Abdominal adiposity and risk of death, or "belly fat'll kill ya'"

We need to take a more proactive approach to helping the rotund become more svelte.

Yes, that would be nice, wouldn't it? But what approach, given that, at least as far as I know, every diet trial has been, pardon the expression, a big fat flop.

Also, looking at the data (never trust the abstract and the conclusions), it looks like a big belly is actually the greatest risk in patients with LOW BMIs. For those with a "normal" BMI, only the top quintile was statistically different from the lowest quintile and for those with high BMIs, the difference never reaches statistical significance. This isn't smoking where the relative risk is something like 40. Despite the highly statistical significance of the trends, I'm not sure how biologically significant this data is.

I almost wonder if they didn't miss a significant variable...were the height, weight, and waist circumference measured more than once? If so, how were people who had significant changes in any of the above considered? That might confound, given that weight loss is more often associated with illness than successful dieting.

Surely you meant "C-reactive protein", not "C-reactive peptide".

Did the study find any benefit to statins for those with normal levels of C reactive protein?

Or maybe it's time to say heck with it, weight is an unmodifiable variable* and just give everyone statins.

*With the exception of gastric bypass surgery which can reduce weight in severe to morbid obesity.

Surely you meant "C-reactive protein", not "C-reactive peptide".

Surely i did. Fixed.

The problem with these studies is they're all correlative. The real test is manipulative experiments, and those actually have been done.

So, if fat is the problem, removing it should remove the problem. That doesn't happen - liposuction on abdominal fat does not reduce diabetes or heart disease risks.

IMHO, much of the obesity studies are just plain lazy science - measure a variable and look for a correlation. Manipulative studies are minimal, and could be done without much difficulty on animal models (and probably without much more difficulty than many other medical studies on humans).

The only other science that gets away with this much reliance on correlation and non-manipulative studies is paleontology, and they're got a much better excuse.

Yes, that would be nice, wouldn't it? But what approach, given that, at least as far as I know, every diet trial has been, pardon the expression, a big fat flop.

Exercise, while difficult, and not having the success that we might desire, is far more effective than diets.

So, if fat is the problem, removing it should remove the problem. That doesn't happen - liposuction on abdominal fat does not reduce diabetes or heart disease risks.

Not necessarily. There may be better and worse ways to lose weight/fat just as there are better and worse ways to increase HDL or decrease arrythmias.

Also, looking at the data (never trust the abstract and the conclusions), it looks like a big belly is actually the greatest risk in patients with LOW BMIs. For those with a "normal" BMI, only the top quintile was statistically different from the lowest quintile and for those with high BMIs, the difference never reaches statistical significance. This isn't smoking where the relative risk is something like 40. Despite the highly statistical significance of the trends, I'm not sure how biologically significant this data is.

That's consistent with the height dependence of BMI.

In the meantime, I'd like to see less people hating on people for being fat. It's like hating on somebody for scratching bug bites too often, or drinking too much soft drinks and having bad looking teeth, or any parallel to the fashion industry really.

If you want people to be healthy, then that's how you approach it. But by the name of ceiling cat, if a person does not have a problem with being fat it's not your business to walk up to them and try to "help" them any more than you can walk up to somebody and criticize their shoes.

Throwing in an anecdote:

I retired almost 4 years ago, I was obese; with little conscious effort I dropped 80 # over 3 years.

There's something going on here with fat and stress and hunter/gatherers in fluorescent-lit cubicles.

The fat doesn't decrease in the same way it increased.

The only other science that gets away with this much reliance on correlation and non-manipulative studies is paleontology, and they're got a much better excuse.

I beg to differ, the science of second had smoke (SHS) or basically the tobacco control crowd gets away with allot of BS in effort to promote smoking bans. Here I may be a denialist, I am not arguing the health risks to smokers is not there, but the health risk of passive smoking have been way over done (that might be my denialism). The only one I largely see in the public health community is Dr. Michael Siegel trying to stay with real science. Here is the problem I have with PalMD:

We need to take a more proactive approach to helping the rotund become more svelte.

Does that mean new laws, in the same sense as smoking bans? We have already seen that in New York City. The banning of certain new restaurants (think fast food) in California (can't remember which city). If you meant something different, please clarify.

Yow, that's scary! This post made me want to get up and jog 5 miles RIGHT NOW! ...'Cept it's raining. And I just drank an enormous glass of milk in response to RDCT's Vitamin D post. Which leads me to conclude that perhaps my health would benefit from my taking a deep breath after reading teh internets.

There may be better and worse ways to lose weight/fat just as there are better and worse ways to increase HDL or decrease arrythmias.

Agreed, but that doesn't mean the failure of manipulative experiments to confirm to the expectation is invalid - it means more manipulative studies are needed that use other methods. If more diverse manipulative studies fail to confirm that reduction in obesity reduces health issues, then the entire hypothesis of obesity causing health problems can be discarded.

What if the supposed health effects of obesity are actually of an underlying metabolic disorder which *also* results in obesity? That would result in the observed correlation, but also explain the lack of effect of liposuction (and, depending upon the nature of the disorder, possibly the failure of most diets). But, to my knowledge, this hypothesis has not been tested.

Basically, the medical community seems to have just latched onto this hypothesis and done a load of correlations without actually, really *testing* it.

The Perky Skeptic: Read the post on your blog. Nice post, tried to leave a comment, but the requirements needed to post exceed those that I have.

I've seen waist-measurement stuff at my doctor's office, and now I'm wondering if they've refined the tools any since whenever it was that my doctor put up the information, because it really seems as though the numbers aren't ready for prime time. For one thing, the charts there say there's no data for any number of groups, such as Native Canadians and some Africans.

For me, personally, I also kind of look at their numbers for the groups I'm in and sort of go, "Yeah, I could lose every microgram of fat on my body and not have that waist measurement." For my group, for women, the "target" is 31" or below...but I lift weights (which broadens my back) and my ribs sit on my hips, so there's no there there. (I don't have an hourglass figure, I have a waterglass figure.)

Seems to me this is an even cruder tool than the BMI, which doesn't work well at the best of times, so is it really smart to be assigning so much import to it?

One thing I'm not entirely clear on: Obesity has several causes. How do they tease out presumably more harmful problems like sedentary lifestyles from other causes, such as a genetic predisposition?

Because while I may try and cut sugars - and am - I don't think I'm going to manage much more than that, so knowing what obesity-related risk factors are the most important and which I can change...

