Thrombocytopenia in the antiphospholipid syndrome

British Journal of Haematology, 1 9 9 6 , 9 3 , 1 –5 An n otation THROMBOCYTOP ENIA IN THE ANTIP HOSP HOLIP ID SYNDROME Th e an tiph osph olipid syn drome (APS) is defin ed by th e association between an tiph osph olipid (aPL) an tibodies – i.e. lu pu s an ticoagu lan t (LA) an d an ticardiolipin (aCL) an tibodies – an d pecu liar clin ical man ifestation s, su ch as ven ou s an d arterial th rombosis, recu rren t abortion s an d th rombocytopen ia. Two forms of APS h ave been described: th e ‘primary’ APS (Ash erson et al, 1 9 8 9 ), th at occu rs in th e absen ce of any u n derlyin g disease, an d th e ‘secon dary’ APS (Alarcon Segovia et al, 1 9 8 9 ), th at develops in association with an oth er path ological con dition , main ly systemic lu pu s eryth ematosu s (SLE). Th romboses are th e most frequ en t clin ical even ts in APS, becau se th ey occu r in abou t 3 0 % of patien ts (Lech n er & Pabin ger-Fasch in g, 1 9 8 5 ). Ven ou s th romboses accou n t for approximately two-th irds of all th e th romboembolic even ts, represen ted main ly by deep ven ou s th rombosis of th e legs an d pu lmon ary embolism (Italian Registry of An tiph osph olipid An tibodies, 1 9 9 3 ). On th e oth er h an d, complete cerebral isch aemic strokes an d tran sien t isch aemic attacks are th e most common arterial th rombotic even t in patien ts with APS (Italian Registry of An tiph osph olipid An tibodies, 1 9 9 3 ). Th rombotic even ts may be recu rren t an d are often spon tan eou s. Obstetrical complication s in clu de abortion , fetal loss, in tra-u terin e growth retardation an d preeclampsia (Barbu i et al, 1 9 8 8 ). Th ey are reported in abou t 1 0 –2 0 % of women with APS (Italian Registry of An tiph osph olipid An tibodies, 1 9 9 3 ) an d are regarded as pecu liar types of th rombotic complication s involvin g th e placen tal vessels (De Wolf et al, 1 9 8 2 ). A variable degree of th rombocytopen ia is reported in as many as 2 0 –4 0 % of th e patien ts with aPL an tibodies (Lech n er & Pabin ger-Fasch in g, 1 9 8 5 ; Italian Registry of An tiph osph olipid An tibodies, 1 9 9 3 ). Even th ou gh th rombocytopen ia is a rath er common man ifestation of APS, u n certain ties still exist regardin g its path ogen esis, treatmen t, an d in flu en ce on th e policy for th e th erapy of th rombosis. Pathogenesis of thrombocytopenia and interactions between antiphospholipid antibodies and platelets Approximately a decade ago, aPL an tibodies h ad been su ggested to cau se th rombocytopen ia, based on th e h igh prevalen ce of aCL an tibodies in patien ts with idiopath ic th rombocytopen ic pu rpu ra (ITP) (Harris et al, 1 9 8 5 ) an d on th e in teraction between platelet membran e ph osph olipidsan d th ese an tibodies (see below). However, recen t fin din gs qu estion ed th e role of aPL an tibodies in th e path ogen esis of th rombocytopen ia. In fact, an tibodies directed again st Correspon dence: Dr Mon ica Galli, Departmen t of Haematology, Ospedali Riu n iti, Largo Barozzi 1 , 2 4 1 0 0 Bergamo, Italy. # 1 9 9 6 Blackwell Scien ce Ltd glycoprotein s (GP) IIb/ IIIa an d Ib/ IX are fou n d in abou t 4 0 % of aPL-positive patien ts (Galli et al, 1 9 9 4 ), a figu re similar to th ose already kn own for ITP (He et al, 1 9 9 4 ). Sin ce th ese an tibodies are path ogen etically lin ked to ITP (Ku n iki & Newman , 1 9 9 2 ), on e may su spect a similar lin kage also in patien ts with aPL an tibodies. Moreover, an tibodies again st a 5 0 –7 0 kD in tern al platelet protein h ave been specifically fou n d in aPL-positive patien ts with th rombocytopen ia bu t n ot in ITP patien ts (Fabris et al, 1 9 9 4 ). Fin ally, it h as been observed th at immu n