About 30% of adult Australians are thought to have non-alcoholic fatty liver disease – fatty liver for short – and its prevalence is increasing. In its early stages fatty liver is of minor concern and there are no symptoms. But it can progress to a stage called NASH where liver cells begin to be damaged and this is associated with inflammation. At the next stage the risk of cirrhosis of the liver increases which may lead to liver failure and even cancer of the liver. Together, fatty liver and NASH are now the number one cause of liver disease in Western countries. However, the major cause of death in people with fatty liver is cardiovascular disease.

What causes fatty liver?

Fatty liver develops when the liver’s normal processes of delivery and removal of fats go awry. Although the exact cause is not known several things could be going wrong:

• Perhaps too much fat is being delivered to the liver
• The liver may have a problem ‘burning’ fats for energy
• Fat (triglyceride) production by the liver may be increased
• The movement of triglycerides from the liver into the bloodstream may be impeded.

But what is the underlying cause?

Fatty liver is now considered to be the liver expression of the metabolic syndrome. This syndrome is a cluster of abnormalities including central obesity (fat around your waist), raised blood glucose, high triglycerides, low HDL-cholesterol and high blood pressure. Usually, not all of these occur in the same person but if you have central obesity and any two of the others then you have the metabolic syndrome. And you are a good candidate for fatty liver too. Not surprisingly, people with type 2 diabetes are twice as likely as the general population to have fatty liver. The association with the metabolic syndrome explains why people with fatty liver are at increased risk for cardiovascular disease.

Image: source

The critical link between metabolic syndrome and fatty liver is insulin resistance i.e. the cells of the body have become less sensitive to insulin. When insulin resistance is present more free fats flow from the body’s fatty tissue to the liver which then packages them up into triglycerides and pushes them out into the bloodstream. This high flux of fat through the liver may be the driver of fatty liver.

Lifestyle management of fatty liver

In the absence of any evidence-based drug treatment of fatty liver the best option is lifestyle intervention but it has to be said that there is little in the scientific literature to go on. With the general acceptance that insulin resistance is the core of the problem strategies that are known to improve insulin sensitivity, such as increased physical activity and weight loss, are generally supported.

There were encouraging findings from a recent meta-analysis that showed a benefit of exercise on liver fat, even when weight loss was minimal or absent. However, the quality of the available trials included in this study was only fair.

Weight loss of 5% or more lessens fatty liver and weight loss of 7% or more lowers inflammation of the liver, according to a recent meta-analysis by Musso and colleagues. Again, there are only a handful of good studies. The American Association for the Study of Liver Diseases recommends weight loss of 5-10% for overweight people with fatty liver.

Diet composition

There is no established diet for fatty liver and no evidence-based clinical guidelines. There are simply too few published randomised controlled trials to provide the data necessary to create such guidelines.

A review by Conlon and colleagues draws on the American Diabetes Association guidelines for the prevention and management of diabetes and cardiovascular disease for inspiration for the dietary management of fatty liver. A dietary intake pattern that emphasises whole grains, fruits, vegetables, legumes and low-fat dairy is recommended, together with a limit on calories, less saturated fat and less refined carbohydrates. No limit on total carbohydrate is recommended. Glycaemic Index is discussed but there are no firm guidelines. It’s all quite general and all extrapolated from recommendations for other conditions.

Image: source

Mediterranean diet

Recently, an Australian randomised controlled trial into the effects of a Mediterranean diet on people with insulin resistance and fatty liver was published. Compared to the control low fat/high carbohydrate diet the Mediterranean diet improved insulin sensitivity and reduced liver fat, even though body weight was not affected. Whether the effect was due to the monounsaturated fats in the Mediterranean diet or the carbohydrates they effectively displaced is unclear. The higher polyphenol content of the Mediterranean diet may also be relevant.

Fructose

It has been hypothesised that fructose intake may affect the risk for fatty liver. To test this hypothesis Chiu and colleagues recently conducted a meta-analysis of controlled feeding trials to determine the effect of fructose on two markers of fatty liver. They considered two types of trials – one where fructose was exchanged for other carbohydrates without affecting calorie intake, and the other where diets were supplemented with fructose thereby increasing the energy intake.

