Br Heart _J 1995;74:531-535

M mode and Doppler echocardiographicassessment of left ventricular diastolic function inprimary antiphospholipid syndrome

Nicolas Coudray, Dominique de Zuttere, Olivier Bletry, Jean-Charles Piette, BertrandWechsler, Pierre Godeau, Jean-Claude Poumy, Yves Lecarpentier, Denis Chemla

AbstractBackground-High titres of serum anti-phospholipid antibodies are a possiblepathogenic factor for cardiac lesions inpatients with systemic lupus erythemato-sus.Objective-To test the hypothesis of acausal link between high titres of antiphos-pholipid antibodies in the serum andmyocardial involvement in patients with-out systemic lupus erythematosus.Patients and design-18 patients with pri-mary antiphospholipid syndrome (recur-rent fetal loss, arterial and/or venousthrombosis, high titres of antiphospholipidantibodies, and no criteria forsystemic lupus erythematosus) wereprospectively studied by cross sectional, Mmode, and pulsed Doppler echocardiogra-phy, and compared with 18 healthy con-trols. The pulsed Doppler indices of leftventricular diastolic function included iso-volumic relaxation time and four mitraloutflow indices: peak velocity of early flow,peak velocity oflate flow, early to late peakflow velocity ratio, and rate of decelerationof early flow. Four computerised M modeindices were also measured: peak rate ofleft ventricular enlargement in diastole,peak rate of posterior wall thinning, peakvelocity of lengthening of the posteriorwall, and velocity of circumferentialchamber lengthening.Results-Compared with controls,patients with primary antiphospholipidsyndrome had higher values for isovolu-mic relaxation time and peak velocity oflate mitral outflow and lower values forearly to late mitral peak outflow velocityratio, rate of deceleration of early mitraloutflow, peak rate of left ventricularenlargement in diastole, peak rate of pos-terior wall thinning, peak velocity oflengthening of the posterior wall andvelocity of circumferential chamberlengthening.Conclusion-This abnormal patternreflects an impairment of myocardialrelaxation and filling dynamics of the leftventricle in patients with primaryantiphospholipid syndrome who were freeof any clinically detectable heart disease.These data suggest that high serum titresof antiphospholipid antibodies may beassociated with subclinical myocardialdamage.

(Br Heartr_ 1995;74:531-535)

Keywords: antiphospholipid antibodies, left ventriculardiastolic function, myocardial damage.

Recurrent venous and/or arterial thombosis,repeated episodes of fetal loss, and thrombo-cytopenia suggest the possibility of primaryantiphospholipid syndrome. Diagnostic con-firmation depends upon the detection of hightitres of serum antiphospholipid antibodies,without evidence of systemic lupus erythe-matosus or malignant tumour. Cardiac mani-festations have been reported in primaryantiphospholipid syndrome, mainly valvelesions.'-3 Valvar, myocardial, and pericardialabnormalities are also often seen in patientswith systemic lupus erythematosus.45 Severalstudies have found a statistical link betweenhigh titres of antiphospholipid antibodies andheart involvement, suggesting that antiphos-pholipid antibodies may be a key pathogenicfactor for cardiac lesions in systemic lupuserythematosus.6 8 In contrast, recent studiesdid not find any significant correlationbetween antiphospholipid antibodies and valvelesions.9-"' In primary antiphospholipid syn-drome antiphospholipid antibodies are presentwithout the other features of systemic lupuserythematosus that complicate the apprecia-tion of intrinsic cardiac involvement (hyper-tension, steroid or immunosuppressortreatment, renal insufficiency).

In several cardiovascular diseases diastolicfunction may be abnormal when systolic func-tion remains preserved'2: this is usuallyregarded as an initial stage of cardiomyopathy.Early impairment of left ventricular diastolicfunction has been reported in systemic lupuserythematosus, especially in patients withactive disease.'3 14 In the present study wetested whether diastolic function was impairedin patients with primary antiphospholipid syn-drome and no overt cardiac disease.

