Efficacy of a new once-daily long-acting inhaled beta2-agonist indacaterol versus twice-daily formoterol in COPD.

Abstract

Indacaterol is a long-acting inhaled beta(2)-agonist (LABA) for the treatment of chronic obstructive pulmonary disease (COPD). In previous studies, indacaterol provided 24 h bronchodilation on once-daily dosing with a fast onset of action. This study compared the efficacy and safety of indacaterol with the twice-daily LABA formoterol and placebo over 1 year.
Patients with moderate to severe COPD were randomised to receive once-daily indacaterol 300 microg (n=437) or 600 microg (n=428), twice-daily formoterol 12 microg (n=435) or placebo (n=432) for 52 weeks in a double-blind double-dummy parallel group study. The primary efficacy variable was forced expiratory volume in 1 s (FEV(1)) measured 24 h postdose after 12 weeks (indacaterol vs placebo). Other outcomes included dyspnoea (transition dyspnoea index, TDI), use of as-needed salbutamol, symptom-based measures recorded on diary cards, exacerbations, health status (St George's Respiratory Questionnaire), BODE index (body mass index, obstruction, dyspnoea, exercise), safety and tolerability.
Indacaterol increased 24 h postdose FEV(1) after 12 weeks by 170 ml (both doses) versus placebo and by 100 ml versus formoterol (all p<0.001). These significant differences were maintained at 52 weeks. Symptomatic outcomes were improved compared with placebo with all active treatments, and indacaterol was more effective than formoterol in improving TDI score and reducing the need for as-needed salbutamol. Indacaterol was well tolerated and had a good overall safety profile, including minimal impact on QTc interval and systemic beta(2)-mediated events.
Once-daily indacaterol is an effective 24 h bronchodilator that improves symptoms and health status and confers clinical improvements over a twice-daily 12 h LABA as a treatment for patients with moderate to severe COPD.
NCT 00393458.

Full text

Page 1

doi: 10.1136/thx.2009.125435
2010 65: 473-479 Thorax

Ronald Dahl, Kian Fan Chung, Roland Buhl, et al.
formoterol in COPD
-agonist indacaterol versus twice-daily
2
β
Efficacy of a new once-daily long-acting inhaled

http://thorax.bmj.com/content/65/6/473.full.html
Updated information and services can be found at:
These include:
References

http://thorax.bmj.com/content/65/6/473.full.html#related-urls
Article cited in:


http://thorax.bmj.com/content/65/6/473.full.html#ref-list-1
This article cites 21 articles, 12 of which can be accessed free at:
service
Email alerting
box at the top right corner of the online article.
Receive free email alerts when new articles cite this article. Sign up in the
Notes

http://thorax.bmj.com/cgi/reprintform
To order reprints of this article go to:

http://thorax.bmj.com/subscriptions
go to: ThoraxTo subscribe to
group.bmj.com on June 7, 2010 - Published by thorax.bmj.comDownloaded from

