Three-Dimensional Accuracy of Plastic Transfer Impression Copings for Three Implant Systems Juin Wei Teo, BDS, MDS1/Keson B. Tan, BDS, MSD2/Jack I. Nicholls, PhD3/ Keng Mun Wong, BDS, MSD4/Joanne Uy, DDM, MDS4 Purpose: The purpose of this study was to compare the three-dimensional accuracy of indirect plastic impression copings and direct implant-level impression copings from three implant systems (Nobel Biocare [NB], Biomet 3i [3i], and Straumann [STR]) at three interimplant buccolingual angulations (0, 8, and 15 degrees). Materials and Methods: Two-implant master models were used to simulate a three-unit implant fixed partial denture. Test models were made from Impregum impressions using direct implant-level impression copings (DR). Abutments were then connected to the master models for impressions using the plastic impression copings (INDR) at three different angulations for a total of 18 test groups (n = 5 in each group). A coordinate measuring machine was used to measure linear distortions, three-dimensional (3D) distortions, angular distortions, and absolute angular distortions between the master and test models. Results: Three-way analysis of variance showed that the implant system had a significant effect on 3D distortions and absolute angular distortions in the x- and y-axes. Interimplant angulation had a significant effect on 3D distortions and absolute angular distortions in the y-axis. Impression technique had a significant effect on absolute angular distortions in the y-axis. With DR, the NB and 3i systems were not significantly different. With INDR, 3i appeared to have less distortion than the other systems. Interimplant angulations did not significantly affect the accuracy of NBDR, 3iINDR, and STRINDR. The accuracy of INDR and DR was comparable at all interimplant angulations for 3i and STR. For NB, INDR was comparable to DR at 0 and 8 degrees but was less accurate at 15 degrees. Conclusions: Three-dimensional accuracy of implant impressions varied with implant system, interimplant angulation, and impression technique. Int J Oral Maxillofac Implants 2014;29:577–584. doi: 10.11607/jomi.3382 Key words: accuracy, implant impression, plastic coping, three-dimensional
M
isfit or distortion in implant prostheses may arise from errors incurred during impression making. The factors affecting the accuracy of the impression may include the impression material, polymerization shrinkage, and impression technique,1–9 as well as the machining tolerance of the implant components.10 1Registrar,
Dental Department, Khoo Teck Puat Hospital, Republic of Singapore. 2 Associate Professor, Faculty of Dentistry, National University of Singapore, Republic of Singapore. 3 Professor Emeritus, Department of Restorative Dentistry, School of Dentistry, University of Washington, Seattle, Washington. 4 Adjunct Senior Lecturer, Faculty of Dentistry, National University of Singapore, Republic of Singapore. Correspondence to: Dr Keson B. Tan, Faculty of Dentistry, National University of Singapore, 11 Lower Kent Ridge Road, Singapore 119083, Republic of Singapore. Fax: +65-67785742. Email: [email protected] ©2014 by Quintessence Publishing Co Inc.
While most studies on implant impressions were based on implants placed parallel to each other, several studies have also investigated divergent implants. Phillips et al2 incorporated a 10-degree divergence in each of the implant abutments in a five-implant model and found cross-arch distortions ranging from 22 to 104 µm. Carr11 evaluated direct and indirect impression techniques with implants at 15 degrees of divergence and found no significant difference. Jang et al12 investigated the accuracy of implant impressions with divergences of up to 20 degrees and found that distortions in the 20-degree divergent models were significantly greater than those in the 15-degree divergent models. However, these studies were based on older implant systems that used both open-tray and closed-tray techniques and screw-on metallic impression copings. More recently, many implant systems have introduced plastic impression copings for abutment-level impressions to simplify impression making and minimize the amount of abutment reconnection/disconnection needed, since this can result not only in tissue The International Journal of Oral & Maxillofacial Implants 577
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Teo et al
Z
Y
X a
The aim of this study was to compare the threedimensional (3D) accuracy of plastic impression copings with the direct implant-level metal impression copings from three implant systems: Brånemark, Nobel Biocare (NB); Nanotite Certain Implant System, Biomet 3i (3i); and Tissue Level Implant System, Straumann (STR). The effects of the three implant systems; buccolingual interimplant angulations of 0, 8, and 15 degrees; and the two impression techniques (indirect plastic [INDR] vs direct metal [DR]) were also determined and compared.
b Fig 1 Master model with gauge block (A). Point B is the origin of the local coordinate system.
MATERIALS AND METHODS Master Model
irritation13 but also changes in abutment location in relation to the implant.14 However, the newer plastic impression copings lack rigidity, and their “snap-on” connection to the implant components can potentially allow deformation under the forces applied during impression making.15 These forces and deformations can be greatly increased when an impression is pulled away from nonparallel abutments. To date, only two studies have investigated the accuracy of plastic impression copings. Walker et al16 studied the accuracy of indirect implant-level impressions using screw-on metal copings and compared it to that of abutmentlevel impressions using snap-on plastic copings in the Nobel Replace implant system. They used a microscope to measure interimplant distances and found that the indirect metal copings were more accurate than the plastic snap-on copings. The percentage difference in the interimplant distances between the master model and casts made from the indirect metal copings ranged from –6% to 2%. The percentage difference in the interabutment distances between the master model and casts made from the plastic snap-on copings ranged from 19% to 24%. However, Akça and Cehreli17 found no significant differences between the metal and plastic impression copings of the ITI Straumann Tissue Level implant system. Most implant systems have introduced simplified impression techniques with plastic snap-on abutmentlevel impression copings. The use of such impression systems is recommended by the manufacturers not only for single implant restorations but also in cases where two or more abutments are involved. Limitations regarding their use, in terms of the number of abutments and the divergence of abutments, have not been highlighted by the manufacturers or reported in the literature. At present, studies of their accuracy and comparisons between different implant systems are still lacking in the literature.
The master models were fabricated using an acrylic resin (Trevalon, Dentsply Corp) partial mandibular dental arch attached to an aluminum block incorporating a gauge block (Part No:MCL-VP100, Matchling Tooling) that defined the local coordinate system. Two implants were placed, in the first premolar and first molar positions, to simulate a three-unit implant prosthesis (Fig 1). Three master models were made of each implant system, and the interimplant angulations were set at 0, 8, and 15 degrees using a universal vice (VUA3, Vertex Machinery Works) and a drill press (Universal Drilling Machine, ARICAS). Implants were secured with self-curing acrylic resin (Orthocryl, Dentaurum) and allowed to set for 24 hours before measurements were made. With the three implant systems (NB, 3i, STR) and three interimplant angulations, a total of nine master models were fabricated.
Implant Impressions
Impressions were made using direct impression copings (DR) connected to the implants (Fig 2). These copings were tightened to a torque of 15 Ncm using a calibrated torque gauge (6BTG, Tohnichi). The respective abutments (Nobel Biocare Snappy, Biomet 3i Provide, and Straumann Solid) were then connected to the implants and a tightening torque was applied to each abutment in accordance with the manufacturer’s instructions (45 Ncm for the NB groups, 20 Ncm for the 3i groups, and 35 Ncm for the STR groups). The abutments were retightened again to the recommended torque values to ensure that the proper torque was achieved and to minimize settling effects.18 The respective plastic snap-on impression copings were then attached (Fig 3), and an indirect impression was made (INDR). The recommended torque was reapplied after every impression to ensure that the abutments remained at the correct level of torque at all times. All impressions were made with a polyether elastomeric impression material (Impregum, 3M/ESPE)19,20 using
578 Volume 29, Number 3, 2014 © 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Teo et al
Fig 2 Direct implant-level impression copings (DR) attached.
