Relationship Between Low Cardiorespiratory Fitness and Mortality in Normal-Weight, Overweight, and Obese Men

Ming Wei, MD, MPH; James B. Kampert, PhD; Carolyn E. Barlow, MS; Milton Z. Nichaman, MD, ScD; Larry W. Gibbons, MD, MPH; Ralph S. Paffenbarger, Jr, MD, DrPH; Steven N. Blair, PED


Context Recent guidelines for treatment of overweight and obesity include recommendations for risk stratification by disease conditions and cardiovascular disease (CVD) risk factors, but the role of physical inactivity is not prominent in these recommendations.

Objective To quantify the influence of low cardiorespiratory fitness, an objective marker of physical inactivity, on CVD and all-cause mortality in normal-weight, overweight, and obese men and compare low fitness with other mortality predictors.

Design Prospective observational data from the Aerobics Center Longitudinal Study.

Setting Preventive medicine clinic in Dallas, Tex.

Participants A total of 25,714 adult men (average age, 43.8 years [SD, 10.1 years]) who received a medical examination during 1970 to 1993, with mortality follow-up to December 31, 1994.

Main Outcome Measures Cardiovascular disease and all-cause mortality based on mortality predictors (baseline CVD, type 2 diabetes mellitus, high serum cholesterol level, hypertension, current cigarette smoking, and low cardiorespiratory fitness) stratified by body mass index.

Results During the study period, there were 1025 deaths (439 due to CVD) during 258,781 man-years of follow-up. Overweight and obese men with baseline CVD or CVD risk factors were at higher risk for all-cause and CVD mortality compared with normal-weight men without these predictors. Using normal-weight men without CVD as the referent, the strongest predictor of CVD death in obese men was baseline CVD (age- and examination year-adjusted relative risk [RR], 14.0; 95% confidence interval [CI], 9.4-20.8); RRs for obese men with diabetes mellitus, high cholesterol, hypertension, smoking, and low fitness were similar and ranged from 4.4 (95% CI, 2.7-7.1) for smoking to 5.0 (95% CI, 3.6-7.0) for low fitness. Relative risks for all-cause mortality in obese men ranged from 2.3 (95% CI, 1.7-2.9) for men with hypertension to 4.7 (95% CI, 3.6-6.1) for those with CVD at baseline. Relative risk for all-cause mortality in obese men with low fitness was 3.1 (95% CI, 2.5-3.8) and in obese men with diabetes mellitus 3.1 (95% CI, 2.3-4.2) and as slightly higher than the RRs for obese men who smoked or had high cholesterol levels. Low fitness was an independent predictor of mortality in all body mass index groups after adjustment for other mortality predictors. Approximately 50% (n = 1674)of obese men had low fitness, which led to a population-attributable risk of 39% for CVD mortality and 44% for all-cause mortality. Baseline CVD had population attributable risks of 51% and 27% for CVD and all-cause mortality, respectively.

Conclusions: In this analysis, low cardiorespiratory fitness was a strong and independent predictor of CVD and all-cause mortality and of comparable importance with that of diabetes mellitus and other CVD risk factors.

JAMA. 1999;282:1547-1553

I have a problem with this conclusion.

I decided to contact the Cooper Institute and ask a few questions. So that you understand I was not trying to be rude when I sent this letter, I did not know I was sending it to Steven Blair, as his email address did not specify his full name. Therefore I wrote "To whom it may concern."

On the JAMA website page where this study was published, the sidebar had a "comments" heading. Clearly they welcomed commentary and/or questions.

The following is a word-for-word dialogue with Stephen Blair. I changed nothing.

11/04/99

To Whom it may concern,

I am curious as to how muscular strength was factored out of the analysis in this study.
Clearly a low-level of lower body strength will cause a poor result on a treadmill fitness
test. Strong legs will perform quite well regardless of VO2 capacity.

How do you know it was not a low level of muscular strength that was the cause of a
higher incidence of all-cause mortality in this study?

