Probe-based confocal laser endomicroscopy for the diagnosis of undetermined biliary stenoses: A meta-analysis

Summary Background Presently, both the accuracy and sensitivity for distinguishing biliary stenosis from benign to malignant are low. In recent articles, the probe-based confocal laser endomicroscopy (pCLE) showed a better sensitivity than traditional methods in diagnosing malignant biliary stenosis. Here, we conducted a meta-analysis to summarise the published literature. Methods A systematic search for literature was conducted in the Medline, Embase and Cochrane Library databases published until November 2015. Further publications were found in the reference lists of the relevant articles. A quality assessment and data extraction were performed by two reviewers independently. A meta-analysis was performed to evaluate the diagnostic efficiency of a pCLE for discriminating benign and malignant biliary stenoses. Results Eight studies involving 280 patients were included in the analysis. Significant heterogeneity in specificity was observed among the studies (Cochran's Q test = 15.89, degrees of freedom [df] = 7, P = 0.0261 and I2 = 55.9%), while the heterogeneity in sensitivity was not obvious (Cochran's Q test = 7.99, df = 7, P = 0.3334 and I2 = 12.4%). The area under the summary receiver operating characteristic (SROC) curve was 0.8968. The meta-regression and subgroup analysis indicated that the outlier was the source of heterogeneity. When analysed in the random-effects model, the pooled sensitivity, specificity, positive likelihood ratio (LR) and negative LR were 0.90 (0.84–0.94), 0.75 (0.66–0.83), 3.17 (2.18–4.61) and 0.17 (0.11–0.26), respectively. No significant publication bias was found in our research. Conclusion A pCLE is a valuable method for the differential diagnosis between malignant and benign biliary stenoses. However, a preferable diagnostic standard should be explored and improvements in specificity are required.