A Belated Earthquake Comment

August 30, 2014

I completely forgot to mention on Thursday, to any reader who might have been concerned about it, that the earthquake last Sunday had no serious impact on me. The epicenter was a little south of Napa (in the next county over).

The shaking was certainly severe enough, and prolonged enough, to give me a good scare, but no damage occurred in my home, and I didn't even lose electricity or water. However, the nature of the wave motion (a kind of lazy sloshing back and forth) led me to conclude that the epicenter was a good distances away, so I knew it might be a more serious event in some other town. At 3:20 AM, it didn't occur to me to do any investigative journalism on the subject; I thought I'd check into it later. I went back to sleep. Later I awoke to a phone call from my sister in New England, and that's when I knew the quake had made the news.

A few of my coworkers live in Napa, but even there, the dangerously severe shaking was confined to a fairly small area and most of them were okay. One of them, though, had a harrowing story to tell me: the quake started out with the same kind of sloshing motion I had experienced, but in his neighborhood it soon escalated to a violent pounding motion, as if the house was being hit by a giant hammer. Parked cars moved two to four feet from their original positions (including his motor home, which leaped forward and banged into his house). His bed jumped away from the wall and ended up in the center of the room; other furniture was similarly tossed around and upended. He said he was glad he was in bed, and stayed in it, during the quake -- if he had stood up, he would certainly have been knocked down, and maybe struck by moving furniture. Downstairs, the refrigerator had migrated to the middle of the kitchen, and every single cup and glass was in pieces on the floor. No noticeable damage to the house itself, apart from where the motor home collided with it. No injuries to his family. Still, a nightmarish experience.

Your first thought is: why live in a part of the country where things like that happen? If by "things like that" you mean earthquakes, things like that can happen in more places than California (although California admittedly is a state especially prone to such problems). However, if "things like that" means "natural disasters that present a frightening risk to life and property", I'm afraid that sort of thing is not so easy to avoid. If you tried to make a map of the United States showing the regions where floods, fires, severe storms, and plagues of locusts don't happen, I don't think you would find that there was enough room on your map for all of us to live there.

I care about risk, but I think it is only worth thinking about risk if you view the problem realistically, as being about risk reduction rather than risk elimination. The former is useful, the latter impossible.

Fourth Thursday Update

August 28, 2014


The Maillard reaction: what's in it for you?

A lot of the diabetes-related questions that people type into search engines give me the strong impression that they are misunderstanding the issue they're inquiring about. For example: "how to flush glucose out from urine". If you have glucose in your urine, that isn't the real problem you need to solve.

Glucose in the urine is significant mostly because it is an indicator of another problem (too much glucose in the blood, some of which is passing into the urine because the kidneys are unable to re-absorb it all). "Flushing it out" of your urine is not the task before you; reducing it in your bloodstream, so that it never leaks into the urine in the first place, is where your attention needs to be focused. More important things are being threatened here than the quality of your urine.

Another example of this sort of thing came to my attention today: "what harm comes from the hemoglobin being glycated?". Presumably this was written by someone who found out that the hemoglobin A1c test, routinely given to diabetes patients, determines what percentage of the patient's hemoglobin (a crucial protein in red blood cells) is "glycated" (bonded with sugar). This person now wants to know the answer to the next, seemingly obvious question: why does it matter how much of your hemoglobin is glycated? What exactly is wrong with having glycated hemoglobin? How does it injure you?

The answer is a great deal more complicated than this person could possibly have expected, so settle in comfortably for one of my longer explanations.

In a way, asking what harm comes from having glycated hemoglobin in your blood is like asking what harm comes from having dead canaries in your coal mine. The reason for taking canaries into a coal mine in the first place is that canaries, owing to their rapid metabolism, will be affected by contaminated air before humans will. If the air in a mine is tainted with carbon monoxide, the canaries will be knocked out by it before the miners are, and then the miners will know that they need to evacuate the mine quickly before they, too, lose consciousness. The canaries are there to serve as an indictor of a problem which would otherwise be hard to detect.

Similarly, the hemoglobin A1c test is used by doctors not because they are specifically worried about your hemoglobin becoming glycated, but because the degree of glycation of your hemoglobin reveals a lot about how high your blood glucose levels have been, on average, over the past few months. The higher your glucose levels get over the course of an average day, the more glycation occurs, and the higher your test result will be.

Glycation is monitored by doctors primarily because it is a handy indicator of something else (average glucose levels). However, I'm not saying glycation itself is irrelevant, except as an indicator of average blood glucose. Glycation does matter in and of itself. Glycation in general, that is -- not just glycation of hemoglobin.

