Obesity has been recognized as one of the most serious public health challenges. Strategies for the prevention and treatment of obesity are a high public health priority. Weight gain that leads to obesity results from a positive energy balance: energy intake exceeds energy expenditure over some period. Moreover, an impaired ability to oxidize lipids may be an important factor in the etiology of obesity, and this may contribute to the development and maintenance of large lipid stores (18,35). Several physiological mechanisms have been suggested for this reduced lipid use in obesity: 1) low activity of β-oxidation enzymes (20), 2) low skeletal muscle lipoprotein lipase activity (15), and 3) impaired mobilization of lipid depots (6). Endurance exercise training and regular physical activity have been shown to increase total energy expenditure (25) and lipid oxidation at rest (32) and during submaximal exercise in obese (37) and healthy lean subjects (32,34). Thus, endurance training may contribute significantly to weight management in obese subjects and individuals at risk for obesity.
However, the mobilization of nonesterified fatty acids (NEFA) from adipose tissue (38) and the extraction of NEFA in the exercising muscles (26) may be the limiting factors for lipid oxidation during exercise, especially at high exercise intensity (29), and may be insufficient to affect lipid stores in adipose tissue (22). On the other hand, it has been proposed that the recycling of NEFA and triacylglycerol (TAG) between adipose tissue and the liver plays a role in excess postexercise oxygen consumption (EPOC) (2). Furthermore, there is a substantial postexercise lipolytic activity and NEFA mobilization from subcutaneous adipose tissue, beginning about 1 h after the exercise period and continuing for more than 3 h (26). Thus, the effect of the relative contribution of lipids to energy expenditure in postexercise recovery may be important in the management of adiposity.
Recently, Kuo et al. (22) and Henderson et al. (17) have shown that the contribution of lipid oxidation to energy expenditure after two levels of isoenergetic exercise (45% and 65% of maximal oxygen uptake [V˙O2max]) is significantly increased during postexercise recovery compared with preexercise and time-matched no-exercise control period values. These authors have suggested that whereas muscle and liver glycogen stores are mobilized to support most the energy expenditure during exercise, lipids become the predominant energy source during postexercise recovery to spare CHO and facilitate subsequent restoration of glucose homeostasis and glycogen repletion (22). Indeed, recent results have shown that despite the elevation of glucose and insulin concentrations after high-CHO meals during recovery, CHO oxidation and pyruvate dehydrogenase activation were decreased, supporting the hypothesis that muscle glycogen resynthesis is a high metabolic priority resulting in a relative shift from CHO to lipid oxidation (21). Plasma NEFA, very low density lipoprotein TAG, and intramuscular acetylcarnitine stores are likely to be important in providing fuel for muscle aerobic metabolism, particularly during the first few hours of recovery (21). However, the role of intramuscular TAG in contributing to enhance lipid oxidation during recovery from exhaustive exercise remains unclear (19,21).
It has been shown that the CHO oxidation rate is 1.2 times higher during interval exercise (12 s at 120% V˙O2max and 18 s of passive recovery for 90 min) compared with submaximal continuous exercise matched for total energy expenditure (11). Thus, it is possible to hypothesize that muscle glycogen stores could be depleted to a greater extent during high-intensity interval exercise, thereby leading to a larger reliance on lipid oxidation during postexercise recovery. To our knowledge, few studies have investigated the metabolism during recovery after high-intensity interval exercise (i.e., submaximal exercise (9,24) and supramaximal exercise (3,23)) in healthy young male subjects. These studies have mainly focused on EPOC and have shown that the respiratory exchange ratio (RER) is lower in postexercise recovery after interval exercise relative to the continuous exercise (24) and no-exercise control trials (3,23). Still, no studies have accurately investigated substrate partitioning during recovery after high-intensity interval submaximal exercise in healthy young men.
Therefore, the purpose of this study was to examine the energy substrate partitioning difference during 3 h of postexercise recovery after continuous and interval exercise matched for mechanical work output on a cycle ergometer. It was hypothesized that greater CHO oxidation during high-intensity interval submaximal exercise could shift the pattern of substrate use toward lipids during postexercise recovery, when lipid oxidation predominates.
Twelve fit young men (mean ± SE: 24.6 ± 0.6 yr, range = 21-29 yr) were recruited from the university by an e-mail announcement. All subjects were nonsmokers, disease free, physically active (7.4 ± 1.1 h·wk−1), accustomed to physical exercise, and not taking any medications. Typical activities were competitive athletics, basketball, badminton, soccer, hockey, volleyball, and swimming, but not cycling. All test procedures and risks were fully explained, and subjects were asked to provide written consent for participation. The protocol and the consent form were approved by the local ethics committee.
