An in vitro comparison of the accuracy of implant impressions with coded healing abutments and different implant angulations.

STATEMENT OF PROBLEM Fabricating implant definitive casts with CAD/CAM technology (Robocasts) from coded healing abutment impressions represents a simpler and innovative alternative to conventional implant impression techniques. However, information about the accuracy of the impressions and the resultant definitive casts is limited. PURPOSE The purpose of the study was to evaluate the accuracy of the Robocasts and compare them to those definitive casts fabricated with conventional implant impression techniques (open tray with splinted impression copings technique). MATERIAL AND METHODS A reference epoxy resin cast was fabricated and shaped to simulate a dental arch. Two regular platform implant replicas (Biomet 3i Certain, 4.1 mm diameter and 15 mm length) with internal connections were placed 10 mm apart with a 10-degree convergence for one side of the reference resin cast and a 30-degree convergence for the other. Coded healing abutments (Encode) were placed at 3 different heights above the level of the soft tissue replication material (approximately 1, 2, and 4 mm) and served as test groups (E1, E2, and E4), and open trays with splinted impression copings (OTSC) served as a control group. The control group was compared to the impressions of the coded healing abutments by using a standardized measurement protocol. Impressions were made for each group (n=18) and poured with vacuum mixed (100 g powder/20 mL water) Type IV dental stone. The vertical discrepancy (Z axis) between 2 prefabricated passively fitting titanium reference frameworks and the platforms of the implant replicas was measured with an optical comparator applying the 1 screw test. Data were analyzed with Kruskal-Wallis and post-hoc Mann-Whitney U tests, as well as the Wilcoxon signed-rank tests. The Bonferroni correction was used to account for multiple comparisons. The significance level (α) used in a given set of tests was equal to .05 divided by the number of tests performed in that set. RESULTS The median vertical discrepancy of each coded healing abutment impression group was higher than the corresponding median of the control group (OTSC) for every combination of angulation and position. Kruskal-Wallis tests indicated a statistically significant difference (P<.001) between groups for each angulation/position combination. All post hoc Mann-Whitney U tests indicated statistically significant differences (all P≤.002) between OTSC and the other groups. Differences between the angulations and positions were not statistically significant when accounting for multiple comparisons. CONCLUSIONS The implant definitive casts fabricated from the coded healing abutment impressions were found to be less accurate than those fabricated from the open tray with splinted impression copings technique for restoring 2 paired (10 or 30 degrees) convergent internal connection implants with nonengaging screw-retained splinted 2-unit implant restorations. Accuracy of fit was not influenced by the implant angulation or position for either impression technique or by the Encode healing abutment height for the Encode impression technique.