Patients and methodsPATIENTSWe studied 18 consecutive patients (two men,16 women, mean (SD) age 36 (9)) withAsherson et ars criteria of primary antiphos-pholipid syndrome.'9 We also studied 18 ageand sex matched healthy controls mean (SD)age 33 (8). All gave informed consent to thestudy. All patients had a physical examination,and patients with primary antiphospholipidsyndrome had an electrocardiogram and achest radiograph. We excluded those with ahistory of previous cardiovascular disease(hypertension, valve disease, myocardialinfarction) and/or cardiac abnormality on

INSERM U426-Serviced'ExplorationsFonctionnelles,Hopital Bichat, Paris,FranceD de ZuttereINSERM U275, LOA-ENSTA-EcolePolytechnique,Palaiseau, FranceN CoudrayJ C PoumyD ChemlaService de MedicineInterne, Hopital de laPitie, Paris, France0 BletryJ-C PietteB WechslerP GodeauService d'ExplorationsFonctionnelies Cardio-respiratoires, Hopitalde Bicetre, Paris,FranceY Lecarpentier

Correspondence to:Dr D de Zuttere, INSERMU25 1-Serviced'Explorations,Fonctionnelles, H6pitalBichat, 46 rue HenriHuchard, 75018 Paris,France

Accepted 1 May 1995

531 on A

ugust 11, 2021 by guest. Protected by copyright.

http://heart.bmj.com

/B

r Heart J: first published as 10.1136/hrt.74.5.531 on 1 N

ovember 1995. D

ownloaded from

Coudray, de Zuttere, Blktry, Piette, Wechsler, Godeau, et al

physical examination. All patients were insinus rhythm with a heart rate <90 beats perminute. None of the patients with primaryantiphospholipid syndrome had abnormalitieson the electrocardiogram or chest x ray. Inpatients with primary antiphospholipid syn-drome antiphospholipid antibody titres weremeasured in blood drawn within a week of theechocardiographic study. We regardedantiphospholipid antibodies as being presentwhen we detected circulating anticoagulantand/or a high titre of anticardiolipin antibod-ies. We defined circulating anticoagulant as apositive diluted thromboplastin titration index>1I3 at 1/200 final dilution and high titres ofanticardiolipin antibodies were defined as aconcentration of immunoglobulin G isotype>20 GPL units or a concentration ofimmunoglobulin M isotype > 10 MPL units byenzyme-linked immunosorbent assay.'6

CROSS SECTIONAL M MODE AND DOPPLERECHOCARDIOGRAPHIC EXAMINATIONEchocardiographic examinations were per-formed by one experienced operator (DdZ)using an Aloka SSD 870 ultrasonoscope with2-5 and 3-5 MHz transducers. Measurementswere made in accordance with recommenda-tions of the American Society ofEchocardiography.17 Time-motion and pulsedDoppler recordings were performed on anAloka stripchart thermal recorder at a paperspeed of 100 mm/s. Timing of the cardiaccycle was documented by simultaneous elec-trocardiogram recording.

Left ventricular geometry and systolic functionStandard parastemal long axis and short axisviews and apical views (four chamber and fourchamber with aorta) were obtained. We usedM mode echocardiography to measure leftventricular end diastolic and end systolicdimensions and interventricular septum andleft ventricular posterior wall thickness in dias-tole. Left ventricular fractional shortening andleft ventricular mass were derived from Mmode measurements. Left ventricular masswas calculated according to the method rec-ommended by Devereux et al (left ventricularmass = 0-80 (American Society ofEchocardiography - cube of left ventricularmass) + 0.60).18 Left ventricular segmentalwall motion was assessed systematically usingcross sectional echocardiography. Wall motionabnormalities were assessed by visually evalu-ating wall motion and endocardial thickeningof the different left ventricular wall segments,as recommended by the American Society ofEchocardiography. 19

Assessment of left ventricularfillingPulsed Doppler recordings were obtained inexpiration, with the sample volume locatedbetween the tips of mitral leaflets.20 We mea-sured the peak velocity of early mitral outflow(cm/s), the peak velocity of late mitral outflow(cm/s), early to late peak outflow velocityratio, and the rate of deceleration of velocity inearly diastole (m/s2). All values were averagedout over three consecutive cardiac cycles.