Each subject completed four test sessions. In the first session (preexperimental testing), the subject was introduced to the experimental procedures and anthropometric measurements (i.e., height, body mass, and body composition) were taken. Each subject then performed a maximal incremental test on a cycle ergometer to determine maximal aerobic power output (W˙max), maximal oxygen uptake (V˙O2max), and ventilatory thresholds. For the remaining sessions, after a 25-min resting period, all subjects were tested in a random order in three conditions separated by approximately 7 d: 1) continuous exercise for 1 h; 2) interval exercise consisting of a 1-min work followed by a 1-min active recovery for a duration adjusted for each subject to match the mechanical work output to continuous exercise; and 3) control trial of a 1-h resting period. Each trial was followed by a 3-h recovery period where subjects were seated quietly at a table. The three trials were always performed in the morning (start of exercise between 8:00 and 9:00 a.m.) after an overnight fast and at the same time to avoid circadian variance. Subjects were asked to fill in a 1-d food diary on the day before their first trial, and they were asked to repeat this diet before all subsequent experiments. Furthermore, subjects were asked to avoid strenuous exercise the day before each experimental trial.
Body mass composition.
Percent body fat was estimated from skinfold thickness measurements at four sites according to the methods of Durnin and Womersley (12).
W˙max, V˙O2max, and ventilatory thresholds were determined by a maximal incremental test until exhaustion on an electronically braked ergometer (Ebike Basic BPlus, General Electric, Niskayuna, NY). The starting workload was 60 W (for 5 min), and this was increased by 30 W·min−1 until exhaustion. Oxygen uptake (V˙O2), CO2 output (V˙CO2), ventilation (V˙E), and RER were measured continuously using a breath-by-breath online system (OxyconPro, Jaeger, Germany). OxyconPro has been shown to be a valid and reliable (i.e., day-to-day and within day variations) system for generating accurate and repeatable respiratory data for V˙O2, V˙CO2, V˙E, and RER during metabolic simulator and exercise compared with Douglas bags (10). Moreover, before the start of study, the validity and the reliability of the metabolic card used in the present study were ascertained over a wide range of simulated breathing flows. Before each experimental period, the gas analyzers were calibrated with air and gases of known concentration (16% O2 and 5.02% CO2), and the volume was automatically calibrated at two different flow rates (0.2 and 2 L·s−1). The heart rate (HR) was also measured continuously using an HR monitor (S810i, Polar Electro OY, Kempele, Finland). V˙O2max was considered to be the point when at least three of the four following criteria were met: 1) a plateau in V˙O2 concurrent with continuous increase in exercise workload (<100 mL·min−1); 2) an HR >90% of the age-predicted maximal heart rate (HRmax = 220 − age); 3) an RER >1.1; and 4) an inability to maintain the minimal required pedalling frequency (i.e., 60 rpm) despite maximum effort and verbal encouragement.
Ventilatory thresholds 1 (VT1) and 2 (VT2) were determined as described in the literature using the Wasserman's ventilatory method, which consists of visually determining the point at which the V˙O2 respiratory equivalent (V˙E/V˙O2) increases whereas the V˙CO2 ventilatory equivalent (V˙E/V˙CO2) remains stable (VT1) and the point at which the respiratory equivalents increase together (VT2) (40). The estimate of VT1 was supported using the Beaver ventilatory method, which consists of visually determining the inflection point of V˙CO2 with respect to V˙O2 (5). Two blinded and independent investigators determined VT1 and VT2.
Three experimental trials each consisted of three phases: preexercise resting phase, exercise phase, and postexercise recovery phase. The pre- and postexercise phases were identical for the three experimental conditions.
Preexercise resting phase.
After 15 min of a seated resting period, each subject remained seated for 10 min on the cycle ergometer and was connected to the metabolic system. Average HR and gas exchange data during the final 5 min were used as the baseline.
The exercise phase consisted of a warm-up followed by either continuous or interval exercises on the cycle ergometer. The warm-up lasted for 5 min at an intensity of 20% of W˙max and was the same for both exercise trials. The continuous exercise lasted 1 h at 45% of V˙O2max (C45%). Workload at 45% of V˙O2max was determined using a linear regression of data obtained during the maximal incremental test. The interval exercise (INT) consisted of 1 min at 80% of W˙max followed by 1 min at 40% of W˙max for a total duration adjusted for each subject to match the mechanical work output to C45%. Subjects were asked to rate their perceived exertion (RPE) at exercise cessation using the 6-20 Borg's category scale. The control trial (CON) was a 65-min resting period seated at a table (i.e., equivalent to 5 min warm-up + 60 min exercise). Gas exchange data and HR were collected continuously throughout the exercise phase of the three experimental conditions. Respiratory and HR values were averaged for each 1-min period.
Postexercise recovery phase.
At the end of each exercise period, subjects were instructed to stop cycling and to remain seated on the cycle ergometer for the first 15 min of recovery while expired air and HR were constantly monitored. After this period, the subject moved to a seat at a nearby table to rest for the remaining 165 min. Gas exchange data and HR were continuously monitored for the duration of the recovery period with the exception of a 10-min period (minutes 15 to 25 of the recovery phase) and a 15-min period (minutes 80 to 95 of recovery phase) when the mask was removed for the subjects' relief. During the second 15-min period, the gas analyzers and volumes were calibrated. Respiratory and HR values were averaged over 15-min periods. Throughout the study period when subjects were not wearing the mask, they were allowed to drink water ad libitum.
