APLS Article

8/4/2019 APLS Article

1/38

Antiphospholipid syndrome (APL syndrome)can be called acquired antibodymediatedthrombophilia from the clinical point of view.The hypercoagulability/thrombotic diathesis of

APL syndrome can manifest in a wide variety ofclinical and/or laboratory abnormal states. Theexpanding heterogeneous spectrum of the disor-der varies from a coincidental abnormality in thehemostasis laboratory to thromboembolic eventssuch as a serious stroke, venous thrombosis, ar-terial occlusion, or a catastrophic obstetrical

complication in the clinic (1).There are many obscurities challenging a bet-

ter understanding of APL syndrome and APL an-tibodyassociated pathologic states. Difficulties

are mainly present in the confusing terminologydescribing the disorders and antibodies, prob-lems in the hemostasis laboratory and their clin-ical relevance, still ongoing revisions in the diag-nostic criteria, heterogeneous patient presenta-tions, chaotic multisystemic involvement, unsat-isfactory treatment strategies, and, of course, theunresolved enigmatic distinct etiopathogenesisof the APL syndrome. The aim of this review is tooutline current dilemmas and their present clar-ifications in the wide clinicopathologic spectrum

of APL syndrome and APL antibodyrelated dis-tinct pathologic conditions.

DIFFICULTIES IN THE CURRENT

TERMINOLOGY OF THE APL SYNDROME

The term APL antibodies represents a het-erogeneous group of antibodies associated with aprothrombotic state and/or a condition with high

89

State-of-the-Art Review

Current Debates in Antiphospholipid Syndrome:The Acquired AntibodyMediated Thrombophilia

M. Akif ztrk, MD,* Ibrahim C. Haznedaroglu, MD, Mehmet Turgut, MD,

and Hakan Gker, MD

*Gazi University School of Medicine Department of Rheumatology, Hacettepe University School of MedicineDepartment of Hematology, Ankara, Turkey

Summary:Antiphospholipid (APL) syndrome is the most commonform of acquired thrombophilia. It can cause significant morbidityand even mortality. The term APL antibodies represents a hetero-geneous group of antibodies associated with this disorder. Currentlyno single assay can identify every APL antibody. Clinically relevantAPL antibodies are mainly anticardiolipin antibodies (ACA) detectedby solid phase enzyme-linked immunosorbent assay (ELISA) andlupus anticoagulants (LA) demonstrated by in vitro coagulationassay. However, there are some other antibodies associated with theAPL syndrome (i.e., subgroup APL antibodies). ACAs, LAs, and sub-group APL antibodies represent intersecting, but non-identical, sub-sets of autoantibodies. Thus, those autoantibodies may coexist ormay occur independently. Any organ system and any size of vesselcan be affected during the clinical course of the disease. Therefore,

the APL syndrome can manifest itself in a wide variety of clinicalthrombotic features. Fetal loss and pregnancy morbidity represent aspecific challenge. Despite tremendous advances in the understand-ing of the pathogenesis of APL syndrome during the past decade, themainstay of management is still anticoagulation. However, there isno general agreement regarding the duration and intensity of anti-coagulant therapy. In this review, we focused on the current dilem-mas and their present clarifications in the wide clinicopathologicspectrum of APL syndrome and APL antibodyrelated distinct patho-logic conditions.Key Words: Antiphospholipid antibody syndromeAntiphospho-lipid antibodiesLupus anticoagulantsAnticardiolipin antibod-iesTerminologyClinical featuresLaboratory investigationPathobiologyManagement.

Clin Appl Thrombosis/Hemostasis 10(2):89126, 20042004 Westminster Publications, Inc., Glen Head, NY

Address correspondence and reprint requests to Dr. M. Akifztrk, Ostim mahallesi 89. sokak, AK-84 sitesi A2 blok no:8, Yenimahalle, Ankara TR-06160, Turkey;e-mail: [email protected].


2/38

risk of recurrent fetal loss. Currently no singleassay can identify every APL antibody. While thepresence of those antibodies is detected with aphospholipid-dependent coagulation reaction,

they are designated as lupus anticoagulants(LA). When they are demonstrated with the solidphase enzyme-linked immunosorbent assay(ELISA) in which microplate wells are coated

with cardiolipin, they are called anticardiolipinantibodies (ACA). Therefore, ACAs are detect-ed by their antigenic specificity irrespective oftheir functional properties. On the other hand,LAs are detected by their effect on in vitro coag-ulation irrespective of their immunologic target.

Despite the famous historical name of the syn-drome, the real antigenic target for APL antibod-ies is not phospholipid itself but some phospho-lipid binding plasma proteins, mainly beta

2gly-

coprotein-1 (2GP-1) and prothrombin. 2GP-1 isrequired for the binding of autoimmune ACAs tocardiolipin. Most LAs depend on the presence ofprothrombin or

2GP-1 directed against anionic

phospholipids (18). Furthermore, the anionicphospholipids probably play important roles forthe binding of APL antibodies to those phospho-lipid binding plasma proteins in in vivo condi-tions. It is not clear whether these antibodies rec-ognize the proteins themselves, the phospho-lipid-protein complex, or neo- or cryptic antigenseventually expressed as a result of the interac-tions. Another hypothesis is that close contact tothe negatively charged surface makes clusters,providing higher antigenic density for the bind-ing of the APL antibodies that have low affinityfor binding under normal conditions (3,713).Because APL antibodies represent a heteroge-nous group, each of those hypotheses could beoperative in different patients.

ACAs and LAs represent two intersecting, butnon-identical, subsets of autoantibodies. Thus,LACs and ACAs may coexist or may occur inde-pendently (1416). From the clinical point of

view, there are some patients with recurrent ve-nous and/or arterial thromboses without APLantibodies via with those routine assays in the

absence of other autoimmune diseases. How-ever they have positivity for the more recentlyrecognized subgroups of APL antibodies (i.e.antibodies against

2GP-1, prothrombin, or an-

nexin-V, and antibodies against phospholipidsthemselves other than cardiolipin includingphosphatidylserine, phosphatidylethanolamine,phosphatidylglycerol, phosphatidylinositol, andphosphatidylcholine) (1720). The term APLantibody describes either of those previously

mentioned antibodies, whereas the term APLsyndrome designates the presence of either ofthose APL antibodies together with the throm-botic manifestations and/or recurrent fetal loss

and/or thrombocytopenia (Fig. 1).The term LA is improper for two major as-pects. First, the LA antibodies are not neces-sarily specific to patients with systemic lupuserythematosus (SLE). Many patients sufferingfrom recurrent thrombotic events or obstetricalcomplications and are positive for LA do notmanifest any of the major clinical or serologicfeatures of SLE. Second, LAs prolong the in vitrocoagulation assays but actually demonstrate pro-coagulant, not anticoagulant, effects in in vivoconditions.

APL antibodies can be associated with clinicalfeatures of APL syndrome without any finding of

an underlying connective tissue disorder (i.e.,primary APL syndrome). APL syndrome can alsocomplicate the clinical course of various connec-tive tissue disorders, most commonly SLE(18,2128), as well as some other autoimmunedisorders such as Sjgrens syndrome, rheuma-toid arthritis, systemic sclerosis, systemic vas-culitis, or dermatomyositis (23). APL syndrome,however, more commonly develops in patients

without any underlying disorder (18,23,2629).Presence of LA or ACAs generally defines two

distinct patient populations, each of which is as-sociated with an increased risk of thrombosis(15,18). However, there appears to be an intenseoverlap between those disorders, challenging

widespread use of the LAs and APL antibodies to-gether. Therefore, we prefer to present those twodisorders together for the sake of clarity in thispaper.

Infection- or Drug-Related APL AntibodiesAPL antibodies are reported to occur in some

infections such as syphilis, malaria, infectiousmononucleosis, tuberculosis, human immunode-ficiency virus, and hepatitis A, B, and C (3032).Infection-related ACAs tend to be transient, oflow titre, and directly bind to the cardiolipin it-

self; i.e., they are not 2GP-1 dependent, andmoreover could be inhibited by

2GP-1 (30,31).

In these conditions, there is generally no appar-ently increased risk for developing manifesta-tions of APL syndrome (1,32), although certain

viral infections can precede the development ofAPL syndrome (3336).

Some drugs are associated with APL antibod-ies and LA activity. Procainamide hydrochloride,quinidine sulfate, phenytoin sodium, chlorpro-

M. AKIFZTRK ET AL90


3/38

mazine hydrochloride, valproic acid, amoxicillin,hydralazine hydrochloride, and propranolol hy-drochloride are all reported for this association.Those drug-induced APL antibodies were associ-ated with thrombotic complications in contrast tothe majority of infectious-related ACAs (15,18).

PATHOBIOLOGY AND GENESIS OF APL

SYNDROME: AN ENIGMA FROM

MOLECULES TO DISEASE

Pathobiology and genesis of the APL syndromestill remain undetermined. However, there have

been enormous efforts and progress for our un-derstanding the molecular pathogenesis of the

APL syndrome in the past decade.The role of APL antibodies as the primary

cause of the syndrome could be questionable, atleast for some patients. APL antibodies are in-creased in patients with autoimmune hemolyticanemia (37). The APL antibodies, thus, could beinnocent bystanders generated secondary toantibody-mediated cell injury in some disorders.

Another possibility is that APL antibodies couldact as a second hit, and could be both an effectand a cause of thrombosis. In that hypotheticalmodel, anionic phospholipids that are exposedduring blood clotting could trigger immunologi-cal recognition and the formation of APL anti-bodies, promoting a vicious cycle through theirthrombogenic properties (38). Finally, the APLantibodies could play direct causal roles in thegeneration of the clinical features of APL syn-drome. There are data supporting this final hy-pothesis. The majority of antigenic structuresthat are targets for the APL antibodies do playroles in the normal hemostasis. Moreover, those

APL antibodies precede the thrombotic events,and the risk of developing the clinical manifes-tations of the APL syndrome correlates directly

with the level of the APL antibodies (3946).Furthermore, passive transfer of immunoglobu-lins from patients with APL syndrome generatesfeatures of APL syndrome in animal models(4753). Therefore, rather than being merely anepiphenomenon, APL antibodies seem to play di-rect roles in the pathogenesis of APL syndrome.

