Respiratory Medicine (2008) 102, 1425e1430

ava i lab le at www.sc ienced i rec t . com

j ourna l homepage : www.e lsev ier . com/ loca te / rmed

Use of 6-min and 12-min walking test for assessingthe efficacy of formoterol in COPD

Mario Cazzola a,*, Gian Luca Biscione b, Franco Pasqua b,Girolmina Crigna b, Massimiliano Appodia a, Vittorio Cardaci c, Luigi Ferri c

a Unit of Respiratory Diseases, Department of Internal Medicine, University of Rome ‘Tor Vergata’, Rome, Italyb Unit of Pulmonary Rehabilitation, San Raffaele Hospital, Velletri, Italyc Unit of Pulmonary Rehabilitation, IRCSS San Raffaele Hospital, Rome, Italy

Received 30 March 2008; accepted 17 April 2008Available online 14 July 2008

KEYWORDSCOPD;6-MWT;12-MWT;Formoterol

* Corresponding author. DipartimUniversita di Roma ‘Tor Vergata’, ViaItaly. Tel.: 39 34 8641 2311; fax: 39 0

E-mail address: mario.cazzola@un

0954-6111/$ - see front matter ª 200doi:10.1016/j.rmed.2008.04.017

Summary

Exercise tolerance is an important outcome measure in patients with COPD, mostly becausethere is evidence that exercise testing is superior to other functional measurements obtainedat rest in demonstrating the positive effect of a specific intervention. We assessed the effectof a 5-day treatment with formoterol 12 mg twice daily on lung function, exercise capacityand dyspnea in 22 stable COPD patients, and compared 6-MWT with 12-MWT in evaluatingformoterol efficacy. All subjects entered a crossover design. They underwent 6-MWT or 12-MWT in a randomised order and soon after started the 5-day treatment. After a 3-day washout,patients who had first performed 6-MWT, underwent 12-MWT, and the contrary. Formoterolinduced a progressively significant increase in pre-drug FEV1 and IC and also significantchanges in these parameters 2 h after its inhalation at each test day. Moreover, it increasedthe walked distance by 53.6 m at the end of 6-MWT and 59.9 m at the end of 12-MWT. For-moterol also induced a significant change in Borg score for dyspnea caused by the 6-MWTwhen compared with the pre-treatment values, whereas it significantly changed dyspnea in-duced by 12-MWT only after the first dose. Our study not only strengthens the importance ofwalking tests as a useful tool for evaluating the impact of a bronchodilator on some COPDpatient-centred outcomes, but also indicates that 6-MWT seems to be a more appropriate in-strument than 12-MWT for assessing the exercise response to a bronchodilator in COPD.ª 2008 Published by Elsevier Ltd.

ento di Medicina Interna,Montpellier 1, 00133 Rome,

6 7259 6621.iroma2.it (M. Cazzola).

8 Published by Elsevier Ltd.

Introduction

Exercise tolerance has become an important outcomemeasure in patients with COPD, mostly because evi-dence has been provided that exercise testing is

Table 1 Patients’ demographics and disease severity(GOLD) classification

Sex Age(yrs)

FEV1

(% predicted)FEV1/FVC

PaO2 PaCO2 GOLDseverityclassification

M 72 40 62 70.1 41.5 IIIM 56 29 51 68.7 45 IVF 77 45 51 68.2 35.6 IIIM 51 45 68 80.2 38.7 IIIM 70 49 65 73.6 44.9 IIIM 71 19 36 68 47.5 IVM 81 25 42 71 51.5 IVF 68 28 50 67 45 IVF 62 28 69 69.4 49.2 IVF 65 36 41 68.1 48.8 IIIM 73 49 60 65.8 38.5 IIIF 72 48 69 68.9 34.4 IIIF 77 53 69 74.7 42.8 IIM 61 14 25 63.3 46 IVF 74 55 40 51.3 34.5 IIM 72 49 43 60.3 40.1 IIIM 75 56 60 84.6 44.4 IIM 67 48 52 85.5 40 IIIM 58 55 60 70.5 41.1 IIM 65 45 55 64.5 38.9 IIIF 71 38 43 59.0 42.3 IIIM 62 42 49 62.8 40.7 III

1426 M. Cazzola et al.

superior to other functional measurements obtained atrest (e.g., forced expiratory volume in 1 s [FEV1]) indemonstrating the positive effect of a specific interven-tion.1 Pepin et al.2 recently showed that walking isa more sensitive exercise modality to evaluate theresponse to bronchodilators than cycling in patientswith COPD. This may be especially true for patientswith advanced disease in whom skeletal muscle dysfunc-tion is more prevalent.