To avoid potential instability of the plasma bicarbonate pool during the INT and the postexercise recovery periods, we used indirect calorimetry for the determination of substrate oxidation rates when constant V˙E, V˙CO2, and arterial carbon dioxide partial pressure (PaCO2) responses were observed (11). Based on mass balance considerations, PaCO2 was calculated as
where 863 is the product of barometric pressure, temperature, and water vapor correction (i.e., factors needed to express V˙E at BTPS, V˙CO2 at STPD, and CO2 as a partial pressure), VD is the physiologic dead space, and VT is the tidal volume.
During INT, data for respiratory gas exchange were interpreted across work and recovery combined because the time lag between the cellular exchange of O2 and CO2 and the measurement of respiratory gas exchange preclude the analysis of work and recovery as distinct metabolic states during intermittent exercise (13). The stability of V˙E, V˙CO2, and PaCO2 in INT was tested as a function of percent of total exercise duration for each subject.
Energy expenditure (kcal·min−1) was calculated using caloric equivalents of the nonprotein respiratory quotient (RQ) (1) and lipid, and CHO oxidation rates was calculated using the stoichiometric equations of Frayn (16), with the assumption that the urinary nitrogen excretion rate was negligible:
where V˙O2 and V˙CO2 are in liters per minute.
During INT for RER data above 1, the CHO oxidation rate was calculated by dividing energy expenditure for RER = 1 (e.g., maximal aerobic CHO metabolism) by the energy equivalent of CHO (1 g = 4 kcal).
To determinate an adequate sample size, we calculated the statistical power (http://www.cs.uiowa.edu/∼rlenth/Power/) to detect changes in RER during postexercise recovery using an alpha level of 0.05. The power analysis assumed two-way repeated-measures ANOVA of the differences between the experimental trials during postexercise recovery over time. Using the RER values of postexercise recovery of McGarvey et al. (24), a minimal sample size of 3 was necessary to detect changes between experimental trials during postexercise recovery with an 80% power. Nevertheless, a sample size of 12 has been chosen to be consistent with previous studies (22,24).
A two-way (time × experimental trial) repeated-measures ANOVA test was used to determine differences in respiratory data (V˙O2, V˙E, V˙CO2, PaCO2, and RER), energy expenditure, and substrate oxidation (i.e., the primary outcome variables of the study: absolute and relative lipid and CHO oxidation values and their relative contribution to energy expenditure). When the assumption of normality of distribution or the equality of variance was violated, a one-way ANOVA or an ANOVA (Kruskal-Wallis) for nonparametric values was used to determine the significance of differences among trials and over time. Significance was located with post hoc analysis using the Tukey test. Data are expressed as mean ± SE for all variables, and the level of significance was set at P < 0.05. Statistical power for the various comparisons among the primary outcome variables was >80% except for the comparisons during the preexercise resting period (5%) and for the comparison of the total energy expenditure during postexercise recovery among trials (42%).
Anthropometric characteristics of the study participants are listed in Table 1. Table 2 shows W˙max, V˙O2max, VT1, and VT2 as determined by the maximal incremental test.
Interval Exercise Characteristics
During INT, the mean V˙O2 and V˙E were constant across the final 60% and 50% of exercise duration, respectively (Figs. 1A and B, respectively; P > 0.05). There were no significant differences between V˙CO2 values during INT (Fig. 1C; P = 0.77). The mean RER was stable across the final 70% of INT duration (Fig. 1D; P > 0.05). There were no significant differences between PaCO2 values during the final 40% of INT period (Fig. 1E; P > 0.05). Thus, respiratory gas exchange data for indirect calorimetry collected in the initial 60% of INT duration were excluded from analysis due to the potential instability of plasma bicarbonate in this period.
Postexercise Recovery Characteristics
For each exercise trial, there were no significant differences in V˙E and V˙CO2 during the 3-h postexercise recovery (Table 3; P > 0.05). After C45%, the mean PaCO2 was significantly lower in the first hour than in the third hour of postexercise recovery (Table 3; P = 0.048). During recovery after INT, the mean PaCO2 was significantly lower in the first hour than in the second and third hours (Table 3; P = 0.041 and 0.003, respectively). In contrast, there was no significant difference in PaCO2 during the second and third hours of recovery after each exercise trial (Table 3; P > 0.05). There were no significant differences in V˙E, V˙CO2, or PaCO2 during the 3-h postexercise recovery of CON (Table 3; P > 0.05). Considering the second and third hours of postexercise recovery, there were no significant differences in V˙E, V˙CO2, or PaCO2 values among trials (Table 3; P = 0.10, 0.14, and 0.61, respectively) and from corresponding preexercise resting values (Table 3; P > 0.05).
Because V˙E, V˙CO2, and PaCO2 values were stable in the second and third hours of postexercise recovery of each trial, respiratory gas exchange data for indirect calorimetry to determinate substrate oxidation rates collected in the first hour of recovery were excluded from statistical analysis due to the potential instability of plasma bicarbonates in this period. Therefore, for all metabolic variables, only the second and third hours of recovery were considered as total postexercise recovery.
