We are using cookies to give you the best experience on our website. Show
Reach out to us and we will get back to you shortly
Thanks for contacting us. One of our team will be in touch shortly. sustained exercise places high demands on body thermoregulatory mechanisms, especially in conditions of high ambient temperature (Ta) and humidity. For competitive athletes and active individuals, the effective dispersal of the heat load generated by contracting muscle bears particular importance. Failure of mechanisms to effectively remove body heat during strenuous exercise would result in substantial decrements in physical performance while posing risk for eventual circulatory collapse, brain dysfunction, and generalized organ failure (21). The dynamics of heat flux during sustained exercise can be briefly summarized (34): heat liberated by contracting muscle fibers is transferred away by its surrounding blood flow, resulting in an increase in core body temperature [estimated as rectal temperature (Tre)]. In response, hypothalamic control centers and peripheral receptors trigger compensatory cooling mechanisms, principally 1) cutaneous vasodilatation to augment skin blood flow (SBF) for convective heat loss to the surrounding air and 2) increased rate of sweating (SR) via sympathetic cholinergic stimulation to dissipate heat by evaporation at the skin-air interface. The magnitude of convective heat loss is governed by the local skin-air temperature gradient as well as adequacy of cutaneous blood flow. This means of heat dispersal is thus most effective in conditions of moderate environmental temperature, and it becomes less so as Ta rises. Heat loss by evaporation is directly related to both rate of sweat production and the skin-air water vapor pressure gradient. In high Ta, then, body heat loss is effected primarily through sweating, particularly in conditions of low ambient humidity. Many factors influence this basic scheme, including level of aerobic fitness, clothing, energy substrate utilization, body composition, and wind velocity. Highly critical, however, is the state of body hydration and plasma volume, because increasing levels of dehydration incurred via sweating during exercise are reflected in decreases in cardiac output, decrements in SR, and rise in Tre (48). In summary, then, thermoregulatory efficacy during exercise is most closely linked to 1) adequacy of circulatory responses, 2) rate of sweat production, and 3) maintenance of body fluid volume, all in response to exercise intensity (19, 47). When these thermoregulatory patterns were initially studied in children, certain maturational differences became evident (5–7). Most particularly, the SR of prepubertal boys during exercise was observed to be significantly less, by almost one-half, than that of young men. Recognized, too, were features unique to the pediatric population that might be expected to negatively influence body temperature regulation during exercise, including a greater body surface area-to-mass ratio (BSA/M), a higher metabolic demand relative to body mass (lower exercise economy) during weight-bearing exercise, slow acclimatization to heat, and a reduced cardiac output at a given metabolic rate compared with adults. Based on these observations, prepubertal children have been traditionally considered “less effective thermoregulators than adults,” at increased risk for exercise-induced heat illness as well as with diminished tolerance for exercise in hot climatic conditions (7, 11). As a consequence of this concern, particular guidelines for fluid intake and sports activities in the heat have been formulated for child athletes (2, 6, 36). This view of physically active youth as an at-risk group for heat injury and exercise intolerance in the heat has been based on a group of early studies that often lacked direct child-adult comparisons, failed to match subjects by relative exercise intensity, or involved extremes of Ta. More recently, a number of studies have assessed maturational differences in thermoregulatory responses to exercise in the heat while avoiding these methodological difficulties (25, 40, 44, 50). This review will reassess thermoregulation during exercise in the heat by children in light of these more recent reports and examine implications of any maturational differences in respect to their effects on physical performance and risk of heat injury. Initial sections will examine evidence for child-adult differences in various factors that bear on thermal regulation during exercise. Following this, the discussion will focus on the “so what?” factor: what evidence exists that any physiological differences between children and adults in thermal regulation during exercise can be translated to maturational differences in core temperature response, exercise tolerance, and risk of heat injury? In this discussion, references to data in “adults” will indicate postpubertal young and middle-aged individuals (excluding elderly subjects, who possess their own unique thermoregulatory responses to exercise). All reported Ta will expressed as dry bulb. MUSCLE HEAT PRODUCTION AND ENERGY ECONOMYThe energy required to perform a given amount of muscle work [as assessed by oxygen uptake (V̇o2), adjusted for substrate utilization] at a given work rate during cycle exercise is similar in children and adults (46). That is, muscular efficiency is not influenced by biological maturation. This implies that children posses qualitatively and quantitatively the same energetics of intracellular energy transfer and contraction coupling as do adults. More pertinent to the present discussion, absolute amounts of heat production can be expected to be similar when a child and adult are performing equal muscular work. When walking or running on a treadmill at the same speed and elevation, children typically demonstrate a higher V̇o2 relative to body mass (i.e., lower exercise economy) than adults (43, 52). For instance, Unnithan and Eston (52) reported mean V̇o2 values of 42.4 and 36.2 ml·kg−1·min−1 during treadmill running at 8.0 km/h in 9- to 10-yr-old boys and 18- to 25-yr-old men, respectively. While a number of explanations have been offered to explain child-adult differences in running economy, the most likely explanation lies in the observation that at a given treadmill speed, the young child is exercising at a greater relative exercise intensity [i.e., percentage of peak work, reflected as V̇o2/maximal oxygen uptake (V̇o2max)] than the adult. For example, Cureton et al. (12) presented cross-sectional treadmill running data (8 km/h) in three groups of boys ages 7–10, 11–14, and 15- yr (12). Relative intensities for the three groups were 74.6, 63.2, and 61.7% of V̇o2max, respectively. In the study by Unnithan and Eston cited above, the relative intensity for the boys was 67.3% and for the men 55.7% of V̇o2max. Another way of viewing this is that metabolic cost (i.e., heat load) per kilogram of body mass during locomotion is most appropriately defined in respect to each stride (49), and at any given treadmill speed, stride frequency (reflecting differences in leg length) is greater in children than adults (43, 52). If relative intensity of exercise is equated in children and adults by adjusting treadmill speed to leg length (and thus stride frequency), adult-child differences in exercise economy disappear (29). Similarly, when energy expenditure during submaximal running at the same treadmill speed is expressed as V̇o2 per kilogram per stride, no differences in economy are seen between children and adults (15, 29, 43). In the popular viewpoint, children suffer an increased heat burden during exercise because metabolic rate per body mass in this age group is greater than the adult exercising at the same speed (5–7). Yet, that adult-child levels of metabolic expenditure and heat production during exercise are instead more appropriately considered in terms of relative exercise intensity is indicated by the following: 1) thermoregulatory mechanisms respond to heat production during exercise in respect to relative (i.e., %V̇o2max) rather than absolute workloads (19, 47), and 2) in the real world of sports play, children do not exercise at the same work rates as adults. Instead, they participate at lower levels of physical activity commensurate with body size (muscle bulk, leg length, etc.). In summary, when exercising at a work rate that is commensurate with body size, heat production per body mass is expected to be equal in prepubertal children and adults. By this argument, then, children should not disadvantaged by excessive heat production during exercise relative their body mass compared with adults. BSA/MThe processes of sweat evaporation and convection eliminate body heat at the skin surface. Thus individuals with a greater body surface area (i.e., heat radiator) relative to body mass (reflecting muscle mass, the heat generator) should be expected to expedite heat dispersion, reduce heat storage, and facilitate thermoregulation during exercise. By geometric principles, BSA/M is inversely related to body mass. Small animals have higher values than large ones, and children are no exception. The BSA/M ratio of the average 8 yr old is almost 50% greater than that of the young adult. However, above age 13 yr, the values in children and adults do not differ appreciably (41). The greater BSA/M in children should be expected to be advantageous to their heat loss and thermal homeostasis during exercise compared with adults. A number of studies performed in adult subjects have supported this concept (16, 30). For example, Marino et al. (30) showed that BSA/M was negatively correlated with heat storage during running in highly trained distance runners, even in Ta of 35°C. Other authors have criticized this concept as being oversimplistic (22, 36). Havenith (22) argued that confounding factors such as body composition, fitness, sex, and type and duration of exercise influence the effect of BSA/M on heat loss and that in certain conditions a low BSA/M may even be associated with lower rather than higher heat strain during exercise. It has been suggested that in very hot climatic conditions, when Ta exceeds that of the skin (e.g., with a reversal of the skin-to-air temperature gradient), a higher BSA/M would act disadvantageously to absorb body heat from the environment (5–7). Several studies have presented skin-to-air temperature gradients for children performing steady-state submaximal exercise in hot climatic conditions (14, 18, 25, 39, 40, 53). These permit an estimation of the level of Ta that might be necessary before a reversal of temperature gradient would become a handicap to children with their higher BSA/M. Average gradients for individual studies are plotted in Fig. 1 against Ta. These data suggest that a reversal of skin-air temperature gradient would only be expected to occur at Ta exceeding 38°C (100°F). This implies that higher BSA/M would only become a potential liability in extremely hot climatic conditions, which are not encountered during sports play.  Fig. 1.Studies reporting skin-air temperature gradient vs. ambient temperature during exercise in the heat in children. Each dot represents a published study, cited by number in parentheses. There exists no experimental evidence, however, that this actually occurs. Two studies have revealed no significant group differences when comparing skin temperatures of children and adults performing exercise in extremely high Ta (41 and 48°C) at the same relative intensity (14, 25). In these reports, then, children and adults had similar skin-air gradients when the BSA/M was on the average 34% greater in the youth. The change in Tre with exercise was similar between groups in both studies. These data do not support a higher BSA/M as a liability for children at high ambient temperatures. Reversal of skin-air gradient during exercise occurs only at marked extremes of Ta and appears to be similar in prepubertal children and adults without consequence to dispersal of body heat relative to BSA/M. SWEAT PRODUCTIONA diminished SR is the most characteristic feature that distinguishes thermoregulatory response to exercise of prepubertal boys from that of adult men. An early study by Kawahata (28) in resting subjects in conditions of 45°C and 97% relative humidity (RH) indicated that whole body SR of 9-yr-old boys (455 ml·m−2·h−1) was approximately one-half that of men 21–27 yr old (815 ml·m−2·h−1) (28). Subsequent studies during both low-grade and highly intense sustained exercise have confirmed the magnitude of this age effect (7, 25, 33, 50). These investigations suggest that the greatest gains in rate of SR during exercise (in male individuals) coincides with the age of puberty (3, 17). Falk et al. (17) compared sweating responses in groups of pre-, mid- and late-pubertal boys who cycled at 50% V̇o2max for two 20-min bouts in 42°C, 20% RH (17). Average SR for the three groups were 4.95 ± 0.23, 5.79 ± 0.20, and 6.70 ± 0.42 ml·min−1·m−2, respectively. Size of sweat drops (drop area) increased with greater pubertal stage. In a similar study, Meyer et al. (33) reported that SR of boys aged 9.1 ± 1.4 and 11.7 ± 0.7 yr while cycling in the heat were similar but were only approximately one-half that of men aged 21.4 ± 3.2 yr. Based on such data, Inoue et al. (26) suggested that the lower SR in boys was related to lack of male hormonal effects that occur at the time of puberty. Supporting this, the studies assessing sex-related SR at rest have indicated no differences in young girls from either prepubertal boys or adult women (28, 38). This sex-related pubertal effect implies that variations in androgenic stimulation are responsible for maturational differences in SR. Still, the role of testosterone in regulation of sweat production has not been firmly established (7). Inoue et al. (26) noted that the frequency in pulsativity of sweat production relative to rate of sweat flow was generally lower in boys compared with men, with no differences in SR in respect to mean body temperature. They concluded that the lower SR in boys relative to young men reflected underdevelopment of peripheral sweating mechanisms rather than any impairment of central-driven sudomotor function. Regional body differences in SR are observed in children as well as adults. But Shibasaki et al. (50) found that local SR values in chest, back, and forearm sites were significantly lower in boys than young men. The maturational differences in SR among males during exercise is not related to sweat gland number, which is fixed by age 3 yr. Instead, the diminished flow rate in prepubertal subjects reflects a lower sweat output per gland as well as a decreased sensitivity of sweat gland output in response to a given Ta (5, 6). Inbar et al. (25) described sweating responses to three 20-min bouts of exercise at 50% V̇o2max in prepubertal and young adult male subjects. SR was 327 ± 11 and 445 ± 30 ml·m−2·h−1 in the two groups, respectively. Sweat production relative to change in Tre was greater in the adults (771 ± 104 vs. 385 ± 26 ml·°C·h−1) as was SR per gland (11.0 ± 0.7 vs. 2.8 ± 0.2 ml/h per gland). Similar differences findings were observed with increasing pubertal stage by Falk et al. (17). Does the lower SR confer a thermoregulatory advantage or disadvantage to young boys exercising in the heat compared with their adult counterparts? The answer is not altogether clear. Compared with adults, children would be expected to be at decreased risk for sweating-induced dehydration, with its adverse effects on heat storage, fitness, and risk of heat injury. On the other hand, evaporative sweat is the principal means of heat dispersal during exercise in hot climatic conditions when a diminishing skin-air temperature gradient limits convective heat loss. Consequently, children might be expected to demonstrate greater increases in heat load and Tre when exercising in conditions of high Ta. As will be discussed in sections that follow, the issue may be moot, because neither of these positive or negative outcomes are, in fact, observed. Davies et al. (13) estimated that average heat loss by evaporation, expressed as percentage of metabolic heat, averaged 65% in young men compared with 51% in children while running at 68% V̇o2max in thermoneutral conditions (13). Subsequent studies suggested, however, that the diminished sweat capacity in young boys does not necessarily imply lower evaporative heat loss during exercise (17, 25). In their comparison of prepubertal boys and young adult men, Inbar et al. (25) estimated that evaporative skin heat losses normalized to body mass were greater in the prepubertal subjects (8.10 ± 0.13 vs. 6.80 ± 0.13 W/kg). They calculated that sweating efficiency (evaporative loss relative to total body sweat) was significantly greater in the boys (0.69 ± 0.02 vs. 0.60 ± 0.04 W·ml−1·h−1). They considered that these findings might be explained by 1) children having smaller, more diffusely spaced drops, which could result in higher evaporative cooling, and/or 2) the possibility that larger drops in adults are more likely to coalesce, providing less cooling. CONVECTIVE HEAT LOSSEarly investigators who observed low SR in boys expected that children might compensate by demonstrating higher levels of convective heat loss during exercise compared with adults. In fact, studies that have examined changes in SBF as a surrogate marker of convective heat loss have generally found greater values in children than adults. Shibasaki et al. (50) compared regional SBF by laser-Doppler flowmetry in prepubertal boys and young men cycling at 