Glycation of hemoglobin is not necessarily more important than glycation of other proteins; elevated glycation of proteins in general is undesirable and potentially harmful. Hemoglobin just happens to be a protein that's easy to collect and measure, and has a fairly reliable mathematical relationship with average blood glucose. (Other glycation tests exist, which target different proteins, but for purposes of diabetes management the hemoglobin A1c test is usually the most useful.)

Glycation can be defined as the unwanted bonding of sugars to proteins. (The same reaction is also called glycosylation, but that latter word is most often used when the reaction is useful -- when it is accidental and undesirable, it is usually called glycation.)

There are other terms for this reaction, too. The most general term is "the Maillard reaction", which is named for a French chemist who first described it in 1912. To refer to it as a single reaction is a simplification, though. There are several varieties of sugar, and a staggering variety of proteins; each of these molecules has its own distinctive shape. Therefore, "bonding of sugars to proteins" is a categorical description that covers a lot of ground. The Maillard reaction refers to a large family of roughly similar glycation reactions (and those reactions trigger a cascade of secondary reactions, resulting in the formation of compounds known as AGEs: "advanced glycation end-products").

For reasons I have not been able to uncover, the AGEs tend to have a brown coloring. Maillard reactions are also known as "browning reactions", because they are responsible for the browning of foods during cooking (heat promotes bonding between the proteins and the sugars in the food). The AGEs ultimately produced by Maillard reactions play a huge role in giving foods their characteristic flavors (that is why artificial flavorings are often produced by means of Maillard reactions in the lab).

Much as we might appreciate what glycation and its resulting AGEs do to make cooked foods flavorful, we don't necessarily want to be carmelizing the living proteins within our own bodies. Anything which promotes Maillard reactions in living tissue has the potential to produce health problems of many kinds, because many crucial proteins are liable to function poorly in a sugar-coated form.

Proteins are the hand-tools of the biochemical world; they are able to perform their functions in living cells because they have distinctive shapes that let them latch onto other molecules and manipulate them. Change the shape of a protein, by welding a sugar molecule onto it somewhere, and it might no longer be able to do its job.

Furthermore, whatever AGEs are produced by secondary browning reactions might gum up the works inside the cell, and inhibit healthy functioning in any of thousand ways. For example: AGEs in the blood may promote inflammation of blood vessel walls -- which is a bit of a problem, because inflammation of blood vessel walls tends to promote the buildup of cholesterol plaques, which in turn promote cardiovascular disease.

Another potential problem: glycation of proteins in the lens of the eye can cause the lens to lose its transparency, resulting in cataracts. I'm sure doctors have not even begun to identify all of the potential health problems that could be triggered by glycation. Maybe it's enough to say that the reason diabetes produces so many "complications" is that there are so many ways for sugar-coated proteins to malfunction.

Of course, a certain amount of glycation goes on in even the healthiest person. We all have some glucose in our blood, and some of it gets bonded to our proteins every day. That is why the body recycles its proteins. A given type of protein in a given tissue gets replaced, at a certain fixed rate, by new (and unglycated) proteins synthesized within the body. Hemoglobin, for example, lasts about 3 months before being recycled (that is why the result of an A1c test is not influenced by conditions going back any further than that).

However, the body only replaces proteins fast enough to keep up with normal glycation rates. If the glycation rate is accelerated, because your blood glucose level is elevated (or because you're lying in a hot skillet), your proteins will be glycated faster than the recycling process can renew them. Your A1c result goes up, because a higher percentage of your hemoglobin is glycated. But not just your hemoglobin! If the Maillard reaction is browning your insides, then proteins in general are being glycated, and if that leads to health problems, the problems could take a great many different forms.

Because human body temperature is maintained within quite a narrow range (if it went up 5% you'd call it a very high fever), the glycation rate is controlled almost entirely by one factor: how much glucose is in your blood. That's what you can control, and that's why you want to control it. Allowing your glucose level to get well above the normal range is a little like climbing into the wok and stir-frying yourself for a while.

Perhaps it would even be useful to diabetes patients to think of the Maillard reaction, and its relation to blood glucose, as a kind of burning. We all burn ourselves from time to time, of course -- a little too much sun on the face now and again, or a little too much hasty consumption of a hot stew. These minor burns usually heal pretty quickly, but more severe ones are harder for the body's maintenance functions to repair -- and even minor burns can have a lasting impact if they happen too often. Anyway: regard an unusually high glucose spike as a kind of burn. Aim to keep your spikes infrequent and mild, so that your body can recover from them properly. Severe spikes, or moderate ones that happen too often, could leave you scarred.


