Diagnostic efficacy of direct and indirect digital imaging for approximal caries.

Data sourcesMedline, Embase, LILACS, PubMed, The Cochrane Library and Web of Science databases. A grey literature search was conducted through Google Scholar, where abstracts from the top 100 results (as sorted by search engine relevance) were examined. Hand searching of reference lists only. No language restrictions were imposed.Study selectionStudies that evaluated the efficacy of bitewing and periapical images produced by photostimulable phosphor plate (PSP) and direct digital sensor (DDS) systems for the diagnosis of approximal dental caries in extracted human teeth. Studies were required to have used histologic analysis as the gold standard comparison. Primary outcomes were sensitivity and specificity regarding detection of dental caries.Data extraction and synthesisTwo reviewers independently evaluated the titles and abstracts of studies identified through the search, selecting articles according to established inclusion criteria. The selected articles were subsequently reviewed full-text by the same two authors. Disagreements regarding article inclusion were resolved by consensus with an additional third reviewer. One reviewer performed initial data extraction using a customised data extraction form based on the PICOS principle, with two other authors independently verifying collected information. Risk of bias was assessed independently by three reviewers using the QUADAS-2 checklist. A meta-analysis was performed on the grouped studies that presented suitably homogeneous data to evaluate diagnostic capability for approximal caries in dentine. Results were presented with 95% confidence intervals.ResultsSix studies were included, with four being used for meta-analysis. Methodologies of all studies were considered low risk of bias. Only one study reported a significant difference between PSP and DDS technologies; remaining studies determined that PSP and DDS were comparable in the clinical detection of caries. The meta-analysis sample total was 668 tooth surfaces. All studies reported poor sensitivity and high specificity. For PSP, sensitivity ranged from 15 to 54%, and specificity from 84 to 100%. For DDS, sensitivity varied from 16 to 56%, and specificity from 90 to 100%.ConclusionsDDS and PSP systems are excellent at identifying caries-free surfaces, but both lack sufficient sensitivity to reliably identify surfaces with caries. There is no significant difference between performances of DDS and PSP digital systems for caries detection.