Does a "calorie not in" equal a "calorie out"?

Apparently not!

According to this study, how one creates a caloric deficient can make a difference.

The Impact of Exercise Training Compared to Caloric Restriction on Hepatic and Peripheral Insulin Resistance in Obesity
Context: It has been difficult to distinguish the independent effects of caloric restriction versus exercise training on insulin resistance.
Objective: Utilizing metabolic feeding and supervised exercise training, we examined the influence of caloric restriction vs. exercise training with and without weight loss on hepatic and peripheral insulin resistance.
Design, Participants, and Intervention: Thirty-four obese, older subjects were randomized to: caloric restriction with weight loss (CR), exercise training with weight loss (EWL), exercise training without weight loss (EX), or controls. Based on an equivalent caloric deficit in EWL and CR, we induced matched weight loss. Subjects in the EX group received caloric compensation. Combined with [6,62H2]glucose, an octreotide, glucagon, multistage insulin infusion was performed to determine suppression of glucose production (SGP) and insulin-stimulated glucose disposal (ISGD). Computed tomography scans were performed to assess changes in fat distribution.
Results: Body weight decreased similarly in EWL and CR, and did not change in EX and controls. The reduction in visceral fat was significantly greater in EWL (−71 ± 15 cm2) compared to CR and EX. The increase in SGP was also almost 3-fold greater (27 ± 2%) in EWL. EWL and CR promoted similar improvements in ISGD [+2.5 ± 0.4 and 2.4 ± 0.9 mg · kg fat-free mass (FFM)−1 · min−1], respectively.
Conclusions: EWL promoted the most significant reduction in visceral fat and the greatest improvement in SGP. Equivalent increases in ISGD were noted in EWL and CR, whereas EX provided a modest improvement. Based on our results, EWL promoted the optimal intervention-based changes in body fat distribution and systemic insulin resistance.
This study illustrates so compellingly the benefits of increasing calories out via exercise for weight loss and improved glucose homeostasis vs. restricting calories in.    Some of this will be repetitive of the summary above, but in plain English they took older (range 50-80, avg age for groups mid-to-late 50's) overweight/obese (BMI ~32) adults and assigned them to one of three study groups and a control group that maintained intake and activity levels to usual amounts.  The diets for all participants were 35% fat, 20% protein, 45% carbohydrate, and all subjects besides the controls were weight stabilized on this diet during a 4 week pre-intervention period.  Controls were instructed to continue on as normal but were adapted to this macro percentage for 4 days prior to testing.  The three study groups were:
  • CR = calorie restriction
  • EX = exercise training with calorie compensation to maintain weight
  • EWL = exercise training with normal intake

In the exercise interventions, they started with a 1000 cal/week expenditure and ramped it up to 2500 cal/week.  The CR group had the intake deficit matched accordingly and the EX group received supplemental calories to compensate and maintain energy balance.  The study lasted 12 weeks and the CR & EWL groups both lost 5kg = 11 lbs.   They used radiolabeled glucose, an octreotide/glucagon infusion, and multistage insulin infusion (MSI) to measure fasting glucose production (hepatic IR)  and glucose disposal (peripheral IR).  I note there was a significant post-intervention period of time where all subjects were stabilized back to energy balance (2 wk refeed + 4 wk stabilizing) before retesting.

The effect on fasting glucose production, a measure of hepatic insulin sensitivity, with exercise induced weight loss was almost 3 times that of the calorie restriction induced weight loss.  Exercise alone, without weight loss produced similar improvements in hepatic insulin sensitivity.   This was mirrored by over twice as much visceral fat lost by the EWL v. CR.  However while  EX had some visceral fat loss, it was about half that in CR.   Lean thigh tissue was also measured.  There was no statistically significant change in EX (+2cm2) while EWL gained a significant 7cm2 as CR lost a significant 7cm2.  

For insulin stimulated glucose disposal (mg per kg FFM per min), both EWL and CR produced similar significant increases from pre to post interventions of 2.45 ± 0.43  and 2.46 ± 0.69.  This was about 2.5X that observed with exercise alone as EX increased 0.95 ± 0.41.  

From this we can see that cardio equivalent to 2500 cal/week expenditure improves hepatic insulin sensitivity with or without weight loss and improved systemic glucose handling without weight loss. It's worth reminding that the EX group also ate 2500 cal/week more than their weight loss counterparts.  For many of us that's 1-2 extra "days" of food.  So eating a "surplus" amount of calories, they still experienced improvements in insulin sensitivity (which yours truly believes to be at the root of many diseases, and not insulin levels per se).   This is something worthy of consideration for those overweight proponents of their healthy diets who shun cardio for calorie burning.   

