Antiphospholipid syndrome: A diagnostic challenge

Abstract

The antiphospholipid syndrome (APS) is an acquired autoimmune thrombophilic disorder that is characterized by thrombosis (venous, arterial and microvascular) and obstetric morbidity due to a diverse family of antibodies against phospholipid-binding proteins present in plasma. The term antiphospholipid antibody is actually a misnomer as the antibodies are not against the phospholipid per se, but target the plasma protein co-factors, which bind to anionic PLs. The exact etiology has not been elucidated and is multifactorial.

The initial guidelines for the diagnosis of APS were laid down in Sapporo, 1999, which were subsequently revised as the Sydney Consensus Conference criteria in 2006. Major changes were the inclusion of β2GPI as independent laboratory criteria, addition of ischemic stroke and transient cerebral ischemia as established clinical criteria and the requirement of repeating the test after 12 weeks. The laboratory tests recommended are coagulation assays, which study the effect of lupus anticoagulant on the clotting time and immunological assays, mostly ELISAs to detect IgG and IgM antibodies against cardiolipin and/or β2 glycoprotein I. For the diagnosis of APS, at least one clinical criterion and one laboratory criterion should be present. Limitations pertaining to the standardization, reproducibility and robustness of the currently recommended diagnostic tests still remain.

Despite elaborate guidelines and syndrome defining criteria, the diagnosis of APS still remains a challenge. A greater interaction between the clinicians and the laboratory professionals is necessary for arriving at the correct diagnosis as a misdiagnosis of APS can have grave consequences.

Introduction

The antiphospholipid syndrome (APS) is an acquired autoimmune thrombophilic disorder characterised by venous/arterial thrombosis and pregnancy morbidity due to a diverse family of antibodies against phospholipid-binding plasma proteins. Although the scientific community has come a long way in understanding APS since its first description in 1983,¹ correct and timely diagnosis of this entity still remains a challenge.

Historical overview

The journey of antiphospholipid antibodies (APLA) began with the discovery of a complement fixation assay for syphilis in 1907 by Wasserman et al.² In 1941 Pangborn demonstrated that the reagin used by Wasserman was an anionic phospholipid (PL) and renamed it cardiolipin.³ However, as serological screening for syphilis became more widespread, it became evident that these tests were positive in many individuals with no symptoms of the disease. In 1952, Moore and Mohr found these transient false-positive tests to be associated with many other contagious diseases besides syphilis, SLE and other autoimmune disorders.⁴ Almost concurrently, Conley and Hartmann described an acquired circulating inhibitorin 2 patients of SLE.⁵ Bowie et al. in 1963 showed that it led to thrombosis and not bleeding.⁶ Finally the term “lupus anticoagulants” was introduced in 1972 by Feinstein and Rapaportfor antibodies against PLs which prolonged the in-vitro clotting time. Harris and co-workers introduced the first radioimmunoassay for the detection of anticardiolipin (aCL) antibodies in 1983 by as well as the first quantitative ELISA two years later.⁷ Using this ELISA they studied SLE patients and reported an increased incidence of thrombosis and pregnancy morbidity in the subgroup with increased aCL antibodies. Professor Graham RV Hughes, an eminent rheumatologist played a central role in describing this clotting disorder. Thus the entity of anticardiolipin syndrome/antiphospholipid syndrome also known as Hughes Syndrome came into existence in 1983.

In the 1990s various studies showed that the term APLA is actually a misnomer as the antibodies are not against the PL per se, but target the plasma protein co-factors which bind to anionic PLs. Beta2 glycoprotein 1 (β2GPI) is the commonest such autoantigen. Other autoantigens such as prothrombin (PT), annexin V and phosphatidylethanolamine are less commonly found.⁸ Further studies showed that β2GPI-dependent antibodies were seen in autoimmune diseases whereas β2GPI-independent APLA were found in patients with infectious diseases.⁹

Etiology

The exact etiology has not been elucidated and is multifactorial. APS can be primary or secondary to an underlying disease. APLAs have been found in autoimmune diseases (most commonly SLE), infections such as hepatitis C, parvovirus B19, human immunodeficiency virus (HIV), adenovirus, human herpes virus, leprosy and syphilis, after treatment with drugs such as quinine, procainamide, oral contraceptives, phenothiazine and anti-tumor necrosis factor, in association with malignancies and even in otherwise healthy individuals. Epitope mimicry has been postulated as a plausible cause in autoimmune diseases. Genetic and environmental factors have also been shown to determine the occurrence and clinical expression of APS.¹⁰ A small amount of antiphospholipid antibody is normal as it helps in removing dying and damaged cells. However people with APLA manifesting as APS have too much of APLA/abnormal variant of APLA and/or abnormal β2-glycoproteinI.

