Article Text

Download PDFPDF

Patterns of exercise-related inflammatory response in sickle cell trait carriers
  1. Julien Tripette1,2,
  2. Philippe Connes1,
  3. Mona Hedreville1,3,
  4. Maryse Etienne-Julan4,
  5. Laurent Marlin1,
  6. Olivier Hue1,
  7. Marie-Dominique Hardy-Dessources2
  1. 1Département de Physiologie, Laboratoire ACTES (EA 3596), Université des Antilles et de la Guyane, Pointe-à-Pitre, Guadeloupe, France
  2. 2Inserm U763, Université des Antilles et de la Guyane, Pointe-à-Pitre, Guadeloupe, France
  3. 3Service de Cardiologie, Centre Hospitalier Universitaire, Pointe-à-Pitre, Guadeloupe, France
  4. 4Centre Caribéen de la Drépanocytose ‘Guy Mérault,’ Pointe-à-Pitre, Guadeloupe, France
  1. Correspondence to Dr Marie-Dominique Hardy-Dessources, Inserm U763, Université des Antilles et de la Guyane, Pointe-à-Pitre, Guadeloupe, France.martigua2000{at}


Objective To clarify whether sickle cell trait (SCT) carriers (SCT group) present a specific postexercise inflammatory response to repeated and strenuous exercise.

Design The patterns of inflammatory markers in response to repeated heavy exercise were investigated in SCT carriers (SCT group: eight men, 20.0±0.7 years) and subjects with normal haemoglobin (CONT group: seven men, 20.6±0.7 years). The exercise consisted of three successive maximal ramp exercise tests, interspaced with 10 min of recovery, and accomplished at room temperature. Blood was sampled at rest (TR), at the end of each of the three tests (T1, T2, T3) and during the immediate (T1 h, T2 h) and late (T24 h, T48 h) recovery periods. Standard haematological parameters and plasma levels of cytokines (TNFα, IL-6) and adhesion molecules: soluble L- and P-selectins (sL-selectin, sP-selectin), soluble vascular cell adhesion molecule-1 (sVCAM-1), soluble intracellular adhesion molecule-1 (sICAM-1) were measured.

Results In both groups, the three successive maximal exercise bouts prompted an inflammatory response (ie, white blood cells and IL-6 levels increased in response to exercise). sICAM-1 and sVCAM-1 levels did not change during or after exercise and presented no difference between groups. However, during exercise, sL-selectin and sP-selectin kinetics differed between groups: sL-selectin increased earlier in the SCT group than in the CONT group, and sP-selectin statistically increased only in the SCT group.

Conclusion Although the data do not indicate an extended exercise inflammatory response in SCT carriers, a specific activation of the L- and P-selectins was observed. Further studies are needed to determine whether the selectins' changes are evidence of greater risk for SCT carriers during physical exercise in specific conditions or an indication of a protective mechanism mediated by the shedding process of adhesion molecules.

Statistics from


Sickle cell anaemia (SCA) is an inherited haemoglobinopathy most commonly seen in people from Africa, India and the Mediterranean and Caribbean areas. The hallmarks of the disease are anaemia and recurrent painful vaso-occlusive crises which result from the physical obstruction of small blood vessels with rigid sickle-shaped red blood cells (RBCs). The vaso-occlusive process is facilitated by the adhesion of circulating cells on the endothelial wall1 and mediated by a pro-inflammatory vascular environment promoting increased cell surface receptor expression or activation on erythrocytes, leucocytes and endothelial cells.1 2

Sickle cell trait (SCT) is the heterozygous form of SCA. It is characterised by the presence of both normal haemoglobin A (HbA) and abnormal haemoglobin S (HbS) and is commonly considered as benign and asymptomatic in routine activities. However, the role of SCT as a cofactor for morbidity and mortality has been suggested,3 and a recent report has proposed reclassifying SCT as a disease state.4 Moreover, several studies have reported cases of exercise-related sudden death.5,,7 Kark et al 6 observed a 30-fold higher rate of exercise-related sudden death among black recruits with SCT compared with those carrying normal haemoglobin. Metabolic or environmental changes such as hypoxia, acidosis, dehydration, hyperosmolality or hyperthermia may prompt polymerisation of HbS and, hence, transform asymptomatic SCT into a syndrome resembling SCA with a vaso-occlusive process.5 7