Sincerely,

Fred Hahn

Here is Mr. Blair’s response:

Dear Mr. Hahn,

I am afraid that I do not agree with you that maximal exercise test
performance on a treadmill (using a protocol similar to the one used in our
study) is determined by lower extremity muscular strength. I am not saying
that there is no influence of muscular strength on cardiorespiratory
fitness, but I think the effect is minimal. I would be interested in
learning about any peer-reviewed articles on lower extremity muscular
strength and treadmill test performance.

Steve Blair


Notice that he did not close with a polite ending.

Here is my response to Mr. Blair:

Dear Mr. Blair,

Thank you for your prompt response. I appreciate it very much.

My point WAS that muscular strength or the (lack thereof) can influence the
outcome of any study attempting to measure CV fitness. And, as I'm sure
you're aware, for a study to suggest anything, all possible variables that
could influence the outcome must be factored out. By your own admission (and
you are indeed correct) you agree that muscular strength could have had an
influence. This was my point.

Keep in mind also that it is literally impossible to prove cause and effect
from epidemiological studies. The most that an epidemiological study can do
is alert us to possible factors that contribute to a disease. We must then
devise and conduct direct, controlled experiments to determine if the
previously observed association does in fact cause or contribute to the
disease. The study simply illustrates an "association." This is not evidence
of "cause and effect."

Having said that, clearly in order to test the heart and vascular system you
have to go through the skeletal muscles. The condition of the skeletal
muscles, therefore, exerts an influence.

For example, if you were to test a sedentary individual on a treadmill test
much like the one in your study, record the outcome, and then for the next 3
months do nothing but strength train the subject *properly* increasing his
muscular strength significantly over the 3 month period (without adding any
traditional aerobic training during the same time), and tested him again at
the end of the 3 months on the same treadmill test, his results would
improve markedly.

Conversely, if a marathon runner sustained a minor knee injury requiring
meniscal surgery, the surgery would severely atrophy his quadriceps muscles.
The surgery itself takes all of an hour. Recovery is quite rapid these days
due to the advanced techniques used and he will more than likely be able to
run again a few days later. However, his injured leg will be very weak
post-op. His run time will now suffer. And suffer not because his
cardiorespiratory/cardiopulmonary/cardiovascular health will have deteriorated
in these few days of not running, but suffer because he is weaker.

Not only could muscular strength "muddy the waters" so to speak, genetically
inherited, already present heart or vascular disease could have caused the
increase in mortality as well. Were the subjects in the study all given
angiograms to see if they already had arteriosclerosis? Or were they simply
tested on a treadmill (which, of course, cannot detect lurking vascular
disease) and assumed disease free? You know as well or better than I do that
just because one can perform feats of incredible vascular fitness, this does
not mean one is free of vascular disease. Runners die from heart attacks all
the time -- almost with the same frequency as non-athletes. To quote
cardiologist Henry Solomon MD from his book 'The Exercise Myth':

"Most of the improvement in functional capacity due to exercise is not even
directly related to the heart. It is due to an effect on the peripheral
muscle cells whereby they more efficiently extract oxygen from the blood."

"Not only is superb physical performance possible in the presence of severe
coronary heart disease, but also the person may himself not feel the
symptoms. I know patients of exceptional fitness who have severe coronary
artery disease. Even people with imminently fatal heart disease can play
sports, exercise and run. They may have no symptoms and may be capable of
outstanding physical performance with hearts that will kill them."

Bruce Charash, MD current head of cardiology at Lenox Hill Hospital in NYC
agrees. From his book Heart Myths:

"When patients participate in exercise programs, they often assume that
their heart becomes stronger. This is not the case. Physical training
results in a sense of well being because of other effects. It improves the
efficiency of the muscles. It improves the hormonal tone of the body. It
improves the control of sugar in people with diabetes. However, exercise
will not make the heart beat more strongly."

