Plasma-Activated Factor VII Level in Patients Positivefor Lupus Anticoagulant

Tutomu Nakase, 1 Hideo Wada, 1* Yoshihiro Wakita, 1 Toshihiro Kaneko, 1 Hiroshi Deguchi, 1

Katuyo Hiyoyama, 1 Minori Shimura, 1 Yositaka Mori, 2 and Hiroshi Shiku 1

1Second Department of Internal Medicine, Mie University School of Medicine, Tsu-city, Japan2Mie Red Cross Blood Center, Tsu-city, Japan

We examined plasma levels of activated factor VII (F VIIa) in 50 patients positive for lupusanticoagulant (LA), in 83 patients negative for LA, and in 10 healthy volunteers as con-trols. Plasma F VIIa was present in healthy volunteers; its level was significantly in-creased, compared to the level in the controls, in patients with thrombosis, collagendiseases, and disseminated intravascular coagulation (DIC), suggesting that it reflecteda thrombotic state. Plasma F VIIa was correlated with thrombin–antithrombin complex(TAT) in patients negative for LA but showed no such correlation in those positive for LA.Plasma F VIIa was negatively correlated with activated partial thromboplastin time (APTT)in patients positive for LA, but not in those negative for LA, suggesting that LA couldinhibit the F VIIa assay system. Plasma F VIIa level was significantly increased in patientswith thrombotic diseases; however, in patients positive for LA, it is possible that in-creased plasma F VIIa level may not be correlated with thrombogenicity. Am. J. Hematol.55:9–14, 1997. © 1997 Wiley-Liss, Inc.

Key words: lupus anticoagulant; plasma activated factor VII; thrombosis; disseminatedintravascular coagulation

INTRODUCTION

High levels of factor VII (FVII) coagulant activity(FVIIc) are associated with an increased risk of coronaryheart disease [1] and with atherosclerotic diseases such ashyperlipidemia [2] and peripheral vascular disease [3].FVII, the first enzyme in the extrinsic pathway of bloodcoagulation, is a serine protease zymogen that is con-verted by limited proteolysis into FVIIa, the enzymati-cally active form [4,5]. Intravenous administration ofFVIIa in animals and humans has shown that its half-lifeis extremely long for an activated plasma protease [6].The measurement of plasma FVIIa levels would there-fore be desirable in patients with thrombotic diseases.FVIIa exhibits very little activity unless it is complexedwith tissue factor (TF), its essential protein cofactor [7].In the presence of TF, FVIIa activates factor X, and FVIIis also autoactivated to FVIIa. A mutant soluble TF, inwhich the transmembrane and cytoplasmic domains weredeleted, was recently developed [8]. This TF retains thecofactor activity toward FVIIa but abolishes the autoac-

tivation of FVII. Plasma FVIIa levels have been mea-sured with this mutant TF in patients with thrombosis [9].

Lupus anticoagulants (LA) were initially reported asassociated with systemic lupus erythematosus (SLE)[10]. However, they have now been identified in a num-ber of other pathological conditions, as well as in appar-ently normal subjects [11]. In recent years, it has becomeclear that the presence of LA is a risk factor for throm-bosis [12,13], for recurrent spontaneous abortion [14],for thrombocytopenia, and for a variety of neurologicaldisorders. LA are immunoglobulins that interfere with avariety of coagulation assays, prolonging phospholipid-dependent coagulation [15–17]. We measured the plasma

Contract grant sponsor: Ministry of Education, Science, and Culture,Japan.

*Correspondence to: Hideo Wada, Second Department of InternalMedicine, Mie University School of Medicine, 2-174 Edobashi, Tsu-city, Mie-ken 514, Japan.

Received 4 March 1996; Accepted 11 December 1996.

American Journal of Hematology 55:9–14 (1997)

© 1997 Wiley-Liss, Inc.

levels of FVIIa in three groups of subjects, patients posi-tive and negative for LA, and healthy controls.

