Losing weight over the long-term is difficult. Many people can lose weight during brief periods of concerted effort but most find they rapidly relapse to previous levels as soon as they take their eye off the ball. This rebound is such a common experience that it’s almost universal. It feels like there’s a powerful force intent on maintaining our body fat. But does the science corroborate this intuition? And if so, what are the physiological processes keeping us overweight? Lastly, and perhaps most importantly, how can we use this knowledge to create effective and lasting weight-loss?
A classic paper on this subject, published in 1995 by Leibel et al., looked at the metabolic effects of weight-loss (and gain). They took a group of 41 individuals, some of whom were obese and others who had never been, and studied them in a highly controlled environment. Their nutritional intake was limited to a precisely titrated liquid formula for the duration of the experiment so that caloric intake and macronutrient splits could be tightly regulated. The set-up allowed researchers to accurately calculate metabolic rates by recording the number of calories required to maintain a certain weight.
Once baseline measurements of total energy expenditure (TEE) had been taken, subjects were limited to 800 kcal/day until they’d lost 10% of their initial bodyweight. Their nutritional intake was then adjusted back to maintenance, until bodyweight was stable within 10g/day for 14 days. At this point, the researchers re-measured energy expenditure. Then the process was repeated until participants had maintained a 20% drop in bodyweight and final measurements were taken.
The results were pretty staggering. You would predict that a 10% loss in bodyweight would result in a 10% reduction in total energy expenditure. You might even think that TEE would fall by a bit less, given that you would expect most of the weight-loss to come from less metabolically active fat. By contrast, Leibel et al. found that a 10% weight-loss actually resulted in a 15% decrease in metabolic rate. In other words, your total energy expenditure falls by 50% more than you would expect under conditions of moderate weight-loss.
These effects were corrected for body composition: non-obese subjects saw their total energy expenditure fall from 45 kcal/kg of fat-free mass (FFM) to 39 kcal/kg FFM. Obese subjects saw theirs fall from 50 kcal/kg FFM to 42 kcal/kg FFM. In calorie terms, these effects are significant. The extra reduction in metabolic rate, over and above what you would predict, equates to 218 kcal/day for the non-obese and 244 kcal/day for the obese. On average, non-obese subjects’ maintenance calories decreased from 2380 kcal/day to 1952 kcal/day and obese subjects’ from 3100 kcal/day to 2549 kcal/day.
Looking at these numbers, it becomes immediately obvious why losing weight, particularly for more overweight individuals, is so difficult. Simply maintaining a reduced body-weight entails a hugely reduced caloric intake. This is counterintuitive and psychologically challenging. After losing weight, you expect the maintenance period to be something of a reward but in fact you have to work almost as hard just to retain your new state. Indeed, the study quantified the comparison between the weight-loss and maintenance periods. They found that daily energy expenditure is only 10-15% less during rapid weight-loss, eating 800 kcal/day, than it is while subjects maintained their lowered weights, eating 2,000-2,500 kcal/day. Losing weight and maintaining a lower weight are therefore similarly difficult.
The data also elucidates another intuition about weight-loss: that’s it’s much harder for some people than others. If you look at the figures above, you’ll see that the standard deviation of the metabolic compensation is 123 kcal/day for the lean subjects and 198 kcal/day for the obese subjects. In percentage terms, these standard deviations equate to 56% and 81% of the mean effect for lean and obese participants, respectively. That’s telling us that there is a huge variation in the degree to which people’s metabolisms compensate during weight-loss. So it is not surprising that some people seem to lose weight much more easily than others. That being said, in all of the studies I’ve seen, ALL subjects saw at least some degree of extra energy expenditure loss after losing weight.
The same applies on the other side of the spectrum. A 10% gain in body-weight resulted in an equivalent compensatory metabolic response: participants’ total energy expenditure increased by 16%, showing that it is just as hard to hang on to ‘gains’ as it is to hang on to abs.
Unfortunately, these effects may not be temporary. In a study comparing weight-stable subjects with those who had recently lost weight and those who had maintained a weight-loss for more than a year, researchers found that the metabolic compensations after sustained weight-loss were similar to those after recent weight-loss. To be fair, this study was conducted exclusively on obese subjects but the lesson is simple: you should not necessarily expect your metabolic rate to speed up, even after extended periods at your new weight.
Things get even more fascinating when you consider what happened when (obese) subjects further reduced their weight to a total loss of 20%. While a 10% drop in body-weight was associated with a 16% decline in total energy expenditure, a 20% drop in body-weight only results in a 23.5% decline in TEE. So at a 10% body-weight reduction, the metabolic response is 60% greater than expected while at a 20% reduction it’s only 35% greater. Again, you might have expected the opposite. As you perturb further away from the equilibrium point, you would anticipate a greater and greater physiological ‘pull’ back. You might have imagined a kind of spring-loaded system. In reality, you see the opposite. So what’s happening here?
