Introduction: The performance of registration materials and methods has been researched by many investigators in an effort to thoroughly understand the patient’s occlusion.1-23 In the dental community, articulating paper has been widely accepted as the gold standard for occlusal analysis.24 However, published studies about its physical properties (thickness, composition, ink substrate, plastic deformation) offer no evidence to suggest that variable articulating paper mark size can be descriptive of variable occlusal loads.1, 3-5, 25-31 In 1987, Maness et al32 first reported the development of the prototype of a new computerized occlusal analysis device (T-Scan® Tekscan Incorporation, Boston, USA). From then to the present, the manufacturer states having improved the system’s accuracy, sensitivity and reproducibility. Still, the latest generation (T-Scan®III HD) lacks of independent overall studies on its improved performance, which shall be tested in this paper. A better diagnostic interpretation of the results from the T-Scan®III HD system should be expected. Materials & Methods: The present study aims to test, under different simulated anatomic circumstances, the performances of a computerized occlusal analysis device (T-Scan®III HD Tekscan Incorporation, Boston, USA). For this purpose, four different occlusal tables were created:  two of 120° created with an artificial inferior first molar (Ivoclar® Vivadent, Vaduz, Liechtenstein) either embedded in a periodontal ligament simulator or not (representing the anatomy of posterior natural teeth vs an implant);  one of 100° (simulating the distortion created to the sensor when anterior teeth occlude);  and finally one plane surface of 180° (control) in static and variable positions. Three levels of force (10N, 50N and 150N) were applied 40 times each by a universal testing machine (Autograph®, AG-I; Shimadzu Co., Kyoto, Japan). A polished spherical bur (diameter=2,2mm) assured the contact on the sensor film. All T-Scan®III HD recordings were compared through a One-way ANOVA statistical analysis with post-hoc tests using Bonferroni corrections for multiple comparisons. Results: According to our study, the following results were obtained:  The sensor film could produce repeatable data for a mean of 85.71 closures with a standard deviation of 35.99;  85% of the outliers are within the 5 first closures, representing the conditioning time required by the initially flat sensor  Graphically and statistically sustained differences (p < .05) could be found : o in the coefficients of variation between tables (180° Variable vs. all other) o in the coefficients of variation between the applied loads (10N vs. 50N vs. 150N); o in the mean RAW-sum between the different tables for the same applied load. Conclusions: An undeniable improvement of this newest T-Scan® system as compared to former designs could be proved. However, when using the T-Scan®III HD system, some points of capital importance have to be considered:  Its sensitivity seems to be improved as compared to former designs, however futher studies on its variability throughout its sensing surface are required;  Its reproducibility could be proved, except for the 5 first values (outliers to the mean values), which shall be used as a conditioning time to both the sensor and the patient;  Its accuracy proved to be dependent of the anatomic circumstances and requires a trained interpretation;  Particular caution has to be taken when interpreting the force % of a rigid vs. non-rigid model, for instance when balancing a mixed occlusion (implant-tooth). Our study shows that despite the technologic advances made in the area of occlusal analysis, a critical interpretation and a careful handling of the depicted values is indispensable, and can only be acquired through a long learning curve.

Trabalho final do 5º ano com vista à atribuição do grau de mestre no âmbito do ciclo de estudos de Mestrado Integrado em Medicina Dentária apresentado à Faculdade de Medicina da Universidade de Coimbra.