Page 2

Efficacy of a new once-daily long-acting inhaled
b2-agonist indacaterol versus twice-daily formoterol
in COPD
Ronald Dahl,1Kian Fan Chung,2Roland Buhl,3Helgo Magnussen,4Vladimir Nonikov,5
Damon Jack,6Patricia Bleasdale,6Roger Owen,6Mark Higgins,6Benjamin Kramer,7
on behalf of the INVOLVE (INdacaterol: Value in COPD: Longer Term Validation of
Efficacy and Safety) Study Investigators
ABSTRACT
Background Indacaterol is a long-acting inhaled b2-
agonist (LABA) for the treatment of chronic obstructive
pulmonary disease (COPD). In previous studies,
indacaterol provided 24 h bronchodilation on once-daily
dosing with a fast onset of action. This study compared
the efficacy and safety of indacaterol with the twice-
daily LABA formoterol and placebo over 1 year.
Methods Patients with moderate to severe COPD were
randomised to receive once-daily indacaterol 300 mg
(n¼437) or 600 mg (n¼428), twice-daily formoterol
12 mg (n¼435) or placebo (n¼432) for 52 weeks in
a double-blind double-dummy parallel group study. The
primary efficacy variable was forced expiratory volume in
1 s (FEV1) measured 24 h postdose after 12 weeks
(indacaterol vs placebo). Other outcomes included
dyspnoea (transition dyspnoea index, TDI), use of as-
needed salbutamol, symptom-based measures recorded
on diary cards, exacerbations, health status (St George’s
Respiratory Questionnaire), BODE index (body mass
index, obstruction, dyspnoea, exercise), safety and
tolerability.
Results Indacaterol increased 24 h postdose FEV1after
12 weeks by 170 ml (both doses) versus placebo and by
100 ml versus formoterol (all p<0.001). These
significant differences were maintained at 52 weeks.
Symptomatic outcomes were improved compared with
placebo with all active treatments, and indacaterol was
more effective than formoterol in improving TDI score
and reducing the need for as-needed salbutamol.
Indacaterol was well tolerated and had a good overall
safety profile, including minimal impact on QTc interval
and systemic b2-mediated events.
Conclusions Once-daily indacaterol is an effective 24 h
bronchodilator that improves symptoms and health
status and confers clinical improvements over a twice-
daily 12 h LABA as a treatment for patients with
moderate to severe COPD.
Trial registration number NCT 00393458.
Current
obstructive pulmonary disease (COPD) provide
a clear treatment algorithm for the maintenance
treatment of COPD, recommending initial treat-
ment with one or more long-acting bronchodilators
for patients with moderate, severe and very severe
disease, with the addition of inhaled corticosteroids
(ICS) for patients who exacerbate frequently.1The
long-actingbronchodilators
managementguidelines forchronic
currentlyavailable
include the twice-daily inhaled b2-agonists formo-
terol and salmeterol. It may be hypothesised that
the sustained bronchodilation provided by a bron-
chodilator with a 24 h duration would reduce
fluctuations in airway patency compared with
twice-daily bronchodilators, and may improve
clinical outcomes.
Indacaterol is a novel inhaled long-acting b2-
agonist providing 24 h bronchodilation on once-
daily dosing.2e4This 1-year study was designed to
provide efficacy and long-term safety information
on indacaterol compared with placebo and formo-
terol. Our primary hypothesis was that indacaterol
would have a greater effect than placebo on forced
expiratory volume in 1 s (FEV1) at 24 h post-dose
(‘trough’) after 12 weeks.
METHODS
Patients
Subjects aged $40 years with a clinical diagnosis of
moderate to severe COPD5and a smoking history of
$20 pack-years, with postbronchodilator (salbu-
tamol 400 mg) FEV1<80% and $30% predicted and
ratio of FEV1to forced vital capacity (FVC) <70%
were enrolled in the study. Factors preventing study
entry included respiratory tract infection or hospi-
talisation for COPD exacerbation within the
previous 6 weeks, oral corticosteroids or change in
ICSduringthepreviousmonth,orhistoryofasthma.
Study design
A 2-week run-in was followed by 52 weeks of
double-blind treatment (see figure S1 in online
supplement), with patients randomised to treat-
ment (1:1:1:1) with stratification for smoking
status (current/ex-smoker) using an automated
interactive system. Data were collected from
outpatient clinics and physicians’ offices.
The study treatments were indacaterol 300 mg
and 600 mg delivered once daily via single-dose dry
powder inhaler (SDDPI), formoterol 12 mg twice
daily (the standard therapeutic dose) via its
proprietary SDDPI and matching placebos to
indacaterol and formoterol taken each morning
(08:00e10:00;bothdevices)
(20:00e22:00; proprietary SDDPI only). Fixed-dose
combinations of ICS plus long-acting b2-agonist
were replaced by monotherapy ICS at an equivalent
dose and regimen plus salbutamol as-needed.
Patients on ICS monotherapy continued treatment
andevening
<Supplementary figures/tables
are published online only. To
view these files please visit the
journal online (http://thorax.bmj.
com).
1Department of Respiratory
Diseases, Aarhus University
Hospital, Aarhus, Denmark
2Airway Disease, National Heart
and Lung Institute, Imperial
College and Royal Brompton
Hospital, London, UK
3Pulmonary Department, Mainz
University Hospital, Mainz,
Germany
4Pulmonary Research Institute
at Hospital Grosshansdorf,
Center for Pneumology and
Thoracic Surgery,
Grosshansdorf, Germany
5Central Clinical Hospital,
Moscow, Russia
6Novartis Horsham Research
Centre, Horsham, West Sussex,
UK
7Respiratory Development,
Novartis Pharmaceuticals
Corporation, East Hanover, New
Jersey, USA
Correspondence to
Professor R Dahl, Department of
Respiratory Diseases, Aarhus
University Hospital, Aarhus,
Denmark; ronadahl@rm.dk
Received 19 August 2009
Accepted 23 March 2010
Thorax 2010;65:473e479. doi:10.1136/thx.2009.125435473
Chronic obstructive pulmonary disease
group.bmj.com on June 7, 2010 - Published by thorax.bmj.comDownloaded from