ISOVOLUMIC RELAXATION TIMEThe isovolumic relaxation time, defined as theinterval between aortic valve closure andmitral opening, was measured from beat tobeat recordings of left ventricular filling andejection flows by pulsed Doppler. The samplevolume was located at the left ventricular out-flow tract close to the anterior mitral leaflet (toallow the recording of both aortic closure andmitral opening sounds). Isovolumic relaxationtime was averaged out over three consecutivecycles.

Endocardium, PericardiumWe looked for valve thickening and/or pericar-dial effusion in each patient.

COMPUTERISED DIGITISATION OF M MODE LEFTVENTRICULAR RECORDINGSThe lengthening capabilities of themyocardium during early diastole were pre-cisely quantified by computer-assisted digitisa-tion of M mode echocardiographicrecordings.2' Recordings from the endo-cardium and epicardium of the posterior walland recordings from the interventricular sep-tum were manually traced on a digitisingtablet connected to a microcomputer (HewlettPackard 1000, self-devised software. datapoint recording at 10 ms intervals). The Pennconvention'8 was applied for left ventricularwall tracings. The indices measured were thepeak rate of left ventricular enlargement indiastole (cm/s), the peak rate of posterior wallthinning (cm/s), the velocity of circumferentialchamber lengthening (s-1), and the peakvelocity of lengthening of the posterior wall(cm/s).2I The accuracy of the digitisation pro-cedure was improved by automatically averag-ing each trace at least three times, and eachmeasurement over three consecutive cardiaccycles, as recommended.22

REPRODUCIBILITY OF MEASUREMENTSTo test the interobserver variability forechocardiographic measurements, we ran-domly selected 15 M mode recordings of theleft ventricle, 15 pulsed Doppler recordingsused for isovolumic relaxation time determina-tion, and 15 pulsed Doppler recordings ofmitral outflow. Indices of left ventricular dias-tolic function were measured independentlyby two observers unaware of patients' diag-noses. Interobserver variability was calculatedas the absolute value of the difference betweenthe two measurements divided by the averageof two measurements multiplied by 100%.Mean values of interobserver variability were2-1% (SD 0 4%) for computerised M modemeasurements, and 6-2% (SD 5-3%) formitral outflow indices. Percentage of interob-server variability for isovolumic relaxationtime measurements was 4-3% (SD 3 6%).

STATISTICAL ANALYSISData are mean (SD) values. Groups werecompared by Student's unpaired t test.Differences between groups of p < 0-05 wereregarded as significant.

532 on A

ugust 11, 2021 by guest. Protected by copyright.

http://heart.bmj.com

/B

r Heart J: first published as 10.1136/hrt.74.5.531 on 1 N

ovember 1995. D

ownloaded from

M mode and Doppler echocardiographic assessment of left ventricular diastolic function in primary antiphospholipid syndrome

Table 1 Clinical and echocardiographic indices (mean(SD)) in patients with primary antiphospholipid syndrome(PAPS) and controls

Controls PAPSIndex (n = 18) (n = 18)

Age (y) 33 (8) 36 (9)Heart rate (beat/min) 66 (9) 71 (6)Systolic BP (mm Hg) 121 (9) 128 (14)Diastolic BP (mm Hg) 68 (8) 73 (12)LV mass index (g/m2) 90 (19) 87 (17)LV EDD (mm) 48 (4) 49 (4)LV ESD (mm) 29 (3) 31 (4)FS (%) 37 (5) 37 (5)

BP, blood pressure; EDD, end diastolic diameter; ESD, endsystolic diameter; FS, fractional shortening; LV, left ventricular.

