LAUSR

OBJECTIVE To investigate whether miRNAs have a remarkable pooled diagnostic accuracy, sensitivity and specificity as non-invasive biomarkers to distinguish endometriosis patients from non-endometriosis women. METHODS A comprehensive literature search of PubMed, Embase and ProQuest was performed through 21 Feb 2021 to find relevant studies. Two reviewers independently screened each article, and discrepancies were resolved by consensus. Deeks' funnel plot asymmetry test was performed to assess the publication bias of included studies. STATA software and RevMan 5.4 were used for data analysis and quality assessment, respectively. RESULTS The overall quality of the studies was moderate to high. In total 87 datasets were assessed miRNAs' performance which results in sensitivity: 0.82, specificity: 0.79, DOR: 18, NPV: 0.80, PPV: 0.78, PLR: 3.97, and NLR: 022. We conducted subgroup analyses, which showed panels of miRNAs (DOR: 54) and serum (DOR: 43) as a target tissue were more reliable to utilize as biomarkers. Deeks' funnel plot showed that there is no publication bias (P-value = 0.25). CONCLUSIONS Panels of miRNAs differentiate endometriosis patients from non-endometriosis women with high sensitivity and specificity; therefore, it has the potential to use as a non-invasive biomarker.