Clinical Chemistry 56:6 930940 (2010)

Review Challenges in the Diagnosis of the Antiphospholipid Syndrome

Katrien Devreese1* and Marc F. Hoylaerts2

The antiphospholipid syndrome (APS) is an important cause of acquired thromboembolic complications and pregnancy morbidity. Its diagnosis is based on clinical and laboratory criteria, defined by strict guidelines. The original clinical and laboratory criteria for the identification of APS patients were published in 1999, in the so-called Sapporo criteria. In 2006 these criteria were revised, and recently more precise guidelines for analysis of the lupus anticoagulant have been provided. However, several questions related to the diagnosis of APS remain unanswered.
CONTENT:

BACKGROUND:

The antiphospholipid syndrome (APS)3 is an autoimmune disease defined by the presence of thromboembolic complications (TEC) and/or pregnancy morbidity, in the presence of persistently increased titers of antiphospholipid antibodies (APAs). Although at present APS is a well-described, difficult-to-diagnose entity, it took many decades to define the diagnostic criteria. Since its first description in 1983 (1 ), APS has engaged hematologists and obstetricians.
HISTORICAL OVERVIEW

In addition to providing a historical perspective, this review covers several challenges in the diagnosis of APS with respect to clinical and laboratory features, while highlighting pathogenic pathways of the syndrome. We discuss ongoing dilemmas in the diagnosis of this complex disease. Although antiphospholipid antibodies are found in association with various clinical manifestations, the older established clinical criteria were not substantively altered in the 2006 update. Several laboratory tests recommended in the latest criteria, including phospholipid-dependent coagulation tests for the detection of the lupus anticoagulant and ELISAs for measuring anticardiolipin and 2glycoprotein I antibodies, still show methodological and diagnostic shortcomings. In addition, antiphospholipid antibodies have been described against other antigens, but their clinical role remains uncertain.

CONCLUSIONS: Despite updated APS criteria, diagnosis of this syndrome remains challenging. Further research on clinically relevant antibodies and standardization of their detection are needed to improve clinical risk assessment in APS.

2010 American Association for Clinical Chemistry

APS first began to be defined as an entity in the 1950s with the recognition of 2 unusual laboratory phenomena: false-positive syphilis tests and the occurrence of a nonspecific coagulation inhibitor (2, 3 ). APA already had been detected earlier in the 20th century by immunological assays used in the diagnosis of syphilis. In 1907, Wasserman et al. (4 ) developed a complement fixation assay for syphilis, using as an antigen (which they named reagin) the phospholipids (PLs) derived from saline liver extracts of fetuses with congenital syphilis. In 1941, Pangborn (5 ) demonstrated that reagin was an anionic PL, and it was renamed cardiolipin because from then on it was isolated from bovine heart muscle. As serological tests for syphilis were used increasingly for the screening of large numbers of patients, it became evident that a subpopulation of individuals with positive serological tests for syphilis showed none of the clinical symptoms of the disease. Moore and Mohr, in 1952 (6 ), were among the first to identify such individuals and found that these transiently false-positive tests were associated with several other infectious diseases besides syphilis. Persistently false-positive tests were also associated with an increased risk for future development of systemic lupus erythematosus (SLE) and several related autoimmune

Coagulation Laboratory, Department of Clinical Chemistry, Microbiology and Immunology, Ghent University Hospital, Ghent, Belgium; 2 Center for Molecular and Vascular Biology, University of Leuven, Leuven, Belgium. * Address correspondence to this author at: Coagulation Laboratory, Laboratory for Clinical Biology, Ghent University Hospital, De Pintelaan, 185 (2P8), B-9000 Gent, Belgium. Fax 00-32-9-332-45-89; e-mail katrien.devreese@uzgent.be. Received November 24, 2009; accepted February 18, 2010. Previously published online at DOI: 10.1373/clinchem.2009.133678

Nonstandard abbreviations: APS, antiphospholipid syndrome; TEC, thromboembolic complications; APA, antiphospholipid antibody; PL, phospholipid; SLE, systemic lupus erythematosus; LAC, lupus anticoagulant; aCL, anticardiolipin; 2GPI, 2-glycoprotein I; SSC, Scientific Standardisation Subcommittee; ISTH, International Society of Thrombosis and Haemostasis; ITP, idiopathic thrombocytopenic purpura; aPT, antiprothombin antibody; aPS/aPT, antibody to phosphatidylserine-prothrombin complex; dRVVT, diluted Russell viper venom test; aPTT, activated partial thromboplastin time; NPP, normal pooled plasma; INR, international normalized ratio; OAC, oral anticoagulant therapy; LMWH, low molecular weight heparin; TG, thrombin generation.

930

Challenges in the Diagnosis of APS

Review
only vascular thrombosis and recurrent fetal loss are included in the current diagnostic criteria (see (22 )). The syndrome is characterized by the occurrence of a heterogeneous population of autoantibodies against PL-binding proteins. APAs are also found in other autoimmune diseases, in patients receiving drugs such as procainamide and chlorpromazine, in children with recent viral infections, in patients with infections (HIV, hepatitis, malaria, and others), in association with malignancy, and even in otherwise healthy individuals (23 ). The mechanisms and pathophysiology of the clinical symptoms associated with APS are highly heterogeneous. Although progress is being made in understanding the pathogenesis, difficulties persist in the identification of patients at risk for TEC. Because the incidence of clinical symptoms is high and these symptoms can be caused by other underlying factors, APS diagnosis relies predominantly on laboratory results where detection of APAs is required, by definition. However, the laboratory diagnosis is complicated by the lack of a gold standard. The Sydney update of the classification criteria for definite APS published in 2006 (22 ) led to a substantial improvement of APS diagnosis, but several difficult issues remain unsolved. Recently, the serological criteria that define APS have been under debate (24 26 ). In addition, the Scientific Standardisation Subcommittee (SSC) of the International Society of Thrombosis and Hemostasis (ISTH) has recently provided useful additional details and specifications about LAC detection, in more stringent guidelines, obtained via international consensus statements (27 ). This review highlights the clinical diagnosis and the potential and limitations of current diagnostic tests, against the pathophysiological background of the syndrome. Diagnostic Criteria Clinical and serological classification criteria for APS were first formulated and published in 1999 and called the Sapporo criteria (28 ). Since their publication, new clinical, laboratory, and experimental insights have been acquired. Consequently, the criteria were revised and published in 2006 as the Sydney criteria (22 ). The updated statement maintains a subdivision of clinical and laboratory criteria requiring at least 1 clinical and 1 laboratory criterion to diagnose APS.
CLINICAL DIAGNOSIS

