Hemorheological alterations related to training and ... - CiteSeerX

Biorheology 47 (2010) 95–115 DOI 10.3233/BIR-2010-0563 IOS Press

95

Review

Hemorheological alterations related to training and overtraining ∗ Jean-Frédéric Brun c,∗∗ , Emmanuelle Varlet-Marie a,c , Philippe Connes b and Ikram Aloulou c a

Laboratoire de Biophysique and Bio-Analyses, Faculté de Pharmacie, Université Montpellier I, Montpellier, France b Laboratoire ACTES, Département de Physiologie, Université des Antilles et de la Guyane, Campus de Fouillole, Pointe-à-Pitre, Guadeloupe, French West Indies c INSERM ERI 25 Muscle et Pathologies, Service Central de Physiologie Clinique, Unité d’Exploration Métabolique, CHU Lapeyronie, Montpellier, France Received 24 December 2008 Accepted in revised form 30 March 2010 Abstract. Alterations of blood rheology related to muscular activity have been extensively studied over the last 20 years. It has been shown that exercise exerts a “triphasic” action on the rheological properties of blood. In the short term, exercise induces a transient hyperviscosity, mostly due to a rise in hematocrit and plasma viscosity, but also to alterations in erythrocyte rheology. Reversal of this hyperviscosity pattern over the following 24 h can be described as an “autohemodilution”. Later, training results in several profiles of “hemorheologic fitness” with a low hematocrit reflecting an expansion in plasma volume, and improvements in red cell rheology (increased deformability, decreased aggregation, reduced disaggregation shear rate). Some specific aspects of these long-term adaptations have been described, such as the intriguing occurrence of a paradoxical improvement in RBC deformability during exercise in some athletes, and overtraining, which is associated with higher plasma viscosity. Given the variety of modes of exercise and the wide heterogeneity of their effects on blood rheology in the short and long term, many investigations remain to be performed in this area of clinical hemorheology. Keywords: Exercise, blood viscosity, erythrocyte deformability, erythrocyte aggregation, training, overtraining

Part I. Effect of exercise on hemorheological parameters 1. Introduction Exercise hemorheology has been the subject of many studies in athletes, in sedentary people and patients suffering from various diseases [9,16,45]. Over the last years, we developed concepts aimed *

This article is based on a paper given by Dr. J.F. Brun in Symposium 6 at the 13th International Congress of Biorheology and the 6th International Conference on Clinical Hemorheology at the Pennsylvania State College, PA, USA, July 9–14, 2008. ** Address for correspondence: Dr. J.F. Brun, INSERM ERI 25 Muscle et Pathologies, Service Central de Physiologie Clinique, Unité d’Explorations Métaboliques (CERAMM), CHU Lapeyronie, 34295 Montpellier Cédex 5, France. Tel.: +46 7338284; Fax: +46 7338986; Telex: CHR MONTP 480 766 F; E-mail: [email protected]. 0006-355X/10/$27.50 © 2010 – IOS Press and the authors. All rights reserved

96

J.-F. Brun et al. / Hemorheology and exercise

at bringing together all this information in order to provide an integrated view of both the physiological mechanisms and the functional consequences of the hemorheological alterations observed during and after exercise. This approach emphasizes the need to delineate hemorheological physiology from hemorheopathy and to integrate hemorheology and hemodynamics. Accordingly, in previous articles [9, 16] we proposed to describe exercise-induced hemorheological alterations as a triphasic phenomenon: acute, delayed and chronic effects. The first phase is the situation of chronic “hyperhydration-dilution status” observed in endurance athletes and resulting in low hematocrit (Hct). The second phase is an improvement in RBC properties induced by metabolic and hormonal changes (insulin sensitivity, growth hormone and IGF-I status). A third phase is related to endurance training-induced decrease in body fat, increase in muscular volume, and alterations of muscular processing of fuels. A fourth phase is the reversal of all these improvements due to the effects of excess exercise or excess training (e.g., the overtraining syndrome). 2. Acute effects of exercise: A short-term increase in whole blood viscosity 2.1. Increase in whole blood and plasma viscosity Both maximal and submaximal exercise, either of short or long duration, almost always increases blood viscosity due to a rise in plasma viscosity (ηp ) and hematocrit (Hct). In most cases (e.g., short acute exercise), these two hemorheological variables explain virtually all the observed increase in whole blood viscosity (ηB ) [13]. Curiously, one study failed to detect these changes [62] but, when looking at its protocol, one can notice that only post exercise (e.g., recovery) values are measured; accordingly, these short-term alterations (in ηp and Hct) were probably not detected, due to a rapid return to pre-exercise values [9,16]. This rise in ηp and Hct is sometimes interpreted as ‘hemoconcentration’ [78]. Such an explanation is incomplete since the observed modifications are due to at least five separate mechanisms: redistribution of red cells (RBC) in the vascular bed; splenocontraction that increases the number of RBC; enrichment of plasma in several proteins, coming presumably from lymphatics; a loss of water in the sweat for thermoregulation; entrapment of water into muscle cells [9,16,85]. 2.2. Positional fluid shifts alter blood viscosity It is important to stress that ηB increases when recumbent subjects become orthostatic, due to an increase in Hct and ηp associated with a rise in plasma proteins (fibrinogen and others). These positional fluid shifts should be taken into account in the analysis of exercise-induced alterations in water status. 2.3. Changes in RBC deformation and aggregation In most (but not all) exercise protocols there are also changes in the rheological properties of RBC. The most classical is a decrease in RBC deformability which is not a specific finding since it is also observed in most stressful events like labor, video film-induced emotional stress, and endogenous depression [8]. These effects are generally not found at exercise when RBC rheology is investigated after resuspension of cells in a buffer, indicating that they are mostly due to plasma factors rather than to intrinsic RBC properties [13,16]. Blood lactate, which experimentally shrinks the RBC and decreases their flexibility, is likely to explain in part this exercise-induced rigidification of RBC, as supported by positive correlations between lactate


97

concentrations and RBC rigidity at exercise. Interestingly, in one study we found a threshold value for this effect which became apparent only when blood lactate increased above 4 mmol · l−1 , i.e. a value which has been proposed to represent approximately the point where lactate induces acidosis [12]. However, some other studies suggest that lactate also exerts some effects at lower concentrations, either in vivo or in vitro [8,9,29]. Lactate is not the only factor explaining this rigidification. Traumatic damage of RBC due to their compression in the foot plantar circulation is likely to be important in sports like running, although this issue remains unclear [45]. Presumably, fluid status has also a major influence on RBC rheology during exercise, as suggested by the preventive effect of fluid intake on RBC rigidification [85]. There is also an acute increase in RBC aggregation and disaggregation shear rate [16]. Little is known about the mechanisms of these latter modifications which are not found in all exercise protocols, and are generally not detected by the most widely used technique, i.e. the light transmission analysis (Myrenne aggregometer) [9,16]. While pre-exercise fibrinogen concentrations are positively correlated to the extent of these changes in RBC aggregation [87], it is very likely that lactate does not play a role in this change in the aggregation properties of the cells [36]. Thus, the most important extracellular determinant of this event is fibrinogen. However, as discussed below, aggregation changes may also reflect leukocyte activation [17]. While RBC rigidity was generally found to be either increased or unchanged during exercise [9,13, 14,16,44,85], there was a surprising report of a decrease of this parameter, when assessed after exercise with ektacytometry (using the LORCA) [16]. This paradox has recently been explained by a study on highly trained athletes during a progressive exercise test conducted to VO2 max . In this case, paxadoxically, RBC rigidity was found to decrease [26]. Moreover, in vitro experiments [15] showed that lactate at concentrations ranging from 2 to 10 mM increased RBC deformability in such athletes while it classically decreased it in blood from sedentary subjects. Thus, in highly trained subjects, the exercise-induced increase in blood lactate does not rigidify the RBC as observed in sedentary subjects or in moderately trained ones (like soccer players [12]) but actually improves RBC deformability. 2.4. Leukocyte activation and oxidative stress Both white cell activation and oxidant stress [2,76,79] are likely to play an important role in the hemorheological effects of exercise. The marked increase in oxygen utilization that occurs during exercise results in production of free radicals by several sources, including the mitochondria and the white cells. Transient tissue hypoxia due to rapid accelerated consumption of oxygen in exercising muscles, and to inadequate oxygen supply at the pulmonary level in some trained people, has also been demonstrated and may lead to free radical formation. Whatever the mechanism, it is well established that oxidative stress during acute exercise is associated with a hemorheological impairment [2]. According to Ajmani [2], exercise-induced oxidative stress can also produce an increase in mean RBC volume and increase plasma fibrinogen levels, thus also increasing aggregation. However, until recently, little was known about the involvement of leukocyte activation in these rheological changes. The number of leukocytes increases after strenuous exercise. This effect is attributed to increased blood flow that recruits the leukocytes from the marginal pool and/or hormonal changes which are likely to be mediated by beta-2 adrenergic receptors. More interestingly, a decrease in filterability of white cells during exercise has been shown, reflecting some degree of leukocyte activation that may interact via several circulating factors with RBC properties. Transient hypoxia might also result in cytokine release and leukocyte activation. When leukocytes (especially polymorphonuclear leukocytes)

