The truth about fructose and metabolic syndrome: Part 3 – Fructose isotope tracer studies – overview

By Dagmar
In Discussion
Sep 6th, 2014

This is a third chapter of the series about the truth of the fructose metabolism in humans. The previous (Part 2) has discussed the trends of the consumption of macronutrients (and energy within) in the U.S. for the past few decades and the available data suggest that the consumption of sugar was decreasing whereas the intake of added fats has increased since 2000. This Part 3 is little more difficult for an average reader. For those who find it challenging, there is a conclusion and Summary at the end.

I have based this review on a single article published in the Nutrition and Metabolism Journal in 2012. I have found this article quite comprehensive and well balanced, despite its authors have admitted a link with the food industry. Nonetheless, the materials that Sun and Empie (the authors) have included in their review are legitimate scientific findings and these should not be ignored. It is possible that a selection bias had occurred here, too, since the authors had certain criteria for inclusion of the studies and also since 2012 some new research might have been published about the isotopic tracer studies of fructose metabolism in humans. In such case I encourage anybody who has the data and findings that were not included here (or published in the original article of Sun and Empie) to come out and present it.

Below I am going to summarize the review of the authors and cite their work when I feel that they said it better than I would. The authors reviewed the studies examining the fructose oxidation, fructose conversion into glucose, glycogen, lactate and lipids in humans. Although the studies examining these outcomes (and other metabolic processes in relation to fructose) are not too abundant, the authors have found 34 studies to look at the outcomes named. The isotopically labelled fructose metabolism was examined in adult participants (exercising and non-exercising) and the fructose was administered orally or perorally, as a single sugar or together with glucose/maltodextrin.

The metabolic pathways of glucose and fructose are interconnected

What we know about fructose metabolism? Countless studies were performed on animals, mainly rodents, but the data obtained from humans studies is less abundant. The metabolic pathways shared by fructose and glucose are numerous and their carbons can be found in the processes such as glycolysis, TCA cycle, Cori cycle, pentophosphate shunt and lipid synthesis. However, fructose carbons can also be found in the process of gluconeogenesis, glycogenolysis, suggesting that there happens the conversion of fructose to glucose in the liver. What is important to understand, is, that fructose and glucose have common end-point of disposal. Therefore, the popular conclusion that fructose is the sweet fat whereas glucose is a primary fuel for body cells and therefore a good one while fructose is an evil (as Dr Lustig and others kept saying), has to be presented with caution. The fact is that as fructose can be converted to glucose, glucose can also be converted to fat, depending on the individual circumstances.

Below is the diagram of the mutually interlinked metabolism of glucose and fructose for those who have some strange like for the metabolic biochemistry. I have copied it from the original article of Sun and Empie (paper). Follow the arrows to see how complex the metabolism is; yet still, it is quite a simplistic diagram after all because the reality is much more complicated. When looking closely at the diagram, you can notice how these pathways are interconnected and that fructose carbons can be mixed with those of glucose and end up virtually in the same outcome molecules. What is more, fructose carbons can enter the newly formed glucose molecule. The intermediates printed in bold are the “major metabolic intermediates or end products of glucose and fructose metabolism” (Sun and Empie, 2012)

glucose and fructose metabolism

Further the paper stated that “the dashed line and arrow represent minor pathways or will not occur under a healthy condition or ordinary sugar consumption“. The amounts considered as ordinary sugar consumption have not been stated, however. Nonetheless, this diagram illustrates the summary of the overall findings about these two sugars metabolism in humans, suggesting that VLDL lipoprotein is a minor outcome of fructose metabolism in healthy humans consuming ordinary (average) sugar amounts. Moreover, the fructose forms roughly 40% of all caloric sweeteners on the U.S. market. Several authors have tried to estimate the average content of fructose in the sweeteners supply and their figures fluctuated around this number.

Limitations of the classic studies of fructose metabolism

The reason why I have decided to share the content of Sun and Empie review (2012) was the statement:

Conventional clinical trials and ecological studies have been conducted to assess the hypotheses, but not all results are found to be supportive.