More food for thought (or thought for food)

The people I work with often bring excess produce from their back-yard vegetable gardens into the office to share. That's how I was able to snag this parsnip today (at least I think that's what it is), which seems to belong to a genre of anthropomorphic vegetable photography which I had formerly assumed was always faked.

No photo-manipulation here, friends. That's what the thing looks like, all right.

Before I toss that into a stir-fry, though, I'm probably going to see if I can have it neutered. I prefer my side-dishes to seem a little more abstract than this. Cooked vegetables should not look as if they have a tragic story to share!

Third Thursday Update

August 21, 2014


Is it stress -- or something more?

Doctors find that a lot of non-diabetic patients, upon being admitted to a hospital, have elevated blood glucose, seemingly as a stress-response to whatever else is ailing them. Inflammation, due to infection or other stress on the body, tends to suppress insulin sensitivity, so anyone who is ill enough to be admitted to hospital may very well be experiencing elevated glucose. This phenomenon is called "stress hyperglycemia".

However! The fact that stress hyperglycemia exists does not make it the one and only explanation for elevated blood glucose in newly-admitted hospital patients. Some of those patients are exhibiting elevated blood glucose not because of stress, but because they're diabetic (or heading that way) and their doctors just don't happen to have identified the problem yet.

Undiagnosed diabetes is a pretty common condition. Because blood glucose is such a volatile quantity, taking an occasional snapshot of it doesn't go that far toward clarifying the situation. People who have excess blood glucose a lot of the time don't have it all the time. The problem can easily be missed by a fasting blood test given in connection with an annual physical. In fact, we should expect it to be missed routinely: a patient's fasting glucose level often remains normal for a long time after post-prandial levels have become seriously elevated. It is unrealistic to think of fasting tests as an early-warning indicator, when the fasting level is typically the last thing to go. Most people with diabetes were diabetic long before their doctors were able to collect any proof of it. Therefore, some percentage of the patients admitted to a hospital with "stress hyperglycemia" actually have a chronic problem with blood glucose, not a temporary one.

But what are the proportions here? How many of the supposedly non-diabetic patients admitted to a hospital with "stress hyperglycemia" actually have a more serious problem with glucose regulation, which will be recognized as diabetes a little further down the road? Some Scottish researchers decided to find out, by following up on former hospital patients from three years ago, to see what blood glucose levels they had upon hospital admission, and find out how many of them had been diagnosed with diabetes during the three years since then.

The results showed a very clear pattern: the higher a hospital patient's blood glucose was when admitted to the hospital, the higher the risk that patient would be diagnosed with diabetes within the next three years.

Here's a color-coded chart showing the results. The tables show age-categorized diabetes prevalence for male and female patients with different glucose levels upon hospital admission three years before. The higher the glucose level upon hospital admission, the higher the probability of being diagnosed with diabetes within three years:

Unfortunately, the glucose levels in the chart are expressed in mmol/l, which is a commonly used measurement unit in Europe but is unfamiliar to Americans. I doctored the chart below to show the glucose levels in mg/dL, which is the commonly used measurement unit in America.

As none of the risk levels are anywhere near 50%, it is clear that most of the hospital patients who show evidence of stress hyperglycemia really are suffering from stress hyperglycemia, rather than from not-yet-unrecognized diabetes.

However, the higher the patient's blood glucose is upon admission, the higher the probability is that the patient will be diagnosed as diabetic within a few years.

This is what makes diabetes so hard to define and so hard to deal with: we want to think of it as an absolute (either you have diabetes or you don't, and everybody who does have it is in the same situation), when the reality of the disease is considerably more vague and shadowy. It's hard to say who has diabetes and who doesn't. It's even harder to say how one person's experience of the disease will compare to someone else's.

I think of diabetes as being like a rainbow, in that it's a strictly individual thing. No two people can see the same rainbow, because of the way the optics of the phenomenon work. Two people looking at what they assume to be the same rainbow are, in fact, having entirely separate visual experiences, because their viewing angles differ, and rainbows are all about viewing angles. In the case of the rainbow, the two experiences won't be very different, at least if the two observers are located near one another. But in the case of diabetes, the difference between my experience of it and your experience of it not only can be very different, but probably will be very different. No two people have exactly the same diabetes, or develop it in exactly the same way.

