Secondary Logo

Journal Logo

Acute High-Intensity Interval Cycling Improves Postprandial Lipid Metabolism


Medicine & Science in Sports & Exercise: August 2018 - Volume 50 - Issue 8 - p 1687–1696
doi: 10.1249/MSS.0000000000001613

Purpose This study aimed to examine the effects of two exercise regimes on physiological and postprandial lipemic responses.

Methods Thirty-six active men (peak oxygen uptake [V˙O2peak], 46.5 ± 6.4 mL·kg−1·min−1) were randomly assigned to a high-intensity interval exercise (HIIE), involving 10 × 60 s cycling at 85% V˙O2peak interspersed with 120 s recovery; a moderate-intensity continuous exercise (MICE), involving 50 min continuous exercise at 65% V˙O2peak; and a nonexercise control (Con). In the next morning after evening exercising, fasting blood samples were obtained. Additional blood samples were obtained 1–4 h after eating a given high-fat meal that based on participants’ body mass. Carbohydrate and fat oxidation rates were measured before and after the meal.

Results After exercise, glucose and insulin concentrations decreased by 33% and 70% in MICE compared with those in HIIE (P = 0.00–0.03). During the 1- to 2-h postprandial periods, the fat oxidation rate increased by 24%–37% in HIIE that that in MICE and Con (P = 0.01–0.03); however, the carbohydrate oxidation rate was not significantly different among the conditions (P = 0.28). During the postprandial period, insulin (P = 0.02–0.04) and triglyceride (P = 0.02–0.03) concentrations were lower in HIIE than those in MICE and Con. No difference was observed in free fatty acid or the total areas under the curve of triglyceride and free fatty acid among the conditions (P = 0.24–0.98).

Conclusion Acute MICE improved glucose and insulin metabolism immediately after exercise. However, HIIE performed in the evening exerts more favorable effects than MICE for decreasing postprandial insulin and triglyceride levels and increasing fat oxidation in the next morning.

1Division of Physical and Health Education, Center for General Education, National Sun Yat-Sen University, Kaohsiung, TAIWAN;

2Department of Physical Education, National Taiwan Normal University, Taipei, TAIWAN; and

3Department of Athletic Performance, National Taiwan Normal University, Taipei, TAIWAN

Address for correspondence: Chia-Lun Lee, Ph.D., Division of Physical and Health Education, Center for General Education, National Sun Yat-sen University, No. 70, Lienhai Rd., Kaohsiung City, Taiwan, ROC; E-mail:

Submitted for publication October 2017.

Accepted for publication March 2018.

Recent research has shown that postprandial lipemia (PPL) is associated with increased risks of atherosclerosis (1) and cardiovascular disease (CVD) (2) in adults. In an early study, Zilversmit (1) indicated that PPL along with cholesterol (Chol), LDL cholesterol (LDL-C), and very LDL cholesterol (VLDL) contributes to the positive association between hypertriglyceridemia and coronary artery disease and the inverse relationship between HDL cholesterol (HDL-C) and CVD. Research demonstrated that the magnitude of PPL or single postprandial triglyceride values can predict asymptomatic and symptomatic atherosclerosis, independent of risk factors measured in the fasting state (3). The fasting PPL and postprandial triglyceride responses also correlated strongly with physical activity (4) and dietary habit (5). Occasionally, a hectic day and a busy working life may lead people to miss one meal a day (e.g., breakfast or dinner); the whole volume of the missed meal is ingested with other meals on the same day. Hence, this type of dietary behavior may be conducive to the ingestion of high-fat food, which disturbs the natural lipid metabolism (6). The ingestion of a high-fat meal (HFM) acutely changes the blood lipid profile and reduces endothelial function for many hours after the meal. Thus, a significant proportion of life is spent in the postprandial state; the factors leading to this transient impairment in endothelial function may play a key role in the development of atherosclerosis (7).

PPL and aerobic exercise have been extensively demonstrated to have antagonistic effects regarding cardiovascular risk; aerobic exercise may decrease PPL, decreasing triglyceride secretion by the liver and increasing plasma triglyceride clearance by the muscle (6). However, the effect of more intense exercise such as high-intensity interval exercise (HIIE) (i.e., 85%–95% peak heart rate or 80%–100% peak work rate) on PPL is not well understood. Sprint interval exercise or HIIE has been proposed as a viable alternative to moderate-intensity continuous exercise (MICE) because it is a time-efficient exercise strategy that elicits improvements in the peak oxygen uptake (V˙O2peak) (8) and muscle insulin sensitivity (9) in healthy adults. HIIE is defined as near maximal efforts or submaximal interval exercise performed at an intensity of 85%–95% of the maximal heart rate (HRmax); by contrast, sprint interval exercise is characterized by efforts performed at intensities equal to or greater than the intensity that elicits V˙O2peak, including all-out or supramaximal efforts (10). An early study found that an HIIE protocol involving 4 × 4 min activity at 85%–95% HRmax or a matched work continuous exercise protocol at 60%–70% HRmax did not affect PPL triglyceride and HDL-C levels on the next day (7). A 20-min bout of 4 × 30 s sprint cycling also had no effect on PPL (11). By contrast, a 20-min bout of 60 × 8 s sprint cycling against a fixed resistance at 60%–65% V˙O2max significantly decreased postprandial triglyceride (12). Moreover, a study compared 3-min sessions at 115% anaerobic threshold with 1.5 min of recovery versus a matched energy expenditure (EE) continuous exercise intervention performed until an EE of 500 kcal was reached; compared with moderate continuous exercise, the more intense exercise intervention reduced PPL 2–4 h after HFM (13). In addition, the HIIE intervention has been reported to improve triglyceride concentration or triglyceride area under the curve (AUC) (14–18), postprandial fat oxidation (19), blood insulin sensitivity (16), and fat oxidation after 6-wk sprint interval training (15). However, some studies have compared the effects of single sessions of HIIE at different intensities and for different durations on the PPL response; these studies have found no significant effects on triglyceride AUC (11,20–22) or that the HIIE results in more decreases in PPL in girls than in boys (18).