Pathophysiology

The APLAs have both a procoagulant as well as an anticoagulant effect (Table 1). Eventually, it is still not clear why the procoagulant effect predominates over the anticoagulant effect.

Table 1.

Mechanisms of APLA-mediated thrombosis.

Procoagulant effect	Anticoagulant effect
Inhibits activated protein C pathway	Inhibits activation of factor IX
Up-regulates TF pathway	Inhibits activation of factor X
Inhibits antithrombin activity	Inhibits activation of factor XI
Disrupts annexin V shield on membranes	Inhibition/deficiency of factor XII
Inhibits anticoagulant activity of β2-glycoprotein I	Inhibits activation of prothrombin to thrombin
Inhibits fibrinolysis
Activates endothelial cells
Inhibition of endothelial cell prostacyclin production
Activates and degranulates neutrophils
Enhances expression of adhesion molecules by endothelial cells and adherence of neutrophils and leukocytes to endothelial cells
Potentiates platelet activation and enhances platelet aggregation
Enhanced binding of β2-glycoprotein I to membranes
Enhanced binding of prothrombin to membranes
Acquired protein S deficiency
Interference with the thrombomodulin-protein S-protein C pathway
Complement activation

Ref: 19.

APS represents about 15% of the cases of recurrent pregnancy losses. The obstetric complications due to APLA cannot, however, be explained only by thrombotic or ischemic mechanisms. It is proposed that APLAs bind to phosphatidylserine which is exposed during trophoblast syncytium formation causing inhibition of its proliferation and induction of apoptosis. Annexin V has been observed on the surface of syncytiotrophoblasts and is crucial for placental development and integrity. Anti-annexin antibodies found in APS patients could affect embryo implantation and cause pregnancy morbidity by causing apoptosis of syncytiotrophoblasts and loss of trophoblastic gonadotropin secretion. In addition up-regulated coagulation and placental inflammation has been suggested to explain both recurrent miscarriage and fetal losses.¹¹

Clinical manifestations

The core clinical manifestations of APS are thrombotic (arterial, venous or microvascular thrombosis) and obstetric complications. Contrary to the other conditions predisposing to hypercoagulability, in APS there is no predilection for any territory and thrombosis can affect both the veins and arteries including capillaries. The commonest venous and arterial sites of thrombosis are the deep veins of lower limbs and the cerebral arteries, respectively. Diagnostic clinical criteria do not include superficial venous thrombosis. The other major clinical manifestations of the APS are obstetrical. As per Sydney criteria they include:

• i.Unexplained fetal loss at or after the 10th week of gestation (normal fetal morphology on ultrasonography or direct examination) or
• ii.Eclampsia or severe preeclampsia causing premature birth (<34th week of pregnancy) of one or more morphologically normal fetus or
• iii.Three or more unexplained, consecutive early pregnancy loss (<10th week of pregnancy) with normal maternal and paternal factors.¹²

The clinical symptoms should be confirmed by appropriate imaging or histopathological examination. Apart from thrombosis, chronic inflammation and thrombocytopenia is an important characteristic in APS. Other less common manifestations include skin disorders like livedo reticularis, skin ulcerations, digital gangrene, and subungual splinter hemorrhages, autoimmune hemolytic anemia, valvular heart disease, nephropathy, transverse myelopathy or myelitis and other neurologic manifestations like migraine headaches, multiple sclerosis-like syndrome and chorea that cannot be explained by thrombosis alone. In rare cases, thrombosis can occur in multiple vascular territories, resulting in multiple organ failure, a condition called catastrophic antiphospholipid syndrome.¹² CAPS is diagnosed when there are thromboses in three or more organs developing in less than a week, confirmation by histopathology of small-vessel occlusion, microthrombosis in at least one organ and persistent aPL positivity. However, if a patient has only three out of these four criteria, then the patient is classified as probable CAPS.¹³ The rapid progression in CAPS mandates early recognition and aggressive management. However the challenge lies in differentiating it from other microangiopathies like Thrombotic thrombocytopenic purpura (TTP), Hemolytic uremic syndrome (HUS) and HELLP which have overlapping features. In a patient who has been on heparin for more than a week, CAPS can be confused with heparin induced thrombocytopenia. If not managed timely CAPS can be rapidly fatal.

