More from Japan

I was googling for an idea of what the current value is for LDL in the general population of the USA at the moment, when I accidentally hit on this paper. It's from the Japan cohort of the infamous Seven Countries Study, focusing on the farming town of Tanushimaru. Apart from the data presented, which are fascinating, there are the data not presented, which say a great deal more. What is also fascinating is the fossilised mindset of the investigators, inheriting from Ancel Keys the ability to look at, and in this case publish, data which destroy the lipid hypothesis as proposed by Keys so many years ago, but ignore what they have found. Lets look at some results tables.

The pdf is copy protected so if you would like the actual data just download the pdf and have a look-see.

I'll start with line one of Table 1, energy intake. This has fallen significantly between 1958 and 1999. Over that period BMI has risen from (Table 3) 21.7 to 23.7. The paper talks about less manual labour etc, suggesting BMI is some simplistic marker of calories in vs calories out. Duh. OK, people on the 1999 diet are storing more energy than they can access, compared to the massive caloric intake in 1958, which was being accessed at a rate which maintained a lower BMI. This to me suggests that the population in 1999 has a higher average insulin level despite lower carbohydrate intake. Think metabolic syndrome, not calories in vs calories out.

You would expect people with lower carbohydrate intake to have lower insulin levels, but anyone who follows Stephan's blog or any of the Kitava posts here will realise that very high carbohydrate diets per se are not the problem. The problem is failure to maintain efficient glucose usage at physiological concentrations of insulin. Something happened between 1959 and 1999 to increase insulin levels despite lowered carbohydrate intake. This means insulin resistance.

So the next thought is; what question do the investigators not ask, or at least not tell us so, if they did?

Table 1 tells us that in 1958 the farmers were eating 2837kcal/d, 84% of which was carbohydrate, ie 2383kcal of carbs. Flicking to table 2 we can see they were eating 593g/d of rice, ie 2668kcal of rice!!!!!!!!!!!!!!!! OK, they were eating more rice calories than carbohydrate calories! The numbers don't quite balance but this is not banking (jk) we're talking here. Assessing dietary intakes and the associated calories is not hard science and these researchers are not exactly famous for precision. Essentially all of this caloric intake was starch. I don't see a lot of scope for fructose intake when you are eating more rice calories than carbohydrate calories!

The same numbers in 1999 were 1365kcal as carbs and only 1062kcal of rice. There is now a deficit of 300kcal of carbs, ie people are now eating non-rice carbohydrate. The authors didn't comment on this. I don't suppose anyone who considers saturated fat to be the reason for elevated cholesterol would consider fructose, probably via sucrose, to have any relevance to cholesterol levels.

My guess is that 8 teaspoons a day of sugar is enough to both do some glycation of LDL and to put enough fat in to the liver to raise hepatic insulin resistance. Probably not enough to increase heart attack risk, just enough to raise LDL cholesterol. Not enough to cause hyperglycaemia. Just looking at the numbers needing hypertension meds (see below) I would expect HbA1c to have begun to rise after 1982. Of course sons of Keys would never think to look at HbA1c evels. Too suggestive of Prof Yudkin's ideas!

Is there any other support for the idea of progressively increasing insulin resistance? Back in table 3 we can see that BP is remarkably stable but there is a sudden increase in the percentage of the population needing hypertension medication to achieve this, from 7% in 1989 to 20% in 1999. The roll of elevated insulin in hypertension is not particularly contentious. If you had to say anything about overall health this line tells me that some sort of threshold was crossed in the 1990s. So where are the heart attacks?

Smoking started to fall around 1980, from around 70% of adults (Kitavan levels!) to 45% in 1999. There was zero drop in heart attack rate with this fall. Why not? Perhaps replacing nicotine with fructose is a balanced trade off!

Now, just to finish, TC levels skyrocketed from 152mg/dl in 1958 to 194mg/dl in 1999 and there was no effect on the incidence of coronary heart disease.

How do the jokers running this cohort view their data, which destroy the hypothesis on which their jobs depend?

"In conclusion, large changes in dietary patterns and remarkable changes in serum cholesterol levels among men aged 40-64 years in a Japanese farming area were demonstrated. Fortunately, incidence of coronary heart disease has not increased in our cohort for a couple of decades. The varied composition of the Japanese diet has probably prevented coronary heart disease. However, careful surveillance is needed in the future because of the increasing intake of fat, especially saturated fatty acids, with the potential of a modern epidemic of coronary heart disease in Japan."

The end.

Cholesterol skyrockets, CHD doesn't. A paradox. Unless Keys was wrong and his "offspring" are still wrong.

Oh, or the "varied composition" of the Japanese diet, of which 1000kcal/d is white rice, saves them (giggle, hysterical) from CHD!

Peter