diseases (2 ). Nearly concurrently, Conley and Hartmann (3 ) observed an acquired circulating inhibitor that prolonged the coagulation time in vitro in 2 patients with SLE. Seven years later, Loeliger (7 ) described the presence of a factor in some plasmas that prolonged the clotting time, even when the plasma was diluted with normal pooled plasma, and proposed prothrombin as a possible candidate. In 1963, Bowie et al. (8 ) first described the association of this anticoagulant with thrombosis instead of bleeding. The association of the anticoagulant with thrombosis and intrauterine death was subsequently described by other investigators (9 ). It was not until 1972, however, that Feinstein and Rapaport (10 ) introduced the term lupus anticoagulants (LACs) for an inhibitor directed against coagulation cascade PLs. Because the majority of patients with LACs do not have SLE, the term is clearly a misnomer but has been retained for historical reasons. The anticoagulant effect is strictly an in vitro phenomenon, although some antibodies have been noted to simultaneously manifest an anticoagulant effect in vitro and a prothrombotic effect in vivo. In 1983, Harris and coworkers developed a radioimmunoassay for the detection of anticardiolipin (aCL) antibodies (11 ), and 2 years later, they developed the first quantitative ELISA (12 ). Using this assay on a large population of SLE patients, they showed that the subgroup with increased aCL antibodies had a higher incidence of thrombosis and pregnancy morbidity (13 ). This led to the first description of the so-called anticardiolipin syndrome, also referred to as the antiphospholipid syndrome by the same investigators (1 ). One of the most intriguing findings in the early 1990s was that the aCL antibody activity in most APS patients depends on the presence of a protein cofactor. Three independent groups showed that so-called aCL antibodies were not directed against cardiolipin per se, but against a cofactor (14 16 ), a plasma apolipoprotein known as 2-glycoprotein I ( 2GPI), a 50-kDa single-chain polypeptide glycoprotein that binds to anionic PLs. Further research uncovered that not all aCL antibodies were 2GPI dependent and that the in vitro anticoagulant effect of LAC-positive aCL antibodies depended on the presence of 2GPI only in a subset of patients. 2GPI-dependent antibodies were found only in patients with autoimmune diseases, whereas patients with infectious diseases had 2GPI-independent APA (16 ). Subsequently, other groups also reported that 2GPI was needed as a cofactor for LAC activity (1719 ). In other patients, the LAC activity was shown to be dependent on the presence of prothrombin (20, 21 ).
CURRENT CHALLENGES

The actual cause of APS is unknown, but thought to be multifactorial. APS can involve almost any organ, but

Because APS patients can present very differently with a diverse range of clinical symptoms, it remains a challenge for clinicians to recognize the underlying disease. The classic presentation of APS includes life-threatening vascular thrombosis, as well as obstetrical complications such as recurrent fetal loss and severe preeclampsia.
Clinical Chemistry 56:6 (2010) 931

Review

Table 1. Mechanisms of APA-mediated thrombosis.

Effect of APAs on the coagulation system Interference with the components of intrinsic pathway Inhibition of factor XI activation Factor XII deficiency or inactivation Inhibition of antithrombin activity Interference with the protein C pathway Inhibition of APC anticoagulant pathway by competition on phospholipid surfacea Disruption of interaction within the APC complex Acquired protein S deficiency Interference with the thrombomodulinprotein Sprotein C pathway Binding to activated cofactors Va and VIIIa resulting in impaired proteolysis by APC Antibodies against the endothelial protein C receptor Inhibition of thrombin formation Neutralization of activated factor X and tissue factor pathway inhibitor Upregulation of tissue factor and tissue factor pathway inhibitor Increasing the amount of Von Willebrand factor Effect of APAs on cell functions Procoagulant effects on platelets by ApoER2 and GPIb recpetors Impairment of fibrinolysis on endothelial cells by annexin A2 receptor Tissue factor expression and proinflammatory cytokine production by monocytes Disruption of the annexin A5 antithrombotic shield on vascular cells Inhibition of endothelial cell prostacyclin production Binding of LDL receptors on renal tubuli cells, neuronal cells Other interactions of APAs Complement activation
a

Table 2. Sydney clinical criteria for APS.

1. Vascular thrombosis One or more clinical episodes of arterial, venous, or small-vessel thrombosis in any tissue or organ confirmed by imaging, Doppler studies, or histopathology, with the exception of superficial venous thrombosis. For histopathologic confirmation, thrombosis should be present without significant evidence of inflammation of the vessel wall. 2. Pregnancy morbidity (a) One or more unexplained deaths of a morphologically normal fetus at or beyond the 10th week of gestation, with normal fetal morphology documented by ultrasound or by direct examination of the fetus, or (b) One or more premature births of a morphologically normal neonate at or before the 34th week of gestation because of severe preeclampsia or eclampsia or severe placental insufficiency, or (c) Three or more unexplained consecutive spontaneous abortions before the 10th week of gestation, with maternal anatomic or hormonal abnormalities and paternal and maternal chromosomal causes excluded.

APC, activated protein C; ApoER, apolipoprotein E receptor.

Nonetheless, the disease may affect virtually any organ, including lungs, heart, skin, brain, kidneys, eyes, adrenal glands, and liver (29 ). Several pathogenic mechanisms for thrombosis in APS have been described (30, 31 ) (Table 1). It is likely that no single mechanism explains thrombosis and that the episodes of acute thrombosis have a multifactorial pathogenesis. Thrombosis plays a key role in most of the clinical manifestations of APS. Inhibition of factors of the anticoagulant system, impairment of fibrinolytic activity, interference with coagulation factors and complement, and the direct effect of APAs on cell function have been proposed to explain the thrombotic tendency found in APS patients (30, 31 ).
932 Clinical Chemistry 56:6 (2010)

Some clinical symptoms including neurological and pregnancy complications cannot be explained by thrombotic or ischemic mechanisms (see (32 )). The pathophysiology of pregnancy failure in APS patients may involve inflammation in the placenta and disruption of normal trophoblast function and survival, rather than occurrence of a prothrombotic event. A disturbed cytokine/chemokine profile and chemotaxis function of trophoblasts induced by anti- 2GPI antibodies may be responsible for the inflammatory infiltrates seen in the maternalfetal interface in these patients (33 ). Chronic inflammation is an important characteristic in APS patients that may dispose them to TEC. Platelets play a key role in inflammation and thrombosis. Plasma levels of platelet receptors may therefore be indicators of a thrombotic risk. Accordingly, soluble P-selectin is higher in patients with venous thromboembolism, but its role as predictor for TEC needs to be further evaluated in prospective studies (34, 35 ). Established clinical criteria were left unchanged in the revised Sydney criteria (22 ) (Table 2), except for transient cerebral ischemia and ischemic stroke, which now are considered to reflect vascular thrombosis. Superficial venous thrombosis is not included in the clinical criteria. Thrombosis of the deep limb veins with or without pulmonary embolism is the most common manifestation of APS, but any part of the venous and arterial circulation can be involved, including mesenteric, portal, intracranial, and retinal vessels. Ischemic stroke is the most common arterial thrombotic manifestation, but myocardial infarction, peripheral gan-

