ARTICLE IN PRESS

Respiratory Medicine (2007) 101, 2454–2463

0954-6111/$ - see frdoi:10.1016/j.rmed.

�Corresponding au

E-mail address: m

Extra-pulmonary features in COPD patientsentering rehabilitation after stratification forMRC dyspnea grade

Martijn A. Spruita,�, Herman-Jan Penningsb, Paul P. Janssena,Joan D. Doesb, Sigrid Scroyenc, Marco A. Akkermansa,Rob Mostertb, Emiel F.M. Woutersd,e

aDepartment of Research, Development & Education, Centre for Integrated Rehabilitation of Organ failure(CIRO), Horn, The NetherlandsbDepartment of Respiratory Medicine, Centre for Integrated Rehabilitation of Organ failure (CIRO), Horn, The NetherlandscDepartment of Physiotherapy, Centre for Integrated Rehabilitation of Organ failure (CIRO), Horn, The NetherlandsdCentre for Integrated Rehabilitation of Organ failure (CIRO), Horn, The NetherlandseDepartment of Respiratory Medicine, University Hospital Maastricht, Maastricht, The Netherlands

Received 31 May 2007; accepted 5 July 2007Available online 31 August 2007

KEYWORDSCOPD;Medical ResearchCouncil;Dyspnea;Disease-specifichealth status;Body composition;Exercise intolerance

ont matter & 20072007.07.003

thor. Tel.: +31 475

artijnspruit@prot

SummaryExperts have stated that referral for rehabilitation of patients with chronic obstructivepulmonary disease (COPD) becomes appropriate when these patients become aware oftheir disability (e.g. usually grade 3 to 5 on the Medical Research Council (MRC) dyspneascale). However, patients with MRC dyspnea grade 1/2 may also suffer from extra-pulmonary features, such as abnormal body composition, exercise intolerance and reduceddisease-specific health status. In the present study, we have studied whether and to whatextent chronic obstructive pulmonary disease (COPD) patients with MRC dyspnea grade 1/2have extra-pulmonary features compared to patients with grade 3, 4 or 5?Pulmonary function, body composition, 6-min walking distance, peak exercise capacity,anxiety, depression and disease-specific health status have been assessed in 333outpatients who had been referred for pulmonary rehabilitation. On average, patientswith MRC dyspnea grade 1/2 had a better exercise tolerance and disease-specific healthstatus compared to patients with grade 4 or 5. Nevertheless, grade 1/2 patients had ahigher prevalence of muscle mass depletion. In addition, these patients did still haveaberrant values in one or more of the aforementioned outcomes.

Elsevier Ltd. All rights reserved.

587 600; fax: +31 475 587 592

eion.nl (M.A. Spruit).

ARTICLE IN PRESS

MRC dyspnea grade in COPD 2455

On average, patients with MRC dyspnea grade 1/2 may clearly suffer from extra-pulmonaryfeatures, indicating the necessity to refer these patients for rehabilitation. Therefore,MRC dyspnea scale alone does not appear to be a suitable measure to identify mostpatients with COPD who have to be referred for rehabilitation.& 2007 Elsevier Ltd. All rights reserved.

Introduction

Patients with moderate to severe chronic obstructivepulmonary disease (COPD) may suffer every day fromdyspnea.1,2 The degree of severity of dyspnea perceptionand its repercussions on daily functioning have beenassessed regularly in patients with moderate to severeCOPD and have been shown to be worse compared to healthyage-matched control subjects.3,4

The Medical Research Council (MRC) scale is asimple measure of dyspnea,4 which has shown to bemore discriminating than staging of disease severityof the American Thoracic Society with respect to5-year survival of patients with COPD.5 Furthermore, theMRC dyspnea scale has been suggested in recent interna-tional guidelines to identify COPD patients that maybenefit from rehabilitation (e.g. usually grades 3 to 5).6,7

This suggestion has been based on experts opinion thatrehabilitation of patients with COPD becomes appropri-ate when these patients become aware of their disability,irrespective of their pulmonary dysfunction.7 In fact, COPDpatients entering rehabilitation with MRC dyspnea grade3 or above have been shown to have a clear exerciseintolerance, reduced disease-specific health status,reduced mood status and reduced self-reported dailyphysical activity, especially in COPD patients with grade5.4,8 Unfortunately, Bestall and colleagues did not enrollCOPD patients with MRC dyspnea grade 1/2, given thatthese grades correspond to mild disability due todyspnea.4 In spite of this, it is still reasonable to hypothesizethat COPD patients with lower MRC dyspnea grades(grade 1 or 2) can still suffer from extra-pulmonary features(i.e. exercise capacity, body composition and disease-specific health status), which may partly improve followingpulmonary rehabilitation.

For instance, a mean 6-min walking distance (6MWD)of only 71% predicted has been found in COPD patientsentering rehabilitation with MRC dyspnea grade 1/2.9

Moreover, mean body mass index (BMI, body weight (kg)divided by squared height (m)) tended to be significantlylower in COPD patients entering pulmonary rehabilitationwith MRC dyspnea grade 1/2 than those of patients withMRC dyspnea grade 3/4.9 Even though this unexpectedfinding was not significant, it is reminiscent for the possibleexistence of disparities in body composition after stratifica-tion for MRC dyspnea grades. Finally, median scores onthe domain symptoms of the St. George’s RespiratoryQuestionnaire (SGRQ) in male COPD outpatients withMRC dyspnea grade 2 were comparable to those of patientswith dyspnea grade 3 or above.10 Actually, the meantotal SGRQ scores of grade 1/2 COPD patients entering

rehabilitation9 were clearly worse than values obtained inhealthy elderly.11

At present, it remains unknown whether and to whatextent extra-pulmonary features are different after strati-fication for MRC dyspnea grades in a large cohort ofconsecutive COPD patients referred for rehabilitation bytheir chest physician.

