International normalized ratio in the prothrombin test: clinical significance and use

The safety and efficiency of therapy by peroral anticoagulants (PA) depend on a laboratory monitoring based on the prothrombin test (PT). The test is distinguished through its variability conditioned by different means of results' presentation as well as through the sensitivity of thromboplastin and a type of a device used in coagulation detection. WHO recommended, 1983, to standardize the thromboplastin preparations through adjusting their sensitivity (the so-called International Sensitivity Index--ISI) to blood coagulation defects induced by PA versus the primary international reference thromboplastin. Thromboplastin ISI as well as the mean normal prothrombin time (MNPT) of blood plasma are used to calculate the international normalized ratio (INR). The presentation of PT results as INR is justified exclusively for the PA-therapy stabilized patients. The INR system makes it possible to optimize the PA therapy only if the laboratory expert and clinician can clearly understand the PT standardization essence and observe the key WHO recommendations, i.e. definition of a coagulometer-specific ISI by manufacturing companies, estimation of MNPT by laboratories and use of the correct anticoagulant concentration.     

肝衰竭患者凝血酶原时间报告形式的探讨

目的 :探讨在肝衰竭病人的凝血酶原时间 (PT)以秒数、比率、活动百分率计 ,还是以国际正常化比率(INR)表示。方法 :采用 ISI 1.11,1.76和 2 .0 5的三种凝血活酶对 5 2例肝衰竭病人和 5 0例口服华法令的换瓣术后的病人进行 PT测定。以上述四种方式表示结果。结果 :肝衰竭病人 ,PT百分率能消除变异的可能性 (P<0 .0 5 ) ,而其他表示形式仍有明显的差异 (P>0 .0 1)。患者口服抗凝治疗后仅 INR能较准确地反映 PT的结果。结论 :INR不能用于表达非抗凝治疗患者的 PT结果。 PT活动的百分率是为肝衰竭病人的最好报告形式。     

Phospholipid composition controls thromboplastin sensitivity to individual clotting factors

BACKGROUND: Tissue factor is the active ingredient in thromboplastin reagents used to perform prothrombin time (PT) clotting tests to monitor oral anticoagulant therapy and to screen for clotting factor deficiencies. Thromboplastins are complex mixtures prepared from extracts of brain or placenta, although newer thromboplastins contain recombinant tissue factor incorporated into phospholipid vesicles. Thromboplastins can vary widely in their sensitivity to reductions in the levels of vitamin K-dependent clotting factors. A system to compensate for this, the International Sensitivity Index (ISI) and International Normalized Ratio (INR), has revolutionized the monitoring of oral anticoagulant therapy. The INR system is also sometimes used to monitor coagulopathies in patients with sepsis or liver failure, applications for which it was not originally designed and for which it has not been rigorously validated. OBJECTIVES: To better understand thromboplastin performance, we systematically investigated which properties of recombinant thromboplastins influence their sensitivities to changes in the levels of specific clotting factors. RESULTS: We now report that relative sensitivities to changes in the plasma levels of factors V, VII, X (FV, FVII, FX) and prothrombin are differentially influenced by a recombinant thromboplastin's content of phospholipid and sodium chloride. Furthermore, thromboplastins of similar ISI values may exhibit quite different sensitivities to each of these clotting factors. CONCLUSIONS: Differing sensitivities of thromboplastin reagents to individual clotting factor levels have implications for monitoring of oral anticoagulant therapy and interpreting results of the PT assay.     

A comparison of INRs after local calibration of thromboplastin international sensitivity indexes.

