- Email: [email protected]
Arm Exercise and Hyperinflation in Patients With COPD
Arm Exercise and Hyperinflation in Patients With COPD* Effect of Arm Training Francesco Gigliotti, MD; Claudia Coli, MD; Roberto Bianchi, MD; Michela Grazzini, MD; Loredana Stendardi, MD; Carla Castellani, RT; and Giorgio Scano, MD, FCCP
Background: Unlike studies on leg exercise, reports on the regulation of dynamic hyperinflation during arm exercise are scanty. We ascertained the following in patients with COPD: (1) whether and to what extent upper-limb exercise results in dynamic hyperinflation, and (2) the mechanism whereby an arm-training program (ATP) reduces arm effort and dyspnea. Patients: Twelve patients with moderate-to-severe COPD were tested during incremental, symptom-limited arm exercise after a nonintervention control period (pre-ATP) and after ATP. Methods: Exercise testing (1-min increments of 5 W) was performed using an arm ergometer. Oxygen uptake (V˙O2), carbon dioxide output, minute ventilation (V˙E), tidal volume, and respiratory rate (RR) were measured continuously during the tests. Inspiratory capacity (IC), exercise dyspnea, and arm effort using a Borg scale were assessed at each step of exercise. Results: Arm exercise resulted in a significant decrease in IC and significant positive relationships of IC with an increase in V˙O2 and exercise dyspnea and arm effort. The results of ATP were as follows: (1) a significant increase in exercise capacity (p < 0.001); (2) no change in the relationships of exercise dyspnea and arm effort with V˙E and IC, and of IC with V˙O2; (3) at a standardized work rate, V˙E, exercise dyspnea, and arm effort significantly decreased, while the decrease in IC was significantly less (p < 0.01) than before the ATP; the decrease in V˙E was accomplished primarily by a decrease in RR; and (4) at standardized V˙E, exercise dyspnea and arm effort decreased significantly. Conclusions: Arm exercise results in the association of dynamic hyperinflation, exercise dyspnea, and arm effort in COPD patients. An ATP increases arm endurance, modulates dynamic hyperinflation, and reduces symptoms. (CHEST 2005; 128:1225–1232) Key words: arm training; COPD; dynamic hyperinflation; dyspnea Abbreviations: ATP ⫽ arm-training program; au ⫽ arbitrary unit; EELV ⫽ end-expiratory lung volume; FRC ⫽ functional residual capacity; HR ⫽ heart rate; IC ⫽ inspiratory capacity; RR ⫽ respiratory rate; TLC ⫽ total lung capacity; V˙co2 ⫽ carbon dioxide output; V˙e ⫽ minute ventilation; V˙o2 ⫽ oxygen uptake; Vt ⫽ tidal volume; WR ⫽ work rate
increase in ventilation during leg exercise is T heaccomplished by dynamic hyperinflation, which helps minimize expiratory flow limitation1 but increases dyspnea in patients with COPD.2,3 Unlike studies on leg exercise, reports on regulation of *From Fondazione Don C. Gnocchi, IRCCS, Pozzolatico, Florence, Italy. This work was supported by a grant from the Fondazione Don C. Gnocchi ONLUS (IRCCS), Florence, Italy. Manuscript received July 23, 2004; revision accepted February 3, 2005. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Giorgio Scano, MD, Section of Pulmonary Rehabilitation, Fondazione Don C. Gnocchi IRCCS, Via Imprunetana 124, 50020 Pozzolatico, Firenze, Italy; e-mail: uopneumo. [email protected] www.chestjournal.org
dynamic hyperinflation during arm exercise are scanty: simple arm elevation is associated with a mild increase in functional residual capacity (FRC)4 or a decrease in inspiratory capacity (IC)5 in COPD patients with moderate-to-severe obstruction. In contrast, arm exercise results in premature termination of expiration and an increase in dynamic hyperinflation in flow-limited patients with cystic fibrosis.6 Pulmonary rehabilitation programs improve exercise capacity by reducing both dynamic hyperinflation and dyspnea in patients with COPD.7,8 An arm-training program (ATP) leads to reduction in ventilatory requirements for simple arm elevation and increases the exercise level.9,10 Ratings of perceived dyspnea and fatigue decreased significantly in one study11 but not in another,9 so that the mechaCHEST / 128 / 3 / SEPTEMBER, 2005
1225
nisms of their improvement, if any, remain to be defined in patients with COPD. Like leg-training programs, the effect of an ATP on dyspnea and arm fatigue could be due to a decrease in ventilatory requirements for a given level of activity,12 a decrease in dynamic hyperinflation,8 and desensitization.7,8 The aims of this study were therefore to ascertain the following in patients with COPD: (1) whether and to what extent upper-limb exercise results in dynamic hyperinflation, and (2) the mechanism whereby an ATP reduces arm effort and dyspnea.
Materials and Methods Subjects We studied 12 consecutive patients with stable, moderate-tosevere COPD who were entering an outpatient pulmonary rehabilitation program. Patients satisfied the following criteria: (1) a long history of smoking and moderate-to-severe, chronic dyspnea (Medical Research Council grade 2 to 4); (2) clinically stable condition, with no exacerbation or hospital admission in the preceding 4 weeks; and (3) no other significant disease(s) potentially contributing to dyspnea. The patients were all motivated to participate in the program and did not currently smoke. Functional Evaluation Routine spirometry with the subjects in a seated position was performed as previously described.13,14 FRC was measured by helium dilution technique. The normal values for lung volumes are those proposed by the European Respiratory Society.15 Before performing exercise, the ventilatory patterns were evaluated with subjects sitting comfortably in an armchair with a mass-flow sensor (V˙max; Sensor Medics; Yorba Linda, CA). From the spirogram, we derived the tidal volume (Vt), respiratory rate (RR), and minute ventilation (V˙e) [Vt ⫻ RR]. Arm Ergometry All patients performed an incremental (5 W/min), symptomlimited arm exercise. A modified stationary ergometer (Monarch Instrument; Varberg, Sweden) was used to deliver precise workload adjustments. The ergometer was secured to a table at shoulder level with the subjects seated in a straight-back chair. Each patient was made familiar with the apparatus days before the test. Expired gas was analyzed for V˙e, oxygen uptake (V˙o2), carbon dioxide output (V˙co2) using breath-by-breath analysis (V˙max System; SensorMedics). V˙co2 and V˙o2 were expressed as standard temperature and pressure, dry, and as a percentage of the predicted maximum V˙o2.16 The ventilatory equivalent for oxygen (V˙e/V˙o2) and ventilatory efficiency (V˙e/V˙co2) were also calculated. For each run, changes in V˙e, Vt, and RR were continuously recorded. There was a continuous monitoring of 12-lead ECG and oxygen saturation by pulse oximetry (NPB 290; Nellcor Puritan Bennett; Pleasanton, CA). BP was recorded at rest and every 2 min during exercise, and recovery to baseline levels. Subjects were asked to perform two IC maneuvers during the last 30 s of each workload for the measurement of end-expiratory lung volume (EELV). The use of IC maneuvers to determine EELV 1226
is based on the assumption that total lung capacity (TLC) does not change during exercise in patients with COPD.17 IC maneuvers at peak exercise are shown to be a reliable measure of change in EELV in patients with COPD.18 The perception of exercise dyspnea and arm effort was evaluated each minute during exercise. Patients described dyspnea as a sensation of labored or difficult breathing experienced during cranking. Subjects were asked to rank the overall sensation of exercise dyspnea and arm effort on a large Borg scale from 0 (none) to 10 (maximal).19 The subjects were instructed that 0 signified no sensation at all, and that 10 signified the most severe sensation that they had ever experienced. ATP Each patient attended a 6-week outpatient pulmonary rehabilitation program. The program included education, breathing retraining, leisure walking, and unsupported arm exercise and arm training with an arm ergometer. For the training on the arm ergometer, the workload corresponding to 80% of the peak work rate (WR) observed in the pretraining incremental exercise test was set as the training intensity. Patients were instructed to maintain this work level until they reached a symptom limit. Sessions were closely supervised by a rehabilitation therapist; during the session, heart rate (HR) and arterial oxygen saturation were monitored. Unsupported arm exercise consisted of the following: (1) repetitive bilateral shoulder abduction and extension holding a light hand weight (0.5 to 1.0 kg) in a rhythm synchronized with breathing for 2-min periods; and (2) threading a set of rings in a series of pegs placed at different heights while the arm is held above the horizontal. The study was approved by the Ethics Committee of the Institution, and informed consent was obtained from subjects. Protocol This is a single-center, two-period, controlled study in which subjects complete a 6-week nonintervention period before entering a 6-week pulmonary rehabilitation program involving regular exercise training. In an initial screening, subjects were tested for pulmonary function and gas exchange. They became familiar with exercise testing procedures and the various scales for rating symptom intensity, and completed an incremental, symptomlimited exercise test. Three experimental visits were held at 6-week intervals immediately before the control period, after the control period (pre-ATP visit), and after ATP. Therefore, the subjects acted as their own controls. All visits were conducted at the same time of day for each subject. Data Analysis To compare responses to an identical level of exercise before and after the rehabilitation program, we selected the highest WR tolerated by a given patient during pre-ATP test, (the standardized WR). To compare responses to an identical level of ventilation, we selected the highest V˙e tolerated by a given patient during the pre-ATP test (the standardized V˙e). Nonparametric ratings of exertional breathlessness were compared before and after intervention using the Wilcoxon test. All other measurements made before and after ATP were analyzed using the paired t test. Pearson correlation coefficients were used to test the strength of the association between measured variables. A value of p ⬍ 0.05 was considered significant.