Methods

Patients

Pulmonary function, dyspnea, body composition, exercisecapacity, mood status and disease-specific health statushave been assessed prospectively between January 1, 2005to November 1, 2006 in 333 consecutive patients withmoderate to severe COPD. Additionally, self-reported co-existing morbidity was assessed using the Charlson comor-bidity index.12

All patients were admitted to the Centre for IntegratedRehabilitation of Organ failure (CIRO) to undergo a3-day intake/assessment program before the start of acomprehensive interdisciplinary pulmonary rehabilitationprogram.

Previously, the use of long-term oxygen therapy has beenshown to be related to hospital admission rate in COPD,13

which, in turn has been shown to clearly affect disease-specific health status.14 Therefore, patients who were onlong-term oxygen therapy were not included in the presentanalyses. All tests were performed in the clinical routine andwere in accordance with World Medical Association declara-tion of Helsinki.15 All participants were Caucasian and gaveinformed consent. The local Medical Ethical Commissionapproved this observational study.

Assessments

Pulmonary function and arterial blood gasesForced expiratory volume in the 1 s (FEV1) and forced vitalcapacity (FVC) were calculated from the flow–volume curveusing spirometry. FEV1 was also calculated 15min afterinhalation of a b agonist via a metered dose inhaler. Carbonmonoxide transfer factor of the lung (DLCO) was determinedusing the single breath method. Lung function parameterswere expressed as a percentage of reference values.16

Arterial oxygen tension (PaO2), arterial carbon dioxidetension (PaCO2) and arterial oxygen saturation (SaO2) wereanalyzed with a blood gas analyzer.

ARTICLE IN PRESS

M.A. Spruit et al.2456

DyspneaSelf-perceived dyspnea in relation to physical disability hasbeen assessed by using the MRC dyspnea scale.4 COPDpatients had to grade their self-perceived dyspnea by usingpre-defined statements, e.g. (1) ‘‘I only get breathless withstrenuous exercise’’; (2) ‘‘I get short of breath whenhurrying on the level or up a slight hill’’; (3) ‘‘I walk slowerthan people of the same age on the level because ofbreathlessness or have to stop for breath when walking atmy own pace on the level’’; (4) ‘‘I stop for breath afterwalking 100 yards or after a few minutes on the level’’; and(5) ‘‘I am too breathless to leave the house or breathlesswhen dressing or undressing’’.

Body compositionBody composition was assessed by using single-frequencybioelectrical impedance assessment after overnight fast-ing.17 Fat-free mass was calculated by using validated COPD-specific equations.18 A priori, a BMI o21 kg/m2 and a fat-free mass index (FFMI, fat-free mass in kilogram divided bysquared height in meters, f/m) o15/16 kg/m2 have beendefined as abnormally low.17

Exercise capacityPeak cycling load and peak aerobic capacity (VO2) have beenassessed by using a symptom-limited incremental ergometrycycling test19 and were normalized for age, sex and height.20

In addition, functional exercise capacity has been assessedby using a 6MWD test21 and were expressed in percentage(%) predicted.22

Mood statusAnxiety and depression have been assessed using theHospital Anxiety and Depression Scale (HADS). A priori, ascore of X10 points for anxiety and/or depression had beendefined as abnormally high and may be suggestive for thepresence of clinically relevant symptoms of anxiety and/ordepression.23 Scores can range from 0 ( ¼ optimal) to 21points ( ¼ worst).

Disease-specific health statusSGRQ has been used to asses disease-specific health status.3

It consists of three domains (symptoms, activity and impact)and a total score. Scores can range from 0 ( ¼ optimal) to100 points ( ¼ worst).

StatisticsAll statistics were done using GraphPad Prism 4.03 and SPSS14.0. Obtained values have been tested for normality andare presented as mean (standard deviation, SD). Patientswere stratified by self-reported MRC dyspnea grades. Due tothe low number of COPD patients with MRC dyspnea grade 1(n ¼ 9), the authors have decided a posteriori to combinethe results of these patients with those of grade 2.Differences between MRC strata have been assessed byusing a one-way analysis of variances (ANOVA). In addition,Fisher’s least significant differences (LSD) test has been usedas post-hoc test because of an unequal group size.Differences in the prevalence of low BMI or low FFMIbetween MRC dyspnea grades (1/2 versus 3, 4, 5, respec-tively; 3 versus 4 and 5; and 4 versus 5) have been assessed

by using Chi square test. A priori, a two-sided level ofsignificance was set at pp0.05.24

Results

General characteristics

Patients generally had mild to severe COPD (FEV1:45.9715.0% predicted; FEV1/FVC: 44.1712.9%), normalarterial blood gases (PaO2: 9.371.2 kPa; PaCO2:5.370.7 kPa), normal body composition (BMI:25.175.0 kg/m2; FFMI (f/m): 15.472.0/17.172.1 kg/m2),functional exercise intolerance (6MWD: 69.7716.3% pre-dicted), peak exercise intolerance (VO2: 64.5724.5% pre-dicted; peak cycling load: 56.7722.5% predicted),normal mood status (anxiety: 6.974.7 points;depression: 6.974.1 points) and reduced disease-specifichealth status (total SGRQ score: 55.0717.0 points). Inaddition, patients had a mean MRC dyspnea gradeof 3.671.2.

Stratification by MRC dyspnea grades

CharacteristicsMean age and scores on the Charlson comorbidity indexwere not significantly different between MRC dyspneagrades (Table 1). Patients with MRC dyspnea grade 1/2had a significantly better FEV1 than patients with grade3, 4 or 5. In addition, patients with MRC dyspnea grade3 had a significantly better FEV1 than patients grade 4 or 5.In contrast, the degree of severity of airway obstructionwas comparable between MRC strata (Table 1). Staticlung hyperinflation was comparable between patientswith MRC dyspnea grade 1/2 or grade 3, but was signifi-cantly lower compared to patients with grade 4 or 5. DLCOwas comparable between patients with MRC dyspneagrade 1/2 or grade 3, but was significantly bettercompared to patients with grade 4 or 5. Differences inDLCO disappeared after correction for alveolar surface(Table 1).

