Veröffentlichungen • Giessen Wholesome Nutrition Study: Relation between a health conscious diet and blood lipids

I. Hoffmann¹, M.J. Groeneveld¹, H. Boeing², C. Leitzmann<sup>1

1</sup> Institute of Nutritional Science, University of Giessen, Wilhelmstrasse 20, D-35392 Giessen, Germany
² German Institute for Human Nutrition, Arthur-Scheunert-Allee 114-116, D-14555 Potsdam-Rehbrücke, Germany

Poster 8th European Nutrition Conference
Scand J Nutr/Näringsforskning 34 (2S), 1999

Introduction

The empirical evidence for the longterm effect of a health conscious diet on blood lipid profils is small. Wholesome Nutrition is a diet taking most of the current recommendations for the prevention of nutrition related diseases into account (e. g. Nutrition Committee, American Heart Association 1996 for the prevention of CHD). It is a primarily lacto-vegetarian diet which mainly consists of vegetables, fruits, potatoes, legumes, whole grain products as well as dairy products. Since the consumption of meat and fish is not expressivly recommended it is possible to practise a vegetarian or a non-vegetarian version of this diet (Koerber et al. 1999).

Objective

Nation-wide cross sectional study in former West Germany with 243 healthy women, aged 25 - 65 y, adhering to Wholesome Nutrition for at least 5 years (with the subgroups of 111 ovo-lacto vegetarians and 132 low-meat eaters) and a corresponding control group of 175 women eating the average German mixed diet. Both groups were recruited through advertisements and selected according to their food consumption.

Methods

To study the relation between a diet taking most of the current recommendations for the prevention of nutrition related diseases into account (Wholesome Nutrition) and blood lipid profile (total cholesterol, LDL-, HDL-cholesterol, LDL/HDL-ratio, triglycerides).

Results

The food consumption of the two Wholesome Nutrition subgroups differs significantly from that of the control group (Table 1) - partly caused by the selection criteria. The Wholesome Nutrition subgroups meet (on average) all of the CHD-preventative recommendations except for fat intake (e.g. they prefer full fat dairy products to nonfat/low fat).

Between the blood lipids (total cholesterol, LDL-, HDL-cholesterol and LDL/HDL-ratio) and belonging to a diet group there is no crude association (Table 2). Only the triglyceride concentrations of the ovo-lacto vegetarians are significantly lower compared to the controls. The stratification for a single confounding variable shows differences between the diet groups in terms of blood lipids not seen without this control. The multivariate logistic regression analysis (Tab. 3), considering potential confounders simultan-eously, reveals that the Wholesome Nutrition group as well as its subgroups (the ovo-lacto vegetarians and low-meat eaters) have higher HDL-cholesterol levels than the control group and that the ovo-lacto vegetarians have lower LDL/HDL-ratios and lower triglyceride levels. No significant differences are observed for total cholesterol and LDL-cholesterol. None of the blood lipids discriminate between ovo-lacto vegetarians and low-meat eaters.

Discussion

The data of this study:- suggest that women adhering to such a preventative diet, especially those who practice an ovo-lacto vegetarian version, exhibit more favourable LDL/HDL-ratios, HDL-cholesterol- and triglyceride concentrations than those who eat the average German diet.- are opposite to findings of other studies on healthy longterm vegetarians (e.g. Richter et al. 1993, Thorogood et al. 1987). Those studies also support the advantages of vegetarian diets concerning the blood lipids, but they predominantly show lower concentrations of total cholesterol and LDL-cholesterol for vegetarians while the results for HDL-cholesterol are not consistent.- question whether a reduction of total fat intake in order to influence blood lipid fractions and hence reduce the risk of CHD might be as decisive as the proportion of foods of plant origin in an overall well planned diet. It appears that individuals can choose a diet with a similar proportion of total fat and total carbohydrate compared to the general population as long as it contains a high proportion of plant foods with more protective nutrients and less risk factors.

Conclusion

The more favourable distribution of HDL-cholesterol, LDL/HDL-ratio and triglycerides in the Wholesome Nutrition group (especially in the vegetarian version) supports the view that a change in nutritional habits has far reaching public health implications, since a lower risk for CHD in the long run may be expected for the Wholesome Nutrition groups.

References

• Hoffmann I (1994). Gießener Vollwert-Ernaehrungs-Studie: Untersuchung auf Bias am Beispiel von Fettstoffwechsel-Parametern. Dissertation, Institute of Nutritional Science, University of Giessen. (German)
• Koerber Kv, Maennle T, Leitzmann C (1999). Vollwert-Ernaehrung - Konzeption einer zeitgemaessen Ernaehrungsweise. 9th ed., Heidelberg: Haug. (German)
• Nutrition Committee, American Heart Association (1996). Dietary guidelines for healthy American adults. Circulation 94, 1795-1800.
• Thorogood M, Carter R, Benfield L, McPherson K, and Mann JI (1987). Plasma lipids and lipoprotein cholesterol concentrations in people with different diets in Britain. Brit. Med. J. 295, 351-353.
• Richter V, Rassoul F, Purschwitz K, Hentschel B, Rotzsch W (1993). Lipid screening on population basis and of vegetarians. Akt. Ernähr. Med. 18, 286-290.

* significant differences with Mann-Whitney U test (U-Test) between:

ovo-lacto vegetarians and control group

low-meat eaters and control group

ovo-lacto vegetarians and low-meat eaters

0.05; adjusted by the Shaffer multiple comparison procedure)

*significant difference between ovo-lacto vegetarians and control group according to 95 % CI

adjusted for age, BMI, alcohol consumption, physical activity, intake of sex hormones, education and per capita income by multivariate logistic regression analysis

changes relate to 10 mg/dl*significant differences with logistic regression at p 5 %

differences with logistic regression at p 5 %

zurück zu Veröffentlichungen