925

CHRONIC OBSTRUCTIVE PULMONARY DISEASE

Severe acute exacerbations and mortality in patients

with chronic obstructive pulmonary disease
J J Soler-Cataluña, M Á Martı́nez-Garcı́a, P Román Sánchez, E Salcedo, M Navarro,
R Ochando
...............................................................................................................................
Thorax 2005;60:925–931. doi: 10.1136/thx.2005.040527

See end of article for Background: Patients with chronic obstructive pulmonary disease (COPD) often present with severe acute
authors’ affiliations exacerbations requiring hospital treatment. However, little is known about the prognostic consequences of
.......................
these exacerbations. A study was undertaken to investigate whether severe acute exacerbations of COPD
Correspondence to: exert a direct effect on mortality.
Dr J J Soler-Cataluña, Methods: Multivariate techniques were used to analyse the prognostic influence of acute exacerbations of
Unidad de Neumologı́a, COPD treated in hospital (visits to the emergency service and admissions), patient age, smoking, body
Servicio de Medicina
Interna, Hospital General mass index, co-morbidity, long term oxygen therapy, forced spirometric parameters, and arterial blood
de Requena, Paraje gas tensions in a prospective cohort of 304 men with COPD followed up for 5 years. The mean (SD) age of
Casablanca s/n 46340, the patients was 71 (9) years and forced expiratory volume in 1 second was 46 (17)%.
Requena (Valencia),
Spain; jjsoler@telefonica. Results: Only older age (hazard ratio (HR) 5.28, 95% CI 1.75 to 15.93), arterial carbon dioxide tension
net (HR 1.07, 95% CI 1.02 to 1.12), and acute exacerbations of COPD were found to be independent
indicators of a poor prognosis. The patients with the greatest mortality risk were those with three or more
Received 13 January 2005 acute COPD exacerbations (HR 4.13, 95% CI 1.80 to 9.41).
Accepted 3 July 2005
Published Online First Conclusions: This study shows for the first time that severe acute exacerbations of COPD have an
29 July 2005 independent negative impact on patient prognosis. Mortality increases with the frequency of severe
....................... exacerbations, particularly if these require admission to hospital.

C
hronic obstructive pulmonary disease (COPD) is a series have examined the specific influence of acute exacer-
chronic illness with great social, health care, and bations of COPD on patient mortality.
economic repercussions. It is the fourth most frequent Some authors17–20 have recently detected a high mortality rate
cause of death after neoplastic disease, ischaemic heart following hospital admission, ranging from 22–43% after 1 year
disease, and cerebrovascular diseases.1 However, of all these to 36–49% after 2 years, depending on the severity of the COPD
illnesses, only COPD mortality has increased in recent years patients studied. These series have specifically investigated the
and by the year 2020 it is expected to become the third risk factors associated with mortality after hospitalisation.
leading cause of death in the world.2 Again, advanced patient age,17 19 20 PaCO2,19 PaO2/FiO2,17 BMI,17
Different authors have investigated the predictive factors serum albumin,17 co-morbidity,18 20 cor pulmonale,17 20 and
associated with increased COPD mortality. Forced expiratory functional status17 18 have been identified as prognostic variables.
volume in one second (FEV1),3 4 patient age,4 hypoxaemia,5 These results suggest that the principal determinant of death
hypercapnia,3 co-morbidity,3 4 6 pulmonary hypertension,7 following hospitalisation is the baseline severity of the disease—
and body mass index (BMI)8 are some of the adverse that is, the greater the severity of COPD, the greater the likeliness
prognostic variables most commonly cited in the literature. of hospital admission and also of death. However, to date no
In recent years other risk factors have also been described study has examined the possibility that acute exacerbations of
including dyspnoea,9 health related quality of life (HRQoL),10 COPD may intrinsically exert a negative effect upon patient
and exercise tolerance.11 A new multifactorial prognostic prognosis independently of the baseline severity of the disease.
classification—the BODE (BMI, airflow obstruction, dys- This study investigates whether severe acute exacerbations
pnoea, exercise performance) index—has recently been of COPD—that is, exacerbation episodes requiring hospital
proposed. This classification, proposed by Celli et al,12 stresses management—exert a direct and independent effect on the
the multicomponent nature of COPD and addresses not only survival of patients with COPD.
its pulmonary consequences but also the systemic manifesta-
tions of the disease. None of these studies has specifically METHODS
examined the prognostic influence of acute exacerbations of Subjects
COPD, despite the fact that they have a very relevant role in A prospective study was performed in a cohort of 304 male
the natural course of the disease. It has been estimated that patients with stable COPD. The patients were recruited in the
COPD patients suffer 1–4 exacerbations per year.13 Such course of 1998 and followed for 5 years. The general
decompensation episodes have an important impact on characteristics of the cohort and the selection criteria used
HRQoL14 and generate great healthcare burdens and eco- are described in detail elsewhere.15 Briefly, the diagnosis of
nomic costs. Between 1% and 2% of all emergency service COPD was based on current or past smoking history (.20
visits and 10% of all medical admissions are attributable to
acute exacerbations of COPD.15 Nearly 60% of the global cost
Abbreviations: BMI, body mass index; COPD, chronic obstructive
of the disease is associated with exacerbation episodes, pulmonary disease; FEV1, forced expiratory volume in 1 second; FVC,
particularly severe acute exacerbations requiring admission forced vital capacity; LTOT, long term oxygen therapy; PaO2, PaCO2,
to hospital.16 Despite this strong impact, however, very few arterial oxygen and carbon dioxide tensions