osu ppressive treatmen t of ITP in creases platelet n u mber an d redu ces th e titres of plateletassociated IgG (PAIgG) bu t n ot th at of aPL an tibodies (Stasi et al, 1 9 9 4 ), in dicatin g th at aPL an d an ti-platelet an tibodies comprise differen t specificities. Th u s, plateletspecific an tibodies, more th an an tiph osph olipid an tibodies, appear to play a role in th e path ogen esis of th rombocytopen ia in th e APS. aPL an tibodies react with n egatively-ch arged ph osph olipids, wh ich are an essen tial con stitu en t of cell membran es. In platelets, an ion ic ph osph olipids are located in th e in n er leaflet of th e plasma membran e (Bevers et al, 1 9 9 4 ), th u s th ey are n ot available for in teraction with aPL an tibodies. However, u n der physiological con dition s, th is asymmetric distribu tion can be lost, resu ltin g in exposu re of an ion ic ph osph olipids on th e ou ter leaflet of platelet membran e (Bevers et al, 1 9 9 4 ). For platelets, th is ph en omen on occu rs u pon activation by differen t agon ists an d is accompan ied by sh eddin g of microvesicles (Bevers et al, 1 9 9 4 ). Sin ce both activated platelets an d platelet-derived microvesicles (PMPs) represen t an importan t in vivo procoagu lan t su rface (Bevers et al, 1 9 9 4 ), th e in teraction between aPL an tibodies an d platelet membran e may be relevant for th e path ogen esis of th e th rombotic complication s of APS. Even th ou gh bin din g of aPL an tibodies to restin g platelets h as occasion ally been reported (Hasselaar et al, 1 9 9 0 ; Ou t et al, 1 9 9 1 ), activation an d/ or damage appear a prerequ isite for th e bin din g of aPL an tibodies, as in dicated by several in vitro experimen ts (Kh amash ta et al, 1 9 8 8 ; Mikh ail et al, 1 9 8 8 ; Sh i et al, 1 9 9 3 ). Conversely, less convin cin g data exist regardin g th e in vivo in teraction between aPL an tibodies an d platelets. Alth ou gh impairmen t of th e th romboxan e/ PGI2 balan ce (Martinu zzo et al, 1 99 3), increased u rinary excretion an d plasma levels of th e platelet-specific -thromboglobulin (Galli et al, 1 9 8 8 ) h ave been reported, poin tin g towards a con dition of in vivo platelet activation , n eith er elu tion of aPL an tibodies from platelet membran e (Biasiolo & Pen go, 1 9 9 3 ) n or eviden ce for circu latin g activated platelets h ave been demon strated (Ou t et al, 1 9 9 1 ). Recen tly, th e presen ce of PMPs h as been investigated in patien ts with aPL an tibodies (Galli et al, 1 9 9 3 b). By mean s of flow cytometry, employin g 1 2 Annotation specific an ti-platelet glycoprotein an tibodies, path ological levels of circu latin g PMPs were observed in approximately h alf of th e patien ts. Th is fin din g was statistically associated – by retrospective an alysis – with th rombosis. Th e con troversial resu lts on th e in teraction between aPL an tibodies an d platelets may stem, at least in part, from th e recen t observation s th at plasma protein ‘cofactors’ bou n d to an appropriate an ion ic su rface (su ch as an ion ic ph osph olipids) are th e an tigen ic target of aPL an tibodies. In particu lar (h u man ) proth rombin is th e ‘cofactor’ for LA an tibodies (Bevers et al, 1 9 9 1 ; Rao et al, 1 9 9 5 ), wh ereas 2 -glycoprotein I ( 2 -GPI) is recogn ized by aCL an tibodies (McNeil et al, 1 9 9 0 ; Matsu u ra et al, 1 9 9 0 , 1 9 9 4 ; Bevers & Galli, 1 9 9 0 ; Galli et al, 1 9 9 0 , 1 9 9 2 , 1 9 9 3 a; Oostin g et al, 1 9 9 2 ). Oth er protein s with affin ity for an ion ic ph osph olipids – an n exin V, th rombomodu lin , protein C, protein S – h ave been proposed to play su ch a role for oth er so-called ‘aPL’ an tibodies (Oostin g et al, 1 9 9 3 ; Gibson et al, 1 9 9 2 ; Matsu da et al, 1 9 9 4 ). It is likely th at th e variable availability of th ese ‘cofactors’ in differen t experimen tal con dition s may greatly in flu en ce th e reactivity of aPL an tibodies with platelets. In con clu sion , th e su m of th e existin g data poin ts towards th e possibility th at platelets may circu late in an activated state in patien ts with aPL an tibodies, even th ou gh n o clear eviden ce th at th ese an tibodies do play a role in th is process h as been reach ed. In th is respect, it h as been hypoth esized th at ch ron ic exposu re of a procoagu lan t su rface (for persisten t activation an d/ or damage of blood cells) may con fer a h igh risk of th rombosis an d, at th e same time, allow bin din g of th e plasma protein ‘cofactors’ with su bsequ en t gen eration of aPL an tibodies (Galli & Bevers, 1 9 9 4 ). Th ese an tibodies th erefore migh t be regarded as a u sefu l marker of an on goin g proth rombotic state. Moreover, on e can n ot exclu de th e possibility th at particu lar types of aPL an tibodies may play a role in main tain in g an d exten din g th e th rombotic process. Th is hypoth esis represen ts th e ration ale for th e con tin u in g investigation of th e immu n ological an d fu n ction al properties of th e family of aPL an tibodies. Prevalence, clinical manifestations and treatment of thrombocytopenia in APS Severe th rombocytopen ia (<5 0 2 1 0 9 platelets/ l) is n ot a frequ en t fin din g in patien ts with aPL an tibodies, becau se its prevalen ce is abou t 5 –1 0 % (Lech n er & Pabin ger-Fasch in g, 1 9 8 5 ) an d it is seldom associated with h aemorrh agic man ifestation s. Amon g th e 2 9 3 cases en rolled in th e Italian Registry of An tiph osph olipid An tibodies (IR-APA) (1 9 9 3 ) th e prevalen ce of th rombocytopen ia was 2 6 %, 3 2 patien ts (1 1 %) were severely th rombocytopen ic an d major bleedin g was experien ced by on ly two ou t of th ese 3 2 cases (6 %). Th erefore th e prevalen ce of major h aemorrh agic complication s in APS does n ot appear sign ifican tly differen t from th at reported in ITP patien ts (Cortelazzo et al, 1 9 9 1 ). Recen tly, th rombocytopen ia h as been reported to be associated with a h igh risk of mortality in patien ts su fferin g from SLE (Dren kard et al, 1 9 9 4 ). By u n ivariate an alysis, Dren kard et al (1 9 9 4 ) demon strated th at th rombocytopen ia an d arterial th rombosis were th e on ly APS-related man ifestation s statistically associated with mortality in 6 5 8 SLE patien ts (P < 0 :0 0 1 an d P ˆ 0 :0 2 , respectively). Un fortu n ately th e au th ors did n ot state wh eth er an d to wh at extent bleedin g complication s accou n ted for th e death of th eir patien ts; th erefore it is possible th at th rombocytopen ia was simply associated with a more severe disease. Some au th ors h ave specu lated abou t th e possibility th at th rombocytopen ia selects a popu lation of aPL-positive patien ts at particu lar risk for th e developmen t of SLE (Harris et al, 1 9 8 5 ), bu t few data exist to su pport th is. In deed, th e first 5 -year follow-u p an alysis of th e patien ts en rolled in th e IR-APA stu dy in dicates th at th e rate of progression to overt SLE is low, 0 .2 8 % patien ts/ year, irrespective of wh eth er th rombocytopen ia is presen t or n ot (Fin azzi, 1 9 9 4 ). Similar data h ave been recen tly reported by Stasi et al (1 9 9 5 ), wh o performed a lon g-term observation of 2 0 8 ITP patien ts. Th ese au th ors iden tified on ly fou r patien ts wh o developed laboratory or clin ical featu res con sisten t with SLE or oth er overt au toimmu n e diseases over a median follow-u p period of 9 2 mon th s. Th rombocytopen ia in APS rarely requ ires treatmen t. However, wh en th is is n ecessary th