Objective: To assess the effects of smoking on serum lipoprotein levels taking into account current nutritional intakes.
Design: Case-control study comparing smokers and non-smokers.
Setting: Ambulatory subjects of the Urban Community of Lille examined at home or at the Health Care Center.
Subjects: Men, between 45 and 65 years of age; n = 89 smokers (cases) and n = 91 non-smokers (controls); non-smokers were randomly selected from the voter's registration lists, 7 smokers and 2 non-smokers were excluded for hyperlipidemia.
Interventions: Blood sampling, medical examination and three-day food records validated by a registered dietician.
Results: Smokers were younger (P < 0.001) and thinner (P = 0.003), chose more frequently visible fats of animal origin (P = 0.03) and reported significantly larger daily meat servings (P = 0.01) than non-smokers. Smokers had a significantly higher intake of non-alcoholic energy (P < 0.05), lipids of animal origin (P = 0.003), saturated fat (P < 0.01) and monounsaturated fat (P < 0.01) and a lower P/S ratio (P = 0.02) than non-smokers. After adjustment on age and body mass index, smokers had significantly higher levels of mean serum triglycerides (P = 0.03), VLDL-cholesterol (P = 0.0003) and systolic blood pressure (P = 0.02) and lower values of HDL-cholesterol (P = 0.02) than non-smokers. Additional adjustment for alcohol consumption showed significantly lower apo A-I values (P = 0.03) in smokers than non-smokers. Further adjustment for the intake of lipids of animal origin revealed higher apo B values (P = 0.04) in smokers than non-smokers.
Conclusions: (1) Smoking is associated with serum lipid and lipoprotein alterations independently of nutritional factors, (2) Nutritional habits of smokers are less 'healthy' than those of non-smokers (3) Smokers is a group of particular importance in terms of multifactorial public health intervention.