Disclosure: The QUALITY Cohort is funded by the Canadian Institutes of Health Research, the Heart and Stroke Foundation of Canada, and the Fonds de la recherche en santé du Québec.
Mechanical efficiency during a cycling test is not lower in children with excess body weight and low aerobic fitness†
Article first published online: 16 MAR 2013
Copyright © 2012 The Obesity Society
Volume 21, Issue 1, pages 107–114, January 2013
How to Cite
Jabbour, G., Lambert, M., O'Loughlin, J., Tremblay, A. and Mathieu, M.-E. (2013), Mechanical efficiency during a cycling test is not lower in children with excess body weight and low aerobic fitness. Obesity, 21: 107–114. doi: 10.1002/oby.20241
See the online ICMJE Conflict of Interest Forms for this article.
- Issue published online: 16 MAR 2013
- Article first published online: 16 MAR 2013
- Manuscript Received: 21 DEC 2012
- Manuscript Accepted: 16 MAY 2012
The aims of this study were to assess the association between (i) body weight status and mechanical efficiency (ME); and (ii) ME and aerobic fitness in children aged 8-10 years.
Design and Methods:
The sample included 464 prepubertal children (258 boys). A total of 288 were normal-weight (NW); 84 overweight (OW); and 92 obese (OB). Subjects performed an incremental maximal cycling test with indirect calorimetry. MEcrude (%) was calculated for the first five stages of the protocol (25, 50, 75, 100, and 125 W) as follows: work produced, in watts total energy consumption, in watts−1·100−1. For MEnet, resting energy consumption was subtracted from total energy consumption. Energy consumption was calculated as follows: (4.94·respiratory exchange ratio + 16.04) · VO2, in ml·min−1·60−1.
MEcrude was significantly higher in NW compared to OW and OB children and in OW compared to OB children at 25, 50, 75, 100, and 125 W. In contrast, MEnet did not differ significantly among NW, OW, and OB children. No statistically significant association was found between crude or net ME and peak oxygen consumption (VO2peak; in ml·kg−1·min−1); therefore, the ability to transfer chemical energy to mechanical work is maintained in children aged 8-10 years old regardless of body weight status and aerobic fitness. Moreover, higher values of MEcrude during exercise are explained by elevated oxygen consumption at rest and not by energy consumed during physical activity.
These results highlight that prepubertal children are equally efficient since they are able to perform a physical task such as cycling using the same proportion of energy regardless of their body weight status.
Obesity among children and adolescents is a serious problem that is present in all industrialized countries. This condition has important psychosocial and medical consequences linked to comorbidities such as early onset dyslipidemia, hypertension, and insulin resistance (1). Childhood represents a vulnerable period for the development of obesity (2). Indeed, this period induces several morphological, hormonal, and physiological modifications that can contribute to gains in fat mass and reductions in physical activity level (3). Physical activity, in addition to a nutritional program, is key to reduce body fat mass and improve physical fitness; thus, it is often recommended to obese (OB) individuals (4).
OB adolescents and adults can exhibit a considerable degree of functional limitation in motor activity due to a reduction in aerobic (5) and anaerobic (6) capacities. It is known that excessive fat mass imposes an unfavorable burden on cardiac function, oxygen uptake by working muscles, motor unit activation, and ratio of muscle strength per unit of muscle mass (7,8). One key factor studied in the context of obesity and physical ability is mechanical efficiency (ME). ME refers to the amount of work performed for a given energy consumption, or stated differently, the ability of an individual to transfer energy consumed in external work (9).
In the context of obesity, studies that examined ME are limited to the adult population. They indicate that ME is lower in OB adults compared to non-OB adults (i.e., 18% vs. 26%) (5,10,11). In OB children and adolescents, no studies looked at cycling ME and thus, none is available to support if ME is impaired. It is important to know if ME is changed in order to understand the energy required to perform a given mechanical work. For example, if a lower ME is found in OB children, it would mean that more energy is consumed at a given work output (e.g., 100 W) and that deterioration in energy transfer occurs. However, one study conducted with normal-weight (NW) children with OB parents (9) suggested that ME did not differ between the two groups. Another study examining the effect of excess body mass in overweight (OW) children on ME while cycling on a stationary bike at a mean of 57 W (60% of the child's maximal oxygen consumption) (12). This study revealed that ME is comparable in OW and NW children. The question therefore remains whether cycling ME is impaired or maintained in OB compared to NW children. If OB children do in fact have preserved ME, this may indicate that the reduction in ME probably occurs through the process of puberty or after a longer period of time is spent at an obesity status.
