Tuesday, December 10, 2013

Does "Metabolically Healthy Obesity" Exist?

Obesity is strongly associated with metabolic alterations and negative health outcomes including diabetes, cardiovascular disease, and some types of cancer (1234).  Excess body fat is one of the primary causes of preventable health problems and mortality in the United States and many other affluent nations, ranking in importance with cigarette smoking and physical inactivity.  Obesity is thought to contribute to disease via the metabolic disturbances it causes, including excess glucose and lipids in the circulation, dysregulated hormone activity including insulin and leptin, and inflammatory effects.  This immediately raises two questions:
  1. Does metabolically healthy obesity exist?
  2. If so, are metabolically healthy obese people at an elevated risk of disease and death?

Does metabolically healthy obesity exist?

Yes.  Several papers have documented that about 10-20 percent of obese people are "metabolically healthy", depending on how that is defined and what population is examined (567).  Metabolically, these people are often indistinguishable from lean, healthy people, and they even appear metabolically healthier than a subset of lean people who are considered "metabolically obese".

Do metabolically healthy obese people have an elevated risk of disease and death?

If we believe that we can measure obesity-related health risk using established metrics such as lipids, glucose, insulin, and blood pressure, then obese people that appear healthy by these measures shouldn't have an elevated risk of disease.  This makes good sense, but it was challenged by two recent studies that had a big impact both in the media and the scientific community (89).  The studies measured diabetes and heart disease risk in metabolically unhealthy vs. healthy obese, both concluding that even metabolically healthy obese people are at an elevated risk of disease*.

This conclusion provoked a backlash from some researchers, including Dr. Arya Sharma of the University of Alberta (8), a major proponent of the metabolically healthy obesity concept.  Dr. Sharma pointed out that the studies' definition of "metabolically healthy" allowed subjects to have metabolic risk factors (up to one in the diabetes study, and up to two in the cardiovascular disease study), and therefore that they were not truly metabolically healthy (9).

The main counter-evidence he presents is his own research, including a 2011 paper that examined long-term mortality risk among overweight and obese US adults according to the number of health-related risk factors they exhibit (10).  These presence or absence of these risk factors places each individual into one of five categories, which collectively form the Edmonton Obesity Staging System (EOSS) for evaluating the need for medical intervention:
Stage 0: No apparent risk factors (e.g., blood pressure, serum lipid and fasting
glucose levels within normal range), physical symptoms, psychopathology,
functional limitations and/or impairment of well-being related to obesity. 
Stage 1: Presence of obesity-related subclinical risk factors (e.g., borderline
hypertension, impaired fasting glucose levels, elevated levels of liver
enzymes), mild physical symptoms (e.g. dyspnea on moderate exertion,
occasional aches and pains, fatigue), mild psychopathology, mild
functional limitations and/or mild impairment of well-being. 
Stage 2: Presence of established obesity-related chronic disease (e.g.,
hypertension, type 2 diabetes, sleep apnea, osteoarthritis), moderate
limitations in activities of daily living and/or well-being. 
Stage 3: Established end-organ damage such as myocardial infarction, heart
failure, stroke, significant psychopathology, significant functional
limitations and/or impairment of well-being. 
Stage 4: Severe (potentially end-stage) disabilities from obesity-related chronic
diseases, severe disabling psychopathology, severe functional limitations
and/or severe impairment of well-being.
The paper is simple but quite meaty, and I could spend several posts discussing the data and implications.  I'm going to try to stay focused for today however.  I think the findings of the paper are fairly well summarized by the following graph, showing survival over time among overweight and obese people at EOSS stages 0-3.  In this graph, the faster a line drops, the faster the people in that group died:


What you can see is that among overweight/obese people with no established risk factors (EOSS stage 0), mortality over time is low (and not significantly different from lean people).  People with elevated risk factors (EOSS stage 3) die at a rapid rate, and everyone else is in between.  The paper showed that any way you slice and dice the data, whether you look only at overweight people, only at obese people, or only at severely obese people, the conclusion is the same: risk factors are the main predictor of mortality, not fatness itself.

This is why Dr. Sharma argues passionately that obesity itself is not a major rationale for treatment in a person who exhibits no established risk factors.  That being said, this study looked at mortality, not morbidity  (disease itself) as the other studies did.  It's possible that this explains part of the discrepancy-- some of the "metabolically healthy" obese people in Dr. Sharma's study could be sick but not dead.