Results Anthropometric and baseline function data of the 12 patients with moderate-to-severe airflow obstrucClinical Investigations
Table 1—Anthropometric and Lung Function Data* Variables Age, yr Male/female gender, No. Height, cm Weight, kg Vital capacity, L Vital capacity, % predicted FEV1, L FEV1, % predicted FEV1/vital capacity, % FRC, L FRC, % predicted TLC, L TLC, % predicted Pao2, mm Hg Paco2, mm Hg pH
Before Control
Before ATP
After ATP
3.48 ⫾ 0.87 86 ⫾ 12.4 1.59 ⫾ 0.58 49 ⫾ 12.3 46 ⫾ 12 4.56 ⫾ 1.2 129 ⫾ 24.5 7.3 ⫾ 0.6 113 ⫾ 9.2 73.5 ⫾ 12.1 42.4 ⫾ 5 7.4 ⫾ 0.03
66.7 ⫾ 7.5 10/2 170 ⫾ 9.9 77.3 ⫾ 18.5 3.52 ⫾ 0.9 86.8 ⫾ 12.8 1.61 ⫾ 0.6 49.1 ⫾ 12.6 46.7 ⫾ 13.9 4.5 ⫾ 1.3 129 ⫾ 25 7.18 ⫾ 0.5 113 ⫾ 9 74.8 ⫾ 12.7 42.5 ⫾ 4.6 7.4 ⫾ 0.03
76.9 ⫾ 18 3.66 ⫾ 1.1 87.5 ⫾ 13 1.63 ⫾ 0.69 51.2 ⫾ 12.8 44.9 ⫾ 13.4 4.6 ⫾ 1.2 127 ⫾ 24 7.6 ⫾ 0.5 115 ⫾ 8 80.6 ⫾ 11.3 41.1 ⫾ 3.2 7.41 ⫾ 0.02
*Data are presented as mean ⫾ SD.
tion and hyperinflation, mild-to-moderate hypoxia, and mild carbon dioxide retention are shown in Table 1. The data did not change during the study. Incremental Exercise Performance At peak arm exercise, V˙e, V˙co2, V˙o2, HR, exercise dyspnea, arm effort, and WR all increased. The ATP did not modify V˙e, V˙co2, V˙o2, HR, exercise dyspnea, and arm effort but increased WR (p ⬍ 0.001) [Table 2]. The relationships of changes in exercise dyspnea and arm effort with changes in V˙e, V˙o2, V˙co2, and the relationships of changes in V˙e with changes in V˙o2 and V˙co2 were not altered with the ATP (Table 3). Changes at Standardized WR With arm exercise, IC decreased by 0.93 ⫾ 0.4 L (from 2.68 ⫾ 0.79 to 1.75 ⫾ 0.63 L, p ⬍ 0.000004)
[mean ⫾ SD]. After the ATP, IC decreased by 0.59 ⫾ 0.27 L (from 2.6 ⫾ 0.83 to 2.01 ⫾ 0.81 L, p ⬍ 0.0001) and was significantly less (p ⬍ 0.01) than before ATP (Table 4). ATP lowered HR (p ⬍ 0.03) and decreased V˙e (p ⬍ 0.01) by lengthening RR (p ⬍ 0.03) but did not modify V˙o2 and V˙co2 (Table 5). Figure 1 (left panel) shows individual changes in both exercise dyspnea (average, from 5.6 ⫾1.3 to 3.6 ⫾ 1.5 au; p ⬍ 0.02) and arm effort (average, from 6.5 ⫾ 2.3 to 5.1 ⫾ 2.3 arbitrary units [au]; p ⬍ 0.01) with the ATP. In most patients, exercise dyspnea and arm exercise decreased at standardized WR. Changes at Standardized V˙E The decrease in IC during arm exercise before ATP (0.68 ⫾ 0.42 L; from 2.69 ⫾ 0.62 to 2.01 ⫾ 0.57 L; p ⬍ 0.03) tended to significantly differ (p ⬍ 0.057) from the decrease after ATP (0.43 ⫾ 0.32 L; from 2.79 ⫾ 0.66 to 2.35 ⫾ 0.75 L; p ⬍ 0.007). ATP reduced exercise dyspnea (from 4 ⫾ 1.7 to 2.75 ⫾1.2 au; p ⬍ 0.02) and arm effort (from 5.6 ⫾ 3 to 4.6 ⫾ 3.1 au; p ⬍ 0.01; Fig 1, right panel), lowered HR (p ⬍ 0.03), but did not modify WR, V˙o2, V˙co2, Vt, and RR (Table 6). Relationships Figure 2 shows the slopes of the relationships of changes in IC with changes in exercise dyspnea (left panel; all patients but patient 11) and arm effort (right panel; all patients but patients 6 and 11) before the ATP (r2 ⫽ 0.25 and r2 ⫽ 0.64, respectively) and after ATP (r2 ⫽ 0.37 and r2 ⫽ 0.60, respectively); no difference was found between before and after ATP. In each patient, a decrease in IC was significantly correlated with increase in V˙o2 (⌬V˙o2/ ⌬IC, 0.01 ⫾ 0.03 L/min/L; from r ⫽ 0.73 to 0.96). This correlation did not significantly change after
Table 2—Peak Exercise Response Before and After Rehabilitation* After Control Variables
Before Control
Before ATP
After ATP
p Value†
WR, W V˙e, L/min V˙co2, L/min V˙o2, L/min Exertional dyspnea, au Arm effort, au IC, L HR, beats/min RR, breaths/min
39.4 ⫾ 12.6 39.5 ⫾ 8.9 1.16 ⫾ 0.3 1.11 ⫾ 0.26 5.3 ⫾ 1.22 6.2 ⫾ 2.3 1.76 ⫾ 0.62 130.7 ⫾ 17.7 31.6 ⫾ 12.9
41.1 ⫾ 13.6 41 ⫾ 9.3 1.17 ⫾ 0.29 1.12 ⫾ 0.23 5.5 ⫾ 1.4 6.4 ⫾ 2.6 1.8 ⫾ 0.6 134 ⫾ 19.9 32.4 ⫾ 11.6
48.3 ⫾ 14.5 44.3 ⫾ 7.6 1.34 ⫾ 0.27 1.23 ⫾ 0.22 5.75 ⫾ 2.1 6.3 ⫾ 2.6 2.1 ⫾ 0.7 135.5 ⫾ 19.6 31.9 ⫾ 10.2
⬍ 0.04 NS NS NS NS NS NS NS NS
*Data are presented as mean ⫾ SD. NS ⫽ not significant. †After vs before ATP. www.chestjournal.org
CHEST / 128 / 3 / SEPTEMBER, 2005
1227
Table 3–Slopes of the Relationships of Changes in Exertional Dyspnea, Arm Effort, and V˙E with changes in V˙O2 and VCO2* After Control Variables
Before Control
Before ATP
After ATP
⌬ED/⌬V˙o2, au/% maximal V˙o2 ⌬ED/⌬V˙co2, au/L/min ⌬ED/⌬V˙e, au/L/min ⌬AE/⌬V˙o2 ⌬AE/⌬V˙co2, au/L/min ⌬AE/⌬V˙e, au/L/min ⌬V˙e/⌬V˙co2 ⌬V˙e/⌬V˙o2, L/min/% maximal V˙o2 ⌬V˙co2/⌬V˙o2, L/min/% maximal V˙o2
0.15 ⫾ 0.02 5.3 ⫾ 2.9 0.2 ⫾ 0.1 0.17 ⫾ 0.15 0.17 ⫾ 0.1 0.2 ⫾ 0.13 28.1 ⫾ 4 0.51 ⫾ 0.1 0.6 ⫾ 0.12
0.13 ⫾ 0.06 5.6 ⫾ 3.2 0.18 ⫾ 0.13 0.19 ⫾ 0.17 0.18 ⫾ 0.15 0.21 ⫾ 0.11 29.2 ⫾ 4.7 0.56 ⫾ 0.12 0.8 ⫾ 0.16
0.12 ⫾ 0.04 4.6 ⫾ 1.6 0.21 ⫾ 0.20 0.13 ⫾ 0.06 0.14 ⫾ 0.07 0.23 ⫾ 0.17 27.4 ⫾ 6.7 0.6 ⫾ 0.1 0.7 ⫾ 0.2
*Data are presented as mean ⫾ SD. ED ⫽ exertional dyspnea; AE ⫽ arm effort.