Laboratory diagnosis

The diagnosis of APS has always been a challenge due the wide plethora of clinical symptoms that the patient can present with, especially atypical or non-classic presentations. For example a patient presenting with transverse myelitis as the only symptom can be referred to a neurophysician who might treat him as a case of multiple sclerosis. Similarly a dermatologist may not suspect APLA as the underlying cause in an isolated dermatological manifestation of this disease. The delay in correct diagnosis and management of APS may predispose the patient to disastrous complications, namely a major arterial or venous thrombotic event. The problem is further compounded by lack of a gold standard test for APS along with issues relating to robustness and standardization of the available tests. Not all specialists are entirely familiar with the myriads of tests which can be performed for APS and the specific pre-analytic requirements for performing these tests. Moreover a single negative test result (especially aCL antibody) cannot rule out APS. A greater interaction between physicians and laboratory professionals is required to enable the pathologist to know the detailed history of the patient along with the medications that he is on, which would affect the test results. A follow up testing is required after 12 weeks and the patient is more likely to comply with his primary physician's advice. Not all physicians are thorough with the interpretation of these test results so it is best that an interactive reporting is done.

The initial guidelines for diagnosis of APS were laid down in Sapporo, Japan, (1999). These have been subsequently revised and promulgated as the Sydney Consensus Conference criteria, 2006. The important differences in both these criteria are the inclusion of β2GPI as independent laboratory criteria, quantification of the aCL antibody titres required for diagnosis, addition of ischemic stroke and transient cerebral ischemia as established clinical criteria and the requirement of repeating the test after 12 weeks in the Sydney criteria instead of 6 weeks in Sapporo criteria. The essential serological criteria as per Sapporo and Sydney criteria are shown in Table 2.

Table 2.

Comparison of laboratory criteria for APS.

	Sapporo criteria	Sydney criteria
LAC	Screening, mixing and confirmation tests (ISTH guidelines)	Screening, mixing and confirmation tests (ISTH guidelines)
	Two or more occasions at least 6 weeks apart	Two or more occasions at least 12 weeks apart
aCL antibodies	Detected by standardized β2GP I dependent ELISA	Detected by standardized ELISA
	IgG and/or IgM	IgG and/or IgM
	Medium or high titer	Medium or high titer >40 units IgG/IgM phospholipid antibody titer or >99th percentile
	Two or more occasions at least 6 weeks apart	Two or more occasions at least 12 weeks apart
Anti β2GP I antibody		IgG and/or IgM antibody titer >99th percentile
		Two or more occasions at least 12 weeks apart

Ref. 12.

There are two broad categories of assays that are used to diagnose APS: coagulation assays which study the effect of LAC on the clotting time and immunological assays, mostly ELISAs to detect IgG and IgM antibodies against cardiolipin and/or β₂ glycoprotein I. For diagnosis of APS, at least one clinical criteria and one laboratory criteria (Sydney) should be present.

Screening asymptomatic individuals to identify those at high risk of thrombotic or gestational complications, results in false positives, which has been reported to be between 3 and 20%.13, 14 In patients, in whom if there is a clinical event that is suspected to be due to APLA, a thorough medical history should be taken along with a detailed physical examination and selected laboratory testing, which are necessary to make the diagnosis. Depending on the antibody profile, patients can be classified into various categories according to Sydney consensus statement: category I: two or more laboratory criteria positive in any combination; IIa: only lupus anticoagulant positive; IIb: only anti-cardiolipin (IgG and/or IgM) positive; IIc: only anti-β2 glycoprotein I (IgG and/or IgM) positive.¹² The importance of differentiating patients with more than one laboratory criterion was emphasized in the Sydney Consensus Conference 2006.

Lupus anticoagulant testing

An accurate laboratory test result for APLA is must, as a false positive test result can expose the patient to unwarranted anti-coagulant therapy and pose a risk of bleeding. Samples should not be collected during acute episode. If the patient is using anticoagulants, it is desirable to use the aCL and anti-β2GPI tests, which are not affected by the use of these drugs. In patients taking vit K antagonists, LAC can be tested using undiluted patient plasma if the INR is less than 1.5. Test can be performed after diluting the test sample 1:1 with normal pooled plasma (NPP) if the INR is between 1.5 and 3.0. However patients on unfractionated heparin should not be tested for LAC. Low molecular weight heparin has a lesser effect on LAC particularly dilute Russel's viper venom time (dRVVT) testing. Newer anticoagulants such as argatroban and dabigatran (direct thrombin inhibitors) interfere with all LAC assays and give a significant proportion of false positive results.¹⁵

The International Society on Haemostasis and Thrombosis (ISTH 2009), the British Committee for standards in Haematology (BCSH 2012) and the Clinical and Laboratory Standards Institute (CLSI 2014) have all given guidelines for LAC detection.¹⁵ There is a general consensus among guidelines on methodology for sample preparation, performance of two or more tests with different assay principles for screening, demonstration of PL dependent prolongation and use of normalized ratios, calculations for the same. The cut-offs in the ISTH guidelines differ from the BCSH and the CLSI guidelines (99th versus 97.5th percentile). The ISTH and the BCSH guidelines define a three-step strategy for the detection of LAC:

• i)Screening test: demonstrate prolongation of a PL dependant clotting time further than the upper limit of reference interval.
• ii)Mixing test: confirmation of the presence of an inhibitor and absence of a clotting factor deficiency.
• iii)Confirmation that the inhibitor is PL dependent and not against a specific clotting factor.