Fig 3 Plastic abutment-level impression copings (INDR) attached to prefabricated abutments. System illustrated is Snappy Abutment (Nobel Biocare).
Probe hit
A
B
Probe hit Fig 4 Measured plane on an implant. Six probe hits were taken on the implant platform to define the plane.
Fig 5 Measured cylinder on an implant. The cylinder was defined by two circles approximately 1 mm apart, each of which was defined by six probe hits. (Not shown: Prefabricated abutment cone features were defined by two circles approximately 3 mm apart.)
custom trays (Tray Resin II, Shofu). The impressions were poured with vacuum-mixed type IV stone (Silky Rock, Whip Mix Corporation), mixed according to the manufacturer’s directions (100 g powder to 23 mL water), and allowed to set for 60 minutes before the test models were separated. For each test group, five impressions were made to yield five test models (n = 5).
Coordinate Measuring Machine
All measurements were made using a coordinate measuring machine (CMM) (Model TESA MH3D DCC, Brown and Sharpe) with an accuracy of 2 µm. For each master model, the local coordinate axes were defined on the gauge block with four probe hits on each of the three exposed orthogonal metal surfaces. The intersection point of these three planes defined the origin for the local coordinate system for each master model. The stone surfaces on the replica of the gauge block in the test models were also used to define the origin and transferred local coordinate axis system. Pilot studies
Fig 6 The intersection points between the measured planes and cylinders (NB and 3i systems) or cones (STR system) defined the centroid positions of the molar (point A [x M1, yM1, and zM1]) and premolar (point B [x M2, yM2, and zM2]) implants.
found that the flatness of the metal gauge block planes and that of the stone replica were not greater than 3 µm and 5 µm, respectively. For the DR groups, a plane was defined on the platform of each implant on the master model and implant analog on the test model by taking six probe hits (Fig 4). A cylinder or cone was then defined on the implant or implant analog by measuring two circles of six probe hits each (Fig 5). For the INDR groups, a plane was defined by taking six hits on the top surface of each abutment on the master model and abutment analog on the test model. A cone was then defined by two circles of six probe hits each around the abutment or abutment analog. The intersection point of the central axis of the measured cone/cylinder and the plane gave the centroid coordinates for the molar implant or abutment (xM1, yM1, and zM1) and the premolar implant or abutment (xM2, yM2, and zM2) (Fig 6). Similar measurements made on the test model analogs gave the centroid coordinates for the molar implant or abutment The International Journal of Oral & Maxillofacial Implants 579
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Teo et al
Table 1 Mean Linear Distortions (Standard Deviations) in Millimeters Group
dy
dz
dR
NBDR0
0.022 (0.008)
dx
–0.010 (0.008)
0.005 (0.005)
0.026 (0.008)
NBDR8
–0.004 (0.014)
–0.018 (0.024)
–0.004 (0.011)
0.031 (0.012)
NBDR15
0.017 (0.015)
0.003 (0.017)
–0.011 (0.010)
0.028 (0.014)
NBINDR0
0.009 (0.008)
–0.009 (0.021)
0.023 (0.016)
0.034 (0.011)
NBINDR 8
0.006 (0.009)
–0.021 (0.018)
0.002 (0.043)
0.043 (0.022)
NBINDR 15
0.054 (0.021)
0.002 (0.019)
0.002 (0.025)
0.061 (0.019)
3iDR0
0.016 (0.029)
0.014 (0.028)
0.018 (0.013)
0.045 (0.014)
3iDR8
–0.004 (0.018)
–0.012 (0.015)
–0.004 (0.009)
0.024 (0.011)
3iDR15
0.033 (0.013)
0.038 (0.056)
0.025 (0.018)
0.075 (0.010)
3iINDR0
0.012 (0.028)
–0.004 (0.051)
–0.010 (0.013)
0.053 (0.017)
3iINDR8
–0.001 (0.023)
0.009 (0.035)
–0.008 (0.023)
0.039 (0.023)
0.003 (0.038)
0.010 (0.030)
0.007 (0.015)
0.040 (0.029)
STRDR0
3iINDR15
–0.018 (0.014)
0.031 (0.027)
0.007 (0.042)
0.055 (0.025)
STRDR8
0.032 (0.023)
–0.027 (0.054)
–0.013 (0.019)
0.070 (0.007)
STRDR 15
0.057 (0.048)
0.078 (0.028)
0.013 (0.037)
0.109 (0.039)
STRINDR0
–0.045 (0.008)
0.004 (0.025)
0.003 (0.023)
0.054 (0.011)
STRINDR8
–0.024 (0.042)
0.009 (0.053)
–0.009 (0.022)
0.066 (0.021)
STRINDR15
–0.014 (0.030)
0.057 (0.028)
0.020 (0.012)
0.071 (0.017)
• Linear distortion in the y-axis (dy) = (yT2 – yT1) – (yM2 – yM1) • Linear distortion in the z-axis (dz) = (zT2 – zT1) – (zM2 – zM1) • Global 3D distortion (dR) = √(dx2 + dy2 + dz2)
0.140 Distortion (mm)
0.120 0.100 0.080 0.060 0.040
0.000
NBD RO NBD R8 NBD R15 3iDR O 3iDR 8 3iDR 15 STRD RO STRD R8 STRD R15 NBIN DRO NBIN DR8 NBIN DR1 5 3iIND RO 3iIND R8 3iIND R1 STRIN 5 DR STRIN O DR8 STRIN DR1 5
0.020
Test group Fig 7 Mean global 3D distortions (dR). Error bars indicate standard deviations.
analog (xT1, yT1, and zT1) and the premolar implant or abutment analog (xT2, yT2, and zT2). For each measured plane and cone/cylinder feature, the respective flatness or roundness criterion was set at no greater than 5 µm. All components used and all features measured fulfilled these criteria. Linear distortions in the three axes and global 3D distortion were defined as follows: • Linear distortion in the x-axis (dx) = (xT2 – xT1) – (xM2 – xM1)
The angular deviation of each implant in relation to the x-axis (xz) and y-axis (yz) were calculated from the measured planes. Hence, angular deviations were given as xzM1 and yzM1 for the molar implant and xzM2 and yzM2 for the premolar implant on the master model. Similarly, the angular deviations of the implant analogs on the test models were given as xzT1 and yzT1 for the molar analog and xzT2 and yzT2 for the premolar analog. Angular distortions with regard to the x-axis and y-axis between the master and test models were calculated as follows. • Angular distortions with regard to the x-axis (dθx) = (xzT2 – xzT1) – (xzM2 – xzM1) • Angular distortions with regard to the y-axis (dθy) = (yzT2 – yzT1) – (yzM2 – yzM1) Because angular distortions measured relative to the local coordinate system can be either positive or negative, the mean angular distortions may not be a true representation of the absolute amount of distortion present in a group. Hence, the mean absolute angular distortions (ABSdθx and ABSdθy) were reported.