www.thoraxjnl.com
926 Soler-Cataluñ a, Martı´nez-Garcı´a, Romá n Sá nchez, et al

pack years), clinical evaluation, and pulmonary function Statistical analysis

testing showing airflow obstruction (FEV1/FVC ,70) with a Descriptive statistics were used to describe the study
change in FEV1 of less than 200 ml and 12% in the population at baseline. The comparison of means among
bronchodilator test.21 The severity of the disease was the three study groups was based on analysis of variance,
established according to the latest GOLD criteria.22 Patients with x2 testing and Bonferroni correction for comparison of
previously diagnosed with bronchial asthma, bronchiectasis, the proportions. Survival of all subjects was assessed after
cystic fibrosis, upper airways obstruction, or bronchiolitis 5 years and all-cause mortality was evaluated. The starting
related to systemic pathology were excluded. All included date was taken as 1 January 1998 and the termination date
patients were required to be in a stable phase of the disease, was 31 December 2002. We first conducted univariate
without acute exacerbations of COPD in the month preceding analyses based on the Cox proportional hazards model using
the study. each of the potential predictors of respiratory mortality as
independent variables and survival as the dependent vari-
Study protocol able.27 Survival curves for the three groups were estimated by
Age, sex, smoking history, co-morbidity, BMI, long term the Kaplan-Meier product limit method and compared using
oxygen therapy (LTOT), forced spirometric parameters, and the log rank test.28 Independent variables associated with
arterial blood gas data were collected in all patients. Co- respiratory mortality with p,0.15 in the univariate analysis
morbidity was quantified according to the index of Charlson were then incorporated into a multivariate analysis also
et al.23 This index has been developed to predict mortality based on the Cox proportional model. An interaction term
among patients with chronic diseases. BMI was calculated by between the variables and time was introduced into the
dividing patient body weight (in kg) by the square of height model to analyse risk proportionality. All statistical analyses
(in m2). FEV1 and forced vital capacity (FVC) were were carried out using a statistical software package (SPSS
determined by forced spirometry (Autospiro AS-600, for Windows, version 11.5; SPSS Inc, Chicago, IL, USA). A p
Minato Medical Science SA, Japan), following the guidelines value of ,0.05 was considered significant.
established by the Spanish Society of Pneumology and Chest
Surgery (Sociedad Española de Neumologı́a and Cirugı́a RESULTS
Torácica, SEPAR).24 The FEV1 and FVC results are expressed Subject characteristics
as percentages of the adult reference values.25 Of the 320 cases in the initial cohort, two (0.6%) were
Postbronchodilator FEV1 was used as airflow limitation excluded because of an absence of smoking history and 11
index because it is regarded as a better predictor of mortality (3.4%) were excluded because they had a smoking history of
than prebronchodilator FEV1.4 ,20 pack years. The only women in the study (n = 3, 0.9%)
were also excluded to increase homogeneity. The study
Exacerbations therefore comprised a total of 304 male patients diagnosed
A severe acute exacerbation of COPD was defined as any with COPD (mean age 71 (9) years). The baseline character-
sustained increase in respiratory symptomatology compared istics of the patients are shown in table 1. One hundred and
with the baseline situation requiring modification of regular sixty three cases (53.6%) had suffered no acute exacerbations
medication and hospital treatment.26 A prospective registry (group A); 105 (34.5%) had two or less acute exacerbations,
was made of all exacerbation episodes requiring hospital 60 of whom had been admitted to hospital (group B); and 36
management during the year of the study. The patients were (11.8%) had suffered three or more exacerbations, 29 having
divided into three groups according to the number of acute been admitted to hospital at least once during the year of the
exacerbations recorded: group A (no acute exacerbations); study (group C). In general terms, the patients belonging to
group B (1–2 acute exacerbations); and group C (>3 acute group C were older and had more advanced disease, with
exacerbations). lower FEV1, FVC, and PaO2 and higher PaCO2 (table 1).

