Clinical Significance of Various ELISA Assays for Detecting Antiphospholipid Antibodies

We have compared the prevalence of antiphospholipid antibodies (APA) measured by enzyme-linked immunosorbent assay (ELISA), in 119 selected patients using five different antigens: bovine cardiolipin, phosphatidylserine, phosphatidylinositol, bovine partial thromboplastin and human brain partial thromboplastin. All the plasmas have been evaluated for the presence of lupus anticoagulant (LA) activity by clotting techniques. We found a significant association between the incidence of LA and APA (p <0.001), only moderate agreement between the prolongation of the activated partial thromboplastin time (APTT) and ELISAs (r around 0.50) and a good agreement between the ELISAs (r around 0.80). The combination of antibodies against cardiolipin (ACA) and human brain partial thromboplastin (AHPTA) allowed the detection of antibodies in most of the LA positive cases. ACA, AHPTA and antiphosphatidylinositol antibodies detected all the positive samples. Six patients (5%) had a single APA detected. The clinical associations of APA according to phospholipid specificity, immunoglobulin isotype and titer are shown.

Partial Neutralization of a Lupus Anticoagulant by Human Immunoglobulin

In a patient with a Lupus Anticoagulant (LA) and recurrent fetal loss, we observed a significant shortening of the APTT after high-dose intravenous immunoglobulin infusion (IVIg).

The LA activity present in patient’s serum and purified IgG was partially neutralized by IVIg in a dose-dependent way. In addition, IgG purified from IVIg and its F(ab′)2 fragment neutralized LA activity of the patient’s IgG. In both cases, the neutralization was dose-dependent and it was obtained with similar molar ratios

The “in vitro” neutralization of LA activity and the immediate shortening of the APTT after IVIg infusion, might be mediated through idiotypel antiidiotype interactions.

On the other hand, the long-lasting effect of IVIg in this patient indicates that it may induce specific inhibition of autoantibody synthesis.

We believe that IVIg should be considered as a therapeutic alternative for LA-related clinical disorders.

“Lupus” Anticoagulant and Thrombosis – Possible Role of Inhibition of Prostacyclin Formation

A “lupus” anticoagulant was discovered in 14 patients over a one year period. Only three of them had systemic lupus erythematosus. Bleeding manifestations were only present in one patient with concomitant severe thrombocytopenia. In contrast, eight patients had a history of thrombosis; five of them presented repeated thrombotic episodes. Obstetrical complications (recurrent abortion, fetal death, or intrauterine growth retardation) were observed in six patients. An inhibitory effect of plasma on the production of prostacyclin by vascular tissue was detected in eight patients, six of whom had thrombosis. We suggest that inhibition of prostacyclin formation could play a major role in the pathogenesis of thrombosis and obstetrical problems in some patients with this type of anticoagulant, even in the absence of systemic lupus erythematosus.

Reactivity to β2 Glycoprotein I Clearly Differentiates Anticardiolipin Antibodies from Antiphospholipid Syndrome and Syphilis

Anticardiolipin antibodies (aCL) detected by standard ELISA are found in association with autoimmune and infectious diseases. It is now recognized that P2 glycoprotein I (β2GPI) is a cofactor for aCL binding to cardiolipin (CL). To examine differences in cofactor requirements, aCL positive sera from patients with the antiphospholipid syndrome (APS) and syphilis were studied. Using an ELISA with human purified P2GPI adsorbed onto irradiated plates, we detected high binding activity in 29 out of 35 samples from APS patients and low in only 1 out of 37 aCL positive syphilis sera. Moreover, a good correlation (r = 0.79) was also observed in the former group between aCL and anti β2GPI. Whole IgG and affinity purified IgG aCL from APS patients did not bind to CL in the absence of β2GPI, but recognized β2GPI on irradiated plates in the absence of phospholipids. In contrast, IgG purified from syphilis patients only bound to CL alone. Taken together, these data indicate that performing both ELISA (aCL and anti β2GPI) it could be possible to distinguish aCL from autoimmune or infectious diseases.

Antiphospholipid Antibodies Enhance Thrombin-Induced Platelet Activation and Thromboxane Formation

In a group of 6 patients with lupus anticoagulant (LA) and antiphospholipid (aPL) antibodies detected by ELISA ovemight urine and blood were simultaneously collected. A significantly increased urinary excretion of the platelet-derived thromboxane (TX) metabolite 11-dehydro-TXB2 was found in this group, as compared to 12 healthy individuals. In contrast, a small but significant reduction of the vascular prostacyclin (PGI2) metabolite 2,3-dinor-6-keto-prostaglandin F1α was observed.

To further elucidate the effect of these antibodies on platelet activation we isolated the F(ab’)2 fragments from IgG of the 6 patients and 5 Controls, and we evaluated the effect of these fragments on the responses of isolated normal platelets to thrombin.

Patients’ F(ab’)2 increased platelet aggregation and serotonin release of platelets stimulated by low dose thrombin (0.01 U/ml). At threshold thrombin concentration (0.05 U/ml) an enhanced TXB2 production was also observed.

In summary, our results show, in addition to the altered TXA2/PGI2 balance observed in vivo, a direct stimulatory effect of aPL antibodies on platelet activation in vitro. This effect is related to recognition of phospholipid epitopes on platelets as shown by its neutralization upon preincubation with phospholipids. This phenomenon may be relevant for the thrombotic tendency of these patients.

Attempts to make sense of the antiphospholipid syndrome

Many investigators have been intrigued by the paradoxical association of a circulating anticoagulant, first called lupus anticoagulant by Feinstein and Rapaport [1], with a tendency to develop thrombosis, as initially described by Walter Bowie [2]. Work in Leuven on this topic started when Luis Carreras, an Argentinian hematologist, joined the laboratory of blood coagulation at this university in 1979. At that time, the head of the laboratory was Marc Verstraete. Luis had a particular interest in antibody-mediated coagulation disorders, and had prepared reviews on thrombosis and thrombocytopenia induced by heparin [3] and on the lupus inhibitor [4]. In Leuven, he joined Jos Vermylen, senior member of the laboratory, and an internist with particular interest in hemostasis, thrombosis and vascular disease. As such, Professor Vermylen was involved in both laboratory research and patient care.