The researchers found that when calorie intake was constant fructose had no effect on markers of fatty liver. But overfeeding with fructose was associated with increases in both markers of fatty liver. They concluded that the adverse effects observed in the latter trials may be due to excess calorie intake rather than fructose per se.

The authors noted that the small number of trials available for inclusion in their meta-analysis was a limitation. They identified a need for larger, longer, high-quality trials of the effect of ‘real-world’ intake patterns of fructose on markers of fatty liver in at-risk populations.

Summary

For people with fatty liver, regular physical activity and 5-10% weight loss in those who are overweight are the best ways of managing the problem. Because these lifestyle measures also improve insulin sensitivity they will have beneficial effects on other elements of the metabolic syndrome and lower the risk for cardiovascular disease.

Obesity is proving to be an intractable public health problem demanding innovative solutions and one idea that is attracting attention is the taxation of sugar-sweetened beverages. The theory is simple enough. Basic economics tells us that if the price of sugary soft drinks were to rise, their consumption would fall; lower intake should mean lower calorie intake which would lead to lower body weights. But would it work in practice?

The Ohio experience

A recent research study conducted in the United States provided some interesting insights. Taxing soft drinks has a long history in the US and occurs in many states today, though historically the rates have been low and the purpose has been to raise revenue. But there was an interesting exception. In 1992, the state of Ohio introduced high taxation of sugary soft drinks which was then repealed at the end of 1994. This provided an opportunity to test the effect of taxation of soft drinks on body weights over a period of two years. The researchers compared changes in body weights in Ohio over this period to (1) all other states that had no increased taxation and (2) a bundle of states with the same mean BMI as Ohio. The researchers found:

… very little evidence that the large tax imposed in Ohio had any detectable effect on population weight … our results cast serious doubt on the assumptions that proponents of large soda taxes make on its likely impacts on population weight.

How come? Why didn’t quite high taxation of sugary soft drinks affect body weights?

How strong is the evidence that soft drinks affect body weight?

It may come as a surprise to many dietitians and nutritionists that there is actually a debate in the scientific literature about whether lowering intakes of sugar-sweetened beverages would reduce the prevalence of obesity. Two excellent papers published in Obesity Reviews last year presented the opposing views.

In one corner was Professor Frank Hu from the Harvard School of Public Health who argued that the issue of whether sugary soft drinks are linked to body weight has been resolved and the answer is yes. In developing his argument Professor Hu first drew on epidemiological studies, citing a Harvard meta-analysis of cohort studies. It has to be said that the association between sugar-sweetened beverage consumption and body weight in children in this study only just reached statistical significance, though an updated meta-analysis by the same research team is also supportive.

The second line of evidence considered was randomised controlled trials, which would be expected to provide better insight but these were relatively few in number and their findings were conflicting. However, Professor Hu was swayed by two recent high quality trials by Ebbeling et al and de Ruyter et al both of which indicated that sugary drinks affect body weight.

Image: source

The alternative view

The alternative view was presented by a team headed by Professor David Allison from the Nutrition Obesity Research Centre at the University of Alabama. Their view was based on data from their updated meta-analysis of randomised controlled trials, which included the two recent trials by Ebbeling et al and de Ruyter et al. Again the trial evidence was equivocal. On the one hand, adding sugar-sweetened beverages to subjects’ diets was linked to increased body weights, but lowering intake of these beverages did not lower body weight, except in overweight subjects. However, these authors noted that the recent high quality trials were ‘tilting the needle’ in favour of an association.

The effect size is small

One thing that had never really registered with me was how small the ‘effect size’ of sugar-sweetened beverages on body weight really is. The authors explained it thus:

Increasing consumption of [sugar-sweetened beverages] explains 1.92% of the variance in body weight or BMI change. Reducing consumption of [sugar-sweetened beverages] in persons of all weight categories explains 0.09% of the variance in body weight or BMI change. Among persons who are overweight or obese at baseline, reducing the consumption of [sugar-sweetened beverages] explains 1.54% of the variance in body weight or BMI change.

