antiphospholipid syndrome: five year follow

Annals ofthe Rheumatic Diseases 1991; 50: 805-810

Antiphospholipid syndrome: five year follow up

Ronald A Asherson, Elaine Baguley, Chobie Pal, Graham R V Hughes

Lupus Arthritis ResearchUnit, The RayneInstitute, St Thomas'sHospital, LondonSEI 7EHR A AshersonE BaguleyC PalG R V HughesCorrespondence to:Dr R A Asherson,Division of Rheumatology andConnective Tissue Diseases,St Lukes/Roosevelt HospitalCenter, Columbia University,New York, NY 10019, USA.

Accepted for publication18 March 1991

AbstractNineteen patients out of 250 subjects withantiphospholipid antibodies, who had initiallypresented to the lupus clinic at St Thomas'sHospital, London five or more years ago witha history of venous/arterial occlusions, wereentered into the study. The patients weredivided into two main groups: I those whoremained well without any further thrombo-embolic complications (n= 10); H those whodeveloped recurrent thrombotic events in thefive year period (n=9).The patients were followed up to determine

the relation between the level or the isotype ofthe anticardiolipin antibodies, or both, to therecurrent thromboembolic events, and theeffect of a variety of treatments (corticoster-oids, immunosuppression, anticoagulation) inthe prevention of further vascular occlusions.Lupus activity over the five year period variedconsiderably between the two groups-thosein group I tending to be relatively inactivecompared with those in group H. For somepatients in group H thromboembolic eventsseemed to occur at the time of lupus activity.

Antiphospholipid antibodies remained posi-tive in all patients, the levels remaining fairlyconstant. Levels fell in only one patient ingroup I and in two in group H. Patients ingroup II had more systemic lupus erythe-matosus related disease than those in group I;most were receiving concomitant steroid andimmunosuppressive therapy, but this did notseem to protect against the development offurther occlusions. All patients were givenanticoagulation treatment (warfarin/heparin)or salicylates (low dose aspinn 75 mg daily),or both. Patients with deep vein thrombosesdeveloped more complications during anti-coagulation therapy than those with cerebro-vascular symptoms. Problems in anticoagu-lation control and recurrent thrombosesconsequent on warfarin withdrawal despitethe administra'tion of subcutaneous heparinwere responsible for complications in mostpatients in group H.

Antiphospholipid antibodies-those antibodiesresponsible for the in vitro lupus anticoagulanttest-as well as antibodies directed againstnegatively charged phospholipids, of whichcardiolipin is one example, are associated with awide variety of clinical complications as aconsequence of thrombosis of veins as well asarteries 2: recurrent fetal loss,3 haematological(thrombocytopenia, Coombs' positivity/hae-molytic anaemia), dermatological (livedo

reticularis, skin nodules/necrosis), and neuro-logical (chorea, transverse myelitis, dementia)complications.4 These complications in about50%/o of patients occur in association withsystemic lupus erythematosus (SLE) or 'lupus-like' disease, a variant of SLE, with patientsessentially showing less than four of the revisedcriteria for the classification of SLE as laiddown by the American Rheumatism Associationin 1982.5

Criteria for an 'antiphospholipid syndrome'were proposed in 1987 by Harris et al,6 compris-ing essentially the combination of vascularocclusions and thrombocytopenia accompaniedby persisting increases of anticardiolipin anti-bodies or presence of the lupus anticoagulant,or both. The antiphospholipid syndrome in50% of patients is seen without any clinical orserological evidence of SLE-the recentlydefined 'primary' antiphospholipid syn-drome.7-9

It is still unclear whether these antibodiesalone are the direct cause of thrombotic eventsby a number of differing mechanisms, orwhether they constitute 'markers' present inmost patients with clotting problems and lupus-type illness.

This study aimed at determining whetherthere was any quantitative change in antibodylevels preceding the recurring thromboticevents in patients originally presenting with anantiphospholipid syndrome and what, if any,other 'trigger' factors might contribute to theoccurrence of such events in these susceptiblepatients over a five year period.