Page 3

at a stable dose throughout the study. Patients were given
salbutamol to use as-needed but were asked not to take it within
6 h before each visit. Other bronchodilators or corticosteroids
were not allowed unless to treat a COPD exacerbation. Partici-
pating physicians were responsible for treating exacerbations.
Assessments
Patients visited the clinics on days 1, 2, 15, 29, 84, 85, 113, 168,
197, 253, 364 and 365. The visits on days 1 and 2, week 12 and
week 52 were on consecutive days in order to provide a precise
trough FEV1 value which was based on the mean of two
measurements taken 23 h 10 min and 23 h 45 min following the
previous day’s morning dose. At other visits the 15 min predose
value was used as a ‘trough’ measurement. In a subset of
patients, spirometry was performed serially over 24 h postdose.
Patients were given a diary to record peak expiratory flow (PEF),
symptoms, use of salbutamol, any change in concomitant
medications and adverse events.
COPD exacerbations were defined as onset or worsening of
more than one respiratory symptom (dyspnoea, cough, sputum
purulence or volume or wheeze) for >3 consecutive days (based
on diary cards or patients’ reports of their health since the
previous visit) plus documented proof of intensified treatment
(eg, systemic steroids, antibiotics or oxygen) and/or hospital-
isation or emergency room visit.
Health status(St George’s
SGRQ)6and dyspnoea (baseline and transition dyspnoea indices,
BDI/TDI7) were measured on day 1 and weeks 4, 8, 12, 24, 44
and 52. The modified Medical Research Council (mMRC)
dyspnoea scale8 9was measured on day 1, week 12 and week 52
followed by the 6-min walk test10as components of the BODE
(body mass index, obstruction (FEV1, % predicted), dyspnoea,
exercise) index.11
Adverse events were recorded at each visit. ECG assessment,
blood pressure and pulse rate measurements and blood sampling
for haematology were performed at regular intervals.
RespiratoryQuestionnaire,
Objectives and outcomes
The primary objective was to test the hypothesis that both
indacaterol doses were superior to placebo in their effect on
trough FEV1after 12 weeks. The important secondary efficacy
outcomes were days of poor COPD control (a composite
measure used in formoterol registration studies,12 13defined as
the number of days with a score of $2 on a 0e3 scale for at least
two symptoms out of cough, wheeze, production/colour of
sputum and breathlessness), SGRQ total score and time to first
exacerbation. Other secondary efficacy outcomes included other
spirometry, TDI scores and percentage of responding patients,
exacerbation rates, BODE index and end points derived from
diary data. Safety outcomes included the incidence of adverse
events and clinically notable values for plasma potassium
(<3.0 mmol/l) and blood glucose ($9.99 mmol/l). QTc interval
was calculated using the Fridericia formula.
Statistical methods
Efficacy results are presented for the modified intention-to-treat
(ITT) population including all randomised patients who received
atleastonedoseofstudydrugbutexcludingpatientsfromsixsites
owing to non-conformance with good clinical practice. Safety
resultsarepresentedforallpatientswhoreceivedatleastonedose
of study drug. All analyses presented here were preplanned.
The primary variable was analysed using a mixed model
analysis of covariance (ANCOVA) with treatment as a fixed
effect and baseline FEV1and FEV1reversibility as covariates. The
same model (with appropriate covariates) was used to analyse
other efficacy variables. The results are presented as estimated
adjusted treatment effects (least squares means) and treatment
contrasts with 95% CIs and two-sided p values. For trough FEV1
at weeks 12 and 52, SGRQ total score, TDI score and daily
salbutamol use, missing values were replaced using last obser-
vationcarried forward.Data
a maximum of 11 weeks, and for trough FEV1no data were
carried forward from before day 15. Exploratory analyses of
week 12 trough FEV1 values were performed in subgroups
defined according to baseline age, disease severity, smoking
history or ICS use. A hierarchical testing procedure to control
type I error for multiplicity was used for the comparisons of
indacaterol and placebo for the primary end point, then in turn
for days of poor COPD control, SGRQ total score at week 12
and time to first COPD exacerbation. Other analyses were not
adjusted for multiplicity.
Time to first COPD exacerbation was analysed using a Cox
regression model with the assumption of proportional hazards
checked with Schoenfeld residuals of treatment effects plotted
against time. The number of COPD exacerbations over 52 weeks
was analysed using a Poisson regression model with a sensitivity
analysis to assess the impact of premature discontinuations by
imputing an additional one exacerbation for these patients.
Neither the Cox nor the Poisson regression model assumes
normality. Normal distribution was assumed for efficacy and
safety variables where ANCOVAs were performed. For the
primary and important secondary variables (excluding exacer-
bations), the normality assumption was checked with a quant-
ileequantile plot of residuals for each treatment group.
Allsafetyendpointsweresummarisedforthesafetypopulation.
were carriedforwardfor
Sample size
With a SD of 270 ml for trough FEV1,12 13108 evaluable patients
per group were needed to detect a 120 ml difference between
indacaterol 300 mg and placebo as significant at the 5% level
(two-sided) with 90% power, with 84% power (due to the
hierarchical testing) for the indacaterol 600 mg versus placebo
comparison. However, the plan for at least 300 patients per
indacaterol dose group to provide a robust safety database
provided 99% power for the primary comparisons.
RESULTS
Patients
Patient disposition is shown in figure 1, which shows that the
ITT population was modified to exclude 129 patients (7.4% of
randomised) before unblinding because of non-conformance
with good clinical practice irrespective of treatment group, the
main reasons for non-conformance being substandard spirom-
etry, unsatisfactory completion of diary cards and health status
questionnaires and inability to verify source data. Baseline data
for all patients (safety population) are shown in table 1.
Approximately 5% of patients were enrolled before an early
protocol amendment stipulating postbronchodilator (rather
than prebronchodilator) spirometry values as entry criteria and
had baseline spirometry outside specified limits.
Spirometry
Trough FEV1at week 12 with both indacaterol doses was 170 ml
higher than placebo (p<0.001) and 100 ml higher than formo-
terol (p<0.001) (table 2). Over the remainder of the study,
significant differences versus placebo were maintained at
a similar level for indacaterol, while the difference between
formoterol and placebo diminished (table 2; figure 2). Analysis of
474 Thorax 2010;65:473e479. doi:10.1136/thx.2009.125435
Chronic obstructive pulmonary disease
group.bmj.com on June 7, 2010 - Published by thorax.bmj.com Downloaded from