580 Volume 29, Number 3, 2014 © 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Teo et al
Table 2 Mean Angular Distortions (Standard Deviations) in Degrees Group
dθx
dθy
ABSdθx
ABSdθy
NBDR0
–0.001 (0.076)
0.041 (0.067)
0.065 (0.021)
0.069 (0.025)
NBDR8
–0.023 (0.112)
0.082 (0.153)
0.086 (0.061)
0.124 (0.113)
NBDR15
–0.012 (0.073)
0.129 (0.118)
0.058 (0.036)
0.129 (0.118)
NBINDR0
0.284 (0.286)
0.007 (0.155)
0.219 (0.203)
0.106 (0.101)
NBINDR 8
–0.158 (0.221)
–0.064 (0.249)
0.160 (0.220)
0.199 (0.133)
NBINDR 15
0.477 (0.287)
0.503 (0.233)
0.477 (0.287)
0.503 (0.233)
3iDR0
0.010 (0.212)
–0.059 (0.070)
0.150 (0.131)
0.078 (0.041)
3iDR8
0.011 (0.135)
0.185 (0.086)
0.107 (0.062)
0.185 (0.086)
3iDR15
0.096 (0.151)
0.226 (0.125)
0.120 (0.127)
0.226 (0.125)
3iINDR0
0.049 (0.289)
0.291 (0.136)
0.219 (0.163)
0.291 (0.136)
3iINDR8
0.078 (0.162)
–0.205 (0.192)
0.138 (0.100)
0.205 (0.192)
3iINDR15
–0.016 (0.242)
0.176 (0.165)
0.198 (0.098)
0.176 (0.165) 0.185 (0.108)
0.677 (0.526)
0.183 (0.114)
0.777 (0.307)
STRDR8
0.520 (0.520)
–0.122 (0.364)
0.572 (0.446)
0.292 (0.211)
–0.658 (0.731)
0.235 (0.467)
0.694 (0.689)
0.449 (0.187)
STRINDR0
0.336 (0.169)
–0.009 (0.166)
0.336 (0.169)
0.129 (0.082)
STRINDR8
–0.145 (0.351)
0.262 (0.511)
0.328 (0.122)
0.420 (0.355)
STRINDR15
–0.170 (0.308)
0.407 (0.157)
0.235 (0.249)
0.407 (0.157)
Prior to this study, a feasibility test was performed to verify the dimensional stability of the acrylic resin master model over a period of 245 days. The measured linear interimplant dimension varied within a range of only 3 µm over this period.
Statistical Analyses
All data were subjected to three-way analysis of variance (ANOVA) and subsequent one-way ANOVA for the independent variables SYSTEM, ANGLE, and TECHNIQUE. The dependent variables were the linear distortions (dx, dy, dz, and dR) and the angular distortions (dθx and dθy) as well as the absolute angular distortions (ABSdθx and ABSdθy). Post hoc analyses were performed using the Tukey test at a 5% significance level. All analyses were performed using SPSS v18.0 (SPSS).
RESULTS For the test groups measured, the mean dx ranged from –0.045 to 0.054 mm, mean dy ranged from –0.027 to 0.078 mm, and mean dz ranged from –0.013 to 0.025 mm. Mean dR ranged from 0.024 to 0.109 mm (Table 1, Fig 7). Mean dθx ranged from –0.658 to 0.677 degrees and mean dθy ranged from –0.205 to 0.503 degrees. Mean ABSdθx ranged from 0.058 to 0.777 degrees, and mean ABSdθy ranged from 0.069 to 0.503 degrees (Table 2, Fig 8).
1.0
ABSdθx
0.8
ABSdθy
0.6 0.4 0.2 0.0
NBD RO NBD R8 NBD R15 3iDR O 3iDR 8 3iDR 15 STRD RO STRD R8 STRD R15 NBIN DRO NBIN DR8 NBIN DR1 5 3iIND RO 3iIND R8 3iIND R15 STRIN DRO STRIN DR8 STRIN DR1 5
STRDR 15
Distortion (degree)
STRDR0
Test group Fig 8 Mean absolute angular distortions (ABSdθx and ABSdθy). Error bars indicate standard deviations.
Three-way ANOVA showed that SYSTEM had a significant effect on dX, dY, dR, ABSdθx, and ABSdθy. ANGLE had a significant effect on dx, dy, dz, dR, and ABSdθy. TECHNIQUE had a significant effect on dx and ABSdθy. For comparisons among the implant systems using the direct impression technique (DR groups), the STR groups showed significantly higher distortions at all three interimplant angulations tested. At 15 degrees of interimplant angulation, significantly lower dR was found in NBDR15. For the INDR groups, no significant differences were found between the three implant systems at all three interimplant angulations tested. The International Journal of Oral & Maxillofacial Implants 581
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Teo et al
When comparing the three interimplant angulations, no significant differences were found for all distortions in the NBDR, 3iINDR, and STRINDR groups. In the NBINDR groups, ABSdθy for NBINDR15 was significantly higher than for NBINDR0, but not significantly different between NBINDR0 and NBINDR8. dR and ABSdθx were not significantly different among the NBINDR groups. In the 3iDR and STRDR groups, dR was found to be significantly higher in 3iDR15 and STRDR15, but no significant differences were found for ABSdθx and ABSdθy. Comparisons were made between the two impression techniques of the NB implant system at each angulation (NB0, NB8, NB15). ABSdθy of NBINDR15 was significantly higher than NBDR15, but it was not significantly different between NBDR0 and NBINDR0 or between NBDR8 and NBINDR8. No significant differences were found for dR and ABSdθx between the impression techniques of NB0, NB8, and NB15. For comparisons between the two impression techniques of 3i at each angulation (3i0, 3i8, and 3i15), no significant differences were found for all translational and angular distortions between the two impression techniques. For comparisons made between the two impression techniques of STR at each angulation (STR0, STR8, and STR15), dR, ABSdθx, and ABSdθy were not significantly different between the two impression techniques of SRT0, STR8, and STR15.
DISCUSSION Other studies that investigated the accuracy of implant impressions using methods such as a micrometer, vernier calipers, strain gauge, or measuring microscope were only able to perform 2D measurements.1,3,21 Furthermore, most of these methods were not able to report the specific 3D directions in which the distortions occurred. Other studies have investigated 3D distortions using the CMM. Tan et al22 investigated the 3D distortions of implant framework castings of two designs using a CMM. They reported translational displacements ranging from 1.5 to 47.5 µm as well as rotational displacements ranging from 0.0128 to 0.305 degrees. Hence, the CMM was chosen because it could examine the complex 3D nature of these distortions with a repeatability within 2 µm, thus allowing the present study to report three linear and two angular distortions. The results of the DR groups for NB and 3i were comparable to the findings of Phillips et al2 and Eliasson et al,23 who found linear distortions ranging from –7.6 to 72.3 µm and 10.9 to 66.1 µm, respectively. In the DR groups, the highest distortions were seen for the Straumann impressions, with a mean dR of 109
µm for the STRDR15 group and high ABSdθx values for all STRDR groups. This may be a result of the implant and metallic impression coping design. Whereas the Nobel Biocare Brånemark and Biomet 3i metallic impression copings have a flat-to-flat connection, where the flat surfaces of the implant platform and impression coping come into contact when tightened, the Straumann metallic impression copings have a sloped surface, which rests against the corresponding sloped surface of the implant. This may lead to a less definite seating of the impression coping onto the implant when tightened, resulting in greater angular distortion. Semper et al14 reported vertical discrepancies between the abutment and implant for the Straumann Tissue-Level implant system ranging from 32 to 83 µm upon repeated connection and reconnection. They also reported larger vertical discrepancies for beveled implant-abutment connections compared to butt-joint implant-abutment connections. However, the methodology used did not allow the reporting of horizontal linear displacements and angular displacements. In the INDR groups, no significant differences were found among the three implant systems tested. This could be a result of the similar nature of the plastic materials used in the impression copings. With respect to each of the three implant systems and two impression techniques, the effect of interimplant angulations up to 15 degrees did not have a significant effect on the NBDR groups. This may be attributed to the high tolerances between the impression coping and implant and between the impression coping and implant analog. However, the NBINDR groups showed that an interimplant angulation of 15 degrees resulted in a significantly higher ABSdθy. And although significant differences were seen only at 15 degrees of angulation, a trend for the distortion to increase as the angulations increased was also observed. This could be attributed to deformation of the plastic impression copings when they were being pulled off the abutments at increased amounts of implant divergence. For the 3iDR groups, an interimplant angulation of 15 degrees resulted in significantly higher dR than angulations at 0 and 8 degrees. The 3iDR0 and 3iDR8 groups showed distortion values similar to those of the NBDR groups. The significantly larger distortion in the 3iDR15 group could be the result of the use of impression copings that engage the internal-hex configuration of the implant, rather than the nonengaging copings that are commonly used in implant impressions when multiple implants are to be splinted by a prosthesis. The internal connection of the Biomet 3i metallic direct impression copings extends approximately 4 mm into the implant cylinder, in contrast to the external-hex connection of the Brånemark System, which extends less than 1 mm into the impression coping. This would result in
