Fifth Thursday Update

July 31, 2014


I'm glad I don't live in one of those places where they have weather...

That threat of partial cloudiness next Monday is apparently going to be out weather-drama for the week. I hope we can get through it untraumatized!

 


Disease with a thousand causes

As I've said before, a new causal mechanism for diabetes seems to be discovered every week or so. Here's another one, right on schedule!

The publication formally called the Proceedings of the National Academy of Sciences (which, bravely and of its own free will, also goes by the highly inadvisable acronym PNAS), reports here on some Yale research into a hitherto-unknown cause of diabetes. Or perhaps I should say, a hitherto-unknown way of causing diabetes artificially -- because I'm not sure this ever happens outside the laboratory.

The Yale scientists were studying an enzyme called prolyl endopeptidase (PREP to its friends), which is produced in an out-of-the-way corner of the human brain called the ventromedial nucleus of the hypothalamus (VMH to its friends). What is the purpose of the PREP enzyme, and what happens if the VMH stops producing it?

As usual these days, the researchers figured this out by means of "knockout" genetic engineering: they created a population of laboratory mice which were unable to produce the PREP enzyme.

It turns out that the PREP enzyme controls the sensitivity of brain cells which are supposed to act as glucose sensors. When blood sugar goes up, these sensor cells should detect the increase, and signal the pancreas to start producing more insulin to bring the glucose level down. However, in the mice that couldn't produce the PREP enzyme, these glucose-sensors in the brain were insensitive to rising glucose levels, and did not send the appropriate signal to the pancreas. Result: insufficient insulin production, resulting in diabetes. The same result, without any genetic engineering, is also produced by treating the mice with a drug which inhibits production of the PREP enzyme.

What's not clear from the three reports I've read on this research is whether or not this ever happens in nature. Are people with diabetes, or at least some people with diabetes, known to produce an abnormally low amount of the PREP enzyme? If so, then this is relevant. If not, then this is somebody monkeying around in the lab because it's cool.

The glucose-regulation system must be the most vulnerable mechanism in the human body. There seem to be a thousand vulnerabilities in that system -- a thousand ways for something to go wrong which can result in diabetes. With so many potential failure modes to chose from, I hope researchers will try to focus on the ones that actually occur in real life. A lot of things can be made to happen in the lab; it's the things that also happen outside the lab that are of primary interest here.

 


Low-carb dieting and diabetes

Two different studies have validated the seemingly obvious idea that limiting carbohydrate intake is a good idea for diabetes patients.

A report from the University of Alabama at Birmingham concludes that carbohydrate restriction has many benefits as a first response to diabetes:

  • Such diets reliably reduce high blood glucose, the most salient feature of diabetes.
  • Benefits do not require weight loss although nothing is better for weight reduction.
  • Carbohydrate-restricted diets reduce or eliminate medication.
  • There are no side effects comparable to those seen in intensive treatment with drugs.

Another report from Australian researchers looked at two diets -- one low in carbs and low in saturated fats, and one that was low-fat but high in unrefined carbs (those famous high-fiber foods that are supposedly helpful for diabetes management). It turned out that both diets had some benefits for both glycemic control and control of cardiovascular risk factors, but the low-carb approach was the better of the two.

Carbohydrates are the main dietary factor that drives up blood glucose during digestion of a meal. Protein and fat have much less impact -- not because the body, crafty devil that it is, can't make sugar out of them, but because it does so very slowly. The conversion process is too slow, in the case of protein or fat, to produce that sudden surge of blood sugar known as a "spike". Sugar or starch, though, can spike you like nobody's business.

For a while there, when low-carb diets for weight loss were a huge fad, life was easier for diabetes patients: restaurants, including fast-food restaurants, were offering all sorts of low-carb options. That's over, now. The biggest problem facing diabetes patients trying to restrict their carbohydrate intake is that most of the foods that are ever going to be offered to them, in any situation, are high-carb foods. Corn and rice and sugar have to be smuggled into everything. Trying to limit your carbs, in today's world, is like going out for a stroll in a rainstorm and trying not to get wet. If we could change that situation, diabetes would be a lot easier to manage.

We may have to add "modern life" to the ever-growing list of causal mechanisms for diabetes. Certainly this one can be observed outside the laboratory, without genetically altering a mouse.