Page 4

trough FEV1at week 12 for the ITT population (ie, including
patients from the centres excluded from the modified ITT
analysis) is provided in table S1 in the online supplement.
Supportive analysis of the primary variable excluding any
patients with protocol deviations (13e15% of patients across
the groups, including those not meeting the postbronchodilator
spirometry entry criteria; see table 1) showed similar results (not
shown). The group mean values and differences between treat-
ment and placebo groups in trough FEV1after 12 weeks were
similar for subgroups of patients analysed according to age,
COPD severity, smoking status and ICS use.
At 5 min postdose on day 1, compared with placebo FEV1
increased by 130 ml (95% CI 110 to 150) with indacaterol
300 mg, by 150 ml (95% CI 140 to 170) with indacaterol 600 mg
and by 140 ml (95% CI 120 to 160) with formoterol (all p<0.001
vs placebo). Serial measurements of FEV1showed a significant
effect of indacaterol over placebo at each time point over 24 h
postdose (figure 3). Other spirometry results are shown in the
online supplement.
Clinical outcomes
Both indacaterol doses were superior to placebo for all secondary
end points controlled for type I error (ie, days of poor control,
SGRQ and time to first exacerbation; differences between inda-
caterol and formoterol for these end points were not statistically
significant; tables 2 and 3, figure 4). Time to first COPD exacer-
bationwasimprovedwithalltreatments,asshownbytheHRsin
table3,althoughthereweretoofeweventstocalculateevent-free
time. Exacerbation rate ratios were similar without imputation
(table 3) and with imputation (not shown).
Bothindacaterol andformoterolincreased TDIscores, withthe
differences between indacaterol and placebo close to or exceeding
the1-pointminimumclinicallyimportantdifference14 15(table2).
TDI scores were significantly higher with indacaterol than with
formoterol at week 12 but not at week 52. At week 12, 63% and
58% of patients treated with indacaterol 300 and 600 mg, respec-
tively, had a clinically important increase of $1 in TDI score
compared with 40% of patients given placebo (p<0.001) and 53%
ofpatientstreatedwithformoterol(p<0.01forindacaterol300 mg
vs formoterol). Efficacy outcomes based on diary data and BODE
index (table 2) were generally significantly improved with inda-
caterolandformoterolversusplacebo,withsignificantdifferences
between indacaterol and formoterol for use of as-needed salbu-
tamol, nights without awakenings and PEF.
Safety
COPD worsening and nasopharyngitis were the only adverse
events reported by >10% of patients in any treatment group
(table 4). Most cases of COPD worsening were mild or moderate
in severity (indacaterol 300 mg, 92%; indacaterol 600 mg, 91%;
formoterol, 84%; placebo, 87%). All cases of nasopharyngitis
were mild or moderate with indacaterol and formoterol.
Eight patients died during treatment and four during follow-
up. Of the deaths during treatment, two were due to cardiac
arrest (indacaterol 300 mg; placebo), one to multiorgan failure
(formoterol), one to respiratory failure (formoterol) and four to
sudden death (one formoterol; three placebo). The indacaterol-
treated patient who died had discontinued treatment 3 days
earlier owing to dyspnoea (suspected as related to treatment).
The three deaths during formoterol treatment were considered
unrelated to formoterol.
Tremor was reported in 0.2%, 1.9%, 1.2% and 0.5% of the
indacaterol 300 mg, 600 mg, formoterol and placebo groups,
respectively, and tachycardia in 0.9%, 0.7%, 0.5% and 1.2%,
respectively. Two cases were severe (tremor in one patient
receiving placebo and atrial tachycardia in one patient receiving
indacaterol 600 mg).
In addition to patients reporting cough as an adverse event
(table 4), investigators were asked to record any instances of
cough occurring within 5 min of drug administration during
clinic visits regardless of whether they considered it an adverse
event. This was observed in an average of 19.1% of patients in
Figure 1
Patient flow through study.
Thorax 2010;65:473e479. doi:10.1136/thx.2009.125435475
Chronic obstructive pulmonary disease
group.bmj.com on June 7, 2010 - Published by thorax.bmj.com Downloaded from