Moreover, it is increasingly recognised that physical exercise induces various changes in the human immune system, stimulating immune cell and adhesion molecule expression.8 It was recently hypothesised that increased cytokine and adhesion molecule expression following intense exercise might promote vaso-occlusive crises in SCT carriers.9 However, the role of the vascular cell adhesion molecule-1 (VCAM-1) may have been overemphasised in this study. First, the vaso-occlusive process characteristic of SCA patients does not depend only on the action of VCAM-1,1 10 and second, α4β1, which is the favourite VCAM-1 ligand, is only present on immature RBCs (ie, reticulocytes), which are increased in SCA patients but not in SCT carriers.11 Other endothelial adhesion molecules such as the selectins can promote the vaso-occlusive process through leucocyte adhesion and have to be brought to our attention.

It is possible that a specific postexercise inflammatory response in SCT carriers may have contributed to deleterious events already reported.5,,7 Therefore, we hypothesised that the pattern of change in inflammatory molecules and blood cells in response to strenuous exercise would be different between SCT carriers and subjects with normal haemoglobin. We examined standard haematological indices as well as the plasma levels of several inflammatory molecules known to be involved in both the immune response to exercise and the vaso-occlusive process of SCA, that is, tumour necrosis factor alpha (TNFα), interleukin-6 (IL-6) and the following soluble adhesion molecules: VCAM-1, intracellular adhesion molecule-1 (ICAM-1), and the P- and L-selectins.


Ethics approval

The ethics committee of the academics hospitals of Pointe-à-Pitre approved the present protocol. The subjects were informed of the procedures and purposes of the study, and they gave their written informed consent. The protocol was in accordance with the guidelines set by the Declaration of Helsinki.


Eight SCT carriers (SCT group) and seven subjects with normal haemoglobin (CONT group) participated in the study. All subjects were male. They were students at the University of the French West Indies and Guyana. They reported regular training in sport activities (8 h/week±2 h) but had never practised sports at a high level. None of them practised endurance sports specifically. The major exclusion criteria included cardiovascular disorders, stroke, obesity and α-thalassaemia. None of the subjects presented a significant medical history. Moreover, the subjects were not under any medical treatment.


The subjects did not participate in any physical activity for the 2 days preceding the study. They then performed exercise on a cycle ergometer, which consisted of three successive maximal ramp exercise tests. The first test was used to determine the maximal exercise responses. Venous blood samples were drawn from the antecubital vein of the non-dominant arm to measure haematological parameters, cytokines and adhesion molecules, at rest (TR), at the end of the three tests (T1, T2 and T3, respectively), and after 1, 2, 24 and 48 h of recovery (T1 h, T2 h, T24 h and T48 h, respectively).

The laboratory was air-conditioned (room temperature 23–25°C). No additional drink or food was allowed before, during and for 2 h after exercise.

SCT diagnosis and haematological parameters

To test for haemoglobin type, venous blood was drawn at rest into tubes containing EDTA and screened by isoelectric focusing. The results were confirmed by citrate agar electrophoresis. The various haemoglobins were isolated and quantified by high-performance liquid chromatography. A test of solubility confirmed the presence of HbS. Positive test results for SCT were determined by the presence of HbS (<50%) and a normal percentage of HbA2. Total counts of leucocytes, polynuclear neutrophils, monocytes, lymphocytes, platelets, RBCs and reticulocytes (expressed in percentages) and haemoglobin concentration ([Hb]) were determined from blood sampled in EDTA tubes at rest, at the end of exercise (T3) and during recovery (T2 h, T24 h and T48 h), using a haematology analyser (Max M-Retic, Beckman Coulter, Miami, Florida, USA). Haematocrit was also measured (data not shown) in order to take into account the possible plasma volume change induced by exercise. Thus, correction of plasma concentrations for blood cells and inflammatory molecules were possible according to the method described by Van Beaumont.12 Resting haematological parameters (notably mean corpuscular volume and mean corpuscular haemoglobin concentration; data not shown) were also used for the indirect diagnosis of anaemia and α-thalassaemia, which is known to result in haematological modifications.13