 


Fourth Thursday Update

July 24, 2014

It was a hot day, and it was still hot by the time I got to the state park in the evening for a trail-run. I wondered if maybe this run wasn't such a great idea, especially as my intended route involved a 3-mile stretch that is not only continuously uphill, but becomes gradually steeper as it approaches the top.

And as if that wasn't enough to make me feel old and feeble, the high school and college track teams were training in the park...

...and so were the hardy young mountain-bikers.

On the plus side, however, I knew that my chosen route would enable me to spend a great deal of time in the shade.

In a region with high humidity, getting out into the shady woods might not have offered me much relief from the summer heat, but here in California, the air is usually dry and getting out of the sun makes a big difference.

And by the time I started emerging back into the sunlight, the heat of the day was largely gone.

It wasn't the easiest run I've ever done, but it wasn't that tough, and I've got a pretty good endorphin rush going on right now, so I'm glad I went.

 


Are doctors turning against pre-D?

The medical profession seems to be drifting in the direction of declaring "prediabetes" an unhelpful diagnostic concept, and an idea whose time has gone. Like Dr. Frankenstein, they are starting to regret having conjured the thing into existence in the first place.

I, too, have concerns about the prediabetes concept -- but not necessarily the concerns that doctors have about it. To put it simply (perhaps too simply), it's the pre in "prediabetes" that I have a problem with, and it's the diabetes in "prediabetes" that doctors have a problem with.

My view is that the situation we call prediabetes is an actual condition, with an actual impact on health in the short term. To call that problem pre-diabetes, or pre-anything-else, implies that it is only a potential problem, with a potential impact on health in the distant future. By naming it to suggest that it refers to a time before diabetes happens, doctors are inevitably suggesting that nothing has happened so far. It's a little like referring to the first trimester as prepregnancy: the clear implication is that "prepregnant" means "not yet pregnant". The clear implication is that being called "prediabetic" means you're not diabetic. Diabetes is not an issue for you. It might become an issue some day, but in the meantime, forget about it.

Well, that's my view, but doctors perceive a different problem with the prediabetes concept. Their view is that calling the condition which typically precedes a diabetes diagnosis "prediabetes" implies that it will necessarily progress to diabetes -- when in fact that might or might not happen. Being prediabetic puts you at risk of developing diabetes later, but that "later" may come only after a lot of time has passed. (According to the Centers for Disease Control, less than a third of people are diagnosed as diabetic within the first five years after being diagnosed as prediabetic.) And if prediabetes is going to progress to diabetes, it will happen even if you try to prevent it -- treating prediabetes patients for diabetes, as if they were already diabetes patients, delays the onset of diabetes, but only for a few years; in the long run it doesn't appear to help. What's the point of diagnosing someone as prediabetic if doing so doesn't produce any important health benefits?

That, at least, is the view of researchers at the University College of London. Professor Yudkin of the University says, "Pre-diabetes is an artificial category with virtually zero clinical relevance. There is no proven benefit of giving diabetes treatment drugs to people in this category before they develop diabetes, particularly since many of them would not go on to develop diabetes anyway."

Notice the casual assumption buried in that comment: it's useless to identify a condition known as prediabetes, because there are no drugs for it that make a difference. The diagnosis has "zero clinical relevance" because doctors can't do anything about it that will work.

Part of the reason I can't agree with the medical profession's viewpoint on this issue is that, to me, "what can be done about diabetes" means "what the docrtor or the patient can do about diabetes". To a doctor, "what can be done about diabetes" means "what the doctor can do about diabetes" -- and what doctors can do about it is frustratingly limited.

It has been said that a movie-studio executives have about as much control over a director during shooting as a hospital administration does over a surgeon during an operation: the people who are nominally supervising what's going on are not even there to see it most of the time, and they certainly are not in direct control. It's more or less the same way for a doctor treating a diabetes patient: most of what is done (or not done) about the disease will be done (or not done) by the patient while the doctor is not around. Doctors can offer guidance, and write prescriptions, but the actual treatment of the disease is going to be done by the patient if it's done at all.

And yet, doctors have no choice but to regard what they themselves are doing as the "treatment", and the effectiveness of that treatment is evaluated by results, even if those results reflect the patient's choices rather than the doctor's.

That is why the medical profession assesses exercise as diabetes therapy not by looking at whether exercise works, but by looking at whether asking people to exercise works. A particular type of dietary recommendation is evaluated not by looking at whether the diet works, but by looking at whether recommending it works.

