Cytohistological correlation in serous effusions using the newly proposed International System for Reporting Serous Fluid Cytopathology: Experience of an oncological center

Abstract

Background Cytological analysis is part of the initial etiological evaluation of serous effusions. The newly proposed International System for Reporting Serous Fluid Cytopathology (ISRSFC) aims to standardize reporting. Methods All pleural and peritoneal effusion samples admitted for cytological analysis at our institution between 2012 and 2016, and pericardial effusion samples admitted between 2008 and 2018, were reviewed and reclassified according to the ISRSFC. Risk of malignancy (ROM) and performance parameters were calculated. Results 1496 pleural effusion samples were reclassified: 12(0.8%) non‐diagnostic (ND), 944(63.1%) negative for malignancy (NFM), 9(0.6%) atypia of undetermined significance (AUS), 54(3.6%) suspicious of malignancy (SFM) and 477(31.9%) malignant (M). 64 pericardial effusion samples were reclassified: 23(35.9%) NFM, 1(1.6%) AUS, 4(6.3%) SFM and 36(56.2%) M. 763 peritoneal effusion samples were reclassified: 5(0.7%) ND, 457(59.9%) NFM, 12(1.6%) AUS, 37(4.8%) SFM and 252(33%) M. The ROM was, respectively, for each of the aforementioned categories, 57.1%, 23.9%, 50%, 76.2%, 100% in pleural effusions, 100%, 26.3%, 62.5%, 91.7%, 100% in peritoneal effusions and 0% for NFM, 0% for AUS and 100% for M in pericardial effusions. Sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy were, respectively, 61.6%, 100%, 100%, 73.3%, 81.3% for pleural, 100%, 100%, 100%, 100%, 100% for pericardial and 61.2%, 100%, 100%, 70%, 79.7% for peritoneal effusion samples. Conclusion Serous effusion cytology has a high specificity and positive predictive value and a modest sensitivity and negative predictive value, supporting its role in confirming the diagnosis of malignancy. The ISRSFC will increase standardization and reproducibility in reporting, leading to improved clinical decision‐making.