Correlation between plasma levels of selenium and antithrombin‐III

Patients with previous myocardial infarction were tested for antithrombin‐III (AT‐III) activity and selenium levels in their plasma and compared with sex‐ and age‐matched healthy control individuals. Patients and controls showed a positive correlation between AT‐III and selenium levels (r = 0.27, p = 0.015). After calculatory adjustment for this correlation, selenium was found to be significantly negatively correlated with disease. Multivariate analysis of differences between patients and controls indicated that triglyceride levels in serum had the greatest discriminatory ability (r² = 0.169), followed by AT‐III (r² = 0.072) and selenium (r² = 0.056). The increased AT‐III levels were correlated with the use of warfarin and beta blockers in the patients, but these drugs could not explain the comparatively low selenium levels in the patients. Serum total cholesterol and plasma fatty acid composition had no discriminatory power in multivariate testing. The various fatty acids did not show co‐variation with the selenium levels. The clinical significance of these observations is not clear, but they are consistent with the hypothesis that selenium is an important determinant in cardiovascular disease. The relation between AT‐III and selenium should be further evaluated.