Study of Macronutrients and Heart Disease Risk

Not Applicable

Completed

Conditions: Body Composition
Blood Pressure
Obesity
Cardiovascular Diseases

Interventions: Behavioral: low carbohydrate diet
Behavioral: low fat diet

Registration Number: NCT00609271

Lead Sponsor: Tulane University Health Sciences Center

Brief Summary: The objective of this trial is to examine the long-term effects of a diet low in carbohydrates, as compared to one low in fat, on cardiovascular disease risk factors, including blood pressure (BP), body weight and composition, serum lipids, plasma glucose, insulin, adipocytokines (adiponectin, leptin, resistin), and C-reactive protein (CRP) among obese adults.

The investigators will test the following hypotheses:

Hypothesis 1: Compared to a low fat diet, a diet low in carbohydrates will reduce systolic and diastolic BP over 12 months; Hypothesis 2: Compared to a low fat diet, a diet low in carbohydrates will reduce body weight, total percent body fat, and waist circumference over 12 months; Hypothesis 3: Compared to a low fat diet, a diet low in carbohydrates will reduce serum levels of LDL-cholesterol and triglycerides and increase serum levels of HDL-cholesterol over 12 months; Hypothesis 4: Compared to a low fat diet, a diet low in carbohydrates will reduce plasma levels of glucose and insulin levels over 12 months; and Hypothesis 5: Compared to a low fat diet, a diet low in carbohydrates will reduce plasma levels of leptin, resistin, and CRP and increase plasma levels of adiponectin over 12 months.

Detailed Description: Cardiovascular diseases (CVD) remain the leading cause of death globally as well as here in the United States. Manipulations of the macronutrient (protein, carbohydrate and fat) contents of diet have been used extensively for weight loss and weight control in the past several decades. Low carbohydrate diets, in particular, have gained popularity for weight loss. However, few studies have examined the effects of a diet low in carbohydrates on traditional and novel cardiovascular risk factors in the long term, particularly in contrast to the current dietary recommendations for decreased fat intake to reduce risk of CVD. In this proposal, we plan to conduct a 12-month, parallel-arm, randomized controlled trial of a diet low in carbohydrates versus the currently recommended low fat diet to reduce CVD risk factors among obese adults. The objective of this trial is to examine the long-term effects of a diet low in carbohydrates, as compared to one low in fat, on CVD risk factors, including blood pressure (BP), body weight and composition, serum lipids, plasma glucose, insulin, adipocytokines (adiponectin, leptin, resistin), and C-reactive protein (CRP) among obese adults. In order to accomplish these objectives we will randomize 130 eligible participants (n=65 in each group) to consume either a diet low in carbohydrates (≤40 g/d) or a diet low in fat (\<7% saturated fat, \<30% total fat). Neither of the diets will be energy-restricted. Participants will meet with a dietitian for one-on-one counseling sessions weekly for the first 4 weeks, then bi-monthly in small group sessions for the next 5 months, and monthly in larger group sessions for the final 6 months of the intervention. Data on both traditional and novel CVD risk factors will be collected at baseline, 3, 6, and 12 months. We hypothesize that a diet low in carbohydrates as compared to a diet low in fat will lower systolic and diastolic BP, body weight, total percent body fat, waist circumference, serum levels of triglycerides, and plasma levels of insulin, glucose, leptin, resistin, and CRP, and increase serum levels of HDL-cholesterol and adiponectin. Because CVD is the most common cause of death here in the U.S. and world-wide, this study has important public health implications. It will provide new information on the potential long-term effects of diets low in carbohydrates on both the traditional risk factors for CVD as well as novel risk factors and inflammatory factors. The results from this study will help to determine if a diet low in carbohydrates as compared to the currently recommended low fat diet can decrease the risk of CVD among obese adults.

Recruitment & Eligibility

Status: COMPLETED

Sex: All

Target Recruitment: 148

Inclusion Criteria

Men or women aged 22 - 75 years, any race/ethnicity
BMI of 30 - 45 k/m2
Willing and able to provide informed consent

Exclusion Criteria

History of self-reported clinical CVD (angina/myocardial infarction, coronary revascularization, heart failure, stroke/transient ischemic attack, peripheral arterial disease)
Medical condition in which a low-carbohydrate diet may not be advised (diabetes, renal disease, cancer requiring treatment during the past year, osteoporosis, untreated thyroid disease, gout)
Current use of more than 2 antihypertensive or more than 2 cholesterol-lowering medications
For women, current pregnancy or breastfeeding or plans to become pregnant during the study period
Consumption of more than 21 alcoholic beverages per week
Currently on a diet or using prescription weight loss medications, underwent weight loss surgery, and/or experienced weight loss >15 pounds within 6 months of study entry
Plans to move out of the study area (>1 hour from study site) or difficulty to come to the study site
Participation of another household member in the study; employees or persons living with employees of the study
Participation in other lifestyle intervention trials currently
At the discretion of the study coordinator