Oxygen Uptake and Rating Perceived Exertion Scale
The mean resting V˙O2 in INT was significantly higher than in CON (Table 4; P = 0.004). The mean exercise duration was significantly lower in INT (34.7 ± 1.2 min) than in C45% (60 min; P < 0.001), and there were no significant differences in mechanical work performed between exercise trials (C45% = 426 ± 24 kJ; INT = 423 ± 24 kJ; P = 0.14). The mean relative intensity was 47.7 ± 1.0% of V˙O2max in C45% and 79.5 ± 1.4% of V˙O2max in INT, and the mean V˙O2 in INT was significantly higher than in C45% (Table 4; P < 0.001) and in CON (Table 4; P < 0.001). The mean RPE after INT (17.5 ± 0.4; range = 15-20) was significant higher than after C45% (11.8 ± 0.4; range = 9-14; P < 0.001). During postexercise recovery, the mean V˙O2 was significantly higher in INT than in C45% and CON (Table 4; P < 0.05).
Respiratory Exchange Ratio
There were no significant differences in RER values during the preexercise resting period among experimental trials (Table 4; P = 0.81). During exercise, the mean RER in C45% was significantly lower than in INT (Table 4; P < 0.001). RER values in both exercise trials were significantly higher as compared with CON and the corresponding preexercise resting values (Table 4; P < 0.001 for both). During postexercise recovery, the mean RER was significantly lower in INT and C45% compared with CON (Table 4; P = 0.002 and 0.003, respectively), and for all trials, RER values were significantly lower relative to the preexercise resting periods (Table 4; P ≤ 0.02). The mean values for RER during preexercise resting period, exercise, and postexercise recovery are shown in Figure 2.
Total Energy Expenditure
During the preexercise resting period, the total energy expenditure was significantly higher in the INT relative to the CON trial (Table 5; P = 0.004). There was no significant difference in total energy expenditure during the C45% and the INT exercise trials (Table 5; P = 0.55). During postexercise recovery, there was no significant difference in total energy expenditure among trials (Table 5; P = 0.054). Considering exercise plus recovery, the mean total energy expenditures in INT and C45% were significantly higher than that in CON (Table 5; P < 0.001 for both).
Absolute substrate oxidation.
During the preexercise resting period, there were no significant differences in absolute substrate oxidation among the various trials (Tables 4 and 5; P > 0.05). The contribution of CHO to energy expenditure and the CHO oxidation rates during exercise were significantly higher in INT than in C45% and CON (Tables 4 and 5; P < 0.001 for all). The contribution of lipids to energy expenditure and the lipid oxidation rate during exercise were significantly higher in C45% than in INT and CON (Tables 4 and 5; P < 0.001 for all). On the other hand, there were no significant differences in these variables between INT and CON (Tables 4 and 5; P = 0.94 and 0.57, respectively). Figures 3 and 4 show the lipid and the CHO oxidation rates during postexercise recovery in the three experimental trials. The mean total lipid oxidation rates and the mean contribution of lipids to energy yield were significantly higher in INT and C45% compared with CON (Tables 4 and 5; P ≤ 0.007). After 120 min postexercise, the lipid oxidation rate was higher in the exercise trials than in CON (Fig. 3; P ≤ 0.03). The mean total CHO oxidation rate and the mean contribution to energy yield were significantly lower in INT and C45% compared with CON (Tables 4 and 5; P ≤ 0.02). Considering exercise plus recovery, the total amount of lipid oxidized was significantly higher in C45% (25.7 ± 2.9 g) than in INT (12.0 ± 1.7 g) and CON (11.9 ± 0.7 g; P < 0.001 for both). The total amount of CHO was significantly higher in INT (136 ± 6.9 g) than in other trials (P < 0.001 for both) and in C45% (109 ± 5.4 g) compared with CON (22.5 ± 1.1 g; P < 0.001). A significantly larger contribution of lipids to energy expenditure was found in C45% compared with INT and CON during exercise plus recovery (Table 5; P < 0.001 for both). The mean total lipid oxidation rate during postexercise recovery was positively correlated with total energy expenditure of exercise trials (Fig. 5; r = 0.47; P = 0.02).
Relative substrate oxidation.
During the preexercise resting period, there were no significant differences in relative substrate oxidation among the various trials (Table 6; P = 0.79). During exercise, a larger relative contribution of CHO oxidation was found during INT than in other trials (Table 6; P < 0.001 for both). A greater relative lipid contribution to energy expenditure was found in C45% compared with INT (Table 6; P < 0.001). Also, relative lipid oxidation was higher in CON than in both exercise trials (Table 6; P < 0.001 for both). During the second and third hours of postexercise recovery, a significantly greater relative lipid oxidation and a lower relative CHO contribution were found in INT and C45% than in CON (Table 6; P ≤ 0.003 for both).
