Abstract
There are no reports concerning the regulation of end-expiratory lung volume (EELV) and flow–volume relationships during upper limb exercise in health and disease. We studied EELV during such exercise in 22 adults with cystic fibrosis (CF) and nine age-matched healthy control subjects. Subjects with CF were grouped according to the severity of their lung disease, as follows: mild = FEV1 > 80% predicted; moderate = FEV1 40 to 80% predicted, and severe = FEV1 < 40% predicted. EELV was calculated from measurements of inspiratory capacity (IC) made at each workload during an incremental arm and leg ergometer test to peak work capacity. In the control group, the decrease in EELV was significantly smaller for arm than for leg exercise at peak work ( − 0.13 L versus − 0.53 L, p < 0.001) and for arm than for leg exercise at an equivalent submaximal ventilation ( − 0.13 L versus − 0.46 L, p < 0.01). In the groups with moderate and severe CF, arm exercise resulted in an increase in EELV from resting levels (dynamic hyperinflation) that was not significantly different from the increase observed for leg exercise. For CF subjects there was a significant inverse relationship between FEV1 and changes in EELV from rest to peak arm exercise (r = − 0.46, p < 0.05). In normal subjects, there was a difference in the EELV response for arm versus leg exercise. In CF subjects with airflow limitation, dynamic hyperinflation occurred with both forms of exercise.
The ability to perform upper-limb exercise is important for daily activities. Despite this, there is limited information about the physiologic response to upper-limb exercise. In particular, there are no reports concerning the regulation of end-expiratory lung volume (EELV) and flow–volume relationships during upper-limb exercise in either health or disease. Changes in EELV are an important determinant of inspiratory muscle length and airway caliber, affecting both the mechanical efficiency of breathing and the distribution of inspired flow (1). The decrease in EELV seen during lower-limb exercise in people with normal lung function (1-4) optimizes diaphragm length, and the abdominal wall recoil following expiration aids the subsequent inspiration (2, 5). The increase in EELV seen during lower-limb exercise in patients with severe obstructive lung disease (6-9) aids in minimizing expiratory flow limitation, but the dynamic hyperinflation in such patients causes the muscles of inspiration to be in a more shortened state. This is mechanically disadvantageous and may be a constraint to exercise performance. EELV during lower-limb exercise has been shown to change according to the severity of lung disease in patients with cystic fibrosis (CF) (10).
Upper-limb exercise capacity is reduced in patients with CF who have more severe lung disease (11). Information about changes in EELV is relevant to possible mechanical constraints to upper-limb exercise in this patient group, and to a better understanding of the physiologic responses to upper-limb exercise across different levels of disease severity.
The aim of the present study was to define the changes in EELV during upper-limb exercise in normal subjects and in CF patients with lung disease of varying severity. In addition, we wished to study these changes in conjunction with flow– volume relationships during upper-limb exercise.
Twenty-two subjects with CF (16 males, and six females), ranging in age from 17 to 44 yr, (mean age ± SEM: 24 ± 2 yr) agreed to participate in the study. All subjects had had CF diagnosed through a positive sweat test and clinical manifestations. The subjects' level of pulmonary impairment ranged from mild to severe, based on spirometric results (i.e., mild = FEV1 > 80% predicted, FEF25–75 = 70 to 80% predicted [mild CF group]; moderate = FEV1 40 to 80% predicted, FEF25–75 = 20 to 69% predicted [moderate CF group]; and severe = FEV1 < 40% predicted, FEF25–75 < 20% predicted [severe CF group]). In addition, nine healthy control subjects (five males, and four females) were recruited (control group). Control subjects had no history of acute or chronic lung disease, and their spirometric and lung volume values were within the normal range of predicted values. None of the control subjects was in regular exercise training.
Informed, written consent was obtained from all subjects, and the study was approved by the Royal Prince Alfred Hospital Ethics Review Committee.
Spirometry was performed with an autospirometer (AS6000 Autospirometer; Minato, Osaka, Japan), which was calibrated prior to each study. Maximum voluntary ventilation (MVV) was calculated from FEV1 × 35 (12). Lung volumes were determined by body plethysmography (model 2800; Gould Electronics, Dayton, OH). TLC was calculated from the mean FRC plus the mean IC in three acceptable tests. Residual volume (RV) was calculated by subtracting the best value of VC from the TLC (13). Predicted normal values for spirometry were taken from Crapo and colleagues (14), and normal values for lung volumes were taken from Goldman and Becklake (15). Bronchodilators, if prescribed, were taken prior to lung function measurements and exercise.
To ascertain whether arm position influenced the EELV at rest, 10 subjects with CF (two from the mild CF group and four from each of the moderate and severe CF groups) also had lung volumes measured in the body plethysmograph, with arms raised to shoulder height and elbows bent.
