Original Article

Chronic Respiratory Disease

10(3) 117–126
Abnormal heart rate recovery and ª The Author(s) 2013
Reprints and permission:
sagepub.co.uk/journalsPermissions.nav
chronotropic incompetence on DOI: 10.1177/1479972313493097
crd.sagepub.com
exercise in chronic obstructive
pulmonary disease

Mansi Gupta1, Vishal Bansal2 and Sunil K Chhabra1

Abstract
Chronotropic incompetence (CI; failure to reach the targeted heart rate (HR) on exercise) and a delayed HR
recovery (HRR; 12 beats decline within the first minute after cessation) reflect autonomic dysfunction (AD)
and predict adverse cardiac prognosis. As chronic obstructive pulmonary disease (COPD) is known to be
associated with AD, we hypothesized that these patients may manifest these responses on exercise. The
prevalence and predictors of these responses in COPD and their association with its severity have not been
evaluated. Normoxemic, stable male patients with COPD (n ¼ 39) and 11 healthy controls underwent lung
function testing and incremental leg ergometry. HR responses were monitored during exercise and recovery
to compute the HRR and CI. Of all the patients, 33 (84.6%) had at least one of the two exercise responses as
abnormal, with the majority (23, 58.9%) having both an abnormal HRR and CI. The frequency of abnormal
responses increased with increasing Global Initiative for Chronic Obstructive Lung Disease stage and body
mass index, airflow obstruction, dyspnoea and exercise capacity index. After adjusting for smoking history and
post-bronchodilator forced expiratory volume in 1 second, only a reduced diffusion capacity for carbon monox-
ide predicted abnormal HRR, though weakly. We concluded that abnormal HRR and CI are common in patients
with COPD. These responses are observed with increasing frequency as the severity of disease increases.

Keywords
Autonomic function, BODE index, chronotropic incompetence, chronic obstructive pulmonary disease, heart
rate recovery