This means that the role of fructose in the development of obesity or other metabolic disruptions was not always proven and there were contradicting results among the studies.  Furthermore,

Conventional studies often cannot reveal details of interconnected metabolic pathways, when testing fructose or fructose containing sugars, but they also cannot clearly distinguish a mechanistic cause associated with an observed physiological consequence linked to the sugar consumed. This is because the ordinary diet contains multiple forms of saccharides which are inter-convertible in the body and share many steps of the carbohydrate metabolism pathways.”

In other words, the team applied a principle of benefit of doubt. I am OK with that. Every scientist should have a healthy level of doubt about things presented to them, unless there is stone solid evidence that THIS is causing THAT, not accepting that THAT happens in the presence of THIS. This is a fallacy, not a science. What is more, something observed on animals (mostly rats) cannot be directly interpreted as typical for humans. More experimental trials were performed on rats and other animals, than on humans, but most of the review papers discussing the effect of fructose on human metabolic health used both: human and animal studies outcomes, mixing and messing about with the assumptions and theories, not the actual evidence. And that was the reason why I left the animal studies behind and keep focusing on humans only.

The authors further highlighted:

In many of the intervention studies … very high doses of sugars over short term were often applied, the study designs were more similar to toxicological studies and the studies were only able to draw associative conclusions between applied dose and observed health-related outcomes… The observed biological changes, although statistically significant by a P-value ruling, were often only fluctuations within normal ranges. These studies rarely measured actual development of disease or the intermediate metabolites characterizing mechanism-based reactions.

And I am adding: even when the metabolites characterizing mechanism-based reactions were studied, the studies provided equivocal results. Moreover, just because some metabolite was affected and this was statistically significant, it does not mean that the disease would develop, since the statistical significance is not equal to the clinical importance (read more). The body has an enormous capacity for adaptation and can keep itself healthy if it is given a chance by other dietary and lifestyle modifications.

The advantage of isotope tracer studies

The isotope tracer studies enable us to understand what happens with what substrate in the body by tracing their fate once consumed and then found in body fluids or solids, or even gas (breath) or their occurrence in the blood or other examined tissues.

To begin to prove true effect of a diet component it is useful to study the component disposal through the common central pathways at the molecular level.”

In other words, these isotopic tracer studies aim to follow the particular nutrient through the metabolic pathways which it shares with other nutrients and where these nutrients can be converted into each other, which makes studying the effect of one of them on the general metabolic outcomes so difficult. There are still some limitations in the isotope tracer study methods, but there is hardly anything better we have at the moment. And this is the reason why I have decided to dedicate one whole chapter to this method and the findings, however limited they are.

The method of Nuclear Magnetic Resonance (NMR) allows to specifically examine the isotopes of various molecules, including sugars. The importance of this technique lies in that its ability to trace the fate of the carbon isotope at specific position within the molecule. As we metabolize more complex molecules, these are broken down and different parts of these molecules enter different metabolic pathways. Knowing where this isotope was allocated in the molecule enables the scientists to trace the metabolic fate of the original molecule and to quantify to what extend this molecule contributes to particular metabolic pathway and therefore the metabolic outcome. This is a more sophisticated method than just feeding  sugar to the participants and observing, what happens. Since the metabolism of the two most common simple sugars (glucose and fructose) in our diet is so inter-linked, it is only a great advantage knowing to what extent which sugar contributes to which metabolic outcome.  Even more, because we normally consume glucose and fructose in sugar together, studying the metabolism of fructose by feeding only fructose to the participants is not sufficient, exactly because of their interlinked metabolism and also because feeding participants a single sugar in comparison to feeding them with a mixed sugar has repeatedly produced different outcomes. In other words: fructose consumed with glucose behaves differently than when fructose is consumed alone.

It is important to understand that feeding participants with a single sugar and observing what happens is great for studying the individual metabolic pathways. However, when examining the metabolic effect on the whole body, administering a single sugar is not sufficient, because this does not reflect the usual living conditions when the combined sugar is consumed. Once we have established the pathways we needed to know, we should rather focus on studying the metabolic effect of fructose in combination with glucose for obtaining a true picture of the role of fructose in metabolic disturbances among the population. In this aspect the toxic doses of isolated fructose in the studies (often as much as 30% of total energy intake and often on top of the balanced energy state)  is not sufficiently relevant to the dietary patterns of the free-living population. Not all people overeat and a diet-related metabolic syndrome can develop in relatively slim individuals, too.