40% V̇o2max for 45 min in Ta of 30°C and 45% RH. The boys demonstrated greater increases in flow at the left chest and back, but values were lower than the adults on the left forearm. Falk et al. (18) found that forearm SBF (by venous occlusion plethysmography) in prepubertal boys both at rest and during exercise in the heat was twice that of postpubertal adolescents. Martin et al. (31) described age differences in maximal skin vascular conductance (FVCmax) at rest in the left forearm that had been sprayed with hot water to create a skin temperature of 42°C. Blood flow was measured venous occlusion plethysmography, and maximal flow was divided by mean arterial blood pressure to obtain FVCmax. FVCmax was inversely related to age, with steepest rate of decline between ages 5 and 17 yr. Mean values at age 10 and 30 yr were 30 and 21 ml·100 ml−1·min−1·100 mmHg−1, respectively. These limited data indicate a higher SBF rate, greater skin vascular conductance, and, by inference, larger rates of relative convective heat loss during exercise in children compared with adults. The mechanisms that might account for these developmental differences remain obscure. DEHYDRATIONThe low SR of children during exercise in the heat might be expected to beneficially limit body fluid losses. However, there exists no evidence that their levels of dehydration during such exercise are any different from that of adults. The only direct child-adult comparison of hydration status during exercise without fluid replacement is that of Drinkwater et al. (14). They found that percent weight loss, rate of weight loss, and change in plasma volume during walking in 28, 35, and 48°C conditions were similar in premenarcheal girls and young adult women. In a review of six child-adult comparison studies, Meyer and Bar-Or (32) estimated levels of hypohydration that would have occurred if fluid replacement had not been given (taking in account body weight and SR). They concluded from these data that the magnitude of expected dehydration during exercise in the heat in these reports was similar in the children and adults. Based on findings in two separate studies, Bar-Or (8, 9) suggested that at any given level of dehydration, a child's Tre will rise more rapidly than that of an adult (8, 9). Eleven 12-yr-old boys cycled with fluid intake at 45% V̇o2max at 39°C and 45% RH (8). On the average, Tre rose by 0.28°C for each 1% increase in weight loss. In the second study, four young adults (2 men, 2 women) performed treadmill walking without fluid replacement in 38–39°C Ta. The rise in Tre for each 1% increase in weight loss was 0.15°C (9). Degree of dehydration during exercise is dictated by fluid intake as well as SR. No experimental data are available regarding maturational differences in thirst drive relative to dehydration thresholds. Limited information suggests, however, that voluntary drinking and dehydration during exercise in the heat is similar in children and adults. The eight boys and eight men studied by Rowland et al. (44) consumed an average of 5.1 and 5.3 ml/kg, respectively, when drinking cool water ad libitum during cycling in 31°C and 50% RH for 30 min. The boys in the study by Bar-Or et al. (8) cited above reach dehydration levels of 1–2% after cycling for 80–100 min. Voluntary drinking amounted to 66% of fluid loss. The authors noted that comparisons with studies in adults was difficult because of different climatic conditions, exercise protocols, and type of ingested fluid. Illustrating this, Rivera Brown et al. (39) found that voluntary intake replaced fluid loss of 78% with water intake but over 100% with intake of a glucose and electrolyte solution in 12 boys cycling in 33°C 58% RH with ad libitum drinking. CIRCULATORY RESPONSESThe cardiovascular system bears a heavy burden during exercise in the heat. While satisfying the blood flow demands of muscle metabolism, circulation must be provided to shunt heat away from contracting skeletal muscle and body core, augment cutaneous flow for convective heat loss, and provide a fluid supply for sweat production. The adequacy of these circulatory responses is defined largely by body fluid content and blood volume. When dehydration ensues from sweat loss during exercise, stroke volume, cardiac output, and blood pressure fall concurrent with increase in Tre and decline in work performance (20). In subjects who remain euhydrated by fluid intake, however, cardiovascular function is typically maintained (44, 48). Findings from early studies suggested to investigators that children's circulatory responses to exercise were inferior to those of adults. Specifically, children demonstrated values of cardiac output at any given level of absolute V̇o2 that clustered at the lower limits of the normal range in adult subjects (45, 51). This “hypokinetic circulatory response” was considered to contribute to impaired thermoregulation of prepubertal subjects during exercise in the heat (6). It has been argued, though, that this observation is “biologically spurious,” because children do not exercise at the same absolute V̇o2 as adults (45). Instead, they participate in physical activities at intensities relative to their body size (i.e., at similar relative intensities as adults). Supporting this, when values of cardiac output and stroke volume during exercise in normothermic conditions are expressed appropriately to body size, no quantitative nor qualitative differences in circulatory responses (cardiac output, stroke volume) are observed between children and adults (37, 45). Moreover, no maturational differences have been observed in myocardial contractility, patterns of stroke volume response, systolic-to-diastolic time intervals, peripheral vascular resistance, or ratio of change in cardiac output to that of V̇o2 during exercise (42). Current research data therefore fail to identify any impairment of children relative to adults in their cardiovascular functional responses during exercise. Three studies (2 in female subjects, 1 in male subject) have directly compared cardiovascular responses of adults and children during sustained exercise in the heat. Drinkwater et al. (14) described findings in five nonacclimatized prepubertal girls and five college-aged women who walked at low intensity (30% V̇o2max) for two 50-min bouts in Ta of 28°C (83°F), 35°C (95°F), and 48°C (118°F). No fluid replacement was given. No significant differences in cardiac index were observed between groups during walking in any of the ambient conditions, although heart rate was lower and stroke index greater in the adults. No decline in cardiac output was observed in either the girls or women despite dehydration levels of 1.8 and 2.7%, respectively. Nonetheless, findings of facial flushing, dizziness, and fatigue in four of the girls walking in 48°C were considered as “overt indictors that the girls were experiencing cardiovascular difficulty.” Rivera Brown et al. (40) tested nine premenarcheal girls and an equal number of young adult women who were considered heat acclimatized. Subjects pedaled outdoors in Ta of ∼33°C until fatigue at 60% V̇o2max while body fluid status was maintained by prescribed drinking. Patterns of cardiovascular responses were identical, and no significant group differences in heart rate, stroke index, or cardiac index were seen at the point of fatigue. Rowland et al. (44) found no differences between eight boys (mean age 11.7 ± 0.4 yr) and adult men (age 31.8 ± 2.0 yr) in cardiac responses to sustained cycle exercise (65% V̇o2max) to exhaustion in ambient conditions of ∼31°C and 50% RH. Dehydration in this study was avoided by voluntary fluid intake. A small rise in cardiac index was observed during exercise in both groups, with values at exhaustion of 11.75 ± 1.91 l·min−1·m−2 in the boys and 10.15 ± 1.75 l·min−1·m−2 in the men (P > 0.05). Stroke index, mean arterial pressure, and arterial venous oxygen differences remained stable in both groups. Comparisons of children and adults exercising in both normothermic and hot environmental conditions indicate that cardiovascular responses to exercise stress are as effective in prepubertal as in mature subjects. These data do not support earlier contentions that inferior cardiovascular responses in children impair their thermoregulatory responses to exercise in the heat. TRE AND BODY HEAT LOADIn 1995 Armstrong and Maresh (4) compiled data from 8 studies indicating no group differences between children and adults in rise of Tre during exercise in the heat. Average increase was 1.24 ± 0.40 and 1.21 ± 0.39°C, respectively. They concluded that children's thermoregulatory response to exercise in hot ambient conditions is not dissimilar to that of adults. More recent investigations directly comparing children and adults have borne this out (Table 1).
Similar findings have been observed by those investigators who have assessed body heat load during exercise (Table 1). Among the earlier reports, Drinkwater et al. (14) reported an average heat storage of 16.8 and 19.7 kcal/m2 in girls and women, respectively, while walking in 35cC conditions and 31.0 and 26.2 kcal·m−2·h−1 in 48°C (14). Falk et al. (18) calculated heat storage of 5.53 ± 0.44 kJ·h−1·kg−1 in prepubertal boys and 6.81 ± 0.27 kJ·h−1·kg−1 in those who were postpubertal while exercising in 42°C and 20% RH ambient conditions. No maturational differences have thus been observed in responses of Tre or accumulation of heat storage during exercise in hot climatic conditions. These findings imply that thermoregulatory outcomes in children and adults are the same, regardless of any physiological and anatomic features unique to prepubertal subjects. PHYSICAL PERFORMANCE IN THE HEATThe idea that children are more intolerant to exercise in the heat compared with adults stems principally from the study of Drinkwater et al. (14) of five premenarcheal girls and five young adult women walking on a treadmill in a climatic chamber (without fluid replacement). As noted above, while walking at the same low relative intensity (30% V̇o2max), subjects were asked to perform two 50-min bouts of exercise in ambient conditions of 28°C and 45% RH, 35°C and 65% RH, and 48°C and 10% RH. All subjects were able to finish the first walking bouts at 28 and 35°C. In 48°C Ta, all women completed the first walk but four of the five girls were removed by the investigators because of high heart rates (>90% maximum), flushed facies, and “marked signs of distress.” In the second 50-min walk, all completed the 28°C condition, but only two of the girls finished the bout in 35°C (compared with all the women). [Other studies that have been cited to support a decreased exercise capacity by children in the heat compared children and adults at the same absolute workload (23, 24, 53). In this situation, the children were working at a higher relative exercise intensity and would thus be expected to demonstrate inferior exercise tolerance.] Two recent studies have indicated no child-adult differences in tolerance to exercise in the heat when subjects are cycling at the same relative intensity. In their comparison of men and boys performing steady-load cycling to exhaustion (∼63% V̇o2max), Rowland et al. (44) could find no significant group differences in endurance performance capacity in either hot or cool ambient conditions. In ∼19.7°C and 60% RH, the boys endured for 41.38 ± 6.30 min and the men for 42.88 ± 11.79 min. In 31.1°C and 54% RH, the boys lasted 29.30 ± 6.19 min and the men lasted 30.46 ± 8.84 min. In the study of Rivera Brown et al. (40), exercise endurance time in 33.4°C and 55% RH at 60% V̇o2max in acclimatized women (76.5 ± 9.9 min) was greater than in young girls (56.9 ± 6.3 min), but the difference between groups was not statistically significant. HEAT ILLNESSChildren have traditionally been considered to be at increased risk for heat illness (heat stroke, heat exhaustion) during physical activities compared with adults, a supposition based on 1) their perceived inferior thermoregulatory mechanisms, and 2) a greater incidence of heat stroke in the pediatric age group recorded during times of heat waves (1, 6, 35). However, these reports of heat stroke have indicated an augmented risk restricted to infants and small children (<4 yr old), which has been ascribed largely to dependency factors (such as parent neglect) and preexisting chronic illness (54). How this vulnerability might be translated to child athletes or older children playing in the heat is not clear. In fact, cases of serious heat illness in child athletes are conspicuously absent from the medical literature, and informal opinion suggests that such events are rare. Brun and Mitchell (10) were unable to find a single case of heat-related illness in a child athlete in their survey of 10 yr of medical records in a tropical region of Australia (Cairns). CONCLUSIONThe characterization of children's physiological responses and performance outcomes during exercise in the heat is far from complete. However, contrary to earlier assumptions, current research information fails to indicate thermoregulatory differences in the heat between children and adults. Physiological and anatomic features that might potentially bear influence on dissipation of heat in youth do not appear to translate into differences between children and adults in accumulation of body heat, changes in Tre, exercise tolerance, or vulnerability to heat illness. Attention to adequate fluid intake and prevention of heat illness in active youth are clearly important, but there exists no convincing evidence that the risk of exercising in high Ta is any greater than that of adults. (27) REFERENCES
Page 2regular participation in physical activity usually begins in youth, and many individuals peak in their sport performance during adolescence. Exercise causes dramatic changes in hormone concentrations and metabolites that may influence growth and development during puberty. For some, intense training occurs in youth, and the anabolic and catabolic processes that occur with regular activity can influence maturation and gross motor development. For nearly all children, regular participation in physical activity is healthy, promoting development of the muscular and skeletal systems, enhancing cardiovascular fitness and insulin sensitivity, while reducing the likelihood of adolescent obesity, dyslipidemia, and insulin resistance. On the other hand, excessive endurance training in adolescence is associated with hypothalamic-pituitary dysfunction that can delay menarche, cause amenorrhea, and lead to immune suppression. Knowledge about specific hormonal and metabolic responses to exercise in youth is critical, therefore, to understand the physiological benefits and potential risks of physical activity and sport participation. In addition, understanding the metabolic response to exercise may help to form better physical activity and nutrient recommendations for youth of all ages. Unfortunately, despite the potential importance of studying pediatric exercise metabolism, research to date has been limited in this area, in part, because of technical and ethical conditions in studying youth. Over the last 70 years, considerable progress has been done to characterize the metabolic and hormonal responses to exercise in youth (Fig. 1). This review highlights the current state of knowledge about the hormonal and metabolic responses to acute exercise in youth and offers some potential areas of future investigation. Fig. 1.Time line of some of the significant discoveries in pediatric substrate metabolism since the 1930s. RER, respiratory exchange ratio; 31P-MRS, phosphorus-31 nuclear magnetic resonance spectroscopy. PUBERTY, INSULIN SENSITIVITY, AND SKELETAL MUSCLE CHARACTERISTICSThe hormones associated with pubertal development, which include growth hormone (GH), insulin-like growth factor (IGF), the sex steroids, and the catecholamines, are also hormones that can influence energy metabolism during exercise. As such, characterizing the endocrine responses to various forms of exercise in pre- and postpubertal adolescents has long been a focus. It is well known that pubertal development is associated with a period of insulin resistance (3) and that insulin-stimulated glucose disposal at rest is lower in pubescents than in prepubertal children (3, 5). Although the cause of insulin resistance during puberty is not entirely clear, the elevation in sex steroids are thought to oppose the effects of insulin in skeletal muscle, adipose tissue, and liver (14). It has been argued that pubertal adolescents also demonstrate insulin resistance during exercise, although more work needs to be conducted in this area. For example, during 30 min of moderate cycling following glucose ingestion (0.5 g/kg body mass), peripubertal adolescents have a higher glucose-to-insulin ratio than adult male subjects (11). Although it is possible that the lower insulin sensitivity is a result of the increased levels of circulating GH, IGF-I, and catecholamines, this remains to be confirmed. Interestingly, orally ingested glucose is oxidized at a higher rate during exercise in adolescents than in adults (67), suggesting that there is not impairment in plasma glucose uptake and oxidation, per se. It may be that the recruitment of contraction-induced glucose four transporters (GLUT4) to the plasma membrane in children and in prepubertal adolescents is higher than in adults during exercise, although this requires investigation. Testosterone, which increases dramatically in adolescent boys during puberty (26), increases sarcotubular and mitochondrial enzyme content, at least