Journal Logo


The Impact of Morning versus Afternoon Exercise on Iron Absorption in Athletes


Author Information
Medicine & Science in Sports & Exercise: October 2019 - Volume 51 - Issue 10 - p 2147-2155
doi: 10.1249/MSS.0000000000002026
  • Free


Iron deficiency (ID) is the most prevalent nutritional disorder worldwide and continues to be a prevailing health issue in athletes (1). Existing literature reports the incidence of ID to be up to 17% in male and 50% in female endurance athletes across various cohort studies (2–6). It has been established that ID impairs an individual’s oxygen transport and energy metabolism, with severe cases (i.e., anemia) linked to decreases in work capacity and maximal oxygen consumption (V˙O2max) (7). The high prevalence of ID in athletes is likely due to a combination of insufficient iron intake and mechanisms of iron loss that are exacerbated by activity. These include sweating, hematuria, gastrointestinal (GI) bleeding, and hemolysis; and in female athletes, menstrual losses (8). Moreover, the low bioavailability of dietary heme (15%–35%) and nonheme (2%–20%) iron (9) may substantiate the difficulty that the athletes experience in replenishing their daily iron losses from food. Recently, exercise-induced inflammation has also been implicated in the elevation of the primary iron regulatory hormone, hepcidin, which suppresses the absorption of dietary iron by duodenal enterocytes and iron recycling by macrophages (10,11). To date, numerous research articles have explored the postexercise response of hepcidin and its impact on iron metabolism in athletes. This work has consistently shown twofold to fourfold increases in hepcidin levels at 3 h postexercise, with the magnitude of this response mediated by exercise duration, and by the athletes’ preexercise serum ferritin (sFer) level (11–13). Of note, both Peeling et al. (11) and Newlin et al. (12) observed a significant increase in interleukin-6 (IL-6) immediately after exercise, which preceded the peak in serum hepcidin concentrations by 3 to 6 h postexercise.

With the time course profile of hepcidin response (3–6 h postexercise) established (11), it is likely that there is a period of impeded iron absorption in the GI tract after exercise, which may (negatively) affect an athlete’s iron status. Since athletes are encouraged to eat immediately after exercise to enhance the rate of glycogen restoration and protein synthesis (14), a possible disparity between mealtimes (i.e., breakfast or dinner) and optimal iron absorption may exist. Consequently, exercise-induced elevations in serum hepcidin concentration, and the prospect of subsequent postexercise reductions in iron absorption, are a likely mechanism of iron regulation that potentially contributes to ID in athletes; however, this hypothesis is yet to be confirmed. When considering iron absorption, Moretti and colleagues (15) recently explored the inflammatory and erythropoietic influences of exercise on plasma hepcidin concentration and iron absorption during a 3-wk training period using recreationally trained runners. In this study, the net effect of exercise was to decrease hepcidin concentrations and mildly increase iron absorption over time. These authors suggested a potential mechanism whereby, during the progression of training (additional 8 km run every second day), the impact of inflammation and hepcidin elevations are offset by the erythropoietic stimuli, to chronically increase iron mobilization and absorption for erythroid expansion, in an attempt to maintain iron balance in active individuals (15). However, the effect of acute elevations in serum hepcidin concentration after exercise on measurements of iron absorption remains to be investigated. Therefore, the aim of this study was to examine the influence of an exercise bout on subsequent serum IL-6, hepcidin concentrations and iron absorption in endurance athletes, and to assess the impact of exercise timing (i.e., morning or afternoon exercise) on this relationship.



Sixteen endurance-trained runners (10 male and 6 female) with suboptimal iron status (16) (Table 1) were recruited for this study. Participants could not be supplementing with iron within 3 wk of participating in the study. Participants were informed of the purpose, requirements and risks associated with their involvement. Written informed consent was obtained before study commencement. Ethics approval for this study was obtained from the Human Research Ethics Committee of The University of Western Australia.

Mean ± SD participant characteristics.

Experimental Overview

This study adopted a randomized, repeated measures crossover design. Participants attended one introductory, and two laboratory-based testing sessions during the experimental period. Sessions were separated by 14 d for males, and 28 d for females (to control for menstrual cycle phase). The introductory trial familiarized participants to the laboratory equipment and was concluded with a graded exercise test (GXT) to determine the individual’s V˙O2max. After the introductory trial, participants undertook two separate experimental trials, inclusive of:

  • (i) A 90-min morning running trial, performed at 65% V˙O2max velocity (vV˙O2max), with a stable iron isotope consumed at 30 min and 10 h postexercise to replicate a breakfast and evening dinner meal after morning exercise (am).
  • (ii) A 90-min afternoon running trial, performed at 65% vV˙O2max, with a stable iron isotope consumed 7.5 h preexercise and at 30 min postexercise to replicate a breakfast and evening dinner meal on a day where exercise was conducted in the afternoon (pm).

