Abstract
INTRODUCTION: Early detection of potentially malignant oral cavity disorders is critical for a good prognosis, and it is unclear whether the use of chemiluminescence as an adjunctive diagnostic screening method improves diagnostic accuracy. OBJECTIVE: This systematic review and meta-analysis was performed to assess the accuracy of chemiluminescence for diagnosis of oral cancer and precancerous lesions. METHODS: Sixteen prospective and retrospective studies from PubMed, Cochrane database, SCOPUS, Web of Science, Embase, and Google Scholar were reviewed. Oral mucosal disorder, as detected by chemiluminescence, was compared with oral mucosal disorder detected by toluidine blue or visual examination. True-positive, true-negative, false-positive, and false-negative rates were extracted for each study. Methodological quality was evaluated using the Quality Assessment of Diagnostic Accuracy Studies tool (ver. 2). RESULTS: Sensitivity, specificity, negative predictive value, and diagnostic odds ratio (DOR) of the use of toluidine blue were 0.832 (95% confidence interval [CI] 0.692-0.917), 0.429 (95% CI 0.217-0.672), 0.747 (95% CI 0.607-0.849), and 4.061 (95% CI 1.528-10.796; I(2)=9.128%), respectively. The area under the summary receiver operating characteristic (SROC) curve was 0.743. Compared with toluidine blue, as used in 12 studies, chemiluminescence had a higher sensitivity (0.831 vs. 0.694); it had a lower specificity (0.415 vs. 0.734), negative predictive value (0.674 vs. 0.729), and DOR (3.891 vs. 7.705). Compared with clinical examination, as used in three studies, chemiluminescence had lower DOR (4.576 vs. 5.499) and area under the curve (0.818 vs. 0.91). CONCLUSION: Although chemiluminescence itself has good sensitivity for diagnostic work-up of oral cancer and precancer, the diagnostic accuracy of chemiluminescence is comparable to or worse than toluidine blue and clinical examination. Diagnostic accuracy was therefore insufficient for reliable use of chemiluminescence alone.