We now know that fructose and glucose share common pathways once they were converted to either a di-hydroxyacetone-phosphate (DHAP) or glyceraldehyde in case of fructose, or glyceraldehyde-triphosphate (GAP) in case of glucose, into which the glyceraldehyde can also be converted. The metabolism of these two sugars and their conversion to the metabolic intermediates vary among different animal species, also in humans. This practically means that both glucose and fructose can be converted to fat if consumed in excess. However, humans are less capable of converting carbohydrates to fat, in contrast to some animal species, which hibernate over the winter and therefore they rely on efficient de-novo-lipogenesis (DNL). What is more, the authors reported that:

“… blood glycerol concentration increased after fructose ingestion in exercising subjects…” and “….the noted glycerol increases after fructose ingestion are either greater or similar compared with the values after glucose ingestion and the produced glycerol can be oxidized for energy.”

However, my aim is not to focus on the sports nutrition but on a general population and what the consumption of sugar does to them. Nonetheless, more studies were performed on exercising subjects than on non-exercising and I will include the results from both for a more complex picture.

The limitations of isotope tracer studies

Although this isotope labelling approach has been pronounced as superior to other methods studying metabolic effect of various nutrients, there are limitations in these studies. The main one is the short period of time over which the metabolic fate of fructose was followed – up to 8 hours. The partial labelling and the position of the isotope in the molecule can influence the rate of appearance of the isotope in the exhaled CO2 and other metabolic outcomes. The authors also outlined that the short time frame of examination of the fructose oxidation may lead to incomplete picture because some portion of the fructose has been diverted to the non-oxidative pathways. Hence, as always, more research is needed to map the fructose metabolic fate in more detail and for longer.

In the following section I will look at different metabolic fates of fructose metabolism and their extent. In most of the cases I relied on the outcomes of the authors, but I have looked at some of the reviewed studies myself so my article adds extra information to the original paper.

1. Fructose oxidation 

This term means how much of the ingested fructose was metabolized in the body to provide direct energy, i.e. not stored as fat or converted to other metabolites, which were not yet oxidized in the measured time period. If some portion of this fructose was converted to fat, that fat has been oxidized for energy and not stored.  The authors reviewed 19 relevant studies that have met their inclusion criteria for this outcome. Four studies were performed on resting subjects, consuming 0.5 – 1 gram of fructose per kg of body weight. To give you a better picture, for a 70kg individual, the range of fructose was between 35 to 70g during the testing period. These volumes were on top of the usual diet of the participants although in some studies the participants may have been fasted for hours before the test. The amounts of oxidized fructose ranged from 30.5% to 60%, the latter was measured by Lecoultre et al. (2010) when examining exercising male subjects.

In the study of Chong et al. (2007) the fructose was oxidized faster when compared to glucose within the monitored period (6 hours), i.e. 30.5% vs. 24.5%. The two possible reasons discussed were those that fructose oxidation is less regulated than oxidation of glucose and that glucose has a wider distribution of metabolism within the body and therefore when they both hit the liver, fructose is primarily metabolized right in the liver, but glucose is directed to the rest of the body. This takes time and the studies reviewed here were performed only within 2 – 8 hours. What happened after 8 hours these isotopic studies did not examine, usually for practical reasons, but some also reported reaching a plateau in the measured outcomes. With the increased dose of ingested fructose, its oxidation also increased – but this was limited by the absorption. Fructose is less effectively absorbed than glucose and above certain absorption rate, the increased ingestion does not lead to increased oxidation. Instead, the individual can experience gastro-intestinal discomfort and diarrhoea. Moreover, the diabetic patients have reduced capacity for fructose oxidation in comparison to healthy subjects (31.3% vs. 38.5%) according to Delarue et al. (1993).

Tran et al. (2010) examined the sex differences in various metabolic outcomes of isotopically traced fructose. They examined non-exercising subjects, healthy men and women of a healthy BMI and not being athletes. Within the 6 hours, the rate of fructose oxidation was almost identical among the subjects of both sexes – accounting to 42.9 and 43%.

