Cardiovascular risk determination: discrepancy between total cholesterol evaluation and two compound laboratory indices in Norway.

OBJECTIVE To compare group classification of cardiovascular risk by two compound laboratory indices with classification according to the serum total cholesterol concentration alone. DESIGN Healthy employees were defined as low and high cardiovascular risk subjects according to their total cholesterol concentration or two compound indices of blood lipid components-the total cholesterol: high density lipoprotein (HDL) cholesterol ratio and an atherogenic index defined as ([total cholesterol-HDL cholesterol]*[apolipoprotein B])/([HDL cholesterol]*[apolipoprotein A-I]). Cut off values to distinguish between low and high risk subjects were as follows: total cholesterol 6.5 mmol/l, HDL cholesterol 0.9 mmol/l, apolipoprotein A = 1.8 g/l, and apolipoprotein B = 1.3 g/l. These gave total: HDL cholesterol ratio and atherogenic index cut off values of 7.2 and 4.5 respectively. SETTING An occupational health service in a non-manufacturing company in Norway. PARTICIPANTS A total of 112 male and 117 female employees. The mean body mass index values were 25.6 and 23.6 kg/m2 and the mean ages 39.8 and 40.1 years in men and women respectively. Those with cardiovascular, diabetic, or renal diseases were excluded. MEAN OUTCOME MEASURES Serum total cholesterol, HDL cholesterol, apolipoproteins A-I and B, lipid peroxidation, blood pressure, smoking, physical activity, and fruit, vegetables, and salt in the diet were determined. RESULTS The cut off values allocated 19%, 7%, and 40% as high risk subjects according to total cholesterol, total: HDL cholesterol, and the atherogenic index respectively. The mean age was two to four years higher in the high risk groups. Cardiovascular risk in siblings and no reported physical activity were more prevalent in those high risk groups defined by the compound indices than by total cholesterol alone, as was a high body mass index and a measure of lipid peroxidation. Grouping according to total cholesterol failed to allocate heavy smokers mainly to the high risk group. Diet variables did not demarcate clearly between indices. CONCLUSIONS There is considerable variability in classification into high and low risk subjects when using the total cholesterol concentration alone compared with compound risk indices. Smoking was more prevalent in the high risk groups defined by the compound indices than by total cholesterol. These findings call for caution when total cholesterol is used to estimate cardiovascular risk in epidemiological studies, and even more so at individual counselling in occupational or primary health care settings.