A previous study found a higher reduction in postprandial triglyceride in participants in the energy deficit condition during a shortened postprandial protocol than during a longer postprandial protocol (16). Thus, EE and energy intake (23) have been suggested to be important variables determining exercise-induced reductions in triglyceride during submaximal interval exercise. Moreover, the time from exercise cessation to test meal consumption may play a role in exercise-induced changes in PPL (24). Studies have provided contradictory results regarding the effect of prior exercise on PPL variations in the HIIE protocol. These contradictory results may be attributed to interindividual variation, inconsistent energy intake after exercise, or insufficient time postexercise for inducing an increase in lipoprotein lipase (LPL) activity. Manipulating energy intake through standardization at pre- and postexercise intervals may reduce the variations in energy deficit and prolong the time for observing lipid metabolism. Research has demonstrated that different intensities of the HIIE protocol may not elicit similar effects on postprandial lipoprotein profile (11,12,14,17,20,22). Studies focused on the acute effects of different exercise modes in PPL after HFM are important for developing a well-rounded exercise prescription to reduce the risk of CVD in healthy individuals, as well as in the sedentary population. Therefore, this study investigated the effect of acute HIIE versus MICE on PPL at time points ranging from immediately after exercise to the next day. We hypothesized that HIIE can decrease PPL and produce a beneficial lipid profile compared with MICE in healthy individuals.

Back to Top | Article Outline


Participants and physical activity

A total of 36 healthy and active men participated in this study. Participants were 25 ± 5 yr old. The average height was 175 ± 6 cm, the average body mass was 69 ± 10 kg, and the average percentage body fat was 12% ± 4%. For this study, participants were considered to perform “regular exercise” if they participated in approximately 60-min sessions of a structured exercise program that combined cardiovascular and resistance exercise, at least three times per week for the past 3 yr. Participants were recreationally active but were not formally trained for any type of professional competition. The exclusion criteria were as follows: smoking, anabolic steroid intake, and a history of CVD, diabetes, hypertension, or any other metabolic disease or illness requiring the ingestion of medications that affect carbohydrate or lipid metabolism. All participants were fully informed of the aims, risks, and discomfort associated with the investigation before providing written informed consent. This study was approved by the Hospital institutional review board.

Back to Top | Article Outline

Initial assessment

To determine participants’ eligibility, 5–7 d before their first familiarization session, fasting blood samples (baseline) were collected for glucose, HDL-C, LDL-C, VLDL, total Chol, and triglyceride measurement to exclude participants presenting abnormal glucose, blood lipids, or frank diabetes. Subsequently, participants’ skinfold thickness was measured twice at three sites, namely, chest, abdomen, and thigh, in rotational order (25). The measurements were conducted to the nearest 0.1 mm on the right side of the body by using Lange calipers (Cambridge Scientific Industries, Inc., Cambridge, MD) and were used to determine body composition. If the difference in measured values at each site exceeded 1 mm, a third assessment was performed by an experienced researcher certified as a health fitness instructor. These measured values were used to calculate body density (25), which was then used to estimate the percentage body fat [% body fat = (495 per density) − 450].

Back to Top | Article Outline

Experimental protocol

The randomized control design was used in this present study. Thirty-six participants were randomly assigned to one of three experimental conditions, including HIIE, MICE, and nonexercise control (Con). At 7 d before the start of the study, each participant reported to the laboratory for a familiarization session (i.e., HIIE and MICE). They also practiced a V˙O2peak test, became accustomed to the gas collection equipment (e.g., gas mask), and determined their preferred seat height on the ergometer. The three experimental conditions were as follows: Con (n = 12), moderate-intensity exercise (MICE, n = 12), and HIIE (n = 12). Each trial involved 2 d of diet and physical activity controls, during which participants consumed a laboratory-provided diet and performed no exercise outside the laboratory. Participants refrained from physical activity and alcohol ingestion for 48 h before each trial and did not consume any caffeine-containing beverage for 24 h before each trial. Participants also recorded their diet 2 d before the V˙O2peak test, and this was replicated during the subsequent visit in exercise test.

On day 1, participants arrived at the laboratory at 3:30 PM and rested for 10 min before their blood was collected from a vein in the antecubital fossa. Subsequently, they consumed a standardized meal for dinner at 4:00 PM. For all participants, meals were consumed at a similar time on the test day. Participants rested for 3 h after consuming a standardized meal. At 7:00 PM, they performed 5 min warm-up and subsequently completed one of the three exercise protocols, namely, HIIE, MICE, or Con intervention. In the morning of day 2, participants reported to the laboratory after a 12-h overnight fast. Upon arrival at the laboratory, fasting blood samples were collected for measurements of plasma glucose, triglyceride, insulin, HDL-C, LDL-C, VLDL, total Chol, and free fatty acid (FFA). Immediately after blood collection, resting EE, carbohydrate oxidation, and fat oxidation were determined in the sitting position. After completion of the fasting measurement, participants consumed a test meal. Blood collection was repeated hourly for 4 h postprandially, and resting EE, carbohydrate oxidation, and fat oxidation measurements were repeated at 1, 2, 3, and 4 h postprandially. The protocol schematic is shown in Figure 1.



