Cochrane Database of Systematic Reviews
Data collection and analysisTwo review authors extracted data independently using a piloted study data extraction form based on the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2). Sensitivity and specificity with 95% confidence intervals (CIs) were reported for each study. This information was displayed as coupled forest plots, and plotted as summary receiver operating characteristic (SROC) plots, displaying the sensitivity-specificity points for each study. Due to variability in thresholds we estimated diagnostic accuracy using hierarchical summary receiver operating characteristic (HSROC) methods.
Main resultsWe included seven studies reporting a total of 719 tooth sites or surfaces, with an overall prevalence of the target condition of 73% (528 tooth sites or surfaces). The included studies evaluated two index tests: the electronic caries monitor (ECM) (four studies, 475 tooth surfaces) and CarieScan Pro (three studies, 244 tooth surfaces). Six studies used histology as the reference standard, one used an enhanced visual examination. No study was considered to be at low risk of bias across all four domains or low concern for applicability or both. All studies were at high (five studies) or unclear (two studies) risk of bias for the patient selection domain. We judged two studies to be at unclear risk of bias for the index test domain, and one study to be at high risk of bias for the reference standard and flow and timing domains. We judged three studies to be at low concern for applicability for patient selection, and all seven studies to be of low concern for reference standard and flow and timing domains.Studies were synthesised using a hierarchical method for meta-analysis. There was variability in the results of the individual studies, with sensitivities which ranged from 0.55 to 0.98 and specificities from 0 to 1.00. These extreme values of specificity may be explained by a low number of healthy tooth surfaces in the included samples. The diagnostic odds ratio (DOR) was 15.65 (95% CI 1.43 to 171.15), and indicative of the variability in the included studies. Through meta-regression we observed no meaningful di erence in accuracy according to device type or dentition. Due to the small number of studies we were unable to formally investigate other potential sources of heterogeneity.We judged the certainty of the evidence as very low, and downgraded for risk of bias due to limitations in the design and conduct of the included studies, imprecision arising from the relatively small number of surfaces studied, and inconsistency due to the variability of results.Electrical conductance for the detection of dental caries (Review)