To some people that might seem like a negative thing, but I see it in a more hopeful light: if my experience of diabetes doesn't have to be like someone else's, than maybe I can do better with it than they do. A lot of people with diabetes get depressed because of some horror story they have heard about somebody else's diabetes -- which they assume will have to be their story, too.

It ain't necessarily so!


Second Thursday Update

August 14, 2014

I thought my fasting number would be lower today, and I felt a little bit cheated that it wasn't. But there's a limit to how fully we can control such things, even when we think we're doing everything right. At least my post-prandial number was low -- and that was before I exercised.

I had established a pattern of going to the state park after work on Thursdays, but tonight I was prevented by a late meeting that didn't start till 6 PM. But I managed to get out for a 5.4-miler before the meeting. Which was good, because a lot of what got discussed at the meeting had great potential to stress me out; the lingering endorphins from the run made it possible for me to deal with it calmly.


Welcome to the club!

Every kind of disease is isolating to some degree, but a lot of people find diabetes especially so. There is no outward sign that you have any kind of serious health issue, so you mix with everyone normally, and they expect you to do things the way they do -- but you can't.

Managing the disease well requires you to run your life differently than other people do. Often they don't want you to do that. Often you don't want to do that. But trying to explain any of this to them is usually futile, so you just have to resign yourself to feeling (and seeming) "different" in an embarrassing way. (Unless, that is, you resign yourself to not managing the disease well -- a choice which simplifies life in many ways, but also leads to, uhm, complications.)

But take heart! I don't know that this is exactly what I would call a silver lining, but people who feel isolated by their diabetes can look forward to having a lot of company. The latest estimate from the Centers for Disease Control says that about 40% of Americans can expect to become diabetic at some point in their lives. Apparently this is a big step up from previous estimates, but previous estimates were based on diabetes prevalence as it stood in the 1990s -- and since then, diabetes prevalence has not stood still. The diagnosis rate has been going up steadily. Lead author Edward Gregg is quoted as follows: "We have found that over the past few decades the chance that the average person will develop diabetes at some point in their lifetime has increased tremendously".

Naturally this raises questions about why there should be a tremendous increase in the prevalence of a non-infectious disease. Imagine how aggressively the situation would be investigated (and covered by the news media) if the prevalence of cancer had increased tremendously over the past few decades!

The Usual Suspect in this case is obesity, of course. Dr. Gregg continues: "Obesity really has been the driving force. There is some encouraging evidence that obesity trends might be flattening, and we hope that will have an effect on diabetes, which follows very closely what happens in obesity." Other suspects involve dietary changes (increased restaurant dining, expanded portion sizes, sugared drinks, refined carbs) and also declining physical activity levels. Whether these things play a direct role in causing diabetes, or merely contribute to the rising obesity rate, is far from settled. Every conceivable hypothesis about what is causing the trend toward increased diabetes prevalence has passionate supporters. Maybe most of them are right to a degree. And, as I've often said before, diabetes is the disease with a thousand causes. The body's glucose-regulation system is extraordinarily vulnerable, and can fail in so many different ways that it's terribly difficult to pin down the cause of an increase in prevalence of the disease.

The 40% lifetime diabetes risk is only an average, by the way. Some demographic segments face a higher risk than that. The risk is over 50% for Hispanic men and non-Hispanic black women. So, depending on your family background, you might actually find yourself in the majority within your own demographic as a diabetes patient. I'm not suggesting that will be a thrill for anybody, but it might make people feel slightly less marginalized by the disease.


Another diabetes cause!

No sooner had I written my comments above, about diabetes having a thousand causes, than I chanced upon another reported causes of diabetes: work stress. German researchers report that stress in the workplace increases your risk of diabetes by 45%.

I don't know what we do with that information exactly. But there it is. I guess my own take-away message from the research is: if you can go for a run before your big stressful meeting at 6 o'clock, do it!


First Thursday Update

August 7, 2014


Health news: the cynic's guide!

It always pays to read the fine print when health news comes your way. These days, health news often comes our way in the form of a link with an intriguing headline, which we click on to read whatever online article it leads to. But reading the fine print of that article, nowadays, often means following links to some other place where the less savory details of the story can be found.

Today I saw a link that read "New Diabetes Drug Has Surprising Health Benefits". I know click-bait when I see it, but I'm not above clicking on it. So I did, and ended up here. It's a summary article about research reported elsewhere. The headline of the article does not match the one I clicked on originally, but it's equally intriguing: "Type 2 diabetics can live longer than people without the disease". But what about the "New Diabetes Drug" the earlier link referred to? What is this new drug?