Table 1 Mean (SD) baseline characteristics of study patients (n = 304)

All patients Group A Group B Group C
(n = 304) (n = 163, 53.6%) (n = 105, 34.5%) (n = 36, 11.8%) p value

Age (years) 71 (9) 69 (10) 72 (7) 71 (9) 0.001

BMI (kg/m2) 28.3 (5.1) 28.8 (5.4) 27.7 (4.7) 27.7 (4.1) 0.350
Co-morbidity index 0.95 (1.34) 0.90 (1.30) 0.92 (1.21) 1.23 (1.82) 0.422
Co-morbidity, n (%) 0.765
0 145 (47.7%) 80 (48.8%) 48 (46.2%) 17 (47.2%)
1 91 (29.9%) 47 (28.6%) 35 (33.6%) 9 (25.0%)
>2 68 (22.4%) 37 (22.6%) 21 (20.2%) 10 (27.8%)
FEV1 (l) 1.18 (0.64) 1.29 (0.76) 1.11 (0.45) 0.88 (0.33) 0.002
FEV1 (% predicted) 46.4 (17.2) 49.2 (16.6) 45.9 (17.6) 34.5 (13.6) ,0.001
GOLD classification, n (%) ,0.001
Stage I 20 (6.6%) 15 (9.2%) 5 (4.8%) –
Stage II 109 (35.9%) 66 (40.5%) 38 (36.2%) 5 (13.9%)
Stage III 102 (33.6%) 57 (35.0%) 33 (31.4%) 12 (33.3%)
Stage IV 73 (24.0%) 25 (15.3%) 29 (27.6%) 19 (52.8%)
FVC (l) 2.22 (0.67) 2.34 (0.72) 2.12 (0.62) 1.93 (0.62) 0.002
FVC (% predicted) 69.4 (19.1) 72.0 (19.3) 68.0 (18.7) 61.1 (17.3) 0.010
PaO2 (kPa) 8.8 (1.6) 9.1 (1.5) 8.6 (1.5) 8.0 (1.6) ,0.001
PaCO2 (kPa) 5.8 (1.0) 5.7 (0.8) 5.9 (1.0) 6.2 (1.4) 0.019

BMI, body mass index; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; PaO2, PaCO2, arterial oxygen and carbon dioxide tensions.
Group A, no acute exacerbations of COPD; group B, 1–2 acute exacerbations of COPD (emergency visits or hospital admissions); group C, >3 acute
exacerbations of COPD.

www.thoraxjnl.com
 Acute exacerbations and mortality in COPD 927

Table 2 Predictors of mortality: univariate analysis

Hazard ratio
(crude) 95% CI p value

Age (years) ,0.001

,65 – –
65–74 1.76 0.88 to 3.52
>75 5.26 2.70 to 10.24
Current smoking 0.97 0.63 to 1.48 0.873
Cumulative smoking (pack years) 1.00 0.99 to 1.01 0.973
Co-morbidity index 1.43 1.28 to 1.60 ,0.001
Charlson index 0.002
0 1.00 –
1 1.32 0.84 to 2.07
>2 2.20 1.41 to 3.43
BMI (kg/m2) 0.94 0.89 to 0.99 0.034
Postbronchodilator FEV1 (% predicted) 0.96 0.94 to 0.98 ,0.001
GOLD stage ,0.001
I 1.00 –
II 3.40 0.81 to 14.21
III 5.34 1.29 to 22.07
IV 8.34 2.01 to 34.55
FVC (% predicted) 0.98 0.97 to 0.99 0.009
PaO2/FiO2 0.99 0.98 to 0.99 ,0.001
PaCO2 (kPa) 1.05 1.04 to 1.09 ,0.001
LTOT 3.42 2.36 to 4.95 ,0.001
Acute exacerbation groups ,0.001
Group A 1.00 –
Group B 2.20 1.45 to 3.33
Group C 4.30 2.62 to 7.02

BMI, body mass index; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; PaO2, PaCO2,
arterial oxygen and carbon dioxide tensions; LTOT, long term oxygen therapy.
Group A, no acute exacerbations of COPD; group B, 1–2 acute exacerbations of COPD (emergency visits or
hospital admissions); group C, >3 acute exacerbations of COPD.

Univariate survival analysis CI 2.62 to 7.02) than for patients requiring no hospital
A total of 116 deaths (38.2%) were recorded; 78 (25.7%) were management (group A). Patients in group B also showed
due to respiratory causes and 38 (12.5%) died of different significant differences in survival compared with group A
causes (12 (3.9%) from cardiovascular disease, seven (2.3%) (hazard ratio (HR) 2.20, 95% CI 1.45 to 3.33).
from cerebrovascular disease, 11 (3.6%) neoplasms, and six Eighty nine patients (29.3%) were admitted to hospital at
(2.0%) from other diseases). In two cases (0.7%) the cause of least once during 1998. Mortality in this group after 12, 24,
death was unknown. Thirty two patients were lost in the 36, 48, and 60 months was 11.6%, 25.9%, 40.2%, 46.6% and
course of the 5 year follow up period (follow up rate 89.5%). 55.2%, respectively. Patients with only one hospital admis-
Table 2 shows the influence on the prognosis of the variables sion had poorer survival than those with no acute exacerba-
included in the univariate analysis. tions of COPD (HR 2.94, 95% CI 1.82 to 4.72) or than patients
Figure 1 shows the survival curves according to the with visits to the emergency service without admission. The
frequency of acute exacerbations of COPD. Patients with lowest survival was observed in the group with readmissions
frequent exacerbations (group C) had the highest mortality (HR 4.31, 95% CI 2.70 to 6.88; fig 2).
rate (p,0.001) with a risk of death 4.30 times greater (95%