Patients can do a lot about diabetes (or prediabetes) if they choose to, but because so many of them will choose not to, asking them to do what's necessary is ineffective: it has "virtually zero clinical relevance". Asking them to take pills is more likely to get results (not great results, but results), so that is what counts as treatment. If the treatment doesn't make much difference in the case of prediabetes, why not just eliminate the diagnosis altogether, and act as if everything's fine up to the minute that you fail a diabetes test?

I have a slight problem with the notion that the way to deal with prediabetes is not to define it better and explain it better, but to pretend that there is no such thing.

Prediabetes, no matter how badly named it may be, and no matter how variously people interpret the term (some assume it means "you're doomed" and some assume it means "nothing to worry about so far"), is an indication of trouble brewing. Blood sugar levels do not start drifting above the upper limit of the normal range because everything is okay. They start doing that because your glycemic regulatory system is trying to compensate for a problem and not quite succeeding.

The trouble is that the compensation itself, even if it "works", can be harmful. "Compensatory hyperinsulinemia" (excessive production of insulin to correct for a loss of sensitivity to the stuff) keeps blood sugar under reasonable control, for a while anyway, but it also tends to take a toll on the arteries. This "metabolic syndrome" has been identified as a cause of heart attacks even in people whose blood sugar appears, so far, to be normal. Giving people diabetes drugs before they are diabetic apparently does not solve the problem, but lifestyle changes designed to boost insulin sensitivity can make a big difference. Unfortunately, those lifestyle changes only make a difference if you actually make them. Being asked to make them doesn't make a difference, so it has "virtually zero clinical relevance".

I think prediabetes is a significant issue which suffers from being poorly understood and poorly named. We need to fix the name first, and then go to work on helping people understand what it means.

Maybe the right approach is contract the five-syllable term "prediabetes" to something shorter and punchier.

"Prius" might work. Do suppose that's taken?

 


Third Thursday Update

July 17, 2014

 


Of mice and men

I'm not sure that being a mouse is an advantage in any other way, but it sure makes diabetes easier to treat. Researchers are constantly discovering amazing new diabetes therapies that work on mice. Finding a way to make them work on humans usually turns out to be trickier, though.

We must learn not to get over-excited when it is reported that a potential diabetes therapy does wonders for tiny rodents in cages, because that report may turn out to be the last thing you will ever hear about that particular therapy. Perhaps it will be quietly abandoned somewhere along the way, when it turns out not to be effective in humans, or when it turns out to cause pancreatic tumors, arterial inflammation, or suicidal depression. It's hard to mess with human metabolism without creating unintended consequences somewhere in the system.

So, don't make too much of this, but researchers at the Salk Institute are reporting that the mouse equivalent of Type 2 diabetes can be "stopped in its tracks" (for two days or more) by a single injection of a protein called FGF1. In the treated mice, blood sugar was restored to a healthy range -- and with repeated injections, insulin sensitivity was similar restored. And this was accomplished without the side effects (such as weight gain and hypoglycemia) which are associated with many diabetes drugs.

Why did the researchers think injecting FGF1 might be effective against diabetes? Because of an accidental discovery from "knockout" research. Genetic engineering is good enough these days that scientists can produce mice in the lab which cannot produce this or that protein because the gene that codes for it has been suppressed ("knocked out"). Often the purpose of a protein is discovered by suppressing its production in a mouse to see what happens to the creature with that protein missing. It turned out that mice without FGF1 became diabetic. It seemed a reasonable guess that, if mice become diabetic when they lack FGF1, then giving diabetic mice an abundance of it might be a useful therapy. And this guess turned out to be right.

Pessimistic as usual, I'm immediately wondering what the catch is. Well, the obvious one is that FGF1, being a protein and therefore broken down by digestion, can only be injected, not taken orally. Plenty of people would see that as a minor problem, especially if you can go a couple of days between shots, but some might see it as a deal-breaker. And I see a more serious area of concern: FGF1 is a growth factor -- one of a family of proteins which regulate cell growth, cell division, and cell life cycles. Whenever a drug aims to intrude into those areas, you have to worry about whether the drug might turn out to promote cancer. The researchers aren't reporting any evidence of that, but it's going to take a lot of evidence from clinical trials to eliminate that as a concern.