Back to Top | Article Outline

Determination of V˙O2peak

One week before the first trial, we determined the V˙O2peak of participants performing a bicycle ramp protocol on an ergometric appliance (Cyclus 2; RBM Elektronik-Automation GmbH, Leipzig, Germany). After a 5-min warm-up period, each participant performed a V˙O2peak test. Briefly, the workload was set at 50 W initially and was then increased by 30 W every 1 min with the aim of reaching each participant’s V˙O2peak. During the test, pulmonary gas exchange and heart rate data were collected using the Cortex Metamax 3B portable metabolic test system (Cortex Biophysik, Leipzig, Germany) and wireless telemetry system (Polar Electro, Kempele, Finland), respectively. Participants were verbally encouraged to continue the maximal cycling test until volitional fatigue. During the maximal protocol, V˙O2peak was determined within the final 30 s before exhaustion. The highest values of oxygen uptake within the final 30 s of the test were used for statistical analysis. The criteria for V˙O2peak were as follows: (a) age-predicted maximum heart rate (208 − 0.7 × age), (b) plateau in oxygen uptake that increased by less than 150 mL·min−1 despite an increase in power output, (c) RER greater than 1.10, and (d) maximal capillary blood lactate concentration greater than 6 mmol·L−1 after exercise. For all participants, at least three of the four criteria were met. The coefficient of variation (CV) for V˙O2peak for active populations in our laboratory was 2.6%. The O2 consumption (V˙O2) data collected at each workload during the V˙O2peak test were further analyzed using a simple linear regression to determine the exercise intensities (i.e., 30%, 65%, and 85% V˙O2peak) of the MICE and HIIE regimes. To verify metabolic commitment, the peak blood lactate concentration was measured 5 min after cessation of the exercise trial. Throughout the testing, oxygen uptake was measured using an open-circuit breath-by-breath Spirograph, and standard algorithms were used to dynamically account for the time delay between gas consumption and volume signal. The Spirograph was calibrated before each test by using calibration gas (15% O2, 5% CO2; Praxair, Dusseldorf, Germany) targeting the range of anticipated fractional gas concentration; a precision 1-L syringe (nSpire, Oberthulba, Germany) was used to inject the calibration gas. The regression equation of oxygen consumption versus the work rate from the V˙O2peak test and the determined V˙O2peak value was used to set the work rates for HIIE and MICE, which corresponded to 85% and 65% V˙O2peak, respectively. Moreover, V˙O2 and CO2 production (V˙CO2) measurements from the V˙O2peak test were used to determine the target EE for the exercise trial based on the energy cost of cycling at high and moderate intensities.

Back to Top | Article Outline

Exercise protocols and EE

The HIIE protocol involved repeated 10 × 60 s sprint cycling with pedaling as fast as possible against a fixed resistance at 85% V˙O2peak, interspersed with a 120-s active recovery at 30% V˙O2peak. The MICE protocol involved 50 min of continuous endurance exercise at 65% V˙O2peak, with a pedaling cadence of 60 or 70 rpm, as selected by the participant. The HIIE protocol was chosen because it was well tolerated by our participants and did not require longer intervals (2–4 min) or brief supramaximal sprints (10–30 s in duration) (26). The nonexercise Con remained in the laboratory during the entire period and expended little energy (i.e., they watched a movie, read a book or magazine, or worked on the computer).

Respiratory gases were collected before exercise and during cycling on day 1, in the fasting condition before breakfast on day 2, and at 1, 2, 3, and 4 h after breakfast. The parameters of V˙O2, V˙CO2, and RER were used to calculate whole-body fat and carbohydrate oxidation. These calculations were executed using the nonprotein RER table and the following equations (27): carbohydrate oxidation = 4.210 (V˙CO2) − 2.962 (V˙O2); fat oxidation = 1.695 (V˙O2) − 1.701 (V˙CO2). To determine resting EE during the HFM test on day 2, an indirect calorimeter was used to collect gas for 15 min after 10 min of rest in a sitting position.

Back to Top | Article Outline

Dietary recording and test meal

Participants were instructed to maintain their usual diet throughout the study period, and they recorded their food and drink intake and all physical activities within 48 h before each test day. This information was used to match their diet and activity patterns across the V˙O2peak test and the one experimental condition. The recorded food intake was analyzed by research staff using a nutritional analysis software package (Nutritional Chamberlain Line, Professional Edition, E-Kitchen Inc., Taiwan). The results were returned to participants so they could mimic this intake on all days before test days.

On day 1, participants consumed a standardized meal, which was devised depending on their body mass (daily macronutrients: 60% carbohydrate, 23% fat, and 17% protein), for dinner at 4:00 PM. The meal consisted of rice, chicken, vegetables, eggs, and beans. Participants rested for 3 h after dinner at 7:00 PM. Subsequently, they performed the acute HIIE, MICE, or Con intervention. To avoid the ingestion of extra food, they ate a small nutrition bar (25 g of carbohydrate, 14 g of fat, and 7 g of protein) at 08:30 PM. Before leaving the laboratory on day 1, participants in each experimental condition were reminded that they could drink plain water but should not consume any food.

For breakfast on day 2, the test meal consisted of bread, cheese, peanut butter, margarine, and fruit juice. The calorie composition size (kcal) of the HFM was determined based on participants’ body mass. The calorie composition of the meal was equal to 1.1 g of fat, 1.0 g of carbohydrate, and 0.3 g of protein per kilogram of body mass, totaling approximately 13 kcal·kg−1 of body mass, similar to the meal used in the study by Freese et al. (16). On the basis of the possible mechanisms responsible for the reduction in postprandial lipid metabolism, the breakfast meal was ingested approximately 12 h after exercise overnight. During the interval between the breakfast meal ingested in the morning and lunch 4 h later, no other meals or calorie-containing beverages were allowed. Water was allowed in limited amounts.