Exercise protocol and measurements

The trial consisted of three maximal ramp exercise tests interspaced with 10 min of seated recovery comprising 5 min of easy pedalling and 5 min of rest. The first test began with a 3 min warm-up at 60 W, and the load was increased by 30 W every minute until exhaustion. The next two tests began immediately with a 1 min step at 60 W, and the load increment was the same as in the first test. Pedalling speed remained constant at 70 rpm during the full trial. Gas exchanges were measured during the first and third tests using a breath-by-breath automated exercise metabolic system (Vmax 229, Sensor Medics, Yorba Linda, California USA) and oxygen uptake was considered maximal (ie, Vo2 max) if at least three of the following four criteria were met: (1) a respiratory exchange ratio greater than 1.10, (2) attainment of age-predicted maximal heart rate (HRmax) (210–(0.65×age)±10%), (3) an increase in oxygen uptake (Vo2) lower than 100 ml with the last increase in work rate and (4) an inability to maintain the required pedalling frequency (70 rpm) despite maximum effort and verbal encouragement. Maximal aerobic power was also determined for the first and third tests, and the total testing time (Ttest) was noted. A 10-lead electrocardiogram (Hellige, Marquette Medical Systems, Freiburg, Germany) was used to continuously monitor heart rate.

Measurements of proinflammatory cytokines and adhesion molecules

For every time-point (TR, T1, T2, T3, T1 h, T2 h, T24 h and T48 h), plasma was separated from the collected blood by centrifugation (1000 g, 4°C, 5 min) and stored at –80°C. ELISA kits were used to assay the plasma concentrations of (1) pro-inflammatory cytokines: TNFα (human TNFα DuoSet kit, range: 15.6–1000 pg/ml, R&D Systems, Minneapolis, Minnesota) and IL-6 (High sensitivity kit; range: 1.56–50 pg/ml; sensitivity: 0.8 pg/ml; Diaclone Systems, Besançon, France) and (2) soluble adhesion molecules: sVCAM-1 (VCAM-1 Eli-Pair; range: 1.56–50 ng/ml; Diaclone Systems), sICAM-1 (ICAM-1 Eli-Pair; range: 1.56–50 ng/ml, Diaclone Systems), sL-selectin (Human sL-selectin ELISA; range: 1.6–60 ng/ml, Diaclone Systems) and sP-selectin (Human sP-selectin ELISA; range: 2.19–140 ng/ml, Diaclone Systems), according to the manufacturer's instructions. These adhesion molecules, which are expressed on the vascular endothelium, platelets and leucocytes, can also be found in their soluble forms in plasma, reflecting endothelial, platelet or leucocyte activation.14

Statistical analysis

The results are presented as mean±SEM. Anthropometric and Ttest results were compared between the two groups using an unpaired Student t test. The other exercise test results (Vo2 max, HRmax, maximal aerobic power) at T1 and T3; the haematological parameters; and the cytokines and soluble adhesion molecules obtained at rest, at the end of each maximal ramp exercise test, and during the recovery were compared between the two groups using a two-way analysis of variance with repeated measures. Regarding cytokines, adhesion molecules and blood cells, the statistical analyses were performed on the values corrected for plasma volume variation. Pairwise contrasts were used when necessary to determine where significant differences had occurred. The significance level was defined as p<0.05. Analyses were conducted using Statistica (v. 5.5, Statsoft, Tulsa, Oklahoma, USA).


Subject characteristics

As shown in table 1, anthropometric data (age, height, weight) were not different between the SCT group and CONT group. The exercise test results (HRmax, Vo2 max, maximal aerobic power and Ttest) did not differ between the groups. In addition, HRmax and Vo2 max were not statistically different between the first and third tests, while the maximal aerobic power significantly decreased in both groups (p<0.05).

Table 1

Subjects' characteristics at rest (TR), and ramp test results at the end of the first (T1) and third (T3) repetition

Haematological measurements

SCT group presented 38.6%±0.9 of haemoglobin S (table 1). As can be seen in table 2, only the lymphocyte count was higher in the CONT group than in the SCT group at TR (p<0.05). Leucocytes and polynuclear neutrophils statistically increased at T3, T2 h and T24 h in the two groups (p<0.05), before returning to resting values at T48 h.

Table 2

Haematological data at rest (TR), at the end of exertion (T3) during the immediate (T2 h) and late (T24 h, T48 h) recovery

A significant increase in the monocyte count over the resting value was measured in the SCT group at T3 and T2 h (p<0.05), whereas this increase was only observed at T2 h in the CONT group (p<0.05). Monocyte counts returned to baseline at T24 h in both groups. The lymphocyte count statistically increased at the end of exercise in both groups. Then, in the CONT group, it decreased under resting values at T2 h, before returning to baseline at T24 h and T48 h (p<0.05), while in the SCT group, it returned to baseline at T2 h, before rising over resting values at T24 h and T48 h (p<0.05). Platelet and RBC counts and [Hb] presented higher values at T24 h and T48 h in both groups compared with TR (p<0.05). No time effect was noted concerning reticulocytes.

Thus, except for the lymphocyte and monocyte counts, no significant group effect in the circulating blood cell counts was observed at any time. However, in order to have a better evaluation of the exercise response in the two groups, the comparisons between groups were also calculated, for each parameter, on the basis of the percentage of changes related to the resting values (ie, TR) (data not shown). Two important results can be identified at the end of exercise (T3): (1) the leucocyte count in the SCT group (156%±5) appeared to be significantly higher than that of the CONT group (137%±7) and (2) a trend towards an increase in the neutrophil counts in the SCT group was also detected (146%±10 vs 130±6 respectively in SCT and CONT).

Measurements of proinflammatory cytokines and adhesion molecules

TNFα was undetectable at all times in the subjects, except for two subjects at the end of exercise (one in the SCT group, one in the CONT group; values not shown), which indicates very low plasma concentrations. The mean plasma concentration of IL-6 was not significantly different between groups at any time. However, IL-6 was significantly higher than resting values at T3 and T1 h (p<0.05; figure 1) in the two groups. No time or group effect was noted for sICAM-1 or sVCAM-1 (table 3). Plasma levels of sL and sP-selectins are both expressed as raw values (table 3) and as percentages of the resting values (figures 2, 3). The plasma concentration of sL-selectin was not significantly different between the two groups at any time (table 3, figure 2). However, when expressed as a percentage of the resting value, its kinetics differed between groups with a group effect detected in the immediate recovery T1 h, T2 h (p<0.05; figure 2). As shown in figure 2, sL-selectin increased above the resting value in the CONT group the day after exercise (T24 h), whereas it was significantly increased above baseline level at T3, T1 h, T2 h and T24 h in the SCT group (p<0.05; figure 2). Similarly, the plasma sP-selectin concentration did not differ significantly between the two groups at any time (table 3), but when expressed as a percentage of resting value (figure 3), sP-selectin increased over the baseline level at T1, T2 and T3 (+33.8%) only in the SCT group with a group effect at T2 and T3 (p<0.05).

Figure 1

Effect of three progressive maximal exercises on IL-6 plasma concentration in the sickle cell trait group (dotted line) and the normal haemoglobin group (solid line). There was no group effect. Asterisks indicate the time effect, which was different from TR in the two groups (p<0.05).

Figure 2

Effect of three progressive maximal exercises on sL-selectin kinetic in the sickle cell trait (SCT) group (dotted line) and the normal haemoglobin (CONT) group (solid line). The values shown in each group are expressed as a fold increase in comparison with the resting value (TR). Filled diamonds indicate differences from TR in the CONT group (p<0.05). Double daggers indicate differences from TR in the SCT group (p<0.05). Asterisks indicate differences between the two groups (p<0.05).

Figure 3

Effect of three progressive maximal exercises on sP-selectin kinetic in the sickle cell trait (SCT) group (dotted line) and the normal haemoglobin group (solid line). The values shown in each group are expressed as a fold increase in comparison with the resting value (TR). Double daggers indicate differences from TR in the SCT group (p<0.05). Asterisks indicate differences between the two groups (p<0.05).