Patients and methodsAll patients were defined as having an 'anti-phospholipid syndrome' according to the criteriaproposed in 1987.6 Patients with recurrent fetallosses were excluded as they will form part of aseparate study. All patients were thrombocyto-penic either initially or at some point during thefive years. Fluctuations in platelet counts werenot considered in the study. Anticardiolipinantibody estimations were performed at three tosix monthly intervals at routine follow up visits.The anticardiolipin antibody levels weredefined as low (for IgG and IgM anticardiolipinantibody) (<20 units); moderate (20-80 units);and high (>80 units).The lupus anticoagulant was not routinely

measured at follow up as most patients werereceiving anticoagulant treatment. Estimationsof protein C, protein S, and antithrombin IIIconcentrations were also not measured forsimilar reasons. Blood of patients with a history

805

on April 23, 2022 by guest. P

rotected by copyright.http://ard.bm

j.com/

Ann R

heum D

is: first published as 10.1136/ard.50.11.805 on 1 Novem

ber 1991. Dow

nloaded from

http://ard.bmj.com/

Asherson, Baguley, Pal, Hughes

of deep vein thromboses unrelated to high risksituations, such as oral contraceptive use orpregnancy, was anticoagulated with warfarin(Coumadin). Long term anticoagulation wascontinued in the presence of positive anticardio-lipin antibody tests. Subcutaneous/intravenousheparin was used when warfarin was withdrawnfor particular reasons-for example, beforebiopsies or in preparation for surgery.

In patients with cerebrovascular accidentsalone the presence of valve abnormalities sug-gested the possibility of embolic events ratherthan in situ thrombosis, and in most of thesepatients aspirin was used initially, with warfarinadded if recurrent cerebrovascular or otherevents supervened.Lupus disease activity was assessed and

further treatments were introduced whenrequired. These therapies were not used solelyfor the attempted treatment of thrombosis.

For analytical purposes the patients weredivided into two groups: I those without furtherthromboembolic complications (n= 10); II thosewho developed recurrent thrombotic events(n=9).

All patients had complete histories taken,including social, medical, and family details, atthe start of the study. The family historyincluded specific questions about the presenceof a connective tissue disease or coagulopathy infamily members. All patients had routine bloodcounts as well as biochemical and immunologicalevaluations during the study period. Anti-cardiolipin antibodies were determined by an

enzyme linked immunosorbent assay (ELISA),as modified by Gharavi et al. 0 The lupusanticoagulant was evaluated by the method ofExner. l

ResultsGROUP I (table 1)Five patients were diagnosed as having SLEinitially. Four had 'lupus-like' disease at theonset, with evolution to SLE after five years inone. One patient was defined as having a

'primary' antiphospholipid syndrome. Fourpatients had a cerebrovascular accident only,which in one had developed during oral contra-ceptive use. A fifth patient had a history of two

deep vein thromboses occurring before thecerebrovascularaccident-one in the postpartumperiod and one during oral contraceptive use.Four patients had deep vein thromboses, and inone this was accompanied by a retinal arterythrombosis. One patient had an arterial occlusionof the leg.Two of the five patients with cerebrovascular

accidents were treated with aspirin as the soleanticoagulant. In one the aspirin was accom-panied by prednisolone and in the other byprednisolone and azathioprine. The third patientwith a cerebrovascular accident was treated withwarfarin alone. The fourth patient with acerebrovascular accident, which followed recur-rent deep vein thromboses, was also treatedwith aspirin accompanied by prednisolone. Thefifth patient who had developed a cerebro-vascular accident while taking dipyridamole wasthen changed to warfarin without the additionof prednisolone or azathioprine and hasremained well. Therefore, of the five patientswith cerebrovascular accidents, none of whomdeveloped further thrombotic complications,three had been treated with aspirin alone as theanticoagulant and two had been treated withwarfarin only.Of the five patients with deep vein throm-

boses, two had received warfarin alone, one

aspirin alone, and two warfarin plus aspirin.Concomitant low dose prednisolone had beengiven to four of these five patients.The patient with arterial occlusion of the leg

despite persistently high levels of IgM anti-cardiolipin antibodies throughout the total fiveyear period has remained well with long termwarfarin treatment and has had no furtherocclusions. This patient is also a heavy smoker,who continued to smoke despite medical adviceto the contrary, as well as having hyper-cholesterolaemia.Of the 10 patients in group I, nine were

positive for anticardiolipin antibodies and onewas negative. All nine patients were positive forIgG anticardiolipin antibodies throughout theperiod, one patient converting from a high to amoderate positive value within the five yearperiod and also from a low to a negative IgManticardiolipin antibody level in the same period.Two patients, one with a cerebrovascular