Another important finding of Dr. Sharma's study lurks in the supplemental materials.  These data show that although some obese people do fall into the apparently healthy EOSS stage 0 classification, this only applies to less than 10 percent.  About 20 percent fall into EOSS stages 0 and 1 combined, meaning they are minimally impaired.  The other 80 percent of obese US adults have "established obesity-related chronic disease", "established end-organ damage", or "severe (possibly end-stage) disabilities".

One finding that was consistent across all three studies is that disease risk increases with body fatness, on average.  Metabolically healthy obesity exists and is fairly common, but it's not typical.

Conclusions

Here are my tentative conclusions:
  1. Obesity remains a major cause of disease and mortality in industrialized nations.
  2. Metabolically healthy obesity exists, although it only represents about 10-20 percent of obese people.
  3. Metabolically healthy obese people are probably not at an elevated risk of death.
  4. Metabolically healthy obese may or may not be at an elevated risk of obesity-related disease relative to lean healthy people, depending on how one defines "metabolically healthy".  
  5. Metabolically healthy obese are at a lower risk of disease than metabolically unhealthy obese. 
  6. Genetics, diet, and lifestyle probably determine who will be metabolically healthy vs. unhealthy at any given level of body fatness.  I believe it's possible to reduce health risks at any level of body fatness by living wisely, although losing fat is preferable if possible, mostly because it improves metabolic health in the majority of obese people.

* Both studies found that metabolically healthy obese people are at a reduced risk of disease compared to metabolically unhealthy obese, but still at a higher risk level than metabolically healthy lean people.  Both studies also confirmed that obese people have a much higher risk of diabetes and heart attack than lean people on average.

17 comments:

Shunderson said...

I appreciated the whole post but "simple but quite meaty" really made me happy.

Gretchen said...

It would be interesting to see a similar graph of people who started out obese and then lost weight.

IOW, does the obesity cause the higher mortality, or do other factors cause both obesity and higher mortality?

RLL said...

While I have always had a high BMI, my suspected obesity status goes back and forth from just obese to almost trim. Blood pressure under good control via meds, as is T2 diabetes (A1C under 5). Last VA visit I was advised leave my weight about where it is. Physically very capable in 8th decade of life. Occasionally in superb shape, strength, sprints etc.

I think my point is underlying genes/ environment/ diet/ exercise may cause a variety of strengths as well as problems. I can even blame some of my fitness on exercise that gives me a lot of night pain.

Steven Tursi said...

Wow, this post is exactly about me - BMI=32 but otherwise "metabolically healthy" by all measures (blood pressure, fasting blood sugar, HDL/LDL ratio & total cholesterol are not just normal, but my doctor was impressed.)

I wonder how much of my improvement is determined by the fact that I have lost a tremendous amount of weight, albeit not recently. Fifteen years ago my BMI was over 46. I've been in the 32 range for 5 years now and my numbers (besides BMI) continue to be spot-on. Daily exercise is also undoubtedly a contributor - I run 25-30 miles per week and 7500 miles over the last 7 years.

It's good to know that I probably won't die anytime soon, but I'm still motivated to lose weight. Even though I run 25-30 MPW, I'd like to be able to run them faster and more easily. Healthy as I am, I'm not going to deny whatever law of physics that states it takes more energy to move more mass.

suspiciouspatterns said...

Interesting - glad you're tackling this subject, since the prevailing wisdom is that obesity simply has to be bad news. I came across this fairly robust meta analysis the other day http://jama.jamanetwork.com/article.aspx?articleid=1555137 which suggests that it's only at BMIs of over 35 that obesity really starts to have a negative impact on health, whereas Grade 1 obesity (BMI 30 - 34) has a lower hazard ratio than being at a normal weight, i.e. could even be health protective?

Patricia Makenson said...

Interesting.
Not going to lie - I immediately checked to see if he's teaching any of my classes next term. Sadly, no.

Jad Patrick said...

Be interesting to know if there was any examination here of the impacts of stigma and psychosocial stress on the metabolically healthy obese vs metabolically unhealthy obese/lean. I know the HAES people have gone to great lengths to point out this is an understudied aspect of the effects of obesity and health. Anyone got any thoughts on this (love all the brains here !)

marchwinds said...

