BACKGROUND--High titres of serum antiphospholipid antibodies are a possible pathogenic factor for cardiac lesions in patients with systemic lupus erythematosus. OBJECTIVE--To test the hypothesis of a causal link between high titres of antiphospholipid antibodies in the serum and myocardial involvement in patients without systemic lupus erythematosus. PATIENTS AND DESIGN--18 patients with primary antiphospholipid syndrome (recurrent fetal loss, arterial and/or venous thrombosis, high titres of antiphospholipid antibodies, and no criteria for systemic lupus erythematosus) were prospectively studied by cross sectional, M mode, and pulsed Doppler echocardiography, and compared with 18 healthy controls. The pulsed Doppler indices of left ventricular diastolic function included isovolumic relaxation time and four mitral outflow indices: peak velocity of early flow, peak velocity of late flow, early to late peak flow velocity ratio, and rate of deceleration of early flow. Four computerised M mode indices were also measured: peak rate of left ventricular enlargement in diastole, peak rate of posterior wall thinning, peak velocity of lengthening of the posterior wall, and velocity of circumferential chamber lengthening. RESULTS--Compared with controls, patients with primary antiphospholipid syndrome had higher values for isovolumic relaxation time and peak velocity of late mitral outflow and lower values for early to late mitral peak outflow velocity ratio, rate of deceleration of early mitral outflow, peak rate of left ventricular enlargement in diastole, peak rate of posterior wall thinning, peak velocity of lengthening of the posterior wall and velocity of circumferential chamber lengthening. CONCLUSION--This abnormal pattern reflects an impairment of myocardial relaxation and filling dynamics of the left ventricle in patients with primary antiphospholipid syndrome who were free of any clinically detectable heart disease. These data suggest that high serum titres of antiphospholipid antibodies may be associated with subclinical myocardial damage.