MATERIALS AND METHODS

We examined 50 patients positive for LA, 83 patientsnegative for LA, and 10 healthy volunteers (normal con-trols). Seven of the 50 LA-positive patients had throm-bosis: 4 deep vein thrombosis (DVT), 1 acute myocardial

embolism (AMI), 1 cerebral thrombosis, and 1 portalvein thrombosis. Of the 73 LA-negative patients, 14 hadthrombosis (5 cerebral thrombosis, 4 DVT, 2 AMI, 2pulmonary embolism, 1 portal vein thrombosis), 12 hadcollagen diseases (6 SLE, 4 progressive systemic sclero-sis, 2 Sjogren syndrome), 23 had disseminated intravas-cular coagulation (DIC), 10 had malignant diseases as-sociated without DIC (non-DIC), and 14 had idiopathicthrombocytopenic purpura (ITP) (Table I). These throm-botic disorders were diagnosed on the basis of clinicalsymptoms, laboratory data, electrocardiogram (ECG),angiography, venography, or computed tomography(CT). None of the patients was being treated with anti-coagulants at the time of blood sampling. LA was con-sidered positive if at least two of the following threecriteria were fulfilled: (1) prolongation of activated par-tial thromboplastin time (APTT); (2) prolongation of ka-olin clotting time (KCT) and mixing test in KCT (KCT–MIX); and/or (3) prolongation of DRVVT and increaseddilute Russell’s viper venom time (DRVVT)/DRVVTwith high lipid concentration (DRVVT-HLC) ratio.

Fig. 1. Plasma factor VIIa levels in patients with variousdiseases. *** P < 0.001, **P < 0.01.

Fig. 2. Plasma factor VIIa levels in patients with dissemi-nated intravascular coagulation. *** P < 0.001, **P < 0.01.

TABLE I. Subjects

Group n

LA-positive 50Thrombosis (−) 43Thrombosis (+) 7

LA-negative 83Thrombosis (+) 14Collagen disease

(thrombosis is 4) 12DIC 23non-DIC 10ITP 14Healthy volunteers 10

LA, lupus anticoagulant; DIC, disseminated intravascular coagulation;ITP, idiopathic thrombocytopenic purpura.

10 Nakase et al.

Blood was collected by venipuncture, and 9 parts ofwhole blood were mixed with one part of 3.8% trisodiumcitrate. APTT was determined by Langdell’s method[18]. Actin (Dade) was used as partial thromboplastin.Prothrombin time (PT) was determined by the one-stagemethod described by Quick [19]. Simplastin (WarnerLambert) was used as tissue thromboplastin. KCT wasexamined as follows: 0.2 ml of plasma was incubatedwith 0.1 ml of kaolin suspension (20 mg/ml), and clottingtime was measured after the addition of 0.1 ml of 1/40 MCaCl2. For KCT–MIX, we measured KCT in a mixtureof 0.1 ml patient plasma and 0.1 ml normal plasma. Morethan 71.1 sec for KCT–MIX was regarded as abnormalprolongation. DRVVT and DRVVT–HLC were mea-sured with the DVVtestR (American Diagnostica) andDVVconfirmR (American Diagnostica), respectively. Aratio of more than 1.2 was taken as abnormal. Antithrom-bin activity and protein C activity were measured byamidolytic assay, using Berichrom-Protein C (Behring-werke AG) and Berichrom AT III (Behringwerke AG),respectively. Protein C and protein S antigens were mea-sured by enzyme-linked immunosorbent assay (ELISA),using the respective polyclonal antibodies (Dakopatts).

Thrombin–antithrombin complex (TAT), plasmin–plasmin inhibitor complex (PPIC), and fibrin-D-dimerwere determined with Enzygnost-TAT (BehringwerkeAG), PIC test (Teijin), and Frelisa D-dimer (Agen), re-spectively. Plasma F VIIa was measured with StaclotVIIa–rTF (Stago), using recombinant soluble tissue fac-tor that has specific cofactor function for F VIIa and doesnot activate FVII to FVIIa.

Values are shown as means ±SD, and statistical analy-sis was performed with Student’st-test. P values of<0.05 were considered to be significant.

