Methods: The samples were consecutive 399 metal cores in 236 patients and systematically (every fourth) selected consecutive 378 resin cores in 207 patients luted by adhesive resin cements (SuperBond, SunMedical; Panavia EX, Kuraray) at our clinic from April 1988 to December 1991. The utilized record protocol, containing predictor variables, e.g., patient age, gender difference, restoration type (resin, metal), tooth location, remaining dentin height of the tooth crown (0, 1/5, 2/5, 3/5), and root canal forms, was filled on site by the operator at their build-up. Fifteen-year after installation, survival/failure of the cores and failure reasons were investigated by 3 calibrated examiners with their clinical records. Failure was defined temporal and permanent loss of the cores by any reasons, e.g., core re-cementation, renewal of core restoration, tooth extraction, etc. The survival curves of the metal and resin cores were drawn by Kaplan-Maier method and the difference between them and the risk factors for the core failure were analyzed by Log-rank and Cox hazards tests.
Results: The 15-year cumulative survival rate of the resin cores (78.1%) showed significantly higher than that of the metal cores (50.9%)(p=0.027, Log rank test). Cox hazards test revealed that remaining lower dentin height (p=0.0196) and restoration type (metal cores, p=0.0507) were significant and marginally significant risk factors for the core failures respectively, while age (p=0.2844), gender difference (p=0.4421), tooth location (p=0.3100) was not significant.
Conclusion: Excluding the effect of remaining dentin height of the tooth crown on the longitudinal core survival, the resin cores survived substantially longer than the metal cores in the 15-year follow-up period.