A sensitivity comparison of the Quick and Owren prothrombin time methods in oral anticoagulant therapy

Prothrombin time (PT) is the leading test for monitoring oral anticoagulation therapy (OAT). According to the World Health Organization recommendation, International Normalized Ratio (INR) results obtained from the same patient samples with the major PT methods (Quick and Owren) should be the same when the therapeutic range is the same. In our study blood samples were obtained from 207 OAT patients. We analyzed the samples using two Quick and two Owren PT (combined thromboplastin) reagents for INR and assessed the sensitivity and true coagulation activity using a new-generation PT method. The INR values with the Quick PT and Owren PT methods were very similar around the normal range, while unacceptable differences were seen within the therapeutic range and at higher INR values. The Quick PT results as INR are clearly lower than those given by Owren PT and the difference increases toward higher INR. The new PT method functions well with both Owren PT reagents, and we can calculate the true active INR. The Quick PT methods show no sensitivity to coagulation inhibition measurement. The harmonization of the INR is an important goal for the safety of OAT patients. More accurate INR results reduce morbidity and mortality, and the therapeutic ranges should be similar worldwide. In this study we found unacceptable differences in INR results produced by the two PT methods. The new method showed a lack of sensitivity to Quick PT. For the global harmonization of OAT therapy and for INR accuracy only the more sensitive Owren PT method should be used.


Introduction
Oral anticoagulation therapy (OAT) is one of the most commonly used medications worldwide. The purpose of the treatment is to balance the risks of hemorrhage and thrombosis in the patient. Arterial and venous complications commonly are involved in morbidity and mortality in OAT patients globally. [1][2][3][4] The vita-min K antagonists coumarin and warfarin are inexpensive and the most widely used medicines in the prevention and treatment of thromboembolism in various clinical situations, and the benefit of OAT has been proved. [5][6][7][8][9][10][11][12] The major drawback with warfarin is a narrow therapeutic window and individually variable responses to the treatment. Thus frequent prothrombin time (PT) checks are required to ensure that anticoagulation remains within the therapeutic range, which is 2.0-3.0 International Normalized Ratio (INR).
New medications for oral anticoagulation therapy have been developed over a number of years and anticipated without laboratory test control in an effort to replace warfarin. The new medicines, however, have not proved superior to warfarin. The new molecules are expensive and involve serious side effects or only narrow indications for OAT. 13 According to a recent review, 14,15 prothrombin time has served as a basis for OAT monitoring since its first description by Quick more than 70 years ago. PT measures vitamin Kdependent coagulation factors II, VII, and X. 3,4 Warfarin (VKAs) inhibits the synthesis of coagulation factors II, VII, IX, and X in the liver, which remain partially inactive unless nine to 13 of the amino-terminal glutamate (Glu) residues are carboxylated to form Ca 2+ -binding γ-carboxyglutamate (Gla) residues. 3,4 Thus it is inevitable that the more medication is administered, the more inactive coagulation factors will result, hampering and inhibiting the PT measurement.
Today, two major PT methods are accepted for anticoagulant medication control and they are used globally for chronic anticoagulation checks: either the Quick PT, which is based on the technique described by Quick and his colleagues in 1935, 14,15 or the Owren PT 16 (combined thromboplastin reagent). The Owren PT is the predominant approach used in the Nordic countries, the Benelux, and Japan. The accuracy and comparability of PT results are essential to the safety of the individual in anticoagulant therapy and to improve the applicability of anticoagulation guidelines. The World Health Organization (WHO) recommendation for the use of INR given in the mid-1980s was aimed to harmonize PT results for OAT, regardless of the laboratory, reagent, instrument, or method used. 17,18 Unfortunately, an increasing body of evidence indicates that this goal has not been achieved. [19][20][21][22][23][24][25][26] Different reagents are known to vary with respect to the source of thromboplastins and other reagent components, 26 and these have not been characterized in detail. The Owren method is reagarded as the "reference method" in which fibrinogen and factor V are added to the reagent, and these do not depend on K-vitamin antagonists. The Quick PT meas-ures both of these coagulation factors, which is a drawback in OAT control. In the Owren method the sample volume in the reaction mixture is 5% only, in the Quick method 33%. On account of this difference in sample dilution the Owren PT is considered to be a more sensitive method than the Quick PT.
In the present study, we sought to compare INR results for warfarin therapy from the same OAT patient samples using both the Owren and the Quick PT methods. The aim was to study the method and reagent sensitivity of Owren and Quick PT using the new-generation PT method developed by Horsti et al., which can measure active coagulation without inhibition from inactive coagulation factors. 13,27,28

Patients and blood sampling
Venous blood samples were obtained from 207 hospital and health-center patients referred to the PT test for the monitoring of oral anticoagulant therapy. In our region a "P-INR" test code is used for this purpose. Hence, the patient samples represented all possible phases of anticoagulation: pretreatment, doseadjusting phase, and steady-state phase. All the procedures were approved by the responsible committee of our institution in accordance with the Helsinki Declaration of 1975. The blood (3.15 mL) was drawn into citrate coagulation tubes (Greiner Labortechnik GmbH, Vacuette cat. No. 454332, 9NC) containing 0.35 mL 0.109 mol/L (3.2%) citrate solution. The sample needle (Terumo, Venoject needle, Quick Fit, cat. No. MN-2138MQ) was 0.8x40 mm. The sample tubes were centrifuged at 1850 g for 10 min at 20°C to separate the plasma. All the measurements were commenced within eight hours from blood collection.