1.0
1.0

0.8
0.8
Probability of surviving

A
Probability of surviving

(1)
p<0.0002 NS
0.6 (2)
0.6
p<0.0001
B p<0.0001 p=0.005
0.4 p=0.069 0.4 p<0.0001
(3)
C NS
0.2 0.2 (4)

0.0 0.0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (months) Time (months)

Figure 1 Kaplan-Meier survival curves by frequency of exacerbations Figure 2 Kaplan-Meier survival curves by severity of exacerbations in
in patients with COPD: group A, patients with no acute exacerbations of patients with COPD: (1) no acute exacerbations of COPD; (2) patients
COPD; group B, patients with 1–2 acute exacerbations of COPD with acute exacerbations of COPD requiring emergency service visits
requiring hospital management; group C, patients with >3 acute without admission; (3) patients with acute exacerbations of COPD
exacerbations of COPD. requiring one hospital admission; (4) patients with readmissions.

www.thoraxjnl.com
928 Soler-Cataluñ a, Martı´nez-Garcı´a, Romá n Sá nchez, et al

Interaction and confounding by variables presenting with >3 exacerbations, particularly if they
No significant interaction was observed among the different required hospitalisation. We consider that these observations
study variables and the prognostic effect of acute exacerba- are very important since the frequency of acute exacerbations
tions of COPD. The confounding variables of this association of COPD is potentially modifiable.
were age, FEV1%, PaO2/FiO2, and LTOT. After adjusting the Different studies have reported a high mortality rate after
model for each of these variables, the presence of acute admission to hospital with acute exacerbations of COPD.17–20
exacerbations of COPD continued to appear as an indepen- In the most important series published to date, Connors et al17
dent prognostic variable in all cases (table 3). recorded an in-hospital mortality rate of 11% in patients with
acute hypercapnic respiratory failure. During subsequent
Multivariate analysis of survival follow up the subjects who survived the hospital admission
Age, co-morbidity index, BMI, FEV1 (%), FVC (%), PaO2/FiO2, had mortality rates of 43% and 49% after 1 and 2 years,
PaCO2, LTOT, and the number of acute exacerbations of COPD respectively. Other series with less severe exacerbations have
during the year of the study were the variables included in also reported a considerable number of deaths after admis-
the Cox multiple regression model. In this multivariate sion, although to a somewhat lesser degree. Almagro et al,18 in
model, the frequency of acute exacerbations, age, and a series of 135 patients, reported respiratory mortality figures
Charlson index were analysed as categorical variables. The of 22% and 35.6% after 1 and 2 years, respectively.
risk proportionality test proved non-significant; hazards Groenewegen et al19 likewise found a 23% mortality rate after
ratios were unchanged over time. Table 4 shows the 1 year in a series of 171 patients. The number of deaths is
regression model adjusted for all the significant prognostic even higher in patients with COPD requiring mechanical
variables. Severe acute exacerbations of COPD, particularly ventilation. Seneff et al29 reported an in-hospital mortality
when multiple, were one of the most relevant independent rate of 24%, with mortality rates after hospital discharge
adverse prognostic variables with an adjusted mortality risk increasing to 59% after 1 year.
four times greater than for patients without acute exacerba- Based on these findings, it has been suggested that
tions. Older age and PaCO2 also had a deleterious effect on admission to hospital for acute exacerbations of COPD allows
survival. the identification of a subgroup of patients with a poorer
prognosis.18 The most widely accepted hypothesis relates
DISCUSSION baseline severity of the disease to an increased likeliness
To our knowledge, this is the first study to show that severe of an exacerbation—a situation which, in turn, would explain
exacerbations requiring hospital management are indepen- the poorer survival recorded after hospitalisation. This
dently associated with all-cause mortality in patients with hypothesis is reinforced by identification of the risk factors
COPD. Our results suggest that the mortality risk increases associated with mortality following admission, since the
with the frequency of severe acute exacerbations. The variables defining the baseline severity of COPD—such as
maximum mortality risk occurred in those individuals advanced patient age,17 19 20 hypoxaemia,17 hypercapnia,19

Table 3 Prognostic value of frequency of acute exacerbations of COPD (AECOPD)

adjusted by confounders
Hazard ratio
(adjusted) 95% CI p value

AECOPD adjusted by age

Age (years) ,0.001
,65 Reference –
65–74 1.62 0.80 to 3.25
>75 4.52 2.31 to 8.83
AECOPD groups ,0.001
Group A Reference –
Group B 1.93 1.27 to 2.93
Group C 3.64 2.22 to 5.98

