Validation of an equine serum amyloid A assay with an unusually broad working range

Background Serum amyloid A (SAA) is a major equine acute phase protein and of great value in detection and monitoring of inflammation. A new immunoturbidometric assay based on monoclonal antibodies (VET-SAA, Eiken Chemical Co., Japan) may be useful for SAA measurements in routine diagnostic laboratories. The aim of the study was to validate the VET-SAA immunoturbidometric assay and use it to measure serum SAA concentrations in a variety of clinical cases. Precision was assessed by intra- and interassay coefficients of variation of repeated measurements of serum pools (low, intermediate, high concentrations of SAA). Accuracy was estimated by linearity under dilution. Detection limit was determined by replicate determinations of ionized water. Measurements were compared to measurements performed in a previously validated SAA assay (LZSAA assay, Eiken Chemical Co., Japan). Subsequently, the VET-SAA assay was used for measuring serum SAA concentrations in horses with and without inflammation. Results Detection limit was 1.2 mg/L. Without modifications, the assay measured SAA concentrations with acceptable reliability in a broad concentration range (0 to > 6000 mg/L). In the 0-3000 mg/L range, the assay demonstrated good precision and accuracy, and concentrations correlated well with those obtained in the LZSAA assay, albeit with a slight systematic bias. Concentrations of SAA assessed in horses with and without inflammation followed the expected pattern, with significantly higher concentrations in horses with systemic inflammation than in healthy horses and horses with non-inflammatory disease. Conclusions The assay was unique in its ability to measure SAA concentrations with acceptable reliability over an extreme concentration range. This is relevant in the equine species, where SAA concentrations may reach very high concentrations.