CURRENT DEBATES IN ANTIPHOSPHOLIPID SYNDROME 91

FIG. 1. Classical diagram representing the intersecting subsets of antiphos-

pholipid antibodies. Subgroup of antiphospholipid antibodies are antibodies

against 2-GP-1, prothrombin, or annexin-V, and antibodies against phospho-

lipids themselves other than cardiolipin including phosphatidylserine, phos-

phatidylethanolamine, phosphatidylglycerol, phosphatidylinositol, and

phosphatidylcholine.


4/38

Although the exact pathogenetic mecha-nism(s) remain to be determined, it is likely thatmore than one mechanism could be operative forthe development of the clinical manifestations

because a wide spectrum of autoantibodies havebeen shown to be associated with the APL syn-drome. Thrombotic events in patients with APLsyndrome segregate into venous or arterial vas-cular tree. Arterial events are generally followedby arterial events and venous events by venousevents in the majority of the cases in APL syn-drome (54). Therefore, molecular mechanismsof thrombosis could differ in different clinicalmanifestations. For instance, venous thromboticmanifestations may be attributed to coagulationfactorassociated mechanisms (i.e., secondaryhemostasis), whereas disorders of platelets (i.e.,primary hemostasis) might be responsible for ar-

terial thromboses.The

2GP-1 protein is the most frequently ob-

served protein cofactor in the APL syndrome.However little is known regarding the physiolog-ic role(s) of this protein. The

2GP-1 has been

identified as a constituent of chylomicrons, and very-low-density and high-density lipoproteins.Roughly 40% of the circulating

2GP-1 is bound

to lipoproteins. Therefore this protein may be in-volved in lipoprotein metabolism (55).

2GP-1

may also play a scavenging role in mediating theclearance of foreign particles and apoptotic cellsin the circulation (5658).

2GP-1 was also sug-

gested as a natural anticoagulant. Circulatinglevels of

2GP-1 are reduced dramatically during

disseminated intravascular coagulation (DIC)along with reductions in other anticoagulant pro-teins including protein C and antithrombin III.These data suggest specific consumption of the2GP-1 during in vivo coagulation (59). A num-

ber of in vitro studies also supported the antico-agulant role of the

2GP-1 molecule.

2GP-1 can

inhibit various platelet functions (60,61), down-regulate the activation of the Hageman factorand the contact phase system of the coagulationcascade (62,63), prevent the binding of protein S

with its plasma inhibitor, the C4b-binding pro-

tein (64), and inhibit tissue factor activity (65).However, the clinical relevance of those in vitroobservations remains questionable becausehereditary deficiency of

2GP-1 does not seem to

be associated with thrombophilia (66). The2GP-1 molecule can bind to negatively charged

macromolecules that are involved in the initia-tion of coagulation. However, its binding affinityis far below that of many other phospholipid-binding proteins (67). Therefore, it seems un-

likely that the binding of2GP-1 can neutralize

the initiation of coagulation cascade. Likewise,anti-prothrombin antibodies do not interfere

with the coagulant properties of the prothrombin

molecule itself (68). Both anti-2GP-1 antibodiesand anti-prothrombin antibodies can display LAactivity (2,46,68). Hence, other than direct in-hibition of function, binding of those antibodiesto their targets should have some common dy-namic influences on the coagulation cascade.

Disruption of the Annexin V Shieldin APL Syndrome

Annexin V is found in a variety of tissues in-cluding the vascular endothelium and placenta.This protein could have potent anticoagulantproperties based on its high affinity for anionicphospholipids. The capacity of annexin V to dis-

place coagulation factors from phospholipid sur-faces is evident (69,70). A number of consecu-tive experiments supported this hypothesis. Thedecrements in the levels of annexin V induced by

APL antibodies were accompanied by a shorten-ing of the coagulation time of plasma. Likewiseincubating endothelial cells with antiannexin Vresulted in faster coagulation of plasma.Moreover, removing annexin V from the cell sur-face by calcium chelator EGTA significantly ac-celerated the coagulation of plasma. Adding ex-ogenous annexin V increased the coagulationtime of plasma applied to the cells (69). APL an-tibodies directly reduced binding of annexin-V tononcellular anionic phospholipid-coated surfacesin a

2GP-1dependent manner interfering with

its anticoagulant activity, resulting in the accel-eration of coagulation (70). Inhibition of annex-in V binding to procoagulant phospholipid sur-faces by APL antibodies was dependent uponanti-

2GP-1 antibodies. This reduction is signifi-

cantly correlated with clinical thrombosis (71).The hypothesis depending on these data is usedto explain the in vivo procoagulant and in vitroanticoagulant effects of APL antibodies. Based onthe hypothesis, annexin V forms a crystal latticeover the anionic phospholipid surface that serves

and shields it from the availability for the assem-bly of the phospholipid-dependent coagulationcomplexes (i.e. tissue factor, factor X, factor IX,factor V, etc.) under physiologic conditions.

Annexin V, thus, has a potent anticoagulant ef-fect in vivo. During the absence of the annexin

V, the complexes of APL antibodies and theirantigenic targets such as

2GP-1 or prothrombin,

bind the phospholipid bilayer, reducing the ac-cess of coagulation factors to anionic phospho-

M. AKIFZTRK ET AL92


5/38

lipids in vitro. Because there is a limiting quanti-ty of anionic phospholipids, this interaction re-sults in the prolongation of the coagulationassay. However, in in vivo conditions where

there is sufficient amount of annexin V, the APLantibodies disrupt the ability of annexin V toform ordered crystals on the phospholipid sur-face and inhibit the anticoagulant effect of an-nexin V. That state results in a net increase of theamount of anionic phospholipid available forpromoting coagulation reactions, which estab-lishes a hypercoagulable state (38,70). This at-tractive hypothesis could link the paradoxical LAphenomenon and the in vivo thrombogenic ten-dency of the APL antibodies. However, there issome conflicting data. In a Russells viper venomassay, co-incubation of annexin V with APL anti-body containing plasma failed to correct the clot-

ting time, but even caused prolongation of theclotting time. The observed clotting time exceedsthat with either annexin V or APL containingplasma alone (72). Moreover, annexin V binding

was unaffected by the presence of ACA-2GP-1

complexes. Annexin V pre-adsorbed to the bilay-ers completely prevented adsorption of ACA-2GP-1 complexes. None of the pre-adsorbed an-

nexin V was displaced by ACA-2GP-1 complexes

(73). Therefore, displacement of the annexin Vshield is not a prerequisite for the thromboticevents in patients with APL syndrome.

Inhibition of the Protein C AnticoagulantPathway in APL Syndrome

The protein C anticoagulant pathway is amajor regulatory mechanism of the coagulationcascade. Upon binding of thrombin to thrombo-modulin, the thrombin molecule loses its proco-agulant properties and cleaves protein C.

Activated protein C then complexes with proteinS and proteolyzes coagulation factors Va and

VIIIa (74). A missense mutation in the factor Vgene (1691 GA) (Leiden mutation) is a com-mon cause of thrombotic tendency by causing anabnormal factor V product. However, factor VLeiden mutation is not common in patients with

APL syndrome (75). On the other hand, some pa-tients with APL syndrome and asymptomatic car-riers of APL antibodies may have acquired freeprotein S deficiency (76,77). Immunoglobulinfractions isolated from patients with APL causesacquired activated protein C resistance (78,79).The presence of LAs can interfere with the pro-tein C pathway (8083). Other APL antibodies,including anti-

2GP-1 antibodies and ACAs, were

also associated with the acquired activated pro-

tein C resistance phenomenon (8486). ACAsbound to protein C in the presence of cardiolipinand

2GP-1, but not in the absence of either

2GP-1 or cardiolipin. These data suggest that

protein C could be a target of ACA by making acomplex with cardiolipin and 2GP-1, leading to

protein C dysfunction (87). Autoantibodies gen-erated against the isolated components of theprotein C pathway (i.e. thrombomodulin, proteinC, or protein S) may also impair the protein Cpathway (88,89). Different steps in the protein Canticoagulant pathway could be targets for APLantibodies. They can inhibit thrombin formation.

APL antibodies interfere with the protein C acti-vation via thrombin-thrombomodulin complex.Inhibition of the protein C complex assemblytakes place. Protein C activity can be downregu-lated directly or via inhibition of the activity of its

cofactor, protein S. Finally, the antibodies direct-ed against the substrates of APC, factors Va and

VIIIa may protect them from inactivation (90,91).A subpopulation of APL antibodies selectively

inhibits the activated protein C complex as a re-sult of differences in the phospholipid require-ments of this complex as compared to those ofthe procoagulant complexes. Phosphatidyl-ethanolamine supports activated protein C anti-coagulant activity but has little influence on pro-thrombin activation. The inhibitory effect of LAson activated protein C pathway was also en-hanced in the presence of phosphatidylethano-lamine (92). The rate of inhibition of the antico-agulant pathway (i.e., inactivation of factor Va)is much higher than the inhibition of the proco-agulant pathway (i.e., thrombin generation) inthe presence of phosphatidylethanolamine in thesystem (92). Therefore, differences betweenmembrane phospholipid requirement of the anti-coagulant and procoagulant reactions could ac-count for the selective inhibition of one pathwayover the other. The variability of different LA as-says may be attributed to the differences in thephospholipid reagents inside them (93). Why theautoantibodies directed against the phospho-lipids themselves can cause hypercoagulability is

possibly explained via this hypothesis.Taken together, those observations suggest

that inhibition of the activated protein C antico-agulation system may be a common mechanismcontributing to the prothrombotic state of APLantibodies in at least a subgroup of patients.