98


are activated, they reduce molecular oxygen enzymatically to generate metabolites such as superoxide anions, hydrogen peroxide or hydroxyl radicals [2]. These metabolites can injure the surrounding tissues by oxidative damage. RBC are vulnerable to oxidative damage, although they are equipped with antioxidant defense mechanisms. Recent studies have indicated that RBC in close contact with activated leukocytes can be damaged, at least in part by oxidative mechanisms, resulting in significant structural and functional alterations [17]. Temiz et al. [79] investigated leukocyte activation and RBC damage after exhaustive exercise in untrained rats. Significant increments in RBC membrane protein oxidation and lipid peroxidation, and decreased membrane enzyme activities were observed during early and late phases after the exercise episode. RBC transit times measured by a cell transit analyzer failed to indicate significant changes in RBC deformability, despite the biochemical evidences of oxidant damage. These alterations were positively correlated with increased leukocyte phagocytic activity [79]. 2.5. RBC rheological changes and exercise performance Theoretically, most of the rheologic changes reviewed above are likely to exert negative effects on exercise performance. This assumption is supported by experiments conducted on both healthy volunteers and rats under hypobaric hypoxic conditions [57]. Those studies have demonstrated that preventing the exercise-induced rise in RBC rigidity by ω3-fatty acids improves maximal aerobic capacity. Thus, in conditions of hypoxia, a rigidification of RBC may represent a limiting factor for muscle oxygen supply and thus impair performance. RBC stiffening has also been shown to augment the pulmonary hemodynamic response to hypoxia [39] (hypobaric simulation of altitude). This hypobaric environment increased ηB via a combination of factors (hypoxia, low pH and high values of blood lactate). These factors apparently generate rigid RBC, which presumably are associated with pulmonary hypertension. The hypobaric-induced pathophysiological phenomena were corrected by the calcium channel blocker, flunarizine. Exercise-induced changes in blood rheology have been reported to be related to the rating of perceived exertion. Increased Hct positively correlated with perceived exertion [11] and was hypothesized to represent a signal among the other well-characterized ones (e.g., heart rate, lactate, blood glucose) that are integrated at a conscious level to generate the feeling of exertion. An attractive hypothesis [16,44] was that an impairment of blood rheology may be involved in the cardiovascular risk of maximal exercise, together with hypercoagulation. In agreement with this hypothesis we recently reported the case of a 50 years old marathon runner who underwent a retinal central vein thrombosis after a marathon run [54]. And, during a subsequent standardized submaximal exercise test, the runner exhibited a disproportionate increase in ηB , Hct and RBC aggregation and disaggregation thresholds. While some of this post-marathon hyperviscosity may be due to the previous thrombotic event, these findings support our hypothesis stating that hemorheological abnormalities during the marathon may be the pathogenesis of the retinal thrombosis [54]. However, it is noteworthy that we observed the same rheological changes during submaximal exercise as those during maximal workloads [14]. This leads us to propose that simple changes in Hct, RBC rigidity and ηp are physiological adaptive modifications which occur during many kinds of exercises and do not imply a risk by themselves. Presumably such changes can be easily overcome by vasodilatation. In our opinion, the risk of maximal or exhausting workloads is more related to other factors, including extensive muscular damage, modifications of hemostasis and white cell activation. But, when vast changes in blood rheology are associated with hemostatic perturbations and inflammatory phenomena without sufficient vasodilation, the risks for medical complications are presumably no longer negligible. According to several authors [2,76,79],


99

these adverse rheological effects may be responsible in part for the enhanced incidence of myocardial infarction and sudden death associated with exercise.

3. Adaptive training-related improvements in blood rheology and the “hematocrit paradox” Cross-sectional studies of athletes compared to sedentary controls have repeatedly shown that athletes have a lower ηB . Both ηp and Hct are lower [47,53,64]. As summarized by Ernst et al. [46,47]: “the fitter the athlete the more fluid his blood”. Koenig et al. [61] studied the self-reported regular leisure time physical activity in connection with ηp data in 3522 men and women age 25–64 years from the Monica–Augsburg cohort. This population-based study shows that regular leisure physical activity is associated with a lower ηp across all age groups. During the hours following exercise, there is an increase in plasma volume [9,16] that represents a reversal of the acute hyperviscosity described above, resulting in an “autohemodilution” [49]. This autohemodilution results in a lower Hct that explains the negative correlations between Hct and fitness that are found in sportsmen [9,16]. Therefore, one should point out an important paradox concerning Hct in exercise physiology [18]. Since sports performance depends on the capacity of oxygen transport to the exercising skeletal muscles, it is not surprising to observe that performance may be increased by an artificial Hct augmentation achieved by either training in high altitude, blood transfusions, or injecting erythropoietin. Since erythropoietin has become largely available due to bioengineering, doping with this hormone has become extremely popular in most sports. However, the rationale for this doping procedure contrasts with the physiological information reported above. Under normal conditions, there is a strong negative correlation between Hct and fitness which is explained by the effect of regular training [9,16,17]. Comparisons between the extreme quintiles of Hct in athletes clearly illustrate this paradox [17]: athletes in the lowest quintile compared to those in the four other quintiles had a lower value of ηB and a higher fitness as reflected by their aerobic working capacity, their relative maximal power output, and their isometric adductor strength. By contrast, athletes in the highest quintile had higher viscosity and higher RBC disaggregation shear rate. On the whole, when Hct increases, there is a decrease in fitness and a higher score of overtraining. Fit athletes have a rather low Hct associated to other metabolic and ergometric improvements, while athletes with a high Hct are frequently overtrained and/or irondeficient, and their ηB (and RBC disaggregation shear rate) tends to be increased [17]. This issue has been further addressed by Gaudard and co-workers in our laboratory [55]. Maximal aerobic capacity defined in power units (Wmax ) exhibited a negative correlation with whole ηB . When it was expressed as the power corresponding to a fixed heart rate of 170 beats/min it exhibited a negative correlation with several hemorheological factors but a stepwise regression analysis only selected Hct as an independent determinant. Similarly the best determinant of the maximal oxygen consumption (VO2max ) was also Hct. Therefore, it is clear that fitness is associated with a low viscosity-low Hct pattern, while unfitness and overtraining (as discussed below) are associated with a mild hyperviscosity. Since this physiological truth disagrees with the popular belief: “the more RBC you have, the fitter you are”, we think that it is important to broadly disseminate this hemorheological paradigm of Hct among exercise physicians, coaches and athletes. In several sports a specific hemorheological pattern differs from that of most sports. Body-builders have no improvement of blood rheology after training [9], while a lower ηp increase in exercising rugbymen seems to be the most prominent characteristic of training and fitness [6]. The increase in plasma volume has been assumed to contribute to the body water pool to help prevent dehydration [9].

100


4. Chronic training-induced “hemorheologic fitness” 4.1. Exercise-induced metabolic and hormonal changes Around 1986, Dudaev et al. [40] reported the beneficial effects of 30 daily cycling sessions in male coronary patients compared to controls. Results indicated that regular exercise decreased RBC membrane levels of triglycerides, fibrinogen and cholesterol, and increased the level of high density lipoprotein cholesterol. Importantly, fibrinogen and triglyceride concentrations correlated with hemorheological and hemodynamic improvements, suggesting that exercise training triggered a beneficial adaptation of lipid metabolism; the altered metabolism promoted blood rheology improvement and, thereby, produced beneficial hemodynamic effects. A pivotal adaptive mechanism, the growth hormone-somatomedin axis, may be (directly) involved in the regulation of training-induced changes in blood rheology. For example, growth hormone-deficient adults have a low insulin sensitivity associated with an increased percentage of body fat with increased circulating lipid and fibrinogen levels, while trained sportsmen exhibit the opposite metabolic profile (low insulin sensitivity, decreased percentage of body fat, decreased circulating lipid and fibrinogen levels) and an increased function of this axis [9,16]. In sports that improve strength rather than endurance, RBC aggregation and deformability are improved without marked changes in body fluid distribution. The improvements in RBC rheology (aggregation and deformability) positively correlate with body composition (percentage of fat) and with the balance of substrate (carbohydrate and fat) oxidation at exercise. Rugby players have been extensively studied with regard to these correlates, which are also known to positively correlate with performance [6]. 4.2. Endurance training: Body fat and muscular volume In male rugby players the isometric adductor strength positively correlated with RBC flexibility, and RBC aggregation positively correlated with fat mass measured by bioelectrical impedance. The aerobic working capacity is negatively correlated with the increase in ηp during exercise, suggesting that strength is less important in Rugby players. This study shows that fat mass, even within a physiological range, is a determinant of RBC aggregation, suggesting that training-induced alterations in body composition play a role in the specific hemorheological profile of these athletes [6]. In female rugby players [22] two subgroups can be considered: forward (FW) and backward (BW). BW are leaner, due to a lower fat free mass and a lower fat mass; they had a faster running speed during field testing and a higher VO2max . Exercise calorimetry showed a higher ability to oxidize fat at exercise expressed by the (carbohydrate and fat) “point of crossover” and the point of maximal lipid oxidation. Comparison of hemorheological parameters indicates a higher ηB in FW explained by higher RBC rigidity while ηp , Hct and RBC aggregation were similar. Adductor strength correlated negatively with RBC aggregation, and handgrip strength correlated negatively with RBC aggregation. The ability to oxidize lipids at exercise correlated negatively with whole ηB and with RBC rigidity. Thus, ηB related negatively with fitness in rugbywomen, and, as previously observed in rugbymen, RBC rheology (deformability and aggregation) are the most important factors. The negative correlations found between RBC deformability and the ability to oxidize more lipids at exercise (i.e., a parameter of endurance performance) may be due to greater lipid oxidation by endurance training. This greater lipid oxidation then modifies lipid metabolism (as noted above) and improves RBC rheology, with possible performance enhancement [19].