The main objective of this study is to document if there are differences in ME and related components (respiratory exchange ratio, oxygen consumption, and energy consumption) among OB, OW, and NW children. Two types of ME were investigated: the first, crude ME, refers to the ratio of external work on energy consumed; and the second is net ME, in which resting energy consumption is withdrawn from the energy consumed. Studies indicate that resting energy expenditure is higher in OB children and adults than in NW individuals (13). If higher energy expenditure at rest is not considered, such as in crude ME, then a lower ME can be observed during a physical task due to higher resting energy consumption and not necessarily higher energy consumption for the exercise component. Thus, net ME, for which resting energy consumption is subtracted from exercise energy consumption, was documented. A secondary objective was to document the association between aerobic aptitude level and ME profile. It is currently unknown if ME is higher in more fit children. It is hypothesized that excess body weight is associated to a lower ME and that a better aerobic capacity could be linked to higher ME.
Methods and Procedures
The Quebec Adipose and Lifestyle InvesTigation in Youth (QUALITY) Cohort study comprises 632 Caucasian children of Western European ancestry between the ages of 8-10 years (www.etudequalitystudy.ca;14). The Ethical Committees of the Sainte-Justine, Laval and CHUS hospitals approved the project. Written informed consent was obtained from the parents of each subject before the study. Participants were instructed to refrain from participation in intense physical activity the day before evaluation day.
A limit of previous studies on ME is the inclusion of children at different stages of pubertal maturity (10,12). Therefore, only prepubertal children at Tanner stage 1 evaluated by a trained nurse were included in the present study (15). Two versions of the McMaster cycling protocols were used in the study to ensure that aerobic tests were not too short or too long; one for children <160 cm and one for children ≥160 cm (16). The protocols differed in the magnitude of the intensity increment from one stage to the other; increments in watts were larger in taller individuals. In the QUALITY study, 74% of children performed the <160 cm protocol, and these children were the ones selected; thus, the final sample for this study comprises 464 children.
Children with a previous diagnosis of type 1 or 2 diabetes; with a serious illness, psychological condition, or cognitive disorder that hindered participation in some or all of the study components; on a treatment regimen with antihypertensive medication or steroids (except if administered topically or through inhalation); or on a very restricted diet (600 kcal/day) were not eligible.
Dual-energy X-ray absorption was used to assess body mass, total body percentage of body fat, and fat-free mass (Prodigy Bone Densitometer System, DF+14664; GE Lunar Corporate, Madison, WI). Body mass was measured to the nearest 0.1 kg with the subject in light clothing without shoes. Height was determined to the nearest 0.1 cm. Body weight status groups were determined based on BMI percentiles derived from the Center for Disease Control and Prevention Clinical Growth Charts for children aged 2 years and older (17) and according to Canadian guidelines (18).
Maximal cycling test
Children performed an incremental maximal test on an upright cycle ergometer to determine peak oxygen consumption (VO2peak in ml·kg−1·min−1; Oxycon pro; Jaeger, Bunnick, the Netherlands in Montreal; Cosmed, Quark B2, Italy, Rome in Quebec City). The test took place in the afternoon, at least 2 h after lunch. Breakfast was standardized. The lunch content was standardized but children were free to eat what they want from the provided lunch. Before beginning the test, children remained seated for 5 min on the bicycle ergometer in the same position used in subsequent exercise. Resting oxygen consumption was measured based on the mean oxygen consumption of the last 30 s of minutes 3, 4, and 5. No proper warm-up was performed. The test started at low intensity (initial power of 25 W) and was progressively increased by 25 W every 2 min until exhaustion (16). The only exception was for the second stage (50 W) during which duration was set at 5 min. This longer stage was performed to address question unrelated to the current study i.e. exercise energy expenditure at a stable state. During the test, children were instructed to pedal at a rate of 50-70 revolutions per minute. This rate of revolution is currently used in the pediatric population performing a cycling test (9,12). The test was terminated, and the subject entered a recovery phase when the subject requested to stop the exercise or could no longer maintain the required pedaling rate (revolutions per minute <40). Breath-by-breath automated metabolic system was used to determine VO2peak. Calibration before each test was performed with standard gases of known oxygen and carbon dioxide concentration for gas composition and a calibration syringe for volume. The data were averaged on a 30-s interval, and oxygen uptake and respiratory exchange ratio, which is the ratio of carbon dioxide produced to oxygen consumed, were obtained. Achievement of VO2peak was accepted when subjects reached a heart rate measured with an electrocardiogram above 195 beats/min or a respiratory exchange ratio greater than 1.0 at VO2peak (19).
