Search

Hemophilia is a rare disorder that is complex to diagnose and to manage. These evidence-based guidelines offer practical recommendations on the diagnosis and general management of hemophilia, as well as the management of complications including musculoskeletal issues, inhibitors, and transfusion-transmitted infections. By compiling these guidelines, the World Federation of Hemophilia aims to assist healthcare providers seeking to initiate and/or maintain hemophilia care programs, encourage practice harmonization around the world and, where recommendations lack adequate evidence, stimulate appropriate studies. © 2012 Blackwell Publishing Ltd.

Background: BAY 81-8973 is a full-length unmodified recombinant factor VIII (rFVIII) in development for the treatment of hemophilia A. BAY 81-8973 has the same amino acid sequence as Bayer's sucrose-formulated rFVIII but is manufactured using the latest technologies. Potency labeling is based on the chromogenic assay. BAY 81-8973 has demonstrated an excellent safety and efficacy profile in the clinical development program. This field study was conducted to evaluate assay variability in BAY 81-8973 measurements compared with a marketed rFVIII (Advate®, Baxter, Westlake Village, CA) using the methods, reference standards, and reagents typically used in clinical laboratories. Methods: Clinical laboratories in North America, Europe, Israel, and South Africa were invited to participate in the study. Each laboratory was provided with 21 blinded samples to analyze using their routine assay (one-stage, chromogenic, or both), reagents, and standards. The 21 samples consisted of 3 aliquots each of spiked hemophilia plasma containing normal von Willebrand factor levels with BAY 81-8973 at 3 different levels: 50 IU/dL (high); 3 aliquots each of spiked hemophilia plasma with Advate at the same levels (low, medium, high); and 3 aliquots of commercially available positive control sample (normal human plasma). Samples were identified by unique numbers and by target FVIII levels (low, medium, high). The nominal spiked target levels of FVIII concentration were 0.043 IU/mL (low), 0.375 IU/mL (medium), and 0.865 IU/mL (high) for BAY 81-8973 and Advate and 0.960 IU/mL for the plasma control. Results were analyzed statistically for intra- and interlaboratory variability. Results: Of 82 laboratories contacted, 41 in 11 countries participated. Thirty-one laboratories used the one-stage assay only, 1 used the chromogenic assay only, and 9 used both assays. Intralaboratory variability was Conclusions: The laboratories participating in this field study used a wide range of methods and reagents for FVIII measurements. The variability of the results was similar for both BAY 81-8973 and Advate and highest for low concentrations. There was no relevant difference in the results between the one-stage and chromogenic assays. Disclosures Maas Enriquez: Bayer Pharma AG: Employment. Beckmann:Bayer Pharma AG: Employment. Katterle:Bayer Pharma AG: Employment. Bruns:Winicker Norimed: Employment. Tseneklidou-Stoeter:Bayer Pharma AG: Employment. Kitchen:Bayer Pharma AG: Other: Advisory fees, Speakers Bureau.

SummaryWhen the International Normalised Ratio (INR) is used for control of oral anticoagulant therapy the same result should be obtained irrespective of the laboratory reagent used. However, in the UK National External Quality Assessment Scheme (NEQAS) for Blood Coagulation INRs determined using different reagents have been significantly different.For 18 NEQAS samples Manchester Reagent (MR) was associated with significantly lower INRs than those obtained using Diagen Activated (DA, p = 0.0004) or Instrumentation Laboratory PT-Fib HS (IL, p = 0.0001). Mean INRs for this group were 3.15, 3.61, and 3.65 for MR, DA, and IL respectively. For 61 fresh samples from warfarin-ised patients with INRs of greater than 3.0 the relationship between thromboplastins in respect of INR was similar to that observed for NEQAS data. Thus INRs obtained with MR were significantly lower than with DA or IL (p We conclude that the differences between INRs measured with the thromboplastins studied here are sufficiently great to influence patient management through warfarin dosage schedules, particularly in the upper therapeutic range of INR. There is clearly a need to address the issues responsible for the observed discrepancies.

Haemophilia A is caused by a reduction in clotting factor VIII (FVIII). FVIII coagulant activity (FVIII:C) can be measured by three methods; the one-stage activated partial thromboplastin time-based clotting assay, the two-stage Xa generation-based clotting assay and the chromogenic Xa generation-based assay. The FVIII:C of most patients with haemophilia A are concordant regardless of the assay method employed. Up to a third of patients show assay discrepancy, usually with the two-stage and chromogenic assays being much lower than the one-stage assay. Very rarely, patients have been reported with lower one-stage compared to two-stage and chromogenic assays, but here we report that the mutation p.Tyr365Cys accounts for most of these patients and, at least in the UK, is not rare. We have identified this mutation in 23 different families. Affected male index individuals had a lower mean one-stage FVIII:C of 27 iu/dl compared to two-stage FVIII:C mean of 77 iu/dl. Affected individuals had minimal or absent bleeding symptoms and when these were present they were usually in patients with another co-inherited bleeding disorder. Affected individuals often had surgery without the need to correct the one-stage FVIII:C.