Body compositionPatients with MRC dyspnea grade 4 had a high mean BMI,which was significantly higher than values obtained inpatients with MRC dyspnea grade 1/2 or grade 3 (Table 2).Additionally, 27.6% of the patients with MRC dyspnea grade1/2 had an abnormal low BMI, which tended to be lessprevalent in patients with MRC dyspnea grade 4 (14.8%,p ¼ 0.086).

On average, FFMI was comparable between the MRCstrata in female and male patients (Table 2). Then again, theprevalence of an abnormal low FFMI was significantly higherin the grade 1/2 patients (48.3%) than in grade 4 patients(27.9%, p ¼ 0.022).

Functional exercise capacityFunctional exercise capacity was clearly reduced in all MRCstrata (Table 2). Nevertheless, patients with MRC dyspneagrade 1/2 had a significantly better absolute 6MWD thanpatients with grade 3, 4 or 5. In addition, grade 3 patients

ARTICLE IN PRESS

Table 1 General characteristics.

Grade 1/2 Grade 3 Grade 4 Grade 5 ANOVA

Male/female 33/25 69/41 33/28 65/39 –

Age, years 62.1 (9.7) 63.9 (9.3) 64.8 (9.7) 64.2 (9.2) 0.4240Charlson index, points 1.45 (0.80) 1.46 (0.80) 1.59 (0.90) 1.61 (0.99) 0.5035

Pulmonary functionFEV1, % predicted 55.2 (12.7) 48.4 (14.8)* 42.0 (13.8)*y 39.8 (14.0)*y o0.0001FEV1/IVC, % 47.4 (11.1) 44.1 (11.1) 44.2 (14.4) 42.0 (14.5) 0.1031RV/TLC ratio 0.53 (0.09) 0.54 (0.09) 0.60 (0.10)*y 0.61 (0.09)*y o0.0001DLCO, % predicted 57.2 (18.6) 54.1 (16.7) 47.5 (14.0)*z 48.1 (17.3)*z 0.0020DLCO/VA, % pred 64.9 (23.2) 64.4 (21.7) 60.5 (20.7) 62.8 (25.4) 0.7353

Arterial blood gasesPaO2, kPa 9.5 (1.2) 9.4 (1.2) 9.2 (1.1) 9.2 (1.3) 0.3452PaCO2, kPa 5.4 (0.6)z 5.2 (0.6) 5.4 (0.7)z 5.4 (0.7)z 0.0459SaO2, % 93.9 (2.5) 93.8 (2.2) 93.4 (2.5) 93.5 (2.8) 0.6540

Values expressed as mean (SD). Forced expiratory volume in the 1 s (FEV1), inspiratory vital capacity (IVC), residual volume (RV), totallung capacity (TLC), carbon monoxide transfer factor (DLCO), carbon monoxide transfer factor after correction for alveolar surface(DLCO/VA), resting arterial oxygen tension (PaO2), resting arterial carbon dioxide tension (PaCO2), resting arterial oxygen saturation(SaO2).Post-hoc: *pp0.01 Versus grade 1/2.ypp0.01 versus grade 3;zpp0.05 versus grade 3.

MRC dyspnea grade in COPD 2457

had a better 6MWD than patients with MRC dyspnea grade 4or 5.

After correction for gender, age, height and bodyweight,22 significant differences were found betweenpatients with grade 1/2 and grade 4 or 5; between patientswith grade 3 and grade 4 or 5; and between patients withgrade 4 and grade 5 (Figure 1).

Changes in transcutaneous oxygen saturation followingthe 6MWD test were comparable between the MRC strata, aswell as peak heart rate and symptom Borg score for fatigueat the end of 6MWD test (Table 2). In contrast, symptomBorg score for dyspnea at the end of 6MWD test wassignificantly lower in patients with grade 1/2 compared topatients with grade 4 or 5. In addition, patients with MRCdyspnea grade 3 perceived less breathlessness at the end ofthe 6WMD test than patients with grade 5.

Peak exercise capacityAbsolute peak cycling load was significantly differentbetween the MRC strata, except for the difference betweenpatients with grades 4 and 5. These differences remainedsignificant after correction for gender, age and height(Table 2).

Patients with MRC dyspnea grade 1/2 had a significantlybetter peak VO2 per kilogram body weight than patientswith grade 3, 4 or 5. Additionally, patients with MRC dyspneagrade 3 had a significantly better peak VO2 per kilogrambody weight than patients with grade 4 or 5. Conversely,after expressing peak oxygen uptake as a percentage of thereference values of Jones and colleagues,20 there were onlysignificant differences between patients with MRC dyspneagrade 1/2 and 5; and grades 3 and 5.

Peak heart rate was significantly higher in patients withMRC dyspnea grade 1/2 compared to grade 3, 4 or 5. Inaddition, patients with MRC dyspnea grade 3 had signifi-cantly higher mean peak heart rate compared to patientswith grade 4 or 5.

Patients with MRC dyspnea grade 3 had significantlyhigher absolute minute ventilation than patients with grade4 or 5. In addition, patients with MRC dyspnea grade 1/2 hadsignificantly higher absolute minute ventilation than pa-tients with grade 4 or 5. In contrast, patients with MRCdyspnea grade 4 or 5 had relatively higher minute ventila-tion than grade 1/2 patients. Symptom Borg scores fordyspnea and fatigue and changes in transcutaneousSaO2 were not significantly different between MRC strata(Table 2).

Mood statusSignificant differences in HADS anxiety score were foundbetween patients with MRC dyspnea grades 3 and 5, andbetween grades 4 and 5 (Table 3). Patients with MRCdyspnea grades 1/2 or 3 had lower ( ¼ better) HADSdepression scores than grade 5 patients. Additionally,40.7% and 43.5% of the grade 5 patients had an abnormalhigh score for anxiety and depression, respectively. Theprevalence of increased HADS scores for anxiety anddepression was lower in patients with grade 1/2 (27.8%,p ¼ 0.006 and 20.8%, p ¼ 0.118, respectively) or grade 3(19.6%, p ¼ 0.001 and 23.1%, p ¼ 0.001, respectively).

