Influence of biopsy specimen size, tissue fixation, and assay variation on copper, iron, and zinc concentrations in canine livers.

OBJECTIVE-To determine whether metal concentrations in canine liver specimens were influenced by specimen size, assay variability, tissue processing (formalin fixation and deparaffinization), or storage in paraffin blocks. SAMPLE POPULATION-Liver specimens (fresh frozen and deparaffinized) from 2 dogs with chronic hepatitis (high copper but unremarkable iron concentration [liver 1] and unremarkable copper but high iron concentration [liver 2]) as well as fresh and deparaffinized-archived liver specimens from 20 dogs with various hepatopathies. PROCEDURES-Fresh frozen liver specimens (obtained via simulated needle-core and wedge biopsy), fresh hepatic tissue, and deparaffinized-archived specimens (0.5 to 14 years old) were analyzed for concentrations of copper, iron, and zinc by atomic absorption flame spectrometry. Clinical severity scores were assigned on the basis of tissue metal concentrations. RESULTS-Interassay variation of metal standards was < 4%. Measurements of liver tissues on 8 consecutive days yielded high coefficients of variation (3.6% to 50%) reflecting heterogenous histologic metal distribution; variation was highest in liver 1 and deparaffinized-archived tissues. Heterogenous metal distribution was confirmed by histologic evaluation. The largest range of metal concentrations was detected in wedge biopsy specimens. In tissues with high metal concentrations, copper and iron concentrations were significantly lower in needle-core versus wedge biopsy specimens. A higher zinc concentration in deparaffinized-archived specimens masked a low zinc concentration in fresh liver tissue of 10 of 20 (50%) dogs. CONCLUSIONS AND CLINICAL RELEVANCE-Retrospective measurement of copper and iron concentrations but not zinc concentrations in deparaffinized-archived liver specimens provided relevant information. The value of needle-core biopsy specimens for measurement of metal concentrations is questionable.