BACKGROUND: Medium chain triglycerides (MCT) are energetically less dense, highly ketogenic, and more easily oxidised than long chain triglycerides (LCT). MCT also differ from LCT in their digestive and metabolic pathways.
OBJECTIVE: To test the effects of MCT supplementation during a very low calorie diet (VLCD).
SUBJECTS AND METHODS: Three groups of tightly matched obese women with body mass index (BMI)>30 kg/m2 received an isoenergetic (578.5 kcal) VLCD (Adinax®, Novo Vital, Sweden) enriched with MCT or LCT (8.0 and 9.9 g/100 g Adinax® respectively) or a low-fat (3 g/100 g) and high-carbohydrate regimen. The diets were administered over 4 weeks. Body composition was measured with DEXA and appetite/satiety-according to Blundell. Beta hydroxybutyric acid concentration in plasma and nitrogen excretion in urine was measured during consecutive days of VLCD. The study was performed in a randomised double-blind manner.
RESULTS: The MCT group showed a significantly greater decrease in body weight during the first 2 weeks. The contribution of body fat to the total weight loss was higher while the contribution of fat-free mass (FFM) was lower. The MCT group had a higher concentration of ketone bodies in plasma and a lower nitrogen excretion in urine. Hunger feelings were less intense while satiety was higher. These differences were observed during the first 2 weeks of treatment and gradually declined during the third and fourth weeks.
CONCLUSIONS: Replacement of LCT by MCT in the VLCD increased the rate of decrease of body fat and body weight and has a sparing effect on FFM. The intensity of hunger feelings was lower and paralleled the higher increase of ketone bodies. These effects gradually declined, indicating subsequent metabolic adaptation. Further studies are required to confirm the protein-sparing and appetite-suppressing effects of MCT supplementation during the first 2 weeks of VLCD treatment.
Patients and doctors recognise the difficulty in controlling body weight with traditional diet regimens.1,2 Very low calorie diets (VLCD) were first reported in 1929 in The American Journal of Medical Science.3 Despite its long history, the broad use of VLCD has met increasing criticism. Some investigators have postulated VLCD regimens to give a higher loss of fat-free mass (FFM) when compared with conventional low calorie diets4,5 (FFM includes muscle, bone and internal organs).
Recent in vivo studies using direct nitrogen measurements have shown that net protein breakdown is reduced by 32% in the ketonic state (ie under VLCD conditions) compared to the pre-diet state.6 Ketones, the products of fat oxidation, have also been shown to be utilised by the brain as an energy source, probably ameliorating hunger feelings but, most importantly, indirectly sparing protein conversion to glucose and increasing protein synthesis.7
Other recent studies have shown that thermogenesis is greater in medium chain triglycerides (MCT) diets compared to long-chain triglyceride (LCT) diets, suggesting a reduced efficiency of utilisation.8 MCT, once hydrolysed, are metabolised more rapidly and completely than LCT. This rapid oxidation leads to greater ketone formation.8,9,10,11,12
The use of MCT in connection with energy-restricted diets has been repeatedly proposed.8,9,10,11,12,13,14,15,16,17,18 However, there are only a few studies in humans and none in connection with VLCD. The purpose of this study was to evaluate MCT supplementation during short-term treatment with VLCD.
Patients and methods
Sixty-six female patients were enrolled in the study. Recruitment was by an advertisement in a popular daily newspaper. The Ethical Committee at Sahlgrenska University Hospital approved the study and all patients were required to give their formal written consent.
The only patients included in the study were those with body mass index (BMI) higher than 30. The exclusion criteria were diabetes, hypertension, heart disease, kidney disease or any other serious disease, existing or past psychiatric problems and active medication. Patients already on a diet regimen were also excluded.
Compliance was assessed during group meetings and each patient gave the questionnaire, VAS ratings and weekly reports regarding side effects and general evaluation of the treatment, tolerance of the diet etc to the physician. There were no drops-out and no side effects were reported during this short-term dietary study.
All patients were peri-menopausal women and all blood samples were taken in the luteal phase of the menstrual cycle. The general characteristics of the patients are shown in Table 1.
Out of 66 patients, 22 were randomly assigned to a control group (VLCD with low fat content). The remaining 44 were pair-matched—VLCD+MCT or VLCD+LCT. Pair-matching involved identifying pairs of patients with close to identical BMI, age, body-fat, waist-to-hip ratio (WHR) and duration of obesity. Normal distribution of these variables was confirmed.