Disease-specific health statusSignificant differences in the three SGRQ domains betweenthe different MRC strata have resulted in significant

ARTICLE IN PRESS

Table 2 Outcomes of physical fitness.

Grades 1/2 Grade 3 Grade 4 Grade 5 ANOVA

Body compositionBMI, kg/m2 24.1 (4.5) 24.6 (4.1) 26.3 (6.1)#,z 25.5 (5.2) 0.0441FFMI (f), kg/m2 14.8 (1.4) 15.2 (1.7) 16.0 (2.7) 15.5 (2.0) 0.1557FFM (m), kg/m2 16.8 (2.0) 17.0 (1.8) 17.7 (2.6) 17.1 (2.2) 0.3104

6MWDMeters 525.0 (92.5) 492.7 (87.9)# 412.4 (95.8)*y 383.2 (104.3)*y o0.0001% predicted 80.2 (14.2) 75.9 (13.4) 65.7 (13.6)*y 59.8 (15.4)*y,zz o0.0001DtSaO2, % �3.4 (4.2) �4.8 (4.6) �4.7 (4.6) �3.7 (4.2) 0.1027Peak HR, bpm 107.9 (17.6) 109.9 (16.4) 104.7 (16.4) 105.1 (17.7) 0.1254Dyspnea end 3.4 (1.8) 4.1 (2.1) 4.7 (2.3)* 4.9 (2.2)*y 0.0002Fatigue end 3.4 (2.2) 3.7 (2.3) 3.9 (2.5) 4.0 (2.4) 0.4800

Cycling testPeak load, watts 94.3 (29.4) 81.0 (30.2)* 62.5 (34.1)*y 59.4 (26.0)*y o0.0001Peak load, % pred 73.2 (21.5) 62.8 (21.2)* 46.7 (18.6)*y 46.8 (18.7)*y o0.0001Peak VO2, ml/min/kg 18.2 (4.4) 16.8 (4.1)# 14.8 (4.5)*y 13.6 (3.2)*y o0.0001Peak VO2, % pred 71.0 (24.0) 66.4 (24.4) 64.9 (25.6) 58.4 (23.4)#,z 0.0135DtSaO2, % �3.5 (3.3) �3.8 (3.3) �3.1 (3.4) �2.7 (3.0) 0.1019HR, bpm 138.8 (21.0) 130.4 (23.4)# 120.5 (18.4)*z 122.9 (18.9)*z o0.0001Peak HR, % max HR 87.0 (11.9) 82.3 (12.2)# 77.6 (10.9)*y 78.9 (11.6)*y o0.0001Peak VE, l/min 47.1 (13.2) 45.1 (16.0) 39.7 (14.5)*z 38.7 (11.8)*y o0.0001Peak VE, % MVV 80.8 (17.7) 87.7 (23.8) 92.1 (26.3)# 93.9 (26.4)* 0.0092Dyspnea end 6.3 (2.2) 6.7 (2.1) 6.5 (1.6) 7.1 (1.6) 0.0506Fatigue end 5.9 (2.3) 5.1 (2.4) 4.8 (2.2) 5.2 (2.5) 0.0871

Values expressed as mean (SD). Body mass index (BMI), kilogram (kg), squared meter (m2), fat-free mass index (FFMI), female (f), male(m), oxygen uptake (VO2), transcutaneous oxygen saturation (tSaO2), heart rate (HR), beats per minute (bpm), ventilation (VE),maximal voluntary ventilation (MVV), liters (l), minute.25

Post-hoc: *pp0.01 versus grade 1/2;#pp0.05 versus grade 1/2;ypp0.01 versus grade 3;zpp0.05 versus grade 3;zzpp0.01 versus grade 4.

1/2 3 4 5

0

25

50

75

100p ≤ 0.01

MRC dyspnea grade

6M

WD

(%

pre

dic

ted)

Figure 1 Functional exercise intolerance.

M.A. Spruit et al.2458

differences in SGRQ total scores between all MRC strata(Table 3). Thus, patients with a lower MRC dyspnea gradehad significantly lower ( ¼ better) SGRQ total scores thanpatients with higher MRC dyspnea grades (Figure 2).

Discussion

The present study demonstrates significant differences inextra-pulmonary features in a large cohort of patients withCOPD entering pulmonary rehabilitation after stratificationfor MRC dyspnea grades.

Body composition

Previously, mean BMI tended to be significantly lower inCOPD patients entering rehabilitation with MRC dyspneagrade 1/2 (26.3 kg/m2) than those of patients with MRCdyspnea grade 3/4 (29.2 kg/m2).9 In the present study, grade4 patients also had a high mean BMI, which was significantlyhigher than the normal mean BMI of patients withMRC dyspnea grade 1/2 or 3 (Table 2). Indeed, a highermean BMI may explain to some extent a higher dyspneaperception in daily life.26,27 Then again, underweightCOPD patients (mean BMI: 18.7 kg/m2) have been shownto have a significantly worse mean modified MRC dyspneagrade than normal weight peers (mean BMI: 24.5 kg/m2).28

ARTICLE IN PRESS

Table 3 Mood and health status.

Grades 1/2 Grade 5 ANOVA

HADS (n ¼ 299)Anxiety, points 6.9 (4.9) 6.0 (4.1) 6.6 (5.2) 8.3 (4.7)y,� 0.0064Depression, points 6.3 (4.0) 6.2 (3.8) 7.0 (4.3) 8.1 (4.3)#,y 0.0098

SGRQ (n ¼ 290)Symptoms, points 54.6 (20.9) 61.7 (18.5)# 66.5 (21.7)* 71.4 (15.6)*y o0.0001Activity, points 50.5 (18.1) 64.8 (17.3)* 74.3 (16.5)*y 82.6 (10.4)*y,zz o0.0001Impact, points 31.0 (17.9) 40.5 (18.4)* 47.0 (16.7)*z 54.7 (15.6)*y,zz o0.0001Total score, points 40.8 (16.6) 51.3 (15.9)* 58.5 (14.8)*y 65.9 (11.2)*y,zz o0.0001

Values expressed as mean (SD). Hospital Anxiety and Depression Scale (HADS), St. George’s Respiratory Questionnaire (SGRQ).Post-hoc: *pp0.01 versus grade 1/2;ypp0.01 versus grade 3;#pp0.05 versus grade 1/2;zpp0.05 versus grade 3;zzpp0.01 versus grade 4;�pp0.05 versus grade 4.

