Which mechanical and physical testing methods are relevant for predicting the clinical performance of ceramic‐based dental prostheses?

Abstract Purpose: The survival and performance of clinical prostheses with a ceramic component are probabilistic in nature. Only under very rare circumstances will all of the prostheses in a group exhibit either 100% successes or 100% failures over a period of 5 years or more. Prosthesis failure may be defined as any condition that leads to replacement. These conditions include secondary caries, irreversible pulpitis, excessive wear of opposing tooth surfaces, excessive erosion and roughening of the ceramic surface, ditching of the cement margin, unacceptable esthetics, cracking, chipping and fracture. A systematic review of the dental literature was performed to determine the extent to which the mechanical and physical properties of dental alloys and ceramics can predict the 5‐year clinical performance of metal‐ceramic and all‐ceramic fixed dental prostheses (FDP) and to determine the associated quality of reported outcomes associated with these clinical studies. Materials and methods: The review was based on clinical research studies of 5 years or greater duration that were published in English dental journals between 1980 and 2006 using the following key words and MeSH terms. Our search strategy was as follows:  Search 1 : Partial fixed denture OR denture, partial, fixed OR denture, partial fixed OR dental porcelain OR metal ceramic alloys OR dental ceramic   Search 2 : Prosthesis failure OR dental restoration failure OR time factors OR survival analysis   Search 3 : Meta‐analysis OR evaluation studies OR review OR clinical trial OR comparative study OR follow‐up studies OR prospective studies OR clinical follow‐up study OR clinical trial OR longitudinal studiesInclusion of searches 1, 2 and 3 and limits placed on the publication date starting on January 1, 1980, English language, and clinical studies involving humans resulted in a total of 684 articles. By restricting the clinical studies to 5 years or more in duration, the number was reduced to 193. By eliminating resin‐bonded FDPs, cantilever designs, implant‐supported prostheses, crowns, inlay‐ or onlay‐supported prostheses, a total of 37 articles remained for detailed review. After excluding review articles and articles involving resin‐bonded bridges, single‐author clinical research articles, cantilever designs and implant‐supported FDPs, 11 clinical research articles remained. For these articles, it was not possible to determine conclusively the probability of failure for three‐unit FDPs compared with four‐unit and larger prostheses or the location of the crowns and pontics. Results: This systematic review of studies on ceramic‐based FDPs confirms the results of previous studies that, in most cases, <15% of these prostheses were removed or were in need of replacement at 10 years. However, there was considerable variability in the number of parameters that were reported as well as the range of details on failures that occurred. In some studies, a standardized evaluation system was used in which USPHS or Ryge criteria were applied. However, there was also great uncertainty in the definition of failure with respect to repairable fractures and whether the identified causative factors were directly or indirectly associated with the replacement of the prostheses. Conclusions: This review indicates that there is no single in vitro test variable that can predict clinical performance in these prostheses. Based on these reviews, there is an urgent need to develop a comprehensive classification system for identifying clinical prosthesis failures, technical complications and biologic complications. Guidelines on the retrieval of fractured prostheses and/or impressions that capture the fracture surface details should also be developed. The predictive power of in vitro data can be increased by finite element stress analysis and computer programs such as the CARES/Life software (NASA Lewis Research Center, Cleveland, OH) that estimates the time‐dependent nature of ceramic structure survival.