So although the consumption of sugar-sweetened beverages appears to be linked to body weight the magnitude of the effect is tiny.

Is substitution a factor?

One of the key assumptions of the taxation strategy is that the higher cost of sugary soft drinks would result in lower intake of these beverages and lower calorie intake. In other words, there would be no substitution with other calorie-containing drinks or foods. This may well be the case in the constrained environment of a randomised controlled trial but would it happen in real life?

A study in 2010 found that although taxing soft drinks resulted in a small fall in soft drink consumption this was completely offset by increases in consumption of other high calorie drinks, such milk-based drinks. Consequently there was no effect on total calorie intake at all. Although most nutritionists would be glad to see nutrient-rich beverages replacing nutrient-poor drinks in children’s diets, this is not the intended purpose of an obesity tax.

Could taxation be made to work?

If taxation is to work as a strategy for obesity prevention it may be worthwhile revisiting the 2011 study by Mozaffarian and colleagues which looked at which foods and drinks were associated with weight gain over time. The surprising findings of this study, which considered three large cohorts of men and women, were that just about every group of foods and drinks was associated with weight gain, positively or negatively.

Image: source

Although sugary soft drinks were positively linked with weight gain, so were fruit juice, potatoes, potato chips, red meat, processed meats, sweets and desserts, butter and refined grains. Rather than tax just one of these, an effective strategy for obesity prevention would probably need to tax them all. As nuts, yoghurt, fruits, vegetables and wholegrains were all negatively associated with weight gain an effective taxation strategy would also need to lower the price of these relative to the ‘fattening foods’ and also those foods with an apparently neutral effect.

Although it sounds complicated it might work. But would the Federal Treasury ever support such an approach?

Last week on ABC radio’s AM current affairs program the former chair of the National Preventative Health Taskforce lamented that the Government had failed to address obesity. In 2009, the Taskforce released a report titled Obesity in Australia: a need for urgent action which was supposed to be the springboard for Government action on the issue. Instead, the report was tossed onto a large pile of obesity reports and recommendations that have been ignored by our politicians.

What went wrong? Why did the then Government, which was favourably disposed to public health and disease prevention, fail to act? My guess is that it was the appalling quality of the report.

Convincing the Minister

Have a read of this report and try to imagine how the conversation went when the good professor presented it to the Minister for Health. Maybe it went something like this:

Professor: Minister, here is our carefully crafted blueprint for addressing the obesity epidemic that has been raging in Australia for the last 30 years.

Minister: Hmmm. How come there is no chapter outlining the causes of the epidemic? How can you address a problem if you don’t understand its cause?

Professor: Oh, we skipped that bit. But everyone knows the cause – its sugar and fat.

Minister: So the fat content of the Australian diet has gone up in the last 30 years?

Professor: No … not exactly. It’s gone down a little.

Minister: But there is scientific evidence of a link between fat intake and obesity, isn’t there?

Professor: Actual evidence? No. But it’s common knowledge that fat makes you fat.

Minister: I see. And sugar intake? Has that gone up in the last 30 years?

Professor: Well, total sugar intake may have come down a bit too. But it’s the sugary drinks we are really worried about.

Minister: So, intake of sugary drinks is going up?

Professor: Um, no. It’s coming down too.

Minister: How come? What’s going on there?

Professor: Dunno …  Minister, I can see where you are going with this …

Minister: Oh, is that the time? Next!

The battle between good and evil

Both the radio interview and the report reflect a belief system among some of our public health colleagues that goes something like this:

• People are getting fat because they are eating unhealthy food.
• Unhealthy food is easily defined, in terms of sugar and fat.
• Rapacious food companies pump food full of sugar and fat because these ingredients are cheap. The capitalists reap the rewards; society pays the price in the form of obesity.
• The food companies are just like Big Tobacco – they know they are wreaking havoc but they deny their guilt and just focus on profit. Bastards!
• The only solution is to regulate everything – food advertising, food labelling, the composition of food and the cost of food, through taxation.
• Alas, the food companies are all-powerful and governments buckle to their will.
• Our cause may seem hopeless but we must fight the good fight.

