- Email: [email protected]
The Factor V (Leiden) Test
Thrombosis Research 96 (1999) 125–133
REGULAR ARTICLE
The Factor V (Leiden) Test: Evaluation of an Assay Based on Dilute Russell Viper Venom Time for the Detection of the Factor V Leiden Mutation Peter Quehenberger1, Sylvia Handler1, Christine Mannhalter1, Paul Alexander Kyrle2 and Wolfgang Speiser1 The Clinical Institute of Medical and Chemical Laboratory Diagnostics and 2 Department for Internal Medicine I, University of Vienna, A-1090 Vienna, Austria.
1
(Received 30 November 1998 by Editor I. Pabinger; revised/accepted 9 April 1999)
Abstract In the present study a new clotting assay for the detection of an increased resistance of coagulation factor V against degradation by activated protein C (Factor V Leiden mutation, FVLM) was evaluated. The Factor V (Leiden) Test (Gradipore, North Ryde NSW, Australia) is based on the dilute Russell Viper Venom Time (DRVVT), which is prolonged when the plasma sample is preincubated with dilute whole Agkistrodon contortrix contortrix venom for activation of protein C (PC). In contrast to the DRVVT based global assay, Protein C Pathway Test (Gradipore, North Ryde NSW, Australia) this new assay is expected to be more specific for FVLM because of optimized amounts of the venom. The test result is expressed as the ratio between the DRVVT with and without addition of the venom. The following precision values were found: intraassay coefficient of variation (CV): 5.53% (n520) in the normal range, 4.30% (n520) in the pathological range; interassay CV: Abbreviations: PC, protein C; PS, protein S; aPTT, activated partial thromboplastin time; OA, oral anticoagulant; FVDP, factor V-deficient plasma; DRVVT, dilute Russell Viper Venom Time; FVLM, Factor V Leiden mutation; CV, coefficient of variation; AT, antithrombin. Corresponding author: W. Speiser, MD, Clinical Institute of Medical and Chemical Laboratory Diagnostics, University of Vienna, AKH-Wien, Leitstelle 5J, Waehringer Guertel 18-20, A-1090 Vienna, Austria. Tel: 1(43) 1-404005369; Fax: 1(43) 1-4036688; E-mail: ,[email protected]..
6.90% (n510) and 7.64% (n510), respectively. A normal range (5th to 95th percentile) of 2.12 to 3.08 was calculated from 50 healthy controls. A ratio below 2.12 was found in all samples from patients with FVLM (n521), in 9 of 12 patients with PC, in 0 of 6 with protein S (PS), and in 0 of 4 with antithrombin (AT) deficiency. There was, however, a good discrimination between carriers of the FVLM (highest ratio 1.44) and patients deficient in PC (lowest ratio 1.59), in particular when samples were prediluted with factor V deficient plasma FVDP (1.16 vs. 1.96, respectively). Predilution of samples with FVDP caused a clear discrimination between controls and patients deficient in PC, PS, AT, and FVLM-positive individuals and also in patients on oral anticoagulant treatment. Our data show that the Factor V (Leiden) Test discriminates well between carriers of the FVLM and healthy controls or patients deficient in PC, PS, and AT. Individuals presenting values between the lower cutoff of controls and the range in which FVLM-positive individuals are found are highly suspicious for protein C deficiency. 1999 Elsevier Science Ltd. All rights reserved. Key Words: Evaluation; Factor V Leiden mutation; Dilute Russell Viper Venom Time
W
hen thrombin, the central coagulation protease, binds to its endothelial receptor thrombomodulin, its mode of action switches from procoagulant to anticoagulant since
0049-3848/99 $–see front matter 1999 Elsevier Science Ltd. All rights reserved. PII S0049-3848(99)00090-0
126
P. Quehenberger et al./Thrombosis Research 96 (1999) 125–133
this activator is now capable of activating the coagulation inhibitor protein C (PC). Together with its cofactor protein S (PS), activated PC inactivates coagulation factors Va and VIIIa [1]. An impaired capacity of the PC/PS pathway results in an increased risk of venous thromboembolism; such a dysfunction may not only result from defects in the PC or PS molecule [2] but can also be caused by an increased resistance of factor V to degradation by activated PC/PS. This decreased susceptibility for inactivation is due to a G→A transition at nucleotide position 1691 of the factor V gene, which leads to an Arg→Gln substitution at residue 506, a cleavage site of factor Va that is important for its inactivation [3]. This increased resistance against activated PC caused by mutated factor V (factor V Leiden or FV:Q506) is the most frequent hereditary abnormality associated with deep vein thrombosis [4]. At present it is a matter of debate whether testing for factor V Leiden mutation should not only be performed on patients suffering from venous thromboembolism, but also in as-yet unaffected individuals before exposure to situations with increased risk for venous thrombosis, such as surgery or treatment with oral contraceptives [5]. Direct detection of factor V is based on DNA analysis [3], a rather labor-intensive and expensive method that is not suitable for screening of high numbers of samples. Therefore several functional tests based on the activated partial thromboplastin time (aPTT) or on modified tissue factor assays were developed [4,6–8]. It was previously shown that these tests are influenced by the presence of anticoagulants, such as oral anticoagulants (OAs), heparin, or lupus anticoagulant, and by the levels of various coagulation factors [9,10]. Recently sensitivity and specificity of these tests were improved by predilution of plasma samples with factor V-deficient plasma (FVDP) [11–13]. The purpose of our study was to evaluate the new assay Factor V (Leiden) Test (Gradipore, North Ryde NSW, Australia), which is based on the dilute Russell Viper Venom Time (DRVVT) [14] with respect to its sensitivity and specificity in the detection of the factor V Leiden mutation (FVLM) in patients with PC, PS, and antithrombin deficiency. Five patients with lupus anticoagulant were also tested. Furthermore, it was our aim to investigate the effect of predilution with FVDP on the values obtained from patients on OA therapy.
1. Patients and Methods 1.1. Patients Plasma of 50 healthy controls (25 males, 25 females; mean 37 years) without any known defects in the blood coagulation inhibition system [PC deficiency, PS deficiency, antithrombin (AT) deficiency, resistance against activated PC/FVLM] and showing normal prothrombin time, activated partial thromboplastin time, and fibrinogen values were used to estimate the normal range for the Factor V (Leiden) Test. The DRVVT ratio was also determined in 21 patients with FVLM (18 patients heterozygous and three patients homozygous for FVLM), in 12 patients deficient solely in PC, in one patient with combined PC deficiency and FVLM, in six patients deficient in PS, in seven patients with combined PS deficiency and FVLM, and in four patients deficient in AT. At the time of investigation none of these patients was on OA therapy. Furthermore, two groups of patients on OA therapy were studied. One group consisted of seven patients heterozygous for FVLM, seven patients deficient solely in PC, four patients with combined PC deficiency and FVLM, four patients deficient in PS, three patients with combined PS deficiency and FVLM, and three patients deficient in AT; the other group of OA-treated patients (n521) had none of the above-mentioned hereditary risk factors for thrombosis and showed international normalized ratio values between 1.7–3.3 (median 2.260.42). All carriers of the FVLM had a ratio below 1.85 measured with Coatest APC Resistance Test (Chromogenix AB, Mo¨lndal, Sweden). All investigated patients deficient for PC (ranges of PC activity: 25–60%) or PS (ranges of free PS antigen: 6–42%) were members of families with at least one person with a defined defect in the corresponding gene. In 10 patients with PC deficiency and eight patients with PS deficiency, the respective gene defects were previously described in detail [15–17]. All patients were in a stable phase without signs of acute illness.
