Bone marrow necrosis in dogs: 34 cases (1996-2004).

Objective: To identify the incidence, potential causes, and clinical and clinicopathologic features of bone marrow necrosis in dogs.
Design: Retrospective study.
Animals: 34 client-owned dogs.
Procedures: Reports of cytologic examinations of bone marrow specimens performed between 1996 and 2004 were reviewed. All reports that indicated the presence of necrosis, stromal disruption, phagocytic macrophages, individual cell necrosis, or myelofibrosis were evaluated further.
Results: Of 609 reports of bone marrow evaluations performed during the study period, 34 (5.6%) had evidence of bone marrow necrosis. Nine dogs had no evidence of associated diseases or drug or toxin exposure, and 25 dogs had associated disease conditions or drug exposures. All 9 dogs with idiopathic bone marrow necrosis were anemic (mean Hct, 14%), but only 3 had neutropenia, and 3 had thrombocytopenia. All 9 had myelofibrosis. Of the 25 dogs with associated disease conditions or drug exposures, only 14 (56%) had anemia (mean Hct, 33%). In addition, 14 (56%) had neutropenia and 18 (72%) had thrombocytopenia. Only 10 (40%) had myelofibrosis.
Conclusions and Clinical Relevance: Results suggest that bone marrow necrosis may be common in dogs with hematologic disorders. In most dogs, bone marrow necrosis was associated with an underlying disease condition or drug exposure, but idiopathic bone marrow necrosis was also identified. Disease conditions that should increase suspicion of possible bone marrow necrosis include sepsis, lymphosarcoma, and systemic lupus erythematosus; drug exposures that should increase suspicion of possible bone marrow necrosis include chemotherapeutic agents, phenobarbital, carprofen, metronidazole, and mitotane.