AECOPD adjusted by FEV1%

FEV1% (predicted) 0.99 0.97 to 0.99 0.028
AECOPD groups ,0.001
Group A Reference –
Group B 2.08 1.33 to 3.25
Group C 3.45 1.99 to 5.99

AECOPD adjusted by PaO2/FiO2

PaO2/FiO2 0.99 0.990 to 0.997 ,0.001
AECOPD groups ,0.001
Group A Reference –
Group B 2.12 1.37 to 3.29
Group C 3.69 2.18 to 6.25

AECOPD adjusted by LTOT

LTOT 2.72 1.84 to 4.02 ,0.001
AECOPD groups
Group A Reference – ,0.001
Group B 1.74 1.14 to 2.67
Group C 2.92 1.74 to 4.88

AECOPD, acute exacerbations of COPD; FEV1, forced expiratory volume in 1 second; PaO2, arterial oxygen
tension; FiO2, fractional inspired oxygen; LTOT, long term oxygen therapy.
Group A, no acute exacerbations of COPD; group B, 1–2 acute exacerbations of COPD (emergency visits or
hospital admissions); group C, >3 acute exacerbations of COPD.

www.thoraxjnl.com
Acute exacerbations and mortality in COPD 929

Table 4 Predictors of mortality: multivariate analysis

Hazard ratio
(adjusted) 95% CI p value

Age (years) 0.001

,65 Reference –
65–74 1.92 0.64 to 5.25
>75 5.28 1.75 to 15.93
Charlson index 0.101
0 Reference –
1 1.17 0.59 to 2.29
>2 2.23 1.06 to 4.69
BMI (kg/m2) 0.96 0.89 to 1.03 0.245
FEV1 (% predicted) 0.97 0.96 to 1.03 0.810
FVC (% predicted) 0.97 0.97 to 1.02 0.695
PaO2/FiO2 1.00 0.99 to 1.01 0.374
PaCO2 1.07 1.02 to 1.12 0.006
LTOT 1.95 0.93 to 4.12 0.078
Exacerbation groups 0.003
Group A Reference –
Group B 2.00 1.01 to 3.98
Group C 4.13 1.80 to 9.45

BMI, body mass index; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; PaO2, PaCO2,
arterial oxygen and carbon dioxide tensions; FiO2, fractional inspired oxygen; LTOT, long term oxygen therapy.
Group A, no acute exacerbations of COPD; group B, 1–2 acute exacerbations of COPD (emergency visits or
hospital admissions); group C, >3 acute exacerbations of COPD.

BMI,17 co-morbidity,18 20 cor pulmonale,17 20 or sustained oral survival was the overall severity of illness on day 3 of the
corticosteroid treatment19—have been described as mortality hospital stay as measured by the acute physiology component
predicting factors after admission. of the APACHE score. Blood gas values on hospital admis-
We also observed a high mortality among patients sion—another measure of acute abnormality—have also been
requiring admission to hospital, with a crude mortality risk reported as an independent predictor of survival.17–20 Other
2.94-fold greater than in subjects without acute exacerba- aspects intrinsically related to acute COPD exacerbations may
tions requiring hospital admission. However, unlike in the also be implicated in the survival of these patients. Fuso et
earlier series, our results indicate that acute exacerbations of al,20 in a retrospective series, showed that the variables
COPD requiring hospital management per se increase the risk reflecting heart dysfunction during exacerbations—such as
of death independently of other classical prognostic factors atrial fibrillation or ventricular arrhythmias—are important
such as FEV1, age, BMI, co-morbidity or respiratory failure. determinants of mortality risk. Connors et al17 likewise
Increased mortality is particularly observed in cases requiring observed that patients presenting with congestive heart
admission to hospital and less so in patients requiring failure as the cause of the acute COPD exacerbation suffered
emergency visits without admission to hospital, which increased mortality. In contrast, an infectious aetiology of the
suggests that the severity of the exacerbation influences acute exacerbation does not seem to be a determinant of
mortality. However, exacerbation recurrence also has an increased mortality.17 18
important effect on prognosis since the risk was particularly One of the most attractive aetiological hypotheses relates
high for patients with three or more acute exacerbations and inflammation (pulmonary or systemic) triggered during an
for those who had been readmitted. A recent prospective acute exacerbation to patient survival. Donaldson et al32 have
study following patients with severe COPD suggests that lung recently shown that the severity of exacerbations increases
function deteriorates more rapidly in those with more over time, as does sputum purulence, which suggests more
frequent exacerbations.30 This effect appears to be small, inflammation in patients with more severe exacerbations. In
however, with a mean decline in FEV1 of 8 ml/year in our study we only analysed acute exacerbations which were
individuals experiencing exacerbations more often than the sufficiently serious to require hospital treatment. Therefore,
median compared with those who had exacerbations less in accordance with the results of Donaldson el al,32 it is
often than the median. Although not specifically designed to possible that our study patients may have had greater
evaluate exacerbations, the Lung Health Study noted a inflammation. A group of British investigators has also
similar effect of exacerbations on the rate of decline of shown a relationship between airway inflammation and the
FEV1 in smokers but no effect in ex-smokers.31 Both studies frequency of exacerbations in patients with COPD.33 34
suggest that exacerbations accelerate the loss of lung According to these authors, patients with frequent exacerba-
function, this being an indirect indicator of mortality. tions have an increased bacterial load in their airways in the
However, to our knowledge, this is the first study to identify stable state. This lower airway bacterial load is associated
an intrinsic deleterious effect of acute exacerbations on with increased airway inflammation and an accelerated
patient survival. decrease in FEV1, an indirect but potent mortality indicator.
The reasons underlying this increase in risk following acute Poorer survival among patients with frequent acute COPD
exacerbations of COPD are not clear. Our study was not exacerbations is currently the subject of debate. In a Dutch
specifically designed to examine this, but the results suggest series19 the authors recorded no differences in survival rates
that, as the severity of exacerbations increases (regardless of between those who had been readmitted and those who had
the baseline severity of the disease), the risk of death also not been readmitted to hospital. In contrast, Connors et al,17
increases. In fact, patients admitted to hospital had higher in a univariate analysis, observed an increase in mortality
mortality rates than those seen in the emergency service who associated with readmission. Mortality after 6 months
did not require admission to hospital (fig 2). Similar results among patients readmitted one, two or more times was
have been reported by other authors. In the series published 27%, 31% and 36%, respectively, while those requiring no
by Connors et al17 the most important predictor of 6 month readmission had a mortality rate of 21% (p = 0.004). More