Link between anti-CD36 antibodies and thrombosis in the antiphospholipid syndrome

OBJECTIVE

Some studies have previously suggested the involvement of antibodies directed against CD36 (anti-CD36) in the pathogenesis of thrombosis. The aim of this study was to evaluate the prevalence of anti-CD36 in patients with antiphospholipid antibodies (aPL) and its relationship with thrombosis.

METHODS

Anti-CD36 were tested using an indirect MAIPA assay in 62 patients with autoimmune aPL but without SLE; there were 38 with and 24 without thrombosis. Nineteen patients with thrombosis served as an aPL(-) control group and 58 healthy subjects as the normal control group.

RESULTS

15 of 62 aPL patients (24.2%) but only 1 of 58 (1.7%) normal controls had anti-CD36 (p < 0.0005). As compared to normal controls, the prevalence of anti-CD36 was significantly higher in aPL patients with (26.3%, p < 0.0005) or without thrombosis (20.8%, p < 0.01). Anti-CD36 were significantly more frequent in aPL patients with thrombosis than in thrombosis aPL(-) subjects (26.3% vs 0%, p = 0.02). The presence of anti-CD36 seems to be more frequent in aPL patients with recurrent thrombosis than in those with a single episode (36.8% vs 15.8%).

CONCLUSION

The presence of anti-CD36 is highly prevalent in patients with autoimmune aPL with a trend to being more frequent in patients with recurrent episodes of thrombosis.

Markers of platelet, endothelial cell and blood coagulation activation in leprosy patients with antiphospholipid antibodies

OBJECTIVE

To evaluate plasma levels of markers of platelet, endothelial cell and blood coagulation activation in leprosy patients with or without antiphospholipid antibodies (aPL) and to compare them to those found in patients with antiphospholipid syndrome (APS).

METHODS

42 patients with leprosy (35 lepromatous and 7 borderline): 29 aPL(+) and 13 aPL(-), as well as 26 healthy subjects as normal controls (NC) and 79 control aPL patients without leprosy (59 with and 20 without APS) were included in the study. Plasma soluble P and E selectin (sPsel and sEsel), and VCAM-1 (sVCAM-1), prothrombin F1 + 2 fragment (F1 + 2), thrombin-antithrombin complexes (TAT) and D dimer (DD) were measured by ELISA. The protein C pathway was assessed by the ProC global test.

RESULTS

Leprosy patients with aPL presented increased median levels of sPsel [ng/ml (82.0 vs 36.0, p < 0.001)] and sVCAM-1 [ng/ml (495 vs 335, p < 0.001)] compared to NC, as observed in control aPL patients without leprosy. Levels of sPsel in aPL(+) patients with leprosy were significantly higher than in aPL(-) ones (52.5 ng/ml), p = 0.005. However, plasma markers of thrombin generation were increased in control aPL patients without leprosy but not in those with leprosy. ProcC global test was abnormal in 24.1% of leprosy patients with aPL compared to 4.4% of NC (p < 0.024), and to 57.2% of control patients with aPL without leprosy (p = 0.005).

CONCLUSIONS

We demonstrated that although patients with leprosy present a high prevalence of aPL, and platelet and endothelial cell activation in vivo to the same extent than patients with APS, they do not show a procoagulant state.

Major and potential prothrombotic genotypes in a cohort of patients with venous thromboembolism

Factor V Leiden (FVL) and the prothrombin 20210A (PT-20210A) variant are well-known risk factors for venous thromboembolism (VT). The thermolabile variant (TT) of the methylenetetrahydrofolate reductase (MTHFR) gene, and homozygosity for the 4G allele of the promoter region of the plasminogen activator inhibitor-1 (PAI-1) are potential genetic polymorphisms that have not been consistently associated with increased risk of VT. A case-control study was performed on 192 consecutive unrelated patients referred for evaluation of thrombophilia because of VT and 200 healthy controls. FVL was found in 10.4% of patients compared to 3.0% of controls, while 6.3% of patients were carriers of the PT-20210A allele compared to 2.0% of controls. The adjusted odds ratios (OR) were 5.92 and 4.03 for FVL (P=.001) and the PT-20210A (P=.033), respectively. The prevalence of homozygotes for MTHFR (TT) and PAI-1 (4G/4G) among patients and controls were 13.7% versus 13.0% and 21.6% versus 23.5%, respectively (P=ns). A total of 121 patients underwent a complete screening for FVL, the PT-20210A, protein C (PC), protein S (PS), antithrombin III (ATIII), levels of factor VIII, and antiphospholipid antibodies (aPL). In 59 patients (48.8%) at least one defect was found, being a single defect in 55 and combined defects in 4 patients. Plasma levels of homocysteine (Hcy) were measured in 138 patients and 144 controls. Subjects from both groups carrying the MTHFR-TT variant had higher Hcy levels than those with the normal genotype. Hyperhomocysteinemia (HHcy) by itself is a risk factor for VT (OR 4.92, P<.0001). We conclude that FVL and the PT-20210A are risk factors for VT as well as Hcy levels, but the MTHFR and PAI-1 polymorphisms do not appear to be associated with VT in our country.

Thrombophilic factors in chronic thromboembolic pulmonary hypertension

Chronic thromboembolic pulmonary hypertension (CTE-PH) is an infrequent cause of pulmonary hypertension that develops in 0.1-0.2% of patients who survive after an acute venous thromboembolic event. According to the largest series so far reported, 15-30% of patients with diagnosis of CTE-PH have an underlying congenital or acquired hypercoagulable state. To determine the prevalence of thrombophilic factors in our population, we analyzed 24 patients admitted to our institution between November 1992 and March 2000 fulfilling criteria for CTE-PH. Eighteen patients disclosed abnormal results in the screening for thrombophilia. The presence of antiphospholipid antibodies (lupus anticoagulant and/or anticardiolipin antibodies) was the abnormality most frequently found (12 out of 24 patients). We found hyperhomocysteinaemia in 7/14, true protein S deficiency in 1/10, protein C deficiency in 1/13, activated protein C resistance in 1/22, antithrombin III deficiency in 1/24, and prothrombin gene G20210A mutation in 1/18 patients. Factor V Leiden was normal in all 18 patients studied. Five patients (20.8%) disclosed more than one thrombophilic abnormality. In conclusion, contrary to the largest series of patients with CTE-PH so far reported, we found that 75% of patients with CTE-PH presented at least one thrombophilic risk factor, being antiphospholipid antibodies in 50% of the cases. We recommend a thorough screening for thrombophilia in all patients with diagnosis of CTE-PH.