All subjects performed an incremental arm and leg exercise test to peak work capacity according to methodology previously used in our laboratory (10, 11). An electrically braked bicycle ergometer (Seimens-Elema, Solina, Sweden) was modified for arm work. The ergometer was mounted on an adjustable table so that the crankshaft was in line with the glenohumeral joint, with the subjects seated in a straight-backed chair. The mouthpiece was adjusted so that the subjects sat fully upright during exercise. For leg exercise, subjects exercised on an electrically braked bicycle ergometer (Seimens-Elema) according to an incremental workload protocol as described by Jones (16). The height of the handlebars was adjusted so that the subjects sat fully upright during the test. The workload was increased each minute by a fixed amount (5 or 10 W for arm exercise; 10 or 20 W for leg exercise) that was chosen according to the severity of the subject's pulmonary disease. Subjects were asked to exercise at 50 to 60 rpm for both types of exercise. Arm exercise and leg exercise were separated by at least 1.5 h, or were performed on separate days.
Subjects breathed through a two-way valve (No. 2700; Hans Rudolph, Kansas City, MO). Inspired volume was measured with a dry gas meter (Vacumetrics, Ventura, CA), and inspiratory and expiratory flows were measured with a pneumotachograph placed at the subject's mouth (Screenmate; Jaeger, Wurzburg, Germany). Mixed expired gases were continually sampled from a mixing chamber (4 L chamber for FEV1 < 1.5 L; 8 L chamber for FEV1 > 1.5 L) and analyzed for oxygen (S3A; Applied Electrochemistry, Sunnyvale, CA) and carbon dioxide (Normocap; Datex, Helsinki, Finland), from which oxygen consumption (Vo 2) and carbon-dioxide production (Vco 2) were calculated. Frequency of breathing (fb) was divided by tidal volume (Vt) to give a breathing index (fb/Vt) that represented the pattern of breathing (17). Heart rate and percent oxygen saturation (SaO2 %) were obtained with a forehead probe attached to a pulse oximeter (N200; Nellcor, Hayward, CA).
For the measurement of EELV, subjects were asked to perform two IC maneuvers during the last 30 s of each workload. The measurements for the best of the two maneuvers was then subtracted from the previously determined TLC to obtain EELV. To ensure the stability of the EELV before any IC maneuver, rib-cage and abdominal volumes were measured through inductance plethysmography (Respitrace; Ambulatory Monitoring Inc., Ardsley, NY) calibrated with the isovolume technique (18). When the end-expiratory value of the sum signal of the Respitrace was stable, the patient was signaled, at end expiration (without previous warning), to perform an IC maneuver. This prevented the patient from anticipating the maneuver and changing his or her EELV. Any recording in which EELV changed in the five breaths before the maneuver was excluded from analysis.
Seven subjects with CF (one from the mild CF group, two from the moderate CF group, and four from the severe CF group) were also asked to perform IC maneuvers while at rest and with their arms in two positions before arm and leg exercise (i.e., hands resting on lap and hands on arm-crank “pedals” or bicycle handlebars). This was done to determine whether a change in arm position alone caused a change in resting EELV.
The use of IC maneuvers to determine EELV is based on the assumption that TLC does not change during exercise. Stubbing and colleagues demonstrated that TLC remained unchanged during submaximal exercise in both normal subjects (19) and patients with chronic obstructive lung disease (9). Recently, IC maneuvers at peak exercise in patients with chronic obstructive pulmonary disease (COPD) were shown to be a reliable measure of changes in EELV (20).
Flow–volume loops at peak exercise were reconstructed from measures of flow and volume, and were placed within the maximum postexercise flow–volume loop, using the EELV calculated from the TLC minus the IC. Expiratory flow limitation at peak exercise for each subject was determined by the degree to which expiratory flow rates met or exceeded the expiratory boundary of the postexercise maximum flow–volume loop. This was reported as a percentage of the respective Vt (21, 22).
All mean data are presented as mean ± SEM. Mixed (between groups and within groups) analysis of variance (ANOVAs) were conducted on the data, with planned comparisons made across both the disease-severity factor (control, mild, moderate, severe) and the limb factor (arm, leg) (23). The between-group comparisons were a linearly independent set in which each CF subgroup was compared with the control group on a particular measure. The within-group comparisons were simple-effects tests in which the arm–leg difference was tested for significance within each subgroup. A value of p < 0.05 was considered significant. Post hoc testing was conducted on data at an equivalent ventilation in order to further analyze differences observed in the outcome of the planned analysis (24). With the Bonferroni's correction used for multiple comparisons, a p < 0.017 was considered significant. Least-squares regression was used to relate measurements of pulmonary function, peak workload, peak Vo 2, and peak ventilation to changes in EELV in the CF group. Testing for differences between slopes of regression lines was done according to the procedure outlined by Bland (25).