Challenges in the Diagnosis of APS

Review
These additional risk factors include age, the established risk factors for cardiovascular disease, inherited thrombophilias (such as factor V Leiden or prothrombin mutation G20210A), use of oral contraceptives, nephrotic syndrome, malignancy, immobilization, and surgery (22 ). Possible genetic predisposition for developing APS and producing APAs has been addressed in family and population studies (41 ). Associations of APS and APAs have been described with human leukocyte antigen-DR and -DQ, as well as with polymorphisms in the 2GPI antigen, the genes encoding for platelet glycoproteins, the signaling pathways for proinflammatory mediators, and genetic defects of the immune system (e.g., IgA or complement) (42 ). A major change in the Sydney vs the Sapporo criteria was the omission of the distinction between primary and secondary APS. Patients with APS often have SLE or related autoimmune diseases and are referred to as having secondary APS. When the syndrome occurs in the absence of such autoimmune diseases, it is then known as primary APS. It is unknown whether APS and SLE are 2 diseases coinciding in an individual, whether underlying SLE offers a setting for the development of APS, or whether APS and SLE represent 2 elements of the same process. Secondary and primary APS are not always distinguishable, since they show some common clinical/serological manifestations (43 ). For the interpretation of laboratory tests, distinction between primary and secondary APS can be useful: a test will have a higher specificity in a population with high prevalence for the targeted disease, but may result in false positives in an unselected population. Testing for APA should be limited to patients with a significant probability of having APS. In the updated guidelines for LAC detection (27 ) the appropriateness of searching for LACs is graded according to the patients clinical characteristics into low, moderate, and high risk, taking into account age, provoked vs unprovoked TEC, location of the TEC, and early vs late pregnancy loss. Generalized screening for APA and screening in patient populations with low pretest probability is discouraged to avoid false-positive laboratory diagnoses, caused by the poor specificity of some assays (see Laboratory diagnosis). The catastrophic antiphospholipid syndrome is regarded as a separate entity with high morbidity and mortality (44 ). The term catastrophic defines an accelerated form of APS, resulting in multiple organ failure, but this form occurs in fewer than 1% of all APS patients.
LABORATORY DIAGNOSIS

Table 3. Other clinical manifestations of APS.

APA-associated cardiac valve disease APA-associated livedo reticularis APA-associated nephropathy APA-associated thrombocytopenia Other skin manifestions Ulcerations, pseudo-vasculitic lesions, digital gangrene, superficial phlebitis, malignant atrophic papulosis-like lesions, subungual splinter hemorrhages, and anetoderma (a circumscribed area of loss of dermal elastic tissue) Other neurological manifestations Cognitive dysfunction, chorea, headache or migraine, multiple sclerosis, transverse myelopathy, and epilepsy

grene, and thrombotic occlusion of any part of the arterial vessels, including mesenteric, adrenal, and renal vessels, can occur (36 ). Clinical features of APS (Table 3) not included in the classification criteria are heart valve disease, livedo reticularis, thrombocytopenia, nephropathy, and neurological manifestations (22 ). Various hematologic manifestations have been described in association with APS, thrombocytopenia being among the most frequent (37 ). The occurrence of thrombocytopenia in patients with persistent APAs should be differentiated from idiopathic thrombocytopenic purpura (ITP), since patients with ITP do not have an increased thrombotic risk. With regard to cardiac manifestations, only coronary artery disease fulfils the thrombosis criterion for APS. Other dermatological manifestations besides livedo reticularis occur and may be the first presenting clinical manifestation in APS, e.g., skin ulcerations, digital gangrene, and subungual splinter hemorrhages (38 ). Besides transient cerebral ischemia and ischemic stroke, no neurological manifestations are included in the clinical criteria. There was insufficient evidence to include cognitive dysfunction, chorea, headache or migraine, multiple sclerosis, transverse myelopathy, and epilepsy in the revised classification criteria; however, the data concerning cognitive dysfunction are suggestive and warrant further study (32 ). Patients persistently positive for APAs only occasionally suffer from thrombotic events. A 2-hit hypothesis offers a potential explanation of this phenomenon by requiring, apart from the increased risk of thrombotic events induced by APAs by cellular/coagulation activation (first hit), another priming factor to trigger thrombus formation (second hit). In the revised criteria, attention is given to additional inherited or acquired risk factors for thrombosis, which are frequent in APS patients, as illustrated in recent studies (39, 40 ).

Appropriate laboratory testing is required to exclude or confirm the clinical diagnosis of APS. The most appropriate laboratory tests as described in the official criteClinical Chemistry 56:6 (2010) 933

Review

Table 4. Comparison of laboratory criteria of APS.

Sapporo criteria Sydney criteria

LACs

Screening, mixing, and confirmation tests (ISTH guidelines) Two or more occasions at least 6 weeks apart

Screening, mixing, and confirmation tests (ISTH guidelines) Two or more occasions at least 12 weeks apart Detected by standardized ELISA IgG and/or IgM Medium or high titer ( 40 units IgG or IgM phospholipid antibody titer or 99th percentile) Two or more occasions at least 12 weeks apart IgG and/or IgM Titer 99th percentile Two or more occasions at least 12 weeks apart

aCL antibodies

Detected by standardized 2GPI-dependent ELISA IgG and/or IgM Medium or high titer Two or more occasions at least 6 weeks apart

Anti- 2GPI antibodies

ria (22 ) fall into 2 categories: (a) immunological assays (usually ELISA formats) that detect antibodies and (b) clotting assays that detect APAs indirectly by measuring their effect on the coagulation system. The APAs constitute a heterogeneous group of autoantibodies directed against a multitude of antigens ( 2GPI, prothrombin, protein C, protein S, annexin V, thrombomodulin, factor XII, and others). Antibodies against 2GPI and prothrombin are found frequently enough in APS patient plasma to ascribe them a pathophysiological role (31 ). Moreover, there is increasing evidence that only high-avidity antibodies are pathogenic (45 ). The laboratory criteria have been substantially modified in the Sydney criteria (22 ) (Table 4). Laboratory criteria consist now of documenting the presence of LACs, aCL antibodies, or anti- 2GPI antibodies (22 ). The inclusion of antibodies against 2GPI in the new criteria as a sufficient laboratory criterion for the diagnosis of definite APS is a major change. Determining antibody profiles with subclassification of patients according to the number and type of positive tests is encouraged (22 ). Categories include patients with more than 1 laboratory criterion present in any combination, LACs present alone, aCL antibodies present alone, or anti- 2GPI antibodies present alone. Also, the agreement of isotype (preferably IgG) of aCL and anti- 2GPI antibodies is believed to identify patients at higher risk (27, 46 ). Clinical studies have confirmed the clinical importance of triple positivity (46 49 ). Moreover, patients with a full positive profile are at high risk of developing future TEC. Patients positive only for aCL and anti- 2GPI antibodies without any history of TEC present with obstetric complications only (46 ). Isolated LAC positivity is present
934 Clinical Chemistry 56:6 (2010)

