Insulin is responsible for pushing nutrients to their respective targets and if necessary creating fat. It also inhibits fat loss.

So given a negative caloric balance while trying to achieve fat loss with adequate protein fish oils and say 100 grams of carbs, how does the body bypass insulin’s effect on fat loss inhibition?

Does the insulin just deplete itself and fat oxidation resumes?

What about differences in GI and its effect on insulin.

Does equal carb intake of say white rice vs .brown rice elicit the same insulin response but at different portions varying over time?

If both white rice and brown rice are equal in calories and nutrients and let’s say right before bed I consume it, but still consuming at negative caloric balance. Would I still burn the fat equally to brown rice?

Possibly a bit lengthy but I would appreciate a response if you could. An insulin column would be fantastic on your new site and would eliminate a lot of repetitive questions! nice job on the site and thanks for all you do.

Answer: This is going to be a long answer because, frankly, there is a lot of misinformation and misunderstanding about insulin and, as usual, I got a lot to say.

This is because, in a lot of ways, insulin is a schizophrenic hormone.  Depending on what folks read (e.g. bodybuilding literature), they will be told that insulin is great, it’s the most anabolic hormone in the body, it’s key to getting big. And if you read other stuff (a lot of mainstream dieting literature),  you’ll hear that insulin is the devil, it makes you fat and ruins your health.  Who’s right?  Well, everybody…sort of.

As the question above states, it’s best to think of insulin as a generalized storage hormone rather than being good or bad; and what it does, as always, depends on the context.   I should mention that insulin not only affects peripheral tissues such as the liver, muscle and fat cell; it also has central effects in the brain.  I discuss this in Bodyweight Regulation: Part 1 and that series of articles.

When elevated (and I’d note here that while carbohydrate has the primary effect on raising insulin, protein also raises insulin; as well, the combination of protein and carbohydrate raises insulin more than either alone), insulin pushes nutrients into cells.  So insulin stimulates glycogen storage in the liver, it also enhances glycogen storage in skeletal muscle.  And while insulin isn’t that critically involved in protein synthesis per se, it does decrease protein breakdown; as discussed in The Protein Book, this is important for maximal increases in muscle mass.  So far so good.

But insulin also is involved in fat storage which is where it gets its ‘bad’ characterization.  Insulin activates an enzyme called lipoprotein lipase which is involved in breaking fatty acids off of chylomicrons for storage.  However, this isn’t the only important step in fat storage.

Contrary to popular belief (espoused by people still reading literature from the 1970′s), insulin is neither the only nor single most important hormone involved in fat storage.  Rather, a little compound called acylation stimulation protein (ASP) has been described as “the most potent stimulator of fat storage in the fat cell”.  And ASP levels can go up without an increase in insulin (although insulin plays a role).

As another effect of insulin on body-fat levels, and this is discussed in some detail in The Stubborn Fat Solution, insulin drastically inhibits lipolysis (fat mobilization) from fat cells. Even fasting insulin levels inhibit lipolysis by up to 50%, even small increases essentially turn off lipolysis completely.   Some could easily interpret this as meaning that ‘eating carbs stops fat loss’.  Or it might lead them to conclude that a carbohydrate based diet would make fat loss impossible.

Tangentially I’d note, and one weird little study supports this, that spiking insulin (and letting it crash back down) might be superior for fat loss than the standard strategy of trying to keep insulin low but stable all day long.  The reason is that even tiny amounts of insulin block lipolysis, if you keep insulin low but stable all day, you are effectively impairing lipolysis.  But the study in question showed that blood fatty acid levels came back up much faster when insulin was spiked (which crashed blood glucose back down, lowering insulin).  The drawback, mind you, is that rapidly falling blood glucose tends to make people hungry and calorie control would be nearly impossible with this strategy.  And, as you’ll see below, in a hypocaloric situation, it probably doesn’t matter a bit.

Anyhow, despite the sometimes seen mentality that you must ‘cut carbs to get lean’, four decades of practical experience (and endless clinical research) show that that is simply not the case: bodybuilders (well, some bodybuilders) have gotten plenty lean on carb-based diets (of course, others have failed miserably) so it’s obviously not as simple as many would make it.  That’s because whether a high-carb, moderate-carb, or low-carb diet is most appropriate for someone depends on the circumstances; a topic I discuss in Comparing the Diets.

Which brings me the long way around to the first question above.  What is happening in terms of fat loss on a diet that is hypocaloric (below maintenance levels, that is the person is burning more calories than they are consuming) but contains sufficient protein and essential fatty acids but with say 100 grams of carbohydrate? Don’t the carbs prevent fat loss by raising insulin?  What’s going on?

