Randomized Crossover Trial

A - Click here for explanation of classification scheme.

POSITIVE: See Quality Criteria Checklist below.

To compare the effects of high-fat and low-fat diets on serum triglyceride concentrations and additionally on serum concentrations of total, LDL-, and HDL-cholesterol as well to determine the effects on LDL susceptibility to oxidation and urinary isoprostanoid excretion as a measure of in vivo oxidative stress. 

Males with fasting triglyceride (TG) concentrations >2.3 mmol/L on at least 2 occasions.

BMI 30 kg/m² or less

History of symptoms or markers of type III hyperlipidemia

Apolipoprotein C-11 deficiency, diabetes mellitus, pancreatitis

Liver, kidney or thyroid dysfunction

Coronary heart disease

Use of tobacco, regular consumption of large amounts of alcohol

Recruitment

Outpatient lipid clinic of the University of Muenster 

 

Design

Randomized crossover

 

Blinding used (if applicable)

 

Intervention (if applicable)

 High-fat diet versus low-fat diet

Statistical Analysis

Number of study subjects was determined by power analysis to detect a difference in serum TG concentration of at least 0.56 mmol/L (50 mg/dL) between subjects consuming the 2 test diets with a power of 80%.  Statistical analyses were conducted using the Statistical Package for the Social Sciences (version 10; SPSS).  The distribution of means was approximateely normal for most variables, as shown by normal plots and histograms of the data and by Kolmogorov-Smirnov tests. In contrast, serum TG concentrations did not appear to be normally distributed and were logarithmically transformed.  Mean differences and 95% CI values were calculated by two-tailed t tests to assess the effect of diet order;  levels did not differ between these 2 time points.  Effects of the high-fat and low-fat diets and differences between them were assessed by two-tailed paired t tests.  Results were considered significant at values of P < 0.05.

 

Timing of Measurements

There were four consecutive dietary periods: a 2 week acclimation period, a 3 week dietary period, a 2 week washout and a second 3 week dietary period.  Three day food records were completed during each period plus a second three day food record completed during each dietary period.  Blood samples were taken at baseline, and before and after each of the diet periods.  Additional venous blood samples were taken on day 7 and 14 of each diet period.  Morning urine samples were collected at each visit.

 

Dependent Variables

• Variable 1: Bloods were drawn after an overnight fast of 10 h.  After 30 to 60 minutes, serum and plasma containing EDTA were obtained by centrifugation at 2000 X g for 15 minutes at 10⁰C.  Serum and EDTA plasma samples were immediately frozen at -80⁰C and stored until analysis.  Serum concentrations of total cholesterol and TG were measured using enzymatic assays.  The HDL-cholesterol concentration was measured by a precipitation method and the LDL-concentration was measured after precipitation of LDL with dextran sulfate.   Serum concentations of apo A-1 and apo B were measured using immunoturbidimetric assays.  All of these measuremnets were performed in series within 1 d, using a Hitachi 917 autoanalyzer.  Serum TG and pancreatic enzyme concentrations were measured immediately after each clinic visit to detect critically high TG levels. 
• Variable 2: LDL was separated from EDTA plama by density gradient centrifugation in a single 2-h session.  After centrifugation, the LDL was collected using a syringe and needle, then filtered through a sterile 0.22- µm filter into sterile vacuum containers and stored in darkness at 4⁰C.  Susceptibility to oxidation was measured on the following day.  All samples from each subject were measured in a single session, yielding intraassay CV values of 2.7, 3.8, and 4.5% for the measurements of lag time, rate of propagation, and maximum amount of conjuated diene formation, respectively.
• Morning urine samples had 0.002% butylated hydroxytoluene added and were frozen at -80⁰C until analysis.  Levels of 8-iso-prostaglandin F_2à were measured with a commercially available competitive enxyme immunoassay.  All samples from each subject were measured on the same plate.  An intraassay CV was 3.0%.  Urinary creatinine was measured by an automated kinetic procedure using a Hitachi 747 autoanalyzer.  All samples were measured within 1 day. 

 