Mokele and khan have excellent points. Visceral fat is an effect, not a cause. The cause is not enough NO, which skews physiology into the response mode for high stress. That stress response mode mobilizes energy (increasing blood sugar and blood lipids), increases ATP production rate (by disinhibiting cytochrome c oxidase by destroying NO via mitochondrial generated superoxide), conserves ATP by turning off long time constant pathways (such as autophagy and mitochondria biogenesis), shifts oxidation of glucose from peripheral tissues to the CNS (by increasing glycolysis which generates lactate which the brain preferentially uses), skews steroid metabolism to a hyperandrogenic profile (by disinhibiting the cytochrome P450 enzymes that produce and metabolize steroids). The visceral fat comes from the need to dispose of lactate when there are not enough liver mitochondria to recycle it back into glucose. It gets converted into fat.

This same thing happens (but much faster) during cachexia. Then skeletal muscle is converted into alanine to provide substrate to make glucose to run everything by glycolysis. Most cachexia is due to immune system activation, and the immune system doesn't use lipid, only glucose, so the lipid mobilization isn't the same as the fight or flight stress one gets from work.

Agreed, but that doesn't mean the failure of manipulative experiments to confirm to the expectation is invalid - it means more manipulative studies are needed that use other methods. If more diverse manipulative studies fail to confirm that reduction in obesity reduces health issues, then the entire hypothesis of obesity causing health problems can be discarded.

What if the supposed health effects of obesity are actually of an underlying metabolic disorder which *also* results in obesity? That would result in the observed correlation, but also explain the lack of effect of liposuction (and, depending upon the nature of the disorder, possibly the failure of most diets). But, to my knowledge, this hypothesis has not been tested.

Obesity is not from an "underlying metabolic disorder". You forget that over the last 50 years we've seen a big change in body habitus and increases in obesity-caused diseases as a result. This is the broad interventional experiment you want. We haven't all acquired a new metabolic disease, we're eating too much. Things unheard of 50 years ago - type II DM being diagnosed regularly in teens, gout at younger and younger ages, obstructive sleep apnea (a remarkably malignant process very much unrecognized by the public yet) etc., are showing us that it's the fat, not some cosmic change in our DNA, or new fantastical disease, or cryptic virus causing all this.

And it's biologically plausible. The animal studies that actually do interventions (transplanting excess fat for instance from legs vs bellies into mice and studying the effects), and the extensive physiologic basic science on adipose tissue teaches us that fat is an endocrine tissue and different fat behaves differently. Dismissing this particular study as correlative, and painting the whole field as such misses the point, this is part of a broad knowledge of physiology and medicine that shows us that an excess of fat is bad, and wishing for some unethical study of fat transplant in humans is an unrealistic expectation of science.

Further, the liposuction thing is a total red herring. Liposuction can not safely remove more than a few pounds of fat at a time, and doesn't exist to control obesity as people mystically seem to believe. Plastic surgeons use it to shape bodies, not reduce them dramatically in size. And the effect of removing less than 1% of your body weight, not surprisingly, doesn't counterbalance all the other fat that remains, not to mention a significant component of the big belly is omental fat - not accessible to the plastic surgeon. However, studies in which people diet and lose weight show that they can reduce their risk factors, cure their NASH, control their diabetes and cholesterol etc. Granted, we don't have a great method of making people lose weight without surgery but anyone who has ever treated diabetes knows that you can control it in individuals who are motivated to make lifestyle changes - cutting out the fat.

An even better interventional example is gastric bypass surgery or gastric banding - there is your intervention. Gastric bypass surgery cures fricking type II diabetes a lot of the time - sometimes before patients even leave the hospital suggesting an interesting mechanism from intestinal rearrangement. Banding does it too - but usually not until significant weight has been lost - an interesting hint at some diabetes physiology here. Now, granted, it's not always cured by these interventions, but it's quite possible that the time spent obese causes cumulative damage to the pancreas that can't be reversed - it's a crappy organ lacking a stem cell population that has very limited regenerative capacity (as suggested by papers such as this one).

Finally, as someone learning medicine and regularly encountering patients with BMIs well into the 60s, 70s and even 80s, and people who are eating themselves to death, you see the steady progression of increasing injury from excess weight. There is no clearer dose-response relationship in medicine than the negative effects of increasing obesity. The diabetes and hypertension get worse, the joints fail, the breathing becomes restricted, mobility declines making exercise more difficult, sleep apnea ensues with resultant pulmonary hypertension and increased right heart strain, and on and on. And to make things worse, medical interventions become more difficult, surgery becomes riskier, infections become harder to treat. We can argue about the gray area between health and overweight, and obsess over the relative risks of someone with BMI of 29 vs 31, but that's missing the point. The more weight you're putting on, the worse your risks. And a significant portion of most physicians' time is spent treating these complications in this patient population. Quite frankly, your description of the obesity-bad health link as a mere hypothesis, or a meaningless correlation, or your suggestion it's going to be dismissed any time soon is complete and total nonsense and denialism.

Now, as far as health interventions the libertarian types can get into hissy fits over government interventions but I welcome them. When I hear of the ball park that sells a double bacon cheese burger with two Krispy Kreme donuts for a bun, I don't think that's just gross, that's an incredibly irresponsible business practice. And the libertarian types can get all upset about personal freedom and choice, but the fact is crappy health is costing all of us money as it results into excessive utilization of health resources. Every man is not an island, there are societal effects to these behaviors and it is not wrong for the government to be involved in trying to curb them. I'm not talking 1984 forced calisthenics, but there are things that make business sell healthier food - like making them list calories, accurate labeling of ingredients (did you know lots of restaurants dip meats in liquid butter before serving them?), and maybe, just maybe, forbidding some ingredients with specific health risks like trans fats. There's a lot of room for argument here - and I think this is why there is a libertarian trend towards obesity denialism - but there is no real argument among health professionals that obesity is a malignant process.

Libertarian slippery-slope/Orwellian paranoia/meltdown in 3...2....1...

I used to mix it up with the extreme pro-fat crowd before I finally understood that 1) they are far more irrational than the most swamp-bound Mississippi gin-chugging Baptist preacher, and 2) I was only making people feel worse about themselves regardless of how gently I couched my words.

I wrote a five part series for SB on this very topic in 2006; interested people can go here to check it out. It's a very good think that the Big Fat Blog crowd represents the opinions of (I'm guessing) under 1% of overweight people everywhere.

Good post. Actually, it's the macrophages that accumulate in the abdominal fat that are the source of inflamatory mediators. Just picking nits.

The other big issue that will likely get more press as research continues is adiponectin. Adipocytes make adiponectin, which in tern seems needed to for insulin to do its job (storing sugar and triglycerides). When the fat cells become dysfunctional, they stop producing adiponectin. The result is less insulin function and less sugar and triglyceride storage. Sugar and triglycerides start going up. It seems to be a feedback mechanism, like the fat cells saying, "hey, we are storing enough fat here, don't sore anymore".

