AbstractBackground/AimIn case of crown fractures after traumatic dental injuries, the affected teeth can be restored either with reattachment of the fractured fragment or with a direct composite restoration. So far, longevity data for reattachments and direct composite restorations with regard to different failure types (pulp necrosis and infection, restoration loss) are scarce. Therefore, the aim of this retrospective study was to evaluate the restorative and biological survival of reattached fragments and composite restorations after crown fractures in permanent teeth.Material and MethodsDental records of patients treated between 2000 and 2018 were retrospectively analysed regarding the restoration (reattachment or direct composite restorations) of teeth with crown fractures. Survival (no further intervention) and restorative and/or biological failure of all restored teeth were recorded. Statistical analysis was performed using Kaplan–Meier statistics, and the mean annual failure rates for two and 5 years were calculated. Furthermore, the effect of potential risk factors on survival was assessed. Log‐rank tests and univariate Cox regression models (likelihood ratio tests) were used to assess the univariate effect of all variables of interest. Variables with a p‐value ≤.10 were included in a multivariate Cox regression model with shared frailty (p < .05).ResultsOverall, 164 patients with 235 teeth (uncomplicated crown fracture: N = 201, complicated crown fracture: N = 34) were included (1.6 ± 2.5 years observation time). Of these, 59 teeth were restored with reattachment of the fragment and 176 with a composite restoration. Overall, composite restorations had a significantly higher survival rate than reattachments (p = .002). The cumulative survival after 2 years was 42.9% and 65.0% for teeth treated with a reattachment (mAFR = 34.5%) and a composite restoration (mAFR = 19.3%), respectively. When differentiating between failure types, restoration failure and pulp necrosis were significantly more frequently detected in reattached crown fractures compared to composite restorations (restorative failure: p = .001; biological failure: p = .036). In the multivariate Cox regression model, the variable jaw and luxation significantly influenced the survival when the tooth was restored with a composite restoration. The survival was not influenced by the fracture type.ConclusionsRestorative and biological failures were more frequently detected when the tooth was restored with a reattached fragment compared to a direct composite restoration. Both, restoration failure and pulp necrosis with infection should be considered as frequent complications after restoration of crown‐fractured teeth which emphasizes the necessity of regular and short follow‐up intervals throughout the first 2 years.