New Theory about Saturated Fat

We all know that dietary saturated fat is bad because it raises blood cholesterol levels, right?

Well, maybe. 

When we get lipid tests like LDL, HDL, and so forth, they draw blood and then measure these things in the blood. They don't measure them inside the cells, or in the cell membranes. And a new study suggests that high blood cholesterol might simply reflect the membrane's need for cholesterol.

Unsaturated fats make membranes more fluid; saturated fats and cholesterol stiffen membranes. What we need is a balance between the two states, and the body is usually pretty good at knowing what we need.

The new theory suggests that when we eat saturated fats instead of unsaturated fats, the membranes don't need a lot of cholesterol to prevent the membranes from being too fluid, so they don't take the cholesterol out of the blood, and hence blood cholesterol levels are higher.

In other words, blood cholesterol levels can fluctuate according to how much cholesterol is needed in membranes. "The effect of dietary fats on blood cholesterol is not a pathogenic response, but rather a completely normal and even healthy adaptation to changes in diet" says the lead author of the study.

The authors call this the homeoviscous adaptation to dietary lipids model. They note that as of now, this is only a theory that needs to be verified. And they distingish between elevated cholesterol levels from dietary changes and elevated cholesterol levels from metabolic disturbances such as inflammation and insulin resistance.

You can read the abstract of the article here. Unfortunately, the full text is behind a paywall. 

Subtypes of Type 2 and Prediabetes

 A recent study has suggested that there are six subtypes of prediabetes, with each subtype having different risks of progressing to type 2 diabetes and different risks of various side effects. The study, which was begun 25 years ago, grouped people into clusters depending on factors like blood glucose levels, liver fat, body fat distribution, blood lipid levels, and genetic risks.

They found that people in different clusters differed in insulin secretion and insulin action in addition to the factors listed above.

People in three of the clusters have a low risk of diabetes. People in another cluster produce too little insulin. Those in another cluster have kidney damage even before overt diabetes is diagnosed.

It would be nice to know what cluster a newly diagnosed patent was in, but many patients today have difficulty getting even basic tests, and its unlikely they could get all the tests required to classify them. Even the authors concede that "our clustering aproach is not designed to provide definitive subphenotypes for individual patients in a clinical seting." However it would be useful for researchers.

I suspect that in the future they'll find even more subtypes, but for now this is a start.

Patients can differ a lot in both physical characteristics and economic and emotional ones. One patient might be willing to go on a strict low-carb diet, and another couldn't tolerate that, or couldn't afford it. Treatment may depend on a particular patient's situation, and good doctors take that into account agaialready, rather than relying on a cookie cutter approach.

The full text of this study, in preprint form, can be found here. 

This is not the first time researchers have tried grouping patients into clusters. Other researchers have grouped people with type 2 diabetes into five subgroups, using GAD autoantibodies, age at diabetes onset, HbA_1c, BMI, and measures of insulin resistance and insulin secretion.

They found that the group with severe insulin-deficient diabetes had increased risk of retinopathy and neuropathy, whereas the severe insulin-resistant diabetes group had the highest risk for diabetic kidney disease and fatty liver

Again, most physicians will probably not have the resources to measure all these parameters for every patient, but again, it's a start.