When Dietary Fat Is Fattening and When It Is Not: The Fat Thermostat

Recently, I wrote an article in which I discussed how inflammation of the brain’s body-fat regulator – or ‘fat thermostat’ – is of critical importance for weight management. In this article, I discuss how certain dietary patterns affect inflammation in the fat thermostat, as well as short- and long-term outcomes of different diet types. At the end of the article, I consider how to reflect this study against broad dietary choices we make, and I also link to a recent radio show where I answer audience questions about health and weight control with Robb Wolf and Stephan Guyenet, PhD.

Remember from the first article that if the fat thermostat becomes insensitive to signals it relies on to keep the quantity of body fat stable, then the result is that fat mass gets ratcheted up to a new, higher stable point. So, what diet properties affect the sensitivity of the fat thermostat?

A study by David McNay and John Speakman explored this. In the study, mice were fed a fattening diet until they became obese. At that point, the obese mice were divided into one of ten groups – five dietary styles, and two food-access conditions:

Quick primer on dietary style number 5: a ketogenic diet is a very low carb diet that stimulates the liver to produce ketones, which are short chain fats the brain and body can use as fuel. Ketones may have unique effects on the fat thermostat, which I will discuss later.

The diet intervention was maintained for four weeks, then all the animals were switched back to a normal diet and monitored for six weeks. The researchers were interested in:

What happened when the obese mice were placed on the different diets and food access conditions?

Calorie-restriction condition

As expected, all groups of obese mice that were calorie restricted, regardless of diet type, lost a significant amount of weight. No surprise there. Calories matter. Also, there was not a significant difference in the amount of weight lost between diet groups in this condition. So, according to this study, when restricting calories, it doesn’t matter what a diet looks like for weight loss. This is consistent with findings in humans.

Free-access condition

Interestingly, only some of the dietary styles – high protein, high carb, and the ketogenic diet – caused the obese mice to lose weight when given free access to food. On the other hand, when obese mice were put on a high fat diet, they did not lose weight.

Taken together, the treatment of obese mice caused significant body weight loss in all groups except mice given free access to a diet relatively high in fat.

What happened when the mice went off the diets?

Even though obese mice lost weight when calorie restricted, not all groups fared similarly when put back on their normal diet. Additionally, neither final body weight or adiposity at the end of calorie restriction predicted long-term body weight. That’s important because it suggests that the ability to sustain a lower body fat level may be independent of the amount of weight lost, which is good news! Importantly, the group exposed to the high fat diet experienced the most rebound weight gain. This suggests that the high-fat diet may have caused an increase in the set point of the fat thermostat.

The ketogenic diet groups, however, showed final body weights similar to the high protein and high carb groups – groups that maintained some of their weight-loss results!

What effect did the diets have on the fat thermostat?

The neurons in the fat thermostat undergo constant regeneration, yet diet-induced obesity can halt this. This reduction in neuron regeneration changes the structure of the fat thermostat and leads to an impaired ability to regulate body fat stores normally. Insensitivity to leptin in the remaining cells, as well as these structural changes could shift the relative balance of power between the fat-gain and fat-loss programs, which could explain what happens when:

The good news is that weight loss itself seems to reinstate at least some new neuron formation. But this is one of the reasons why this study was so interesting. The high fat diet did not reinstate remodeling even with weight loss, which is exactly what you’d expect by the observation of rapid weight regain once the mice went back on a normal diet. On the other hand, all the other diets did partially reinstate neuronal remodelling under both calorie-restriction and free-access conditions. Although, the calorie-restriction condition was less beneficial at this. The fact that these beneficial diets only partially reinstated remodeling may explain why many people who have gained weight can lose some but not all of it, or after weight loss, why some people rebound up to a new stabilization point somewhere between their high point and low point. Also, an important point here is that deliberate calorie restriction is not the same thing as spontaneous calorie reduction. In the later scenario, a person feels like consuming fewer calories for one of a multitude of possible reasons, but ultimately, it is that type of calorie control that may indicate a much more sustainable approach to weight control; a concept we aim to leverage in the Ideal Weight Program.