Table I Antiphospholipid syndrome: group I*

Case Diagnosis Initial APS Further Treatment Antiphospholipid antibodiesNo associated complications

Onset At 5 years manifestations Anticoagulation Prednisolonel IgG IgM LAtimmunosuppression aCLt aCL

1 SLEt SLE CVAt during OCt use Nil Aspirin Prednisolone Neg Neg +2 SLE SLE CVA Nil Aspirin Prednisolone+ LS Mu-L +

azathioprine3 LLDt SLE DVTt Nil Warfarin Prednisolone M - +4 1° APSt 1° APS CVA while taking Nil Warfarin Nil M - ND

dipyridamole5 LLD LLD CVA Nil Warfarin Nil HS - +6 SLE SLE DVT Nil Warfarin Prednisolone H*M L*Neg +7 LLD LLD DVTx2,t CVA Nil Aspirin Prednisolone M - +8 SLE SLE DVT Nil Warfarin+aspirin Prednisolone L - ND9 SLE SLE DVT, retinal artery thrombosis Nil Warfarin+aspirin Nil M - +10 LLD LLD Arterial occlusion (leg), Nil Warfarin Nil H +

(arteriosclerosis, smoker + +)

.Patients without further thromboembolic complications during the five year follow up.tSLE=systemic lupus erythematosus; LLD=lupus-like disease; 1° APS=primary antiphospholipid syndrome; CVA=cerebrovascular accident, OC=oralcontraceptive; DVT=deep vein thromboses; aCL=anticardiolipin antibody; LA=lupus anticoagulant.tOne episode post partum, once during oral contraceptive use.§L=low (<20 units); M=moderate (20-80 units), H=high (>80 units).

806



j.com/

Ann R

heum D


ber 1991. Dow

nloaded from

http://ard.bmj.com/

Antiphospholipid syndrome

accident and one with deep vein thromboses,had low IgG anticardiolipin antibodies; in thefirst patient this was accompanied by a moderateincrease of the IgM anticardiolipin antibodies,which fell to a low level at the end of the fiveyear period. The lupus anticoagulant test was

positive in all eight patients in whom it was

performed.

GROUP II (table 2)Seven of the 9 patients had had deep veinthromboses (accompanied by pulmonary emboliin two), while in two the cerebral vessels were

affected (transient ischaemic attacks in one

patient and cerebrovascular accident in one). Inthe patients with recurrent deep vein throm-boses, despite long term treatment with war-

farin, further thrombotic events occurred in all.Further deep vein thromboses were associated

with warfarin withdrawal in five patients.Warfarin was withdrawn for one of the follow-ing reasons: (a) it was thought that the anti-phospholipid antibody levels in one patienthad reverted to negative. After withdrawal,however, it was found that although the anti-cardiolipin antibody levels were in fact negative,a lupus anticoagulant was nevertheless stilldemonstrable; (b) as a precaution, because ofcaesarean section in a pregnant patient; (c) toundertake renal biopsies in two further patients;and (d) bloody diarrhoea in one patient.

The use of subcutaneous heparin (15 000units daily) was ineffective in preventing furtherdeep vein thromboses and pulmonary emboli intwo patients. Absorption of heparin subcut-aneously may be erratic and the bioavailabilityreduced. Poor compliance was responsible forinadequate anticoagulation control in one patientand transiently inadequate anticoagulationcontrol (international normalisation ratio <2) ina further three patients. In one of these, despitebeing stable with anticoagulation therapy, twoepisodes of apparent warfarin resistance occur-