VLCD comprised a formula diet, which was a protein-rich meal-substitute, Adinax® (Novovital, Vimmerby, Sweden). All patients consumed three meals of 30 g Adinax® powder (ca. 105.3 kcal per portion) dissolved in 2.5 dl skimmed milk (87.5 kcal) with a total daily intake of approximately 578.4 kcal. Three different groups received three different formulas:
All groups maintained their daily intake on the isoenergetic level of approximately 578.4 kcal/day. It was assumed that the energy value of MCT corresponds to 8.0 kcal/g and LCT 9.0 kcal/g. The LCT group and the low-fat group were treated as a placebo and a control group, respectively. All but two patients from the control group continued with the entire 4 week treatment. The two who dropped out did so because of family reasons not related to the diet.
The Adinax® powder was fortified with vitamins, minerals and dietary fibre (guar gum, xanthan gum and oat bran fibres, totalling 16 mg per day). MCT comprised caprylic (C:8) and capric (C:10) acids to the amount of 9.92 g per 100 g of Adinax®.
In all groups, the total daily energy intake was maintained at 578.4 kcal. The whole study was performed double blind.
DEXA (Lunar DPX, Scanexport Medical, Helsinborg, Sweden) was used to measure body composition (FFM and body fat (BF)), as described before.19
Height, body weight and body mass index
Height was measured to the nearest 1.0 cm and weight to the nearest 1.0 kg. The measurements were performed in the morning after an overnight fast. BMI was calculated as body weight (kg) divided by height (m2).
Waist and hip circumference measurements
WHR was determined as the ratio of the circumference (cm) at the level of the umbilicus to the greatest circumference at the level of the hips.20 Before commencing the study, all patients were assessed for weight stability. The requirements were a stable body weight (change below±0.3 kg) over 3 weeks. Anthropometric and body weight measurements were performed every week while body composition assessment was performed and blood samples were taken before and after 2 and 4 weeks of VLCD.
Blood samples for routine hospital blood chemistry were taken at the start and after 2 and 4 weeks of treatment. Venous blood samples were drawn from an antecubital vein in the morning, after an overnight fast. Blood glucose was analysed using glucose-6-phosphate dehydrogenase method (Beckman Instruments, Fullerton, CA, USA).
Insulin and C-peptide levels were determined by radio-immunoassay (RIA) using the Phadebas insulin kit (Pharmacia, Uppsala, Sweden and NOVO, Copenhagen, Denmark). C-peptide was assayed with an inhouse RIA method using commercially available reagents with iodinated C-peptide (NOVO, Bagsvaerd, Denmark) and antiserum (Diagnostica, Falkenberg, Sweden). Cholesterol was measured using the CHOD PAP enzymatic calorimetric kit and triglycerides (as esterified glycerol) using the enzymatic calorimetric kit (Boehringer Mannheim Diagnostics).
The results of different measurements are reported only in the case of finding of relevant, significant changes related directly to the main thrust of the paper.
Determination of beta hydroxy-butyrate in serum
Beta hydroxy-butyrate concentrations were determined in HCLO4-neutralised extracts of serum by urine enzymatic spectrophotometric assay as described by Bergmeyer.21
Hunger and satiety
Hunger, appetite, fullness and prospective consumption were assessed using a model developed by Blundell and Hill.22 According to these authors, appetite is defined as a process which begins at the commencement of eating and guides the moment-to-moment selection of foods. Hunger is defined as a conscious sensation that is linked to a desire to obtain and eat food. The process that causes eating to stop is defined as satiation. Satiety is a state of inhibition of further eating.
The following questions were presented to each patient and rated on a VAS (visual analogue scale; the questions were related to the above definitions of appetite, hunger, satiation and satiety):
How strong is your desire to eat? (‘Very strong’ at one end, ‘Very weak’ at the other).
How hungry do you feel? (‘As hungry as I have ever felt’ at one end, ‘Not at all hungry’ at the other).
How full do you feel? (‘Very full’ at one end, ‘Not at all full’ at the other).
How much food do you think you could eat (prospective consumption; ‘A large amount’ at one end, ‘Nothing at all’ at the other).
Do you fancy something to eat? (‘Very much’, ‘Not at all’).
How satisfied are you after a meal? (‘Very satisfied’, ‘Not satisfied at all’).