Participants did not exercise for 24 h before each experimental trial and were provided with a standardized low iron, high carbohydrate (CHO) diet to consume during this period. Similarly, during each experimental trial, all food and water consumption was standardized. On the day of each experimental trial (Fig. 1), participants arrived at the laboratory at 6:00 am, having fasted overnight. Basic anthropometric measures were taken, and a cannula was placed inside the participant’s forearm vein before a baseline venous blood sample was drawn (B1). Participants consumed 300 mL of sports drink (340 kJ, 21 g CHO) immediately after the baseline blood sample. During the am trial, participants commenced the exercise intervention at 6:30 am, alternatively, when undertaking the pm trial, participants remained in a rested state at the exercise physiology laboratory. Blood lactate (BLa) was sampled at the beginning and conclusion of the 90-min running task. Additionally, heart rate (HR) measurements were continuously monitored, and documented every 30 min of the run, and a rate of perceived exertion (RPE) was recorded immediately after the run. At the immediate conclusion of the run/rest task, a second venous blood sample was taken from the participant’s forearm (B2), and 30 min later, a low iron-containing meal was consumed with 200 mL of standardized water containing 5 mg of 57Fe as ferrous sulfate (FeSO4). At 3 h postexercise/rest, participants provided a third venous blood sample (B3), and at 3:30 pm, a fourth venous blood sample was collected (B4), which served as a preexercise blood sample for participants during the pm trial. During the pm trial, participants commenced the exercise intervention (as described above) at 4:00 pm, and a postexercise venous blood sample was collected immediately after finishing the run (B5). At 6:00 pm, participants consumed the same low iron-containing meal with 200 mL of standardized water containing 5 mg of 58Fe as FeSO4. A final venous blood sample was collected at 8:30 pm (3 h postexercise in the pm trial; B6), before the cannula was removed and the participant was free to leave the laboratory. Fourteen days after each administration of the iron labels, a final venous blood sample was collected to determine the erythrocyte incorporation of the absorbed stable iron isotopes (17).

Diagrammatic representation of experimental overview.

Experimental Procedures

Graded exercise test

The running GXT was conducted on a motorized treadmill (h/p/Cosmos Venus 200/100r, Germany) utilizing 3-min work and 1-min rest periods. The initial work velocity was set to 11.3 ± 1.4 km⋅h−1 with subsequent 1 km⋅h−1 increments over each work period until volitional exhaustion. During the GXT, ventilation and expired air was analyzed for concentrations of O2 and CO2 using a TrueOne 2400 Metabolic Measurement System (ParvoMedics, Sandy, UT). This system was calibrated pretest according to the manufacturer’s specifications. The V˙O2max was determined as highest 30 s V˙O2 reached during the final 3 min of the GXT. Participant’s vV˙O2max was determined as the weighted average of the velocity in the final 3 min of the test, and 65% of the associated vV˙O2max was calculated for the experimental trials.


The day before, and throughout both trials, participants adhered to a standardized diet (Table 2) that was devised in collaboration with a sports dietitian using Foodworks software (version 8.0.3553; AusBrands 2015 and AusFoods 2015 databases). The standardized iron-labeled breakfast and dinner meals were identical and consisted of eggs and toasted bread (Table 2). In addition to the iron-labeled breakfast and dinner, participants were provided with supplementary low iron containing meals and snacks.

Mean ± SD breakdown of standardized diet.

Blood lactate

Blood lactate concentration was assessed via a capillary sample taken from the fingertip of the participant. The site of collection was cleaned with a sterilized alcohol swab, before a small incision was made into the fingertip. The first blood droplet was discarded to ensure the integrity of the sample, and the blood sample was collected into a lactate pro strip to be analyzed by a Lactate Pro II Analyzer (Arkray Inc., Kyoto, Japan).

Blood collection

Venous blood was collected via a cannula inserted into a forearm vein by a trained phlebotomist with the athlete lying down for 5 min beforehand to standardize postural shifts in plasma volume. Blood was collected into a 5-mL EDTA and 8.5-mL SST II Gel vacutainer tubes. Subsequently, the sample in the SST II Gel vacutainer tube was allowed to clot for 30 min at room temperature before being centrifuged at 10°C and 3000 rpm for 10 min. Serum was divided into 1 mL aliquots and stored at −80°C until further analysis. Frozen serum samples were analyzed for circulating concentrations of sFer, IL-6, and hepcidin-25, and whole blood samples were analyzed for hemoglobin (Hb) concentration and erythrocyte incorporation of the absorbed stable iron isotopes. Concentrations of sFer were determined using a sandwich immunoradiometric assay (Roche Diagnostics). Serum IL-6 samples were measured in duplicate using a commercially available ELISA (Quantikine HS; R&D Systems, Minneapolis, MN). The analytical coefficient of variation for IL-6 was 3.3%. Serum hepcidin-25 concentration was measured by weak cation exchange enrichment of hepcidin coupled to time-of-flight mass spectrometry as described previously (18,19).