Part of the rationale for using the walking tests ratherthan the bicycle or treadmill exercise in clinical trials isthat it is a more natural form of exercise and may betterreflect daily activity. Walking is one of the simplest andmost common activities, and consequently it is an in-tegrative useful test to assess the physical, psychological,and emotional capabilities of patients. It is not a surprise,therefore, that field tests are widely used in trialsexploring the benefits of rehabilitation, pharmaceuticalintervention, oxygen supplementation and surgery incardiorespiratory disease.3 In particular, timed walkingtests have been shown to predict survival4 and utilizationof health-care resources5 in COPD patients. The 6-minwalking test (6-MWT) and 12-min walking tests (12-MWT)are the most frequently used field tests to assess exercisecapacity.

The 6-MWT, which has been shown to be a submaximalhigh-intensity constant-load test,1 has been widely used inmany large trials exploring the benefits of pharmaceuticalintervention,6 but only modest changes in 6-MWT of littleclinical significance have been reported following acutebronchodilation in COPD.7 Moreover, recent research hasbeen suggested that the 6-MWT test may not be as respon-sive to bronchodilation as the endurance shuttle walk.8

The 6-MWT is a self-paced test (e.g., patients determinetheir own walking speed) of a fixed duration. In contrast,the endurance shuttle walk is an externally paced test(e.g., walking speed is dictated to patients) of an indefi-nite duration. Improvements in performance are, there-fore, achieved differently for the two tests.8 Patientshave to increase walking speed to cover more distanceon the 6-MWT, while they have to increase endurancetime to achieve the same outcome on the endurance shut-tle walk.8

It has been highlighted that patients usually startwalking at a faster pace before settling to a remarkablyconstant speed.9 Shortening the walking time thus tends tomeasure the initial spurt rather than the true exercisetolerance. Moreover, it has been suggested that a longertest (12-MWT) is the best walking test for assessing thefunctional capabilities of a patient.10 The basis for this con-clusion is the observation that changes in Vo2max correlatemore closely with changes in the 12-MWT than with changesin other walking tests.10

Nonetheless, to our best knowledge no study has directlycompared the changes induced by a treatment witha bronchodilator in exercise performance detected by the6-MWT with those detected by the 12-MWT.

Therefore, the present study was undertaken inpatients with stable COPD to directly compare thechanges in exercise performance detected by the 6-MWT test with those detected by the 12-MWT test inthese patients.

Methods

Subjects

Twenty-two clinically stable COPD patients capable ofperforming a walking test were included in the study.Patients with overt comorbidity preventing them fromsafely performing an exercise test could not participate inthis trial. In particular, patients did not suffer fromcardiovascular, neurological disorders or advanced osteo-arthrosis. As part of the characterisation procedures,resting pulmonary function testing and a practice-walkingtest were carried out in all patients. Patients who requiredO2 during the walking test were excluded from the study.All patients were ex-smokers and all had experience ofwalking tests being included in a long-term respiratory re-habilitation programme, but none practiced a practice-test immediately before being enrolled in the study.

The selected patients gave their consent to participatein the study. All trial procedures were conducted accordingto the Declaration of Helsinki at the San Raffaele Hospital,Velletri, and IRCSS San Raffaele Hospital, Rome, Italy.

Table 1 illustrates patients’ demographics and diseaseseverity (GOLD) classification.