when given to adult men (61). Whether the increase in testosterone at the time of puberty also increases mitochondrial content is unknown. With exercise, there is also a small but significant increase in circulating testosterone levels in adolescent boys (26). However, a high aerobic capacity already exists in preadolescent youth compared with adults. Moreover, no evidence exists that mitochondrial content changes during puberty, although only a few studies have investigated this areas because of the need for muscle biopsy sampling. In one key study that used the muscle biopsy technique, Bell and colleagues (7) report similar mitochondrial volume ratios in prepubertal and adult muscle samples. It therefore remains unclear whether puberty influences mitochondrial content and, if so, whether it has any influence on fuel utilization during exercise. Maturation of skeletal muscle fiber type at the time of puberty, specifically a pattern change from slow to fast twitch, might explain some of the differences in the metabolic responses to exercise between children and adults. Compared with fast-twitch (or type 2) fibers, slow-twitch red (or type 1) fibers are fatigue resistant with a higher mitochondrial content that relies more on oxidative metabolism of fat and carbohydrates. The limited muscle data available for newborns and infants indicate a trend for a higher percentage of type 1 distribution in infants than in either newborns or adults (18). Evidence is lacking, however, that adolescence causes a shift in fibre type composition. Examination of muscle biopsy samples taken from human diaphragm (38) and leg muscle (7) demonstrates that fiber type differentiation and mitochondrial content are unchanged during adolescents. Comparison of muscle fiber type among adolescence and adults is challenging, however, not only because of ethical and technical difficulties in obtaining muscle biopsy samples but also because of the heterogeneous nature of muscle fiber type composition is in humans. Muscle fiber type composition is known to be influenced by a number of factors, including genetic variation and differences in muscle use (30, 31). It may be difficult, therefore, to determine whether fiber type changes occur with age, unless a relatively large longitudinal study is conducted. In one Swedish study, the muscle fiber type composition was determined in 26 female subjects and 55 male subjects all at 16 yr of age and again at age 26 yr. Both sexes had ∼52% type 1 fibers in their vastus lateralis muscles while the type 2A (a subfraction of type 2 fibers with a high oxidative capacity) and 2X (a subfraction of type 2 fibers with a lower oxidative capacity and a high glycolytic capacity) percentage were 33 and 15%, respectively (29). According to Saltin and Gollnick (62), this is similar to what is observed in adults and in younger children with a similar genetic background. Thus, based on the limited information available, very little muscle fiber type changes occur with puberty, although it may be that motor unit recruitment patterns during exercise may change with pubertal development and that these differences might alter fuel selection. Despite similar phenotypical characteristics in muscle between adolescents and adults, their metabolic responses to exercise differ considerably. One of the first to show that metabolic responses were unique in children was Bar-Or (6), who revealed that the anaerobic capacity, as measured by his recently developed all-out 30-s Wingate test, was lower in young children compared with adolescents and adults. Compared with adults, adolescents were also shown to have lower circulating lactate levels during exercise, suggesting that the younger individuals had a lower glycolytic capacity. The commonly held belief that children had an immature glycolytic capacity was based initially on the small number of muscle biopsy studies conducted by Eriksson and his colleagues more than 35 yr ago. In those classical studies, using muscle biopsies of a small numbers of volunteers from local schools in Sweden, Eriksson proposed that children have an inferior anaerobic (or glycolytic) capacity for supplying ATP during high-intensity exercise. This hypothesis was based principally on the observation that phosphofructokinase activity was lower in the boys compared with the men. Muscle glycogen content was also found to be 50–60% that of adults (for a review of that work see Ref. 20). Based on these findings, it was proposed that adolescents were more “oxidative” and less “glycolytic” compared with adults. Some time later, using the phosphorus-31 nuclear magnetic resonance spectroscopy (31P-NMRS) technique in calf muscle, Taylor and colleagues (64) confirmed that muscle oxidative capacity was indeed higher in children and adolescents compared with young adults. In that study, children also had higher pH and ADP levels during muscular contraction and a more rapid recovery from the exercise (64), which fit with the belief that children have a greater reliance on aerobic metabolism and, perhaps, an immature glycolytic capacity. One must note that other biopsy studies from maturing children have failed to confirm the relationship between maturation and glycolytic enzyme capacities (33). However, children have been reported to have ∼3.5-fold lower lactate dehydrogenase activity compared with adults in the obliques internus abdominis muscle in a recent study by Kaczor et al. (37). This finding of a lower lactate dehydrogenase activity in an abdominal muscle of children compared with adults confirms what had been reported previously in lower limb biopsies done on boys and adults (9, 21). The lower nonoxidative glycolytic flux though lactate dehydrogenase is thought to, at least partially, explain the considerably lower circulating lactate levels that are commonly observed during exercise in children compared with adults (34, 52, 58, 67). Based on the biopsy samples done by Eriksson and his colleagues (22–25), muscle lactate levels after maximal exercise are also lower in adolescents than in men. The lower blood lactate concentrations in children during exhaustive exercise may be a result of a lower relative muscle mass, a higher relative total lactate water space, and/or a higher reliance on aerobic metabolism (8). Children and young adolescents clearly oxidize both fat and exogenous carbohydrate at a higher rate during exercise than adults (see substrate utilization during exercise below). Surprisingly, no studies have been published on the key regulatory enzyme pyruvate dehydrogenase in children. A higher activity in this key regulatory enzyme that links glycolytic flux to oxidative metabolism might help to explain the higher oxidative phosphorylation and greater exogenous glucose oxidation in youth compared with adults. Enzymes of the tricarboxylic acid cycle have been investigated in children but to a limited extent. Early studies found significantly higher NADP-isocitrate dehydrogenase, fumarase, and malate dehydrogenase levels in skeletal muscle of pubescent children compared with adults, with no differences in citrate synthase activity (33). The activity of 2-oxoglutarate dehydrogenase (OGDH) in skeletal muscle, which is a rate-limiting enzyme in the tricarboxylic acid cycle, is similar in children and in adults (37). Thus, based on the limited muscle biopsy data in healthy children, it is unclear whether differences in enzyme content or activity explain the differences in energy metabolism during exercise between children and adults. The use of noninvasive techniques to study substrate metabolism in children and adolescents has increased our knowledge on the topic considerably, while allowing researchers to avoid the risks associated with needle biopsy and blood sampling. In the first study to use 31P-NMR in children to measure calf muscle Pi, PCr, and pH, Zanconato et al. (72) showed that preadolescent children (aged 7–10 yr) are less able than adults to generate ATP via rephosphorylation through anaerobic metabolism. Moreover, these young adolescents did not attain as low a muscle pH as the adults did during intense calf muscle contractions, supporting the notion that the children do not have the same anaerobic capacity as adults. Using the same technique, Kuno et al. (39) also published that trained and sedentary boys aged 12–15 yr had reduced anaerobic capacity when compared with young men. On the other hand, no difference was found in glycolytic metabolism between prepubertal and pubertal girls using magnetic resonance spectroscopy (MRS) techniques (49). This latter study cautioned that the morphological differences in muscle size and composition between younger children, adolescents, and adults significantly impacts on the interpretation of the MRS data (49). The work by Zanconato et al. (72) and Kuno et al. (39), cited above, is consistent with the notion that, compared with adults, children and adolescents have a higher rate of muscle oxidative phosphorylation during heavy exercise. Oxygen kinetics measured at the mouth during exercise are indeed different in children compared with adults, suggesting that children are more “oxidative” (4, 72, 73). As these authors point out, the greater O2 utilization during heavy exercise could result from a greater delivery of O2 to the working muscle, a greater mitochondrial content, or a great reliance on oxidative fuels, such as fat, during exercise, although these possibilities all require confirmation. As mentioned above, however, a higher mitochondrial capacity in youth is unlikely, at least based on the studies using the muscle biopsy technique to measure mitochondrial content and activity. Based on the available biochemical factors cited above, and the other important contributions in the field, Inbar and Chia (35) suggest in a recent review that prepubertal adolescents are at relative disadvantage when performing short-term strenuous activities because of their “immature anaerobic metabolism.” Moreover, based on the limited data available (49), they conclude that the maturational changes in the anaerobic pathways may be more pronounced in male than in female subjects. Even if at a relative disadvantage, children can often be observed doing brief intense activities and then taking short breaks to recover from the activity. Whether this type of activity helps to “mature” the glycolytic pathways in muscle is unclear. ENDOCRINE RESPONSES TO EXERCISE AND GLUCOSE HOMEOSTASISBecause physical activity increases some of the hormones associated with pubertal development and because other hormones that are released during exercise influence growth and development, characterization of the hormonal responses to acute exercise in youth is important. Surprisingly, systematic characterization of the endocrine response to exercise in children and adolescents has not been done nearly as carefully as it has been studied in adults [for a historical review of the studies done in adults see Riddell et al. (57)]. One of the first to measure hormones during exercise in children was Fahey in 1979 (26). No differences in hormonal responses to exercise were found among 27 boys at different stages of puberty, perhaps because only one type of exercise was examined and because only a few hormones were measured (testosterone, GH, and insulin). Since then, only a handful of studies have reported circulating levels of select hormones in children and adolescents, and often the responses are measured only at moderate-intensity exercise. Exercise has long been known to be a potent stimulus of the GH and IGF-i axis in children and adolescents. Indeed, physical activities that include intermittent bursts of high intensity play do cause dramatic peaks in circulating GH concentrations that add considerably to the normal circadian rhythm of this potent developmental hormone. It was thought in the 1960s by Ekblom (17) that elevations in pubertal hormones, including GH and testosterone, caused by regular exercise, contribute to increased growth and development in adolescents. This hypothesis remains to be confirmed. In adolescent boys and girls, intermittent high-intensity cycling at 80% peak oxygen consumption (V̇o2peak), with 1-min rest intervals every 10 min, causes an ∼10-fold increase in serum GH levels, although IGF-I levels are unaffected (27). Interestingly, a high-fat meal before high-intensity intermittent exercise blunts the GH response considerably (27). Carbohydrate ingestion also blunts the GH response to exercise in boys and girls (68). Thus the timing of a nutrient-rich meal before exercise may be an important consideration in young growing adolescents. Based on a large number of sophisticated studies done in animal models and in humans conducted in the 1980s and 1990s, it is known that the main glucoregulatory hormones during moderate-intensity prolonged exercise are insulin and glucagon, with minimal contributions from catecholamines, GH, cortisol, and thyroid hormone (70). With increasing intensities, catecholamines are the main regulators of hepatic glucose production and glycogen utilization (43). It is unclear whether these same endocrine factors regulate substrate mobilization in a similar fashion in children, although evidence described below suggests that differences do exist. Based on a number of pediatric-focused studies, insulin levels drop during prolonged exercise in adolescent boys (27, 28, 53, 55, 66) and girls (40), similar to what is typically observed in adults during prolonged exercise. In a study by Laaneots et al. (40) in Viru's laboratory, examining girls at various stages of puberty, postexercise insulin levels were shown to be highest in girls at the last stage of puberty (40), perhaps to compensate for their elevations in pubertal hormones. In contrast, a drop in insulin was observed only in mature boys but not in less mature boys in an earlier study by Fahey et al. (26). Therefore, it may be that the insulin response to exercise differs with pubertal stage and sex, although this requires confirmation. When adults perform exercise above 80% maximal oxygen consumption (V̇o2 max), insulin levels increase after exercise to counter the dramatic rise in catecholamines and the transient hyperglycemia that ensues (43). It is unclear whether the same endocrine responses exist in adolescents. In fact, Galassetti et al. (27) report that 30 min of heavy intermittent exercise (at 80% V̇o2 peak) in adolescents increases glucagon levels but fails to increase the other counterregulatory hormones. In a related study at the same relative work rates, glucagon failed to increase and a compensatory rise in insulin also failed to occur (28). In contrast, during resistance exercise, the catecholamine and cortisol response is reported to be higher in boys compared with men or women (51), suggesting that adolescents may have a higher stress response to this form of exercise. This might have implications in blood glucose homeostasis during intense exercise in youth, causing more of a transient hyperglycemia. At rest, plasma concentration of epinephrine and metanephrine decrease, whereas norepinephrine increases with advancing puberty (71). It is unclear whether these changes in baseline catecholamine levels influence glucose homeostasis at the onset of exercise in adolescents. In a series of classic studies conducted in Germany, Lehmann and colleagues (41) reported similar catecholamine excretion between younger and older boys but found ∼30% lower peak norepinephrine levels in 12-yr-old boys compared with men during maximal treadmill running (42). Rowland et al. (60) found no significant differences in norepinephrine levels during two different submaximal exercise intensities nor at maximal exercise between 10- to 12-yr-old boys and men. The catecholamine responses to 60 min of exercise (∼70% V̇o2max) was also identical in pre- and postpubertal girls, despite marked differences in energy utilization (65). Thus more research is needed to clarify whether the adrenal responses to different forms of exercise change with puberty. Regular exercise is known to influence dehydroepiandrosterone, testosterone, progesterone, β-endorphin, somatotropin as well as leptin and other adipokines, although the examination of the hormonal response to exercise in adolescence is limited. Pomerants et al. (50) found no changes in circulating leptin or ghrelin in response to 30 min of exercise performed at 95% of the ventilatory threshold, although prepubertal children had significantly higher basal values for serum ghrelin compared with the more mature boys. Plasma β-endorphin, corticotrophin, and ACTH all increase proportional to the exercise intensity similarly in pubertal and pre pubertal adolescents, whereas the GH response is higher in pubertal vs. prepubertal adolescents in some (12) but not all (26) studies. Recently, more attention has been focused on how obesity influences the endocrine responses to exercise in youth. Based on work in Cooper's (19) laboratory, obesity blunts the GH and catecholamine responses to exercise in adolescents, whereas insulin levels are higher, suggesting that these youth are insulin resistant as expected. The attenuated GH, the attenuated catecholamine, and the failure to suppress insulin release during exercise in obese adolescents, to levels observed in nonobese youth, support the hypothesis that obesity causes impairment in the adrenergic response to exercise. Interestingly, despite the blunted counterregulatory hormone responses and the higher circulating insulin levels during exercise, obese adolescents do not develop hypoglycemia (19), suggesting