The aim of this study was to examine the energy substrate partitioning difference during 3 h of postexercise recovery after continuous or interval exercise matched for mechanical work output on a cycle ergometer in fit young men. The main finding of this investigation indicated that the lipid oxidation rate and the contribution of lipids to energy yield during postexercise recovery were increased by a similar amount on high-intensity interval submaximal exercise and moderate-intensity continuous exercise compared with the no-exercise control trial. Contrary to our preliminary hypothesis, the greater CHO oxidation rate during interval exercise was probably too small to lead to larger muscle glycogen depletion and to enhance lipid oxidation during postexercise recovery. Moreover, the postexercise lipid oxidation rate was positively correlated with total energy expenditure of exercise trials. Therefore, postexercise lipid oxidation may be more dependent upon the energy expenditure of exercise than on the intensity of the prior isoenergetic exercise (17,22).
In the present study, substrate oxidation was quantified by indirect calorimetry. The calculation of substrate oxidation rates by this technique is based on the assumption that V˙O2 and V˙CO2 measured at the mouth reflect the O2 consumption and the CO2 production at the tissue level, and thus the RER equals the cellular respiratory quotient (RQ) (14). Although O2 uptake measured at the mouth follows whole-body O2 consumption very quickly due to the absence of large O2 reserves in the body, cellular production of CO2 is not instantaneously translated to changes in V˙CO2 at the mouth because the body stores of CO2 are large and have complex kinetics (4). Indeed, in exercise intensity above VT1, as with INT in this study, lactate production and hyperventilation affect bicarbonate kinetics (i.e., buffering of the lactic acid by bicarbonates). Therefore, V˙CO2, as measured in expired air, reflects the metabolic production of CO2 as well as the depletion of bicarbonate pools (14). This will result in RER exceeding RQ and can potentially lead to an overestimation of CHO and an underestimation of lipid oxidation during exercise. Consequently, the presence of a stable bicarbonate pool is required for the reliable estimation of tissue CO2 production from measurement of expired V˙CO2 during high-intensity interval exercise (11,28). Nevertheless, during acute metabolic acidosis of high-intensity exercise, RER exceeds RQ until a new steady state is attained (i.e., the CO2 size is again constant although depleted), at which time RER again equals RQ (40). In fact, it has been suggested that constant lactate concentrations are associated with a stable bicarbonate level during sustained constant heavy work rate exercise, despite an initial decrease in bicarbonate concentration corresponding to initial increase in lactate (39). Christmass et al. (11) have shown that capillary bicarbonates, PaCO2, and pH decrease in the initial 10 min of interval exercise (12 s at 120% V˙O2max and 18 s of passive recovery for 90 min) but subsequently remain stable in the presence of a constant blood lactate as well as V˙E and V˙CO2. Taken together, these results suggest the presence of stable plasma lactate and bicarbonate pools and support the use of indirect calorimetry for estimating substrate oxidation rates across the final 75 min of high-intensity supramaximal interval exercise (11). Therefore, in the present study, indirect calorimetry was used for INT when constant V˙E, V˙CO2, and PaCO2 responses were observed (i.e., the final 40% of exercise duration) to avoid potential instability of the plasma bicarbonates due to the high exercise intensity during work periods. Moreover, it has been shown that close agreement exists between the estimates of substrate oxidation rates measured with indirect calorimetry and by an isotope method during strenuous exercise at ∼85% of V˙O2max (i.e., >80% V˙O2max of the present study) (28) and that there was good agreement between the leg RQ and the RER during high-intensity interval exercise (15 s at V˙O2max and 15 s of rest for 60 min) (13).
Skeletal muscle (30) and arterial blood bicarbonate pools (33) that are depleted during exercise in an intensity-dependent manner (33) would be replenished during the postexercise recovery, providing a nonpulmonary destination for metabolically produced CO2. Therefore, RER is lower than RQ, and the lipid oxidation rate may be overestimated during recovery. Recently, Henderson et al. (17) have shown that bicarbonate/CO2 retention is transiently increased for the first hour of postexercise recovery after 1 h of exercise at 65% V˙O2max but is not increased after isoenergetic bouts of exercise at a lower endurance (45% V˙O2max). In addition, it has been demonstrated that after exercise at 75% V˙O2max, blood bicarbonate levels return to baseline during the 30-min postexercise recovery period (27). In the present study, there were no significant differences in V˙E, V˙CO2, and PaCO2 for each experimental trial during the second and third hours of postexercise recovery. Combining these results with those of previous studies (17,27), respiratory gas exchange data for indirect calorimetry to determine substrate oxidation rates collected in the first hour of recovery were excluded from statistical analysis, and the use of the mean RER value during the second and third hours of postexercise recovery thus appears to be a valid indicator of substrate partitioning.
During the preexercise resting period, V˙O2 and the total energy expenditure in INT were significantly higher than in CON (4.5% and 6%, respectively). To evaluate the day-to-day variability of these variables, we calculated the coefficient of variation among trials. The resting V˙O2 and the total energy expenditure exhibited variations of approximately 10.5 ± 1.1% and 10.3 ± 1.1%, respectively. These values are similar to the 9% coefficient of variation for resting V˙O2 reported by Carter and Jeukendrup (10), who suggested that this variation is probably linked to significant biological variability in the responses (10). Moreover, the differences in the mean resting V˙O2 and the total energy expenditure between INT and CON during the preexercise resting period are 15.6 mL·min−1 and 0.5 kcal, respectively. These magnitudes are very small and confirm the lack of any significant physiological change.
