Effect of Diet on Vascular Disease in Pre-Menopausal Women

African Americans have a higher prevalence of vascular disease than Caucasians. Vascular
disease can lead to heart attacks, strokes and even amputations. Insulin, a hormone which is
secreted by the pancreas, affects not only glucose and fat metabolism but also vascular
disease. Impairment of insulin s ability to remove glucose from the circulation is known as
insulin resistance. To overcome insulin resistance the pancreas secretes extra insulin. These
high levels of insulin affect circulating triglyceride levels by both promoting production of
triglyceride by the liver and interfering with clearance of triglyceride from the
circulation. Triglyceride in turn contributes to the development of vascular disease by
causing both inflammation and hypercoagulability.

Surprisingly African Americans are more insulin resistant and have a higher rate of vascular
disease than Caucasians but have lower triglyceride levels. Because of the high rate of
vascular diseases in African Americans, our aim is to determine if the adverse effects of
triglyceride occur at a lower level in African Americans than Caucasians. To achieve this
goal we will determine if there are differences in the effect of a meal on triglyceride
levels and vascular function in a representative cohort of African American and Caucasian
women.

For this study we will enroll 96 women (48 African American and 48 Caucasian women). We are
recruiting women because ethnic differences in triglyceride are even greater in women than
men. We are enrolling women between the ages of 18 and 65 years. The study will involve
several outpatient visits to the NIH Clinical Center. The first visit will be a screening to
determine eligibility. At the second visit a test to measure insulin resistance will be
performed. This test is called a frequently sampled intravenous glucose tolerance test. The
third visit will be for the test meal. Before and at 2, 4 and 6 hours after the meal, blood
will be drawn and vascular function measured. Vascular function is determined by taking blood
pressure and then measuring blood flow in the arm with ultrasound. It is possible that
individual differences in diet could affect the results of the vascular study on the day of
the test meal. Therefore for 7 days prior to the test meal, the NIH Clinical Center will
provide to each participant all their meals in the form of either trays or meals in a box.
These meals will be consistent with the typical American diet and be 33% fat, 15% protein and
52% carbohydrate. In designing these meals, the dietician will take into account individual
food preferences.

This study is being performed in collaboration with the Harvard School of Public Health, the
University of Texas Southwestern Medical Center and Indiana University. Therefore some blood
drawn during Visits 2 and 3 will be sent coded, without personal identifiers, to each
institution for analyses.

- INCLUSION CRITERIA:

African American Women: Individuals will be considered to be African American if they
self-identify as African American and were born in the United States. Further they must
describe both parents as being African American.

Caucasian Women: Individuals will be considered to be Caucasian if they self-identify as
Caucasian.

Healthy Volunteers: The potential enrollee must self identify as a normal volunteer and
have this confirmed by having at the screening visit a normal complete blood count,
glucose, BUN creatinine, liver and thyroid panel.

Age between 18 and 65 years: This age range is chosen because TG levels across the lifespan
from early adulthood through to postmenopausal status. Weare using 65 years of age as a
conventional upper limit. In addition, 65 years of age has been used as an age category
cut-off by National Health and Nutrition Examination Surveys. Our goal is to detect ethnic
differences in the pathways that allow vascular disease to develop. Enrolling women between
the ages of 18 and 65 years, maximizes our ability to detect differences in mechanism of
action that are truly secondary to ethnicity.

Weight less than 136 kg (300 lbs): This weight restriction is necessitated by the
limitations of the DXA scanner. The DXA platform cannot accommodate subjects who weigh more
than 136 kg.

BMI between 20 and 45 kg/m2: Women in both ethnic groups will be recruited across BMI
nonobese and obese categories. Therefore we will enroll approximately equal numbers of
women from each ethnic group in both BMI categories BMI between 20 and 30 kg/m(2) and BMI
between 30 and 45 kg/m(2).

EXCLUSION CRITERIA:

Refusal to agree to use barrier contraception: The DXA scan and CT scan should not be
performed if the participant is pregnant. In addition, TG levels are directly affected by
pregnancy. Even though we will perform pregnancy tests within 7 days of each visit, we also
require that the subject agree to abstinence or barrier contraception throughout the study.

Medications which affect parameters under investigation: Examples include corticosteroids,
oral contraceptives, hypoglycemic (oral or injection), hypolipidemic, antihypertensive and
antipsychotic agents. Oral contraceptives, in particular, are exclusion criteria because
the estrogen component of oral contraceptives increase TG levels and would therefore
obscure the effect of the test meal on TG levels.

Medical Conditions which affect parameters under investigation: Examples include diabetes,
thyroid disease, liver disease, pancreatitis, nephrotic syndrome.

Hispanics: The relationship between TG and insulin resistance in Hispanic White women is
midway between African American and Caucasian women. Therefore in this initial study to be
able to determine the maximum ethnic difference in the relationship between TG and insulin
resistance, we will enroll in this protocol study women who self-identify as African
American and Caucasian and not enroll women who self-identify as Hispanic.