101

4.3. Therapeutic exercise: Hormonal and metabolic changes In sedentary, markedly obese, insulin resistant patients submitted to a therapeutic training protocol, ηp improved more than other hemorheological factors, which reflects an appropriate plasma protein pattern (increased fibrinogen and lipoproteins, etc.) related to the metabolic disorders [71]. We recently demonstrated [41] that training in sedentary insulin resistant patients, induced significant improvements in body composition (loss of fat mass with a stability of fat free mass) associated with improvements in exercise capacity (as assessed with VO2max ). The metabolic improvements indicate a markedly increased ability to oxidize fat at exercise. The insignificant blood lipids and insulin sensitivity levels probably were not significantly improved probably due to the low sample size. As expected, blood rheology also improved, due to a decrease in ηp ; Hct, RBC deformability and RBC aggregation did not change. Thus, consistent with observations in athletes, the metabolic and ergometric improvements induced by training reduces ηp in sedentary insulin resistant patients also. But, at the low levels of training that were applied, ‘autohemodilution’ (as reflected by Hct) did not occur; nor did it improve RBC deformability or aggregation. The reliability of ηp as a simple and inexpensive marker of efficiency of training in insulin resistant patients should be further evaluated [41]. Eterovic [50], extending a work of Dintenfass [38], demonstrated that the values of ηp is explained by a combination of cholesterol, fibrinogen, triglycerides, Hct (reflecting the degree of dilution) and HDL, that may be combined in a predictive equation [50]. In addition, insulin sensitivity is positively correlated with fitness, probably because training improves both glucose and lipid processing by muscle, and body composition. This helps to explain why exercise is an effective treatment of the insulin resistance syndrome [41]. Exercise training improves the lipid pattern of patients suffering from this syndrome [3]. Controversy clouds the effect of training on fibrinogen; some studies failed to show a training effect. Brun et al. [9,16] hypothesized that this variability may be due to genetic subtypes of fibrinogen that respond differently to training. Training unequivocally reduces fibrinogen [16], an argument supported by negative correlations of fibrinogen with both fitness [5] and insulin sensitivity [72]. Literature evidence clearly documents that training decreases the plasma concentrations of metabolic constituents that impair blood rheology (second step of hemorheologic fitness); training also induces a body composition pattern characterized by a low percentage of fat (third step of hemorheologic fitness). All these modifications are likely to play a major role in the improvement of blood rheology induced by regular physical activity [9].

5. Beyond training: Overtraining The overtraining syndrome (OTS), a condition wherein an athlete’s performance deteriorates despite excessive training affects mainly endurance athletes. Thus, chronic fatigue, under performance, and an increased vulnerability to infection leads to recurrent infections. It is not yet known exactly how the stress of hard training and competition leads to the observed spectrum of symptoms. Psychological, endocrinogical, physiological and immunological factors all play a role in the failure to recover from exercise. OTS remains a controversial issue since its clinical symptoms are not specific [51]. Recently, the French consensus group on overtraining, Société Française de Médecine du Sport (SFMS), proposed a standardized questionnaire of early clinical symptoms of this elusive syndrome, allowing the calculation of a ‘score’ that may help to clinically classify sportsmen submitted to a heavy training program

102


[10]. This score appears to be positively correlated with markers of muscular damage (creatine kinase, myosin) or neuroendocrine dysfunction (somatotropic axis), and also with some hematological markers like ferritin. There appears to be a mild dehydration with increased Hct, serum Na+ and serum K+ . These markers presumably represent an excess plasma water loss. Concentrations of blood urea nitrogen and serum glutamic-oxaloacetic transaminase were also increased without evidence of a water-electrolyte deficiency syndrome, renal dysfunction or liver cell damage. We interpreted these findings as reflecting a persistent mild degree of dehydration and catabolic state noted after intense training [1,89]. In a recent investigation [89] of elite male athletes, we documented a positive relationship between the OTS score and an increased ηB due to elevated ηp and Hct; RBC deformability and RBC aggregation were unchanged. Therefore, the early signs of overtraining in elite sportsmen are associated with a hemorheological pattern that suggests some degree of reversal of the fitness-associated ‘autohemodilution’ discussed above. In addition, overtrained athletes are frequently iron depleted, a mechanism that may induce additional hemorheological alterations but is unlikely to explain the early hemorheological signs of the overtraining syndrome [60]. Current pathophysiologic concepts of OTS explain the observed mild hyperviscosity and mild hemoconcentration. Cytokines released by the “over-stressed” muscle now appear to be responsible for most of the symptoms [7]. According to this “cytokine hypothesis of overtraining”, high volume/intensity training, with insufficient rest, will produce muscle and/or skeletal and/or joint trauma. Circulating monocytes are then activated by injury-related cytokines, and in turn produce large quantities of proinflammatory IL-1beta and/or IL-6, and/or TNF-alpha, producing systemic inflammation. Elevated circulating cytokines then co-ordinate the whole-body response by: (a) communicating with the central nervous system and inducing a set of behaviors referred to as “sickness” behavior, which involves mood and behavior changes that support resolution of systemic inflammation; (b) adjusting liver function, to support the up-regulation of gluconeogenesis, as well as de novo synthesis of acute phase proteins, and a concomitant hypercatabolic state; and (c) impacting on immune function. Theoretically, OTS is viewed as the third stage of Selye’s general adaptation syndrome, with the focus being on recovery/survival and not adaptation, and is deemed to be “protective”, occurring in response to excessive physical/physiological stress. OTS appears to be a systemic low grade inflammatory condition which can be monitored by markers of inflammation including hemorheological ones [89], thus the concept holds the interest of hemorheologists. The hemorheological profile in OTS can also be relevant to some of the clinical symptomatology of overtraining. For instance, the feeling of having “heavy legs” (FHL) is one of the most commonly reported signs. Since FHL is also a sign of chronic venous insufficiency and can be corrected by hemorheologically-active drugs, we recently investigated whether the FHL is associated with a hemorheological profile. It appeared that FHL subjects complaining from OTS signs had higher ηp (1.43 ± 0.047 vs. 1.32 ± 0.02 mPa s, p < 0.05) and a higher RBC aggregation as measured with laser backscattering [88]. These findings suggest that the feeling of heavy legs in overtrained athletes is related to OTS-related hemorheological disturbances. We recently addressed the question of whether hemorheological measurements may provide a marker of the early stages of overtraining (overreaching). The logical candidate for this, ηp [90], appeared to be a rather specific, but a poorly sensitive predictor of overreaching (early OTS). Plasma viscosity had no diagnostic value for the overtraining syndrome according to the following calculations [90]: (a) prediction of overreaching: sensitivity 28.57%; specificity 90%; positive predictive value 80%; negative predictive value 47.4%; (b) prediction of chronic overtraining: sensitivity 2.70%; specificity 18.2%; positive predictive value; 10.00%; negative predictive value 5.3%.


103

6. Exercise-induced hemorheological changes influenced by nutrition and metabolism 6.1. Dehydration Nutritional factors influence hemorheological changes associated with exercise, with possible effects on muscular performance itself. A first important issue is water [16]: dehydration reduces blood and plasma volume, increases Hct, plasma osmolality, ηp and plasma protein, while it dramatically increases red blood cell aggregation proportional to a rise in plasma globulin. Accordingly, an adequate water intake almost completely prevents the increase in RBC rigidity induced by 1 h submaximal strenuous exercise [85]. 6.2. Caloric intake Caloric intake has also a pivotal importance. In athletes there is a tendency to consume fewer calories than expended and to avoid fats. This tendency may further compromise antioxidant mechanisms protecting RBC against the exercise stress [56]. Muscle glycogen depletion compromises exercise performance and also increases the stress. Glycogen stores can be protected by increased fat oxidation (glycogen sparing). The diets of athletes should be balanced so that total caloric intake equals expenditure, and so that the carbohydrates and fats utilized during exercise are replenished. Many athletes do not meet these criteria and have compromised glycogen or fat stores, have deficits in essential fats, and do not take sufficient micronutrients to support exercise performance, immune competence and antioxidant defence. From all nutritional variables optimal energy supply is considered as most vital for human performance. It is postulated that lack of energy homeostasis is a basic problem in the development of overtraining [7]. Most if not all clinical symptoms of this syndrome are directly or indirectly related to the physiological mechanisms of energy homeostasis. Lack of available energy has a much greater impact than exercise stress alone. Dietary insufficiencies should be compensated for by supplementation with nutrients, with care not to over compensate. In a recent study we addressed the hemorheological side of this problem [89]. In 41 elite athletes exercising 13±0.9 h/week a standardized nutritional questionnaire suitable for sports medicine was applied. We found a negative correlation between fibrinogen and protein intake. Accordingly, the RBC disaggregation threshold was also correlated negatively with protein intake. Caloric intake was positively correlated with RBC rigidity and negatively correlated with the RBC disaggregation threshold. Lipid intake was negatively correlated with the RBC disaggregation threshold. Carbohydrate intake was positively correlated with whole ηB and negatively to the Hct/viscosity ratio. Therefore, fibrinogen levels and RBC rheology (deformability and aggregation) exhibit correlations with nutritional status in athletes. Low protein intake appeared to be associated with (mildly) raised fibrinogen and aggregation and caloric restriction with lowered RBC deformability [56]. These data are thus consistent with the concept of a pivotal role of adequate nutrition to prevent the effects of the chronic exercise-induced inflammation in people on the edge of the overtraining syndrome [89]. 6.3. Pre-exercise feeding In contrast to the above described effects of nutritional status on baseline hemorheology of people subject to more than 10 h per week of training, there are few reported effects of a single pre-exercise feeding on blood rheological response to exercise [9,16]. In thirty-one male triathletes van der Brug [84] investigated the effects of different kinds of feedings on the hemorheological response to prolonged exercise. While exercise caused the expected increase in whole blood and ηp , Hct, and osmolality, and a