ATP (⌬V˙o2/⌬IC, 0.009 ⫾ 0.01 L/min/L; r ⫽ 0.67 to 0.98). Two representative subjects with the highest (patient 6, left panel) and the lowest (patient 7, right panel) response slopes of this relationship during arm exercise before and after ATP are shown in Figure 3. The relationship between IC and V˙o2 indicates that dynamic hyperinflation predicts a good amount of the variability in V˙o2 during arm exercise. Discussion Arm exercise resulted in dynamic hyperinflation, which directly correlated with increases in dyspnea, arm effort, and V˙o2. The ATP resulted in the following: (1) a significant increase in exercise capacity at peak exercise; (2) a decrease in both ventilation and dynamic hyperinflation, primarily due to a de-
Table 4 —IC at Standardized WR (41.1 ⴞ 13.6 W) IC After Control
Patient No.
IC Before Control, L
Before ATP
After ATP
Start
End
Start
End
Start
End
1 2 3 4 5 6 7 8 9 10 11 12 Mean SD p Value*
3 3.5 2.78 2.6 2.75 3.5 1.4 3.3 1.6 3.8 3.4 1.4 2.75 0.85 ⬍ 0.00004
1.68 2.5 2.4 1.1 1.2 3.0 1.2 2.6 0.8 2.0 2.1 0.8 1.78 0.75
3.1 3.26 2.9 2.4 2.7 3.16 1.58 3.53 1.47 3.5 3.16 1.4 2.68 0.79 ⬍ 0.000004
1.72 2.2 2.3 1.2 1.5 1.98 1.1 3.0 1 2.1 1.98 0.9 1.75 0.63
3.3 2.8 2.95 2.68 2.62 2.49 2.36 3.74 1.34 3.81 2.5 1.57 2.60 0.83 ⬍ 0.0001
2.3 2.6 2.1 2.4 1.9 1.36 1.34 3.34 0.98 3.3 1.36 1.08 2.01 0.81
*Start vs end IC. 1228
crease in RR at standardized WR; and (3) a decrease in dyspnea and arm effort at standardized WR and ventilation. The novel finding of this study is the dynamic hyperinflation, ie, the increase in dynamic EELV during supported-arm exercise in patients with COPD. To the extent that TLC does not change appreciably during exercise in COPD,17 a change in IC accurately reflects the change in EELV.18 Dynamic hyperinflation minimizes expiratory flow limitation during lower-limb exercise in patients with severe COPD,2,3,8,20 –22 but puts the respiratory muscles in a disadvantageous portion of their length/ tension relationship and imposes an elastic load on the muscles during inspiration.2,3 Alison et al6 found that arm exercise may either increase EELV or reduce its decrease compared to leg exercise in healthy subjects, and diminish exercise capacity in flow-limited patients with cystic fibrosis. We explain dynamic hyperinflation with arm exercise in patients with COPD as being due to inadequate time to exhale the volume required to maintain EELV before the next inspiration begins. Due to the load it places on an already stressed system, arm exercise contributes to dyspnea and disability in many patients with COPD.23,24 In particular, the following are seen: (1) hyperinflation decreases maximal force-generating capacity of the respiratory muscle,25,26 thereby increasing the central motor output to a weakened muscle; (2) the decrease in force-generating capacity increases the sense of inspiratory effort25; and (3) the association of an increased motor output with an increased respiratory system impedance increases the respiratory muscle load and may affect the coupling between inspiratory effort and concurrent inspiratory volume (neuromuscular dissociation of the ventilatory pump).2,3 Another finding of this study is the association Clinical Investigations
Table 5—Exercise Response at Standardized WR (41.1 ⴞ 13.6 W)* After Control Variables
Before Control
Before ATP
After ATP
p Value†
V˙e, L/min Vt, L/min RR, breaths/min V˙co2, L/min V˙o2, L/min Exertional dyspnea, au Arm effort, au HR, beats/min V˙e/V˙co2 V˙e/V˙o2
40.1 ⫾ 8.6 1.26 ⫾ 0.1 30.6 ⫾ 12.9 1.16 ⫾ 0.3 1.11 ⫾ 0.26 5.3 ⫾ 1.22 6.2 ⫾ 2.3 130 ⫾ 13 34.8 ⫾ 6.1 36.2 ⫾ 6.2
41.1 ⫾ 8.9 1.26 ⫾ 0.1 29.5 ⫾ 10.8 1.16 ⫾ 0.3 1.1 ⫾ 0.2 5.6 ⫾ 1.3 6.5 ⫾ 2.3 126 ⫾ 25 35.9 ⫾ 6.9 37.1 ⫾ 6.8
37.7 ⫾ 8.5 1.32 ⫾ 0.5 27 ⫾ 1 1.15 ⫾ 0.3 1.1 ⫾ 0.18 3.6 ⫾ 1.5 5.1 ⫾ 2.3 119 ⫾ 20 33 ⫾ 6.8 33.7 ⫾ 6.4
⬍ 0.01 NS ⬍ 0.03 NS NS ⬍ 0.02 ⬍ 0.01 ⬍ 0.03 ⬍ 0.001 ⬍ 0.001
*Data are presented as mean ⫾ SD. See Table 2 legend for expansion of abbreviation. †Before ATP vs after ATP.
between arm effort and dynamic hyperinflation. The rib cage and abdominal wall must be fixed to maintain the position of the torso during upper-limb exercise.27 This may result in a stiffer rib cage and the need for maintaining a particular ventilation by increasing the RR, and thereby dynamic hyperinflation, while the central output to the exercising arms increases. Last but not least, the increase in dynamic hyperinflation predicted an average 69% of the variability in V˙o2 (from 36 to 92%; Fig 3). These data confirm the results by Diaz et al28 during incremental, symptom-limited exercise on a cycle ergometer in stable patients with COPD. In that study, resting IC (IC percentage of predicted) explained 56% of the variance in maximum V˙o2. Our data extend those re-
sults, in that dynamic hyperinflation during exercise largely predicts V˙o2. Factors other than mechanical factors, such as deconditioning, are likely to explain the residual variance in V˙o2. The present data also show that, regardless of leg or arm cycling, dynamic hyperinflation is a predictor of exercise tolerance in patients with COPD.28 Training that includes both upper- and lower-limb exercise may improve upper-extremity exercise endurance7,9 –11 and ratings of dyspnea.7,11 The effects of the ATP on dyspnea reduction must, however, be defined. Although respiratory function has generally been reported not to change after pulmonary rehabilitation programs, a mild decrease in FRC after upper-extremity exercise training has been found in patients with COPD.11 Ries et al11 showed, however,
Figure 1. Changes in exercise dyspnea (ED) and arm exercise (AE) with an ATP at standardized WR (left panel) and standardized V˙e (right panel). www.chestjournal.org
CHEST / 128 / 3 / SEPTEMBER, 2005
1229
Table 6 —Exercise Response at Standardized V˙E (37.3 ⴞ 9.5 L/min)* After Control Variables
Before Control
Before ATP
After ATP
p Value†
WR, W Vt, L/min RR, breaths/min V˙co2, L/min V˙o2, L/min Exertional dyspnea, au Arm effort, au HR, beats/min
37.8 ⫾ 12.3 1.41 ⫾ 0.4 29 ⫾ 10.8 1.1 ⫾ 0.3 1.0 ⫾ 0.25 4 ⫾ 1.7 5.6 ⫾ 3 121 ⫾ 19.2
35.5 ⫾ 16.4 1.46 ⫾ 0.5 29 ⫾ 10.8 1.1 ⫾ 0.3 1.0 ⫾ 0.25 4 ⫾ 1.7 5.6 ⫾ 3 122 ⫾ 21
37.2 ⫾ 17.1 1.51 ⫾ 0.5 27.1 ⫾ 8.8 1.1 ⫾ 0.35 1.1 ⫾ 0.25 2.75 ⫾ 1.2 4.6 ⫾ 3.1 113 ⫾ 14
NS NS NS NS NS ⬍ 0.02 ⬍ 0.01 ⬍ 0.03
*Data are presented as mean ⫾ SD. See Table 2 legend for expansion of abbreviation. †Before vs after ATP.