1/2 3 4 5

0

10

20

30

40

50

60

70

80

p ≤ 0.004

MRC dyspnea grade

Tota

l S

GR

Q s

core

(poin

ts)

Figure 2 Disease-specific health status.

MRC dyspnea grade in COPD 2459

The present authors were not able to corroborate thesefindings.

Even though the mean FFMI was similar amongst the MRCstrata after stratification for gender, 48.3% of the COPDpatients with MRC dyspnea grade 1/2 had an abnormal lowFFMI, which may partly be related to daily physical inactivityin patients with COPD.29 This suggests the possible existenceof a reduced daily physical activity level in a subgroup ofCOPD patients with MRC dyspnea grade 1/2. Indeed,patients may modify their daily behavior (e.g. less weight-bearing activities and/or a lower effort patients may exertin completing activities of daily life) to avoid unpleasantsensations of dyspnea.30 In fact, an a posteriori analysis hasbeen shown that grade 1/2 patients with a normal FFMItended to have a higher FEV1 than grade 1/2 patients withan abnormal FFMI (58.3 versus 52.3% predicted, p ¼ 0.08).Then again, daily physical activity level has been shownto be reduced in patients with COPD compared to healthyage-matched subjects, irrespective of the degree ofpulmonary function impairment.29 Therefore, other under-

lying factors may to some degree have contributed to FFMdepletion in a subgroup of COPD patients with MRC dyspneagrade 1/2.31–33

Exercise intolerance

In the present study, patients with MRC dyspnea grades 1/2had the best 6MWD compared to patients with a higher MRCdyspnea grade (Table 2, Figure 1). These findings are in linewith previous findings.9,34 These data suggest that func-tional exercise capacity is related to the degree of severityof dyspnea perceived during daily life. In fact, patients withMRC dyspnea grade 4 or 5 had also the worst symptom Borgscores for dyspnea at the end of the 6MWD. Furthermore,the mean difference in 6MWD between patients with MRCdyspnea grade 1/2 and grade 4 (112.7m) or grade 5(141.9m) clearly exceeded the minimum clinically impor-tant difference of 54m.35

COPD patients with MRC dyspnea grade 5 had theworst peak exercise capacity of all MRC strata, as seenpreviously.9 Nevertheless, a decreased mean peak exercisecapacity has been found in all MRC strata (Table 2). Indeed,mean peak oxygen uptake of all MRC strata was clearlybelow the lower limit of normal as proposed in the lateststatement on cardiopulmonary exercise testing by theAmerican Thoracic Society and the American College ofChest Physicians (lower limit of normal for peak VO2: 85%predicted).36

A clear difference has been observed between MRC strataconcerning peak heart rate and peak minute ventilation atthe end of the symptom-limited incremental exercise test(Table 2). In fact, patients with MRC dyspnea grade 1/2generally have been shown to have a cardio-circulatorylimitation, which makes them most probably good candi-dates for an endurance-type of exercise training.6,19,37 Incontrast, patients with grade 4 or 5 were generallyventilatory limited, which makes them most probably good

ARTICLE IN PRESS

M.A. Spruit et al.2460

candidates for an interval-type of training to improveexercise tolerance.38 The latter patients may even improvetheir exercise tolerance by performing progressive resis-tance training39 or transcutaneous neuromuscular electricalstimulation.40

Mood status

Recently, the mean scores on the Brief Assessment ScheduleDepression Cards were not different between COPD patientsentering rehabilitation with MRC dyspnea grade 1/2, 3/4 or5.9 On the contrary, higher HADS scores for anxiety anddepression have been reported in COPD patients enteringpulmonary rehabilitation with MRC dyspnea grade 5 com-pared to lower grades.4 The latter finding has beencorroborated in the present study (Table 3). It seemstherefore reasonable to hypothesize that the presence ofanxiety and/or depression is, at least in part, related to theperceived dyspnea during daily life in COPD. In fact, patientswith severe COPD had a 2.5 times greater risk for depressionthan healthy control subjects.41 Moreover, dyspnea hasshown to be significantly related to depression in elderlypatients with recent hospitalization for acutely decompen-sated heart failure.42

Disease-specific health status

Previously, male COPD outpatients with MRC dyspnea grade2 have shown to have lower median total SGRQ scorescompared to patients with severe dyspnea (e.g. combinationof grades 3, 4 and 5).10 The present study is the first toobserve significant differences in SGRQ total scores betweenall MRC strata in a large group of consecutive patients withCOPD (Table 3, Figure 2). In fact, mean differences in totalSGRQ scores between patients with MRC dyspnoea grade 1/2and patients with grade 3, 4 or 5 were approximately a threefold, four fold and six fold of the minimum clinicallyimportant difference of 4 points, respectively.43 It istherefore reasonable to conclude that dyspnea is astrong determinant of disease-specific health status.44

However, patients who do not experience extreme dyspneaduring activities of daily life (e.g. MRC dyspnea grade 1/2)appear to still have an increased ( ¼ worse) mean totalSGRQ score compared to values obtained in healthyelderly.11,45 This is in line with Garrod and colleagues.9

These findings point strongly toward the fact that COPDpatients with MRC dyspnea grade 1/2 may still have areduced disease-specific health status. Therefore, otherfactors besides dyspnea may, at least in part, also contributeto a decreased disease-specific health status in patients withCOPD.14,46–52

