Tomoko Shimakawa

Dietary Antioxidants and Carotid Artery Wall Thickness The ARIC Study

BACKGROUND Evidence that dietary antioxidants may prevent atherosclerotic disease is growing. The relationship between the intake of dietary and supplemental vitamin C, alpha-tocopherol, and provitamin A carotenoids and average carotid artery wall thickness was studied in 6318 female and 4989 male participants 45 to 64 years old int he Atherosclerosis Risk in Communities Study. METHODS AND RESULTS Intake was assessed by use of a 66-item semiquantitative food-frequency questionnaire. Carotid artery intima-media wall thickness was measured as an indicator of atherosclerosis at multiple sites with B-mode ultrasound. Among men and women > 55 years old who had not recently begun a special diet, there was a significant inverse relationship between vitamin C intake and average artery wall thickness adjusted for age, body mass index, fasting serum glucose, systolic and diastolic blood pressures, HDL and LDL cholesterol, total caloric intake, cigarette use, race, and education (test for linear trend across quintiles of intake, P = .019 for women and P = .035 for men). An inverse relationship was also seen between wall thickness and alpha-tocopherol intake but was significant only in women (test for linear trend, P = .033 for women and P = .13 for men). There was a significant inverse association between carotene intake and wall thickness in older men (test for linear trend, P = .015), but the association weakened after adjustment for potential confounders. No significant relationships were seen in participants < 55 years old. CONCLUSIONS These data provide limited support for the hypothesis that dietary vitamin C and alpha-tocopherol may protect against atherosclerotic disease, especially in individuals > 55 years old.

Vitamin intake: A possible determinant of plasma homocyst(e)ine among middle-aged adults

PURPOSE Many epidemiologic studies have identified elevated plasma homocyst(e)ine as a risk factor for atherosclerosis and thromboembolic disease. To examined the relationship between vitamin intakes and plasma homocyst(e)ine, we analyzed dietary intake data from a case-control study of 322 middle-aged individuals with atherosclerosis in the carotid artery and 318 control subjects without evidence of this disease. METHODS All of these individuals were selected from a probability sample of 15,800 men and women who participated in the Atherosclerosis Risk in Communities (ARIC) Study. RESULTS Plasma homocyst(e)ine was inversely associated with intakes of folate, vitamin B6, and vitamin B12 (controls only for this vitamin)--the three key vitamins in homocyst(e)ine metabolism. Among nonusers of vitamin supplement products, on average each tertile increase in intake of these vitamins was associated with 0.4 to 0.7 mumol/L decrease in plasma homocyst(e)ine. An inverse association of plasma homocyst(e)ine was also found with thiamin, riboflavin, calcium, phosphorus, and iron. Methionine and protein intake did not show any significant association with plasma homocyst(e)ine. CONCLUSIONS In almost all analyses, cases and controls showed similar associations between dietary variables and plasma homocyst(e)ine. Plasma homocyst(e)ine among users of vitamin supplement products was 1.5 mumol/L lower than that among nonusers. Further studies to examine possible causal relationships among vitamin intake, plasma homocyst(e)ine, and cardiovascular disease are needed.

Relation of Hormone-Replacement Therapy to Measures of Plasma Fibrinolytic Activity

BACKGROUND The mechanisms by which replacement hormones may reduce the risk of coronary heart disease are not fully understood. Of specific interest is a potential effect of replacement hormones on plasma fibrinolytic activity, a putative determinant of thrombotic events. METHODS AND RESULTS We investigated the relation of current use of replacement hormones to three measures of plasma fibrinolytic activity: tissue-type plasminogen activator (TPA) antigen, plasminogen activator inhibitor-1 (PAI-1) antigen, and D-dimer. The sample was composed of 288 women, free of clinical cardiovascular disease, who were selected for a case-control study of atherosclerosis: 142 women with ultrasonographic evidence of carotid intimal-medial thickening (cases) and 146 control subjects. Twenty percent (59 women) reported current use of replacement hormones. TPA antigen and PAI-1 antigen were highly correlated with each other (r = .67), whereas D-dimer correlated only weakly with TPA or PAI-1. Compared with nonusers, current users of replacement hormones had lower mean levels of TPA and PAI-1 antigens, suggesting enhanced fibrinolytic potential. In the entire sample, the multivariate-adjusted geometric mean values of TPA antigen were 6.3 and 7.3 ng/mL among current users and nonusers, respectively (P = .01); the corresponding values for PAI-1 antigen were 6.1 and 7.5 ng/mL (P = .13). These results were generally consistent for both atherosclerosis cases and their control subjects. D-dimer levels were lower in current hormone users than in nonusers, but the difference was not statistically significant (P > .15) in any of the analyses. CONCLUSIONS The use of replacement hormones appears to be associated with enhancement of endogenous fibrinolytic potential. Enhanced plasma fibrinolytic activity among hormone users may explain, in part, the inverse association between hormone replacement therapy and coronary heart disease.

Reliability of a food frequency questionnaire by ethnicity, gender, age and education

Abstract The reproducibility of nutrient intakes measured by a semi-quantitative food frequency questionnaire were examined in 418 African American and white men and women between the ages of 45 and 64 years. Subjects were a subset of participants in the Atherosclerosis Risk in Communities (ARIC) study. Reliability coefficients were calculated for nutrient intakes reported over a mean elapsed period of three years. Reliability coefficients tended to be higher in men than in women and higher in white Americans than in African Americans. The median reliability coefficients were 0.63 in white men, 0.48 in white women. 0.50 black men, and 0.45 in black women. Reliability coefficients tended to be higher in subjects with more than 12 years of education compared to those with less education. Also, age was associated with the size of the reliability coefficients, and participants who were 45–49 years of age tended have higher coefficients than older participants.

COMPARISON OF DIETS OF NIDDM AND NON-DIABETIC AFRICAN AMERICANS AND WHITES: THE ATHEROSCLEROSIS RISK IN COMMUNITIES STUDY

Abstract To compare dietary intakes of participants with non-insulin-dependent diabetes (NIDDM) with nondiabetic participants, to determine whether those with previously diagnosed NIDDM were complying with the 1986 American Diabetes Association (ADA) guidelines and to examine associations between diet and new cases of NIDDM. Cross-sectional data were obtained by trained interviewers from participants aged 45–64 years, chosen by random probability sampling of four U.S. communities. Dietary intakes were assessed by food frequency questionnaire in 475 African Americans and 674 whites with NIDDM and 3,255 African American and 10,439 white non-diabetic participants. Participants with diagnosed NIDDM consumed less calories, carbohydrates, sucrose, alcohol, candy, added sugar and soft drinks, but more fiber, protein, cholesterol, low calorie drinks, fruits, vegetables, cheese, poultry, red meat, and eggs compared to the ethnic-specific non-diabetic group. African Americans diagnosed with NIDDM consumed more total and monounsaturated fats (both as %) and cereal, and less fried foods and carbohydrate compared with non-diabetic African Americans; and whites with diagnosed NIDDM consumed more milk, yogurt or ice cream; margarine; fish; bread and peanut butter and less sugared foods compared with non-nutrient intakes of participants with new NIDDM were not different from non-diabetic individuals 1986 ADA guidelines. Participants with diagnosed NIDDM were eating a low calorie, high protein, high cholesterol, and low carbohydrate diet. Dietary intakes of new cases of NIDDM were not implicated in its etiology.