To understand what’s going on, I need to explain two terms which are the post-prandial and post-absorptive phases.

Post-prandial phase: this is just a technical term for ‘after you’ve eaten a meal’.  In this situation, nutrients are being absorbed and digested from the gut and released into the bloodstream, a whole host of hormones are being released (depending on the macronutrient content of the meal) and the body will generally be in an anabolic state (meaning that more nutrients are being stored than are being released from storage).

Post-absorptive stage: This is what happens between post-prandial phases.  Eventually what you’ve eaten has all been digested, absorbed and either burned for energy or stored in various tissues.  When this happens, hormone levels change an the body starts shifting to an overall catabolic state (I’m using this term generally here to indicate that the body is releasing more nutrients from storage than are being stored).

So throughout the day, the body is shifting between the post-prandial phase and the post-absorptive phase as you eat, that food gets digested and absorbed, and the body starts to draw on stored nutrients (hopefully stored fat in fat cells).

And when you lower caloric intake, over a 24 hour period, the body will end up spending relatively more time in the post-absorptive (remember: body burning stored nutrients) than post-prandial (remember: body storing ingested nutrients) phase.  This is simply a consequence of having less nutrients coming in relative to what’s being burned.

On a diet, meals are smaller (or activity is higher, or both) so any given meal will only maintain an anabolic state for so long (and that time period will be shorter than if the person were eating more) before the body shifts back to burning stored nutrients.  So even in the face of dietary carbohydrate intake, the body still will tap into stored fat; hence fat loss.

I’d note that theoretically this might mean that eating less frequently would improve fat loss, since the body would spend more time between meals in the post-absorptive stage.  Of course, this is probably offset by each meal being larger and therefore taking longer to digest and I tend to doubt it matters in the long-run.  Some interesting research into intermittent fasting suggests that there is more to it than that but that’s another topic for another day.

And this brings me to the second part of the above question, the glycemic index (GI) and insulin.  Which requires another long explanation.  The GI was developed back in the 80′s to help with diabetic meal planning.  Basically it involves feeding folks a fixed amount of a reference carbohydrate (studies have typically used either 50 grams or 100 grams of digestible carbs and while glucose was the original test food, they now use white bread) with blood glucose being measured over a several hour period.  The glucose response to the reference food is defined as having a GI of 100.

Then, whatever food was being tested (again either 50 or 100 grams of digestible carbs were given) and blood glucose was measured, researchers compared the blood glucose response of the test food to the reference food.  If the blood glucose response was say, 80% of the reference food, the test food was given a GI of 80.  If the blood glucose response was 120% of the test food, that’s a GI of 120.  You get the idea. And lower GI values basically meant that the test food was generating a smaller blood glucose response than the reference food.

GI is far from perfect, there is massive individual variability, many foods will show a different GI depending how you cook them  and, as soon as you start mixing foods or adding things like protein, fiber and fat, GI changes (almost always going down).  So GI in and of itself ends up not saying very much in the big scheme of things.    An additional confound is training.  As I discuss in the research review The Influence of the Subjects’ Training State on the Glycemic Index, people who are better aerobically trained show a lower GI response than those who are less well trained.

Now, it was always pretty much assumed that the GI was indicative of the insulin response and that lower GI foods caused a lower insulin response than higher GI foods; this is part of where dieters originally got fixated on the issue.  However, it looks like it’s not quite that simple.  While there was some brief interest in an Insulin Index (II) which measured the insulin response to foods in the same way GI does, research seems to have stopped as soon as it started.

As well as I discuss this in detail in the research review article Different Glycemic Indexes of Breakfast Cereals Are Not Due to Glucose Entry into Blood but to Glucose Removal by Tissue there is some evidence that low GI foods are low GI because they generate a fast initial insulin response.

That is, it’s important to realize that the blood glucose response of a food is determined by both its rate of digestion and entry into the bloodstream as well as the rate of glucose storage in tissues such as muscle.  And it looks like low GI foods are not necessarily digesting more slowly but that a fast initial insulin response is clearing more blood glucose.  To quote from the summary of that research article:

“Bran cereal has a low GI because a more rapid insulin-mediated increase in tissue glucose uptake attenuates the increase in blood glucose concentration, despite a similar rate of glucose entry into the blood.”

