The progress of prothrombin time measurement

Abstract

Warfarin is the most widely used medicine for oral anticoagulant therapy (OAT). It inhibits the synthesis of coagulation factors II, VII, IX, and X in the liver and results in the production of inactive or partially active versions of these factors. Inactive coagulation factors interfere with prothrombin time measurement (Quick and Owren PT) measuring the sum of coagulation activity and inhibition. The narrow therapeutic range here involves a danger of serious complications and the risk of bleeding or thrombosis. The new-generation PT method can measure coagulation activity and inhibition separately. This new technique promotes patient care and anticoagulant medication (warfarin, dicoumarol) based on coagulation activity in vivo. Both therapy and laboratory controls should be unquestionably accurate and based solely on in vivo coagulation activity. Inactive coagulation factors (inhibition) render measurement, calibration, and harmonization. The use of the new-generation PT method based on measurement of coagulation activity in vivo could develop vitamin K antagonist (VKA) therapy for the marked benefit of patients.

History

Discovery of vitamin K antagonists (VKAs) in anticoagulation occurred in the 1920s, when the veterinarian Frank Schofield studied the hemorrhagic disease affecting cattle consuming sweet clover. He ascribed the bleeding to a toxin in the clover.¹ Karl Link and his team studied spoiled sweet clover and, in 1939, extracted dicumarol and identified the hemorrhagic agent, which was the substance in the sweet clover affecting coagulation.² The discovery of dicoumarol made it possible to inhibit thrombosis and to study anticoagulation therapy for humans in the early 1940s.^3,4 In the decades 1930 and 1940 Professor Armand Quick developed a routine prothrombin time (PT) coagulation test, which preceded the use of VKAs.^5,6 This has served as a basis for oral anticoagulant therapy (OAT) monitoring from the onset. However this first drug for OAT subsequently was found to have drawbacks by reason of its long half-life. Karl Link⁷ synthesized more than 150 anticoagulant compounds and found one particularly active molecule, which was named warfarin (a 4-hydroxy compound) after the patent holder, the Wisconsin Alumni Research Foundation. Professor Paul Owren found factor V and developed a new PT method⁸ subsequent to the “Quick method” to overcome its drawbacks. The Owren reagent (combined thromboplastin reagent, Thrombotest) is used mainly in the Nordic countries, Benelux, and Japan. By reason of its different reagent composition, the Quick technique is sensitive to the coagulation factors fibrinogen, II, V, VII, and X, while the Owren technique is affected by deficiencies in factors II, VII, and X. The Quick method measures factor V and fibrinogen, which are not dependent on warfarin therapy, and this constitutes a drawback for OAT.

Warfarin

The coumarins or VKAs have been the mainstay of OAT for over 50 years. Warfarin is the most widely used drug for OAT world-wide, and it has a predictable onset and duration of action and excellent bioavailability. The therapy is effective for a variety of clinical indications of anticoagulation. The dosage and the clinical response still vary markedly among patients, depending on genetic inheritance, age, and metabolism.

Warfarin is a racemic mixture of two optically active isomers, the R and S forms. It is rapidly absorbed from the gastrointestinal tract (within 90 min) and its half-life is 36–42 h. In the circulation warfarin is bound to plasma proteins (mainly albumin). It accumulates rapidly in the liver, where it is metabolized.⁹ Warfarin interferes with the cyclic intercon-version of vitamin K and its 2,3 epoxide (vitamin K epoxide) and thus is a VKA. Vitamin K is a cofactor in the carboxylation of vitamin K-dependent coagulation factors. VKA inhibits the synthesis of coagulation factors II, VII, IX, and X in the liver, and they remain partially inactive unless 9 to 13 of the amino terminal glutamate (Glu) residues are carboxylated to form the Ca²⁺-binding γ-carboxyglutamate (Gla) residues (Figure 1).^9–11 This carboxylation step renders the coagulation factors functionally active: binding to Ca²⁺ and the phospholipid surface.¹⁰ Therapeutic dosages of warfarin decrease, by 30–50%, the total amount of each vitamin K-dependent coagulation factor synthesized by the liver. The secreted molecules from the liver are under-carboxylated, resulting in diminished biological activity (10–40% of normal).¹¹

Figure 1.