Study design

After baseline assessment of pulmonary function, andwalking capacity (before exercise testing, subjects werefamiliarized with the Borg scale11 and its endpoints were

Walking test for assessing the efficacy of formoterol 1427

anchored such that zero represented ‘‘no breathlessness’’and 10 was ‘‘the most severe breathlessness that they hadever experienced or could imagine experiencing’’), and 1week of bronchodilator washout during which the patientscould use salbutamol 200 mg as needed as ‘‘rescue’’ medica-tion, all subjects entered a crossover design where theyunderwent the 6-MWT or the 12-MWT in a randomised orderand soon after started a 5-day treatment with formoterol12 mg bid. FEV1, inspiratory capacity (IC), walked distance(m), and changes in Borg score from rest to the end of thewalking test were the variables measured on day 1, beforeand 2 h after the first inhalation of formoterol, and on day6 before and 2 h after the last inhalation of formoterol.After a further 3-day washout during which, again, thepatients could use salbutamol 200 mg as needed as ‘‘rescue’’drug, patients who had first performed 6-MWT, underwent12-MWT, and the contrary. In this second phase, we followedthe same protocol used in the first phase.

Walking test protocol

Each clinical site used the same walking course for allparticipants. The timed walking tests were conducted in anenclosed corridor on a flat course 30 m in length accordingto the procedures recommended by the American ThoracicSociety (ATS).12 To avoid interference with the patients’exercise performance, patients walked unaccompaniedand no encouragement was given.13 Oxygen saturation(SpO2) and heart rate (HR) were recorded continuously bypulse oximetry. At the end of the walking tests, the partic-ipant was told to stop, and total distance covered was cal-culated to the nearest metre.

Pulmonary function testing

Spirometric data were obtained according to recommendedtechniques.14 The patients were investigated in a sittingposition. IC was determined by a slow manoeuvre (slow inspi-ration until maximum volume after regular tidal breathing).

Statistical analysis

The results of the study are expressed as mean and 95% CIunless otherwise stated. The extent to which the two timedfield tests were able to capture changes in the measuredvariables after bronchodilation was examined by the abilityof the two exercise modalities to detect statisticallysignificant changes using a paired t test. The end-exercisevalues for the measured variables obtained during 6-MWTand 12-MWT were compared using paired t tests and Blandand Altman analyses.15 A p< 0.05 level of significance wasused for all analyses.

Results

In general, formoterol induced a progressively significantincrease in pre-drug FEV1 and IC and also significantchanges in these parameters 2 h after its inhalation ateach test day (Table 2).

After the first formoterol dose, both 6-MWT and 12-MWTimproved walked distance (35.3 m, 95% CI: 7.7e62.9,

p Z 0.0145, and 45.8 m, 95% CI: 24.1e67.4, p Z 0.0003,respectively) (Fig. 1). This improvement was observedalso after the 5-day treatment, before the inhalation ofthe last dose that induced a further increase to 53.6 m(95% CI: 17.6e89.3, p Z 0.0055) and 59.9 m (95% CI:11.2e108.5, p Z 0.0182) in 6-MWT and 12-MWT, respec-tively, when compared with the pre-treatment values.However, we were unable to record a further significantincrease after the inhalation of the last dose in 12-MWTwhen compared with the pre-inhalation values (Fig. 1). Itis intriguing that the difference in the walked distancebetween 6-MWT and 12-MWT, which was 303.5 m (95% CI:263.6e343.5) before and 314.0 m (95% CI: 272.6e355.4)after the first inhalation of formoterol, decreased from338.6 m (95% CI: 267.0e410.2) before to 309.8 m (95% CI:247.7e371.9) after the inhalation of the last dose of thesame bronchodilator.

There was a statistically significant difference in therate of change in Borg score for dyspnea after the first doseof formoterol (�1.5, 95% CI: �2.4 to �0.5, p Z 0.0043) andat the end of the 5-day treatment (�2.3, 95% CI: �3.6e1.1,p Z 0.0010) when compared with the pre-treatment valuesonly with the 6-MWT, although the last dose did notinduced a further significant decrease when comparedwith the pre-inhalation of the last dose values (Fig. 2). For-moterol significantly changed dyspnea induced by the12-MWT only after the first dose (�0.6, 95% CI: �1.1e0.1,p Z 0.0241) (Fig. 2). The end-exercise values for the Borgscore obtained during 6-MWT and 12-MWT were signifi-cantly different at the end of the 5-day treatment bothbefore (�1.7, 95% CI: �2.9e0.6, p Z 0.0045) and after(�1.8, 95% CI: �2.8 to �0.8, p Z 0.0011) the inhalationof the last dose.

