Objective: The aim of this retrospective study was to evaluate the long-term clinical outcomes of posterior composite resin sandwich restorations, and secondarily to assess the influence of potential factors on survival and causes of failure. Method and Materials: Two hundred and four posterior Herculite XRV restorations due to primary caries performed between 1991 and 1997 were included. The restorations were assessed after 18 years, by two calibrated examiners, according to USPHS criteria. The survival of the restorations was estimated by the Kaplan-Meier estimator. Cox regression was applied to evaluate the influence of the cavity size, location of the tooth, caries risk, and gender on survival rate. The predictive power of the analyzed variables on survival rate was studied with multiple linear regression analysis. Results: After 10 years the survival rate was 92.6%, and 82.4% at the end of the study. Thirty-six (17.6%) restorations failed during the evaluation period, 21 (10.3%) of them after more than 10 years. The most common failure was secondary caries (69.4% of the failures). There were statistically significant differences in survival rate depending on caries risk (P = .000), but not between Class I and II (P = .106), and the type and localization of the tooth (P = .115). Conclusion: Posterior Herculite XRV restorations due to primary caries have high long-term survival rates. Generally, failures occur by secondary caries and are more common in molars. The patient's caries risk is the variable that best predicts the survival of posterior restorations. Schlagwörter: caries, clinical evaluation, composite resin, longevity, posterior restoration, survival rate