The mechanism by which FGF1 works its magic has not yet been clarified, but some of its effects are known. This photograph shows liver cells in a diabetic mouse not treated with FGF1. Notice all those white balloon-like calls: those cells have absorbed a lot of fat. But they're not very willing to absorb sugar.

Here are liver cells of a mouse treated with FGF1; there are fewer of the pale, fat-swollen cells, and more small dark cells (which are willing to absorb sugar).

I have no idea if anything really useful will come of this research, but obviously it is promising enough that it needs to be followed up on.

Even if it works as well as it now seems to, FGF1 faces an uphill battle for acceptance: drugs that have to be injected have a hard time winning friends.

 


Learning from car trouble

Between last Saturday night (when I was out of town) and the following Monday morning (when I was back home) I had four visits from AAA tow-truck drivers, and by the time the whole mystery of what was wrong with my car was solved, I had been given four very different views of what was going on, ranging from "there's nothing wrong with your car, you just need a jump start" to "there's a lot more wrong with your car than a dead battery, and you need to be towed to a garage because it's not safe to drive it".

The fourth driver knew something that the others didn't: there is a common failure mode in Honda Civics, a problem which is actually located in the compressor switch for the air conditioner, but has the effect of draining the battery. The battery-draining effect occurs whether the air conditioner is used or not. Once I was able to get the car to a Honda mechanic who also knew about this problem, I was finally able to get my car back into a state of reliability. A series of jump-starts might have kept my car limping along for a good while, even if I did nothing about the problem in the air conditioner, but obviously that's no way to operate.

It seemed to me like a good metaphor for health problems.

It is very tempting to define your health problem as whatever symptom it is causing you to suffer -- and then to seek a solution which relieves (dare I say "hides"?) that symptom. The trouble with this approach is that you aren't addressing the real problem. You're dealing with the symptom, but not dealing with whatever dysfunction within the body is causing the symptom.

If you view your battery-drainage problem as a battery problem and nothing else, then it's hard to see what you can do about it other than get a jump-start every time you want to start the engine. However, if you expand your view of the problem to include whatever is causing the battery to drain, then you can actually find a solution -- at least, you can find it once you figure out that the compressor switch in the air conditioner is the underlying issue.

Most approaches to diabetes treatment assume that the real problem is high blood sugar (not whatever is driving it), and address that problem by trying to force blood sugar down by means of temporary chemical interventions. In other words, most approaches to diabetes treatment amount to a series of jump-starts. I'm trying to take a different approach -- in which something is done about the underlying problem hidden within the air-conditioner. I can't just buy a replacement compressor-switch, of course. I guess no metaphor is perfect!

 


Second Thursday Update

July 10, 2014

 


A Neuropathy Controversy

Like a surprising number of other medical terms, "neuropathy" is very vaguely defined. It pretty much means "something wrong with nerves". Neuropathy is commonly a complication of diabetes, and in that case it is usually called "peripheral neuropathy", which doesn't add much to the meaning, except for a suggestion that the nerve problem is found in the extremities (especially below the knee) or is distributed across various parts of the body (sometimes this is called polyneuropathy rather than peripheral neuropathy).

When doctors talk about peripheral neuropathy they are most often talking about degenerative nerve damage associated with diabetes. The two primary symptoms of the condition are, somewhat paradoxically, a sensation of pain that is not associated with any real injury, combined with insensitivity to pain which an actual injury should be causing. Neuropathy patients have to endure the sensation of being pricked by imaginary needles, and yet when they have an actual wound they are dangerously unaware of it. The worst of both worlds, in terms of pain sensitivity: you hurt when nothing's wrong, and you don't hurt when something is. Pain no longer serves its useful function of warning you to get that rock our of your shoe; it only serves the useless function of ruining your day.

People living with neuropathy have to deal with unpleasant symptoms caused directly by it, and they also have to deal with tissue damage (potentially leading to amputations) which can result indirectly from not realizing you have an injury. Obviously no diabetes patient wants to develop neuropathy, and avoiding that problem is one of the better motives for maintaining good glycemic control. But how good does your control actually have to be? We usually associate neuropathy with diabetes which is poorly controlled over a long period. Is there any indication that it can develop in patients who are merely "prediabetic", or whose diabetes is well-controlled?

