Semin Thromb Hemost 1999; 25(1): 17-25
DOI: 10.1055/s-2007-996419
Copyright © 1999 by Thieme Medical Publishers, Inc.

Standardization of Prothrombin Time for Laboratory Control of Oral Anticoagulant Therapy

Steven Kitchen, Francis Eric Preston
  • From Sheffield Hemophilia and Thrombosis Center, Royal Hallamshire Hospital, Sheffield, United Kingdom.
Further Information

Publication History

Publication Date:
06 February 2008 (online)


Laboratory monitoring of oral anticoagulants is mandatory to ensure efficacy and safety of therapy. The test of choice is the prothrombin time (PT), using thromboplastins, which vary markedly in respect of the defect induced by oral anticoagulants. Standardization is improved by the use of the WHO calibration model for thromboplastins. This system defines the International Sensitivity Index (ISI) of a reagent in relation to a reference material. This is then used together with a locally determined mean normal PT to derive the International Normalized Ratio (INR). There is an inverse relationship between sensitivity and ISI, the lower the ISI the greater is the sensitivity of the reagent. There are important practical advantages to using a sensitive reagent with ISI of 1.2 or less. The ISI system was originally designed for manual tests and in some cases the ISI is influenced by the use of coagulometers, which generally have the effect of lowering the ISI. For this reason the ISI should be assigned for the combination of reagent and endpoint detection system. A thromboplastin should not be used with techniques for which the ISI is unknown. In some instances there is variation between instruments of the same type, in which case a local calibration using plasma calibrants with assigned values can be considered. The precise requirements for such local calibrations are presently the subject of research in a number of centers. The INR/ISI system is increasingly being adopted in many countries for anticoagulant therapy. Some past difficulties leading to discrepancies between results with different reagents have now been resolved and the view that this is the best available system for standardization of the PT for monitoring oral anticoagulants is widely held.