Increased lipid peroxidation correlates with platelet activation but not with markers of endothelial cell and blood coagulation activation in patients with antiphospholipid antibodies

Recent studies have shown that patients with antiphospholipid antibodies (aPL) have increased lipid peroxidation. We evaluated the urinary excretion of 11-dehydro thromboxane B2 (11-DH-TXB(2) and isoprostane F(2alpha)III (IPF(2alpha)III), reflecting platelet activation and lipid peroxidation in vivo, and plasma soluble markers of endothelial cell, platelet and blood coagulation activation: soluble vascular cell adhesion molecule-1 (sVCAM-1), P- and E-selectin (sPsel and sEsel), F1 + 2 fragment of prothrombin (F1 + 2), thrombin-antithrombin complexes (TAT) and D-Dimer (DD). We studied 79 patients with aPL (47 with previous thrombosis), 45 healthy volunteers (normal controls, NC), 12 patients with systemic lupus erythematosus (SLE) without aPL and a thrombosis control group (TCG) without thrombophilia (n = 16). Urinary levels (mean, range) of eicosanoids and isoeicosanoids were significantly increased in 39 patients with aPL compared with 25 NC, 11-DH-TXB(2) 164.0 ng/mmol creatinine (9.5-1162.8) versus 43.4 ng/mmol creatinine (4.2-87.6), P < 0.001; IPF(2alpha)III 56.9 pg/mg creatinine (5.5-388.7) versus 27.0 pg/mg creatinine (4.6-87.6), P = 0.03. Both metabolites were significantly correlated (rho = 0.49, P = 0.014), but none correlated with any clinical manifestation or antibody profile. The aPL group presented increased levels of sPsel, sEsel, sVCAM-1, TAT, F1 + 2 and DD, but any soluble marker correlated with IPF2alphaIII. Urinary 11-DH-TXB(2) correlated with sPsel (rho = 0.39, P = 0.04). Compared with SLE controls, the SLE group with aPL had higher levels of F1 + 2. Plasma levels of F1 + 2 and DD were significantly increased and a trend to higher sPsel was found in aPL patients with thrombosis compared with the TCG. Platelet activation, lipid peroxidation and blood coagulation activation seem to be important in the pathophysiology of antiphospholipid syndrome.

The combination of thrombophilic genotypes is associated with definite antiphospholipid syndrome

BACKGROUND AND OBJECTIVES

Thrombosis and pregnancy morbidity are clinical features of the definite antiphospholipid syndrome (APS). These clinical complications are also associated with the presence of inherited thrombophilias. Interactions between acquired and genetic risk factors are becoming increasingly related to a higher thrombotic risk. The aim of our study was to determine the prevalence of four common gene polymorphisms in patients with antiphospholipid antibodies (aPL).

DESIGN AND METHODS

A series of 105 consecutive unselected patients with aPL grouped as having APS (n= 69) and not having APS (n= 36) was studied. A control group of 200 healthy subjects was also investigated for the presence of factor V Leiden (FVL), the 20210A allele of the prothrombin (PT-20210A) gene, the thermolabile variant (677TT) of methylenetetrahydrofolate reductase (MTHFR), and the 4G/4G genotype of the plasminogen activator inhibitor (PAI-1) promoter.

RESULTS

Two patients who belong to the APS group carried the FVL while PT-20210A was found in 6 patients with APS (8.7%) and in 1 of the non-APS group (2.8%). The prevalence of FVL was similar to that found in the control group whereas PT-20210A was significantly more frequent in APS patients than in normal controls (2.0%, p=0.02). The MTHFR-677TT was found in 22.0%, 15.1% and 13.0%, and the PAI-1 (4G/4G) in 27.5%, 22.8% and 23.5% of APS, non-APS patients and normal controls, respectively. Furthermore, combinations of PT-20210A or FVL with PAI-1 (4G/4G) were significantly more frequent in APS patients (5.8%) than in normal controls (0.5%, p=0.016). This difference was not found between non-APS patients and normal subjects.

INTERPRETATION AND CONCLUSIONS

Present data indicate that testing for heritable thrombophilia would be important to identify aPL subjects with an increased risk of developing APS.

High prevalence of antiphospholipid antibodies in leprosy: evaluation of antigen reactivity

Antiphospholipid antibodies (aPL) have been reported not only in autoimmune disorders but also in various infectious diseases. Accumulating evidence indicates that beta2 glycoprotein I (beta2GPI) and prothrombin are the main proteins to which autoimmune aPL bind. The aim of this study was to evaluate the prevalence of different aPL in patients with leprosy. We included 51 outpatients (42 lepromatous and 9 borderline leprosy) without any clinical feature of the antiphospholipid syndrome (APS). 35 had lupus anticoagulant and 31 had anticardiolipin antibodies (aCL). Anti-beta2GPI antibodies were highly positive in 29/51 and anti- prothrombin antibodies (anti-II) were detected in 23/51. Almost all aCL and anti-beta2GPI were of IgM isotype, while IgG isotype was more frequent among anti-II. No statistical difference was found when aPL were evaluated in patients grouped according to their bacteriological status. Furthermore, patients under treatment (n=33) had a similar frequency of positive aPL compared to patients in vigilance (n=14). Assessing the specificity of antibody binding to CL and beta2GPI in ELISA by means of inhibition studies with cardiolipin-beta2GPI liposomes, leprosy and APS sera showed a similar behaviour. Comparable results were also found in both groups of patients when inhibition experiments with lysate of Mycobacterium leprae were carried out. In summary, leprosy-related aPL resemble those found in patients with APS but the immunoglobulin isotype is different, with IgM much more prevalent in leprosy patients.

Which are the best biological markers of the antiphospholipid syndrome?