www.thoraxjnl.com
930 Soler-Cataluñ a, Martı´nez-Garcı´a, Romá n Sá nchez, et al

.....................
recently, a Spanish multivariate study18 reported a 1.85-fold Authors’ affiliations
greater mortality rate among readmitted subjects after J J Soler-Cataluña, M Á Martı́nez-Garcı́a, Hospital General de
adjusting for other predictive variables. We also found that Requena, Unidad de Neumologı́a, Servicio de Medicina Interna,
patients with multiple acute exacerbations admitted to Requena (Valencia), Spain
hospital or seen in the emergency service had a markedly P Román Sánchez, E Salcedo, M Navarro, Hospital General de
Requena, Servicio de Medicina Interna, Requena (Valencia), Spain
greater mortality risk and, after adjusting for different R Ochando, Hospital General de Requena, Servicio de Urgencias,
prognostic factors, mortality in these patients was 4.1-fold Requena (Valencia), Spain
greater than in those with no acute COPD exacerbations.
Competing interests : none declared
These results could support the inflammatory hypothesis
whereby, as the frequency of severe acute exacerbations
increases, inflammation and mortality also increase.
REFERENCES
However, we did not assess pulmonary or systemic inflam- 1 World Health Organization. World Health Report. Geneva: World Health
mation so further studies in this direction are needed. Organization, 2000.
To date, the only treatments that have been shown to 2 Murray CJL, Lopez AD. Mortality by cause for eight regions of the world:
Global Burden of Disease Study. Lancet 1997;349:1269–76.
improve survival in COPD patients have been smoking 3 Burrows B, Earle RH. Prediction of survival in patients with chronic airway
cessation35 and oxygen therapy in subjects with respiratory obstruction. Am Rev Respir Dis 1969;99:865–71.
failure.5 If our results are confirmed, treatments capable of 4 Anthonisen NR, Wright EC, Hodgkin JE. Prognosis in chronic obstructive
pulmonary disease. Am Rev Respir Dis 1986;133:14–20.
reducing the frequency of acute exacerbations of COPD could 5 Nocturnal Oxygen Therapy Trial Group. Continuous or nocturnal oxygen
potentially reduce mortality inherent to exacerbations. Some therapy in hypoxemic chronic obstructive pulmonary disease: a clinical trial.
drugs such as inhaled steroids,36 inhaled long acting b2 Ann Intern Med 1980;93:391–8.
6 Incalzi RA, Fuso L, De Rosa M, et al. Co-morbidity contributes to predict
agonists,37 and tiotropium bromide38 have resulted in a mortality of patients with chronic obstructive pulmonary disease. Eur Respir J
significant reduction in the number of exacerbations. This 1997;10:2794–800.
reduction is associated with a marked improvement in 7 Oswald-Mammosser M, Weitzenblum E, Quoix E, et al. Prognostic factors in
COPD patients receiving long-term oxygen therapy. Importance of pulmonary
HRQoL,36–38 and confirmation of our results might suggest artery pressure. Chest 1995;107:1193–8.
that this treatment could also contribute to a reduction in 8 Landbo C, Prescott E, Lange P, et al. Prognostic value of nutritional sattus in
mortality. In a recent pharmacoepidemiological study chronic obstructive pulmonary disease. Am J Respir Crit Care Med
1999;160:1856–61.
Soriano et al39 found that patients treated with a combination 9 Nishimura K, Izumi T, Tsukino M, et al. Dyspnea is a better predictor of 5-year
of fluticasone and salmeterol had better survival than those survival than airway obstruction in patients with COPD. Chest
receiving monotherapy. An important multicentre study is 2002;121:1434–40.
currently underway to assess the impact of this combined 10 Domingo-Salvany A, Lamarca R, Ferrer M, et al. Health-related quality of life
and mortality in male patients with chronic obstructive pulmonary disease.
treatment on patient survival.40 If our results are confirmed, Am J Respir Crit Care Med 2002;166:680–5.
other management strategies such as measures to combat the 11 Oga T, Nishimura K, Tsukino M, et al. Analysis of the factors related to
high therapeutic failure rate or early treatment could also mortality in chronic obstructive pulmonary disease. Role of exercise capacity
and health status. Am J Respir Crit Care Med 2003;167:544–9.
afford benefit in terms of survival. 12 Celli BR, Cote CG, Marı́n JM, et al. The body-mass index, airflow obstruction,
Our study has some limitations. Firstly, the prognostic dyspnea, and exercise capacity index in chronic obstructive pulmonary
influence of acute exacerbations of COPD requiring hospital disease. N Engl J Med 2004;350:1005–12.