e same treatmen t policy as for ITP sh ou ld be con sidered. Recen tly, a large series of ITP patien ts h as been evalu ated for treatmen t ou tcome accordin g to th eir aPL statu s (Stasi et al, 1 9 9 4 ): corticosteroids, splen ectomy, h igh -dose IvIg, an ti-D immu n oglobu lin s an d in terferon produ ced respon ses irrespective of th e presen ce of aCL an tibodies. Most of th e patien ts were treated with steroids an d splen ectomy. As expected, corticosteroids produ ced a similar low rate of su stain ed respon ses in patien ts with or with ou t aCL an tibodies (1 5 % an d 1 7 %, respectively). Splen ectomy was performed in 2 3 patien ts: su stain ed respon ses were obtain ed in approximately 6 0 % in both grou ps of ITP patien ts. Th erefore aPLan tibodies did n ot predict th e ou tcome of treatmen t. Acqu ired defects of platelet fu n ction h ave been estimated to occu r in as many as 4 0 % of th e patien ts with aPL an tibodies (Hasselaar et al, 1 9 9 0 ). Impairmen t of platelet aggregation h as been fou n d in th e cou rse of opportu n istic in fection s in AIDS patien ts with aPL an tibodies (Coh en et al, 1 9 8 6 ), wh ereas redu ced adh esion to su ben doth eliu m h as been observed by Orlan do et al (1 9 9 2 ). Oth er defects in clu de storage pool disease, isolated defects in respon se to epin eph rin or collagen , an d impairmen t of th romboxan e B2 syn th esis (Hasselaar et al, 1 9 9 0 ). In vitro experimen ts performed with th e immu n oglobu lin fraction s isolated from patien ts’ plasma led to th e su ggestion th at aPL an tibodies migh t be respon sible for th ese acqu ired qu alitative platelet defects (Coh en et al, 1 9 8 6 ; Orlan do et al, 1 9 9 2 ). However, th e eviden ce is in con clu sive, becau se on ly in a few cases were affin itypu rified aCL an tibodies tested (Orlan do et al, 1 9 9 2 ). Even th ou gh th e bleedin g time is gen erally prolon ged, th e defects are n ot associated with sign ifican t h aemorrh agic complication s, u n less an oth er risk factor coexists. Th erefore defective platelet fu n ction n eed on ly be sou gh t in cases of u n explain ed bleedin g in aPL-positive patien ts on an tith rombotic treatmen t. Treatmen t of h aemorrh agic complication s n ot associated with th rombocytopen ia remain s an open qu estion . # 1 9 9 6 Blackwell Scien ce Ltd, British Journal of Haematology 9 3 : 1 –5 Annotation On ly a few case reports h ave been pu blish ed; corticosteroid admin istration an d plasma exch an ge h ave been fou n d to be u sefu l an ecdotally (Man oh aran & Gottlieb, 1 9 8 4 ; Ordi et al, 1 9 8 4 ). Prevalence and treatment of thrombosis in relation to thrombocytopenia Ven ou s an d arterial th rombosis are th e most common man ifestation s of APS. Wh en th eir prevalen ce is evalu ated accordin g to platelet n u mber, it appears th at severe, bu t n ot moderate, th rombocytopen ia protects, at least in part, again st th romboembolic complication s (Italian Registry of An tiph osph olipid An tibodies, 1 9 9 3 ). Th e IR-APA stu dy reported th at 3 2 % of moderately th rombocytopen ic patien ts (platelets 5 0 – 1 0 0 2 1 0 9 / l) experien ced at least on e th rombotic even t compared with 4 0 % of patien ts with n ormal platelet cou n t. Conversely, severe th rombocytopen ia (platelets <5 0 2 1 0 9 / l) was associated with a lower prevalen ce of th rombosis (9 %). It is importan t to emph asize th at th rombotic complication s may occu r in spite of a very low platelet cou n t, th u s complicatin g th e an ticoagu lan t man agement. Th ese retrospective data h ave been con firmed an d exten ded by th e first prospective IR-APA su rvey (Fin azzi, 1 9 9 4 ). No sign ifican t differen ces in th e in ciden ce of th rombosis were fou n