red with an inadequate international normalis-ation ratio despite doubling the oral warfarindose. These episodes lasted for three weeks, thepatient then reverting to normal control withthe previous warfarin dosage. Eight of the ninepatients were anticardiolipin antibody positiveand remained positive throughout the fiveyears. Four were high positive for IgG anti-cardiolipin antibodies, one of whom hadchanged to a moderate positive level at the endof the five years. Two were moderate IgGanticardiolipin antibody positive and two were

low positive. Initially low positive IgG anti-cardiolipin antibodies in one patient hadreverted to negative at the end of the period. Inthis patient moderately increased IgM anti-cardiolipin antibodies had changed to low levelsof the same isotype. One patient with moderatelyincreased IgG anticardiolipin antibodies had an

accompanying low level of the IgM isotype.

Table 2 Antiphospholipid syndrome: group II*

Case Diagnosis Initial APS Treatment Further Treatment AntiphospholipidNo associated complications antibodies

Onset At S manifestations Anticoagulation Prednisolonel Anticoagulation Prednisolonelyears immunosuppresston immunosuppression IgG IgM LAt

aCLt aCL

1 SLEt SLE TIAt Warfarin + Azat TIA Warfarin+ + Aza H - +aspirin

2 SLE SLE DVTt Warfarin + - DVT (after Warfarin + - L*N MwL +(withdrawn: warfarin reinstitutedrenal biopsy) withdrawal)

3 LLDt SLE DVT (4), Warfarin - - Axillary artery Short term - - M - +PEt thrombosis intravenous

heparin andepoprostenol

Reocclusion Phenindioneaxillary artery

PE Phenindione4 LLD SLE DVT, fetal Warfarin + Aza PE Phenindione + Aza H*M - +

losses5 SLE SLE DVT Warfarin + Aza DVT Warfarin (withdrawn: L - ND

(withdrawn: renal biopsy),caesarean SC heparinsection) (5000 U 8 hourly)

DVT Intravenous heparim + +(48 h), warfarin (pulseand aspirin (cyclophosphamide)

6 SLE SLE DVT (3) Warfarin, + Aza Axillary vein Warfarin + Aza - - +nicoumalone thrombosis (INRt low)

DVT and PE Warfarin(dose omitted)

Angina/Mlt Warfarin ( T dose)CVA Warfarin+

aspirinSubclavian

vein occlusion7 LLD LLD DVT Warfarin + Aza DVT Warfarin - - H - +8 1IAPSt 1°APS DVT (3) Warfarin - - DVT Warfarin - - H - +

(poor control)CVA T Warfarin

9 1° APS LLD CVAt (2) Warfarin - - DVT Warfarin+aspirin + Aza M L +Warfarin+ CVA Phenindione,

aspirin aspirin

*Patients who developed recurrent thrombotic events during the five year foliow up.tSLE=systemic lupus erythematosus; LLD=lupus-like disease; I' APS=primary antiphospholipid syndrome; TIA=transient ischaemic attacks; DVT=deep veinthromboses; PE=pulmonary embolus; CVA=cerebrovascular accident, Aza=azathioprine; MI=myocardial infarction; aCL=anticardiolipin antibodies-H=high,M=moderate, L=low, N=negative; LA=lupus anticoagulant.tINR=intemational normalisation ratio.

807



j.com/

Ann R

heum D


ber 1991. Dow

nloaded from

http://ard.bmj.com/


DiscussionThe patients were divided into two groups-withand without thromboembolic complications-toassess whether this arbitrary division mightperhaps provide some clues as to the reasonswhy some patients with antiphospholipid anti-bodies developed further complications andwhether these complications bore any relationto the levels or isotype of anticardiolipinantibody, or to the type of treatment given. Itcan be seen from the tables that in severalpatients the initial thrombotic event was relatedto factors such as oral contraceptive use (in twopatients), occurred in the postpartum period inthree, at which time (together with late preg-nancy) a hypercoagulable state exists. Inadequatetreatment with warfarin was responsible forrecurrent thromboembolic events in fourpatients. An international normalisation ratio 9fless than three is clearly ineffective in this groupof patients, and on further investigation it wasfound that most doctors attending anticoagu-lation clinics were content with a ratio in therange of 2 to 2 5. It is therefore necessary thatattending doctors should be aware of the highrisk of thrombosis reoccurring in this group ofpatients and they should maintain the ratio at ahigher therapeutic level than necessary inpatients without antiphospholipid antibodies.

