It’s funny how things seem to come in threes, and this week was an example of this maxim in that I saw three patients all of whom had deranged liver function tests. All of them had raised levels of at least two liver ‘enzymes’. This, in medicine, is generally taken as a sign of damage to the liver. While there are lots of reasons for why liver enzymes may be raised, our first thought in medicine is usually to ask about alcohol consumption. None of these individuals was abstemious, but at the same time none of them had alcohol consumptions that could be described as excessive.
When I see individuals with deranged liver function who do not drink much alcohol, my next thought is usually their diet. And in particular, I start thinking about whether they may have a touch of ‘metabolic syndrome’. This condition, characterised by excess weight around the middle of the body, is common, and tends to be a sign of an excess of carbohydrate and therefore insulin in the body. Insulin promotes fatty production in the body, and some of this can end up being dumped in the liver. As a result, liver function can become deranged, and if the problem persists, it may eventually lead to a condition known as ‘fatty liver’.
As it happens, each of the three patients I saw this week were carrying excess weight around their middles. My advice to them was to get control over blood sugar and insulin levels. Basically, that means eating less carb, particularly those carbs that tend to cause most disruption in blood sugar and insulin levels including bread, potatoes, rice, pasta and breakfast cereals.
I was reminded about these three patients today on reading a study published yesterday in the journal Gut [1]. In this study, Swedish researchers took a group of adults (average age 26) and put them on a regime which involved limiting their physical activity and getting them to eat two fast food meals each day for four weeks. The participants were monitored in term of, among other things, liver function and weight. Their results were compared with a group of individuals who were not subjected to the regime (these individuals acted as ‘controls’).
Over the course of the 4-week study, those on the fast-food regime put on an average of about 6.5 kg in weight. In particular, waist size increased significantly. The level of the liver enzyme known as ALT (alanine aminotransferase) went up from an average of 22.1 U/L (normal) to 97.0 U/L (abnormally raised). This would be taken, generally speaking, as a sign of liver damage. Not only that, but the fat level in the liver cells of these individuals increased by over 150 per cent. One of the 18 participants developed full blown fatty liver (quite a feat in just four weeks of unhealthy eating).
The findings of this study are reminiscent of Morgan Spurlock antics in the film ‘Supersize Me’. His one-month fast food diet experiment led to significant derangement in his liver enzymes.
So, perhaps not surprisingly, fast food turns out to be bad for our weight and liver. But what was interesting about this study is that the authors looked at the relationship between different elements of the diet and changes in ALT levels. In other words, they wanted to see if they could find out what it was about fast food that seemed to damaged the liver.
Here’s what they found:
Intake of FAT was NOT associated with ALT levels
Intake of PROTEIN was NOT associated with ALT levels
Total CALORIE INTAKE was NOT associated with ALT levels
Intake of CARBOHYDRATE WAS associated with ALT levels
Rather oddly, the authors conclude that this shows that raised liver enzymes can be caused by not just alcohol, but also sedentary behaviour and higher than usual caloric intake. I don’t know why the authors felt the need to draw conclusions that were not supported by their data: remember, it was not an excess of calories that was associated with deranged liver function, but an excess of carbohydrate.
Those of you who watched Supersize Me may remember that the doctor who was keeping a medical eye on Morgan Spurlock at one point told him that his diet was causing his liver to turn into foie gras. The evidence from this study suggests that it was not mere overindulgence nor a high intake of fat that was responsible for the fatty degeneration of Morgan’s liver, but a glut of carbohydrate in this diet.
Any of you wanting to remember that it’s carbs that cause fatty deposition in the liver can do this contemplating the making of foie gras. What is it that geese are force-fed to turn their livers into something that is mainly fat? The answer, of course, is grain.
References:
1. Kechagias S, et al. Fast-food-based hyper-alimentation can induce rapid and profound elevation of serum alanine aminotranferase in healthy subjects. Gut 2008 [epub Feb 14th]
I have just read the article in ” GUT” ( available free on their website)
Carbohydrate intake was correlated with rising ALT (a liver enzyme), not with increased fat in the liver, and started returning to normal before the subjects returned to normal diets. They postulate that the raised liver enzyme is not a sign of liver ill health but of enzyme induction by being presented with a large load. Increased fat in the liver which may be a sign of liver disease.was correlated with overall intake of calories,
I think it is a step too far to conclude from this study that carbohydrates alone were responsible for anything other than a clinically insignificant change in a liver enzyme.
Esther
On what basis do you conclude that the rise in ALT levels was ‘clinically insignificant’? The authors themselves state that ‘most participants developed PATHOLOGICAL [emphasis mine] ALT levels during study’
Just a bit of trivia concerning foie gras. Nowadays the geese are stuffed mostly with corn (maize) but historicly in roman times (yes the romans enjoyed already foie gras) figs were used for that purpose. The word foie in french derives from it.
I can’t help but wonder if it was the High Fructose Corn Syrup portion of the carbohydrate intake that was most responsible for the increase in ALT and liver fat, since fructose goes to the liver to be metabolized, unlike glucose. Many fast foods have HFCS as an ingredient: non-diet carbonated beverages, catsup, mayonnaise, etc. The content in a full meal could be quite high.
