Diagnostic accuracy of International System for Reporting Serous Fluid Cytopathology: A systematic review and meta-analysis in malignancy diagnosis

Abstract

This study conducts the first meta-analysis to assess the aggregated risk of malignancy associated with each category of the International System for Reporting Serous Fluid Cytopathology (ISRSFC) for reporting serous effusion cytology, while also evaluating diagnostic accuracy. PubMed/MEDLINE and Embase were systematically searched using the keywords “(pleural, peritoneal, and pericardial effusions) AND (serous effusion cytology) OR (International System for Reporting Serous Fluid Cytopathology)”. Articles underwent risk of bias assessment using the QUADAS-2 tool. After excluding inadequate samples, a meta-analysis determined sensitivity and specificity for different cutoff points, including "atypical considered positive," "suspicious of malignancy considered positive," and "malignant considered positive." Summary receiver operating characteristic curves assessed diagnostic accuracy, and the diagnostic odds ratio was pooled. Sixteen retrospective cross-sectional studies, totaling 19,128 cases, were included. Sensitivity and specificity for the “atypical and higher risk categories” considered positive were 77% (95% confidence interval [CI], 68%–84%) and 95% (95% CI, 93%–97%) respectively. For the “suspicious for malignancy and higher risk categories” considered positive, sensitivity and specificity were 57% (95% CI, 49%–65%) and 100% (95% CI, 99%–100%) respectively. Sensitivity and specificity for the “malignant” category considered positive for malignancy were 70% (95% CI, 60%–77%) and 99% (95% CI, 98%–99%), respectively. The pooled area under the curve ranged from 85% to 89.5% for each cutoff. This meta-analysis underscores the ISRSFC's accuracy in reporting serous fluid cytology. It emphasizes the diagnostic importance of the "suspicious" and "malignant" categories in identifying malignancy, and the role of the "benign" category in ruling out malignancy.