During exercise, lipid oxidation rates were approximately three times lower in INT as compared with C45%, despite similar overall energy expenditure in both exercise trials. Higher CHO oxidation rates during INT (∼2 times) compensated for the lower lipid contribution. These results agree with previous findings, which showed similar substrate partitioning during supramaximal interval exercise on a treadmill (12 s at 120% V˙O2max and 18 s of passive recovery) compared with continuous submaximal exercise (60% V˙O2max) matched for total energy expenditure (11). However, to obtain two isoenergetic exercise bouts and to compare INT with a continuous exercise with a recommended duration (60 min) and intensity (40-50% V˙O2max) for weight management (8) and to enhance lipid oxidation (37) in obese individuals, the mean exercise duration was significantly lower in INT (∼35 min) than in C45% (60 min). Thus, the contribution of CHO to energy expenditure was significantly but only slightly higher in INT compared with C45% (+20%; ΔCHOINT-C45% = 106.4 kcal ∼ 26 g). Moreover, it has been shown that the duration and the intensity of exercise influence EPOC and postexercise substrate use (7). However, the mean RPE (∼17 = "very hard") and the mean relative intensity (∼80% V˙O2max) in INT indicated that our healthy fit subjects were close to the upper limit of the exercise duration that may be borne during INT session.
The results of the present study show that during the postexercise recovery period, the mean lipid oxidation rate and the mean contribution of lipids to energy yield were increased by a similar amount in INT and C45% compared with CON, and the contribution of CHO to energy expenditure and the relative CHO oxidation after INT and C45% were lower as compared with CON. Moreover, the postexercise lipid oxidation rate in INT and C45% was positively correlated with total energy expenditure of exercise trials. These results are in line with previous findings (17,22), which showed that the contribution of lipid oxidation to energy expenditure after isoenergetic low- and moderate-intensity exercise (45% and 65% V˙O2max, respectively) was higher during postexercise recovery compared with time-matched no-exercise control period values. These authors further suggested that isoenergetic exercises result in similar elevations in postexercise lipid oxidation with no significant effect of exercise intensity on substrate oxidation in recovery. To our knowledge, our study is the first to confirm this evidence comparing two isoenergetic exercises of different forms (i.e., interval vs continuous) and intensities (i.e., 45% vs 80% V˙O2max) in healthy male subjects. In the present study, the mean RER after exercise was lower than preexercise and CON values in both trials. These results confirm the difference in substrate partitioning between exercise and no-exercise control trials and corroborate previous findings concerning postexercise recovery (1 h , 2 h , and 4 h ) after high-intensity interval exercise. It has been shown that RER is lower in recovery after supramaximal interval exercise (20 × 1-min intervals at 105% V˙O2max with intervening 2-min rest periods ; 3 × 2-min intervals at 108% V˙O2max with intervening 3-min rest periods ) compared with the no-exercise control trial. Moreover, our findings are in line with those of Saris and Schrauwen (31), which showed that 24 h of postexercise RER was similar after INT (3 × 30 min of 2.5-min intervals at 80% W˙max with intervening 2.5-min active rest periods at 50% W˙max) and continuous exercise (38% W˙max) in obese men. On the other hand, our results partially agree with findings of McGarvey et al. (24) who have shown that RER is lower in postexercise recovery after interval submaximal exercise (7 × 2-min intervals at 90% V˙O2max with intervening 3-min active recovery at 30% V˙O2max) compared with preexercise values but also in recovery after continuous exercise (∼30 min at 65% V˙O2max). This difference may be explained by the fact that exercise bouts were not matched by energy expenditure and that authors did not verify the validity of indirect calorimetry to determine substrate oxidation across postexercise recovery. In fact, the dip in RER values during the first hour of recovery after both interval and continuous exercise is indicative of CO2 retention to replenish bicarbonate stores with the more pronounced dip after interval likely the result of greater lactate buffering.
As a result, the data in the present study support the assertion that greater lipid oxidation during postexercise recovery may be due to sparing CHO and facilitating subsequent muscle glycogen restoration, which represents a priority during recovery (19,21). Recently, Henderson et al. (17) have shown that lipolysis and NEFA mobilization are elevated in postexercise recovery and plasma NEFA oxidation is also increased, leading to an elevation of total lipid oxidation after exercise. It has been suggested that noradrenaline and growth hormone are involved in the elevated postexercise lipolysis in men. In addition, the increased NEFA use in recovery may have been related to depletion of CHO removing energy substrate competition between oxidative glycolysis and β-oxidation (17).