Table 3 shows changes in HR and SpO2 induced by formo-terol and walking tests. Formoterol influenced neither pre-test HR and SpO2 nor exercise-induced HR and SpO2

changes.There was no evidence that a specific order of testing

influenced our data.

Discussion

Our study has documented that a 5-day treatment withformoterol 12 mg twice daily increased the walked distanceby 53.6 m at the end of the 6-MWT and 59.9 m at the end ofthe 12-MWT. This is, in our opinion, an important findingbecause, at least for the 6-MWT, none of the studies thathave explored the effects of bronchodilators in COPDreached the minimal clinical significance limits and, conse-quently, the relevance of the observed effect to the patientremains debatable.8 It must be pointed out that Redelmeieret al.16 suggested that subjects have to improve their walk-ing distance by 54 m after a 6-MWT in order to appreciatethis increase as a beneficial effect, whereas Troosterset al.17 reported a mean increase of 52 m in a study ofthe short- and longer-term benefits of pulmonary rehabili-tation delivered over a 6-month period. As far as we areaware, the minimal clinical significance levels in the othertests of exercise capacity have not been defined. There-fore, also for the 12-MWT, the minimal clinically relevantdistance is unknown.

Table 2 Changes in FEV1 and IC induced by formoterol 12 mg twice daily

6-MWT 12-MWT

FEV1 (L) IC (L) FEV1 (L) IC (L)

1st day before first dose 1.019 (0.8372e1.201) 1.797 (1.511e2.082) 1.027 (0.826e1.229) 1.676 (1.437e1.916)1st day 2 h after first dose 1.186 (0.953e1.419) 2.182 (1.956e2.408) 1.200 (1.004e1.396) 2.005 (1.728e2.281)6th day before last dose 1.135 (0.949e1.322) 1.945 (1.695e2.194) 1.139 (0.936e1.342) 2.063 (1.775e2.351)6th day after last dose 1.185 (0.979e1.391) 2.045 (1.737e2.353) 1.231 (1.016e1.446) 2.205 (1.952e2.457)

Values are mean and 95% CI.

1428 M. Cazzola et al.

It is interesting that our data show that a 5-daytreatment with formoterol 12 mg twice daily has allowednot only to reach the minimal clinical significance limit pro-posed by Troosters et al.,17 but also to obtain a significantchange in Borg score for dyspnea induced by the 6-MWTwhen compared with the pre-treatment values.

We believe that these results support the use of the6-MWT as a relatively simple test to be used extensively intrials to evaluate possible benefits of pharmacologicalintervention as suggested by the ATS/ERS Task Force onoutcomes for COPD pharmacological trials.18 Obviously, thisopinion contrasts with the documentation that the 6-MWT isnot really responsive for detecting changes in exerciseperformance following bronchodilation.6 It is well knownthat the 6-MWT is not physiologically pure because it isself-paced and, unless appropriately conducted, may besubject to learning and motivation effects.19 Furthermore,it is a physiological hybrid test that may produce erraticmetabolic demand that may include peak or endurance per-formance.19 Nonetheless, the 6-MWT is probably the mostpopular investigation because of its simplicity, relevanceto daily life, widespread use and availability of referencevalues.19

We must highlight that in this study we have usedformoterol as test bronchodilator because there is evidencethat inhaled formoterol 12 mg twice daily is effective inameliorating exercise intolerance even in patients with ad-vanced COPD.20 In effect, it is able to reduce the level of

1D before 1D after 6D before 6D after

300

325

350

375

400

425

600

625

650

675

700

725***

* ***

***

#

6-M

WT

, m

12-M

WT

, m

6-MWT12-MWT

Figure 1 Changes in distance walked with formoterol for the6 min walking test (6-MWT) and the 12 min walking test(12-MWT). Values are mean� SE. *p< 0.05, **p< 0.01,***p< 0.001 vs. pre-treatment; #p< 0.05 vs. pre-last dose. 1Dbefore, first day before inhalation of formoterol; 1D after, firstday after inhalation of formoterol; 6D before, sixth day beforeinhalation of formoterol; 6D after, sixth day after inhalation offormoterol.