RESULTS

Plasma FVIIa level in healthy volunteers was 53 ± 12mU/ml. It was significantly increased in the patients withcollagen diseases, those with DIC, and those with throm-bosis, both those who were LA-negative and those whowere LA-positive, and in the LA-positive patients, butnot in patients with ITP or in the non-DIC patients (whowere all LA-negative). Plasma F VIIa level in patientswith DIC was significantly higher than in those withoutDIC (non-DIC) (Figs. 1, 2). In both LA-positive and

TABLE II. Hemostatic Abnormalities in Patients With and Without LA

LA-positive LA-negative Healthyvolunteersth(+) th(−) Collagen ITP

Factor VIIa (mU/ml) 189 ± 196 89 ± 69 173 ± 86 59 ± 18 53 ± 12APTT (sec) 37.1 ± 9.4 35.0 ± 6.5 26.7 ± 2.1 27.8 ± 3.1 30.9 ± 2.4PT (sec) 12.7 ± 1.3 11.9 ± 1.1 11.3 ± 0.5 11.9 ± 0.6 11.8 ± 0.6DRVVT (sec) 73.8 ± 10.4 62.0 ± 14.3 35.8 ± 16.5 28.1 ± 4.7 31.1 ± 3.4KCT (sec) 124.8 ± 47.4 121.0 ± 46.1 76.1 ± 33.2 70.6 ± 20.2 63.1 ± 4.0TAT (ng/ml) 6.4 ± 5.0 3.1 ± 1.7 8.7 ± 7.9 2.5 ± 0.9 1.4 ± 0.4PPIC (mg/ml) 1.5 ± 1.3 0.8 ± 0.4 1.2 ± 0.5 0.8 ± 0.2 0.6 ± 0.2D-Dimer (mg/ml) 91.5 ± 76.9 50.4 ± 41.4 115.6 ± 88.3 39.2 ± 15.7 23.5 ± 10.8Antithrombin activity (%) 85.1 ± 17.7 109.3 ± 24.3 101.3 ± 30.3 101.4 ± 9.6 101.4 ± 15.6PC activity (%) 78.1 ± 12.8 83.8 ± 19.3 114.2 ± 31.8 116.2 ± 29.7 95.2 ± 13.3PS antigen (%) 87.5 ± 18.3 96.6 ± 21.5 80.0 ± 24.2 114.7 ± 33.6 101.3 ± 13.8

LA, lupus anticoagulants; th(+), th(−), thrombosis-positive and thrombosis-negative; ITP, idiopathic thrombocytopenic purpura; APTT, activated partialthromboplastin time; PT, prothrombin time; DRVTT, dilute Russell’s viper venom time; KCT, kaolin clotting time; TAT, thrombin–antithrombincomplex; PPIC, plasma–plasmin inhibitor complex; PC, protein C; PS, protein S.

TABLE III. Hemostatic Abnormalities in LA-Negative Patients in Relation to DIC

ThrombosisLA-negative

DIC Non-DICHealthy

volunteers

Factor VIIa (mU/ml) 115 ± 46 140 ± 64 68 ± 32 53 ± 12APTT (sec) 32.3 ± 5.5 33.5 ± 7.3 36.0 ± 12.7 30.9 ± 2.4PT (sec) 11.9 ± 0.7 12.9 ± 1.9 12.9 ± 2.8 11.8 ± 0.6TAT (ng/ml) 7.7 ± 10.2 50.8 ± 60.9 10.9 ± 8.0 1.4 ± 0.4PPIC (mg/ml) 1.0 ± 0.7 3.6 ± 3.8 1.0 ± 0.4 0.6 ± 0.2D-Dimer (mg/ml) 108.4 ± 111.2 772.2 ± 659.7 294.4 ± 457.5 23.5 ± 10.8Antithrombin activity (%) 100.5 ± 23.1 83.3 ± 24.3 81.1 ± 18.8 101.4 ± 15.6PC activity (%) 98.9 ± 27.1 82.5 ± 37.3 78.0 ± 37.2 95.2 ± 13.3PS antigen (%) 106.4 ± 27.9 87.7 ± 28.2 102.0 ± 34.3 101.3 ± 13.8

DIC, disseminated intravascular coagulation; see Table II for all other abbreviations.