The remaining 15 studies were performed on exercising subjects with the 50-75% of the VO2Max, i.e. within the aerobic capacity. The oxidized amounts of fructose were slightly higher, within a range 37.5% – 62 %. Except of one, all of these studies have found a higher glucose oxidation rate than of fructose, which was the opposite to non-exercising subject.

However, a very interesting thing was observed when comparing the rate of oxidation of glucose or fructose only with both of these sugars supplied together. According to Adopo et al (1994), within two hours of exercise, the amount of oxidized glucose increased from 37.8 % to 58.3% when ingested in 50g and 100g, respectively, whereas the amount of fructose increased from 32.2% to 45.8%, when administered in the same amounts. However, the 50g of both sugars were given together, whereas the 100g were given each sugar separately. So in all three cases the participants received a total 100g of various sugars each time. It is apparent that the consumption of both sugars together enhanced their absorption and therefore the metabolism quite significantly: adding up the 37.8% and 32.2% gives total 70% of oxidation from 100g of combined sugars (compare to the 58.3% and 45.8% for 100g of individual glucose and fructose, respectively). This has a practical impact on sporting people, especially in endurance activities, where there is needed a fast supply of carbohydrates for the activity to continue.  A similar comparison of oxidation in non-exercising subject is not know to me, but it is well established that the combined sugars ingestion have a higher absorption rate than each sugar alone in non-exercising people, too (Riby et al., 1993), so a similar effect is likely to be observed, but perhaps in different outcomes for individual sugars.

2. Fructose conversion to glucose

Based on several studies on non-exercising subjects, the authors have summarized that about 41% (ranging from 28.9 to 54%) of ingested fructose was found to be converted into blood glucose within 2-6 hours from the ingestion. Women, obese and diabetic subjects again showed a significantly reduced conversion rate of fructose to glucose than men, non-obese and healthy subjects, respectively. Under exercise condition, only 29% of ingested fructose was converted to glucose, but this was only within 2 hours. Jandrain et al (1993) examined exercising subjects for up to 3 hours and found that that at a later stages of exercise the conversion of fructose to glucose increased from 29% to up to 60%. Similarly, in non-exercising subjects when examining two different doses of fructose (0.5-1 g/kg body weight) for 6 hours, over 50% of fructose was converted to glucose (Delarue et al., 1993).

Overall, both studied fates of fructose: oxidation and conversion to glucose showed similar scenarios in terms of metabolic health, gender and physical activity. The team concluded, that:

“Fructose is converted to glucose in variable extents, depending on exercise condition, gender and health status”.

3. Fructose conversion to glycogen

As the authors noted, this area is less explored. However, from the small amount of data there is, it appears that infused fructose translated into 3.6 times higher formation of liver glycogen than infused glucose of the same concentration whereas the difference in muscle glycogen did not reach statistical significance (Nilsson and Hultman, 1974). It is important to bear in mind that this fructose was not consumed by mouth and therefore carried by portal vein directly to the liver, but it was injected directly into the systemic bloodstream, which is where the nutrients normally occur after being processed by the liver (except of fats). This form of substrate administration therefore might not be too representative of the usual dietary sugar intake. Moreover, this study, based on the abstract freely available, does not mention the isotopic tracing of the substrates as did not the study of Blom et al. (1987). The team studied the rate of muscle glycogen synthesis after consumed fructose in post-exercise subjects. It was found that fructose is only half efficient than glucose in replenishment of muscle glycogen in the post-exercise subjects (0.32 and 0.58 mmol/kg per hour). When compared to sucrose ingestion in the same amounts as both individual sugars alone, the resultant glycogen synthesis of 0.62 mmol/kg suggest that there was again the synergistic effect of these two sugars, which has been likely affected by the enhanced absorption of both sugars consumed together. Nonetheless, a study of  Coss-Bu et al. (2009) examined the surge of isotopically labelled glucose release after previously ingested isotopically labelled fructose. This released glucose was the converted fructose and was released from the glycogen stores after administration of glucagon, a hormone breaking down glycogen when blood glucose levels drop.  This was the proof that fructose does contribute to the glycogen stores. Overall, regarding this paragraph, Sun and Empie concluded that

“a portion of fructose is incorporated into glycogen after conversion to glucose but the extent is not known”.