Mean anthropometric data and resting lung function results for each subgroup are presented in Table 1. The body mass index (BMI) of subjects decreased with increasing lung disease, whereas air trapping increased (RV/TLC ratio of 57% ± 3 for the severe CF group).
| Control (n = 9) (5M) | Mild (n = 5) (4M) | Moderate (n = 9) (6M) | Severe (n = 8) (6M) | |||||
|---|---|---|---|---|---|---|---|---|
| Age, yr | 25 ± 1 | 26 ± 4 | 23 ± 2 | 27 ± 4 | ||||
| Height, cm | 173 ± 3 | 173 ± 4 | 171 ± 3 | 168 ± 2 | ||||
| Weight, kg | 72 ± 3 | 66 ± 3 | 60 ± 2 | 53 ± 3 | ||||
| BMI, kg/m2 | 24 ± 1 | 22 ± 1 | 20 ± 1 | 19 ± 1 | ||||
| FEV1, % pred | 108 ± 3 | 101 ± 4 | 68 ± 3 | 27 ± 3 | ||||
| FEF25–75, % pred | 100 ± 7 | 78 ± 12 | 34 ± 4 | 14 ± 2 | ||||
| FVC, % pred | 103 ± 4 | 104 ± 5 | 75 ± 4 | 40 ± 4 | ||||
| RV/TLC % | 22 ± 2 | 24 ± 2 | 36 ± 2 | 57 ± 3 |
No differences in resting EELV, measured in the body plethysmograph, were found with arm position (EELV with arms down: 4.24 ± 0.27 L; EELV with arms up: 4.25 ± 0.25 L). Similarly, when the effect of arm position on EELV was measured before exercise, there were no significant differences between arm positions (EELV with hands on lap: 3.80 ± 0.07 L; EELV with hands on arm-crank pedals or bicycle handlebars: 3.81 ± 0.08 L).
Workload, oxygen consumption, ventilation, and pattern of breathing. At peak exercise, subjects in all subgroups were able to achieve significantly higher workloads for leg exercise than for arm exercise (Table 2). Minute ventilation (V˙e) at peak leg exercise was significantly higher than at peak arm exercise in the control group and in the mild and moderate CF groups (Figure 1A). In all these groups the higher ventilation was a result of a significantly higher Vt (Figure 1D), with fb (Figure 1C) not being significantly different. In the severe CF group, there was no significant difference in ventilation, Vt, or frequency of breathing between peak arm and peak leg exercise.
| Control (n = 9) | Mild (n = 5) | Moderate (n = 9) | Severe (n = 8) | |||||
|---|---|---|---|---|---|---|---|---|
| Wpeak arm, (W) | 92 ± 12* | 98 ± 17* | 74 ± 12* | 38 ± 6*,† | ||||
| Wpeak leg, (W) | 219 ± 24 | 242 ± 51 | 149 ± 18† | 72 ± 10† | ||||
| Vo 2peak arm, L · min−1 | 1.92 ± 0.25* | 1.94 ± 0.27* | 1.50 ± 0.25* | 0.90 ± 0.08† | ||||
| Vo 2peak leg, L · min−1 | 2.92 ± 0.31 | 3.28 ± 0.59 | 1.99 ± 0.27† | 1.06 ± 0.13† | ||||
| fb/Vt index arm, br · min−1 · L−1 | 18 ± 1 | 18 ± 3 | 29 ± 4† | 51 ± 5*,† | ||||
| fb/Vt index leg, br · min−1 · L−1 | 16 ± 2 | 15 ± 2 | 24 ± 4 | 44 ± 3† | ||||
| %FL arm | 0 | 1 ± 1 | 8 ± 4 | 48 ± 12*,† | ||||
| %FL leg | 0 | 4 ± 3 | 16 ± 6† | 66 ± 8† |
Fig. 1. Changes in ventilation (A), EELV (B), fb (C ), and Vt (D), for control group and mild, moderate, and severe CF groups at peak arm and leg exercise. ⊥ = SEM. Values of p show level of significance for difference between arm exercise and leg exercise measurements within a subgroup. *Significantly different from resting levels; #significantly different from control.
[More] [Minimize]The subjects in the severe CF group had a limited increase in Vt throughout both arm and leg exercise (mean ± SEM : 37% ± 10% increase for arm and 57% ± 11% increase for leg exercise) as compared with the control group (206 ± 36% increase for arm and 331 ± 39% increase for leg exercise). Increases in ventilation in the severe CF group were achieved predominantly by increases in fb which resulted in a significantly higher fb/Vt index at peak arm and leg exercise than in the control group (peak arm exercise: 51 ± 5 br · min−1 · L [severe CF group]) versus 18 ± 1 br · min−1 · L−1 (control), p < 0.001; peak leg exercise: 44 ± 3 br · min−1 · L [severe CF group] versus 16 ± 2 br · min−1 · L−1 (control), (p < 0.001) (Table 2). The fb/Vt index was significantly greater for arm exercise than for leg exercise only in the severe CF group (p < 0.05) (Table 2). In the severe CF group, the Ve/MVV ratio was 98 ± 9% at peak arm exercise and 101 ± 10% at peak leg exercise. All other subgroups had a mean Ve/MVV ratio of less than 80%.