significant bronchodilator responsiveness at baseline, which may account for some of this improvement during training period. Previous studies29,30 of inhaled bronchodilator therapy in patients with COPD have shown that even a moderate reduction in dynamic hyperinflation may contribute to an improvement in dyspnea. Likewise, we have shown that rehabilitation programs based on leg exercise training decrease dynamic hyperinflation and reduce dyspnea.8 The present data clearly show the association between decreases in dynamic hyperinflation, RR, dyspnea, and arm effort at standardized WR. We explain the decrease in dynamic hyperinflation as being due to the increased time to exhale the volume required to maintain EELV before the next inspiration begins. We found a significant relationship between IC and dyspnea and arm effort ratings during arm exercise (Fig 2), with the correlation coefficients being, however, as low as that found during leg exercise.8 These data suggest that, although dynamic
hyperinflation contributes to exertional dyspnea in patients with COPD, additional physiologic and/or sensory factors are also important to the perception of dyspnea and arm effort during exercise. The reasons are as follows: (1) unlike the lack of changes in the slopes of ventilation with both V˙o2 and V˙co2 (Table 3), ventilation decreased at standardized WR (Table 4); (2) changes in ventilation explained a consistent amount of changes in both dyspnea and arm effort before (r2 ⫽ 0.62 and r2 ⫽ 0.64, respectively) and after the ATP (r2 ⫽ 0.44 and r2 ⫽ 0.60, respectively); and (3) tolerance or desensitization to dyspnea and arm effort may allow patients to perform higher levels of work with reduced symptoms.7,31,32 And indeed, at standardized ventilation, both arm effort and dyspnea were lower after the ATP as compared to before (Table 6). These findings indicate that both mechanical constraints and desensitization contribute to arm exercise-induced dyspnea after the training program. Furthermore,
Figure 2. Left panel: Slopes of the relationships of changes in exercise dyspnea (⌬ED) with changes in IC (⌬IC) before and after ATP. Right panel: Slopes of the relationships of changes in arm exercise (⌬AE) with changes in IC before and after the ATP. Individual data points are shown. Bars are mean ⫾ SD. 1230
Clinical Investigations
Figure 3. Relationships of changes in IC with changes in V˙o2 before and after the ATP in two representative subjects (patient 6, left panel; and patient 7, right panel). sl ⫽ slope.
breathing retraining may have actually improved the breathing pattern, slowing the RR.33 A variable percentage of COPD patients stop leg exercise because of both dyspnea and leg effort, while others mostly stop because of either one or the other.7,8,34 The present results showing in each subject a similar amount (increase and decrease) of both dyspnea and arm effort are consistent with the neural motor output being balanced between respiratory muscles, and arm and torso muscles. Even though the small series of patients in this study should prevent us from drawing definite conclusions, our data indicate that arm exercise may result in the association of dynamic hyperinflation, dyspnea, and arm effort in patients with COPD. The ATP increases arm endurance, modulates dynamic hyperinflation, and reduces symptoms.
7
8
9
10
11 12
References 1 Pride N. Respiratory muscle activation during exercise in chronic obstructive pulmonary disease. In: Jones NL, Killian KJ, eds. Breathlessness: the Campbell symposium. Hamilton, ON, Canada: Boehringer Ingelheim, 1992; 52–56 2 O’Donnell DE, Webb KA. Exertional breathlessness in patients with chronic airflow limitation: the role of lung hyperinflation. Am Rev Respir Dis 1993; 148:1351–1357 3 O’Donnell DE, Revill SM, Webb KA. Dynamic hyperinflation and exercise intolerance in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2001; 1624:770 –777 4 Martinez FJ, Couser JI Jr, Celli BR. Respiratory response to arm elevation in patients with chronic obstructive pulmonary disease. Chest 1991; 143:476 – 480 5 Dolmage TE, Maestro L, Avendano MA, et al. The ventilatory response to arm elevation of patients with chronic obstructive pulmonary disease. Chest 1993; 104:1097–1100 6 Alison JA, Regnis JA, Donnelly PM, et al. End-expiratory lung volume during arm and leg exercise in normal subjects and www.chestjournal.org
13 14 15 16 17 18
19
patients with cystic fibrosis. Am J Respir Crit Care Med 1998; 158:1450 –1458 O’Donnell DE, McGuire MA, Samis L, et al. General exercise training improves ventilatory and peripheral muscle strength and endurance in chronic airflow limitation. Am J Respir Crit Care Med 1998; 157:1489 –1497 Gigliotti F, Coli C, Bianchi R, et al. Exercise training improves exertional dyspnea in patients with COPD: evidence of role of mechanical factors. Chest 2003; 123:1794 – 1802 Couser J Jr, Martinez FJ, Celli BR. Pulmonary rehabilitation that includes arm exercise reduces metabolic and ventilatory requirements for simple arm elevation. Chest 1993; 103: 37– 41 Martinez FJ, Vogel PD, Dupont DN, et al. Supported arm exercise vs unsupported arm exercise in the rehabilitation of patients with severe chronic airflow obstruction. Chest 1993; 103:1397–1402 Ries AL, Ellis B, Hawkins RW. Upper extremity exercise training in chronic obstructive pulmonary disease. Chest 1988; 93:688 – 692 Casaburi R, Wasserman K. Exercise training in pulmonary rehabilitation. N Engl J Med 1986; 314:1509 –511 Scano G, Garcia-Herreros P, Stendardi L, et al. Cardiopulmonary adaptation to exercise in coal miners. Arch Environ Health 1980; 35:360 –366 Gorini M, Duranti R, Misuri G, et al. Aminophylline and respiratory muscle interaction in normal humans. Am Rev Respir Dis 1994; 149:1227–1236 European Community for Coal and Steel. Standardization of lung function test. Eur Respir J 1993; 6(Suppl 16):1–100 Jones NL, Campbell AJM. Clinical exercise testing. 2nd ed. Philadelphia, PA: W.B. Saunders, 1982 Stubbing DG, Pengelly LD, Morse JC, et al. Pulmonary mechanics during exercise in subjects with airflow obstruction. J Appl Physiol 1980; 49:511–515 Yan S, Kaminski D, Sliwinski 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 1997; 156:55–59 Borg GAV. Psychophysical basis of perceived exertion. Med Sci Sports Exerc 1982; 14:377–381 CHEST / 128 / 3 / SEPTEMBER, 2005
1231
20 Grimby G, Elgefors B, Oxhoj H. Ventilatory levels and chest wall mechanics during exercise in obstructive lung disease. Scand J Respir Dis 1973; 54:45–52 21 Potter WA, Olafsson S, Hyatt RE. Ventilatory mechanics and expiratory flow limitation during exercise in patients with obstructive lung disease. J Clin Invest 1971; 50:910 –918 22 Dodd DS, Brancatisano T, Engel LA. Chest wall mechanics during exercise in patients with severe chronic airflow obstruction. Am Rev Respir Dis 1984; 129:33–38 23 Celli B, Rassulo J, Make BJ. Dyssynchronous breathing during arm but not leg exercise in patients with chronic airflow obstruction. N Engl J Med 1986; 314:1485– 490 24 Criner G, Celli BR. Effect of unsupported arm exercise on ventilatory muscle recruitment in patients with severe chronic airflow obstruction. Am Rev Respir Dis 1988; 138:856 – 861 25 Leblanc P, Summers E, Inman MD, et al. Inspiratory muscles during exercise: a problem of supply and demand. J Appl Physiol 1988; 64:2482–2489 26 El-Manshawi A, Killian KJ, Summers E, et al. Breathlessness during exercise with and without resistive loading. J Appl Physiol 1986; 61:896 –905 27 Couser JI Jr, Martinez FJ, Celli BR. Respiratory response and ventilatory muscle recruitment during arm elevation in normal subjects. Chest 1992; 101:336 –340 28 Diaz O, Villafranca C, Ghezzo H, et al. Role of inspiratory
1232
29
30
31 32 33
34
capacity on exercise tolerance in COPD patients with and without tidal expiratory flow limitation at rest. Eur Respir J 2000; 16:269 –275 Belman MJ, Botnick WC, Shin JW. Inhaled bronchodilators reduce dynamic hyperinflation during exercise in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1996; 153:967–975 O’Donnell DE, Lam M, Webb KA. Spirometric correlates of improvement in exercise performance after anticholinergic therapy in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1999; 160:542–549 Belman MJ, Brooks LR, Ross DJ, et al. Variability of breathlessness measurement in patients with chronic obstructive pulmonary disease. Chest 1991; 99:566 –571 Carrieri-Kohlman V, Gormley JM, Douglas MK, et al. Exercise training decrease dyspnea and the distress and anxiety associated with it. Chest 1996; 110:1526 –1535 Casaburi R, Porzasz J, Burns MR, et al. Physiologic benefit of exercise training in rehabilitation of patients with severe chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1997; 155:1541–1551 Killian KJ, Leblanc P, Martin DH, et al. Exercise capacity and ventilatory, circulatory and symptom limitation in patients with chronic airflow limitation. Am Rev Respir Dis 1992; 146:935–940
Clinical Investigations
