Components versus whole foods

[Guest guest]

Hi All,

The below brief review is a fairly interesting to me

article regarding using our whole of in our diets versus

use of neutraceuticals that represent their components

that are thought to provide their benefit.

How beneficial is the use of the component versus control

whole food trials that are described in the below?

For references [2] and [5], access information is provided

in the reference list at the end of the full-test below.

The issue of lycopene being not used in the one experiment

[2] appears to be to me not an issue of relevance, since the

other reference given additionally in the list indicated that

glutathione peroxidase is not affected by lycopene.

DC.

Nutrition and metabolism: antioxidants, the food matrix and

methodological

considerations - editorial comment.

Curr Opin Lipidol. 2005 Feb;16(1):111-3. No abstract available.

PMID: 15650569 [PubMed - in process]

This section provides a rapid update service, covering the whole

field of lipidology, and lists papers entered into the publisher's

database between 1 October 2003 and 30 September 2004.

Key articles are selected for comment by specialists and are

highlighted in the following way:

• Papers considered to be of specialist interest

•• Papers considered to be of outstanding interest

A bibliographic listing follows each comment and selected papers

are accompanied by an annotation in which the scope and context of

the article are summarized briefly.

Despite best efforts by many to increase vegetable and fruit

intake across the industrialized nations, consumers persist in

ignoring the advice. Perhaps emphasizing the benefits in a different

way may get through to the health-conscious consumer. Demonstrating

the functionality of their components in addressing disease risk

reduction may be an option and is the starting point of this

commentary.

The delightfully named sweetie fruit, a hybrid citrus variety, was

trialled [1] in a small, short-duration, secondary prevention study

of coronary heart disease. Consumption of two fruit per day for 30

days, during which time patients also omitted their lipid-lowering

medication, resulted in small but significant falls in total and LDL

cholesterol compared with controls consuming the usual

antiatherogenic diet but without citrus fruit. The authors pointed to

the fibre and antioxidant content of sweetie fruit as contributing to

the result. It remains to be seen whether longer-term consumption of

the citrus fruit would have maintained the effect from this whole

food approach.

A second small, short-term study [2] but with complete control of

food intake, and in healthy nonsmokers, tested the effects of food

matrix compared with supplements. Volunteers consumed as part of

their diet either 600 g daily of vegetables and fruit for 25 days or

a supplement which attempted to mimic the micronutrient content of

the food. The controls were deprived of all vegetables and fruit.

Biomarkers of antioxidant damage or defence were the outcomes

monitored. Critically, the supplement matching excluded lycopene and

there were few outcomes between the supplement and the food that

could not be explained by this lack of matching. The one exception

was the antioxidant defence marker, glutathione peroxidase, which was

higher on vegetables and fruit compared with the supplement and

control diets.

This study did not resolve the issue of whether vegetables and

fruit are better than supplements, but does emphasize the value of

the contained micronutrients. Better designed studies may help and

this issue was raised in a review [3••] questioning the divergent

outcomes reported for antioxidants and certain diet-related diseases.

Specifically the authors question whether it should be a requirement

in studies of antioxidant action that participants are matched on

their prior antioxidant status. They argue that failure to have done

so in previous large-scale trials may have accounted for the varied

results seen, despite the best theoretical and in-vitro evidence of

effectiveness.

Perhaps including a measure or index of dietary antioxidants would

be useful in studies that also matched the antioxidant status of

individuals. The authors [4] used data from a large (27 111

participants) trial (the Alpha Tocopherol Beta Carotene prevention

study) to construct an index of dietary antioxidants that was

predictive of lung cancer risk in male smokers. The index, made up of

dietary carotenoids, flavonoids, and vitamin E, was only marginally

better than daily fruit and vegetable intake in predicting the

outcome from this cohort study. Perhaps a further analysis

correlating the antioxidant status of participants at entry to the

cohort with the dietary index would have identified outliers and

increased the predictability of the model.

Which brings me to the final paper in this commentary. To develop

their index, the authors of the previous paper used a statistical

technique known as principal component analysis. This technique does

have its detractors because it sometimes fails to identify meaningful

dietary patterns. Using a technique called reduced rank regression

and applying it to a dataset derived from the European Prospective

Investigation of Cancer study, the authors [5••] were able to

identify food patterns associated with risk of diabetes. The four

factors extracted by reduced rank regression explained 93% of

response variation, compared with 42% using principal component

analysis.

Using better methodology and improving matching of individuals for

intervention trials may remove some of the confusion around the roles

of antioxidants and enable clearer messages to be delivered on the

benefits of vegetables and fruit and their contained nutrients.

References

1 Gorinstein S, Caspi A, Libman I, et al. Preventive effects of diets

supplemented with sweetie fruits in hypercholesterolemic patients

suffering from coronary artery disease. Prev Med 2004; 38:841–847.

PMID: 15193907 [PubMed - indexed for MEDLINE]

2 Dragsted LO, Pedersen A, Hermetter A, et al. The 6-a-day study:

effects of fruit and vegetables on markers of oxidative stress and

antioxidative defense in healthy nonsmokers. Am J Clin Nutr 2004;

79:1060–1072. PMID: 15159237 [PubMed - indexed for MEDLINE]

http://tinyurl.com/6rr7m = http://www.nutritionj.com/content/3/1/15

3•• Violi F, Loffredo L, Musella L, Marcoccia A. Should antioxidant

status be considered in interventional trials with antioxidants?

Heart 2004; 90:598–602.

4 ME, Mayne ST, Stolzenberg- RZ, et al. Development of

a comprehensive dietary antioxidant index and application to lung

cancer risk in a cohort of male smokers. Am J Epidemiol 2004; 160:68–

76. [Context Link]

5•• Hoffmann K, Schulze MB, Schienkiewitz A, et al. Application of a

new statistical method to derive dietary patterns in nutritional

epidemiology. Am J Epidemiol 2004; 159:935–944. PMID: 15128605

[PubMed - indexed for MEDLINE]

Recommended reading

•• Hoffmann K, Schulze MB, Schienkiewitz A, et al. Application of a

new statistical method to derive dietary patterns in nutritional

epidemiology. Am J Epidemiol 2004; 159:935–944. Using a technique

called reduced rank regression, the authors were able to better

identify food patterns associated with disease risk than when using

the more commonly available method of principal component analysis. A

method to watch for in future studies.

•• Violi F, Loffredo L, Musella L, Marcoccia A. Should antioxidant

status be considered in interventional trials with antioxidants?

Heart 2004; 90:598–602. Questioning the confusing relationships found

between antioxidants and diet-related diseases, the authors recommend

that individuals are matched on their prior antioxidant status in

studies of antioxidant action. They argue that failure to have done

so in previous large-scale trials may have accounted for the

disparate results.

Al Pater.