Page 5

both indacaterol groups, 0.8% of the formoterol group and 1.8%
of the placebo group. The cough typically started within 15 s of
inhaling indacaterol, had a median duration of 12 s or less, and
was not associated with bronchospasm. Importantly, the pres-
ence of this cough was not associated with any increase in
discontinuation rates.
An increase in QTc interval of >60 ms from baseline occurred in
one patient (0.2%) in each of the indacaterol and formoterol groups
and in no patient in the placebo group. No patient had an absolute
value >500 ms. Serum potassium concentrations <3.0 mmol/l
were recorded for two patients (0.5%) in each of the indacaterol
300 mg and placebo groups and not in the other two groups. Group
mean values of serum potassium 1 h postdose at week 52 were
not significantly different. Blood glucose concentrations of
>9.99 mmol/l were recorded for 8.0%, 9.0%, 6.5% and 7.5% of the
indacaterol 300 mg, 600 mg, formoterol and placebo groups,
respectively. Group mean values of blood glucose 1 h postdose at
week 52 were statistically significantly different for indacaterol
600 mg compared with placebo (5.86 vs 5.61 mmol/l, p¼0.033).
DISCUSSION
This study provides clear evidence of effective 24 h broncho-
dilation with once-daily administration of indacaterol 300 or
600 mg. The effect of indacaterol on trough FEV1 relative to
placebo exceeded both the 120 ml level versus placebo that was
prespecified as clinically significant for this study and the
100e140 ml range proposed recently as a minimal important
difference.17Bronchodilator efficacy was independent of age,
severity of airways obstruction and smoking status. Results for
trough FEV1at week 12 were very similar when analysed for the
ITT population, suggesting that the exclusion of patients from
six sites for non-conformance with good clinical practice had
little impact on efficacy outcomes.
The bronchodilator effect of indacaterol was accompanied
by significant improvements compared with placebo in dysp-
noea and health status (both by a clinically meaningful
margin15 16), time to first exacerbation, BODE index, use of as-
needed salbutamol and symptom-based variables derived from
diary data. Although this study was not primarily powered to
detect significant differences between indacaterol and formo-
terol, indacaterol was more effective for several outcomes
including bronchodilator efficacy. The 100 ml difference in
trough FEV1 between the two treatments after 12 weeks
occurs in the early morning, a time that patients describe as
worst for COPD symptoms.18Among the clinical outcomes,
differences favoured indacaterol over formoterol in dyspnoea
score, use of as-needed salbutamol and nights without awak-
enings. Indacaterol had a greater effect than formoterol on
SGRQ total score, although the difference did not reach
statistical significance.