There are approximately 300 reagent/instrument combinations for performing prothrombin times/international normalized ratios (PT/INR) in the United States. Manufacturers and laboratories continually struggle to ensure that the International Sensitivity Index (ISI) of their thromboplastin is accurate for assaying PT/INR. OBJECTIVE: This study reports the feasibility of a new method to locally calibrate ISI of thromboplastin using the mechanical STA automated coagulation analyzer (Diagnostica-Stago Inc.) and two photo-optic coagulation analyzers, the BCS (Dade-Behring) and CA-540 (Sysmex). DESIGN: Neoplastine CI+ (CI+) (Diagnostica-Stago Inc); Thromboplastin C+ (TC+); Thromborel S (TRS); and Innovin (I) (Dade-Behring) were used in this study. A mean normal PT (MNPT) was determined for each reagent/instrument combination using samples from 25 normal individuals. Manufacturer instrument specific ISI values were not available for the STA with TC+, TRS and I. The CA540 had no ISI value for CI+ and the BCS system had no manufacturer assigned ISI values for TC+ and I; generic photo-optic and mechanical ISI manufacturer values were used for these two systems. Local on-site calibration was performed using frozen plasma calibrators to determine ISI values for each thromboplastin. Post-calibration, 95 patient samples were assayed for each reagent/instrument system combination using the manufacturer ISI and the local calibrated ISI to determine the INR result. PATIENTS: Patients from whom samples were obtained included five with a lupus anticoagulant, 30 on heparin therapy, and 60 on coumadin therapy. RESULTS: Differences between manufacturer versus local calibrated ISI ranged from 0.9% to 18.9% for normal sample INRs and from 0.8% to 16.4% for patient sample INRs. The number (or proportion) of patient specimens with clinically significantly different INR values (>10.0% difference) ranged from zero for several reagent combinations to more than half (or >50.0%) of those tested for several other combinations. CONCLUSION: Our results indicated that by locally calibrating ISI values, each laboratory may eliminate variability and guesswork between different reagent/instrument systems for ISI values when performing PT/INR assays and potentially improve the clinical accuracy of their patients' PT/INR results.     

Traces of factor VIIa modulate thromboplastin sensitivity to factors V, VII, X, and prothrombin

BACKGROUND: Thromboplastin reagents are used to conduct prothrombin time (PT) clotting tests to monitor oral anticoagulant therapy and screen for clotting factor deficiencies. Thromboplastins made from purified, recombinant tissue factor are generally more sensitive to changes in plasma factor (F) VII levels than are thromboplastins prepared from tissue extracts. This may be problematic as FVII's short plasma half-life can result in day-to-day fluctuation during oral anticoagulant therapy. We hypothesized that trace contamination of tissue-derived thromboplastins with FVII(a) blunts sensitivity to plasma FVII levels. METHODS: Traces of purified FVIIa were added to thromboplastin reagents prepared using recombinant human tissue factor and the effect on sensitivity to individual clotting factors was quantified in PT clotting assays. RESULTS AND CONCLUSIONS: Adding 5-100 pm FVIIa not only decreased thromboplastin sensitivity to plasma FVII, it surprisingly increased sensitivity to plasma levels of FV, FX and prothrombin. In addition, traces of FVIIa interacted with changes in the salt content and phospholipid composition of recombinant thromboplastins to further modulate their sensitivities to individual clotting factors. These results help explain how thromboplastin reagents of differing composition exhibit differing sensitivities to individual clotting factor levels. Implications of our results for monitoring oral anticoagulant therapy and other uses of the PT assay are discussed.     

Properties of recombinant human thromboplastin that determine the International Sensitivity Index (ISI)

Prothrombin Time (PT) clotting tests are widely used to monitor oral anticoagulation therapy and to screen for clotting factor deficiencies. The active ingredient in PT reagents (thromboplastins) is tissue factor, the integral membrane protein that triggers the clotting cascade through the extrinsic pathway. Several years ago, a system for calibrating and using thromboplastin reagents, known as the International Sensitivity Index (ISI) and the International Normalized Ratio (INR), was developed to standardize monitoring of oral anticoagulant therapy. The ISI/INR method, while revolutionizing the monitoring of coumarin therapy, has been criticized for a number of perceived shortcomings. We have undertaken a series of studies aimed at achieving a detailed understanding of which parameters influence the ISI values of thromboplastin reagents, with an ultimate goal of creating 'designer thromboplastins' whose sensitivities to the various clotting factors can be individually tailored. In this study, we demonstrate that ISI values of thromboplastin reagents based on relipidated, recombinant human tissue factor can be controlled by a combination of changes in the phospholipid content (in particular, the levels of phosphatidylserine and phosphatidylethanolamine) and ionic strength. The sensitivity of a given thromboplastin reagent can be increased (i.e. its ISI value decreased) by decreasing the content of phosphatidylserine and/or increasing the ionic strength. The molar ratio of phospholipid to tissue factor, on the other hand, had essentially no impact on ISI value.     

A modified method of prothrombin time/International Normalised Ratio determination in capillary blood and monitoring oral anticoagulant therapy.