High sugars and triglycerides result. Couple this with the inflamatory mediators, some hypertension to damage the arteries, high LDL cholesterol to put plaques in the damaged arteries, and you have recepie for disaster. Don't even get me going on smoking!

These bad chains of events can be reduced or cured with regular exercise and diet. For those unable to reach goal due to poor effort and/or genetics, then statins are very helpful, and potentially life saving. They decrease LDL and appear to decrease the inflamatory mediators. In obese diabetics, combining simvastatin and pioglitazone (stimulates adiponectin production) has been shown to dramatically lower CRP levels, sugar and LDL.

Many people seem open to understanding these pathways if explained in the exam room. Some even start to take the advice to eat better and exercise. The natural phobia of these medications is sometimes enough to get people doing exactly what is needed. The problem is the people who say, "just give me pills" and are not willing to get off the couch.

Unfortunately Mark, many are suffering from an underlying metabolic disorder, not enough basal nitric oxide. That is the fundamental difference between the rural undeveloped world and the developed world. It actually started 150 years ago with the advent of the germ theory and the acquisition of abundant supplies of clean water which people used for bathing. This removed the surface biofilm of ammonia oxidizing bacteria which normally sets the basal NO level.

This got worse starting in the 1960s with the advent of synthetic anionic detergents, the alkyl benzene sulfonate detergents. These are toxic to ammonia oxidizing bacteria at ppm levels. The next hit was the advent of conditioning shampoos which allowed for hair washing every day. The increased use of municipal water supplies treated with disinfectants replacing wells which allowed ammonia oxidizing bacteria to survive was another hit. The current craze of anti-microbial everything is the latest hit.

Sleep apnea is quite clearly caused by not enough NO. That is why Sildenafil makes it worse. Inhibiting the PDE5 increases cGMP levels and causes feedback inhibition of NO production. For things which are mediated by NO but not through cGMP, that causes problems. Breathing is triggered by SNO-thiols secondary to hypoxia in addition to low O2 and high CO2. When the NO pathway is screwed up, the low O2 and high CO2 pathways have to get further out of range to activate. That is what causes OSA. Chronic hypoxia makes low NO worse because one of the compensatory pathways of hypoxia is the production of superoxide which lowers NO levels.

People eat more because their bodies are telling them they are starving. They are starving because the cells too far from a capillary dont have enough glucose because the intervening cells have consumed it. That is what causes the hypertension of the metabolic syndrome, compensation to increase extravascular flow of fluid, where essentially all cells actually get their glucose (not from bulk blood which is only in contact with endothelium). That is what causes the insulin resistance too. Insulin gets consumed by the intervening cells too, and unless the first cells are glucose resistant and insulin resistant the cells too far from a capillary cant get enough glucose or insulin.

Banding peoples stomachs does get them to lose weight. That is ok as long as they are in ketosis as they consume their excess depot fat. Once that is gone, the liver and the vasculature doesnt have enough capacity to supply enough glucose so they develop an ATP crisis in the brain that they cant get out of, so they get depressed.

However, studies in which people diet and lose weight show that they can reduce their risk factors, cure their NASH, control their diabetes and cholesterol etc.

Can you cite? I'm unaware of any studies that have shown great results in controlling obesity with diet/exercise alone. The few I've seen have mentioned weight loss on the level of 5-10 pounds with regain within a year or so. Gastric bypass can be useful in severe obesity, but isn't really safe enough for mild obesity or people who are overweight.

Daedalus, I love you man. But while NO might be the proximate cause of a lot of things, the problem with sleep apnea is physics.

In obesity you have a restrictive pulmonary picture (pickwickian as it were), as well as floppy, restricted and collapsing airways (snoring). This results in apnea - periods of not breathing and decreasing O2 sats - that doesn't resolve until sufficient pCO2 buildup causes a bump in inspiratory drive. The molecular mechanisms of how the airways and pulmonary vasculature remodel in response in this are separate from the cause - crappy airways due to an excess of adipose tissue increasing intrabdominal pressure and the work the diaphragm needs to do as well as narrowed airways from obese body habitus.

I'm not talking 1984 forced calisthenics, but there are things that make business sell healthier food - like making them list calories, accurate labeling of ingredients (did you know lots of restaurants dip meats in liquid butter before serving them?), and maybe, just maybe, forbidding some ingredients with specific health risks like trans fats

I generally agree, particularly with the idea of making restaurants list calories and maybe other nutritional ingredients (who is it supposed to be oppressing to do that anyway: you can still buy or sell whatever you want, you just have more information about what's in it). However...

We haven't all acquired a new metabolic disease, we're eating too much.

This is an oversimplification so severe that it ends up as a distortion. Eating too much--and the wrong things--is not the only lifestyle change that has occurred over the last 50 years and it's only part of the problem. People sleep less than they did 50 years ago and not getting enough sleep has been linked to obesity. Longer working hours also mean less leisure time and so less time for exercise. Heck, the prevalence of A/C means less energy use going towards sweating. There are few supermarkets in poor neighborhoods and many fast food joints. And so forth. The problem goes beyond "oh, we've just become a bunch of lazy slobs".

BTW: You're a fourth year medical student. You haven't seen much yet. In another 10 years you will have seen skinny type II diabetics, underweight people with sleep apnea, and people whose cholesterol goes up when they lose weight. (Yep, I've seen all of those, more than once. No, I'm not going to prove it. Haven't you ever heard of HIPAA?) Obesity is not THE risk factor for DMII, OSA, or hypercholesterolemia in the same way that smoking is THE risk factor for lung cancer. Assuming that it is will eventually end up in a lawsuit when you fail to diagnose someone's DM despite their polydipsia and polyuria because their BMI is 18 and everyone knows that only fat people get DMII.

Here's a good cite - right of the top of my search. Leave it to the Chinese to do a wicked population study like that. Exercise - even without weight loss can lower your HA1C Or this or this cochrane review or this one which shows that diet alone probably is ineffective probably do to patient noncompliance in my opinion. People often don't lose a lot of weight in these studies and when it comes to changing a lifelong culture of food, no amount of dietary advice ever seems to suffice. Another good diet + exercise trial.

I think the message is that weight loss studies on US subjects suck. But other countries have been successful. Ha!

Obesity is not from an "underlying metabolic disorder". You forget that over the last 50 years we've seen a big change in body habitus and increases in obesity-caused diseases as a result. This is the broad interventional experiment you want. We haven't all acquired a new metabolic disease, we're eating too much. Things unheard of 50 years ago - type II DM being diagnosed regularly in teens, gout at younger and younger ages, obstructive sleep apnea (a remarkably malignant process very much unrecognized by the public yet) etc., are showing us that it's the fat, not some cosmic change in our DNA, or new fantastical disease, or cryptic virus causing all this.