Introduction has been observed in studies on cardiac responses to

Cardiovascular disease is a major cause of mortality
in chronic obstructive pulmonary disease (COPD).1
A probable cause of mortality is autonomic dysfunc-
tion (AD) that may predispose to arrhythmias.2,3
Though there is a lack of direct evidence, AD predicts
fatal cardiac events in diabetes mellitus, post-
myocardial infarction state and heart failure4 and may
play a similar role in COPD.
The rate of recovery of heart rate (HR) within the
first minute after cessation of exercise, termed ‘heart
rate recovery’ (HRR), reflects parasympathetic reacti-
vation and an abnormally delayed (12 beats) decline
implies its dysfunction.5,6 It has been shown to predict
all-cause and cardiac mortality in population studies
and in patients with cardiac disease.6–10 Though it has
not been systematically investigated, a delayed HRR
exercise in COPD.11,12 In a retrospective analysis,
Lacasse et al.13 found abnormal HRR to be associated
with all-cause mortality in COPD. Chronotropic
incompetence (CI), an attenuated HR response to
exercise, is believed to represent an impaired sympa-
thetic response and is another independent predictor
1
Department of Cardiorespiratory Physiology, Vallabhbhai Patel
Chest Institute, University of Delhi, Delhi, India
2
Department of Physiology, Vallabhbhai Patel Chest Institute,
University of Delhi, Delhi, India
Corresponding author:
Sunil K Chhabra, Department of Cardiorespiratory Physiology,
Vallabhbhai Patel Chest Institute, University of Delhi, Delhi
110 007, India.
Email: skchhabra@mailcity.com
118
of cardiac mortality.14–16 It is also associated with
atherosclerosis17 and diastolic heart failure.18 In a
recent retrospective study, it was observed that CI
improved after lung volume reduction surgery in
patients with COPD.19 A prospective study to docu-
ment and characterize CI has not been carried out in
COPD.
Stewart et al.2 demonstrated AD in patients with
hypoxemic COPD, and later, we reported it even in
normoxemic patients and in those with mild COPD.3
As both an abnormal HRR and CI reflect AD, we
hypothesized that an abnormal HRR and CI may
occur in COPD. Being established as independent
predictors of cardiac mortality, these markers of AD
may find application as risk stratification tools in
COPD and therefore need to be characterized. The
prevalence and predictors of these markers in COPD
and their association with severity of the disease have
not been evaluated. With these objectives, the present
study was carried out.
Materials and methods
The study was conducted in an outpatient setting at the
Vallabhbhai Patel Chest Institute (New Delhi, India). It
was approved by the Ethics Committee of Vallabhbhai
Patel Chest Institute (MD Pulmonary Medicine project
3/2008). A total of 39 stable patients with a clinical
diagnosis of COPD based on Global Initiative for
Chronic Obstructive Lung Disease (GOLD) criteria20
were included after a written informed consent.
Subjects
We intended to include patients with COPD due to
their smoking habit. COPD in women in India is more
often due to biomass fuel exposures than due to smok-
ing and considering our hospital attendance patterns, it
was felt that sufficient numbers of female smokers with
COPD may not be found within the study period.
Therefore, we selected only male patients with a his-
tory of smoking of 10 pack-years. Consecutive stable
patients were screened for inclusion in the study.
Patients with chronic respiratory failure (pulse oxime-
try saturation < 90% and partial pressure of oxygen
(PaO2) < 80 mm Hg with or without partial pressure
of carbon dioxide (PaCO2) > 45 mm Hg), recent acute
exacerbation (in the previous 4 weeks), diabetes melli-
tus, evidence of any systemic disease including hyper-
tension or coronary artery disease and any neurological
disorder or musculoskeletal limitation that could pre-
vent successful performance of leg ergometry were
Chronic Respiratory Disease 10(3)
excluded. All the patients had been attending the out-
patient clinic and were on treatment with inhaled
long-acting beta agonists (LABAs, salmeterol/formo-
terol), tiotropium and corticosteroids according to
severity.20 Eleven non-smoker healthy males in the
matched age range and not known to have any chronic
disease or any medication were included as controls
from among the hospital staff and attendants of the
patients.
Initial assessment
After blood counts, routine biochemistry panel, plain
chest radiograph (posteroanterior view) and a 12-lead
standard electrocardiogram, the patients underwent
lung function tests including spirometry and measure-
ment of single-breath diffusion capacity for carbon
monoxide (DLCO) and a 6-min walk test. Dyspnoea was
graded according to the Modified Medical Research
Council scale. The severity of COPD was staged
according to the GOLD criteria.20 Body mass index, air-
flow obstruction, dyspnoea and exercise capacity
(BODE) index was calculated.21 The BODE index
scores each of these components in an ordinal manner
and a total composite score is calculated. As it is not
scored on a ratio/interval scale, we did not use it as a
continuous variable but divided the patients into two
categories, above and below the 50th percentile; these
were labelled as ‘low’ and ‘high’ BODE index groups,
respectively.
Inhaled bronchodilators were withdrawn 24 hours
before investigations but corticosteroids were allowed.
Spirometry was performed on a dry rolling seal spi-
rometer (Benchmark lung function equipment, P.K.
Morgan, Kent, UK) according to current recommenda-
tions.22 DLCO was corrected for alveolar volume to
obtain the Krogh’s constant (KCO). The 6-minute walk
test was performed in accordance with the American