Image: source

These folks really need to stop talking to themselves and get out more.

Equating food companies with the tobacco industry is just a dumb strategy that only serves to make its proponents look extreme and its opponents appear moderate. Ideological conflict in relation to food may provide some with a warm inner glow but the recent evidence suggests that it doesn’t achieve very much. The success of public health activities depends on outcomes, not on the purity of the cause. In particular, ideological battles enable and encourage politicians to sit back and do nothing while opposing sides slug it out.

How much failure is enough?

When the same approach fails time after time, what should happen next? Obviously, try it again. In this morning’s Sydney Morning Herald there is an article titled “Big Junk is just as evil as Big Tobacco” penned by a leading public health academic. The belief system is on display again and the author comes to the same hand-wringing conclusion:

“There is little evidence that governments of any colour will take serious action on obesity. They will continue to involve the junk food industry in policy discussions, run soft education programs, and back off from any tough measures. We should not kid ourselves that there is any real intent to tackle the problem.”

Image: source

You see, everyone else is to blame. At no stage do our public health experts consider that their adversarial strategy may be flawed. Given the recent election of a Government even less inclined to respond to the ‘regulate everything’ approach, I respectfully suggest that the public health strategy needs to change. Some thoughts:

• Drop the ideological crusade. It will fall on deaf ears.
• Step across the aisle and partner with an industry, company or association around specific projects. Show the politicians that you can work collaboratively with other sectors in the food and nutrition system.
• Conduct research. Develop some programs that actually work.
•  Avoid using terms such as ‘unhealthy foods’, based on meaningless criteria like sugar and fat. Work up some sophisticated models for classifying foods. Be scientific.

This consensus approach won’t appeal to those who define themselves by what they hate. However, I am sure the new Health Minister in Canberra would respond well to an obesity prevention proposal that is soundly based in science, broadly supported by various constituencies and practical to implement.

If only his predecessors had received one.

The Look AHEAD Study

People with type 2 diabetes have increased risk for cardiovascular disease. As overweight is often associated with cardiovascular risk factors it has always been assumed that excess body weight contributes to the burden of cardiovascular disease and that weight loss would lower it.

The Look AHEAD Study was conceived to test this hypothesis. This large, multi-centre, randomised controlled trial began 11 years ago in the United States. Over 5000 overweight men and women with type 2 diabetes were recruited – half received active weight loss intervention and the control group received a general program of support and education. The intervention group successfully lost weight. After one year they had lost nearly nine per cent of their initial body weight and at 11 years their average weight was still about five per cent less than baseline.

Over the years the good news started to flow from the Look AHEAD Study. A string of papers showed that weight loss in the intervention group was linked to improvements in fitness, mobility, blood glucose, sleep apnea, urinary incontinence, symptoms of depression, blood clotting and body image. And use of medications was reduced. All very positive, but the real test was yet to come: would weight loss lower rates of cardiovascular disease?

Late last year the National Institutes of Health announced that the Look AHEAD Study had been stopped due to ‘futility’. After 11 years there had been no significant reduction in heart attacks or strokes in the weight loss group and there was no prospect of success if the trial was continued – a stunning finale to this much anticipated trial.

Image: source

Contrast with the DPP

The Look AHEAD Study was preceded by the successful Diabetes Prevention Program, which showed that weight loss and exercise reduced the incidence of diabetes by 58 per cent in overweight subjects with impaired glucose tolerance. This result was virtually identical to that of a Finnish diabetes prevention study published the previous year. Look AHEAD took these important findings further and was the first study to examine the long-term effects of a lifestyle intervention on major cardiovascular disease events and death in adults with type 2 diabetes. But it didn’t work.