The diagnosis of antiphospholipid syndrome (APS) requires the presence of both clinical and biological features. Due to the heterogeneity of anti-phospholipid antibodies (aPL) the laboratory approach for their detection includes clotting-based tests for lupus anticoagulant (LA) as well as solid-phase assays for anticardiolipin antibodies (aCL). In addition, as it has been shown that autoimmune aPL recognize epitopes on phospholipid (PL)-binding plasma proteins, assays detecting antibodies to beta 2-glycoprotein I (beta 2-GPI) or prothrombin have been developed. The association between venous or arterial thrombosis and recurrent fetal loss with the presence of conventional aPL (LA and/or aCL) has been confirmed by many studies. The LA and IgG aCL at moderate/high titre seem to exhibit the strongest association with clinical manifestations of the APS. Several reports indicate that LA is less sensitive but more specific than aCL for the APS. Assays against PLs other than CL as well as the use of mixtures of PLs have been proposed to improve the detection of APS-related aPL. Concerning antibodies to PL-binding proteins (detected in the absence of PLs), there is evidence that anti-beta 2-GPI are closely associated with thrombosis and other clinical features of the APS. Moreover, these antibodies may be more specific in the recognition of the APS and in some cases may be present in the absence of aPL detected by standard tests. Many issues are still under debate and are discussed in this review, such as the problems of standardization of anti-beta 2-GPI assays, detection of the IgA isotype of aCL and anti-beta 2-GPI, the coagulation profiles of LA in the recognition of the thrombotic risk and the association of particular markers with subsets of patients with APS.

Occurrence of anti-prothrombin and anti-beta2-glycoprotein I antibodies in patients with history of thrombosis

New evidence indicates that antibodies to beta2-glycoprotein I (anti-beta2GPI) or to human prothrombin (anti-II)(or to both of these) are specific markers of the antiphospholipid syndrome (APS). They have been mainly associated with thrombotic complications in patients with APS. However, some studies have reported that elevated levels of anti-II, but not of anfi-beta2GPI, imply a risk of venous thrombosis (VT) or arterial thrombosis (AT) in subjects with no previous thrombosis and no antiphospholipid antibodies (aPL) by ELISA. The present study Included 180 patients with a history of thrombosis, 83 of them without aPL (group I) and the remaining 97 diagnosed as having APS (group II). Anti-beta2GPI was found in only 1 of the 83 patients from group I but was found in approximately 50% of those from group II (P < .0001). In contrast, positive anti-II was detected with a high prevalence in patients from group I (VT, 22.6%; AT, 26.7%) and in those from group II (VT, 37.5%; AT, 14.6%). No statistical differences were found in the prevalence of anti-II between the two groups of patients. On the other hand, such a difference was significant when compared with results in a normal group (1/67, 1.4%, P < .0001). These data Indicate that anti-II occurs frequently in patients with previous thrombosis either with or without lupus anticoagulant activity. Accordingly, testing of anti-II might be clinically useful in the evaluation for thrombophilla.

Binding properties of antibodies to prothrombin and beta2-glycoprotein I (beta2-GPI) assayed by ELISA and dot blot

Most anti-phospholipid antibodies (aPL) associated with the anti-phospholipid syndrome are autoantibodies with specificity towards beta2-GPI (anti-beta2-GPI) or prothrombin (anti-II). They are mainly screened by ELISA using polyoxygenated plates. However, some authors have claimed that immunoblotting can also be used. Exposure of cryptic epitopes or increase of antigen density on its binding to either phospholipids or suitable plastic surfaces are the two hypotheses proposed for the interaction of beta2-GPI or prothrombin with their antibodies. Forty-five patients with aPL were studied: 25 with lupus anti-coagulant (LA) and anti-cardiolipin antibodies (aCL), 10 with LA alone and 10 with aCL but negative LA. All patients with LA and aCL were positive for anti-beta2-GPI by ELISA and dot blot, while 15/25 had anti-IIELISA and 14 of them also had anti-II by dot blot assay. No patient with LA alone tested positive for anti-beta2-GPI by ELISA or dot blot, whereas 6/10 had anti-IIELISA (five of them were also positive by dot blot). Four out of 10 aCL-positive patients had anti-beta2-GPI by ELISA and dot blot, while none of this group had anti-II by ELISA or dot blot. Antibody binding to beta2-GPI or prothrombin in both ELISA and dot blot was significantly reduced by phospholipid liposomes mixed together with beta2-GPI or prothrombin, whereas liposomal eluants retained it in both assays. Parallel fluid-phase inhibition experiments using increasing concentrations (up to 200 microg/ml) of beta2-GPI or prothrombin demonstrated that antibody binding reduction was more evident on dot blot than on ELISA. It was almost completely abolished on dot blot, while on ELISA a moderate inhibition was achieved even at the highest protein concentration. However, antibody binding on ELISA was virtually abolished when diluted sera were incubated with high protein concentrations applied to nitrocellulose membranes. We could infer that ELISA and dot blot detect antibodies with some differences in avidity but directed against native epitopes on beta2-GPI and prothrombin.

Early occlusion of coronary by-pass associated with the presence of factor V Leiden and the prothrombin 20210A allele: case report

The presence of factor V Leiden mutation and the variation of the prothrombin gene 20210GA have been described as additional risk factors for arterial thrombosis when other acquired or metabolic risk factors are present. We report here a 56-year-old man who developed coronary artery disease since 1980 without any known risk factor and underwent a cardiopulmonary by-pass in 1997. In the first month after surgery, he became symptomatic, and an angiography showed complete occlusion of the grafts and some native coronary arteries. Three months after the second cardiopulmonary by-pass, a thrombophilic state was searched, and plasma levels of lipoprotein (a) (LPa) were measured. The patient is heterozygous for factor V Leiden mutation and has the variation 20210GA of the prothrombin gene and high levels of LPa. These findings induced us to add oral anticoagulation to the aspirin treatment, and the patient is in a good condition 11 months later.

