Thrombosis Research 111 (2003) 29–32

Regular Article

Does the anti-h2-glycoprotein I antibody provide additional information

in patients with thrombosis?

Eun Young Lee, Chang-Keun Lee, Tae Hoon Lee, Son Mi Chung, Seong Ho Kim,You Sook Cho, Bin Yoo*, Hee-Bom Moon

Division of Allergy and Rheumatology, Department of Internal Medicine, University of Ulsan College of Medicine,

Asan Medical Center, 388-1 Pungnap-Dong, Songpa-Gu, Seoul 138-736, South Korea

Received 2 June 2003; received in revised form 30 July 2003; accepted 1 August 2003

Abstract

Objective: To investigate whether the anti-h2-glycoprotein I (anti-h2GPI) antibody may provide additional information in patients with

thrombosis in conjunction with the lupus anticoagulant (LAC) or anticardiolipin (aCL) antibody. Methods: We selected 235 patients whose

plasma were tested for the presence of all three antiphospholipid (aPL) antibodies (LAC, aCL, and anti-h2GPI) and were positive for at least one

aPL antibody from January 2000 to December 2001. The LAC test was performed using dilute activated thromboplastin time reagent (dAPTT)

and dilute Russell viper venom time reagent (dRVVT). ACL (IgG/IgM) and anti-h2GPI (IgG/IgM) were detected by enzyme-linked

immunosorbent assay (ELISA). Clinical data were collected and analysed in all patients with aPL antibody. Results: Of the 235 patients with

aPL, thrombosis was detected in 76 patients (28.0%). Of the 76 patients with thrombosis, 29 were positive for LAC, 9 for aCL, 7 for anti-h2GPI,

3 for LAC+ aCL, 9 for aCL + anti-h2GPI, 11 for LAC+ anti-h2GPI, and 8 for LAC+ aCL+ anti-h2GPI. The rate of thrombosis was

significantly different ( p = 0.01) among single positive patients (45/163, 27.6%), double positive patients (23/60, 38.3%), and triple positive

patients (8/12, 66.7%). In single positive patients, the rate of thrombosis was highest in LAC positive patients (29/85, 34.1%). In double positive

patients, the LAC+ anti-h2GPI positive group (11/24, 45.8%) and aCL+ anti-h2GPI positive group (9/22, 40.9%) had higher rates of

thrombosis than the LAC+ aCL positive group (3/14, 21.4%). Conclusion: Single positivity for anti-h2GPI explained 9.2% of thrombotic

events in the absence of LAC or aCL. Double or triple positivity for aPLs were associated with a higher rate of thrombosis than single positivity

for aPL. Our results suggest that anti-h2GPI provides additional information in patients with thrombosis in conjunction with LAC or aCL.

D 2003 Elsevier Ltd. All rights reserved.

Keywords: Antiphospholipid antibody; Anti-h2-glycoprotein I antibody; Thrombosis

1. Introduction apart) is needed [1]. However, recent advances have shown

Antiphospholipid (aPL) antibodies are a family of auto-

antibodies that exhibit a broad range of target specificities

and affinities, all recognizing various combinations of phos-

pholipids, phospholipid-binding proteins, or both. It origi-

nally included the anticardiolipin (aCL) antibody, detected

by enzyme-linked immunosorbant assay (ELISA), and the

lupus anticoagulant (LAC), detected by clotting tests. These

antibodies are associated with the clinical manifestations of

venous or arterial thrombosis and recurrent fetal losses. To

diagnose antiphospholipid syndrome (APS), the association

of one clinical criterion with one biological criterion (aCL

and/or LAC on two or more occasions at least six weeks

0049-3848/$ - see front matter D 2003 Elsevier Ltd. All rights reserved.

doi:10.1016/j.thromres.2003.08.023

* Corresponding author. Tel.: +82-2-3010-3282; fax: +82-2-3010-6969.

E-mail address: byoo@amc.seoul.kr (B. Yoo).

that most pathogenic antiphospholipid antibodies are direct-

ed to phospholipid-binding proteins such as h2-glycoprotein

I (h2GPI) [2,3] and prothrombin [4].

The aim of our study was to (i) investigate whether the

anti-h2GPI antibody may provide additional information in

patients with thrombosis in the absence the LAC or aCL

antibody, and (ii) determine the clinical significance of the

double or triple positivity of aPL antibodies in thrombosis.