Table 3

Raw data for soluble CAMs and selectins, at rest (TR), during exercise (T1, T2, T3) and during recovery (T1 h, T2 h, T24 h, T48 h)


The main findings of this study were that three successive progressive and maximal exercise tests (1) provoked the same plasma IL-6 response in SCT carriers and a control group, (2) did not produce any significant variations in sICAM-1 or sVCAM-1 level in either group, (3) increased sL-selectin earlier in the SCT group than in the CONT group and (4) induced an increase in sP-selectin only in SCT carriers.

As reported in previous studies, maximal aerobic power, Vo2 max and HRmax were not statistically different between SCT carriers and the control group.15,,18 The two groups presented a low maximal aerobic power, since none of them was highly involved in endurance activities such as cycling or running. Although HRmax and Vo2 max observed at T3 did not differ from the values observed at T1 in either group, the maximal aerobic power declined between the first and third exercise tests in both groups, indicating that all subjects had reached exhaustion.

Except for lymphocyte count, which was in the normal range for the two groups, no difference was found at rest between the two groups for haematological parameters, IL-6 or adhesion molecule levels. These results agree with those of Duits et al,19 who showed no difference in plasma concentration of soluble adhesion molecules between SCT carriers and subjects with normal haemoglobin, and differ from those of Monchanin et al9 with regard to sVCAM-1 plasma concentration. However, in the present study, statistically similar concentrations of IL-6 and soluble adhesion molecules indicate the absence of any distinguishing signs of inflammation with these markers between groups.

In agreement with previous studies,20,,22 the three successive maximal exercises increased IL-6 plasma concentration and total leucocyte and platelet counts in both groups, suggesting an acute postexercise inflammatory response. Nevertheless, the percentage increase in total leucocytes at T3 as compared with resting values (ie, TR) was significantly higher in the SCT group than in the control group. In addition, for neutrophils, the percentage increase at T3 compared with TR tends to be higher in the SCT group. These results argue in favour of a greater inflammatory response magnitude in SCT carriers. Peak IL-6 levels were comparable with those observed by Ronsen et al23 in response to repeated versus single bouts of prolonged cycling. Some data also suggest that IL-6 might exert anti-inflammatory effects through an inhibition of the production of TNFα and IL-1 and a stimulation of the production of IL-1ra and IL-10.24 Unfortunately, the size of the blood samples did not allow us to measure the anti-inflammatory cytokines IL-1ra or IL-10. However, the absence of detectable levels of TNFα in the present study could be mediated by this anti-inflammatory effect of IL-6.

The exercise intensity-dependent increase in circulating adhesion molecules is thought to occur as a result of shedding via adrenergic mechanisms.25 However, we did not observe sICAM-1 or sVCAM-1 changes during exercise or recovery, suggesting no exercise-dependent effect on the activation state of the endothelium in the present study. These results are in contrast to those reported by Signorelli et al22 and Monchanin et al,9 who observed an increase in sICAM-1 and sVCAM-1 after a treadmill exercise test in volunteers with peripheral vascular disease,22 and in sVCAM-1 after progressive and maximal exercise in volunteers with SCT,9 respectively. The reasons for these discrepancies are unclear, but it seems that the changes in plasma soluble adhesion molecules depend on the exercise mode and intensity.9 22 26 27 Additionally, the measured values of the cytokines and adhesion molecules are affected by changes in plasma volume, (ie, the effect of haemoconcentration) independently of true endothelial activation. No evidence of greater endothelial activation in SCT carriers than in control subjects was consequently observable in the present study when variation of plasma volume was taken in account. Moreover, the α4β1 integrin, VCAM-1's ligand, is especially present on immature RBCs (ie, reticulocytes),28 and the percentage of reticulocytes was normal (about 1% only) in our two populations and did not change with exercise. Therefore, the role of sVCAM-1 in the vaso-occlusive process and exercise-related sudden death may have been overemphasised in Monchanin et al's study.9

