Analytical quality assessment and method comparison of two immunoassays for the measurement of serum cardiac Troponin I in dogs and cats.

Abstract

BACKGROUND: Cardiac troponin I (cTnI) is a sensitive and specific biomarker for myocardial injury. Validation of point-of-care (POC) analyzers for cTnI measurement is valuable to the critical care setting, in which rapid results can facilitate prompt diagnoses. An immunoassay for detecting cTnI is available for the POC AIA-360 analyzer (Tosoh Bioscience), but this has not been validated using canine and feline serum. OBJECTIVES: The objectives were (a) to determine precision, accuracy, and linearity of cTnI measurement using the AIA-360 immunoassay in pooled canine and feline samples, and (b) to compare results for individual canine and feline samples with those obtained using a reference chemiluminescence method (Immulite 1000, Siemens). METHODS: Intra- and inter-assay repeatability was determined using pooled canine and feline samples, and the coefficient of variation (CV) was calculated for each. Pooled samples were also serially diluted to assess linearity. A modified spike and recovery analysis was performed by mixing pooled samples with different concentrations. Bland-Altman and Deming regression analyses were used to determine bias for individual samples, and the total observed error (TE) was calculated. RESULTS: Coefficient of variation values were well within the required maximum of 20%. Linearity was demonstrated over the range of samples tested, and the recovery study showed minimal proportional inaccuracies. Although the correlation between the analyzers was excellent, there was a large mean bias due to relative proportional bias. Total observed error consequently exceeded the total allowable error (TE). CONCLUSION: Although, in most respects, the analyzer demonstrated adequate performance, pronounced bias contributed to the large TE, indicating a requirement for analyzer-specific reference intervals.