582 Volume 29, Number 3, 2014 © 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Teo et al
more distortion upon withdrawal of the impression, especially at this level of implant divergence. A similar finding was reported by Mpikos et al,24 who found that the accuracy of implant impressions was significantly affected by implant angulations when internal-hex implants were used, compared to those with externalhex connections. Internal-hex engaging impression copings were used in this study to standardize the type of impression coping used. The 3iINDR groups apparently showed no significant differences in the various distortions at interimplant angulations up to 15 degrees. It was interesting to note that the 3i plastic impression copings produced the most definite “snap” when connected to the abutments and laboratory analogs. Although these findings are subjective in nature, it may explain why their accuracy was not affected by the interimplant angulations. In the STRDR groups, although only the 15-degree interimplant angulation had a significant effect on distortion, it can be noted that the translational distortions gradually increased from 0 to 15 degrees of angulation, indicating a probable trend. Similarly, for the STRINDR groups, a trend was observed in which the distortions increased as the interimplant angulations increased. However, these differences were not significant. With regard to the two impression techniques, the NBDR groups performed better than the NBINDR groups at 15 degrees of interimplant angulation. The distortion was seen in the buccolingual angulation of the implant. Distortion in this plane would be expected, since the impression material and impression coping would go through the greatest deformation in the buccal or lingual direction during withdrawal of the impression. No significant differences were found between the 3iDR and 3iINDR groups or between the STRDR and STRINDR groups at each interimplant angulation tested. Limitations of this study include the use of a sectional master model instead of a full-arch model. Impressions made of a full arch are potentially subject to more distortion. In addition, the soft tissue level in this in vitro study does not replicate the clinical situation when excessive soft tissue height or incorrect abutment height may interfere with the seating or visual verification of the plastic snap-on impression copings. Also, the use of abutments of the same height in the present study does not replicate the clinical situation, in which abutment heights may differ greatly. The effect of this height discrepancy between angulated abutments on the withdrawal of plastic snap-on impression copings may be significant. These alternative situations should be analyzed in future investigations.
CONCLUSION This study compared the accuracy of plastic indirect abutment-level impression copings and metallic direct implant-level impression copings. Implant system, interimplant angulation, and impression technique were found to have a significant effect on the translational and angular distortions measured. Within the limitations of this study, the following conclusions were made. 1. The accuracy of implant impressions was dependent on the implant system, interimplant angulations, and impression technique. 2. The Straumann metallic direct implant-level impression coping appeared to be the least accurate, compared to those of Nobel Biocare and Biomet 3i, at interimplant angulations of up to 15 degrees. 3. No significant differences were found among the three implant systems tested with the plastic abutment-level impressions with 0, 8, and 15 degrees of interimplant angulation. 4. Interimplant angulations up to 15 degrees had no significant effect on the accuracy of the Nobel Biocare metallic direct implant-level impression copings, the Biomet 3i plastic abutment-level impression copings, or the Straumann plastic abutment-level impression copings. 5. The accuracy of the plastic abutment-level impression copings was comparable to that of the metallic direct implant-level impression copings for the Nobel Biocare Brånemark implant system at 0 and 8 degrees of interimplant angulation and for the Biomet 3i and Straumann implant systems at 0, 8, and 15 degrees of interimplant angulation.
ACKNOWLEDGMENTs The authors reported no conflicts of interest related to this study.
REFERENCES 1. Spector MR, Donovan TE, Nicholls JI. An evaluation of impression techniques for osseointegrated implants. J Prosthet Dent 1990;63:444–447. 2. Phillips KM, Nicholls JI, Ma T, Rubenstein JE. The accuracy of three implant impression techniques: A three dimensional analysis. Int J Oral Maxillofac Implants 1994;9:533–540. 3. Liou AD, Nicholls JI, Yuodelis RA, Brudvik JS. Accuracy of replacing three tapered transfer impression copings in two elastomeric impression materials. Int J Prosthodont 1993;6:377–383.
The International Journal of Oral & Maxillofacial Implants 583 © 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Teo et al
4. Vigolo P, Majzoub Z, Cordioli G. Evaluation of the accuracy of three techniques used for multiple implant abutment impressions. J Prosthet Dent 2003;89:186–192. 5. Vigolo P, Fonzi F, Majzoub Z, Cordioli G. An evaluation of impression techniques for multiple internal connection implant prostheses. J Prosthet Dent 2004;92:470–476. 6. Herbst D, Nel JC, Driessen CH, Becker PJ. Evaluation of impression accuracy for osseointegrated implant supported superstructures. J Prosthet Dent 2000;83:555–561. 7. Assif D, Marshak B, Schmidt A. Accuracy of implant impression techniques. Int J Oral Maxillofac Implants.1996;11:216–222. 8. Kim S, Nicholls JI, Han CH, Lee KW. Displacement of implant components from impressions to definitive casts. Int J Oral Maxillofac Implants 2006;21:747–755. 9. Humphries RM, Yaman P, Bloem TJ. The accuracy of implant master casts constructed from transfer impressions. Int J Oral Maxillofac Implants 1990;5:331–336. 10. Ma T, Nicholls JI, Rubenstein JE. Tolerance measurements of various implant components. Int J Oral Maxillofac Implants 1997;12:371–375. 11. Carr AB. Comparison of impression technique for a two-implant 15-degree divergent model. Int J Oral Maxillofac Implants 1992;7:468–475. 12. Jang HK, Kim S, Shim JS, Lee KW, Moon HS. Accuracy of impressions for internal-connection implant prostheses with various divergent angles. Int J Oral Maxillofac Implants 2011;26:1011–1015. 13. Abrahamsson I, Berglundh T, Lindhe J. The mucosal barrier following abutment dis/reconnection. An experimental study in dogs. J Clin Periodontol 1997;24:568–572. 14. Semper W, Heberer S, Mehrhof J, Schink T, Nelson K. Effects of repeated manual disassembly and reassembly on the positional stability of various implant-abutment complexes: An experimental study. Int J Oral Maxillofac Implants 2010;25:86–94.