d between patien ts with a n ormal platelet cou n t (2 .5 4 % patien ts/ year) an d th ose with moderate th rombocytopen ia (3 .2 9 % patien ts/ year). Again , a lower in ciden ce (0 .9 5 % patien ts/ year) was observed in severely th rombocytopen ic patien ts. Th is differen ce did n ot reach statistical sign ifican ce, probably becau se of th e relatively low n u mber of patien ts with th rombosis (7 / 5 6 patien ts with mild th rombocytopen ia versu s 2 / 5 2 severely th rombocytopen ic patien ts). Secon dary prophylaxis of th rombotic even ts with an tithrombotic dru gs is still con troversial in APS. By retrospective an alysis, two stu dies (Kh amash ta et al, 1 9 9 5 ; Rosove & Brewer, 1 9 9 2 ) sh owed th at lon g-term h igh -in ten sity warfarin th erapy con fers better an tith rombotic protection th an lowin ten sity warfarin an d/ or aspirin . However, th ese th erapeutic recommen dation s remain to be con firmed in prospective clinical trials. The two stu dies reported an inciden ce of sign ificant h aemorrhagic complication s of 3 .1 an d 7 .1 % patients/year, with an inciden ce of life-threaten in g bleedin g even ts of 1 .9 an d 1 .7 % patien ts/ year, respectively. Th ese figu res are somewh at h igh er th an th e 1 % patien ts/ year in ciden ce of major bleedin g observed in patien ts with prosth etic h eart valve replacemen t, wh o are similarly on h igh -in ten sity oral an ticoagu lation (PT INR 3 .0 –4 .5 ) (Cortelazzo et al, 1 9 9 3 ). In an ticoagu lated APS patien ts on e can n ot exclu de th e possibility th at th rombocytopen ia (or, less frequ en tly, a platelet fu n ction defect) may play a role, at least in part, in th e developmen t of th e h aemorrh agic complication s. Un fortu n ately n eith er stu dy specifically addressed th is problem, becau se n o statemen t was made regardin g platelet cou n ts at th e times of bleedin g even ts. Th e IR-APA su rvey (Fin azzi et al, 1 9 9 4 ) reported major bleedin g in th ree patien ts u n der oral an ticoagu lation ; on e of th em was th rombocytopen ic an d experien ced a fatal cerebral h emorrh age. Th erefore th e decision abou t both th e du ration # 1 9 9 6 Blackwell Scien ce Ltd, British Journal of Haematology 9 3 : 1 –5 3 an d th e in ten sity of oral an ticoagu lation th erapy mu st be weigh ed again st th e risk of bleedin g complication s (Barbu i & Fin azzi, 1 9 9 4 ). In ou r in stitu tion , moderate th rombocytopen ia does n ot modify th e policy for th e treatmen t of th rombosis in APS-positive patien ts, wh ich follows th e gen eral gu idelin es for th e oth er con dition s of so-called ‘primary th romboph ilia’ (Greaves & Preston , 1 9 9 1 ). Con versely, we do n ot recommen d an ticoagu lan t treatmen t in cases of severe th rombocytopen ia. In con clu sion , moderate th rombocytopen ia is frequ en t in APS an d, as a ru le, does n ot modify th e policy for treatmen t of th rombosis. Severe th rombocytopen ia is relatively u n common an d is seldom associated with bleedin g even ts. Wh en requ ired, its treatmen t is similar to th at of ITP. Fin ally, mu ch clin ical an d laboratory work is still requ ired to elu cidate th e role of platelets an d aPL an tibodies in th e path ogen esis of th e th rombotic complication s typical of APS. Department of Haematology, Ospedali Riuniti, Bergamo, Italy M ON I CA G A LLI G UI DO F I N A ZZI T I ZI A N O BA R BUI REFERENCES Alarcon-Segovia, D., Deleze, M. & Oria, C.V. (1 9 8 9 ) An tiph osph olipid an tibodies an d th e an tiph osph olipid syn drome in systemic lu pu s eryth ematosus: a review of 5 0 0 con secutive cases. Medicine, 6 8 , 3 5 3 –3 6 5 . Ash erson , R.A., Kh amash ta, M.A., Ordi-Ros, J., Derksen, R.H.W.M., Mach in , S.J., Barqu in ero, J., Ou