FVIIa Does Not Increase in LA Significantly 11

LA-negative patients with thrombosis, plasma F VIIa,TAT, PPIC, and D-dimer levels were markedly increasedrelative to controls; plasma F VIIa was not increased inLA-negative patients with ITP and in the non-DIC pa-tients relative to the healthy volunteers (controls). In LA-positive patients without thrombosis, plasma F VIIa wasmarkedly increased, but plasma TAT level was not.

Plasma FVIIa level in patients with collagen diseaseswas significantly increased, and plasma TAT and D-dimer levels were slightly increased compared to con-trols (Tables II, III). In all patients, plasma F VIIa waspoorly correlated with PT, TAT, PPIC, D-dimer, anti-thrombin, protein C, and protein S; however, it was nega-tively correlated with APTT (Fig. 3A; r4 −0.254,P <0.01). In the LA-negative patients, plasma F VIIa wascorrelated only with TAT (r4 0.288,P < 0.05), and itwas not correlated with APTT (Fig. 3B). In patients posi-tive for LA, plasma F VIIa was correlated with PPIC (r4 0.468, P < 0.001), PT (r4 0.315, P < 0.05), andAPTT (Fig. 3C: r4 −0.394,P < 0.01) (Table IV).

DISCUSSION

Plasma F VIIa is present in healthy subjects becausethere is no effective plasma inhibitor of free F VIIa [20];furthermore, as stated above, the intravenous administra-tion of F VIIa in animals or humans has shown that it hasan extremely long half-life for an activated plasma pro-tease [6]. Compared to the level in the healthy volunteersin our study, plasma F VIIa level was significantly in-creased in patients with collagen diseases, those withDIC, those with thrombosis, and the LA-positive patientswho had thrombi, suggesting that F VIIa may reflectthrombosis. Plasma F VIIa level in the patients with DICwas significantly higher than in those without DIC, in-dicating that the increase of plasma F VIIa, similarly tothe increase of TAT, PPIC, and D-dimer, could dependon the degree of disseminated coagulation. Plasma FVIIa, TAT, PPIC, and D-dimer levels were markedlyincreased in the patients with thrombosis; we thereforetook F VIIa to be an indicator of thrombosis. However, inthe LA-positive patients without thrombosis, plasma FVIIa was also markedly increased, but plasma TAT levelwas not. Reduced plasma protein C and increased plasmaactivated protein C–protein C inhibitor complex havebeen reported previously in patients with LA [21] and thepresence of LA is considered a risk factor for thrombosis[12,13]. As these LA-positive patients are considered toexhibit a hypercoagulable state, it is conceivable thatplasma F VIIa could be more sensitive than TAT for thediagnosis of such a state. Elevated plasma F VII coagu-lant activity has been reported to be a significant predic-tor of risk for ischemic heart disease and cardiovasculardeath [22,23]. However, the conventional F VII coagu-lant assay measures the aggregate of both F VII and F

VIIa levels in plasma, whereas our results suggest thatthe measurement of F VIIa is important for the diagnosisof thrombotic disease. Plasma TAT, PPIC, and fibrin-D-dimer levels have been reported to be important markers

Fig. 3. A: Relationship between factor VIIa and APTT in allpatients open circles, patients with thrombosis. B: Relation-ship between factor VIIa and APTT in patients negative forLA. Open circles, patients with thrombosis. C: Relationshipbetween factor VIIa and APTT in patients positive for LA.Open circles, patients with thrombosis.