The determination of functional Antithrombin is a central part of thrombophilia screening. In this multicenter study, a new FXa-based method (INNOVANCE® Antithrombin) was evaluated on four different analyzers.The INNOVANCE Antithrombin method was evaluated by precision and reference interval studies and by comparing the new method with established methods through parallel measurement of samples from 249 patients and 151 apparently healthy individuals.The INNOVANCE Antithrombin assay demonstrated on all analyzers repeatability coefficients of variation (CVs)<

Point-of-care testing (POCT) in haematology has seen a significant increase in both the spectrum of tests available and the number of tests performed annually. POCT is frequently undertaken with the belief that this will reduce the turnaround time for results and so improve patient care. The most obvious example of POCT in haemostasis is the out-of-hospital monitoring of the International Normalized Ratio in patients receiving a vitamin K antagonist, such as warfarin. Other areas include the use of the Activated Clotting Time to monitor anticoagulation for patients on cardio-pulmonary bypass, platelet function testing to identify patients with apparent aspirin or clopidogrel resistance and thrombelastography to guide blood product replacement during cardiac and hepatic surgery. In contrast to laboratory testing, POCT is frequently undertaken by untrained or semi-trained individuals and in many cases is not subject to the same strict quality control programmes that exist in the central laboratory. Although external quality assessment programmes do exist for some POCT assays these are still relatively few. The use of POCT in haematology, particularly in the field of haemostasis, is likely to expand and it is important that systems are in place to ensure that the generated results are accurate and precise.

SummaryA number of different methods are available for the measurement of factor VIIa. Almost all of these employ ratios of two different measurements of factor VII. In order to determine which is the most sensitive to activated factor VII we have compared currently available methods in the following groups: two patients with haemophilia A following treatment with activated recombinant factor VII (rVII a); 6 normal plasmas during cold promoted activation of factor VII; normal individuals (n = 23); and patients with unequivocal disseminated intravascular coagulation (DIC, n = 19). Factor VII was measured in an amidolytic assay (VII: Amid) and an antigen assay (VII:Ag). Clotting activity was measured using rabbit (VII:C Rab), human (VII:C Hum) and bovine (VII:C Bov) thromboplastin.Of the clotting assays the most sensitive to the presence of factor VIIa was that which utilised bovine thromboplastin. Amidolytic and immunological measurements were unaffected by the activity state of factor VII. The ratios VII:C Rab/VII: Ag and VII:C Rab/VII:Amid were insensitive to activated factor VII. The ratios most sensitive to the presence of factor VII a were VII:C Bov/VII: Amid and VII: C Bov/VII:Ag. The ratios VII:C Bov/VII:C Rab and VII:C Bov/VII:C Hum are less sensitive but have the advantage for epidemiological studies of narrower reference ranges.

Summary. Background: Warfarin reversal is a common clinical situation. This is commonly performed using vitamin K and, depending on the urgency, fresh frozen plasma (FFP), prothrombin complex concentrates (PCCs), or activated factor VII. Even though PCCs are widely used, the ideal dosing regimen is far from established. Objectives: To verify differences in warfarin reversal patterns using FFP, recombinant FVIIa (rFVIIa), and PCC; and to test the hypothesis that supratherapeutic International Normalized Ratios (INRs) might not correlate with thrombin generation (TG) and identify the ideal concentrations of PCC required to reverse various INR thresholds. Methods: We studied the effects of FFP, rFVIIa and Beriplex P/N on the INR and TG, using the calibrated automated thrombography assay in ex vivo warfarinized plasma. Plasmas with different INRs were spiked with different concentrations of Beriplex P/N. Results: Beriplex P/N was the only agent that completely normalized TG and the INR. The endogenous thrombin potential (ETP) and the peak thrombin showed a significant negative correlation with all INRs. The ETP and velocity of TG reached a plateau at an INR of ∼ 4.0. A concentration equivalent to a dose of 30 IU kg−1 Beriplex P/N normalized the ETP, the INR, FII, FVII, FIX and FX of samples with INRs ≥ 4.0. Higher doses resulted in hypercoagulable TG patterns. A concentration equivalent to a dose of 20 IU kg−1 was sufficient to reverse warfarin at an INR range of 2.0–3.9, as judged by the same tests. Conclusions: Warfarin reversal algorithms could be simplified with the adoption of this strategy utilizing two doses of PCC, depending on the INR of the patient. This would also lead to cost reductions and, possibly, a reduction in thrombotic risk.