Table 2 Pulsed Doppler and computerisedM modeindices (mean (SD)) in patients with primaryantiphospholipid syndrome and in controls

Control group PAPS(n = 18) group

E wave (cm/s) 74 (12) 69 (14) (n = 18)A wave (cm/s) 45 (6) 54 (16) (n = 18)*E/A 1-68 (0 37) 1-35 (0 45) (n = 18)*IVRT (ms) 85 (15) 95 (15) (n = 18)*EF slope (m/s2) 5 8 (1-5) 4-8 (1-3) (n = 18)*D+ (cm/s) 15-9 (3 3) 11-5 (3 5) (n = 13)tPWT (cm/s) 10-6 (3-1) 8-4 (2.8) (n = 13)*PVL (cm/s) 12-7 (2 8) 9-8 (3 3) (n = 13)*VCL (s-') 4-6 (0 9) 3-4 (1-2) (n = 13)t

E, peak early diastolic filling velocity; A, peak filling velocity atatrial contraction; IVRT, isovolumic relaxation time; EF slope,rate of deceleration of velocity in early diastole; D+, peak rate ofleft ventricular enlargement in diastole; PWT, posterior wallthinning rate; PVL, peak velocity of lengthening of the posteriorwall; VCL, velocity of circumferential chamber lengthening.*P < 0-05 and tP < 0-01 v controls.

EC.

P < 0.05I 7

0110 _

90K

70 +

50

Controls PAPSn= 18 n= 18

Figure 2 Peak late filling velocities (E) of the controlgroup, and ofpatients with primary antiphospholipidsyndrome (PAPS). Shaded area as in fig 1.

P<0.05I 7370

2

wL

@0

oControlsn = 18

PAPSn = 18

Figure 3 EIA ratios of the control group and ofpatientswith primary antiphospholipid syndrome (PAPS). Shadedarea as infig 1.

ResultsThere were eight cases of mild valve disease(three mitral and two aortic regurgitation) andfive small pericardial effusions in the primaryantiphospholipid syndrome group. There wereno differences between the two groups in heartrate, systolic or diastolic arterial blood pres-sure, left ventricular end diastolic and end sys-tolic dimensions, or left ventricular mass andfractional shortening (table 1). This meantthat diastolic indices could be validly com-pared.

PULSED DOPPLER STUDYCompared with controls, patients with pri-mary antiphospholipid syndrome had a pro-longed isovolumic relaxation time (P < 0 05),an increased peak velocity of late mitral out-flow (P < 0 05), and decreased values for therate of deceleration of velocity in early diastole(P < 0 05) and for the early to late peak out-flow velocity ratio (P < 0 05) (table 2 and figs1-5).

P < 0.05I I~~~125Fr

coE

Hr

100

75

50Controls PAPSn= 18 n= 18

Figure 4 Isovolumic relaxation times (IVRT) of thecontrol group and ofpatients with primaryantiphospholipid syndrome (PAPS). Shaded area as infig1.

P < 0.05

10 7

Figure 1 Peak earlyfilling velolcities (E) of thecontrol group, and ofpatients with primaryantiphospholipid syndrome(PAPS). Shaded areaindicates the normal rangewith 95% confidenceintervals.

110 F0

90

COE 700

0

Controlsn = 18

PAPSn = 18

Controls PAPSn= 18 n= 18

Figure 5 Rates of deceleration of velocity in early diastole(EF slope) of the control group and ofpatients withprimary antiphospholipid syndrome (PAPS). Shaded areaas in fig 1.

0

E

U-w

8

6

4

2L

533

30

on August 11, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.74.5.531 on 1 Novem

ber 1995. Dow

nloaded from

Coudray, de Zuttere, Bletry, Piette, Wechsler, Godeau, et al

COMPUTERISED M MODE STUDYThe patients with primary antiphospholipidsyndrome had lower values than the controlsfor the four measured indices: peak rate of leftventricular enlargement in diastole (P < 0-01),posterior wall thinning rate (p < 0 05), peakvelocity of lengthening of the posterior wall(P < 0 05) and velocity of circumferentialchamber lengthening (P < 0-01) (table 2 andfigs 6-9).