The CLSI guideline does not necessitate a mixing study in all cases.

The blood samples of patients should be collected in 3.2% sodium citrate vacutainers in 1:9 ratio (anticoagulant: venous whole blood). The sample should be processed by double centrifugation to get platelet poor plasma (platelet count <10 × 10⁹ L⁻¹) as PLs present on their surface, can neutralize part of the APLAs present, compromising the test result. The filtration of the sample is not recommended as it causes reduction of the levels of von Willebrand factor. Before performing LAC assays, a thrombin time or anti-factor Xa assay should always be performed to exclude the presence of heparin. The preferred tests for screening are dRVVT and activated partial thromboplastin time (aPTT) using low concentration of PLs. A local reference range should ideally be established for each test performed. The mixing study involves combining patients’ plasma with NPP (1:1) and then studying clotting time. For confirmation, platelet neutralization procedure, use of modified aPTT reagent containing hexagonal (II) phase PLs or use of LA-insensitive PL can be done.

Solid phase APLA assays

Anti cardiolipin antibodies

Commercially available IgG & IgM ELISA coated with cardiolipin and β2GPI are used for the detection of aCL antibodies in the presence of bovine serum. IgG and IgM antibody titers should be expressed in international standardized GPL and MPL units respectively. The problem with aCL testing is a wide inter and intra kit variability, which is seen more often in low titer antibodies. As per the Sydney criteria a positive result is defined as a medium or high titer (i.e. >40 units GPL/MPL or >99th percentile) which has increased the test specificity.¹²

The role of aCL testing as a diagnostic criterion for APS is currently under debate. However, exclusion of aCL testing can lead to failure to diagnose APS in a small fraction of patients. Since the test is highly sensitive, it can be used as a screening test and the results should be read in conjunction with the other diagnostic tests and clinical features of the patient.

Anti-β2GPI antibodies

β2GPI is a PL-binding apolipoprotein which exists as open S type or closed loop conformation. The β2GPI structure is divided into five domains. In the closed type, cryptic epitope on domain 1 is hidden preventing autoantibodies (aAbs) from binding. When β2GPI is immobilized on negatively charged PL surfaces via domain V, it changes to open form-opening the cryptic domain1 to which the aAbs bind. It is the epitope 1 of β2GPI which is the target in APS patients. Domain V has been related to patients with leprosy and childhood atopic dermatitis. The drawback with anti-β2GPI ELISA is that all antibodies reactive with β2GPI including the non-pathogenic and PL independent ones also bind during the assay which is why a positive result may not always correlate with the clinical picture. To offset this issue, assays employing recombinant domain I of β2GPI have been developed. The antibody titer should be >99th percentile for the results to be positive. Further, these antibodies can be of low or high avidity. It is the latter which are related to thrombosis.¹⁶

Other APLA

Assays for other autoantigens such as PT, annexin V and phosphatidylethanolamine, glycerol, serine, inositol also exist. The clinical utility of these assays remains to be validated by large prospective studies.

Clinical considerations of the APS antibody profile

Many studies have reported a strong association between antibody profile showing positivity on multiple assays (LAC, aCL and β2GPI-ELISA) and clinical manifestations of APS, both thrombotic and obstetric. It is speculated that such patients would have higher levels of β2GPI antibodies compared to patients with single test positivity. Also the effect of multiple distinct pathogenic antibodies may be additive. Considering the antibody isotypes, it has been seen that it is the IgG isotype in both aCL and β2GPI-ELISA that is associated with venous and arterial thromboses.⁸

LAC has been more consistently associated with thrombosis in both venous and arterial circulation, as compared to aCL and β2GPI-ELISA.¹⁷ In the obstetric population also, LAC has been found to have a stronger association with recurrent pregnancy loss than other antibodies. However aCL ELISA was found to be associated with both early (<13 weeks) and late (>24 weeks) pregnancy loss, whereas both LAC and β2GPI-ELISA were associated with late gestation miscarriages.¹⁸

Conclusion

Since the time it has been defined, APS has held the attention of both clinicians and pathologists alike. Despite elaborate guidelines and syndrome defining criteria, the diagnosis of APS still remains a challenge. Limitations pertaining to the standardization, reproducibility and robustness of the currently recommended diagnostic tests still remain. Larger prospective trials will be required to study the significance of the newer laboratory parameters. Documented clinical symptoms along with either/combination of the three diagnostic tests (LAC, β2GPI-ELISA and aCL) should be used for arriving at the correct diagnosis as a misdiagnosis of APS can have grave consequences.

Conflicts of interest

The authors have none to declare.