resting and exercise-related dynamic hyperinflation withconsequential benefits on effort tolerance and breathless-ness on daily living in severely disabled patients with ad-vanced COPD.20 The degree of dynamic hyperinflation canbe assessed by measuring reduction in IC.21 It has been sug-gested that changes in IC, which might imply improvementsin dynamic hyperinflation, can reduce exercise-induceddyspnea.22 In our patients, formoterol induced a significantimprovement in IC. However, this improvement elicitedchanges in dyspnea and walked distance only after the6-MWT. In fact, formoterol significantly changed dyspneainduced by the 12-MWT only after the first dose whereasthere was a significant difference between the 6-MWT and12-MWT in favour of the 6-MWT at the end of the 5-daytreatment with formoterol. Moreover, the difference inthe walked distance between 6-MWT and 12-MWTdecreased after the inhalation of the last dose of thisbronchodilator.

It is intriguing that three double-blind studies usinganother long-acting b2-agonist salmeterol, 50e100 mg bid,did not find a significant effect on walking distance (after4, 16, and 12 weeks, respectively), despite a significantimprovement in FEV1 in the first two studies8 and thefact that salmeterol is able to reduce lung hyperinflationat rest and during exercise with associated improvementsin exercise endurance time.23 It must be highlighted thatO’Donnell et al.24 studied the effect of ipratropiumbromide on exercise tolerance in patients with COPD.

1D before 1D after 6D before 6D after

2.02.53.03.54.04.55.05.56.0

6-MWT12-MWT

*

*****

**

B

org

sco

re

Figure 2 Changes in the exercise-induced dyspnea (Borgscore) with formoterol for the 6 min walking test (6-MWT)and the 12 min walking test (12-MWT). Values are mean� SE.*p< 0.05; **p< 0.01; ***p< 0.001 vs. pre-treatment. 1Dbefore, first day before inhalation of formoterol; 1D after, firstday after inhalation of formoterol; 6D before, sixth day beforeinhalation of formoterol; 6D after, sixth day after inhalation offormoterol.

Table

3Change

sin

heart

rate

and

SpO

2

6-M

WT

12-M

WT

HR

(beats

/min

)Sp

O2

(%)

HR

(beats

/min

)Sp

O2

(%)

Pre

-test

Nadir

Pre

-test

Nadir

Pre

-test

Nadir

Pre

-test

Nadir

1st

day

befo

refirs

tdose

82.6

(76.

3e88

.8)

103.

5(9

8.3e

108.

8)94

.8(9

3.7e

95.6

)90

.5(8

8.9e

92.2

)80

.3(7

6.3

e84

.3)

101.

2(9

6.8

e10

5.6)

95.6

(94.

0e

96.0

)91

.1(8

9.7e

92.4

)

1st

day

2h

aft

er

firs

tdose

84.1

(78.

9e89

.2)

105.

9(1

01.2

e11

0.6)

93.9

(92.

8e94

.9)

90.3

(88.

9e91

.7)

80.9

(76.

2e

85.7

)99

.8(8

5.8

e10

3.8)

94.9

(94.

0e

95.8

)90

.6(8

9.3e

91.9

)

6th

day

befo

rela

stdose

83.1

(78.

1e88

.1)

101.

0(9

6.4e

105.

6)94

.5(9

3.4e

95.5

)89

.9(8

8.3e

91.5

)83

.4(7

9.0

e87

.8)

104.

3(1

00.2

e10

8.3)

95.0

(93.

7e

96.6

)91

.1(8

9.7e

92.6

)

6th

day

aft

er

last

dose

83.9

(79.

8e88

.0)

102.

7(9

7.4e

107.

7)94

.3(9

3.2e

95.4

)91

.0(8

9.6e

92.4

)86

.3(8

2.0

e90

.5)

106.

1(1

01.9

e11

0.3)

94.6

(93.

7e

95.6

)91

.6(9

0.3e

92.8

)

Valu

es

are

mean

and

95%

CI.