Ah, 'the environment has changed, so it can't be a gene' fallacy. Ignoring the fact that all genes interact with the environment. In addtion you seem to suffer from the delusion that 'eating too much' cannot be influenced by genes.

Further, the liposuction thing is a total red herring. Liposuction can not safely remove more than a few pounds of fat at a time, and doesn't exist to control obesity as people mystically seem to believe. Plastic surgeons use it to shape bodies, not reduce them dramatically in size. And the effect of removing less than 1% of your body weight, not surprisingly, doesn't counterbalance all the other fat that remains, not to mention a significant component of the big belly is omental fat - not accessible to the plastic surgeon.

And you failed to read the linked article on liposuction, which contains:

"In the past, we usually removed no more than about 5 liters of fat, but in this study we showed you can safely remove four times that amount."

20 liters of fat is roughly 40 pounds. So it's not '1% of your body weight'.

The rest of your screed can be ignored.

Dianne, I've seen plenty of skinny type II diabetes. Hell, my grandfather was one. If you live long enough your pancreas craps out. Like I said, it's a crummy organ. I have had skinny people with OSA, quite a few really - short thick necks on a skinny person can be just as bad as pickwickian body habitus. I'm not saying these diseases are only caused by obesity.

As far as my simple distortion we're starting to confuse lots of statements and what I am responding to.

People are overweight because they eat too much and exercise too little. It's physics. Energy goes in, what is not used gets stored as fat. All of the factors you mentioned contribute, certainly, to obesity. The fast food among the poor is an especially pertinent point. Sometimes the cheapest food is the worst for you - another intervention in which I think the government could do a lot of good would be making healthful food more inexpensive.

Diseases like these are multifactorial but I'll disagree with you. Obesity is the factor the majority of the time that people can control. Yes, you will have diabetes due to genetics, environment, whatever, but the leading cause of preventable diabetes is unquestionably obesity and sedentary lifestyles. Same with OSA. You can't control your airways but you can control the stress put on them by exercising, keeping the pounds off your belly and not smoking.

Finally, the point of citing the lap band and laparoscopy studies (ah, citing, note that I do this when I write), is to show that an intervention to reduce weight (or rework the pipes interestingly) can cure diabetes. What more information do we need? We have correlative studies showing increased diabetes in the overweight. We have animal studies showing the effects of adiposity on glycemic control. We have human studies of weight loss and exercise intervention and surgery to induce weight loss decreasing the incidence of the disease or curing it completely. If you can't be convinced by that, well, you know what that means.

As far as saying I'll be incompetent because I supposedly think obesity is the only risk factor for diabetes is a bit much. I really don't need this "you'll be a bad doctor if you don't believe my BS" crap. I'm looking at the science that drives these risk factors it is undeniable. You're hedging, throwing out red herrings and getting a little bit cranky, and the personal attack was offensive. Knock it off.

Mark: Thanks! I can see you're experienced at pulling references for lazy attendings. Nonetheless, a cursory reading (I'm at home without full text access) suggests that weight loss was not the important factor in improving diabetic control. (Side note to anyone reading this who has DM and little medical background: Diet and exercise MAY help CONTROL your diabetes. Do NOT under any circumstances assume that your diabetes is "cured" because you started exercising more or lost five pounds. Glucose monitoring is for life or until a better therapy comes along.)

The ethics of that first trial left me gasping. Informed consent, feh, who needs it? That particular trial was low risk, but what are they doing with higher risk trials like phase I drug studies? I probably don't want to know, at least not until/unless I'm in a position to do anything about it.

Obesity is the factor the majority of the time that people can control.

Well, no, they can't. A sedentary lifestyle is something they can control. Eating too much junk food is something they can control. But serious weight loss without regain within several years is really quite a rare achievement. Humans are complex systems and their bodies make adjustments when caloric intake falls, whether due to intentional dieting or famine.

As far as saying I'll be incompetent because I supposedly think obesity is the only risk factor for diabetes is a bit much. I really don't need this "you'll be a bad doctor if you don't believe my BS" crap. I'm looking at the science that drives these risk factors it is undeniable.

Cut the crap, kid. I wasn't saying that you'll be a bad doctor, just that you should check your assumptions. I will say that you'll be an unsuccessful resident if you get this offended when an attending challenges you. Honestly, is this the first time you've ever made an overly broad statement and had someone point out to you that you could make a bad mistake if you take your assumption too literally? Shame on your attendings if so.

And please note that the studies you cited did not universally show what you initially claimed. Your claim was "studies in which people diet and lose weight show that they can reduce their risk factors, cure their NASH, control their diabetes and cholesterol etc." You cited several studies demonstrating that diet and/or exercise can sometimes help control diabetes or reduce the risk of progression from impaired glucose tolerance to DM. Several of the studies cited explicitly state that weight loss was NOT critical for successful reduction of HgbA1C. Nor did the average HgbA1C reductions typically bring the DM into a range that could be called "controlled." In general, when you look at DM over time, it can sometimes initially be controlled by diet and exercise but inevitably progresses to requiring drug therapy, no matter how diligent the patient is. So while I appreciate your citations, I think that maybe you're taking too broad an interpretation of them. Again, I do not mean this as an attack on you, but more of a warning to look at the data carefully and not draw overly broad conclusions.

Now you guys are starting to piss me off. I'm showing you studies with an intervention to remove weight from people, and it cures their freaking diabetes. What have you got. "oh, liposuction doesn't work, it can't be the fat" using a cherry pick of the freaking literature. Meanwhile you ignore the animal studies and the basic physiology of adipose tissue as an endocrine organ. Give me a break.

And Llewelly. If there is a gene that makes us eat too much, guess what? If it's a gene, it's one we've had all along, I'm pretty sure we didn't evolve into a new species in the last 50 years. The "environment" we live in is one of ease and plenty. One in which we eat a lot, and don't exercise. So yes, call it environmental. But at the end of the day the problem is too many calories in, not enough out and you guys are distracting from the central point. Obesity is causing diabetes. Whatever is causing the obesity besides - space aliens, lead in the water, whatever, the result is that when you eliminate the obesity, the disease goes away.

My understanding of liposuction is that the upper safe limit hovers around 20 lbs. If people are pushing that limit that's fine, but we're still talking a small percentage of the total fat in the body being removed. The use of liposuction for control of obesity is controversial and not routinely performed. Further, that is just one study with n=7 of diabetic patients (we are talking about the NEJM one in that news article right?), others suggest conflicting results, oh and another study with a larger n showing better diabetic control after liposuction. A cherry-pick of the literature is not going to convince me, sorry.