Background: Unlike studies on leg exercise, reports on the regulation of dynamic hyperinflation during arm exercise are scanty. We ascertained the following in patients with COPD: (1) whether and to what extent upper-limb exercise results in dynamic hyperinflation, and (2) the mechanism whereby an arm-training program (ATP) reduces arm effort and dyspnea. Patients: Twelve patients with moderate-to-severe COPD were tested during incremental, symptom-limited arm exercise after a nonintervention control period (pre-ATP) and after ATP. Methods: Exercise testing (1-min increments of 5 W) was performed using an arm ergometer. Oxygen uptake (V˙O2), carbon dioxide output, minute ventilation (V˙E), tidal volume, and respiratory rate (RR) were measured continuously during the tests. Inspiratory capacity (IC), exercise dyspnea, and arm effort using a Borg scale were assessed at each step of exercise. Results: Arm exercise resulted in a significant decrease in IC and significant positive relationships of IC with an increase in V˙O2 and exercise dyspnea and arm effort. The results of ATP were as follows: (1) a significant increase in exercise capacity (p < 0.001); (2) no change in the relationships of exercise dyspnea and arm effort with V˙E and IC, and of IC with V˙O2; (3) at a standardized work rate, V˙E, exercise dyspnea, and arm effort significantly decreased, while the decrease in IC was significantly less (p < 0.01) than before the ATP; the decrease in V˙E was accomplished primarily by a decrease in RR; and (4) at standardized V˙E, exercise dyspnea and arm effort decreased significantly. Conclusions: Arm exercise results in the association of dynamic hyperinflation, exercise dyspnea, and arm effort in COPD patients. An ATP increases arm endurance, modulates dynamic hyperinflation, and reduces symptoms. (CHEST 2005; 128:1225–1232) Key words: arm training; COPD; dynamic hyperinflation; dyspnea Abbreviations: ATP ⫽ arm-training program; au ⫽ arbitrary unit; EELV ⫽ end-expiratory lung volume; FRC ⫽ functional residual capacity; HR ⫽ heart rate; IC ⫽ inspiratory capacity; RR ⫽ respiratory rate; TLC ⫽ total lung capacity; V˙co2 ⫽ carbon dioxide output; V˙e ⫽ minute ventilation; V˙o2 ⫽ oxygen uptake; Vt ⫽ tidal volume; WR ⫽ work rate
increase in ventilation during leg exercise is T heaccomplished by dynamic hyperinflation, which helps minimize expiratory flow limitation1 but increases dyspnea in patients with COPD.2,3 Unlike studies on leg exercise, reports on regulation of *From Fondazione Don C. Gnocchi, IRCCS, Pozzolatico, Florence, Italy. This work was supported by a grant from the Fondazione Don C. Gnocchi ONLUS (IRCCS), Florence, Italy. Manuscript received July 23, 2004; revision accepted February 3, 2005. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Giorgio Scano, MD, Section of Pulmonary Rehabilitation, Fondazione Don C. Gnocchi IRCCS, Via Imprunetana 124, 50020 Pozzolatico, Firenze, Italy; e-mail: uopneumo. [email protected] www.chestjournal.org
dynamic hyperinflation during arm exercise are scanty: simple arm elevation is associated with a mild increase in functional residual capacity (FRC)4 or a decrease in inspiratory capacity (IC)5 in COPD patients with moderate-to-severe obstruction. In contrast, arm exercise results in premature termination of expiration and an increase in dynamic hyperinflation in flow-limited patients with cystic fibrosis.6 Pulmonary rehabilitation programs improve exercise capacity by reducing both dynamic hyperinflation and dyspnea in patients with COPD.7,8 An arm-training program (ATP) leads to reduction in ventilatory requirements for simple arm elevation and increases the exercise level.9,10 Ratings of perceived dyspnea and fatigue decreased significantly in one study11 but not in another,9 so that the mechaCHEST / 128 / 3 / SEPTEMBER, 2005
1225
nisms of their improvement, if any, remain to be defined in patients with COPD. Like leg-training programs, the effect of an ATP on dyspnea and arm fatigue could be due to a decrease in ventilatory requirements for a given level of activity,12 a decrease in dynamic hyperinflation,8 and desensitization.7,8 The aims of this study were therefore to ascertain the following in patients with COPD: (1) whether and to what extent upper-limb exercise results in dynamic hyperinflation, and (2) the mechanism whereby an ATP reduces arm effort and dyspnea.
Materials and Methods Subjects We studied 12 consecutive patients with stable, moderate-tosevere COPD who were entering an outpatient pulmonary rehabilitation program. Patients satisfied the following criteria: (1) a long history of smoking and moderate-to-severe, chronic dyspnea (Medical Research Council grade 2 to 4); (2) clinically stable condition, with no exacerbation or hospital admission in the preceding 4 weeks; and (3) no other significant disease(s) potentially contributing to dyspnea. The patients were all motivated to participate in the program and did not currently smoke. Functional Evaluation Routine spirometry with the subjects in a seated position was performed as previously described.13,14 FRC was measured by helium dilution technique. The normal values for lung volumes are those proposed by the European Respiratory Society.15 Before performing exercise, the ventilatory patterns were evaluated with subjects sitting comfortably in an armchair with a mass-flow sensor (V˙max; Sensor Medics; Yorba Linda, CA). From the spirogram, we derived the tidal volume (Vt), respiratory rate (RR), and minute ventilation (V˙e) [Vt ⫻ RR]. Arm Ergometry All patients performed an incremental (5 W/min), symptomlimited arm exercise. A modified stationary ergometer (Monarch Instrument; Varberg, Sweden) was used to deliver precise workload adjustments. The ergometer was secured to a table at shoulder level with the subjects seated in a straight-back chair. Each patient was made familiar with the apparatus days before the test. Expired gas was analyzed for V˙e, oxygen uptake (V˙o2), carbon dioxide output (V˙co2) using breath-by-breath analysis (V˙max System; SensorMedics). V˙co2 and V˙o2 were expressed as standard temperature and pressure, dry, and as a percentage of the predicted maximum V˙o2.16 The ventilatory equivalent for oxygen (V˙e/V˙o2) and ventilatory efficiency (V˙e/V˙co2) were also calculated. For each run, changes in V˙e, Vt, and RR were continuously recorded. There was a continuous monitoring of 12-lead ECG and oxygen saturation by pulse oximetry (NPB 290; Nellcor Puritan Bennett; Pleasanton, CA). BP was recorded at rest and every 2 min during exercise, and recovery to baseline levels. Subjects were asked to perform two IC maneuvers during the last 30 s of each workload for the measurement of end-expiratory lung volume (EELV). The use of IC maneuvers to determine EELV 1226
is based on the assumption that total lung capacity (TLC) does not change during exercise in patients with COPD.17 IC maneuvers at peak exercise are shown to be a reliable measure of change in EELV in patients with COPD.18 The perception of exercise dyspnea and arm effort was evaluated each minute during exercise. Patients described dyspnea as a sensation of labored or difficult breathing experienced during cranking. Subjects were asked to rank the overall sensation of exercise dyspnea and arm effort on a large Borg scale from 0 (none) to 10 (maximal).19 The subjects were instructed that 0 signified no sensation at all, and that 10 signified the most severe sensation that they had ever experienced. ATP Each patient attended a 6-week outpatient pulmonary rehabilitation program. The program included education, breathing retraining, leisure walking, and unsupported arm exercise and arm training with an arm ergometer. For the training on the arm ergometer, the workload corresponding to 80% of the peak work rate (WR) observed in the pretraining incremental exercise test was set as the training intensity. Patients were instructed to maintain this work level until they reached a symptom limit. Sessions were closely supervised by a rehabilitation therapist; during the session, heart rate (HR) and arterial oxygen saturation were monitored. Unsupported arm exercise consisted of the following: (1) repetitive bilateral shoulder abduction and extension holding a light hand weight (0.5 to 1.0 kg) in a rhythm synchronized with breathing for 2-min periods; and (2) threading a set of rings in a series of pegs placed at different heights while the arm is held above the horizontal. The study was approved by the Ethics Committee of the Institution, and informed consent was obtained from subjects. Protocol This is a single-center, two-period, controlled study in which subjects complete a 6-week nonintervention period before entering a 6-week pulmonary rehabilitation program involving regular exercise training. In an initial screening, subjects were tested for pulmonary function and gas exchange. They became familiar with exercise testing procedures and the various scales for rating symptom intensity, and completed an incremental, symptomlimited exercise test. Three experimental visits were held at 6-week intervals immediately before the control period, after the control period (pre-ATP visit), and after ATP. Therefore, the subjects acted as their own controls. All visits were conducted at the same time of day for each subject. Data Analysis To compare responses to an identical level of exercise before and after the rehabilitation program, we selected the highest WR tolerated by a given patient during pre-ATP test, (the standardized WR). To compare responses to an identical level of ventilation, we selected the highest V˙e tolerated by a given patient during the pre-ATP test (the standardized V˙e). Nonparametric ratings of exertional breathlessness were compared before and after intervention using the Wilcoxon test. All other measurements made before and after ATP were analyzed using the paired t test. Pearson correlation coefficients were used to test the strength of the association between measured variables. A value of p ⬍ 0.05 was considered significant.