12 Nakase et al.

for DIC [24], pre-DIC [25], and hyperlipidemia [26]; it isalso possible, in the light of our findings, that F VIIacould be an important indicator of these diseases. In allpatients, plasma F VIIa was poorly correlated with PT,TAT, PPIC, D-dimer, antithrombin, protein C, and pro-tein S but was negatively correlated with APTT. In theLA-negative patients, plasma F VIIa was correlated onlywith TAT and was not correlated with APTT. In patientspositive for LA, plasma F VIIa was correlated with PPICand was negatively correlated with APTT, suggestingthat LA could inhibit the F VIIa assay system. The re-combinant TF used in this assay activates F VIIa, but notF VII, and it has been reported that F VIIa levels were notwell correlated with F VII coagulant activity [9]. This FVIIa assay is measured by clotting assay. Determinationof the presence of LA has been performed by variousclotting time methods, the sensitivities of which dependon the amount or nature of the phospholipids involved[27,28]. LA is an anticoagulant inhibitor-dependentphospholipid that inhibits various coagulation tests, sug-gesting that it could inhibit clot formation in the F VIIaassay. In patients positive for LA, increased F VIIa levelswere not detected in this assay, probably due to the in-terference of LA. In patients positive for LA, we shouldevaluate the risk of thrombosis, not only with F VIIa butwith other hemostatic markers as well.

ACKNOWLEDGMENTS

This work was supported, in part, by a Grant-In-Aidfor Cancer Research from the Ministry of Education,Science, and Culture, Japan.

REFERENCES

1. Meade TW: Hypercoagulability and ischemic heart disease. Blood Rev1:2–8, 1987.

2. Carvalho de Sausa J, Bruckert E, Giral P, Soria C, Truffert J, MirshahiMC, de Gennes JL, Caen JP: Plasma factor VII, triglyceride concen-tration and fibrin degradation products in primary hyperlipidemia: Aclinical and laboratory study. Haemostasis 19:83–90, 1989.

3. Orland M, Leri O, Maciocc G, Mattia G, Ferri GM: Factor VII insubjects at risk for thromboembolism: Activation or increased synthe-sis? Haemostasis 17:340–343, 1987.

4. Fair DS: Quantitation of factor VII in the plasma of normal and war-farin-treated individuals by radioimmunoassay. Blood 62:784–791,1983.

5. Hagen FS, Gray GL, O’Hara P, Grant FJ, Saari GC, Woodbury RG,Hart CE, Insley M, Kisiel W, Kurachi K, Davie EW: Characterizationof a cDNA coding for human factor VII. Proc Natl Acad Sci USA83:2412–2416, 1986.

6. Seligsohn U, Kasper CK, Osterud B, Rapaport SI: Activated factorVII: Presence in factor IX concentrate and persistence in the circula-tion after infusion. Blood 53:828–837, 1978.

7. Nemerson Y: Tissue factor and hemostasis. Blood 71:1–8, 1988.

8. Neuenschwander PF, Morrissey JH: Deletion of the membrane an-choring Region of tissue factor abolishes autoactivation of factor VIIbut not cofactor function. J Biol Chem 267:14477–14482, 1992.

9. Morrissey JH, Macik BG, Neuenschwander PF, Comp PC: Quantita-tion of activated factor VII levels in plasma using a tissue factormutant selectively deficient in promoting factor VII activation. Blood81:734–744, 1993.

10. Conley CL, Hartmann RC: A hemorrhagic disorder caused by circu-lating anticoagulant in patients with disseminated lupus erythemato-sus. J Clin Invest 31:621–622, 1952.

11. Schleider MA, Nachman RL, Jaffe EA, Coleman MA: A clinical studyof the lupus anticoagulant. Blood 48:499–509, 1976.

12. Lechner K, Pabinger-Fasching I: Lupus anticoagulants and thrombo-sis. A study of 25 cases and review of the literature. Haemostasis215:293–298, 1985.

13. Feinstein DI: Lupus anticoagulant, thrombosis and fetal loss. N Engl JMed 313:1348–1350, 1985.

14. Lubbe WF, Liggins GC: Lupus anticoagulant and pregnancy. Am JObstet Gynecol 153:322–327, 1985.

15. Thiagarajan P, Shapiro SS, De Marco L: Monoclonal immunoglobulinM coagulation inhibitor with phospholipid specificity; mechanism of alupus anticoagulant. J Clin Invest 66:379–405, 1980.