Different types of antiphospholipid antibodies in AIDS: a comparison with syphilis and the antiphospholipid syndrome

Alloimmune antiphospholipid antibodies react with phospholipids and are an epiphenomenon of an infectious disease. Most autoimmune antiphospholipid antibodies recognise phospholipid-protein complexes or proteins, such as beta2 glycoprotein I or prothrombin and are related to the clinical features of the antiphospholipid syndrome. Lupus anticoagulant, anticardiolipin antibodies, antiprothrombin, and anti-beta2 glycoprotein I antibodies were studied in 61 human immunodeficiency virus (HIV) patients, 55 syphilis patients, and 45 selected patients with antiphospholipid syndrome. Lupus anticoagulant was present in 72% of HIV and 81% of antiphospholipid syndrome patients. None of the syphilis patients had lupus anticoagulant. Anticardiolipin antibodies were found at comparable prevalence in the three groups (HIV 67%, syphilis 67%, antiphospholipid syndrome 84%). HIV had more frequently anti-beta2 glycoprotein I (13%) and antiprothrombin (12%) antibodies than syphilis (0 and 4%, respectively), but significantly less than antiphospholipid syndrome (61 and 40%, respectively). Autoimmune antiphospholipid antibodies in HIV without clinical features of antiphospholipid syndrome might be a reflex of the immunological chaos and/or the constant antigenic virus stimulus.

Antiplatelet factor 4--heparin antibodies in patients with antiphospholipid antibodies

Antibodies directed against platelet factor 4-heparin are present in patients with heparin-induced thrombocytopenia (HIT). Additionally, it has been suggested that heparin can be an antigenic target of antiphospholipid antibodies (aPL). We investigated the presence of heparin-platelet factor 4-induced antibodies (HPIA) in 33 patients with aPL. There were 30 patients with lupus anticoagulant, 25 with anticardiolipin antibodies, 21 with anti-beta2 glycoprotein I, and 18 with antiprothrombin antibodies. 20 patients had a history of thrombosis and 19 had received heparin during the last 60 months. We found 7 (21.2%) who had HPIA; 5 of them also had anti-beta2 glycoprotein I antibodies. Four patients had severe thrombocytopenia and suspicion of HIT. Among them, two presented high positive HPIA results, one of them with positive platelet aggregation test. The third patient showed grey zone HPIA and borderline aggregation test and the fourth one had negative results. Among patients without a history of HIT, 2 who had never received heparin presented high positive, one a moderate positive, and one a grey zone HPIA result; all of them with negative aggregation tests. Five positive sera samples were incubated with cardiolipin liposomes in the presence of beta2 glycoprotein I, and whereas an inhibition greater than 50% was achieved in anticardiolipin and anti-beta2 glycoprotein I activities, HPIA results did not change. We demonstrate that HPIA could be frequently found in patients with aPL. They are responsible for HIT in some cases but can also be found in patients who have not received heparin. Whether they predispose patients with aPL to HIT is not known; nevertheless, a close follow-up of heparin treatment in these patients seems to be mandatory.

Lupus anticoagulant, high levels of anticardiolipin, and anti-beta2-glycoprotein I antibodies are associated with chronic thromboembolic pulmonary hypertension

OBJECTIVE

To evaluate the prevalence of lupus anticoagulant (LAC) and anticardiolipin antibodies (aCL), and that of anti-beta2- glycoprotein I (anti-beta2-GPI) and prothrombin antibodies in patients with pulmonary hypertension (PH).

METHODS

Fifty-four consecutive patients with PH were studied: 23 with primary, 20 secondary, and 11 chronic thromboembolic PH. LAC was diagnosed by screening and confirmatory coagulation tests, while aCL, anti-beta2-GPI, and prothrombin antibodies were measured by ELISA.

RESULTS

Prevalence of aPL was higher in patients with chronic thromboembolic PH compared to the other 2 groups. The prevalence in chronic thromboembolic PH vs primary and secondary PH was: LAC 63.6 vs 13.0 and 10.0%, p < 0.001; aCL-IgG 54.5 vs 17.4 and 15.0%, p < 0.02; anti-beta2-GPI-IgG 36.4 vs 0 and 0%, p < 0.001; and prothrombin antibodies-IgG 36.4 vs 8.7 and 5.0%, p < 0.05. No differences between groups were found for any antibody of IgM isotype. Antibodies detected in patients with primary and secondary PH were of low titer, so considering only moderate or high titers these differences were greater for aCL-IgG (odds ratio, OR 24.6, confidence interval, CI 3.0-282, p = 0.0004) and IgM (OR 35.0, CI 2.9-1692, p = 0.0007) and remained significant for anti-beta2-GPI-IgG (OR = undefined, p = 0.006). Multivariate analysis showed that only LAC and aCL-IgG at moderate or high levels were independent variables associated with chronic thromboembolic PH.

CONCLUSION

The presence of LAC, moderate or high levels of aCL-IgG, or anti-beta2-GPI-IgG was strongly associated with that of chronic thromboembolic PH. These data are in agreement with the close relationship observed among these 3 variables and thromboembolism in patients with aPL.

Anti-beta2 glycoprotein I antibodies and platelet activation in patients with antiphospholipid antibodies: association with increased excretion of platelet-derived thromboxane urinary metabolites

Platelet activation may contribute to the increased risk of thrombotic complications in patients with antiphospholipid antibodies (aPL). The increased urinary excretion of 11-dehydro-thromboxane B2 (11-DH-TXB2) reported in patients with lupus anticoagulant (LA) and/or anticardiolipin antibodies (aCL) reflects in vivo platelet activation. However the majority of autoimmune aPL are directed to beta2 glycoprotein I (beta2GPI) or prothrombin (II). We investigated the relationship of these antibodies with 11-DH-TXB2 urinary excretion in 34 patients with aPL. The urinary 11-DH-TXB2 was measured by EIA after extraction on octadecyl columns and purification on silica gel columns, which was validated by thin-layer chromatography/EIA procedure. A significantly increased excretion of 11-DH-TXB2 was found in aPL patients as compared to 18 normal controls (p <0.01). But no differences were seen in the excretion of 11-DH-TXB2 between patients with or without LA, or aCL. The number of patients with anti-II antibodies was too small to draw any conclusion. In contrast, patients with anti-beta2GPI antibodies IgG at moderate/high titre (group A, n = 14) had higher levels of urinary 11-DH-TXB2 than those at low titre or negative (group B, n = 20) (p = 0.01). The group A of patients presented an increase in 11-DH-TXB2 compared to controls (p <0.001), but no statistically significant difference was found between patients from the group B and normal controls. A correlation between levels of urinary 11-DH-TXB2 and titre of antibodies was only found for anti-beta2GPI-IgG (r(s) = 0.51, p <0.005). Our data show that the observed platelet activation in aPL patients is related to the presence of antibodies reacting with beta2GPI.