more frequently in subjects without clinical events and may even be false positive in elderly patients or when diagnosed for the first time (50, 51 ). The new criteria increase the elapsed time between the initial antibody test and the confirmatory test from 6 to 12 weeks (22 ), thereby increasing the probability of excluding temporary infection-associated antibodies. Persistent positivity of laboratory tests is important (52 ), since transient presence of epiphenomenal APAs may give rise to misclassification. It has been suggested not to classify APS when the time interval between the clinical event and the positive laboratory test exceeds 5 years; a time interval exceeding 12 weeks between symptom and test requires careful assessment of the relationship between clinical manifestations and APAs (22 ). A few studies have compared the original Sapporo criteria with the updated Sydney criteria (39, 53, 54 ). Analysis of an APA-positive patient population demonstrated that the updated criteria are more stringent than the original ones, resulting in a smaller, more highly selected APA-positive patient population. By contrast, between 1.5% and 6% of the patients who did not fulfill the original Sapporo criteria met the revised criteria owing to isolated anti- 2GPI antibody positivity (39, 49, 53 ). Swadzba et al. (49 ) confirmed that anti- 2GPI antibodies are strongly associated with clinical complications of APS, but these appeared less frequently in their study population than aCL antibodies, illustrating the high specificity and quite low sensitivity previously reported. Moreover, the presence of anti- 2GPI antibodies was closely related to the presence of other APAs. All laboratory tests considered for the classification of APS (aCL antibodies, anti- 2GPI antibodies, and LACs) have some limitations related to robustness,

Challenges in the Diagnosis of APS

Review
bodies are associated with preeclampsia and eclampsia, as well as with abortion (see online Supplemental Data). The circulating 2GPI protein is not able to interact with cellular receptors until after its dimerization by autoantibodies (31 ). Receptor-bound 2GPIAPA complexes bound to target cells induce an intracellular signaling and/or a procoagulant/proinflammatory phenotype that predisposes to clinical symptoms. Thus, the deregulated activation of endothelial cells, platelets, and monocytes by anti- 2GPI/ 2GPI complexes may explain the thrombotic predisposition in APS (30, 31, 42 ). Several receptors have been advanced in this activation, such as those activating platelets (apolipoprotein E receptor 2 and the glycoprotein Ib ). In a recent study, platelet factor 4 appeared to be an important binding protein for 2GPI, leading to a stabilized 2GPI dimeric structure facilitating antibody binding with procoagulant potency (59 ). Annexin A2 is postulated in endothelial and monocyte activation, as well as Toll-like receptors (30, 60, 61 ). Other binding sites for anti- 2GPI/ 2GPI complexes are the LDL receptorrelated protein, megalin, and very low density lipoprotein, as well as P-selectin glycoprotein ligand 1 (31 ). In theory, anti- 2GPI antibody ELISAs are preferred over aCL antibody ELISAs, since the microtiter plates used are coated with a single and well-defined glycoprotein. Although there is room for improved standardization (55 ), better agreement is found between different assays compared with aCL antibody assays (51 ). Laboratories measuring anti- 2GPI antibodies are encouraged to follow the guidelines proposed by the standardization group of the European Forum on Antiphospholipid Antibodies (62 ). The anti- 2GPI antibody ELISA still has some diagnostic weaknesses. In particular, when anti- 2GPI antibody detection is the only positive test, results should be related to clinical and other laboratory findings. Despite their higher specificity compared with the aCL ELISA, the anti- 2GPI antibody ELISAs detect all antibodies reactive with 2GPI, including nonpathogenic antibodies and phospholipid-independent anti2GPI antibodies, which makes them less suitable as a general diagnostic test. Furthermore, the current anti2GPI antibody ELISAs also detect low-affinity anti2GPI antibodies (45 ). Decreasing detection of lowaffinity antibodies improves the specificity of the assay and correlation of the test with thrombotic risk. Antibodies targeting 2GPI domain I appear to correlate better with thrombosis than antibodies against other domains of 2GPI, but odds ratios vary from study to study, and whether these are the only pathogenic antibodies is still uncertain (63, 64 ).
Clinical Chemistry 56:6 (2010) 935

reproducibility, standardization, and clinical relevance. Clinicians not familiar with the diagnosis of APS often have difficulties with interpretation of the results, necessitating interpretative comments or advice and clear conclusions on the laboratory report (27 ).
APAs MEASURED BY ELISA

Anticardiolipin antibodies. The diagnostic value of aCL antibodies is currently under debate (24, 25 ), partly owing to methodological problems. In addition to several factors that contribute to variability in pre-, post-, and analytical conditions, many factors are related to the assay itself and its calibration (55 ). Currently, most of the commercially available assays have added cardiolipin and 2GPI, which increases the specificity of the assay. Recently, a detailed protocol delineated guidelines for the performance of aCL antibody ELISAs (56 ). One of the critical issues in the standardization process is the definition of a cutoff value. Disagreement between different assay kits and methods is observed particularly in the lower range of antibody levels. In the current guidelines, only medium and high levels (i.e., greater than the 99th percentile or 40 units IgG or IgM phospholipid antibody titer) of aCL antibodies are included as a diagnostic criterion, which improves the specificity of the test (22 ). The idea that an aCL test would be better replaced by an anti- 2GPI assay was based on the results of a metaanalysis (57 ) illustrating that LACs are a stronger risk factor for thrombosis than APAs detected by aCL assays. Although this study elicited some criticism, other studies also found a lack of correlation between increased aCL antibodies and thrombosis or pregnancy loss. One report indicated that omission of aCL testing from the investigation of APS could lead to failure to diagnose the syndrome in a fraction of patients (see Supplemental Data, which accompanies the online version of this article at www.clinchem.org/content/ vol56/issue6). Myocardial infarction and stroke and recurrent fetal loss were found to be associated with aCL antibodies in 2 reviews (57, 58 ). Although the discussion about the role of aCL antibodies continues, the aCL assay deserves its place in the diagnostic work-up for APS, owing to its high diagnostic sensitivity, albeit low specificity. In view of the low specificity of assays for aCL antibodies, they are to be considered an additional diagnostic tool and their interpretation should be made in the light of related clinical manifestations. Anti- 2GPI antibodies. Several studies have shown a relation between thrombosis and the presence of anti2GPI antibodies. Recently, the role of anti- 2GPI antibodies has been described in young women with ischemic stroke. Evidence exists that anti- 2GPI anti-