The late George Sheehan MD known as the "Guru of Running" had this to say:

"You might suspect from the emphasis on cardiopulmonary fitness that the
major effect of training is on the heart and lungs. Guess again. Exercise
does nothing for the lungs that has been amply proved...Nor does it
especially benefit your heart. Running, no matter what you have been told,
primarily trains and conditions the muscles."

Devoid of peer-reviewed studies, it is fairly common knowledge these days
that increasing muscular strength will increase aerobic capacity -- perhaps
not as much as performing formal aerobics, but increases it nonetheless.
(This is one of the reasons that runners and other endurance athletes these
days are taking up strength training at an ever increasing rate.) Dr. Ted
Lambrinedes, head of research at MedX Corporation in Ocala Florida,
www.medxonline.com can steer you towards a large body of peer-reviewed
studies which support this. Dr. Ellington Darden www.classicx.com can also
aid you in your search for such information. One study that I know of is by
Messier and Dill, Research Quarterly for Exercise and Sport, 1985 Vol. 56,
No. 4 pp.345-351. Their conclusion:

"The results of this study suggest that for a training period of short duration, Nautilus
circuit weight training appears to be an equally effective alternative to standard free-weight (strength) and aerobic (endurance) training programs for untrained individuals."

The authors state on page 348 second paragraph, that there was a significant
increase in VO2 max in the Nautilus group and that "There was no significant
difference between the Nautilus and Run groups." in VO2 max.

Another is by Goldberg, Elliot, Kuehl, Human Performance Laboratory,
Division of General Medicine, Oregon Health Sciences Services, Portland
Oregon published in the Journal of Applied Science Research 1988 Vol. 2, #3,
pp. 42-45. Their conclusion:

"Traditional, non-circuit weight training for both the athlete and the
general population can be viewed as a method of reducing myocardial oxygen
demand during usual daily activities. This cardioprotective benefit allows
the individual to perform isometric exertion combined with dynamic work with
lower cardiac oxygen requirements, and, thus, improvement in cardiovascular
efficiency. Although standard methods of weight training and strength
acquisition may not improve running, cardiovascular benefits do occur."

I have others.

And also keep in mind that obese people find it much harder to run on a
treadmill than to cycle on an exercise bike. The reason I think is pretty
obvious -- they don't have to cart around all the extra weight on the
bicycle. Perhaps the outcomes would be different using different devices.
Perhaps not, but I think it likely.

It would be interesting for your organization to perform a study using
proper strength training techniques as to its influence on aerobic capacity.
Of course, the strength training would have to be performed under controlled
conditions, with good equipment, using a high intensity protocol and expert
supervision. Otherwise, left to their own devices, most people will not
strength train hard enough.

Assuming the outcome to be positive, perhaps we would learn that what we
should be teaching the general public is not that they need to be physically
active or cardiovasculary fit, but muscularly fit -- best achieved through
proper strength training.

I'd be more than happy to help with such an endeavor. I know many other
physiologists who own and run successful fitness facilities who'd
participate as well. After all, like any legitimate science, it's all about
truth, furthering knowledge and learning what's best and what's safest for
people in order to live longer, healthier and happier, isn't it?

I'd enjoy keeping our dialogue open and ongoing.

Sincerely,

Fredrick Hahn, President
Serious Strength Inc.
National Council for Exercise Standards

Here is Mr. Blair’s response:

Dear Fred,

I repeat, can you refer me to articles in peer-reviewed scientific journals
in which lower extremity muscular strength is strongly correlated with
aerobic power? I am not impressed by the quotes from those such as Henry
Solomon, who has not conducted and published research studies.

Since you state that epidemiological studies cannot prove cause and effect,
I take it that you do not believe that smoking causes lung cancer or that
thalidomide is a cause of birth defects. Neither of these examples were
based on controlled experimental trials.

This time no closing at all.

And my response:

Dear Steve,

I believe the two examples I included in my letter to you were/are peer reviewed and
published in scientific journals. Am I incorrect on this? Please explain.