1.2. Blood Sampling Between 8:00 and 10:00 a.m. after overnight fasting, citrated blood (Na citrate, final concentration 0.013 M) for coagulation testing and for DNA analysis
P. Quehenberger et al./Thrombosis Research 96 (1999) 125–133
was taken into siliconized glass tubes (Vacutainer Becton Dickinson, Meylan Cedex, France). Platelet-poor plasma was prepared from citrated blood by centrifugation at 30003g for 20 minutes. Plasma was then assayed immediately or frozen at 2708C.
1.3. Assay Systems Prothrombin time (Thromborel R, Dade-Behring, Marburg, Germany), aPTT (Dade Actin-FS, Baxter Diagnostics Inc, Deerfield, IL, USA), antithrombin activity (STA Antithrombin III, Diagnostica Stago, Asnieres-Sur-Seine, France; normal range: 75–125%), PC activity (COAMATIC Protein C, Chromogenix AB, Mo¨lndal, Sweden; 65– 130%), and resistance against activated PC (Coatest APC Resistance, Chromogenix AB; ratio. 1.85) were determined on the analyzer STA (Stago Diagnostica, Asnieres-Sur-Seine, France); aPTT mixing studies for detection of Lupus anticoagulant were carried out with a 1:1 mixture of patients plasma with pooled normal plasma and considered as positive when the aPTT value from pooled normal plasma was prolonged for more than 6 seconds through the patient’s plasma. Lupus anticoagulant confirmation test STACLOT LA and STACLOT PNP were determined on ST4 coagulation analyzer (Stago Diagnostica). Samples were considered as positive when the clotting time was shortened after incubation with hexagonal phase phosphatidylethanolamine (STACLOT LA) to more than 2 seconds and after incubation with platelet lysate (STACLOT PNP) to more than 0.5 seconds; free PS antigen was determined by enzyme-linked immunosorbent assay (Asserachrom PS, Diagnostica Stago; 50–160%). For direct detection of the factor V gene mutation, genomic DNA was isolated from citrated blood by a standard method. The DNA region containing guanine 1691 was amplified by polymerase chain reaction as described previously [11] with primers described by Bertina et al [3]. The polymerase chain reaction product was digested by the restriction enzyme MnlI and analyzed on polyacrylamide gels.
1.4. Factor V (Leiden) Test The Factor V (Leiden) Test (Gradipore, North Ryde NSW, Australia) was performed fully automated on STA analyzer (Stago Diagnostica, Asnieres-
127
Sur-Seine, France) as follows: 50 mL of plasma were mixed with 50 mL 0.9% NaCl (DRVVT 2) or 50 mL of venom activator reagent (whole venom from Agkistrodon contortrix; DRVVT 1) for activation of endogenous PC, and incubated for 5 minutes at 378C; thereafter 50 mL of diluted phospholipid enriched Russell’s Viper Venom for activation of factor X and subsequent initiation of the coagulation process were added. The result of the assay is expressed as the ratio between DRVVT 1 and DRVVT 2. FVDP (Chromogenix AB, Mo¨lndal, Sweden) was used for predilution (114) of samples. Freezing samples once at 2708C and thawing was found to significantly reduce the DRVVT ratio (immediate testing: ratio of healthy controls 3.826 0.42; ratio determined after once freezing and thawing: 2.8260.45, p,0.05, n510). For the evaluation of the assay only once-frozen and thawed plasma was used. Measurement of PC activity, free PS antigen, and resistance against activated PC using the Coatest APC Resistance Test and the Factor V (Leiden) Test were performed as duplicates. Results of these tests are given as mean values. Sensitivity is defined as true positive results in patients with disease/total patients with disease. Specificity is defined as true negative results in patients without disease/total patients without disease.
2. Results 2.1. Precision Studies Intraassay imprecision of the assay was estimated from 20 determinations of a plasma sample with a ratio within the normal range (A) and a sample with a reduced ratio (B). For sample A, the mean DRVVT ratio was 2.42, standard deviation (SD) was 0.13, and coefficient of variation (CV) was 5.53%. For sample B, the mean DRVVT ratio was 1.15, SD was 0.05, and CV was 4.30%. Interassay imprecision (n510) was calculated from determinations performed on pooled normal plasma (C) and pooled abnormal plasma (D). For sample C, the mean DRVVT ratio was 2.42, SD was 0.17, and CV was 6.90%. For sample D, the mean DRVVT ratio was 1.07, SD was 0.08, and CV was 7.64%.
128
P. Quehenberger et al./Thrombosis Research 96 (1999) 125–133
Fig. 1. The DRVVT ratios obtained with Factor V (Leiden) Test of samples from healthy controls, patients with antithrombin, PC, or PS deficiency, and/or FVLM are given. Samples of homozygous patients for FVLM are given as full squares. A lower cutoff of 2.12 was calculated from the values of healthy controls.
2.2. Normal Range Normal ranges for DRVVT 1, DRVVT 2, and the ratio were determined from samples obtained from 50 healthy volunteers who showed PC, PS, and AT levels within the normal range and were negative for FVLM: mean DRVVT ratio was 2.44, range 2.1–3.52 (5th to 95th percentile 2.12–3.08).
2.3. Effect of PC, PS, AT Deficiency and FVLM on the Factor V (Leiden) Test Fifty-one patients not on OA therapy and with known mutations in PC (n512), PS (n56), AT (n54), factor V gene (n521), or combined mutations (PC/FV Leiden: n51; PS/FV Leiden: n57) were tested with the Factor V (Leiden) Test. Results are shown in Figure 1. All 29 carriers of the FVLM (the combined defects are included in this group) revealed a DRVVT ratio of ,2.12; thus the Factor V (Leiden) Test shows a sensitivity of 1.0 in the diagnosis of the FVLM. In contrast to the aPTT-based assay, the ratios of heterozygous and homozygous patients were not markedly different from each other in the Factor V (Leiden) Test.