Table 1
Patient demographic data and other characteristics at baseline (safety population)
Indacaterol 300 mg
n[437
Indacaterol 600 mg
n[425
Formoterol
n[434
Placebo
n[432
Age, years
Male sex, n (%)
BMI, kg/m2
64.0 (57.0, 71.0)
351 (80.3)
25.9 (23.3, 28.9)
63.0 (57.0, 69.0)
327 (76.9)
26.0 (23.0, 29.7)
64.0 (58.0, 69.0)
348 (80.2)
25.7 (22.9, 29.0)
63.0 (57.5, 69.0)
352 (81.5)
26.4 (23.1, 30.0)
COPD severity, n (%)*
At risky
Mildy
Moderate
Severe
Very severe
9 (2.1)
2 (0.5)
226 (51.7)
190 (43.5)
9 (2.1)
3 (0.7)
6 (1.4)
212 (49.9)
188 (44.2)
15 (3.5)
7 (1.6)
7 (1.6)
226 (52.1)
182 (41.9)
10 (2.3)
10 (2.3)
10 (2.3)
216 (50.0)
186 (43.1)
9 (2.1)
ICS users, n (%)
Smoking history, pack-years
243 (55.6)
40.0 (30.0, 53.0)
226 (53.2)
40.0 (30.0, 58.0)
221 (50.9)
40.0 (30.0, 50.0)
224 (51.9)
43.0 (31.0, 53.5)
History of, n (%)
CCV condition
Diabetes mellitus
Hypertension
Hyperlipidaemia
80 (18.3)
33 (7.6)
227 (51.9)
145 (33.2)
93 (21.9)
39 (9.2)
204 (48.0)
138 (32.5)
88 (20.3)
37 (8.5)
213 (49.1)
133 (30.6)
82 (19.0)
50 (11.6)
216 (50.0)
131 (30.3)
FEV1, % predictedz
FEV1/FVC, %z
Prebronchodilator FEV1, l
Postbronchodilator FEV1, lz
Reversibility, %z
SGRQ total scorex
BDI total scorex
6-min walking distance, mx
BODE indexx
Data are median (upper and lower quartiles) unless stated otherwise.
*COPD severity based on GOLD 2005 criteria.
yPatients enrolled before a protocol amendment of entry criteria stipulating postbronchodilator rather than prebronchodilator spirometry values.
zMeasured 30 min after salbutamol 400 mg inhalation. Reversibility was calculated as the difference between the prebronchodilator and postbronchodilator values of FEV1(in litres) as
a percentage of the prebronchodilator value.
xReported for the modified intention-to-treat population.
BDI, baseline dyspnoea index; BMI, body mass index; BODE index, body mass index, obstruction, dyspnoea, exercise; CCV, cardio- or cerebrovascular; COPD, chronic obstructive pulmonary
disease; FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity; ICS, inhaled corticosteroid; SGRQ, St George’s Respiratory Questionnaire.
51.5 (42.2, 62.9)
50.7 (43.5, 59.5)
1.29 (1.01, 1.58)
1.44 (1.14, 1.78)
9.8 (3.0, 18.3)
43 (32, 58)
6.0 (6.0, 8.0)
379 (300, 448)
3.0 (2.0, 4.0)
50.8 (41.2, 60.2)
51.1 (43.8, 59.1)
1.28 (0.98, 1.60)
1.41 (1.11, 1.75)
10.9 (3.8, 20.6)
44 (31, 56)
6.0 (5.0, 8.0)
370 (300, 447)
3.0 (1.0, 4.0)
52.5 (41.2, 63.1)
51.2 (43.5, 59.0)
1.30 (1.00, 1.62)
1.47 (1.12, 1.80)
10.1 (3.4, 18.3)
44 (32, 57)
6.0 (5.0, 8.0)
380 (300, 450)
3.0 (1.0, 4.0)
52.0 (41.9, 63.6)
52.0 (44.1, 60.5)
1.28 (1.00, 1.70)
1.44 (1.12, 1.85)
10.8 (4.7, 19.1)
43 (30, 57)
6.0 (5.0, 8.0)
390 (307, 444)
3.0 (2.0, 4.0)
476 Thorax 2010;65:473e479. doi:10.1136/thx.2009.125435
Chronic obstructive pulmonary disease
group.bmj.com on June 7, 2010 - Published by thorax.bmj.comDownloaded from