that liver and/or muscle are resistant to these endocrine signals that regulate glucose balance. Future studies should focus on how age, maturation, and body adiposity influence not only hormonal responses to exercise but also the specific tissues metabolic responses to these endocrine factors. In some (15, 16, 24, 55), but not all (44, 47) studies, children have been shown to have a small (1–1.5 mmol/l) and transient fall in blood glucose concentration during the first few minutes after the start of aerobic exercise. The reason for the drop in blood glucose is unclear but may be a result of what Boisseau and Delamarche (10) refer to as an “immature” hepatic glycogenolytic system. The clinical relevance (or performance-related implications) of this small drop in glycemia is unclear, however. The decrease in glycemia at the onset of exercise in adolescent boys is not attenuated by exogenous carbohydrate intake (55). During moderate-intensity cycling, blood glucose levels and catecholamines levels are lower in adolescent girls than in adolescent boys (15), which suggests that girls either have a lower hepatic glucose production rate than boys or that girls utilize more plasma glucose during exercise than boys. Both possibilities require investigation. Although the belief that children have an immature glucoregulatory system, at least for the metabolic pathways involved at the start of exercise, remains a possibility, there is little evidence that healthy youth develop exercise-associated hypoglycemia. It may be that the drop in glycemia at the start of exercise is caused by the increase in glucose disposal into working muscle, which is independent of insulin signaling. The effect of age, maturation, and sex on hepatic glycogen stores and on hepatic glucose production rates during exercise requires investigation, although these studies are currently challenging because of technical and ethical considerations. As pointed out in a review by Boisseau and Delamarche (10), infants have only ∼15 g of hepatic glycogen with the majority of that used to maintain the central nervous system's high demand for glucose. Knowledge of hepatic glycogen stores in older children and adults is scarce because of the invasiveness of these sorts of studies. If children and young adolescents have a low amount of hepatic glycogen, it would be expected that they would be hampered in their ability to exercise for prolonged periods because of increased risk for hypoglycemia. This is clearly not the case. Young adolescents can exercise for at least 120 min at a high intensity (∼60% V̇o2max) without developing hypoglycemia, even when they do not ingest carbohydrate before or during the activity (24, 44, 47, 53–55). However, it is worth mentioning that in a majority of these studies, the participants were exercising in a fed state (i.e., within hours of a meal), rather than in a fasted state when hypoglycemia may be more likely. It is also important to note, that although orally ingested carbohydrate may not be required to prevent hypoglycemia during prolonged exercise in children and adolescents, these energy source can improve performance and contribute significantly to the energy supply during the activity (see substrate utilization during exercise below). SUBSTRATE UTILIZATION DURING EXERCISEUsing the methods developed by Dill and colleagues at the Harvard Fatigue Laboratory, Robinson (58) was the first to publish in 1938 that the physiological responses to treadmill walking and running differed among children and adults. In that pioneering work, Robinson suggested that a lower respiratory exchange ratio (RER) during exercise in younger adolescents compared with older adolescents and adults was the result of a diminished reserve of carbohydrates. However, he believed that the prolonged 15-h fast before the test caused a reduction in endogenous carbohydrate stores in the younger volunteers and therefore a lower RER. Follow-up studies by Morse and colleagues (46) confirmed that RER levels are lower in children than in adults and that RER increases with increasing age in adolescents. Although RER does have its limitations when used to quantify substrate utilization (36), this technique has been used extensively to characterize substrate oxidation in children and in adults. A study by Montoye (45) in the early 1980s is also of note in this context because of the large numbers of children and adolescents studied (∼180 examined the 10- to 14-yr age range and ∼190 in the 15- to 19-yr age range). Based on these, and numerous other studies to date, the general consensus is that the lower RER frequently observed in children, when compared with adults, indicates that the former utilize considerably more fat and less carbohydrate for energy at a given relative exercise intensity. This difference between boys and men also holds true for girls and women (44). As an example of the magnitude of difference, Timmons et al. (67) found that during cycling at 70% V̇o2 peak, pre- and early perpetual boys oxidize ∼70% more fat and ∼23% less carbohydrate compared with men (67). The higher relative contribution from fat in adolescents persists even when the exercise is performed during carbohydrate feeding. In line with the notion that maturation influences substrate utilization are a number of studies examining children at different stages of puberty. Compared with older girls (aged 14 yr), younger girls (aged 12 yr) have a higher fat oxidation rate during moderate intensity exercise (65). Compared with older boys, younger boys have higher relative rate of fat oxidation than post pubertal boys (66). Zunquin et al. (74) also recently reported that the development of puberty reduces the ability to oxidize fat during exercise in obese boys. Recently, investigation on the exercise intensity that elicits peak fat oxidation has been revisited by Jeukendrup's group, showing that peak fat oxidation rates occur between 40 and 50% of V̇o2max in untrained men and women (69) and at 63% of V̇o2max in well-trained athletes (1). Finding the intensity at which point maximal fat oxidation occurs in children and adolescents is important for both performance and health benefit reasons. Recently, Stephens et al. (63) showed that maximal fat oxidation is higher in prepubertal boys compared with pubertal boys, and based on their RER values collected, maximal fat oxidation occurs at somewhere between 40 and 70% V̇o2max in prepubertal boys. Moreover, based on their findings, these authors propose that the development of an “adult-like metabolic profile” occurs between mid to late puberty and is complete by the end of puberty (63). Recently, peak fat oxidation rate during cycling in boys was calculated to be ∼8 mg·kg lean body mass−1·min−1, occurring at 60% V̇o2 peak. In untrained young men, peak fat oxidation rate was determined to be ∼5 mg·kg lean body mass−1·min−1, occurring at ∼40% V̇o2 peak (56). In that same study of a small cohort of adolescent boys, examined in a 3-yr longitudinal design, both the relative fat oxidation rate and the exercise intensity that elicited peak fat oxidation decreased as boys develop through puberty, with the most significant changes occurring between mid- to late puberty (56). Because of the high rate of fat oxidation occurring during intense exercise, the exercise intensity at which point carbohydrate dominates as the fuel source is considerably higher in children than in adults, and this drops considerably as adolescents develop though puberty. The mechanism for the higher rate of fat oxidation in childhood compared with adulthood is unknown. During prolonged exercise, nonesterified fatty acids (NEFA) are released from triacylglycerol stores in adipose tissue and muscle. Their release during moderate exercise is regulated by the sympathetic nervous system through a gradual increase in epinephrine release and a gradual reduction in insulin release by the pancreatic beta cells. The high epinephrine-to-insulin ratio increases circulating NEFA levels severalfold, which then facilitates the amount oxidized by skeletal muscle. It is not clear, however, whether children and adolescents have higher circulating NEFAs than adults during endurance exercise. Delamarche et al. (16) indirectly reported a higher NEFA turnover in boys than what is typically reported for men; however, Martinez and Haymes (44) found no difference in plasma NEFA and glycerol levels in girls and women during a 30-min run. Interestingly, increases in circulating lactate levels during high-intensity exercise is thought to limit NEFA release by adipose tissue and children do have lower lactate production during heavy exercise compared with adults. The higher oxidation of long-chain fatty acids in the skeletal muscle mitochondria of children may play a key regulatory role in lowering carbohydrate oxidation during exercise. Long-chain fatty acids are shuttled across the mitochondrial membrane by two carnitine palmitoyltransferases (CPT I and CPT II), and these are thought to be rate limiting for fatty acid oxidation in muscle. However, there are no age-related changes in CPT activity in children compared with adults (37), and there are also no major differences in enzyme activities of fatty acid metabolism (acetoacetyl-CoA thiolase and 3-hydroxyacyl dehydrogenase) in pubescent children compared with young adults (33). It may be that the higher fat oxidation in boys may be a default mechanism as a result of an underdeveloped glycogenolysis and/or glycolytic system as has been proposed by Timmons et al. (67), although experimental evidence to support this hypothesis is lacking. Interestingly, the CPT/OGDH ratio of enzyme activities in skeletal muscle tends to be higher (16%) in children vs. young adults (37), suggesting that there is preferential oxidation of fatty acids over other substrates such as carbohydrate in children compared with adults. The influence of childhood disease and metabolic dysregulation on substrate metabolism has been of interest to some investigators. Adolescents with Type 1 diabetes have even higher rates of fat oxidation during moderate-intensity exercise compared with nondiabetic adolescents (53). Fat oxidation rate is correlated to the level of fat-free mass (FFM) in obese boys (13). When expressed per unit of FFM, or as a percentage of total fuel oxidation, fat utilization is lower in postpubertal than prepubertal obese boys (13). It does not appear, therefore, from this study and others that childhood obesity is caused by a reduced rate of fat oxidation during exercise. Estrogen has long been thought to explain the higher rate of fat oxidation in women compared with men. An increase in estrogen during puberty was therefore thought to increase fat oxidation rates in female subjects. As cited above, however, a more than twofold higher rate of fat oxidation occurs during exercise in 12-yr-old girls compared with 14-yr-old girls (65), despite nearly 50% higher circulating estradiol levels in the latter group. These data, along with the observation that fat oxidation rate is inversely related to estradiol concentrations in girls (65), suggest that it is likely not estrogen that is responsible for the higher rate of fat oxidation in adult women, as has been proposed by some (32). Boys and girls appear to have similar rates of fat oxidation during treadmill walking performed at the same absolute intensity, at least until the final stages of puberty at which point RER values are slightly higher in girls than in boys (59). Despite a lower whole body rate of carbohydrate oxidation during exercise, and a much higher rate of fat oxidation, children have considerably higher rates of orally ingested glucose oxidation. During moderate to intense cycling (ranging from 50 to 70% V̇o2 peak), orally ingested 6–8% glucose beverage (sometimes called exogenous carbohydrate) provides between 15 and 22% of the overall energy supply in pre- and early pubertal boys (53–55, 67). These values are considerably higher than what is reported for adults (48) and is counter to the widely held belief that children would not oxidize significant amounts of exogenous carbohydrate because of their greater reliance on fat as a fuel. In the one study that compared adolescent boys with male subjects directly, pre- and early pubertal boys had a 50% higher relative contribution from exogenous carbohydrate to the total energy supply when compared with men (67). To further investigate the effects of pubertal status and age on exogenous carbohydrate utilization, Timmons et al. (66) recently took 20 boys of the same chronological age (12 yr) and divided them into three groups based on their pubertal stage (i.e., prepubertal vs. early pubertal vs. mid- to late pubertal). The boys consumed either a placebo drink or a 13C-enriched 6% carbohydrate drink, while cycling for 60 min at 70% of their V̇o2max. Interestingly, fat oxidation was similar in all three pubertal groups, but the energy supply from exogenous carbohydrate oxidation differed. Specifically, exogenous carbohydrate oxidation contributed to ∼30% of the total energy expenditure in the pre- and early pubertal boys, but to only 24% in mid- to late pubertal boys, identical to what was observed in a group of more mature 14-yr-old boys. Exogenous carbohydrate oxidation rate, expressed as a percentage of total energy expenditure, was inversely related to testosterone levels (r = −0.51, P = 0.005, n = 29), suggesting that the androgen might lower the relative contribution from this important fuel source. Thus it appears that that the reliance on exogenous carbohydrate during exercise is particularly sensitive to pubertal status, with the highest oxidation rates observed in pre- and early pubertal boys and lowest in mid- to late puberty, independent of chronological age. Interestingly, relative exogenous carbohydrate oxidation rate during exercise is not different between younger (i.e., 12 yr old) and older (i.e., 14 yr old) girls despite large differences in circulating estradiol levels (65). Differences between adults and children may also exist in their ability to oxidize fructose during exercise. A beverage combining fructose and glucose has considerably less of an insulin and glucose response compared with glucose alone in adolescent boys (55). In addition, compared with glucose alone, fructose plus glucose has a similar rate of oxidation, although the former provides more of an ergogenic effect during aerobic exercise (55). Beverages containing glucose and fructose have a higher energy contribution during exercise than glucose alone in adults (2). All-out cycling sprint performance after intermittent cycling for 90 min in boys can be increased by 20–40% with a combined 6% glucose and fructose beverage, compared with either placebo or 6% glucose alone (55). Although exogenous carbohydrate intake spares endogenous carbohydrate and fat in boys, to a greater extent than in adults, the location of this spared fuel (i.e., intramuscular or extramuscular stores) is unknown. SUMMARY AND IMPLICATIONSDuring play and structured sport, young people derive energy from the interplay between anaerobic and aerobic metabolism, with a greater reliance on aerobic metabolism as fuel compared with adults. In the resting state, adolescents developing through puberty have some degree of insulin resistance that carries over during exercise. Compared with adults, adolescents have lower circulating lactate levels likely because of a reduced activity of the enzyme pyruvate dehydrogenase, a lower lactate production rate, a higher rate of oxidative phosphorylation, and a great reliance on lipid as an energy source. Glucose production at the onset of exercise may be inadequate to prevent a transient drop in glycemia, although the likelihood of the development of hypoglycemia in healthy adolescents even during long-duration exercise seems remote. Throughout a wide range of exercise intensities, lipid utilization is considerably higher in childhood than in adult hood, and adolescents have attenuated rates of lipid oxidation as they develop through puberty. The greatest change in substrate utilization in boys and girls occurs during the final stages of puberty. Despite a higher rate of lipid utilization compared with adults, adolescent boys oxidize exogenous carbohydrate at a higher relative rate, perhaps because of their lower endogenous carbohydrate stores. As a result, endogenous energy sources are spared with carbohydrate intake and exercise performance may be enhanced. In young prepubertal female subjects, exogenous carbohydrate intake reduces the high rate of lipid oxidation during exercise, although this effect is lost as girls develop through puberty. A schematic summary of the major fuel fluxes during prolonged exercise in youth and the possible mechanisms for the observed higher reliance on lipid and exogenous carbohydrate as a fuel source during exercise is shown in Fig. 2. Fig. 2.Substrate flux during prolonged moderate-intensity exercise in youth. Fuel utilization during exercise differs in children compared with adults. Before puberty, youth rely less on carbohydrate and more on lipid as a fuel source. When carbohydrate intake (i.e., exogenous carbohydrate) occurs during exercise, prepubertal children have a higher relative rate of exogenous carbohydrate oxidation compared with adults. As children develop through puberty, the greater reliance on lipid and exogenous carbohydrate as the 2 main fuel sources during exercise is lost. The possible mechanisms for the unique patterns of fuel utilization in youth may include 1) a greater provision of nonesterified fatty acids (NEFA) and a reduced provision of hepatic-derived glucose into the circulation, 2) a greater reliance on intramyocellular lipids for oxidation, 3) a reduced muscle glycogen content, 4) reductions in key glycolytic enzymes in skeletal muscle, or 5) increased fatty acid transport into mitochondria. Because considerable differences exist in the mix of fuels used during exercise between children and adults, one wonders whether their might be some performance or health related implications. A greater proportional reliance on exogenous carbohydrate as fuel during exercise, in addition to a greater utilization of endogenous fat, leads one to conclude that they may be developed in a way that protects endogenous substrates for growth and development of the musculoskeletal and the central nervous systems. Further investigation is required to determine whether the unique metabolic responses to exercise in youth confer any performance or health-related advantages, however. First, the lower lactate and H+ production and a greater reliance on aerobic metabolism may favor activities that are more aerobic in nature rather than those that are of a higher intensity and for a brief duration. For this reason, consideration should be given to develop exercise recommendations for the type, intensity, and duration of physical activity for healthy growing youth. Second, work is needed to determine whether the greater reliance on fat as fuel during prolonged activities helps to advantage children for activities of longer duration, while helping to protect precious carbohydrate stores for the developing central nervous system. Third, because it appears that children have already in place many of the metabolic advantages of a well-trained athlete, more research is needed to determine whether there may be an optimal training regime (frequency, duration, intensity) for the maintenance of this apparent metabolic advantage though development into adulthood. Finally, consideration should be given as to what are the appropriate preexercise nutritional requirements of growing youth based on their unique metabolic profile. Future investigations, using minimally invasive techniques, are needed to determine the mechanisms for the altered substrate utilization during exercise in adolescents. Specifically, the various components of macronutrient oxidation during exercise require investigation, with techniques developed that can compartmentalize the sources of fat and carbohydrate utilized during a variety of exercise intensities and durations. Moreover, protein oxidation during exercise in youth and the influence of protein utilization on growth and development requires investigation. Finally, more work needs to be done to clarify whether adolescence causes altered endocrine responses to prolonged moderate intensity exercise and heavy exercise. GRANTSM. C. Riddell is supported by the National Science and Engineering Council of Canada and the Canadian Foundation for Innovation. I would acknowledge the contributions of Dr. Oded Bar-Or (deceased) for his work in pediatric exercise physiology and substrate metabolism. I thank Anna Standish for her artwork in Fig. 2 and Dr. Brian Timmons for review of the manuscript. REFERENCES
Page 3several factors influence the normal biological growth and maturation of the child and adolescent. Although genes, nutrients, and hormones are viewed as the chief determinants, the level of habitual physical activity (PA) is thought to be one of several ancillary factors contributing to the growth and maturation of body size and composition (11, 51). Recently, concern has been expressed regarding the low levels of habitual PA and associated negative health outcomes among contemporary youth (49). Most of the previous studies examining the influence of PA on body composition of children and adolescents have focused on either adiposity (4) or bone accrual (2), given the concern of obesity and osteoporosis, respectively. In contrast, few studies have considered the influence of PA on lean body mass (LBM) in children and/or adolescents. LBM is often considered a surrogate for skeletal muscle mass, which is an important predictor of several physiological capacities expressed in absolute terms (e.g., maximal oxygen uptake, neuromuscular strength, anaerobic capacity, etc.) (1, 54) and health indicators (e.g., bone mass, insulin resistance, and obesity) (18, 56). Previous studies have shown that significant differences have been reported in LBM between athletes and nonathletes (35), changes following exercise (14–16), and longitudinal studies or exercise training studies of special populations (e.g., cystic fibrosis, Prader-Willi syndrome, etc.) (28, 42, 44, 50). However, few studies have examined the influence of free-living, habitual PA on LBM accrual during adolescence, a critical period of growth affecting body composition and those that have, for the most part, been cross-sectional in design and thus unable to tease out the independent effects of PA from those of growth and maturation (12, 21, 25). To fully explore the influence of PA on the development of LBM during adolescence, longitudinal studies are required so that the individual trajectories of LBM can be examined by taking into consideration the timing and tempo of growth and maturation. This is important as there are known age- and sex-associated variations in lean mass development. During childhood, sex differences are minimal, but become more apparent during adolescence. Young adult values are reached earlier in girls (15–16 yr) compared with boys (19–20 yr). In late adolescence and young adulthood, boys have an average LBM that is about 1.5 times larger than that of girls. Partitioning the effects of PA on lean mass accrual from normal growth and maturation is, therefore, a major challenge (6). The introduction of multilevel statistical models (19) has enabled researchers to fit individual growth curves to measurements over time. Essentially, multilevel modeling is an extension of multiple regressions, which is appropriate for analyzing hierarchical data (20). In the multilevel framework, each individual has their own straight-line growth trajectory, with intercepts and slope coefficients varying between individuals. Using this technique, the independent effects of growth, maturation, and sex on LBM accrual can be identified, and the independent time-dependent effects of PA can be identified (5). To our knowledge, no studies have documented the independent influence of habitual free-living PA on LBM accrual in boys and girls. Clearly, further study is warranted to examine the effects of PA on lean mass development, taking into consideration normal growth and maturation and sex differences. The purpose of this study was to investigate the independent effects of PA on total body and regional lean mass accrual, while accounting for the confounding effects of growth and maturation. It was hypothesized that PA would have a small but significant influence on the development of lean mass accrual during growth and maturation in both boys and girls. The uniqueness of this study is that serial measures of habitual PA and LBM were observed in a free-living group of boys and girls for 7 consecutive yr during the adolescent growth period. METHODSSubjects were drawn from the Saskatchewan Pediatric Bone Mineral Accrual Study (PBMAS) (2, 3). The study used a mixed-longitudinal cohort design. In 1991, seven age cohorts were recruited and followed for up to 6 consecutive yr. Because the cohorts overlapped, it is possible to establish developmental patterns from 8 to 20 yr of age. In 1991, of the 375 eligible students (ages 8–15 yr) attending two elementary schools in the city of Saskatoon (population 200,000), the parents of 228 students (113 boys and 115 girls) provided written consent for their children to be involved in this study, and 220 were scanned by dual-energy X-ray absorptiometry (DXA). From 1992 to 1993, an additional 31 subjects were recruited and scanned. After 6 yr of data collection, 109 boys and 113 girls had been measured on one or more occasions (median 6 occasions). These subjects represent the study population for the present investigation. Ninety-eight percent of subjects were of Caucasian descent. Before participation in the study, informed consent and child assent were obtained. The human subject's research protocol was approved by the University of Saskatchewan Biomedical Research Ethics Board (Bio no. 88-102). DXA scans of the total body were performed in October or November of each year using a QDR2000 scanner (Hologic, Waltham, MA). The array mode was used for all scans, employing enhanced global software version 7.10. Total body scans were analyzed for body composition using software version 5.67A. Precision of the QDR2000 scanner was tested in vitro using a lumbar spine phantom (scanned daily) and in vivo using a test-retest design with 20 healthy male and female university students. The retest occurred either the same day (for short-term precision) or 4 wk later (for long-term precision); coefficients of variation of duplicate measurements were calculated (47). The short-term precision (%) in vivo was 0.54% for total body bone mineral free lean mass (BMFL), 4.09% for arm BMFL, 1.19% for leg BMFL, and 0.67% for trunk BMFL. The precision for total body fat mass (FM) and bone mineral content (BMC) was 2.95 and 0.60%, respectively. These values are in line with other studies utilizing the QDR 2000 in the array mode (48). Site-specific soft tissue (fat and lean mass) values were taken from the total body scans. Anthropometric measurements were taken at 6-mo intervals by trained personnel following a standard protocol (40). Stature was recorded without shoes as stretch stature to 0.1 cm using a wall-mounted stadiometer. Body mass was measured to 0.01 kg on a calibrated electronic scale. A PA questionnaire was administered a minimum of three times per year (fall, winter, and spring) for the first 3 yr of the study and two times per year (fall and spring) thereafter for all subjects. The PA questionnaires for children (PAQ-C) and adolescence (PAQ-A) consist of nine items designed to provide a measure of a child's general PA level during the school year. PA is described as “sports, games, gym, dance or other activities that make you breathe harder, make your legs feel tired and make you sweat.” Each item is scored on a 5-point scale, with higher scores indicating higher levels of activity. The mean of these items forms a composite activity score. In diverse samples of children, the scale has consistently demonstrated acceptable internal consistency. Validity has been examined by comparing results with teacher evaluation of activity, Caltrac motion sensors, 7-day activity recalls, step tests of fitness, and leisure time activity scales. Results have been generally favorable with moderate relationships reported (26, 27). In any 1 yr, a subject had either two or three activity assessments. The mean of these assessments was used as the activity score for that year. A biological maturity age was determined for each individual to control for sex-related maturational differences. The age of peak linear growth [age at peak height velocity (APHV)] is an indicator of somatic maturity, representing the time of maximum growth in stature during adolescence. It occurs when linear growth is ∼92% of adult height (36). To establish APHV for each child, whole-year velocity values were calculated for each subject by dividing the difference between the annual distance measurements by the age increment (the mean age increment was 0.998 ± 0.048 yr). A cubic spline fit was then applied to the whole-year velocity values for each child. A spline is interpolating polynomials, which uses information from neighboring points to obtain a degree of global smoothness. The cubic spline procedure was chosen over other curve-fitting protocols, because it maintains the integrity of the data without transforming or modifying the underlying growth characteristics. A biological maturity age (years) was calculated by subtracting the chronological age at the time of measurement from the chronological APHV. Lean mass values were considered in terms of time before and after APHV. Thus a continuous measure of biological age was generated. Biological age categories were constructed using 1-yr intervals, such that −1 APHV age group included observations between −0.49 and −1.50 yr from (i.e., before) APHV. Statistical analysis was performed using SPSS software version 15.0. Values are reported as means ± SE, unless otherwise noted, a level of significance of P < 0.05 was used, and all statistical tests were two-tailed. For the longitudinal analyses, hierarchical (multilevel) random-effects models were constructed using a multilevel modeling approach (MlwiN version 1.0, Multilevel Models Project; Institute of Education, University of London, UK) (5, 8, 9, 17, 19, 33). Detailed description of multilevel modeling, as applied to the PBMAS, has been previously reported (9, 17), and complete details of this approach are presented elsewhere (5). In brief, lean mass development was measured repeatedly in individuals (level 1 of the hierarchy) and between individuals (level 2 of the hierarchy). Analysis models that contain variables measured at different levels of the hierarchy are known as multilevel regression models. Specifically, the following additive, sex-specific, random-effects multilevel regression models were adopted to describe the developmental changes in lean mass parameters with biological age.
RESULTSStature and body mass values were within normal reference standards (34) for all chronological ages in both sexes. Anthropometric, body composition, and PA data for the adolescents aligned by biological age (years from PHV) and sex are presented in Table 1. Boys were significantly older, taller, heavier, and had greater total body BMC than girls at all biological maturity ages, −4 yr pre-PHV to +4 yr post-PHV (P < 0.05). No significant sex differences were found in total body FM until 1 yr after the attainment of PHV (+1) (P > 0.05). Figure 1 shows the developmental curves for the lean mass parameters. When aligned by biological age, boys had greater lean mass than girls at all ages and at all body sites (P < 0.05) (Fig. 1 and Table 1). The mean differences between body weight and sum of DXA parts (bone mineral, fat and lean mass) (Table 1) was 0.7 ± 0.4 kg, when outliers (>3 SD) were removed. No significant differences in PA scores were found, apart from at −3 and −1 yr from PHV. For subsequent analysis, the sexes were split. Fig. 1.Development of boys' and girls' lean mass for total body (A), trunk (B), arms (C), and legs (D) aligned by biological maturity age [years from age at peak height velocity (PHV)]. Values are means ± SE. Table 2 summarizes the results from the multilevel models for total body, trunk, arm, and leg lean mass development for boys. To shape the individual curves, and thus make the models nonlinear, power functions of biological age (biological age2 and biological age3) were added as fixed effects. The model for boys' total body lean mass (Table 2) indicates that, once biological age (1 yr predicts 1,119 g of lean mass) and height (1 cm predicts 738.7 g of lean mass) are controlled, a significant independent PA effect is found (a score of 1 predicts 484.7 g of lean mass) (P < 0.05). This indicates that a boy who had a PA score of 1 (1 * 484.7 = 484.7) had 1,938.8 g less total body lean mass than a boy of the same biological age and height with a PA score of 5 (5 * 484.7 = 2,423.5, 2,423.5 − 484.7 = 1,938.8). Once the effects of biological age and height were controlled at the other three sites, PA was also found to be a significant independent predictor of arm (69.6 ± 27.2 g, P < 0.05), leg (197.7 ± 60.5 g, P < 0.05), and trunk lean mass (249.1 ± 91.4 g, P < 0.05). A biological age by PA group interaction coefficient was added to the models, but was not significant (i.e., the estimated mean for the coefficient was <2 * SEE, P > 0.05), indicating that the PA group difference was the same at all biological ages.
Biological age was also added as a random coefficient (Table 2). The random effects coefficients describe the two levels of variance [within individuals (level 1 of the hierarchy) and between individuals (level 2 of the hierarchy)]. In boys, at all sites, the significant variances at level 1 of the models indicate that lean mass was increasing significantly at each measurement occasion within individuals (estimate < 2 * SEE; P < 0.05). The between-individuals variance matrix (level 2) for each model indicated that individuals had significantly different lean mass growth curves, both in terms of their intercepts (constant/constant, P < 0.05), and the slopes of their lines (biological age/biological age, P < 0.05). The variance of these intercepts and slopes was positively and significantly correlated (constant/biological age, P < 0.05) in all four models. The variance around the average line was, therefore, different at different biological ages. Similar results were found in girls for both fixed and random effects (Table 3). The fixed effects indicated that PA was a significant independent predictor of total body (306.9 ± 96.6 g, P < 0.05), arm (31.4 ± 15.5 g, P < 0.05), leg (162.9 ± 40.0 g, P < 0.05), and trunk lean mass (119.6 ± 58.2 g, P < 0.05), although it is noted that the coefficients were between 18 and 55% lower than those observed in the boys. Similar to the boys, biological age by PA group interaction coefficients were not significant.
The significant effects of PA on total body lean mass models are illustrated in Fig. 2. In this figure, values from Tables 2 and 3 were used to predict the average growth curve for a boy A and girl C, who had an activity score of 5 and average statures at each biological age (Table 1). These data were compared with a boy B and girl D, who had an activity score of 1.0 and who had average statures at each biological age (data from Table 1). Both scenarios reflect the extremes of the observed distribution of these factors and serve to illustrate their influences on total body lean mass growth in adolescence. The significant difference between the two boys and girls is associated with PA levels. Fig. 2.Predicted total body lean mass accrual for a boy A and girl D, who had an activity score of 5 (high) at each biological age (years from PHV) compared with a boy B and girl C, who had an activity score of 1.0 (low) at each biological age (years from PHV). The height values are taken as the mean values shown in Table 1. Values are predicted means from models in Tables 2 and 3. DISCUSSIONThe main finding of this study is that an increase of 1 SD in habitual PA score (SD = 0.71) will increase lean mass by 344 g in boys and 218 g in girls, when the confounders of biological maturity and stature are controlled. The results indicate a sex difference in that, with an increase in the PA score of 1 SD of PA score, there is >50% greater increase in total body lean mass accrual for boys compared with girls (344 vs. 218 g), when biological maturity and stature are controlled. The average PA level score for boys was 2.9 ± 0.71; thus increasing PA by 1 SD above the mean at each maturity age would add an additional 1.3% total body lean mass at −4 yr from PHV, 0.8% at PHV, and 0.6% 4 yr after PHV, when growth in stature is held constant. In contrast, the average PA level for girls was 2.7 ± 0.71, and increasing PA by 1 SD above the mean at each maturity age would add an additional 1.0% total body lean mass at −4 yr from PHV, 0.7% at PHV, and 0.5% 4 yr after PHV, when growth in stature is held constant. Few longitudinal studies have examined the influence of habitual PA on lean mass accrual among free-living adolescence, with the majority only partitioning out fat mass (FM) from fat free mass (FFM). For example, a study of 40 Polish boys grouped the boys as regularly trained (>6 h/wk), moderately trained (4 h/wk but not on a regular basis), and untrained (<2 h of sport/wk) and followed them from 11 to 18 yr of age (35). The results showed that FFM was comparable at age 11 yr, but then increased more in the regularly trained boys compared with the other groups. The differences in FM remained minimal between the regularly active and untrained boys during adolescence. Others have shown the cross-sectional association between PA and lean mass or FFM in children (12, 21, 25). In general, these studies suggest that habitual PA explains a small portion (<5–10%) of the total variance in lean mass, which is consistent with our findings. Exercise training studies also show small changes in FFM in adolescent subjects. For example, Eliakim and colleagues (14–16) reported a consistent mean increase of 3–4% in thigh muscle volume following a 5-wk training program in boys and girls. However, it is important to consider that most children do not engage in highly structured training regimens such as those imposed in exercise training studies; therefore, the importance of considering the influence of habitual, free-living PA over a period of time (e.g., the adolescent growth spurt) is a novel and practical aspect of the present study. On another note, our findings showing a lack of difference in PA between boys and girls is inconsistent with the literature (53). Nevertheless, our findings suggest the importance of PA during the adolescent growth period on lean mass accrual. This finding is important, since adolescence represents not only the period of the lifespan when PA levels decrease substantially (41, 45, 52), but also a time when substantial changes in body composition are occurring, so it becomes more important to ascertain the positive influence of habitual PA. Furthermore, the development of lean mass has important implications for metabolic health (56). Given our previous work on bone mass, a discussion of the link between lean mass and bone mass development will be considered below. However, it should also be acknowledged that lean mass plays a potential role in obesity and insulin resistance. The mechanisms by which PA or mechanical stress influences lean mass accrual during growth and maturation are uncertain (11). However, it is generally agreed that the hormonal milieu plays an important role in the anabolic effects of PA. The primary basis for skeletal muscle growth during childhood and adolescence is the growth hormone (GH)/insulin-like growth factor I (IGF-I) axis (11, 38), androgens (estrogen and testosterone) and their interaction with GH (39), and insulin (22). Cooper (11) has proposed a conceptual model of the exercise modulation of growth, which includes both central and local components. The central component includes mechanisms that affect cellular growth throughout the body, namely through the actions of GH-IGF-I. The local component includes mechanisms that stimulate growth of tissue specific to the exercised tissue. These mechanisms may act through either autocrine or paracrine actions of IGF-I or fibroblast growth factors in response to a variety of “signals” (i.e., stretch, changes in tissue Po2, etc.). However, in the growing and maturing child, the combination of relatively high levels of habitual PA and puberty-related increases in anabolic hormones make it difficult to specify the independent effects of PA and hormones on lean mass accrual. On the other hand, it is also possible that there are integrated mechanisms by which PA influences these metabolic processes (11). As previously mentioned, the results indicate that only a modest portion of the total variance in lean mass can be independently accounted for by PA. It is known that lean mass is highly heritable, and previous studies in adults suggest that genetic factors account for as much as 80–90% of the variance in lean tissue (23, 32, 43). Limited genetic studies of children and adolescence are available. One study (29) of 105 twin pairs between 10 and 14 yr of age showed several important findings that may be related to the present study. First, univariate models revealed that the largest part (87–95%) of the variance for muscle circumferences at most ages was explained by additive genetic factors. Second, sex differences were observed for some age categories. Third, multivariate models showed age- and sex-specific patterns, which may suggest pubertal influences. Besides showing that muscle circumferences are highly heritable characteristics, this study also indicated that pubertal events in boys and girls explain some of the variation in lean mass accrual. It is well known that muscle mass and bone mass are closely associated (13). The correlation between LBM and bone mineral content (BMC) is especially close during growth and maturation (30, 56). It has been postulated that the statistical association between LBM and BMC reflects a direct cause-and-effect relationship, the mechanostat theory (18). According to this hypothesis, the skeleton continually adapts its strength to the loads to which it is exposed to keep bone deformation within safe limits. The largest physiological loads on the skeleton result from muscle contraction, which puts severalfold larger stresses on the skeleton than the simple effect of gravity (10). Mechanostat theory, therefore, predicts that the increasing muscle mass (and thus force) during development creates the stimulus for bone to increase its mass (and thus its strength). We have previously shown in this cohort that the maximal rate of LBM accrual occurs a few months before the maximal increase in BMC (24) and that the peak rates of change in these two measures are closely correlated (37); these observations are in accordance with the mechanostat theory. In relation to PA, our laboratory has previously shown, when the PBMAS subjects were quartiled for activity, that the most active quartile had greater total body and femoral neck BMC peak accrual, and greater accrual over 2 yr around the peak, than their inactive peers, when the confounders of maturation and size were controlled (3). However, when the data were analyzed using the multilevel modeling procedure, an age-dependent effect of PA on bone mineral accrual was only found in girls at the total body (7). These findings suggest that the higher bone mineral accruals observed in the most active children are compatible with the view that bone development is driven by muscle development. Given the present findings, we further postulate that the muscle development driving bone development is driven in part by PA. However, it is important to note that the data do not exclude the hypothesis that the two processes are independently determined by genetic mechanisms. A major limitation in this study pertains to the assessment of habitual PA. There are several tools to quantify PA, including subjective and objective measures (46). These tools also fall across a spectrum of feasibility and accuracy. In 1991 when the study was initiated, the technology with regards to objective measures was not as sophisticated as it is today; thus this study used a subjective survey to quantify habitual PA. This is a common limitation of longitudinal studies in that the instruments chosen at the beginning of the study are not necessarily the ones that would be chosen at the end of the study. Although this instrument (PAQ-C/A) has consistently demonstrated acceptable internal consistency in diverse samples of children and modest validity compared with a variety of other instruments, other objective instruments (e.g., accelerometry) would provide more accurate measures. However, two or three activity assessments on any given subject were provided in any one year. Thus the estimate of habitual PA was generally more representative of typical activity. In addition, it is also important to note that this study examined habitual, free-living PA as opposed to exercise training or sport. Here it is important to recognize that an exercise training program is brief in duration (e.g., 3–6 mo) in the lives of children and adolescence, and sport occupies only a portion of the total daily activity (55). In summary, this study is unique in examining longitudinally the influence of PA on the development of lean mass during adolescence. The results of this study and previous studies from the PBMAS showing the influence of PA on bone mass (3) and FM (31) indicate the positive influence of PA on the growth of body composition during adolescence. GRANTSThis study was supported in part by grants from the Canadian Institute of Health Research, the Saskatchewan Health Research Foundation, and the Canadian National Health and Research Development Program. FOOTNOTESPBMAS group members include D. A. Bailey, A. D. G. Baxter-Jones, P. E. Crocker, K. S. Davison, D. T. Drinkwater, E. Dudzic, R. A. Faulkner, K. Kowalski, H. A. McKay, R. L. Mirwald, W. M. Wallace, and S. J. Whiting. REFERENCES
Page 4the two main energy sources oxidized during aerobic exercise are fat and carbohydrate, and their relative proportions depend primarily on the intensity of exercise (45). As the work rate increases, there is a progressive increase in the ratio of carbohydrate to fat oxidation (17). Although the relative rate of fat oxidation decreases with increasing exercise intensity, the absolute rate (e.g., mg/min) increases from low to moderate intensity and then decreases as the exercise becomes more intense (1, 45, 55). Using a graded exercise protocol that allows for the estimation of fat oxidation rates over a wide range of intensities, Achten et al. (1) found that fat oxidation in endurance trained men peaks at 64% of peak aerobic power (V̇o2 peak), with maximal rates of ∼0.60 g/min observed. High rates of fat oxidation in these trained cyclists were observed over a wide range of exercise intensities (i.e., between 55 and 72% V̇o2 peak) with rates decreasing rapidly toward zero by ∼90% V̇o2 peak. In addition to the exercise intensity, other variables such as the duration of exercise (45), sex (51), nutrient intake (43), fitness level (13), mode of exercise (5), and the level of body adiposity (29) also alter the mix of substrates oxidized during prolonged exercise. The exercise intensity that elicits maximal fat oxidation (Fatmax) is influenced by sex, the amount of fat-free mass and fitness level, according to a large cross-sectional study, although these variables only explain ∼12% of the variance (55). Our laboratory (52–54, 54) and others (9, 24, 25, 37, 39, 40, 44, 48) have shown that the effects of biological age and/or maturational status also impact the mix of fuel utilization during exercise. Indeed, for nearly 70 yr, it has been known that children have a lower respiratory exchange ratio (RER) than adults during submaximal exercise performed at the same relative intensity (i.e., percentage of V̇o2 peak) (41). The lower RER in children, compared with adults, suggests that children oxidize more fat and less carbohydrate at a given intensity of exercise. Although important, these studies are limited because they frequently compare substrate oxidation at only one exercise intensity (i.e., moderate) and do not attempt to quantify peak fat oxidation rates or the exercise intensity that corresponds to that peak rate (i.e., Fatmax). Recently, Stephens et al. (48) found that early and midpubertal boys have higher relative rates of fat oxidation than either late pubertal or adult male subjects during cycling at 30, 40, 50, 60, and 70% V̇o2 peak. Although important, this study did not determine Fatmax precisely, nor did it assess whether maturational status or age influences Fatmax in youth. Further study of pubertal effects on energy metabolism during exercise is important for a number of reasons. First, puberty is associated with an increase in insulin resistance (6) and alterations in substrate partitioning in response to lipid (7) and glucose (8) infusions. These studies suggest that pubertal development is associated with a conservation of carbohydrate stores, most likely for the energy requirements of growth. Second, knowledge of substrate metabolism in relation to pubertal status in healthy children is likely to be of clinical relevance in a variety of metabolic-related conditions. For example, our laboratory has assessed substrate utilization in children with Type 1 diabetes mellitus and found that, despite elevated circulating insulin levels, both endogenous and exogenous carbohydrate oxidation during exercise is lower than in children without diabetes (42). Finally, a better understanding of how puberty influences substrate utilization during exercise in healthy-weight children will also help clarify the potential consequences of excess adiposity on fuel selection during exercise in overweight children. To date, all studies of the influence of pubertal status and age on fuel metabolism during exercise have been conducted using cross-sectional designs, thereby limiting the ability to account for between subject variability in any “genetic” variation in substrate metabolism. The primary purpose of this longitudinal study was to profile carbohydrate and fat oxidation rates over a wide range of relative exercise intensities in a small cohort of growing, healthy, normally active prepubertal boys and compare these values with that of young adult male subjects. We hypothesized that, when expressed in relative terms (mg·kg lean body mass−1·min−1), fat oxidation rates would be higher in prepubertal boys compared with men throughout a wide range of exercise intensities and that peak fat oxidation would occur at a higher percentage of their V̇o2 peak. We also hypothesize that as these boys matured, fat oxidation rates would decrease to values similar to those observed in young men. METHODSInitially, six healthy, normally active boys, all prepubertal (i.e., Tanner stage 1, ages 11 or 12 yr) were recruited to participate in a 3-yr longitudinal study that began in late 2003. Because of a non-life-threatening injury, one boy dropped out of the study between year 1 and year 2, leaving n = 5 for the analysis shown. The reason for the small sample size was to pilot the effects of maturational status on substrate utilization using a previously established protocol. Substrate oxidation rates were assessed on an annual basis (see below) as they developed through puberty over a period of 3.5 yr. As a comparator, nine healthy, normally active, adult male subjects (ages 22–26 yr) also participated, but they were studied only on one occasion (5 men were measured in 2003, and 4 men were measured in 2006). This study was approved by the York University Human Participants Ethics Review Committee. The pubertal status of the boys was established annually according to pubic hair development by the criteria of Tanner (50), assessed by a parent and the child. Tanner self-staging has been validated as a reasonable tool for the assessment of pubertal development (38) and correlates with serum testosterone levels in maturing boys (unpublished observations). Our laboratory has used this self-assessment method previously (53, 54), which we consider appropriate for use in a research setting because it minimizes feelings of embarrassment for the subjects. All of the boys progressed to Tanner stage 4 at a similar rate by the end of the study, according to their own self-reported staging. All participants were healthy, of normal weight, not taking any medication, and recreationally active but not endurance-trained athletes. After the purpose, procedures, and risks of the study were explained to each subject, the men signed an informed consent, while the boys assented verbally to participate. Each boy's parent then signed an informed consent on his or her son's behalf. Subjects reported to the laboratory in a fasted state (∼0800) on each occasion for substrate oxidation measures using a standardized protocol (see below). Also during each visit, the following measurements were conducted: height (stadiometer), body mass (Seca digital scale), percent body fat [boys: sum of 2 skinfolds (46); men: sum of 3 skinfolds (31)]. Subjects arrived at the laboratory in the morning of the experimental trial in a fasted state. They were asked to avoid strenuous exercise the day before the trial. Before beginning the graded exercise test, subjects would practice breathing into the metabolic mouthpiece and cycle at a low wattage to become accustomed with the cycle ergometer. After sitting quietly for 20 min, subjects performed a graded exercise test to volitional fatigue on an electromagnetically braked cycle ergometer with continuous gas collection and heart rate monitoring, according to a modified protocol developed by Achten et al. (1). For this, boys (Tanner stage 1 and 2/3) started cycling at 12.5 W, and the work rate was increased 12.5 W every 3 min. The men started at 25 W, and the work rate was increased 25 W every 3 min. To help keep the exercise task duration to ∼30 min, the watt increment was changed to 20 W when the boys were tested at Tanner stage 4. In all subjects, when the RER was ≥1.00, indicating zero fat oxidation, the work rate was increased by the same increments at 1-min intervals until volitional fatigue. The purpose of this modified protocol was to profile both substrate oxidation, at least to the anaerobic threshold, and to determine V̇o2 peak, as has been described by Venables et al. (55) for use in adults. V̇o2 peak was considered to have been reached when the RER was >1.05 and the subject achieved his age-predicted maximal heart rate [heart rate maximum/(220−age)]. It is important to note that this method may underestimate the true aerobic capacity (i.e., V̇o2 max) of the subjects because of the exercise modality (i.e., cycling compared with treadmill running) and because of the somewhat prolonged test duration. Peak work rate during the incremental test was calculated from the last completed work rate plus the fraction of time spent in the final noncompleted work rate multiplied by the work rate increment. Single-lead heart rate ECG and breath-by-breath volume and gas analysis were recorded continuously using a calibrated SensorMedics metabolic cart (Vmax 29). The volume sensor and gas analyzers of the system were calibrated using a 4-liter calibration syringe and commercially prepared gas mixtures of known concentration certified to within 0.02% (cylinder 1: 25.9% O2,-0.0% CO2,-balance N2; cylinder 2: 16.00% O2-4.00% CO2,-balance N2). O2 consumption (V̇o2) and CO2 production (V̇co2) were averaged over the final 30 s of each work rate, and the results of this test were used to calculate fat oxidation over a wide range of exercise intensities for each subject (1). Indirect calorimetry, as was used in this study, is the standard method to quantify substrate oxidation rates at rest and during exercise (26), although it does have its limitations during high-intensity exercise (34) (see discussion). Fat oxidation was calculated using the following equation (26) with the assumption that the urinary nitrogen excretion rate was negligible:
For each individual, a best-fit polynomial curve was constructed of fat oxidation rate (expressed as mg·kg lean body mass−1·min−1) vs. exercise intensity (expressed as %V̇o2 peak), as illustrated in the example in Fig. 1. Because of the likelihood that indirect calorimetry underestimates fat oxidation at exercise intensities above the “anaerobic threshold” because of a reduction in the functional size of the bicarbonate pool (34), we profiled the curves only up to 75% of the subjects V̇o2 peak. Each individual curve was then used to determine the peak fat oxidation rate and the exercise intensity that was associated with maximal fat oxidation rate (Fatmax), according to the protocol of Achten et al. (1). The average R2 for the fitted curves in the boys were: Tanner 1 = 0.75 ± 0.12, Tanner 2/3 = 0.85 ± 0.16 and Tanner 4 = 0.79 ± 0.13. The average R2 for the fitted curves in the men was 0.81 ± 0.10. Individual results were then used to compose average fat oxidation curves for the men and for the boys at each stage of maturation. To do this, the Fatmax and the exercise intensities that elicited fat oxidation rates at 95, 90, and 80% of peak fat oxidation to the left side of Fatmax were determined for each individual (1). In addition, the exercise intensity that 95% of fat oxidation rates above Fatmax was also determined (1) (Fig. 1). These specific points were averaged separately for both the boys (at each Tanner stage) and for the men and then plotted against the average fat oxidation rates (expressed as mg·kg body mass−1·min−1). Similar graded exercise tests in children (16, 48, 59) and in adults (1, 4, 5, 55) have been used to quantify substrate oxidation with the following limitations: 1) substrate oxidation rates may not be valid because of bicarbonate buffering at high intensities (i.e., >70% V̇o2 peak), 2) the duration of the exercise may have been too long to have allowed for subjects to reach their true cycling V̇o2 peak, and 3) the progressive nature of the exercise may not allow for an accurate measure of substrate oxidation at steady state, because time and duration are known to influence substrate utilization rate. Fig. 1.Example fat oxidation rate profile during 12.5-W increments in an adolescent boy, plotted against peak aerobic capacity (%peak V̇o2). Fat oxidation rates were averaged over the final 30 s of each increment up to 75% of the individual's peak O2 consumption (V̇o2peak), and a best-fit polynomial curve was created to determine the following points: peak fat oxidation rate and fat oxidation rates at 95% [both to the left and right side of peak fat oxidation (Fatmax)], 90% (left of Fatmax only) and 80% (left of Fatmax only) of peak fat oxidation rate. The corresponding exercise intensities to these points were also determined, including the exercise intensity that elicited Fatmax. Experimental data are expressed as means ± SD. Significant differences in anthropometric data, functional data, Fatmax, and the calculated fat oxidation rates between the boys at different stages of maturation were determined using a one-way repeated-measures ANOVA with three levels of maturation (Tanner 1 vs. Tanner 2/3 vs. Tanner 4). For comparisons between the boys at different maturational stages and the men, a one-way between-factor ANOVA (4 levels of group) was used in a separate analysis. This two-step analysis allowed us to accommodate our mixed-methods design (repeated measures in the boys and between measures in the boys vs men). Differences between mean values were determined if a significant F-ratio was found by using Tukey's multiple comparison test. All statistical procedures were computed using GraphPad Prism software (GraphPad Software 7, San Diego, CA). RESULTSLean body mass increased in the boys from Tanner stage 1 (33.5 ± 5.6 kg) to Tanner stage 4 (48.3 ± 8.6 kg) (P < 0.05). As expected, the boys during Tanner stage 1 and Tanner stage 2/3 had a lower lean body mass than the men (65.7 ± 9.0 kg) (both P < 0.05). Other physical and functional characteristics of the subjects are summarized in Table 1. During exercise, boys at Tanner stage 1 had a lower absolute peak work rate (P < 0.001) but were similar in fat mass, relative V̇o2 peak, and percentage of age-predicted heart rate maximum to the men. As expected, over the study period, the boys grew in height (P < 0.001), weight (P < 0.001), and lean body mass (P < 0.001) but still remained shorter (P < 0.05) and lighter (P < 0.05) than the male subjects by the time they reached Tanner 4. The exercise duration was 27.5 min (range 24.5–35.3 min) for the boys at Tanner stage 1. The exercise duration was 28.4 min (range 22–32 min) in the men (not significantly different). The duration was similar at 34.5 min (range 27.5–39 min) by the time the boys reached Tanner stage 4 (P > 0.05). Peak power output was higher in the men compared with the boys in Tanner 4 (P < 0.05). Peak heart rate during exercise and percentage of age-predicted heart rate maximum was similar between the boys at all stages of puberty and the men.
Figure 2 illustrates the relationship between relative fat oxidation rate (expressed in mg·kg lean body mass−1·min−1) and exercise intensity (expressed as a percentage of V̇o2 peak) as determined by the polynomial curve-fitting method described above. The peak fat oxidation rate was approximately twofold higher (P < 0.05) in the boys in Tanner stage 1 (8.6 ± 1.5 mg·kg lean body mass−1·min−1) compared with the men (4.2 ± 1.1 mg·kg lean body mass−1·min−1). Peak fat oxidation rate was similar between Tanner 1 and Tanner 2/3 (7.6 ± 0.6 mg·kg lean body mass−1·min−1), but it decreased significantly (P < 0.05) by Tanner 4 (5.4 ± 1.8 mg·kg lean body mass−1·min−1) to values similar to those observed in the men. The exercise intensity that elicited peak fat oxidation rate (i.e., Fatmax) was higher in the boys at Tanner stage 1 (56 ± 6% V̇o2 peak), stage 2/3 (55 ± 2 V̇o2 peak), and stage 4 (45 ± 10% V̇o2 peak), compared with the men (31 ± 4% V̇o2 peak) (main effect of group P < 0.0001). Repeated-measures ANOVA on the effect of Tanner stage revealed a trend for a decrease in Fatmax with maturation (P = 0.058). High rates of fat oxidation (i.e., within 5% of peak) were observed between 47 ± 6 to 65 ± 7% V̇o2 peak in the boys in Tanner 1 compared with 27 ± 4% to 37% ± 5% V̇o2 peak in the men (both P < 0.05). The mean heart rate at Fatmax was also higher (P < 0.05) in the boys at Tanner stage 1 (131 ± 16 beats/min) than in the men (100 ± 15 beats/min), corresponding to 68 ± 9 and 59 ± 10% of peak heart rate, respectively (P < 0.05). Fig. 2.Fat oxidation rates versus exercise intensity expressed as percentage of V̇o2 peak in boys (n = 5) at Tanner stage 1 (•), the identical boys a year later at Tanner stage 2/3 (▴), and during the last year of testing at Tanner stage 4 (▪). Untrained men (n = 9) were used for a comparison (◊). Values are mean ± SD. The 3 symbols to the left of the apex of the curve indicate 80, 90, and 95% of peak fat oxidation, while the 2 symbols to the right indicate 95% of peak fat oxidation. Peak fat oxidation rate was unchanged between Tanner stages 1 and 2/3 but decreased significantly by Tanner stage 4 (P < 0.05) to values similar to the men. Fatmax was higher in the boys at Tanner stage 1 and 2/3 compared with the men (between-measures ANOVA, main effect of group, P < 0.0001). There was a tendency for Fatmax to decrease as the boys developed from Tanner stage 1 to Tanner stage 4 (repeated-measures ANOVA, main effect of group, P = 0.06). As a second method of determining Fatmax and peak fat oxidation rates, independent of the polynomial curve-fitting technique described above, we also averaged the highest recorded fat oxidation rates and their corresponding exercise intensities, collapsed over the final 30 s of each workload. Using this simple technique, peak fat oxidation rates were 9.6 ± 2.2, 8.4 ± 0.6, 5.8 ± 1.3, and 4.7 ± 0.8 mg·kg lean body mass−1·min−1 in the boys at Tanner stage 1, 2/3, 4 and men, respectively (P < 0.0001). The relative exercise intensities corresponding to these peak fat oxidation rates were 59 ± 17, 63 ± 13, 50 ± 21, and 38 ± 17% of V̇o2 peak in the boys at Tanner stage 1, 2/3, 4 and men, respectively (P = 0.06). DISCUSSIONTo our knowledge, this is the first study to examine the effects of maturation on peak fat oxidation rate and the exercise intensity that elicits Fatmax during exercise. We found that, compared with men, prepubertal boys had considerably higher relative rates of fat oxidation (mg·kg lean body mass−1·min−1) throughout a wide range of exercise intensities (Fig. 2). Furthermore, we found that the peak fat oxidation rate in prepubertal boys occurs at a significantly higher relative metabolic rate than in men. Finally, we demonstrated that in these same healthy growing adolescent boys, both the peak fat oxidation rate and Fatmax decrease with advancing pubertal stage, with the greatest decreases occurring during the final stages of puberty. It has long been known that children have a higher rate of fat oxidation during exercise compared with adults. Higher relative rates of fat oxidation in children, compared with adults, have been reported previously using a variety of steady-state (9, 24, 37, 54) and non-steady-state (39, 40, 44) exercise protocols. Previously, our laboratory found that boys use ∼70% more fat and ∼23% less carbohydrate compared with men during prolonged moderate-intensity exercise, performed after a small standardized meal (54). The higher relative rate of fat oxidation in children, compared with adults, occurs even when exogenous carbohydrates are consumed during exercise (53, 54). In this study, we extend these observations by showing that throughout a wide range of relative exercise intensities, fat oxidation rate, expressed relative to lean body mass, is approximately twofold higher in prepubertal boys compared with men (Fig. 2). This dramatically higher fat oxidation rate drops as the boys develop through puberty to values similar to what is observed in the men. In addition to the higher relative rate of fat oxidation in prepubertal boys, we also found that Fatmax was considerably higher in the boys at the early stages of puberty compared with the men (Fig. 2). Duncan and Howley (20) also reported fat oxidation rates in prepubertal children during five, 6-min steady-state exercise tasks and estimated that the peak fat oxidation rate occurs somewhere between 50 and 65% V̇o2 peak, although the nature of their protocol did not allow them to determine the Fatmax in their subjects. Stephens et al. (48) recently showed that fat oxidation rate was higher in prepubertal boys compared with pubertal boys at the same relative exercise intensity, and based on their RER values collected, it appeared that maximal fat oxidation occurred at somewhere between 40 and 70% V̇o2 max in prepubertal boys. Our observation that peak fat oxidation rate occurs at 56% V̇o2 peak in normally active prepubertal boys, based on our curve-fitting technique, is also comparable to the Fatmax of 64% of V̇o2 peak reported for elite male cyclists using a similar protocol (1). Moreover, like the elite male athletes studied previously, the untrained boys in our study had high rates of fat oxidation (i.e., within 5% of the highest fat oxidation rate observed) up to at least 65% V̇o2 peak. Together, these data suggest that normally active prepubertal boys possess the capacity for high fat oxidation rates even at high work rates similar to that observed in endurance-trained men. The average peak fat oxidation rate in the normally active boys (Tanner stages 1–4, 6.1 mg·kg−1·min−1) in the present study and in the trained men (∼8.0 mg·kg−1·min−1) in a previous study (4) contrasts with the considerably lower Fatmax (∼4.0 mg·kg−1·min−1) observed in the untrained men in the present study. Compared with untrained individuals, trained individuals have higher absolute rates of fatty acid oxidation during exercise performed at the same relative work rate, in men (33) and in women (49). This greater reliance on fat as fuel occurs because of a higher absolute work rate performed in the trained vs. untrained individuals and possibly because of training-induced adaptations in the skeletal muscle of trained individuals (30, 32). We also found that the Fatmax is lower in the untrained men in our study (31% V̇o2 peak) compared with the trained subjects used by Achten et al. (1) (64% V̇o2 peak). In a large cross-sectional study conducted by Venables et al. (55), Fatmax during treadmill running occurred at 45 and 52% V̇o2 max in men and women, respectively. Taken together, these findings suggest that either the level of habitual physical activity and/or the aerobic capacity influences Fatmax. It is unclear whether these variables would influence the