Considering exercise plus recovery, these results show that the total amount of lipid oxidized and the contribution of lipid to energy expenditure were approximately 2.5 times higher in C45% than in INT due to the higher lipid oxidation in C45% during exercise. The mean relative intensity of C45% was close to the level at which lipid oxidation reached a maximum in healthy subjects (i.e., Fatmax) and may represent a crucial intensity to recommend exercise training to promote lipid oxidation in healthy and obese subjects. Furthermore, it has been shown that at this intensity, 12-wk low-intensity endurance training (three training sessions per week: approximately 50 min at 40% V˙O2max on a cycle ergometer) increases the ability to oxidize lipids during both rest and exercise in healthy male subjects (32). However, although INT does not acutely generate a greater postexercise lipid oxidation compared with C45%, the chronic influence of such an approach may lead to a different outcome. Indeed, Tremblay et al. (36) have shown that the increase in lipid oxidation is greater after a 15-wk high-intensity interval training program than after a 20-wk moderate-intensity continuous training program. The metabolic adaptations taking place in the skeletal muscle in response to the high-intensity interval training program appear to favor the process of lipid oxidation. Moreover, it has recently been shown that only seven sessions of high-intensity submaximal interval training (4 min at 90% V˙O2max with 2-min rest between intervals) over 2 wk induced an increase in whole body lipid oxidation during exercise in moderately active women (34). These findings reveal the potency of high-intensity interval exercise associated with short duration sessions and training periods (i.e., 3 h·wk−1 for 2 wk) to improve the skeletal muscle capacity to oxidize lipids, which has implications for improving fitness and health (34). Obviously, INT requires a high level of motivation and causes a feeling of severe fatigue resulting in poor adherence, soreness, and significant risk of injury for individuals with health problems or for obese subjects who are not accustomed to exercise. Especially in these specific populations, low-intensity continuous training should be performed during the first training period to develop the endurance necessary to support high-intensity interval exercise sessions. Subsequently, INT associated with short duration sessions and low volume may be added to a training program to further improve metabolic effects. Thus, high-intensity interval exercise may be an important complementary training tool to traditional low-intensity continuous exercise and could make an important contribution to weight management in individuals at risk for obesity and in obese subjects (31).
Some methodological limitations exist and need to be addressed. First, participants were healthy, active, and normal-weight individuals, so the response observed may not be representative of very physically fit, sedentary, overweight, or obese subjects. Furthermore, the precise sources of lipids (i.e., NEFA and intramuscular or plasma TAG) and CHO (i.e., muscle glycogen, blood glucose, blood, or muscle lactate) oxidized during exercise and recovery could not be determined or differentiated.
In conclusion, this study shows that lipid oxidation during postexercise recovery was increased by similar amount on high-intensity interval submaximal exercise and moderate-intensity continuous exercise compared with the no-exercise control trial. Despite a significant but slightly greater CHO oxidation that likely leads to larger muscle glycogen depletion during interval exercise, the two isoenergetic exercises of different forms and intensities promote similar shifts in the pattern of substrate use toward lipids during postexercise recovery, when lipid oxidation predominates. Thus, postexercise lipid oxidation may be more dependent upon the energy expenditure of exercise than on the intensity of the prior isoenergetic exercise. Because the data reported are of potential interest for exercise prescription in obese individuals, future research should compare the effect of these two forms of exercise between fit young and obese subjects.
This work was supported by Office Fédéral du Sport Macolin (Switzerland). The results of the present study do not constitute endorsement by ACSM.
1. Astrand PO, Rodahl K. Textbook of Work Physiology
. New York: McGraw-Hill Book Company; 1986. p 704.
2. Bahr R. Excess postexercise oxygen consumption
-magnitude, mechanisms and practical implications. Acta Physiol Scand Suppl
3. Bahr R, Gronnerod O, Sejersted OM. Effect of supramaximal exercise on excess postexercise O2 consumption. Med Sci Sports Exerc
4. Barstow TJ, Cooper DM, Sobel EM, Landaw EM, Epstein S. Influence of increased metabolic rate on [13C]bicarbonate washout kinetics. Am J Physiol
. 1990;259(1 Pt 2):R163-71.
5. Beaver WL, Wasserman K, Whipp BJ. A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol
6. Blaak EE, Van Baak MA, Kemerink GJ, Pakbiers MT, Heidendal GA, Saris WH. Beta-adrenergic stimulation of energy expenditure
and forearm skeletal muscle metabolism in lean and obese men. Am J Physiol
. 1994;267(2 Pt 1):E306-15.
7. Borsheim E, Bahr R. Effect of exercise intensity, duration and mode on post-exercise oxygen consumption. Sports Med
8. Bouchard C, Depres JP, Tremblay A. Exercise and obesity. Obes Res
9. Brockman L, Berg K, Latin R. Oxygen uptake during recovery from intense intermittent running and prolonged walking. J Sports Med Phys Fitness
10. Carter J, Jeukendrup AE. Validity and reliability of three commercially available breath-by-breath respiratory systems. Eur J Appl Physiol
11. Christmass MA, Dawson B, Passeretto P, Arthur PG. A comparison of skeletal muscle oxygenation and fuel use in sustained continuous and intermittent exercise. Eur J Appl Physiol Occup Physiol
12. Durnin JV, Womersley J. Total body fat, calculated from body density, and its relationship to skinfold thickness in 571 people aged 12-72 years. Proc Nutr Soc
13. Essen B, Hagenfeldt L, Kaijser L. Utilization of blood-borne and intramuscular substrates during continuous and intermittent exercise in man. J Physiol
14. Ferrannini E. The theoretical bases of indirect calorimetry
: a review. Metabolism
15. Ferraro RT, Eckel RH, Larson DE, et al. Relationship between skeletal muscle lipoprotein lipase activity and 24-hour macronutrient oxidation. J Clin Invest
16. Frayn KN. Calculation of substrate oxidation rates in vivo from gaseous exchange. J Appl Physiol
17. Henderson GC, Fattor JA, Horning MA, et al. Lipolysis and fatty acid metabolism in men and women during the postexercise recovery period. J Physiol
. 2007;584(Pt 3):963-81.