Figure 6 Peak rates ofleft ventricular enlargementin diastole (D+) of thecontrol group and ofpatients with primaryantiphospholipid syndrome(PAPS). Shaded area asin fig 1.

p < 0.01

25F

20

15

10

Controlsn = 18

Figure 7 Posterior wallthinning rates (PWT) ofthe control group and ofpatients with primaryantiphospholipid syndrome(PAPS). Shaded area asin figure 1.

IPAPSn = 13

P < 0.0520

15

10

5

Controlsn = 18

Figure 8 Peak velocitiesof lengthening of theposterior wall (PVL) of thecontrol group and ofpatients with primaryantiphospholipid syndrome(PAPS). Shaded area asin fig 1.

Figure 9 Velocities ofcircumferential chamberlengthening (VCL) of thecontrol group and ofpatients with primaryantiphospholipid syndrome(PAPS). Shaded area asin fig 1.

PAPSn = 13

P<0.05207

15

10

5 00

Controlsn = 18

PAPSn = 13

P < 0.01I8

2

Controlsn = 18

DiscussionThis prospective, non-invasive study showedan abnormal Doppler and computerised Mmode pattern of left ventricular diastolic func-tion in a group of 18 patients with primaryantiphospholipid syndrome none of whom hadany clinically detectable manifestation of car-diac disease.We found a prolonged isovolumic relax-

ation time, a disturbed left ventricular earlyfilling pattern, and a decreased myocardiallengthening rate in the primary antiphospho-lipid syndrome group. This may indicateimpaired myocardial relaxation and/or fillingdynamics of the left ventricle23 in thesepatients whose systolic performance indiceswere normal. Similarly, diastolic dysfunctionwith preserved systolic function has beenreported at the early stages of cardiovasculardiseases including hypertension and coronaryartery disease.'2

All except one of our patients were less than45 years old, all of them had normal heartrate, blood pressure, and left ventricular sys-tolic function and mass index; none of themhad any previous history of coronary heart dis-ease. Thus the main physiopathological condi-tions known to alter left ventricular relaxationor filling dynamics were excluded, so that thediastolic function abnormalities can beregarded as being directly related to the illnessitself.We found only two reports of myocardial

involvement in patients with primary antiphos-pholipid syndrome.1524 Asherson et al foundfive cases of myocardial infarction in a series of70 patients with primary antiphospholipid syn-drome.'5 Murphy and Leach reported the caseof a 33 year old man with primary antiphos-pholipid syndrome who died of cardiac failure.Histological examination of the myocardiumshowed extensive platelet thrombi within smallintracardiac arteries with areas of microinfarc-tion. Antiphospholipid antibodies may disruptthe normal coagulation cascade by interferingwith phospholipid surface interactions amongendothelial cells, platelets, and circulating pro-teins.

Myocardial involvement has been moreextensively studied in patients with systemiclupus erythematosus. Left ventricular diastolicdysfunction has been recently reported inlUpUS,'3 14 25 but without any reference to a pos-sible link with high titres of antiphopholipidantibodies. On the other hand, segmental orglobal left ventricular hypokinesis has beenobserved in lupus patients, and has been cor-related to the presence of high titres ofantiphospholipid antibodies.67 Histologicalevidence of inflammation, immune deposits inthe wall of blood vessels of myocardium,26 andwidespread intramyocardial arteriolar throm-bosis with surrounding myocardial necrosisl 25have been reported. A link between the pres-ence of antiphospholipid antibodies andmicrovascular lesions resulting in myocardialischaemia has thus been suspected in patientswith lupus.' 27 Myocardial ischaemia mediatedby antiphospholipid antibody may slow iso-metric relaxation and impair left ventricular

nC'

E0

Ec;

CeE-j

>-

C',

-jC)

534

50

PAPSn = 13

on August 11, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.74.5.531 on 1 Novem

ber 1995. Dow

nloaded from

M mode and Doppler echocardiographic assessment of left ventricular diastolic function in primary antiphospholipid syndrome

filling, in the same manner as reported inpatients with coronary heart disease.'2We fournd an abnormal filling pattern in

patients affected by primary antiphospholipidsyndrome without any previous history ofmyocardial infarction or coronary artery dis-ease. This is further evidence of a causal linkbetween high serum titres of antiphospholipidantibodies and cardiac involvement in primaryantiphospholipid syndrome and, by extension,in systemic lupus erythematosus.