For all the talk about how a calorie in is not a calorie in based on hormonal status and fuel partitioning, and for all the talk of dietary manipulations to achieve a "fat burning metabolism" and lower insulin and all that ... we see that a calorie out is also not a calorie out!  Which doesn't do squat to discredit "Team CICO", but it should give pause to the Gateking who plans to roll out an Exercisegate bood/website/franchise in the future?!  What it does do, or should anyway, is give all of the "it's 80% diet" and "I proudly sit on my a$$" crowd on "Team TWICHOO" pause as to where calories are distributed or taken from depending on the "source" of the calorie deficit.  

Exercise doesn't make everyone hungry afterwards and not everyone compensates for exercise.  I believe that most of us can use exercise only to lose weight with just some attention to intake to avoid compensation.  If we don't maintain intake and therefore not lose weight, we still get some health benefits.  If we lose the same amount of weight as we would by diet alone, we lose more of it from fat and -- surprisingly -- may gain lean mass.  (I've seen this in other studies as well where lean mass is not only preserved but increased in caloric deficit/weight loss).  A caloric deficit through exercise improves hepatic insulin sensitivity, as measured by suppression of glucose production, by almost three-fold what a caloric deficit through calorie restriction produces.    More food for thought for those trying to achieve weight loss through diet alone (which amounts to calorie restriction whether you want to acknowledge reality or not).  How about if you're hopping into ice baths to burn a few extra calories?  I don't know, my money would be on it being more like calorie restriction than exercise over the long haul, but that's not a bet I'd make at this point either way.  Can we get a metabolic ward trial?  :-)  

Lastly, sure the weight loss is nothing to write home about, but 10 lbs over 3 months equates to 40 lbs in a year.  Since RMR correlates so highly with lean mass, guess which group is more likely to attain and maintain long term success with weight loss?  My bet would be on the EWL group.  Imagine, however combining caloric restriction and exercise expenditure to produce calorie deficit.  Nah, that's too much like eating less and moving more!  


Unknown said…
More evil cardio!! More evil CICO!! They need to do a study where people only eat road kill (with lots of butter for JM) and lift rocks!!
river rance said…
Great post Evelyn!
Unknown said…
The perils of Chronic Cardio, if you're not careful you might get healthier.

Says they were exercising at 50% of VO2 max on a bicycle, anyone know what intensity that would translate to? Walking briskly? Slow jog?
Interesting ... the exercise arms "began supervised cycle exercise training (50% of VO2peak) by expending 1000 kcal/wk, and gradually increased their training (500 kcal each week) until they were expending 2500 kcal/wk."

I'm not familiar with VO2 as a measure of exercise intensity, but if it's all similar to heart rate, this level of cardio appears to be in the lower/slower range. If so, other low/slow exercise, e.g., brisk walking, would presumably be just as beneficial.

I think that's important, as I think it's not just about how many calories you can crank out on the Stairmaster or elliptical. That's my read of this abstract (and the NYT's take). I think it's about being more active period.
Tsimblist said…
@Beth: I believe you are correct about the exercise intensity. 50% VO2peak would be on the low side of low/moderate according to this source:
Kindke said…
BBC Horizon showed us that high intensity exercise improves glucose homeostasis that is seemingly independent from total calorie expenditure, most of the people here have seen this right?

Not sure of the mechanism though, either something to do with crossing the lactate threshold, glycogen churning, or ROS from anaerobic metabolism?
Sonnenschein said…
I am surprised that cardio exercise was helpful to maintain/gain (?!) lean mass. I know about several other studies that focused on resistance training in order to maintain lean mass while restricting calories. Participants who only did cardio exercises also lost some muscle. The result here is very different from what other studies concluded!
steve said…
"insulin sensitivity (which yours truly believes to be at the root of many diseases, and not insulin levels per se)"
Is it possible that this may explain the following:
"Our studies have focused on a common, genetically
influenced metabolic profile, characterized by a predominance of small, dense LDL particles (subclass pattern B), that is associated with a two- to threefold increase in risk for coronary artery disease. We have found that healthy normolipidemic individuals with this trait show a greater reduction in LDL cholesterol and particle number in response to low-fat, high-carbohydrate diets than do unaffected individuals (subclass pattern A)." -from

I seem to fit this profile: normal weight, exerciser,TRGS only 60, BG 83 and yet produce lots of small particles even on a higher fat diet than this study. If i dramatically lower carbs, I virtually eliminate small LDL particles; on a Jaminet diet I do not. Maybe at times I do not have the fat level high enough, but I don't think it is the case. What is interesting that where carbs are not an issue in the weight realm, they may be in the heart disease area.
Thanks, that's helpful!!
Anonymous said…
One thing that I've changed: I now stand at my computer instead of sit. The effects of this (several months' worth) - too soon to tell.
Kindke said…
Just want to add that yeh all calories out are certainly not equal, my thesis would be that only exercise above a certain intensity is beneficial. Not all forms of "Move More" are equal.

