マイクロRNAを歯周病の診断バイオマーカーとして：系統的レビューとメタアナリシス

MicroRNAs (miRNAs) are key gene regulators, and their dysregulation is linked to periodontitis. This systematic review and meta-analysis assessed miRNAs consistently associated with periodontitis, focussing on their diagnostic accuracy and potential as biomarkers for early detection. Following PRISMA guidelines (PROSPERO ID: CRD42024520884), electronic searches were conducted in PubMed, Scopus, Web of Science, and Wiley Online Library up to 31 May 2025. The meta-analysis included studies reporting common miRNAs in at least three independent investigations. Diagnostic accuracy was determined by pooled sensitivity, specificity, and area under the hierarchical summary receiver operating characteristic curve (HSROC). Heterogeneity (Cochran's Q, I², τ²), publication bias, influence diagnostics, and leave-one-out analyses were performed. Stratified analyses by specimen type and assay platform were conducted, and predictive values (PPV, NPV) were estimated across plausible prevalence scenarios using a bivariate random-effects Reitsma model in R (version 2025.05.1). Visual outputs, including forest plots, were generated using RevMan (version 5.4). Certainty of evidence was assessed using the GRADE-DTA framework. Of 552 records identified, 32 studies were included in the review, and 14 studies were eligible for quantitative analysis, with miR-146, miR-155, and miR-223 being the most consistently reported. miR-146 and miR-155 showed high diagnostic accuracy, with area under the curve (AUC) values of 0.928 and 0.903, respectively, while miR-223 demonstrated moderate performance (AUC: 0.787). Heterogeneity was substantial for miR-146 and miR-223 and moderate for miR-155. These findings highlight the potential of these miRNAs as non-invasive biomarkers for distinguishing periodontitis from healthy individuals; however, standardised protocols and further validation are necessary for clinical translation.