13 Miratvilles M, Mayordomo C, Artés M, et al. Treatment of chronic obstructive
management was analysed and it is not clear whether other pulmonary disease and its exacerbations in general practice. Respir Med
less severe exacerbation episodes could also exert a prog- 1999;93:173–9.
nostic influence. Secondly, although we have attempted to 14 Seemungal TAR, Donaldson GC, Paul EA, et al. Effect of exacerbation on
quality of life in patients with chronic obstructive pulmonary disease.
incorporate the prognostic variables most commonly cited in Am J Respir Crit Care Med 1998;157:1418–22.
the literature, some parameters that have not been examined 15 Soler JJ, Sánchez L, Latorre M, et al. Impacto asistencial hospitalario de la
(such as the degree of dyspnoea, HRQoL, exercise tolerance, EPOC. Peso especı́fico del paciente con EPOC de alto consumo sanitario.
Arch Bronconeumol 2001;37:375–81.
or pulmonary artery pressure) could also interact with 16 Hilleman DE, Dewan N, Malesker M, et al. Pharmacoeconomic evaluation of
exacerbations and their prognostic repercussions. COPD. Chest 2000;118:1278–85.
Nevertheless, in an earlier study41 in which we considered 17 Connors AF, Dawson NV, Thomas C, et al. Outcomes following acute
exacerbation of severe chronic obstructive lung disease. Am J Respir Crit Care
some of these variables, hospital admission as an expression Med 1996;154:959–67.
of severe acute exacerbations was likewise shown to be an 18 Almagro P, Calbo E, Ochoa de Echagüen A, et al. Mortality after
independent predictive variable. Another limitation of our hospitalization for COPD. Chest 2002;121:1441–8.
19 Groenewegen KH, Schols AMWJ, Wouters E. Mortality and mortality-related
study is the cross sectional data collection. Although most factors after hospitalization for acute exacerbation of COPD. Chest
studies adopt this same design, longitudinal data (changes in 2003;124:459–67.
treatment, impairment of lung function, loss of muscle mass, 20 Fuso L, Incalzi RA, Pistelli R, et al. Predicting mortality of patients hospitalized
for acutely exacerbated chronic obstructive pulmonary disease. Am J Med
etc), which could equally condition mortality and interact 1995;98:272–7.
with future exacerbations, were not available. Lastly, we lack 21 Global Initiative for Chronic Obstructive Lung Disease. Global strategy for the
detailed data on the acute exacerbations such as clinical diagnosis, management and prevention of chronic obstructive pulmonary
disease, NIH Publication No 2701. NHLBI/WHO Workshop Report.
aetiology and treatment, so no in-depth evaluation can be Bethesda, MD: National Heart, Lung and Blood Institute, 2001:1–100.
made of the causes underlying their future course. The aim of 22 Fabbri LM, Hurd SS, for the GOLD Scientific Committee. Global strategy for
our study was to draw attention to the prognostic role of the diagnosis, management and prevention of COPD: 2003 update. Eur
acute exacerbations of COPD per se, rather than to define the Respir J 2003;22:1–2.
23 Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic
aetiopathogenic mechanisms associated with the increased comorbidity in longitudinal studies: development and validation. J Chronic Dis
mortality rate of such exacerbations. Further studies are 1987;40:373–83.
needed which are designed specifically to address this matter. 24 Sanchis J, Casan P, Castillo J, et al. Normativa para la práctica de la
espirometrı́a forzada. Arch Bronconeumol 1989;25:132–42.
In conclusion, our results suggest for the first time that 25 Roca J, Sanchis J, Agustı́-Vidal A, et al. Spirometric reference values for a
severe acute exacerbations of COPD have an independent Mediterranean population. Bull Eur Physiopathol Respir 1986;22:217–24.
negative prognostic impact, with mortality increasing with 26 Rodrı́guez-Roisin R. Toward a consensus definition for COPD exacerbation.
Chest 2000;117:398–401s.
the frequency of severe exacerbations, particularly if these 27 Cox DR. Regression models and life tables. J R Stat Soc 1972;B34:187–220.
require admission to hospital. If these results are confirmed, 28 Kaplan EL, Meier P. Nonparametric estimation from incomplete observation.
a reduction in the number and severity of exacerbations J Am Stat Assoc 1969;53:457–81.
29 Seneff MG, Wagner DP, Wagner RP, et al. Hospital and 1-year survival of
could be regarded as priority objectives in the management of patients admitted to intensive care units with acute exacerbation of chronic
COPD. obstructive pulmonary disease. JAMA 1995;274:1852–7.