All patients were asked to mark their answers with a cross on the VAS scale in connection with the second meal (dinner) twice a week. Patients answered the questions 5 min before the meal and at 5, 20, 40 and 120 min after the meal.23
For statistical calculations, all ratings relating to appetite and satiety were accumulated to create one cumulative variable for answers relating to appetite and hunger feelings. The same procedure was applied for satiety and satiation.
Nitrogen concentration in urine
Twenty-four hour urine samples were obtained to determine urinary nitrogen.24 These were collected on consecutive days of VLCD treatment commencing on day two.
Mean values and standard deviation (s.d.) were calculated using conventional methods. Non-parametric tests (Mann–Whitney and Wilcoxon) were used to calculate intra-individual differences within each group. The data were expressed as mean±standard error of the mean (s.e.m.). A P-value of<0.05 was considered significant. The relationship between the different variables was calculated using linear regression analysis and the Pearson correlation coefficient. Multiple stepwise regression as well as analysis of variance (ANOVA) was also used for intergroup comparison.
All three versions of the VLCD were perfectly tolerated, no serious side effects being reported and no drop-outs being registered during the study.
Body weight and body composition
Application of MCT-enriched VLCD resulted in a significantly greater decrease of body weight after the first and second week of dietary treatment. The difference between the groups gradually decreased throughout the subsequent 2 weeks (Table 2).
The percentage contribution of body fat to the total body weight loss was found to be significantly higher in the MCT group. It was most prominent after 1 week of treatment (56% in the MCT, 32 and 35% in the LCT and low-fat group respectively; Table 2).
The reverse was observed with FFM. The percentage contribution of FFM to the total body weight loss was lowest in the MCT group (44% in the MCT, 68 and 65% in the LCT and low-fat group, respectively after one week of treatment; Table 2).
Blood chemistry variables were measured before and after 2 and 4 weeks (Table 3). After 2 weeks, there was a significant decrease in the concentration of glucose and insulin in all three groups. There was a significant decrease in plasma cholesterol concentration in the MCT and low fat VLCD group.
After 4 weeks, the decrease in concentrations of blood glucose and plasma insulin was significant in all three groups. There was a significant decrease in plasma triglyceride concentrations in the LCT and MCT groups.
Appetite and satiety
Appetite and satiety were monitored in the tightly matched LCT and MCT groups only. After 1 week, the MCT group demonstrated significantly lower hunger feelings at all times before and after the meal (Figure 1). Similarly, a significantly higher satiety was observed at all times after the meal but not before (Figure 1).
After 2 weeks of dietary treatment, appetite and hunger feelings were still lower in the MCT group before and 40 min after the meal while satiety was higher 5 and 40 min after the meal (not shown). After 4 weeks of treatment, the difference between the groups diminished.
However, appetite was still significantly lower in the MCT group before a meal and satiety higher 5 and 40 min after the meal (not shown).
Plasma concentration of ketone bodies and urinary nitrogen excretion
The concentration of beta hydroxy-butyric acid (HBA) increased rapidly during the first week and decreased gradually in subsequent weeks. The MCT group showed significantly higher concentrations of ketone bodies (HBA) over 24 days of treatment (Figure 2).
The 24 h nitrogen excretion in urine was significantly higher in the LCT group during the first 2 weeks of treatment. In the LCT and MCT groups, nitrogen excretion gradually decreased after the first 4 days of treatment. However, in the MCT group, the decrease occurred earlier and was greater in terms of percentage of baseline values (Figure 3).
Apart from lower appetite and hunger feelings, there was no difference in compliance and side effects reported by the patients in the three groups. Moreover, because of the lower level of hunger feelings, the MCT-supplemented VLCD was well tolerated and well liked by patients during the whole period of 4 weeks.
Supplementation of VLCD with MCT resulted in a higher body weight loss than that obtained in the tightly matched group supplemented with LCT. The difference was significant only after the first and second week and diminished in the third and fourth week parallel with the decline in ketone body concentrations. The difference in total body weight appears to be mainly due to the greater decrease of BF and a higher preservation of FFM.
Changes in nitrogen excretion confirm the observed associated difference in body composition. Because FFM contains approximately 25% protein, it is possible to estimate the expected nitrogen losses during weight reduction with the assumption that protein is 16% nitrogen. However, when calculations are based on the FFM decrease, nitrogen losses are somewhat higher than actually observed. The mean nitrogen loss corresponding to weight loss was 7.6±0.9 g/kg, of the lost weight and was significantly lower in the MCT group compared to the LCT group, 9.1±0.8 g/kg of the weight loss.