End-expiratory lung volume. The mean changes in EELV at peak exercise for each group are shown in Figure 1B. As compared with resting levels, there was a significant decrease in EELV in the control group at peak leg exercise (p < 0.001), and a significant increase in EELV in the moderate CF group for arm exercise (p < 0.05) and in the severe CF group for both arm (p < 0.05) and leg exercise (p < 0.05). In the control group, the decrease in EELV from rest to peak exercise was significantly smaller for arm exercise than for leg exercise (−0.134 L versus −0.526 L, p < 0.001). In all other subgroups the change in EELV from rest to peak exercise was not significantly different for arm versus leg exercise. The mean increases in EELV from rest to peak arm exercise for the moderate and severe CF groups were significantly different from the mean decrease in EELV for the control group (Figure 1B) (moderate CF group versus control group: 0.318 L versus −0.134 L, p < 0.05; severe CF group versus control group: 0.346 L versus −0.134 L, p < 0.05). The mean increases in EELV from rest to peak leg exercise for the moderate and severe CF groups were significantly different from the mean decrease in EELV for the control group (moderate CF group versus control group: 0.172 L versus −0.526 L, p < 0.001; severe CF group versus control group = 0.365 L versus −0.526 L, p < 0.001).
For subjects with CF, there was a significant relationship between the change in EELV from rest to peak exercise (ΔEELV) and lung function (FEV1 and FEF25–75%) for both arm and leg exercise (Figures 2A and 2B, respectively). The slopes of the relationships for arm and leg exercise were not significantly different. There was also a significant relationship between ΔEELV and Wmax, Vo 2max, and Ve max at peak leg exercise but not at peak arm exercise (WmaxLeg versus ΔEELVLeg, r = 0.50, p < 0.025; Vo 2maxLeg versus ΔEELVLeg, r = 0.52, p < 0.025; Ve maxLeg versus ΔEELVLeg r = 0.52, p < 0.025).
Fig. 2. Relationship between FEV1% predicted (A) and FEF25–75% (B) and the change in EELV from rest to peak arm and peak leg exercise (Wpeak) for subjects with CF.
[More] [Minimize]Data from individual subjects showed that the changes in EELV throughout arm and leg exercise varied considerably between groups (Figure 3). In the control group, EELV decreased in all subjects during incremental leg exercise, whereas the response to incremental arm cranking was variable, with EELV increasing in three subjects. The majority of subjects in the mild CF group had a decrease in EELV, whereas the majority of subjects in the moderate and severe CF groups had an increase in EELV throughout both arm and leg exercise.
Fig. 3. Changes in EELV throughout arm and leg exercise for each subject within the subgroup categories.
[More] [Minimize]The change in EELV and end-inspiratory lung volume (EILV) throughout arm and leg exercise within each subgroup is shown in Figure 4. In the severe CF group, both EELV and EILV constituted a significantly higher percentage of TLC at rest than in the control group (EELV/TLC% at rest prior to arm exercise, severe CF group versus control group: 72% versus 54%, p < 0.001; EILV/TLC% at rest prior to arm exercise, Severe CF Group versus Control Group = 83% versus 64%, p < 0.001). EELV and EILV increased throughout exercise, with peak EILV/TLC% in the moderate CF and severe CF groups being significantly higher than that in the control group for both arm and leg exercise (EILV/TLC% at peak arm exercise: 86% for the moderate CF group versus 81% for the control group, p < 0.05; 91% for the severe CF group versus 81% for the control group, p < 0.001; EILV/TLC% at peak leg exercise: 90% for the moderate CF group versus 84% for the control group, p < 0.01; 91% for the severe CF group versus 84% for the control group, p < 0.01) (Figure 4).
Fig. 4. Changes in EELV and EILV expressed as a percentage of TLC throughout arm and leg exercise for each subgroup.
[More] [Minimize]Flow limitation. The percent expiratory flow limitation at peak exercise for each subgroup is presented in Table 2. Group data showed that the control group experienced no expiratory flow limitation at peak arm or leg exercise; however, as the severity of lung disease increased, expiratory flow at peak exercise reached the maximum expiratory flow–volume curve during some portion of expiration (Table 2). Representative samples of flow–volume loops at peak exercise, placed within the maximum postexercise flow–volume loop for a subject from each subgroup, are shown in Figure 5.