Dietary fat – unpacking a complex problem

Previous work by these authors showed similar undesirable effects from a diet that was 58% fat (the high fat diet in this current study was 45% fat). Keep in mind, dietary fat is not a singular thing – it’s chemical category containing many different molecules, each having potentially different biological effects. A lot of confusion results from this. For example, in rodent models, palmitic acid – a saturated fat found in high proportions in animal fats – causes cellular lipid accumulation, inflammation, and insensitivity in the fat thermostat, while oleic acid – a monounsatured fat found in olive oil and avocados – doesn’t (Benoit et al., 2009).

Having noted that caveat for dietary fat subtypes, this study did show multiple negative effects from a diet of 45% fat. But, it also showed that a diet of almost 95% fat – the diet that produced ketones – had completely different effects. In fact, the trends of the ketogenic diet suggest that it might be the most effective for weight loss when given free access to food, amongst the best diets for weight-loss maintenance, and also one of the best for stimulating new neurons in the fat thermostat. All of which are very positive signs for someone who wants to get and keep fat loss results. These findings are also in line with previous research in humans showing that a ketogenic diet suppresses typical weight-loss induced increases in the hunger hormone ghrelin, as well as in subjective appetite, which are both signs that ketones could have a favorable effect on the fat thermostat. Other research has shown that certain ketone bodies (beta hydroxybutyrate) can reduce inflammation, which may offer a possible mechanistic link for the observations. Whether ketones cause or simply correlate with these favorable physiological effects, a ketogenic diet may be among the best diets we have for sustainable weight loss. I look forward to more long-terms studies evaluating this in humans.

In the McNay and Speakman study, the high carb diet fared well for weight loss, and fared moderately well for weight maintenance and neuron remodeling. However, many people reduce carbs because they believe they are fattening. When you reduce carbs, you make up for some of the deficit by eating more protein and fat. And simply restricting access to any macronutrient class will cause some spontaneous reduction in calorie intake. Restricting carbs, more than the other macronutrients, will produce the most noticeable short-term weight loss (first week effect), which reinforces the idea that carbs are indeed fattening. However, most of this initial weight loss is water weight. This is because carbs are stored in the muscle and liver as glycogen, which clings to water (hydrophilic). When you take in less carb, you will lose a few pounds of water within the first few days as your glycogen stores are depleted and you excrete water. When you add carbs back, you rehydrate and gain a few pounds. These water dynamics have a profound impact on the perception of weight loss efficacy. And this perception, in turn, can have a profound impact on compliance to the diet.

For some people, however, the reduction in carbs also causes their calorie intake to decline, causing slower but more meaningful changes in fat and weight. It is possible that this effect has do with what else you’re eating, instead of what you’re not eating, or both. For example, replacing refined flour-based products with vegetables and protein will have a positive impact on energy balance. Having said that, this positive effect can also be achieved by choosing different carb sources – think fresh whole starchy tubers and fruits over cookies and soda.

In today’s world, the reality is that there are so many tempting-but-poor carbohydrate choices. Approaches that limit all carbs, even forsaking good options (due to the behavioral empowerment of simplicity – ‘carbs are bad’), may be one of the more practical options we have for many people. Consider this: according to a recent panel of 13 nutrition experts charged with helping develop federal nutrition standards – the Dietary Guidelines Advisory Committee – the number one source of calories in the United States is “Grain-based desserts (cakes, cookies, donuts, pies, crisps, cobblers, and granola bars)”. Additionally, each of the remaining carb-based calorie sources have the qualities (e.g., highly processed, highly palatable) of fattening foods. Yikes!

Here are the top 10 sources of calories in the US diet:

Only 4 of the top 10 sources of calories could remain on the list if you eliminate the carbs:

If you replace these carb sources with good carb sources, I guarantee we would not have an obesity problem in this country. Picture this:

Restricting 60% of the top calorie categories can and does have an impact on overall calorie intake. Personally, I do restrict carbs, but it’s type not quantity that I restrict. After that, I let my palate guide the way and I choose not to over-complicate the matter.

When switching to a higher fat diet, it’s possible that a person could maintain a lower calorie intake and attain fat loss, but that person might be putting themselves in a position for accelerated fat gain later (extrapolating this deduction from the findings of this study). That is a scary proposition. So many people are switching to a higher fat diet, eating a lot of problematic dietary fats along the way, and this situation could promote early success and downstream problems when dietary practices become more lax.

