N02026: A randomised trial to examine the public health impact of changes in the dietary n-6:n-3 PUFA ratio

Study Duration: October 2000 to September 2003

Contractor: MRC Human Nutrition Research

Recent changes in dietary habits have led to a reduction in saturated fat and an increase in polyunsaturated fatty acids (PUFA), especially n-6 PUFAs, together with a marked increase in the n-6:n-3 ratio. Data from diverse sources, including epidemiological analysis, experimental research and human intervention trials have suggested that this change may have inadvertently led to impairments in a number of risk factors. These include decreased insulin sensitivity, hyperlipidaemia and increased inflammatory mediators, which may in turn be associated with adverse health outcomes including cardiovascular disease, diabetes and chronic inflammatory disorders. Recent evidence also suggests an effect on the risk of weight gain and we hypothesise that imbalances in the n-6:n-3 ratio have the potential to provide a critical link between obesity and its co-morbid conditions. The spectrum of co-morbid conditions appears broad but there are clear mechanisms which provide plausible links, especially through effects on insulin resistance and pro-inflammatory cytokines.

The evidence to date mostly relates to the protective effect of n-3 PUFAs, but its impact has been reduced by the seemingly inconsistent results from different studies. We hypothesise that this confusion has arisen from the failure to account for the background intake of n-6 PUFAs and that the logical key intervention is a reduction in the ratio of n-6 to n-3 PUFAs. The impact of this dietary change can be most effectively evaluated in a population of overweight and obese subjects who have an increased incidence of a abnormal pathology.

This study will provide new information on the metabolism and health effects of polyunsaturated fatty acids and will test the public health impact of a change in the ratio of n6:n3 fatty acids, using a simple intervention with widespread applicability.

Objectives and rationale

1. Key Objective

To measure the impact of changes in the n-3:n-6 ratio, achieved by modifiying habitual fats and oils used in the home, and/or the addition of long chain n-3 PUFA into the diet, on obesity and it's co-morbidities including:

(i) Body weight and fatness
(ii) Insulin sensitivity
(iii) Fasting blood lipoproteins
(iv) Haemostatic factors
(v) Inflammatory mediators
(vi) Markers of immune function

2. Six month dietary intervention in 150 overweight middle aged adults

150 middle aged (35-65 years) volunteers with a body mass index (BMI) of 25-35 (kilograms per square metre) will be recruited and baseline plasma phospholipid composition assessed. Using a stratified randomisation procedure, based on fatty acid status, subjects will be allocated to a seasonal control group (no intervention) or one of four intervention groups:

(i) Sunflower oil and spreads (high n-6:n-3; approx. 19:1) + white fish (low long-chain n-3).

(ii) Sunflower oil and spreads (high n-6:n-3; approx. 19:1) + oily fish (high long chain n-3).

(iii) Low erucic acid rapeseed oil and spreads (low n-6:n-3; approx. 3:1) + white fish (low long-chain n-3).

(iv) Low erucic acid rapeseed oil and spreads (low n-6:n-3; approx 3:1) + oily fish (high long chain n-3).

Compliance will be assessed primarily by changes in plasma phospholipid status and qualitatively by 4-day dietary records. Outcome measurements will be made primarily at 0, 12 and 24 weeks.

3. To measure the effects of the dietary intervention on:

i) Body weight and fatness
Body weight will be measured every 2 weeks. Total body fat , lean tissue mass and bone mineral will be measured using dual energy X-ray absorptiometry at 0,12 and 24 weeks.

ii) Insulin sensitivity
Two fasting blood samples will be taken at baseline and after 12 and 24 weeks of intervention to measure glucose and insulin. Insulin sensitivity will be assessed using the HOMA model. Additionally, on one occasion an oral glucose tolerance teat will be performed with measurements of glucose and insulin at 0, 30, 60, and 120 mins. One fasting sample will be used to measure HBA1c.

iii) Fasting blood lipoproteins
Two fasting blood samples will be taken at baseline and after 12 and 24 weeks of intervention to measure triglycerides, total cholesterol, HDL and LDL cholesterol and NEFA.