Relationship of anti beta2-glycoprotein I and anti prothrombin antibodies to thrombosis and pregnancy loss in patients with antiphospholipid antibodies

The lupus anticoagulant (LA) and anticardiolipin antibodies (aCL) are clinically relevant because of their association with thrombosis and pregnancy loss. The group of antiphospholipid antibodies (aPL) includes antibodies primarily directed against various phospholipid-binding proteins, mainly beta2-glycoprotein I (beta2GPI) and prothrombin. Some studies suggest that there is an association between the presence of anti beta2GPI antibodies (alphabeta2GPI) of IgG isotype and thrombosis. Therefore, aPL defined according to the plasma protein to which they are directed appear to be more appropriate for the evaluation of their clinical importance. Using home-made ELISAs we evaluated the presence of alphabeta2GPI and antiprothrombin antibodies (anti-II) of both isotypes (IgG and IgM) in a group of 233 patients with LA and/or aCL. Forty-four women had a history of pregnancy loss, 45 patients had a history of venous thrombosis (VT) and 32 of arterial thrombosis (AT). Patients from the autoimmune group (systemic lupus erythematosus and antiphospholipid syndrome) had a higher prevalence of alphabeta2GPI and/or anti-II than those from the miscellaneous group. In the univariate analysis, a significant association was shown between the presence of alphabeta2GPI-IgG (OR 3.2; 95% CI 1.5-6.6) and previous VT, but not AT. Anti-II were related to VT but the multivariate analysis showed that alphabeta2GPI-IgG are the only independent risk factor for VT (OR 3.0; 95% CI 1.3-6.2). The presence of alphabeta2GPI-IgM correlates well with a history of pregnancy loss (OR 2.6; 95% CI 1.1-6.1). The coagulation tests profile showed that the clotting assays were more prolonged in patients having aCL, alphabeta2GPI or anti-II. But a higher prevalence of abnormal results was only found for the dilute Russell viper venom time in patients with VT, as compared to those without thrombosis (94.4% vs. 58.7%, p <0.02). The measurement of alphabeta2GPI of both isotypes could help to identify aPL-positive patients with a higher risk for thrombosis and pregnancy loss, although this association should be confirmed by prospective studies.

Pregnancy loss and autoantibodies against phospholipid-binding proteins

OBJECTIVE

To evaluate the relationship between antibodies against beta 2-glycoprotein I or prothrombin and pregnancy losses in women with antiphospholipid antibodies.

METHODS

Women with antiphospholipid antibodies, (lupus anticoagulant and/or anticardiolipin antibodies), with (n = 41) and without (n = 61) a history of pregnancy loss were evaluated. Thirty-one out of the forty-one patients with pregnancy loss had early miscarriages (at less than 13 weeks) and ten patients had late miscarriages. Immunoglobulin (Ig)-G and IgM anti-beta 2-glycoprotein I and anti-prothrombin antibodies were measured by an enzyme-linked immunosorbent assay method.

RESULTS

A significant association between pregnancy loss and positive IgM anti-beta 2-glycoprotein I antibodies was found (odds ratio 2.6; 95% confidence interval 1.03, 6.6; P = .043). Women with late pregnancy loss had higher levels of both IgG and IgM anti-beta 2-glycoprotein I antibodies compared with controls (P < .05). There was a good correlation between anticardiolipin and anti-beta 2-glycoprotein I antibodies levels (IgG: r = 0.75; IgM: r = 0.73). In contrast, there was no correlation between the levels of anticardiolipin or anti-beta2-glycoprotein I antibodies and the levels of anti-prothrombin antibodies. Furthermore, the presence of anti-prothrombin antibodies was not associated with a history of pregnancy loss.

CONCLUSION

The result of our study shows that there is a relationship between the presence of IgM anti-beta 2-glycoprotein I and previous miscarriages in women with anti-phospholipid antibodies.

Activated protein C resistance in patients with anti-beta 2 glycoprotein I antibodies

To evaluate if the presence of anti-beta 2GPI antibodies (a beta 2GPI) is associated with activated protein C resistance (APC-R) phenotype, we performed the APC-R APTT-based assay in 74 plasma samples from patients with antiphospholipid antibodies (aPL). Samples were diluted 1:5 in factor V-deficient plasma. Lupus anticoagulant (LA), anticardiolipin antibodies (aCL) and a beta 2GPI (IgG and IgM) were also performed. A control group of 22 healthy volunteers was used. The prevalence of reduced APC-R ratio in patients with aPL was significantly higher than in normal controls (31.1 vs 4.5%, P < 0.05) and the mean APC-R ratio was lower (mean +/- SD; 2.32 +/- 0.40 vs 2.55 +/- 0.21, P < 0.02). There were no differences in the prevalence of APC-R and the ratio values between LA(+) and LA(-). Among the LA(+), the aCL(+) had a higher prevalence of APC-R than the aCL(-) (P < 0.01) and lower APC-R ratios (P < 0.01). The latter group was no different to normal controls. Anti-beta 2GPI antibodies were associated with a higher prevalence of APC-R (50.0 vs 19.6%, P < 0.001), and lower APC-R ratios (2.15 +/- 0.41 vs 2.42 +/- 0.35, P < 0.005), compared with a beta 2GPI(-). In conclusion, the acquired APC-R in patients with aPL seems to be associated with aCL and a beta 2GPI rather than an in vitro interference by LA.

Pathogenic role of antiprotein-phospholipid antibodies

Antiphospholipid antibodies (aPL) are a heterogeneous family of antibodies, including those specific for a variety of phospholipid (PL)-binding proteins and also those reacting with PL molecules. The former seem to be associated with the antiphospholipid syndrome (APS). At present, the main proteins proposed as antigens are beta 2 glycoprotein I, prothrombin, protein C, protein S, kininogens and annexin V. Anionic PL might play a key role "in vivo" in the binding of aPL to PL-bound proteins. Different mechanisms may be involved in the pathogenesis of the APS, including effects of aPL on the protein C system and antithrombin III and also on platelets, endothelial cells and monocytes. Recent data on experimental animal models have provided support for a causative role of aPL in the clinical complications of the APS.