Nine of 12 patients with PC deficiency showed a DRVVT ratio of ,2.12 (lower cutoff calculated for nonanticoagulated controls). The three patents with a DRVVT ratio in the normal range had PC activities ranging from 60 to 50%; none of the patients with PS or with AT deficiency showed an abnormal result. The ratios of FVLM-positive individuals were between 0.83 and 1.44; neither healthy controls nor patients deficient in AT, PC, or PS had a ratio of <1.44. Taking a diagnostic ratio of ,1.44 for FVLM, a specificity of 1.0 is thus calculated. Taking the 5th percentile of healthy controls (2.12), the specificity of the Factor V (Leiden) Test in the group of patients deficient in AT or with hereditary defects in the PC/PS pathway was only 0.52. Our data suggest that individuals showing a ratio between 2.12 (lower cutoff of untreated healthy controls) and 1.44 (highest value determined in the group of carriers of the FVLM) are highly suspect for PC deficiency. We also investigated samples from individuals on OA therapy who were free of hereditary thrombosis risk factors. Our results suggest that OA therapy causes a decrease in the ratio: median of untreated controls 2.44 (range 2.1–3.52), median of patients on OA
P. Quehenberger et al./Thrombosis Research 96 (1999) 125–133
129
Fig. 2. DRVVT ratios of samples from healthy controls, from OA-treated patients without any known hereditary risk factors for thrombosis (5OA controls), and from patients with antithrombin, PC, or PS deficiency, and/or FVLM are given. The lower cutoff values of the DRVVT ratio for nonanticoagulated healthy controls (2.12) and for OA controls (1.37) are shown.
therapy 2.25 (1.19–3.94). In the group of OAtreated individuals (21 controls, 28 patients) we found no good discrimination between controls, patients deficient in PC, and carriers of the FVLM, even if a lower 5th percentile specific for OA treatment is calculated (Figure 2). One out of six PSdeficient patients on OA therapy tested abnormal.
2.4. Effect of Plasma Predilution with FVDP on the Factor V (Leiden) Test After predilution of plasma samples (n550) with FVDP, the following normal ranges were obtained: mean DRVVT: ratio 2.11, range 1.95–2.43 (5th to 95th percentile 2.0–2.34). As shown in Figure 3 two of 12 PC-deficient patients showed, even after sample predilution with FVDP, ratios (1.99 and 1.96) below the lower cutoff calculated for FVDP prediluted normal samples (2.0). Samples of AT- and PS-deficient patients showed normal results, whereas all carriers of the FV Leiden mutation had ratios<1.16, values clearly lower than the lowest ratio of healthy controls or PC-deficient patients. Thus predilution
with FVDP leads to an improved discrimination of carriers of the FVLM compared with nonprediluted samples. As shown in Figure 4 predilution of samples from OA-treated individuals revealed the following results: a nearly similar lower cutoff was calculated for OA-treated (1.92) and non-OA-treated (2.0) control individuals. All AT- and PC-deficient patients on OA therapy showed a ratio above 2.0, whereas one patient deficient in PS showed a markedly reduced ratio even after predilution. All carriers of the FVLM had ratios clearly below 1.92. Also in OA-treated patients predilution with FVDP caused an excellent discrimination between carriers of the FVLM (highest ratio 1.09) on the one hand and controls and patients deficient in PC, PS, and AT (lowest ratio in this group 1.67) on the other hand. In one patient deficient in PS, who is negative for the FVLM and shows a normal result in an aPTT-based assay (Coatest APC Resistance, Chromogenix AB), the ratio measured with the Factor V (Leiden) Test was not normalized after sample predilution with FVDP. This particular patient should probably be excluded from the group
130
P. Quehenberger et al./Thrombosis Research 96 (1999) 125–133
Fig. 3. DRVVT ratios of samples prediluted with FVDP from healthy controls, patients with antithrombin, PC, or PS deficiency, and/or FVLM are given. Samples of homozygous patients for FVLM are given as full squares. A lower cutoff of 2.0 was calculated from the values of prediluted healthy controls.
of PS-deficient patients, as he may, in addition to PS deficiency, suffer from a factor V mutation different from factor V Leiden.
2.5. Effect of Lupus Anticoagulant on the Factor V (Leiden) Test Factor V (Leiden) Test was performed on samples of five patients with lupus anticoagulant. Four of them were free of FVLM and one was positive for FVLM. Both in the prediluted and nonprediluted assay procedures the Factor V (Leiden) Test showed results corresponding to the molecular analysis (Table 1).
3. Discussion It was previously shown that functional tests for detection of FVLM based on the Russell Viper Venom method give a good discrimination between carriers and noncarriers of this mutation [18]. Therefore such a DRVVT-based test seems to be a reliable method to screen for FVLM. At
least two DRVVT-based assays for the detection of defects within the PC/PS pathway are available at the moment: the Protein C Pathway Test and the Factor V (Leiden) Test, both from Gradipore (Gradipore, North Ryde NSW, Australia). In a previous study we found that the Protein C Pathway Test, which was conceived as a global estimation of the PC pathway, is sensitive to PC and PS deficiency with an overlap between the values obtained from PC-deficient individuals and carriers of the FVLM (manuscript in preparation). Similar results were obtained with an other global assay system, ProC GLOBAL [19]. Due to low specificity the Protein C Pathway Test assay is not appropriate to screen for FVLM. According to informations obtained from the manufacturers, the newly developed Factor V (Leiden) Test is suitable for screening for FVLM as higher concentrations of the Agkistrodon contortrix venom are used for activation of PC compared with the Protein C Pathway Test assay to make the test more specific. In the present study the Factor V (Leiden) Test was evaluated with respect to its assay characteristics and its sensitivity an specificity in the diagnosis
131
P. Quehenberger et al./Thrombosis Research 96 (1999) 125–133
Fig. 4. DRVVT ratios of samples prediluted with FVDP from healthy controls, from control patients on OA therapy without any known hereditary risk factors for thrombosis (5OA controls), and from patients under OA therapy suffering from antithrombin, PC, or PS deficiency, and/or FVLM are given. Lower DRVVT ratio cutoff values in prediluted samples of 2.0 for nonanticoagulated healthy controls and of 1.92 for OA controls are shown.
of the FVLM. This assay, performed on STA coagulation analyzer, showed satisfying intra- and interassay precision values. Freezing and thawing once of plasma samples significantly influences the test results in lowering the ratio calculated from the DRVVTs in the absence and in the presence of the PC activator; this phenomenon is also observed with the Coatest APC Resistance Test from Chromogenix [11]. Therefore one should test patient samples in the same way as samples of healthy controls from which the reference values and the cutoffs were calculated. In the present study all
samples were frozen and thawed once before determination. Reference values and cutoffs calculated in our study should not be used in general. Due to possible influences of different analyzers on test results each laboratory should define its own reference and diagnostic ranges. We found an excellent sensitivity of 1.0 of the Factor V (Leiden) Test in the diagnosis of the FVLM. While the ratios obtained from patients suffering from PS and AT deficiency were clearly within the normal range, 9 of 12 PC-deficient patients showed abnormally low values. Although
Table 1. DRVVT ratios of five patients with Lupus anticoagulant without and with predilution with FVDP Without predilution with FVDP Patient 1 2 3 4 5 * s5seconds.
