Type of smear may influence thrombopoietic cell counts in the bone marrow of clinically healthy dogs

Background: The expected number of thrombopoietic cells in normal canine bone marrow is poorly defined and there is no consensus on the most appropriate way to prepare cytologic smears to evaluate these cells nor on the optimum method for their quantification. Objective: The purpose of this study was to determine total and differential counts of thrombopoietic cells in the bone marrow of clinically healthy Beagle dogs by comparing 4 different smear types and bone marrow core biopsies. Methods: Twenty-two clinically healthy, male Beagle dogs, 10 to 12 months old, were used in the study. Following bone marrow aspiration and core biopsy from the iliac crest, Giemsa-stained smears were prepared by 4 techniques: drop-squash, particle-squash, buffy coat, and fat-layer smears. Thrombopoietic cells were counted in up to 100 low-power fields (LPF, x10 objective) in the aspiration smears and in all possible high-power fields (HPF, x40 objective) in H&E-stained biopsy sections. Results: Mean total thrombopoietic cell counts were 2.76 cells/LPF (drop-squash), 1.55 cells/LPF (particle-squash), 8.05 cells/LPF (buffy coat), and 3.08 cells/LPF (fat-layer). Core biopsies yielded 5.31 cells/HPF but frequently failed to provide interpretable specimens. There was a significant difference in cell counts among the 4 smear types (P < .001). Based on evaluation of buffy coat smears, thrombopoietic cells included 1.23% megakaryoblasts, 8.77% promegakaryocytes, and 90% megakaryocytes, with a mean maturation index of 0.11. Conclusion: Thrombopoietic cell counts in canine bone marrow are influenced by the smear technique. Buffy coat and fat-layer smears may be useful to obtain cellular smears in hemodiluted or small aspirate samples. ©2005 American Society for Veterinary Clinical Pathology