LIMITATIONSPulsed Doppler echocardiography is a simple,reproducible method for studying the relax-ation period and the dynamics of filling of theleft ventricle considered as a pump, but hasseveral limitations, in particular ignorance ofleft atrial pressure level.20 Left ventricularcomputerised M mode recordings describediastolic properties of the left ventricle consid-ered as muscle,20 but for one cross sectionaldiameter only. However, both methodsshowed diastolic abnormalities in patients withprimary antiphospholipid syndrome and thisstrengthens the relevance of our results.

CONCLUSIONWe report previously undescribed abnormali-ties of left ventricular relaxation and early fillingdynamics in a group of patients with primary

antiphospholipid syndrome and no clinicalsymptoms of cardiac disease. Further studiesare needed to determine the relation betweenthe impaired diastolic function associated withhigh titres of serum antiphospholipid antibodiesand the later development of contractile statedisorders. It also remains to be establishedwhether antiplatelet or anticoagulant drugtreatment can prevent diastolic dysfunction inpatients with primary antiphospholipid syn-

drome.

This study was supported by grants from INSERM/CNAMTSand Fondation pour la Recherche Medicale and presented atthe IVth Conference on Non-invasive Cardiology, Limassol,Cyprus, October 1993 de Zuttere D, Coudray N, Bletry 0,

Piette JC, Wechsler B, Godeau P, Glaser E, Pourny JC,Lecarpentier Y, Chemla D. Left ventricuar diastolic dysfunc-tion in primary antiphospholipid syndrome (abstr). Am YCardiac Imaging 1994;8:324.

1 Kaplan SD, Chartash EK, Pizzarello RA, Furie AF.Cardiac manifestations of the antiphospholipid syn-drome. Am HeartJ 1992;124:1331-8.

2 Brenner B, Blumenfeld Z, Markiewicz W, Reisner SA.Cardiac involvement in patients with primary antiphos-pholipid syndrome. JAm Coi Cardiol 1991;18:931-6.

3 Galve E, Ordi J, Barquinero J, Evangelista A, Vilardell M,Soler-Soler J. Valvular heart disease in the primaryantiphospholipid syndrome. Ann Intern Med 1992;116:293-8.

4 Chia BI, Mah EP, Feng PH. Cardiovascular abnormalitiesin systemic lupus erythematosus. Clin Ultrasound1981;9:237-43.

5 Galve E, Candell-Riera J, Pigrau C, Permayer-Miralda G,

Del-Castillo HG, Soler-Soler J. Prevalance, morphologictypes, and evolution of cardiac valvular disease in sys-temic lupus erythematosus. N Engl Jf Med 1988;319:817-23.

6 Leung WH, Wong KL, Lau CP, Wong CK, Liu HW.Association between antiphospholipid antibodies andcardiac abnormalities in patients with systemic lupus ery-thematosus. Am Jf Med 1990;89:411-9.

7 Nihoyannopoulos P, Gomez PM, Joshi J, Loizou S,Walport MJ, Oakley CM. Cardiac abnormalities in sys-temic lupus erythematosus, association with raised anti-cardiolipin antibodies. Circulation 1990;82:369-75.

8 Khamashta MA, Cervera R, Asherson RA, Font J, Gil A,Coltart DJ, et al. Association of antibodies against phos-pholipids with heart valve disease in systemic lupus ery-thematosus. Lancet 1990;335:1541-4.

9 Roldan CA, Shively BK, Lau CC, Gurule FT, Smith EA,Crawford MH. Systemic lupus erythematosus valve dis-ease by transesophageal echocardiography and the role ofantiphospholipid antibodies. J Am Coll Cardiol 1992;20:1127-34.

10 Li EK, Crozier IG, Milne MJ, Nicholls MG, Cohen MG.Lack of association between anticardiolipin antibodiesand heart valve disease in Chinese patients with systemiclupus erythematosus. Lancet 1990;336:504-5.