Pulmonary rehabilitation

In the recent past, the MRC dyspnea scale has beensuggested and used to select COPD patients who maybenefit from rehabilitation.6 This suggestion has beenbased on experts opinion that rehabilitation of patientswith COPD becomes appropriate when these patientsbecome aware of their disability, irrespective of theirpulmonary dysfunction.7 Actually, COPD patients with MRC

grade 1 or 2 have been excluded from a rehabilitationprogram as being ‘too fit’.8 However, the present resultsclearly indicate that COPD patients referred for rehabilita-tion with MRC grade 1/2 may have extra-pulmonaryfeatures, which most probably can be addressed during acomprehensive interdisciplinary pulmonary rehabilitationprogram.2,6,37,39,53,54 Indeed, Garrod and colleagues haveshown significant mean improvements in 6MWD (+55m)and total SGRQ score (�7.5 points) in COPD patients withbaseline MRC dyspnea grade 1/2 following 7 weeks of twice-weekly outpatient exercise training and patient education.9

The latter improvements were comparable to those ofCOPD patients with baseline MRC dyspea grade 3/4 and weresignificantly better than those of COPD patients withbaseline MRC dyspnea grade 5.9 In addition, COPDpatients with somewhat ventilatory reserve during a base-line symptom-limited incremental cycling test have beenshown to be the best responders during a 6-monthexercise training program.55 In the present study, this wouldbe the grade 1/2 patients (Table 2). Finally, healthyelderly may also benefit from physical exercise training,while they most probably will have a MRC dyspnea grade 1 atbaseline.56 Therefore, it appears reasonable to hypothesizethat the MRC dyspnea scale alone is not an ideal measureto identify COPD patients who may benefit from rehabilita-tion. Based on the present results, it may be reasonableto add the bioelectrical impedance assessment. Indeed,the MRC dyspnea scale (e.g. grade 3 or above) togetherwith the overnight fasting bioelectrical impedanceassessment (e.g. an abnormal low FFMI) identifiedalready 91% of the present patients entering pulmonaryrehabilitation.

Limitations of the present studyPreferably, healthy controls should have been included inthe present study to demonstrate the presence of extra-pulmonary features in COPD patients, in particular in thosewith MRC dyspnea grade 1/2. Nevertheless, to determinethe presence of extra-pulmonary features in the presentCOPD patients have been based on internationally well-established cut-off values obtained in healthy age-matchedcontrol subjects.

The presence of whole-body muscle mass depletionmay have been an important reason to refer COPDpatients with MRC dyspnea grade 1/2 for rehabilitation.This, in turn, may have resulted in a selection bias.However, the prevalence of cachexia in the presentCOPD patients with MRC dyspnea grade 1/2 is in agreementwith findings in Dutch COPD outpatients with MRCdyspnea grade 1/2 who volunteered to participate in theCOSMIC trial.57

Statistically significant differences between self-reportedMRC dyspnea grades need to be interpreted in the light ofthe number of comparisons that were made in the presentstudy. Nonetheless, the current statistical procedures havebeen used to test pre-defined hypotheses.

Finally, the present authors want to stress that they havebeen using the original MRC dyspnea scale as donebefore,4,5,8–10,40 while others have been using variousmodifications of the MRC scale.28,58–63 Although thedifferent versions of the MRC dyspnea scales are rather

ARTICLE IN PRESS

MRC dyspnea grade in COPD 2461

comparable to the original MRC dyspnea scale, theirscoring is slightly different and may therefore be somewhatmisleading.

Conclusions

At present, the MRC dyspnea scale does provide anindication of the degree of severity concerningextra-pulmonary features in patients with moderate tosevere COPD entering rehabilitation. Nevertheless, theMRC dyspnea scale is inadequate to identify extra-pulmonary features in the clinical practice in individualpatients with COPD. Therefore, MRC dyspnea scalealone does not appear to be a suitable measure toidentify most patients with COPD who have to bereferred for rehabilitation. Future studies are warrantedto determine whether chest physicians can use theMRC dyspnea scale and the overnight fasting bioelectri-cal impedance assessment together to identify COPDpatients at the outpatient consultation for pulmonaryrehabilitation.

Acknowledgments

The present authors are grateful to the colleagues ofCIRO Horn who have contributed to the collectionof the present data. In addition, they are grateful toRobin van Hout, Raoul Widdershoven and Tim van Til fortheir support with the development of an electronicdatabase.

Conflict of interest statementThe authors do not have any conflict of interest with the

contents of the present manuscript.

References

1. Oga T, Nishimura K, Tsukino M, Sato S, Hajiro T,Mishima M. Longitudinal deteriorations in patient reportedoutcomes in patients with COPD. Respir Med 2006;101(1):146–53.

2. Wouters EF. Management of severe COPD. Lancet2004;364(9437):883–95.

3. Jones PW, Quirk FH, Baveystock CM, Littlejohns P. A self-complete measure of health status for chronic airflow limita-tion. The St. George’s Respiratory Questionnaire. Am Rev RespirDis 1992;145(6):1321–7.

4. Bestall JC, Paul EA, Garrod R, Garnham R, Jones PW, WedzichaJA. Usefulness of the Medical Research Council (MRC)dyspnoea scale as a measure of disability in patients withchronic obstructive pulmonary disease. Thorax 1999;54(7):581–6.

5. Nishimura K, Izumi T, Tsukino M, Oga T. Dyspnea is a betterpredictor of 5-year survival than airway obstruction in patientswith COPD. Chest 2002;121(5):1434–40.

6. Nici L, Donner C, Wouters E, Zuwallack R, Ambrosino N,Bourbeau J, et al. American thoracic society/european respira-tory society statement on pulmonary rehabilitation. Am J RespirCrit Care Med 2006;173(12):1390–413.

7. Pulmonary rehabilitation. Thorax 2001;56(11):827–34.8. Wedzicha JA, Bestall JC, Garrod R, Garnham R, Paul EA, Jones

PW. Randomized controlled trial of pulmonary rehabilitation in

severe chronic obstructive pulmonary disease patients, strati-fied with the MRC dyspnoea scale. Eur Respir J 1998;12(2):363–9.

9. Garrod R, Marshall J, Barley E, Jones PW. Predictors of successand failure in pulmonary rehabilitation. Eur Respir J 2006;27(4):788–94.

