Introduction: Implant abutments (zirconia) have been marketed recently that can be prepared directly in the mouth (prepable) to overcome esthetic and angulation problems of anterior maxillary implants. However, preparation introduces ceramic surface defects that potentially reduce the overall resistance of the assembly (abutment, abutment screw, analog or implant).
Objective: Assess the failure load of simulated implant assemblies after different degrees of clinical reduction of abutments.
Methods: Ceramic abutments (yttrium-stabilized
tetragonal zirconia, ZirDesign, AstraTech)
were prepared with 0, 0.5, or 1mm of external axial reduction starting 1mm
above height-of-contour. Abutments were attached to implant analogs (25 N-cm torque) with titanium screws to simulate implant
attachment. Analogs were firmly attached into a hole in a stainless steel cylinder
using low melting metal alloy. Assemblies were loaded at 60° off-axis until
fracture (Instron, CHS= 1mm/min). Groups were statistically compared (ANOVA,
p<0.05, superscripts=row differences).
|
Abutment Reduction --Prepped |
0mm |
0.5mm |
1.0mm |
|
Assembly Fracture Load (N) ±sd |
429±140A |
576±120A |
547±139A |
Results: Fracture loads for all assemblies displayed large variations within groups (see standard deviations). There were no statistically significant differences (p>0.05) among different groups and no logical relationship of strength to increasing amount of reduction. All fractures occurred at the interface where abutments connected to analogs, suggesting that fractures were unrelated to actual abutments. Assembly failures were best described as ceramic fracture after extensive yielding of the titanium implant screw. It was obvious the titanium screw was the weak link in the assembly. Clinical failures more likely will involve fatigue rather than single cycle loads to failure, but the same effect probably occurs there as well.
Conclusion: Preparation of zirconia abutments did not significantly impair the fracture resistance of simulated implant assemblies.
Acknowledgments: Supported by an ACPEF 3M-ESPE grant and materials contributed from AstraTech and Premier.