As the general and dietary experimental conditions in all groups were identical, the observed difference between the groups relating to FFM and nitrogen excretion in urine most likely indicates a protein-sparing effect of MCT. The decrease in nitrogen excretion indicates a transition to a greater anabolic protein metabolism. This decrease occurred earlier and was greater in the MCT group. The decrease in nitrogen concentration and an associated significant increase in ketone body concentration was most probably a result of oxidation of ketogenic MCT.
The conclusion that can be drawn from these findings is that MCT supplementation during short-term VLCD is effective in preventing the erosion of FFM that is predominantly muscles. The loss of FFM after the first week of VLCD was nearly twice as high in both the LCT and low-fat VLCD group. The difference decreased in the subsequent weeks but the decrease of FFM was still lower in the MCT group after the fourth week.
According to Howard,4 assuming the average weight loss during VLCD corresponds to 300 g/day, the protein available in the metabolic pool from the breakdown of FFM can be calculated to be about 18 g/day.
The diet used in all three groups contained 31.5 g protein (28.5 g from the VLCD itself and 3.0 g from skimmed milk). This is lower than the US recommendations for women (44.0 g/day) but compensates for protein catabolism, eliminating a negative nitrogen balance, especially during the second phase of treatment.
Muscle combustion usually begins very early during semistarvation as glycogen stores are used up and gluconeogenesis commences, supplying glucose to the brain. During the first days of VLCD, the liver converts fat to ketone bodies, which are then oxidised by many tissues, replacing glucose as the main energy source for the brain.
In less obese or lean subjects, ketone body concentrations rise more rapidly.25 A delay in the rise in plasma ketone bodies is therefore to be expected in obese patients during fasting. In our group of grossly overweight patients, it seems likely that MCT supplementation eliminated this delay. Later on, FFA replaced ketone bodies as the major source of energy for muscles and protein breakdown simultaneously diminished. This is indicated by the lower loss of FFM in all three groups. Ketone bodies have been implicated in sparing protein directly or indirectly by replacing glucose as a fuel for the brain. The latter is generally thought to be responsible for the suppression of hunger and even for the ‘euphoric-like feelings’ commonly associated with VLCD. Whatever speculations are made about the mechanism of this phenomenon, significantly less hunger feelings and a higher degree of satiety were observed in the MCT group. The changes of these variables paralleled the changes in concentrations of ketone bodies. This makes them a likely factor responsible for the difference between the MCT and the low-fat and LCT groups.
The ratio of glucose derived from protein to glucose derived from glycerol during VLCD is lower in grossly obese women. As MCT are metabolised more rapidly, it seems probable that the ratio is even lower in the MCT group, in contrast to both lean subjects and low-fat and LCT Adinax® treated patients.
MCT feeding has been reported to increase thermogenesis to a greater extent than LCT, resulting in a reduced efficiency of utilization.11,27 This has been well illustrated in a study by Hill, who demonstrated that the MCT diet produced a higher thermal effect of food (TEF).28
TEF increased by 50% after 6 days on an MCT diet, but was unchanged after 6 days on an LCT diet. Similar observations of enhanced postprandial thermogenesis were reported for obese males when LCT was given in a mixed meal with MCT.28 Experiments in laboratory animals have shown that diets high in MCT increase thermogenesis and lead to less fat deposition compared with diets high in LCT.8,13,14 This has also been demonstrated in a number of acute metabolic rate studies in humans.15 Scalfi et al have reported similar findings with a mixed meal high in MCT.16 Flatt et al also reported greater postprandial thermogenesis and a lower respiratory quotient after a mixed meal high in MCT.29 These observations agree with the findings of Hill et al who demonstrated a higher postprandial thermogenesis acid and fat oxidation after 7 days of overfeeding with MCT.28 Recently, Dullo et al reported that a low to moderate intake of MCT leads to a 500 kJ increase in 24 h energy expenditure and an increased 24 h urinary excretion of noradrenaline.27
The net energy value of MCT can be calculated from the increase in daily energy expenditure of about 500 kJ with 15–30 g MCT per day substituted isoenergetically for LCT. The net energy value of MCT when consumed in moderate to high amounts was estimated by different groups as corresponding to 5–8.2 kcal/g. Thus MCT if given in the same amount provides less total energy than LCT in the LCT group. To eliminate this difference in energy density, the MCT group received somewhat higher amounts of the fat to assure the equality in the supplied energy.