15. Shen C. Viscoelastic properties. In: Anusavice KL (ed). Philips’ Science of Dental Materials, ed 11. St Louis, MO: Elsevier Science;2003:211–212. 16. Walker MP, Ries D, Borello B. Implant cast accuracy as a function of impression techniques and impression material viscosity. Int J Oral Maxillofac Implants 2008;23(4):669–674. 17. Akça K, Cehreli MC. Accuracy of 2 impression techniques for ITI implants. Int J Oral Maxillofac Implants 2004 Jul–Aug;19(4):517–523. 18. Kim KS, Lim YJ, Kim MJ, et al. Variation in the total lengths of abutment/implant assemblies generated with a function of applied tightening torque in external and internal implant-abutment connection. Clin Oral Implants Res 2011 Aug;22(8):834–839. 19. Lee H, So JS, Hochstedler JL, Ercoli C. The accuracy of implant impressions: A systematic review. J Prosthet Dent 2008;100:285–291. 20. Wee A. Comparison of impression materials for direct multi-implant impression. J Prosthet Dent 2000;83:323–331. 21. Wee AG, Schneider RL, Aquilino SA, Huff TL, Lindquist TJ, Williamson DL. Evaluation of the accuracy of solid implant casts. Int J Prosthodont 1998;7:161–169. 22. Tan KB, Rubenstein JE, Nicholls JI, Yuodelis RA. Three-dimensional analysis of the casting accuracy of one-piece, osseointegrated implant-retained prostheses. Int J Prosthodont 1993;6:346–363. 23. Eliasson A, Ortorp A. The accuracy of an implant impression technique using digitally coded healing abutments. Clin Implant Dent Relat Res 2012 May;14(suppl 1):e30–38. 24. Mpikos P, Tortopidis D, Galanis C, Kaisarlis G, Koidis P. The effect of impression technique and implant angulation on the impression accuracy of external- and internal-connection implants. Int J Oral Maxillofac Implants 2012;27:1422–1428.F
584 Volume 29, Number 3, 2014 © 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
M
isfit or distortion in implant prostheses may arise from errors incurred during impression making. The factors affecting the accuracy of the impression may include the impression material, polymerization shrinkage, and impression technique,1–9 as well as the machining tolerance of the implant components.10 1Registrar,
Dental Department, Khoo Teck Puat Hospital, Republic of Singapore. 2 Associate Professor, Faculty of Dentistry, National University of Singapore, Republic of Singapore. 3 Professor Emeritus, Department of Restorative Dentistry, School of Dentistry, University of Washington, Seattle, Washington. 4 Adjunct Senior Lecturer, Faculty of Dentistry, National University of Singapore, Republic of Singapore. Correspondence to: Dr Keson B. Tan, Faculty of Dentistry, National University of Singapore, 11 Lower Kent Ridge Road, Singapore 119083, Republic of Singapore. Fax: +65-67785742. Email: [email protected] ©2014 by Quintessence Publishing Co Inc.
While most studies on implant impressions were based on implants placed parallel to each other, several studies have also investigated divergent implants. Phillips et al2 incorporated a 10-degree divergence in each of the implant abutments in a five-implant model and found cross-arch distortions ranging from 22 to 104 µm. Carr11 evaluated direct and indirect impression techniques with implants at 15 degrees of divergence and found no significant difference. Jang et al12 investigated the accuracy of implant impressions with divergences of up to 20 degrees and found that distortions in the 20-degree divergent models were significantly greater than those in the 15-degree divergent models. However, these studies were based on older implant systems that used both open-tray and closed-tray techniques and screw-on metallic impression copings. More recently, many implant systems have introduced plastic impression copings for abutment-level impressions to simplify impression making and minimize the amount of abutment reconnection/disconnection needed, since this can result not only in tissue The International Journal of Oral & Maxillofacial Implants 577
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Teo et al
Z
Y
X a
The aim of this study was to compare the threedimensional (3D) accuracy of plastic impression copings with the direct implant-level metal impression copings from three implant systems: Brånemark, Nobel Biocare (NB); Nanotite Certain Implant System, Biomet 3i (3i); and Tissue Level Implant System, Straumann (STR). The effects of the three implant systems; buccolingual interimplant angulations of 0, 8, and 15 degrees; and the two impression techniques (indirect plastic [INDR] vs direct metal [DR]) were also determined and compared.
b Fig 1 Master model with gauge block (A). Point B is the origin of the local coordinate system.
MATERIALS AND METHODS Master Model
irritation13 but also changes in abutment location in relation to the implant.14 However, the newer plastic impression copings lack rigidity, and their “snap-on” connection to the implant components can potentially allow deformation under the forces applied during impression making.15 These forces and deformations can be greatly increased when an impression is pulled away from nonparallel abutments. To date, only two studies have investigated the accuracy of plastic impression copings. Walker et al16 studied the accuracy of indirect implant-level impressions using screw-on metal copings and compared it to that of abutmentlevel impressions using snap-on plastic copings in the Nobel Replace implant system. They used a microscope to measure interimplant distances and found that the indirect metal copings were more accurate than the plastic snap-on copings. The percentage difference in the interimplant distances between the master model and casts made from the indirect metal copings ranged from –6% to 2%. The percentage difference in the interabutment distances between the master model and casts made from the plastic snap-on copings ranged from 19% to 24%. However, Akça and Cehreli17 found no significant differences between the metal and plastic impression copings of the ITI Straumann Tissue Level implant system. Most implant systems have introduced simplified impression techniques with plastic snap-on abutmentlevel impression copings. The use of such impression systems is recommended by the manufacturers not only for single implant restorations but also in cases where two or more abutments are involved. Limitations regarding their use, in terms of the number of abutments and the divergence of abutments, have not been highlighted by the manufacturers or reported in the literature. At present, studies of their accuracy and comparisons between different implant systems are still lacking in the literature.
The master models were fabricated using an acrylic resin (Trevalon, Dentsply Corp) partial mandibular dental arch attached to an aluminum block incorporating a gauge block (Part No:MCL-VP100, Matchling Tooling) that defined the local coordinate system. Two implants were placed, in the first premolar and first molar positions, to simulate a three-unit implant prosthesis (Fig 1). Three master models were made of each implant system, and the interimplant angulations were set at 0, 8, and 15 degrees using a universal vice (VUA3, Vertex Machinery Works) and a drill press (Universal Drilling Machine, ARICAS). Implants were secured with self-curing acrylic resin (Orthocryl, Dentaurum) and allowed to set for 24 hours before measurements were made. With the three implant systems (NB, 3i, STR) and three interimplant angulations, a total of nine master models were fabricated.
Implant Impressions
Impressions were made using direct impression copings (DR) connected to the implants (Fig 2). These copings were tightened to a torque of 15 Ncm using a calibrated torque gauge (6BTG, Tohnichi). The respective abutments (Nobel Biocare Snappy, Biomet 3i Provide, and Straumann Solid) were then connected to the implants and a tightening torque was applied to each abutment in accordance with the manufacturer’s instructions (45 Ncm for the NB groups, 20 Ncm for the 3i groups, and 35 Ncm for the STR groups). The abutments were retightened again to the recommended torque values to ensure that the proper torque was achieved and to minimize settling effects.18 The respective plastic snap-on impression copings were then attached (Fig 3), and an indirect impression was made (INDR). The recommended torque was reapplied after every impression to ensure that the abutments remained at the correct level of torque at all times. All impressions were made with a polyether elastomeric impression material (Impregum, 3M/ESPE)19,20 using
578 Volume 29, Number 3, 2014 © 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Teo et al
Fig 2 Direct implant-level impression copings (DR) attached.
Fig 3 Plastic abutment-level impression copings (INDR) attached to prefabricated abutments. System illustrated is Snappy Abutment (Nobel Biocare).
Probe hit
A
B
Probe hit Fig 4 Measured plane on an implant. Six probe hits were taken on the implant platform to define the plane.
Fig 5 Measured cylinder on an implant. The cylinder was defined by two circles approximately 1 mm apart, each of which was defined by six probe hits. (Not shown: Prefabricated abutment cone features were defined by two circles approximately 3 mm apart.)
custom trays (Tray Resin II, Shofu). The impressions were poured with vacuum-mixed type IV stone (Silky Rock, Whip Mix Corporation), mixed according to the manufacturer’s directions (100 g powder to 23 mL water), and allowed to set for 60 minutes before the test models were separated. For each test group, five impressions were made to yield five test models (n = 5).