FV Leiden
DRVVT 1 (s)*
DRVVT 2 (s)
2 2 2 1 2
102.7 98.4 108.6 63.9 91.7
36.1 38.3 33.6 39.2 31.8
With predilution with FVDP
Ratio
DRVVT 1 (s)
DRVVT 2 (s)
Ratio
2.84 2.57 3.23 1.63 2.88
101.8 97.9 119.3 68.4 96.4
46.5 45.0 44.4 49.9 42.9
2.19 2.18 2.69 1.37 2.25
132
P. Quehenberger et al./Thrombosis Research 96 (1999) 125–133
only a limited number of patients was investigated, we were able to find a good discrimination between PC-deficient individuals and carriers of the FVLM, as the latter had lower ratios and no overlap was detected between these two groups of patients. The discrimination of the carriers of the FVLM could even be increased when the plasma samples were prediluted with FVDP. Calculating a diagnostic range for the FVLM, a specificity of 1.0 of the Factor V (Leiden) Test can be achieved. Our results suggest that samples with ratios between those of carriers of the FVLM and the lower cutoff of normals are highly suspect for PC deficiency. Patients to be tested for FVLM are often under OA therapy. OA therapy causes a lowering of the DRVVT ratio in the evaluated Factor V (Leiden) Test. We investigated two approaches to overcome this phenomenon in order to find an adequate procedure for performing the Factor V (Leiden) Test in OA-treated patients. First we determined the DRVVT ratio in 21 patients under OA therapy who had no abnormality in the PC/PS pathway and thus calculated a lower cutoff specific for OA treatment. This approach did not result in an adequate discrimination between carriers of the FVLM on the one hand and PC-deficient patients and controls on the other hand. In a second approach samples were prediluted with FVDP. As expected, predilution resulted in a clear discrimination between carriers of the FVLM and the other individuals investigated. We believe that, in contrast to the manufacturer’s instructions, patients on OA therapy who show a ratio within the diagnostic range of the FVLM should be retested after predilution with FVDP. Surprisingly we found one OAtreated patient deficient in PS without FVLM and a normal value in an aPTT-based factor V (Leiden) test who showed in repeated tests an abnormally low DRVVT ratio even after predilution with FVDP. This FVLM-negative patient may suffer from a factor V mutation different from the FVLM, which is only detected in a DRVVT- and not in an aPTT-based assay. The values of this patient in the figures are therefore given in parenthesis and should not be considered in the assessment of the sensitivity and the specificity of the Factor V (Leiden) Test; this patient should probably be included into the group defective in the factor V gene. Our results show that the Factor V (Leiden) Test evaluated in the present study is an adequate test
for the diagnosis of defects in the factor V gene that leads to an inadequate inactivation of factor V by activated PC. We believe that this test adequately distinguishes between patients with such a defect and those with deficiencies in the coagulation inhibitors PC and PS, in particular when samples are prediluted with FVDP. Provided that the samples are prediluted with FVDP, the test can also be used in patients treated with OAs. The relatively limited number of patients investigated in the present study, in comparison to similar test evaluations, is explained by the fact that the present investigation was restricted to patients with defined genetic defects and affected family members of these patients. This approach was chosen because it is known that functional tests for coagulation inhibitors show an overlap between affected individuals and normals [20]. Thus using the results of functional tests only, the group of patients would have been less well-defined.
References 1. Walker FJ, Fay PJ. Regulation of blood coagulation by the protein C system. FASEB J 1993;6:2561–7. 2. de Stefano V, Finazzi G, Mannucci PM. Inherited thrombophilia: pathogenesis, clinical syndromes, and management. Blood 1996;87: 3531–4. 3. Bertina RM, Koeleman BPC, Koster T, Rosendaal FR, Dirven RJ, de Ronde H, van der Velden PA, Reitsma PH. Mutation in blood coagulation factor V associated to activated protein C. Nature 1994;369:64–7. 4. Dahlba¨ck B, Carlsson M, Svensson PJ. Familial thrombophilia due to a previously unrecognized mechanism characterized by poor anticoagulant response to activated protein C: Prediction of a cofactor to activated protein C. Proc Natl Acad Sci 1993;90:1004–8. 5. Vandenbroucke JP, Koster T, Briet E, Reitsma PH, Bertina RM, Rosendaal FR. Increased risk of venous thrombosis in oral contraceptive users who are carriers of factor V Leiden mutation. Lancet 1994;344:1453–7. 6. Gouault-Heilmann M, Leroy-Matheron C. Evaluation of a new chronometric assay for factor
P. Quehenberger et al./Thrombosis Research 96 (1999) 125–133
7.
8.
9.
10.
11.
12.
13.
V Leiden-dependent APC resistance. Thromb Res 1997;85:357–62. Le DT, Griffin JH, Greengard JS, Mujumdar V, Rapaport SI. Use of a generally applicable tissue factor-dependent factor V assay to detect activated protein C-resistant factor Va in patients receiving warfarin and in patients with a lupus anticoagulant. Blood 1995;85:1704–11. Arkel YS, Ku DH, Kamiyama M, Pajaro R, Alder H, Marchand A. A highly specific functional test foe factor V Leiden: A modified tissue factor assay for activated protein C resistance. Haemostasis 1997;27:290–304. de Ronde H, Bertina RM. Laboratory diagnosis of APC-resistance: A critical evaluation of the test and the development of diagnostic criteria. Thromb Haemost 1994;72:880–6. Ehrenforth S, Radtke KP, Scharrer I. Acquired activated protein C-resistance in patients with lupus anticoagulant. Thromb Haemost 1995; 74:797–8. Kapiotis S, Quehenberger P, Jilma B, Handler S, Pabinger-Fasching I, Mannhalter C, Speiser W. Improved characteristics of aPC-resistance assay Coatest aPC resistance by predilution of samples with factor V deficient plasma. Am J Clin Pathol 1996;106:588–93. Jorquera JI, Motoro JM, Fernandez MA, Aznar JA, Aznar J. Modified test for activated protein C resistance. Lancet 1994;344:1162–3. Trossaert M, Conard J, Horellou MH, Samama MM, Ireland H, Bayston TA, Lane DA. Modi-
14.
15.
16.
17. 18.
19.
20.
133
fied aPC resistance assay for patients on oral anticoagulated therapy. Lancet 1994;344: 1709. Exner T, Murray B, Chong BH, Chesterman CN. Improved APC-resistance method based on a Russell’s viper venom clotting test. Thromb Haemost 1995;73:1119(Abstract). Poort SR, Pabinger-Fasching I, Mannhalter C, Reitsma PH, Bertina RM. Twelve novel and two recurrent mutations in 14 Austrian families with hereditary protein C deficiency. Blood Coagul Fibrin 1993;4:273–80. Wittmann E, Walter J, Pabinger-Fasching I, Watzke HH. Symptomatic hereditary type-II protein C deficiency caused by a missense mutation in exon IX of the protein C gene (Gly381 to Ser). Ann Hematol 1994;68:255–9. Mustafa S, Pabinger I, Mannhalter C. PS deficiency type I: Identification of point mutations in 9 of 10 families. Blood 1995;86:3444–51. Tripodi A, Negri B, Bertina RM, Mannucci PM. Screening for the FV:Q506 mutation— Evaluation of thirteen plasma-based methods for their diagnostic efficacy in comparison with DNA analysis. Thromb Haemost 1997;77: 436–9. Ruzicka K, Kapiotis S, Quehenberger P, Handler S, Pabinger-Fasching I, Mannhalter C, Jilma B, Speiser W. Evaluation of a new screening assay ProC GLOBAL for identification of defects in the protein C/PS anticoagulant pathway. Thromb Res 1997;87:501–10. Miletich J, Sherman L, Broze G Jr. Absence of thrombosis in subjects with heterozygous protein C deficiency. N Engl J Med 1987; 317:991–6.