iv) Haemostatic factors
At baseline, 12 and 24 weeks a range of haemostatic factors will be measured including clotting time, ESR, prothrombin, fibrinogen, APTT, PAI antigen and activity, t-PA antigen and activity. Additional measurements will be performed by Prof. G. Miller (sub-contractor) including CRP, D-dimer, PMN elastase, activated factor XII, activated factor VII and factor VII antigen, von Willebrand factor,

v) Inflammatory mediators
At baseline, 12 and 24 weeks a range of inflammatory mediators will be measured including sialic acid, orosomucoid, ACT, IL-6, TNF alpha and TNF alpha r2, MIF.

vi) Markers of immune function
At week 24 a cross-sectional analysis of immune function will be performed based on endotoxin stimulated IL-1 and TNF alpha release from cultured peripheral blood mononuclear cells and Tissue Factor measurements.

The aim of this study is to measure the impact of changes in the n-6:n-3 ratio, achieved by modifying habitual fats and oils used in the home, and/or the addition of long chain n-3 PUFA into the diet, on obesity and its co-morbidities including: body weight and fatness; insulin sensitivity; fasting blood lipoproteins; haemostatic factors; inflammatory mediators; and markers of immune function. A six month dietary intervention in 150 overweight middle aged adults (35-65 years) assigned to a control group or one of four intervention groups:

• Sunflower oil and spreads (n-6:n-3 ratio approx 19:1) and white fish (low long chain n-3)
• Sunflower oil and spreads (n-6:n-3 ratio approx 19:1) and oily fish (high long chain n-3)
• Low erucic acid rapeseed oil and spreads (n-6:n-3 ratio approx 3:1) and white fish (low long chain n-3)
• Low erucic acid rapeseed oil and spreads (n-6:n-3 ratio approx 3:1) and oily fish (high long chain n-3)

Introduction

There is accumulating evidence that the type of dietary fat may have an effect on disease risk. Health recommendations to substitute vegetable-based fats and oils for saturated fat have been linked to reductions in cardiovascular disease (CVD) over the last 20-30 years. However, these dietary changes have also had incidental effects on the dietary balance of polyunsaturated fatty acids (PUFA). There are two families of PUFA, the n-6 series and the n-3 series. Linoleic acid (n-6) and alpha-linolenic acid (n-3) are essential fatty acids since they cannot be produced in the body and must be obtained from dietary sources. Linoleic acid (LA) and alpha-linolenic acid (LNA) are converted in the body to other members of their respective families. In western societies there has been a considerable increase in the availability and use of oils rich in n-6 PUFA (e.g. sunflower oil) with consequent rises in the dietary n-6:n-3 ratio. In the UK, this ratio is currently estimated at around 6:1 in the adult population. This marks a dramatic increase over the past 50 years corresponding to changes in the food supply. PUFA are metabolically active and are believed to exert potent effects on many health outcomes. At present, it is not clear precisely what absolute quantities or relative proportions should be consumed for optimal health.

To date, evidence of the health impact of PUFA mostly relates to the positive effects of the long chain n-3 fatty acids, especially EPA and DHA. The most concentrated dietary source of these fatty acids is oily fish and they are often referred to as fish oils. Around half the adult population take fish oil supplements. EPA and DHA can also be made in the body from LNA. However, the conversion of LNA to EPA and DHA is limited and may be restricted by high intakes of n-6 LA. It is not clear if increased intakes of LNA or decreases in the n-6:n-3 ratio can offer similar health benefits to long chain n-3 PUFA, or whether there are additional effects.

Rationale and Objectives

The study examined the health impact of reducing the dietary n-6:n-3 ratio, both by changes in the proportion of LA and LNA or increases in long chain n-3 PUFA. In particular it focused on the risk of CVD and diabetes. These risks are strongly associated with body weight and can be grouped into a cluster of conditions, which are collectively described as the metabolic syndrome (MS). MS is characterised by obesity and CVD risk factors including high blood pressure, elevated blood fat levels and abnormalities in blood glucose control (insulin resistance). The study was conducted in overweight and obese subjects who have a pre-existing susceptibility to these diseases.