In this regards, I’d note that adding protein to carbs has been known to lower the GI for a couple of decades.  However, it’s also been established that adding protein to carbs increases the insulin response.   Which is consistent with the conclusions of the paper above, by increasing insulin, protein lowers blood glucose levels giving a lower effective GI.  Just not for the reason that most people think.  And I daresay that most of the ‘insulin is evil’ people are going to argue that eating more protein hurts fat loss, yes protein increases the insulin response to carbs.  While increasing the insulin response.  Go figure.

Which is a long way of saying that I don’t think the GI and insulin response matter much (although see my final comments below).  If there is much effect of GI on fat loss, it’s more likely to be mediated through food intake and fullness as lower GI foods generally make people feel fuller and often cause decreased food intake.  As I discuss in detail in Is a Calorie a Calorie, this is the typical confound in these types of studies: certain food types often make people spontaneously eat less, causing fat and weight loss and people confuse the food itself with the reduction in food intake that it causes.

It’s also worth noting that a 2006 review paper titled Glycaemix Index Effects on Fuel Partitioning in Humans examined this issue and concluded that:

“Apparently, the glycaemic index-induced serum insulin differences are not sufficient in magnitude and/or duration to modify fuel oxidation.”

Basically, at least outside of the absolute extremes (where it’s possible that some of this stuff might matter), it just doesn’t really seem to matter much outside of any influence on food intake (e.g. if a certain food keeps you fuller and you eat less, it’s good for fat loss; if it doesn’t, it’s not).  Basically:

The GI doesn’t truly indicate the insulin response in the first place, if it does it appears that low-GI foods may be generating a faster initial insulin response in the first place, and none of this seems to meaningfully impact on fuel utilization anyhow.  Certainly any tiny differences in GI between brown and white rice are going to be utterly irrelevant for 99% of cases.

Now, to wrap this up, I’d note that most studies done on this topic are drawing conclusions from average responses and emerging evidence suggests that it’s a bit more complicated than this.  In the article Insulin Sensitivity and Fat Loss, I detail some recent work suggesting that the insulin sensitivity of a given individual interacts with diet; the punchline of that article is that individuals who are insulin resistant (and/or show a pronounced early insulin response to food intake) seem to get superior results from a lower GI/lower-carbohydrate diet.  In contrast, individuals with high insulin sensitivity show superior results on a carb-based diet.  Which is something  I’ve observed for the last 15 years since writing my first book The Ketogenic Diet.

Ok, I know that was long but, as noted initially, there’s a lot of confusion over insulin and I have a lot to say on the topic.  Hopefully I answered your question.

On which note, if you’d like to submit a question for the Q&A, please email me at: questions@bodyrecomposition.com.  Due to the volume of responses, I can’t guarantee a personal response so please check the site to see if I’ve answered it.

Objective:To determine the glycemic index (GI) dependence on the training state of healthy adult males.Subjects and design:Young, adult males of normal body mass index and normal glucose tolerance were tested twice with a 50 g reference glucose solution and twice with a breakfast cereal containing 50 g of available carbohydrates in a randomized order. Ten subjects were sedentary (SE), 12 were moderately trained (MT) and 12 were endurance trained (ET). Blood glucose, insulin and glucagon were measured.Results:The GI differed significantly between SE and ET subjects (P=0.02, mean difference: 23 GI units, 95% CI=3-42 GI units). The GI of the MT subjects was intermediary, but did not differ significantly from the SE or ET subjects. The insulin index did not differ significantly between the groups (P=0.65).Conclusion:The GI of the commercially available breakfast cereal depended on the training state of the healthy males. The training state is the first reported factor influencing the GI that is subject specific rather than food specific.European Journal of Clinical Nutrition advance online publication, 12 July 2006

My comments: For readers who aren’t familiar with the concept, the glycemic index (GI) is a measure of how a given food affects blood glucose levels.  It was introduced over 25 years ago as a more accurate measure of foods (as opposed to earlier schemes that simply used simple versus complex carbs) for diabetics and has been researched extensively since that time.

To determine GI, first a standard food is given.  This used to be 50 or 100 grams of pure glucose but now 50 grams of white bread is used as the standard.  Blood glucose is tracked over time and the area under the curve for blood glucose is given a value of 100.  Other foods are then tested (again, 50 grams of digestible carbs are given) and the ratio of area under the curve to the test food gives the GI. So a food that has 75% of the area under the curve is given a GI of 75, a food which has 20% of the area under the curve is given a GI of 20.  A food that gave 120% of the area under the curve has a GI of 120; you get the idea.