Prothrombin time determination
The calculation formula is for the International Normalized Ratio: INR = (sample sec /normal sec ) ISI , where ISI is the International Sensitivity Index. When ISI is close to 1.0, the reagent is sensitive and ISI is insignificant in the INR calculation. The PT coagulation times were measured using a fully automated ACL TOP CTS coagulation analyzer (Instrumentation Laboratory, IL, Lexington, MA, USA). For the Owren PT (combined thromboplastin reagent) the coagulation reaction contained 10 µL of citrated sample plasma, 60 µL of diluent, and 140 µL of reagent. The volumes for "dilution measurement" were 5 µL + 65 µL + 140 µL.

The new-generation prothrombin time method
We constructed PT (sec) vs. C (C = plasma or calibrator dilution factor) plots for normal and OAT plasmas. This is consistent with an uncompetitive inhibition principle with oral anticoagulants. From the line equation the yaxis intercept is calculated. This is the socalled minimal clotting time (t min ) with an infinite number of clotting factors. The difference in intercepts (y-axis) between normal plasma and OAT plasma indicates the action of uncompetitive inhibition in seconds without a calibration effect. We went on to calculate the dif-

Analytical imprecision and statistics
The within-run precision of the PT tests was measured using normal plasma (n = 10 determinations  (Figure 1). In contrast, marked differences were seen in the therapeutic range ( Diff  Reco  Diff  Diff  PT INR  PT INR %  INR  INR  %  INR  2G INR  %  INR   1,00  1,00 0,05  0,00  1,08  8,03  -1,01  0,74  -0,08  0,   as well as clinicians. WHO originally envisaged INR as a reliable and safe method for obtaining results from anticoagulation trials and as a basis for expert guidelines for treatment. However, the results of earlier studies and this present study demonstrate that the goal has not been achieved. The narrow therapeutic range and the increase in complications and mortality outside this range call for the development of more accurate laboratory analytics. A computer simulation study of serial INR measurements has been conducted within the most widely used therapeutic range (INR 2.0 -3.0); the authors concluded that analytical imprecision should be <5% and analytical bias < ± 0.2 INR units. 29 In this study, agreement between the Quick PT and Owren PT methods was particularly poor and failed to meet the qualification requirements. For this study we selected a newly developed recombinant PT reagent, but the result was not acceptable. We may wonder whether the difference in sample volume in the reaction mixture (5% and 33%) accounts for the measurement sensitivity. Both Owren PT reagents are sensitive to inactive coagulation factor measurement (inhibition), while the Quick PT reagents are not.

Nycotest Owren's Diff Diff PT-Fib Diff
In our earlier study we used DadeBehring Innovin with human placenta thromboplastin (Quick PT reagent), which is the most sensitive reagent to inactive coagulation factors and inhibition. The sensitivity thus does not depend only on differences in principle between the Quick PT or Owren PT methods. The thromboplastin and other components used in the reagent affect the reagent sensitivity. Horsti 24 25 The unacceptable situation for oral anticoagulation was confirmed further by this present investigation.
How can the harmonization of INR results be improved worldwide? For global use, WHO should recommend only the Owren PT, which is the superior method. The advantages of Owren PT are the fact that it measures only vitamin K-dependent coagulation factors F II, F VII, and F X (not fibrinogen and F V), and provides greater sensitivity and extensive dilution of the interfering matrix substances in the final reaction mixture. 19,25 The patient samples also contain inactive coagulation factors, the amount of which increases concomitant with increased anticoagulant medication and higher INR levels, interfering with and causing error in the coagulation measurement. The inhibition depends on the medication and the patient's individual metabolism, and thus each patient's sample must be corrected individually. 27,28 The Owren PT methods are sensitive, but more susceptible to the inactive coagulation factors and inhibition in measurement. In an earlier study we measured only the active coagulation factors, INRAct (F II, F VII, FX) without inhibition, which provided a new possibility to develop anticoagulant therapy and more appropriate care for OAT patients. The active coagulation factors are responsible for thrombosis or bleeding in vivo and medication should be based on this principle. 13,27,28 The inhibition interferes with the calibration procedure and measurement.
This study was conducted from the point of view of a clinical laboratory without clinical data and outcomes are missing between methods (Quick PT, Owren PTTOT, Owren PTACT), which would give the final answer as to the superiority of PT methods. This study revealed the lack of sensitivity in both Quick PT methods but good sensitivity and correlation in the Owren PT activity (INRAct) measurements. The editorials of Clinical Chemistry have posed the critical question and sought answers: "Has the Time Arrived to Replace the Quick Prothrombin Time Test for Monitoring Oral Anticoagulant Therapy?" 19 On the basis of recent studies and our opinion we can answer: "Yes, with the Owren PT without inhibition."