REGULAR ARTICLE
The Factor V (Leiden) Test: Evaluation of an Assay Based on Dilute Russell Viper Venom Time for the Detection of the Factor V Leiden Mutation Peter Quehenberger1, Sylvia Handler1, Christine Mannhalter1, Paul Alexander Kyrle2 and Wolfgang Speiser1 The Clinical Institute of Medical and Chemical Laboratory Diagnostics and 2 Department for Internal Medicine I, University of Vienna, A-1090 Vienna, Austria.
1
(Received 30 November 1998 by Editor I. Pabinger; revised/accepted 9 April 1999)
Abstract In the present study a new clotting assay for the detection of an increased resistance of coagulation factor V against degradation by activated protein C (Factor V Leiden mutation, FVLM) was evaluated. The Factor V (Leiden) Test (Gradipore, North Ryde NSW, Australia) is based on the dilute Russell Viper Venom Time (DRVVT), which is prolonged when the plasma sample is preincubated with dilute whole Agkistrodon contortrix contortrix venom for activation of protein C (PC). In contrast to the DRVVT based global assay, Protein C Pathway Test (Gradipore, North Ryde NSW, Australia) this new assay is expected to be more specific for FVLM because of optimized amounts of the venom. The test result is expressed as the ratio between the DRVVT with and without addition of the venom. The following precision values were found: intraassay coefficient of variation (CV): 5.53% (n520) in the normal range, 4.30% (n520) in the pathological range; interassay CV: Abbreviations: PC, protein C; PS, protein S; aPTT, activated partial thromboplastin time; OA, oral anticoagulant; FVDP, factor V-deficient plasma; DRVVT, dilute Russell Viper Venom Time; FVLM, Factor V Leiden mutation; CV, coefficient of variation; AT, antithrombin. Corresponding author: W. Speiser, MD, Clinical Institute of Medical and Chemical Laboratory Diagnostics, University of Vienna, AKH-Wien, Leitstelle 5J, Waehringer Guertel 18-20, A-1090 Vienna, Austria. Tel: 1(43) 1-404005369; Fax: 1(43) 1-4036688; E-mail: ,[email protected]..
6.90% (n510) and 7.64% (n510), respectively. A normal range (5th to 95th percentile) of 2.12 to 3.08 was calculated from 50 healthy controls. A ratio below 2.12 was found in all samples from patients with FVLM (n521), in 9 of 12 patients with PC, in 0 of 6 with protein S (PS), and in 0 of 4 with antithrombin (AT) deficiency. There was, however, a good discrimination between carriers of the FVLM (highest ratio 1.44) and patients deficient in PC (lowest ratio 1.59), in particular when samples were prediluted with factor V deficient plasma FVDP (1.16 vs. 1.96, respectively). Predilution of samples with FVDP caused a clear discrimination between controls and patients deficient in PC, PS, AT, and FVLM-positive individuals and also in patients on oral anticoagulant treatment. Our data show that the Factor V (Leiden) Test discriminates well between carriers of the FVLM and healthy controls or patients deficient in PC, PS, and AT. Individuals presenting values between the lower cutoff of controls and the range in which FVLM-positive individuals are found are highly suspicious for protein C deficiency. 1999 Elsevier Science Ltd. All rights reserved. Key Words: Evaluation; Factor V Leiden mutation; Dilute Russell Viper Venom Time
W
hen thrombin, the central coagulation protease, binds to its endothelial receptor thrombomodulin, its mode of action switches from procoagulant to anticoagulant since
0049-3848/99 $–see front matter 1999 Elsevier Science Ltd. All rights reserved. PII S0049-3848(99)00090-0
126
P. Quehenberger et al./Thrombosis Research 96 (1999) 125–133
this activator is now capable of activating the coagulation inhibitor protein C (PC). Together with its cofactor protein S (PS), activated PC inactivates coagulation factors Va and VIIIa [1]. An impaired capacity of the PC/PS pathway results in an increased risk of venous thromboembolism; such a dysfunction may not only result from defects in the PC or PS molecule [2] but can also be caused by an increased resistance of factor V to degradation by activated PC/PS. This decreased susceptibility for inactivation is due to a G→A transition at nucleotide position 1691 of the factor V gene, which leads to an Arg→Gln substitution at residue 506, a cleavage site of factor Va that is important for its inactivation [3]. This increased resistance against activated PC caused by mutated factor V (factor V Leiden or FV:Q506) is the most frequent hereditary abnormality associated with deep vein thrombosis [4]. At present it is a matter of debate whether testing for factor V Leiden mutation should not only be performed on patients suffering from venous thromboembolism, but also in as-yet unaffected individuals before exposure to situations with increased risk for venous thrombosis, such as surgery or treatment with oral contraceptives [5]. Direct detection of factor V is based on DNA analysis [3], a rather labor-intensive and expensive method that is not suitable for screening of high numbers of samples. Therefore several functional tests based on the activated partial thromboplastin time (aPTT) or on modified tissue factor assays were developed [4,6–8]. It was previously shown that these tests are influenced by the presence of anticoagulants, such as oral anticoagulants (OAs), heparin, or lupus anticoagulant, and by the levels of various coagulation factors [9,10]. Recently sensitivity and specificity of these tests were improved by predilution of plasma samples with factor V-deficient plasma (FVDP) [11–13]. The purpose of our study was to evaluate the new assay Factor V (Leiden) Test (Gradipore, North Ryde NSW, Australia), which is based on the dilute Russell Viper Venom Time (DRVVT) [14] with respect to its sensitivity and specificity in the detection of the factor V Leiden mutation (FVLM) in patients with PC, PS, and antithrombin deficiency. Five patients with lupus anticoagulant were also tested. Furthermore, it was our aim to investigate the effect of predilution with FVDP on the values obtained from patients on OA therapy.
1. Patients and Methods 1.1. Patients Plasma of 50 healthy controls (25 males, 25 females; mean 37 years) without any known defects in the blood coagulation inhibition system [PC deficiency, PS deficiency, antithrombin (AT) deficiency, resistance against activated PC/FVLM] and showing normal prothrombin time, activated partial thromboplastin time, and fibrinogen values were used to estimate the normal range for the Factor V (Leiden) Test. The DRVVT ratio was also determined in 21 patients with FVLM (18 patients heterozygous and three patients homozygous for FVLM), in 12 patients deficient solely in PC, in one patient with combined PC deficiency and FVLM, in six patients deficient in PS, in seven patients with combined PS deficiency and FVLM, and in four patients deficient in AT. At the time of investigation none of these patients was on OA therapy. Furthermore, two groups of patients on OA therapy were studied. One group consisted of seven patients heterozygous for FVLM, seven patients deficient solely in PC, four patients with combined PC deficiency and FVLM, four patients deficient in PS, three patients with combined PS deficiency and FVLM, and three patients deficient in AT; the other group of OA-treated patients (n521) had none of the above-mentioned hereditary risk factors for thrombosis and showed international normalized ratio values between 1.7–3.3 (median 2.260.42). All carriers of the FVLM had a ratio below 1.85 measured with Coatest APC Resistance Test (Chromogenix AB, Mo¨lndal, Sweden). All investigated patients deficient for PC (ranges of PC activity: 25–60%) or PS (ranges of free PS antigen: 6–42%) were members of families with at least one person with a defined defect in the corresponding gene. In 10 patients with PC deficiency and eight patients with PS deficiency, the respective gene defects were previously described in detail [15–17]. All patients were in a stable phase without signs of acute illness.