Background: In vitro hemolysis, the prevailing cause of preanalytical error in routine laboratory diagnostics, might influence the reliability of several tests, affect the quality of the total testing process and jeopardize patient safety. Although laboratory instrumentation is now routinely equipped with systems capable of automatically testing and eventually correcting for hemolysis interference, to our knowledge there are no reports that have compared the efficiency of different analytical platforms for identifying and classifying specimens with hemolysis. Methods: Serum from a healthy volunteer was spiked with varying amounts of hemolyzed blood from the same volunteer, providing a serum free hemoglobin concentration ranging from 0.0 g/L to 2.0 g/L as measured by the reference cyanmethemoglobin assay. The spiked serum samples were shipped to seven separate laboratories and the hemolysis index (HI) was tested in triplicate on the following analytical platforms: Roche Modular System P (n=4) and Integra 400 Plus (n=1), Siemens Dimension RxL (n=3), ADVIA 2400 (n=1) and ADVIA 1800 (n=1), Olympus AU 680 (n=1) and Coulter DXC 800 (n=1). Results: Satisfactory agreement of HI results was observed among the various analytical platforms, despite a trend toward overestimation by the ADVIA 2400 and 1800. After normalizing results according to the instrument-specific alert value, discrepancies were considerably reduced. All instruments except for the Dimension RxL gave values normalized to the instrument-specific alert value, 1.0 for the sample with 0.075 g/L free hemoglobin. The results of the four Modular System P tests were also highly reproducible among the different facilities. When evaluating instruments that provided quantitative HI results, the mean intra-assay coefficient of variation (CV) calculated for the triplicate determinations was always between 0.1% and 2.7%. Conclusions: The results of this multicenter evaluation confirm that efficiency of different analytical platforms to correctly identify and classify unsuitable samples is satisfactory. However, more effort should be placed on the standardization of reporting HI. All the instruments that we tested provide either quantitative or qualitative results that are essentially comparable, but which should always be compared with the instrument-specific alert values to harmonize their efficiency. Clin Chem Lab Med 2009;47:934–9.

Laboratory diagnostics, a pivotal part of clinical decision making, is no safer than other areas of healthcare, with most errors occurring in the manually intensive preanalytical process. Patient misidentification errors are potentially associated with the worst clinical outcome due to the potential for misdiagnosis and inappropriate therapy. While it is misleadingly assumed that identification errors occur at a low frequency in clinical laboratories, misidentification of general laboratory specimens is around 1% and can produce serious harm to patients, when not promptly detected. This article focuses on this challenging issue, providing an overview on the prevalence and leading causes of identification errors, analyzing the potential adverse consequences, and providing tentative guidelines for detection and prevention based on direct-positive identification, the use of information technology for data entry, automated systems for patient identification and specimen labeling, two or more identifiers during sample collection and delta check technology to identify significant variance of results from historical values. Once misidentification is detected, rejection and recollection is the most suitable approach to manage the specimen.Clin Chem Lab Med 2009;47:143–53.

To examine the relationship between diabetic vascular disease and haemostasis, a set of sensitive assays has been used to assess in vivo activation of coagulation in 62 diabetic patients (41 Type 1 and 21 Type 2), aged 19-68 years, who had been screened for the presence of complications. Fibrinopeptide A, an index of thrombin activity, was significantly increased in diabetic patients compared with control subjects (p less than 0.05), in both plasma (with complications mean 8.04 +/- 11.87 (+/- SD); without complications 7.21 +/- 10.13; control subjects 2.11 +/- 1.40 micrograms l-1) and urine (with complications mean 1.48 +/- 0.74; without complications 1.35 +/- 0.62; control subjects 0.98 +/- 0.39 micrograms l-1). Activated factor VII (VIIa ratio 1.21 +/- 0.39; 1.13 +/- 0.23; 1.01 +/- 0.11) and fibrinogen (3.15 +/- 0.59; 3.11 +/- 0.69; 2.70 +/- 0.57 g l-1) were also elevated in diabetic patients with and without complications (VIIa p less than 0.05, fibrinogen p less than 0.01). The only difference between Type 1 and Type 2 patients was in fibrin degradation products (Type 1 0.28 +/- 0.18; Type 2 0.40 +/- 0.18 mg l-1, p less than 0.01). Plasma levels of fibrin degradation products were elevated in diabetic patients (p less than 0.05 vs control subjects), and correlated with age (r = 0.44, p less than 0.01) but were unrelated to the presence of complications. There were no significant differences in any coagulation variables between diabetic patients with and without complications.(ABSTRACT TRUNCATED AT 250 WORDS)

Summary Congenital protein C deficiency significantly increases the risk of venous thromboembolism, a serious and potentially lethal condition. Protein C levels can be determined by chromogenic, clotting and antigenic assays, each type of assay has differences in specificity and sensitivity to protein C deficiency. In principle, clotting-based assays of protein C are preferred over chromogenic assays, as they can detect some rare mutations that are missed by the chromogenic assay, however, clotting-based assays may be prone to inaccuracy because of poor specificity. We have evaluated a new venom-based clotting assay of protein C, and optimized it for use on Sysmex CA-1500 analyser. The assay was linear from 0 to 130 U/dl, a normal plasma demonstrated good inter-assay precision, with a coefficient of variation of 4.8%. The assay compared well with antigenic- and venom-based chromogenic protein C assay in normal individuals, subjects with lupus anticoagulant, and subjects with FV Leiden. Median protein C levels by clotting, chromogenic and antigen for the three subject groups were 108 U/dl, 108 IU/dl and 109 IU/dl for normal subjects, 94 U/dl, 106 IU/dl and 103 IU/dl for subjects with lupus anticoagulant, and 102 U/dl, 104 IU/dl and 100 IU/dl for subjects heterozygous for FV Leiden. Comparing levels of clotting protein C with protein C antigen by ratio (clotting/antigen), the three groups showed small differences that did not quite reach statistical significance, (mean ratios ranged from 0.95 to 1.01, anovaP = 0.0561), the lowest ratio was with the lupus anticoagulant group. Comparing clotting assay with chromogenic assay by ratio (clotting/chromogenic), the three groups did show a statistically significant difference (P = 0.0033) which was due to a difference in mean ratios between normal and lupus anticoagulant groups (ratios 1.00 and 0.91, respectively, P