Back to Top | Article Outline

Blood collection and analyses

Blood samples for analyses were collected from the antecubital vein before the exercise or control (i.e., no exercise) and immediately after exercise on day 1. Additional blood samples were collected before meal ingestion at 8:00 AM (i.e., after approximately 12 h overnight fasting) and at 1, 2, 3, and 4 h after meal ingestion. A 5-mL blood sample was collected in a tube containing a serum clot activator (Vacuette®, St. Gallen, Switzerland) and was centrifuged at 3000 rpm for 15 min at 4°C. The sample was placed on ice during experiments and was stored frozen at −80°C. The collected serum was subsequently analyzed for measuring insulin, triglyceride, total Chol, HDL-C, LDL-C, VLDL, and FFA levels; the interassay CV values were 2.2%, 1.5%, 1.8%, 2.1%, 2.2%, 5.0%, and 2.1%, respectively. Serum insulin levels were determined using a commercial kit (DIAsource INS-IRMA kit, Nivelles, Belgium). Lipid parameters (triglyceride, total Chol, HDL-C, LDL-C, and VLDL) were measured by photometric method (UnicelDxC 800, Beckman Coulter autoanalyzer). The serum FFA levels were determined using a nonesterified fatty acid kit (Randox Laboratories Canada Ltd., Ontario, Canada). All assays were performed in duplicate on the first thaw. However, lactate and glucose assays were performed once in a single analysis using fresh whole blood. Blood lactate concentrations were measured immediately using an automated analyzer (Lactate Pro TM LT-1730; Arkray KDK Corp., Kyoto, Japan), and blood glucose concentrations were measured using a portable device (Breeze2; Bayer, Munich, Germany); the interassay CV values were 1.2% and 1.1%, respectively. Using the trapezoidal method, postprandial responses for glucose, triglyceride, FFA, and insulin were measured by summing the 17-h AUC for serum/plasma concentration versus time (15). With n + 1 measurements yi at time ti (i = 0, 1, 13, 14, 15, 16, and 17 h), the total AUC (TAUC, mg·dL−1·h−1; μU·mL−1·h−1; mmol·L−1·h−1) was calculated as follows: 1 × [(y0 + y1)/2] + 12 × [(y1 + y2)/2] + 1 × [(y2 + y3)/2] + 1 × [(y3 + y4)/2] + 1 × [(y4 + y5)/2] + 1 × [(y5 + y6)/2].

Back to Top | Article Outline

Statistical analysis

All data are presented as means ± SD. The Shapiro–Wilk normality test was applied to determine the homogeneity of all data. Preexercise, postexercise, fasting, and postprandial plasma glucose, serum insulin, triglyceride, Chol, FFA, HDL-C, LDL-C, and VLDL levels were assessed using two-way repeated-measures ANOVA for condition and condition–time interactions. Fasting and postprandial V˙O2, V˙CO2, EE, carbohydrate oxidation, and fat oxidation were also analyzed using two-way repeated-measures ANOVA for condition–time interactions. The Bonferroni post hoc correction was applied if a significant difference was found. One-way ANOVA was used to compare total EE and AUC (calculated using the trapezoidal method) among the conditions. Statistical significance was set at P ≤ 0.05. The data were analyzed using SPSS for Windows, Version 17.0 (SPSS, Chicago, IL). Informed by a previous study detailing the changes in PPL after intense interval cycling intervention in active healthy adults (16), we hypothesized that triglyceride values would be more influenced by different exercise regimes, and therefore, we expected to see a 9% decrease on triglyceride in exercise regimes relatively to control condition. Thus, our study had 80% power at a 0.05 significance level to detect a 9% decrease in triglyceride in the exercise conditions. A sample size of 12 participants per condition was required to detect the differences on the changes in PPL profiles, including glucose, Chol, insulin, HDL-C, LDL-C, triglyceride, and FFA, between experimental and control conditions. Standardized effect sizes were also calculated using Cohen equations (28) with the following threshold values: <0.2 (trivial); >0.2 and <0.6 (small); >0.6 and <1.2 (moderate); >1.2 and <2.0 (large); >2.0 and <4.0 (very large); and <4.0 nearly perfect (29).

Back to Top | Article Outline


Physiological measurements and dietary intake

Participants’ characteristics are presented in Table 1. Participants in the three conditions had a normal range of blood pressure, body mass index, body fat, and blood lipid profile at the baseline. V˙O2peak values were not significantly different among HIIE, MICE, and Con (48.2 ± 6.3, 45.8 ± 5.7, and 45.5 ± 7.2 mL·kg−1·min−1, respectively; P = 0.2, d = 0.3). During the exhaustion phase of the V˙O2peak test, the ratings of perceived exertion were not significantly different among HIIE, MICE, and Con (19 ± 1, 19 ± 1, and 18 ± 1, respectively; P = 0.5, d = 0.4). Moreover, the peak heart rates were not significantly different among HIIE, MICE, and Con (190 ± 9, 190 ± 16, and 189 ± 10 bpm, respectively; P = 0.9, d = 0.1). Blood lactate concentrations after the V˙O2peak test were not significantly different among HIIE, MICE, and Con (8.7 ± 2.4, 9.8 ± 2.8, and 9.3 ± 3.6 mmol·L−1, respectively; P = 0.2, d = 0.1).



During the exercise intervention on day 1, the average heart rate values were significantly different among HIIE, MICE, and Con (134 ± 11, 138 ± 12, and 68 ± 8 bpm, respectively; P = 0.00, d = 6.2). During the HFM intervention on day 2, no significant condition–time interaction effect (P = 0.48, d = 0.3) and main effect of condition (P = 0.16, d = 0.4) were found for average heart rates from premeal to postmeal. However, a main effect of time (P = 0.00, d = 1.3) showed that average heart rates during the postprandial 1–2 h had higher values compared with premeal, postprandial 3–4 h (premeal: 62 bpm; postprandial 1 h: 70 bpm; postprandial 2 h: 68 bpm; postprandial 3 h: 65 bpm; postprandial 4 h: 62 bpm, respectively).

During the 48 h before day 2, the estimated average energy intake levels were similar among HIIE, MICE, and Con (10.6 ± 1.2, 10.8 ± 1.3, and 11.0 ± 1.6 MJ·d−1, respectively). Moreover, during the aforementioned period, the macronutrient intake levels were similar among HIIE, MICE, and Con (protein: 108 ± 22 vs 110 ± 23 vs 112 ± 25 g·d−1; carbohydrate: 380 ± 42 vs 387 ± 45 vs 394 ± 53 g·d−1; fat: 65 ± 15.4 vs 66 ± 14.3 vs 67 ± 15.8 g·d−1).

Back to Top | Article Outline

Metabolic rate and substrate utilization

The average premeal and postmeal values are reported in Table 2. Significant condition–time interactions were found for V˙O2 (P = 0.01, d = 0.6). The post hoc test indicated that the V˙O2 value of HIIE was significantly higher than that of Con (P = 0.05). However, this value was not statistically different from that of MICE (P = 0.2). No significant condition–time interaction (P = 0.33, d = 0.3) or main effect of condition (P = 0.79, d = 0.1) was found for V˙CO2. During the postprandial period, no significant condition–time interaction (P = 0.07, d = −0.6) was found for RER; however, a main effect of condition was found for RER (P = 0.02, d = −0.8). The post hoc test indicated that at the 1-h postprandial time point, RER was significantly lower in HIIE than that in Con (P = 0.00) and MICE (P = 0.01). A main effect of time was found for all metabolic rates and substrate partitioning values (P = 0.00).