Measurement of iron bioavailability

An iron-fortified fluid labeled with stable isotopes was administered under the supervision of trained research personnel, together with a standardized meal consisting of eggs and toasted bread. Meals and fluid were provided 30 min postexercise/rest and contained 5 mg of 57Fe or 58Fe as FeSO4 in 200 mL of water. The labeled solutions were prepared by dissolving isotopically enriched elemental iron (57Fe: 97.9%, 58Fe:99.8%; Chemgas, Boulogne-Billancourt) in diluted sulfuric acid (0.1 mol⋅L−1) and by flushing the containers with argon to keep Fe in the (II) oxidation state (20). The stable iron isotope solution (5 mg⋅mL−1) was added to the water just before consumption. Erythrocyte incorporation of the oral iron dose was determined from the amount of the respective iron isotope in the red blood cells 14 d after administration. Blood samples were prepared and analyzed using procedures previously described by Hotz et al. (21) with a high-resolution, multicollector ICP-MS instrument at ETH Zürich, Switzerland. Iron absorption was calculated assuming that 80% of the absorbed iron is incorporated into erythrocytes. Total circulating iron in erythrocytes was calculated based on the measured Hb concentration, the iron content of Hb (3.47 mg Fe⋅g−1 Hb) and the individual blood volume, which was estimated using previously established formulas for females (22) and males (23).

Data analysis

All data were initially checked for normal distribution. Iron absorption and hepcidin-25 were log-transformed before analysis to stabilize the variance and back transformed. Iron absorption relative to run condition (two levels: am, pm) and meal time was analyzed using linear mixed-effects models with a random intercept for each participant. Covariates considered were sex, age, mass, meal time (2 levels: breakfast, dinner), hepcidin-25 measures (4 levels: B1, B3, B4, B6), sFer measures, IL-6 measures (4 levels: B1, B2, B4, B5), run velocity, and HR. Linear mixed-effects models with a random intercept for each participant were also used to analyze the effect of exercise on IL-6 and hepcidin-25. Covariates considered were sex, age, mass, meal time (two levels: breakfast, dinner), sFer measures, exercise (two levels: yes, no), run velocity, HR, and concentrations of IL-6 (four levels: B1, B2, B4, B5) or hepcidin-25 (4 levels: B1, B3, B4, B6) (in their respective models). All covariates were initially included in the linear mixed-effects models and were subsequently removed in a stepwise fashion based on their P value (retained if P < 0.05). Results from all linear mixed-effects models have been expressed using Beta coefficients and residual standard deviation (residual SD). Subsequent paired-samples t tests were then used to analyze any within-condition effects over time for IL-6 and hepcidin-25 and have been expressed as mean differences and 95% confidence intervals (95% CI). Paired-samples t tests were also used to explore any condition effects between am and pm trials for sFer, dietary analysis, and the physiological markers measured during the two running trials (HR, RPE, and BLa) and expressed as mean ± SD. The alpha level was defined as α ≤ 0.05.


Group characteristics

Participant demographics are shown in Table 1. Baseline sFer was 27.88 ± 11.46 μg·L−1 and 26.38 ± 9.37 μg⋅L−1 in the am and pm run trials, respectively. There were no significant differences between trials (P = 0.511). Furthermore, there were no significant differences in sFer between males and females in the am or pm trial (P = 0.847 and P = 0.292, respectively). However, female participants had significantly lower baseline Hb concentrations than male participants (P < 0.001).

Dietary analysis

Table 2 presents the nutrient analysis of the standardized diet. There were no significant differences (all P > 0.05) in relative energy intake, CHO, iron, vitamin C, or calcium intake between am and pm run trials.


The average HR during the am and pm run trials was 149 ± 13 bpm and 149 ± 11 bpm, respectively. The corresponding RPE for both runs was 13 ± 1. Concentrations of BLa prerun and postrun were 2.1 ± 0.7 mmol⋅L−1 and 1.6 ± 0.8 mmol⋅L−1 in the am trial, and 2.1 ± 0.9 mmol⋅L−1 and 1.8 ± 1 mmol⋅L−1 in the pm trial. There was insufficient evidence to reject the null hypothesis for a difference in HR, RPE or BLa between the am and pm run trials (all P > 0.05).

Serum IL-6 concentrations

Serum IL-6 concentrations are illustrated in Figure 2. Exercise significantly increased the concentration of IL-6 (P = 0.006). The IL-6 concentration was 4.938 pg⋅mL−1 (residual SD = 6.82 pg⋅mL−1) higher after exercise. No other covariates were found to have a significant effect. The am trial showed IL-6 concentrations increased 4.482 pg⋅mL−1 from B1 to B2 (P < 0.001; 95% CI, 3.144 < μ < 5.820 pg⋅mL−1), remaining elevated at B4 (6.981 pg⋅mL−1 increase from B1, P < 0.001; 95% CI, 3.701 < μ < 10.262 pg⋅mL−1) and at B5 (9.274 pg⋅mL−1 increase from B1, P = 0.003; 95% CI, 3.702 < μ < 14.846 pg⋅mL−1). The pm trial showed IL-6 concentrations were increased 4.699 pg⋅mL−1 from B1 to B4 (P < 0.001; 95% CI, 2.389 < μ < 7.008 pg⋅mL−1), and rose 7.752 pg⋅mL−1 further from prerun to postrun (B4 to B5; P = 0.004; 95% CI, 2.895 < μ < 12.610 pg⋅mL−1).

IL-6 concentrations recorded across the day, corresponding to a preexercise (6:00 am and 3:30 pm) and immediately postexercise (8:00 am and 5:30 pm) measure (and the analogous rested measures) for both the morning (am) and afternoon (pm) running trials. Note: Time points on the x-axis correspond with B1 (6:00 am), B2 (8:00 am), B4 (3:30 pm), and B5 (5:30 pm). *Significant difference between preexercise and postexercise in each trial (P < 0.05). †Significant increase from baseline (P < 0.05).

Serum hepcidin-25 concentrations

Serum hepcidin-25 concentrations are illustrated in Figure 3. Exercise had a significant effect on the change in hepcidin-25 from B1 to B3 and B4 to B6. Overall, the hepcidin-25 concentration was 0.80 nM (residual SD = 0.65 nM) higher postexercise (P < 0.001). No other covariates were found to have a significant effect. The am trial showed hepcidin-25 increased 0.86 nM from B1 to B3 (P < 0.001; 95% CI, 0.62 < μ < 1.18 nM), remaining elevated (0.80 nM increase from B1) at B4 (P < 0.001; 95% CI, 0.55 < μ < 1.13 nM), before returning to 1.05 ± 0.82 nM at B6, which was not different to baseline concentrations (P = 0.091; 95% CI, 0.32 < μ < 0.59 nM). The pm trial showed hepcidin-25 increased 0.55 nM from B1 to B3 (P = 0.007; 95% CI, 0.39 < μ < 0.78 nM) and continued to increase a further 0.54 nM from B3 to B4 (P = 0.005; 95% CI, 0.40 < μ < 0.73 nM) and 0.68 nM from B4 to B6 (P < 0.001; 95% CI, 0.52 < μ < 0.89 nM).

Hepcidin-25 concentrations recorded across the day, corresponding to a preexercise (6:00 am and 3:30 pm) and 3 h postexercise (11:00 am and 8:30 pm) measure (and the analogous rested measures) for both the morning (am) and afternoon (pm) running trials. Note: Time points on the x-axis correspond with B1 (6:00 am), B3 (11:00 am), B4 (3:30 pm), and B6 (8:30 pm). *Significant difference between preexercise and postexercise in each trial (P < 0.05). †Significant increase from baseline (P < 0.05).

Fractional iron absorption (erythrocyte iron incorporation)

Fractional iron absorption is illustrated in Figure 4. The linear mixed-effects model found that significantly more iron (0.67%, residual SD = 1.47%) was absorbed from the breakfast meal compared to the dinner meal (P = 0.011) when participants ran in the am. The covariates age (−0.93%; P = 0.001), mass (−0.97%; P = 0.032) and sFer (−0.96%; P = 0.002) were also found to have a significant effect on iron absorption. The linear mixed-effects model found there was no significant difference in the amount of iron absorbed from the breakfast meal compared to the dinner meal (P > 0.05) when participants ran in the pm. It was found that pre-pm run hepcidin-25 (B4) was the only significant covariate to explain iron absorption (−0.33%, residual SD = 1.39%; P = 0.01) when participants ran in the pm. The linear mixed-effects model also found that iron absorption at breakfast was significantly higher (0.78%, residual SD = 1.17%; P = 0.02) when they ran in the am compared to when they ran in the pm. The covariate run velocity (1.15%; P = 0.032), was also found to have a significant effect. The linear mixed-effects model estimates that iron absorption at dinner was higher (0.46%, residual SD = 1.12%; P < 0.001) when they ran in the am compared to when they ran in the PM. Within this model, the covariates age (0.94%; 0.005), B1 hepcidin (0.41%; P = 0.001), B3 hepcidin (0.58%; P = 0.009), B6 hepcidin (2.98%; P < 0.001), and sFer (0.98%; P = 0.009) were also found to have a significant effect on iron absorption at dinner.

Amount of iron absorbed at breakfast and dinner for both morning (6:30 am run) and afternoon (4:00 pm run) running trials. *Significant difference between breakfast and dinner absorption for the respective run trials (P < 0.05). †Significant difference in absorption at a meal between the am and pm run trials (P < 0.05).


Our results reveal that the interaction of exercise and time-of-day influences the amount of iron absorbed by athletes with suboptimal (sFer ≤50 μg·L−1) (16) iron stores. We have shown that despite a postexercise increase in hepcidin concentrations, more iron was absorbed at breakfast after morning exercise, as compared with breakfast in a rested state, or when compared with the absorption from an evening meal. This outcome suggests that, although the regulatory mechanism of hepcidin is at the forefront of iron absorption, there are other exercise-induced physiological changes that influence iron uptake.

Given exercise has been consistently shown to induce an increase in hepcidin concentrations (11–13), the impact on subsequent iron absorption is of interest. In agreement with prior research, which consistently report hepcidin concentrations to increase twofold to fourfold at 3 h after exercise (11,12), the 90-min running protocol used in the current study elicited an approximately threefold increase in hepcidin at 3 h after the am run, and an approximately twofold increase at 3 h after the pm run. Similarly, the magnitude of the hepcidin increase (~4 nM) was consistent with those previously reported in athletes with sFer in the range of 30 to 50 μg⋅L−1 (16). The present data also illustrate a twofold to fourfold increase in serum IL-6 concentration following the 90-min running protocol, confirming that an inflammatory response was induced by this exercise protocol and suggests that IL-6 was the likely mechanism responsible for the increase in hepcidin concentration 3 h thereafter. Of note, this elevation in serum IL-6 concentration persisted across the day after the am run, an outcome likely linked to the circadian rhythm of IL-6, which has previously been shown to increase threefold to fivefold between the hours of 8:00 am and 7:00 pm and is exacerbated by sleep deprivation and fatigue (24), both factors to consider here.

Similar to IL-6, serum hepcidin concentrations (in a nonexercise setting) also exhibit a diurnal variation in healthy adults (25,26). The kinetics of serum hepcidin’s circadian rhythm were shown by Kemna et al. (26), who reported a twofold to sixfold rise in hepcidin from 6:00 am to 3:00 pm, which subsided into the evening. Although there was no completely rested condition in this study, an analogous diurnal increase in hepcidin was observed in the pm run trial in the present study, with resting concentrations of hepcidin increasing approximately 2.7 times from 6:00 am to 3:30 pm in the absence of exercise. Our results suggest that the greater hepcidin response seen after the pm (compared with the am) run trial is an aggregate effect of elevated hepcidin concentrations in response to the combined effects of the exercise-induced inflammation and the diurnal nature of hepcidin. Here, hepcidin was found to be elevated approximately twofold greater after exercise in the afternoon (as compared to in the morning), indicating a potentially larger inhibitory effect on iron absorption, thereby helping to explain the (comparatively) reduced iron absorption postrun in the afternoon. Of note, however, despite a significantly greater serum hepcidin concentration at 11:00 am, the iron absorption postexercise in the morning was greater than at rest, whereas the amount of iron absorbed postexercise in the afternoon not substantially different to the quantity of iron absorbed from breakfast consumed at rest. Such findings support the premise that iron absorption during the postexercise period may be a net effect of one or more exercise-induced mechanisms that promote iron absorption, opposing the inhibitory effects of hepcidin activity, to influence the overall iron absorption. In the context of the present investigation, the net effect of the interaction between exercise-induced mechanisms and postexercise hepcidin concentration is an overall positive influence on iron absorption in the morning, whereas in the afternoon, greater increases in hepcidin concentrations appear to negate these potential exercise driven mechanisms. Recently, only one study has explored iron absorption in an exercise setting (15). These authors found that over a 3-wk period of increased exercise intensity and erythropoiesis, hepcidin concentrations decreased, and iron absorption increased by 24% compared to a control period of less intense exercise activity. Such outcomes corroborate that there is an exercise-induced mechanism that regulates iron balance, which is likely yet to be completely understood.

It is well known that exercise prompts a number of transient physiological changes to the cardiovascular, metabolic, and hormonal function of an individual, one or more of which may influence iron absorption. Interestingly, the statistical modeling of the present investigation identified age, mass and run velocity (but not sex) as contributing covariates to iron absorption. Intuitively, however, each of these variables are likely associated to sex, and as such, more research is required to specifically establish the potential sex-differences in iron absorption after exercise. In the context of exercise, hemolysis is one avenue of exercise-induced iron-loss (27) that may prompt changes in iron metabolism via an upregulation of intestinal iron transport proteins, as shown in mice induced with hemolytic anemia (28); however, this concept is yet to be confirmed in humans. Alternatively, exercise has been shown to create a transient increase in the permeability of the small intestine (“leaky gut”) as a result of splanchnic hypoperfusion and repetitive mechanical movement during exercise, such outcomes are thought to increase the transfer of iron across the intestinal border (29,30). This may also be reflected in the outcomes of Nachtigall et al. (17) who reported both a threefold increase in GI blood loss in periods of intensive training, compared with periods of rest, and an upregulation of iron absorption in ID (sFer <35 μg⋅L−1), male distance runners. However, no measures of inflammation were made during this study, and therefore, it is also possible that exercise-induced inflammation is at the origin of these changes in iron absorption.

In consideration of this, it might also be possible that there may exist an exercise-induced upregulation of iron transport proteins in the intestine, similar to those found in mice and human small-bowel cultures, in response to inflammatory cytokines, particularly tumor necrosis factor alpha (TNFα) (31,32). Of interest, Sharma et al. (32) demonstrated that TNFα transiently increases the expression and localization of enterocyte iron transport proteins, divalent metal transporter 1 (DMT1) and ferroportin in a human intestinal cell line and ex vivo small bowel cultures. One hour after TNFα treatment, ferroportin and DMT1-mRNA expression were increased, resulting in significant increases to iron import and export (32). As TNFα is a proinflammatory cytokine related to IL-6, cytokine-mediated changes to iron transport proteins may stimulate iron absorption immediately postexercise in the morning. However, this effect appears to be an acute response to inflammation, and in fact, prolonged treatment with TNFα appeared to block iron transport and increase iron storage (32). As a result, it is possible that the acute benefits to iron absorption may exist more after morning exercise, and that the prolonged diurnal increase in IL-6 concentration before exercise in the pm trial may explain the lack of similar effect in the afternoon.

Although the specific mechanisms stimulating iron absorption after exercise are currently unknown, similar physiological phenomena have already founded contemporary nutritional strategies. Athletes will consume combinations of nutrients during and around exercise to optimize performance and recovery, with a specific aim to maximize muscular adaptation and to facilitate the repair of damaged tissue (33). For example, the postexercise period, often termed the “anabolic window of opportunity” (33), is a limited time (after exercise) of super-compensated rebuilding of damaged tissue and restoration of energy reserves (33). Within this paradigm, the highest rates of glycogen resynthesis appear to occur when large amounts of carbohydrates (>1 g⋅(kg⋅h)−1) are consumed within 2 h of exercise cessation due to the activation of glycogen synthase and increased permeability of the muscle cell membrane to glucose (14,34). In parallel, based on the present findings, athletes with suboptimal iron status might be advised to consume/supplement with iron during this postexercise window in the morning to achieve optimal iron absorption outcomes. However, further research should aim to establish the best time to ingest iron during and around morning exercise, because preexercise iron ingestion and subsequent rates of absorption remain to be explored. Furthermore, the impact of extending the time postexercise to feeding should also be quantified, because there may be a critical point at which the postexercise elevation in hepcidin concentration begins to impact on the rate of iron absorption if the food is consumed at a time closer to the peak in hormone response.

In line with this thinking, the rate of gastric emptying and the exact time the test iron arrived at the site of absorption (relative to both exercise-induced changes in iron transport and hepcidin activity) should be considered. Previously, Siegel and colleagues (35) measured the gastric emptying rate of a comparable test meal consisting of eggs and bread, in both the solid and liquid state, recording a half-emptying time of 77.6 min for the solid meal. In the context of our investigation, the postexercise meals would have reached the absorption site in just under 2 h postexercise, and over an hour before the established (11) peak postexercise hepcidin concentration (i.e., B3 and B6). Additionally, the iron isotope we provided was consumed immediately before the solid test meal in the liquid state, formerly shown to bypass an initial lag-phase in the stomach accompanying solid meals (35). Moreover, rates of gastric emptying have been shown to be equivalent, or even enhanced, during moderate intensity exercise up to 70% V˙O2max, such as the 90-min protocol used here (36,37). Consequently, the test-iron dose provided in the current study may have reached the site of absorption at a time foregoing hepcidin’s peak obstructive influence. Lastly, gastric emptying half-time of an evening meal (8:00 pm) has previously been shown to be 54% longer compared with morning emptying half times (8:00 am) (38). Therefore, it is possible that with faster gastric emptying and lower overall hepcidin concentrations in the morning, the aforementioned mechanisms, such as the postexercise anabolic window, may be more influential, and thus summate to overall greater iron absorption after morning exercise.


In summary, these findings reveal that, despite a postexercise increase in hepcidin concentration, more iron is absorbed at breakfast after morning exercise, as compared with breakfast in a rested state, or when compared with iron absorption from an evening meal. Although the physiological mechanisms at play promoting iron absorption postexercise remains elusive, the current investigation demonstrates that overall iron absorption is impacted by inflammation and a cumulative effect of hepcidin, in addition to one or more transient postexercise physiological changes. Such transient mechanisms may likely include increased gut permeability or upregulation of intestinal iron transport proteins postexercise; however, further research is required to elucidate these prospects. The practical outcomes of this study would be to advise active individuals to consume/supplement with iron shortly after morning exercise for optimal iron absorption. Moving forward, future research should aim to find the optimal timeframe of morning iron consumption to maximize iron uptake.

The authors thank all the participants involved in this study for their commitment to the exercise and diet protocol. The authors also thank the West Australian Institute of Sport for facilitating the implementation of this study. This project was supported by the Australian Institute of Sport High Performance Sport Research Fund and Federation University Australia Seed Funding Grant. The authors declare no conflicts of interest. The results of the present study do not constitute endorsement by the American College of Sports Medicine. The results of the study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation.


1. Goodman C, Peeling P, Ranchordas M, Burke L, Stear S, Castell L. A to Z of nutritional supplements: dietary supplements, sports nutrition foods and ergogenic aids for health and performance—part 21. Br J Sports Med. 2011;45(8):677–9.
2. DellaValle DM, Haas JD. Impact of iron depletion without anemia on performance in trained endurance athletes at the beginning of a training season: a study of female collegiate rowers. Int J Sport Nutr Exerc Metab. 2011;21(6):501–6.
3. Koehler K, Braun H, Achtzehn S, et al. Iron status in elite young athletes: gender-dependent influences of diet and exercise. Eur J Appl Physiol. 2012;112(2):513–23.
4. Selby GB, Eichner ER. Endurance swimming, intravascular hemolysis, anemia, and iron depletion. Am J Med. 1986;81(5):791–4.
5. Sinclair LM, Hinton PS. Prevalence of iron deficiency with and without anemia in recreationally active men and women. J Am Diet Assoc. 2005;105(6):975–8.
6. Haymes EM, Puhl JL, Temples TE. Training for cross-country skiing and iron status. Med Sci Sports Exerc. 1986;18(2):162–7.
7. Haas JD, Brownlie T 4th. Iron deficiency and reduced work capacity: a critical review of the research to determine a causal relationship. J Nutr. 2001;131(2S-2):S676–90.
8. Peeling P, Dawson B, Goodman C, Landers G, Trinder D. Athletic induced iron deficiency: new insights into the role of inflammation, cytokines and hormones. Eur J Appl Physiol. 2008;103(4):381–91.
9. Craig WJ. Iron status of vegetarians. Am J Clin Nutr. 1994;59(5S):S1233–7.
10. Nemeth E, Rivera S, Gabayan V, et al. IL-6 mediates hypoferremia of inflammation by inducing the synthesis of the iron regulatory hormone hepcidin. J Clin Invest. 2004;113(9):1271–6.
11. Peeling P, Dawson B, Goodman C, et al. Effects of exercise on hepcidin response and iron metabolism during recovery. Int J Sport Nutr Exerc Metab. 2009;19(6):583–97.
12. Newlin MK, Williams S, McNamara T, Tjalsma H, Swinkels DW, Haymes EM. The effects of acute exercise bouts on hepcidin in women. Int J Sport Nutr Exerc Metab. 2012;22(2):79–88.
13. Roecker L, Meier-Buttermilch R, Brechtel L, Nemeth E, Ganz T. Iron-regulatory protein hepcidin is increased in female athletes after a marathon. Eur J Appl Physiol. 2005;95(5–6):569–71.
14. Burke LM, van Loon LJ, Hawley JA. Postexercise muscle glycogen resynthesis in humans. J Appl Physiol. 2016;122(5):1055–67.
15. Moretti D, Mettler S, Zeder C, et al. An intensified training schedule in recreational male runners is associated with increases in erythropoiesis and inflammation and a net reduction in plasma hepcidin. Am J Clin Nutr. 2018;108(6):1324–33.