4. Fructose conversion to lactate

The authors announced that

“ Earlier tracer studies observed that blood lactate concentration was increased after fructose or fructose and glucose ingestion when compared to glucose ingestion alone.”

The consumption  of sucrose produced similar result than of glucose alone. Lecoultre et al. (2010) examined the conversion of fructose to lactate in exercising men, giving them a mixed amount of fructose and glucose in ratio 1:1.5. The team found that 28% of ingested fructose was converted to lactate. Of this lactate, almost 90% was oxidized by the exercising muscles. Further 29% of fructose was converted to glucose. It was also stated, that fructose was oxidized faster than glucose and this contributed to the increased overall oxidation of the mixed sugar than of the glucose alone. A similar study of Rowlands et al. (2008) examined different amounts of fructose added to maltodextrin in subjects exercising for two hours. At medium fructose supply (0.5g/min) the increase of blood lactate concentration was 31% in comparison to maltodextrin only, but at a higher supply (0.7g/min) of fructose, while the maltodextrin content remained constant in all cases, the lactate concentration was only 24%, . The non-oxidative metabolism of higher amounts of fructose was again suggested in the latter case.  However, the study did not quantify how much the labelled fructose contributed to the increased levels of blood lactate in contrast to the labelled glucose/maltodextrin.  In this particular paper it was stated: “From these studies, it appears that ingested fructose is less immediately available to the skeletal muscle than glucose, but the liver acts as a reservoir for releasing fructose-derived metabolites that are utilized by the muscle later as a carbon source for oxidation.” In addition, the increased carbohydrate oxidation in higher fructose doses led to a decreased oxidation of endogenous fat. This is not surprising since the carbohydrate is preferentially metabolized when available while the body tends to spare the fat, which is metabolically more ‘expensive’ to get energy from. Nonetheless, the higher fructose ingestion was characteristic with reduced fatigue and enhanced performance than when no or less fructose was consumed, indicating a buffering effect of additional fructose-originating substrates from the liver continuously provided for the exercising muscles.

No study on non-exercising subjects was reviewed in terms of evaluating the contribution of isotopically labelled fructose to the lactate concentrations.  Sun and Empie concluded that

“the effects of fructose dose, administration method, physical activity and subject characteristics on fructose-lactate metabolism remain to be further studied”.

Also, the

“labelled patterns of isotope tracer in fructose will have an influence on the measured isotope appearance in lactate, if the studied sugar is not uniformly labelled”.

This is what I have mentioned earlier: that different parts of the sugar molecule can end up in different metabolic pathways, which makes studying the fate of these carbohydrates so complicated.

 5. Fructose conversion to lipids and its influence on lipoproteins 

The authors have started this section with highlighting the challenges in studying the conversion of sugars into lipids, which can appear in blood as VLDL triglycerides, or remain in the liver cells.

“There are currently no convenient methods to quantify overall DNL and intrahepatic lipid deposition.”

They further state:

“DNL can also occur in adipose tissue or muscles, but there are no methods to quantify it.”


“The time periods of liver DNL from sugars and the factors influencing it are not completely understood.”

Overall, Sun and Empie identified only two studies investigating the labelled dietary fructose conversion into plasma lipids. One of them was Chong et al. (2007) studying labelled fructose and glucose along a labelled oil mix for the duration of 6 hours. Fructose significantly increased the formation of plasma triglycerides in comparison to glucose when reaching a plateau, starting from the same baseline level. However, within the monitoring period, the origin of these fats was predominantly not from fructose – suggesting that fructose acts rather as a factor for the assembly of available fatty acids into the triglycerides than significantly contributing to the fatty acids via DNL. Within the 4 hours period only 0.05% of fructose was converted into the free fatty acids and only 0.15% into the triglycerides. Further conversion of fructose into fats was not examined in this particular study and to my knowledge the data on this is not available. I encourage everyone who has the data from a more recent research, to come out with that.