Even though new anticoagulants are being devised with the notion that they do not require regular monitoring, when bleeding occurs, it is important to have an antidote and a reliable test to confirm whether the anticoagulant effects are persisting. We examined the effects of five heparinoids, unfractionated heparin (UFH), tinzaparin, enoxaparin, danaparoid and fondaparinux on the traditional APTT and anti-Xa assays as well as on the calibrated automated thrombogram (CAT). We also studied the ability of protamine sulphate (PS), NovoSeven, FEIBA and FFP to reverse maximum anticoagulation induced by the different heparinoids. The CAT was the only test to detect the coagulopathy of all the anticoagulants. PS produced complete reversal of UFH, and this could be monitored with all three tests. Tinzaparin can also be completely neutralised in vitro with high doses of PS, but the maximum enoxaparin reversal achieved with PS is only approximately 60%. Fondaparinux does not significantly affect the APTT and PS has no significant effect on its reversal. Only NovoSeven was able to correct the fondaparinux induced CAT abnormalities whilst having no effect on the anti-Xa level. None of the reversal agents was very effective in danaparoid spiked plasma but NovoSeven, at high dose, increased the ETP by 40% and reduced the anti-Xa level from 0.93 to 0.78 IU/ml. We conclude that the CAT is superior to the traditional coagulation tests in that it not only detects the coagulopathy of all the heparinoids but can be also be used to monitor their reversal.

Aims: Platelet function testing forms an important part of the laboratory investigation of a bleeding tendency; however, little standardisation and quality control is available for these tests. A UK National External Quality Assessment Scheme (UK NEQAS) for Blood Coagulation exercise sought to identify current practice among laboratories performing platelet function tests. Methods: A questionnaire was circulated in March 2006 to establish the current status of platelet function testing practice among participants of UK NEQAS. Participants were asked specifically about practice in bleeding time testing, PFA-100 analyser use, platelet aggregometry methodology and additional tests of platelet function. Results: 169 returned questionnaires revealed that 26 centres used bleeding time, the PFA-100 analyser and platelet aggregometry in their investigations; 13 used bleeding time and the PFA-100 only; 33 used bleeding time and platelet aggregometry; and 23 used the PFA-100 with platelet aggregometry. 58 centres reported that they performed only bleeding times in their investigations, 10 reported use of the PFA-100 only, and 6 reported use of aggregometry only. Marked variability was observed in methodology for each of these tests, and in many cases no form of quality control was employed. Conclusions: The data confirmed the lack of standardisation in methodology employed in different centres. Updated guidelines and standardisation of platelet function assessment are required to facilitate comparability between centres.

Atrial fibrillation (AF) is a common cardiac arrhythmia with a 5-20% annual risk of stroke. Warfarin reduces this risk by at least 60%. Despite adequate anticoagulation within the target International Normalized Ratio (INR) range of 2.0-3.0, some patients still experience thrombotic and bleeding events. It is now possible to assess the intensity of anticoagulation with automated thrombin generation (TG) tests, such as the calibrated automated thrombogram (CAT). These tests were compared and an inverse relationship was found between the INR and CAT in 143 elderly AF patients. There was equally good correlation between the concentration of factors II, VII, IX and X and the INR and TG parameters. The peak thrombin was most strongly associated with the concentration of prothrombin fragment 1 + 2 in plasma. There was wide variability in TG parameters in patients with identical INR values, sometimes up to a fourfold difference. This TG variability in individuals with the same INR is not due to inflammation, at least when the latter is measured as the concentration of factor VIII coagulant activity, von Willebrand factor antigen, high sensitivity C-reactive protein and fibrinogen. It was concluded that, although the TG and INR were closely correlated there was wide variability in peak thrombin and endogenous thrombin potential in patients within the INR therapeutic range, the cause of which remains unclear.

Introduction Global coagulation tests may have a better relation with phenotype in haemophilia than traditional coagulation tests. These include the Calibrated Automated Thrombin generation assay (CAT) and modified thromboelastometry using low tissue factor triggering. Both have shown marked variability in thrombin generation and clot formation profiles respectively despite similar FVIII:C levels and have been suggested as means to monitor treatment. Data with modified thromboelastometry are largely limited to severe and moderate haemophiliacs. CAT measurements in haemophilia are generally performed at low TF concentrations (1 pM) because of a higher sensitivity for the intrinsic pathway at this concentration but is also sensitive for FVIII at higher concentrations (5 pM) and this has the advantage that inhibition of contact factor activation can be avoided. No formal comparison of both TF concentrations has been reported and the data on modified thromboelastometry in mild haemophilia are limited. Methods In this study we compared thrombin generation at 1 and 5 pM in 57 haemophilia patients without exposure to treatment and 41 patients after treatment. We also assessed the sensitivity of thromboelastometry for haemophilia A in 29 patients. Results and conclusion We found that CAT discriminates well between normal individuals and haemophilia patients; also FVIII:C correlates well with the ETP/peak. We found no clear advantages of measurements at 1 compared to 5 pM but found increased variation over time at 1 pM. The sensitivity of modified thromboelastometry for haemophilia A was less than CAT with abnormal measurements largely limited to severe and moderate patients. Larger studies correlating both methods with clinical outcome are required.