And simply because you do not know of or possess peer reviewed studies which
indicate that stronger muscles increase aerobic power does not mean that these
studies do not exist. With all due respect, you need to become more familiar with
the research on the subject. No offense intended but it is not for me to educate you.
I am merely pointing out the facts. The information is out there whether you are aware
of it or not.

But for the record here's another. This is a peer-reviewed study printed in a scientific
journal:

The Usefulness of Weightlifting Training in Improving Strength and Maximal
Power Output in Coronary Artery Disease, McCartney Ph.D., McKelvie MD,
Haslam MSc, Jones MD, American Journal of Cardiology May 1991 pp. 939-945.

The authors conclusion:

"In patients with CAD, combined weight lifting and aerobic training was a
more effective method of increasing aerobic performance and strength than
traditional aerobic training alone."

Again I have many more studies in my library but I think I've done more than enough
to enlighten you to the fact that muscular strength can greatly influence CV fitness.

As for your examples, smoking and Thalidomide, realize that not all people
who smoke contract lung cancer. Many people who don't smoke DO get lung
cancer. Therefore, it is not smoking per se that causes lung cancer. That's
why the side of cigarette packs say MAY cause lung cancer and MAY cause
birth defects. Your study likewise should state that a low level of CV
fitness MAY cause premature death since you did not factor out many other
influential variables. Again, this is my point. I am not saying you are
wrong. What I'm saying is that it is misleading to state absolutes when the
evidence does not yet exist. The study's conclusion that a low level of CV
fitness IS a risk factor for premature death is incorrect. The conclusion
should be stated as:

"It appears that people who perform poorly on a particular treadmill test MAY be
at risk for premature death."

The two are quite different.

As for the Thalidomide example, this is different and you know it. Please let's
refrain from "playing the party game" as it were. I'd like to keep this dialogue
rational, logical and professional. I'm sure you do to.

Sincerely,

Fred Hahn


So far I have not received a response from Mr. Blair.

I leave you all to draw your own conclusions.

Final Comments:

By Tim Ryan

"We only have baseline data on fitness, other exposures, and weight, so we do not
know if changes in any of these variables occurred during follow-up or from the
influence of possible changes on the results."

In the above quote the authors admit they only have the baseline data and have no
knowledge of what if any change occurred in these individual's health, fitness, lifestyle,
etc., since the time they received their physical. Furthermore, the subjects were not
screened for the presence of disease at the outset. 

Without knowing what changes occurred with these individuals over the years before
their death, they don't know what their level of cardiovascular fitness was at the time
of their death. Maybe some of these individuals took up an exercise program and
became very fit over the years and then still died anyway?  Their CV fitness didn't save
them. Maybe some of them already were simply unhealthy for a variety of different reasons
from the outset and never could exercise because of their poor health.  Most importantly,
since subjects were not tested for heart disease at the outset, it is quite possible that those
who originally tested poorly on the treadmill VO2 test may have already possessed heart
disease and this was the cause of the poor cardiovascular fitness level. Therefore,
their low fitness level did not cause their poor health and/or death, rather, their poor health
caused their poor score on the fitness test.  They didn't die from low CV fitness. They died
from disease.

Bear in mind that this study was NOT a direct, laboratory controlled study designed to
isolate and test the hypothesis that low CV fitness leads to premature death. No.   This was simply a post hoc epidemiological review of patient data and statistical analysis.  Therefore,
it is literally impossible to determine cause and effect of anything with this study.   However,
despite this fact of science, and despite having no knowledge of any of a multitude of factors
that may have influenced the outcomes studied, Blair and his associates have formed the conclusion that low cardiovascular fitness leads to an increase in mortality.

It is our position that this study, and others like it, are irresponsible and give science a
bad name.  They misinform and mislead the general public.  In this case the public is
frightened into believing that if they don't perform regular aerobic exercise and score well
on a particular treadmill test, they will die prematurely.