Objectives: Loosening and fracture of screws used on implant-abutments as well as  inflammatory reactions in the periimplant tissues (caused by microbial leakage at the interface) are relatively often experienced clinical problems. It was suggested that dynamic loading decreases the stability of  differently constructed implant-abutment connections and thereby leads to early microleakage at the interfaces.

Methods: Standard implant-abutment combinations for single-tooth-restorations were tested for the Brånemark-System (Nobel Biocare, S-Göteborg), the Frialit-2/Hermetics-System (Friadent, D-Mannheim), the Replace-Select-System (Nobel Biocare, S-Göteborg), the Camlog-System (Altatec, D-Wurmberg) and the Screw-Vent-System (Centerpulse Dental, USA-Carlsbad). Eight abutments of each system were restored with a simplified molar crown. Inner surfaces of the implants were inoculated with 5µl Escherichia coli suspension. Thereafter they were loaded in a chewing-simulator (1,200,000 cycles of 120N). When microbial leakage or abutment loosening was detected, accomplished chewing cycles were recorded. Survival times while chewing simulation were evaluated by  Kaplan-Meier method and statistically analyzed with Log-Rank test. For microbial leakage Kruskal-Wallis test followed by Mann-Whitney test was used. In order to find a correlation, the differences between survival times and times of microleakage were analyzed by Kaplan-Meier method and Log-Rank test.

Results: Median chewing cycles until bacterial leakage occurred were: 172,800(Br), 43,200(Fr), 64,800(Re), 345,600(Ca), 24,300(Sc). Statistically significant differences were found between Fr/Ca(p=0.004) and between Sc/Ca(p=0.005). For mechanical strength median survival times (loading cycles) were: 954,300(Br), 388,800(Fr), 1,200,000(Re, Ca) and  820,800(Sc). Significant differences were found between groups Fr/Re(p=0.0023), Fr/Ca(p=0.0023), Sc/Re(p=0.0025) and Sc/Ca(p=0.0025). No significant correlation was  found between survival times and times of microleakage within the groups.

Conclusions: Tested implant systems could not reliably prevent microbial leakage at the implant-abutment-interface, but some systems showed significantly later bacterial leakage than others. Implant systems with different implant-abutment connections showed significantly different survival times. The time when microleakage occurred did not correlate with the fatigue survival time of the implant-abutment-connection.