Reactivity to beta 2 glycoprotein I clearly differentiates anticardiolipin antibodies from antiphospholipid syndrome and syphilis

Anticardiolipin antibodies (aCL) detected by standard ELISA are found in association with autoimmune and infectious diseases. It is now recognized that beta 2 glycoprotein I (beta 2GPI) is a cofactor for aCL binding to cardiolipin (CL). To examine differences in cofactor requirements, aCL positive sera from patients with the antiphospholipid syndrome (APS) and syphilis were studied. Using an ELISA with human purified beta 2GPI adsorbed onto irradiated plates, we detected high binding activity in 29 out of 35 samples from APS patients and low in only 1 out of 37 aCL positive syphilis sera. Moreover, a good correlation (r = 0.79) was also observed in the former group between aCL and anti beta 2GPI. Whole IgG and affinity purified IgG aCL from APS patients did not bind to CL in the absence of beta 2GPI, but recognized beta 2GPI on irradiated plates in the absence of phospholipids. In contrast, IgG purified from syphilis patients only bound to CL alone. Taken together, these data indicate that performing both ELISA (aCL and anti beta 2GPI) it could be possible to distinguish aCL from autoimmune or infectious diseases.

Increased expression of inducible cyclooxygenase-2 in human endothelial cells by antiphospholipid antibodies

The effect of IgGs from 4 patients with antiphospholipid antibodies and elevated excretion of urinary 11-dehydro-thromboxane B2 was evaluated on the production of prostacyclin by human endothelial cells in culture. After 6 h incubation, there was no change in 6-keto-prostaglandin F1 alpha in the supernatant. However patients' IgGs induced a marked increase in cyclooxygenase (Cox) activity compared to IgGs from 2 normal individuals or a commercial pool of IgGs from normal donors, tested by adding exogenous arachidonic acid. Western blot analysis of the cellular Cox content using antibodies specific for the different forms of the enzymes revealed that patients' IgGs stimulated the synthesis of the newly described inducible Cox-2 without affecting the constitutive Cox-1. This effect was partially neutralized by preincubating the IgGs with phospholipids. The induction was dependent on the amount of IgGs; it was visible at 2 h and persisted up to 24 h. Analysis of mRNA levels showed a pattern of variation in good agreement with the results obtained for protein. The protein kinase inhibitor H-7 or long-term incubation of cells with PMA strongly reduced the induction. These results suggest that antiphospholipid antibodies may not prevent the potential of the vascular cells from generating higher amounts of prostacyclin in response to acute episodes of thrombosis.

Anti beta 2 glycoprotein I antibodies: detection and association with thrombosis

It has been demonstrated that antiphospholipid antibodies (aPL) recognize epitopes formed by anionic phospholipids and protein cofactors. beta 2 glycoprotein I (beta 2GPI) is accepted as the cofactor of anticardiolipin antibodies (aCL). In the present study we explored the presence and clinical associations of anti beta 2GPI antibodies of IgG isotype (a beta 2GPI-IgG), measured by ELISA. We studied sera from 169 patients with aCL and/or lupus anticoagulant (LA), including 52 patients with systemic lupus erythematosus and 49 with primary antiphospholipid syndrome (PAPS). We found 31.9% positive sera for a beta 2GPI-IgG in the whole population and 48.6% in the aCL-IgG(+) group. There was a good correlation between the titre of aCL-IgG and the optical density for a beta 2GPI-IgG (r = 0.69, P < 0.01). The presence of a beta 2GPI-IgG was associated with the presence of aCL-IgG (P < 0.0001) and LA (P < 0.0005). However, none of 23 LA (+) patients without aCL had a beta 2GPI-IgG. We found a statistically significant association between the presence of a beta 2GPI-IgG and a history of venous thromboembolism (VTE) in our patients (P < 0.005). This association was observed in PAPS (P < 0.05) but not in secondary antiphospholipid syndrome (SAPS). Our study confirms that some aPL(+) sera react with beta 2GPI in special experimental conditions. In addition, the presence of these antibodies is associated with a history of VTE.

Evaluation of recently described tests for detection of the lupus anticoagulant

It is known that lupus anticoagulants (LA) are antibodies which interfere with phospholipid-dependent coagulation tests, but due to the heterogeneity of LA and the differences in sensitivity of reagents and tests, the diagnosis of LA remains difficult. Recently, Triplett et al. (26) have proposed a new test based on two venoms, Textarin (T) and Ecarin (E), that activate prothrombin but differ in their phospholipid requirements. By testing this new assay we have evaluated 36 patient plasmas containing LA according to standard tests (activated partial thromboplastin time, dilute Russell viper venom time and platelet neutralization procedure) and our results confirm a high sensitivity for LA of the T/E test. In addition, we observed a greater sensitivity of the tissue thromboplastin inhibition test using a recombinant thromboplastin instead of a human placenta thromboplastin. Our study also showed that the T/E test seems to be a useful assay in confirming the diagnosis of LA in patients with an unexplained prolonged APTT.

Differences in protein S and C4b-binding protein levels in different groups of patients with antiphospholipid antibodies

Total protein S (tPS), free protein S (fPS) and C4b-binding protein (C4b-BP) were measured by immunological assays in 73 patients with antiphospholipid (aPL) antibodies, in order to determine whether the previously reported abnormalities in PS levels in this group of patients could be related to the presence of lupus anticoagulant (LA) or anticardiolipin (aCL) antibodies. As compared with the normal controls (n = 44), the authors found a significant decrease of tPS, fPS and C4b-BP in 45 LA(+)aCL(+) patients (P < 0.001), a decrease of tPS (P < 0.001), fPS and C4b-BP (P < 0.01) in eight LA(-)aCL(+) patients and a decrease of only fPS (P < 0.05) in 20 LA(+)aCL(-) patients. There was no difference in the levels of tPS, fPS and C4b-BP between LA(+)aCL(+) and LA(-)aCL(+) patients. In contrast, the LA(+)aCL(+) patients had lower values of tPS, fPS and C4b-BP than LA(+)aCL(-) patients (P < 0.05). In some patients, protein S activity (PSact) was also measured and a high correlation was observed between fPS antigen and PSact (r = 0.93, P < 0.001). The data show that the presence of aCL antibodies is associated with a probably acquired deficiency of PS and C4b-BP. On the other hand, in LA patients without a CL antibodies, the fPS deficiency is unrelated to an increase in C4b-BP levels and may be due to abnormal binding of PS to C4b-BP.

