Monitoring Oral Anticoagulant Therapy by the Prothrombin Time: Reporting the Coagulation Activity or the International Normalized Ratio?

In 1935, Quick et al. [1] described the one-stage prothrombin time (PT) test of the extrinsic pathway of the coagulation cascade. In the present nomenclature, the PT should be referred to as the tissue factor-induced coagulation time. With modifications, this test continues to be the principal method for monitoring oral anticoagulant therapy. Quick [2] recommended a detailed procedure for the preparation and use of rabbit brain thromboplastin to ensure consistent results at any time or place. In defiance of Quick’s detailed description, many modifications of the test were developed and used. As a consequence of the modifications of the original test system, the results obtained by different laboratories could not be compared directly. Ideally, all laboratories should use the same well standardized thromboplastin preparation and should employ comparable techniques. This would require some form of central control but it is unlikely that this can be accomplished on a wide scale. Instead, standardization of the PT may be achieved by transformation of the PT values obtained with any reagent and technique into standardized values. It was soon realized that the percentage of prothrombin complex activity obtained with reference to dilutions of pooled normal plasma depends heavily on the thromboplastin reagent and the diluent used [3, 4]. For example, a therapeutic target range of 510% Thrombotest (Nycomed) activity was found to be equivalent to 10-20% Hepato Quick (Boehringer) and to 18-30% Thrombo-plastin-C (Dade). In recent years, the international normalized ratio (INR) has been recommended as the universal scale for reporting the PT in oral anticoagulant monitoring [5]. It should be emphasized that the INR was defined only for this purpose and not for screening of the extrinsic pathway factors. The INR is based on two principles: (a) the establishment of an international reference preparation for thromboplastin together with a method to use this preparation, and (b) the calibration of other PT systems against the reference system by testing fresh samples from both healthy individuals and patients stabilized on oral anticoagulant therapy. Although the INR system meets a number of practical problems (see below), it can reduce the differences between laboratories considerably [6]. In the present issue of Haemostasis two articles are published describing so-called simplified prothrombin time standardization methods [7, 8]. The proposed ‘modified coag-