated for the frameworks with pressable porcelain (Luminesse Low-Fusing & Pressable Porcelain; Talladium Inc). The
maxilla with 4 implants was restored with a 12-unit CAFDP (all-on-four) (group AO4), and the maxilla with 6 implants was restored with a 14-unit CAFDP (all-on-six) (group AO6) (n¼5). Interim luting agent (RelyX Temp; 3M ESPE Dental Products) was used for the cementation of the pressed porcelain crowns (Fig. 2). The vertical microgap between the CAFDPs with the prosthetic abutments was evaluated under 2 reading conditions (Sheffield test)32: Condition 1 (1-screw test): 1A, the screw located on implant 1 was tightened, and the readings were made on all the implants, including implant 1; 1B, the screw on implant 4 (implant 6 for AO6) was tightened and the readings were again made on all the implants,
2 Twelve-unit complete-arch fixed dental prosthesis specimen supported by 4 implants.
including implant 4 (implant 6 for AO6). Condition 2 (all screws tightened): 2A, the vertical microgap was measured after tightening all screws in the framework. The screws were tightened to 10 Ncm for all reading conditions. An optical microscope at 80 and a precision of 1 mm (Leica DFC295 attached to a Leica S8 APO; Leica Microsystems) was used for the measurements. The vertical microgap was inspected around each implant on the lingual surface of the CAFDP frameworks for each reading condition (Fig. 3); 3 readings were made on each cylinder to ensure the precision of the microgap measurements. A calibrated 200 N cyclic loading was applied to all specimens at 2 Hz for 1106 cycles (40 months of simulated function)33 in an electromechanical mastication fatigue testing machine (MSFM; Elquip, Equipamentos para Pesquisa Odontológica). The opposing mandibular dental arch was fabricated with bis-acrylic composite resin (Luxatemp; DMG Dental-Material GmbH), copying the anatomy of the patient’s initial implant-supported mandibular complete-arch prosthesis. The opposing occlusion was adjusted for each specimen before cyclic loading. Both the replicas of the maxilla and mandible and the CAFDP specimens were immersed in distilled water under constant circulation at 37 2 C for the fatigue testing. Only microgap reading condition 2 was repeated after the cyclic loadings for all specimens in the study (2B).
3 A, Specimen under microscope for microgap measurement. B, View of microgap measurement showing readings made on each abutment.
Tiossi et al
4
Volume The statistical comparison between the groups was performed by using orthogonal contrasts in a linear mixedeffects model (a¼.05). The linear mixed-effects model was selected because the results found for 1 specimen were grouped and the assumption of independence between the measurements within the same group was inadequate.34 The residuals should present normal distribution with a mean equal to zero and constant variance for the linear mixed-effects model. Because this was not the case for the results found in the study, a logarithmic transformation of the data was applied to the results. This transformation succeeded in providing a normal distribution and equal variances. The model was adjusted with the PROC MIXED function in specialized statistical software (SAS 9.1; SAS Institute Inc).
RESULTS Three microgap measurements were made on each cylinder for this study. The intraclass correlation coefficient between measurements was calculated for each reading condition, and the results are as follows: 1A, 0.992; 1B, 0.996; 2A, 0.910; and 2B, 0.896. Table I presents the mean and standard deviation results found for the vertical microgap (mm) measurements before the cyclic loadings and under all reading conditions. The statistical comparison (a¼.05) between the groups tested in this study before the cyclic loadings and for all reading conditions is presented in Table II. The CAFDPs supported by 4 implants (AO4) presented a significantly lower mean microgap than the CAFDPs supported by 6 implants (AO6) under all reading conditions (P