Fig. 5. Representative samples of flow–volume loops at peak arm and peak leg exercise placed within the maximum postexercise flow–volume loop for a subject from each subgroup. The tidal flow–volume loop at rest is also shown.
[More] [Minimize]EELV and pattern of breathing. For subjects within each subgroup, changes in EELV were compared at an equivalent ventilation throughout arm and leg exercise. These comparisons were made at fixed percentages of the peak ventilation attained during leg exercise. We have reported results at 40% and 60% of the peak ventilation attained during leg exercise. Only in the control group were the mean changes in EELV significantly different for arm and leg exercise (−0.17 L arm versus −0.39 L leg at 40% peak leg exercise Ve, p < 0.017; −0.13 L arm versus −0.46 L leg at 60% peak leg exercise Ve, p < 0.01). In all the CF subgroups, matching ventilation during arm and leg exercise produced no significantly different changes in EELV for arm versus leg exercise (Figure 6). Vt at 60% peak leg exercise Ve was significantly lower for arm exercise than for leg exercise in the control group (1.9 L arm versus 2.3 L leg, p < 0.01) and in the mild CF group (1.9 L arm versus 2.3 L leg, p < 0.01). Although the fb was higher for arm exercise than for leg exercise, this difference was not significant in these groups. In the moderate and severe CF groups, no significant differences were found in Vt or fb for arm versus leg exercise at equivalent ventilations.
Fig. 6. Mean changes in EELV at an equivalent ventilation of 60% of the peak ventilation during leg exercise for each subgroup. ⊥ = SEM. Value of p shows level of significance for the difference between arm exercise and leg exercise.
[More] [Minimize]This study examined changes in lung volume during incremental upper- and lower-limb exercise in healthy control subjects and subjects with CF. The major new findings of this study were first, that in normal subjects, changes in EELV were significantly different during arm and leg exercise, both at peak work and at an equivalent submaximal ventilation; and second, that, in CF subjects with airflow limitation, arm exercise resulted in dynamic hyperinflation, which was not significantly different from that observed during leg exercise for these subjects.
Control group. In the control group, the mean decrease in EELV from rest to peak exercise was small and significantly lower than that observed during leg exercise. This may reflect the lower relative peak workloads of arm exercise as compared with leg exercise, resulting in a lower peak ventilation for arm exercise and therefore less of a requirement to decrease EELV as much as for leg exercise. However, when changes in EELV during arm and leg exercise were compared at an equivalent ventilation in the control group, the change in EELV for arm exercise remained significantly smaller than that for leg exercise, indicating that the level of ventilation was not the sole reason for the smaller decrease in EELV at peak arm exercise. One explanation for this smaller decrease might be that arm cranking in normal subjects results in a lower Vt than does leg exercise at an equivalent ventilation. The lower Vt may be the result of fixation of the rib cage and abdominal wall, which is required to maintain the position of the torso during upper-limb exercise (26). This may result in a “stiffer” rib cage and the need to maintain a particular ventilation by increasing fb rather than Vt. The consequent, relatively smaller Vt would not require such a large decrease in EELV. Because the expiratory respiratory muscles (e.g., abdominal muscles) are mainly responsible for the control of EELV in upright humans on a cycle ergometer (1), another consequence of the abdominal muscles being used for torso stabilization may be that they are less available to contribute to decreasing EELV. This would result in smaller decreases in EELV for arm exercise than for leg exercise at an equivalent Ve.
It is unknown whether passive arm cranking contributes to changes in EELV during arm exercise. It has been shown that in leg exercise in normal subjects, passive movement of the leg on a cycle ergometer results in a decrease in EELV similar to that with mild exercise (27). However, passive movements of the limbs stimulate ventilation (28), and the resulting increase in ventilation may be responsible for the changes in EELV. In the study by Kagawa and Kerr (27), ventilation was not reported. We know of no studies in which the effect of passive arm movement on EELV has been examined.
Cystic fibrosis. For the subjects with CF, there was a significant inverse relationship between lung function and changes in EELV from rest to peak arm exercise. With arm exercise, subjects with CF who had an FEV1 of less than 90% predicted and an FEF25–75% of less than 65% predicted (Figure 2) were likely to dynamically hyperinflate during arm cranking. Moreover, the degree of increase in EELV was larger the greater the reduction in FEV1 or FEF25–75%. The changes in EELV from rest to peak arm exercise in relation to FEV1% predicted were similar to those for incremental leg exercise (Figure 2), and support previously reported data for leg exercise in CF patients (10).
