Background: We tried to evaluate the diagnostic utility of HBME-1 in distinguishing between reactive meso-thelial cells and adenocarcinoma in body serous effusions.

Methods: We examined 52 cytologic specimens of serous effusions processed by cell block technique retrieved from the pathology archive of Al-zahra hospital (Isfahan). They were categorized in two groups: Group I. 26 effusions containing reactive mesothelial cells from patients with no evidences of malignancy based on cyto-morphology, clinical data and imaging; and Group II. 26 effusions containing adenocarcinomatous cells from patients with diagnosis established by routine histology. Immunostaining with HBME-1 was performed using an Envision technique. Statistical analysis was performed with SPSS software.

Findings: Statistical significance was found with HBME-1 when comparing both adenocarcinoma versus meso-thelial cells (P = 0.001). Also, we found HBME-1 outlined cell membranes in reactive mesothelial cells versus cytoplasmic pattern in adenocarcinoma cells (P = 0.001). The staining intensity for adenocarcimoma cells in-cluded: Negative in 7 cases (26.9%); score + in 4 cases (15.38%), score ++ in 11 cases (42.30%) and score +++ in 4 cases (15.38%). In mesothelial cells, Negative and score + was not seen, score ++ was in 3 cases (11.5%), and score +++ in 23 cases (88.5%) (P = 0.001).

Conclusion: The staining pattern and intensity for HBME-1 is a usufull panel for differentiation of adenocarci-noma and mesothelial cells. The only limitation of this marker is ovarian carcinoma that shows the same pattern as reactive mesothelial cells.

Keywords: Serous fluid, adenocarcinoma, reactive mesothelial, HBME-1 antigen, immunocytochemistry.