Results Anthropometric and baseline function data of the 12 patients with moderate-to-severe airflow obstrucClinical Investigations
Table 1—Anthropometric and Lung Function Data* Variables Age, yr Male/female gender, No. Height, cm Weight, kg Vital capacity, L Vital capacity, % predicted FEV1, L FEV1, % predicted FEV1/vital capacity, % FRC, L FRC, % predicted TLC, L TLC, % predicted Pao2, mm Hg Paco2, mm Hg pH
Before Control
Before ATP
After ATP
3.48 ⫾ 0.87 86 ⫾ 12.4 1.59 ⫾ 0.58 49 ⫾ 12.3 46 ⫾ 12 4.56 ⫾ 1.2 129 ⫾ 24.5 7.3 ⫾ 0.6 113 ⫾ 9.2 73.5 ⫾ 12.1 42.4 ⫾ 5 7.4 ⫾ 0.03
66.7 ⫾ 7.5 10/2 170 ⫾ 9.9 77.3 ⫾ 18.5 3.52 ⫾ 0.9 86.8 ⫾ 12.8 1.61 ⫾ 0.6 49.1 ⫾ 12.6 46.7 ⫾ 13.9 4.5 ⫾ 1.3 129 ⫾ 25 7.18 ⫾ 0.5 113 ⫾ 9 74.8 ⫾ 12.7 42.5 ⫾ 4.6 7.4 ⫾ 0.03
76.9 ⫾ 18 3.66 ⫾ 1.1 87.5 ⫾ 13 1.63 ⫾ 0.69 51.2 ⫾ 12.8 44.9 ⫾ 13.4 4.6 ⫾ 1.2 127 ⫾ 24 7.6 ⫾ 0.5 115 ⫾ 8 80.6 ⫾ 11.3 41.1 ⫾ 3.2 7.41 ⫾ 0.02
*Data are presented as mean ⫾ SD.
tion and hyperinflation, mild-to-moderate hypoxia, and mild carbon dioxide retention are shown in Table 1. The data did not change during the study. Incremental Exercise Performance At peak arm exercise, V˙e, V˙co2, V˙o2, HR, exercise dyspnea, arm effort, and WR all increased. The ATP did not modify V˙e, V˙co2, V˙o2, HR, exercise dyspnea, and arm effort but increased WR (p ⬍ 0.001) [Table 2]. The relationships of changes in exercise dyspnea and arm effort with changes in V˙e, V˙o2, V˙co2, and the relationships of changes in V˙e with changes in V˙o2 and V˙co2 were not altered with the ATP (Table 3). Changes at Standardized WR With arm exercise, IC decreased by 0.93 ⫾ 0.4 L (from 2.68 ⫾ 0.79 to 1.75 ⫾ 0.63 L, p ⬍ 0.000004)
[mean ⫾ SD]. After the ATP, IC decreased by 0.59 ⫾ 0.27 L (from 2.6 ⫾ 0.83 to 2.01 ⫾ 0.81 L, p ⬍ 0.0001) and was significantly less (p ⬍ 0.01) than before ATP (Table 4). ATP lowered HR (p ⬍ 0.03) and decreased V˙e (p ⬍ 0.01) by lengthening RR (p ⬍ 0.03) but did not modify V˙o2 and V˙co2 (Table 5). Figure 1 (left panel) shows individual changes in both exercise dyspnea (average, from 5.6 ⫾1.3 to 3.6 ⫾ 1.5 au; p ⬍ 0.02) and arm effort (average, from 6.5 ⫾ 2.3 to 5.1 ⫾ 2.3 arbitrary units [au]; p ⬍ 0.01) with the ATP. In most patients, exercise dyspnea and arm exercise decreased at standardized WR. Changes at Standardized V˙E The decrease in IC during arm exercise before ATP (0.68 ⫾ 0.42 L; from 2.69 ⫾ 0.62 to 2.01 ⫾ 0.57 L; p ⬍ 0.03) tended to significantly differ (p ⬍ 0.057) from the decrease after ATP (0.43 ⫾ 0.32 L; from 2.79 ⫾ 0.66 to 2.35 ⫾ 0.75 L; p ⬍ 0.007). ATP reduced exercise dyspnea (from 4 ⫾ 1.7 to 2.75 ⫾1.2 au; p ⬍ 0.02) and arm effort (from 5.6 ⫾ 3 to 4.6 ⫾ 3.1 au; p ⬍ 0.01; Fig 1, right panel), lowered HR (p ⬍ 0.03), but did not modify WR, V˙o2, V˙co2, Vt, and RR (Table 6). Relationships Figure 2 shows the slopes of the relationships of changes in IC with changes in exercise dyspnea (left panel; all patients but patient 11) and arm effort (right panel; all patients but patients 6 and 11) before the ATP (r2 ⫽ 0.25 and r2 ⫽ 0.64, respectively) and after ATP (r2 ⫽ 0.37 and r2 ⫽ 0.60, respectively); no difference was found between before and after ATP. In each patient, a decrease in IC was significantly correlated with increase in V˙o2 (⌬V˙o2/ ⌬IC, 0.01 ⫾ 0.03 L/min/L; from r ⫽ 0.73 to 0.96). This correlation did not significantly change after
Table 2—Peak Exercise Response Before and After Rehabilitation* After Control Variables
Before Control
Before ATP
After ATP
p Value†
WR, W V˙e, L/min V˙co2, L/min V˙o2, L/min Exertional dyspnea, au Arm effort, au IC, L HR, beats/min RR, breaths/min
39.4 ⫾ 12.6 39.5 ⫾ 8.9 1.16 ⫾ 0.3 1.11 ⫾ 0.26 5.3 ⫾ 1.22 6.2 ⫾ 2.3 1.76 ⫾ 0.62 130.7 ⫾ 17.7 31.6 ⫾ 12.9
41.1 ⫾ 13.6 41 ⫾ 9.3 1.17 ⫾ 0.29 1.12 ⫾ 0.23 5.5 ⫾ 1.4 6.4 ⫾ 2.6 1.8 ⫾ 0.6 134 ⫾ 19.9 32.4 ⫾ 11.6
48.3 ⫾ 14.5 44.3 ⫾ 7.6 1.34 ⫾ 0.27 1.23 ⫾ 0.22 5.75 ⫾ 2.1 6.3 ⫾ 2.6 2.1 ⫾ 0.7 135.5 ⫾ 19.6 31.9 ⫾ 10.2
⬍ 0.04 NS NS NS NS NS NS NS NS
*Data are presented as mean ⫾ SD. NS ⫽ not significant. †After vs before ATP. www.chestjournal.org
CHEST / 128 / 3 / SEPTEMBER, 2005
1227
Table 3–Slopes of the Relationships of Changes in Exertional Dyspnea, Arm Effort, and V˙E with changes in V˙O2 and VCO2* After Control Variables
Before Control
Before ATP
After ATP
⌬ED/⌬V˙o2, au/% maximal V˙o2 ⌬ED/⌬V˙co2, au/L/min ⌬ED/⌬V˙e, au/L/min ⌬AE/⌬V˙o2 ⌬AE/⌬V˙co2, au/L/min ⌬AE/⌬V˙e, au/L/min ⌬V˙e/⌬V˙co2 ⌬V˙e/⌬V˙o2, L/min/% maximal V˙o2 ⌬V˙co2/⌬V˙o2, L/min/% maximal V˙o2
0.15 ⫾ 0.02 5.3 ⫾ 2.9 0.2 ⫾ 0.1 0.17 ⫾ 0.15 0.17 ⫾ 0.1 0.2 ⫾ 0.13 28.1 ⫾ 4 0.51 ⫾ 0.1 0.6 ⫾ 0.12
0.13 ⫾ 0.06 5.6 ⫾ 3.2 0.18 ⫾ 0.13 0.19 ⫾ 0.17 0.18 ⫾ 0.15 0.21 ⫾ 0.11 29.2 ⫾ 4.7 0.56 ⫾ 0.12 0.8 ⫾ 0.16
0.12 ⫾ 0.04 4.6 ⫾ 1.6 0.21 ⫾ 0.20 0.13 ⫾ 0.06 0.14 ⫾ 0.07 0.23 ⫾ 0.17 27.4 ⫾ 6.7 0.6 ⫾ 0.1 0.7 ⫾ 0.2
*Data are presented as mean ⫾ SD. ED ⫽ exertional dyspnea; AE ⫽ arm effort.