BACKGROUND: Oral anticoagulant therapy is monitored by a prothrombin time (PT) assay. The PT is standardised by the International Normalised Ratio (INR). The purpose of this study was to work out a modified method of PT/INR measurement in capillary blood for monitoring anticoagulation treatment. METHODS: Healthy donors, subjects with high or low haematocrit values, and oral anticoagulant-treated patients were included in the study. Plasma and capillary blood PT/INRs were determined by the standard Quick clotting assay, by the modified approach and with the CoaguChek S analyser. RESULTS: The performance characteristics of the developed method were accuracy, due to taking into account whole capillary blood haematocrit values, and precision, due to a decrease in the viscosity of the analysed samples. Implementation of the modified method showed that it is possible to use PT values of normal plasma for capillary blood INR calculation. The developed method allowed the determination of PT in capillary blood within the haematocrit value range from 0.15 up to 0.7. For capillary blood, the results of the modified method closely correlated with PT/INR values determined by the reference Quick method in venous plasma (r=0.99) and with the CoaguChek S analyser (r=0.97). CONCLUSIONS: The modified method of capillary blood PT/INR determination could be recommended for oral anticoagulant therapy monitoring.     

Preanalytical variables and off-site blood collection: influences on the results of the prothrombin time/international normalized ratio test and implications for monitoring of oral anticoagulant therapy

BACKGROUND: The quality of oral anticoagulant therapy management with coumarin derivatives requires reliable results for the prothrombin time/International Normalized Ratio (PT/INR). We assessed the effect on PT/INR of preanalytical variables, including ones related to off-site blood collection and transportation to a laboratory. METHODS: Four laboratories with different combinations of blood collection systems, thromboplastin reagents, and coagulation meters participated. The simulated preanalytical variables included time between blood collection and PT/INR determinations on samples stored at room temperature, at 4-6 degrees C, and at 37 degrees C; mechanical agitation at room temperature, at 4-6 degrees C, and at 37 degrees C; time between centrifugation and PT/INR determination; and times and temperatures of centrifugation. For variables that affected results, the effect of the variable was classified as moderate when <25% of samples showed a change >10% or as large if >25% of samples showed such a change. RESULTS: During the first 6 h after blood collection, INR changed by >10% in <25% of samples (moderate effect) when blood samples were stored at room temperature, 4-6 degrees C, or 37 degrees C with or without mechanical agitation and independent of the time of centrifugation after blood collection. With one combination of materials and preanalytical conditions, a 24-h delay at room temperature or 4-6 degrees C had a large effect, i.e., changes >10% in >25% of samples. In all laboratories, a 24-h delay at 37 degrees C or with mechanical agitation had a large effect. We observed no clinically or statistically relevant INR differences among studied centrifugation conditions (centrifugation temperature, 20 degrees C or no temperature control; centrifugation time, 5 or 10 min). CONCLUSIONS: We recommend a maximum of 6 h between blood collection and PT/INR determination. The impact of a 24-h delay should be investigated for each combination of materials and conditions.     

不同血凝分析仪检测结果的一致性研究

目的 探讨不同实验室以及同一实验室的不同血凝分析仪检测结果的一致性.方法 将不同实验室的14台血凝分析仪分为3组,分别为STA系列(A组)、ACL系列(B组)、CA系列(C组),同时检测同一批号不同水平质控品(水平1、2、3)的凝血酶原时间(PT)、国际标准化比值(INR)、活化部分凝血活酶时间(APTT)、纤维蛋白原含量(FIB)及凝血酶时间(TT);并以同一实验室检测原理基本一致的2台血凝分析仪同时检测139份受检血浆的PT、INR、APTT、PT演算法测定FIB(PT-FIB)、Clauss法测定FIB(FIB-C).结果 3组血凝分析仪检测INR水平3的结果分别为5.35±0.20、4.35±1.00、4.46±0.30,差异无统计学意义(P＞0.05);检测TT水平3的结果分别为(17.1±0.3)s、(15.5±1.1)s、(14.8±1.8)s,差异无统计学意义(P＞0.05);其他各检测指标结果间差异有统计学意义(P＜0.05);两两比较的结果表明,B组和C组检测结果的符合率高达66.7%(10/15).同一实验室的ACL Futura和CA 510血凝分析仪检测PT的结果分别为(17.7±6.7)s、(20.1±10.9)s,检测INR的结果分别为1.75±1.07、1.64±0.91,检测PT-FIB的结果分别为(3.51±1.50)g/L、(3.68±1.93)g/L,检测FIB-C的结果分别为(2.61±1.31)g/L、(2.58±1.45)g/L,上述指标间差异均无统计学意义(P＞0.05);而检测APTT的结果分别为(49.9±21.5)s、(39.1±16.7)s,差异具有统计学意义(P＜0.05);同时,二者检测PT、INR、APTT、PT-FIB和FIB-C结果的相关性良好,r值分别为0.984 3、0.988 8,0.987 0,0.975 6,0.994 0;偏倚分析结果显示,2台血凝分析仪检测PT、INR、PT-FIB和FIB-C结果的一致性较好.结论 检测原理基本相同的不同血凝分析仪,其检测结果具有较好的一致性.不同血凝分析仪应通过定期比对和试剂的标准化,以改善和保证其检测结果的一致性.     