10. Hajiro T, Nishimura K, Tsukino M, Ikeda A, Oga T, Izumi T. Acomparison of the level of dyspnea vs disease severity inindicating the health-related quality of life of patients withCOPD. Chest 1999;116(6):1632–7.

11. Spencer S, Calverley PM, Sherwood Burge P, Jones PW. Healthstatus deterioration in patients with chronic obstructivepulmonary disease. Am J Respir Crit Care Med 2001;163(1):122–8.

12. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method ofclassifying prognostic comorbidity in longitudinal studies:development and validation. J Chronic Dis 1987;40(5):373–83.

13. Miravitlles M, Calle M, Alvarez-Gutierrez F, Gobartt E, Lopez F,Martin A. Exacerbations, hospital admissions and impairedhealth status in chronic obstructive pulmonary disease. QualLife Res 2006;15(3):471–80.

14. Miravitlles M, Ferrer M, Pont A, Zalacain R, Alvarez-Sala JL,Masa F, et al. Effect of exacerbations on quality of life inpatients with chronic obstructive pulmonary disease: a 2 yearfollow up study. Thorax 2004;59(5):387–95.

15. World Medical Association declaration of Helsinki. Recommen-dations guiding physicians in biomedical researchinvolving human subjects. J Am Med Assoc 1997;277(11):925–6.

16. Quanjer PH, Tammeling GJ, Cotes JE, Pedersen OF, Peslin R,Yernault JC. Lung volumes and forced ventilatory flows. Reportworking party standardization of lung function tests, Europeancommunity for steel and coal. Official statement of theEuropean respiratory society. Eur Respir J 1993;16(Suppl):5–40.

17. Schols AM, Broekhuizen R, Weling-Scheepers CA, Wouters EF.Body composition and mortality in chronic obstructive pulmon-ary disease. Am J Clin Nutr 2005;82(1):53–9.

18. Schols AM, Wouters EF, Soeters PB, Westerterp KR. Bodycomposition by bioelectrical-impedance analysis compared withdeuterium dilution and skinfold anthropometry in patients withchronic obstructive pulmonary disease. Am J Clin Nutr1991;53(2):421–4.

19. Franssen FM, Broekhuizen R, Janssen PP, Wouters EF, Schols AM.Effects of whole-body exercise training on body compositionand functional capacity in normal-weight patients with COPD.Chest 2004;125(6):2021–8.

20. Jones NL, Makrides L, Hitchcock C, Chypchar T, McCartney N.Normal standards for an incremental progressivecycle ergometer test. Am Rev Respir Dis 1985;131(5):700–8.

21. ATS statement: guidelines for the six-minute walk test. Am JRespir Crit Care Med 2002;166(1):111–7.

22. Troosters T, Gosselink R, Decramer M. Six minute walkingdistance in healthy elderly subjects. Eur Respir J 1999;14(2):270–4.

23. Zigmond AS, Snaith RP. The hospital anxiety and depressionscale. Acta Psychiatr Scand 1983;67(6):361–70.

24. Altman DG, Gore SM, Gardner MJ, Pocock SJ. Statisticalguidelines for contributors to medical journals. Br Med J (ClinRes Ed) 1983;286(6376):1489–93.

25. Redelmeier DA, Goldstein RS, Min ST, Hyland RH. Spirometryand dyspnea in patients with COPD. When small differencesmean little. Chest 1996;109(5):1163–8.

26. Berner YN. The contribution of obesity to dyspnoea in elderlypeople. Age Ageing 2001;30(6):530.

ARTICLE IN PRESS

M.A. Spruit et al.2462

27. Ho SF, O’Mahony MS, Steward JA, Breay P, Buchalter M, Burr ML.Dyspnoea and quality of life in older people at home. AgeAgeing 2001;30(2):155–9.

28. Sahebjami H, Sathianpitayakul E. Influence of body weight onthe severity of dyspnea in chronic obstructive pulmonarydisease. Am J Respir Crit Care Med 2000;161(3 Pt 1):886–90.

29. Pitta F, Troosters T, Spruit MA, Probst VS, Decramer M, GosselinkR. Characteristics of physical activities in daily life in chronicobstructive pulmonary disease. Am J Respir Crit Care Med2005;171(9):972–7.

30. Jones P, Lareau S, Mahler DA. Measuring the effects of COPD onthe patient. Respir Med 2005;99(Suppl B):S11–8.

31. Engelen MP, Schols AM, Lamers RJ, Wouters EF. Differentpatterns of chronic tissue wasting among patients withchronic obstructive pulmonary disease. Clin Nutr 1999;18(5):275–80.

32. Spruit MA, Gosselink R, Troosters T, Kasran A, Gayan-Ramirez G,Bogaerts P, et al. Muscle force during an acute exacerbation inhospitalised patients with COPD and its relationship with CXCL8and IGF-I. Thorax 2003;58(9):752–6.

33. Van Vliet M, Spruit MA, Verleden G, Kasran A, Van Herck E,Pitta F, et al. Hypogonadism, quadriceps weakness, andexercise intolerance in chronic obstructive pulmonarydisease. Am J Respir Crit Care Med 2005;172(9):1105–11.

34. McGavin CR, Artvinli M, Naoe H, McHardy GJ. Dyspnoea,disability, and distance walked: comparison of estimates ofexercise performance in respiratory disease. Br Med J1978;2(6132):241–3.

35. Redelmeier DA, Bayoumi AM, Goldstein RS, Guyatt GH.Interpreting small differences in functional status: the SixMinute Walk test in chronic lung disease patients. Am J RespirCrit Care Med 1997;155(4):1278–82.

36. ATS/ACCP Statement on cardiopulmonary exercise testing. Am JRespir Crit Care Med 2003;167(2):211–77.

37. Spruit MA, Troosters T, Trappenburg JC, Decramer M, GosselinkR. Exercise training during rehabilitation of patients withCOPD: a current perspective. Patient Educ Couns 2004;52(3):243–8.

38. Sabapathy S, Kingsley RA, Schneider DA, Adams L, Morris NR.Continuous and intermittent exercise responses in individualswith chronic obstructive pulmonary disease. Thorax2004;59(12):1026–31.

