Thursday, 13 August 2009

Fizzy drinks and Liver disease

Confusing Cause and Correlation.

The Daily Express had a front page story about the Danger of Just Two Fizzy Drinks. It reported work by researchers at a liver unit in Israel who had apparently found that drinking a litre of fizzy drinks or fresh juice were five times more likely to develop non-alcoholic fatty liver disease (NAFLD).

Although the Express story gives some details about the study design, there aren't many. Reading it we don't know how many people took part in the study, if it was a dietary intervention, or just an observational study. We have no information about how long the study continued.

There are a couple of suggestions from the researcher, Dr Nimer Assy, about problems that drinks high in fructose could cause, but these are not backed up by data.

The NHS Behind the Headlines site has, as usual, an excellent breakdown of what was done, and what it all means.

The paper itself is a conference poster report from April 2009. There isn't much to go on in it, so I had a look to see where the other information in the article could have come from. There was no sign of a press release on the news services Science Daily or EurekaAlert. Nothing from the Ziv Medical Centre (where Dr Assy works) itself. Strange.

Eventually I found a link to Israel 21C where there is more detail about the study and quotes from Dr Assy.

Interestingly, when you read the paper, and the interview, you find that the Daily Express have got a few, quite important, details wrong.

Israeli scientists at the Ziv Liver unit in Haifa compared two groups of
volunteers – neither of which was at risk of developing the condition.

When they finished the study they found that 80 per cent of those who had consumed fizzy drinks and fruit juices had fatty liver changes.
But only 17 per cent of the control group – who had not drunk fizzy drinks – developed fatty ­livers.

What isn't clear here, is that there were three groups of people taking part in the test.
  • 32 patients with NAFLD who had typical metabolic risk symptoms (such as obesity and diabetes)
  • 28 patients with NAFLD who didn't have risk symptoms
  • 17 patients without NAFLD

The amount of fat in the liver was measured using ultrasound. At the beginning and end of the six month study the volunteers were asked to keep a 7 day food and drink diary.

What the researchers found was that 70% of the patients with NAFLD drank more than 1/2 litre of soft drinks compared with 20% of the controls.

So, there is a correlation between drinking lots of soft drinks and having NAFLD. Does that mean that drinking lots of soft drinks will cause liver disease?

No, not necessarily. This study can't tell us if it does. Unfortunately, the Daily Express (and to be fair, the Daily Telegraph who also carried the story) didn't realise that.

Abed A, Nseir W, Ali T et al. Soft drink consumption linked with fatty liver independently by metabolic syndrome. Journ of Hepat. April 2009