MCT are metabolised differently from LCT. After absorption by enterocytes, MCT are released into the portal circulation and transported to the liver, where they are largely oxidised. Their intramitochondrial transport does not require carnitine palmityltransferase, unlike dietary LCT.
The oxidation of LCT is described as reduced in obesity.11 This limiting factor in LCT oxidation is probably associated to an enhanced uptake of diet-derived fatty acids and their subsequent storage in the adipose tissue, in contrast to MCT, which are totally oxidised and not stored.
These findings, together with additional evidence of enhanced postprandial thermogenesis and lipolytic effects of MCT, seem to indicate that MCT could be treated as potential supplements to hypoenergy diets.17 These results are in accordance with our findings related to differences in weight loss as well as with previous reports on MCT replacement of LCT in lean and obese patients.18 However, it seems reasonable to assume that the effect of MCT is too moderate and the amount too small for recommendation of MCT use in low-calorie diets. Large amounts of MCT would be needed, but reduced palatability of such diets would make the use of MCT in LCD less suitable, and the effect on ketone bodies a priori less evident.
This study demonstrates that supplementation of VLCD with MCT induces after 2 and 4 weeks higher glucose and insulin levels decrease than LCT-supplemented VLCD.
Although requiring confirmation in further studies, these findings corroborate the observation by Yost et al12 of enhanced insulin sensitivity and glucose disposal after 4 weeks of MCT-supplemented hypoenergy feeding. According to these authors, supplementation with MCT eliminates an inhibitory effect of LCT-derived fatty acids on insulin-mediated carbohydrate metabolism known as Randle's effect. However, this reasoning may not necessarily apply to a very low calorie diet, where even small changes in different components of the diet can contribute to relatively large differences in the rate and degree of body weight loss and body composition.
Supplementation with MCT increased the contribution of body fat to total weight loss during the first 2 weeks and decreased that of FFM. The metabolic differences, ie the concentrations of ketone bodies and nitrogen excretion in urine, gradually diminished at a later stage. This explains the parallel decreasing differences in body weight loss and body composition during the subsequent weeks. The same pattern is observed with respect to appetite and satiety. The difference between the groups becomes less evident during longer treatment, which corresponds well to the diminishing difference in the concentrations of beta hydroxybutyric acids.
The amount of MCT was relatively small. It was limited due to the influence of MCT on the palatability of the formula diet. Furthermore, it seems likely that the metabolic impact of MCT was evident under conditions of ketonic diet only. To obtain similar effects during normal intake and/or LCD would require several times higher amounts of MCT, which hardly seems feasible from a practical point of view.
It seems tempting to speculate that the gradual exclusion of MCT-rich palm oil from the diet in countries of south Asia has contributed to the epidemic of obesity observed recently in this region.
In conclusion, the results of the present study seem to indicate that supplementation of VLCD with MCT can initially accelerate the rate of body weight loss and decrease the contribution of FFM to the total weight loss. The differences in the first 2 weeks correspond to a decrease in appetite and an increase in satiety. The effects of supplementation with MCT gradually diminished along with the adaptive changes in metabolism with longer lasting VLCD. Further confirmatory studies are required to substantiate the value of MCT supplementation in short-term VLCD treatment.
Stunkard AJ, McClaren-Hume M . The results of treatment of obesity Arch Intern Med 1959 103: 79–85.
Atkinson RL . Low and very low calorie diets Med Clin N Am 1989 73: 203–215.
Evans FA, Strong JM . A departure from the usual methods of treating obesity Am J Med Sci 1929 177: 339–348.
Howard AN . The historical development of very low calorie diet Int J Obes 1989 13: (Suppl 2): 1–9.
Wadden TS, Van Itallie TB, Blacburn GC . Responsible and irresponsible use of very low calorie diets in the treatment of obesity JAMA 1990 263: 83–85.
Ryde SIS, Saunders NH, Birks JL, Ali PA, Thomas DW, Morgan VD, Evans CJ, Al-Zeibak A, Dutton J, Sivyer A . The effects of VLCD on body composition. In: Krietzman SN, Howard AN (eds). The Swansea trial. Smith Gordon: London 1993 31–54.
Hoffer LJ, Bistrian BR, Young VR, Blackburn GL, Matthews DE . Metabolic effects of very low calorie weight reduction diets J Clin Invest 1982 73: 750–758.