The second of the two studies which Sun and Empie mentioned was Tran et al. (2010) who examined a conversion of labelled fructose into fats for the duration of 6 hours, with a “small but significant increase of 13C enrichment in VLDL palmitate, but only in men, not in women”. However, it is important to note, that in comparison to baseline, after the fructose intake the levels of plasma triglycerides and FFA have decreased in both genders to a similar extend, which again suggests rather a functional activity of fructose in assembly of VLDL from already available lipid sources than significantly contributing to them by its own carbons – aka being converted to fat. The team stated twice that the DNL activity of fructose is “quantitatively minor” to other fates of this sugar and also that “only small amount of fructose is converted into fatty acids”. The team also discussed that the ingestion of fructose leads to acute suppression of NEFA (non-esterified fatty acids or FFA: free fatty acids) oxidation and suggested the mechanism via minor increase of insulin levels. I just add that if that was the reason, then the same mechanism should apply even more on glucose or starch, depending on the body insulin response after their consumption.

The advantage of this second study was that it examined non-athletic and healthy young participants, which is more representing the average population than if they were athletes or suffering a metabolic disease. My focus was to find the evidence that fructose can act as a sole factor in the development of metabolic disorders in healthy people. One of the limitations of this study was that they administered the amount of fructose according to fat-free mass (0.3g/kg FFM), of which men have proportionally more to the body weight than women do. According to Kyle et al. (2001), the average FFM of healthy young men, matched by age with the participants of Tran et al. study, is about 60% whereas for the women it is cca 43%, representing a 17% difference in fructose feeding between the participants. In other words, a man weighing 70 kg would have consumed average 37.8 g of fructose in three installments over the period of 4 hours, while a woman of the same weight had only consumed 27.1 g of fructose during the same time.  Overall, the women consumed 10 g fructose less on average than men, and this on top of the fructose in their normal diet, which contributed by average 4% to their total energy intake. The reason why have I dug into the figures so much is that fructose is primarily metabolized by the liver, not the muscles. So, if there is such a significant difference in the FFM between men and women, the difference in size and therefore the metabolic capacity of their liver is not that big. The livers of men had to work extra hard than of women, which might have added to the different outcomes. Moreover, pre-menopausal women also seem to be protected by oestrogen from various metabolic disturbances, in comparison to men of the same age, health status, size, etc. Women also showed a higher net lipid oxidation rate than men, who were more efficient in oxidizing the carbohydrates/fructose whereas the oxidation of fats was blunted in men. Men also showed higher increase of energy expenditure in comparison to women, suggesting that the futile cycles took place after the higher fructose consumption and also because of a higher muscle mass than women have. On the other hand, the oxidation of fructose and its conversion to glucose was very similar among both sexes.

Here you can see how difficult it is to assess the effect of fructose on various metabolic outcomes, especially in a non-homogenous population comprised of both sexes, different ages and menopausal state, body weight, body composition, physical activity level, etc.

Further reviewed studies listed by Sun and Empie were based on the labelled acetate, not directly assessing the labelled fructose and its incorporation into body lipids, so I will not continue further at this point. Similarly, many other studies examining fructose influence on metabolism used indirect assessment by labelled palmitate (fatty acid) and I am not going to use these here either.


Reviewing the studies in this original paper was a challenging task due to a high heterogeneity of the study designs, making the comparison between them and the overall conclusion difficult. The gathered information is often only partial, not giving a whole picture comprised of the assessed metabolic outcomes. Classic studies (other than isotopic tracer ones) have many limitations in the assessment of a direct effect of fructose consumption on the specific metabolic outcomes in humans because they cannot directly establish the individual pathways of fructose carbons disposal. Most of the isotopic tracer studies of fructose metabolism on humans in this review were performed on exercising subjects, hence the information applicable on the general population with little or no exercise is rather scarce or missing. Even the isotopic tracer studies have their limitations due to various forms of labelled substrates available. Different position of the isotope within the fructose or any other studied molecule influences the outcomes because different parts of the fructose molecule follow different metabolic pathway, at a different rate and during a different length of time until they reach common outcomes. This largely affects studying the fates of the substrate.