APA and EndotheliumVascular endothelial cells play a critical pro-

tective role in the whole body defense against



6/38

thrombosis. Other than being a mechanical bar-rier, the endothelium plays an active role in thephysiologic regulation of hemostasis throughsynthesis of numerous procoagulant and antico-

agulant mediators. Immunologic injury to en-dothelial cells and alterations of the endothelialcell functions could have a place in the patho-genesis of the APL syndrome. Binding of APL an-tibodies suppressed vasodilator prostacyclin re-lease by vascular endothelium. This inhibitiongenerates an imbalance between vasodilator and

vasoconstrictor prostoglandins toward the vaso-constrictor state (94). Likewise, urinary excre-tion of the major thromboxane metabolite ofplatelet origin (11-dehydro-thromboxane B2) is

very significantly increased while the urinarymetabolite reflecting the vascular production ofprostacyclin (2,3-dinor-6-keto-prostaglandin F1

alpha) is much less enhanced (95). LAs induceapoptosis in endothelial cells (96,97). Apoptoticcells expose phosphatidylserine on the outer sur-face of the plasma membrane, further activatingcoagulation cascades (98). APL antibodies inducethe production of monocyte chemoattractant pro-tein-1 in human umbilical vein endothelial cellsin a

2GP-1dependent manner, which might in

turn promote thrombus formation either by en-hancing the influx of monocytes to the endothe-lium or by increasing tissue factor expression inmonocytes (99). Binding of APL antibodies to theendothelial cell surface generates a functionalcell activation and a procoagulant state. APL an-tibodies induce upregulation of certain cell ad-hesion molecules on the surface of endothelialcells. E-selectin, vascular cell adhesion molecule-1 (VCAM-1), and intracellular adhesion mole-cule-1 (ICAM-1) are mediated through

2GP-1

(100103). Anti-2GP-1 antibodies also signifi-

cantly increase the secretion of proinflammatorycytokine IL-6 and augment prostacyclin metabo-lism by endothelial cells in a dose-dependentmanner (100). APL antibodies increase leukocyteadhesion to the endothelium and enhancethrombus formation as demonstrated in an in

vivo model of leukocyte adhesion and microcir-

culation. This increment was accompanied by en-hanced adhesion molecule expression as an in

vitro marker of endothelial cell activation (104).Circulating VCAM-1 is increased in patients withprimary APL syndrome, which is further evidentin patients with repeated thrombotic events(105). APL antibody induced endothelial activa-tion and enhanced thrombosis is impaired inICAM-deficient and ICAM-1/P-selectin deficientmice. These data demonstrate that pathogenic ef-

fects of APL antibodies could be mediated bythose cell adhesion molecules (106). Moreover,leukocyte adhesion to endothelial cells was sig-nificantly decreased after infusion of anti-VCAM-

1 antibodies (106). Taken together, those datastrongly indicate that pathologic activation of theendothelium may contribute to the hypercoagu-lable state of APL syndrome.

The Tissue Factor Pathway in APL SyndromeTissue factor (TF) is the principle (patho)bio-

logic initiator of the coagulation cascade.Endothelial cells and monocytes can generate tis-sue factor upon stimulation by various sub-stances. Tissue factor is inducibly synthesized inendothelium and monocytes (107).

Tissue factor pathway is upregulated in theAPL syndrome. Plasma tissue factor is higher in

patients with APL syndrome compared to healthypeople and patients with thrombosis without APLantibodies (108,109). Cell surface expression oftissue factor and tissue factor mRNA levels arealso increased in the monocytes of APL syn-drome. These data suggest that circulatingmonocytes in APL syndrome express increasedamounts of tissue factor (109,110). Moreover,

APL antibodies induce tissue factor generationand enhance tissue factor activity in monocytesand vascular endothelial cells (108,111113). Onthe other hand, APL antibodies can also interfere

with the inhibitory activity of tissue factor path-way inhibitor in a group of patients (114,115),despite the circulating tissue factor pathway in-hibitor levels were increased in APL syndrome(108). The increased levels of tissue factor path-

way inhibitor could either reflect a compensato-ry response or endothelial activation or damage(108). Therefore, enhancement of the tissue fac-tor pathway either by increasing tissue factorgeneration and activity or by interfering with itsinhibition can be an important mechanism thatleads to activation of coagulation and thrombusformation in the APL syndrome.

Molecular Mimicry as a Potential Triggering

Mechanism in APL SyndromeImmunization with phospholipids alone does

not induce the production of APL antibodies(116). However, immunization of normal miceand rabbits with purified

2GP-1 induced pro-

duction of high levels of APL antibodies in addi-tion to antibodies against

2GP-1 (117119).

Characteristics of those 2GP-1induced APL an-

tibodies were similar to those of the human au-toimmune APL antibodies (118,119). Moreover,

M. AKIFZTRK ET AL94


7/38

immunization with 2GP-1 induced the develop-

ment of the clinical and histopathologic featuresof APL syndrome in PL/J mice, which are genet-ically predisposed to autoimmune disease (120).

A major PL-binding site of2GP-1, calledGDKV, induced production of pathogenic APL an-tibodies alone without the rest of the

2GP-1

molecule (121). Some common human virusesshared structural similarities with this phospho-lipid-binding region of

2GP-1, the GDKV. Those

peptides showed greater degrees of binding to PLcompared to GDKV. Interestingly, immunizationof mice with those peptides caused production ofhigh levels of APL antibodies (122). A CMV pep-tide-induced monoclonal APL antibody showedsignificantly high binding to cardiolipin and vary-ing degrees of binding to other phospholipids inthe presence of

2GP-1. Injection of those mono-

clonal APL antibodies in mice resulted in a sig-nificant increase in the number of leukocytes ad-hering to endothelial cells and enhanced throm-bus formation in vivo. These data suggest thatthose antibodies could be pathogenic (123).Certain viral infections can precede the develop-ment of APL syndrome (3236). Taken together,those experiments proposed that the tolerance toself

2GP-1 could be broken after certain micro-

bial exposure by molecular mimicry. This mech-anism could be operative in at least a subgroupof APL patients (124). A wide variety of diseasesand immunologic alterations affect the endothe-lial phospholipids. These effects promote thepathogenicity of otherwise harmless APL anti-bodies. Therefore hypercoagulability is unpre-dictable in APL syndrome. Likewise healthypeople with APL antibodies could remain asymp-tomatic for many years.

Infection-related APL antibodies are usuallytransient and do not offer an increased risk forthrombotic events (1,32). The affinity of autoim-mune APL antibodies for their protein antigensin the circulation is often at most moderate,

which is also a characteristic of natural autoanti-bodies (125). Indeed, patients with antibodiesagainst either

2GP-1 or prothrombin do usually

have normal circulating levels of these proteins.Therefore, some other authors concluded thatthe origin of pathogenic autoimmune APL anti-bodies resides in deregulated overproduction ofnatural antibodies rather than in antigen-drivenformation of immune antibodies (125).

APL Antibodies and ApoptosisAnionic phospholipids are almost exclusively

located in the intracellular surface of cell mem-

branes, whereas cardiolipin is located primarilyon the mitochondial membranes. Therefore, APLantibodies cannot react with their targets undernormal conditions. Redistribution of the anionic

phospholipids from the intracellular to the ex-tracellular compartment can occur in certainphysiologic conditions such as platelet activationor apoptosis (98,126,127). Cardiolipin molecules

were expressed on the surface of apoptotic cellsand were recognized by APL antibodies (128).2GP-1 bound selectively to the surface of apop-

totic cells (57,127,129). The binding of2GP-1

to the surface of the apoptotic cells generated anepitope for the binding of APL autoantibodies(127). Heterologous human

2GP-1 bound to the

surface of apoptotic cells induced the productionof ACAs and LA activity in non-autoimmune mice(130). Importantly, neither apoptotic cells nor

2GP-1 alone could be able to induce generationof ACAs after intravenous injection (130).

Subcutaneous or intradermal administrationof the heterologous

2GP-1 molecule alone could

also induce generation of APL antibodies(5,117119). Slow absorption of the

2GP-1 over

several days may induce such a phenomenon.During the sustained absorption time,

2GP-1

could bind to apoptotic cells of the host. This in-teraction could be induced by the generation oflocal inflammatory response secondary to co-ad-ministration of the adjuvant (namely CFA, com-plete Freunds adjuvant) (130). Therefore apop-totic cell-bound

2

GP-1 appears to be a true im-munogen for the production of APL antibodies.2GP-1 complexed with native cardiolipin is

structurally altered. Likewise, intravenous im-munization of mice with

2GP-1 in the presence

of cardiolipin vesicles induced a high level ofanti-

2GP-1 antibodies, ACAs, and LA activity

(131). On the other hand, binding of APL anti-bodies to apoptotic cells can occur in both in a2GP-1dependent and independent way (132).

Taken together, these data provide evidence thatexposure of phospholipids during apoptosis maybe an early event in the apoptotic cellular pro-gram leading to specific interactions with circu-

lating phospholipid-binding plasma proteins suchas

2GP-1. The binding of the

2GP-1 molecule

to those phospholipids exposed on the apoptoticcells could be immunogenic. That phenomenoncan break tolerance and trigger the generationof APL antibodies. Once an immune response isinitiated against such an immunodominant epi-tope, the immunologic response can expance toother epitopes within the complex by the processof epitope spreading. As the end, the APL anti-



8/38

body response eventually includes autoantibod-ies against

2GP-1 or phospholipids alone, or

against a new epitope generated after the inter-action between those molecules (130,131,133).