Warfarin resistance with a dose four to fivetimes greater than normal (20 mg/day) wasnecessary to achieve normalisation ratios ofmore than three in two patients. One of thesealso had a similar resistance to phenindione andmore than 100 mg daily was necessary. Thisphenomenon has been noted in several otherpatients with antiphospholipid antibodies, whoare not included in this study.

Subcutaneous heparin seemed to be ineffectivein the prevention of recurrent thromboses in adosage of 5000 units eight hourly. Warfarin hadbeen withdrawn and SC heparin started as atherapeutic anticoagulation 'cover' in severalpatients undergoing caesarean section or renalbiopsy. Intravenous heparin in a dose of 40 000units daily seemed to prevent total occlusion ofan axillary artery as well as a threatenedmyocardial infarction in another patient. Onepatient had developed a deep vein thrombosis atthe onset of acute nephritis in the absence of anephrotic syndrome. Deep vein thrombosisoccurred in another patient in the postpartumperiod concomitant with a flare of her lupuswith renal disease and a nephrotic syndrome.

Patients with a nephrotic syndrome may havea spectrum ofhaemostatic disorders, particularlythromboembolism. Four major abnormalitiesmay contribute to this hypercoagulable state,including abnormalities in the coagulationcascade, deficiency of specific coagulationinhibitors, such as antithrombin III, protein C,and protein S, abnormalities of the fibrinolyticsystem, or alterations in platelet function,particularly increased platelet adhesion andaggregation. 12-18The levels of anticardiolipin antibody in both

groups were almost identical, with similarnumbers of patients being in the high, medium,and low positive IgG anticardiolipin antibodyrange. IgM anticardiolipin antibodies were only

present in two patients in group I and two ingroup II.

Anticardiolipin antibody levels changed inonly three patients over the five year period,and in these, levels actually fell. There did notseem to be any sudden rise of anticardiolipinantibody levels in the three to six weeks beforethe recurrent thrombotic events. It is thereforeconcluded that the actual levels ofanticardiolipinantibody as monitored in the low, medium, andhigh ranges did not influence the developmentof complications in these patients.

Six patients in group I and six in group IIwere receiving prednisolone treatment. Onlyone patient was receiving azathioprine in groupI, whereas four in group II were receiving it.Pulse cyclophosphamide had been given to twopatients in group II but not to any in group I.Background azathioprine treatment did notseem to prevent recurrent complications, nordid the long term administration ofprednisolone.Because only two patients had received pulsecyclophosphamide treatment (for acute renaldisease in both) no assessment of the efficacy ofthis form of treatment in the prevention ofthrombotic complications is possible. The useof dipyridamole seemed to be ineffective alonein the prevention of thrombosis. Salicylates (lowdose aspirin 75 mg daily) were effective in theprevention of further thrombotic events in threepatients in group I (three with previous cerebro-vascular accidents-one additionally had twoprevious deep vein thromboses: during oralcontraceptive use and in the postpartum period).The salicylates had been combined with pred-nisolone in two ofthese patients and azathioprineas well in the third. Warfarin alone preventedfurther complications in three patients (twowith cerebrovascular accidents, one with peri-pheral arterial occlusion) in group I. Additionalaspirin was required to prevent recurrenttransient ischaemic attacks in one patient ingroup II, however.There have been few previously published

chronological or sequential studies on patientswith antiphospholipid antibodies. Elias andEldor found similar results in their study donein 1984,19 anticoagulant treatment being effec-tive in the prevention of further vascularocclusions. Out et al recently followed up 53patients with SLE over a four year period, themean duration of follow up being 26 months,and compared the anticardiolipin antibodylevels with disease activity and treatment withprednisolone.20 Forty per cent (for the IgGisotype) and 45% (for the IgM isotype) ofpatients remained in the same anticardiolipinantibody category detected at entrance to thestudy. Marked increases from negative to highpositive were not accompanied by any evidenceof thrombosis or thrombocytopenia. Fluctu-ations of levels occurred irrespective of changesin prednisolone dose. Patients receiving lowdose prednisolone or none at all also changedtheir category. In 23% of patients, increase ofIgG anticardiolipin antibodies was significantlyassociated with thrombosis, whereas duringmoderate or severe disease activity the corre-lation was not significant. Association of theIgM anticardiolipin antibody levels with