16. Exner T, Triplett DA, Taberner DA, Howard MA, Harris EN: Com-parison of test methods for the lupus anticoagulant: International sur-vey of lupus anticoagulants-1(ISLA-1). Thromb Haemost 64:478–484,1990.

17. Thiagarajan P, Pengo V, Shapiro SS: The use of the dilute Russell’sviper venom time for the diagnosis of lupus anticoagulants. Blood68:869–874, 1986.

18. Langdell PD, Wagner RH, Brinkhous KM: Effect of antihemophilicfactor on one-stage clotting time. J Lab Clin Med 47:637–647, 1953.

19. Quick AJ, Brown MS, Bancroft FW: A study of the coagulation defectin the hemophilia and in jaundice. Am J Med Sci 190:501–511, 1935.

20. Kondo S, Kisiel W: Regulation of factor VIIa in plasma: Evidence thatantithrombin III is the sole plasma protease inhibitor of human factorVIIa. Thromb Res 46:325, 1987.

21. Nakase T, Wada H, Minamikawa K, Wakita Y, Shimura M, HiyoyamaK, Tamaki S, Shirakawa S, Deguchi K, Nishioka J, Hayashi T, SuzukiK: Increased activated protein C–protein C inhibitor complex level inpatients positive for lupus anticoagulant. Blood Coagul Fibrinolysis5:173–177, 1994.

22. Meade TW, Brozovic M, Chakrabarti RR, North WRS, Haines AP,Stirling Y, Imenson JD, Thompson SG: Haemostatic function andischaemic heart disease: Principal results of the Northwick Park heartstudy. Lancet 2:533–537, 1986.

23. Meade TW, North WRS, Chakrabarti R, Stirling Y, Haines AP,Thompson SG, Brozovic M: Haemostatic function and cardiovasculardeath: Early results of a prospective study. Lancet 1:1050–1053, 1980.

24. Wada H, Minamikawa K, Wakita Y, Nakase T, Kaneko T, Ohiwa M,

TABLE IV. Relationship Between factor VIIa and HemostaticMarkers in Patients Positive and Negative for LA

All LA-positive LA-negative

APTT −0.254** −0.394** −0.117PT −0.188 −0.315* −0.130TAT 0.196 0.117 0.288*PPIC 0.182 0.468*** 0.166D-Dimer 0.026 −0.154 0.022Antithrombin activity 0.074 0.041 0.167PC activity 0.146 0.011 0.226PS antigen −0.101 −0.040 −0.160

*P < 0.05, **P < 0.01, ***P < 0.001. See Table II for abbreviations.

FVIIa Does Not Increase in LA Significantly 13

Tamaki S, Deguchi K, Shirakawa S, Hayashi T, Suzuki K: Increasedvascular endothelial cell markers in patients with disseminated intra-vascular coagulation. Am J Hematol 44:85–88, 1993.

25. Wada H, Minamikawa K, Wakita Y, Nakase T, Kaneko T, Ohiwa M,Tamaki S, Deguchi A, Mori Y, Deguchi K, Shirakawa S, Suzuki K:Hemostatic study before onset of disseminated intravascular coagula-tion. Am J Hematol 43:190–194, 1993.

26. Wada H, Mori Y, Kaneko T, Wakita Y, Nakase T, Minamikawa K,Ohiwa M, Tamaki S, Tanigawa M, Kageyama S, Deguchi K, Nakano

T, Shirakawa S, Suzuki K: Elevated plasma levels of vascular endo-thelial cell markers in patients with hypercholesterolemia. Am J He-matol 44:112–116, 1993.

27. Harris EN, Loizou S, Englert H, Derue G, Chan JK, Gharavi AE,Hughes GRV: Anticardiolipin antibodies and lupus anticoagulant.Lancet 2:1099, 1984.

28. Lesperance B, David M, Rauch J, Infante-Rivard C, Rivard GE: Rela-tive sensitivity of different tests in the detection of low titer lupusanticoagulants. Thromb Haemost 60:217–219, 1988.

14 Nakase et al.