Review
Especially high titers of anti- 2GPI antibodies are associated with a high risk of thrombosis. Until an international consensus is reached, the same threshold ( 99th percentile of controls) is recommended for positive anti2GPI antibodies for patients with thrombosis and patients with obstetrical complications (22 ). Antiprothrombin antibodies. Antiprothrombin antibodies (aPTs) are a heterogeneous group, including antibodies against prothrombin (aPT) and antibodies to the phosphatidylserine-prothrombin complex (aPS/PT). Evidence is accumulating that antibodies against aPS/PT, rather than against prothrombin alone, are closely associated with APS and the presence of LACs. Some studies have established aPT antibodies as a risk factor for thrombosis, especially, but not always, in conjunction with the presence of a LAC or the presence of SLE. aPTs have also been described in association with pregnancy loss. Other studies fail to indicate these antibodies as pathogenic in APS (see online Supplemental Data). So far, and despite growing evidence, these antibodies have not (yet) been included in the Sydney criteria. Therefore aPS/PT antibodies remain to be explored further as candidates in the classification of APS (65 ). Other antiphospholipid antibodies. Many other antiphospholipid antibodies are described in APS, such as those reactive with phosphatidic acid and phosphatidylcholine, -ethanolamine, -glycerol, -inositol, and -serine (66 ). The measurement of antiannexin A5 antibodies has appeared useful in some situations (30, 67 ). Although a few studies report the association of antiphosphatidylethanolamine, antiprotein S antibodies, and antiannexin A2 antibodies with thrombosis and fetal loss, experience with such antibodies is limited (66, 67 ). It is premature to include assays of other aPL antibodies in the daily work-up for APS because these assays have not been evaluated for precision and interlaboratory variation and standardization is lacking (68 ). Further studies are needed to address the importance of nonclassic APAs found in APS. Isotype of APAs. Current criteria recommend increased levels of IgG and IgM APA antibodies to confirm APS. IgM APAs are less often associated with clinical events of APS than IgG (57 ). A large prospective study on IgG and IgM aCL, anti- 2GPI, and aPT antibodies did not find any significant association between IgM isotype and thrombosis, even when a higher cutoff value was used (69 ). A recent prospective study confirmed that only a minority of APS patients carried IgM aCL or anti- 2GPI antibodies; moreover, these patients also had other thrombotic risk factors (46 ).
936 Clinical Chemistry 56:6 (2010)

IgA anti- 2GPI antibodies are not included in the current criteria because their association with the clinical manifestation of APS is unclear (70 ). IgA anti2GPI antibodies are frequently detected in specific ethnic groups such as African Americans (22 ). Case reports have been published of exclusive IgA anti2GPI seropositivity with concomitant clinical manifestations (71 ).
LUPUS ANTICOAGULANTS

LACs are detected on the basis of their functional activity through interference with PL-dependent steps in the coagulation cascade, thereby detecting various APA. LACs interfere with coagulation reactions via more than 1 mechanism (Table 1). LAC detection is performed according to the revised criteria proposed by the SSC of the ISTH in a multistep procedure, including screening, mixing, and confirmation tests (72 ). LAC assays are highly specific for the diagnosis of APS and show strong association with thrombotic events and fetal loss (see online Supplemental Data). On the other hand, LAC detection has several disadvantages, and the methodology is complicated and labor intensive (55 ). Until recently, precise guidelines for performance and interpretation of LAC assays were lacking. In 2009, the SSC provided useful additional details and specifications for the LAC detection in more stringent guidelines, based on knowledge and experience collected over several decades (27 ) (summarized in Table 5). Specifications are provided on adequate plasma preparation and storage. Double centrifugation is considered the most effective way to prepare platelet-poor plasma ( 10 000/ L). Filtration is discouraged because it may affect other coagulation proteins such as von Willebrand factor and factor VIII. Plasma storage for longer periods should be done by quick freezing (at least 70 C). Before performing LAC assays, a thrombin time or antifactor Xa assay should always be performed to exclude the presence of heparin, even when tests include a heparin neutralizer (27 ). To reduce the risk of false positives, the current recommendation is to use 2 tests with different assay principles, because no test has 100% sensitivity. LAC should be considered positive when 1 of the 2 tests is positive. The first choice is a dRVVT (diluted Russell viper venom test) and the second one an aPTT (activated partial thromboplastin time) with silica as activator and a low concentration of PLs. Other tests are not recommended. Mixing tests should be performed by mixing patient plasma in a 1:1 proportion with normal pooled plasma (NPP) and without preincubation. NPP should preferentially be homemade, but commercial lyophilized or frozen plasmas can be used when they fulfill specifications or when they are validated for

Challenges in the Diagnosis of APS

Review

Table 5. Update of the guidelines for LAC detection: technical specifications.

Recommendations Discouraged

Blood collection

0.109 mol/L sodium citrate (9:1) Double centrifugation Quick freezing ( 70 C) Unthawing at 37 C for 5 min in warm water bath by total immersion, mix before testing Plasma filtration

Test procedure

Thrombin time, anti-Xa, specific clotting factor dosage if clinical history of bleeding Screening, mixing, and confirmation tests Two tests with different principle dRVVT as first test aPTT as second test Kaolin as activator Ellagic acid as activator dPT KCT Ecarin and textarin clotting time More than 2 tests

Choice of assays

Screening test Mixing test

Cutoff value by 99th percentile 1:1 proportion patient plasma/NPP No preincubation Homemade NPP ( 10 000 plt/ L, 100% clotting factors, Commercial NPP can be used if adequate Cutoff value by 99th percentile or Rosner index 70 C)

Mean plus 2 SDs

Confirmation test

Increase of PL of screening test Bilayer or hexagonal PL Cutoff value by % correction [(screen or LA ratio (screen/confirm) confirm)/screen] 100 Frozen/thawed platelets

LAC detection. For the confirmation step, bilayer or hexagonal-phase PLs are recommended over other sources, such as rabbit brain extract, PL vesicles, platelet extracts, and frozen-thawed platelets (27 ). The use of normalized ratios for all LAC test results is recommended to compensate for inter- and intraassay variation. Calculation of cutoff values by the individual laboratory is mandatory in all test steps, using local reagents and coagulometers. Fewer false positives will be obtained by taking the 99th percentile of a healthy donor population instead of the mean plus 2 SDs (22 ). Interpretation of mixing studies is now well defined by considering a local cutoff value calculated by the 99th percentile of a normal population or by the Rosner index (27 ). A percentage correction formula is not considered in this update (73 ). Alternatively, for confirmation tests a percentage correction formula is recommended (27 ). The Sydney criteria have recommended integrated test systems, including screening and confirmation tests into 1 assay, making LAC testing less time consuming and increasing accuracy and interlaboratory

agreement (22, 55 ). Although mixing tests are still considered mandatory for LAC detection in the 3-step procedures (22, 27, 72 ), the updated guidelines (27 ) recommend specifically for integrated tests that, in principle, no mixing test is required. However, additional considerations are needed. To increase diagnostic efficacy, mixing tests cannot be omitted from integrated test systems and should be applied when the clotting time of the confirmation assay is outside the reference interval (74 ). One of the major drawbacks of the coagulation tests used for LAC detection is their sensitivity to the presence of anticoagulant therapy. It is recommended to postpone these tests until the international normalized ratio (INR) is 1.5. Bridging oral anticoagulant therapy (OAC) with low molecular weight heparin (LMWH) can be an alternative, with blood sampling 12 h after the last administration. If the INR is between 1.5 and 3, a dilution of patient plasma with NPP can be considered. Nonetheless, testing in the acute phase of the TEC is discouraged (27 ). The detection of LAC via the combination of PL-dependent and
Clinical Chemistry 56:6 (2010) 937