Energy consumption and ME calculations
MEcrude was calculated for the first five workloads (25, 50, 75, 100, and 125 W) and at peak power output (Ppeak) as follows: work produced, in watts · (total energy consumption, in watts)−1 · 100-1 (10). For MEnet, resting energy consumption (Erest) was subtracted from total energy consumption at each exercise stage (9). The energy consumption (E) in watts, was calculated as follows: (4.94 · respiratory exchange ratio + 16.04) · (VO2, in ml·min)−1 · 60−1 (20). Ecrude and Enet for each workload and peak power output were also calculated. For Enet, resting oxygen consumption (VO2rest) was subtracted from the total oxygen consumption at each exercise stage. Finally, the changes in crude ME (ΔMEcrude) were obtained from the ratio of work performed above the previous workload to E above the previous work level: ΔME (%) = ΔWork produced (watts) · ΔE-1 (watts) (21). The ΔMEnet from the ratio of work performed above the previous workload to net energy consumed (Enet) above the previous work level was also calculated.
In this study, the percent contribution of lipid (% LO) and carbohydrate (%CHO) oxidation to energy yield was computed from gas exchange measurement according to the nonprotein respiratory quotient technique for the rest and the five workloads (25, 50, 75, 100, and 125 W) and at and at peak power output (Ppeak) as follows: % LO = ((1 − RER)/0.29) · 100, % CHO = ((RER − 0.71)/0.29) × 100 (22).
Data are presented as means ± SD. Analyses were performed using IBM SPSS Statistics 19 software. After testing for normality (Kolmogorov-Smirnov test), differences within and between the groups were analyzed using two-way analysis of variance, and Scheffé's test post hoc was performed. Pearson correlations were used to assess the association between aerobic fitness and net ME values. A value of P < 0.05 was set as the level of statistical significance.
The subject characteristics are presented in Table 1. Age and height were similar across body weight categories. The repartition of boys and girls among the three body weight status group was similar. As expected, OB children showed statistically significantly higher values of body mass, percentage of body fat, BMI percentiles, and fat-free mass than NW and OW children. In addition, OW children had higher values of body mass, percentage of body fat, BMI percentiles, and fat-free mass than NW children. Absolute peak power output, in watts, and VO2peak, in ml·min−1, values were statistically significantly higher in OB compared to NW children.
Respiratory exchange ratio, energy, and oxygen consumption
At rest and at all exercise levels, crude oxygen consumption was statistically significantly higher in OB group compared to NW and OW individuals (Table 2). For two stages, OW individuals also have higher crude oxygen consumption than NW children. In contrast, VO2net did not differ statistically significantly between groups for submaximal and maximal levels. No difference in respiratory exchange ratio was detected between the three body weight status groups at rest and at 25, 50, 75, 100, and 125 W.
Energy consumption (E), an indicator that takes oxygen consumption and the respiratory exchange ratio into account, followed the same trend as oxygen consumption: (i) Erest was statistically significantly higher in OB children compared to OW and NW children; (ii) Ecrude for OB children was higher at 25, 50, 75, 100, and 125 W compared to NW and OW children and was higher at 25, 50 and 75 W in OW compared to NW children; and (iii) Enet determined at 25, 50, 75, 100, and 125 W did not differ among the three groups. Finally, Peak Ecrude was statistically significantly higher in OB and OW children compared to NW children (P < 0.01). However, peak Enet did not differ among the three groups. No difference was noted in the percentage of substrate oxidation (Table 2) as well as in carbohydrates and lipids oxidation in mg per minute (data not presented).
Crude and net ME measured at submaximal and peak effort are reported in Table 3. Net and crude ME increased with increasing ergometer workload in all body weight status groups. When expressed in crude terms, the OB group exhibited a lower MEcrude at all workload stages and at peak power and this was statistically significant when compared to the OW and NW individuals. In addition, MEcrude was consistently lower in OW children compared to NW ones. When expressed per net value, ME was similar across the three groups for all submaximal stages and at peak power output. The changes in net and crude ME (ΔMEcrude and ΔMEnet) were similar at all stages among the three groups. Finally, no statistically significant relationship was found between MEnet and aerobic fitness (r = 0.01, P = 0.8). However, a positive and statistically significant relationship was found between MEcrude and VO2rest (r = 0.7, P <0.05).
