- Click here for explanation of classification scheme.

To compare the efficacy of two palatable dietary regimens on the the treatment of hypertriglyceridemia which can be practised by patients on a daily basis.  The secondary objective was to investigate the effects of these diets on subclasses of triglyceride-rich lipoproteins and HDL, on LDL size, on plasma activities of lipolytic enzymes and lipid transfer proteins and on insulin and obesity markers which regulate the metabolism of triglyceride-rich lipoproteins and HDL. 

Fasting serum triglyceride (TG) concentrations >2.3 mmol/l on at least two occasions

Obesity

History with symptoms or markers indicative of type III hyperlipidemia, apoC-II deficiency, diabetes mellitus, liver, kidney or thyroid dysfunction, pancreatitis, or coronary heart disease. 

Recruitment Outpatient lipid clinic of the University of Munster

 

Design Nonrandomised Trial

 

Blinding used (if applicable) None

 

Intervention (if applicable) High-fat diet, rich in monounsaturated fatty acids (MUFA) and n-3 fatty acids followed by a low-fat diet.

 

Statistical Analysis

Statisical analyses were perfomed using SPSS-X.  Values are expressed as means ± standard deviations.  Since most parameters were not normally distributed, the level of significance for changes in lipid parameters between the different dietary periods and their correlations were calculated by the non parametric paired Wilcoxon U-test and Spearmon analysis, respectively.  In order to adjust changes in lipid concentrations for baseline lipid concentrations, changes in body mass index (BMI) and alcohol intake, multiple regreesion analysis was conducted to take into account these variables.  All  P values are two-tailed.  P values <0.05 were considered significant. 

Timing of Measurements

 Fasting bloods were drawn at screening, at the end of the baseline period and at days 4, 7, 14, and 21 of each period.  Three day food records were completed athe end of the baseline period and in the first week of each diet period, two 3 day food diaries were completed.

Dependent Variables

• Variable 1: Blood without additives was taken for the quantification of lipids, lipoproteins, apolipoproteins and hormones.  After 30-60 min coagulation serum was obtained by centrifugation at 1800 g for 10 min at room temperature.  EDTA blood was taken for the measurement of lipid transfer proteins, placed on ice and cetrifuged at 1800 g for 10 min at 10²C.  For the determination of lipolytic enzymes, EDTA blood was collected 15 min after intravenous injection of 70 IU heparin per kilogram body weight.  EDTA blood, EDTA plasmas and sera were stored frozen at -70^oC until analysis
• Variable 2: Concentrations of cholesterol, TG and glucose were analyzed by the use of an Hitachi 917 autoanalyser and with enzymatic tests. 
• HDL was determined after precipitation of apo B containing lipoproteins with phosphotungstic acid and magnesium chloride. 
• VLDL was measured in the supernatant which was obtained by ultracentrifugation of 1 ml serum for 2h at 435.680 g at 4^oC.  The recovery was controlled by the measurement of cholesterol in the infranatant and by comparison of the sum of cholesterol in the supernatant and infranatant with total cholesterol.
• LDL was calculated as the difference of total cholesterol minus VLDL minus HDL. 
• Serum concentrations of apolipoproteins (apo) A-1, A-II, and B were determined by the use of immunoturbidimetric tests.   Apo C-III in serum and HDL as well as HDL-subclass lipoprotein A-1 (LpA-1) were measured by electroimmunodiffussion assays.  ApoE in serum and HDL was determined by enzyme immunoassay.  LpA-I, A-II was calculated from the difference between apoA-1 and LpA-I; the apoC-III and apoE contents of apo B-containing lipoproteins were calculated from the differences between total apoC-III and apoC-III in HDL and total apoE and apoE in HDL, respectively.  All assays for the determination of lipoprotein subclasses were performed in series which included complete sets of 3 samples of every volunteer.  The apo E polymorphism and the presence of normal apoC-II isoforms were analyzed by isoelectric focusing to delipidated serum and subsequent anti-apo E- and anti-apoCII-immunoblotting.
• The size of LDL was determined by the use of a commercially available polyacrylamide gradient gel electrophoresis kit where lipoproteins are visualized by lipid staining.  LDL size in samples was estimated by log-linear regression analysis of R_f values which were calculated as the ratio of the electrophoretic migration distance of albumin to the electrophoretic migration distance of LDL.  Complete sets of 3 samples per individual were analyzed within one gel. 
• Lipoprotein (LPL) and hepatic lipase (HL) activities were determined as the activity of postheparin EDTA-blood to hydrolyse radioactive triolein.  Lecithin:cholesterol acyltransferase (LCAT) was determined as the activity of plasma to esterify radio-labeled cholesterol in exogenous reconstituted HDL.  Cholesteryl ester transfer protein (CETP) activity was determined as the transfer of radiolabeled cholesteryl oleate from reconstituted HDL to LDL.  Phospholipid transfer protein (PLTP) activity was monitored as the transfer of radiolabled phosphatidylcholine from vesicles to HDL₃.  All adiometric assays for the determination of lipolytic or lipid transfer activities were performed as duplicates and in series which included complete sets of 3 samples of every patient.  The intraassay and interassay coeffieients of variation for all assays were below 8% and 15%, respectively. 