As with Gretchen, I wonder whether other factors affect the health of obese people - what about poverty? My understanding is that obesity is more prevalent in poor populations (at least in the US). Are more of the metabolically unhealthy obese poor? It would make sense that they would be, because poor people have less money to spend on exercise and sports, and less income for purchasing good food. They also have more stress. So how is their health affected by poverty? How does their health compare to thin poor people's health?

I am healthy and technical obese. Although if it weren't for the BMI, I would not consider myself obese. Just...typically middle aged! But apparently I am an unusual person - fat and healthy. I practice martial arts and do super-intense exercise all the time. I don't feel impeded in any way. I hope I don't die young :-)

Mark Rand said...

@steven: It's been shown that "fitness trumps fatness" In other words overweight and fit, better outcomes than thin and sedentary.

George Henderson said...

According to Jude Oben – who also works as a researcher at University College London – the key indication that a person is dangerously obese is not provided by their weight or by their body mass index, which is calculated by combining a person's height and weight. "You can have a high body mass index but be quite fit," he said. "In fact, the most reliable factor is simply your waist circumference."

FOR WOMEN

With waist circumferences of under 31 inches (79cm), there should be no major health problems. For those between 31 inches and 35 inches (89cm), the outlook is worrying. For those over 37 inches (94cm), the prognosis is alarming.

FOR MEN

The danger zone begins for those whose waists are greater than 37 inches, while those over 40 inches (102cm) face prospects that Oben also describes as alarming.

http://www.theguardian.com/society/2013/dec/14/obesity-tidal-wave-cirrhosis

Ed said...

Stephan, years ago you wrote about the difference in health between people with ectopic vs peripheral fat accumulation. More recently, Ned Koch proposed the waist-to-weight ratio as the single best weight-related health marker. Are you or other researchers still thinking that ectopic or belly fat to be meaningful risk markers?

richard hale said...

Great post! I know what you're talking about and I can tell you know also. There's a big difference between weight loss and reducing fat. I read a book by Phil Hoffman that discussed the big differences and how important nutrition is. I agree with Gretchen, I'd love to see a chart in that manner. Awesome post though, we'll be sure to revisit your blog. Thanks

Grinch said...

Stephan,

What do you think of Dr. Glenn Gaesser's work? Below is a paper he published regarding metabolic health and exercise being independent of body weight.

http://www.ncbi.nlm.nih.gov/pubmed/21673488

He went as far as to state on the Dr. Oz show that obesity is just a marker of a poor lifestyle but isn't the actual cause of metabolic derangement and disease. Do you think there could be some truth behind this, or do you think the obesity itself has to be a major cause from how you interpret the literature?

dbk_999 said...

This is theoretically interesting, but you would have to ask , how likely is it for someone to be obese in be 'stage 0' as opposed to 'stage 2 ' or higher? In other words, if they simply sampled random obese people for the study, how many would be in each group?

Travis Culp said...

I wonder what the gender breakdown would be for something like this. I suspect that the vast majority of healthy obese are females whose genetically-determined fat deposition pattern results in very little visceral fat accumulation.

Jane said...

I think we need to know whether healthy obese people have healthy adipose tissue. I just read a paper which suggests they don't.

RESULTS: Adipocyte expression of most markers of ER stress ... were significantly correlated with BMI and percent fat (r > 0.5; P < 0.00001). ... ER stress markers were only modestly correlated with SI [insulin sensitivity].

'Endoplasmic reticulum stress markers are associated with obesity in nondiabetic subjects'
http://www.ncbi.nlm.nih.gov/pubmed/18728164

This wasn't visceral fat, which one might expect to be unhealthy, it was subcutaneous fat. ER stress means the cell is having trouble producing secretory proteins, which include leptin. It could also mean mitochondria aren't working too well. If they aren't, the cell might store more fat than it burns.

Yoni Freedhoff said...

Hi Stephan,

Am late to your post here, but there was an appendix in the EOSS study that was quite interesting too wherein it would seem that the proportion of folks with EOSS 0 is actually going down.

Wrote a quick post back when discussing same and wanted to steer you (not to the post per se, but to the graphs in the appendix) there as if true, it's both fascinating and important.

Best,
Yoni