2. Patients and methods

2.1. Patients

We consecutively selected 527 patients whose plasma

were tested for all three aPL antibodies (LAC, aCL anti-

Table 1

Site of thrombosis among single, double, or triple aPL positive patients

Artery Vein Both Fetal loss

Single positive patients 21 13 9 2

Double positive patients 13 6 3 1

Triple positive patients 3 4 1

Total 37 23 13 3

Values are absolute number of subjects in each group.

Statistical analysis was performed by using v2 test ( P= 0.69).

E.Y. Lee et al. / Thrombosis Research 111 (2003) 29–3230

body, and anti-h2GPI antibody) at Asan Medical Center

from January 2000 to December 2001. APL antibodies were

checked in these patients under the suspicion of aPL

associated disease such as systemic lupus erythematosus

(SLE), APS, avascular necrosis (AVN), stroke in young age,

pulmonary hypertension, thrombosis, fetal loss, etc. Clinical

data were collected from 235 patients who were positive for

at least one aPL antibody: positive LAC test and/or raised

IgG (z 12 U GPL/ml)/IgM (z 6 U MPL/ml) aCL anti-

bodies and/or raised IgG (z 20 SMU)/IgM (z 20 SMU)

anti-h2GPI antibodies.

Patients with the aPL antibody included those with SLE

(N = 52), primary APS (N = 34), AVN (N = 27), malignancy

(N = 21), atherosclerotic cerebrovascular disease (N = 21),

infectious disease (N = 10), deep vein thrombosis (N = 8),

primary pulmonary hypertension (N = 5), and others

(N = 57). All SLE patients fulfilled four or more criteria of

the American College of Rheumatology for the classifica-

tion of SLE [5]. Diagnosis of APS was made according to

preliminary criteria for the classification of APS [1]. Throm-

bosis was confirmed by venography, arteriography, Doppler

ultrasonography, or magnetic resonance (MR) angiography

in each patient.

Table 2

Antibody profiles in patients with the aPL antibody with or without

thrombosis

Thrombosis

(N= 76)

No thrombosis

(N= 159)

Single positive LAC 29 (38.2%) 56 (35.2%)

patients aCL 9 (11.8%) 26 (16.4%)

anti-h2GPI 7 (9.2%) 36 (22.6%)

Double positive LAC+ aCL 3 (3.9%) 11 (6.9%)

patients LAC+ anti-h2GPI 11 (14.5%) 13 (8.2%)

aCL+ anti-h2GPI 9 (11.8%) 13 (8.2%)

Triple positive

patients

LAC+ aCL+

anti-h2GPI

8 (10.5%) 4 (2.5%)

3. Methods

All three aPL antibody tests were performed simulta-

neously in each patient. As recommended by the Interna-

tional Society of Thrombosis and Hemostasis (ISTH), the

screening tests for LAC were performed using dilute acti-

vated thromboplastin time (dAPTT) reagent (Diagnostica

Stago, France) and dilute Russell viper venom time

(dRVVT) reagent (Gradipore, North Ryde, Australia) [6].

To confirm the presence of LAC, dRVVT confirm test

(Gradipore) was performed. ACL antibodies (IgG/IgM)

were detected by enzyme-linked immunosorbent assay

(Pharmacia Diagnostics, Freiburg, Germany). Results were

expressed in IgG and IgM units, and reported as positive for

raised IgG (z 12 U GPL/ml)/IgM (z 6 U MPL/ml) aCL

antibodies. Anti-h2GPI antibodies (IgG/IgM) were detected

by enzyme-linked immunosorbent assay (INOVA Diagnos-

tics, San Diego, CA, USA). Results were expressed in

standard h2GPI Ig units (SMU) and reported as positive

for raised IgG (z 20 SMU)/IgM (z 20 SMU) anti-h2GPI

antibodies.

Single aPL positivity was determined when only one aPL

test was positive, double positivity was determined when

two aPL tests (LAC+ aCL or LAC+ anti-h2GPI or anti-

h2GPI + aCL) were positive, and triple positivity was deter-

mined when all three aPL tests (LAC+ aCL + anti-h2GPI)

were positive.