808



j.com/

Ann R

heum D


ber 1991. Dow

nloaded from

http://ard.bmj.com/

Antiphospholipid syndrome

thrombotic events was, however, independentof lupus flares and remissions and was signifi-cant. These authors suggested that because ofthe great variability in anticardiolipin antibodyconcentrations (which is only partially explainedby disease activity and prednisolone treatment)at least two samples should be taken forscreening purposes-one during low dose pred-nisolone treatment and one in its absence. Theauthors also concluded that the IgG anticardio-lipin antibody levels were of limited value inidentifying patients at risk during exacerbations.The IgM anticardiolipin antibody, however,was thought by them to be more reliable as apredictor of future risks. Other authors havealso studied variations ofanticardiolipin antibodylevels during disease activities. Kalunian et alalthough finding a positive association betweenanticardiolipin antibodies and thrombosis, fetalloss, and thrombocytopenia, found no corre-lation of disease activity with anticardiolipinantibody levels.21

Recently, a retrospective study on patientswith SLE was performed by Ishii et al.22 Eightypatients were classified as having active disease.Sixty nine of 155 (45%) were positive for IgGanticardiolipin antibodies. Serial measurementsof IgG anticardiolipin antibodies allowed thepatients to be further classified into subgroups:(a) those who were persistently positive for IgGanticardiolipin antibodies, and (b) those whowere only positive in the active phases of thedisease. The level of IgG anticardiolipin anti-bodies was significantly higher in group (a) andthe incidence of thromboses, spontaneous abor-tions and lupus anticoagulant positivity was alsosignificantly higher in this group. The incidenceof these events was very low in group (b),suggesting that those patients who presentedwith persistently high levels of IgG anticardio-lipin antibodies, irrespective of disease activityand treatment, are at risk for thromboses andspontaneous abortion. From this study it alsoseemed that IgG anticardiolipin antibodies ingroup (a) patients could not be reduced bytreatment with corticosteroids, nor apparentlywere they reducible by immunosuppressanttherapy. In group (b), however, with moreactive disease and a high incidence of renaldisease, anticardiolipin antibodies decreasedeasily in response to either corticosteroids,immunosuppressants, or both.

Drenkard et al studied six patients with SLEand found that falls of antiphospholipid anti-bodies occurred within two weeks (averageeight days) of a thrombotic episode.23 Allpatients had high titres of IgG anticardiolipinantibodies before the thrombosis. High levels ofIgM anticardiolipin antibodies in three patientsalso fell coincidentally with the thrombo-occlusive event. A similar occurrence had beenpreviously reported by Sturfelt et al.24 Drenkardet al postulated that consumption of anti-phospholipid antibody occurred at the time ofthe thromboses. Treatment was excluded as avariable by these investigators.Gladman and Urowitz in an earlier study25

reviewed the incidence of venous syndromesreported by other authors as well as reportingon 17 of their own patients, who had 21

episodes of venous thromboses or pulmonaryembolism, or both. In 16 of the 21 episodes theSLE was considered to be active in 12 patients.Of great interest was the finding that focalglomerulonephritis, mesangial involvement aswell as mesangial sclerosis, was present in fivepatients with nephrotic syndrome and yet allhad experienced venous thromboses or pul-monary embolism, or both. The lupus anti-coagulant was detected in only three of theirpatients at the time of the occlusion. Other riskfactors in their patients, such as the use of oralcontraceptives, excessive smoking, raisedcholesterol and triglyceride concentrations,were also assessed. Only one patient had takenoral contraceptives for two years up to the timeof the thrombosis. Eight of the 17 smoked andin three increased lipid concentrations weredetected. Angles-Cano etal subsequently studied28 patients with SLE,26 five of whom had hadthrombotic episodes, which had occurred con-currently with a flare of the disease in four ofthese five patients. In three, the lupus anti-coagulant was present but 12 patients (43%) hadevidence of prolongation of the activated partialthromboplastin time and prolonged pro-thrombin time in the absence of any thrombo-embolic complications. Peck et al reported on14 of 114 patients who had episodes of thrombo-phlebitis with SLE27 and concluded that thepresence of active SLE alone might be sufficientto increase the risk of thrombosis, though otherpredisposing risk factors might have beenoperating. Three of their 14 patients wereseriously ill and in bed when the occlusionsoccurred. Nine, however, had no known pre-disposing factors.