Independent Variables

The subjects continued their habitual diet for at least two weeks.  Those who were previously treated with lipid-lowering drugs took 4 weeks to allow for wash-out of these drugs.  At the end of this time the subjects completed a 3-day food diary which was used to analyze their habitual intake and to estimate each person's energy intake.  They then ingested a high-fat diet, rich in MUFA and n-3 fatty acids for 3 weeks followed by another 3 weeks ingesting a low-fat diet.  The prescribed daily amounts of foods were calculated for levels of energy intake ranging from 1800-3200 kcal.  They were designed to differ only in the amount of toal fat, MUFA, n-3 fatty acids and carbohydrates.  With the exception of fatty fish, the same foodstuffs were generally allowed in both dietary periods, differing only in their daily amounts.  To standardize the intake of n-3 fatty acids in the high-fat diet, the patients were provided with canned herring and mackerel and were required to consume one can of either herring or mackerel per day, which provided 3.0 or 4.6 g of marine n-3 fatty acids (sum of eicosapentaenoic and docosahexaenoic acid).  In both dietary periods, the patients were supplied with rapeseed oil as the main source of MUFA and alpha-linolenic acid (1.3% of energy = 3.5 g), with a MUFA-rich commerical margarine, and with a mixed whole-grain cereal flakes.  The main components of both diets were vegetable foods, primarily whole-grain bread and cereals, vegetables, crudites and fruit.  They were instructed to consume only these products in the prescribed amounts and not to use any other dietary fat, oil or muesli.  Alcoholic drinks, sugar, sweets and soft drinks were not permitted throughout the entire study.  As sweeteners only aspartame, cyclamate or saccharine were allowed.  To facilitate adherence to the study diets, the patients were permitted to consume 100kcal per day in the form of "free-choice items', which could be chosen from a given list. 

At the beginning of each dietary period the patients and their wives received extensive instructions about the respective study diet by a trained nutritionist.  They also received detailed insturction booklets indicating in particular the foods to be consumed daily with their respective amounts, non-allowed foods, and the list of free-choice items.  Instructions for preparation were also given.   During the whole study the patients recorded their daily free-choice items and any deviations from their diets in diaries.  A nutritionist was in regular contact with the patients both at their visits to the laboratory and by phone to provide support and motivation. 

 

Control Variables

 

 

Initial N: 19 males

Attrition (final N): 19

Age: 39.2 ± 8.9 years

Ethnicity:

Other relevant demographics:

Anthropometrics All were non-smokers with BMI 25.45±1.39, TG 3.65 mmol/l (range 2.40-16.82), Total cholesterol 6.84±1.66, HDL 0.94±0.23 and HbAlc 4.46±0.37% 

Location: University of Munster, Germany

 

 