104


very small, but significant decrease in RBC deformability, irrespective of the food consumed, the intake of different amounts of carbohydrate had no influence on the hemorheological parameters. Therefore, if water supply is sufficient, carbohydrates have no major influence to the hemorheological response to exercise. By contrast, polyunsaturated fatty acids of the omega 3 family (ω3PUFA) increase exercise performance by improving RBC flexibility [8,57]. Toth and co-workers [81] have also found that ω3PUFA increases aerobic exercise capacity in patients suffering from ischemic heart disease and hyperlipoproteinemia. The exercise performance improvement in cardiac compromised patients was related to an improvement in hemorheology (ηB ) and circulation (decrease in total peripheral resistance). A recent well-conducted study [68] challenged this literature since it shows that a 3 week of fish oil supplementation (6 g/day), without or with vitamin E (300 IU/day), has no effect on either RBC rheology or exercise performance. Eating a breakfast before exercise has a significant influence on the hemorheological response to exercise [15]. After a 495 kcal breakfast (8.9% proteins, 27.3% lipids; 63.9 % glucids, i.e. mimicking a “French breakfast”), the increase in RBC rigidity that occurs at fast is prevented, while ηp is higher and increases more when subjects were fed than when they were fasting. Therefore, the standardized (French-style) breakfast modifies the rheologic response to exercise, a response found also in other studies [9,16]. 6.4. Mineral and trace elements Recent studies have underlined the importance of mineral and trace element status in sports hemorheology. Zinc, which in vitro increases the deformability of artificially hardened RBC [8], is frequently low in the serum of athletes, a situation reflecting some degree of deficiency. Athletes with low serum zinc have a higher ηB and an impairment in RBC deformability that is associated with a decrease in performance. Experimentally, a double blind randomized trial of oral zinc supply in healthy volunteers improves ηB [59] while the effects on performance are not significant. Zinc seems also to reduce RBC aggregation both in vitro and in vivo [8,59]. Another mineral, iron, is frequently lacking in athletes. Even without anemia, a low iron level is likely to impair performance, although there is still some controversy concerning the benefit of iron supplementation in athletes. In elite athletes plasma ferritin has been observed to be negatively correlated with ηB [60]. Subjects with low ferritin levels, suggesting mild iron deficiency, thus exhibit a higher ηB explained as being due to a higher ηp while Hct and RBC rigidity are unchanged. RBC aggregation is also higher in iron-deficient subjects [60]. These data suggest that mild iron deficiency as commonly seen in athletes, before anemia occurs, is associated with an increase in ηp and RBC aggregation, together with an increased subjective feeling of exercise overload. Finally, body builder studies show evidence of hemorheological abnormalities, suggested to be at risk for thromboembolic phenomena because of increased ηB [45]. Those abnormalities could reflect the use of diet, exercise and ergogenic drug regimens, which are common among competitive athletes. 6.5. Alcohol consumption Recently the issue of alcohol consumption and blood rheology during exercise has been extensively studied (see review [44]). According to El-Sayed [44], alcohol continues to be the most frequently consumed drug among athletes and appears to evoke detrimental effects on exercise performance capacity.


105

Alcohol consumption decreases the use of glucose and amino acids by skeletal muscles, adversely affecting energy supply during exercise. Although moderate alcohol consumption has favourable effects on blood coagulation and fibrinolysis, occasional and chronic alcohol consumption is usually linked with unfavourable alterations in platelet aggregation and function and may be associated with plateletrelated thrombus formation. Concerning the effects of alcohol on the rheological properties of blood evidence suggests that alcohol use following exercise is associated with unfavourable changes in the main determinants of ηB [43]. In moderately active young men, alcohol (0.7 g/kg body mass) was given 1 h after exercise. During recovery, while the increase in Hct post-exercise returned to the baseline level in a similar manner in control and alcohol trials, ηp and plasma fibrinogen remained significantly high during recovery in the alcohol trial compared to the controls. Authors conclude that the consumption of alcohol after exercise delays the normal return of ηp , plasma fibrinogen to resting baseline levels during recovery.

Part II. Influence of hemorheological parameters on exercise performance 1. Is blood rheology a physiological determinant of exercise performance? We previously reported the well-known classical observations that differentiated between high-fitness and low-fitness groups: the high-fitness group had a lower ηp [9,16]. RBC flexibility correlated positively with adductor isometric strength [6]. Positive correlations of blood fluidity with aerobic working capacity W170 , time of endurance until exhaustion [16], maximal exercise test derived VO2max [9], have been demonstrated. In these correlations the best correlate appears to be ηp . In addition, blood lactate response is negatively correlated to ηB [9,16,86], and the power intensity at which blood lactate reaches 4 mmol · l−1 (the so-called 4 mmol · l−1 lactate threshold) is negatively correlated with erythocyte aggregation [9]. Studies on patients with sickle cell trait (SCT) demonstrated a reduced capacity for prolonged competitive exercise under hypobaric hypoxia, which we interpreted to have resulted from reduced RBC flexibility [56]. In addition, Connes et al. [33] recently reported lower aerobic capacity in SCT carriers performing submaximal exercise at sea level, a finding probably attributed to reduced RBC deformability, increased RBC aggregation, reduced blood fluidity and increased vascular adhesion phenomenon observed in that population in comparison with healthy subjects [31,34,65,67,82]. On the other hand, when RBC fluidity is improved by ω3 fatty acid supply (see below) there is an increase in VO2max under hypobaric hypoxia, suggesting that a prevention of RBC rigidification during exercise improves aerobic capacity in these conditions. On theoretical grounds, increased blood fluidity improves O2 delivery to muscle in trained exercising individuals [35]. However, this hypothesis remains unclear. There are several biological indicators of fitness, which are relevant to different kinds of exercise. The most popular is maximal oxygen uptake (VO2max ), which has not been widely studied in connection with blood rheology despite the unclear theoretical link noted above. As reported previously, VO2max correlated negatively with ηB due to its negative correlation with ηp [16,17]. Another important phenomenon of fitness involves the delay of hyperlactacidemia (so-called ‘anaerobic threshold’ or ‘lactate threshold’) to higher workloads [16]. In three separate studies [9,16,86], we observed that ηB and RBC aggregation were positively correlated with lactate accumulation into blood during exercise [9]. The possible meaning of the relationships between resting blood fluidity and lactate response will be discussed later.

106


1.1. Correlation between maximal oxygen consumption and hematocrit Since maximal oxygen consumption (VO2max ) and aerobic working capacity (W170 ) are highly correlated and both parameters are indices of aerobic exercising capacity, they are affected by the same ηB determinants [9]. We found ηp to be the best statistical determinant of these two measurements of aerobic performance [9,17]. However, Hct also correlates negatively with aerobic performance [17,55], reflecting the beneficial effect of autohemodilution. The maximal oxygen consumption (VO2max ) reflects the body’s ability to increase O2 transfer from the surrounding atmosphere to muscles and depends on several steps. The limiting step (noted below) varies among sportsmen. VO2max equals the maximal value of Q · CaO2 , Q being cardiac output and CaO2 the O2 content of blood. If one applies the Hagen–Poiseuille law [77], the formula, VO2max = Q · CaO2 , can be written as a function of Hct, Φ and viscosity, η. The formula then becomes VO2max = constant × (Φ/η) × (ΔP/Z), with ΔP being the drop in blood pressure and Z being vascular hindrance (dilation/contraction). Thus the value (Φ/η) should be a limiting factor for VO2max . In experimental studies, VO2max did not correlate with Φ/η, but, instead, negatively correlated with Φ (Hct), a determinant of ηB that is negatively related to fitness [16,17]. This suggests that the negatively correlated VO2max -Hct relationship reflects fitness since fitness accompanies autohemodilution that lowers Hct and results in increased cardiac output [55]. Curiously, Deem et al. [37] using rabbits found that a decrease in Hct from 30% to 11% by hemodilution improved PaO2 and oxygen exchange (alveolar to arterial O2 difference, A-aΔO2 ), and enhanced the alveolar ventilation-to-perfusion (VA/Q) matching. Franck et al. [52] also studied the effects of pulmonary Hct on pulmonary oxygen diffusion capacity for O2 (DLO2 ) by using a two-dimensional finite-element model developed to represent the sheet-flow characteristics of pulmonary capillaries. Although DLO2 increased as the Hct increased, DLO2 reached a plateau near a Hct of 35% [52]. Together, the results from Deem et al. [37] and Franck et al. [52] indicate that there is an optimum value for Hct that allows the highest pulmonary oxygen diffusion capacity. It should be emphasized that in [37], alveolar oxygen exchange improved as the Hct level went from low normal to anemic, while, in [52], oxygen exchange (diffusion capacity) improved as Hct went from anemic to a low normal level. This suggests a biphasic effect of Hct. If Hct is too low, RBC cannot uptake enough O2 to maintain normal arterial oxygen pressure but, if Hct is too high, the same phenomenon is possible due to competition between RBC for O2 influx [52]. Moreover, by testing the effects of RBC shape and deformability on DLO2 and resistance to flow in rabbit lungs, Betticher et al. [4] have demonstrated that impairment in RBC deformability decreased DLO2 and increased blood flow resistance whereas improvement in RBC deformability caused the opposite. Optimal RBC deformability allows a more homogeneous RBC distribution within pulmonary capillaries and other microvessels, which might explain the beneficial effects of RBC deformability on pulmonary oxygen diffusion [4]. At a given Hct, Hsia et al. [58] have reported that both local Hct and RBC distribution affected membrane diffusing capacity for carbon monoxide (DMCO) and diffusive uptake for CO (DLCO). At a given Hct within a capillary segment, a more homogeneous distribution of RBC could increase DLCO by up to 33% [58]. Caillaud et al. [21] proposed that a combination of high Hct level and impaired RBC rheology, particularly during exercise, might cause excessive pulmonary capillary wall tension causing capillary rupture, capillary fluid leakage leading to interstitial edema promoting an oxygen diffusion limitation. Together, these studies suggest that hemorheology impairment alters oxygen diffusion from alveoli to capillaries that might contribute to decreased exercise performance. We further speculate that blood rheology impairment such as RBC rigidity or an increased degree of RBC aggregation, could lead to vessel blockade