Thrombin generation has been suggested as a method to monitor treatment with factor eight inhibitor bypassing activity (FEIBA) or recombinant FVIIa (rFVIIa). The sensitivity of the assay for individual coagulation factors is dependent on the tissue factor (TF) concentration. An inverse relation between the rFVIIa concentration needed to shorten the clotting time and TF concentration has been shown but the data on thrombin generation are inconsistent. Information on TF concentration in measurements with FEIBA is limited. We studied the influence of TF concentration (1 and 5 pM) on thrombin generation through spiking experiments with rFVIIa and/or FEIBA in the plasma of severe haemophilia A patients and after four and three treatment episodes, respectively, using the calibrated automated thrombin generation assay (CAT) in platelet poor plasma. Spiking with FEIBA showed a linear relation with the endogenous thrombin potential (ETP)/peak at 1 pM but substrate depletion at 5 pM. Spiking with rFVIIa showed a near linear dose-response relation with the ETP/peak at 1 pm but only a shortening of the initiation phase at 5 pM. Similar effects were present in post-treatment samples. FEIBA acted synergistically with rFVIIa. This suggest a role for CAT in monitoring inhibitor bypass treatment but low TF concentrations are required to avoid substrate depletion with FEIBA and to demonstrate the effect of rFVIIa.

Prevention of medical errors is a major goal of healthcare, though healthcare workers themselves have not yet fully accepted or implemented reliable models of system error, and neither has the public. While there is widespread perception that most medical errors arise from an inappropriate or delayed clinical management, the issue of laboratory errors is receiving a great deal of attention due to their impact on the quality and efficiency of laboratory performances and patient safety. Haemolytic specimens are a frequent occurrence in clinical laboratories, and prevalence can be as high as 3.3% of all of the routine samples, accounting for up to 40%-70% of all unsuitable specimens identified, nearly five times higher than other causes, such as insufficient, incorrect and clotted samples. This article focuses on this challenging issue, providing an overview on prevalence and leading causes of in vivo and in vitro haemolysis, and tentative guidelines on identification and management of haemolytic samples in clinical laboratories. This strategy includes continuous education of healthcare personnel, systematic detection/quantification of haemolysis in any sample, immediate clinicians warning on the probability of in vivo haemolysis, registration of non-conformity, completing of tests unaffected by haemolysis and request of a second specimen for those potentially affected.

Summary. We have investigated the effects of purified IgG fractions from plasma containing the lupus anticoagulant (LAC) and/or IgG anticardiolipin antibody (ACA) on the degradation of factor Va by an activated protein C-protein S complex. Plasma samples from 10 patients were studied. LAC was detected by a Russell's Viper venom technique. ACA was determined by ELISA. IgG fractions were obtained from each plasma sample by protein A-Sepharose fractionation. This fraction was shown to exhibit ACAILAC activity. Using purified activated protein C (APC), protein S and phosphati-dylserine/phosphatidylcholine, factor Va degradation was assessed in the presence and absence of IgG fractions from LAC/ACA containing plasmas. After 2 min incubation the mean factor Va degradation by APC and protein S in the presence of IgG LAC/ACA fractions was 14% compared with 52% with normal IgG. A similar effect was seen when phospholipid was substituted by washed freeze-thawed platelets. Experiments employing varying concentrations of protein S and phospholipid revealed marked differences in respect of the inhibitory specificity of the different antiphospholipid antibodies. These results indicate that antiphospholipid antibodies have an inhibitory effect on the activated protein C/protein S complex and provide some explanation for a relationship between antiphospholipid antibodies and thrombosis.

Accurate and precise measurement of plasma D-dimer levels is important in the diagnosis and management of venous thromboembolism. Considerable variability in D-dimer results obtained using different methods has, however, been reported in multicentre studies. This study explored in two separate multicentre exercises the degree of precision amongst laboratory D-dimer measurements, and the degree by which inter-method agreement could be improved using a calibration curve model. The first exercise demonstrated generally good within-centre precision, with 82% of the centres reporting results for two identical but differently coded samples within 10% of each other. However, six centres reported results which would have excluded deep vein thrombosis (DVT) for one sample but failed to exclude DVT for the other, identical sample. In the second exercise, overall between-method precision of D-dimer results for two samples was shown to improve markedly when a calibration model was applied, using the consensus median values obtained by all participants for three "calibration plasmas" to recalculate D-dimer values. For centres reporting results in fibrinogen equivalent units (FEUs), between-centre coefficients of variation (CVs) fell from 25.9% to 11.6% and 22.4% to 7.7%, respectively, for the two samples. For centres reporting in ng/ml, CVs fell from 45.3-21.6% and 40.8-11.6%, respectively. In conclusion, improved harmonisation of D-dimer results by different methods may be achieved by a calibration model and common calibrant plasmas.

