To test an immunological hypothesis proposed to explain the pathogenes
is of cerebrovascular thrombosis in steroid users, circulating immune
complexes were assayed in the sera from 6 control subjects, 14 ever us
ers of oral contraceptive having developed a neurological ischaemic ac
cident, and 7 patients with the same clinical history during use of ot
her sex steroid not containing ethinylestradiol. Beaumont's ammonium s
ulfate and polyethylene glycol precipitation methods, together with a
specific method of isolation of circulating immune complexes using aff
inity chromatography on Protein A, were used. Radioactivity from label
ed ethinylestradiol added to the sera before precipitation was monitor
ed in the precipitates to detect anti-ethinylestradiol antibodies. The
re were no significant differences for these parameters in the three g
roups. However, protein content and H-3-EE activity in the precipitate
s were equally and dramatically reduced after affinity chromatography
in the three groups. These latter results do not support the presence
of antibodies against ethinylestradiol in steroid users with cerebrova
scular thrombosis. Moreover, our data suggest a lack of specificity of
Beaumont's method for the isolation of immune complexes containing an
ti-ethinylestradiol antibodies.