Comparison of quick and owren prothrombin time with regard to the harmonisation of the International Normalised Ratio (INR) system.

Prothrombin time (PT) is tested mostly to monitor patients on oral anticoagulant treatment. The International Normalised Ratio (INR) was introduced to improve and harmonise PT results and therapeutic range globally for patient care and the scientific literature. We studied the Quick PT in 179 patients and the Owren PT in 137 patients on oral anticoagulant therapy using two different reagents for the two methods of measuring PT. We assessed the clinical significance of the INR results obtained by each method using the two reagents and compared the Quick and Owren methods. We conclude that with the Quick method individual INR results differed from each other too much clinically, while using the Owren method individual INR results were clinically acceptable. Our opinion is that we should develop the INR system using the Owren PT method rather than the Quick to improve patient care.     

乙型病毒性肝炎患者中末期凝血酶原时间标准化报告方式

目的 探讨乙型病毒性肝炎患者凝血酶原时间(PT)的标准化报告方式。方法 选择乙型病毒性肝炎患者61例,其中肝炎后肝硬化41例,慢性重型肝炎20例。20例口服华法令抗凝药患者作为对照组。采用来源不同、ISI值不同的6种凝血活酶试剂进行PT测定,以秒数、比率、活动度百分率以及国际正常化比率4种方式表示PT结果。结果 病毒性肝炎患者PT结果,当以活动度百分率和比率形式表示时,不同凝血活酶试剂之间差异没有显著性意义(F=1.289,P=0.268;F=I.992,,J=3.079),当以秒数和INR报告方式表示时,差异有显著意义(F=8.491,P=0.0001;F=2.497.P=0.031)。通过Neoplastin与其他5种试剂的PT结果作线性回归分析,当结果以活动度百分率表示时,Neoplastin与其他5种试剂之间存在高度一致性;而以秒数,比率和INR表示时,试剂之间不存在一致性。提示PT活动度百分率能使乙型病毒性肝炎中末期患者PT报告方式标准化。口服抗凝剂治疗的患者仅INR能使PT的报告方式标准化。结论 PT活动度百分率能使乙型病毒性肝炎中末期患者PT报告方式标准化,INR仅适用于抗凝治疗患者PT结果的报告。     

The effect of instruments for prothrombin time testing on the international normalized ratio

Prothrombin time (PT) testing is used for monitoring oral anticoagulant therapy, its result being usually expressed as international normalized ratio (INR). This is done using the international sensitivity index (ISI) specific of thromboplastin employed to carry out the test. In this way a good PT standardization may be achieved although the instruments used to calibrate the thromboplastins might influence the ISI value. Presented at the ‘2nd International Symposium on Standardization and Quality Control of Coagulation Tests: Implications for the Clinical Laboratory’, Rome, September 28–29, 1989.     

A review of variables affecting PTs/INRs.

The prothrombin time is a common method of monitoring patients undergoing oral anticoagulant therapy. The proliferation of commercial thromboplastin brands with different international sensitivity indices (ISI) in conjunction with wider availability of automated coagulation analyzers has elevated the need for standardization in monitoring therapy.     