A significant main effect of condition was found for exercise EE per minute among HIIE, MICE, and Con (11.0 ± 2.1 vs 12.2 ± 1.7 vs 1.45 ± 0.21 kcal·min−1; P = 0.00, d = 6.4). The post hoc test showed that EE values in both HIIE and MICE were significantly higher than those in Con (P = 0.00), but HIIE was not different from MICE (P = 0.22). However, total EE for the exercise session was significantly different between HIIE and MICE (330 ± 64 vs 610 ± 85 kcal; P = 0.00, d = 8.6). During the HFM test on day 2, significant condition–time interaction (P = 0.03, d = 0.5) was found for EE; however, post hoc showed that no significant difference (P = 0.1–0.95) before an HFM meal and during the postprandial period among HIIE, MICE, and Con (Table 2).



The results show that there was a significant condition–time interaction for fat oxidation rate (P = 0.02, d = 0.64). The post hoc test showed that at the 1-h postprandial time point, the fat oxidation rate was significantly higher in HIIE than that in MICE and Con (P = 0.00). Moreover, at the 2-h postprandial time point, the fat oxidation rate was significantly higher in HIIE than that in Con (P = 0.02). No significant difference was observed between MICE and Con (P = 1.00). On the other hand, no significant condition–time interaction (P = 0.44, d = −0.3) or main effect of condition (P = 0.3, d = −0.23) was found for the carbohydrate oxidation rate. However, a main effect of time (P = 0.00, d = −1.85) was found for HIIE and MICE.

Back to Top | Article Outline

Changes in blood parameters in response to exercise

Significant differences were observed in the condition–time interactions for blood glucose (P = 0.00, d = 0.15), Chol (P = 0.00, d = 1.92), insulin (P = 0.00, d = 1.26), HDL-C (P = 0.00, d = 0.08), and LDL-C (P = 0.00, d = 1.19), as illustrated in Figures 2A, 2B, 2D, 2E, and 2F, respectively. However, according to post hoc test results, only HDL-C did not significantly differ among the conditions. After exercise, the glucose level was markedly lower in MICE than that in Con (17%, P = 0.02) and HIIE (33%, P = 0.03). After exercise, the Chol level was higher in HIIE than that in Con (19%, P = 0.01), and the Chol level tends to increase in MICE compared with Con (13%, P = 0.06). After exercise, the insulin level in MICE was different from that in Con (65%, P = 0.00) and HIIE (70%, P = 0.00). However, on day 2, the insulin level in HIIE was lower than that in Con before the meal (30%, P = 0.03) and that in MICE at the 3-h postprandial time point (43%, P = 0.02).



A significant condition–time interaction effect (P = 0.02, d = −0.4) was found for the triglyceride level. The post hoc test showed that at 4 h after HFM, the triglyceride level in HIIE decreased to approximately 53% and 46% compared with those in MICE (P = 0.02) and Con (P = 0.03), respectively. Moreover, the percentage changes in triglyceride levels from premeal to 4 h postmeal did not differ significantly among the conditions (HIIE vs MICE vs Con; 110% ± 76% vs 213% ± 261% vs 216% ± 210%, P = 0.34, d = −0.27).

No significant interaction effect (P = 0.54, d = 0.08) or main effect of condition (P = 0.74) was found for FFA. However, significant differences were observed in the main effect of time (P = 0.00) for FFA. This finding indicated that after HFM intake, FFA progressively increased during the postprandial period. The percentage change from premeal to the 4-h postprandial time point revealed a significant difference in FFA among the conditions (HIIE vs MICE vs Con; 30% ± 50% vs 29% ± 68% vs 115% ± 142%, P = 0.05, d = −0.89). The post hoc test showed a trend to reach significant difference in HIIE and MICE compared with Con (P = 0.08).

Regarding the lipemic response, glucose TAUC (P = 0.00, d = −0.27) and insulin TAUC (P = 0.00, d = −0.9) were significantly different among the conditions. The post hoc test showed that glucose TAUC and insulin TAUC were significantly lower in MICE than those in HIIE (P = 0.00) and Con (P = 0.00). However, no significant difference was observed in triglyceride TAUC (P = 0.98, d = −0.01) or FFA TAUC (P = 0.24, d = 0.09) among HIIE, MICE, and Con (Table 3).



Back to Top | Article Outline


The main purpose of this investigation was to determine the effectiveness of a single session of high-intensity interval cycling and a single session of continuous moderate-intensity cycling for blood lipemia profiles during rest and postprandial periods. The main finding of this study is that blood glucose and insulin concentrations were higher in HIIE than those in MICE immediately after exercise. LDL-C and Chol concentrations in HIIE increased compared with those in Con and MICE. However, after HFM intake, insulin and triglyceride concentrations in HIIE decreased compared with those in MICE or Con. Triglyceride TAUC and FFA TAUC did not change in response to the HIIE and MICE interventions. Moreover, compared with MICE and Con, fat oxidation rate in HIIE significantly increased to 27%–43% after HFM intake.