To our knowledge, this study is the first to examine the ME profile in a large sample of prepubertal children with three different body weight statuses. When resting energy expenditure is not removed from exercise values, children with excess body weight exhibit a lower ME. It is thus important to know if this difference results from exercise efficiency or simply from resting profile. In fact, excess body weight does not seem to be associated to ME during a cycling task when the energy consumed at rest is subtracted. This study also suggests that ME is not related to the fitness level of children. Two important conclusions emerge, the first of which has clinical implications. The ability to transfer energy consumed to an output of energy during a physical activity task is preserved in prepubertal children during an incremental maximal cycling test regardless of the body weight status and aerobic fitness level. Second, higher values of crude VO2, E, and ME during exercise in children with excess body weight are explained by elevated oxygen consumption at rest, and not by higher energy consumed during cycling or differences in substrate utilization.
At rest, energy consumption in prepubertal children was statistically significantly higher in the OB group compared to the NW and OW groups and this could be related to their increased body mass, which requires a greater metabolic energy exchange (24). This result concord with findings obtained previously for OB adults (10) and children (23). Specifically, greater energy consumption at rest is the consequence of higher fat-free mass and muscle mass (24), increased work breathing (25), and altered substrate utilization (26). In this study, respiratory gas exchange and substrate oxidation were not affected by body weight status at rest. This finding indicates that substrate oxidation at rest seems to be not important to explain energy expenditure and ME profiles in prepubertal children. Because higher resting values could have an important impact on exercise, clinicians and researchers should take resting values into consideration. The findings of this study support the use of net rather than crude values, to better determine the physical activity response.
At all workloads, crude VO2, ME, and E were statistically significantly higher with increasing body weight status, while net values to sustain a given bicycle workload were similar regardless of body weight status (Tables 2 and 3). This result concords with those of Ayub et al. (27) showing that the net energy cost of locomotion in adolescent boys who walk at lower speeds was not affected by body weight status. Similar results were obtained in younger children performing a maximal cycling test (9). However, pediatric studies in this area have focused only on NW children at risk of obesity due to parental history. Several studies conducted with OB adults showed higher values of oxygen consumption and net energy cost compared with their leaner counterparts in response to locomotion or a cycling test (5,10,11,28). For Lafortuna et al. (10), both crude and net rates of oxygen consumption and energy consumption were higher in OB women compared to NW women in response to a cycling test. This finding represents a major difference with the present study because only crude values are different across the different body weight status groups. Then, it is possible that the energy contribution is preserved in prepubertal children during an incremental maximal cycling test regardless of their body weight status.
In previous studies, OB adults and children also exhibited significantly higher respiratory exchange ratio values at all workloads, which is a potential indicator of impaired fat oxidation capability during exercise (23,29). In the current study, no difference was detected between the three groups at 25, 50, 75, 100, or 125 W.
A rise in ME with workload is explained by a constant amount of energy required to move the pedaling legs, regardless of the ergometer load, to maintain cycling posture and to overcome the internal ergometer friction (9). As the workload increases, the proportion of this energy relative to the total energy requirement decreases, resulting in higher ME values (30). Accordingly, ME increased in this study from one stage to the next, and this increase was observed in all body weight status groups (Table 3). For the OB group, crude ME values were statistically significantly lower than those of the NW and OW groups. However, when these values were reported in net terms, there were no differences between groups. Because no differences were noted in the respiratory exchange ratio, higher values of crude ME during exercise are explained by elevated oxygen consumption at rest and not by energy consumed for physical activity. Our results concord with findings obtained previously in young children at risk but not yet OB (9). However, our results are discordant with the study of Lafortuna et al. (10) that was conducted with OB adults and showed significantly higher values of crude and net ME in NW compared to OB individuals. Again, divergence might be explained by the age difference between the various studies or longer duration of the obesity status in adults. In adults, obesity was associated with an increased proportion of glycolytic muscle fibers (31), which are substantially less efficient than type I fibers, during cycling (32). This profile could contribute to a higher cost of cycling in OB adults (10). In children, no such data are available to our knowledge.
Butte et al. (12) showed lower net ME values in OW children compared to NW children, a difference that they attributed to an excess body mass presented in OW children. However, it is important to note that the results of this study conducted with children from 4 to 19 years of age were no longer statistically significant when factors such as age were taken into account This study is thus concordant with the current work. In fact, in the present study, a well-controlled sample of children between 8 and 11 years of age and at the same pubertal stage was used. This sample is a strength of the present study, as well as inclusion of both OW and OB individuals.