There seems to be a dispute going on regarding this question. The opposing sides are represented, at least in the article I read, by Omar Asghar (University of Manchester) and Peter J. Dyck (Mayo Clinic).

Who is right, Omar Asghar or Peter J. Dyck? I think it's clear enough that Dr. Dyck has the funnier of the two names, but perhaps that's not a sound basis on which to decide the question. Perhaps I should consider their competing claims.

Asghar's contention is that, if you use sufficiently sophisticated tools (including a microscopic examination of the eye known as corneal confocal microscopy), you can find evidence of neuropathy in patients who have not even been diagnosed with diabetes yet, but do have "impaired glucose tolerance". The 37 patients in Asghar's study had results on a 2-hour oral glucose tolerance test of 140-200 mg/dL, which is considered abnormally high but not high enough for a diabetes diagnosis. And 40% of these patients showed subtle signs of neuropathy (specifically, reduced density of nerve fibers).

Not so fast says Dr. Dyck, whose own research found that impaired glucose tolerance was not associated with neuropathy. Dyck says that the evidence presented by Asghar is not compelling; the indicators of neuropathy found by studying patient's corneas were not strong enough to produce symptoms, and if the patients in the study did have neuropathy, it was likely to be from some cause other than diabetes, and should be investigated accordingly.

I have not been given an oral glucose tolerance test (OGTT) formally, but I have followed the standard protocol and given it to myself a couple of times, just out of curiosity, and in neither case did I do nearly as badly as the patients in the study. This leads me to wonder if they are truly as non-diabetic as the study assumes. Plenty of people with diabetes have not been diagnosed with it, often because they were tested on a day which was very atypical for them, and they got a seemingly normal result. They are categorized as non-diabetic because of insufficient data. If many of the people in the study showed subtle, asymptomatic signs of neuropathy developing, maybe it's an indication that their glcyemic control is worse than their doctors realize.

Obviously I have a personal interest in believing research which says I'm pretty safe from neuropathy, at least for now -- and not believing research that says I'm a sitting duck. I'm prejudiced here. I know which side of the argument I'm hoping will win. That can make it difficult to exercise good judgment in deciding what to believe.

Anyway, thirteen years into the diabetes adventure, I'm not experiencing any symptoms of neuropathy; when I've got a rock in my shoe I can't stop thinking about it for an instant. So, I do have some reasonable basis for accepting Dyck's opinion that impaired glucose tolerance is not enough to cause neuropathy.

I hope I'm backing the winning horse here!

 


A Cool Google Trick

People have complained that my site doesn't have a search engine. Having one would certainly be a convenience for people who are curious to see what I've had to say about this or that specific issue over the years. Sometimes I'm curious to see what I've had to say about this or that specific issue over the years, and when I find out, I'm occasionally startled to find that I've changed my mind about an issue that isn't trivial.

Nobody should hold his breath waiting for me to add a search engine to the site -- every time I've looked into it, I've been unable to find a solution that didn't have serious drawbacks. However, the folks at Google have pretty much taken care of the problem for me, because you can use a little-known capability of their browser to do a search of my site.

Here's how it works: in Google, enter your search subject in the form:
site:notmedicatedyet.com subject

For example, to see what I've had to say about the whole cinnamon thing, enter:
site:notmedicatedyet.com cinnamon

...and Google brings up links to 18 pages of my blog where I have mentioned cinnamon.

To find exactly where on these (admittedly large) pages I mentioned cinnamon, you'll probably have to use your browser's find-in-this-page function, but that's usually pretty easy to do. For example, in the Google Chrome browser (generally the least annoying of them), you click on the mysterious three-horizontal-lines icon on the far right, and click "Find" on the dropdown menu.

Good hunting! But please don't hold it against me that I've changed my mind about certain things over time. It's healthy, really, if you think about it.

 


First Thursday Update

Thursday, July 3, 2014

 


The CREDIT Study: Glycemic Control Matters!

The story I've been hearing for years is that keeping blood sugar under control is an important thing, but it doesn't solve every problem diabetes presents. Good glycemic control protects you from diabetes complications such as neuropathy (nerve damage) and retinopathy (eye damage)... which is good, of course... but it doesn't protect you from cardiovascular disease. That is a bit disappointing, considering that cardiovascular disease is the likeliest way for diabetes to kill you. I have taken this to mean that glycemic control is important but not sufficient: it needs to be supplemented with other measures (chiefly exercise) which are protective against cardiovascular disease. Still, it seemed a little strange, a little counter-intuitive, that getting control of blood sugar wouldn't help at all in reducing your risk of dying from cardiovascular disease.