107

and complications in any organ: hypertension (pulmonary) or ischemia, degeneration, necrosis (systemic organs). 1.2. Links between blood rheology and cardiac function Classical heart physiology assumes, after Hill (1923) [77] that maximal cardiac output (Qmax ), the product of heart rate (HR) by stroke volume (SV), constitutes the primary factor explaining individual differences in maximal oxygen consumption (VO2max ), a main determinant of endurance performance. Textbooks generally state that 70–85% of the limitation in VO2max is linked to Qmax . Besides, several studies showed that training induced improvement in VO2max resulted from an increase in maximal Qc rather than a widening of A-aΔO2 . However, any increase in ηB is likely to increase cardiac afterload and thus decrease maximal stroke volume. Recently, an investigation on blood rheology parameters and heart rate variability (HRV) at rest in athletes with sickle cell trait showed a loss of parasympathetic activity associated with hyperviscosity syndrome [32]. These results support our contention that blood rheology and cardiac function could be linked. Sickle cell trait athletes usually have a lower aerobic capacity than normal athletes suggesting that blood rheology impairment and cardiac efficiency are limiting factors for endurance performance [63]. The most important hemorheological factor affecting oxygen carrying capacity is Hct. Doping studies have shown that reinfusion of 900–1350 ml blood elevated both the oxygen carrying capacity and VO2max of the athlete by 4–9%. Several studies reported the effects of recombinant human erythropoietin (rHuEPO) on exercise performance. One reported that VO2max increased 7–10% after 4 weeks of repeated injections; the rHuEPO effect was associated with an 8–10% increase in Hct and hemoglobin concentration (Hb) [30]. In another study, Connes [27] also reported that 4 weeks of rHuEPO treatment led to faster oxygen uptake kinetics at the onset of submaximal cycling exercise. These studies [27,30] demonstrated that Hct, if it does not exceed the optimal Hct range, plays a significant role in endurance performance even if the increased Hct might also increase ηB and partly counteract the beneficial effect of an increased Hct on oxygen carrying capacity. 1.3. Blood rheology and oxygen delivery to the tissues Several investigators reported significant positive correlations between blood fluidity and indices of physical fitness such as: time-to-exhaustion, aerobic working capacity W170 or VO2max [9,16]. As underlined by Brun et al. [9,16], on a theoretical point of view, increased blood fluidity may improve O2 delivery to muscle during exercise in trained individuals. This may explain why VO2max is negatively correlated to ηB , due to negative correlation of VO2max with ηp [9,16]. Moreover, Charm et al. [23] reported reduced ηp among joggers compared with non-joggers; accordingly, ηp could be a determinant of aerobic performance. In capillaries, including muscle capillaries, RBC flow in a single file and are surrounded by a plasma layer. Indeed, if ηp is too high, this could alter this normal blood flow structuring in capillaries, increase blood flow resistance, disorganize normal RBC orientation inside microvessels and thus impair O2 delivery to tissues. Hct could also play a role in O2 supply to muscles. Although higher Hct values are responsible for a greater blood oxygen carrying capacity, several studies reported negative correlations between Hct and aerobic performance [9,16]. Hct beyond the optimal range increases blood flow resistance in capillaries and interferes with tissue O2 supply [9,16]. RBC properties, i.e. RBC deformability and aggregation, may also improve or impair O2 supply to tissues. As we have discussed earlier, a greater RBC deformability allows a more homogeneous RBC distribution within pulmonary capillaries and other microvessels that might allow adequate oxygen diffusion at the pulmonary level

108


or at the muscular level. Parthasarathi et al. [70] studied the effect of reduced RBC deformability on microvessel recruitment attendant to a reduction in tissue PO2 in rat cremaster muscle and found that impairment in RBC deformability may adversely affect capillary recruitment and delivery of oxygen to tissue. Studies of the rheological behaviour of RBC in the capillary network clearly demonstrated that capillary flux and velocity are strongly dependent on the ability of RBC to deform on entry into the capillaries. RBC deformability impairment adversely affects capillary perfusion. Parthasarathi et al. [70] observed that less deformable RBC were excluded from the bulk of the capillaries and followed more centralized pathways within the microvasculature that could alter O2 delivery to tissues. Currently, microcirculatory effects of RBC aggregation are a matter of controversy. Yet, the majority of evidence holds that intensely aggregated RBC could be responsible for blood hyperviscosity, blood flow stasis and increased blood flow resistance, which might alter O2 delivery to muscles resulting in a metabolic shift toward anaerobiosis and limit endurance performance. This part of the discussion will be more developed later in this review. 1.4. Blood rheology and the central governor theory Classic theory in exercise physiology considers that maximal exercise performance is limited by the cardiorespiratory system as a result of specific metabolic changes in the exercising skeletal muscle (i.e., peripheral fatigue). According to the classical model of Hill [66], peripheral fatigue occurs only after the onset of heart fatigue or failure. Thus, this hypothesis predicts that it is the heart, not the skeletal muscle that it is at risk of anaerobiosis or ischemia during heavy exercise. As reported by Noakes et al. [66], Hill proposed the existence of a “governor” in either the heart or brain to limit heart work when myocardial ischaemia developed and to prevent fatal cardiac events during intense exercise. Noakes et al. [66] reviewed and reported several experimental results which are compatible with the theory that the reduced maximal heart rate, maximal stroke volume and maximal cardiac output during acute and chronic hypoxia are due to central regulation. The central governor theory proposes that afferent sensory information from the heart, but also perhaps from the brain and respiratory muscles, informs the brain of any threat that hypoxia or ischemia may develop in those organs. In response, the central governor acts via the motor cortex to reduce the efferent neural activation of the exercising muscles, thereby reducing the mass of muscle that can be recruited, and hence, reducing the exercise intensity that can be sustained [66]. By limiting the muscle mass that can be activated, the central governor limits the peripheral peak VO2 to a level that will not induce hypoxia in any of the vital organs. The brain tissue is very vulnerable to hypoxia and its high metabolic rate needs a high oxygen supply. Because cerebral cortex capillaries are very narrow, any blood rheological impairment (increased local Hct, intense RBC aggregation or loss of RBC deformability) alters blood flow structuring and hence adequate blood oxygenation. Blood flow structuring disturbance in cerebral cortex capillaries results in a slow down to a full stop of the blood flow, despite a preserved arteriolovenular pressure difference. Indeed, we propose that blood rheological impairment induced by exercise and the subsequent metabolic changes (such as lactic acidosis and oxidative stress) could impede blood flow in the brain. A resulting oxygen deprivation might reduce the efferent neural activation of the exercising muscles to limit metabolic changes and hence to limit further blood rheological impairment that might cause severe brain ischemia. Although afferent sensory information from the heart and other organs inform the brain to stop or reduce exercise, as proposed by Noakes et al. [66], hemorheology (in relation with metabolic changes induced by exercise) could also play a great role in this process. In addition, Khaled et al. [59] by studying the effects of oral zinc supplementation on blood rheological behaviour and Borg’s rating of perceived exertion (RPE) during


109

exercise in healthy subjects, found that blood rheology and RPE were improved after zinc supplementation suggesting that hemorheology might influence exercise tolerance and hence, endurance performance [59]. Finally, Brun et al. [11] also support this hypothesis by reporting a significant positive correlation between Hct and perceived exertion in exercising healthy subjects. 2. Viscosity parameters as factors of blood flow homogeneity or heterogeneity during exercise Schmid–Schönbein has pointed out that the classical concepts of total flow and total peripheral resistance cannot adequately explain the relationships between viscosity factors and circulation, and proposed to describe them with contemporary paradigms used in the non-linear sciences in general, namely the “percolation theory” [73–75]. According to this approach, ηB does not behave in vivo as a continuous variable, since it rapidly oscillates between two opposite states: ηB is very low under high shear-high flow conditions and dramatically increases under low shear-low flow conditions. These changes in ηB are induced by local flow conditions and by viscosity factors like RBC deformability, RBC aggregation, local Hct, ηp and fibrinogen concentration. All these factors govern the transition from a highly fluid status towards a ‘near-solid’ status of blood. Evidence that these concepts are relevant for muscular physiology has been provided by ultrasound Doppler in the femoral artery of awake human subjects [72–74]. These measurements show that, in the exercising muscle, the local fluidity of blood remains very high regardless of the level of systemic Hct. The only viscosity factor that may increase local resistance under these conditions is ηp . By contrast, at rest, Hct, RBC rigidity and RBC aggregation also promote local hyperviscosity and impair local blood flow [17,55,72–74]. 2.1. RBC aggregation and blood lactate during exercise These newer concepts provide an explanation for several clinical observations in exercise hemorheology. First, several papers from our group have demonstrated a link between RBC aggregation at baseline and the rise in blood lactate during exercise [86]. These papers suggest that RBC aggregation may influence muscular lactate metabolism. As experimentally shown by Vicaut [91], increased RBC aggregation may impede microcirculatory blood flow in muscles. Although aggregation is beneficial to some extent for microvascular perfusion [69], its increase, even within a physiological range, might impair aerobic metabolism in muscle, resulting in higher blood lactate. If this assumption is correct, lactate accumulation, that was classically described as an “anaerobic process”, but is now explained by a shift in the balance of fuel oxidations from fats to carbohydrates (CHO), could be also influenced by RBC aggregates impeding or blocking microcirculatory vessels resulting in foci of tissue oxygen deprivation. While the microcirculatory effects of RBC aggregation are a matter of controversy, experiments by Johnson and co-workers [20], suggest that RBC aggregation represents 60% of blood flow resistance in venules of the cat gastrocnemius muscle. RBC aggregation could, therefore, be the major modifier of venous resistance in skeletal muscle [20]. Muscle hypoxia experiments [92] in exercising anemic (25% Hct reduction) dogs demonstrated a disproportionate increase in lactate accumulation. This increase in lactate was associated with higher muscular glucose consumption and with an increase in glucagon, norepinephrine, epinephrine and cortisol while insulin and free fatty acids were unchanged. The investigators suggested that the availability of oxygen to tissues should be considered. In humans suffering from peripheral obliterative arterial disease, RBC aggregation is negatively correlated with transcutaneous oxygen pressure, further supporting the concept that aggregation impairs oxygen supply to tissues [42].