A number of reasons have been postulated for the increase in EELV during exercise in subjects with airflow limitation. Johnson and colleagues (21), in a study of an aging, physically active population, suggested that the increase in EELV during exercise may be due to inadequate time to exhale the volume required to maintain EELV before the next inspiration begins, resulting in an increase in EELV despite activation of the expiratory muscles. This was supported by Pelligrino and coworkers (29), who applied an expiratory threshold load to normal subjects and observed increases in EELV. Pellegrino and coworkers contend that flow limitation leads to the premature termination of expiration, resulting in an increase in EELV. We believe that this is the most likely explanation for the increase in EELV seen in our subjects with CF and expiratory airflow limitation. A neural mechanism for the increase in EELV, through tonic activation of inspiratory muscles in asthmatic individuals, was suggested by Martin and associates (30).
In the CF group with mild pulmonary impairment in our study, the response to incremental arm and leg exercise with regard to changes in EELV was similar to that observed in the normal subjects. However, one subject did show dynamic hyperinflation at peak arm and leg exercise (Figure 3). Individual data also showed that one subject at peak arm exercise, and three subjects at peak leg exercise, reached or exceeded the postexercise maximum expiratory flow–volume curve. These findings in the mild CF group indicate that in a few subjects, there were slight variations from the pulmonary responses seen in normal subjects.
The moderate CF group had a wide range of lung function, with FEV1 ranging from 40 to 80% predicted. There was individual variation in the EELV response for arm and leg exercise, with eight of the nine subjects showing increased EELV at peak arm exercise but only five subjects showing increased EELV at peak leg exercise. No significant difference was found in the increase in EELV from resting levels when arm and leg exercise were compared at either peak work or at an equivalent submaximal ventilation. This was also true for the subjects in the severe CF group, suggesting that in CF subjects with airflow limitation, the degree of dynamic hyperinflation was not affected by the type of exercise used in our study (i.e., arm or leg). This is in contrast with the finding for our normal subjects, in whom the degree to which EELV fell below resting levels was different for arm and leg exercise.
The CF group with severe lung disease had a significantly lower work capacity for both arm and leg exercise than did the normal control group. A number of factors indicate that this reduced work capacity was at least partly due to mechanical constraints limiting ventilation. Evidence that ventilation was limited in the severe CF group is provided by the values for Ve/MVV% at peak arm and leg exercise of 98% and 101%, respectively, indicating that subjects in this group were breathing at or near their maximum possible ventilation. Also, ventilation in the severe CF group was almost identical at peak arm and leg exercise, and was significantly lower than that for the control group. This differed from the finding in all other subgroups that ventilation at peak leg exercise was significantly higher than at peak arm exercise. The limitation to ventilation in the severe CF group was most likely due to expiratory flow limitation, since six of the eight subjects at peak arm work, and all of the subjects at peak leg work, reached or exceeded their maximum expiratory flow–volume curve, resulting in a mean percent expiratory flow limitation of 48% for arm exercise and 66% for leg exercise. It should be noted, however, that patients with more severe lung disease may have very inhomogeneous lung units. The rapidly emptying units may contribute more to tidal flow than to forced expiratory flow at the same lung volume, thus resulting in higher flow rates during tidal breathing than during forced expiratory maneuvers (31). Because we compared exercise tidal flows to maximal forced expiratory flows at rest, this may mean that we overestimated the level of expiratory flow limitation. Nonetheless, subjects with severe lung disease due to CF were arguably more flow limited.
The ventilatory constraints in the severe CF group are illustrated in Figure 4. The severe CF group had significantly higher values for EELV/TLC% and EILV/TLC% at rest than did the control group. Although EILV increased during exercise, EELV also increased, resulting in only small changes in Vt. At a fixed Vt, further increases in ventilation could be achieved only by increases in fb. This is reflected in the significantly higher fb in relation to Vt in the severe CF group than in the control group at both peak arm and leg exercise, and probably resulted in increased resistive loads and dead space ventilation (4).
Limitation to increasing ventilation in the severe CF group may also have been due to limitations to the increase in EILV. The mean peak values of EILV/TLC% for arm and leg exercise in the severe CF group were 91 ± 2% and 91 ± 1% respectively. These values are of a similar magnitude to those reported by O'Donnell and colleagues (32) in patients with chronic airflow limitation (CAL) (FEV1 = 36% predicted) at peak leg exercise (EILV/TLC% ≈ 91%), and were significantly higher than those of our normal control group (81 ± 2% arm, p < 0.001; and 84 ± 2% leg, p < 0.01). Mansell and associates (33) have shown that subjects with moderate to severe lung disease due to CF have excessive stiffness of the lung at volumes above approximately 90% TLC. Breathing with tidal volumes constrained between a high EELV and EILV suggests that subjects may have been forced to breathe on this stiffer portion of the pressure–volume relationship, which would substantially increase the elastic load. That the high EILV/TLC% prevented further increases in EELV is suggested by the EELV/TLC% having reached 77 ± 2% at 75% of peak arm work capacity and 76 ± 2% at 75% of peak leg work capacity, without inceasing further with increasing arm or leg work. If further increases in EELV had occurred, maintenance of Vt would have required EILV to increase. At an EILV above approximately 90% of TLC (33), the substantial increase in elastic work may not have been tolerated (4).
