Morphometric analysis of AgNORs in thin-layer, liquid-based liver specimens.

Objective: To detect argyrophilic nucleolar organizer regions (AgNORs) in ThinPrep (Cytyc Corp., Boxborough, Massachusetts, U.S.A.) liver fine needle aspiration (FNA) specimens and to define the diagnostic value of their quantitative analysis in the evaluation of hepatic lesions.
Study Design: ThinPrep liquid-based FNA biopsy specimens from 49 malignant and benign liver lesions were resampled, fixed in 95% ethanol and stained with the AgNOR technique in accordance with the 1-step colloid method. The specimens included 11 benign and 38 malignant lesions (23 poorly differentiated hepatocellular carcinomas [HCCs] and 15 poorly differentiated metastatic adenocarcinomas [MCs]). Morphometric analysis was performed using a Zeiss Axiolab microscope (Carl Zeiss GmbH, Jena, Germany) with a mechanical stage fitted with a Sony-iris CCD videocamera (Tokyo, Japan). The videocamera was connected to a Pentium III P/C (Intel Corp., Santa Clara, California, U.S.A.) loaded with the appropriate image analysis software. The measurements were performed with ImageScan software (Jandel Scientific, Erkrath, Germany). The number of AgNORs per nucleus (NN) and the total area per nucleus occupied by AgNORs (AR) were calculated semiautomatically. Statistical analysis was performed using the SPSS software package (Chicago, Illinois, U.S.A.).
Results: The least significant deviance test for multiple comparisons revealed that NN differed significantly between the 3 groups of samples examined (P < .0001). The mean NN values in HCCs and MCs were significantly different (P < .0001). Logistic regression model demonstrated that as NN increased, the probability of a MC diagnosis decreased (<4%). AR values were different at a statistically significant level only between benign and malignant specimens (P = .00006), not between HCCs and MCs (P = .933).
Conclusion: Quantitative analysis of AgNORs in ThinPrep specimens could be a diagnostically useful method in liver disease.