Dianne, take a look at that Chinese study again - "The cumulative incidence of diabetes at 6 years was 67.7% (95% CI, 59.8-75.2) in the control group compared with 43.8% (95% CI, 35.5-52.3) in the diet group"

Diet alone can prevent diabetes - if successful. The difference between these studies is that you see weight loss that is transient. They avoid worsening diabetes when they lose the weight, and then when the weight returns, shockingly the diabetes returns. I think it reinforces my point that weight loss is key. It's just hard and we don't have a diet that works in a large amount of the population. One of my attendings likes to say, in a competition between doctor's advice and culture, culture always wins. We're talking about taking people eating one way for their whole lives then expect dietary advice to change them in a 15 minute office visit.

But that doesn't mean that when diet is adhered to, and especially with exercise added in to improve the insulin resistance in muscle the effect isn't better glycemic control! The intervention works, but noncompliance is high. That's what you see in terms of the american vs foreign studies. Look at the chinese and swedish study. When they kept the weight off, they kept the diabetes in control.

I'm showing you studies with an intervention to remove weight from people, and it cures their freaking diabetes.

No, you aren't. Pan et al shows a decreased risk of progression to DM when either lean or overweight patients with impaired glucose tolerance were treated with diet or exercise. No mention of weight loss was made in the abstract. (I hate not having full text.) The Boule et al study showed that HgbA1C was reduced from 8.3% to 7.65% with exercise (hardly a "cure" for DM-7.65% is still a diabetic level of HgbA1C) despite a lack of difference in weight in the exercise group. The Thomas et al Cochrane review showed a decrease in HgbA1C of 0.6% with exercise, again, significant but not exactly a cure. Interestingly, they do note a decrease in body fat, though not necessarily weight. The bottom line on the Nield Cochrane review is "No high quality data on the efficacy of diet alone exists for treatment of type 2 diabetes mellitus." Certainly no evidence of a cure for DM with diet alone. The Eriksson/Lingaerde paper probably gives the best evidence for your claim and even there, while the diabetes went into remission in some (>50% per the abstract) subjects, remission is not cure and the weight loss involved was minimal.

Better diet and more exercise are good things, particularly in DM and IGT. But you're way overclaiming for their effects.

Diet alone can prevent diabetes - if successful.

Moving goalposts. Just a post ago you said that diet could CURE diabetes. Now you're making the much more reasonable claim that it can prevent diabetes, in at least some cases. (Note that in the Pan et al study quite a large number of patients progressed even with diet and exercise.) Yes, better diet and more exercise can help prevent or, at least, delay the onset of diabetes. That doesn't mean it can cure diabetes. Colonoscopy can prevent colon cancer by finding and eliminating polyps before they become cancerous, but it can't do crap (er...pardon the expression) for an established cancer. Keep your claims reasonable. Only the woo crowd believes that diet is the key to living forever.

The intervention works, but noncompliance is high

Ok, this is the oncologist in me talking, but...when the noncompliance is that high, the therapy essentially doesn't work. One basic requirement for an effective treatment is that it has to be something a patient is willing to deal with. People used to die of testicular cancer because the effects of taking cisplatinum therapy were so horrible that they preferred to die in relative comfort. Since the invention of better anti-emetics, this problem is no more. (Zofran is wonderful stuff, no matter how much money the manufacturer makes on it!) If the vast majority of people are unwilling or unable to comply with a dietary and exercise regimen then it is a failure, no matter how effective it is in theory.

Ah, Dianne, you've hit it on the head, but it's gonna really bother the altmed folks. Diet and exercise, while desirable, rarely do the trick. Treating people with impaired fasting glucose or dm with drugs early is beneficial. If someone is able to reduce their risk by losing significant weight as part of a global treatment plan, terrific, but they're probably still going to need a statin and metformin, at the least.

Losing significant abdominal mass will help with the other factors, such as osteoarthritis, sleep apnea, arterial hypertension, and pulmonary htn.

I feel like some of you are missing a pretty important point.

If people with low BMI's still have high waist measurements, and therefore an increased risk of various diseases...then how do you fix that? As far as I know, pattern of fat deposition is a lot harder to control than overall amount of fat.

As I mentioned above, one possible mechanism for improved glucose control in overweight diabetics is increasing adiponectin production from the adipocytes. This happens in short order in response to aerobic exercise. Diet does not do this by itself. This is one reason that obese people improve the diabetes control by exercising even without substantial weight loss. They must continue this. If they then eat properly and continue exercising, there will be a reduction in the omental fat cell size. Adiponectin secretion will then normalize do to shutting off of this feedback loop. Unfortunately, for many, this takes too much discipline (ever watch "The Biggest Loser"? - that is what we are talking about). For most, it is statins and DM meds (when is Actos going generic?).

http://care.diabetesjournals.org/cgi/content/full/27/2/629

http://jcem.endojournals.org/cgi/content/abstract/89/9/4312

http://www.springerlink.com/content/85231nmuh8520212/

http://jcem.endojournals.org/cgi/content/abstract/91/5/1698

And Llewelly. If there is a gene that makes us eat too much, guess what? If it's a gene, it's one we've had all along,

You're obviously unfamiliar with theory that a gene's expression can be affected by a change in the environment.
It's well known that food was far more difficult to obtain (for those of us living in developed nations) for our ancestors. Famine was more common. In addition - daily living typically required far more calories. (For example: farmers using pre-1930s farming methods probably expended over 3500 calories a day.) A gene that gives a strong desire to seek out food - especially large amounts of fatty food - has naturally has a very different effect in an environment where such food is easy to obtain than an environment where such food is very hard to obtain.

My apologies, MarkH, I misread your post.

@ dianne:

"never trust the abstract and the conclusions"

Then what would it take to convince you? If none of the studies yielding an apparently unpalatable conclusion are worth looking at, what else is there?

Well, why not look at how really heavy people move through their lives? They often can't climb a flight of stairs. They complain of joint problems, they populate physicians' offices in vastly disproportionate numbers, and yes, they manifest all sorts of woeful conditions from diabetes to gout to PCOD to pregnancy complications and more far in excess of leaner folk.

All discussion of why we've gotten bigger in America in the past 50 years is just a hijack. Yes, it's because of increased food availability, but that's not exactly the point. How increased food consumption triggers weight gain in more susceptible people is not irrelevant, but in the context of how these arguments go, it's moot.

Now, this is NOT a moral judgment. That's the problem that supercharges every single discussion of obesity--from those on the elementary-school playground to those emanating from the Harvard School of Public Health--and leads to rhetorical death-spirals like the one here.