Previous studies have reported that increased PPL is associated with an increased risk of CVD (2,30). Thus, comparing the efficacy of different interventions in reducing postprandial triglyceride concentration and determining the variations in physiological parameters after the interventions are worthwhile. A previous study demonstrated that during the postprandial period after exercise, a high-intensity exercise intervention results in higher oxygen consumption and lipid oxidation than a matched low-intensity intervention with equivalent EE (31). However, in the present study, during the 1- to 2-h postprandial period, V˙ O2 and fat oxidation rate were higher and RER was lower in HIIE than those in MICE, although the two exercise regimes were not isocaloric. These parameters also differed between HIIE and Con. The EE of the HIIE intervention was significantly lower than that of the MICE intervention (279 ± 60 vs 513 ± 80 kcal, P = 0.00, d = 2.7). Hence, in addition to the effect of equal EE during exercise on PPL (13), there have been some indications that exercise intensity plays an important role in PPL responses (19,31). The present study used an exercise intensity of 65% V˙O2peak for 50 min as the protocol for the MICE intervention (MICE: 152 ± 26 W) and of 85% V˙O2peak for 20 min with 10 min interspersed rest as the protocol for the HIIE intervention (248 ± 42 W). The exercise intensity in the HIIE intervention differed from that in the MICE intervention. Highly intense exercise or lower-intensity exercise for longer duration can cause a substantial net reduction in muscle glycogen content; moreover, exercise-induced muscle glycogen depletion has been shown to activate glycogen synthase and increase muscle oxidative capacity (9). Our finding is consistent with that of a previous study (19), in which postprandial fat oxidation was higher in the isoenergetic high-intensity exercise intervention (a 2-min intense exercise at 90% V˙O2peak with 2-min rest for a total exercise time of 42 min) than that in moderate-intensity exercise intervention (50% V˙O2peak for 60 min). Nevertheless, the time of each exercise trial and the total exercise time in the previous study (19) were longer than ours. The protocols for the exercise regimes were established under the consideration of time-efficient exercise. Thus, during each training session, only approximately 10 min of exercise was performed over a 15- to 30-min period (9). The HIIE intervention increased postprandial V˙O2 and fat oxidation and reduced RER, although no change was observed in EE or the carbohydrate oxidation rate in active men performing this intervention. Thus, HIIE is more effective for increasing the fat oxidation during postprandial period than MICE. Given the greater improvement in postprandial RER and fat oxidation after HIIE in the current study, exercise intensity may be an important mediator of PPL responses, and high-intensity exercise offers a low-volume alternative to moderate-intensity exercise.

In the present study, after a single session of exercise without subsequent meal intake, blood glucose and insulin concentrations in MICE decreased compared with those in HIIE and Con (Figs. 2A and 2D). Moderate-intensity endurance exercise results in muscle contractions that improve insulin-mediated glucose uptake; moreover, exercise increases glucose transport capacity in muscle membrane but simultaneously increases glucose delivery through increased muscle blood flow, in addition to increased enzymatic activity related to glucose metabolism (32). Decreases in blood glucose and insulin concentrations immediately after exercise might be attributed to rapid increases in glucose transporter 4 expression, which in turn potentiates insulin-stimulated glucose transport capacity; thus, these decreases may provide a survival advantage by inducing the more rapid replenishment of muscle glycogen stores (33). Furthermore, regarding lipid metabolism, elevated LDL-C and Chol levels after exercise might be mediated by β-adrenergic stimulation. In the present study, we did not measure the epinephrine concentration in the exercise conditions. However, based on previous work, elevations in plasma epinephrine and norepinephrine concentrations immediately after endurance exercise or HIIE (34) may increase whole-body lipolysis. Therefore, increases in fat oxidation induced by a high-intensive cycling exercise might attribute to the activations of adrenalin and LPL (35).

The two interventions had similar effects on PPL (i.e., HDL-C, LDL-C, VLDL, Chol, and FFA), except for insulin and triglyceride. This discrepancy is likely because most parameters of PPL did not rapidly change in response to the acute HIIE and MICE interventions. The lack of changes in lipemic responses including FFA, Chol, HDL-C, LDL-C, and VLDL on day 2 during the postprandial period may partially explain the unaltered LPL activity after acute exercise (36). In general, the regulation of LPL may be complex because exercise (35), feeding, and fasting (37) have been individually shown to have antagonistic effects. Furthermore, although the exercise-induced regulation of muscle LPL expression may be pretranslational, LPL regulation in adipose tissue under different conditions has been shown to be transcriptional, translational, and posttranslational (35). In addition, exercise duration or postexercise period until HFM consumption maybe factors affecting LPL activity. A study confirmed a decrease in VLDL and an increase in FFA at 4.5 h after exercising for 90 min at 58% ± 5% V˙O2max (38). Another study directly compared endurance exercise (60%–65% V˙O2max with 60 min) with resistance exercise (3 sets and 10 repetitions of each exercise at 90%; a maximum time of approximately 60 min) and investigated their effects on postprandial FFA metabolism. Compared with the control trial, the exogenous plasma FFA content increased significantly after the ingestion of a liquid test meal after the endurance and resistance exercise; however, no significant difference in total plasma FFA was observed between endurance and control trials (39). The present results are consistent with those of previous studies, in which the total FFA concentration either immediately after exercise (34) or during the postprandial period (39,40) did not change between the experimental conditions; FFA TAUC also did not change.

Acute high-intensity interval cycling exercise decreased insulin and triglyceride concentrations 3 and 4 h after HFM, respectively. Nevertheless, these changes that temporarily appeared were lower in HIIE than those in Con and MICE. Compared with Con and MICE, the insulin concentration in HIIE decreased significantly by approximately 33% and 40%, respectively, at the 3-h postprandial time point. A previous study found improvements in insulin sensitivity at 14 d after the final bout of high-intensity cycle exercise on an ergometer (41), and the improvements were ascribed to chronic training adaptations (9). Insulin is known to play a pivotal role in triglyceride metabolism. It regulates triglyceride uptake in skeletal muscle and adipose tissue and VLDL release from the hepatic tissue. Insulin sensitivity is also an important biomarker of type 2 diabetes and metabolic syndrome, and it is a primary target for preventative intervention (42). In the present study, insulin sensitivity from the preexercise to the 4-h postprandial time points was calculated using the McAuley index (43) (data not shown). However, we found that insulin sensitivity was higher in HIIE than that in MICE, with a strong trend observed at the 4-h postprandial time point (2.30 ± 0.12 vs 2.12 ± 0.21, P = 0.06). No differences were also observed between either HIIE and Con (P = 0.12) or MICE and Con (P = 0.93) for the McAuley index at 4 h postprandial time point. The HIIE intervention may decrease the insulin concentration and improve insulin sensitivity during the postprandial period. These effects may be because skeletal muscle contractions cause the transport of glucose transporter 4 molecules to the cell membrane and increase glucose and triglyceride uptake in an intensity-dependent manner. As mentioned, the transient increase in postprandial triglyceride after the ingestion of a meal rich in fat may be a more favorable predictor of CVD. Compared with MICE and Con, triglyceride was significantly decreased to 46%–53% in HIIE at the 4-h postprandial time point. This finding indicates that the HIIE intervention is more effective than the MICE and Con interventions in decreasing triglyceride. Consistent evidence shows that performing repeated HIIE at an intensity of 85%–90% V˙O2peak for ~12–16 h before HFM ingestion can reduce PPL in healthy people (14,18,19). By contrast, previous studies have shown that acute low-volume HIIE cycling performed 12 h before HFM ingestion cannot reduce triglyceride TAUC over a 4-h postprandial period in healthy men (18,20). The reason for the lack of effects of this exercise protocol is unclear, although several suggestions can be made. For example, because PLP data were not collected over a period longer than 4 h after HFM ingestion, changes during the later period may not be able to be perceived. The result for the triglyceride TAUC during the postprandial period of 4 h differs from that during the postprandial period of approximately 6.5 h used in other studies (14,21).