Variables	Baseline Measures and confidence intervals	High-fat Diet	Low-fat diet Measures and confidence intervals
BMI	25.55 ± 1.31	25.11 ± 1.37**	24.86 ± 1.45t
TG (mmol/l)	4.04 (2.27-9.94 range)	1.49 (0.62-3.87 range)**	2.43 (0.77-7.89 range)+
Total cholesterol(mmol/l)	6.68 ± 1.45	5.20 ± 1.00**	5.40 ± 0.94
VLDL  (mmol/l)	1.68 (0.67-3.10)	0.78 (0.34-1.34)**	0.93 (0.49-3.08)tt
LDL (mmol/l)	4.15 ± 1.01	3.47 ± 0.91*	3.39 ± 0.71
HDL (mmol/l)	0.93 ± 0.24	1.00 ± 0.22*	0.88 ± 0.18+
Cholesterol/HDL ratio	7.44 ± 1.89	5.45 ± 1.60**	6.40 ± 1.87+
LDL size (nm)	24.18 ±0.95	25.63 ± 1.38**	25.43 ±1.39
Apo A-I (mg/l)	1246±314	1193±214	1169±226
Lp A-I (mg/l)	410±76	344±62*	356±58
Lp A-I, A-II (mg/l)	836±286	849±187	814±195
Apo A-II (mg/l)	355±93	313±64*	315±72
Apo B (mg/l)	1126±311	1061±255	1008±226
Total apo C-III (mg/l)	48 (23-116)	35 (19-75)*	43 (23-86)t
Apo C-III non Lp B	34 (12-73)	24 (13-46)*	28 (14-61)t
Apo C-III Lp B	13 (3-43)	9 (3-30)*	13 (4-27)t
Total apo E (mg/l)	46 (6-101)	31 (8-57)#	40 (8-56)+
HDL-E	24 (4-12)	22 (5-42)	20 (6-48)
Non HDL-E	12 (2-36)	4 (1-36)#	10 (2-35)+

 ( ) range

^#Significantly different from baseline P<0.05

*Significantly different from baseline p<0.01

**Significantly different from baseline p<0.001

^tSignificantly different from high=fat diet p<0.05, ⁺<0.01, ^tt <0.001

Other Findings

The high-fat diet was higher in total fat than the habitual diet.  During the habitual diet, the intake of alcohol, SFA and dietary cholesterol was higher, while the intake of MUFA, n-3 and n-6 PUFA and idetary fiber was lower than during the high-fat diet.  No differences existed between the high-fat and the low-fat diets in intake of alcohol, dietary cholesterol and fiber. 

During the high-fat diet, a TG level of <2.3 mmol/l was achieved in 16 of the 19 subjects, whereas on the subsequent low-ffat diet 12 subjects again showed values above 2.3 mm/l.  The main reduction of TG on the high-fat diet occurred during the first week.  In the second period, TG rose rapidly. 

On the high-fat diet, lipolytic activities of LPL and HL in postheparin blood increased significantly by 11% (p<0.001) and 13% (p<0.05), respectively.  The changes in LPL activity but not the changes in HL activity had significant correlations with the changes in TG (r=-0.7-18, p<0.001).  The activity of LCAT increased by 2% (p<0.01) and the actiivity of PLTP decreased by 4% (p<0.05).  The changes in LCAT activity strongly correlated with the changes in HDL cholesterol (r=0.7822, p<0.001).  The low-fat diet was not associated with any significant changes in enzyme activities except a decrease in LCAT activity by 3% (p<0.001), which again correlated with the changes in HDL cholesterol (r=0.7921, p<0.001). 

On the high-fat diet, serum levels of leptin and insulin decreased significantly by 26% (p<0.001) and 36% (p<0.01), respectively.  The changes in insulin correlated with the changes in TG (r=0.4855, p<0.05) and the changes in leptin with changes in LPL activity (r=0.5684, p<0.05) and PLTP activity (r=0.5359, p<0.05).  There was no change in leptin on the low-fat diet and the increase in insulin was not statistically significant. 

 

In nonobese men with hypertriglyceridemia, a modified diet low SFA, but enriched in MUFA and marine n-3 fatty acids is more effective in correcting an atherogenic lipoprotein profile than a carbohydrate-rich low fat diet. 

Industry:	Walter RAU Lebensmittel-werke GmbH & Co KG (Hilter), NORDA, Union Deutsche Lebensmittelwerke GmbH (Cuxhaven) Food Company:
University/Hospital:	University of Munster (Germany), Interdisziplinares Zentrum fur Klinische Forschung
In-Kind support reported by Industry:	Yes