The unfortunate fact is that even if a fatter individual were somehow to be scientifically found to be incapable of reducing his or her belly adiposity, his or her increased risk of succumbing to all sorts of nastiness remains firmly in place. Looking at a known extreme, people with familial hypercholesterolemia don't have a damned say in what goes on in their bodies, but they're still basically screwed. Taken even further, should people with CF be urged to simply breathe harder? What about people with MS? Maybe if they just focus a bit, they can actually keep their neurological function intact, right?

By establishing up front a categorical mistrust of medical studies, you effectively remove yourself from meaningful dialogue.

Ok, this is the oncologist in me talking, but...when the noncompliance is that high, the therapy essentially doesn't work. One basic requirement for an effective treatment is that it has to be something a patient is willing to deal with.

And here you and I agree. It also shows the problem with the Cochrane studies. When you average together the studies with high compliance to low compliance the effect will disappear. Meta-analysis is meta-ing the results into insignificance. But look at the individual studies and you see for instance the Swedish and Chinese studies showing a high degree of efficacy at 6 years when compliance is high.

If you want to discuss the worthlessness of diet intervention in the population that is valid and good and true - diet is hard. Hell, even exercise is easier because exercise can at least be fun. But that doesn't change the fact that studies with diet and high compliance show slowed progression of diabetes or return to glycemic control.

Granted, diet and exercise are interventions that are unlikely to take in many patients. But so is counseling for smoking cessation - does that mean that quitting smoking doesn't reduce odds of cancer - no matter how infrequent people quit based on counseling (the figure I've seen is less than 2% will quit with doctor advice compared to 0.5% for those not counseled to quit).

Of course not. So it's still responsible to recommend the cheap and easy intervention of diet and exercise for diabetics even if you know it might benefit only 1 in 50 patients. The NNT is high, yes, but it is effective when done.

Finally, dismissing the banding studies I think is an error as it reinforces the underlying point that adiposity itself is contributing to diabetes. These studies with banding and bypass are showing efficacy of preventing diabetes or return to glycemic control in as many as 80% of patients - and these are reviews of studies inclusive of thousands of patients. There is no other effective treatment for cure of diabetes that is more efficacious. And in the ones that fail to control their diabetes the studies indicate this is related to duration of diabetes in that population - again suggesting a long term damage of adiposity on insulin secretion. Granted, eventually their pancreas might crap out too - but it is a crappy organ after all. But the studies prove removing adiposity through surgery can cure the disease in many folks. It's the adipose tissue, simple as that.

"when the noncompliance is that high, the therapy essentially doesn't work. One basic requirement for an effective treatment is that it has to be something a patient is willing to deal with."

I hear you. What this implies is that drug addicts and alcoholics who go through exactly the same kind of "yo-yo" success/relapse cycles as overweight people who go through loss/gain cycles have "cocaine set points" and should simply accept that their bodies are meant to carry a certain burden of pestilential substances, consequences be damned. If certain people have a naturally harder time pursuing healthful outcomes than others do, then the best, most responsible course of action a practitioner can take is to give up and let them be.

At least I think that's the message.

"every diet trial has been, pardon the expression, a big fat flop."

The word "diet" itself essentially implies failure because it suggests nothing more than a temporary (say, for the duration of a clinical study) measure. If people undertake a given nutritional regimen only to later abandon it and return to former eating patterns, I don't think anyone should be shocked if they return to their pre-diet weights.

Here's what would be a stunner: A sedentary person with severe NIDDM who weighs 250 pounds and takes in 3500 kcal a day starts walking 3 to 5 miles a day and reduces intake to 2500 kcal/day. After six months, said person is down to 180 pounds, with a marked reduction in diabetic symptoms. This person maintains his new lifestyle, yet balloons back up to 250 and is again severely diabetic. Anyone ever hear of this happening?

There is a role for common sense in this problem.

Another point I'm surprised hasn't been made: the choice of species in the animal models is wrong. Rats and mice, while vaguely similar to humans, have a vastly different ecology, and thus many physiological responses will be different. We're an omnivorous, K-selected, large mammal with minimal predation risk. Rodents are primarily herbivorous, r-selected, small mammals with high predation risk. I'm by no means trying to undermine animal studies, just pointing out that you cannot blithely assume that "a mammal is a mammal is a mammal". I guarantee that you'll get different results of digestive physiology from a rat, a dog, and a raccoon. Use raccoons, or even better, primates.

Also, the "energy in - energy out" model is severely deficient, as it fails to account for how basal metabolism can change, or how the digestive system and upregulate or downregulate the absorbtion of fats, sugars, etc.

Finally, one thing I'm amazed hasn't been mentioned: stress. That is, hands down, the biggest change in our society in the past 50 years, and is known to have wide-ranging effects.

One thing that Mokele mentioned is the lack of metabolic change after a large amount of liposuction. Note that liposuction only takes out the subcutaneous fat - in this case fat surrounding the abdomen [which is called abdominal fat in the ScienceDaily article]. It doesn't touch the intraperitoneal and pericardial fat that is the problem. Also, what remains are adipocytes that remain the same size. So there should be little or no change in the metabolic effects of the fat that remains.

The real study on reducing fat for positive metabolic changes is to remove intraperitoneal fat - through major abdominal surgery. This might be done during abdominal surgery for some other reason, but it would increase risks to the patient probably to an unacceptable level in many cases.

And there is more to the problem than clories in/out. Caloric intake and output is certainly the major agent in obesity, but other factors need to be mentioned: stress, lack of sleep, genetics, bone growth/stature/somatotype, yo-yo effects, etc.

kemibe:

Here's what would be a stunner: A sedentary person with severe NIDDM who weighs 250 pounds and takes in 3500 kcal a day starts walking 3 to 5 miles a day and reduces intake to 2500 kcal/day. After six months, said person is down to 180 pounds, with a marked reduction in diabetic symptoms. This person maintains his new lifestyle, yet balloons back up to 250 and is again severely diabetic. Anyone ever hear of this happening?

This shouldn't happen - if a person maintains a lifestyle to get from 250 to 180, h/she will not gain the weight back while maintaining the acquired lifestyle. The will stay around the new weight. If weight is gained, then the lifestyle was not maintained and old habits have come back.

There seems to be a definitional problem with the idea of what should be considered an effective treatment. Which is more correct?:

"The treatment was not successful because most of the subjects couldn't remain on the program".
or
"The treatment was effective because a few of the subjects remained compliant and succeeded while those who could not be compliant did not succeed".

This has been the long story of weight loss regimes See here for more very successful anecdotal cases. The fact that some have succeeded with weight loss seems to show successful treatments exist - the programs work, it's just that the patients don't.

Finally, dismissing the banding studies I think is an error as it reinforces the underlying point that adiposity itself is contributing to diabetes.