Role of Lipid Peroxidation in the Initiationand/or Propagation of APL Syndrome

Oxidative injury has been suggested in thepathogenesis of a variety of disorders including

vasculitic syndromes and atherosclerosis. Like- wise, a number of experiments provided evi-dence that oxidative damage could have a rolein the pathobiology of the APL syndrome. Nativecardiolipin, the naturally occurring form of car-diolipin, is highly susceptible to oxidation, par-ticularly upon exposure to air. In contrast, hy-drogenated cardiolipin is unable to undergo lipidperoxidation. Cardiolipin is highly sensitive to

lipid peroxidation under conditions of solid-phase immunoassay. Both sera and affinity puri-fied ACA IgG from APL syndrome patients boundto oxidized cardiolipin, but did not bind to hy-drogenated cardiolipin. These data suggest that

APL antibodies are directed at neoepitopes gen-erated when cardiolipin undergoes the processof peroxidation (134). Moreover,

2GP-1 was

recognized by the APL sera only when bound tooxidized cardiolipin, but not when bound to thereduced cardiolipin analog unable to undergo ox-idation (135). ACAs could bind to

2GP-1 only

when it was bound to a microtiter plate that waspreviously irradiated, a condition that introducesoxygen radicals (9). These data support the hy-pothesis that a simple phosholipid-

2GP-1 com-

plex is not sufficient for the generation of APLantibodies unless the PL first undergoes oxida-tion. Therefore, many of the APL antibodiescould actually be antibodies directed against ox-idized phospholipids (134,135).

Oxidative stress is a potential component ofthe final common pathway leading to apoptosis.On the other hand, cells sustained progressivelipid peroxidation following an apoptotic signal(136,137). Therefore, the apoptotic cell surfaceis a site of increased oxidative activity. Immuno-

genicity of apoptotic cell-bound 2GP-1 depend-ed upon prolonged interaction between

2GP-1

and the apoptotic cell surface (130). Hence bothapoptosis and oxidative injury may play a dualrole in the pathogenesis of APL antibodies.

Antiplatelet Antibodies in APL SyndromeThere is some evidence suggesting that APL

autoantibodies can cross-react with platelets andcan affect both their quantity and function.

Platelet activation induces the exposure of an-ionic phospholipids on the outer leaflet of themembrane (126,138). Both ACAs and LAs boundto thrombin activated, but not unactivated rest-

ing, platelets (139). The most likely binding sitewas phospholipids because there was no cross-reactivity of these antibodies with GPIIb/IIIa(139). Likewise,

2GP-1 preferentially bound to

activated platelets. Binding of ACAs to plateletsurface is also

2GP-1dependent (138,139).

Moreover, some APL antibodies can induceplatelet activation and aggregation (140145).Urinary excretion of the platelet-derived throm-boxane metabolite 11-dehydro-thromboxane B2

was significantly increased in patients with APLantibodies (140). Activated platelets were de-tected by flow cytometry in the majority of APLsyndrome patients with neurologic involvement

(146). In contrast, although APL antibodies didbind to activated platelets, they may fail to in-duce platelet activation (139,147). Therefore,the role of platelet activation in the pathobiologyof thrombosis in APL syndrome is not universal.Platelets circulating in an over-activation statein APL syndrome could either be a cause or aneffect of the thrombosis or vascular injury(148).

Specific anti-platelet antibodies directedagainst major platelet membrane glycoproteins

were higher in APL syndrome patients withthrombocytopenia compared to patients withnormal platelet counts (149,150). Those anti-platelet antibodies did not demonstrate cross-re-activity with ACAs (149). Therefore, anti-plateletantibodies could represent a distinct group ofauto-antibodies other than APL antibodies. Theycould play a significant role in the thrombocy-topenia observed in APL syndrome. However,

whether binding of these anti-platelet antibodiesmay play a pathogenic role in thrombosis on thebasis of hyperaggregability remains obscure.

Heparin-induced thrombocytopenia (HIT) and APL syndrome have similar clinical presenta-tions. Enhanced platelet activation at the vascu-lar wall and endothelial injury are similar in both

HIT and APL. Deposition of the immune com-plexes on slightly activated platelets further in-duces cellular activation. Likewise, generation ofmicrovesicles provides much larger phospholipidsurface area and results in enhanced thrombingeneration (125,151,152). In a recent mild radi-cal-induced injury model in hamsters, anti-

2GP-1

antibodies were mainly localized within theplatelet thrombus. This observation suggests thatafter a slight endothelial damage (first hit), acti-

M. AKIFZTRK ET AL96


9/38

vated platelets exposed negatively charged phos-pholipids, leading to deposition of APL antibodycomplexes perpetuating the platelet activationand thrombus growth (second hit) (125,153).

Mechanisms of Pregnancy Loss in APL SyndromeIntervillous thrombosis, intravillous infarc-

tions, and decidual vasculopathy disturbing pla-cental circulation were proposed as the pathobi-ologic basis of recurrent miscarriages in women

with APL syndrome (154,155). The previously-mentioned hypotheses regarding the prothrom-botic state of APL syndrome could contribute tothe placental thrombosis. The disruption of theannexin V shield hypothesis could offer a logicalexplanation for the placental thrombosis (69).

Annexin V is normally located on the apical sur-face of syncytiotrophoblasts lining the placental

villi (156,157). This protein is constitutively pro-duced to maintain the fluidity of the intervillouscirculation and consequently to maintain fetal vi-ability (156,157). Infusion of anti-annexin V an-tibodies caused placental thrombosis, necrosis,and fetal loss in animal models (157).Interestingly the concentrations of this placentalanticoagulant factor are markedly reduced onplacental villi in APL patients with spontaneousrecurrent abortion (158). Moreover, APL anti-bodies decreased the levels of villous surface an-nexin-V on placental villi (158,159). APL antibod-ies both reduced the levels of annexin V and alsoaccelerated the coagulation of plasma on culturedtrophoblasts and endothelial cells (69). Therefore,deficiency of placental surface annexin-V causedby APL antibodies may complicate the placentalthrombosis observed in those patients (69,158).In contrast, ACA-

2GP-1 complexes failed to dis-

place preadsorbed annexin V from lipid bilayersin another study (73). No difference was observedregarding the intensity of immunostaining for an-nexin V between placentas from women with APLsyndrome and those from control subjects (160).Therefore, placental thrombosis cannot be solelyattributed to the displacement of the annexin Vshield, and other prothrombotic mechanisms

could be operative as well.On the other hand, placental infarction is not

present in all the placentae of patients with APLsyndrome. Furthermore, although placentalthrombosis could be responsible for late fetallosses, early miscarriages cannot be explained byplacental thrombosis (155,161). Recent reportshave suggested that APL antibodies could inducetrophoblast dysfunction such as decreased pla-cental hormone production or trophoblast inva-

sion. In a choriocarcinoma model of pregnancy,phosphatidylserine exposed during syncytiumformation was targeted by APL antibodies. Theyinhibited the intercytotrophoblast fusion process

and trophoblast invasiveness and secretion ofhuman chorionic gonadotropin (159). APL anti-bodycontaining sera suppressed human chori-onic gonadotropin secretion by cultured tro-phoblast cells (162). Anti-annexin V antibody in-duced trophoblast apoptosis and significantly re-duced trophoblast gonadotropin secretion (163).Moreover,

2GP-1dependent APL antibodies as

well as APL antibodies against the phospholipidsthemselves attacked the trophoblasts, and bothgroups of antibodies negatively affected tro-phoblast implantation and development (164).In this regard, APL antibodies can affect tro-phoblast gonadotropin secretion and invasive-

ness by binding directly to anionic phospholipidsand through adhered

2GP-1. Alternatively they

can induce apoptosis and lead to the defectiveplacentation in APL syndrome (159,163,164).

Establishment of the human placenta requiresinvasiveness of fetal cytotrophoblast stem cellsin anchoring chorionic villi. Trophoblast inva-siveness and differentiation are dependent on ad-herence to the extracellular matrix, response toexternal cytokine signals, and expression and thealteration of adhesion proteins. During normalcytotrophoblast differentiation along the invasivepathway, the differentiating cytotrophoblastsdramatically transform their adhesion receptorphenotype. These data suggest that a unique ad-hesion phenotype switch is required for the suc-cessful endovascular invasion and normal pla-centation (165,166). The effects of APL antibod-ies on trophoblast adhesion molecules (alpha1and alpha5 integrins, E- and VE-cadherins) wereinvestigated in primary cytotrophoblast cell cul-tures. APL antibodies modulated the trophoblastcadherins repertoire in vitro, suggesting that re-current pregnancy loss in APL syndrome couldbe associated with abnormal cytotrophoblast ex-pression of adhesion molecules (167).

The Role of Genetic Predispositionin APL Syndrome

Familial clustering of increased APL antibodylevels has been described (168). Moreover, cer-tain polymorphisms of the

2GP-1 molecule were

associated with the presence of anti-2GP-1 anti-

bodies and APL syndrome (169,170). Associa-tions with HLA DR4, DR5, and HLA-DQ7 anti-gens and primary APL syndrome were reported(171174). Frequency of HLA-DR7 was in-



10/38

creased in SLE patients with ACAs (173,175).However, those studies were carried out in a lim-ited number of patients and genetic marker(s)for APL syndrome have not yet been demon-

strated in large cohorts of subjects from differ-ent ethnic populations.

Other Thrombophilic Triggers Complicatingthe Genesis of APL Syndrome

The fibrin polymerization rate is increased inpatients with APL syndrome (176). Fibrinolyticactivity could be suppressed by APL antibodies.Impaired fibrinolysis could be either due to theincrements in PAI-1 activity or to the inhibitionof factor XII activation together with the inhibi-tion of XIIa (177,178). APL antibodies also in-hibit heparin-accelerated formation of an-tithrombin III-thrombin complexes (179).

In conclusion, for the thrombophilia mecha-nism of APL syndrome, several different and/orrelated pathogenic mechanisms could be opera-tive concurrently, and even in the same patientsto induce the generation of the features of APLantibodies. Each of the previously mentionedmechanisms could be in part operative duringthe pathobiologic course of APL syndrome.Moreover, it is likely that those mechanisms aretightly integrated and can display complex dualinteractions as a vicious cycle. For instance, while

apoptosis induces generation of APL antibodies,LAs can trigger endothelial cell apoptosis.