110


In fact, a possible explanation for the relationship between rheology and lactate blood accumulation, rather than a hemorheological ‘Pasteur-like’ effect (so-called ‘anaerobiosis’), may be an influence of RBC aggregation on lactate removal, as demonstrated by modelling of postexercise lactate kinetics [86]. The mathematical analysis of postexercise lactate allows a reasonably close estimate of lactate production by muscles (γ1 ) and lactate clearance (γ2 ). In a sample of subjects exhibiting a wide range of γ2 (from 2 to 7.7 × 10−2 min−1 ) we observed that postexercise RBC aggregation index Myrenne “M1 ” (measured at VO2max ) was the only hemorheological parameter negatively correlated to γ2 . Thus microcirculatory adaptations influenced by RBC aggregation may influence lactate disposal and clearance (as reflected by γ2 ), adding its effect to that of the balance between carbohydrates and fat oxidation which is the major determinant of blood lactate concentrations at exercise in physiological conditions [86]. Another oxygen-related parameter which can be influenced by blood rheology could be the oxygen equivalent of the watt. This parameter is theoretically close to 10.3 ml · W−1 but is higher in sedentary subjects when they exhibit a low fat-free mass or a high waist-to-hip ratio [86]. Interestingly, it is increased in individuals with elevated ηB parameters [86] and the improvement of these parameters by prostaglandin E1, naftidrofuryl or hemodilution partially corrects it. According to Wolff and Witte, the measurement of this waste of oxygen during submaximal steady state workloads may allow a direct clinical determination of microcirculatory performance in involved muscle tissue as a function of ηB [86]. 2.2. Exercise-induced hypoxemia Finally, an amazing issue in current exercise physiology is exercise-induced hypoxemia (EIH), i.e., the arterial oxygen pressure decreases during intense exercise. This situation has some similarities with exercise-induced pulmonary hemorrhage (EIPH) in horses that is frequently observed during competitive races [21]. In both situations a ventilation/perfusion inequality and/or pulmonary diffusing capacity limitation may occur as a result of an interstitial pulmonary edema. In horses, a host of literature has dealt with investigations into a possible role for blood rheology in EIPH, but a clear demonstration of a role of blood rheology in this process is still lacking [21]. In humans, episodes of pulmonary hemorrhage following ultra marathon races have been reported, supporting the hypothesis of some pathophysiological similarities between EIPH and EIH. Actually, pulmonary capillary pressure during maximal exercise does not reach the high levels observed in horses, and the high capacity of shear-dependent rheofluidification found in horses despite their high RBC aggregation [21] indicates that horse and human rheology are extremely different. However several recent lines of evidence support a role for blood rheology in the pathophysiology of EIH. First, a comparison between hypoxemic and non hypoxemic athletes shows that exercise increases ηB to higher levels in EIH athletes. The greater increase in ηB in these athletes was attributed to the lack of RBC deformability improvement during exercise whereas RBC became surprisingly more deformable in non-EIH athletes. In addition, improvement of RBC deformability by dietary poly-unsaturated fatty acids reduces hypoxemia in athletes at maximal exercise [21]. We can hypothesize that there is a training-induced adaptation in high level athletes that apparently decreases the exercise-induced hyperviscosity, as shown by in vitro experiments on the effect of lactate on RBC [29] and by the paradoxical lack of hyperviscosity at exercise sometimes reported in athletes [21,25]. In EIH-prone athletes, this mechanism may be blunted and hyperviscosity may thus result, at maximal exercise, in hypoxemia [25,28]. It is possible that training promotes an adaptive mechanism in RBC that allows them to cope with lactate and oxidative stress. Such an adaptation might maintain optimal RBC deformability and allow a higher endurance performance [28].


111

3. Exercise as a “hemorheological therapy” In sedentary patients, regular exercise also improves blood rheology [16,41,45]. In fact, an improvement in blood fluidity can be induced by regular physical exercise regardless of whether the blood rheology was normal or abnormal at baseline. Thus, regular exercise might be a way of therapeutically increasing blood flow in ischemic vascular diseases. Training compensates not only for the potential damage risk factors represent but also for the physical stress provoked by vigorous exercise. A large body of literature on the therapeutic effects of exercise in peripheral obliterative arterial disease shows that the therapeutic effect of training in this disease may be largely explained by rheological improvements [48]. Non-insulin dependent diabetes represents an extreme example of the insulin-resistance syndrome in which all the metabolic abnormalities are overtly expressed. Exercise has been proposed as a preventive treatment for this disease, mostly explained by a decrease in muscular fat and glucose processing. In these patients there is also a link between unfitness and cardiovascular risk [83]. Recently, Aloulou et al. [3] reported on the hemorheological effects of low intensity endurance training in sedentary patients suffering from the metabolic syndrome. Twenty four patients were submitted to a 2 months targeted training designed for increasing exercise lipid oxidation. Variations of ηB at high shear rate were explained by two statistically independent determinants: Hct and RBC rigidity. A reduction in RBC rigidity appeared to reflect weight loss and a decrease in LDL cholesterol. ηp was related to cholesterol and its traininginduced changes are related to those of VO2max but not to lipid oxidation. RBC aggregation reflected both the circulating lipids (cholesterol and its subfractions HDL and LDL) and the ability to oxidize lipids at exercise. Factors associated with a post-training decrease in aggregation were weight loss and more precisely decrease in fat mass, improvement in lipid oxidation, rise in HDL-cholesterol and decrease in fibrinogen. On the whole, the hemorheologically improved decrease in ηp was attributed to an increase in cardiorespiratory fitness (VO2max ). RBC aggregation and deformability exhibited clear relationships with lipid metabolism. This study shows that the return to “metabolic fitness” in subjects prone to develop metabolic diseases and diabetes, is closely associated with a parallel improvement in blood rheology, aerobic capacity and metabolism. The effects of cardiac rehabilitation and exercise training on blood rheology in patients with coronary heart disease (CHD) have been recently investigated by Church et al. [24]. After rehabilitation, patients with CHD had reductions in ηp (from 1.85 ± 0.18 to 1.77 ± 0.11 mPa s) and in whole ηB . Thus cardiac rehabilitation improves blood rheology in patients with CHD. However, it is not yet known whether these improvements actually contribute to the increased functional capacity and reduced morbidity and mortality that are associated with participation in cardiac rehabilitation and exercise.

4. Conclusions A major historical issue in Sports Medicine is the research of accurate “markers of fitness”, which may help to predict performance and/or under performance, and thus avoid some unexpected health problems in athletes. Although less studied than other aspects of Sports Medicine and Exercise Physiology, Exercise Hemorheology is likely to provide some interesting possibilities fitting in with this scope. As explained above, the training-induced improvement in blood rheology provides an integrated reflection not only of “autohemodilution” (i.e., training-related decrease in Hct) which is clearly a sign of fitness, but also some in plasma and RBC rheology that reflect metabolic improvements and thus in what we

112


propose to term “hemorheologic fitness”. A first unresolved question is thus the exact value in sports medicine of those alterations as “integrated markers of fitness”. The issue of overtraining is another field where clinical assessment is difficult and where there is absolutely no consensus on biological signs. Interestingly, our findings on the hemorheological signs of overtraining fits in with the more recent pathophysiological concepts of this disease, which involve exercise-induced muscle histologic microlesions leading to post exercise inflammation in order to govern muscle repair. Close to this topic is the issue of so-called “sports anemia”: a quite confusing question that hemorheologists may help to clarify. Hemorheologists observe that: (a) low Hct in athletes generally does not mean anemia but hemodilution; (b) iron deficiency, that initially impairs blood rheology in athletes, even before anemia occurs, and may be associated with a paradoxical rise in Hct and viscosity. That leads to what we proposed to call “the paradox of Hct in athletes”. On the whole, both physiological and clinical fields of sports medicine are full of questions still to be resolved by hemorheologists. This means that hemorheology may have some clinical applications in this domain of biological sciences.