However, if this higher fat diet is one that you can sustain – as some people do comfortably when they maintain a non junk-food version of a ‘Paleo’-style diet, for example – and if there isn’t too high of a fraction of problematic dietary fats, then it’s entirely possible that the higher fat diet won’t cause downstream weight issues. In fact, if the dietary pattern naturally leads to better food choices and improved energy balance, then the opposite may be true.

Also, because some of the negative effects of dietary fat are due to the promotion of inflammatory signaling in energy balance circuits, then we must also consider how other factors that affect inflammatory signaling influence the equation. Exercise is one of these factors. Exercise indirectly produces an anti-inflammatory effect that may mitigate dietary fat-based inflammation. Therefore, if you’re an avid crossfitter on a high fat Paleo diet, you may have a very different response to someone on the exact same diet who doesn’t get sufficient physical activity, even when controlling for calorie expenditure. This idea is emphasizing the importance of the net-inflammatory response in the fat thermostat.

On that note, the highest fat diet was one of the most effective in this study. In the real world, someone could interpret this phenomenon to mean that carbs are so bad that one must eradicate all of them – even limiting protein intake since protein can convert to carbs in the body – since they are so fattening. Not getting results on your low carb diet? It’s because you haven’t reduced carbs enough!

But, for so many reasons, that logic falls flat. For example, many cultures around the globe maintain health and leanness on very high carbohydrate diets. Could it be the production of ketones, not the restriction of carbs, that explains the efficacy? Why is this detail crucially important? Because while it’s true that you need to severely reduce carbs in order to produce ketones – and that the production of ketones might be beneficial for weight loss – it could also be true that a high carb diet might be the best diet to adopt post weight loss. On the other hand, if you think carbs are exceptionally fattening, continuing with a high fat diet post weight loss (given all the caveats noted above), could be the worst diet to use if you want to maintain your results.

With dietary fat, it’s possible that it’s darkest before the dawn. In other words, high dietary fat is increasingly fat promoting until you get to the point where you are producing ketones. At that point, the diet may change from fat promoting to fat reducing.

Deliberate calorie restriction is effective for weight loss, but spontaneous caloric reduction via a variety of dietary styles may produce better long term results (beyond the timeframe measured in this study). A diet that produces ketones seems effective for weight control, but a high protein and a high carb diet are effective too. Dietary fat is a broad topic, and different types can produce different results, but quantity alone may also matter. Also dietary fat may demonstrate a u-shaped response curve, where weight control is worsened with higher levels of intake, unless factors are present – like exercise and ketones – to mitigate, eliminate, or even reverse its negative effects.

Thinking carbs are fattening and dietary fat is not is a simple heuristic that has helped some people lose weight, but it is not an accurate reflection of reality. This is one of the reasons I suggest you don’t get your health information from extremist-style internet health personalities. By the way, as a behaviorist, I don’t take issue with the idea of using simple messaging (e.g., carbs are bad) to attempt to induce a favorable behavior. Some people, independent of intelligence level, will simply not be able to act on information more sophisticated than ‘carbs are bad.’ However, it turns into a problem when these oversimplified messages become zealotry-based dogmatism, and good solutions, rational thinking, and reasonable analysis are rejected wholesale by influential health personalities. If you’re a health authority promoting a low carb diet, don’t slam other good options because it doesn’t fit into your facile efficacy framework.

At the end of the day, this study reinforces the idea that there may be more than one good way to lose weight, but not all diets are the same to help you spontaneously reduce your calorie intake, and are also not the same for long term efficacy even when controlling calories. If you’re eating a higher fat diet, aim to have much of it come from monounsaturated sources, and be extra vigilant to attain sufficient levels of physical activity (in other words, always maintain an Exercise Score of 100% or greater on the Dan’s Plan Activity Tracker). Plain and simple, a ketogenic diet might be one of the most effective diets to lose weight. I look forward to more long-term clinical trials in humans to evaluate this. But, regardless of the diet you choose, good sources of carbs remain one of the most effective for weight control and healthy food sources one can consume, and even if you use a higher fat diet to lose weight, you shouldn’t overlook a diet high in excellent carbs sources for long-term weight maintenance.

Lastly, Robb Wolf, Stephan Guyenet, PhD, and I recently answered audience questions about our Ideal Weight Program and other health topics, including Bulletproof coffee, ketones, and weight tracking. You can listen here: