INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

STUDIES HAVE DEMONSTRATED that children are less economical [i.e., greater submaximal oxygen consumption (O_2 submax) in ml · kg¹ · min¹] than adults at a given submaximal treadmill speed and grade (10-15). There is some evidence that these differences are a function of exercise intensity, because the change in oxygen consumption (O₂) becomes larger with increasing treadmill speeds and/or grades (12, 14). In addition to these apparent submaximal economy differences, children typically have greater maximal O₂ (O_2 max) values (ml · kg¹ · min¹), even though maximal work rates may be less than those of adults (9). The reason(s) for these apparent differences remain unclear. Researchers have suggested differences in body surface-to-mass ratio, stride frequency and other gait kinematics, substrate utilization, and respiratory function may play a role (4, 6, 10, 11). The influence of these (or other) factors on economy may be related to the fact that adult subjects have been taller and heavier than child or adolescent comparison groups (10-15). We have found no previous reports that included subjects matched for height and body weight. Therefore, the effect of body size on differences in exercise economy between children and adults is not known.

The purpose of this investigation was to examine differences in walking and running economy between adolescent female (12-14 yr) and young adult female (18-25 yr) subjects who were matched for height and body weight. We hypothesized that there would be no differences in walking and running economy between the two groups matched for body size.

A cross-sectional research design was used in this study. Data collection included several anthropometric dimensions and respiratory gas analysis during rest and exercise. The independent variable was categorical: young adolescent female or young adult female subjects. The primary outcome variable was continuous (O_2 submax in ml · kg¹ · min¹). Secondary outcome variables of interest were submaximal heart rate (HR_submax), ventilation, and stride frequency.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects. Adult subjects were recruited primarily from the undergraduate and graduate student population at a large, midwestern university. The women's ages ranged from 18 to 25 yr. Adolescent participants, between the ages of 12 and 14 yr, were recruited from middle schools in the local area. All girls self-reported as being postmenarcheal. Women and girls who ran >10 miles/wk or were members of organized cross-country or track teams within the past year were excluded from the study. Other types of training such as swimming, aerobics, cycling, and team sport practices were not considered exclusionary criteria. All adult subjects, and the parents or guardians of the adolescents involved in the study, gave written and verbal consents or assents to participate. Approval for this project was given by the University Committee on Research Involving Human Subjects.

Anthropometry. Several anthropometric dimensions were obtained before treadmill testing. All measurements followed the recommended techniques found in the Anthropometric Standardization Reference Manual by Lohman et al. (8) unless otherwise stated.

Preexercise O₂. After anthropometric measures, each participant's respiratory gases were collected for 10 min before a treadmill exercise test. The subject was seated in chair and encouraged to relax. Pilot testing in our laboratory showed <3% minute-to-minute variation in preexercise O₂ measures. After 10 min of resting gas collection, the participant was given instructions for the treadmill test.

Treadmill test. Treadmill testing was performed in a temperature-controlled laboratory (ambient temperature = 21-23°C, relative humidity = 35-50%). Before testing, the treadmill was calibrated, and speed was found to be within 1% of the expected values. Each participant had opportunity for a practice session (depending on experience) until she was comfortable with treadmill walking and running. Then, each participant performed a continuous treadmill protocol consisting of submaximal walking, submaximal running, and running to volitional exhaustion. First, each subject walked at 80 m/min at 0.0% grade for 6 min. After the walk, treadmill speed increased to 147 m/min for the 6-min running test. At the end of the submaximal treadmill run stage, grade was increased by 2.5%/min until the participant reached volitional fatigue. Each participant was verbally encouraged to continue until she could no longer maintain the appropriate treadmill speed. The subject grabbed the rails (per instructions before test initiation) when she was unable to continue. At this time, the treadmill speed and grade were decreased to 2.0 miles/h and 0.0% grade during a cooldown period.

Data collection. Each subject's expired respiratory gases were measured continuously during both the resting and exercise phases of the test by using open-circuit indirect calorimetry (model 2900, SensorMedics, Yorba Linda, CA). The flowmeter was calibrated by using a 3-liter syringe. Metabolic analyzers were calibrated before each test by using certified standard gases of known concentrations. O₂, carbon dioxide production, and respiratory exchange ratios (RER) were recorded as 20-s averages. Preexercise O₂ was defined as the average of the last 3 min of resting measures obtained. O_{2 submax} was defined as the average O₂ in the last 3 min of each 6-min exercise stage. Minute-to-minute variability averaged <3% within a given subject's O₂ values. The criteria for O_2 max included any two of the following: 1) RER 1.00; 2) maximum heart rate (HR_max) = (220  age) ± 5%; and 3) plateau of O₂ (defined as a <2 ml · kg¹ · min¹ increase with increasing workload). O_2 max was obtained by averaging the highest three consecutive 20-s O₂ values recorded in the last 2 min of the test. Heart rate (HR) was monitored continuously by using a telemetric system (Polar Vantage, Gays Mill, WI), and values were recorded each minute. Stride frequency was counted as the number of right foot strikes in 30 s. Stride frequency was measured in both the third and fourth minute of walking and running to ensure reliability. If the values were not within two strides, it was measured again during the fifth minute, and the closest two values were averaged.

Data analysis. Between-groups ANOVA was used to compare our primary (O_2 submax) and secondary (HR_submax, respiratory frequency, and stride frequency) outcome variables. Because of multiple comparisons, a significance () level of P < 0.01 was used a priori. Anthropometric variables were also examined statistically via between-groups ANOVA.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Age and anthropometric data are shown in Table 1. We were successful in matching our groups by standing height and weight, because mean values differed by <1 and 2%, respectively. Further evidence of matching is apparent because there were no significant differences between groups in any other anthropometric variables measured. Similarities in standing and seated heights resulted in leg lengths that did not differ between girls (75.3 ± 2.0 cm) and women (74.9 ± 2.6 cm).

Preexercise O₂ values did not differ significantly between groups (P > 0.01). However, mean values were slightly greater in the girls (4.4 ± 0.8 ml · kg¹ · min¹) compared with the women (3.9 ± 0.5 ml · kg¹ · min¹).

Responses to submaximal steady-state treadmill exercise are shown in Table 2. Except for tidal volume and RER, submaximal values were significantly (P < 0.005) greater in girls than in women during both walking and running. Stride frequency was also measured during steady-state exercise. Results showed similar values between girls and women during walking (60 ± 2 vs. 60 ± 3 steps/min) and running (82 ± 2 vs. 85 ± 2 steps/min), respectively.

Table 3 shows physiological responses to maximal treadmill exercise. Findings showed similar O_2 max (ml · kg¹ · min¹) values in girls and women. Other than HR_max, which was ~7 beats/min higher in the girls, there were no significant differences in any other variables measured at maximal exercise.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Our purpose was to examine walking and running economy in adolescent and young adult female subjects who were matched for height and body weight. We hypothesized that matching the groups for body size would eliminate any differences in exercise economy that had been found in previous studies. However, despite our successful matching strategy, significant differences occurred in O_2 submax (ml · kg¹ · min¹) between subject groups. The adolescent girls were less economical (greater O_2 submax) in both walking and running. These findings indicate that factors other than body size are responsible for differences in energy expenditure during exercise.

Our study design was similar to one used by Rowland and Green in 1988 (12). They compared girls and young women at three treadmill speeds, including two that were used in the present study. As previously indicated, in their study the girls were shorter and lighter than the women. The authors found that O_{2 submax} did not differ during walking at 80 m/min, but the values of the girls values were significantly higher than those of the women during running at 147 m/min. O_2 submax values of the women participating in our study, and those of the study of Rowland and Green (12), were nearly identical at ~34 ml · kg¹ · min¹. However, the values of the girls in their study (~42 ml · kg¹ · min¹) were higher than those of our study (38 ml · kg¹ · min¹). On the basis of these data, it appears that matching our subjects by size has eliminated some, but not all, the difference in O₂ between groups during running exercise.

In contrast to many previous reports (9), O_2 max did not differ between our female adolescent and women subjects. Although none of our subjects were trained distance runners, most exercised regularly. Thus both girls and women had O_2 max values that were higher than predicted on the basis of their ages. We know of no reason why similarities in O_2 max between groups would have contributed to the differences found in exercise economy.