What alarms me even more is increase in use of agave syrup (or nectar) as a “low-glycemic” and “natural” sweetener by those wishing to avoid HFCS and refined table sugar. Very few seem to realize that agave syrup is as much as 92% fructose, depending on brand, which could put a terrific load on the liver if used liberally. I have seen a huge increase in the availability of bottled agave syrup/nectar and processed food products sweetened with agave syrup at the “natural” and “health” food stores. Additionally, lots of health-oriented bloggers have been indicating they use this “healthy” sweetener, perhaps liberally, since the the low-glycemic factor or “raw” claims give the appearance of being a better sweetener. But I think that claim is questionable, given the extremely high concentrated fructose factor. There is nothing natural about overloading the liver with concentrated fructose.
Does this leave me OK with my breakfast of organic porridge oats (made with goat’s milk), dried fruit, linseeds, a few pumpkin seeds and a dob of borage honey?
If you tell me this is appalling, I’m going to kill myself, if nothing because of the washing up, or the fact that I’d rather eat a bacon butty!! Jeez
Actually I feel great on the above…slow energy release and all that.
Excellent post Dr, as ever.
Over at Hyperlipid,
http://tinyurl.com/2dwf8l
Peter has pointed out that the other interesting thing about the study was that it found that healthy HDL cholesterol actually increased over the four-week period.
It is crazy how this was reported – all the newspapers have said that it proves that fat is bad for you when it says nothing of the sort.
Towards the end of my pregnancy they did some blood tests at the hospital and I had high liver function for a while and they never found out why.
I was on low GI diet before I got pregnant because I have pcos but the diet went out of the window throughout the pregnancy and I had a lot of sweet treats and bready stuff. Maybe thats why my ALTs were off the scale???
I am sure Dr Briffa is spot on. On a previous info reference a supressed report re. a cholesterol lowering product – opur GP rang my husband two weeks after I had read the whole report, to tell him to stop taking the drug. He already had.
There are established links between insulin resistance and a fatty liver and excess carbs can lead to insulin resistance. The problem is in establishing what is an excess of carbs and to what degree low GI carbs are ‘safe’.
You still hear dieticians recommending pasta as a key part of a healthy diet. I know someone morbidly obese whose dietician told her to eat pasta and she’s not lost any weight because she eats huge plates of white pasta with sugary processed sauces.
Hmm, can anyone translate post no. 8?
Farmers swear that producing Foie Gras isn’t cruel because the geese come running to have the grain poured down their throats. We’re doing the same thing as we enthusiastically stuff all those croissants and cornflakes and pizzas down. I’m guessing it’s down to the feel good sensations that carbs create by producing serotonin in the brain.
Apparently, geese don’t have a ‘gag reflex’ so the force feeding is not so forceful as it would be for a human. Still makes me shudder to look at it, but on the other hand the geese don’t seem too perturbed.
Tiggy, post 8 is from a Swedish lowcarb blog. The text basically says that dr Briffa has commented the study and links to here, nothing more exciting than that…
‘particularly those carbs that tend to cause most disruption in blood sugar and insulin levels including bread, potatoes, rice, pasta and breakfast cereals.’
In your blog you mention rice and pasta as high GI foods. According to Jennie Brand-Miller there are some types of rice, like japanese rice or basmati, that do not have a high GI and pasta does not have a high GI either. I agree that these foods should not form the basis of the diet in the way the ‘nutritional pyramid’ is presented nowadays, but I do not think it is correct to put all these carbs in the same group.
Simona
I’m actually not the bothered about the GI, it’s the glycaemic load (GL) that I believe is more important. And the reality is that because of the quantity in which they tend to be eaten, rice (even basmati, say) and pasta tend to have high GLs.
Plus, the nutritional value of a food is not just determined by a food’s GI/GL, but other things too including the nutrient density of the food. Here, grains (including rice and pasta) come up short, generally speaking, compared to, say, fruits and vegetables (other than the potato).
For these reasons, I believe elevating certain varieties of rice and pasta solely on the basis of their GIs would be quite misleading.
Does glucose have a GL? Or has that not been calculated?
Neil
Not to my knowledge. The GL depends on the size of a ‘portion’ of food, and people don’t eat ‘portions’ of glucose. So I suspect for these reasons, glucose doesn’t have a GL as such.
In terms of GI, Gary Taubes has a great chapter in his Good Calories, Bad Calories. Glucose goes straight into the blood (only 30-40% pass to the liver), causing a high insulin response (high glycemic) and is used as fuel; excess glucose is converted to fat and stored. Fructose passes directly to the liver to be metabolized instead, so has a low GI, but the liver converts it to triglycerides to be stored as fat. Thus while glucose leads to chronically high insulin levels (producing a myriad of problems, obviously), more fructose consumption leads to higher triglycerides in the blood and more fat storage. Further, fructose apparently blocks the metabolism of glucose in the liver as well as the synthesis of glucose into glycogen. Consequently, the pancreas secretes more insulin to overcome the excess glucose at the liver, and muscles thus compensate by becoming more insulin resistant.
So either way, insulin resistance is the result of a high-carb diet. There are no carbs that are “better” than others in my opinion. (Fiber has it’s own set of problems…)
As for pasta and rice, I don’t know anything about basmati or other types of supposedly healthier rice, but most starches are made up of long-chain glucose molecules that produce high insulin response. I’ve never heard of pasta that does not have high GI.
Also, I forgot to mention, anything that’s not glucose will lower the GI. That’s why ice ream, which is full of fat as well as sugar, has a much lower GI (36) than white bread (69) and white rice (72) among other starches.