Review
-independent tests, such as textarin and ecarin clotting times, is not recommended. Because LAC cannot be tested reliably in patients receiving anticoagulants, alternative coagulation assays can be a valuable tool. We demonstrated that thrombin generation (TG) techniques are well suited for the monitoring of patients on anticoagulants, via the determination of TG parameters in diluted patient plasma (75 ). We found that lag time and peak height measured by use of calibrated automated thrombinography are the parameters of choice to monitor the anticoagulant effect of APAs. During TG in 1:1 mixtures of anticoagulated patient plasma and normal plasma, either their plasma peak height (thrombin concentration) was lowered or their lag time delayed, or both. Selected LAC plasmas showed reduced normalized peak height/lag time ratios, irrespectively of their treatment status (75 ). Coagulation tests have been developed to discriminate between thrombosis-related 2GPI-dependent and nonpathogenic non 2GPIdependent LACs (76 78 ). However, these tests are used for research purposes only and are not applied in daily practice. LAC activity should be quantitatively expressed to indicate the strength of activity and to allow numeric differentiation between high and low titers. Quantification of LACs in clinical samples has been complicated by the lack of a suitable standard of activity. We recently demonstrated that TG calibration curves constructed with mixtures of monoclonal antibodies against 2GPI and prothrombin allow expression of the strength of LACs in arbitrary units. A broad variation in LAC-positive patients was observed, with titers ranging from 0 to 200 U/mL (35 ). High titers discriminated well between LAC patients with and without TEC, with an odds ratio of 3.5. This analysis, however, even with the determination of anti- 2GPI antibodies, was clinically insufficient to discriminate between LAC patients at risk for thrombosis or not. This study demonstrated that if 1 of these 2 parameters had a low titer, additional procoagulant markers needed to be analyzed to reach clinically useful information in weak LAC patients (35 ). Conclusions The detection of APAs in patients with a history of thrombosis or pregnancy complications is an essential step in the diagnosis and management of APS (79 ). The detection of APAs assigns patients with a common event (thrombosis) to a group with a high risk for recurrence, which is a prerequisite for long-term OAC. The assays for the detection of APAs must be sufficiently sensitive to classify patients correctly as APS positive. They also need to be highly specific, since false-positive results may have an impact on clinical decisions. Falsely diagnosed patients risk being exposed to indefinite OAC with a high risk of bleeding, without having any benefit from such treatment. Despite updates of the diagnostic criteria, the diagnosis of APS remains difficult. Some aspects of the Sydney criteria were debated after their introduction. The need for more guidelines regarding the detection of LAC is now fulfilled by the SSC updated guidelines. Evidence is growing that only IgGs play a pathogenic role, arguing against the continued detection of IgM antibodies for aCL and anti- 2GPI. Detection of anti2GPI antibodies specific against the first domain of 2GPI seems to be a useful tool to define thrombotic risk in APS patients. Wide variability in strength of this association has been reported in several studies, however, and therefore it seems premature to focus solely on these antibodies. Future research and development of standardized assays that detect only the clinically relevant antibodies will be of great value. Antibody profiles are encouraged to identify thrombotic risk and are more useful than individual tests in identifying the patients risk. The thrombogenicity assessment in APS patients is still hampered by laboratory tests that have poor predictive value for the thrombotic phenotype (22 ). A recent epidemiological study was unable to identify clinical or immunological predictors of thrombotic events, pregnancy morbidity, or mortality (80 ). Even a quantitative LAC assay is only partially informative in the prediction/exclusion of thrombosis in APS patients, and additional hemostasis markers may have to be adopted to accomplish this task (35 ). Large prospective clinical trials examining the relevance of current laboratory tests, combined with potential new prognostic laboratory parameters, will help identify better means for stratification of patients into risk groups. Meanwhile, we will have to rely on the combination of LACs, aCL, and anti- 2GPI antibodies to diagnose APS.

Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article. Authors Disclosures of Potential Conflicts of Interest: No authors declared any potential conflicts of interest. Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, or preparation or approval of manuscript.

938 Clinical Chemistry 56:6 (2010)

Challenges in the Diagnosis of APS

Review
References

1. Hughes GRV. Thrombosis, abortion, cerebral disease, and the lupus anticoagulant. BMJ 1983; 287:1088 9. 2. Moore JE, Lutz WB. Natural history of systemic lupus erythematosus: approach to its study through chronic biologic false positive reactions. J Chron Dis 1955;1:297316. 3. Conley MR, Hartmann RC. A hemorrhagic disorder caused by circulating anticoagulant in patients with disseminated lupus erythematosus. J Lab Clin Invest 1952;31:6212. 4. Wasserman A, Neisser A, Bruck C. Eine serodiagnostische reaktion bei syphilis. Deutsche Med Wochenschr 1906;32:745 6. 5. Pangborn MC. A new serologically active phospholipid from beef heart. Proc Soc Exp Biol Med 1941;48:484 6. 6. Moore JE, Mohr CF. Biologically false-positive tests for syphilis. JAMA 1952;150:46373. 7. Loeliger A. Prothrombin as co-factor of the circulating anticoagulant in systemic lupus erythematosus? Thromb Diath Haemorrh 1959;3:23756. 8. Bowie EJW, Thompson JH Jr, Pascuzzi CA, Owen CA Jr. Thrombosis in systemic lupus erythematosus despite circulating anticoagulants. J Lab Clin Med 1963;62:416 30. 9. Lechner K. A new type of coagulation inhibitor. Thromb Diath Haemorrh 1969;21:48299. 10. Feinstein DI, Rapaport SI. Acquired inhibitors of blood coagulation. Prog Hemost Thromb 1972;1: 7595. 11. Harris EN, Gharavi AE, Boey ML, Patel BM, Mackworth-Young CG, Loizou S, et al. Anticardiolipin antibodies: detection by radioimmunoassay and association with thrombosis in systemic lupus erythematosus. Lancet 1983;2:1211 4. 12. Loizou S, McCrea JD, Rudge AC, Reynolds R, Boyle CC, Harris EN. Measurement of anticardiolipin antibodies by an enzyme-linked immunosorbent assay (ELISA): standardization and quantitation of results. Clin Exp Immunol 1985; 62:738 45. 13. Hughes GR. Autoantibodies in lupus and its variants: experience in 1000 patients. Br Med J 1984;289:339 42. 14. McNeil HP, Simpson RJ, Chesterman CN, Krillis SA. Antiphospholipid antibodies are directed against a complex antigen that includes a lipidbinding inhibitor of coagulation: 2-glycoprotein I (apolipoprotein H). Proc Natl Acad Sci U S A 1990;87:4120 4. 15. Galli M, Comfurius P, Maassen C, Hemker MC, de Baets MH, van Breda-Vriesman PJ, et al. Anticardiolipin antibodies (ACA) directed not to cardiolipin but to serum protein cofactor. Lancet 1990; 335:1544 7. 16. Matsuura E, Igarashi Y, Fujimoto M, Ichikawa K, Koike T. Anticardiolipin cofactor(s) and differential diagnosis of autoimmune disease. Lancet 1990;336:177 8. 17. Oosting JD, Derksen RH, Entjes HT, Bouma BN, de Groot PG. Lupus anticoagulant activity is frequently dependent on the presence of beta 2-glycoprotein I. Thromb Haemost 1992;67:499 502. 18. Roubey RA, Pratt CW, Buyon JP, Winfield JB. Lupus anticoagulant activity of autoimmune an-