In a 2001 paper with a similar experimental set-up, Rosenbaum et al. set out to answer this question. One hypothesis was that there is a threshold mechanism operating in metabolic adaptation. The idea is that the body wants to protect energy stores above a certain minimum level (the threshold). If energy stores dip below this threshold, physiological processes are activated to maximally reduce energy expenditure. Beyond this point, energy expenditure simply falls linearly with loss of weight since the body has already deployed its most drastic homeostatic tools (see graph below). If the average threshold level occurs before a loss of 10% of body-weight, that would explain why we see much stronger metabolic compensations at this level than at the 20% level.
Change in energy stores from initial
You can see this mechanism in action if you look at the way the components of energy expenditure vary at different levels of weight loss. After a 10% body-weight loss, half the drop in metabolic rate is accounted for by resting energy expenditure (REE) and half by non-resting energy expenditure (NREE). By contrast, after a 20% loss, only 30% of the drop in metabolic rate is due to REE, while 70% is down to NREE. You might explain this by saying that at the threshold, the body makes adjustments to its resting metabolism. But these are one-time, step-wise adaptations that cannot be further tuned to decrease energy output.
If this is what’s happening, the next natural question is: how is this happening? A lot of research has been done to characterise this weight-reduced phenotype. There are hormonal, nervous system and muscular aspects to the picture. Hormonally, weight-reduced people appear to have lowered levels of leptin, thyroxine and triiodothryonine (two of the thyroid hormones) which coordinate to favour weight regain by lowering metabolic rate and increasing appetite. After weight-loss, the sympathetic nervous system (the “fight-or-flight” branch of the NS) is less active. Finally, muscles become more efficient meaning that they consume fewer calories to perform the same amount of work. This last effect is fascinating and, in case you’re wondering, this increased efficiency is not just a result of carrying less weight; there are actually biochemical adaptations that occur at the level of the muscle cell.
Leptin appears to be the protagonist in this story. Leptin, as you have probably heard, is a hormone that regulates appetite. It is secreted predominantly by fat cells and is primarily interpreted by the brain (although also by other cells, e.g. muscle cells) as a signal of the body’s energy storage status. Low levels of the hormone are read as a warning of decreasing reserves and potential starvation. This is translated into a neurological drive to eat. In a study in 2005, Rosenbaum et al. investigated what happens to weight-reduced subjects when they are given exogenous leptin to bring the hormone back up to its pre-weight-loss level. As you can see in the diagrams below, increased levels of leptin reversed many of the detrimental changes in thyroid hormone levels, muscular work efficiency and sympathetic nervous system tone.
This resulted in an overall increase in total energy expenditure, closing the metabolic gap left by weight-loss. So then leptin is a wonder drug that can help us stay buff? Well, sadly, no. Leptin has been a dismal failure in obesity trials; extended periods of exogenous hormone use are rarely a good idea. Just ask the 22 year-old bodybuilder at your gym who looks 40. The problem is that, just as with insulin, the body can build up resistance to leptin signalling. Just as exogenous insulin is not your best bet to improve glucose metabolism, so leptin is not the answer to fixing your weight. This experiment merely shows us how the compensations of weight-loss are mediated in the body. In fact, when you think about it, many of these adaptations to weight-loss could be positive in the long-run. The overweight state of lowered muscular efficiency and heightened sympathetic nervous system tone is not one we necessarily want to protect.
This is all well and good, but if we can’t use leptin to improve our odds of holding on to weight-loss, how can we use this knowledge in our efforts to get, and stay, lean? In the first instance, it’s useful to be aware that you can’t slack off following weight-loss. You now know that the period after weight-loss is just as crucial as the preceding period. Armed with this information, you can prepare yourself for the extra effort; which will be well worth it if you can avoid the dreaded weight ‘yo-yo’. More than that, this data is the best argument I can think of in support of a gradualist approach to weight loss. If you rapidly lose 10% of your body weight, you face a situation where you have to quickly reduce your daily food intake by ~15%. That’s a huge and very difficult adjustment to make. But if you instead lose weight gradually over time, perhaps aiming for a 10% loss over a number of months, you can gradually lower your food intake making the adjustment that much easier. Reducing calories by 100 kcal per month over 6 months is much easier than reducing by 600 kcal in 1 month. The science shows us clearly why crash diets are destined to fail: they simply require too much discipline to maintain the weight lost. The great advantage of the gradualist approach is that you may also stay above your body’s threshold fat-storage levels, avoiding the most potent metabolic counter-punches. Finally, since we now know that to lose weight we necessarily have to find a way to eat significantly less, finding a nutritional approach which promotes satiety at lower caloric levels is hugely important. This is why whole food approaches are much more effective than eating processed foods and it may also be why low-carb (and even keto) diets are powerful for many.
If you’re trying to lose weight, I hope this helps you on your journey. If you’re trying to gain weight, the same lessons apply. Knowledge is power! As ever, if you have any questions or comments, please feel free to drop a comment below or email me at firstname.lastname@example.org