1.2. Blood Sampling Between 8:00 and 10:00 a.m. after overnight fasting, citrated blood (Na citrate, final concentration 0.013 M) for coagulation testing and for DNA analysis
P. Quehenberger et al./Thrombosis Research 96 (1999) 125–133
was taken into siliconized glass tubes (Vacutainer Becton Dickinson, Meylan Cedex, France). Platelet-poor plasma was prepared from citrated blood by centrifugation at 30003g for 20 minutes. Plasma was then assayed immediately or frozen at 2708C.
1.3. Assay Systems Prothrombin time (Thromborel R, Dade-Behring, Marburg, Germany), aPTT (Dade Actin-FS, Baxter Diagnostics Inc, Deerfield, IL, USA), antithrombin activity (STA Antithrombin III, Diagnostica Stago, Asnieres-Sur-Seine, France; normal range: 75–125%), PC activity (COAMATIC Protein C, Chromogenix AB, Mo¨lndal, Sweden; 65– 130%), and resistance against activated PC (Coatest APC Resistance, Chromogenix AB; ratio. 1.85) were determined on the analyzer STA (Stago Diagnostica, Asnieres-Sur-Seine, France); aPTT mixing studies for detection of Lupus anticoagulant were carried out with a 1:1 mixture of patients plasma with pooled normal plasma and considered as positive when the aPTT value from pooled normal plasma was prolonged for more than 6 seconds through the patient’s plasma. Lupus anticoagulant confirmation test STACLOT LA and STACLOT PNP were determined on ST4 coagulation analyzer (Stago Diagnostica). Samples were considered as positive when the clotting time was shortened after incubation with hexagonal phase phosphatidylethanolamine (STACLOT LA) to more than 2 seconds and after incubation with platelet lysate (STACLOT PNP) to more than 0.5 seconds; free PS antigen was determined by enzyme-linked immunosorbent assay (Asserachrom PS, Diagnostica Stago; 50–160%). For direct detection of the factor V gene mutation, genomic DNA was isolated from citrated blood by a standard method. The DNA region containing guanine 1691 was amplified by polymerase chain reaction as described previously [11] with primers described by Bertina et al [3]. The polymerase chain reaction product was digested by the restriction enzyme MnlI and analyzed on polyacrylamide gels.
1.4. Factor V (Leiden) Test The Factor V (Leiden) Test (Gradipore, North Ryde NSW, Australia) was performed fully automated on STA analyzer (Stago Diagnostica, Asnieres-
127
Sur-Seine, France) as follows: 50 mL of plasma were mixed with 50 mL 0.9% NaCl (DRVVT 2) or 50 mL of venom activator reagent (whole venom from Agkistrodon contortrix; DRVVT 1) for activation of endogenous PC, and incubated for 5 minutes at 378C; thereafter 50 mL of diluted phospholipid enriched Russell’s Viper Venom for activation of factor X and subsequent initiation of the coagulation process were added. The result of the assay is expressed as the ratio between DRVVT 1 and DRVVT 2. FVDP (Chromogenix AB, Mo¨lndal, Sweden) was used for predilution (114) of samples. Freezing samples once at 2708C and thawing was found to significantly reduce the DRVVT ratio (immediate testing: ratio of healthy controls 3.826 0.42; ratio determined after once freezing and thawing: 2.8260.45, p,0.05, n510). For the evaluation of the assay only once-frozen and thawed plasma was used. Measurement of PC activity, free PS antigen, and resistance against activated PC using the Coatest APC Resistance Test and the Factor V (Leiden) Test were performed as duplicates. Results of these tests are given as mean values. Sensitivity is defined as true positive results in patients with disease/total patients with disease. Specificity is defined as true negative results in patients without disease/total patients without disease.
2. Results 2.1. Precision Studies Intraassay imprecision of the assay was estimated from 20 determinations of a plasma sample with a ratio within the normal range (A) and a sample with a reduced ratio (B). For sample A, the mean DRVVT ratio was 2.42, standard deviation (SD) was 0.13, and coefficient of variation (CV) was 5.53%. For sample B, the mean DRVVT ratio was 1.15, SD was 0.05, and CV was 4.30%. Interassay imprecision (n510) was calculated from determinations performed on pooled normal plasma (C) and pooled abnormal plasma (D). For sample C, the mean DRVVT ratio was 2.42, SD was 0.17, and CV was 6.90%. For sample D, the mean DRVVT ratio was 1.07, SD was 0.08, and CV was 7.64%.
128
P. Quehenberger et al./Thrombosis Research 96 (1999) 125–133
Fig. 1. The DRVVT ratios obtained with Factor V (Leiden) Test of samples from healthy controls, patients with antithrombin, PC, or PS deficiency, and/or FVLM are given. Samples of homozygous patients for FVLM are given as full squares. A lower cutoff of 2.12 was calculated from the values of healthy controls.
2.2. Normal Range Normal ranges for DRVVT 1, DRVVT 2, and the ratio were determined from samples obtained from 50 healthy volunteers who showed PC, PS, and AT levels within the normal range and were negative for FVLM: mean DRVVT ratio was 2.44, range 2.1–3.52 (5th to 95th percentile 2.12–3.08).
2.3. Effect of PC, PS, AT Deficiency and FVLM on the Factor V (Leiden) Test Fifty-one patients not on OA therapy and with known mutations in PC (n512), PS (n56), AT (n54), factor V gene (n521), or combined mutations (PC/FV Leiden: n51; PS/FV Leiden: n57) were tested with the Factor V (Leiden) Test. Results are shown in Figure 1. All 29 carriers of the FVLM (the combined defects are included in this group) revealed a DRVVT ratio of ,2.12; thus the Factor V (Leiden) Test shows a sensitivity of 1.0 in the diagnosis of the FVLM. In contrast to the aPTT-based assay, the ratios of heterozygous and homozygous patients were not markedly different from each other in the Factor V (Leiden) Test.