The available research shows that fructose is handled differently in various conditions: sex, body mass, body composition, physical activity level, energy status, family medical history or the health status of the studied subjects. There were also observed different metabolic patterns when the glucose and fructose were consumed alone or together. This is having an impact on applicability of the results from fructose-only feeding studies to the metabolic outcomes of the general population consuming the sugars combined in form of sucrose or HFCS. In addition, the subjects were observed only for a short period of time: between 2 to 8 hours. This has the implications that we do not know what happens with fructose that was not oxidized or converted to other monitored compounds within this period of time.

In order to obtain more complex map of fructose carbons disposal in the human metabolism, more studies have to be conducted. These should simultaneously test different forms of labelled fructose and for a longer period of time, in participants of different metabolic conditions and physiologic characteristics. The labelled fructose should be consumed along glucose to obtain results representative for the free-living population instead of studying each sugar separately; and the studies should be as similar to the every-day living and dietary patterns of the average population as possible, while still meeting the criteria for the experimental research.


The following main outcomes were observed from the available data:

  • Fructose is oxidized at a larger rate in non-exercising subjects than glucose, whereas the exercising subjects use glucose as the primary source of energy.
  • The combined sugars (fructose and glucose) led to a higher oxidation rate than glucose alone, but this might have been influenced by increased absorption of the combined sugar than each of them separately.
  • Diabetic and obese subjects are less efficient in oxidizing fructose, which means that they express a higher capacity for a non-oxidative disposal of fructose – potentially worsening their already disrupted metabolism.
  • The largest portion of ingested fructose was found to be converted to glucose. Depending on the duration of exercise, the subjects were able to convert up to 60% of ingested fructose into glucose within 3 hours, whereas for the non-exercising subjects it was ‘only’ 50%.
  • The amount of fructose converted to glucose when consumed along glucose (as is the case of sugar of HFCS, but here it was mixed with maltodextrin) was 29%. However the ratio of fructose to maltodextrin was 1:1.5 whereas in the sucrose it is 1:1 and in HFCS the ratio varies, depending on whether it was HFCS-55 or HFCS-42, which contain still significantly higher ratio of fructose to glucose than 1:1.5.
  • Fructose also contributes to glycogen stores (depending on the energy status of the participants), but the extend of this has not been assessed quantitatively.
  • Lactate is another substrate to which the ingested fructose is converted: 28 % in exercising subjects. The data for non-exercising subjects was not available.
  • The extend to which fructose is converted into lipids is minor. In fasted participants only 0.05% of fructose was converted into free fatty acids (FFA or NEFA in the U.S.) and only 0.15% was found incorporated into the triglycerides. In fact, fructose ingestion led rather to decreased concentration of glycerol and FFAs while increasing concentration of blood lipids – suggesting a lipid assembly function rather than contributing to the endogenous lipids by its carbons.
  • Increased fructose oxidation in tested subjects has been found to be coupled with reduced endogenous fat oxidation, especially in men, whereas pre-menopausal women were not affected. This may have been partly due to a lower fructose dose administered to women per body weight, and the size of their liver – the main organ for fructose metabolism.
  • For now, to my knowledge, there is no direct evidence that fructose per se contributes to the endogenous pool of lipids (by DNL) in humans on such a massive scale as is often mentioned in various media or other studies using isotopes of different compounds than fructose (acetate, palmitate).

A final word: I remember Dr Lustig saying in one of the videos on Youtube, that 30% of fructose is converted to fat and this is behind the increasing incidence of non-alcoholic fatty liver disease (NAFLD) in the population and also perhaps the obesity rates. Well, based on the studies here, it is true that we have an idea of how much fructose is converted to glucose and lactate and how much is oxidized within a certain period of time, but we are not sure about the rest, which could roughly correspond to those 30%. However, we do not have a solid proof for that claim as above as far as humans are concerned. If anybody of the readers have such information and a scientific evidence (in terms of isotope labelled fructose study), please come out. I will be more than happy to look at it or even accept it. For now, it appears that the effect of fructose on fat metabolism in humans has to be studied further to make such conclusions. Some studies suggest tissue damage by various metabolic consequences of fructose administration but the evidence applicable to the normal population is not straigtforward.

