[Evaluation of the quality of prothrombin time technique in the Basque Country].

PURPOSE

The Programme for Quality Assessment in Anticoagulant Therapy, which was started in the Basque Country in 1984, included the manufacturing of a standard human thromboplastin reagent plus periodical quality controls of the prothrombin time. A rabbit thromboplastin reagent was available in 1994; it was called Thromboplastin Bilbao, or TBi, and had to be used as a pattern reagent. Its stability, sensitivity and composition were plotted against the Manchester Reagent, of similar nature and composition. A new phase of quality quality controls was simultaneously started with this new reagent, a group of hospitals of the Autonomous Community taking part in the programme. The aim was to compare this reagent with the others used by the hospitals and to evaluate the variations with regard to the stages previous to the introduction of the human reagent.

MATERIAL AND METHODS

Nine centres joined the study. Controls were performed every fourth month including plasma from patients under anticoagulant treatment, plus the reference reagents. The partaking centres used their usual reagents and routine procedures, as well as the reference reagents and manual technique.

RESULTS

The coefficient of variation (CV) of the INR of anticoagulated plasmas used for these studies when following the centres' own methods and thromboplastins was 9.5 +/- 6.40%, versus 9.94 +/- 1.57% when TBi was used. These results are in accordance with those attained with the previously manufactured human reagent, whose CV was 8.75 +/- 2.19% in the quality assurance methods performed in the period 1984-88. Concurrently with the programme, the reagent was sent as a part of the Spanish Programme for Quality Assessment in Anticoagulant Treatment to the 51 centres taking part in the study. The CV of the INR of the two lyophilized plasmas sent was, respectively, 14.97% and 16.5%. The average value for the healthy subjects was 14.77 seconds, with 8% inter-centres variation when using manual methods.

CONCLUSIONS

These results suggest that a thromboplastin is now available whose sensitivity and reproducibility make it suitable as the national standard for usage in the quality assessment of prothrombin time.