Assessment of serum amyloid A testing of horses and its clinical application in a specialized equine practice.

OBJECTIVE To compare serum amyloid A (SAA) concentration, plasma fibrinogen concentration, total WBC count, and serum albumin-to-globulin concentration ratio (A:G ratio) in clinically normal (CN) and clinically abnormal (CA) horses. DESIGN Prospective cohort study. ANIMALS 111 CN horses and 101 CA horses hospitalized at a specialty clinical practice. PROCEDURES Shortly after admission, a blood sample (20 mL) was collected from each horse for a CBC, serum protein electrophoresis, and determination of plasma fibrinogen concentration; SAA concentration was assessed with a previously validated immunoturbidometric assay. Similar testing of a subset of CA horses was conducted at various points during treatment. RESULTS Total WBC count, A:G ratio, and SAA concentration were determined for all 212 horses; data regarding plasma fibrinogen concentration were available for 127 horses (of which 47 were CN and 80 were CA). Median SAA concentration, total WBC count, and plasma fibrinogen concentration and mean A:G ratio differed significantly between CN horses and CA horses. Correlations between these variables were poor to weak. For discrimination of CN horses from CA horses, the SAA assay had sensitivity of 53% and specificity of 94% (diagnostic accuracy, 75%); for the other assessments, accuracy ranged from 59% to 62%. Repeated assessment of SAA concentration in some CA horses revealed a gradual return to normal concentrations. CONCLUSIONS AND CLINICAL RELEVANCE Results indicated that assessment of SAA concentration can provide valuable information regarding the clinical state of horses and may be more useful for patient monitoring and as a prognostic indicator than are traditional markers of inflammation.