Apoptosis is able to induce oxidative damage.Oxidative damage is a well-known inducer for

apoptosis. APL antibodies can induce pathologicactivation and/or injury to platelets and en-dothelium. The activation of those cells causesexposure of anionic phospholipids that can sec-ondarily amplify the generation of APL antibod-ies. Therefore, complex interactions of biologicstructures and molecules are operative in thegenesis of APL syndrome. A list of proposedmechanisms in the pathobiology of APL syn-drome is given in Table 1.

LABORATORY DIAGNOSIS OF THE

ANTIBODIES IN THE APL SYNDROME

The laboratory diagnosis of APL syndrome de-pends on the detection of autoantibodies direct-ed against anionic phospholipids and/or a varietyof phospholipid binding proteins. Solid-phase im-munoassays and phospholipid-dependent tests ofhemostasis are used for this purpose (21,67,180182). There still remain some problems in thedetection, standardization, clinical utility, rele-

vance, and significance in the laboratory diagno-sis of APL syndrome.

M. AKIFZTRK ET AL98

TABLE 1. Proposed Mechanisms in the Complicated Pathobiology of APL Syndrome

Proposed Mechanism Reference

Disruption of the annexin V shield 38, 69, 70

Inhibition of the protein C anticoagulant pathway 7890

Endothelial injury and/or pathological activation 9497, 99106

Alterations in the tissue factor pathway 108115

Molecular mimicry 122124

Deregulated overproduction of natural antibodies 125

Induction of apoptosis 130, 132, 133

Oxidative injury and lipid peroxydation 134, 135

Platelet activation 140146

Anti-platelet antibodies 149, 15

Genetic predisposition 168175

Induction of trophoblast dysfunction and/or trophoblast apoptosis in pregnancy 159, 162164

Modulation of cytotrophoblast expression of adhesion molecules in pregnancy 167

Increased fibrin polymerization rate 176

Impaired fibrinolytic activity 177, 178

Inhibition of formation of antithrombin III-thrombin complexes 179


11/38

Solid-Phase Tests of APLSyndromeAssociated Antibodies

Solid-phase tests are developed and widelyused to detect antibodies against

2GP-1 (anti-

2GP-1), cardiolipin (ACA), prothrombin, annex-in V, phosphatidylserine, phosphatidyletanola-mine, and many others of which their numbersare growing (180,183). Therefore, more specificmeasurements of antibodies associated with APLsyndrome are currently available via solid-phaseassays (180). A detailed summary regarding thecurrent knowledge and clinical applications ofnovel antigenic targets in APL syndrome is givenin Heterogeneous clinical and laboratory mani-festations of the patient(s) with APL syndrome: alogical approach for the use of various APL anti-bodies section of this review. Nevertheless, thereare many complicated problems restricting the

predictive ability of current diagnostic tests interms of the lack of standardization, repro-ducibility, and lack of prospective and multivari-ate epidemiologic analysis (21). The inter-labo-ratory coefficient of variation for IgG APL units

was found to be higher than 70%, though thewithin-run coefficient of variation had a medianvalue of 10% in the First French Anticardiolipin Antibodies Standardization Workshop (184).ACA assay is currently a routine standard in thediagnosis of APL syndrome. However, criticallaboratory tests should be repeated at least oncebefore reaching a clinical decision in APL syn-drome (185). Another study focusing on ACAassay usefulness suggested exceedingly high in-terlaboratory result variation, combined with ageneral lack of test result grading consensus andmethod-based variation (185). In that study,cross-laboratory testing of samples yielded inter-laboratory coefficients of variation for lgG ACAand IgM ACA that were higher than 50% incases. General consensus (interlaboratory agree-ment, 90% or more) was obtained in approxi-mately 40% of cases only. Different laboratoriesusually failed to agree on whether a sample was

ACA-positive or ACA-negative. Distinct methodsin various laboratories tended toward higher or

lower ACA values (185). Hence, the laboratorydata obtained by ACA assays should be inter-preted together with the clinical presentation ofthe patients, as discussed in various sections ofthis review.

Phospholipid-Dependent Hemostatic Tests ofAPL Syndrome-Associated Antibodies

The term LAs denotes some of the APL anti-bodies that prolong phospholipid-dependent clot-

ting reactions in vitro. The main critical functionof hemostasis laboratory in the diagnosis of APLsyndrome is to measure LA. Essentially, a phos-pholipid-dependent screening assay such as acti-

vated partial thromboplastin time (aPTT) is per-formed to test LA. If this screening test is pro-longed, normal plasma is added to the testedsample in a 1:1 mixture. If the LA is present, ad-dition of normal plasma does not correct the pro-longed assay. The next step for LA detection isto add excess phospholipid to the tested sample.Naturally, addition of those phospholipids doescorrect the prolonged assay. In cases of coagula-tion factor deficiency, addition of normal plasmacorrects the prolonged assay at the second step ofthe experiment. In the presence of anti-coagula-tion factor inhibitors, prolonged assay is nevercorrected in all steps of the tests. Specific mea-

surements of anti-factor inhibitors are mandato-ry for the diagnosis of those cases (Fig. 2) (186).The sensitivity of numerous screening or confir-matory assays for the diagnosis of LA may vary.

Attention to the ratio of tested patient plasma tonormal plasma is important to detect a circulat-ing inhibitor especially in a minimally prolongedaPTT. Furthermore, the normal plasma sourcemust be platelet poor to maximize sensitivityin the case of a weak LA (187). Dilute Russell

viper venom time (dRVVT) is an important con-firmatory test for the LA. Russell viper venom di-rectly activates factor X. Therefore, dRVVT is notaffected from the intrinsic factor deficiencies be-cause they are bypassed. Thus, dRVVT is moresensitive than prolonged aPTT for the detectionof LA (188) with varying reference ranges (189).The sensitivity and specificity of commercialreagents for the detection of LA may exhibitmarked differences in distinct coagulometers(190). Several tests are developed for the diag-nosis of LA, including platelet neutralization pro-cedures (191), platelet-derived microvesiclebased tests (192), Textarin/Ecarin ratio (193),hexagonal phase phospholipid neutralizationassay (194), silica clotting time (195), kaolinclotting time (196), tissue thromboplastin inhi-

bition test (197), dilute prothrombin timebasedlupus ratio test (198), Taipan snake venom time(199), and lupus ratio test, a mixture of a lupus-sensitive and a lupus-insensitive aPTT-reagent

with normal plasma (200). The use of platelets inconfirmatory tests may lead to false-positive re-sults especially in anti-factor V and heparin-likeinhibitors (201).

A normal aPTT result does not exclude thepresence of LA. In approximately half of APL syn-



12/38

drome cases, aPTT is prolonged. LA testing, onthe other hand, is the most difficult task in theeveryday practice of the hemostasis laboratory.Not only the enigmatic clinical course of the APLsyndrome, but also certain technical issues com-plicate the detection of LA (201). Contamination

with platelets, instrumental effects, pH drift dur-ing incubation, and difficulties in standardizationrepresent some of the problems for LA laborato-ry identification (201). Anti-

2GP-1 and anti-pro-

thrombin monoclonal antibodies offered betterunderstanding of the mechanism by which LA

prolongs in vitro clotting. LA-positive monoclon-al antibodies also improved the LA diagnosis(67). Mixing studies are widely used to identifyLA or any other circulating coagulation inhibitor(187). There are some important variables incase of mixing. The normal plasma sourceshould be platelet-poor to maximize sensitivityas mentioned previously. The patient plasma tonormal plasma ratio is extremely important, par-ticularly in the suspected APL syndrome case of a

minimally prolonged aPTT (187). A hemostatictest should be optimally combined for the detec-tion of LA or exclusion of a coagulation inhibitor.Twenty-one patients who initially presented witha prolonged prothrombin time (PT) or aPTT orboth were tested for the presence of LA in a study(202). The authors used a battery of coagulationtests, including both immediate and 2-hour mix-ing studies, a platelet neutralization procedure, atissue thromboplastin inhibition test, and dRVVT.Ten percent of the patients only had a prolongedPT, 33% only had a prolonged aPTT, and in 57%

both test results were abnormal. In 15 patients,inhibition was evident on immediate assay ofequal-volume mixture studies of patient plasmaand normal pooled plasma. However, in somepatients, it was evident only after a 2-hour incu-bation period. Therefore, proper incubation isimportant in mixing studies for LA or inhibitordetection. Fifteen of 18 samples showed correc-tion of the abnormal screening study whenplatelets were used as a source of phospholipid.

M. AKIFZTRK ET AL100

FIG. 2. Simplified practical schema for the laboratory diagnosis of antiphospholipid syndrome. APL: antiphos-

pholipid, ACA: anticardiolipin antibody, LA: lupus anticoagulant, ELISA: enzyme-linked immunosorbent assay.


13/38


14/38

Some cases may lead to multiinfarct dementia.Other stroke risk factors such as cigarette smok-ing or hyperlipidemia may further complicate therisk of recurrent ischemic events in patients with

APL syndrome (29). A less common cerebral vas-cular thrombotic manifestation was cerebral sinusthrombosis (1,23,212214). Other less commonneurologic manifestations of APL syndrome aremigraine-type headache, chorea, transversemyelopathy, myasthenia gravis, Guillain Barrsyndrome, transient global amnesia, seizures,motor neuron disease, and depression. Howevertheir associations with APL syndrome have been

less well-determined (29,212214). Cerebral dys-function in APL syndrome may range from mildcognitive dysfunction to severe dementia. Retinaland choroidal vessel involvement may causeacute ocular ischemia and transient blurred visionor amaurosis fugax, transient diplopia, and de-creased vision. Transient field loss may also de-

velop (1,29,212216). Sudden sensorineuralhearing loss has been rarely reported in APL syn-drome (217).