www.thoraxjnl.com
Acute exacerbations and mortality in COPD 931

30 Donaldson GC, Seemungal TAR, Bhowmik A, et al. Relationship between 36 Burge PS, Calverley PMA, Jones PW, et al. on behalf of the ISOLDE
exacerbation frequency and lung function decline in chronic obstructive study investigators: randomised, double blind, placebo controlled study
pulmonary disease. Thorax 2002;57:847–52. of fluticasone propionate in patients with moderate to severe chronic
31 Kanner RE, Anthonisen NR, Connett JE. Lower respiratory illnesses promote obstructive pulmonary disease; the ISOLDE trial. BMJ
FEV(1) decline in current smokers but not exsmokers with mild chronic 2000;320:1297–303.
obstructive pulmonary disease. Am J Respir Crit Care Med 37 Calverley P, Pauwels R, Vestbo J, et al. Combined salmeterol and fluticasone
2001;164:358–64. in the treatment of chronic obstructive pulmonary disease: a randomised
32 Donaldson GC, Seemungal TAR, Patel IS, et al. Longitudinal changes in the controlled trial. Lancet 2003;361:449–56.
nature, severity and frequency of COPD exacerbations. Eur Respir J 38 Brusasco V, Hodder R, Miravitlles M, et al. Health outcomes following
2003;22:931–6. treatment for six months with once daily tiotropium compared with twice daily
33 Patel IS, Seemungal TAR, Wiks M, et al. Relationship between bacterial salmeterol in patients with COPD. Thorax 2003;58:399–404.
colonisation and the frequency, character, and severity of COPD 39 Soriano JB, Vestbo J, Pride NB, et al. Survival in COPD patients after regular
exacerbations. Thorax 2002;57:759–64. use of fluticasone propionate and salmeterol in general practice. Eur Respir J
34 Wilkinson TA, Patel IS, Wiks M, et al. Airway bacterial load and FEV1 decline 2002;20:819–25.
in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care 40 Vestbo J, and the TORCH Study Group. The TORCH (Towards a Revolution in
Med 2003;167:1090–5. COPD Health) survival study protocol. Eur Respir J 2004;24:206–10.
35 Scanlon PD, Connett JE, Waller LA, et al. Smoking cessation and lung function 41 Soler JJ, Sánchez L, Martı́nez MA, et al. Mid-arm muscle area is a better
in mild-to-moderate chronic obstructive pulmonary disease: the Lung Health predictor of mortality than body mass index in chronic obstructive pulmonary
Study. Am J Respir Crit Care Med 2000;161:381–90. disease. Chest 2005 (in press).

LUNG ALERT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The role of erlotinib in advanced NSCLC
m Shepherd FA, Rodrigues Pereira J, Ciuleanu T, et al. Erlotinib in previously treated non-small-cell lung cancer. N Engl J
Med 2005;353:123–32

m Tsao M-S, Sakurada A, Cutz JC, et al. Erlotinib in lung cancer—molecular and clinical predictors of outcome. N Engl J
Med 2005;353:133–44

R
elapsed advanced non-small cell lung cancer (NSCLC) carries a dismal prognosis and
new treatments are particularly welcome. These papers evaluate the use of erlotinib, a
tyrosine kinase inhibitor of the epidermal growth factor receptor (EGFR) in patients
with NSCLC that has advanced despite prior chemotherapy.
The study was double blind and 731 patients with good performance status who had had
at least one prior course of chemotherapy were randomised to receive either oral erlotinib or
placebo. In the treated group overall survival was improved by 2 months (6.7 months v
4.7 months, p,0.001). Factors predicting response to treatment were being a non-smoker,
adenocarcinoma on histology, and expression of EGFR in biopsy specimens. Survival was
not affected by EGFR receptor expression or EGFR mutations. This is the first study to
demonstrate a survival advantage with an oral tyrosine kinase inhibitor of EGFR. Further
studies investigating the use of these agents in early lung cancer are now warranted.
N Goldsack
Consultant Chest Physician, East Kent Hospitals NHS Trust, UK

www.thoraxjnl.com