The consequences of an erroneous thrombophilia diagnosis may be serious if it is used to determine clinical management. Therefore careful selection, assessment, and control of laboratory tests for thrombophilia are essential. As for other coagulation tests, the pre-analytical phase must be carefully controlled with attention to the specific problems associated with each type of assay. The investigator must then recognize that for most laboratory tests of thrombophilia, there are a number of assay types available, often based on different principles of analysis. This creates the potential for different users to obtain varying results depending on the technique employed. Such problems can occur in assays of antithrombin activity, depending on whether the assay employs factor Xa, human thrombin, or bovine thrombin. In clot-based assays of protein C and protein S, there can be specificity problems related to interference by factor V Leiden (FVL), antiphospholipid antibodies, and other substances. Even genetic tests can give erroneous results and should not automatically be seen as absolute without supporting evidence and careful quality-control measures. Whatever technique is selected, it is mandatory to incorporate sufficient concurrent quality-control samples to validate the results of thrombophilia tests. These should include assessment of the parameter at normal and abnormal levels to give confidence in results across the measurement range that would normally be encountered in routine practice. This should be used in conjunction with regular participation in external quality assessment (EQA) (which has been linked to improved laboratory performance in thrombophilia testing). Larger EQA programs can provide information concerning the relative performance of analytical procedures, including the method principle, reagents, and instruments. Herein, we describe many of the methodologic effects in detail. We use specific examples to illustrate the general principle that, in performing laboratory testing for thrombophilia, one must always consider the performance characteristics and limitations of the assay in use.

Molecular genetic analysis of families with haemophilia and other inherited bleeding disorders is now a common laboratory investigation. In contrast to phenotypic testing in which strict quality control is adhered to, in haemophilia molecular genetic testing there has been a lack of any external quality assurance schemes. In 1998 the UK National External Quality Assessment Scheme (UK NEQAS) established a pilot quality assurance scheme for molecular genetic testing in haemophilia. Results from three initial surveys highlighted problems with the quality of samples when used to screen for the intron 22 inversion within the F8 gene. The scheme was re-launched in 2003, and since that time there have been five exercises involving whole blood or immortalised cell line DNA. The results together with an overall summary of the exercise are subsequently returned to participants. Exercises to date have focused exclusively on haemophilia A and QA, material has included screening for the intron 1 and intron 22 inversions as well as sequence analysis. A paper exercise circulated in 2003 highlighted problems with the format of reports and, following feedback to participants, only a single error has been made in the subsequent four exercises. Participating laboratories now receive QA material every six months. Immortalised cell line material was introduced in 2005 and was shown to perform well. This will allow expansion of the scheme and a reduction in the dependence on blood donation.

SummaryForty-one laboratories participated in an international collaborative study to assess the suitability of a panel of three genomic DNA samples as the 1st International Genetic Reference Panel for the Factor V Leiden (FVL) variant.The code numbers of the materials were 03/254 (FV wild type), 03/260 (FVL homozygote) and 03/248 (FVL heterozygote).The participants evaluated the panel against their in-house controls which were known patient samples and commercial controls. In total, 859 genotype tests were carried out on the panel, with an error rate of 0.7%.The errors were not related to specific samples of the panel or to any specific techniques.The findings of this study have indicated that this panel is suitable to be used as a reference material for genotyping of factor V Leiden. It was therefore recommended that the three genomic DNA samples be established as the 1st International Genetic Reference Panel for Factor V Leiden, Human gDNA, 04/224. This recommendation was approved by the Scientific and Standardization Committee (SSC) of the ISTH (International Society on Thrombosis and Haemostasis) in June 2004 and the Expert Committee on Biological Standardization (ECBS) of theWorld Health Organization (WHO) in November 2004.

The Scientific Committee of Molecular Biology Techniques (C-MBT) in Clinical Chemistry of the IFCC has initiated a joint project in co-operation with the European Commission, Joint Research Centre, Institute of Reference Materials and Measurements to develop and produce plasmid-type reference materials (RMs) for the analysis of the human prothrombin gene G20210A mutation. Although DNA tests have a high impact on clinical decision-making and the number of tests performed in diagnostic laboratories is high, issues of quality and quality assurance exist, and currently only a few RMs for clinical genetic testing are available. A gene fragment chosen was produced that spans all primer annealing sites published to date. Both the wild-type and mutant alleles of this gene fragment were cloned into a pUC18 plasmid and two plasmid RMs were produced. In addition, a mixture of both plasmids was produced to mimic the heterozygous genotype. The present study describes the performance of these reference materials in a commutability study, in which they were tested by nine different methods in 13 expert laboratories. This series of plasmid RMs are, to the best of our knowledge, the first plasmid-type clinical genetic RMs introduced worldwide.

The rare coagulation disorders are heritable abnormalities of haemostasis that may present significant difficulties in diagnosis and management. This review summarizes the current literature for disorders of fibrinogen, and deficiencies of prothrombin, factor V, FV + VIII, FVII, FX, the combined vitamin K-dependent factors, FXI and FXIII. Based on both collective clinical experience and the literature, guidelines for management of bleeding complications are suggested with specific advice for surgery, spontaneous bleeding, management of pregnancy and the neonate. We have chosen to include a section on Ehlers-Danlos Syndrome because haematologists may be consulted about bleeding manifestations in such patients.