Cardiac Features in APL SyndromeThe most common cardiac manifestation of

APL syndrome is valvular pathologies. They in-clude verrucous endocarditis, valvular thicken-ing, and insufficiency. Valvular stenosis israrely seen. The mitral valve is the most com-monly affected site, followed by the aortic

valve, although the tricuspid or pulmonaryvalve can also be affected (1,23,214,218220).A strong association between valvular abnor-malities and arterial brain infarcts in APL syn-drome patients does exist. This association re-flects the embolization from the damaged valve

(219). Patients with APL syndrome have in-creased risk for myocardial infarction, in-creased risk for graft occlusion after coronaryartery bypass surgery, and restenosis after an-gioplasty (18,26,214,221). APL antibodies mayrarely cause myocardial infarction even in thepresence of normal coronary arteries (222). Inaddition, there have been some reports of in-tracardiac thrombi or cardiomyopathy in asso-ciation with the APL syndrome (23,218).


FIG. 3. Schematic representation of the clinical manifestations in antiphospholipid antibody syndrome.


15/38

Cutaneous Lesions in APL SyndromeCutaneous lesions appear in approximately

one third of patients with APL syndrome. Livedoreticularis is by far the most common cutaneous

manifestation. It is frequently the initial clinicalfeature of the syndrome (23,223). Other cuta-neous findings of the APL syndrome includenecrotizing vasculitis, livedoid vasculitis, throm-bophlebitis, cutaneous ulceration, necrosis, ery-thematous macules, purpura, ecchymoses,painful skin nodules, and subungual splinter he-morrhages (214,223).

Pulmonary Involvement in APL SyndromeThe most common pulmonary complications

of APL syndrome are pulmonary thromboem-bolism and pulmonary hypertension (23,214,224). Pulmonary embolism occurs in approxi-

mately 14% to 30% of the patients. It may be theinitial feature in approximately 9% of the pa-tients (23,214). Thrombosis in calf veins, inferi-or vena cava, tricuspid valve vegetations, andright-sided intracardiac thrombosis may be thesource of pulmonary emboli (214). Recurrentpulmonary embolism may lead to thromboem-bolic pulmonary hypertension. Severe cases withpulmonary hypertension may be accompanied byisolated tricuspid valve insufficiency (224).Several cases ofprimary (non-thromboembol-ic) pulmonary hypertension complicating prima-ry APL syndrome have been described. The out-come in patients with pulmonary hypertensionand APL syndrome is usually fatal (224). Rarecases with diffuse alveolar hemorrhage, fibros-ing alveolitis, adult respiratory distress syn-drome, and pulmonary artery thrombosis haveall been reported (23,214,224).

Renal Involvement in APL SyndromeRenal manifestations of the APL syndrome are

relatively rare, occurring in 2.7% of all cases(23). Vasoocclusive events can occur at any levelof the renal vasculature. The disease may affectthe main renal artery and its branches, arterioles,glomerular capillaries, and renal veins.

Characteristic histologic findings of renal APLsyndrome are intrarenal vascular lesions, arte-riosclerosis, fibrous intimal hyperplasia, partialor complete vascular occlusions, and thromboticmicroangiopathy. Focal cortical atrophy is an-other finding. Inflammatory vascular lesions aretypically absent (225). These histopathologicfindings are clinically manifested by hyperten-sion, acute or chronic renal insufficiency, pro-teinuria, nephrotic syndrome, and inconsistent

hematuria in the majority of the APL syndromepatients (1,214,225,226). Among those manifes-tations, hypertension is of particular importancebecause it is almost invariably present among pa-

tients with renal APL syndrome (225). Mild, severe, or even malignant hypertension may bepresent. All patients with APL syndrome and hy-pertension should be investigated for renal in-

volvement (especially renal artery stenosis).Hypertension is often the only early clinical man-ifestation of renal APL syndrome. Other nephro-logic manifestations of APL syndrome include he-modialysis vascular access thrombosis, and pri-mary graft non-function in renal transplant re-cipients (214,226).

Hematologic Manifestations in APL SyndromeThrombocytopenia in the APL syndrome is a

frequent finding, occurring in 20% to 30% of pa-tients (23,214). It is usually chronic and mild,and does not cause a severe bleeding complica-tion (23,214). On the other hand, elevated levelsof APL antibodies are common in immune throm-bocytopenic purpura, and the persistent presenceof APL antibodies is an important risk for the de-

velopment of thrombosis or fetal loss (227).Autoimmune hemolytic anemia may develop

in some patients with APL syndrome. The directCoombs test result is positive in 14% of patients

with primary APL syndrome and 40% of patientswith APL syndrome secondary to SLE. However,hemolytic anemia is not common in those pa-tients (214). Autoimmune hemolytic anemia ismore common in secondary APL syndrome (23).DIC, hemolytic-uremic syndrome and thrombot-ic thrombocytopenic purpura are uncommoncomplications of APL syndrome. However, theyrepresent the drastic rare clinical manifestationcalled catastrophic APL syndrome (1).

Other Manifestations of APL SyndromeNasal septum perforation, pancreas infarction,

splenic infarction, Budd-Chiari syndrome,Addisons disease due to adrenal infarction, andpituitary failure after necrosis of the pituitary

gland have all been reported in APL syndrome(1,23,228). Musculoskeletal manifestations in-cluding arthralgia or arthritis are very common(38.7% and 27.1%, respectively) (23). Avascularnecrosis of bone may occur in approximately 2%of the patients (23). However, this necrotic com-plication can be detected by magnetic resonanceimaging in approximately one fifth of the pa-tients with primary APL syndrome who did notreceive steroids before (229).



16/38

Obstetric Complications in APL SyndromeAPL syndrome is associated with both fetal

and maternal complications. Fetal complicationsinclude both early abortions before the 10th

week of gestation and fetal deaths at or after the10th week of gestation. Preterm labor at or be-fore the 34th week of gestation may also takeplace (1,230). These obstetric features werespecifically mentioned in the 1999 InternationalConsensus Statement Classification Criteria for

APL syndrome (231). Other obstetric complica-tions include intrauterine growth retardation,placental abruption, pre-eclampsia, and eclamp-sia (1,23,226,230). Association with HELLP syn-drome (hemolytic anemia, elevated liver en-zymes, low platelet count in association with pre-eclampsia) has been rarely reported (1).

Catastrophic APL SyndromeCatastrophic APL syndrome is an acute anddrastic manifestation of the APL syndrome char-acterized by clinical involvement of at least threedifferent organ systems over days or weeks withhistopathologic evidence of multiple occlusions oflarge or small vessels (232). It is fortunately arare complication (less than 1% of all patients

with APL syndrome). The mortality is as high as50% and most patients die as a result of a com-bined cardiac and respiratory failure (232,233).Respiratory failure can develop after acute respi-ratory distress syndrome (ARDS) and diffuse alve-olar hemorrhage. Precipitating factors include in-

fections, trauma, surgery, drug administration,and warfarin withdrawal in 22% of the patients

with catastrophic APL syndrome. The majority ofthose patients develop widespread vascular oc-clusion mainly affecting small vessels of organs,particularly kidney, lungs, brain, heart, adrenalglands, and liver, causing microangiopathy. Onlya minority of these patients experience occlusionof a single large vessel. Digital ischemia withfrank gangrene, superficial skin necrosis, and is-chemic ulceration of the limbs may develop.Thrombocytopenia was reported in 68% of thepatients, hemolytic anemia in 26%, and DIC in28%. They all have the potential to contribute tothe multiorgan thrombotic microangiopathy incatastrophic APL syndrome (232,233).

HETEROGENEOUS CLINICAL AND LABORATORY

MANIFESTATIONS OF THE PATIENT(S) WITH

APL SYNDROME: A LOGICAL APPROACH FOR

THE USE OF VARIOUS APL ANTIBODIES

The spectrum of antigens and autoantibodiesin the APL syndrome is expanding. Conventional

approach for the detection of APL antibodies in-cludes the clotting-based assay LA and solidphase immunoassays for ACAs. However, a sig-nificant number of patients with clinical features

of APL syndrome remain negative for those con-ventional APL assays. Hence, tremendous effortsare being spent to demonstrate novel antigenictargets and to develop novel laboratory assays toimprove the diagnostic yield in the APL syn-drome. However, those efforts in the laboratorymay adversely affect critical clinical decisionmaking in the clinic. Therefore, a hierarchicaland rational approach is necessary for effectiveclinical management of APL syndrome patients.

Since the first documentation in 1983 (234), alarge number of clinical studies documented theassociation between the ACAs and the develop-ment of arterial and/or venous thrombotic events

(4043,235237). ACAs can also significantly in-crease the recurrence risk following an episodeof vascular thrombosis (25,238,239). The risk fordevelopment of APL antibodyrelated disordersis higher in patients with high titer IgG isotype(40,41,44,45,240,241). However, IgM isotypesand low titer ACAs could be associated with anincreased thrombosis risk as well (236,237).Moreover, there is growing evidence for the as-sociation of IgA isotype ACAs and APL anti-bodyrelated disorders (242,243). It is generally,however, not assayed routinely and not suggest-ed as a diagnostic criterion because other authorsfailed to confirm this association (244246).

The association of ACAs with vascular throm-bosis is not, however, universally confirmed(15,46,241,246248). A recent meta-analysisconcluded that only high-titer IgG isotype ACA isclearly associated with thrombosis. Less consis-tent results are available for IgM isotype ACAsand low titer IgG isotype ACAs (241). However,there are significant differences between thosestudies regarding the methods of the research,associated clinical conditions, clinical end-points,laboratory methods for the detection of ACAs,and cut-off levels for the antibodies. Thus, thecomparison of studies is really problematic. The

titers for APL antibodies can fluctuate.Furthermore patients who had initially had neg-ative test results for APL antibodies could havepositive test results for ACAs later (20,249).Therefore, prospective well-designed studies viaserial measurements of ACAs with better inter-laboratory harmonization at certain critical timepoints during the clinical course of APA syn-drome is strictly needed to draw more depend-able conclusions. However, in everyday clinical



17/38


18/38

LA activity than the anti-prothrombin antibodies.They are significantly correlated with the clini-cal manifestations of the APL syndrome whileanti-prothrombin antibodies were not (287,288).

