Synovial fluid analysis is utilized to diagnose septic synovitis. However, not all cases are clearly and rapidly discernible with the diagnostic tools available in the laboratory. Serum amyloid A (SAA), an acute phase protein, has been shown to be elevated in synovial fluid from inflamed synovial structures. The goal of this study is to describe the correlation between two diagnostic tests measuring equine SAA levels in septic and non-septic synovial structures and to understand the correlation between an elevated SAA result and synovial sepsis. Prospective estimation of sensitivity (Se) and specificity (Sp) of two tests, handheld and ELISA, measuring SAA in synovial fluid was completed in 62 horses presented with injured synovial structures. The comparison was made to a reference diagnosis based on white cell count, percentage of neutrophils, intracellular bacteria and bacterial culture on synovial fluid. Handheld test levels were classified as: 4 lines visible-SAA level negative; 3 lines visible-SAA level mild; 2 lines visible-SAA level moderate; and 1 line visible-SAA level severe and compared to the numerical value obtained with ELISA test. The ELISA SAA test had an area under the curve of 0.88 (0.78-0.98). An ELISA cut-off of 23.95 μg/mL maximized Se and Sp. This cutoff gave a Se of 0.93 (0.66-1.00) and Sp of 0.77 (0.63-0.88). The handheld test was highly correlated with the ELISA SAA test (Spearman rank correlation 0.96) and at a cutoff of moderate or higher for positive results gave identical Se and Sp. Se and Sp of synovial fluid SAA are very reliable when clinical signs of synovitis are present for >6 h. This test, in conjunction with traditional methods, can assist practitioners to rapidly diagnose and expedite appropriate intervention of synovial sepsis.