Actually, I don't dismiss banding. Sorry if I gave that impression. It's one of the few really effective methods for inducing weight loss with more than anecdotal success levels. However, it has several problems: The surgery is dangerous in itself, there can be sequelae including vitamin deficiencies and, of course, scarring. In short, it's not suitable for situations other than severe/morbid obesity. There might be room for further study of improved surgical techniques (laproscopic minimally invasive banding of small sections of the stomach?) that might expand the usefulness of the technique, but I doubt we'll ever have a good surgical option for people with BMIs of 25-28. (Of course, one could question whether they need any intervention: the CI on mortality at BMIs of greater than 18 and lower than 28 includes one.)

The basic problem is that we're fighting evolution on this one: for most of human existence lack of food, not excess, was the common problem. So the survivors are likely to have genes telling them to eat when they can...Gene therapy for weight loss? Not until we get the gene insertion under better control, but maybe some day.

"[Banding is] one of the few really effective methods for inducing weight loss with more than anecdotal success levels. However, it has several problems: The surgery is dangerous in itself"

I totally agree. And as I recall from my days as a diabetes/metabolic syndrome X researcher, the resolution of diabetic symptoms with banding--even before weight loss had yet to set in--was astounding and nothing any sane person could discount.

Yet I have to wonder what will become of these people in 50 years, for we have no cohort.

Sorry, that last one was from me and the anti-anti-spy interface let me get away with it.

@ Mokele - FAIL

Another point I'm surprised hasn't been made: the choice of species in the animal models is wrong. Rats and mice, while vaguely similar to humans, have a vastly different ecology, and thus many physiological responses will be different. We're an omnivorous, K-selected, large mammal with minimal predation risk. Rodents are primarily herbivorous, r-selected, small mammals with high predation risk. I'm by no means trying to undermine animal studies, just pointing out that you cannot blithely assume that "a mammal is a mammal is a mammal". I guarantee that you'll get different results of digestive physiology from a rat, a dog, and a raccoon. Use raccoons, or even better, primates.

Ahh, the animal models excuse. You know, as a scientist who works with animal models, do you think we don't know this? But the cellular physiology is often preserved across mammals and teaches us a lot. Yes, we know mice are different, but probably not so different that their fat has endocrine properties and ours does not. After all, we're not really talking about organismal diet studies in rodents here, we're talking about the basic physiology. And we don't just use rats and mice, we also use pigs, and dogs, and even non human primates.

Also, the "energy in - energy out" model is severely deficient, as it fails to account for how basal metabolism can change, or how the digestive system and upregulate or downregulate the absorbtion of fats, sugars, etc.

I was not aware that basic metabolism could change in such a way that energy is created from nothing. Are you perhaps cooking your burrito on the edge of an event horizon of a black hole?

Calories are a measure of energy contained in food. The basic metabolism of humans is measureable, known, and not hugely different between individuals. Some people who are constitutionally skinny, if anything, don't have higher metabolisms so much as they do defective absorptive capacity. The number of molecules of ATP you can get from a glucose molecule is constant, and individual variation in "metabolism" is often just a weak excuse for "sedentary". This has, in fact, been studied extensively. Using techniques like the double-water technique or measuring CO2 production you can estimate the basal metabolic rate of a human being - we do it sometimes in the hospital. What do we find is the case in the obese? Their metabolic rates are higher! More tissue, more work on the heart and lungs, more work to move around and consuming more calories results in a higher basal metabolic rate. Thus, the obese require the consumption of greater amounts of energy than a thin person who has a low basal heart rate, less mass to move and less tissue to perfuse. It's not a slow metabolism in the obese, it's a fast one that demands more input. And when you diet, the basal metabolic rate decreases - making diet alone a difficult endeavor.

So what is up with this slow metabolism thing that people claim. They eat like birds and never lose weight! Well, there are dozens of studies showing this if you want to pubmed it - that is that overweight and obese people (and lean people when asked about consumption specifically of unhealthful foods) underestimate their intake. They do the studies by watching patients who have failed multiple weight loss trials and compare their estimations of intake and activity to actual activity measured in physical work and nutritional intake or in a proxy measurement of metabolism like CO2 production or double water experiments. They aren't even aware of their difficulty in assessing their behavior - poor personal insight into personal behavior is a common bias that humans share. So "I have a slow metabolism" is about as good an excuse as "I have big bones", the variation in individual metabolism or absorption is not an explanation and would not explain how someone could get more energy from food than actually exists or why the obese require more energy and expend more energy to support their body mass.

It's not the metabolism it's the intake and output, and yes, it is that simple. Human bodies don't violate laws of physics or biology and there is simply no support in the scientific literature for any of your arguments.

See Here here or here, or here or here, or here for some examples of this research.

I'm amazed the Swedish Obese Subjects Study (SOS) hasn't come up.

You want an interventional study where the intervention (obesity surgery which causes 23% weight loss) causes regression of diabetes, hypertryglyceridemia, HDL, and hypertension?

Not to mention lower mortality.

Sjostrom et al NEJM 2004 v 351 p2683
Sjostrom et al NEJM 2077 v 357 p741

Back in the day, when obesity rates in the U.S. were more manageable, you never saw talk like this. People were given the unfortunate medical facts and accepted them.

It's different now. I'm certain that if 6% of the population rather than ten times that were not overweight, none of this denialist bullshit would be afoot. But the reality is that America chooses to react strongly in terms of its own self-interest: "Well, if that many of us are fat, it can't be bad!"

Yes, it can. Do something about it, or not. But at a minimum, quit blaming others.

Some people just have to work harder and make more effort to be healthy. Life is not fair and too many people resent that fact enough that they look for a blame excuse rather than suck it up and deal. We do a lot of harm when we coddle fragile egos from the facts.

Mark, the reason that the obese have a higher metabolic rate is because they dont have enough mitochondria. That pushes the mitochondria potential higher to produce the same ATP, which causes increased slip, and reduced ATP production per molecule of O2 consumed.

Not enough mitochondria causes more ATP production from glycolysis which takes 19 times more glucose than oxidative phosphorylation. Switching 5% of ATP production from mitochondria (because there are not enough) to glycolysis takes twice as much glucose delivery.

The only way to get twice as much glucose delivery is by increasing glucose concentration (hyperglycemia). But glucose transport into cells is active and only through GLUT transporters. To get more glucose into cells more GLUT transporters are needed. Insulin causes expression of GLUT transporters, so you need more insulin.

Cells dont get glucose and insulin from bulk blood, they get them from the extravascular fluid that is actually in contact with those cells. Intervening cells consume glucose and insulin, so the first cells in contact with extravascular fluid have to not take in that glucose and insulin, they have to be come glucose resistant and insulin resistant so as to leave some glucose for the cells that are farther away.