To assess the practicality of the recent Scientific and Standardization committee (SSC) of the International Society on Thrombosis and Haemostasis (ISTH) recommendations in respect of the classification of hemophilia we distributed samples from three untreated subjects with hemophilia A to 91 UK hemophilia centers (HCs), comprising 20 comprehensive care centers (CCCs) and 71 HCs. Laboratories were requested to perform their routine factor (F)VIII:C assays and to classify the severity of hemophilia. Median values of1 U dL-1 were obtained on two samples. However, for each of the two, approximately 30% of laboratories obtained results in the range 1-29 U dL-1 and 1-33 U dL-1 respectively. For one of these samples 17 laboratories diagnosed severe hemophilia despite obtaining FVIII:C levels in the range 1-5 U dL-1. The median FVIII:C for the third sample was 5.8 U dL-1 with a range of 1.5-36 U dL-1. For this sample eight centers diagnosed severe hemophilia. Fifty-four laboratories obtained a result5 U dL-1; 21 of these diagnosed mild hemophilia, 31 moderate hemophilia and two severe hemophilia. Results from CCCs were more accurate and more precise than those from HCs. Our results indicate a need for improved standardization of FVIII assays. In the UK there remains a lack of consensus in respect of the laboratory diagnostic criteria for the classification of hemophilia A.

See also Shermock KM, Kraus P, Streiff MB. Novel analysis of clinically relevant diagnostic errors in point-of-care devices: a reply to a rebuttal. This issue, pp 1192–4; Shermock KM, Streiff MB, Pinto BL, Kraus P, Pronovost PJ. Novel analysis of clinically relevant diagnostic errors in point-of-care devices. J Thromb Haemost 2011; 9: 1769–75.

SummaryThe wide availability of fibrinogen estimations based on the prothrombin time (PT-Fg) has caused concern about the variability and clinical utility of fibrinogen assays. In a multi-centre study, we investigated fibrinogen assays using various reagents and analysers. Clauss assays generally gave good agreement, although one reagent gave 15–30% higher values in DIC and thrombolysis. Two commercial reference preparations had much lower potencies than the manufacturers declared, and plasma turbidity influenced parallelism in some Clauss assays. PT-Fg assays gave higher values than Clauss and showed calibrant dependent effects, the degree of disparity correlating with calibrant and test sample turbidity. Analyser and thromboplastin dependent differences were noted. The relationship between Clauss and PT-Fg assays was sigmoid, and the plateau of maximal PT-Fg differed by about 2 g/l between reagents. ELISA and immunonephelometric assays correlated well, but with a high degree of scatter. Antigen levels were higher than Clauss, but slightly lower than PT-Fg assays, which appeared to be influenced by degraded fibrinogen. Clauss assays are generally reproducible between centres, analysers and reagents, but PT-Fg assays are not reliable in clinical settings.

The coagulation laboratory has a vital role in the diagnosis and management of patients with familial and acquired haemorrhagic and thrombotic disorders. Wherever possible the laboratory methods employed must reflect the state of the art, and the results generated, including locally derived reference values, should be reliable, reproducible and unambiguous. These issues are of particular importance when investigations are undertaken for the diagnosis of a possible familial disorder. When an error occurs, there is possibility of misdiagnosis and, irrespective of whether this causes an individual to be misdiagnosed as either having or not having a familial defect, the clinical consequences of such a mistake are likely to be serious. Within recent years the workload and scope of the routine laboratory has increased substantially, and this has been accompanied by the introduction of automated equipment employing a variety of different technologies. The potential for laboratory error is therefore considerable, and monitoring of laboratory performance through a programme of quality assurance is an essential laboratory requirement.

In a recent publication Poller et al. (1) described quality assessment (QA) of CoaguChek (CUC) point-of-care (POC) international normalized ratio (INR) monitors and compared 2 procedures. The study compared the proportion of 523 CUC monitors having unsatisfactory performance as indicated by either a >15% deviation of INR results from the INR assigned to 5 lyophilized QA plasmas or a >15% deviation from the overall median of INRs obtained on the monitors included in the survey. The INRs assigned to QA plasmas had been derived from a full calibration in compliance with WHO-recommended procedures and a local mean normal prothrombin time. The CUC device is calibrated by the manufacturer for use with non-anticoagulated whole blood. Patients using the device analyze whole blood and not plasma. Furthermore, the calibration procedure employed by Poller et al. (1) used whole blood samples to derive the International Sensitivity Index for assigning INRs to the lyophilized plasma samples. We believe that caution …