Previous studies showing the effect of dietary fat composition on health improvements have often achieved large decreases in the n-6:n-3 ratios through the use of high dose fish oil supplements, the use of specially formulated foods and/or the provision of unusual types or quantities of foods. These approaches are useful in understanding the metabolic impact of different fatty acids but may not be the most appropriate in formulating dietary recommendations for the general population. This study was designed to inform policy makers of the health impact of a realistic change in the ratio of n-6:n-3 ratio using a simple food-based intervention with widespread applicability. Subjects were instructed to consume particular types of oils and fats, widely available in supermarkets and to eat two portions of either white or oily fish each week. The health of these subjects was compared with a group who continued with their usual dietary habits. This ensured that any positive results from the study could be easily translated into public health recommendations.

Additionally the study set out to explore how the health benefits of n-3 PUFA may be achieved, specifically the role of inflammation. There is growing evidence that high circulating concentrations of inflammatory markers are associated with an increased risk of disease. Arachidonic acid (AA), a member of the n-6 PUFA family, and EPA, a member of the n-3 PUFA family, are important precursors of a range of eicosanoids, which have broadly competitive actions. Eicosanoids derived from AA are generally pro-inflammatory, whereas those from EPA tend to oppose their inflammatory action. The balance between AA and EPA may determine the balance of these eicosanoids and hence the inflammatory status of an individual. Indeed EPA is known to have some anti-inflammatory properties, for example in the relief of symptoms of rheumatoid arthritis. This study measured the impact of changes in dietary fat composition on a range of circulating markers of inflammation.

Approach

One hundred and forty two overweight men and women, aged between 35 and 65 years were recruited into a 24-week dietary intervention study. Subjects either followed their normal diet or one of four specifically designed diets. Two groups were supplied with 2 portions of white fish/week (0.7g/week EPA+DHA), and the other two groups with 2 portions of oily fish/week (4.5g/week EPA+DHA). Each fish group was split into two groups in which usual fat spreads and cooking oils were replaced by those based on either sunflower (high LA) or rapeseed oil (high LNA) to manipulate LA:LNA ratio. Changes in risk factors for CVD and diabetes and the inflammatory response were assessed over the course of the study in each of the groups.

Results

The study was successful in achieving specific changes in the dietary n-6:n-3 ratio in each of the diet groups. Changes in dietary habits translated into significant changes in the circulating levels of certain fatty acids. Both groups receiving the rapeseed-rich oils and spreads showed a decrease in n-6:n-3 ratio relative to the group receiving sunflower oils and spreads. Both groups receiving oily fish showed increases in the proportion of EPA and DHA.

Both groups consuming oily fish showed significant reductions in plasma triglycerides. Triglycerides are a type of blood fat, known to be an important risk factor for CVD. There were no other significant differences in health outcomes, including blood pressure, cholesterol, insulin or glucose in any of the other groups when compared to each other or to the control group. Furthermore, measures of inflammatory status did not show a differential response to changes in the n-6:n-3 ratio. No adverse effects were measured in any group. This is broadly analogous to previous data, often with high dose fish oil capsules, where the most consistent and characteristic effect is to lower triglycerides, whereas effects on other risk factors are less consistent.

This data does not necessarily imply that such diets are ineffective in improving these other health parameters. It may be that a much longer period of dietary change is required in order to overcome the impact of prior dietary habits. Dietary surveys tend to show that people eating the most fish, especially oily fish, have a lower risk of cardiovascular disease, but this reflects lifelong dietary habits, rather than a short-term dietary intervention.

Alternatively it may be that this group, although overweight, were relatively healthy and there was limited scope for further health improvements. Indeed baseline glucose, insulin and lipids were all within the normal range. Previous studies that have demonstrated benefits of oily fish or decreases in the LA:LNA ratio have often been conducted in people with established heart disease.

Conclusions

This study has shown that simple food-based interventions can lead to significant changes in n-6:n-3 status. Consumption of rapeseed-rich oils and spreads and oily fish were associated with fatty acid profiles previously linked to health benefits. In this study these changes in dietary intakes led to a significant improvement in blood triglyceride concentrations. However there were no other measurable effects on risk factors for CVD and diabetes in this particular group over the 6-month period of dietary change.

The final report is available from the FSA Library and Information centre.
To obtain a copy, please contact the Enquiry Desk, Dr Elsie Widdowson Library and Information Services, Food Standards Agency ( tel: 020 7276 8181/8182 or email: infocentre@foodstandards.gsi.gov.uk )

Contact: Email: science@foodstandards.gsi.gov.uk