tt, H.H., Harris, E.N., VilardellTorres, M. & Hu gh es, G.R.V. (1 9 8 9 ) Th e ‘primary’an tiph osph olipid syn drome: major clin ical an d serological featu res. Medicine, 6 8 , 3 6 6 -3 7 4 . Barbu i, T., Cortelazzo, S., Galli, M., Parazzin i, F., Radici, E., Rossi, E. & Fin azzi, G. (1 9 8 8 ) An tiph osph olipid an tibodies in early repeated abortion s: a case–con trol stu dy. Fertility and Sterility, 5 0 , 5 8 9 –5 9 2 . Barbu i, T. & Fin azzi, G. (1 9 9 4 ) Clin ical trials on an tiph osph olipid syn drome: wh at is don e an d wh at is n eeded? Lupus, 3 , 3 0 3 –3 0 8 . Bevers, E.M. & Galli, M. (1 9 9 0 ) 2 -glycoprotein I for bin din g of an ticardiolipin an tibodies to cardiolipin . (Letter). Lancet, 3 3 5 , 9 5 2 – 953. Bevers, E.M., Galli, M., Barbu i, T., Comfu riu s, P. & Zwaal, R.F.A. (1 9 9 1 ) Lu pu s an ticoagu lan t IgG’s (LA) are n ot directed to ph osph olipids on ly bu t to a complex of lipid-bou nd h u man proth rombin . Thrombosis and Haemostasis, 6 6 , 6 2 9 –6 3 2 . Bevers, E.M., Smeets, E.F., Con fu riu s, P. & Zwaal, R.F.A. (1 9 9 4 ) Physiology of membrane lipid asymmetry. Lupus, 3 , 2 3 5 –2 4 0 . Biasiolo, A. & Pen go, V. (1 9 9 3 ) An tiph osph olipid an tibodies are n ot presen t in th e membrane of gel-filtered platelets of patien ts with IgG an ticardiolipin an tibodies, lu pu s an ticoagu lan t an d th rombosis. Blood Coagulation and Fibrinolysis, 4 , 4 2 5 –4 2 8 . Coh en , A.J., Ph ilips, T.M. & Kessler, C.M. (1 9 8 6 ) Circu latin g coagu lation in h ibitors in th e acqu ired immu n odeficien cy syn drome. Annals of Internal Medicine, 1 0 4 , 1 7 5 –1 8 0 . Cortelazzo, S., Fin azzi, G., Bu elli, M., Molten i, A., Viero, P. & Barbu i, T. (1 9 9 1 ) High risk of severe bleedin g in aged patien ts with ch ron ic idiopath ic th rombocytopen ic pu rpu ra. Blood, 7 7 , 3 1 –3 3 . Cortelazzo, S., Fin azzi, G., Viero, P., Galli, M., Remu zzi, A., Paren zan , L. & Barbu i, T. (1 9 9 3 ) Th rombotic an d h emorrh agic complication s in patien ts with mech anical h eart valve prosth esis atten din g an an ticoagu lation clin ic. Thrombosis and Haemostasis, 6 9 , 3 1 6 –3 2 0 . 4 Annotation De Wolf, F., Carreras, L.O., Moerman, P., Vermylen , J., Van Assch e, A. & Ren aer, M. (1 9 8 2 ) Decidu al vascu lopathy an d extensive placen tal in farction in a patien t with repeated th romboembolic acciden ts, recu rren t fetal loss an d a lu pu s an ticoagu lan t. American Journal of Obstetrics and Gynecology, 1 4 2 , 8 2 9 –8 3 4 . Dren kard, C., Villa, A.R., Alarcon -Segovia, D. & Perez-Vazqu ez, M.E. (1 9 9 4 ) In flu en ce of th e an tiph ospholipid syn drome in th e su rvival of patien ts with systemic lu pu s eryth ematosus. Journal of Rheumatology, 2 1 , 1 0 6 7 –1 0 7 2 . Fabris, F., Steffan , A., Cordian o, I., Borzin i, P., Lu zzatto, G., Ran di, M.L. & Girolami, A. (1 9 9 4 ) Specific an tiplatelet au toan tibodies in patien ts with an tiph osph olipid an tibodies an d th rombocytopen ia. European Journal of Haematology, 5 3 , 2 3 2 –2 3 6 . Fin azzi, G. (1 9 9 4 ) A prospective stu dy of 3 6 0 patien ts with an tiph ospholipid an tibodies. Lupus, 3 , 3 6 0 a. Galli, M. & Bevers, E.M. (1 9 9 4 ) In h ibition of ph osph olipid-depen den t coagu lation reaction s by ‘an tiph ospholipid an tibodies’: possible modes of action . Lupus, 3 , 2 2 3 –2 2 8 . Galli, M., Bevers, E.M., Comfu riu s, P., Barbu i, T. & Zwaal, R.F.A. (1 9 9 3 a) Effect of an tiph ospholipid an tibodies on procoagu lan t activity of activated platelets an d platelet-derived microvesicles. British Journal of Haematology, 8 3 , 4 6 6 –4 7 2 . Galli, M., Comfu riu s, P., Barbu i, T., Zwaal, R.F.A. & Bevers, E.M. (1 9 9 2 ) An ticoagu lan t activity of 2 -glycoprotein I is poten tiated by a distin ct