Differences in O_2 submax between groups was not due to leg length or stride frequency because these values were similar in the girls and women. Also, similarities in biacromial breadth, biiliac breadth, and skinfold thicknesses suggest that anthropometrics were not related to economy differences. Similar skinfold thickness values do not guarantee that percent fat did not differ between groups. However, we believe that differences in intramuscular and other internal fat stores between groups was not likely responsible for decreased exercise economy in the girls.

Although leg length and stride frequency did not differ between groups, we cannot rule out the possibility that other gait dynamics (which were not measured) played a role in the higher O₂ seen in the girls. Hausdorff et al. (6) studied children (3-14 yr) and found that stride-to-stride variability decreased as a function of age. This was apparent even when the data were adjusted for standing height. No comparisons were made with adult subjects. Frost et al. (4) measured cocontraction (simultaneous activation of an agonist with its antagonistic muscle) of the leg muscles while children walked and ran on a treadmill. The authors found greater cocontraction activity in younger, compared with older, children. This was coincident with higher O₂ during the treadmill exercise. However, height and weight differences were not controlled for, and no adults were used as a comparison group.

Differences in resting energy expenditure (REE) could also affect O₂ during exercise. That is, O₂ measured during exercise includes both REE and the energy cost of the activity itself. Although we did not measure our subjects in true resting conditions (i.e., fasted state, early morning testing, and so forth), seated preexercise values indicated only a small (0.5 ml · kg¹ · min¹) difference in O₂ between groups. This accounted for ~25% of the difference in walking O₂ and ~11% of the difference during running.

As was the case with previous studies, HR_submax was higher in the girls, compared with the women. The difference averaged 14 and 17 beats/min during walking and running, respectively. Although not significant at the P < 0.01 level, HR values differed also when expressed as a percentage of HR_max. Specifically, girls' percentage of HR_max values were higher than those of our women subjects during both walking (58 vs. 54%) and running (89 vs. 84%).

Although HR values were significantly higher in the girls, it is doubtful whether this resulted in a measurable contribution to differences in exercise economy. Kitamura et al. (7) have shown that HR is positively correlated with myocardial O₂ (r = 0.88 in their study). However, when myocardial O₂ is related to whole body O₂, the difference of 14-17 beats/min seen in our study would amount to only ~10 ml/min, or ~0.2 ml · kg¹ · min¹. This estimate was derived from the myocardial O₂ data of Kitamura et al. and the left ventricular mass data of Rowland et al. (13) and Turley et al. (14).

Higher respiratory rate and E in the adolescent girls may also explain at least some of the difference in economy between our subject groups. In our study, E was ~18% higher in the girls, compared with the women. This was due to increased respiratory rate, which was 6 (during walking) and 9 (during running) breaths/min higher in the girls. Dempsey et al. (3) have shown that the energy required to increase ventilation during exercise can contribute significantly to overall O₂. The cost of exercise hyperpnea can approach 10% or more of total O₂ during intense exercise. However, as was the case with the girls' higher HR, the difference in O₂ cost of ventilation between groups did not likely exceed 0.5 ml · kg¹ · min¹ of total O₂. This estimate was derived from E differences between our groups and the data of Dempsey et al.

The girls in our study showed higher E during exercise despite the fact that submaximal RER did not differ between groups, indicating that increased breathing frequency was not likely a respiratory compensation for metabolic acidosis. More likely, regardless of their body size, the adolescent girls have an increased sensitivity to CO₂ compared with the women, resulting in greater E and lower arterial PCO₂ values. This has been shown previously by Armon et al. (1) and Cooper et al. (2) when they compared adolescent with adult subjects.

In summary, our results show that matching adolescent girls and young adult women by body size does not remove differences in exercise economy during treadmill walking and running exercise. Differences in REE, HR, and E appear to be responsible for only a fraction of the higher O₂ found in the girls. However, we did not obtain a "true" basal metabolic rate in our subjects, nor were myocardial O₂ or energy cost of breathing measured directly during exercise. Future research should examine biomechanical factors or other anatomic and physiological variables that may contribute differences in exercise economy between children and adults.