Meanwhile I would suggest to focus on distinguishing what role in NAFLD or other metabolic disorders plays the increasing consumption of dietary fats and oils, as it was the case for the past decades in the U.S. and also worldwide. Based on the data, we cannot blame fructose for the increased prevalence of NAFLD and other metabolic disorders, including obesity rates, when the sugar and fructose consumption has been decreasing on the population level, can we?

In the next chapter (Part 4) I will look into details of the metabolic consequences of fructose in humans and compare some other studies with the information presented by Dr Lustig.


Adopo, E., Peronnet, F., Massicotte, D., Brisson, G.R., Hillaire-Marcel, C. (1994) Respective oxidation of exogenous glucose and fructose given in the same drink during exercise. Journal of Applied Physiology 76, pp: 1014-1019

Blom, P.C., Hostmark, A.T., Vaage, O., Kardel, K.R., Maehlum, S. (1987) Effect of different post-exercise sugar diets on the rate of muscle glycogen synthesis. Medicine and Science in Sports and Exercise 19, pp: 491-496

Chong, M.D., Fielding, B.A. Frayn, K.N. (2007) Mechanisms for the acute effect of fructose on postprandial lipemia. American Journal of Clinical Nutrition 85, pp: 1511-1520

Coss-Bu, J.A., Sunehag, A.L., Haymond, M.W. (2009) Contribution of galactose and fructose to glucose homeostasis. Metabolism 58, pp: 1050-1058

Delarue, J., Normand, S.,  Pachiaudi, C., Beylot, M., Lamisse, F., Rou, J.P. (1993) The contribution of naturally labelled 13C fructose to glucose appearance in humans. Diabetologia 36, pp: 338-345

Jandrain, B.J., Pallikarakis, N., Normand, S., Pirnay, F., Lacroix, M., Mosora, F., Pachiaudi, C., Gautier, J.D., Scheen, A.J., Riou, J.P, et al. (1993) Fructose utilisation during exercise in men: rapid conversion of ingested fructose to circulating glucose. Journal of Applied Physiology 74, pp: 2146-2154

Kyle, U.G., Genton, L., Slosman, D.Ol, Pichard, C. (2001) Fat-free and fat mass percentiles in 5225 healthy subjects aged 15 to 98 years. Applied Nutritional Investigation 17, pp: 534-541

Lecoultre, V., Benoit, R., Carrel, G., Schutz, Y., Millet, G.P., Tappy, L., Schneiter, P. (2010) Fructose and glucose co-ingestion during prolonged exercise increases lactate and glucose fluxes and oxidation compared with an equimolar intake of glucose. American Journal of Clinical Nutrition 92, pp: 1071-1079

Nilsson, L.H., Hultman, E. (1974) Liver and muscle glycogen in man after glucose and fructose infusion. Scandinavian Journal o Clinical and Laboratory Investigation 33, pp: 5-10

Riby, J.E, Fujisawa, T., Kretchmer, N. (1993) Fructose  absorption. American Journal of Clinical Nutrition 58, pp: 748S-753S

Rowlands, D.S., Thorburn, M.S., Thorp, R.M., Broadbent, S., Shi, X (2008) Effect of graded frucose coingestion with maltodextrin on exogenous 14C-fructose and 13-C glucose oxidation efficiency and high-intensity cycling performance.  Journal of Applied Physiology 104. pp: 1709-1719

Sun, S.Z., Empie, M.W. (2012) Fructose metabolism in humans – what isotopic tracer studies tell us. Nutrition and Metabolism. 9(1):89. doi: 10.1186/1743-7075-9-89

Tran, C., Jacot-Descombes, D., Lecoultre, V., Fielding, B.A., Carrel, G., Le, K-A., Schneiter, P., Bortolotti, M., Frayn, K.N., Tappy, L. (2010) Sex differences in lipid and glucose kinetics after ingestion of an acute and oral fructose load. British Journal of Nutrition, 104, pp: 1139-1147

About "" Has 48 Posts

Graduated at London Metropolitan University: BSc (Hons) Human Nutrition in 2014. Working as a research assistant at the MRC, The University of Cambridge.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.