Nine of 12 patients with PC deficiency showed a DRVVT ratio of ,2.12 (lower cutoff calculated for nonanticoagulated controls). The three patents with a DRVVT ratio in the normal range had PC activities ranging from 60 to 50%; none of the patients with PS or with AT deficiency showed an abnormal result. The ratios of FVLM-positive individuals were between 0.83 and 1.44; neither healthy controls nor patients deficient in AT, PC, or PS had a ratio of <1.44. Taking a diagnostic ratio of ,1.44 for FVLM, a specificity of 1.0 is thus calculated. Taking the 5th percentile of healthy controls (2.12), the specificity of the Factor V (Leiden) Test in the group of patients deficient in AT or with hereditary defects in the PC/PS pathway was only 0.52. Our data suggest that individuals showing a ratio between 2.12 (lower cutoff of untreated healthy controls) and 1.44 (highest value determined in the group of carriers of the FVLM) are highly suspect for PC deficiency. We also investigated samples from individuals on OA therapy who were free of hereditary thrombosis risk factors. Our results suggest that OA therapy causes a decrease in the ratio: median of untreated controls 2.44 (range 2.1–3.52), median of patients on OA
P. Quehenberger et al./Thrombosis Research 96 (1999) 125–133
129
Fig. 2. DRVVT ratios of samples from healthy controls, from OA-treated patients without any known hereditary risk factors for thrombosis (5OA controls), and from patients with antithrombin, PC, or PS deficiency, and/or FVLM are given. The lower cutoff values of the DRVVT ratio for nonanticoagulated healthy controls (2.12) and for OA controls (1.37) are shown.
therapy 2.25 (1.19–3.94). In the group of OAtreated individuals (21 controls, 28 patients) we found no good discrimination between controls, patients deficient in PC, and carriers of the FVLM, even if a lower 5th percentile specific for OA treatment is calculated (Figure 2). One out of six PSdeficient patients on OA therapy tested abnormal.
2.4. Effect of Plasma Predilution with FVDP on the Factor V (Leiden) Test After predilution of plasma samples (n550) with FVDP, the following normal ranges were obtained: mean DRVVT: ratio 2.11, range 1.95–2.43 (5th to 95th percentile 2.0–2.34). As shown in Figure 3 two of 12 PC-deficient patients showed, even after sample predilution with FVDP, ratios (1.99 and 1.96) below the lower cutoff calculated for FVDP prediluted normal samples (2.0). Samples of AT- and PS-deficient patients showed normal results, whereas all carriers of the FV Leiden mutation had ratios<1.16, values clearly lower than the lowest ratio of healthy controls or PC-deficient patients. Thus predilution
with FVDP leads to an improved discrimination of carriers of the FVLM compared with nonprediluted samples. As shown in Figure 4 predilution of samples from OA-treated individuals revealed the following results: a nearly similar lower cutoff was calculated for OA-treated (1.92) and non-OA-treated (2.0) control individuals. All AT- and PC-deficient patients on OA therapy showed a ratio above 2.0, whereas one patient deficient in PS showed a markedly reduced ratio even after predilution. All carriers of the FVLM had ratios clearly below 1.92. Also in OA-treated patients predilution with FVDP caused an excellent discrimination between carriers of the FVLM (highest ratio 1.09) on the one hand and controls and patients deficient in PC, PS, and AT (lowest ratio in this group 1.67) on the other hand. In one patient deficient in PS, who is negative for the FVLM and shows a normal result in an aPTT-based assay (Coatest APC Resistance, Chromogenix AB), the ratio measured with the Factor V (Leiden) Test was not normalized after sample predilution with FVDP. This particular patient should probably be excluded from the group
130
P. Quehenberger et al./Thrombosis Research 96 (1999) 125–133
Fig. 3. DRVVT ratios of samples prediluted with FVDP from healthy controls, patients with antithrombin, PC, or PS deficiency, and/or FVLM are given. Samples of homozygous patients for FVLM are given as full squares. A lower cutoff of 2.0 was calculated from the values of prediluted healthy controls.
of PS-deficient patients, as he may, in addition to PS deficiency, suffer from a factor V mutation different from factor V Leiden.
2.5. Effect of Lupus Anticoagulant on the Factor V (Leiden) Test Factor V (Leiden) Test was performed on samples of five patients with lupus anticoagulant. Four of them were free of FVLM and one was positive for FVLM. Both in the prediluted and nonprediluted assay procedures the Factor V (Leiden) Test showed results corresponding to the molecular analysis (Table 1).
3. Discussion It was previously shown that functional tests for detection of FVLM based on the Russell Viper Venom method give a good discrimination between carriers and noncarriers of this mutation [18]. Therefore such a DRVVT-based test seems to be a reliable method to screen for FVLM. At
least two DRVVT-based assays for the detection of defects within the PC/PS pathway are available at the moment: the Protein C Pathway Test and the Factor V (Leiden) Test, both from Gradipore (Gradipore, North Ryde NSW, Australia). In a previous study we found that the Protein C Pathway Test, which was conceived as a global estimation of the PC pathway, is sensitive to PC and PS deficiency with an overlap between the values obtained from PC-deficient individuals and carriers of the FVLM (manuscript in preparation). Similar results were obtained with an other global assay system, ProC GLOBAL [19]. Due to low specificity the Protein C Pathway Test assay is not appropriate to screen for FVLM. According to informations obtained from the manufacturers, the newly developed Factor V (Leiden) Test is suitable for screening for FVLM as higher concentrations of the Agkistrodon contortrix venom are used for activation of PC compared with the Protein C Pathway Test assay to make the test more specific. In the present study the Factor V (Leiden) Test was evaluated with respect to its assay characteristics and its sensitivity an specificity in the diagnosis
131
P. Quehenberger et al./Thrombosis Research 96 (1999) 125–133
Fig. 4. DRVVT ratios of samples prediluted with FVDP from healthy controls, from control patients on OA therapy without any known hereditary risk factors for thrombosis (5OA controls), and from patients under OA therapy suffering from antithrombin, PC, or PS deficiency, and/or FVLM are given. Lower DRVVT ratio cutoff values in prediluted samples of 2.0 for nonanticoagulated healthy controls and of 1.92 for OA controls are shown.
of the FVLM. This assay, performed on STA coagulation analyzer, showed satisfying intra- and interassay precision values. Freezing and thawing once of plasma samples significantly influences the test results in lowering the ratio calculated from the DRVVTs in the absence and in the presence of the PC activator; this phenomenon is also observed with the Coatest APC Resistance Test from Chromogenix [11]. Therefore one should test patient samples in the same way as samples of healthy controls from which the reference values and the cutoffs were calculated. In the present study all
samples were frozen and thawed once before determination. Reference values and cutoffs calculated in our study should not be used in general. Due to possible influences of different analyzers on test results each laboratory should define its own reference and diagnostic ranges. We found an excellent sensitivity of 1.0 of the Factor V (Leiden) Test in the diagnosis of the FVLM. While the ratios obtained from patients suffering from PS and AT deficiency were clearly within the normal range, 9 of 12 PC-deficient patients showed abnormally low values. Although
Table 1. DRVVT ratios of five patients with Lupus anticoagulant without and with predilution with FVDP Without predilution with FVDP Patient 1 2 3 4 5 * s5seconds.