The enzyme reactions involved in the metabolic function of vitamin K. Vitamin K-dependent carboxylase catalyzes the transformation of peptide- (Factors II, VII, IX, and X) bound glutamate residues (Glu) to γ-carboxyglutamate (Gla) residues in the presence of vitamin K hydroquinone, carbon dioxide, and molecular oxygen (I). Vitamin K hydroquinone is oxidized in the reaction to vitamin K 2,3 epoxide. The reduction of the latter to vitamin K quinone is catalyzed by vitamin K epoxide reductase, which can use certain dithiols as the reductants (II). Vitamin K quinone can be reduced to vitamin K hydroquinone in the reactions catalyzed by either a dithiol-dependent (III) or an NAD(H)P-dependent (IV) entzyme. Warfarin (WARF) blocks reactions II and III. (Uotila L. Recent findings on the functions and requirements of vitamin K in humans. Klinlab 1998; 3: 97–101)

Warfarin treatment reduces the number of Gla residues (normal, 9 to 13) per clotting factor molecule, with a concomitant fall in coagulant activity. When the number of residues decreases from 13 to 9, only 70% of the activity of the clotting factor remains; when one molecule contains six carboxylated residues, only 2% activity is present.¹² The liver excretes both active and inactive coagulation factors to the plasma and both affect International Normalized Ratio (INR) measurement,¹³ which possibly has escaped notice in the World Health Organization (WHO) recommendation for the prothrombin time methodology.

International Normalized Ratio

The calculation formula is for: INR = (sample_sec/normal_sec)^ISI, where ISI is the International Sensitivity Index. When the ISI is near 1.0, the reagent is sensitive and ISI has little meaning in the INR calculation.

The Quick and Owren PT methods are the most common and generally accepted means of monitoring VKA therapy. A comparison between the PT methods has been published in a recent article¹⁴ and the Owren PT was superior. WHO recommends the use of the INR to harmonize PT results and therapeutic ranges globally both for patient care in clinical practice and in the scientific literature, as the units used formerly proved inadequate for international communication.^15,16 The challenge is for global clinical laboratories to harmonize INR testing further to a level where INR results are consistent regardless of the methods used.

The use of the INR system still involves difficulties with sample citrate concentration,^17,18 different reagents and thromboplastins,¹⁹ and instruments,^20,21 and with ISI and “local ISI” calibration^22–28 to harmonize results. Horsti and colleagues measured 150 samples from patients on oral anticoagulation using seven commercial reagents and four different calibrator kits. Agreement between INR results was poor.²⁹

ISI calibration

The reagent manufacturer informs the ISI value or the laboratory can measure the ISI with an ISI calibration kit (local calibration) for a PT reagent lot. The meaning of ISI as power in the INR calculation equation is displayed in the previous section.

First, the aim in the original recommendation^16,30 was to harmonize INR results by calibrating reagent ISI values with Human Combined, which was the primary reference preparation (IRP code 67/40). Thromboplastins from different sources (human brain, rabbit brain, rabbit lung, and ox brain) yield quite different levels of PT. The hierarchy of reference thromboplastin preparations was presented by van den Besselaar and associates.³¹ As the WHO calibration procedure was complex and demanding in the second stage, the recommendation for ISI calibration was local calibration using certified lyophilized plasmas.^32–37 A third possible approach for calibration was presented by van den Besselaar and group. This alternative means of determining the ISI involves the use of freshly pooled plasmas from 20 normal individuals and 60 patients receiving coumarin (OAT). Such numbers of samples are necessary to obtain a precise calibration line for ISI calculation. Freshly pooled plasma can be used to determine reagent or instrument ISI with acceptable precision, or as good a result as with the WHO calibration model.³⁸

Poller and colleagues (European Concerted Action on Anticoagulation, ECAA) have compared local ISI calibration and “Direct INR” in the correction for locally reported INRs. For local INR correction (harmonization) they used seven normal plasmas and 20 from patients on warfarin. The results were variable.³⁹ It is clear that plasmas from patients on warfarin therapy always involve varying amounts of inactive coagulation factors, depending on the level of anticoagulation and individual patient characteristics (see section on Warfarin).