Accurate and precise measurement of plasma D-dimer levels is important in the diagnosis and management of venous thromboembolism. Considerable variability in D-dimer results obtained using different methods has, however, been reported in multicentre studies. This study explored in two separate multicentre exercises the degree of precision amongst laboratory D-dimer measurements, and the degree by which inter-method agreement could be improved using a calibration curve model. The first exercise demonstrated generally good within-centre precision, with 82% of the centres reporting results for two identical but differently coded samples within 10% of each other. However, six centres reported results which would have excluded deep vein thrombosis (DVT) for one sample but failed to exclude DVT for the other, identical sample. In the second exercise, overall between-method precision of D-dimer results for two samples was shown to improve markedly when a calibration model was applied, using the consensus median values obtained by all participants for three "calibration plasmas" to recalculate D-dimer values. For centres reporting results in fibrinogen equivalent units (FEUs), between-centre coefficients of variation (CVs) fell from 25.9% to 11.6% and 22.4% to 7.7%, respectively, for the two samples. For centres reporting in ng/ml, CVs fell from 45.3-21.6% and 40.8-11.6%, respectively. In conclusion, improved harmonisation of D-dimer results by different methods may be achieved by a calibration model and common calibrant plasmas.

We present a two centre study designed to assess the sensitivity of Actin FS and Actin FSL to deficiencies of factor VIII, IX, XI or XII. The study was undertaken at two centres to avoid bias due to the investigations being undertaken on one analyser. Samples from patients with a factor VIII (n = 36, F VIII = < 1.0-50 iu/dl), factor IX (n = 22, F IX = 2-48 iu/dl), factor XI (n = 23, F XI = 5-50 u/dl) or a factor XII (n = 18, F XII = 1-50 u/dl) deficient state were studied. Activated partial thromboplastin times (APTT) were determined using two batches of Actin FS and of Actin FSL; comparison of APTT results between centres was facilitated by the conversion of clotting times to ratios (test divided by geometric mean normal clotting time). APTT ratios were considered to be elevated if greater than two standard deviations above the mean normal. The factor deficient status of each sample was verified by assaying all samples for factors VIII, IX, XI and XII. Clotting factor assays were performed on a Sysmex CA-1000 fitted with research software, which permitted the auto-dilution and testing of three serial dilution of both a reference preparation and each patient's sample. Assay results were calculated using parallel-line Bioassay principles. This procedure allowed for variation in clotting times due to the effect of temporal drift of any of the reagents within the assay system. Actin FS and Actin FSL demonstrate acceptable sensitivity to factor VIII deficiency, however, both reagents failed to detect a large proportion of factor XI (17.4% and 30.4% of samples, respectively) and factor XII (66.7% and 72.2%, respectively) deficiencies. The detection rate with Actin FSL for factor IX deficiency was also poor (36.4% not detected). As factor IX and XI deficiencies are both associated with haemorrhagic disorders, the inability of these reagents to detect such abnormalities gave cause for concern.

AIM: To assess the sensitivity of activated partial thromboplastin time (APTT) reagents for monitoring heparin dosage using data from the UK National External Quality Assessment Scheme (NEQAS) for blood coagulation. METHODS: Data were reviewed from four surveys using samples prepared by addition of heparin to normal plasma in vitro and from two surveys in which samples were prepared using plasma from patients receiving heparin therapy (ex vivo samples). RESULTS: For both in vitro and ex vivo samples, notable differences between APTT reagents with respect to heparin sensitivity were noted. This indicates that a uniform therapeutic range of 1.5-2.5 calculated by the APTT ratio may not be appropriate for all reagents. Reagent sensitivity in ex vivo samples was substantially different to that in in vitro samples. CONCLUSIONS: The results of this large series of laboratories clearly indicate that reagent specific therapeutic ranges may be necessary, and that samples prepared by the addition of heparin to normal plasma in vitro can be misleading and should not be used.

Further Evidence that Activated Protein C Resistance Affects Protein C Coagulant Activity Assays

We have studied the potential thrombogenicity of two different heat-treated prothrombin complex concentrates (PCC) in patients with Haemophilia B. Seven patients were studied on nine separate occasions. Four of the patients had chronic hepatitis C (HCV) associated liver disease and three were HIV-antibody positive. The PCCs were Profilnine (Alpha Therapeutics, Thetford, UK) and 9A (Bio-Products Laboratory, Elstree, UK) and the dose administered ranged from 35 to 60 U/kg. Blood samples were taken on ten separate occasions; twice before the infusion and at 15, 40, 60, 75 and 120 min and 4, 8 and 24 h after the infusion of PCC. Investigations included prothrombin time, kaolin cephalin clotting time, thrombin time, fibrin(ogen) degradation products, factor VIII, factor IX, antithrombin III and fibrinopeptide A (FPA). Fibrinopeptide A rises were seen following two of six infusions of 9A and one of three infusions of Profilnine. On all three occasions the rise in FPA was transient, returning to baseline levels within 120 min. Plasma beta-thromboglobulin (BTG) was assayed in three patients and in one patient, the rise in FPA was followed by an increase in BTG. No other changes were observed and there were no clinical features of disseminated intravascular coagulation. Our results indicate that even with normal clinical doses of PCC, intravascular thrombin generation can occur in patients with Haemophilia B. However, this effect is inconsistent both with respect to PCC batch and patient, but may occur in the absence of HIV infection and HCV liver disease.

A UK National External Quality Assessment Scheme (UK Neqas) for Molecular Genetic Testing for the Diagnosis of Familial Thrombophilia