References [1] A. Aissa Benhaddad, D. Bouix, S. Khaled et al., Early hemorheologic aspects of overtraining in elite athletes, Clin. Hemorheol. Microcirc. 20 (1999), 117–125. [2] R.S. Ajmani, J.F. Fleg and J.M. Rifkind, Role of oxidative stress on hemorheological and biochemical changes induced during treadmill exercise, J. Mal. Vasc. 25(Suppl. B) (2000), 144–145. [3] I. Aloulou, E. Varlet-Marie, J. Mercier and J.F. Brun, Hemorheologic effects of low intensity endurance training in sedentary patients suffering from the metabolic syndrome, Clin. Hemorheol. Microcirc. 35 (2006), 333–339. [4] D.C. Betticher, W.H. Reinhart and J. Geiser, Effect of RBC shape and deformability on pulmonary O2 diffusing capacity and resistance to flow in rabbit lungs, J. Appl. Physiol. 78 (1995), 778–783. [5] D. Bouix, C. Peyreigne, E. Raynaud et al., Fibrinogen is negatively correlated with aerobic working capacity in football players, Clin. Hemorheol. Microcirc. 19 (1998), 219 (abstract). [6] D. Bouix, C. Peyreigne, E. Raynaud et al., Relationships among body composition, hemorheology and exercise performance in rugbymen, Clin. Hemorheol. Microcirc. 19 (1998), 245–254. [7] V. Bricout, M. Guinot, M. Duclos et al., Apports des examens biologiques dans le diagnostic du surentrainement, Sci. Sports 21 (2006), 319–350. [8] J.F. Brun, Hormones, metabolism and body composition as major determinants of blood rheology: Potential pathophysiological meaning, Clin. Hemorheol. Microcirc. 26 (2002), 63–79. [9] J.F. Brun, Exercise hemorheology as a three acts play with metabolic actors: is it of clinical relevance?, Clin. Hemorheol. Microcirc. 26 (2002), 155–174. [10] J.F. Brun, Le surentraînement: à la recherche d’un outil d’évaluation standardisé utilisable en routine, Sci. Sports 18 (2003), 282–286. [11] J.F. Brun, C. Lagoueyte, C. Fédou and A. Orsetti, A correlation between hematocrit increase and perceived exertion in exercising healthy subjects, Rev. Port. Hemorheol. 4 (1990), 51–64. [12] J.F. Brun, C. Fons, E. Raynaud et al., Influence of circulating lactate on blood rheology during exercise in professional football players, Rev. Port. Hemorheol. 5 (1991), 219–229. [13] J.F. Brun, C. Fons, I. Supparo et al., Could exercise-induced increase in blood viscosity at high shear rate be entirely explained by hematocrit and plasma viscosity changes?, Clin. Hemorheol. 13 (1993), 187–199. [14] J.F. Brun, J.P. Micallef and A. Orsetti, Hemorheologic effects of light prolonged exercise, Clin. Hemorheol. 14 (1994), 807–818. [15] J.F. Brun, I. Supparo and A. Orsetti, Effects of a standardized breakfast compared to fasting on the hemorheologic responses to submaximal exercise, Clin. Hemorheol. 15 (1995), 213–220. [16] J.F. Brun, S. Khaled, E. Raynaud, D. Bouix et al., Triphasic effects of exercise on blood rheology: which relevance to physiology and pathophysiology?, Clin. Hemorheol. Microcirc. 19 (1998), 89–104. [17] J.F. Brun, C. Bouchahda, D. Chaze, A. Aïssa Benhaddad et al., The paradox of hematocrit in exercise physiology: which is the ‘normal’ range from an hemorheologist’s viewpoint?, Clin. Hemorheol. Microcirc. 22 (2000), 287–303.


113

[18] J.F. Brun, C. Bouchahda, A. Aissa Benhaddad, C. Sagnes et al., Aspects hémorhéologiques de l’activation leucoplaquettaire dans les pathologies athéromateuses, applications cliniques, J. Mal. Vasc. 25 (2000), 349–355. [19] J.F. Brun, E. Varlet-Marie, D. Cassan, J. Manetta and J. Mercier, Blood fluidity is related to the ability to oxidize lipids at exercise, Clin. Hemorheol. Microcirc. 30 (2004), 339–343. [20] M. Cabel, H.J. Meiselman, A.S. Popel and P.C. Johnson, Contribution of red cell aggregation to venous vascular resistance in skeletal muscle, Am. J. Physiol. Heart Circ. Physiol. 272 (1997), H1020–H1032. [21] C. Caillaud, P. Connes, D. Bouix and J. Mercier, Does haemorheology explain the paradox of hypoxemia during exercise in elite athletes or thoroughbred horses?, Clin. Hemorheol. Microcirc. 26 (2002), 175–181. [22] D. Cassan, J.F. Brun, J.P. Micallef, M.J. Alemany et al., Relationships between hemorheological and metabolic adaptation to exercise in female rugby players: importance of erythrocyte rheology, in: 3rd Intern. Conf. Haemorheology, A. Muravyov, I. Tikhomirova, V. Yakusevitch, L.G. Zaitsev et al., eds, Yaroslavl, Russia, July 29–31, 2001, p. 58. [23] S.E. Charm, H. Paz and G.S. Kurland, Reduced plasma viscosity among joggers compared with non-joggers, Biorheology 15 (1979), 185–191. [24] T.S. Church, C.J. Lavie, R.V. Milani and G.S. Kirby, Improvements in blood rheology after cardiac rehabilitation and exercise training in patients with coronary heart disease, Am. Heart J. 143 (2002), 349–355. [25] P. Connes, Exercise-induced hypoxemia as a consequence of hemorheological alterations?, Biorheology 42 (2005), 92 (abstract). [26] P. Connes, C. Caillaud, D. Bouix, P. Kippelen et al., Red cell rigidity paradoxically decreases during maximal exercise in endurance athletes unless they are prone to exercise-induced hypoxaemia, J. Mal. Vasc. 25(Suppl. B) (2000), 165. [27] P. Connes, S. Perrey, A. Varray, C. Prefaut and C. Caillaud, Faster oxygen uptake kinetics at the onset of submaximal cycling exercise following 4 weeks recombinant human erythropoietin (r-HuEPO) treatment, Pflügers Arch. 447 (2003), 231–238. [28] P. Connes, D. Bouix, F. Durand, P. Kippelen et al., Is hemoglobin desaturation related to blood viscosity in athletes during exercise?, Int. J. Sports Med. 25 (2004), 569–574. [29] P. Connes, D. Bouix, G. Py, C. Préfaut et al., Opposite effects of in vitro lactate on erythrocyte deformability in athletes and untrained subjects, Clin. Hemorheol. Microcirc. 31 (2004), 311–318. [30] P. Connes, C. Caillaud, J. Mercier, D. Bouix and J.F. Casties, Injections of recombinant human erythropoietin increases lactate influx into erythrocytes, J. Appl. Physiol. 97 (2004), 326–332. [31] P. Connes, F. Sara, M.D. Hardy-Dessources, M. Etienne-Julan and O. Hue, Does higher red blood cell (RBC) lactate transporter activity explain impaired RBC deformability in sickle cell trait?, Jpn. J. Physiol. 55 (2005), 385–387. [32] P. Connes, C. Martin, J.C. Barthelemy, G. Monchanin et al., Nocturnal autonomic nervous system activity impairment in sickle cell trait carriers, Clin. Physiol. Funct. Imaging 26 (2006), 87–91. [33] P. Connes, G. Monchanin, S. Perrey, D. Wouassi et al., Oxygen uptake kinetics during heavy submaximal exercise: effect of sickle cell trait with or without alpha-thalassemia, Int. J. Sports Med. 27 (2006), 517–525. [34] P. Connes, F. Sara, M.D. Hardy-Dessources, L. Marlin et al., Effects of short supramaximal exercise on hemorheology in sickle cell trait carriers, Eur. J. Appl. Physiol. 97 (2006), 143–150. [35] P. Connes, O. Yalcin, O. Baskurt, J.F. Brun and M. Hardeman, In health and in a normoxic environment, VO2 max is/is not limited primarily by cardiac output and locomotor muscle blood flow, J. Appl. Physiol. 100 (2006), 2099. [36] P. Connes, C. Caillaud, G. Py, J. Mercier et al., Maximal exercise and lactate do not change red blood cell aggregation in well trained athletes, Clin. Hemorheol. Microcirc. 36 (2007), 319–326. [37] S. Deem, R.G. Hedges, S. McKinney, N.L. Polissar et al., Mechanisms of improvement in pulmonary gas exchange during isovolemic hemodilution, J. Appl. Physiol. 87 (1999), 132–141. [38] L. Dintenfass, Influence of plasma proteins on the in vivo and in vitro rheological properties of blood, Clin. Hemorheol. 5 (1985), 191–206. [39] M.P. Doyle and B.R. Walker, Stiffened erythrocytes augment the pulmonary hemodynamic response to hypoxia, J. Appl. Physiol. 69 (1990), 1270–1275. [40] V.A. Dudaev, I.V. Diukov, E.F. Andreev, V.V. Borodkin and M. Al’-Murabak, Effect of physical training on lipid metabolism and the rheologic properties of the blood of patients with ischemic heart disease, Kardiologiia 26 (1986), 55–60. [41] M. Dumortier, A. Perez-Martin, E. Pierrisnard, J. Mercier and J.F. Brun, Regular exercise (3 × 45 min/week) decreases plasma viscosity in sedentary obese, insulin resistant patients parallel to an improvement in fitness and a shift in substrate oxidation balance, Clin. Hemorheol. Microcirc. 26 (2002), 219–229. [42] C. Dupuy-Fons, J.F. Brun, F. Pellerin, J.C. Laborde et al., Relationships between blood rheology and transcutaneous oxygen pressure in peripheral occlusive arterial disease, Clin. Hemorheol. 15 (1995), 191–199. [43] M.S. El-Sayed, Adverse effects of alcohol ingestion post exercise on blood rheological variables during recovery, Clin. Hemorheol. Microcirc. 24 (2001), 227–232. [44] M.S. El-Sayed, N. Ali and Z. El-Sayed Ali, Interaction between alcohol and exercise: physiological and haematological implications, Sports Med. 35 (2005), 257–269.