All these former calibration models involve the principle that the calibrator contains an average amount of inactive coagulation factors (inhibition), which means an average correction for patient INR results. Is the average correction of inactive coagulation factors appropriate for individual patient samples in a measuring range from 1 to 5 INR? Is it a fact that calibrators, according to biochemical principles, contain inactive coagulation factors (inhibition)? Do different reagents behave identically to active and inactive coagulation factors?²⁹ In our earlier studies we measured inactive coagulation factors (inhibition) in four different kits and noted conspicuous variation in inhibition.^40,41 ISI calibration should be based on normal plasma and normal plasma dilutions⁴² without inactive coagulation factors. In principle, calibrators should not contain inhibitory coagulation factors, which cause error in calibration.⁴⁰

The new-generation prothrombin time method

Today the Quick and Owren PT methods measure the sum of active coagulation factors (FII, FVII, FX) and inhibition by their inactive coagulation factor counterparts. The new-generation PT method can measure separately both active coagulation factors (FII, FVII, FX) and the inhibition caused by inactive or partially inactive coagulation factors.⁴⁰ Inhibition varies markedly between individual patient samples, depending on the medication and metabolism. Correction of inhibition must be made individually for each patient sample.⁴⁰ In this method two measurements for one sample are needed and the cost per sample is twice as much. The average correction does not suffice for accurate INR results. In addition, the number of inactive coagulation factors varies markedly in different calibrator kits (manufactured or local). This means difficulties in calibration and harmonization of different reagents.^29,41 The harmonization of INR results using different PT methods and reagents is problematic if attempted according to the WHO recommendation.^29,43 Measuring only the active coagulation factors without inhibition harmonization between different reagents succeeds well.⁴¹

Conclusions

New medications for OAT have been developed over a number of years and anticipated without laboratory test control in an effort to displace warfarin medication. The new medicines, however, have not proved their superiority over warfarin. The new molecules are too expensive for global use or involve serious side-effects and possibly will never be as popular as warfarin, which is an old and cheap means widely used and accepted globally. The only drawback with warfarin medication is regular laboratory control. It would be important to develop warfarin therapy with greater attention to laboratory control, which helps patient care. Different reagents and thromboplastins react variably with inactive coagulation factors and cause difficulties in calibration using patient plasmas that contain inhibiting coagulation factors. Data on inactive coagulation factors are individual and should be corrected individually. Every new anticoagulant therapy patient should be tested for inhibition at the commencement of therapy.

The accuracy and harmonization of patient INR results for different reagents is of the utmost importance for scientific publications, medication, and patient care. Today's measuring principle is taking the sum of active coagulation factors (FII, FVII, FX), and inhibition of inactive coagulation factors is not a satisfactory approach from the standpoint of accurate patient care. Active coagulation factors in vivo are responsible for thrombosis and bleeding. The errors in INRs are too great, which in many ways affects the success of medication and patient well-being, also using the same calibration (local calibration).²⁹ The new-generation PT offers possibilities of more accurate INR results and patient care based on control of active coagulation factors. The inhibition renders measurement, calibration, and harmonization.^29,40,41 The therapeutic INR ranges guiding anticoagulant therapy using “old methods” are based on the principle that both calibrators and patient samples have inactive coagulation factors (inhibition) on average, which compensate for each other in the final INR result. Thus the therapeutic ranges are available for the new-generation PT method. The therapeutic range lies between thrombosis and bleeding, and complications are serious and general.⁴⁴ The medication and care of OAT patients must be rendered better and safe using a more sensitive PT test.