Arm position itself did not significantly alter resting EELV (FRC), whether measured inside or outside the body plethysmograph. Criner and Celli (34) described a personal observation that in patients with CAL, FRC measured in a body plethysmograph was not different with the arms elevated or dependent. In a study by Dolmage and colleagues (35), arm elevation with clasping of the hands on top of the head did not result in any significant difference in FRC from sitting with the hands on the lap in patients with CAL. However, Martinez and coworkers (36) (also in patients with CAL) demonstrated a small but significant increase in FRC (mean: 120 ml) with 2 min of arm elevation with the patient inside the body plethysmograph. Maintaining the arms in such a position for 2 min also requires an increase in ventilation, which could explain the increase in FRC. From the results of our present study and the cited literature, it can reasonably be concluded that arm position itself does not alter resting FRC to any great extent.
In summary, supported arm exercise in subjects with CF resulted in dynamic hyperinflation in those subjects with moderate to severe lung disease, whereas those subjects with mild lung disease had a decrease in EELV in a manner similar to that of normal control subjects. Normal control subjects had a significantly smaller decrease in EELV at peak arm exercise than with peak leg exercise. This difference in the change in EELV for arm versus leg exercise persisted at an equivalent submaximal ventilation. A possible explanation for this is that during arm exercise, abdominal muscles are required for stabilization of the torso, thus possibly reducing their contribution to the regulation of EELV. In the CF subjects, in our study, the slopes of the inverse relationship between FEV1 and changes in EELV from rest to peak exercise did not differ significantly for arm and leg exercise. Also, in the moderate and severe CF groups, there were no significant differences in the increases in EELV from resting levels for arm and leg exercise. We conclude that in normal subjects, arm exercise results in a different the EELV response than does leg exercise; and that in CF subjects with airflow limitation, dynamic hyperinflation occurs with arm exercise and to an extent that is not significantly different from that observed with leg exercise.
The authors wish to thank Andrea Wong for assistance with aspects of the data analysis.
| 1. | Henke K. G., Sharratt M., Pegelow D., Dempsey J. A.Regulation of end-expiratory lung volume during exercise. J. Appl. Physiol641988135146 |
| 2. | Grimby G., Goldman M., Mead J.Respiratory muscle action inferred from rib cage and abdominal V-P partitioning. J. Appl. Physiol411976739751 |
| 3. | Sharratt M. T., Henke K. G., Aaron E. A., Pegelow D. F., Dempsey J. A.Exercise-induced changes in functional residual capacity. Respir. Physiol701987313326 |
| 4. | McClaran S. R., Babcock M. A., Pegelow D. F., Reddan W. G., Dempsey J. A.Longitudinal effects of aging on lung function at rest and exercise in healthy active fit elderly adults. J. Appl. Physiol78199519571968 |
| 5. | Grassino A. E., Derenne J. P., Almirall J., Milic-Emili J., Whitelaw W.Configuration of the chest wall and occlusion pressures in awake humans.J. Appl. Physiol501981134142 |
| 6. | Grimby G., Striksa J.Flow-volume curves and breathing patterns during exercise in patients with obstructive lung disease. Scand. J. Clin. Lab. Invest251970303313 |
| 7. | Potter W. A., Olafsson S., Hyatt R. E.Ventilatory mechanics and expiratory flow limitation during exercise in patients with obstructive lung disease. J. Clin. Invest501971910919 |
| 8. | Dodd D. S., Brancatisano T., Engel L. A.Chest wall mechanics during exercise in patients with severe chronic air-flow obstruction. Am. Rev. Respir. Dis12919843338 |
| 9. | Stubbing, D. G., L. D. Pengelly, J. L. C. Morse, and N. L. Jones. 1980. Pulmonary mechanics during exercise in subjects with chronic airflow obstruction. J. Appl. Physiol. (Respir. Environ. Exercise Physiol.) 49: 511–515. |
| 10. | Regnis J. A., Alison J. A., Henke K. G., Donnelly P. M., Bye P. T. P.Changes in end-expiratory lung volume during exercise in cystic fibrosis relate to severity of lung disease. Am. Rev. Respir. Dis1441991507512 |
| 11. | Alison J. A., Regnis J. A., Donnelly P. M., Adams R. D., Sutton J. R., Bye P. T. P.Evaluation of supported upper limb exercise capacity in patients with cystic fibrosis. Am. J. Respir. Crit. Care Med156199715411548 |