su bgrou p of an ticardiolipin an tibodies. Thrombosis and Haemostasis, 6 8 , 2 9 7 –3 0 0 . Galli, M., Comfu riu s, P., Maassen , C., Hemker, H.C., de Baets, M.H., van Breda Vriesman, P.J.C., Barbu i, T., Zwaal, R.F.A. & Bevers, E.M. (1 9 9 0 ) An ticardiolipin an tibodies (ACA) directed n ot to cardiolipin bu t to a plasma protein cofactor. Lancet, 3 3 4 , 1 5 4 4 –1 5 4 7 . Galli, M., Cortelazzo, S., Viero, P., Fin azzi, G., de Gaetan o, G. & Barbu i, T. (1 9 8 8 ) In teraction between platelets an d lu pu s an ticoagu lan t. European Journal of Haematology, 4 1 , 8 8 –9 4 . Galli, M., Daldossi, M. &Barbu i, T. (1 9 9 4 ) An ti-glycoprotein Ib/ IX an d IIb/ IIIa an tibodies in patien ts with an tiph osph olipid an tibodies. Thrombosis and Haemostasis, 7 1 , 5 7 1 –5 7 5 . Galli, M., Grassi, A. & Barbu i, T. (1 9 9 3 b) Platelet-derived microvesicles in patien ts with an tiph ospholipid an tibodies. (Abstract). Thrombosis and Haemostasis, 6 9 , 5 4 1 . Gibson , J., Nelson , M., Brown , R., Salem, H. & Kron en berg, H. (1 9 9 2 ) Au toan tibodies to th rombomodu lin : developmen t of an en zyme immu n oassay an d a su rvey of th eir frequ ency in patien ts with th e lu pu s an ticoagu lan t. Thrombosis and Haemostasis, 6 7 , 5 0 7 –5 0 9 . Greaves, M. & Preston , F.E. (1 9 9 1 ) Th e hypercoagu lable state in clin ical practice. British Journal of Haematology, 7 9 , 1 4 8 –1 5 1 . Harris, E.N., Gh aravi, A.E., Hedge, U., Deru e, G., Morgan , S.H., En glert, H., Ch an , J.K.H., Ash erson , R.A. & Hu gh es, G.R.V. (1 9 8 5 ) An ticardiolipin an tibodies in au toimmu ne th rombocytopen ic pu rpu ra. British Journal of Haematology, 5 9 , 2 3 1 –2 3 4 . Hasselaar, P., Derksen , R.H.W.M., Blokzijl, L. & de Groot, P. (1 9 9 0 ) Crossreactivity of an tibodies directed again st cardiolipin , DNA, en doth elial cells an d blood platelets. Thrombosis and Haemostasis, 6 3 , 1 6 9 –1 7 3 . He, R., Reid, D.M., Jon es, C.E. & Sh u lman , N.R. (1 9 9 4 ) Spectru m of Ig classes, specificities an d titers of seru m an tiglycoprotein s in ch ron ic idiopath ic th rombocytopen ic pu rpu ra. Blood, 8 3 , 1 0 2 4 –1 0 3 2 . Italian Registry of An tiph osph olipid An tibodies (IR-APA) (1 9 9 3 ) Th rombosis an d th rombocytopen ia in an tiph ospholipid syn drome (idiopath ic an d secon dary to SLE): first report from th e Italian Registry. Haematologica, 7 8 , 3 1 3 –3 1 8 . Kh amash ta, M.A., Cu adrado, M.J., Mu jic, F., Tau b, N.A., Hu n t, B.J. & Hu gh es, G.R.V. (1 9 9 5 ) Th e man agemen t of th rombosis in th e an tiph ospholipid syn drome. New England Journal of Medicine, 3 3 2 , 9 9 3 –9 9 7 . Kh amash ta, M.A., Harris, E.N., Gh aravi, A.E., Deru e, G., Gil, A., Vazqu ez, J.J. & Hu gh es, G.R.V. (1 9 8 8 ) Immu n e mediated mech an ism for th rombosis: an tiph ospholipid an tibody bin din g to platelet membran es. Annals of Rheumatic Diseases, 4 7 , 8 4 9 –8 5 2 . Ku n icki, T.J. & Newman , P.J. (1 9 9 2 ) Th e molecu lar immu n ology of h u man platelet an tigen s. Blood, 8 0 , 1 3 8 6 –1 4 1 7 . Lech n er, K. & Pabin ger-Fasch in g, I. (1 9 8 5 ) Lu pu s an ticoagu lan t an d th rombosis. Hemostasis, 1 5 , 2 5 4 –2 6 2 . Man oh aran , A. & Gottlieb, P. (1 9 8 4 ) Bleedin g in patien ts with lu pu s an ticoagu lan t. (Letter). Lancet, ii, 1 7 1 . Martin u zzo, M.E., Maclou f, J., Carreras, L.O. & Levy-Toledan o, S. (1 9 9 3 ) An tiphosph olipid an tibodies en h an ce th rombin -in du ced platelet activation an d th romboxan e formation . Thrombosis and Haemostasis, 7 0 , 6 6 7 –6 7 1 . Matsu da, J., Saitoh , N. & Goh ci, K. (1 9 9 4 ) An ti-an n exin Van tibody in systemic lu pu s eryth ematosus patien ts with lu pu s an ticoagu lan t an d/ or an ticardiolipin an tibody. American Journal of Hematology, 