114 [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74]

J.-F. Brun et al. / Hemorheology and exercise M.S. El-Sayed, N. Ali and Z. El-Sayed Ali, Haemorheology in exercise and training, Sports Med. 35 (2005), 649–670. E. Ernst, Changes in blood rheology produced by exercise, J. Am. Med. Assoc. 253 (1985), 2962–2963. E. Ernst, Influence of regular physical activity on blood rheology, Eur. Heart J. 8(Suppl. G) (1987), 59–62. E. Ernst and A. Matrai, Intermittent claudication, exercise, and blood rheology, Circulation 76 (1987), 1110–1112. E. Ernst, L. Daburger and T. Saradeth, The kinetics of blood rheology during and after prolonged standardized exercise, Clin. Hemorheol. 11 (1991), 429–439. D. Eterovic, I. Pintaric, J. Tocilj and Z. Reiner, Determinants of plasma viscosity in primary hyperlipoproteinemias, Clin. Hemorheol. 15 (1995), 841–850. A. Favre-Juvin, et les membres du groupe surentraînement de la SFMS (X. Bigard, J.F. Brun, V. Bricout, J.L. Chatard et al.), Le surentraînement: place de l’interrogatoire dans le diagnostic, Sci. Sports 20 (2005), 314–316. A.O. Frank, C.J. Chuong and R.L. Johnson, A finite-element model of oxygen diffusion in the pulmonary capillaries, J. Appl. Physiol. 82 (1997), 2036–2044. G. Galea and R.J. Davidson, Hemorheology of marathon running, Int. J. Sports Med. 6 (1985), 136–138. A. Gaudard, E. Varlet-Marie, J.F. Monnier, C. Janbon et al., Exercise-induced central retinal vein thrombosis: possible involvement of hemorheological disturbances. A case report, Clin. Hemorheol. Microcirc. 26 (2002), 115–122. A. Gaudard, E. Varlet-Marie, F. Bressolle, J. Mercier and J.F. Brun, Hemorheological correlates of fitness and unfitness in athletes: moving beyond the apparent “paradox of hematocrit”?, Clin. Hemorheol. Microcirc. 28 (2003), 161–173. A. Gaudard, E. Varlet-Marie, F. Bressolle, J. Mercier and J.F. Brun, Nutrition as a determinant of blood rheology and fibrinogen in athletes, Clin. Hemorheol. Microcirc. 30 (2004), 1–8. C.Y. Guézennec, J.F. Nadaud, P. Satabin, C. Léger and P. Laffargue, Influence of polyunsaturated fatty acid diet on the hemorheological response to physical exercise in hypoxia, Int. J. Sports Med. 10 (1989), 286–291. C.C. Hsia, C.J. Chuong and R.L. Johnson Jr., Red cell distortion and conceptual basis of diffusing capacity estimates: finite element analysis, J. Appl. Physiol. 83 (1997), 1397–1404. S. Khaled, J.F. Brun, G. Cassanas, L. Bardet and A. Orsetti, Effects of zinc supplementation on blood rheology during exercise, Clin. Hemorheol. Microcirc. 20 (1997), 1–10. S. Khaled, J.F. Brun, A. Wagner, J. Mercier et al., Increased blood viscosity in iron-depleted elite athletes, Clin. Hemorheol. Microcirc. 18 (1998), 309–318. W. Koenig, M. Sund, A. Doring and E. Ernst, Leisure-time physical activity but not work-related physical activity is associated with decreased plasma viscosity. Results from a large population sample, Circulation 95 (1997), 335–341. C. Lacombe, C. Bucherer, J.C. Lelievre, E. Perreaut et al., Effets hémorhéologiques consécutifs a des exercices physiques controlés sur ergocyle (Hemorheologic effects following ergometric-controlled physical exercise), J. Mal. Vasc. 16 (1991), 79–82. D. Le Gallais, C. Prefaut, J. Mercier, A. Bile et al., Sickle cell trait as a limiting factor for high-level performance in a semi-marathon, Int. J. Sports Med. 15 (1994), 399–402. R.L. Letcher, T.G. Pickering, S. Chien and J.H. Laragh, Effects of exercise on plasma viscosity in athletes and sedentary normal subjects, Clin. Cardiol. 4 (1981), 172–179. G. Monchanin, L.D. Serpero, P. Connes, J. Tripette et al., Effects of a progressive and maximal exercise on plasma levels of adhesion molecules in athletes with sickle cell trait with or without {alpha}-thalassemia, J. Appl. Physiol. 102 (2007), 169–173. T.D. Noakes, J.E. Peltonen and H.K. Rusko, Evidence that a central governor regulates exercise performance during acute hypoxia and hyperoxia, J. Exp. Biol. 204 (2001), 3225–3234. P.C. Obiefuna, Rouleaux formation in sickle cell traits, J. Trop. Med. Hyg. 94 (1991), 42–44. G.S. Oostenbrug, R.P. Mensink, M.R. Hardeman, T. De Vries et al., Exercise performance, red blood cell deformability, and lipid peroxidation: effects of fish oil and vitamin E, J. Appl. Physiol. 83 (1997), 746–752. K. Osterloh, P. Gaehtgens and A.R. Pries, Determination of microvascular flow pattern formation in vivo, Am. J. Physiol. Heart Circ. Physiol. 278 (2000), H1142–H1152. K. Parthasarathi and H.H. Lipowsky, Capillary recruitment in response to tissue hypoxia and its dependence on red blood cell deformability, Am. J. Physiol. 277(6 Pt. 2) (1999), H2145–H2157. A. Perez-Martin, M. Dumortier, E. Pierrisnard, E. Raynaud et al., Multivariate analysis of relationships between insulin sensitivity and blood rheology: is plasma viscosity a marker of insulin resistance?, Clin. Hemorheol. Microcirc. 25 (2001), 91–103. E. Raynaud, A. Pérez-Martin, J.F. Brun, A. Aissa Benhaddad et al., Relationships between fibrinogen and insulin resistance, Atherosclerosis 150 (2000), 365–370. H. Schmid-Schönbein, Wege des Infarzierens. Neue Erkenntinisse durch die in-vivo-Rheoskopie (Paths to infarction. New findings through in-vivo-rheoscopy), Internist 42 (2001), 1196–1200. H. Schmid-Schönbein, L. Desai-Röhrig, H. Röhrig, A. Scheffler et al., Synergetics of blood flow in exercising skeletal muscle in man: cooperative effects of rheology, neurogenic vasoconstriction and mechanical effects of muscle pumps, Biorheology 39 (2002), S10-2 (abstract).


115

[75] H. Schmid-Schönbein, M. Foerster, H. Heidtmann, J. Hektor et al., Percolation theory as the rational basis of clinical hemorheology: the role of “abnormal” flow behaviour of blood elements on the homogeneity of perfusion, gas exchange and metabolism, Biorheology 39 (2002), (abstract). [76] U.K. Senturk, O. Yalcin, F. Gunduz, O. Kuru et al., Effect of antioxidant vitamin treatment on the time course of hematological and hemorheological alterations after an exhausting exercise episode in human subjects, J. Appl. Physiol. 98 (2005), 1272–1279. [77] D.E. Silverthorn, Human Physiology: An Integrated Approach, Pearson Education, San Francisco, CA, 2009. [78] T. Teillet, P. Pilardeau, P. Libercier, J. Vaysse and J.L. Hermant, Variation des protéines plasmatiques pendant un exercice de courte durée, Sci. Sports 6 (1991), 173–178. [79] A. Temiz, O.K. Baskurt, C. Pekcetin, F. Kandemir and A. Gure, Leukocyte activation, oxidant stress and red blood cell properties after acute, exhausting exercise in rats, Clin. Hemorheol. Microcirc. 22 (2000), 253–259. [80] P. Thiriet, J.Y. Le Hesran, D. Wouassi, E. Bitanga et al., Sickle cell trait performance in a prolonged race at high altitude, Med. Sci. Sports Exerc. 26 (1994), 914–918. [81] K. Toth, E. Ernst, T. Habon, I. Horvath et al., Hemorheological and hemodynamical effects of fish oil (Ameu) in patients with ischemic heart disease and hyperlipoproteinemia, Clin. Hemorheol. 15 (1995), 867–875. [82] J. Tripette, M.D. Hardy-Dessources, F. Sara, M. Montout-Hedreville, C. Saint-Martin, O. Hue and P. Connes, Does repeated and heavy exercise impair blood rheology in carriers of sickle cell trait?, Clin. J. Sport Med. 17 (2007), 465–470. [83] J. Tuomilehto, J. Lindstrom, J.G. Eriksson, T.T. Valle et al., Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance, N. Engl. J. Med. 344 (2001), 1343–1350. [84] C. van der Brug, H. Peters, M. Hardeman, G. Schep and W. Mosterd, Hemorheological response to prolonged exercise: no effects of different kinds of feedings, Int. J. Sports Med. 16 (1995), 231–237. [85] H. Vandewalle, C. Lacombe, J.C. Lelièvre and C. Poirot, Blood viscosity after a 1-h submaximal exercise with and without drinking, Int. J. Sports Med. 9 (1988), 104–107. [86] E. Varlet-Marie and J.F. Brun, Reciprocal relationships between blood lactate and hemorheology in athletes: another hemorheologic paradox?, Clin. Hemorheol. Microcirc. 30 (2004), 331–337. [87] E. Varlet-Marie, A. Gaudard, J.F. Monnier, J.P. Micallef et al., Reduction of red blood cell disaggregability during submaximal exercise: correlation with fibrinogen, Clin. Hemorheol. Microcirc. 28 (2003), 139–149. [88] E. Varlet-Marie, A. Gaudard, F. Bressolle, J. Mercier and J.F. Brun, Is the feeling of heavy legs in overtrained athletes related to impaired hemorheology?, Clin. Hemorheol. Microcirc. 28 (2003), 151–159. [89] E. Varlet-Marie, F. Maso, G. Lac and J.F. Brun, Hemorheological disturbances in the overtraining syndrome, Clin. Hemorheol. Microcirc. 30 (2004), 211–218. [90] E. Varlet-Marie, J. Mercier and J.F. Brun, Is plasma viscosity a predictor of overtraining in athletes?, Clin. Hemorheol. Microcirc. 35 (2006), 329–332. [91] E. Vicaut, X. Hou, L. Decuypere, A. Taccoen and M. Duvelleroy, Red blood cell aggregation and microcirculation in rat cremaster muscle, Int. J. Microcirc. 14 (1994), 14–21. [92] D.H. Wasserman, H.L.A. Lickley and M. Vranic, Effect of hematocrit reduction on hormonal and metabolic responses to exercise, J. Appl. Physiol. 58 (1985), 1257–1262.

Copyright of Biorheology is the property of IOS Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.