18. Kelley DE, Goodpaster B, Wing RR, Simoneau JA. Skeletal muscle fatty acid metabolism in association with insulin resistance, obesity, and weight loss. Am J Physiol
. 1999;277(6 Pt 1):E1130-41.
19. Kiens B, Richter EA. Utilization of skeletal muscle triacylglycerol during postexercise recovery in humans. Am J Physiol
. 1998;275(2 Pt 1):E332-7.
20. Kim JY, Hickner RC, Cortright RL, Dohm GL, Houmard JA. Lipid oxidation is reduced in obese human skeletal muscle. Am J Physiol Endocrinol Metab
21. Kimber NE, Heigenhauser GJ, Spriet LL, Dyck DJ. Skeletal muscle fat and carbohydrate metabolism during recovery from glycogen-depleting exercise in humans. J Physiol
. 2003;548(Pt 3):919-27.
22. Kuo CC, Fattor JA, Henderson GC, Brooks GA. Lipid oxidation in fit young adults during postexercise recovery. J Appl Physiol
23. Laforgia J, Withers RT, Shipp NJ, Gore CJ. Comparison of energy expenditure
elevations after submaximal and supramaximal running. J Appl Physiol
24. McGarvey W, Jones R, Petersen S. Excess post-exercise oxygen consumption following continuous and interval cycling exercise. Int J Sport Nutr Exerc Metab
25. Melanson EL, Sharp TA, Seagle HM, et al. Effect of exercise intensity on 24-h energy expenditure
and nutrient oxidation. J Appl Physiol
26. Mulla NA, Simonsen L, Bulow J. Post-exercise adipose tissue and skeletal muscle lipid metabolism in humans: the effects of exercise intensity. J Physiol
. 2000;524(Pt 3):919-28.
27. Phelain JF, Reinke E, Harris MA, Melby CL. Postexercise energy expenditure
and substrate oxidation in young women resulting from exercise bouts of different intensity. J Am Coll Nutr
28. Romijn JA, Coyle EF, Hibbert J, Wolfe RR. Comparison of indirect calorimetry
and a new breath 13C/12C ratio method during strenuous exercise. Am J Physiol
. 1992;263(1 Pt 1):E64-71.
29. Romijn JA, Coyle EF, Sidossis LS, et al. Regulation of endogenous fat and carbohydrate metabolism in relation to exercise intensity and duration. Am J Physiol
. 1993;265(3 Pt 1):E380-91.
30. Sahlin K, Alvestrand A, Brandt R, Hultman E. Intracellular pH and bicarbonate concentration in human muscle during recovery from exercise. J Appl Physiol
31. Saris WH, Schrauwen P. Substrate oxidation differences between high- and low-intensity exercise are compensated over 24 hours in obese men. Int J Obes Relat Metab Disord
32. Schrauwen P, van Aggel-Leijssen DP, Hul G, et al. The effect of a 3-month low-intensity endurance training program on fat oxidation and acetyl-CoA carboxylase-2 expression. Diabetes
33. Stringer W, Casaburi R, Wasserman K. Acid-base regulation during exercise and recovery in humans. J Appl Physiol
34. Talanian JL, Galloway SD, Heigenhauser GJ, Bonen A, Spriet LL. Two weeks of high-intensity aerobic interval training increases the capacity for fat oxidation during exercise in women. J Appl Physiol
35. Thyfault JP, Kraus RM, Hickner RC, Howell AW, Wolfe RR, Dohm GL. Impaired plasma fatty acid oxidation in extremely obese women. Am J Physiol Endocrinol Metab
36. Tremblay A, Simoneau JA, Bouchard C. Impact of exercise intensity on body fatness and skeletal muscle metabolism. Metabolism
37. van Aggel-Leijssen DP, Saris WH, Wagenmakers AJ, Senden JM, van Baak MA. Effect of exercise training at different intensities on fat metabolism of obese men. J Appl Physiol
38. Vukovich MD, Costill DL, Hickey MS, Trappe SW, Cole KJ, Fink WJ. Effect of fat emulsion infusion and fat feeding on muscle glycogen utilization during cycle exercise. J Appl Physiol
39. Wasserman K. The anaerobic threshold: definition, physiological significance and identification. Adv Cardiol
40. Wasserman K, Hansen JE, Sue AY, Whipp BJ, Casaburi R. Principles of Exercise Testing and Interpretation
. 2nd ed. Philadelphia: Lippincott Williams & Wilkins; 1994. p. 479.