11 Gleason CB, Stoddard MF, Wagner SG, Longaker RA,Pierangeli S, Harris EN. A comparison of cardiac valvularinvolvement in the primary antiphospholipid syndromeversus anticardiolipin-negative systemic lupus erythe-matosus. Am Hearty 1993;125:1123-9.

12 Grossman W. Diastolic dysfunction in congestive heart fail-ure. NEnglJMed 1991;325:1557-64.

13 Crozier IG, Li E, Milne MJ, Nicholls MG. Cardiacinvolvement in systemic lupus erythematosus detected byechocardiography. AmJ Cardiol 1990;65:1 145-8.

14 Sasson Z, Rasooly Y, Chow CW, Marshall S, Urowitz MB.Impairment of left ventricular diastolic function in sys-temic lupus erythematosus. Am J Cardiol 1992;69:1629-34.

15 Asherson RA, Khamashta MA, Ordi-Ros J, DerksenRHWM, Marchin SJ, Barquinero J, et al. The "primaryantiphospholipid syndrome": Major clinical and serologi-cal features. Medicine (Baltimore) 1989;68:366-74.

16 Harris EN, Gharavi AE, Boey ML, Patel BM, Mackworth-Young CG, Loizou S, et al. Anticardiolipin antibodies:detection by radioimmunoassay and association withthrombosis in systemic lupus erythematosus. Lancet1983;i: 1211-4.

17 Sahn DJ, DeMaria A, Kislo J, Weyman A. Recommenda-tions regarding quantitation in M-mode echocardio-graphy: results of a survey of echocardiographic measure-ments. Circulation 1978;58:1072-7.

18 Devereux RB, Alonso DR, Lutas EM, Gottlieb GJ, CampoE, Sachs I, et al. Echocardiographic assessment of leftventricular hypertrophy: comparison to necropsy find-ings. Am J Cardiol 1986;57:450-8.

19 Schiller NB, Shah PM, Crawford M, DeMaria A,Devereux R, Feigenbaum H, et al. Recommendations forquantitation of the left ventricle by two-dimensionalechocardiography. JAm Soc Echo 1989;2:863-8.

20 de Zuttere D, Touche T, Saumon G, Nitenberg A,Prasquier R. Doppler echocardiographic measurement ofmitral flow volume: validation of a new method in adultpatients. JAm Coll Cardiol 1988;11:343-50.

21 St John Sutton MG, Reichek N, Kastor JA, Giuliani ER.Computerized M-mode echocardiographic analysis ofleft ventricular dysfunction in cardiac amyloid.Circulation 1982;66:790-9.

22 Pollick C, Fitzgerald PJ, Popp RL. Variability of digitizedechocardiography: size, source and means of reduction.Am _J Cardiol 1983;51:576-82.

23 Nishimura RA, Housmans PR, Hatle LK, Tajik AJ.Assessment of diastolic function of the heart: backgroundand current applications of Doppler echocardiography.Part I. Physiologic and physiopathologic features. MayoClin Proc 1989;64:71-81.

24 Murphy JL, Leach IH. Findings at necropsy in the heart ofa patient with anticardiolipin syndrome. Br Heart J1989;62:61-4.

25 Leung WH, Wong KL, Lau CP, Wong CK, Cheng CH,Tai YT. Doppler echocardiographic evaluation of leftventricular diastolic function in patients with systemiclupus erythematosus. Am HeartJ 1990;120:82-7.

26 Bidani AK, Roberts JL, Schwartz MM, Lewis EJ.Immunopathology of cardiac lesions in fatal systemiclupus erythematosus. Am J Med 1980;69:849-58.

27 Brown JH, Doherty CC, Allen DC, Morton P. Fatal car-diac failure due to myocardial microthrombi in systemiclupus erythematosus. B MedJ 1988;296:1505.

535 on A

ugust 11, 2021 by guest. Protected by copyright.

http://heart.bmj.com

/B

r Heart J: first published as 10.1136/hrt.74.5.531 on 1 N

ovember 1995. D

ownloaded from