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

tiphospholipid antibodies is dependent upon beta 2-glycoprotein I. J Clin Invest 1992;90:1100 4. Galli M, Comfurius P, Barbui T, Zwaal RF, Bevers EM. Anticoagulant activity of beta-2-glycoprotein I is potentiated by a distinct subgroup of anticardiolipin antibodies. Thromb Haemost 1992;68: 297300. Bevers EM, Galli M, Barbui T, Comfurius P, Zwaal RF. Lupus anticoagulant IgGs (LA) are not directed to phospholipids only, but to a complex of lipid-bound human prothrombin. Thromb Haemost 1991;66:629 32. Permpikul P, Rao LV, Rapaport SI. Functional binding studies of the roles of prothrombin and beta 2-glycoprotein I in the expression of lupus anticoagulant activity. Blood 1994;83:2878 92. Miyakis S, Lockshin MD, Atsumi D, Branch DW, Brey RL, Cervera R, et al. International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS). J Thromb Haemost 2006;4:295306. Rand JH. Antiphospholipid syndrome and lupus anticoagulants. In: Loscalzo J, Schafer A, eds. Thrombosis and hemorrhage. 3rd ed. Philadelphia: Lippincott Williams and Wilkins; 2003. p 612. Galli M, Reber G, de Moerloose P, de Groot PG. Invitation to a debate on the serological criteria that define the antiphospholipid syndrome. J Thromb Haemost 2008;6:399 401. Pengo V. A contribution to the debate on the laboratory criteria that define the antiphospholipid syndrome. J Thromb Haemost 2008;6: 1048 9. Swadzba J, Musial J. More on: the debate on antiphospholipid syndrome criteria. J Thromb Haemost 2009;7:5012. Pengo V, Tripodi A, Reber G, Rand JH, Ortel TL, Galli M, de Groot PG. Official communication of the Scientific and Standardization Committee on lupus anticoagulant/phospholipid-dependent antibodies: update of the guidelines for lupus anticoagulant detection. J Thromb Haemost 2009;7:1737 40. Wilson WA, Gharavi AE, Koike T, Lockshin MD, Branch DW, Piette JC, et al. International consensus statement on preliminary classification criteria for definite antiphospholipid syndrome: report of an international workshop. Arthritis Rheum 1999;42:1309 11. Hughes G. Hughes syndrome: the antiphospholipid syndromea clinical overview. Clin Rev Allergy Immunol 2007;32:311. Giannakopoulos B, Passam F, Rahgozar S, Krilis S. Current concepts on the pathogenesis of the antiphospholipid syndrome. Blood 2007;109: 42230. Urbanus RT, Derksen RHMW, de Groot PG. Current insight into diagnostics and pathophysiology of the antiphospholipid syndrome. Blood Rev 2008;22:93105. Miesbach W. Neurological symptoms as a feature of the antiphospholipid syndrome. Semin Thromb Hemost 2008;34:286 9. Abrahams VM. Mechanisms of anti-phospholipid antibody-associated pregnancy complications. Thromb Res 2009;124:52152.

34. Bugert P, Pabinger I, Stamer K, Vormittag R, Skeate RC, Wahi MM, et al. The risk for thromboembolic disease in lupus anticoagulant patients due to pathways involving P-selectin and CD154. Thromb Haemost 2007;97:573 80. 35. Devreese K, Peerlinck K, Hoylaerts M. Thrombotic risk assessment in the antiphospholipid syndrome requires more than the quantification of lupus anticoagulants. Blood 2010;115:870 8. 36. Robertson B, Greaves M. Antiphospholipid syndrome: an evolving story. Blood Rev 2006;20: 20112. 37. Uthmana I, Godeaub B, Taherc A, Khamashta M. The hematologic manifestations of the antiphospholipid syndrome. Blood Rev 2008;22:18794. 38. Blume JE, Miller CC. Antiphospholipid syndrome: a review and update for the dermatologist. Cutis 2006;78:409 15. 39. Kaul M, Erkan D, Sammaritano L, Lockshin MD. Assessment of the 2006 revised antiphospholipid syndrome classification criteria. Ann Rheum Dis 2007;66:92730. 40. Giron-Gonzalez JA, Garcia del Rio E, Rodriguez C, Rodriguez-Martorell J, Serrano A. Antiphospholipid syndrome and asymptomatic carriers of antiphospholipid antibody: prospective analysis of 404 individuals. J Rheumatol 2004;31:1560 7. 41. Castro-Marrero J, Balada E, Vilardell-Tarres M, Ordi-Ros J. Genetic risk factors of thrombosis in the antiphospholipid syndrome. Br J Haematol 2009;147:289 96. 42. Meroni PL. Pathogenesis of the antiphospholipid syndrome: an additional example of the mosaic of autoimmunity. J Autoimmun 2008;30:99 103. 43. Tincani A, Andreoli L, Chighizola C, Meroni PL. The interplay between the antiphospholipid syndrome and systemic lupus erythematosus. Autoimmunity 2009;42:2579. 44. Asherson RA, Cervera R, de Groot PG, Erkan D, Boffa MC, Piette JC, Khamashta MA, et al. Catastrophic antiphospholipid syndrome: international consensus statement on classification criteria and treatment guidelines. Lupus 2003;12: 530 4. 45. de Laat B, Derksen RH, De Groot PG. High-avidity anti-beta glycoprotein I antibodies highly correlate with thrombosis in contrast to low-avidity anti-beta glycoprotein I antibodies, J Thromb Haemost 2006;1619 21. 46. Pengo V, Ruffatti A, Legnani C, Gresele P, Barcellona D, Erba N, et al. Clinical course of high risk patients diagnosed with antiphospholipid syndrome (APS). J Thromb Haemost 2010;8:237 42. 47. Pengo V, Biasiolo A, Pegoraro C, Cucchini U, Noventa F, Iliceto S. Antibody profiles for the diagnosis of antiphospholipid syndrome. Thromb Haemost 2005;93:114752. 48. Ruffatti A, Tonello M, Del Ross T, Cavazzana A, Grava C, Noventa F, et al. Antibody profile and clinical course in primary antiphospholipid syndrome with pregnancy morbidity. Thromb Haemost 2006;96:337 41. 49. Swadzba J, Iwaniec T, Szczeklik A, Musial J. Revised classification criteria for antiphospholipid syndrome and the thrombotic risk in patients with autoimmune disease. J Thromb Haemost