If they were, people running heavy manual labour jobs would tend to be the healthiest alive. Is That true? They tend to be obese in my experience.

Below study shows that only exercise above a certain intensity increases markers for mitochondrial biogenesis.
120% lactate threshold for an hour? Yikes. I'm certainly hoping that HIIT gets you a similar bang for the buck.
CarbSane said…
Perhaps mitochondrial biogenesis is overrated ;-)

I would say construction work qualifies as heavy manual labor work. Not too many obese construction workers -- it's kinda prohibitive especially if you're talking work that involves climbing. The heavy ones tend to also be heavy drinkers after hours.

You can't out-exercise a shitty diet and excessive fructose, alcohol or fat consumption.
CarbSane said…
I think the jury is still out on particle size. Even triglycerides are highly diet dependent and excessive trigs on a Kitivan-like diet do not seem to be harmful. Seems glucose homeostasis -- good fasting and postprandial glucose handling -- indicates proper insulin signaling/reception with wide ranging lipid profiles. I'd like to see more work done on an oral fat tolerance test, as effective dietary fat trapping and clearance into adipocytes seems to be where the trouble starts there, and for the most part the liver "mops up" and redistributes what comes to it.
CarbSane said…
Re: VO2 max vs. heart rate ... from:

What results show, is that there is a very close correlation between heart rate reserve (HRR) and vo2 max (Shepard & Astrand, 2000, p.g., 14). To elaborate, HRR is calculated by subtracting your maximum heart rate by your resting heart rate. A classic example of maximum heart rate is the Karvonen formula, which is 220 minus your age. Say you have a 20 year old, with a resting heart rate of 60 beats per minute, this individuals maximum HR would be 200 (220-20) and his HRR would be 140 (200-60). So, say this individual wants to train at 65% of his VO2 max. He simply would times his HRR by .65, which would be 91 beats per minute. To translate this to a normal
heart rate, simply add the resting heart rate to the heart rate reserve. This would add to 150 beats per minute, or 75% of his maximum heart rate.

These folks averaged just over 50 yoa, but ranged up to 80. Using age 55, 220 - 55 = 165 and if resting heart rate is 60 as above, this makes HRR = 105. Thus 50% VO2 max = 53 so heart rate would be 103. 103/165 = 62% max heart rate.
CarbSane said…
The difference may be that some studies measure whole body fat and lean mass, whereas in this cycling study they saw increased lean thigh mass and measured only abdominal (visceral & subQ) fat depots. Still, this is promising IMO.
CarbSane said…

I think it has to do with efficiency. Anaerobic respiration is involved in high intensity activity and for each molecule of glucose you only about 5% of the ATP from anaerobic respiration vs. aerobic respiration. Use up glucose quite fast and have a little "metabolic advantage" for "wasting" carb calories.
CarbSane said…
Hmmmmphh ... in order to do the proper n=1 on this you have to grease the floor with 3 pounds of butter and wear teflon shoes. ;-)
"I believe that most of us can use exercise only to lose weight with just some attention to intake to avoid compensation."

I agree, Evelyn. However, in real world situations (as opposed to clinical studies) it doesn't work very well at all. Tends to work better for men than for women. Don't get me wrong: I'm a huge advocate of exercise.

The study at hand has significant implications for folks who really need to improve their insulin sensitivity: most prediabetics and type 2 diabetics. That being said, I note this is a very small study with only 8-9 participants in each study cohort.

Thanks for analyzing this research report.

ProudDaddy said…
I can't seem to find any reference to time exercised. As somebody in the same age group and in pretty good shape, I can maybe expend 10 kcal/minute going full out on a bike for a short period of time. So, it would seem reasonable that the subjects probably didn't expend more than 5 kcal/min. 2500 kcal divided by 5 yields 500 minutes per week or over 8 hours -- not likely for most people! I tell my friends to diet to lose weight and exercise to be healthy.

Please correct any math errors.

I was surprised by the failure to lose muscle until I realized the exercise was far from "cardio" -- more like almost-brisk walking (often called "fat burning" on exercise bikes). And 20% protein for this group might have been an increase from their pre-study intake. (Most of us seniors don't get enough protein.)
Tsimblist said…
500 minutes per week. That's 100 minutes/day for 5 days/week. Maybe that's why they call it chronic cardio.
Tsimblist said…
Your numbers are inline with what I came up with on this calculator:
Unknown said…
100 minutes of fairly low intensity exercise five days a week.

It's pretty hard if you aren't used to it. But there really is no easy way out. There is no effortless solution. You could approach it through serious caloric restriction but the problem is that we are hard-wired to eat when we are hungry, you can't continue to exist on 800 calories a day.

