BA8 - Predictive value of nutrient biomarkers for CHD death

Investigator Names and Contact Information

Alice Lichtenstein (Alice.Lichtenstein@Tufts.edu)

Introduction/Intent

Dr. Lichtenstein and colleagues will determine the value of selected nutrient biomarker plasma concentrations (trans fatty acids, very long chain omega-3 fatty acids, phylloquinone, dihydrophylloquinone) in women who participated in the WHI Observational Study and died of CHD. The predictive value of these data relative to self-reported food intake will also be assessed. The results of this work will be of value in defining optimal approaches to evaluating the impact of dietary patterns on CHD risk in large groups of individuals.

The overall objective of this research proposal is to determine whether the predictive value of using plasma concentrations of selected nutrient biomarkers of food intake determined using a single plasma sample either alone or in combination are stronger, objective predictors of subsequent death from coronary heart disease (CHD) or myocardial infarct (MI) compared to selected food intake data derived from subjective, self-reported food frequency questionnaires. The nutrient biomarkers (phospholipid [PL] eicosapentaenoic acid [EPA], PL docosahexaenoic acid [DHA], PL trans fatty acids, phylloquinone, dihydrophylloquinone) and foods (fish, dark fish and tuna, vegetables, fruits and whole grains, and unsaturated fat rich foods) targeted have previously been either directly or indirectly associated with CVD risk. We propose to test our overall objective by conducting a nested case-control study using plasma samples and food frequency data from the observational cohort of the WHI. Our cases will be selected from the subset of women who did not report dietary supplement use and who died of CHD or MI (collectively referred to as WHI CHD cases). The control subjects will be selected from the subset that were free of CHD or MI events and matched with cases for standard National Cholesterol Education Program (NCEP) risk factors (WHI controls). Nutrient biomarker data will be newly generated by the HNRCA team of investigators using stored specimens as part of this proposal whereas the selected food intake data have previously been collected by the WHI investigators. The plasma samples and food frequency data were both collected at the start of the observational period.

Hypothesis 1: Nutrient biomarkers previously established to be surrogate indicators of dietary patterns associated with CVD risk are significantly different between WHI CHD cases and WHI controls.

Aims:

1a. To measure baseline (at entry into the WHI study) concentrations of the plasma nutrient biomarkers; PL EPA, PL DHA, PL trans fatty acids, phylloquinone and dihydrophylloquinone, in WHI CHD cases and WHI controls.

1b. To determine the difference in the concentrations of the plasma nutrient biomarkers; PL EPA, PL DHA, PL trans fatty acids, phylloquinone and dihydrophylloquinone, between WHI CHD cases versus WHI controls.

1c. To develop an algorithm containing two or more nutrient biomarkers of CVD risk (PL EPA, PL DHA, PL trans fatty acids, phylloquinone and dihydrophylloquinone) that will be a stronger predictor of WHI CHD cases compared to any given individual biomarker.

Hypothesis 2: Self-reported intakes of foods associated with CVD risk; fish, dark fish and tuna, vegetables, fruits and whole grains, and unsaturated fat rich foods, determined using a food frequency questionnaire are significantly different between WHI CHD cases and WHI controls.

Aims:

2a. To determine whether intakes of fish, dark fish and tuna, vegetables, fruits and whole grains, and unsaturated fat rich foods determined using a food frequency questionnaire are different between WHI CHD cases and WHI controls.

2b. To develop an algorithm containing two or more foods associated with CVD risk determined using a food frequency questionnaire (fish, dark fish and tuna, vegetable or fruit, whole grains, and unsaturated fat rich foods) that will be a stronger predictor of WHI CHD cases compared to any given individual food.

Hypothesis 3: The predictive value of nutrient biomarkers of CVD risk will be greater than the predictive value of self-reported intakes of foods associated with CVD risk between WHI CHD cases and WHI controls.

Aims:

3a. To compare the predictive value of nutrient biomarkers of CVD risk (PL EPA, PL DHA, PL trans fatty acids, phylloquinone and dihydrophylloquinone), alone and in combination, to the predictive value of self-reported intake of foods associated with CVD risk (fish, dark fish and tuna, vegetable or fruit, whole grains, and unsaturated fat rich foods), alone and in combination, in WHI CHD cases and WHI controls.

3b. To determine whether the predictive value in WHI CHD cases and WHI controls is greatest when nutrient biomarker(s) and self-reported intake of food(s) associated with CVD risk are considered independently or together.

Results/Findings

Some of the publications related to this ancillary study are: 1151.

For a complete, up-to-date list of WHI papers related to this ancillary study, please use the searchable Papers section of this website.