If your body processes 100% more glucose to make ATP via glycolysis, where does the lactate go? Does the liver have the metabolic capacity to make twice as much glucose via the Cori cycle? ATP to run the Cori cycle has to come from mitochondria. If there are not enough mitochondria, the liver cant make glucose from lactate.

So what can the body do with lactate? It can make fat. Every cell can make fat. As bad as subcutaneous depot fat is, visceral fat is worse. As bad as visceral fat is, ectopic fat is worse. When you start forming ectopic fat in the liver (aka NASH), you are in doodoo. When you start making ectopic fat in skeletal muscle, you are in deep doodoo. When you start making ectopic fat in the heart and kidneys, you need to make out your will.

It turns out that it is NO that triggers mitochondrial biogenesis and it is low NO that causes you to not have enough and starting this cascade. If you are morbidly obese and go into ketosis following stomach banding, the ketone bodies bypass the GLUT transporters. Once all the excess fat is gone, and you are not in ketosis any more, the vasculature still cant supply enough glucose to those cells that need it, so they send out starvation signals. But you cant eat enough to supply enough glucose so you become hypoglycemic where it really matters (in the peripheral tissues where glucose levels cant be measured). This is what causes the depression that often follows stomach banding.

How do we know that obese people don't have enough mitochondria?

Not enough mitochondria causes more ATP production from glycolysis which takes 19 times more glucose than oxidative phosphorylation. Switching 5% of ATP production from mitochondria (because there are not enough) to glycolysis takes twice as much glucose delivery.

So where is all that lactic acid going?

Obese people are observed to not have enough mitochondria because they are exercise intolerant. They get tired when they try to exercise. That is an absolutely clear sign of not enough mitochondria in their muscles. If they had enough mitochondria in their muscles, those mitochondria would have the capacity to produce enough ATP to run their muscles and they wouldn't get fatigued so quickly.

The increased basal metabolic state observed in obesity is also indicative of too few mitochondria because when fewer mitochondria make the same ATP they do so less efficiently. Needing the same ATP but made less efficiently looks like increased basal metabolism because more substrate (O2 and glucose) must be consumed to make the same ATP. Basal metabolism is measured by O2 consumption, not ATP production. What cells need is ATP production, not O2 consumption.

The lactic acid (the same as lactate) is turned into fat.

A quickie: Why do big adipocytes release more free fatty acids?

BTW, thanks for the physiological explanations! The studies make a lot more sense to me now.

Exercise intolerance is seen in a wide variety of cardiac and pulmonary disorders. I suppose you could see it if a patient had some sort of mitochondrial problem, but it's certainly not specific for that sort of thing.

It's true that decreased mitochondrial deficiency would result in an increased oxygen consumption, but this could be a result of a defect in the mitochondria themselves, rather than an actual decrease in number. To show that they have fewer mitochondria, I think you'd need to compare actual electron micrographs of muscle biopsies from obese and normal people.

Daedalus, cite a paper that shows the obese have fewer mitochondria per cell or drop it. You're reasonng backwards from effects, not from primary observation.

Exercise intolerance is only caused by not enough ATP inside muscle cells for the muscle cells to continue to generate work after a period of ATP consumption due to exertion. In virtually all cases there is no loss in strength (i.e. acute ability of muscle to exert a static force) because there is no decrease in ATP levels at rest. As the ATP is consumed, it is not replaced at a sufficient rate, so ATP depletion occurs and fatigue happens.

Generation of ATP at a high rate requires mitochondria which require O2 and substrate. Fatigue can occur due to not enough O2 as well as due to not enough substrate. O2 can be supplied as inhaled pure O2, which does not resolve the fatigue of exercise intolerance. Similarly exercise intolerance is observed even under conditions of hyperglycemia. If O2 or substrate are not limiting ATP production, something else must be. The necrotic lesions that occur during an acute myocardial infarction occur due to ATP depletion due to ischemia. Heart muscle has the ability to work itself to death, to consume what ever ATP it has the capacity to generate until it dies from ATP depletion. Skeletal muscle has the same ability, but dying muscle signals its overload status by sending pain signals.

Visual identification of mitochondria under a microscope does not indicate their operational status. Mitochondria are replaced because they get tired. They become damaged and dysfunctional and are unable to produce ATP at a high rate efficiently. A tired mitochondrion and a fresh mitochondrion look much the same.

As I mentioned, a 5% reduction in ATP production from mitochondria would require twice as much glucose delivery to the cell to make up for that with glycolysis. Can a 5% deficiency in mitochondria number or activity be measured by looking at or by counting mitochondria? No, it cant. Mitochondria change in volume, and it takes much more sophisticated measurements to tell if a particular mitochondrion is working at what level of efficiency. These measurements are only destructive, there is no non-destructive way to measure the details of mitochondria function.

Mitochondria number is regulated in each cell independently of every other cell, although there is cross-talk so that different cells work "in sync". Mitochondria turn-over regularly, in the rat CNS they have a lifetime of about a month. In rat heart muscle it is about two weeks. If exercise intolerance persists for longer than a few weeks, the problem must relate to the regulation of mitochondria turn-over because a few weeks is enough time to substantially replace mitochondria.

Because multiple muscle cells regulate mitochondria number independently, for global exercise intolerance to occur, there must be some systemic effect that is interfering with the regulation of normal mitochondria biogenesis. Nitric oxide is the main trigger of mitochondria biogenesis. Systemic low basal NO levels would result in a reduced number of mitochondria in essentially every tissue. This would result in exercise intolerance and (to me) does explain the exercise intolerance observed in many different disorders.

There is no need to postulate multiple unknown mechanisms for exercise intolerance when a single known mechanism explains all the data.

Mark, Here is a nice paper, a twin study where the were discordant for BMI.

http://medicine.plosjournals.org/perlserv/?request=get-document&doi=10…

The heavier twin had reduced mitochondria DNA in fat indicating reduced mitochondria number. There was time series data for one individual over a BMI change, where they found: Longitudinal data accessible to us for one obese male further supported this suggestion: after this twins weight increased by 10 kg within 3 y, mtDNA copy number in his fat tissue declined by 44%.

Mark, here is another nice paper, specific to muscle and weight loss.

http://diabetes.diabetesjournals.org/cgi/content/full/57/4/987

Loss of ~10% of body mass via diet didnt change mitochondria number or activity in muscle, and didnt change exercise capacity. Exercise combined with diet did change mitochondria number and metabolic capacity, and it decreased exercise intolerance (increased VO2max).

This study was pretty small. They didnt find differences in mtDNA copy number, but did in terms of mitochondrial capacity (which is the more important parameter). The lifetime of mitochondria in various human tissues has not been reported.

Mark, "exercise intolerance" is a primary observation, it is data, as is accumulation of visceral fat. Which is cause and which is effect remains unknown.