Poor agreement among prothrombin time international normalized ratio methods: comparison of seven commercial reagents

BACKGROUND: Prothrombin time (PT) has long been the most popular test for monitoring oral anticoagulation therapy. The International Normalized Ratio (INR) was introduced to overcome the problem of marked variation in PT results among laboratories and the various recommendations for patient care. According to this principle, all reagents should be calibrated to give identical results and the same patient care globally. This is necessary for monitoring of single patients and for application of the results of anticoagulation trials and guidelines to clinical practice. METHODS: We took blood samples from 150 patients for whom oral anticoagulation had been prescribed. Plasmas were separated and PTs determined by use of seven commercial reagents and four calibrator sets. The differences in results were assessed by plotting, for each possible pair of methods, the differences in INR values for each sample against the mean INR value (Bland-Altman difference plots). RESULTS: Mean results differed significantly (P <0.001) for 17 of 21 possible paired comparisons of methods. Only two pairs of methods produced very similar results when assessed for problems of substantial differences in INR values; a significant, systematic increase in the difference with INR; and a significant systematic increase in the variation in difference with increasing INR values. CONCLUSIONS: The agreement among several (and perhaps most) commercial INR methods is poor. The failure of current calibration strategies may severely compromise both the monitoring of individual patients and the application of oral anticoagulation guidelines and trial results to clinical practice.     

Treatment of coumarin-associated coagulopathy: a systematic review and proposed treatment algorithms

BACKGROUND: An excessive anticoagulant effect because of coumarins is frequently encountered. Objective: To review available literature on the management of warfarin-associated coagulopathy and to propose evidence-based treatment algorithms. METHODS: Data sources were Medline and Embase. Papers published between 1966 and December 2005 describing randomized trials or prospective cohort studies evaluating treatments for coumarin-associated coagulopathy were abstracted. RESULTS: Low dose oral vitamin K rapidly and reliably returns the international normalized ratio (INR) to the usual therapeutic range in non-bleeding patients. Simple withholding of acenocumarol results in rapid correction of its anticoagulant effect. The impact of oral vitamin K on phenprocumon-associated coagulopathy cannot be determined from available literature. Intravenous vitamin K and coagulation factors should be given to patients with major or life-threatening hemorrhage. The optimal dose and type of coagulation factor is not known. CONCLUSIONS: Vitamin K therapy is an effective treatment for INR prolongation in patients with coumarin-associated coagulopathy; coagulation factor replacement is required, in addition, in patients with major bleeding or with indication for immediate correction of their INR. Clinical trials powered to detect differences in rates of bleeding and thrombosis are now required to determine if vitamin K reduces the risk of bleeding without causing thrombosis in non-bleeding patients with prolonged INR.     

Variability of prothrombin time and activated partial thromboplastin time in the diagnosis of increased surgical bleeding

BACKGROUND: Prothrombin time (PT) and activated partial thromboplastin time (APTT) are used to diagnose causes of increased surgical bleeding and to guide treatment of acquired coagulation factor deficiency. This study compared the sensitivity of various commercial PT and APTT tests in patients with dilutional coagulopathy. STUDY DESIGN AND METHODS: A prospective study was used to identify patients who experienced increased surgical bleeding during elective extensive (>10 spinal segments) spinal fusion and instrumentation. In patients with clinical signs of increased bleeding, blood was obtained to compare the sensitivity of various commercial PT and APTT tests. PT, PT ratio, the International Normalized Ratio (INR), APTT, and APTT ratio were compared for their sensitivity in the diagnosis of a dilutional coagulopathy. RESULTS: Sixteen patients experienced increased bleeding during surgery. Mean estimated blood volume lost exceeded 1 blood volume (1.14 +/− 0.28). PT and APTT test results varied markedly. In the most sensitive PT and APTT tests, the results were 1.5 times the mean reference range values in all but on of the patients. The least sensitive combination of tests had results that were 1.5 times the mean reference range values in only 2 of 16 patients. Variability among tests was not reduced by the use of the PT or the APTT ratio, by the use of INR, or by incorporation of a measure of PT or APTT test sensitivity to factor-deficient serum. CONCLUSION: In surgical patients with dilutional coagulopathy, diagnostic and treatment decisions could depend on which PT and APTT test was used to determine the etiology of increased bleeding. This study indicates that the relationship between increased bleeding and an increased PT and APTT may be more difficult to define than is suggested by current practice guidelines. Each laboratory must establish guidelines based on reagent and instrument sensitivity to coagulation factor dilution.     

Guidelines on preparation, certification, and use of certified plasmas for ISI calibration and INR determination.