4 7 , 5 6 –5 8 . Matsu u ra, E., Igarash i, Y., Yasu da, T., Triplett, D.A. & Koike, T. (1 9 9 4 ) An ticardiolipin an tibodies recogn ize 2 -glycoprotein I stru ctu re altered by in teracting with an oxygen modified solid ph ase su rface. Journal of Experimental Medicine, 1 7 9 , 4 5 7 –4 6 2 . Matsu u ra, E., Igarash i, Y., Fu jimoto, M., Ich ikawa, K. & Koike, T. (1 9 9 0 ) An ticardiolipin cofactor(s) an d differen tial diagn osis of au toimmu ne disease. (Letter). Lancet, 3 3 5 , 1 7 7 –1 7 8 . McNeil, H.P., Simpson , R.J., Ch esterman , C.N. & Krilis, S.A. (1 9 9 0 ) An tiph osph olipid an tibodies are directed again st a complex an tigen th at in clu des a lipid-bin ding in h ibitor of coagu lation: 2 -glycoprotein I (apolipoprotein H). Proceedings of the National Academy of Sciences of the United States of America, 8 7 , 4 1 2 0 –4 1 2 4 . Mikh ail, M.H., Szczech , L.A.M. & Sh apiro, S.S. (1 9 8 8 ) Th e bin din g of lu pu s an ticoagu lan t (LACs) to h u man platelets. Blood, 7 2 , 3 3 3 a. Oostin g, J.D., Derksen, R.H.W.M., Bobbin k, I.W.G., Hacken g, T.M., Bou ma, B.N. & de Groot, P.G. (1 9 9 3 ) An tiphosph olipid an tibodies directed again st a combin ation of ph osph olipids with proth rombin , protein C or protein S: an explanation for th eir path ogen etic mech an isms. Blood, 8 1 , 2 6 1 8 –2 6 2 5 . Ooostin g, J.D., Derksen , R.H.W.M., En tjes, H.T., Bou ma, B.N. & de Groot, P. (1 9 9 2 ) Lu pu s an ticoagu lan t activity is frequ en tly depen den t on th e presen ce of 2 -glycoprotein I. Thrombosis and Haemostasis, 6 7 , 4 9 9 –5 0 2 . Ordi, J., Vilardel, M., Oristell, J., Valdes, M., Kn obel, A., Alijotas, J., Mon asterio, Y. & Flores, P. (1 9 8 4 ) Bleedin g in patien ts with lu pu s an ticoagu lan t. (Letter). Lancet, ii, 8 6 8 –8 6 9 . Orlan do, E., Cortelazzo, S., March etti, M., San fratello, R. & Barbu i, T. (1 9 9 2 ) Prolon ged bleedin g time in patien ts with lu pu s an ticoagu lan t. Thrombosis and Haemostasis, 6 8 , 4 9 5 –4 9 9 . Ou t, H.J., de Groot, P., van Vliet, M., de Gast, G.C., Niewen h u is, H.K. & Derksen , R.H.W.M. (1 9 9 1 ) An tibodies to platelets in patien ts with an tiph ospholipid an tibodies. Blood, 7 7 , 2 6 5 5 –2 6 5 9 . Rao, V.M., Hoan g, A.D. & Rapaport, S.I. (1 9 9 5 ) Differen ces in th e in teraction s of lu pu s an ticoagu lan t IgG with h u man an d bovin e proth rombin . Thrombosis and Haemostasis, 7 3 , 6 6 8 –6 7 4 . Rosove, M.H. & Brewer, P.M.C. (1 9 9 2 ) An tiph osph olipid an tibodies: clin ical cou rse after th e first th rombotic even t in 7 0 patien ts. Annals of Internal Medicine, 1 1 7 , 3 0 3 –3 0 8 . Sh i, W., Ch on g, B.H. & Ch esterman, C.N. (1 9 9 3 ) 2 -glycoprotein I is a requ iremen t for an ticardiolipin an tibodies bin din g to activated platelets: differen ces with lu pu s an ticoagu lan ts. Blood, 8 1 , 1 2 5 5 – 1262. Stasi, R., Stipa, E., Masi, M., Ceccon i, M., Scimò, M.T., Oliva, F., Sciarra, A., Perrotti, A.P., Adomo, G., Amadori, S. & Papa, G. (1 9 9 5 ) Lon g-term observation of 2 0 8 adu lts with ch ron ic # 1 9 9 6 Blackwell Scien ce Ltd, British Journal of Haematology 9 3 : 1 –5 Annotation idiopath ic th rombocytopen ic pu rpu ra. American Journal of Medicine, 9 8 , 4 3 6 –4 4 2 . Stasi, R., Stipa, E., Oliva, F., Sciarra, A., Perrotti, A., Olivieri, M., Zaccari, G., Gan dolfo, M.G., Galli, M., Barbu i, T. & Papa, G. (1 9 9 4 ) Prevalen ce an d clin ical sign ifican ce of elevated an tiph ospholipid # 1 9 9 6 Blackwell Scien ce Ltd, British Journal of Haematology 9 3 : 1 –5 5 an tibodies in patien ts with idiopath ic th rombocytopen ic pu rpu ra. Blood, 8 4 , 4 2 0 3 –4 2 0 7 . Keywords: an tiph osph olipid an tibodies, an tiph osph olipid syn drome, th rombocytopen ia, th rombocytopen ia treatmen t.