ConclusionsFrom this series it seems that there were nobroad significant differences in the levels ofanticardiolipin antibodies between those patientswho developed recurrent thromboembolic events(group II) and those who did not (group I).There was no rise in anticardiolipin antibodytitre preceding any thrombotic events and thisfact concurs with the findings of Out et al20 thatmarked increases from negative to high positivelevels were not associated with thromboticevents. These antibodies, nevertheless, stillrepresent excellent markers for patients at riskfor vascular occlusions. It should be emphasised,however, that anticardiolipin antibodies werenot determined immediately preceding thethrombotic events.Thrombotic events tended to occur at times

of hypercoagulability-for example, in the post-partum period and after the administration oforal contraceptives in several patients in groupI, and were precipitated by problems in anti-coagulation control in the group II patients.Some patients required large doses of anti-coagulants, but this apparent insensitivity toanticoagulants was transient. It is suggested thatideally, the international normalisation ratioshould be kept above three in patients withantiphospholipid antibodies. Subcutaneousheparin in a dosage of 5000 units eight hourlydid not prevent recurrent thromboses in one

809



j.com/

Ann R

heum D


ber 1991. Dow

nloaded from

http://ard.bmj.com/


patient at a time of identified risk. The longterm use of azathioprine or prednisolone treat-ment, or both, to control activity of the SLEalso did not seem to prevent further throm-boembolic events in the group II patients.Patients with recurrent deep vein thrombosesdid less well than those with cerebrovascularaccidents, with problems related to maintainingan adequate normalisation ratio being common.The use of low dose salicylate in some patientswith cerebrovascular accidents was effective andits addition to warfarin treatment controlledrecurrent transient ischaemic attacks in onepatient with valvular disease.

It is also of interest that of the three patientsconforming to a 'primary' antiphospholipidsyndrome, only one changed category over thefive year period of the study to lupus-likedisease.

This work was supported by the British SLE Aid Group and theJean Shanks Foundation. Thanks are due to Mrs Barbara Issomfor typing the manuscript.

1 Boey M L, Colaco C B, Gharavi A E, Elkon K B, Loizou S,Hughes G R. Thrombosis in systemic lupus erythematosus:striking association with the presence of circulating anti-coagulant. BMJ 1983; 287: 1021-3.

2 Harris E N, Gharavi A E, Boey M L, et al. Anticardiolipinantibodies: detection by radioimmunoassay and associationwith thrombosis in systemic lupus erythematosus. Lancet1983; ii: 1211-4.

3 Derue G J, Englert H J, Harris E N, et al. Fetal loss insystemic lupus: association with anticardiolipin antibodies.J Obstet Gynecol Neonatal Nurs 1985; 5: 207-9.

4 Asherson R A, Harris E N. Anticardiolipin antibodies-clinical associations. Postgrad Med J 1987; 62: 1081-7.

5 Tan E M, Cohen A S, Fries J F, et al. The 1982 revisedcriteria for the classification ofsystemic lupuserythematosus.Arthritis Rheun 1982; 25: 1271-7.

6 Harris E N, Baguley E, Asherson R A, Hughes G R V.Clinical and serological features of the "antiphospholipidsyndrome" (APS) [abstract]. BrJ Rheunatol 1987; 26: 19.

7 Asherson R A, Khamashta M A, Ordi-Ros J, et al. The"primary" antiphospholipid syndrome: major clinical andserological features. Medicine [Baltimore] 1989; 68: 366-74.