While I’m at it, I want to mention a related/similar concept called the insulin index (II) which is functionally identical to the GI but looking at insulin response.  Although research into the II seems to have stopped almost immediately after it started, it will come up in this week’s review.

Over the years, an absolutely tremendous number of foods have been tested and GI lists can be found in a variety of places (one of THE most thorough sources for GI information is Rick Mendosa’s amazing GI site).  GI becomes interesting because the results aren’t always what you expect.

Some ‘simple’ carbs have a much lower GI than other ‘complex’ carbs. For example, sucrose (table sugar) has a medium high GI (about 70) while some GI measurements for potatos are much higher.  Full fat ice cream has a low GI (due to the fat content) and fructose (fruit sugar) has an extremely low GI (about 20) because of how it is metabolized by the liver.

However, the use of the GI is still debated in terms of its utility.  Factors such as how the food is prepared, the presence of other nutrients (fiber, fat, protein), and the effects of previous meals all impact on GI.  For non-diabetics, it’s especially questionable how relevant the GI actually is.  Bodybuilders and athletes often use GI as a proxy for insulin response under the assumption that low GI means low insulin which is good.

Up until this point, essentially all of the research into GI has suggested that the impact of a given food is independent of the individual, having only to do with the food itself.

However, this week’s research review brings that into question.  Following up on an earlier paper (by the same group) which found a difference in GI between sedentary and endurance trained individuals, it sought to measure the GI in three different groups of individuals: sedentary folks, moderately trained individuals (defined as aerobic exercise 2-3X/week on average) and trained (aerobic exercise 4X/week on average with some being competitive endurance athletes). Only males were studied so its unknown if these results apply to females.

A standard GI test was done with 50 grams of glucose and then a meal of (seriously) Kellog’s Special K with partial skim milk providing 50 grams of carbohydrate, 14 grams of protein, 4.6 grams of fat and 1.7 grams of fiber was given and blood glucose and insulin response was measured.  GI and II were calculated for all groups.

And, as with the previous study, the endurance trained group showed a significantly lower glycemic response for both the glucose and cereal meal as well as an average GI response that was 23 points (with a range of 3-42 units) less for the cereal meal (GI was about 80 for sedentary, 65 for moderately trained and 57 for endurance trained).  This was nearly identical to the previous study which found a 25 point difference.  The moderately trained group was square in the middle.

Although the absolute insulin response was different across groups (highest in sedentary, medium in moderately trained, lowest in endurance trained), the insulin index was not different between the groups.  Nor was the absolute insulin response different between the glucose and cereal trial, the same amount of insulin was released despite the difference in GI.

Unfortunately, the researchers couldn’t pin down a mechanism for this response although it’s clearly related somehow to training status, probably insulin sensitivity (how well skeletal muscle responds to the hormone insulin).  Given the differences in absolute insulin response (although the insulin index was no different), this would seem the logical conclusions.

So what does this paper tell us?  First and foremost, in endurance trained individuals, choosing foods based on glycemic index may be that much less relevant; regular endurance training will decrease the effective GI significantly.  Additionally, endurance training reduces the absolute amount of insulin released in response to a given carb load. Given the effect of regular training on insulin sensitivity, this makes a certain level of sense.

An unanswered question by this paper is whether regular resistance training will have the same effect.  While weight training can certainly improve insulin sensitivity, there is also evidence that extremely heavy lifting (which causes significant muscle damage) can negatively impact on insulin sensitivity. So the question is still unanswered.

Perhaps more interestingly is how little endurance training is needed to have at least some impact on both glycemic and insulin responses. Even two to three days/week had an impact although a greater frequency had a larger impact.  Strength/power athletes (including bodybuilders) might want to include a moderate amount of aerobic work (2-3X/week at low intensity) solely for that purpose, to improve glycemic and insulin responses to carb intake.