Therefore the place of anti-phosphatidylserine-prthrombin complex antibodies as a marker forAPL syndrome remains to be elucidated.

Annexin V is a tissue and circulating phospho-lipid-binding protein with anticoagulant proper-ties. It has high affinity for anionic phospholipidsand can displace coagulation factors from phos-pholipid surfaces (38). Displacement of the an-nexin V shield by APL antibodies has been pro-posed in the pathogenesis of APL syndrome asdiscussed previously (38). Antibodies directedagainst the annexin V molecule do have LA prop-erties (289). Anti-annexin V antibodies have alsobeen reported in some patients with APL syn-

drome (277,290,291). Moreover anti-prothrom-bin and anti-annexin V antibodies could be morespecific for the diagnosis of APA syndrome (277).Furthermore, anti-annexin V IgG antibodies wereassociated with higher incidences of arterial or

venous thrombosis, intrauterine fetal loss, andprolonged aPTT in patients with SLE (292).However, conflicting data exist regarding the as-sociation of anti-annexin V antibodies and theclinical manifestations of the APL syndrome(291). Hence, strong evidence for the functionaland clinical significance of those autoantibodiesis not yet available.

Low density lipoprotein (LDL) oxidation maycontribute to the development of atherosclerosisin vivo (293,294). Oxidized LDL is more im-munogenic, and led to the generation of autoan-tibodies (295). Antibodies directed against oxi-dized LDL were associated with myocardial in-farction in prospective studies (296). CirculatingLDL contains significant amounts of cardiolipinsuggesting that many ACAs and anti-oxidizedLDL antibodies may be directed at similar epi-topes (297). Moreover, APL antibodies are cross-reactive to oxidized LDL (298). On the otherhand, anti-oxidized LDL antibodies are higher inpatients with APL syndrome compared to healthy

controls and patients with SLE without clinicalfeatures of APL syndrome (299-301). The levelsof anti-oxidized LDL did not show strong correla-tion with those of anti-

2GP-1 or ACAs. Therefore

anti-oxidized LDL may represent a distinct subsetof antibodies (299). Elevated levels of anti-oxi-dized LDL antibodies are potential markers forthe development of arterial thrombosis in patients

with APL syndrome (299,300). However, follow-ing studies failed to demonstrate an association

between high anti-oxidized LDL antibody levelsand arterial thrombosis (301,302). On the con-trary, high anti-oxidized LDL antibody titer wascorrelated with venous thrombosis in another

study (301). Moreover, the association of anti-ox-idized LDL antibodies and primary APL syndromealso remains controversial (301). Therefore, ad-ditional studies are needed to determine the exactclinical relevance of anti-oxidized LDL antibodiesin the APL syndrome.

Pathologic activation of the endothelial cellsby APL antibodies contributes to the prothrom-botic state in APL syndrome as discussed previ-ously. APL antibodies themselves bind to en-dothelial cells (100,303,304). Moreover, a dis-tinct group of endothelial cell reactive antibod-ies other than well-known APL antibodies canoccur in the plasma of patients with APL anti-

bodies (305,306). However, whether those anti-bodies are initiators of the endothelial injury orendothelial cell surface alterations lead to theformation of those antibodies remains to be elu-cidated. Although the presence of those anti-en-dothelial antibodies was associated with a his-tory of thrombotic events (306), the utility ofthe anti-endothelial cell antibodies in the diag-nostic workup of the APL syndrome remains tobe elucidated.

There are additional potential targets for APLantibodies, including protein C and protein S,thrombomodulin, factor XI, factor XII, high-mol-ecular-weight kininogen, low-molecular-weightkininogen, tissue plasminogen activator, comple-ment component C4, complement factor H, co-agulation factor VII/VIIa, antimitochondrial an-tibody type 5, platelet-activating factor, and mal-ondialdehyde-modified lipoprotein(a) (8890,283,307315). However, the association of thoseauto-antibodies and APL syndrome has not beenconfirmed. Accordingly a firm evidence for theclinical use of those antibodies for investigatingthe APL syndrome is currently absent.

Cardiolipin is the most widely used phospho-lipid in the APL antibody assays. There is exten-sive cross reactivity between ACAs and antibod-

ies against other phospholipids (316318).Therefore, serum anti-noncardiolipin antibodiescould be identified by their cross-reactivity with

ACAs. There may be no rational to performELISAs for other phospholipids. However, someof the antibodies against phospholipids otherthan cardiolipin cannot be detected by conven-tional ACA assays. ELISA methods based on car-diolipin as target antigen may not be sensitiveenough to detect all APL antibody-positive sub-



19/38

jects (19,316,319323). Moreover, a significantcorrelation was found between those antibodiesand thrombosis and thrombocytopenia (321,324,325). Recently, a multitarget ELISA assay using a

coating mixture of cardiolipin and non-cardi-olipin phospholipids including phosphatidylinos-itol, phosphatidylserine, and phosphatidylethan-olamine was suggested to enhance the ability toidentify patients with APL syndrome (316). Onthe contrary, testing for APL antibodies otherthan LA and ACA is not clinically useful in theevaluation of APL syndrome (317,326). How-ever, some patients with clinical features for APLsyndrome who are negative for conventional APAassays (ACA and/or LA assays) can be positivefor antibodies againts non-cardiolipin phospho-lipids (316,319323). Therefore, patients whoseclinical symptoms suggest APL syndrome but

whose sera are negative for conventional APA as-says should be screened for antibodies againstnon-cardiolipin phospholipids including phos-phatidylserine, phosphatidylethanolamine, phos-phatidylglycerol, phosphatidylinositol, and phos-phatidylcholine (18,26,28).

In summary, in a patient with unexplainedthrombosis of any type or recurrent fetal lossboth LA and ACA, all three assays should be per-formed. There is general agreement that solid-phase ELISA is the method of choice for the de-tection of ACAs (18,26,28,327). For the detec-tion of LAs, dRVVT seems to be the most sensi-tive assay (18,26,28). The titers for APL anti-bodies can fluctuate and patients who had ini-tially had negative test results for LAs or ACAscould be later positive, or patients who initiallyhad either IgM or low IgG can develop higherlevels of IgG or LAs (20,240,249). Therefore, pa-tients should be re-tested for those assays in caseof strong suspicion (21). For the patients withhigh suspicion for APA syndrome who disclosenegative results for LA and ACA assays, the pres-ence of isolated subgroup antibodies against2GP-1, prothrombin, annexin-V, phosphatidyl-

serine, phosphatidylethanolamine, phosphatidyl-glycerol, phosphatidylinositol, or phosphatidyl-

choline should be tested (18,21,26,28,250).

THE BICK CLASSIFICATION

OF APL ANTIBODYMEDIATED

THROMBOSIS SYNDROMES

Bick has proposed to divide the APL anti-bodymediated thrombosis syndromes into sixsubgroups (18,2628) (Table 2). He stated that

there is little overlap (approximately 10% orless) among these subtypes. Moreover, type IVpatients (mixtures of types I, II, and III) are alsouncommon, with most patients fitting into one

of the first three types. This classification couldbe important for choosing the initial and long-term therapeutic alternatives from the clinicalpoint of view. Recommended antithrombotictherapy regimens for each subgroup is given inTable 2 and also discussed in Clinical strategiesagainst APL syndrome section.

DIAGNOSTIC CRITERIA OF APL SYNDROME:

A STILL ONGOING PROCESS

The diversity of the clinical and laboratory fea-tures of the APL syndrome necessitated a set of

criteria for diagnosis. Preliminary classificationcriteria for APL syndrome were formulated dur-ing a post-conference workshop held in Sapporoin 1998 following the Eighth InternationalMultidisiplinary Symposium on APL Antibodies(231) (Table 3). The workshop focused on defin-ing a category ofdefinite APL syndrome. Thesecriteria were intended to provide a uniform basisfor clinical and research studies. The intention

was not to guide clinical diagnosis or treatmentof APL syndrome in individual patients. There-fore, the criteria encompass those clinical andlaboratory features that were most closely asso-ciated with the APL syndrome in prospective clin-ical and experimental studies. IgG and IgM aCLantibodies can be detected in approximately 5%of normal subjects, while persistently elevatedlevels can be present in less than 2% on repeattesting (328). Hence, the persistent presence of

APL antibodies for 6 or more weeks is an essen-tial requirement for APL syndrome classification.Other features of APL syndrome such as throm-bocytopenia, hemolytic anemia, transient cere-bral ischemia, transverse myelopathy, livedoreticularis, cardiac valve disease, multiple scle-rosislike syndrome, chorea, and migrane wasnot included in the Sapporo criteria because their

associations were not strongly confirmed by clin-ical and experimental investigations. Likewise,anti-

2GP-1 antibodies, low-positive titers of IgG

or IgM ACAs, IgA isotype of ACA, and antibodiesagainst other phospholipids or phospholipid-binding proteins were not included as criteria inthis classification (231,329).

Recently, the Sapporo criteria for APL syn-drome were tested by experienced clinicians.Two hundred forty-three consecutive patients



20/38

who had clinical diagnoses of primary APL syn-drome (n=49), secondary APL syndrome(n=26), systemic lupus erythematosus (SLE)

without clinical APL syndrome (n=131), andlupus-like disease without clinical APL syndrome(n=37) were classified according to the classifi-cation criteria. Sensitivity, specificity, positivepredictive value, and negative predictive value

were found to be 0.71, 0.98, 0.95, and 0.88, re-

spectively (330). Specificity and positive predic-tive value were high and these preliminary crite-ria could be useful for clinical studies. Sensitivityand negative predictive value were lower com-pared with physician-diagnosed APL syndromebecause some other diagnostic features of thedisease such as livedo reticularis, thrombocy-topenia, low-titer aCL, IgA ACA, or anti-

2GP-1

were not included in the criteria for definite APL


Data from references 18, 2628.

*Antithrombotic therapy should not be stopped unless the ACAs has been absent for the preceding 4 to 6 months.

Based on The FDA Safety Information and Adverse Event Reporting Program, 2002 Safety Alert-Lovenox.

Deep venous thrombosis with or withoutpulmonary embolus

Acute treatment with heparin/LMWH followed by

long-term* self-administration of subcutaneous

porcine heparin/LMWH

Clopidogrel (long term if stable)

Type I syndrome

Coronary artery thrombosis

Peripheral artery thrombosis

Aortic thrombosis

Carotid artery thrombosis

Acute treatment with heparin/LMWH followed bylong-term* self-administration of subcutaneous

porcine heparin/LMWH

Clopidogrel (long term if stable)

Type II syndrome

Retinal artery thrombosis

Retinal vein thrombosis

Cerebrovascular thrombosisTransient ischemic attacks

Cerebrovascular

Long-term (clopidogrel) plus long-term* self-admin-

istration of subcutaneous porcine heparin/LMWH

Retinal

Clopidogrel;if failure, add long-term* self-adminis-tration of subcutaneous porcine heparin/LMWH

Type III syndrome

Mixtures of types I, II, and III Therapy depends on types and sites of thrombosis,as per preceding recommendations

Type IV syndrome

Placental vascular thrombosis

Fetal wastage common in first trimester

Fetal wastage can occur in second and third

trimesters

Maternal thrombocytopenia (uncommon)

Low dose aspirin (81 mg/day) before conception

and add fixed, low dose porcine heparin at 5000 U

every 12 hours immediately after conception,

Dalteparin (but not enoxaparin)

Type V (fetal wastage)

syndrome

Antiphospholipid antibody

No apparent clinical manifestations

No clear indications for antithrombotic therapyType VI syndrome

TABLE 2. Bick Classification of APL Antibody-Mediated Thrombosis Syndromes and Recommended Antithrombotic Therapy Regimens

Type of Thrombotic Syndrome Clinical Presentation Recommended Antithrombotic Therapy


21/38

syndrome. Therefore, classification criteria forprobable or possible APL syndrome should be

described to better understand the boundaries ofAPL syndrome and the full spectrum oftrueAPL syndrome (329). At the subsequent work-shop on classification criteria held in Tours,France in September 2000, no alterations wereadded to the Sapporo criteria. However, futureefforts should be focused on the following to im-prove them: 1) further evaluation of the interna-tional (Sapporo) criteria for definite APL syn-drome; 2) definition of other categories of APL

syndrome, such as probable and possible APLsyndrome; 3) guidelines for the clinical diagnosis

as distinct from classification of APL syndrome;4) strategies to improve the compliance of labo-ratories worldwide, with recommended proce-dures for LA and ACA assays; 5) development ofmonoclonal antibody standard reagents for ACAand LA assays; and 6) refinement and subse-quent evaluation of anti-

2GP-1 assays for use in

identification of APL syndrome (331). Becausethe major features of primary and secondary APLsyndromes are similar and SLE is by far the most


TABLE 3. Preliminary Criteria for Classification of Definite APL Syndrome

Definite antiphospholipid antibody syndrome is considered to be present if at least 1 of the clinical criteria and 1 of the

laboratory criteria are met.

Clinical criteria

1. Vascular thrombosis

One or more clinical episodes of arterial, venous, or small vessel thrombosis in any tissue or organ. Thrombosis must be confirmed by

imaging or Doppler studies or histopathology, with the exception of superficial venous thrombosis. For histopathologic confirmation,

thrombosis should be present without significant inflammation in 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

2. (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

2. (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.

In studies of populations of patients who have more than 1 type of pregnancy morbidity, investigators are strongly encouraged to stratify

groups of subjects according to a, b, or c above.

Laboratory criteria

1. Anticardiolipin antibody of IgG and/or IgM isotype in blood, present in medium or high titer, on 2 or more occasions, at least 6 weeks

apart, measured by a standardized enzyme-linked immunosorbent assay for beta2glycoprotein I-dependent anticardiolipin antibodies.

2. Lupus anticoagulant present in plasma, on 2 or more occasions at least 6 weeks apart, detected according to the guidelines of the

International Society on Thrombosis and Hemostasis (Scientific Subcommittee on Lupus Anticoagulants/Phospholipid- Dependent

Antibodies), in the following steps:

(a) Prolonged phospholipid-dependent coagulation demonstrated on a screening test, e.g. activated partial thromboplastin time, kaolin

clotting time, dilute Russells viper venom time, dilute prothrombin time, Textarin time.

2. (b) Failure to correct the prolonged coagulation time on the screening test by mixing with normal platelet-poor plasma.

2. (c) Shortening or correction of the prolonged coagulation time on the screening test by addition of excess phospholipid.

2. (d) Exclusion of other coagulopathies, e.g. factor VIII inhibitor or heparin, as appropriate.

Data from reference 231.

No exclusions other than those contained within the above criteria are needed. However, because of the likelihood that thrombosis may bemultifactorial in patients with the APL syndrome, the workshop participants recommend that (a) patient populations being studied should beassessed for other contributing causes of thrombosis, and (b) such populations should be stratified according to identifiable or probable riskfactors, e.g., age or comorbidities. Specific limits were not placed on the interval between the clinical event and the positive laboratoryfindings. However, it was the view of many at the workshop that (a) information about such intervals should be assessed when relevant, and(b) the relatively strict definition of laboratory criteria (including the requirement that results again be positive on repeat tests performed atleast 6 weeks after the initial test) would help to exclude antiphospholipid antibody positivity that represents an epiphenomenon to theclinical events.


22/38

common cause of secondary APL syndrome, ex-clusion criteria have been proposed for the diag-nosis of primary APL syndrome (Table 4) (332).

CLINICAL STRATEGIES AGAINST

APL SYNDROME: OBSERVATION

AND RATIONAL CLINICAL MANAGEMENT

Despite tremendous advances in our under-standing the pathogenesis of APL syndrome, themainstay of the management is still anticoagula-tion. Available data for making a decision onhow to treat APL syndrome patients largely comefrom retrospective studies, and data fromprospective clinical trials are limited. Moreover,there is no consensus regarding the duration andintensity of anticoagulant therapy after thethrombotic event occurred. Management strate-gies based on Bick classification of APL syn-drome have been given in Table 2. This classifi-cation offers a useful tool for the management of

APL in clinical practice.

Whether or not to treat asymptomatic healthyindividuals with APL antibodies (type VI diseasein Bicks classification) is also a matter of debate.

A weakly or transiently elevated positive APL an-tibody test result needs no prophylaxis. Asymp-tomatic healthy subjects with persistently posi-tive moderate-to-high titer APL antibodies aregenerally given low-dose aspirin (ASA 75100mg/day). Low-dose aspirin alone could offersome benefit in recurrent miscarriage syndrome

with APL antibodies (333335). Aspirin prophy-laxis could also provide protection against post-partum nonpregnancy-related vascular throm-bosis in APL syndrome patients who present withpregnancy loss as their only manifestation (336).Prophylactic aspirin is suggested to all patients

with SLE to prevent both arterial and venousthrombotic manifestations, especially those with

APL antibodies (337). On the other hand, the useof low dose aspirin in people with ACAs did notprotect against deep venous thrombosis or pul-monary embolism (41). Currently there is still nosatisfactory prospective evidence that low-doseaspirin provides an adequate prophylaxis againstthrombosis especially in venous site in asympto-matic carriers of APL antibodies. Hydroxychloro-quine can decrease the titers of APL antibodiesand could protect against thrombotic risk in SLEpatients. It was also tried in primary APL syn-drome in some studies (338,339). However, use-fulness in asymptomatic carriers of APL antibod-ies has not been proven by prospective trials.Moreover, the majority of asymptomatic carriers

of APL antibodies do not develop thromboticmanifestations. Therefore targeted education andclose monitoring of those persons would be ap-propriate (18,26,340). Clearly, any acquired pro-thrombotic risk factors should be removed if pos-sible. Asymptomatic individuals with APL anti-bodies should be examined for the co-existenceof hereditary thrombophilia (factor V Leiden,prothrombin 20210A, MHTFR, and others) and astrong family history of arterial and/or venous


TABLE 4. Proposed Empirical Exclusion Criteria to Distinguish Primary and SLE Related APL Syndrome

The presence of any of these criteria excludes the diagnosis of primary APL syndrome:

Malar rash

Discoid rash

Oral or pharyngeal ulceration, excluding nasal septum ulceration or perforation

Frank arthritis

Pleuritis, in the absence of pulmonary embolism or left-sided heart failure

Pericarditis, in the absence of myocardial infarction or uremia

Persistent proteinuria greater than 0.5 gram per day, due to biopsy-proven immune complex-related glomerulonephritis

Lymphopenia less than 1.000/L

Antibodies to native DNA, by radioimmunoassay or Crithidia fluorescence

Antiextractable nuclear antigen antibodies

ANA of more than 1:320

Treatment of drugs known to induce APL antibodies

A follow-up longer than 5 years after the first clinical manifestation is necessary to rule out the subsequent emergence of SLE

Data from reference 332.


23/38

thrombotic tendency. Selected cases among thatspecific subpopulation are subject to long-termlow-dose aspirin prophylaxis. Likewise, if anyasymptomatic subject does develop some clinical

features of APL syndrome such as migraine, livedoreticularis, dizzy or confusional episodes, nonspe-cific visual disturbance, or very early pregnancyloss, low-dose aspirin may be given (340,341).

Treatment of acute thrombotic events of pa-

Date post:	07-Apr-2018
Category:	Documents
Upload:	karanajm
View:	226 times
Download:	0 times

APLS Article

Documents