Walking test for assessing the efficacy of formoterol 1429

Although bronchodilation had a positive effect on exercisetolerance on the group as a whole, the improvement inendurance during submaximal exercise was only moder-ately correlated with the change in FEV1 and IC inducedby bronchodilation. In fact, ipratropium bromide in ap-proximately 20% of patients, failed to improve endurancetime despite positive effects on lung function. Theseinteresting findings suggest that factors other than impair-ment in lung function must be involved in exercise limita-tion in COPD. The duration of exercise likely contributesto muscle fatigue and it is well known that peripheralmuscle fatigue is a potential contributor to exercise intol-erance in COPD.2 Peripheral muscle fatigue, which wehave not evaluated in this study, might explain why formo-terol was unable to improve dyspnea and walked distanceafter the 12-MWT.

In any case, our findings do not allow to exclude thatthe supposed impact of a learning experience whenrepetitive tests are performed25 has influenced the resultsthat we have obtained when we have evaluated the im-pact of formoterol on 6-MWT, but they also indicate thatthis learning experience is not important when the impactof the same bronchodilator is evaluated by the changes in12-MWT.

We have considered only a 5-day treatment becauseour aim was not that of confirm the usefulness of a regulartreatment with formoterol in improving exercise capacityin COPD patients, but that of directly comparing thechanges in exercise performance detected by the 6-MWTtest with those detected by the 12-MWT test thesepatients. For doing it, we preferred to use a fast butalso long-acting agent but we also considered that otherknown muscle effects of ß2-agonists are those on skeletalmuscle force. ß2-Agonists ‘muscle type’ increase the con-traction force of the fast-contracting muscle type, whichcan influence the walking tests. Unfortunately, there isdocumentation that a marked tolerance develops for ß2-agonist-induced increases in force generation of thefast-contracting muscle type by transmural stimulationex vivo.26 It can be envisaged that such downregulationor uncoupling of skeletal muscle ß2-receptors also occursafter maintenance formoterol treatment and may leadto fewer functional ß2-receptors being available to endog-enous catecholamines released during exercise.

In conclusion, our study strengths the importance ofwalking tests as a useful tool for evaluating the impact ofa bronchodilator on some COPD patient-centred outcomes,but also indicates that 6-MWT seems to be a more appro-priate instrument than 12-MWT for assessing the exerciseresponse to a bronchodilator in COPD. This is an intriguinginformation that deserves further investigation consideringthat recently it has been documented that in patients withsevere airflow limitation (FEV1< 50% predicted), 6-MWT de-clines over time, whereas the rate of FEV1 loss over time islower at the more severe stages of the disease.27

Conflict of interest statement

All Authors have no conflicts to disclose for this paper thatdescribes a spontaneous study, which was not supported byany Drug Company.

1430 M. Cazzola et al.

References

1. Palange P, Ward SA, Carlsen KH, et al. Recommendations onthe use of exercise testing in clinical practice. Eur Respir J2007;29:185e209.

2. Pepin V, Saey D, Whittom F, LeBlanc P, Maltais F. Walkingversus cycling: sensitivity to bronchodilation in chronicobstructive pulmonary disease. Am J Respir Crit Care Med2005;172:1517e22.

3. ATS Committee on Proficiency Standards for Clinical PulmonaryFunction Laboratories. ATS statement: guidelines for the six-minute walk test. Am J Respir Crit Care Med 2002;166:111e7.

4. Gerardi DA, Lovett L, Benoit-Connors ML, Reardon JZ,ZuWallack RL. Variables related to increased mortality followingout-patient pulmonary rehabilitation.EurRespir J 1996;9:431e5.

5. Garcia-AymerichJ,FarreroE, FelezMA, IzquierdoJ,MarradesRM,Anto JM. Risk factors of readmission to hospital for a COPD exac-erbation: a prospective study. Thorax 2003;58:100e5.

6. Paggiaro PL, Dahle R, Bakran I, Frith L, Hollingworth K,Efthimiou J. Multicentre randomised placeboecontrolled trialof inhaled fluticasone propionate in patients with chronicobstructive pulmonary disease. International COPD StudyGroup. Lancet 1998;351:773e80.