In the study they gradually increased the amount of exercise time on a weekly basis, that's the only way it works. You may have to take a more gradual approach. Start out walking 30 minutes a day, and add 2 minutes per week to your walk. Where does that put you in a year? Far in excess of 100 minutes per day. If you can't do 30 minutes, start out with 10 minutes, and add 2 minutes per week. You are still in excess of 100 minutes well within a year.
Harry said…
The whole "exercise contributes x% to weight loss while nutrition contributes y%" is a category error (i.e. like asking how important breathing is to maintaining life, as compared to how important eating is to maintaining life).

A better way to look at it, IMO, is to talk in terms of necessary and sufficient conditions.

An appropriate caloric intake is both a necessary and sufficient condition to maintain a given weight.

That is, caloric intake alone (without exercise) can sustain a desired body mass* (i.e. a sufficient condition).

In addition, an appropriate caloric intake is required in all cases of weight management, irrespective of the amount of exercise being done (i.e. a necessary condition) Evelyn put it, "You can't out-exercise a shitty diet" (strictly speaking, you can out-exercise a calorie-rich diet, but only to a point...there is a threshold caloric intake above which no amount of exercise will redress).

Meanwhile, exercise is neither a sufficient condition for weight management (exercise alone cannot ensure weight management irrespective of caloric intake) nor is it a necessary condition (you can maintain weight without doing any exercise at all).

So, in a logical sense, appropriate nutrition (in the narrow sense; I'm just referring to caloric intake) is prior to exercise, as it is both necessary in all cases of weight management, and it is sufficient to ensure weight management without exercise.

However, while the logic is instructive, it's not the whole story (unsurprisingly). The empirical story is equally as instructive. The experiences of most long-term weight loss maintainers is that exercise plays a crucial role in both encouraging dietary compliance, and in providing some extra 'eating space' in the weekly caloric allotment (allowing for psychological respite/enjoyment).

It also improves metabolic factors (as per the study) and improves mobility, aesthetics, strength, social life, confidence, bone mass, confidence, mood...etc.

So, here's my take-home message...nutrition is 100% of the weight management equation, and exercise is the other 100% ;-)


*Note that this does not account for body composition, or general health; only for body mass.
Mr. Wrong said…
Evelyn, Great post. I am however curios as to your thoughts on this study.
If calorie in/out is needed then how did the low-carb group lose a greater amount of weight than the calorie restricted group with greater retention? My only guess is that either through a lack food reward or a production of satiety via the diet types possibly led the low carb group to eat less calories?
Mr. Wrong said…
This comment has been removed by the author.
Mr. Wrong said…
I have also read that mitochondrial bio-genesis is triggered more so by aerobic activity and not so much by anaerobic. I have yet to see the construction worker dancing around the work site (ignoring village people.
Mr. Wrong said…
Would not the exercise also deplete glycogen stores allowing for glucose calories to be stored as glycogen vs. being converted to fat via lipogenesis?
Asclepius said…
'Is a "calorie out" a "calorie out"? '

Surely the question is 'Does a "calorie not in" equal a "calorie out"?
CarbSane said…
By gosh you're right! I'm going to change it!!
CarbSane said…
Ad libitum low carb always outperforms regimented low fat diets in the relative short run, which I consider 6 months to be. Since there was little monitoring and it has been shown time and again that LC and/or higher protein % spontaneously reduces intake. While self reports are for statistically similar calories, I have a follow-up study on another LC/LF comparison that produced similar results where they measured energy expenditure, etc. There were no differences in REE/TDEE and TEF actually favored the LF diet considerably. Since even the staunchest of low carbers will agree you can't lose mass without an energy deficit the only conclusion once everything else has been measured and ruled out is that the self-reported intakes had to be in error. Purely anecdotal, but most low carbers, especially their first time, will overestimate how many calories they really ate.
ProudDaddy said…
Ah, but the real question is what the heck are you doing up at 4:32 in the morning?!?
CarbSane said…
Yeah, the real world sucks ;-) In accordance to Leighan's comment on the NEAT post, I don't get, however, why so many seem to be more inclined to suffer eating less rather than moving more. I can see initially, when it's not really possible, but later on? This is where weight loss challenges, gym memberships and the like can be helpful for accountability which seems to help many (think Weight Watchers meetings).
Leighan said…
That's what I meant Evelyn. When weight loss is the goal I thought people would be more motivated to move more rather than look for other ways out.
Shalalie said…
I clean houses as a part time job. When I clean a house I'm constantly moving some part of my body for anywhere between one and a half hours to three hours at a time. I am able to maintain my weight or lose weight depending on how well I am watching my diet just with the hours per week I clean.
I am going to guess that the obese heavy labor people are taking in way more calories than they burn.
A lot of people don't choose healthy lunches while on the job and are tired when they come home and reach for whatever is closest to them for a meal. I don't believe it is all about intensity. I fully believe it is calories in and calories least for my body.