Antiphospholipid antibodies enhance thrombin-induced platelet activation and thromboxane formation

In a group of 6 patients with lupus anticoagulant (LA) and antiphospholipid (aPL) antibodies detected by ELISA overnight urine and blood were simultaneously collected. A significantly increased urinary excretion of the platelet-derived thromboxane (TX) metabolite 11-dehydro-TXB2 was found in this group, as compared to 12 healthy individuals. In contrast, a small but significant reduction of the vascular prostacyclin (PGI2) metabolite 2,3-dinor-6-keto-prostaglandin F1 alpha was observed. To further elucidate the effect of these antibodies on platelet activation we isolated the F(ab')2 fragments from IgG of the 6 patients and 5 controls, and we evaluated the effect of these fragments on the responses of isolated normal platelets to thrombin. Patients' F(ab')2 increased platelet aggregation and serotonin release of platelets stimulated by low dose thrombin (0.01 U/ml). At threshold thrombin concentration (0.05 U/ml) an enhanced TXB2 production was also observed. In summary, our results show, in addition to the altered TXA2/PGI2 balance observed in vivo, a direct stimulatory effect of aPL antibodies on platelet activation in vitro. This effect is related to recognition of phospholipid epitopes on platelets as shown by its neutralization upon preincubation with phospholipids. This phenomenon may be relevant for the thrombotic tendency of these patients.

[Antiphospholipid antibodies and cerebral ischemic infarction in a 6-year-old boy]

Lupus anticoagulant activity and anti-phospholipid antibodies (aPL) were found in a six-year-old child with cerebral ischemic infarction in the absence of any underlying disease. The association of these antibodies with thrombosis has been well documented in adult patients. In view of our observation, we believe that aPL may also be involved in the pathogenesis of arterial thrombotic events in childhood, and aPL should be systematically searched in these cases.

Natural inhibitors of blood coagulation and fibrinolysis in patients with lupus anticoagulant

We studied the natural inhibitors (NI) of blood coagulation and fibrinolysis in 50 patients with lupus anticoagulant (LA), in order to identify possible alterations of these NI, that could favour thrombotic manifestations. We found no statistically significant difference in antithrombin III, protein C and alpha 2-antiplasmin between controls and patients with LA, irrespective of their clinical manifestations. We found an increase of plasminogen activator inhibitor (PAI, P < 0.001) and a decrease of free protein S (PSf, P < 0.001) and total protein S (PSt, 0.01 < P < 0.05) in the patients with LA when compared with the control group. We found no difference in the levels of NI between patients with thrombosis (n = 19) and without thrombosis (n = 31) nor between patients with (n = 25) or without thrombosis and/or foetal loss (n = 25). In contrast, we observed a decrease of PSf in women with foetal loss (n = 10) as compared with women without foetal loss (n = 22, 0.01 < P < 0.05) and a decrease of PSf when comparing 19 patients with systemic lupus erythematosus (SLE) with 31 patients without SLE (0.01 < P < 0.05). These findings show that the patients with LA had several abnormalities in the NI system, but there was no significant association between levels of PAI, PSf, PSt and a history of thrombosis.

Increased production of platelet-derived thromboxane in patients with lupus anticoagulant

The in vivo production of thromboxane A2 and prostacyclin was assessed in 31 samples from 25 patients with lupus anticoagulant and in 32 controls. The urinary excretion of 11-dehydro-thromboxane B2 (a major thromboxane metabolite of platelet origin) was very significantly increased (p less than 0.0003) in the patients contrasting with a lesser increase of urinary 2,3-dinor-6-keto-prostaglandin F1 alpha reflecting the vascular production of prostacyclin (p less than 0.02). Our study shows that in patients with lupus anticoagulant, platelet activation may occur without a compensatory increment in the vascular biosynthesis of prostacyclin suggesting an increased risk for thrombosis.

Imbalance of thromboxane/prostacyclin biosynthesis in patients with lupus anticoagulant

The mechanism involved in the association between antiphospholipid antibodies and thrombosis or fetal loss remains unclear. We assessed the biosynthesis of thromboxane A2 and prostacyclin in 31 samples from 25 patients with lupus anticoagulant and in 32 controls. The urinary excretion of the major thromboxane metabolite of platelet origin (11-dehydrothromboxane B2) was very significantly increased (P less than .0003) in the patients. In contrast, the urinary metabolite reflecting the vascular production of prostacyclin (2,3-dinor-6-keto-prostaglandin F1 alpha) was much less increased (P less than .02). We found no correlation between the levels of anticardiolipin antibodies and the urinary excretion of 11-dehydro-thromboxane B2. Six patients with elevated urinary 11-dehydrothromboxane B2 were treated with low-dose aspirin (20 mg/d during 7 days). In these patients, there was a close relationship between the extent of inhibition of the thromboxane urinary metabolite (72%) and serum thromboxane B2 (79%). In contrast, the urinary excretion of 2,3-dinor-6-ketoprostaglandin F1 alpha was nearly unchanged (13% reduction). In addition, the F(ab')2 fragments isolated from six patients presenting increased urinary 11-dehydro-thromboxane B2 enhanced the generation of thromboxane B2 (P = .04) and the release of 14C serotonin (P = .009) by normal washed platelets, as compared with F(ab')2 from controls. In summary, our study shows that in patients with lupus anticoagulant, platelet activation may occur without a compensatory increment in the vascular biosynthesis of prostacyclin. This observation may be crucial to cause or reflect an increased risk for thrombosis. In addition, our results may suggest a rationale for antiplatelet agents for the prophylaxis of thrombosis in many patients with the antiphospholipid syndrome.