8 Mackworth-Young C G, David J, Louizou S, Walport M J."Primary" antiphospholipid syndrome: features of patients

with raised anticardiolipin antibodies and no other dis-orders. Ann Rhewm Dis 1989; 48: 362-7.

9 Alarcon-Segovia D, Sanchez-Guerrero J. Primary anti-phospholipid syndrome. J7 Rheumatol 1989; 16: 482-8.

10 Gharavi A E, Harris E N, Asherson R A, Hughes G R V.Anticardiolipin antibodies: isotype distribution andphospholipid specificity. Ann Rheum Dis 1987; 46: 1-6.

11 Exner T. Similar mechanisms of lupus anticoagulants.Thromb Haemost 1985; 53: 15-18.

12 Llach F. Nephrotic syndrome: hypercoagulability, renal veinthrombosis and other thromboembolic complications. In:Brenner B M, Stein J H, eds. Nephrotic syndrome. Vol 9.Contemporary issues in nephrology. New York: ChurchillLivingstone, 1981: 12144.

13 Vasiri D, Brenton H E, Ness R. Changes of coagulationfactors IX, VIII, VII, X and V in nephrotic syndrome. AmJ Med Sci 1980; 280: 167-71.

14 Kaufmann R H, Veltkamp J J, Van Tilburg N H, Van EsL A. Acquired antithrombin III deficiency in thrombosis inthe nephrotic syndrome. AmJ Med 1978; 65: 607-13.

15 Vigano D'Angelo S, D'Angelo A, Kaufmann C, et al. ProteinS deficiency occurs in the nephrotic syndrome. Ann InternMed 1987; 107: 42-7.

16 Morrison B W. Protein S and the nephrotic syndrome. AnnIntern Med 1988; 108: 152.

17 Thomson C, Forbes C D, Prentice C R, Kennedy A C.Changes in blood coagulation and fibrinolysis in nephroticsyndrome. AmJ Med 1974; 43: 399-407.

18 Kuhlmann U, Steurer J, Siegenthaler W. Platelet aggregationand B-thromboglobulin levels in nephrotic patients withand without thrombosis. Clin Nephrol 1981; 15: 229-35.

19 Elias M, Eldor A. Thromboembolism in patients with the"lupus" type circulating anticoagulant. Arch Intern Med1984; 144: 510-5.

20 Out J H, De Groot P G, Hasselaar P, van Vliet M, DerksenR H W M. Fluctuations of anticardiolipin antibody levelsin patients with systemic lupus erythematosus: a prospec-tive study. Ann Rheum Dis 1989; 48: 1023-8.

21 Kalunian K C, Peter J B, Middlekauff H R, et al. Clinicalsignificance of a single test for anticardiolipin antibodies inpatients with systemic lupus erythematosus. Am J Med1988; 85: 602-8.

22 Ishii Y, Nagasawa K, Mayumi T, Niho Y. Clinical importanceof persistence of anticardiolipin antibodies in .systemiclupus erythematosus. Ann Rheum Dis 1990; 49: 387-90.

23 Drenkard C, Sanchez-Guerrero J, Alarcon-Segovia D. Fall inantiphospholipid antibody and time of thrombo occlusiveepisodes in systemic lupus erythematosus. J Rheumatol1989; 16: 614-7.

24 Sturfelt G, Nived 0, Norberg R, Thorstensson R, Krook K.Anticardiolipin antibodies in patients with systemic lupuserythematosus. Arthritis Rheum 1987; 30: 382-8.

25 Gladman D D, Urowitz M B. Venous syndromes andpulmonary emoblism in systemic lupus erythematosus. AnnRhewn Dis 1980; 39: 340-3.

26 Angles-Cano E, Sultan Y, Clauvel J P. Predisposing factors tothrombosis in systemic lupus erythematosus. Possiblerelation to endothelial cell damage. J7 Lab Clin Med 1979;94: 312-23.

27 Peck B, Hoffman G S, Frank W A. Thrombophlebitis insystemic lupus erythematosus. JAMA 1978; 240: 1728-30.

810



j.com/

Ann R

heum D


ber 1991. Dow

nloaded from

http://ard.bmj.com/

antiphospholipid syndrome: five year follow

Documents