ATP (⌬V˙o2/⌬IC, 0.009 ⫾ 0.01 L/min/L; r ⫽ 0.67 to 0.98). Two representative subjects with the highest (patient 6, left panel) and the lowest (patient 7, right panel) response slopes of this relationship during arm exercise before and after ATP are shown in Figure 3. The relationship between IC and V˙o2 indicates that dynamic hyperinflation predicts a good amount of the variability in V˙o2 during arm exercise. Discussion Arm exercise resulted in dynamic hyperinflation, which directly correlated with increases in dyspnea, arm effort, and V˙o2. The ATP resulted in the following: (1) a significant increase in exercise capacity at peak exercise; (2) a decrease in both ventilation and dynamic hyperinflation, primarily due to a de-
Table 4 —IC at Standardized WR (41.1 ⴞ 13.6 W) IC After Control
Patient No.
IC Before Control, L
Before ATP
After ATP
Start
End
Start
End
Start
End
1 2 3 4 5 6 7 8 9 10 11 12 Mean SD p Value*
3 3.5 2.78 2.6 2.75 3.5 1.4 3.3 1.6 3.8 3.4 1.4 2.75 0.85 ⬍ 0.00004
1.68 2.5 2.4 1.1 1.2 3.0 1.2 2.6 0.8 2.0 2.1 0.8 1.78 0.75
3.1 3.26 2.9 2.4 2.7 3.16 1.58 3.53 1.47 3.5 3.16 1.4 2.68 0.79 ⬍ 0.000004
1.72 2.2 2.3 1.2 1.5 1.98 1.1 3.0 1 2.1 1.98 0.9 1.75 0.63
3.3 2.8 2.95 2.68 2.62 2.49 2.36 3.74 1.34 3.81 2.5 1.57 2.60 0.83 ⬍ 0.0001
2.3 2.6 2.1 2.4 1.9 1.36 1.34 3.34 0.98 3.3 1.36 1.08 2.01 0.81
*Start vs end IC. 1228
crease in RR at standardized WR; and (3) a decrease in dyspnea and arm effort at standardized WR and ventilation. The novel finding of this study is the dynamic hyperinflation, ie, the increase in dynamic EELV during supported-arm exercise in patients with COPD. To the extent that TLC does not change appreciably during exercise in COPD,17 a change in IC accurately reflects the change in EELV.18 Dynamic hyperinflation minimizes expiratory flow limitation during lower-limb exercise in patients with severe COPD,2,3,8,20 –22 but puts the respiratory muscles in a disadvantageous portion of their length/ tension relationship and imposes an elastic load on the muscles during inspiration.2,3 Alison et al6 found that arm exercise may either increase EELV or reduce its decrease compared to leg exercise in healthy subjects, and diminish exercise capacity in flow-limited patients with cystic fibrosis. We explain dynamic hyperinflation with arm exercise in patients with COPD as being due to inadequate time to exhale the volume required to maintain EELV before the next inspiration begins. Due to the load it places on an already stressed system, arm exercise contributes to dyspnea and disability in many patients with COPD.23,24 In particular, the following are seen: (1) hyperinflation decreases maximal force-generating capacity of the respiratory muscle,25,26 thereby increasing the central motor output to a weakened muscle; (2) the decrease in force-generating capacity increases the sense of inspiratory effort25; and (3) the association of an increased motor output with an increased respiratory system impedance increases the respiratory muscle load and may affect the coupling between inspiratory effort and concurrent inspiratory volume (neuromuscular dissociation of the ventilatory pump).2,3 Another finding of this study is the association Clinical Investigations
Table 5—Exercise Response at Standardized WR (41.1 ⴞ 13.6 W)* After Control Variables
Before Control
Before ATP
After ATP
p Value†
V˙e, L/min Vt, L/min RR, breaths/min V˙co2, L/min V˙o2, L/min Exertional dyspnea, au Arm effort, au HR, beats/min V˙e/V˙co2 V˙e/V˙o2
40.1 ⫾ 8.6 1.26 ⫾ 0.1 30.6 ⫾ 12.9 1.16 ⫾ 0.3 1.11 ⫾ 0.26 5.3 ⫾ 1.22 6.2 ⫾ 2.3 130 ⫾ 13 34.8 ⫾ 6.1 36.2 ⫾ 6.2
41.1 ⫾ 8.9 1.26 ⫾ 0.1 29.5 ⫾ 10.8 1.16 ⫾ 0.3 1.1 ⫾ 0.2 5.6 ⫾ 1.3 6.5 ⫾ 2.3 126 ⫾ 25 35.9 ⫾ 6.9 37.1 ⫾ 6.8
37.7 ⫾ 8.5 1.32 ⫾ 0.5 27 ⫾ 1 1.15 ⫾ 0.3 1.1 ⫾ 0.18 3.6 ⫾ 1.5 5.1 ⫾ 2.3 119 ⫾ 20 33 ⫾ 6.8 33.7 ⫾ 6.4
⬍ 0.01 NS ⬍ 0.03 NS NS ⬍ 0.02 ⬍ 0.01 ⬍ 0.03 ⬍ 0.001 ⬍ 0.001
*Data are presented as mean ⫾ SD. See Table 2 legend for expansion of abbreviation. †Before ATP vs after ATP.
between arm effort and dynamic hyperinflation. The rib cage and abdominal wall must be fixed to maintain the position of the torso during upper-limb exercise.27 This may result in a stiffer rib cage and the need for maintaining a particular ventilation by increasing the RR, and thereby dynamic hyperinflation, while the central output to the exercising arms increases. Last but not least, the increase in dynamic hyperinflation predicted an average 69% of the variability in V˙o2 (from 36 to 92%; Fig 3). These data confirm the results by Diaz et al28 during incremental, symptom-limited exercise on a cycle ergometer in stable patients with COPD. In that study, resting IC (IC percentage of predicted) explained 56% of the variance in maximum V˙o2. Our data extend those re-
sults, in that dynamic hyperinflation during exercise largely predicts V˙o2. Factors other than mechanical factors, such as deconditioning, are likely to explain the residual variance in V˙o2. The present data also show that, regardless of leg or arm cycling, dynamic hyperinflation is a predictor of exercise tolerance in patients with COPD.28 Training that includes both upper- and lower-limb exercise may improve upper-extremity exercise endurance7,9 –11 and ratings of dyspnea.7,11 The effects of the ATP on dyspnea reduction must, however, be defined. Although respiratory function has generally been reported not to change after pulmonary rehabilitation programs, a mild decrease in FRC after upper-extremity exercise training has been found in patients with COPD.11 Ries et al11 showed, however,
Figure 1. Changes in exercise dyspnea (ED) and arm exercise (AE) with an ATP at standardized WR (left panel) and standardized V˙e (right panel). www.chestjournal.org
CHEST / 128 / 3 / SEPTEMBER, 2005
1229
Table 6 —Exercise Response at Standardized V˙E (37.3 ⴞ 9.5 L/min)* After Control Variables
Before Control
Before ATP
After ATP
p Value†
WR, W Vt, L/min RR, breaths/min V˙co2, L/min V˙o2, L/min Exertional dyspnea, au Arm effort, au HR, beats/min
37.8 ⫾ 12.3 1.41 ⫾ 0.4 29 ⫾ 10.8 1.1 ⫾ 0.3 1.0 ⫾ 0.25 4 ⫾ 1.7 5.6 ⫾ 3 121 ⫾ 19.2
35.5 ⫾ 16.4 1.46 ⫾ 0.5 29 ⫾ 10.8 1.1 ⫾ 0.3 1.0 ⫾ 0.25 4 ⫾ 1.7 5.6 ⫾ 3 122 ⫾ 21
37.2 ⫾ 17.1 1.51 ⫾ 0.5 27.1 ⫾ 8.8 1.1 ⫾ 0.35 1.1 ⫾ 0.25 2.75 ⫾ 1.2 4.6 ⫾ 3.1 113 ⫾ 14
NS NS NS NS NS ⬍ 0.02 ⬍ 0.01 ⬍ 0.03
*Data are presented as mean ⫾ SD. See Table 2 legend for expansion of abbreviation. †Before vs after ATP.