Study & Design

Study Type: INTERVENTIONAL

Study Design: PARALLEL

Arm && Interventions

Group	Intervention	Description
1	low carbohydrate diet	low carbohydrate diet
2	low fat diet	low fat diet

Primary Outcome Measures

Name	Time	Method
Predicted Mean Difference in Body Weight From Baseline, by Assigned Dietary Group	12 months	Predicted mean difference from random-effects models that included diet, time, and diet-by-time interaction term. Markov-chain Monte Carlo techniques were used to impute missing values.
Predicted Mean Differences in Fat Mass From Baseline, by Assigned Dietary Group	12 months	Mean Difference in Fat Mass predicted from random-effects models that included diet, time, and diet-by-time interaction term. Markov-chain Monte Carlo techniques were used to impute missing values.
Predicted Mean Differences in HDL Cholesterol From Baseline, by Assigned Dietary Group	12 months	Predicted from random-effects models that included diet, time, and diet-by-time interaction term. Markov-chain Monte Carlo techniques were used to impute missing values.
Predicted Mean Differences in Serum Insulin Level From Baseline, by Assigned Dietary Group	12 months	Mean Difference in Serum Insulin Level predicted from random-effects models that included diet, time, and diet-by-time interaction term. Markov-chain Monte Carlo techniques were used to impute missing values.
Predicted Mean Differences in Total-HDL Cholesterol Ratio From Baseline, by Assigned Dietary Group	12 months	Predicted from random-effects models that included diet, time, and diet-by-time interaction term. Markov-chain Monte Carlo techniques were used to impute missing values.
Predicted Mean Differences in Systolic Blood Pressure From Baseline, by Assigned Dietary Group	12 months	Predicted from random-effects models that included diet, time, and diet-by-time interaction term. Markov-chain Monte Carlo techniques were used to impute missing values.
Predicted Mean Difference in Diastolic Blood Pressure, by Assigned Dietary Group	12 Months	Mean Difference in Diastolic Blood Pressure predicted from random-effects models that included diet, time, and diet-by-time interaction term. Markov-chain Monte Carlo techniques were used to impute missing values
Predicted Mean Differences of Waist Circumference From Baseline, by Assigned Dietary Group	12 months	Predicted from random-effects models that included diet, time, and diet-by-time interaction term. Markov-chain Monte Carlo techniques were used to impute missing values.
Predicted Mean Difference in Plasma Glucose Level, by Assigned Dietary Group	12 months	Mean Difference in Plasma Glucose Level predicted from random-effects models that included diet, time, and diet-by-time interaction term. Markov-chain Monte Carlo techniques were used to impute missing values.
Predicted Mean Differences of 10-y Framingham Risk Score From Baseline, by Assigned Dietary Group	12 Months	Mean Difference in 10-y Framingham Risk Score predicted from random-effects models that included diet, time, and diet-by-time interaction term. Markov-chain Monte Carlo techniques were used to impute missing values.
Predicted Mean Differences in Lean Mass From Baseline, by Assigned Dietary Group	12 months	Mean Difference in Lean Mass predicted from random-effects models that included diet, time, and diet-by-time interaction term. Markov-chain Monte Carlo techniques were used to impute missing values.
Predicted Mean Differences in Total Cholesterol Level From Baseline by Assigned Dietary Group	12 months	Predicted from random-effects models that included diet, time, and diet-by-time interaction term. Markov-chain Monte Carlo techniques were used to impute missing values.
Predicted Mean Differences in LDL Cholesterol Level From Baseline, by Assigned Dietary Group	12 months	Predicted from random-effects models that included diet, time, and diet-by-time interaction term. Markov-chain Monte Carlo techniques were used to impute missing values.
Predicted Mean Differences in C-reactive Protein Level From Baseline, by Assigned Dietary Group	12 Months	Mean Difference in C-reactive Protein Level predicted from random-effects models that included diet, time, and diet-by-time interaction term. Markov-chain Monte Carlo techniques were used to impute missing values
Predicted Mean Differences in Triglycerides From Baseline, by Assigned Dietary Group	12 months	Predicted from random-effects models that included diet, time, and diet-by-time interaction term. Markov-chain Monte Carlo techniques were used to impute missing values.
Predicted Mean Differences in Serum Creatinine Level From Baseline, by Assigned Dietary Group	12 Months	Mean Difference in Serum Creatinine Level predicted from random-effects models that included diet, time, and diet-by-time interaction term. Markov-chain Monte Carlo techniques were used to impute missing values.