Baba N, Bracco EF, Hashim SA . Enhanced thermogenesis and diminished deposition of fat in response to overfeeding with a diet containing medium chain triglycerides Am J Clin Nutr 1982 35: 678–682.
Bach AC, Babayan VK . Medium chain triglycerides. An update Am J Clin Nutr 1982 36: 950–962.
Yeh YY, Zee P . Relation of ketosis to metabolic changes induced by acute medium chain triglycerides feeding, in rats J Nutr 1973 160: 58–67.
Binnart CH, Pachiandi CH, Beylot M, Hans D, Vandermander J, Chantre Ph, Riou JP, Laville M . Influence of human obesity on the metabolic fate of dietary long and medium chain triglycerides Am J Clin Nutr 1998 67: 595–601.
Yost TJ, Eckel RH . Hypoenergy feeding in obese women. Metabolic effects of medium chain triglycerides substitution Am J Clin Nutr 1989 49: 326–330.
Geliebter A, Torboy N, Bracco FE, Hashim SA, Van Itallie TB . Overfeeding with medium chain triglyceride diet results in diminished deposition of fat Am J Clin Nutr 1983 37: 1–4.
Rothwell NJ, Stock MJ . Stimulation of thermogenesis and brown fat activity in rats fed medium chain triglyceride Metabolism 1987 36: 128–130.
Seaton TB, Welle SL, Warenko MK, Campbell RG . Thermic effect of medium chain and long chain triglycerides in man Am J Clin Nutr 1986 44: 630–684.
Scalfi L, Coltorti A, Contaldo F . Postprandial thermogenesis in lean and obese subjects after meals supplemented with medium chain and long chain triglycerides Am J Clin Nutr 1991 53: 1130–1133.
Papaniandjaris AA, MacDougall DE, Jones PJH . Medium chain fatty acid metabolism and energy expenditure. Obesity treatment implications Life Sci 1998 62: 1203–1215.
Kaunitz H, Slanetz CA, Johnson RE, Babayan VK, Barsky G . Relation of saturated, medium and long chain triglycerides to growth, appetite, thirst and weight maintenance requirements J Nutr 1958 64: 513–524.
Krotkiewski M, Landin K, Mellström D, Tölli J . Loss of total body potassium during rapid weight loss does not depend on the decrease of potassium concentration in muscles. Different methods to evaluate body composition during a low energy diet Int J Obes Relat Metab Disord 1999 23: 1–7.
Krotkiewski M, Björntorp P, Sjöström L, Smith U . Impact of obesity on metabolism in men and women. Importance of regional adipose tissue distribution J Clin Invest 1983 72: 1150–1162.
Bergmeyer FW . Methods of enzymatic analysis, 2nd edn. Academic Press: New York 1974.
Hill AJ, Blundell JE . Macro-nutrients and satiety. The effects of a high protein or a high carbohydrate meal on subjective motivation to eat and food preferences Nutr Behav 1986 3: 133–144.
Blundell JE, Burley VJ . Satisfaction, satiety and the action of dietary fibre on food intake Int J Obes 1987 11(Suppl): S9–S25.
Di Gorcio J . Non-protein nitrogenous contribents. In: Henry RJ, Cannon DC, Winkelman JW (eds). Clinical chemistry. Principles and techniques Harper & Row: Hagerstown 1974 pp 503–563.
Elia M, Stubbs RJ, Henry CJK . Differences in fat, carbohydrate and protein metabolism between lean and obese subjects undergoing total starvation Obes Res 1999 7: 597–604.
Peters JC, Holcombe BN, Hiller LK, Webb DR . Absorption and calorie value in adult humans J Am Coll Toxicol 1991 10: 357–367.
Dullo AG, Fathi M, Mensi N, Girardier L . Twenty-four energy expenditure and urinary catecholamines of humans consuming low fat to moderate amounts of medium chain triglycerides. A dose–response study in human respiratory chamber Eur J Clin Nutr 1996 50: 152–188.
Hill JO, Peters JC, Yang D, Sharp T, Kaler M, Abumrad NN, Greene HL . Thermogenesis in man during overfeeding with medium chain triglycerides Metabolism 1989 38: 641–648.
Flatt JP, Ravussin E, Acheson KJ, Jéquier E . Effects of dietary fat on postprandial substrate oxidation and on carbohydrate and fat balance J Clin Invest 1985 76: 1019–1024.
About this article
Lipids in Health and Disease (2018)
European Journal of Clinical Nutrition (2017)