BACKGROUND: The glycemic index (GI) of a food is thought to directly reflect the rate of digestion and entry of glucose into the systemic circulation. The blood glucose concentration, however, represents a balance of both the entry and the removal of glucose into and from the blood, respectively. Such direct quantification of the postprandial glucose curve with respect to interpreting the GI is lacking in the literature. OBJECTIVE: We compared the plasma glucose kinetics of low- and high-GI breakfast cereals. DESIGN: On 2 occasions, plasma insulin concentrations and plasma glucose kinetics (by constant-rate infusion of [6,6-(2)H(2)]glucose) were measured in 6 healthy males for 180 min after they fasted overnight and then consumed an amount of corn flakes (CF) or bran cereal (BC) containing 50 g available carbohydrate. RESULTS: The GI of CF was more than twice that of BC (131.5 +/- 33.0 compared with 54.5 +/- 7.2; P < 0.05), despite no significant differences in the rate of appearance of glucose into the plasma during the 180-min period. Postprandial hyperinsulinemia occurred earlier with BC than with CF, resulting in a 76% higher plasma insulin concentration at 20 min (20.4 +/- 4.5 compared with 11.6 +/- 2.1 micro U/mL; P < 0.05). This was associated with a 31% higher rate of disappearance of glucose with BC than with CF during the 30-60-min period (28.7 +/- 3.1 compared with 21.9 +/- 3.1 micro mol. kg(-)(1). min(-)(1); P < 0.05). CONCLUSION: The lower GI of BC than of CF was not due to a lower rate of appearance of glucose but instead to an earlier postprandial hyperinsulinemia and an earlier increase in the rate of disappearance of glucose, which attenuated the increase in the plasma glucose concentration.

My comments: This is another older paper that I wanted to talk about since it ties in somewhat with the feedback on milk below.   In way of introduction, I should probably define glycemic index (GI) for readers who aren’t familiar with it.
The GI is used to rate carbohydrates by examining the blood glucose response to 50 grams of digestible carbohydrates.  After fasting, subjects are first given some reference food; this used to be glucose but researchers now use white bread.  The blood glucose response to white bread is defined as 100.  Then, the test food is given and the blood glucose response is measured and compared to that of the test food.  A food that shows 60% of the blood glucose response to white bread is given a GI of 60.

It has commonly been assumed that GI and insulin response are related and bodybuilders and athletes commonly use GI to determine which foods are or are not acceptable to eat (especially on a fat loss diet).  Low GI foods are usually assumed to digest slowly and it is the slow rate of glucose into the bloodstream which causes the low GI.

A massive number of foods have been tested for GI although there is still much debate as to the validity of GI in meal planning.  GI can vary significantly by food and how it is prepared, as well as between individuals.  Also, GI is measured for 50 gram quantities of foods (that’s 50 grams of digestible carbohydrates).  But this can be misleading; for example, carrots are very high on the GI scale but few people would eat 50 grams of digestible carbohydrates worth of carrots in a sitting.  To counter this, some researchers have proposed a measure called the glycemic load (GL) which is the total amount of digestible carbohydrate multiplied by the GI.  This at least recognizes that, in the real world, carbohydrate intake varies.  GL can be lowered by either picking lower GI foods or by eating less total carbohydrate, or some combination of the two.

Additionally, GI tends to be affected by other nutrients (protein, fat and fiber) although not always in the way you’d think (and not all research finds a significant impact of protein and fat).  For example, protein tends to lower the GI of carbohydrates but insulin levels often increase when you add protein to carbs.

For reference, the most complete site on the web for information about GI is Rick Mendosa’s Site.

Which brings us to the above study.  As mentioned above, bodybuilders and athletes usually assume that a low GI means a low insulin response but the study above draws that conclusion into question.  Rather it found that the low GI food showed a lower blood glucose response because it generated a higher early insulin response (clearing blood glucose out of the bloodstream) at the 30 minute mark (by 60 minutes, both foods showed similar insulin levels).  Quoting directly from the paper "Bran cereal has a low GI because a more rapid insulin-mediated increase in tissue glucose uptake attenuates the increase in blood glucose concentration, despite a similar rate of glucose entry into the blood."

That is to say, both foods released glucose into the bloodstream at similar rates, but the bran cereal showed faster uptake due to a higher initial insulin spike, which lowered the overall GI response.

The researchers also noted that the bran cereal contained more protein than the corn flakes and this is probably what caused the higher insulin response (and lower blood glucose) which ties into my comments above.

Somehow, I don’t think bodybuilders would argue that combining low GI carbs with protein is bad for fat loss, yet here we have an (as of yet unreplicated paper) showing that the initial insulin response is higher; essentially, the higher initial insulin response caused the lower GI in this case.  Yet most bodybuilders also believe that high insulin is detrimental to fat loss.  Here we have a study that I think questions that idea. At the very least, the small initial insulin spike certainly wouldn’t appear to be hurting things, it’s likely that sustained insulin levels would be more problematic by limiting the ability to mobilize fat for fuel.

Then again, at least one study found that spiking insulin (high GI condition) resulted in a larger rebound in blood fatty acid levels (after blood glucose crashed) compared to keeping insulin low but stable (low GI condition) so maybe there’s more to this picture than we yet realize.