Multi-Center Study of Replacement of the International Reference Preparation for Thromboplastin, Rabbit, Plain

Abstract

A multi-center study of the three W.H.O. international reference preparations (IRP) for thromboplastin has been performed. The main purpose of the study was the determination of the mean international sensitivity index (ISI) of a proposed replacement IRP for thromboplastin, rabbit, plain. Twenty laboratories in Europe, North America, and Australasia participated in the study. All clotting times were determined with the manual technique. The inter-laboratory variation of the ISI was considered as the relevant parameter for assessment of ISI precision. Low interlaboratory variation was observed in the calibration of rabbit thromboplastin against rabbit thromboplastin (2.9-3.0% coefficient of variation of calibration line slope). Relatively high interlaboratory variation was observed in the calibration of rabbit against human thromboplastin (3.7–7.2 CV). Intermediary interlaboratory variation was obtained in the comparison of rabbit or human plain thromboplastin with bovine combined thromboplastin (3.5-4.0% CV). These findings confirm the principle of like-to-like calibration in biological standardization. The mean ISI values based on the IRP for rabbit thromboplastin (RBT/79) are in excellent agreement with those obtained with the IRP for bovine thromboplastin (OBT/79). In contrast, the mean ISI values determined with the IRP for human thromboplastin (BCT/253) were 6% lower than those with RBT/79. This bias may be caused by a difference in design of the two historical calibration studies. The mean ISI of the proposed replacement IRP for thromboplastin, rabbit, plain obtained by calibration against RBT/79 was 1.035 (standard error: 0.027). It is recommended to round the mean ISI off to 1.0. Opportunity was also taken to compare a lyophilized normal pooled plasma with fresh individual normal plasmas. The mean PT-ratio of this lyophilized plasma obtained with RBT/79, and BCT/253 was 1.02, which was slightly but significantly higher than the mean PT-ratio obtained with OBT/79 (0.99). INR equivalents were determined for a lyophilized pooled plasma derived from patients treated with oral anticoagulants. There were small but significant differences between the various thromboplastins, which may be caused by slight activation of the plasma or by the lyophilization process.