fat oxidation rates in children who are already achieving high rates of fat oxidation during high-intensity exercise. Similar to other studies measuring fat oxidation rates over a wide range of exercise intensities, we found considerable interindividual differences in the range of exercise intensities at which high rates of fat oxidation occur (Fig. 2). Differences in V̇o2 max, daily physical activity levels, nutritional status, and perhaps a large genetic component appear to contribute to the variance in Fatmax and overall fat oxidation rates during exercise (55). It is important to note, however, that we attempted to match the men in our study to the boys measured in Tanner 1 in percent body fat and relative V̇o2 peak (Table 1) in an attempt to limit the influence that these variables may have on fat oxidation rates. Even as the boys progressed through puberty, V̇o2 peak values remained relatively constant (Table 1), yet both the Fatmax and the relative fat oxidation rate dropped significantly. It is currently unclear whether changes in spontaneous physical activity patterns, dietary factors, growth-related changes, and/or changes in skeletal muscle gene expression caused by pubertal hormones might influence whole body fat oxidation during physical activity in adolescents (41). Several hypotheses may explain the higher rate of fat oxidation commonly observed during exercise in young children compared with adults. First, based on limited biopsy data collected from 6-yr-old children, prepubertal children may have an enhanced ability to oxidize fat because they have higher intramuscular triglyceride (IMTG) stores (11). This hypothesis is supported by a recent study of adult women in whom a higher resting concentration of IMTG resulted in a greater use of this fuel during subsequent submaximal exercise (47). Muscle biopsy samples may not be reflective of whole muscle IMTG stores, however, and the relative contribution of this fuel source to whole body RER during exercise is unclear. Second, higher rates of fat oxidation in children may result from a higher free fatty acid turnover during exercise (18), although elevated plasma lipids in youth compared with adults are generally not reported (14, 37). Finally, higher rates of fat oxidation in children may be a consequence of an underdeveloped glycogenolytic and/or glycolytic system (12, 21, 22, 28). In support of the later point, children have consistently been reported to have lower lactate levels during exercise (23, 36, 57), and, as pointed out by Astrand (10), the utilization of fatty acids is thought to be reciprocally related to the rate of anaerobic glycolysis and lactate production. Interestingly, plasma lactate levels correlate negatively with fatty acid oxidation rates during exercise (3, 15), suggesting that the lower lactate production in children helps facilitate higher rates of fat oxidation. Finally, it is important to note that a lower RER during exercise might not be explained entirely by a greater oxidation of fatty acids in lower limb exercising muscle per se as measurements of lipid oxidation by muscle only explain ∼25% of the total lipid utilization as measured by RER (27). Future studies are needed to determine the mechanism by which the high rate of fat oxidation during exercise is diminished with physical maturation. This study has some important clinical considerations that need to be acknowledged. In children, high rates of fat oxidation during physical activity may have some beneficial health effects. High rates of fat oxidation during physical activity may limit fat mass gain and appear to be protective against the development of obesity and obesity-related disorders. Indeed, the inability to oxidize fat appears to be an important factor in the etiology of obesity and Type 2 diabetes mellitus (35, 60). Specifically, an elevated 24-h RER, indicative of reduced levels of fat oxidation, is associated with a high rate of weight gain in adult Pima Indians (60) and has been shown to be associated with increased insulin resistance (35). In contrast, a high rate of fat oxidation during exercise is associated with increased insulin sensitivity at rest as well as reduced hypertension and lower levels of plasma low-density lipoproteins (2). Obese children have been shown to have lower rates of fat oxidation compared with lean controls, and this “metabolic defect” could be reversed by 2 mo of endurance exercise training (16). In obese adults, exercise training at Fatmax increases peak fat oxidation rate and improves insulin sensitivity more than eucaloric interval training at an exercise intensity corresponding to a lower rate of fat oxidation (56). Future studies should emphasize how pubertal status influences fat oxidation rates during rest and physical activity and whether decreases in fat oxidation with age contribute to increases in fat storage and obesity-related disorders. However, because there is little evidence of age- or maturation-related differences in fat oxidation at rest, our findings highlight the importance of regular physical activity during childhood to maintain the capacity for fat oxidation and reduce the risk of metabolic disease. This study also has some important limitations that need to be recognized. First, we had a very small sample of maturing boys that may not be representative of that population. A larger sample size with an adequate representation of both sexes is clearly needed. Moreover, we asked these participants to self-report their Tanner stage, which is not as accurate as having a trained pediatric endocrinologist conduct the staging (19). Second, because of our study design that utilized a prolonged and progressive exercise test to determine both Fatmax and V̇o2 peak in one visit, we may have overestimated Fatmax in our subjects. The average duration of the exercise test was ∼30 min in all of the subjects, rather than the 10- to 12-min range that is recommended for measurement of aerobic capacity with a progressive exercise task to exhaustion. As a result, it may be that had the rates of fat oxidation were plotted against a more accurate estimate of V̇o2 peak, as determined by a more traditional exercise test that we typically use to measure aerobic capacity in adolescents, the Fatmax would shift left on the figures shown (Figs. 1 and 2). However, we are confident that all of the subjects achieved at least near V̇o2 peak for cycling, because all subjects had RER values over 1.00 at exhaustion and had heart rates that were close to their age prediction (Table 1). Moreover, as mentioned above, our determination of Fatmax appears well in line with what has been estimated previously by Duncan and Howley (20) and by Stephens et al. (48). Third, although indirect calorimetry is used extensively to determine substrate oxidation during exercise in both children and adults, it does have some important limitations. Care must be taken to ensure that subjects attain a steady state at each progressive increase in work rate. To address this issue, we choose to have participants exercise at each stage that profiled fat oxidation rates for 3 min to allow for achievement of steady state. It is important to note, however, that the achievement in steady state may occur more quickly in children compared with adults because the former tend to store less CO2 during exercise (58). In addition, changes in the size of the bicarbonate pool at higher exercise intensities invalidate the calculations of carbohydrate and fat oxidation, particularly at intensities above the anaerobic threshold. Because of the concern that the use of indirect calorimetry may be inaccurate in assessing fuel utilization above the “anaerobic threshold,” we choose not to profile fat oxidation rates at exercise intensities above 75% of V̇o2 peak in this study, as has been done by others previously (1, 4, 55). Indeed, fat oxidation rates may be underestimated at exercise intensities above the ventilatory threshold because RER increases in response to the bicarbonate buffering of the H+ production and the increased release of CO2 in the expired gas. In summary, we found that compared with men, young normally active boys have higher relative rates of fat oxidation throughout a wide range of exercise intensities and that the exercise intensity that elicits peak fat oxidation rate is considerably higher in boys than in men. Using these same boys studied in a longitudinal design, we show that the high rate of fat oxidation is diminished during maturation, particularly as boys develop from midpuberty (Tanner stage 2/3) to full maturation (Tanner stage 4). This comparatively higher rate of fat oxidation during exercise in prepubertal and early pubertal boys may have important consequences in storing less fat in adipose tissue and may be protective in the development of obesity and its related disorders. GRANTSThis work is supported by Natural Sciences and Engineering Research Council of Canada Operating Grant 261306 and by a grant from the Canadian Foundation for Innovation (to M. C. Riddell). FOOTNOTESThe extraordinary effort and time of our subjects are gratefully acknowledged. REFERENCES
Page 5No doubt dynamic hyperinflation and lack of oxidative capacity of skeletal muscles are important causes of exercise limitation in COPD. O'Donnell and Webb and Debigaré and Maltais will convince the reader of this by the elegant experiments they have performed. The thesis we will put forward is that during the natural history of COPD the primary factors leading to impairment of exercise performance are an increase in energy demands combined with a decrease in supplies and that both of these result from excessive recruitment of expiratory muscles. We argue that both dynamic hyperinflation and reduced oxidative capacity are secondary adaptations resulting from this primary abnormality. Energy demands are increased in COPD because of the high O2 cost of breathing (V̇o2resp). In health, V̇o2resp is only 1–3 ml O2/l breathed, whereas in COPD it has been reported variously to average 6.3, 9.7, and 16.4 ml/l breathed with individual values ranging from 3.0 to 19.5 ml/l (8, 17). The large between patient range in V̇o2resp probably reflects variation in the work of breathing (Wresp). During exercise, a large variation in Wresp certainly exists. In two studies, COPD patients formed two distinct groups: those that strongly recruited abdominal muscles and those that did not (5, 10). In the first (5), at an exercise workload of 10 W the work performed on the lung averaged 754 cmH2O·l−1·min−1 in recruiters but only 277 cmH2O·l−1·min−1 in nonrecruiters, although ventilation was similar. Expiratory muscle activation is the normal response to exercise (1) so the recruiters behaved normally. The problem is that in COPD, it fails to increase ventilation, because expiratory flow becomes limited by high pleural pressures. While abdominal muscle recruitment is beneficial during exercise in health (1), it is definitely harmful in COPD (4, 5). Because Wresp was 2.7-fold greater in recruiters, we can assume that their V̇o2resp was twice as high as the nonrecruiters. Let's also assume that it was 12 ml O2/l in the former and 6 ml/l in the latter. The maximal exercise workload (Wmax) was 20 and 35 W in recruiters and nonrecruiters (P < 0.05), while V̇e at Wmax was 35.9 and 37.9 l/min, respectively (5). Thus the estimated V̇o2resp was 430.8 ml/min in recruiters but only 227.4 ml/min in nonrecruiters. From the measured values of V̇o2 at rest and during 10 W exercise and assuming that V̇o2 increased linearly (dV̇o2/dwatt is constant) the V̇o2 at maximal exercise workload (V̇o2max) was 830.0 and 1,327.5 ml O2/min, respectively, in recruiters and nonrecruiters. Subtracting V̇o2resp from V̇o2max reveals that if the respiratory muscles received all their demands there was only 399.2 ml O2 available to locomotor muscles and other body tissues in recruiters but 1,100.1 ml in nonrecruiters. The respiratory muscles demanded 53% of V̇o2max in recruiters but only 17%, a value close to normal (6), in nonrecruiters. The nonrecruiters' breathing pattern was abnormal because abdominal muscles were not recruited during exercise. As a result, their exercise performance was better. However, their resting lung function was worse. Both the FEV1 and FEV1/FVC were significantly lower in nonrecruiters. This strongly suggests that as COPD progresses, patients eventually realize that abdominal muscles recruitment is bad and somehow they learn to derecruit them. Alas, without abdominal muscle contraction they dynamically hyperinflate. They can exercise a bit more, but not much (15). Thus we believe that dynamic hyperinflation results from a learned response to an inadequate supply of energy to meet demands. When normal subjects breathe with a Starling resistor in the expiratory line, which limits expiratory flow to ∼1 l/s, exercise is limited by severe dyspnea; abdominal pressure (Pab) increases abnormally; duty cycle decreases; CO2 retention occurs, increasing Pab even more (3, 13, 14); the high expiratory pressures and short duty cycle act like a Valsalva maneuver and decrease cardiac output (Q′c) (2); as a result, O2 debt is increased by 52% (22). Expiratory flow limitation (EFL) decreases the shortening velocity of abdominal muscles, and, in accordance with their force velocity characteristics Pab increases (3). Expiratory muscle recruitment can account for 66% of the variation in Borg scale ratings of difficulty in breathing (14). None of these abnormalities can be attributed to either dynamic hyperinflation or impaired oxidative capacity of skeletal muscles. Does this scenario occur in COPD? There is strong evidence that it does. First, there is uniform agreement that lactic acid production occurs at a very low exercise level in COPD. This suggests an imbalance between energy supply and demand, resulting in competition between respiratory and locomotor muscles for limited energy supplies (9, 12, 20). Administration of O2 improves exercise performance probably by decreasing V̇o2resp (7), thereby releasing more energy for locomotor muscles. This improvement should not occur if skeletal muscles were unable to use the energy available to them. Richardson et al. (19) showed that in small muscle mass exercise in COPD there was a 2.2-fold greater mass-specific power output than during whole body exercise. Locomotor muscles have a greater maximal power output in the absence of respiratory-locomotor muscle competition, Oelberg et al. (18) reported a Q′c of only 39% of predicted during exercise in COPD and when heliox was breathed, decreasing V̇o2resp and increasing the energy available to locomotor muscles, V̇o2 increased by 15% without any change in Q′c (18). If the respiratory muscles in recruiters demand 53% of V̇o2max, they probably demand the same share of Q′c (6), and if Q′c is only 39% predicted, locomotor muscles must be pretty ischemic. Finally Francois (21) himself reported a plateau in lower limb perfusion while exercise workload increased in COPD. If inadequate energy to meet demands limits exercise in COPD, why is the oxidative capacity of skeletal muscles reduced? The obvious answer is that disuse and lack of energy supplies (tissue hypoxia) cause the enzymatic changes and mitochondrial abnormalities responsible for decreasing oxidative capacity. Again there is strong evidence that this is so [see Gosker et al. (11) for an outstanding review]. The myopathic changes in congestive heart failure and COPD are almost identical. They do not occur in the diaphragm because there is no disuse of this muscle. There is no reason to believe that myopathy is a primary abnormality in COPD and congestive heart failure and every reason to believe that it is secondary to disuse and tissue hypoxia. Francois refers to this when he states “…a comparable disorder has been described in chronic heart failure. Chronic reduction in oxygen availability at the cellular level…could contribute to…skeletal muscle dysfunction” (16). Francois also recognized the potential importance of respiratory-locomotor muscle competition when he wrote that in COPD “…the respiratory muscles, with [high] V̇o2 during exercise…might…compete with lower limb muscles for the available blood flow and O2” (21). Yes, reduced oxidative capacity, like dynamic hyperinflation, can limit exercise performance in COPD, but it is secondary to a longstanding imbalance between energy supply and demand. We believe the long natural history of COPD results in the sequence of events during exercise shown in Fig. 1. The primary event, EFL during exercise, probably occurs when the disease is still mild and exercise is not seriously impaired. This in turn leads to an increase in force generation of expiratory muscles increasing expiratory pressures from which all the pathophysiology described above derives. Fig. 1.Natural history of COPD and its results in the sequence of events during exercise. REFERENCES
|
Why do children have difficulty dissipating heat and maintaining a normal body temperature during exercise than adults?
Postagens relacionadas
direito autoral © 2024 cemle Inc.