39. Spruit MA, Gosselink R, Troosters T, De Paepe K, Decramer M.Resistance versus endurance training in patients with COPDand peripheral muscle weakness. Eur Respir J 2002;19(6):1072–8.

40. Neder JA, Sword D, Ward SA, Mackay E, Cochrane LM, Clark CJ.Home based neuromuscular electrical stimulation as a newrehabilitative strategy for severely disabled patients withchronic obstructive pulmonary disease (COPD). Thorax2002;57(4):333–7.

41. van Manen JG, Bindels PJ, Dekker FW, CJ IJ, van der Zee JS,Schade E. Risk of depression in patients with chronic obstructivepulmonary disease and its determinants. Thorax2002;57(5):412–6.

42. Ramasamy R, Hildebrandt T, O’Hea E, Patel M, Clemow L,Freudenberger R, et al. Psychological and social factors thatcorrelate with dyspnea in heart failure. Psychosomatics2006;47(5):430–4.

43. Jones PW. Interpreting thresholds for a clinically significantchange in health status in asthma and COPD. Eur Respir J2002;19(3):398–404.

44. Stahl E, Lindberg A, Jansson SA, Ronmark E, Svensson K,Andersson F, et al. Health-related quality of life is relatedto COPD disease severity. Health Qual Life Outcomes 2005;3:56.

45. Ferrer M, Villasante C, Alonso J, Sobradillo V, Gabriel R, VilagutG, et al. Interpretation of quality of life scores from the StGeorge’s Respiratory Questionnaire. Eur Respir J 2002;19(3):405–13.

46. Broekhuizen R, Wouters EF, Creutzberg EC, Schols AM. ElevatedCRP levels mark metabolic and functional impairment inadvanced COPD. Thorax 2005.

47. Snoeck-Stroband JB, Postma DS, Lapperre TS, Gosman MM,Thiadens HA, Kauffman HF, et al. Airway inflammationcontributes to health status in COPD: a cross-sectional study.Respir Res 2006;7(1):140.

48. de Torres JP, Cordoba-Lanus E, Lopez-Aguilar C, Muros deFuentes M, Montejo de Garcini A, Aguirre-Jaime A, et al. C-reactive protein levels and clinically important predictiveoutcomes in stable COPD patients. Eur Respir J2006;27(5):902–7.

49. Mostert R, Goris A, Weling-Scheepers C, Wouters EF, Schols AM.Tissue depletion and health related quality of life in patientswith chronic obstructive pulmonary disease. Respir Med2000;94(9):859–67.

50. Montes de Oca M, Torres SH, Gonzalez Y, Romero E, HernandezN, Mata A, et al. Peripheral muscle composition andhealth status in patients with COPD. Respir Med 2006;100(10):1800–6.

51. Ketelaars CA, Schlosser MA, Mostert R, Huyer Abu-Saad H,Halfens RJ, Wouters EF. Determinants of health-related qualityof life in patients with chronic obstructive pulmonary disease.Thorax 1996;51(1):39–43.

52. Hajiro T, Nishimura K, Tsukino M, Ikeda A, Oga T. Stages ofdisease severity and factors that affect the health status ofpatients with chronic obstructive pulmonary disease. RespirMed 2000;94(9):841–6.

53. Trappenburg JC, Troosters T, Spruit MA, Vandebrouck N,Decramer M, Gosselink R. Psychosocial conditions do not affectshort-term outcome of multidisciplinary rehabilitation inchronic obstructive pulmonary disease. Arch Phys Med Rehabil2005;86(9):1788–92.

54. Spruit MA, Gosselink R, Troosters T, Kasran A, Van Vliet M,Decramer M. Low-grade systemic inflammation and the re-sponse to exercise training in patients with advanced COPD.Chest 2005;128(5):3183–90.

55. Troosters T, Gosselink R, Decramer M. Exercise training in COPD:how to distinguish responders from nonresponders. J Cardio-pulm Rehabil 2001;21(1):10–7.

56. American College of Sports Medicine position stand. Therecommended quantity and quality of exercise fordeveloping and maintaining cardiorespiratory and muscularfitness in healthy adults. Med Sci Sports Exerc 1990;22(2):265–74.

57. Vermeeren MA, Creutzberg EC, Schols AM, Postma DS, PietersWR, Roldaan AC, et al. Prevalence of nutritional depletion in alarge out-patient population of patients with COPD. Respir Med2006.

58. Celli BR, Cote CG, Marin JM, Casanova C, Montes de Oca M,Mendez RA, et al. The body-mass index, airflow obstruction,dyspnea, and exercise capacity index in chronicobstructive pulmonary disease. N Engl J Med 2004;350(10):1005–12.

59. Mahler DA, Wells CK. Evaluation of clinical methods for ratingdyspnea. Chest 1988;93(3):580–6.

60. Appleton S, Adams R, Porter S, Peacock M, Ruffin R. Sustainedimprovements in dyspnea and pulmonary function 3 to 5 yearsafter lung volume reduction surgery. Chest 2003;123(6):1838–46.

61. de Torres JP, Pinto-Plata V, Ingenito E, Bagley P, Gray A, BergerR, et al. Power of outcome measurements to detect clinicallysignificant changes in pulmonary rehabilitation of patients withCOPD. Chest 2002;121(4):1092–8.

ARTICLE IN PRESS

MRC dyspnea grade in COPD 2463

62. Marin JM, Carrizo SJ, Gascon M, Sanchez A, Gallego B, Celli BR.Inspiratory capacity, dynamic hyperinflation, breathlessness,and exercise performance during the 6-minute-walk test inchronic obstructive pulmonary disease. Am J Respir Crit CareMed 2001;163(6):1395–9.

63. Hajiro T, Nishimura K, Tsukino M, Ikeda A, Koyama H,Izumi T. Analysis of clinical methods used to evaluatedyspnea in patients with chronic obstructive pulmonarydisease. Am J Respir Crit Care Med 1998;158(4):1185–9.