FV Leiden
DRVVT 1 (s)*
DRVVT 2 (s)
2 2 2 1 2
102.7 98.4 108.6 63.9 91.7
36.1 38.3 33.6 39.2 31.8
With predilution with FVDP
Ratio
DRVVT 1 (s)
DRVVT 2 (s)
Ratio
2.84 2.57 3.23 1.63 2.88
101.8 97.9 119.3 68.4 96.4
46.5 45.0 44.4 49.9 42.9
2.19 2.18 2.69 1.37 2.25
132
P. Quehenberger et al./Thrombosis Research 96 (1999) 125–133
only a limited number of patients was investigated, we were able to find a good discrimination between PC-deficient individuals and carriers of the FVLM, as the latter had lower ratios and no overlap was detected between these two groups of patients. The discrimination of the carriers of the FVLM could even be increased when the plasma samples were prediluted with FVDP. Calculating a diagnostic range for the FVLM, a specificity of 1.0 of the Factor V (Leiden) Test can be achieved. Our results suggest that samples with ratios between those of carriers of the FVLM and the lower cutoff of normals are highly suspect for PC deficiency. Patients to be tested for FVLM are often under OA therapy. OA therapy causes a lowering of the DRVVT ratio in the evaluated Factor V (Leiden) Test. We investigated two approaches to overcome this phenomenon in order to find an adequate procedure for performing the Factor V (Leiden) Test in OA-treated patients. First we determined the DRVVT ratio in 21 patients under OA therapy who had no abnormality in the PC/PS pathway and thus calculated a lower cutoff specific for OA treatment. This approach did not result in an adequate discrimination between carriers of the FVLM on the one hand and PC-deficient patients and controls on the other hand. In a second approach samples were prediluted with FVDP. As expected, predilution resulted in a clear discrimination between carriers of the FVLM and the other individuals investigated. We believe that, in contrast to the manufacturer’s instructions, patients on OA therapy who show a ratio within the diagnostic range of the FVLM should be retested after predilution with FVDP. Surprisingly we found one OAtreated patient deficient in PS without FVLM and a normal value in an aPTT-based factor V (Leiden) test who showed in repeated tests an abnormally low DRVVT ratio even after predilution with FVDP. This FVLM-negative patient may suffer from a factor V mutation different from the FVLM, which is only detected in a DRVVT- and not in an aPTT-based assay. The values of this patient in the figures are therefore given in parenthesis and should not be considered in the assessment of the sensitivity and the specificity of the Factor V (Leiden) Test; this patient should probably be included into the group defective in the factor V gene. Our results show that the Factor V (Leiden) Test evaluated in the present study is an adequate test
for the diagnosis of defects in the factor V gene that leads to an inadequate inactivation of factor V by activated PC. We believe that this test adequately distinguishes between patients with such a defect and those with deficiencies in the coagulation inhibitors PC and PS, in particular when samples are prediluted with FVDP. Provided that the samples are prediluted with FVDP, the test can also be used in patients treated with OAs. The relatively limited number of patients investigated in the present study, in comparison to similar test evaluations, is explained by the fact that the present investigation was restricted to patients with defined genetic defects and affected family members of these patients. This approach was chosen because it is known that functional tests for coagulation inhibitors show an overlap between affected individuals and normals [20]. Thus using the results of functional tests only, the group of patients would have been less well-defined.
References 1. Walker FJ, Fay PJ. Regulation of blood coagulation by the protein C system. FASEB J 1993;6:2561–7. 2. de Stefano V, Finazzi G, Mannucci PM. Inherited thrombophilia: pathogenesis, clinical syndromes, and management. Blood 1996;87: 3531–4. 3. Bertina RM, Koeleman BPC, Koster T, Rosendaal FR, Dirven RJ, de Ronde H, van der Velden PA, Reitsma PH. Mutation in blood coagulation factor V associated to activated protein C. Nature 1994;369:64–7. 4. Dahlba¨ck B, Carlsson M, Svensson PJ. Familial thrombophilia due to a previously unrecognized mechanism characterized by poor anticoagulant response to activated protein C: Prediction of a cofactor to activated protein C. Proc Natl Acad Sci 1993;90:1004–8. 5. Vandenbroucke JP, Koster T, Briet E, Reitsma PH, Bertina RM, Rosendaal FR. Increased risk of venous thrombosis in oral contraceptive users who are carriers of factor V Leiden mutation. Lancet 1994;344:1453–7. 6. Gouault-Heilmann M, Leroy-Matheron C. Evaluation of a new chronometric assay for factor
P. Quehenberger et al./Thrombosis Research 96 (1999) 125–133
7.
8.
9.
10.
11.
12.
13.
V Leiden-dependent APC resistance. Thromb Res 1997;85:357–62. Le DT, Griffin JH, Greengard JS, Mujumdar V, Rapaport SI. Use of a generally applicable tissue factor-dependent factor V assay to detect activated protein C-resistant factor Va in patients receiving warfarin and in patients with a lupus anticoagulant. Blood 1995;85:1704–11. Arkel YS, Ku DH, Kamiyama M, Pajaro R, Alder H, Marchand A. A highly specific functional test foe factor V Leiden: A modified tissue factor assay for activated protein C resistance. Haemostasis 1997;27:290–304. de Ronde H, Bertina RM. Laboratory diagnosis of APC-resistance: A critical evaluation of the test and the development of diagnostic criteria. Thromb Haemost 1994;72:880–6. Ehrenforth S, Radtke KP, Scharrer I. Acquired activated protein C-resistance in patients with lupus anticoagulant. Thromb Haemost 1995; 74:797–8. Kapiotis S, Quehenberger P, Jilma B, Handler S, Pabinger-Fasching I, Mannhalter C, Speiser W. Improved characteristics of aPC-resistance assay Coatest aPC resistance by predilution of samples with factor V deficient plasma. Am J Clin Pathol 1996;106:588–93. Jorquera JI, Motoro JM, Fernandez MA, Aznar JA, Aznar J. Modified test for activated protein C resistance. Lancet 1994;344:1162–3. Trossaert M, Conard J, Horellou MH, Samama MM, Ireland H, Bayston TA, Lane DA. Modi-
14.
15.
16.
17. 18.
19.
20.
133
fied aPC resistance assay for patients on oral anticoagulated therapy. Lancet 1994;344: 1709. Exner T, Murray B, Chong BH, Chesterman CN. Improved APC-resistance method based on a Russell’s viper venom clotting test. Thromb Haemost 1995;73:1119(Abstract). Poort SR, Pabinger-Fasching I, Mannhalter C, Reitsma PH, Bertina RM. Twelve novel and two recurrent mutations in 14 Austrian families with hereditary protein C deficiency. Blood Coagul Fibrin 1993;4:273–80. Wittmann E, Walter J, Pabinger-Fasching I, Watzke HH. Symptomatic hereditary type-II protein C deficiency caused by a missense mutation in exon IX of the protein C gene (Gly381 to Ser). Ann Hematol 1994;68:255–9. Mustafa S, Pabinger I, Mannhalter C. PS deficiency type I: Identification of point mutations in 9 of 10 families. Blood 1995;86:3444–51. Tripodi A, Negri B, Bertina RM, Mannucci PM. Screening for the FV:Q506 mutation— Evaluation of thirteen plasma-based methods for their diagnostic efficacy in comparison with DNA analysis. Thromb Haemost 1997;77: 436–9. Ruzicka K, Kapiotis S, Quehenberger P, Handler S, Pabinger-Fasching I, Mannhalter C, Jilma B, Speiser W. Evaluation of a new screening assay ProC GLOBAL for identification of defects in the protein C/PS anticoagulant pathway. Thromb Res 1997;87:501–10. Miletich J, Sherman L, Broze G Jr. Absence of thrombosis in subjects with heterozygous protein C deficiency. N Engl J Med 1987; 317:991–6.