The failure of Look AHEAD has led to much soul-searching and debate. What went wrong? It can’t be argued that the intervention was not strong enough. Maintaining a five per cent weight loss for 11 years was a significant achievement. Some say the control group was just too damn healthy, yet if the intervention just needed a bit more time to demonstrate a significant result the researchers would never have pulled the plug.

Actually, nothing went wrong with the Look AHEAD Study. A hypothesis was tested and disproved, so now we have new knowledge. Although overweight is the public health nutrition issue of our age, addressing it is sometimes seen as a panacea for all ills. The lesson from the Look AHEAD Study may be that we just have to accept that while weight loss can reduce the incidence of diabetes, or slow its progression, and has many health benefits, it just doesn’t prevent cardiovascular disease in this at-risk group.

Implications for diet: quantity vs quality

For me, the insight is that while overweight is essentially about diet quantity the focus for the prevention of cardiovascular disease needs to be on diet quality. Although a heart healthy diet can take many forms the key aspects of diet quality are quite clear – more fruit, vegetables, legumes, nuts, unsaturated vegetable oils, fibre-rich cereals and fish, and less saturated fats and poor quality carbohydrate foods. The recent PREDIMED trial showed the effectiveness of diet quality in preventing heart disease, in the absence weight loss or increased physical activity. The same trial also showed that better diet quality lowered the risk of developing diabetes – killing two birds with one stone.

For diets for overweight people with diabetes it’s not a matter of focussing on either quantity or quality. It’s got to be both.

Image: source

New meta-analysis

A new meta-analysis of the effects of overweight and obesity on all-cause mortality was published this month in the Journal of the American Medical Association. The analysis included data from 97 studies from around the world, providing a combined sample size of more than 2.88 million subjects and more than 270,000 deaths.

The risk of death of all obese subjects (BMI>30) was substantially and significantly higher (18%) than that of subjects of normal weight (BMI 18.5-25) – the sort of finding that we have come to expect. However, the results told a different and perplexing story when narrower weight categories were considered. For example, those who were overweight (BMI 25-30) experienced 6 per cent lower risk of mortality than subjects of normal weight. Subjects with grade 1 obesity (BMI 30-35) had the same mortality risk as subjects of normal weight. Consequently, the mortality risk associated with obesity appeared to be due entirely to higher grades of obesity – BMI greater than 35, which was associated with a 29% increase in risk.

How can it be?

Previously, a J-shaped relationship between body weight and mortality has been observed i.e. very low body weight was linked with higher risk; normal body weight was associated with the lowest risk; and mortality risk climbed again as body weight increased above the normal range. However, the latest study suggests that we may be looking at a long, drawn out J shape with low mortality risk associated with a broad range of body weights, including normal weight, overweight and grade 1 obesity.

But how come? Isn’t obesity supposed to be associated with increased risk for type 2 diabetes, coronary heart disease, stroke and some cancers? Maybe relying on BMI alone is just not accurate enough. As highlighted in the accompanying editorial, BMI is crude measure of health status. People with the same BMI can have widely differing metabolic health depending on their fitness, the amount of fat on their bodies and, importantly, the distribution of that fat. Waist circumference has been suggested as a better measure of risk than BMI as it is more likely to capture central obesity and its associated risks.

Image: source

An obesity paradox?

More challengingly, the editorial suggests that there appears to be an ‘obesity paradox’ i.e. a protective effect of overweight or grade 1 obesity in old age and in the presence of chronic conditions, such as heart disease and diabetes. They speculate that small excess amounts of adipose tissue may provide energy reserves needed during acute catabolic illnesses, have beneficial mechanical effects with some types of traumatic injuries and convey other effects that need to be investigated. The implication is that it can’t be assumed that the life expectancy of all overweight subjects, even those with chronic disease, will increase with weight loss. Another paradox – that’s just what we need!

What about morbidity?

Another recent study found no association between overweight, obesity and mortality though it was a different story when the researchers looked at the risk for disability. Being overweight was associated with a 33% increase in the risk for becoming disabled and the risk was doubled for those with obesity. Furthermore, recovery from disability was negatively associated with obesity.

So, although being overweight may not kill you, it could make you sick.