16. Peeling P, Sim M, Badenhorst CE, et al. Iron status and the acute post-exercise hepcidin response in athletes. PLoS One. 2014;9(3):e93002.
17. Nachtigall D, Nielsen P, Fischer R, Engelhardt R, Gabbe EE. Iron deficiency in distance runners. A reinvestigation using 59Fe-labelling and non-invasive liver iron quantification. Int J Sports Med. 1996;17(7):473–9.
18. Kroot JJ, Laarakkers CM, Geurts-Moespot AJ, et al. Immunochemical and mass-spectrometry–based serum hepcidin assays for iron metabolism disorders. Clin Chem. 2010;56(10):1570–9.
19. Laarakkers CM, Wiegerinck ET, Klaver S, et al. Improved mass spectrometry assay for plasma hepcidin: detection and characterization of a novel hepcidin isoform. PLoS One. 2013;8(10):e75518.
20. Cercamondi CI, Egli IM, Mitchikpe E, et al. Total iron absorption by young women from iron-biofortified pearl millet composite meals is double that from regular millet meals but less than that from post-harvest iron-fortified millet meals. J Nutr. 2013;143(9):1376–82.
21. Hotz K, Krayenbuehl PA, Walczyk T. Mobilization of storage iron is reflected in the iron isotopic composition of blood in humans. J Biol Inorg Chem. 2012;17(2):301–9.
22. Brown E, Hopper J Jr, Hodges JL Jr, Bradley B, Wennesland R, Yamauchi H. Red cell, plasma and blood volume in healthy women measured by radiochromium cell-labelling and hematocrit. J Clin Invest. 1962;41:2182–90.
23. Wennesland R, Brown E, Hopper J Jr, et al. Red cell, plasma and blood volume in healthy men measured by radiochromium (Cr51) cell tagging and hematocrit: influence of age, somatotype and habits of physical activity on the variance after regression of volumes to height and weight combined. J Clin Invest. 1959;38(7):1065–77.
24. Vgontzas AN, Papanicolaou DA, Bixler EO, et al. Circadian interleukin-6 secretion and quantity and depth of sleep. J Clin Endocrinol Metab. 1999;84(8):2603–7.
25. Ganz T, Olbina G, Girelli D, Nemeth E, Westerman M. Immunoassay for human serum hepcidin. Blood. 2008;112(10):4292–7.
26. Kemna EH, Tjalsma H, Podust VN, Swinkels DW. Mass spectrometry-based hepcidin measurements in serum and urine: analytical aspects and clinical implications. Clin Chem. 2007;53(4):620–8.
27. Weight LM, Byrne MJ, Jacobs P. Haemolytic effects of exercise. Clin Sci (Lond). 1991;81(2):147–52.
28. Latunde-Dada G, Vulpe C, Anderson G, Simpson R, McKie A. Tissue-specific changes in iron metabolism genes in mice following phenylhydrazine-induced haemolysis. Biochim Biophys Acta. 2004;1690(2):169–76.
29. Van Wijck K, Lenaerts K, Van Loon LJ, Peters WH, Buurman WA, Dejong CH. Exercise-induced splanchnic hypoperfusion results in gut dysfunction in healthy men. PLoS One. 2011;6(7):e22366.
30. Peters H, De Vries W, Vanberge-Henegouwen GP, Akkermans LM. Potential benefits and hazards of physical activity and exercise on the gastrointestinal tract. Gut. 2001;48(3):435–9.
31. Laftah AH, Sharma N, Brookes MJ, et al. Tumour necrosis factor α causes hypoferraemia and reduced intestinal iron absorption in mice. Biochem J. 2006;397(1):61–7.
32. Sharma N, Laftah AH, Brookes MJ, Cooper B, Iqbal T, Tselepis C. A role for tumour necrosis factor α in human small bowel iron transport. Biochem J. 2005;390(2):437–46.
33. Aragon AA, Schoenfeld BJ. Nutrient timing revisited: is there a post-exercise anabolic window? J Int Soc Sports Nutr. 2013;10(1):5.
34. Ivy J, Katz A, Cutler C, Sherman W, Coyle E. Muscle glycogen synthesis after exercise: effect of time of carbohydrate ingestion. J Appl Physiol. 1988;64(4):1480–5.
35. Siegel J, Urbain J, Adler L, et al. Biphasic nature of gastric emptying. Gut. 1988;29(1):85–9.
36. Costill D, Saltin B. Factors limiting gastric emptying during rest and exercise. J Appl Physiol. 1974;37(5):679–83.
37. Neufer P, Young AJ, Sawka MN. Gastric emptying during walking and running: effects of varied exercise intensity. Eur J Appl Physiol Occup Physiol. 1989;58(4):440–5.
38. Goo R, Moore J, Greenberg E, Alazraki N. Circadian variation in gastric emptying of meals in humans. Gastroenterology. 1987;93(3):515–8.


Copyright © 2019 by the American College of Sports Medicine