Effects of exercise on postprandial lipemia
It is well-established that consumption of a fatty meal can result in a substantial rise in blood lipids, and that this lipemia has a basis in atherosclerosis [
1,
27]. It is also widely acknowledged that acute aerobic exercise performed 10–12 h prior to a high-fat meal is typically effective in attenuating postprandial lipemia [
15]. In this context, over the past few decades, research has been dedicated to better understanding the specific relationship between acute pre-meal exercise and postprandial lipemia via modification of variables such as exercise duration, intensity, mode, and timing, as well as meal and subject variables [
15]. To our knowledge, the present investigation is the first to specifically test the effectiveness of an acute bout of exercise reflective of current physical activity recommendations [
18] on postprandial lipemia and inflammation in a group of at-risk individuals (i.e. overweight men). We found that the historically recommended 30 min of aerobic exercise, when performed 12 h prior to a high-fat meal, was ineffective in blunting both fasting triglycerides and postprandial lipemia in overweight men. This finding is in agreement with the findings of Pfeiffer et al. [
28] who also found that 30 min of walking at 50 % VO
2max was ineffective in blunting postprandial lipemia in healthy young men that performed cycling exercise immediately prior to the ingestion of two mixed test meals. On the other hand, several studies have shown a beneficial effect of 30 min of aerobic exercise on postprandial lipemia [
29,
30]. However, differences in protocols are the likely explanations for the disagreement in findings, considering that Klein et al. [
30] used a higher-intensity exercise (75 % VO
2max) performed after the high-fat meal and Miyashita and Tokuyama [
29] used a moderate-fat meal in healthy participants (versus a high-fat meal in overweight participants, as in the present study). Similarly, one study found that an exercise bout producing an energy expenditure of less than 500 kcal significantly blunted postprandial lipemia, although the exercise bout was 90 min in duration and the participants had low aerobic capacities (VO
2max ~ 30 ml/kg/min) [
19].
However, the more surprising finding in our study was that 60 min of exercise was not effective for blunting postprandial lipemia (AUC-tot or AUC-inc). This null finding disagrees with several studies that have reported a postprandial lipid-lowering effect of 60 min of exercise [
20‐
22]. There are a noteworthy number of investigations in addition to the present study that have not found reduced postprandial lipemia after an hour of exercise [
28,
31‐
34]. Clearly, the lipid-lowering effect of 60 min of exercise is not universal, nor consistently confirmed in the literature. These discrepancies are certainly affected by such study design variables as meal timing, caloric content, and composition; exercise duration, intensity, and energy expenditure, as well as participant factors such as fitness and health status. However, while the present study did not find a significantly blunted lipemic response in the EX-60 condition relative to the other conditions, there were qualitative improvements in postprandial lipemia following 60 min of exercise (Fig.
2). Specifically, there was a “moderate effect” in the EX-60 condition, with the effect size (Cohen’s
d) for CON vs EX-60 in terms of AUC-tot and AUC-inc lipemic responses being 0.51 and 0.50, respectively.
Energy balance can play a critical role in determining whether exercise blunts postprandial lipemia [
35‐
37]. In the present study, participants consumed a small snack of 270 kcal after the exercise session/before leaving the laboratory. Thus, in EX-30 the energy deficit (291.3 ± 70.9 kcal) would have been nearly completely replaced, and in EX-60 the energy deficit (582.5 ± 141.8 kcal) would have been partially replaced. This replacement of energy deficit may explain the lack of effect of exercise on postprandial lipemia and the contradictory findings between the present study and previous studies that found a lipid-lowering effect of 60 min of exercise [
20‐
22].
There are also other study design features that may explain why previous investigations found that 60 min of exercise lowered postprandial lipemia and the current investigation did not. For example, Dekker et al. [
22] studied obese, hypertriacylglycerolemic men, who would display an elevated, prolonged postprandial lipemia and thus a more pronounced effect for exercise. Kolifa et al. [
21] had participants cycle at 70–75 % of maximal heart rate, leading to a greater exercise energy expenditure, and Thomas et al. [
20] used a test meal of 1.5 g fat/kg (150 % fat content of present study). We believe that our protocol is consistent with the majority of postprandial lipemia investigations [
15], reflective of realistic meal consumption and exercise intensity and duration for overweight individuals, and in line with current physical activity guidelines [
18]. In light of this, our findings that 30 and 60 min of moderate-intensity aerobic exercise were ineffective in attenuating postprandial lipemia have important public health implications, and could be considered a caveat to our traditional physical activity recommendations for adults. It appears that exercise 12 h prior to a high-fat meal, even when expending more than 500 kcal, may be insufficient for attenuating postprandial lipemia if calories are completely or partially replaced following exercise. Less fit individuals may indeed require less exercise energy expenditure to blunt postprandial lipemia [
19], but this may need to be in the context of non-replacement of the exercise-induced energy deficit; otherwise a greater exercise energy expenditure may be required.
Circulating markers of inflammation
Previous studies have suggested that inflammatory cytokines increase after a HFM in overweight participants [
38], however this is not a consistent finding due to variability in many studies, as well as differential time course for measurements [
39]. In the current study, we assessed cytokines from 10 min prior to a HFM (1 g/kg fat and 1 g/kg CHO) to 8 h post-HFM. With the large time span from baseline to 8 h after the high-fat meal, it was unexpected to find an increase in PPL without significant changes in the cytokines measured. However, the inflammatory cytokine literature is conflicting and several factors may explain the increase in PPL without a change in pro-inflammatory cytokines. Specifically, although the rise postprandial triglycerides has been reported to induce inflammation [
40], this relationship between changes in triglycerides may not always be associated with changes in inflammation [
41‐
43].