| 12. | Clarke T. J. H., Freedman S., Campbell E. J. M., Winn R. R.The ventilatory capacity of patients with chronic airways obstruction. Clin. Sci361969307316 |
| 13. | Donnelly P. M., Yang T.-S., Peat J. K., Woolcock A. J.What factors explain racial differences in lung volumes? Eur. Respir. J.41991829838 |
| 14. | Crapo R. O., Morris A. H., Gardner R. M.Reference spirometric values using techniques and equipment that meet ATS recommendations. Am. Rev. Respir. Dis1231981659664 |
| 15. | Goldman H. I., Becklake M. R.Respiratory function tests: normal values at median altitudes and the prediction of normal results. Am. Rev. Tuberc791959457467 |
| 16. | Jones, N. L. 1988. Clinical Exercise Testing, 3rd ed. WB Saunders, Philadelphia. 129–130. |
| 17. | Epstein S. K., Celli B. R., Williams J., Tarpy S., Roa J., Shannon T.Ventilatory response to arm elevation: its determinants and use in patients with chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med.1521995211216 |
| 18. | Konno K., Mead J.Measurement of the separate volume changes of rib cage and abdomen during breathing. J. Appl. Physiol.221967407422 |
| 19. | Stubbing D. G., Pengelly L. D., Morse J. L. C., Jones N. L.Pulmonary mechanics during exercise in normal males. J. Appl. Physiol. (Respir. Environ. Exerc. Physiol.)491980506510 |
| 20. | Yan S., Kaminski D., Sliwinki P.Reliability of inspiratory capacity for estimating end-expiratory lung volume changes during exercise in patients with chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med.15619975559 |
| 21. | Johnson B. D., Reddan W. G., Pegelow D. F., Seow K. C., Dempsey J. A.Flow limitation and regulation of functional residual capacity during exercise in a physically active aging population. Am. Rev. Respir Dis1431991960967 |
| 22. | Regnis J. A., Donnelly P. M., Robinson M., Alison J. A., Bye P. T. P.Ventilatory mechanics at rest and during exercise in patients with cystic fibrosis. Am. J. Respir. Crit. Care Med.154199614181425 |
| 23. | Winer, B. J., D. R. Brown, and K. M. Michels. 1991. Statistical Principles in Experimental Design. McGraw-Hill, New York. 562–567. |
| 24. | Hays, W. L. 1988. Statistics, 4th ed. Holt, Rinehart & Winston, New York. 411–414. |
| 25. | Bland, M. 1995. An Introduction to Medical Statistics, 2nd ed. Oxford University Press, Oxford. 203–204. |
| 26. | Crouser J. I., Martinez F. J., Celli B. R.Respiratory response and ventilatory recruitment during arm elevation in normal subjects. Chest1011992336340 |
| 27. | Kagawa J., Kerr H. D.Effects of brief graded exercise on specific airways conductance in normal subjects. J. Appl. Physiol281970138144 |
| 28. | West, J. B. 1974. Respiratory Physiology—The Essentials. Waverly Press, Baltimore. 127. |
| 29. | Pellegrino R., Brusasco V., Rodarte J. R., Babb T. G.Expiratory flow limitation and regulation of end-expiratory lung volume during exercise. J. Appl. Physiol.74199325522558 |
| 30. | Martin J., Powell E., Shore S., Emrich J., Engel L. A.The role of respiratory muscles in the hyperinflation of bronchial asthma. Am. Rev. Respir. Dis1211980441447 |
| 31. | Landau L. I., Taussig L. M., Macklem P. T., Beaudry P. H.Contribution of inhomogeneity of lung units to the maximal expiratory flow-volume curve in children with asthma and cystic fibrosis. Am. Rev. Respir. Dis.1111975725731 |
| 32. | O'Donnell D. E., Webb K. A.Exertional breathlessness in patients with chronic airflow limitation: the role of lung hyperinflation. Am. Rev. Respir. Dis148199313511357 |
| 33. | Mansell A., Dubrawsky C., Levison H., Bryan A. C., Crozier D. N.Lung elastic recoil in cystic fibrosis. Am. Rev. Respir. Dis1091974190197 |
| 34. | Criner G. J., Celli B. R.Effect of unsupported arm exercise on ventilatory muscle recruitment in patients with severe chronic airflow obstruction. Am. Rev. Respir. Dis1381988856861 |
| 35. | Dolmage T. E., Maestro L., Avendano M. A., Goldstein R. S.The ventilatory response to arm elevation of patients with chronic obstructive pulmonary disease. Chest104199310971100 |
| 36. | Martinez F. J., Couser J. I., Celli B. R.Respiratory response to arm elevation in patients with chronic airflow obstruction. Am. Rev. Respir. Dis1431991476480 |
Jennifer A. Alison was supported by a Biomedical Scholarship from the National Health and Medical Research Council of Australia.
