Objectives: The objective of this meta-analysis was to determine the diagnostic accuracy of cone-beam computed tomography (CBCT) for tooth fractures in vivo. Methods: PubMed, Embase, Web of Science, ProQuest Dissertations & Theses, CNKI and SIGLE were searched from January 1990 to April 2013 for eligible studies. Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) was employed to assess the quality of the included studies. Meta-analyses were performed in MetaDisc 1.4, Stata 12.1 and StatsDirect 2.7.9. Results: Finally, 12 studies were included in this meta-analysis. The pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio and summary receiver operating characteristic were 0.92 (95% CI = 0.89-0.94), 0.85 (95% CI = 0.75-0.92), 5.68 (95% CI = 3.42-9.45), 0.13 (95% CI = 0.09-0.18) and 0.94 (95% CI = 0.90-0.98), respectively. The pooled prevalence of tooth fractures in patients with clinically-suspected but periapical-radiography-undetected tooth fractures was 91% (95% CI = 83%-97%). Positive and negative predictive values were 0.98 and 0.43 (subgroup analysis: 0.98 and 0.28 for endodontically-treated teeth; 0.99 and 0.77 for non-endodontically-treated teeth). Conclusion: We suggest that CBCT has a high diagnostic accuracy for tooth fractures and could be used in clinical settings. We can be very confident with positive test results but should be very cautious with negative test results, especially for endodontically treated teeth. (C) 2013 Elsevier Ltd. All rights reserved.