Coordinate Measuring Machine
All measurements were made using a coordinate measuring machine (CMM) (Model TESA MH3D DCC, Brown and Sharpe) with an accuracy of 2 µm. For each master model, the local coordinate axes were defined on the gauge block with four probe hits on each of the three exposed orthogonal metal surfaces. The intersection point of these three planes defined the origin for the local coordinate system for each master model. The stone surfaces on the replica of the gauge block in the test models were also used to define the origin and transferred local coordinate axis system. Pilot studies
Fig 6 The intersection points between the measured planes and cylinders (NB and 3i systems) or cones (STR system) defined the centroid positions of the molar (point A [x M1, yM1, and zM1]) and premolar (point B [x M2, yM2, and zM2]) implants.
found that the flatness of the metal gauge block planes and that of the stone replica were not greater than 3 µm and 5 µm, respectively. For the DR groups, a plane was defined on the platform of each implant on the master model and implant analog on the test model by taking six probe hits (Fig 4). A cylinder or cone was then defined on the implant or implant analog by measuring two circles of six probe hits each (Fig 5). For the INDR groups, a plane was defined by taking six hits on the top surface of each abutment on the master model and abutment analog on the test model. A cone was then defined by two circles of six probe hits each around the abutment or abutment analog. The intersection point of the central axis of the measured cone/cylinder and the plane gave the centroid coordinates for the molar implant or abutment (xM1, yM1, and zM1) and the premolar implant or abutment (xM2, yM2, and zM2) (Fig 6). Similar measurements made on the test model analogs gave the centroid coordinates for the molar implant or abutment The International Journal of Oral & Maxillofacial Implants 579
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Teo et al
Table 1 Mean Linear Distortions (Standard Deviations) in Millimeters Group
dy
dz
dR
NBDR0
0.022 (0.008)
dx
–0.010 (0.008)
0.005 (0.005)
0.026 (0.008)
NBDR8
–0.004 (0.014)
–0.018 (0.024)
–0.004 (0.011)
0.031 (0.012)
NBDR15
0.017 (0.015)
0.003 (0.017)
–0.011 (0.010)
0.028 (0.014)
NBINDR0
0.009 (0.008)
–0.009 (0.021)
0.023 (0.016)
0.034 (0.011)
NBINDR 8
0.006 (0.009)
–0.021 (0.018)
0.002 (0.043)
0.043 (0.022)
NBINDR 15
0.054 (0.021)
0.002 (0.019)
0.002 (0.025)
0.061 (0.019)
3iDR0
0.016 (0.029)
0.014 (0.028)
0.018 (0.013)
0.045 (0.014)
3iDR8
–0.004 (0.018)
–0.012 (0.015)
–0.004 (0.009)
0.024 (0.011)
3iDR15
0.033 (0.013)
0.038 (0.056)
0.025 (0.018)
0.075 (0.010)
3iINDR0
0.012 (0.028)
–0.004 (0.051)
–0.010 (0.013)
0.053 (0.017)
3iINDR8
–0.001 (0.023)
0.009 (0.035)
–0.008 (0.023)
0.039 (0.023)
0.003 (0.038)
0.010 (0.030)
0.007 (0.015)
0.040 (0.029)
STRDR0
3iINDR15
–0.018 (0.014)
0.031 (0.027)
0.007 (0.042)
0.055 (0.025)
STRDR8
0.032 (0.023)
–0.027 (0.054)
–0.013 (0.019)
0.070 (0.007)
STRDR 15
0.057 (0.048)
0.078 (0.028)
0.013 (0.037)
0.109 (0.039)
STRINDR0
–0.045 (0.008)
0.004 (0.025)
0.003 (0.023)
0.054 (0.011)
STRINDR8
–0.024 (0.042)
0.009 (0.053)
–0.009 (0.022)
0.066 (0.021)
STRINDR15
–0.014 (0.030)
0.057 (0.028)
0.020 (0.012)
0.071 (0.017)
• Linear distortion in the y-axis (dy) = (yT2 – yT1) – (yM2 – yM1) • Linear distortion in the z-axis (dz) = (zT2 – zT1) – (zM2 – zM1) • Global 3D distortion (dR) = √(dx2 + dy2 + dz2)
0.140 Distortion (mm)
0.120 0.100 0.080 0.060 0.040
0.000
NBD RO NBD R8 NBD R15 3iDR O 3iDR 8 3iDR 15 STRD RO STRD R8 STRD R15 NBIN DRO NBIN DR8 NBIN DR1 5 3iIND RO 3iIND R8 3iIND R1 STRIN 5 DR STRIN O DR8 STRIN DR1 5
0.020
Test group Fig 7 Mean global 3D distortions (dR). Error bars indicate standard deviations.
analog (xT1, yT1, and zT1) and the premolar implant or abutment analog (xT2, yT2, and zT2). For each measured plane and cone/cylinder feature, the respective flatness or roundness criterion was set at no greater than 5 µm. All components used and all features measured fulfilled these criteria. Linear distortions in the three axes and global 3D distortion were defined as follows: • Linear distortion in the x-axis (dx) = (xT2 – xT1) – (xM2 – xM1)
The angular deviation of each implant in relation to the x-axis (xz) and y-axis (yz) were calculated from the measured planes. Hence, angular deviations were given as xzM1 and yzM1 for the molar implant and xzM2 and yzM2 for the premolar implant on the master model. Similarly, the angular deviations of the implant analogs on the test models were given as xzT1 and yzT1 for the molar analog and xzT2 and yzT2 for the premolar analog. Angular distortions with regard to the x-axis and y-axis between the master and test models were calculated as follows. • Angular distortions with regard to the x-axis (dθx) = (xzT2 – xzT1) – (xzM2 – xzM1) • Angular distortions with regard to the y-axis (dθy) = (yzT2 – yzT1) – (yzM2 – yzM1) Because angular distortions measured relative to the local coordinate system can be either positive or negative, the mean angular distortions may not be a true representation of the absolute amount of distortion present in a group. Hence, the mean absolute angular distortions (ABSdθx and ABSdθy) were reported.
580 Volume 29, Number 3, 2014 © 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Teo et al
Table 2 Mean Angular Distortions (Standard Deviations) in Degrees Group
dθx
dθy
ABSdθx
ABSdθy
NBDR0
–0.001 (0.076)
0.041 (0.067)
0.065 (0.021)
0.069 (0.025)
NBDR8
–0.023 (0.112)
0.082 (0.153)
0.086 (0.061)
0.124 (0.113)
NBDR15
–0.012 (0.073)
0.129 (0.118)
0.058 (0.036)
0.129 (0.118)
NBINDR0
0.284 (0.286)
0.007 (0.155)
0.219 (0.203)
0.106 (0.101)
NBINDR 8
–0.158 (0.221)
–0.064 (0.249)
0.160 (0.220)
0.199 (0.133)
NBINDR 15
0.477 (0.287)
0.503 (0.233)
0.477 (0.287)
0.503 (0.233)
3iDR0
0.010 (0.212)
–0.059 (0.070)
0.150 (0.131)
0.078 (0.041)
3iDR8
0.011 (0.135)
0.185 (0.086)
0.107 (0.062)
0.185 (0.086)
3iDR15
0.096 (0.151)
0.226 (0.125)
0.120 (0.127)
0.226 (0.125)
3iINDR0
0.049 (0.289)
0.291 (0.136)
0.219 (0.163)
0.291 (0.136)
3iINDR8
0.078 (0.162)
–0.205 (0.192)
0.138 (0.100)
0.205 (0.192)
3iINDR15
–0.016 (0.242)
0.176 (0.165)
0.198 (0.098)
0.176 (0.165) 0.185 (0.108)
0.677 (0.526)
0.183 (0.114)
0.777 (0.307)
STRDR8
0.520 (0.520)
–0.122 (0.364)
0.572 (0.446)
0.292 (0.211)
–0.658 (0.731)
0.235 (0.467)
0.694 (0.689)
0.449 (0.187)
STRINDR0
0.336 (0.169)
–0.009 (0.166)
0.336 (0.169)
0.129 (0.082)
STRINDR8
–0.145 (0.351)
0.262 (0.511)
0.328 (0.122)
0.420 (0.355)
STRINDR15
–0.170 (0.308)
0.407 (0.157)
0.235 (0.249)
0.407 (0.157)
Prior to this study, a feasibility test was performed to verify the dimensional stability of the acrylic resin master model over a period of 245 days. The measured linear interimplant dimension varied within a range of only 3 µm over this period.