significant bronchodilator responsiveness at baseline, which may account for some of this improvement during training period. Previous studies29,30 of inhaled bronchodilator therapy in patients with COPD have shown that even a moderate reduction in dynamic hyperinflation may contribute to an improvement in dyspnea. Likewise, we have shown that rehabilitation programs based on leg exercise training decrease dynamic hyperinflation and reduce dyspnea.8 The present data clearly show the association between decreases in dynamic hyperinflation, RR, dyspnea, and arm effort at standardized WR. We explain the decrease in dynamic hyperinflation as being due to the increased time to exhale the volume required to maintain EELV before the next inspiration begins. We found a significant relationship between IC and dyspnea and arm effort ratings during arm exercise (Fig 2), with the correlation coefficients being, however, as low as that found during leg exercise.8 These data suggest that, although dynamic
hyperinflation contributes to exertional dyspnea in patients with COPD, additional physiologic and/or sensory factors are also important to the perception of dyspnea and arm effort during exercise. The reasons are as follows: (1) unlike the lack of changes in the slopes of ventilation with both V˙o2 and V˙co2 (Table 3), ventilation decreased at standardized WR (Table 4); (2) changes in ventilation explained a consistent amount of changes in both dyspnea and arm effort before (r2 ⫽ 0.62 and r2 ⫽ 0.64, respectively) and after the ATP (r2 ⫽ 0.44 and r2 ⫽ 0.60, respectively); and (3) tolerance or desensitization to dyspnea and arm effort may allow patients to perform higher levels of work with reduced symptoms.7,31,32 And indeed, at standardized ventilation, both arm effort and dyspnea were lower after the ATP as compared to before (Table 6). These findings indicate that both mechanical constraints and desensitization contribute to arm exercise-induced dyspnea after the training program. Furthermore,
Figure 2. Left panel: Slopes of the relationships of changes in exercise dyspnea (⌬ED) with changes in IC (⌬IC) before and after ATP. Right panel: Slopes of the relationships of changes in arm exercise (⌬AE) with changes in IC before and after the ATP. Individual data points are shown. Bars are mean ⫾ SD. 1230
Clinical Investigations
Figure 3. Relationships of changes in IC with changes in V˙o2 before and after the ATP in two representative subjects (patient 6, left panel; and patient 7, right panel). sl ⫽ slope.
breathing retraining may have actually improved the breathing pattern, slowing the RR.33 A variable percentage of COPD patients stop leg exercise because of both dyspnea and leg effort, while others mostly stop because of either one or the other.7,8,34 The present results showing in each subject a similar amount (increase and decrease) of both dyspnea and arm effort are consistent with the neural motor output being balanced between respiratory muscles, and arm and torso muscles. Even though the small series of patients in this study should prevent us from drawing definite conclusions, our data indicate that arm exercise may result in the association of dynamic hyperinflation, dyspnea, and arm effort in patients with COPD. The ATP increases arm endurance, modulates dynamic hyperinflation, and reduces symptoms.
7
8
9
10
11 12
References 1 Pride N. Respiratory muscle activation during exercise in chronic obstructive pulmonary disease. In: Jones NL, Killian KJ, eds. Breathlessness: the Campbell symposium. Hamilton, ON, Canada: Boehringer Ingelheim, 1992; 52–56 2 O’Donnell DE, Webb KA. Exertional breathlessness in patients with chronic airflow limitation: the role of lung hyperinflation. Am Rev Respir Dis 1993; 148:1351–1357 3 O’Donnell DE, Revill SM, Webb KA. Dynamic hyperinflation and exercise intolerance in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2001; 1624:770 –777 4 Martinez FJ, Couser JI Jr, Celli BR. Respiratory response to arm elevation in patients with chronic obstructive pulmonary disease. Chest 1991; 143:476 – 480 5 Dolmage TE, Maestro L, Avendano MA, et al. The ventilatory response to arm elevation of patients with chronic obstructive pulmonary disease. Chest 1993; 104:1097–1100 6 Alison JA, Regnis JA, Donnelly PM, et al. End-expiratory lung volume during arm and leg exercise in normal subjects and www.chestjournal.org
13 14 15 16 17 18
19
patients with cystic fibrosis. Am J Respir Crit Care Med 1998; 158:1450 –1458 O’Donnell DE, McGuire MA, Samis L, et al. General exercise training improves ventilatory and peripheral muscle strength and endurance in chronic airflow limitation. Am J Respir Crit Care Med 1998; 157:1489 –1497 Gigliotti F, Coli C, Bianchi R, et al. Exercise training improves exertional dyspnea in patients with COPD: evidence of role of mechanical factors. Chest 2003; 123:1794 – 1802 Couser J Jr, Martinez FJ, Celli BR. Pulmonary rehabilitation that includes arm exercise reduces metabolic and ventilatory requirements for simple arm elevation. Chest 1993; 103: 37– 41 Martinez FJ, Vogel PD, Dupont DN, et al. Supported arm exercise vs unsupported arm exercise in the rehabilitation of patients with severe chronic airflow obstruction. Chest 1993; 103:1397–1402 Ries AL, Ellis B, Hawkins RW. Upper extremity exercise training in chronic obstructive pulmonary disease. Chest 1988; 93:688 – 692 Casaburi R, Wasserman K. Exercise training in pulmonary rehabilitation. N Engl J Med 1986; 314:1509 –511 Scano G, Garcia-Herreros P, Stendardi L, et al. Cardiopulmonary adaptation to exercise in coal miners. Arch Environ Health 1980; 35:360 –366 Gorini M, Duranti R, Misuri G, et al. Aminophylline and respiratory muscle interaction in normal humans. Am Rev Respir Dis 1994; 149:1227–1236 European Community for Coal and Steel. Standardization of lung function test. Eur Respir J 1993; 6(Suppl 16):1–100 Jones NL, Campbell AJM. Clinical exercise testing. 2nd ed. Philadelphia, PA: W.B. Saunders, 1982 Stubbing DG, Pengelly LD, Morse JC, et al. Pulmonary mechanics during exercise in subjects with airflow obstruction. J Appl Physiol 1980; 49:511–515 Yan S, Kaminski D, Sliwinski 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 1997; 156:55–59 Borg GAV. Psychophysical basis of perceived exertion. Med Sci Sports Exerc 1982; 14:377–381 CHEST / 128 / 3 / SEPTEMBER, 2005
1231
20 Grimby G, Elgefors B, Oxhoj H. Ventilatory levels and chest wall mechanics during exercise in obstructive lung disease. Scand J Respir Dis 1973; 54:45–52 21 Potter WA, Olafsson S, Hyatt RE. Ventilatory mechanics and expiratory flow limitation during exercise in patients with obstructive lung disease. J Clin Invest 1971; 50:910 –918 22 Dodd DS, Brancatisano T, Engel LA. Chest wall mechanics during exercise in patients with severe chronic airflow obstruction. Am Rev Respir Dis 1984; 129:33–38 23 Celli B, Rassulo J, Make BJ. Dyssynchronous breathing during arm but not leg exercise in patients with chronic airflow obstruction. N Engl J Med 1986; 314:1485– 490 24 Criner G, Celli BR. Effect of unsupported arm exercise on ventilatory muscle recruitment in patients with severe chronic airflow obstruction. Am Rev Respir Dis 1988; 138:856 – 861 25 Leblanc P, Summers E, Inman MD, et al. Inspiratory muscles during exercise: a problem of supply and demand. J Appl Physiol 1988; 64:2482–2489 26 El-Manshawi A, Killian KJ, Summers E, et al. Breathlessness during exercise with and without resistive loading. J Appl Physiol 1986; 61:896 –905 27 Couser JI Jr, Martinez FJ, Celli BR. Respiratory response and ventilatory muscle recruitment during arm elevation in normal subjects. Chest 1992; 101:336 –340 28 Diaz O, Villafranca C, Ghezzo H, et al. Role of inspiratory
1232
29
30
31 32 33
34
capacity on exercise tolerance in COPD patients with and without tidal expiratory flow limitation at rest. Eur Respir J 2000; 16:269 –275 Belman MJ, Botnick WC, Shin JW. Inhaled bronchodilators reduce dynamic hyperinflation during exercise in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1996; 153:967–975 O’Donnell DE, Lam M, Webb KA. Spirometric correlates of improvement in exercise performance after anticholinergic therapy in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1999; 160:542–549 Belman MJ, Brooks LR, Ross DJ, et al. Variability of breathlessness measurement in patients with chronic obstructive pulmonary disease. Chest 1991; 99:566 –571 Carrieri-Kohlman V, Gormley JM, Douglas MK, et al. Exercise training decrease dyspnea and the distress and anxiety associated with it. Chest 1996; 110:1526 –1535 Casaburi R, Porzasz J, Burns MR, et al. Physiologic benefit of exercise training in rehabilitation of patients with severe chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1997; 155:1541–1551 Killian KJ, Leblanc P, Martin DH, et al. Exercise capacity and ventilatory, circulatory and symptom limitation in patients with chronic airflow limitation. Am Rev Respir Dis 1992; 146:935–940
Clinical Investigations