A retrospective case-control study was conducted using the records of 80 dogs with visceral hemangiosarcoma (HSA) and 200 dogs with various diseases that had features similar to HSA. All dogs were older than one year of age, had histologically confirmed disease, and had a complete blood count performed prior to the final diagnosis. A standard protocol was used to count acanthocytes in the blood film of each dog. Acanthocyte count had a diagnostic sensitivity of 53.8% (and specificity of 61.5%) at a cutpoint of $\ge$1 acanthocyte/2000 red blood cells. A diagnostic specificity of 100% (and sensitivity of 7.5%) was achieved at a cutpoint of $>$71 acanthocytes/2000 red blood cells, which represented a marked acanthocytosis. The precision of acanthocyte count was poor to fair due to the subjective nature of identifying acanthocytes. Although dogs with acanthocytes were more likely to have HSA (P = 0.02), and dogs with HSA had higher acanthocyte counts than controls (P = 0.003), acanthocyte count had limited utility as a diagnostic test. There was no level of acanthocytosis at which HSA could be ruled out and, although HSA could be ruled in at counts $>$71 acanthocytes/2000 red blood cells, only a small percentage of dogs with HSA would be detected at this level. A multivariable logistic regression analysis was performed using demographic, hematology, and serum biochemistry data for the study population. The resulting predictive model for HSA included age greater than 8 years, Golden retriever breed, anemia, and thrombocytopenia. The model had a sensitivity of 83.9% and a specificity of 80.9% in predicting the presence of HSA. In a related model, acanthocyte count was retained as a significant variable, demonstrating that although acanthocyte count may not be useful as a separate diagnostic test for HSA, it can contribute to diagnostic certainty in the context of other findings. Hypoalbuminemia and elevated serum alkaline phosphatase were associated with the absence of HSA, but inclusion of these variables did not improve the diagnostic performance of the logistic model. Therefore, although select serum biochemistry variables may help to rule out HSA, a biochemistry profile is unlikely to contribute to the certainty of the diagnosis.