Statistical Analyses
All data were subjected to three-way analysis of variance (ANOVA) and subsequent one-way ANOVA for the independent variables SYSTEM, ANGLE, and TECHNIQUE. The dependent variables were the linear distortions (dx, dy, dz, and dR) and the angular distortions (dθx and dθy) as well as the absolute angular distortions (ABSdθx and ABSdθy). Post hoc analyses were performed using the Tukey test at a 5% significance level. All analyses were performed using SPSS v18.0 (SPSS).
RESULTS For the test groups measured, the mean dx ranged from –0.045 to 0.054 mm, mean dy ranged from –0.027 to 0.078 mm, and mean dz ranged from –0.013 to 0.025 mm. Mean dR ranged from 0.024 to 0.109 mm (Table 1, Fig 7). Mean dθx ranged from –0.658 to 0.677 degrees and mean dθy ranged from –0.205 to 0.503 degrees. Mean ABSdθx ranged from 0.058 to 0.777 degrees, and mean ABSdθy ranged from 0.069 to 0.503 degrees (Table 2, Fig 8).
1.0
ABSdθx
0.8
ABSdθy
0.6 0.4 0.2 0.0
NBD RO NBD R8 NBD R15 3iDR O 3iDR 8 3iDR 15 STRD RO STRD R8 STRD R15 NBIN DRO NBIN DR8 NBIN DR1 5 3iIND RO 3iIND R8 3iIND R15 STRIN DRO STRIN DR8 STRIN DR1 5
STRDR 15
Distortion (degree)
STRDR0
Test group Fig 8 Mean absolute angular distortions (ABSdθx and ABSdθy). Error bars indicate standard deviations.
Three-way ANOVA showed that SYSTEM had a significant effect on dX, dY, dR, ABSdθx, and ABSdθy. ANGLE had a significant effect on dx, dy, dz, dR, and ABSdθy. TECHNIQUE had a significant effect on dx and ABSdθy. For comparisons among the implant systems using the direct impression technique (DR groups), the STR groups showed significantly higher distortions at all three interimplant angulations tested. At 15 degrees of interimplant angulation, significantly lower dR was found in NBDR15. For the INDR groups, no significant differences were found between the three implant systems at all three interimplant angulations tested. The International Journal of Oral & Maxillofacial Implants 581
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Teo et al
When comparing the three interimplant angulations, no significant differences were found for all distortions in the NBDR, 3iINDR, and STRINDR groups. In the NBINDR groups, ABSdθy for NBINDR15 was significantly higher than for NBINDR0, but not significantly different between NBINDR0 and NBINDR8. dR and ABSdθx were not significantly different among the NBINDR groups. In the 3iDR and STRDR groups, dR was found to be significantly higher in 3iDR15 and STRDR15, but no significant differences were found for ABSdθx and ABSdθy. Comparisons were made between the two impression techniques of the NB implant system at each angulation (NB0, NB8, NB15). ABSdθy of NBINDR15 was significantly higher than NBDR15, but it was not significantly different between NBDR0 and NBINDR0 or between NBDR8 and NBINDR8. No significant differences were found for dR and ABSdθx between the impression techniques of NB0, NB8, and NB15. For comparisons between the two impression techniques of 3i at each angulation (3i0, 3i8, and 3i15), no significant differences were found for all translational and angular distortions between the two impression techniques. For comparisons made between the two impression techniques of STR at each angulation (STR0, STR8, and STR15), dR, ABSdθx, and ABSdθy were not significantly different between the two impression techniques of SRT0, STR8, and STR15.
DISCUSSION Other studies that investigated the accuracy of implant impressions using methods such as a micrometer, vernier calipers, strain gauge, or measuring microscope were only able to perform 2D measurements.1,3,21 Furthermore, most of these methods were not able to report the specific 3D directions in which the distortions occurred. Other studies have investigated 3D distortions using the CMM. Tan et al22 investigated the 3D distortions of implant framework castings of two designs using a CMM. They reported translational displacements ranging from 1.5 to 47.5 µm as well as rotational displacements ranging from 0.0128 to 0.305 degrees. Hence, the CMM was chosen because it could examine the complex 3D nature of these distortions with a repeatability within 2 µm, thus allowing the present study to report three linear and two angular distortions. The results of the DR groups for NB and 3i were comparable to the findings of Phillips et al2 and Eliasson et al,23 who found linear distortions ranging from –7.6 to 72.3 µm and 10.9 to 66.1 µm, respectively. In the DR groups, the highest distortions were seen for the Straumann impressions, with a mean dR of 109
µm for the STRDR15 group and high ABSdθx values for all STRDR groups. This may be a result of the implant and metallic impression coping design. Whereas the Nobel Biocare Brånemark and Biomet 3i metallic impression copings have a flat-to-flat connection, where the flat surfaces of the implant platform and impression coping come into contact when tightened, the Straumann metallic impression copings have a sloped surface, which rests against the corresponding sloped surface of the implant. This may lead to a less definite seating of the impression coping onto the implant when tightened, resulting in greater angular distortion. Semper et al14 reported vertical discrepancies between the abutment and implant for the Straumann Tissue-Level implant system ranging from 32 to 83 µm upon repeated connection and reconnection. They also reported larger vertical discrepancies for beveled implant-abutment connections compared to butt-joint implant-abutment connections. However, the methodology used did not allow the reporting of horizontal linear displacements and angular displacements. In the INDR groups, no significant differences were found among the three implant systems tested. This could be a result of the similar nature of the plastic materials used in the impression copings. With respect to each of the three implant systems and two impression techniques, the effect of interimplant angulations up to 15 degrees did not have a significant effect on the NBDR groups. This may be attributed to the high tolerances between the impression coping and implant and between the impression coping and implant analog. However, the NBINDR groups showed that an interimplant angulation of 15 degrees resulted in a significantly higher ABSdθy. And although significant differences were seen only at 15 degrees of angulation, a trend for the distortion to increase as the angulations increased was also observed. This could be attributed to deformation of the plastic impression copings when they were being pulled off the abutments at increased amounts of implant divergence. For the 3iDR groups, an interimplant angulation of 15 degrees resulted in significantly higher dR than angulations at 0 and 8 degrees. The 3iDR0 and 3iDR8 groups showed distortion values similar to those of the NBDR groups. The significantly larger distortion in the 3iDR15 group could be the result of the use of impression copings that engage the internal-hex configuration of the implant, rather than the nonengaging copings that are commonly used in implant impressions when multiple implants are to be splinted by a prosthesis. The internal connection of the Biomet 3i metallic direct impression copings extends approximately 4 mm into the implant cylinder, in contrast to the external-hex connection of the Brånemark System, which extends less than 1 mm into the impression coping. This would result in
582 Volume 29, Number 3, 2014 © 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Teo et al
more distortion upon withdrawal of the impression, especially at this level of implant divergence. A similar finding was reported by Mpikos et al,24 who found that the accuracy of implant impressions was significantly affected by implant angulations when internal-hex implants were used, compared to those with externalhex connections. Internal-hex engaging impression copings were used in this study to standardize the type of impression coping used. The 3iINDR groups apparently showed no significant differences in the various distortions at interimplant angulations up to 15 degrees. It was interesting to note that the 3i plastic impression copings produced the most definite “snap” when connected to the abutments and laboratory analogs. Although these findings are subjective in nature, it may explain why their accuracy was not affected by the interimplant angulations. In the STRDR groups, although only the 15-degree interimplant angulation had a significant effect on distortion, it can be noted that the translational distortions gradually increased from 0 to 15 degrees of angulation, indicating a probable trend. Similarly, for the STRINDR groups, a trend was observed in which the distortions increased as the interimplant angulations increased. However, these differences were not significant. With regard to the two impression techniques, the NBDR groups performed better than the NBINDR groups at 15 degrees of interimplant angulation. The distortion was seen in the buccolingual angulation of the implant. Distortion in this plane would be expected, since the impression material and impression coping would go through the greatest deformation in the buccal or lingual direction during withdrawal of the impression. No significant differences were found between the 3iDR and 3iINDR groups or between the STRDR and STRINDR groups at each interimplant angulation tested. Limitations of this study include the use of a sectional master model instead of a full-arch model. Impressions made of a full arch are potentially subject to more distortion. In addition, the soft tissue level in this in vitro study does not replicate the clinical situation when excessive soft tissue height or incorrect abutment height may interfere with the seating or visual verification of the plastic snap-on impression copings. Also, the use of abutments of the same height in the present study does not replicate the clinical situation, in which abutment heights may differ greatly. The effect of this height discrepancy between angulated abutments on the withdrawal of plastic snap-on impression copings may be significant. These alternative situations should be analyzed in future investigations.