Well, a new study is saying that glycemic control actually does reduce the risk of cardiovascular disease. Like most studies these days, it goes by a cute acronym: CREDIT. This is alleged to stand for "Cardiovascular Risk Evaluation in People with Type 2 Diabetes on Insulin Therapy", although the acronym of that would be CREPT2DIT and they're not fooling anybody. I guess they had no choice; I believe it's an actual funding requirement now: no cute name, no research grant. (I'm thinking of setting up something called Diabetes Intervention Mode Polyphase Longitudinal Endpoint Study, just so I can call it DIMPLES and wait for the grant dollars to come rolling in.)

Anyway. The study followed 2999 diabetes patients (I'm sure they wanted an even 3000 but some flake bailed on them) over a period of nine years, and found that the risk of dangerous cardiovascular events (especially strokes) was indeed lower in patients with better control of their blood glucose. For example, a 1% higher hemoglobin A1c level was associated with 31% more risk of cardiovascular death and with 36% more risk of a first stroke.

Another interesting finding of the study was that good glycemic control was protective against cardiovascular disease even if control was achieved by means of insulin treatment resulting in a significant number of hypoglycemic episodes. A lot of doctors have been concerned that insulin treatment might be a bad tradeoff, if it improved glycemic control only at the cost of triggering hypoglycemic episodes which were themselves potential causes of cardiovascular problems. According to this study, that doesn't seem to be a valid concern; good glycemic control was protective even if the patient had to put up with hypoglycemic episodes as an accidental side-effect of therapy.

I'm still holding on to the idea that exercise-centered therapy is the ideal approach, because it (1) reduces blood glucose, (2) does not cause the sort of problems with hypoglycemia that insulin injections often do, and (3) reduces the risk of cardiovascular disease more than glycemic control alone can do.

Still, it's nice to know that those who are not able to exercise are still getting some cardiovascular benefit from other therapies that reduce blood sugar. It's always reassuring to learn that a given therapy works better than you had heard it did.

 


Bone Marrow is Back!

One of the more peculiar word-origin tales I've ever heard had to do with the word "restaurant". The word is French, and it dates back to the days when anyone who wanted a meal outside the home would have gone to an inn or a pub -- the idea of a business that existed for no other purpose than to serve meals to the public has not always been around. "Restaurant" meant "restorative", and it referred to an intense, boiled-down broth made from bone marrow, which was thought to restore people to health after they had been ill. The stuff became popular enough that businesses started opening up, a little like modern coffee-shops, which sold "restaurant". Once you had recovered enough from the latest virus to be able to venture out at all, you headed for the local place that served "restaurant", and sat there sipping your marrow-broth and presumably giving your battered body the strength to face life fully once more. In time, these businesses were named after the restorative broth they sold, and somewhere along the way it occurred to them to sell other sorts of foods, to people who weren't recovering from an illness and were up for something more solid than marrow-broth. The name given to such businesses lingered on, even though restorative broth is no longer what they're in business to sell.

However, restorative broth might be in a position to make a bit of a comeback. A new study from the University of Michigan suggests that bone marrow actually does contain stuff which is beneficial to health: "The fat tissue in bone marrow is a significant source of the hormone adiponectin, which helps maintain insulin sensitivity, break down fat, and has been linked to decreased risk of cardiovascular disease, diabetes, and obesity-associated cancers."

I don't know if subsequent research will show that adiponectin really doesn't provide the health benefits described, or does so only at the cost of causing other problems. But in case you want to do your own study of adiponectin, here's what the protein looks like:

Actually, that's more of schematic diagram, outlining the protein's structure; the real thing probably would look different to you... if human eyes could see it at all. But anyway, there's a lot of that stuff hanging around in bone marrow, and apparently it has some power to prevent diabetes and ameliorate its effects.

I'm sure somebody will isolate adiponectin and start selling it as a supplement (and that, a few years later, some study will show that it doesn't work in that denatured form). But, in the meantime, if a French chef wants you to try some kind of soup or broth made from bone marrow, say yes and see if helps.

 



Older Posts:

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