Bone marrow (BM) cytology and histopathology are complementary tools used to investigate hematological diseases. The purpose of this study was to determine if there are site-dependent differences in the diagnostic quality, myeloid to erythroid ratio (MER), and discordant findings in samples from different sites in the same dog. Eighteen apparently healthy dogs were used in the study. The sequence of sample acquisition was randomized according to a Latin square, and samples for BM cytology and histology were collected from both humeri and both ilial crests immediately after death. Board-certified clinical and anatomical pathologists read the cytology and histology, respectively. The data were analyzed using a mixed-effect model. The site of BM acquisition did not affect BM sample quality. The rate of discordant clinical findings between sites was 0.05 (95% confidence interval, 0.01-0.13). In general, by cytology, the MERs were slightly but significantly greater in samples from the ilial crests than from the humeri ( P = .01). The measured MER for histology was nearly twice that for cytology for all sites ( P < .001). In conclusion, there was a low-rate, site-dependent discordance in diagnostic findings in BM samples and differences in MER between the ilial crest and the humerus. A similar study is justified in sick dogs with hematological disease to determine the effect of sampling site on discordant findings between sites.
Keywords: anatomy; biopsy; bone marrow; canine; cytology; hematology; quality assurance.