• Independent Variables

The high-fat (39.9% energy ± 1.4) and low-fat (28.8% energy ± 1.5) diets were isocaloric (2940 kcal ± 215 vs 2892 ± 239), rich in monounsaturated fatty acids (MUFAs) (17.6 ± 0.7 vs. 12.6 ± 0.7% energy), long-chain (n-3)-polyunsaturated fatty acids (PUFAs) (1.6 ± 0.2 vs. 1.8 ± 0.2% energy), fiber (50 vs. 44 gm), and complex carbohydrates (26.2 ± 1.3 vs. 34.3 ± 2.9 % energy),  and without alcohol.  The diets consisted of conventional mixed foods which the free-living subjects ate throughout the study after receiving detailed dietary insturctions from a trained nutritionist.  The subjects and their wives also received a study booklet with information on amounts and types of food items for meal preparation and snacks.  Refined sugars, sweets, soft drinks and alcoholic drinks were not permitted.  The subjects were permitted 100 kcal each day as free-choice foods from an approved list.  The daily amounts of food were calculated for levels of energy intake ranging from 1800 to 3200 kcal/d.  To standardized the intake of long-chain (n-3)-PUFAs, the subjects were provided with canned herring and mackerel and were requested to consume 1 can/d of either during both dietary periods.  Rapeseed oil, margarine, nuts, cereals, and snacks were also supplied.  Compliance was assessed by 3-d food records as described above.  A nutritionist was in regular contact with the sujbects at clinic visits and by phone to provide support and motivations.  The subjects also recorded their daily free-choice items and any deviations from the diets in the study booklet.  The diets were calculated on the basis of standard German food tables.

Control Variables

 

 

Initial N:  17 males

Attrition (final N): 14 (1 withdrew because of conflicts with work demands and 2 withdrew becasue they were unwilling or unable to comply with the dietary regimen)

Age: 32 to 57 y (mean ± SD = 43.8 ± 7.0 y)

Ethnicity: not described

Other relevant demographics: 

Anthropometrics: baseline levels of TG, cholesterol (total, LDL, HDL) and BMI were similar.  Two subjects had a baseline TG too high to meaure HDL and apo B. 

Location: Muenster, Germany

 

 

Variables	High Fat Diet Baseline Measures and confidence intervals	High Fat Diet End	Stat Signif of Group Difference	Low Fat Diet Baseline Measures and confidence intervals	Low Fat Diet End	Stat Signif of Group Difference	Stat Signif of High-fat vs. Low-Fat
Triglycerides,  mmol/L	3.5±2.30	2.33±1.49	<0.001	3.96±2.74	2.46±0.88	0.048	0.366
Total cholesterol, mmol/L	6.07±1.31	5.40±1.34	0.006	6.51±1.26	5.39±1.27	0.004	0.972
LDL-cholesterol	3.15±1.16	3.06±1.02	0.568	3.46±1.35	3.02±0.85	0.163	0.823
HDL-cholesterol	0.98±0.28	1.01±0.28	0.439	1.01±0.27	1.01±0.37	1.000	0.970
Apo-A-1, mg/L	1306±220	1261± 221	0.009	1329±254	1272±317	0.066	0.819
Apo B, mg/L	1146±297	1109±307	0.326	1207±317	1091±253	0.088	0.746
Body Weight, kg	83.0±8.0	82.9±7.9	0.604	83.0±8.1	83.1±8.3	0.752	0.559

 

Other Findings

Subjects were split into 2 subgroups in response to the diet.  Two subjects had similar serum TG levels at the end of each diet period while 7 had lower mean TG levels at the end of the high-fat diet (1.41 vs. 2.66 mmol/L for the high-fat and low-fat diet periods, respectively).   Five subjects had lower mean TG levels at the end of the low-fat diet period (1.89 vs. 3.32 mmol/L for the low-fat and high-fat diet periods, respectively).  Subjects whose serum TG concentrations were lower at the end of the low-fat diet period had higher initial mean TG concentrations than subjects whose TG concentrations were lower at the end of the high-fat diet period (9.56 and 5.19 mmol/L, respectively; p=0.046).

Urine concentations of PGF_2à did not differ among the different study periods. 

The low-fat diet lowered the susceptibility of plasma LDL to oxidation in subjects compared to the high-fat diet.  The rate of propagation mean difference = -2.07 nmol conjugated dienes/(min ^. mol LDL-cholesterol), 95% CI = -.98 to -3.34 nmol conjugated dienes/(min ^. mol LDL-cholesterol).  The lower maximal formation of conjugated dienes mean difference = -106 nmol conjugated dienes/mol LDL-cholesterol, 95% CI = -21 to -191 nmol conjugated dienes/mol LDL-cholesterol.  The lag time until the start of oxidation tended to be longer (p=0.06) after consumption of the low-fat diet (mean difference = +2.3 min, 95% CI = -0.11 to +4.69 min).

 

A diet rich in unsaturated fatty acids [particularly long-chain (n-3)-PUFAs], fiber, and complex carbohydrates and low in saturated fatty acids, simple carbohydrates, and alcohol can be used effectively to treat hypertriglyceridemias.  Individuals with serum TG concentrations <4.5 mmol/L benefit more from such a diet if its fat content is relatively high.  A considerable number of individuals with serum TG concentration >4.5 mmol/L achieve lower serum TG levels by consuming a low-fat diet. 

University/Hospital:

Univesity of Miami School of Medicine, Miami VA Medical Center