RESUMO
OBJECTIVE: To determine the diagnostic accuracy of fine-needle aspiration cytology (FNAC) in liver masses to isolate malignant from benign tumours and hepatocellular carcinoma (HCC) from metastatic tumours. STUDY DESIGN: Cross-sectional, observational. PLACE AND DURATION OF STUDY: Department of Histopathology, Combined Military Hospital, Peshawar, from June 2004 to June 2005. METHODOLOGY: All the patients with liver masses confirmed by ultrasonography, irrespective of age and gender, were included. Patients with inflammatory lesions were excluded from the study. Selected patients underwent fine-needle aspiration under ultrasound guidance followed by needle biopsy. The cytological slides were stained by haematoxylin and eosin (H & E) stain, while Papanicolaou's stain was employed in selective cases. Needle biopsy fragments were fixed in formalin followed by paraffin embedding and staining with H and E stain. Sensitivity and specificity of FNAC in the diagnosis of liver masses was determined using histological diagnosis on liver biopsy as gold standard. RESULTS: There were one hundred subjects. The mean age at presentation was 55 + or - 12 years with male to female ratio of 1.7:1. Cytological diagnosis in 19 cases was benign/non-neoplastic and 81 was malignant. Out of the latter, 49 (60.49%) were HCC and 32 (39.51%) were metastatic tumours on cytology. The overall sensitivity, specificity and accuracy of FNAC in the diagnosis of malignant lesions was 95.2%, 100% and 96% respectively using histological diagnosis on liver biopsy as gold standard. Sensitivity of FNAC to differentiate HCC from metastatic tumours in liver was 96% while specificity was 100% having a diagnostic accuracy of 97.5%. The discrepancy in cyto-histological comparison was mainly seen in welldifferentiated and poorly-differentiated HCCs. CONCLUSION: FNAC of the liver masses is a simple, safe, accurate, economical screening test without significant morbiditythat can be used to identify the vast majority of hepatic neoplasms of primary or metastatic nature with high sensitivity, specificity and diagnostic accuracy.