Clinical Chemistry 56:6 (2010) 939

Review
2007;5:18839. 50. Pengo V, Biasiolo A, Gresele P, Marongiu F, Erba N, Veschi F, et al. A comparison of lupus anticoagulant-positive patients with clinical picture of antiphospholipid syndrome and those without. Arterioscler Thromb Vasc Biol 2007;27: 309 10. 51. Pengo V, Biasiolo A, Gresele P, Marongiu F, Erba N, Veschi F, et al., on behalf of participating Centres of Italian Federation of Thrombosis Centres (FCSA). Survey of lupus anticoagulant diagnosis by central evaluation of positive plasma samples. J Thromb Haemost 2007;5:92530. 52. Male C, Foulon D, Hoogendoom H, Vegh P, Silverman E, David M, et al. Predictive value of persistent versus transient antiphospholipid antibody subtypes for the risk of thrombotic events in paediatric patients with systemic lupus erythematosus. Blood 2005;106:4152 8. 53. Pourat O, Jollit C, Gombert J, Boinot C, Pierre F. Clinical relevance of the recent update of the classification criteria for definite antiphospholipid syndrome: an obstetric medicine clinic series of 107 patients. J Thromb Haemost 2006;4:2276 7. 54. Bobba RS, Johnson SR, Davis AM. A review of the Sapporo and the revised Sapporo criteria for the classification of antiphospholipid syndrome: where do the revised Sapporo criteria add value? J Rheumatol 2007;34:15227. 55. Devreese K, Hoylaerts MF. Laboratory diagnosis of the antiphospholipid syndrome: a plethora of obstacles to overcome. Eur J Haematol 2009;83: 116. 56. Pierangeli SS, Harris EN. A protocol for determination of anticardiolipin antibodies by ELISA. Nat Protoc 2008;3:840 8. 57. Galli M, Luciani D, Bertolini G, Barbui T. Lupus anticoagulants are stronger risk factors for thrombosis than anticardiolipin antibodies in the antiphospholipid syndrome: a systematic review of the literature. Blood 2003;101:182732. 58. Opatrny L, David M, Kahn SR, Shrier I, Rey E. Association between antiphospholipid antibodies and recurrent fetal loss in women without autoimmune disease: a meta-analysis. J Rheumatol 2006;33:2214 21. 59. Sikara MP, Routsias JG, Samiotaki M, Panayotou G, Moutsopoulos HM, Vlachoyiannopoulos PG. 2 Glycoprotein I ( 2GPI) binds platelet factor 4 (PF4): implications for the pathogenesis of antiphospholipid syndrome. Blood 2010;115:71323. 60. Romay-Penabad Z, Montiel-Manzano MG, Shilagard T, Papalardo E, Vargas G, Deora AB, et al. Annexin A2 is involved in antiphospholipid antibody-mediated pathogenic effects in vitro and in vivo. Blood 2009;114:3074 83. 61. Dunoyer-Geindre S, Kwak BR, Pelli G, Roth I, Satta N, Fish RJ, et al. Immunization of LDL receptor-deficient mice with beta2-glycoprotein I or human serum albumin induces a more inflammatory phenotype in atherosclerotic plaques. Thromb Haemost 2007;97:129 38. 62. Tincani A, Allegri F, Balestrieri G, Reber G, Sanmarco M, Meroni P, et al. Minimal requirements for antiphospholipid antibodies ELISAs proposed by the European Forum on antiphospholipid antibodies. Thromb Res 2004;114:553 8. 63. de Laat B, Derksen RHWM, Urbanus RT, de Groot PG. IgG antibodies that recognize epitope Gly40Arg43 in domain I of 2-glycoprotein I cause LAC, and their presence correlates strongly with thrombosis. Blood 2005;105:1540 5. 64. de Laat B, Pengo V, Pabinger I, Musial J, Voskuyl AE, Bultink IE, et al. The association between circulating antibodies against domain I of beta2glycoprotein I and thrombosis: an international multicenter study. J Thromb Haemost 2009;7: 176773. 65. Oku K, Atsumi T, Amengual O, Koike T. Antiprothrombin antibody testing: detection and clinical utility. Semin Thromb Hemost 2008;34:3359. 66. Shoenfeld Y, Twig G, Katz U, Sherer Y. Autoantibody explosion in antiphospholipid syndrome. J Autoimmun 2008;30:74 83. 67. Galli M, Borrelli G, Jacobsen EM, Marfisi RM, Finazzi G, Marchioli R, et al. Clinical significance of different antiphospholipid antibodies in the WAPS (warfarin in the antiphospholipid syndrome) study. Blood 2007;110:1178 83. 68. Tebo AE, Jaskowski TD, Phansalkar AR, Litwin CM, Branch DW, Hill HR. Diagnostic performance of phospholipid-specific assays for the evaluation of antiphospholipid syndrome. Am J Clin Pathol 2008;129:870 5. 69. Forasteiro R, Martinuzzo M, Pombo G, Puente D, Rossi A, Celebrin L, et al. A prospective study of antibodies to beta2-glycoprotein I and prothrombin, and risk of thrombosis. J Thromb Haemost 2005;3:1231 8. 70. Danowski A, Kickler TS, Petri M. Anti-beta2glycoprotein I: prevalence, clinical correlations, and importance of persistent positivity in patients with antiphospholipid syndrome and systemic lupus erythematosus. J Rheumatol 2006;33: 17759. Kumar S, Papalardo E, Sunkureddi P, Najam S, Gonzalez EB, Pierangeli SS. Isolated elevation of IgA anti-beta2glycoprotein I antibodies with manifestations of antiphospholipid syndrome: a case series of five patients. Lupus 2009;18: 1011 4. Brandt JT, Triplett DA, Alving B, Scharrer I. Criteria for diagnosis of lupus anticoagulants: an update. On behalf of the Subcommittee on Lupus Anticoagulant/Antiphospholipid Antibody of the Scientific and Standardisation Committee of the ISTH. Thromb Haemost 1995;74:118590. Devreese K. Interpretation of normal plasma mixing studies in the laboratory diagnosis of lupus anticoagulants. Thromb Res 2007;119:369 76. Devreese K. No more mixing tests required for integrated assay systems in the laboratory diagnosis of lupus anticoagulants? J Thromb Haemost [Epub ahead of print 2010 Feb 11]. Devreese K, Peerlinck K, Arnout J, Hoylaerts MF. Laboratory detection of the antiphospholipid syndrome via calibrated automated thrombography. Thromb Haemost 2009;101:18596. Simmelink MJ, Derksen RHWM, Arnout J, de Groot PG. A simple method to discriminate between beta2-glycoprotein I- and prothrombindependent lupus anticoagulants. J Thromb Haemost 2003;1:740 7. Pengo V, Biasiolo A, Pegoraro C, Iliceto S. A two-step coagulation test to identify anti- 2glycoprotein I lupus anticoagulants. J Thromb Haemost 2004;2:7027. Devreese KMJ. A functional coagulation test to identify anti- 2-glycoprotein I dependent lupus anticoagulants. Thromb Res 2007;119:7539. Garcia D, Khamashta M, Crowther M. How we diagnose and treat manifestations of the antiphospholipid syndrome: a case-based review. Blood 2007;110:31227. Cervera R, Khamashta MA, Shoenfeld Y, Camps MT, Jacobsen S, Kiss E, et al., Euro-Phospholipid Project Group (European Forum on Antiphospholipid Antibodies). Morbidity and mortality in the antiphospholipid syndrome during a 5-year period: a multicentre prospective study of 1000 patients. Ann Rheum Dis 2009;68:1428 32.

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

940 Clinical Chemistry 56:6 (2010)