A possible limitation of this study is the variation between participants in the conditions. This study used the randomized control design to determine the effects of HIIE and MICE on the PPL responses in healthy and active population. Fortunately, the characteristics of our participants at baseline, i.e., initial fitness level, exercise experiences and habits, blood parameters, and dietary intake before exercise test, were not significantly different among conditions. In addition, the familiarization trial for HIIE or MICE was performed to reduce any learning effects before the formal experiment. These data might partly attenuate the effect of interindividual variation. However, further studies with crossover design are needed to clarify the effects of HIIE or MICE on the postprandial lipoprotein profile.

In conclusion, compared with the HIIE and Con interventions, the acute MICE intervention reduces insulin levels after exercise and decreases insulin TAUC. Thus, the MICE intervention has a positive effect on glucose and insulin metabolism. However, the HIIE intervention induces greater decreases in postprandial insulin and triglyceride concentrations and greater increases in fat oxidation than the MICE and Con interventions. Therefore, a single session of HIIE in the evening exerts more favorable effects than continuous moderate-intensity exercise for decreasing insulin and triglyceride levels and increasing postprandial fat oxidation in the next morning. Because much of human life is spent in the postprandial state, these findings are of significance for the reduction of CVD risk and offer insight into the important mechanisms through which exercise reduces the risk.

The authors thank all the participants for their active participation and Mr. Shih-Feng Ting for his assistance with the testing procedure. The authors also thank the Polypact International Co., Ltd., for sponsoring the consumable materials used for the Cortex metabolic analysis system. The results of the study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation, and do not constitute. This study was supported by a grant from the Ministry of Science and Technology, Taiwan (MOST 105-2410-H-110-042).

The authors declare no conflicts of interest or source of funding in this study. The results of the present study do not constitute endorsement by the American College of Sports Medicine.