Reliable international normalized ratio (INR) determination depends on accurate values for international sensitivity index (ISI) and mean normal prothrombin time (MNPT). Local ISI calibration can be performed to obtain reliable INR. Alternatively, the laboratory may determine INR directly from a line relating local log(prothrombin time [PT]) to log(INR). This can be done by means of lyophilized or frozen plasmas to which certified values of PT or INR have been assigned. Currently there is one procedure for local calibration with certified plasmas which is a modification of the WHO method of ISI determination. In the other procedure, named 'direct' INR determination, certified plasmas are used to calculate a line relating log(PT) to log(INR). The number of certified plasmas for each procedure depends on the method of preparation and type of plasma. Lyophilization of plasma may induce variable effects on the INR, the magnitude of which depends on the type of thromboplastin used. Consequently, the manufacturer or supplier of certified plasmas must assign the values for different (reference) thromboplastins and validate the procedure for reliable ISI calibration or 'direct' INR determination. Certification of plasmas should be performed by at least three laboratories. Multiple values should be assigned if the differences between thromboplastin systems are greater than 10%. Testing of certified plasmas for ISI calibration may be performed in quadruplicate in the same working session. It is recommended to repeat the measurements on three sessions or days to control day-to-day variation. Testing of certified plasmas for 'direct' INR determination should be performed in at least three sessions or days. Correlation lines for ISI calibration and for 'direct' INR determination should be calculated by means of orthogonal regression. Quality assessment of the INR with certified plasmas should be performed regularly and should be repeated whenever there is a change in reagent batch or in instrument. Discrepant results obtained by users of certified plasmas should be reported to manufacturers or suppliers.     

机械瓣膜替换术后抗凝血浆蛋白C含量改变的意义

目的　探讨机械瓣膜替换术后口服华法令者，血浆蛋白C，PC)的改变以及与其他凝血因子改变的关系。方法　采用酶联免疫吸附法测定PC含量，采用经典一期法测定凝血因子X∶C、凝血酶原时间并计算国际正常化比值(INR)。　结果　正常对照组(n=45)，口服华法令组(n=78)PC含量分别为(5.00±0.64)mg/L、(2.54±2.18)mg/L，两者差异有极显著性(P＜0.001)。服药时间1～6个月、7～24个月、25～132个月者PC含量分别为(4.02±3.60)mg/L、(2.52±1.56)mg/L、(2.40±1.66)mg/L。PC含量改变与因子X、凝血酶原时间(PT)、INR间无明显相关性。　结论　机械瓣膜替换术后口服华法令者血浆蛋白C含量减低，并与服药时间长短有关，蛋白C的改变存在较大的个体差异。在抗凝治疗初期，对蛋白C含量的监测有着特别重要的意义。     

广东省凝血酶原时间测定的室间质量评价

目的 探讨广东省临床实验室监测凝血酶原时间（PT）的检测状况及影响PT测定室间可比性的因素.方法 通过每年两次定期向全省参评实验室寄发质控样品（每次5个样品）,然后对其回报的数据进行统计分析,作出实验室检验水平的评价.结果 近年来全省实验室间PT（INR）测定结果的CV值尽管有逐渐下降的趋势,但室间变异仍然相当高.同一凝血活酶试剂的测定结果,PT（INR）的CV明显大于PT（sec）的CV,特别是对异常水平质评物的测定.结论 凝血活酶试剂敏感度指数（ISI）值标定的不准确性和INR计算不正确的是引起广东省临床实验室间凝血酶原时间测定变异大的主要原因.     

Inaccuracy of international normalized ratio (INR) measurements and suggestions for improved WHO guidelines for calibration of reference preparations – a statistical study

Summary.  Numerous empirical studies show that the accuracy of international normalized ratio (INR) measurements is unsatisfactory and worse than generally expected. We demonstrate that a plausible reason for this large inaccuracy is a conventional calibration procedure of reference preparations with (i) an erroneous assumption that the line relating logarithmic prothrombin times (log PTs) of patients passes through the mean log PT of the 'normal' population (mean normal PT); (ii) non-perceived interactions between patients and PT systems; and (iii) systematic exclusions of 'outliers'. The same conventional procedure also results in serious overestimation of the accuracy of INR measurements, thus leading to a false sense of security in oral anticoagulant therapy. In an example with data from WHO guidelines, we show that the systematic overprediction of INR (which is believed to be 0) may be as large as 5%, when prediction is performed under the conventional WHO model. Under the same model the CV of the predicted vs. the true INR is believed to be only about 1% when it in reality is more than 4%. We suggest that the conventional calibration procedure is modified in order to reduce the twofold negative impact of lower true accuracy and overestimated reported accuracy on oral anticoagulant therapy and to allow for an unambiguous definition of true INR values.