Plasma levels of protein C and protein S were measured

by enzyme immunoassay (EIA) using monoclonal antibody

(Diagnostica Stago) in aPL antibody positive patients to

exclude protein C or protein S deficiency. The activity of

antithrombin III was also measured by colorimetric assay

using commercial kit (Diagnostica Stago).

4. Statistical analysis

Rate of thrombosis in patient groups was compared using

v2 test or logistic regression analysis. Comparisons among

single, double, and triple aPL antibody positive groups were

expressed as an odds ratio with its 95% confidence interval.

P values < 0.05 were considered statistically significant.

5. Results

Of the 235 patients with aPL antibodies, thrombosis was

detected in 76 patients (28.0%). According to the site of

thrombosis, 37 patients had arterial thrombosis (48.7%), 23

had venous thrombosis (30.3%), 13 had arterial and venous

thrombosis (17.1%), and 3 had fetal loss (3.9%). Table 1

shows the relationships of site of thrombosis and the

positive rate of thrombosis. There were no differences in

site of thrombosis among single, double, or triple aPL

positive patients (P= 0.69). According to the number of

total thrombotic events, 51 patients had one episode of

thrombosis (67.1%), 18 patients had two episodes of throm-

bosis (23.7%), 4 patients had three episodes of thrombosis

(5.3%), and 3 patients had four episodes of thrombosis

(3.9%). There was no correlation between the number of

thrombotic events and the positive rate of aPL antibodies

(r =� 0.10, P= 0.39). Plasma levels of protein C or protein

Fig. 1. The rate of thrombosis in single aPL antibody positive patients.

Table 3

The rate of thrombosis in aPL positive patients according to the positive

rate of the aPL antibody

N +Thrombosis �Thrombosis OR (95% CI) P

aPL positivity

Single 163 45 (27.6%) 118 (72.4%)

Double 60 23 (38.3%) 37 (61.7%) 1.63 (0.87–3.04) 0.25

Triple 12 8 (66.7%) 4 (33.3%) 5.24 (1.50–18.28) 0.009

The relative risk based on the single aPL positive group is shown.

OR= odds ratio; 95% CI = 95% confidence interval.

E.Y. Lee et al. / Thrombosis Research 111 (2003) 29–32 31

S and plasma antithrombin III activity in patients with

thrombosis were within normal range.

Table 2 shows antibody profiles of single, double, and

triple positive groups in patients with or without thrombosis.

Of the 76 patients with the aPL antibody and thrombosis, 45

were single aPL antibody positive (59.2%), 23 were double

aPL antibody positive (30.3%), and 8 were triple aPL

antibody positive (10.5%). Of the 76 patients with throm-

bosis, 7 patients were anti-h2GPI only positive (9.2%). Of

the 159 patients with the aPL antibody but no evidence of

thrombosis, 118 were single aPL antibody positive (74.2%),

37 were double aPL antibody positive (23.3%), and 4 were

triple aPL antibody positive (2.5%).

In single aPL positive patients with or without thrombo-

sis, LAC was most frequently detected (51.5%) and the rate

of thrombosis was highest in the LAC positive group (Fig.

1). In double aPL positive patients, the rate of thrombosis

was highest in the LAC+ anti-h2GPI group (45.8%) and

lowest in the LAC+ aCL group (21.4%) (Fig. 2).

Table 3 shows differences in the rate of thrombosis in

patients according to the positive rate of the aPL antibody.

The rate of thrombosis was higher in triple positive patients

than in single positive patients (odds ratio 5.2, 95% CI 1.5–

18.3; P= 0.009). In double positive patients, rate of throm-

bosis was higher than in single positive patients, but the

Fig. 2. The rate of thrombosis in double aPL antibody positive patients.

difference was not statistically significant (odds ratio 1.6,

95% CI 0.9–3.0; P= 0.125).

6. Discussion

The detection of anti-h2GPI in autoimmune disease has

been the topic of several reports. Its detection in patients

with SLE has been associated with an increased frequency

of a history of thrombosis [3]. In patients with primary APS,

IgG anti-h2GPI that appear to differ from aCL were detected

[2]. These studies were focused on the detection of anti-

h2GPI which recognize cardiolipin and/or its cofactor

h2GPI in patients’ plasma. Although there have been several

reports on the presence of the anti-h2GPI antibody in APS,

international classification criteria for definite antiphospho-

lipid syndrome did not include the test for the anti-h2GPI

antibody because of lack of standardization [1]. The aim of

our study was to investigate whether the conventional anti-

h2GPI antibody test may provide additional information in

patients with thrombosis in conjunction with the LAC or

aCL antibody test.

