Abstract

Objectives To assess the imprecision and stability of the point-of-care troponin I assay in the Emergency Department and its correlation and bias to two central laboratory troponin I assays (Siemens Atellica and Abbott Alinity). Methods Imprecision and stability testing was performed on opportunistically selected samples using whole blood in the emergency department by non-laboratory trained personnel. Assay comparisons were undertaken on samples taken from participants of the Mersey Acute Coronary syndrome Rule Out Study. Results The coefficient of variation (95 % confidence interval), at the 99th percentile for the point-of-care assay, was 8.1 % (6.1–12.1 %) but with a wide confidence interval reflective of considerable scatter at values just below the 99th percentile. The 10 % limit of quantification was 7.5 ng/L (1.7–61.8 ng/L). All samples met the ≤2 ng/L stability criteria for a duration of 4 h and under. The point-of-care assay very strongly correlated and had a negative bias with the Siemens Atellica and Abbott Alinity assays, Pearson’s R=0.99 and 0.95, mean difference −29.7 ng/L and −13.3 ng/L respectively. Conclusions The Siemens VTLi point-of-care assay fulfils high-sensitivity criteria when operated by non-laboratory trained staff using whole blood in its intended environment. Lithium heparin samples are likely stable up to 4 h. Significant bias between the point-of-care and two central laboratory assays negates the use of these assays interchangeably.