Although sP-selectin levels were not significantly different between the two groups, we observed several kinetic differences during exercise. Plasma sP-selectin increased during exercise in the SCT group only, a result which was close to the results obtained by Signorelli et al22 in individuals with peripheral arterial disease after a treadmill exercise test. The kinetics of sL-selectin also differed between groups. Although the three successive maximal exercises increased sL-selectin in both groups, the increase occurred earlier in the SCT group (ie, from T3 to T24 h) than in the CONT group (ie, at T24 h), showing a significantly higher percentage of increase in carriers of SCT at the end of exercise and during the immediate recovery period. This increase in sL-selectin could be explained by the increase in leucocytes during exercise (ie, at T3), but the time correspondence between these two parameters is only observable for the SCT group. Further studies taking into account the leucocyte and platelet selectin expression are needed to better understand the entire phenomenon of sL- and sP-selectin increases. Chiang and Frenette1 described the different processes occurring during SCA complications and the important role of platelets and leucocytes in the propagation of blood vessel occlusion.1 The selectins are indeed known to promote the rolling of circulating cells on the endothelial wall,29 leading to a slowdown in the microcirculatory blood flow. The specific sL- and sP-selectin responses to exercise in our SCT carriers selectins may indicate a specific activation of platelets and leucocytes in SCT carriers and potential disturbance in blood flow structuring in the microcirculation. In addition, previous investigations have reported a higher blood viscosity and RBC rigidity at the end of exercise and during recovery in SCT carriers.17 30 The cumulative effects of disturbed haemorheology and leucocyte/platelet activation in SCT carriers could promote a microvaso-occlusive process in this population and increase the risks for fatal complications, as suggested by several studies.5,,7 31 32 Moreover, the SCT group exhibited higher monocyte counts than control subjects at the end of exercise and during late recovery. This supports the hypothesis of a greater postexercise risk in SCT carriers, since these blood cells are known to adhere to the endothelium,29 promoting microvascular disturbances.

Nevertheless, considering that the increase in sL- and sP-selectin does not bring the selectin levels up to the levels seen in the CONT group, an alternative hypothesis can be considered. The shedding of P- and L-selectins may reduce the exposure of platelets and leucocytes to endothelium-derived inflammatory mediators, thereby restricting adhesion processes.33 Nielsen and Lyberg33 observed a decrease in L-selectin expression on leucocytes after a marathon and a half-marathon in healthy athletes, while the soluble form was increased.33 The authors concluded that exercise has an immunosuppressive effect.33 Moreover, previous studies have indicated that soluble forms of adhesion molecules may downregulate the adhesion process because they are competitive ligand inhibitors blocking cell adhesion to endothelium.34 35 Despite the postexercise haemorheological alterations usually observed in SCT carriers,17 30 one may suggest that the earlier increase in sL-selectin and the specific increase in sP-selectin in response to exercise might protect them from additional microcirculatory disorders. Further investigations focusing on adhesion molecule expression on leucocytes are thus clearly needed. This alternative hypothesis, in addition to the putative anti-inflammatory role of IL-6, might explain why exercise-related complications in SCT carriers are not more frequently reported, despite their low RBC deformability and high blood viscosity at rest and in response to exercise. However, the environmental conditions in the present study were not very stressful, since the exercise was conducted in a well-air-conditioned laboratory. It is possible that stressful contributing factors such as hyperthermia, hyperosmolality and dehydration could disturb vascular function more significantly in SCT carriers and could explain why this population could be prone to clinical complications in response to exercise.

In conclusion, repeated heavy exertion seems to trigger a specific inflammatory response in SCT carriers. Further studies are needed to determine whether this specific reaction is responsible for the complications already described in the literature in this population.5,,7 We indeed propose that under specific stressful conditions (warm climate, altitude), inflammatory mediators such as selectins could participate in microvascular disturbances. However, the specific ‘shedding process’ observed through the soluble forms of adhesion molecules could also be an effective preventive mechanism against microcirculatory disturbances in SCT carriers under less challenging environmental conditions (without the effects of heat or altitude). Studies need to be designed to determine if the selectins' changes observed in the present study for SCT carriers are evidence of greater risk during physical exercise or an indication of a protective mechanism.


We sincerely thank the Caribbean Sickle Cell Center for their financial support (fellowship to TJ) and M Romain and M-L Persain for their technical help. The authors also thank all the subjects who participated in the present study.



  • Competing interest None.

  • Ethics approval Ethics approval was provided by the ethics committee of the academics hospitals of Pointe-à-Pitre.

  • Patient consent Obtained.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.