Back to Top | Article Outline


1. Zilversmit DB. Atherogenesis: a postprandial phenomenon. Circulation. 1979;60(3):473–85.
2. Nordestgaard BG, Benn M, Schnohr P, Tybjaerg-Hansen A. Nonfasting triglycerides and risk of myocardial infarction, ischemic heart disease, and death in men and women. JAMA. 2007;298(3):299–308.
3. Patsch W, Esterbauer H, Foger B, Patsch JR. Postprandial lipemia and coronary risk. Curr Atheroscler Rep. 2000;2(3):232–42.
4. Homer AR, Fenemor SP, Perry TL, et al. Regular activity breaks combined with physical activity improve postprandial plasma triglyceride, nonesterified fatty acid, and insulin responses in healthy, normal weight adults: a randomized crossover trial. J Clin Lipidol. 2017;11(5):1268–1279.e1.
5. Emerson SR, Kurti SP, Teeman CS, et al. Realistic test-meal protocols lead to blunted postprandial lipemia but similar inflammatory responses compared to a standard high-fat meal. Current Developments in Nutrition. 2017;1(4):e00232.
6. Katsanos CS. Prescribing aerobic exercise for the regulation of postprandial lipid metabolism: current research and recommendations. Sports Med. 2006;36(7):547–60.
7. Tyldum GA, Schjerve IE, Tjønna AE, et al. Endothelial dysfunction induced by post-prandial lipemia: complete protection afforded by high-intensity aerobic interval exercise. J Am Coll Cardiol. 2009;53(2):200–6.
8. Milanovic Z, Sporis G, Weston M. Effectiveness of high-intensity interval training (HIT) and continuous endurance training for V˙O2max improvements: a systematic review and meta-analysis of controlled trials. Sports Med. 2015;45(10):1469–81.
9. Gillen JB, Gibala MJ. Is high-intensity interval training a time-efficient exercise strategy to improve health and fitness? Appl Physiol Nutr Metab. 2014;39(3):409–12.
10. MacInnis MJ, Gibala MJ. Physiological adaptations to interval training and the role of exercise intensity. J Physiol. 2017;595(9):2915–30.
11. Tan MS, Mok A, Yap MC, Burns SF. Effect of sprint interval versus continuous cycling on postprandial lipaemia. J Sports Sci. 2013;31(9):989–95.
12. Tan M, Chan Moy Fat R, Boutcher YN, Boutcher SH. Effect of high-intensity intermittent exercise on postprandial plasma triacylglycerol in sedentary young women. Int J Sport Nutr Exerc Metab. 2014;24(1):110–8.
13. Ferreira AP, Ferreira CB, Souza VC, et al. The influence of intense intermittent versus moderate continuous exercise on postprandial lipemia. Clinics (Sao Paulo). 2011;66(4):535–41.
14. Thackray AE, Barrett LA, Tolfrey K. Acute high-intensity interval running reduces postprandial lipemia in boys. Med Sci Sports Exerc. 2013;45(7):1277–84.
15. Freese EC, Gist NH, Acitelli RM, et al. Acute and chronic effects of sprint interval exercise on postprandial lipemia in women at-risk for the metabolic syndrome. J Appl Physiol (1985). 2015;118(7):872–9.
16. Freese EC, Levine AS, Chapman DP, Hausman DB, Cureton KJ. Effects of acute sprint interval cycling and energy replacement on postprandial lipemia. J Appl Physiol (1985). 2011;111(6):1584–9.
17. Gabriel BM, Pugh J, Pruneta-Deloche V, Moulin P, Ratkevicius A, Gray SR. The effect of high intensity interval exercise on postprandial triacylglycerol and leukocyte activation–monitored for 48 h post exercise. PLoS One. 2013;8(12):e82669.
18. Bond B, Williams CA, Isic C, et al. Exercise intensity and postprandial health outcomes in adolescents. Eur J Appl Physiol. 2015;115(5):927–36.
19. Trombold JR, Christmas KM, Machin DR, Kim IY, Coyle EF. Acute high-intensity endurance exercise is more effective than moderate-intensity exercise for attenuation of postprandial triglyceride elevation. J Appl Physiol (1985). 2013;114(6):792–800.
20. Allen E, Gray P, Kollias-Pearson A, et al. The effect of short-duration sprint interval exercise on plasma postprandial triacylglycerol levels in young men. J Sports Sci. 2014;32(10):911–6.
21. Thackray AE, Barrett LA, Tolfrey K. High-intensity running and energy restriction reduce postprandial lipemia in girls. Med Sci Sports Exerc. 2016;48(3):402–11.
22. Panissa VL, Julio UF, Diniz TA, et al. Postprandial lipoprotein profile in two modes of high-intensity intermittent exercise. J Exerc Rehabil. 2016;12(5):476–82.
23. Maraki MI, Sidossis LS. The latest on the effect of prior exercise on postprandial lipaemia. Sports Med. 2013;43(6):463–81.
24. Silvestre R, Kraemer WJ, Quann EE, et al. Effects of exercise at different times on postprandial lipemia and endothelial function. Med Sci Sports Exerc. 2008;40(2):264–74.
25. Heyward VH, Gibson AL. Advanced Fitness Assessment and Exercise Prescription. Champaign (IL): Human Kinetics; 2014.
26. Weston M, Taylor KL, Batterham AM, Hopkins WG. Effects of low-volume high-intensity interval training (HIT) on fitness in adults: a meta-analysis of controlled and non-controlled trials. Sports Med. 2014;44(7):1005–17.
27. Jeukendrup AE, Wallis GA. Measurement of substrate oxidation during exercise by means of gas exchange measurements. Int J Sports Med. 2005;26(1 Suppl):S28–37.
28. Cohen J. Statistical Power Analysis for the Behavioral Sciences. New York: Academic Press, Inc.; 1988.
29. Hopkins WG. A scale of magnitudes for effect statistics [Internet]. 2002 [cited 2018 Jan 03]. Available from http://sportsciorg/resource/stats/effectmaghtml.
30. Pirillo A, Norata GD, Catapano AL. Postprandial lipemia as a cardiometabolic risk factor. Curr Med Res Opin. 2014;30(8):1489–503.
31. Yoshioka M, Doucet E, St-Pierre S, et al. Impact of high-intensity exercise on energy expenditure, lipid oxidation and body fatness. Int J Obes Relat Metab Disord. 2001;25(3):332–9.
32. Richter EA, Derave W, Wojtaszewski JF. Glucose, exercise and insulin: emerging concepts. J Physiol. 2001;535(Pt 2):313–22.
33. Ren JM, Semenkovich CF, Gulve EA, Gao J, Holloszy JO. Exercise induces rapid increases in GLUT4 expression, glucose transport capacity, and insulin-stimulated glycogen storage in muscle. J Biol Chem. 1994;269(20):14396–401.
34. Williams CB, Zelt JG, Castellani LN, et al. Changes in mechanisms proposed to mediate fat loss following an acute bout of high-intensity interval and endurance exercise. Appl Physiol Nutr Metab. 2013;38(12):1236–44.
35. Seip RL, Angelopoulos TJ, Semenkovich CF. Exercise induces human lipoprotein lipase gene expression in skeletal muscle but not adipose tissue. Am J Physiol. 1995;268(2 Pt 1):E229–36.
36. Grandjean PW, Crouse SF, Rohack JJ. Influence of cholesterol status on blood lipid and lipoprotein enzyme responses to aerobic exercise. J Appl Physiol (1985). 2000;89(2):472–80.
37. Doolittle MH, Ben-Zeev O, Elovson J, Martin D, Kirchgessner TG. The response of lipoprotein lipase to feeding and fasting. Evidence for posttranslational regulation. J Biol Chem. 1990;265(8):4570–7.
38. Børsheim E, Knardahl S, Høstmark AT. Short-term effects of exercise on plasma very low density lipoproteins (VLDL) and fatty acids. Med Sci Sports Exerc. 1999;31(4):522–30.
39. Davitt PM, Arent SM, Tuazon MA, Golem DL, Henderson GC. Postprandial triglyceride and free fatty acid metabolism in obese women after either endurance or resistance exercise. J Appl Physiol (1985). 2013;114(12):1743–54.
40. Trilk JL, Singhal A, Bigelman KA, Cureton KJ. Effect of sprint interval training on circulatory function during exercise in sedentary, overweight/obese women. Eur J Appl Physiol. 2011;111(8):1591–7.
41. Richards JC, Johnson TK, Kuzma JN, et al. Short-term sprint interval training increases insulin sensitivity in healthy adults but does not affect the thermogenic response to beta-adrenergic stimulation. J Physiol. 2010;588(Pt 15):2961–72.
42. Little JP, Gillen JB, Percival ME, et al. Low-volume high-intensity interval training reduces hyperglycemia and increases muscle mitochondrial capacity in patients with type 2 diabetes. J Appl Physiol (1985). 2011;111(6):1554–60.
43. Gutch M, Kumar S, Razi SM, Gupta KK, Gupta A. Assessment of insulin sensitivity/resistance. Indian J Endocrinol Metab. 2015;19(1):160–4.


© 2018 American College of Sports Medicine