Menschliche obere Front- und Eckzähne wurden endodontisch versorgt und anschließend mit verschiedenen Glasfaserstiften, identischen Aufbauten und denselben Vollkeramikkronen versorgt. Ziel der Studie war es die klinischen Überlebensraten stiftversorgter Zähne mit den Ausfällen in der in-vitro Simulation zu vergleichen, um eine Korrelation zwischen den klinischen Daten und der Kausimulation durch Variation des Simulationsparameters �Kaukraft� abzuschätzen.
Es stellt sich die Frage, in welchem Maße die klinisch beobachteten Ausfälle im Rahmen einer in-vitro Kausimulation nachgestellt werden können, um bereits vor dem Einsatz am Patienten aussagekräftige Daten über die zu erwartende Haltbarkeit im Patientenmund zu gewinnen.

This investigation is based on the question whether it is possible to achieve evident data for estimating the longevity of dental restorations by means of an in-vitro study. Therefore the aim of the study was to simulate the clinical situation trying to optimize the quality of the simulation by adjusting the parameter �chewing force�. Survival rates were calculated by correlating up to 180 months (15 years) or 3.6 million loading cycles. The data which were received in an in-vitro simulation should be used to predict the clinical behaviour of the restorations. The survival rates of clinical examinations (NAUMANN 2005 and 2008) were used for comparison. It was examined whether linear or exponential mathematical models showed a better curve fitting.
48 human upper incisors and canines were endodontically treated and restored with fiber reinforced resin posts, identical cores and all-ceramic crowns. A first simulation with 25 N chewing force and 2.4 million cycles including a period of four weeks water storage was compared to the clinical data. A further simulation with 50 N and 1.2 million cycles without any water storage was started. All remaining teeth were loaded to fracture. Fracture force, fracture pattern and number of failures were evaluated and statistics (Median (25%/75%); Mann Whitney U-test and Kaplan-Meier (α=0.05)) including mathematical correlations were performed.
After simulation with 50 N both, fracture forces and failure rates were higher than after simulation with 25 N. The longer lasting simulation with 25 N caused higher aging effects. Using simulation parameters (KREJCI 1990; 5°-55°C, 50 N, 1.6Hz, 1.2 million cycles and 6000 thermo cycles) failure rates after simulation were comparable or even higher than the in vivo failure rates. 50 N-simulation failure patterns were comparable to patterns after fracture testing. Therefore loading to fracture seems important for the evaluation of a restoration. A mathematical exponential curve fitting model was more suitable for describing survival rates than linear curve fitting. The simulation with 50 N can be used for predicting clinical longevity of post and core restorations of upper incisors and canines. The accuracy of a prognosis by in-vitro experiments may be improved by performing both, 25 N and 50 N simulations and by increasing the number of mechanical loadings up to 3.6 million.