Net ME for OB children did not differ compared to OW and NW individuals. Indeed, no significant correlation was obtained between net ME and aerobic fitness level across body weight groups. Therefore, the ability to transfer energy consumed to an output of energy during a physical activity task is preserved in prepubertal children during an incremental maximal cycling test, and this is true regardless of body weight status and aerobic fitness level. To our knowledge, no study has determined the effect of aerobic fitness on ME in OB individuals. There is only one available study that contradicts our findings. In fact, the study of Boone et al. (33) revealed that highly trained cyclists have a ME that is positively and significantly correlated to aerobic fitness. However, no data until now are available to explain the divergence between our finding and those of Boone et al. (33).
The delta ME, calculated by the changes of work and energy between two different levels of exercise, is an excellent indicator of the intrinsic muscle efficiency in power generation (32). No differences were noted in this study across body weight status groups. Similar findings were reported by Lafortuna et al. (10) who studied cycling in OB and NW women and by Berry et al. (28) who studied a group of women ranging from NW to OW. For these authors, the higher cost of cycling observed in OB adults may be due more to the higher energy entailed in limb movements and possibly higher muscle activation aimed at body stabilization during cycling than to intrinsic differences in muscle performance. Our results indicated that delta ME was not altered in children with OW or OB body weight status.
At maximal peak power
Ayub et al. (27) demonstrated that OB children have higher E and ME than NW children at maximal load in a running test. This finding agrees with those reported by Rowland et al. (34) who studied OB adults during cycling and walking. Both groups suggested that the energy cost of breathing is excessive in OB people. The current findings show that no differences were obtained at peak power output across body weight status groups for net oxygen consumption, E, or ME. These results suggest that in prepubertal children, impaired energy consumption is not yet present. Consequently, it is important to further determine whether these alterations appear after a long period of obesity and to identify the mechanisms responsible for these deteriorations. Another novel finding is that ME is not associated with fitness levels in prepubertal children, which generates great interest given that unfit children still have a good ability to transform chemical energy into mechanical work. This finding reinforces the fact that in young children, physical abilities while cycling are preserved. It should, however, be noted that current findings are limited to cycling and that results might differ for activities where body weight is not supported such as running.
This study is reinforced by a large sample size. There are however some limits. Firstly, the energy intake was not standardized during the lunch. Secondly, our finding might apply only to children between the ages of 8-10 years with a family history of obesity. Thirdly, the energy intake at the standardized meal prior to the test was not standardized.
To our knowledge, this study is the first to investigate ME in children of three body weight status groups. The strengths of the study include the large sample size, representation of the main body weight groups, and inclusion of children of the same pubertal stage. Our results clearly indicate that the muscular capacity to perform work for a given energy input is maintained in prepubertal children aged 8-10 years. One particularity of OB and OW prepubertal children is that they consume more energy at rest. This can be an advantage from an energy expenditure point of view, but also a disadvantage in the context of physical activity performance. From a clinical point of view, these data are valuable for the evaluation and management of OW and OB children. Indeed, the results from the present study imply that higher body weight and lower aerobic fitness do not affect the energy required to perform a given activity on a cycle ergometer when resting values are withdrawn. A longitudinal follow-up of children would enable assessment as to whether or not an alteration in energy use occurs after a prolonged period, as is observed in adults. On the basis of this study, it might be speculated that prepubertal children would benefit from physical activity intervention to prevent deterioration in the ability to transfer energy consumed into mechanical work.
The authors thank all the families and the research staff who have given their time so generously over the course of this study and Mr. Lubomir Alexandrov for his statistical support. The QUALITY Cohort is funded by the Canadian Institutes of Health Research, the Heart and Stroke Foundation of Canada, and the Fonds de la recherche en santé du Québec. J.O.L. holds a Canada Research Chair in the Early Determinants of Adult Chronic Disease.
- 14Cohort Profile: The Quebec Adipose and Lifestyle Investigation in Youth Cohort. Int J Epidemiol 2011., , et al.
- 17Clinical Growth Charts for children age 2 years and older. Centers for Disease Control and Prevention (CDC) http://www.cdc.gov/growthcharts 2002.
- 19208-209.Measurement in pediatric exercise science. Human Kinetics: Canadian Society for Exercise Physiology: Windsor, 1996 pp
- 21Physical activity patterns and aerobic fitness among prepubescents. Eur Phys Educ Rev 1996; 2: 19-29., , , .
- 22Biochemistry: A Function Approach. Saunders: Philadelphia, PA, 1983, pp. 810-976.and .
- 23Energy expenditure and physical activity of obese and non-obese Thai children. J Med Assoc Thai 2001; 84 Suppl 1: S314-S320., , , .