7. Pepin V, Brodeur J, Lacasse Y, et al. Six-minute walking versusshuttle walking: responsiveness to bronchodilation in chronicobstructive pulmonary disease. Thorax 2007;62:291e8.

8. Liesker JJ, Wijkstra PJ, Ten Hacken NH, Koeter GH, Postma DS,Kerstjens HA. A systematic review of the effects of bronchodilatorsonexercisecapacity inpatientswithCOPD.Chest2002;121:597e608.

9. Butland RJ, Pang J, Gross ER, Woodcock AA, Geddes DM. Two-,six-, and 12-minute walking tests in respiratory disease. BMJ(Clin Res Ed) 1982;284:1607e8.

10. Bernstein ML, Despars JA, Singh NP, Avalos K, Stansbury DW,Light RW. Reanalysis of the 12-minute walk in patients withchronic obstructive pulmonary disease. Chest 1994;105:163e7.

11. Borg G. Psychophysical bases of perceived exertion. Med SciSports Exerc 1982;14:377e81.

12. ATS statement: guidelines for the six-minute walk test. Am JRespir Crit Care Med 2002;166:111e7.

13. Guyatt GH, Pugsley SO, Sullivan MJ, et al. Effect of encourage-ment on walking test performance. Thorax 1984;39:818e22.

14. American Thoracic Society. Standards for the diagnosis andcare of patients with chronic obstructive pulmonary disease(COPD) and asthma. Am Rev Respir Dis 1987;136:225e44.

15. Bland JM, Altman DG. Statistical methods for assessing agree-ment between two methods of clinical measurement. Lancet1986;1:307e10.

16. Redelmeier DA, Bayoumi AM, Goldstein RS, Guyatt GH. Inter-preting small differences in functional status: the six-minutewalk test in chronic lung disease patients. Am J Respir CritCare Med 1997;155:1278e82.

17. Troosters T, Gosselink R, Decramer M. Short- and long-termeffects of outpatient rehabilitation in patients with chronicobstructive pulmonary disease: a randomized trial. Am J Med2000;109:207e12.

18. Cazzola M, MacNee W, Martinez FJ, et al. Outcomes for COPDpharmacological trials: from lung function to biomarkers. EurRespir J; 2008.

19. Morgan MD, Singh SJ. Assessing the exercise response to a bron-chodilator in COPD: time to get off your bike? Thorax 2007;62:281e3.

20. Neder JA, Fuld JP, Overend T, et al. Effects of formoterol onexercise tolerance in severely disabled patients with COPD.Respir Med 2007;101:2056e64.

21. O’Donnell DE, Webb KA. Exertional breathlessness in patientswith chronic airflow limitation: the role of lung hyperinflation.Am Rev Respir Dis 1993;148:1351e7.

22. Boni E, Corda L, Franchini D, et al. Volume effect and exer-tional dyspnoea after bronchodilator in patients with COPDwith and without expiratory flow limitation at rest. Thorax2002;57:528e32.

23. O’Donnell DE, Voduc N, Fitzpatrick M, Webb KA. Effect ofsalmeterol on the ventilatory response to exercise in chronicobstructive pulmonary disease. Eur Respir J 2004;24:86e94.

24. O’Donnell DE, Lam M, Webb KA. Spirometric correlates ofimprovement in exercise performance after anticholinergictherapy in chronic obstructive pulmonary disease. Am J RespirCrit Care Med 1999;160:542e9.

25. Eastwood PR, Hillman DR, Finucane KE. Ventilatory responsesto inspiratory threshold loading and role of muscle fatigue intask failure. J Appl Physiol 1994;76:185e95.

26. Holmberg E, Jeppsson AB, Waldeck B. Selective developmentof tolerance to beta-adrenoceptor agonists in skeletal muscleas compared with airway smooth muscle from the guinea pig.Clin Exp Pharmacol Physiol 1981;8:49e56.

27. Casanova C, Cote CG, Marin JM, et al. The 6-min walkingdistance: long-term follow up in patients with COPD. EurRespir J 2007;29:535e40.