The drug in question turns out to be metformin -- a drug which, on any time scale other than a geologic one, is not remotely "new". But forget about how new or old the drug is, the claim reported here is that a large scale study of 180,000 people has shown that Type 2 patients treated with metformin can live longer than people without diabetes (doctors normally expect diabetes to shorten your life by 8 years on average). The drug seems to have benefits to health which may benefit even people who don't have diabetes. The article also pointedly compares metformin to sulfonylurea drugs, which have more serious side-effects than metformin, and don't seem to provide the general health benefits that metformin does. Take that, non-metformin pharmaceuticals!

Because the article is such a love-letter to metformin, I wondered who payed for the research. The article didn't say. Time to read the fine print, clearly!

Fortunately, the article did link to another report on the research, and this one did indicate the funding source: "The study was funded by Bristol Myers-Squibb". Hey, that's a pharmaceutical company! You don't suppose they make metformin, do you? Time to read more fine print, which in this case involved Googling "who makes metformin?", and getting the extremely unsurprising answer that Bristol Myers-Squibb does.

Science funded by organizations with a vested interest in getting a particular result is not necessarily wrong, but it's not necessarily science, either. Systematically eliminating sources of bias is what makes science science. So far as I can see, science is the only human endeavor in which what is actually true is more important than what we would like to be true. Criminal law, for example, is focused on what can be made to sound true to a jury; what is actually true is of no interest to the lawyers on either side. The same is true of political argument: "is it good for our side?" is the question we ask, not "is it true?". Only in science does reality matter more than what we would like reality to be. But that kind of focus on reality is possible only when the researchers are not under pressure to focus instead on what the people who are paying for the research would like reality to be.

Maybe the results of this research are perfectly valid, but I'll believe that when it's confirmed by scientists whose livelihoods do not depend on keeping Bristol Myers-Squibb happy. I don't think we can afford to take any other attitude toward research that is payed for by parties who stand to benefit if the answer comes out a certain way.

I followed similar trails of evidence today in the case of two different studies which found that eating pistachios provided health benefits to Type 2 diabetes patients. As I happen to be extremely fond of pistachios, I was certainly ready to welcome evidence that eating these things counted as a virtue and not a vice. One of the studies found that regular pistachio consumption improved control of blood sugar; the other found that it improved markers of cardiovascular risk (including resting blood pressure and constriction of blood vessels as a response to stress).

Who funded this research? Two organizations representing the interests of people who grow pistachios.

As I say, I really like pistachios. I would like nothing better than to have evidence that eating pistachios regularly will help regulate my blood sugar and blood pressure. But evidence isn't quite the same when it's gathered by people whose paychecks are funded by the "American Pistachio Growers" and the "International Tree Nut Council Nutrition Research & Education Foundation".

I would also like to believe the study which found that consuming a whey-protein supplement before a meal reduces the blood sugar spike after the meal. But I'd feel a lot more confident if the Milk Council had not been involved in funding the research.

Look, I don't like doubting health news that seems encouraging, just because the funding setup creates an obvious conflict of interest. But when a conflict of interest is clearly there, it has to be taken into account. Because so much research is commercially funded, there's a lot of conflict to be taken into account.

Some would argue that, if commercially funded research didn't happen, there wouldn't be nearly as much research done. That's true. But, of the research that was left, a lot more of it would be credible.


Older Posts:

July 2014
June 2014
May 2014
April 2014
March 2014
February 2014
January 2014
December 2013
November 2013
October 2013
September 2013
August 2013
July 2013
June 2013
May 2013
April 2013
March 2013
February 2013
January 2013
December 2012
November 2012
October 2012
September 2012
July 2012
June 2012
May 2012
April 2012
March 2012
February 2012
January 2012
December 2011
November 2011
October 2011
September 2011
August 2011
July 2011
June 2011
May 2011
April 2011
March 2011
February 2011
January 2011
December 2010
November 2010
October 2010
September 2010
August 2010
July 2010
June 2010
May 2010
April 2010
March 2010
February 2010
January 2010
December 2009
November 2009
October 2009
September 2009
August 2009
July 2009
June 2009
May 2009
April 2009
March 2009
February 2009
January 2009
December 2008
November 2008
October 2008
September 2008
August 2008
July 2008
June 2008
May 2008
April 2008
March 2008
Jan/Feb 2008