According to our study, the rate of thrombosis was highest

in LAC positive patients among single aPL positive groups.

This result was consistent with a previous report in which the

LAC test was a more specific test than the aCL test in APS

[7]. Although the rate of thrombosis was lower in anti-h2GPI

positive patients than in aCL or LAC positive patients, single

positivity for the anti-h2GPI antibody explained 9.2% (7/76)

of thrombotic events in the absence of the LAC or aCL

antibody. This result shows the need of the anti-h2GPI

antibody test in patients with thrombosis even in an aCL

or LAC negative state. Previous reports about the anti-h2GPI

antibody test were limited to the detection of the anti-h2GPI

antibody in patients who were diagnosed APS and/or had the

aCL antibody [2,8]. In our study, seven patients with only the

anti-h2GPI antibody and thrombosis included two patients

with SLE, two patients with deep vein thrombosis, two

patients of mesenteric vein thrombosis, and one patient with

recurrent lacunar infarction. If we had added the anti-h2GPI

antibody test to the laboratory criteria for APS, more APS

patients would have been diagnosed.

Another important finding in our study was the relation-

ship between aPL antibody positivity and the rate of

thrombosis. The rate of thrombosis in double or triple aPL

E.Y. Lee et al. / Thrombosis Research 111 (2003) 29–3232

antibody positive patients was higher than in single aPL

antibody positive patients. Moreover, LAC + anti-h2GPI

patients or aCL+ anti-h2GPI patients had a higher rate of

thrombosis than LAC+ aCL positive patients in the double

aPL antibody positive group (45.8%, 40.9% vs. 21.4%).

These results may increase the need for all three aPL

antibody tests, including the anti-h2GPI antibody test, in

patients who have thrombosis or thrombotic risk.

Similar to some previous studies, aPL antibody positive

patients in this study included patients with primary APS,

SLE, malignancy, primary pulmonary hypertension, and

avascular necrosis [9,10]. The rate of thrombosis was 28%

of all aPL positive patients in our retrospective study, which

was much lower than a previous study in SLE patients [11].

This difference may be associated with two factors: First, our

study cohort included SLE and nonSLE patients. Second, we

added the anti-h2GPI test to the aPL antibody test which was

not included in the previous study. Whether aPL antibody

positive but nonthrombotic patients would have new throm-

botic events in the future should be pursued by further study.

The anti-h2GPI test that we used in this study was

standardized enzyme-linked immunosorbent assays (ELI-

SAs) for anti-h2GPI of the two major immunoglobulin

classes:IgG and IgM. Lewis et al. [12] confirmed the

precision of this test in APS patients and demonstrated the

correlation with other methods. They measured IgG, IgA,

and IgM anti-h2GPI antibodies in APS patients and healthy

controls and demonstrated that APS patients’ anti-h2GPI

levels were significantly different from those of the control

population. We did not measure IgA anti-h2GPI, but the

percentage of increased IgA anti-h2GPI in APS patients was

lower than that of increased IgG and IgM anti-h2GPI with

this method [12].

One of the limitations of this study is that we did not

check other antibodies against phospholipid-binding plasma

proteins such as prothrombin, protein C, protein S, and

annexin V. We measured plasma levels of protein C and

protein S which were within normal range in our patients

with thrombosis. These antibodies against phospholipid-

binding plasma proteins were frequently observed in patients

with SLE in a recent study [13]. Especially, anti-h2GPI and

antiprothrombin antibodies were significant risk factors for

arterial thrombosis, and the presence of antiprotein S anti-

bodies was a significant risk factor for venous thrombosis.

Whether the heterogeneity of these antiphospholipid anti-

bodies may explain the various clinical manifestations in

patients with definite APS or even in seronegative APS will

need to be confirmed through further studies.

Another limitation is the retrospective design of this

study, similarly to the majority of studies on this topic: the

aPL antibodies were measured some time after the throm-

botic episode. This is important because these antibodies

fluctuate over the time. Therefore, it is difficult to assess the

real value of a single determination of aPL some time after

the thrombotic event. However, it would be very interesting

to see how many patients with anti-h2GPI antibody but

without thrombosis at the time of aPL test may develop

thrombosis in the future.

