Whether lipoproteins are better markers than lipids and lipoproteins for coronary heart disease is widely debated. Our aim was to compare the apolipoproteins and cholesterol as indices for risk of acute myocardial infarction.

We did a large, standardised case-control study of acute myocardial infarction in 12 461 cases and 14 637 age-matched (plus or minus 5 years) and sex-matched controls in 52 countries. Non-fasting blood samples were available from 9345 cases and 12 120 controls. Concentrations of plasma lipids, lipoproteins, and apolipoproteins were measured, and cholesterol and apolipoprotein ratios were calculated. Odds ratios (OR) and 95% CI, and population-attributable risks (PARs) were calculated for each measure overall and for each ethnic group by comparison of the top four quintiles with the lowest quintile.

The apolipoprotein B100 (ApoB)/apolipoprotein A1 (ApoA1) ratio had the highest PAR (54%) and the highest OR with each 1 SD difference (1·59, 95% CI 1·53–1·64). The PAR for ratio of LDL cholesterol/HDL cholesterol was 37%. PAR for total cholesterol/HDL cholesterol was 32%, which was substantially lower than that of the ApoB/ApoA1 ratio (p<0·0001). These results were consistent in all ethnic groups, men and women, and for all ages.

The non-fasting ApoB/ApoA1 ratio was superior to any of the cholesterol ratios for estimation of the risk of acute myocardial infarction in all ethnic groups, in both sexes, and at all ages, and it should be introduced into worldwide clinical practice.

Canadian Institutes of Health Research, the Heart and Stroke Foundation of Ontario, the International Clinical Epidemiology Network (INCLEN). Unrestricted grants from pharmaceutical companies (with major contributions from AstraZeneca, Novartis, Aventis, Abbott, Bristol Myers Squibb, King Pharma, and Sanofi-Synthelabo), and by various national bodies.