Evaluation of the Hemochron Junior Signature Analyser for Oral Anticoagulant Point of Care Testing.

The community anticoagulant service in Tayside, Scotland provides point of care testing for around 5000 patients at 25 sites using 43 analysers. Since December 2004 this service has used the Hemochron Junior Signature (HJS) analyser (International Technidyne Corporation) for INR measurement.

Following satisfactory initial laboratory evaluation, we have compared results of capillary samples analysed using the HJS to venous samples collected simultaneously and analysed in Ninewells Hospital using a Sysmex CA1500 analyser. Reagents used were Citrate PT test cuvettes (International Technidyne Corporation) which uses rabbit brain thromboplastin for the HJS and Dade Innovin (Dade Behring) for the CA1500. Samples for comparison were collected randomly from approximately every tenth patient tested in the anticoagulant clinics for a 12 month period. Results of the comparison are given for 3 separate periods within the 12 months of study as ongoing analysis of data during the study period led to the introduction of a correction factor for results obtained using the HJS analyser. The first correction factor was introduced after the first 4 weeks of study and there was a subsequent revision of this correction factor after a further 5 months.

Results for these 3 periods are as follows (all results given are INR units).

Period 1:

Mean INR 2.8 CA 1500 vs. 3.8 HJS, mean difference −1.0, 95% upper limit of agreement 0.5, 95% lower limit of agreement −2.5 (n=145).

Period 2:

Mean INR 2.6 CA 1500 vs. 2.7 HJS, mean difference −0.1, 95% upper limit of agreement 1.1, 95% lower limit of agreement −1.2 (n=345).

Period 3:

Mean INR 2.7 CA 1500 vs. 2.8 HJS, mean difference 0.0, 95% upper limit of agreement 1.3, 95% lower limit of agreement −1.3 (n=662).

Additionally, local external quality assurance was performed on 13 occasions (683 total tests) during the period of study. Cumulative results demonstrate that 86% of results were within 15% of the median.

We have demonstrated an approach to quality assurance for oral anticoagulant point of care testing and conclude that the HJS is a suitable analyser for this purpose.