This study compared the clinical performance of indirectly manufactured ceramic Evopress inlays with those of directly placed, fine particle hybrid Filtek Z250 composite restorations in posterior teeth. From January 2000 to October 2003, 109 patients received 264 Evopress (Wegold) ceramic inlays and 68 patients received 145 Filtek Z250 (3M ESPE) composite restorations in a dental office. Two-hundred and fifty ceramic inlays (95%) and 135 composite restorations (93%) were re-examined up to 57 months after placement. Modified USPHS criteria were used for the study. The worst finding of all the assessments was the overall assessment of individual restorations. On the basis of these criteria, 220 (88%) Evopress ceramic inlays were assessed as Alpha at the time of clinical re-examination, 26 (10%) were judged Beta and four ceramic inlays (2%) were rated Delta in the re-examination interval and thus categorized as failures. At the time of re-examination, 91 of the 135 composite restorations (67%) were judged Alpha, 36 restorations (26%) were rated Beta and three restorations (2%) were judged Charlie. Five restorations (4%) were categorized as failures (Delta). In two cases, there were marginal gap formations; there were also two cases of secondary caries after 28 and 35 months, as well as a fracture after 13 months. According to Kaplan and Meier, the survival rate after 57 months was 94% for ceramic inlays and 93% for composite restorations. The log rank test showed no significant differences in the survival curves. The current study showed that indirectly manufactured Evopress ceramic inlays performed better than direct Filtek Z250 composite restorations in marginal adaptation, color match and anatomic form. However, with regard to survival probability, there was no significant difference.