Diagnostic yield of cytopathology in evaluating pericardial effusions: Clinicopathologic analysis of 419 specimens.

BACKGROUND

Pericardial effusions can cause considerable morbidity and potentially may lead to mortality. Malignant pericardial effusions are uncommon, and data on malignancies encountered in pericardial effusion cytology specimens are limited.

METHODS

Relevant records of all pericardial effusions from January 2008 to September 2014 were examined and compared with pericardial biopsy results when performed. Discrepant cases were reviewed to determine the cause of the disagreement.

RESULTS

In total, 419 pericardial effusion specimens obtained from 364 patients were examined. Cytologic diagnostic categories included: negative for malignancy (332 specimens; 79%), equivocal (25 specimens; 6%), and positive (62 specimens from 51 patients; 15%). Forty-seven patients who had positive effusions were known to have malignancy. The most common primary malignancies were breast (39.3%) and lung (39.3%) cancers in women and lung cancer (47.4%) in men. A concurrent pericardial biopsy was performed in 46% of patients. Excluding equivocal cytologic diagnoses, cytology and biopsy were concordant in 153 of 173 paired samples (88.4%). The sensitivity of cytology in diagnosing malignancy was 92.1% compared with 55.3% for pericardial biopsy.

CONCLUSIONS

Cytologic examination has significant diagnostic utility in the evaluation of pericardial effusions and exhibits a lower false-negative rate compared with pericardial biopsy. Submission of pericardial biopsy alongside effusion cytology is associated with increased sensitivity for detecting malignancy and may be especially useful in the setting of low-volume pericardial effusion. Cancer Cytopathol 2017;125:128-137. © 2016 American Cancer Society.