CONCLUSION This study compared the accuracy of plastic indirect abutment-level impression copings and metallic direct implant-level impression copings. Implant system, interimplant angulation, and impression technique were found to have a significant effect on the translational and angular distortions measured. Within the limitations of this study, the following conclusions were made. 1. The accuracy of implant impressions was dependent on the implant system, interimplant angulations, and impression technique. 2. The Straumann metallic direct implant-level impression coping appeared to be the least accurate, compared to those of Nobel Biocare and Biomet 3i, at interimplant angulations of up to 15 degrees. 3. No significant differences were found among the three implant systems tested with the plastic abutment-level impressions with 0, 8, and 15 degrees of interimplant angulation. 4. Interimplant angulations up to 15 degrees had no significant effect on the accuracy of the Nobel Biocare metallic direct implant-level impression copings, the Biomet 3i plastic abutment-level impression copings, or the Straumann plastic abutment-level impression copings. 5. The accuracy of the plastic abutment-level impression copings was comparable to that of the metallic direct implant-level impression copings for the Nobel Biocare Brånemark implant system at 0 and 8 degrees of interimplant angulation and for the Biomet 3i and Straumann implant systems at 0, 8, and 15 degrees of interimplant angulation.
ACKNOWLEDGMENTs The authors reported no conflicts of interest related to this study.
REFERENCES 1. Spector MR, Donovan TE, Nicholls JI. An evaluation of impression techniques for osseointegrated implants. J Prosthet Dent 1990;63:444–447. 2. Phillips KM, Nicholls JI, Ma T, Rubenstein JE. The accuracy of three implant impression techniques: A three dimensional analysis. Int J Oral Maxillofac Implants 1994;9:533–540. 3. Liou AD, Nicholls JI, Yuodelis RA, Brudvik JS. Accuracy of replacing three tapered transfer impression copings in two elastomeric impression materials. Int J Prosthodont 1993;6:377–383.
The International Journal of Oral & Maxillofacial Implants 583 © 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Teo et al
4. Vigolo P, Majzoub Z, Cordioli G. Evaluation of the accuracy of three techniques used for multiple implant abutment impressions. J Prosthet Dent 2003;89:186–192. 5. Vigolo P, Fonzi F, Majzoub Z, Cordioli G. An evaluation of impression techniques for multiple internal connection implant prostheses. J Prosthet Dent 2004;92:470–476. 6. Herbst D, Nel JC, Driessen CH, Becker PJ. Evaluation of impression accuracy for osseointegrated implant supported superstructures. J Prosthet Dent 2000;83:555–561. 7. Assif D, Marshak B, Schmidt A. Accuracy of implant impression techniques. Int J Oral Maxillofac Implants.1996;11:216–222. 8. Kim S, Nicholls JI, Han CH, Lee KW. Displacement of implant components from impressions to definitive casts. Int J Oral Maxillofac Implants 2006;21:747–755. 9. Humphries RM, Yaman P, Bloem TJ. The accuracy of implant master casts constructed from transfer impressions. Int J Oral Maxillofac Implants 1990;5:331–336. 10. Ma T, Nicholls JI, Rubenstein JE. Tolerance measurements of various implant components. Int J Oral Maxillofac Implants 1997;12:371–375. 11. Carr AB. Comparison of impression technique for a two-implant 15-degree divergent model. Int J Oral Maxillofac Implants 1992;7:468–475. 12. Jang HK, Kim S, Shim JS, Lee KW, Moon HS. Accuracy of impressions for internal-connection implant prostheses with various divergent angles. Int J Oral Maxillofac Implants 2011;26:1011–1015. 13. Abrahamsson I, Berglundh T, Lindhe J. The mucosal barrier following abutment dis/reconnection. An experimental study in dogs. J Clin Periodontol 1997;24:568–572. 14. Semper W, Heberer S, Mehrhof J, Schink T, Nelson K. Effects of repeated manual disassembly and reassembly on the positional stability of various implant-abutment complexes: An experimental study. Int J Oral Maxillofac Implants 2010;25:86–94.
15. Shen C. Viscoelastic properties. In: Anusavice KL (ed). Philips’ Science of Dental Materials, ed 11. St Louis, MO: Elsevier Science;2003:211–212. 16. Walker MP, Ries D, Borello B. Implant cast accuracy as a function of impression techniques and impression material viscosity. Int J Oral Maxillofac Implants 2008;23(4):669–674. 17. Akça K, Cehreli MC. Accuracy of 2 impression techniques for ITI implants. Int J Oral Maxillofac Implants 2004 Jul–Aug;19(4):517–523. 18. Kim KS, Lim YJ, Kim MJ, et al. Variation in the total lengths of abutment/implant assemblies generated with a function of applied tightening torque in external and internal implant-abutment connection. Clin Oral Implants Res 2011 Aug;22(8):834–839. 19. Lee H, So JS, Hochstedler JL, Ercoli C. The accuracy of implant impressions: A systematic review. J Prosthet Dent 2008;100:285–291. 20. Wee A. Comparison of impression materials for direct multi-implant impression. J Prosthet Dent 2000;83:323–331. 21. Wee AG, Schneider RL, Aquilino SA, Huff TL, Lindquist TJ, Williamson DL. Evaluation of the accuracy of solid implant casts. Int J Prosthodont 1998;7:161–169. 22. Tan KB, Rubenstein JE, Nicholls JI, Yuodelis RA. Three-dimensional analysis of the casting accuracy of one-piece, osseointegrated implant-retained prostheses. Int J Prosthodont 1993;6:346–363. 23. Eliasson A, Ortorp A. The accuracy of an implant impression technique using digitally coded healing abutments. Clin Implant Dent Relat Res 2012 May;14(suppl 1):e30–38. 24. Mpikos P, Tortopidis D, Galanis C, Kaisarlis G, Koidis P. The effect of impression technique and implant angulation on the impression accuracy of external- and internal-connection implants. Int J Oral Maxillofac Implants 2012;27:1422–1428.F
584 Volume 29, Number 3, 2014 © 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