Although a 60-min moderate bout of exercise has been reported to attenuate PPL as previously stated, an acute bout of physical activity may still have anti-inflammatory effects without attenuating PPL. These anti-inflammatory actions may occur by increasing IL-6 from the contracting muscle [
13]. Release of IL-6 may promote an anti-inflammatory environment by increasing downstream release of anti-inflammatory cytokines such as IL-10, and attenuating subsequent increases in TNF-α [
13]. Although changes in cytokines in the current study were not significantly different between trials and there was large inter-subject variability, peak anti-inflammatory IL-4 displayed marginal differences in EX-60 compared to EX-30 and CON (IL-4 95 % CIs; EX-60: 39.895–166.904 pg/mL: EX-30: 48.988–120.508 pg/mL: CON: 36.336–132.860 pg/mL). Therefore, although there were no statistically significant changes over time, exercise may have increased the anti-inflammatory environment, thereby increasing IL-4.
It is well established that the generation of reactive oxygen species and subsequent increases in circulating inflammatory cytokines occur following consumption of a HFM [
44,
45] and also following acute exercise [
46] in overweight individuals compared with lean controls. Specifically, Aljada and colleagues [
44] reported that IL-6 increased from baseline to 2 h post-HFM in lean and obese individuals; however, IL-6 remained elevated in obese individuals only. In Fig.
3 from our data, it appears the peak IL-6 occurred similarly to the lean participants described in Aljada and colleagues’ previous work [
44], whereas the peak IL-6 response in EX-30 qualitatively increased more slowly and remained elevated until 6 h post-HFM. These considerations may also help to elucidate why IL-6 was only associated with a change in triglycerides in the EX-30 condition. Although IL-6 may act as an anti-inflammatory cytokine post-exercise [
13], it can also act as a pro-inflammatory cytokine post-HFM [
47]. Our laboratory has reported that exercise, either 12 h before a HFM or 30 min after a HFM, may not attenuate postprandial lipemia [
48‐
50]. In both studies, there were no reported changes in cytokines post-HFM. In the current study, the finding that IL-6 did not increase post-HFM in CON was surprising, although it was alignment with recent studies [
39] that showed a decrease in IL-6 from baseline to 2 h following a high-fat meal before becoming elevated above baseline. Therefore it appears that many factors influence the actions of these cytokines, and further investigations should be performed to elucidate the time course of inflammation (both in terms for pro- and anti-inflammatory cytokines), as well as interactions between cytokines when investigating postprandial inflammation.
Strengths and experimental considerations
The present study is novel in that it employs exercise bouts that may be reasonable for individuals in the general population to complete. As stated previously, the majority of previous postprandial lipemia and inflammation investigations have used exercise protocols that may not be realistic for many individuals to take part in on a regular basis. Also, the current study included at-risk (overweight or obese) individuals, a relevant study population. Other strengths of the present study are the 8-h measurement duration in the postprandial period, and the large number of metabolic and inflammatory markers assessed. In addition, the current study fills a gap in previous research by examining postprandial lipemia and postprandial inflammation together.
To our knowledge, the present study is among the first to examine the effects of exercise on both postprandial lipemia and inflammation using an exercise duration and intensity that reflect current physical activity guidelines. However, certain considerations should be made when drawing conclusions from the current data. First, partial replacement of the exercise-induced energy deficit (via standard small snack consumption before leaving the laboratory) may have diminished the lipid-lowering effects of the exercise sessions. This post-exercise energy replacement may explain why there was no significant blunting of postprandial lipemia in either exercise trial [
35‐
37]. However, in our opinion consumption of a small snack after the exercise sessions was realistic and reflective of true-to-life behavior. We do acknowledge that the composition of the food item used to replace exercise-induced energy expenditure can have substantial effects on the degree to which the lipid-lowering effects of exercise are negated [
51]. Our utilization of a Little Debbie Swiss Roll as a post-exercise snack could have produced effects on postprandial lipemia that may be different from other food items, such as a post-exercise protein shake, and this should be considered when interpreting our results. We assert that our usage of a highly-processed snack as a means of replacing post-exercise energy expenditure coincides with the Western-based test meal, as well as the notion that our studied population (overweight, insufficiently active individuals) may be more likely to consume such a snack compared to a protein shake or something similar. In addition, without a significant effect of exercise on postprandial lipemia, we could not effectively evaluate the association between postprandial changes in triglycerides and inflammation, weakening our ability to support or reject our third hypothesis that there would be a correlation between the postprandial lipemic and systemic inflammatory responses to the high-fat meal. Next, as the test meal was comprised of ice cream and whipping cream, the findings of the present study cannot be extrapolated to other meals with differing macronutrient distribution or nutrient content. Similarly, the test meal was quite large (~1120 kcal), and may not necessarily have been reflective of daily, habitual food intake. Finally, the issue of sample size needs to be addressed. While there were qualitative improvements in the lipemic response during the EX-60 condition compared to CON and EX-30, we did not find any significant statistical differences. It could be suspected that the present investigation was simply under-powered to detect trial differences. However, a
post-hoc power analysis based on differences between CON and EX-60 (for both AUC-tot and AUC-inc) revealed that an additional 20 participants (32 total) would need to be recruited to the study in order to sufficiently detect trial differences and reject the null hypothesis. The effect size (Cohen’s
d) for CON vs EX-60 in terms of AUC-tot and AUC-inc lipemic responses were 0.51 and 0.50, respectively, reflecting a moderate effect. Thus, despite a potential lipid-lowering effect of exercise, it is unlikely that the present study was merely underpowered, but rather this issue highlights the ability of post-exercise (partial) caloric replacement to negate the lipid-lowering effects of acute exercise.