In summary, single positivity for anti-h2GPI explained

9.2% of the thrombotic events in the absence of the LAC or

aCL antibody. Double or triple positivity for aPL antibodies

were associated with a higher rate of thrombosis than single

positivity for the aPL antibody. Our results suggest that anti-

h2GPI antibody provides additional information in patients

with thrombosis in conjunction with the LAC or aCL

antibody.

References

[1] Wilson WA, Gharavi AE, Koike T, Lockshin MD, Branch DW, Piette

JC, et al. International consensus statement on preliminary classifica-

tion criteria for definite antiphospholipid syndrome: report of an in-

ternational workshop. Arthritis Rheum 1999;42(7):1309–11.

[2] Cabral AR, Cabiedes J, Alarcon-Segovia D. Antibodies to phospho-

lipid-free beta 2-glycoprotein-I in patients with primary antiphospho-

lipid syndrome. J Rheumatol 1995;22(10):1894–8.

[3] Tsutsumi A, Matsuura E, Ichikawa K, Fujisaku A, Mukai M, Kobaya-

shi S, et al. Antibodies to beta 2-glycoprotein I and clinical manifes-

tations in patients with systemic lupus erythematosus. Arthritis

Rheum 1996;39(9):1466–74.

[4] Atsumi T, Ieko M, Bertolaccini ML, Ichikawa K, Tsutsumi A, Mat-

suura E, et al. Association of autoantibodies against the phosphatidyl-

serine–prothrombin complex with manifestations of the antiphospho-

lipid syndrome and with the presence of lupus anticoagulant. Arthritis

Rheum 2000;43(9):1982–93.

[5] Hochberg MC. Updating the American College of Rheumatology

revised criteria for the classification of systemic lupus erythematosus

[letter]. Arthritis Rheum 1997;40:1725.

[6] Brandt JT, Barna LK, Triplett DA. Laboratory identification of lupus

anticoagulants: results of the Second International Workshop for Iden-

tification of Lupus Anticoagulants. On behalf of the Subcommittee on

Lupus Anticoagulants/Antiphospholipid Antibodies of the ISTH.

Thromb Haemost 1995;74(6):1597–603.

[7] Amigo MC, Khamashta MA. Antiphospholipid (Hughes) syndrome

in systemic lupus erythematosus. Rheum Dis Clin North Am 2000;

26(2):331–48.

[8] Audrain MA, El-Kouri D, Hamidou MA, Mioche L, Ibara A, Lan-

glois ML, et al. Value of autoantibodies to beta(2)-glycoprotein 1 in

the diagnosis of antiphospholipid syndrome. Rheumatology (Oxford)

2002;41(5):550–3.

[9] Schved JF, Dupuy-Fons C, Biron C, Quere I, Janbon C. A prospective

epidemiological study on the occurrence of antiphospholipid anti-

body: the Montpellier Antiphospholipid (MAP) Study. Haemostasis

1994;24(3):175–82.

[10] Tsutsumi A, Ichikawa K, Matsuura E, Sawada KI, Koike T. Hetero-

geneous behavior of anti-beta2-glycoprotein I antibodies on various

commercially available enzyme immunoassay plates coated with be-

ta2-glycoprotein I. J Rheumatol 2000;27(2):391–6.

[11] Petri M. Thrombosis and systemic lupus erythematosus: the Hopkins

Lupus Cohort perspective. Scand J Rheumatol 1996;25(4):191–3.

[12] Lewis S, Keil LB, Binder WL, DeBari VA. Standardized measure-

ment of major immunoglobulin class (IgG, IgA, and IgM) antibodies

to beta2glycoprotein I in patients with antiphospholipid syndrome.

J Clin Lab Anal 1998;12(5):293–7.

[13] Nojima J, Kuratsune H, Suehisa E, Futsukaichi Y, Yamanishi H,

Machii T, et al. Association between the prevalence of antibodies to

beta(2)-glycoprotein I, prothrombin, protein C, protein S, and annexin

V in patients with systemic lupus erythematosus and thrombotic and

thrombocytopenic complications. Clin Chem 2001;47(6):1008–15.