Thoracic Society guidelines.23 The 6-minute walk
distance (6MWD) was used to calculate the volume
of oxygen consumed and the exercise workload to be
applied. The maximum HR (MHR) on ergometry was
calculated (MHR ¼ 220  age). The target HR (THR)
was set at 80% of MHR.
Leg ergometry
After 4 hours post-prandial and a 30-minute rest, a 3-
minute stepwise test of increasing intensity of exer-
cise was performed by leg ergometry on a manually
braked cycle (Rehab Trainer 881E, Monark Exercise
AB, Sweden). It was supervised by the same operator
Gupta et al.
Vishal Bansal. The patients were instructed to pedal at
frequencies of 50–55 cycles/min. In this range, the
cycle delivered a constant workload. This frequency
was maintained throughout the exercise and in the
cooling phase. The predicted maximal oxygen con-
sumption (VO2) was calculated from the 6MWD and
the workload to be applied was obtained from equiva-
lent charts (Monark) available for cycle ergometer
workload estimation. After a 3-minute warm-up period
of unloaded pedalling (0 watts), the workload was
increased to the calculated level and maintained for
3 minutes. If the THR was not achieved, the load was
further increased in steps of 10 watts every 3 minutes
until it was achieved or patient reported fatigue, chest
pain or dyspnoea. At this point, the subject continued
to pedal for 20 seconds to ensure maintenance of a con-
stant HR (17: + 5 beats per minute) in a steady state.
This was followed by the recovery phase, where the
load was removed and the patient continued unloaded
pedalling to cool down for another 3 minutes. HR was
recorded at baseline, throughout the exercise and dur-
ing the cooling-down period on a Nellcor pulse oxi-
meter (Model N 560, Nellcor Puritan Bennett,
Pleasanton, California, USA). The HR recorded repre-
sented an average of last six beats. The HR was read off
from the display at specific time points as required.
HRR was calculated as the difference between the
HR at peak exercise and at first minute into the cool-
ing period. A cut-off point of 12 beats was taken as
an abnormal HRR.6 The utilization of HR reserve at
peak exercise was expressed as the chronotropic
response index (CRI)15 and was calculated as follows:
CRI ¼ ððpeak HR  resting HRÞ  100Þ=ðð220  ageÞ
ðresting HRÞÞ. CRI is independent of age and exercise
capacity, and independently predictive of mortality.15
CI was diagnosed if there was a failure to reach the
THR with a CRI below 80.16,24
Statistical analysis
Statistical analysis was carried out using SPSS 16.0
(SPSS Inc., Chicago, Illinois, USA) and Graph Pad
Prism 4.01 (GraphPad Software, Inc., La Jolla, Califor-
nia, USA) softwares. The data were examined for nor-
mality of distribution by Shapiro–Wilk’s test and for
homogeneity of variance by Levene’s test. Data on
continuous variables are presented as mean + SD. Stu-
dent’s unpaired t test/Mann–Whitney’s U test was used
for independent group comparisons. Comparison of
data on continuous variables among multiple groups
of patients was carried out by Kruskal–Wallis analysis
119
of variance. Chi-square test was used to compare the
association between abnormal HRR and CI. Correlation
analysis was carried out to determine Pearson’s correla-
tion coefficient. Multiple logistic regression was carried
out to identify significant predictors of HRR/CI after
adjusting for other variables. Abnormal/normal HRR
and presence/absence of CI were the dependent binary
variables and age, pack-years of smoking, post-
bronchodilator forced expiratory volume in 1 second
(FEV1) percentage predicted, percentage predicted
DLCO and percentage predicted KCO were the predictor
variables. A two-tailed p value of <0.05 was considered
statistically significant.
Results
The clinical characteristics and baseline laboratory
parameters are shown in Table 1. Patients with severe
COPD had a lower body mass index as compared to
those with mild COPD (p < 0.01), while age, smoking
intensity and duration of disease were similar across
categories by GOLD severity. There was a significant
progressive deterioration of lung function parameters
and BODE index with increasing severity.
The duration of exercise (in seconds) was
444.7 + 173.7 in patients and 637.7 + 229.3 in controls.
The HR response during exercise in patients and controls
is shown in Table 2. The resting HR was not different
(p > 0.05), but the latter achieved a higher peak HR
(p < 0.001). The CRI and HRR were significantly
reduced in patients (p < 0.01 and p < 0.0001, respec-
tively). Only 13 patients could achieve the THR, 6 of the
10 mild and only 2 of the 12 severe COPD patients
(p ¼ 0.05). Majority of the patients (22, 56%) stopped
exercise due to leg fatigue and only four stopped because
of dyspnoea. All the normal subjects reached the THR.
Figure 1 shows the HRR and CRI across GOLD
stages. The HRR in mild COPD was 17.8 + 11.67
beats, 12.7 + 8.86 beats in moderate and 3 + 4.97 beats
in severe COPD (p ¼ 0.09). The CRI in these patients
was 49.3 + 15.89, 46.96 + 13.45 and 37.3 + 10.93
(p ¼ 0.08), respectively. Comparison of patients with
low and high BODE index scores (Figure 2) revealed
that the latter had a lower HRR (15.42 + 10.15 and
9.13 + 5.45 beats, respectively, p < 0.05) and a lower
CRI (49.19 + 14.28 and 37.22 + 10.08, respectively,
p < 0.01). The HRR was significantly correlated with
CRI (r ¼ 0.54, p < 0.001).
Of them, 29 (74.4%) patients but none of the con-
trols had an abnormal HRR (p < 0.0001). CI was found
in 27 (69.2%) patients but in only one control subject
120 Chronic Respiratory Disease 10(3)

Table 1. Clinical characteristics and lung function across GOLD categories.

Stage I, mild Stage II, moderate Stage III, severe
COPD (n ¼ 10) COPD (n ¼ 17) COPD (n ¼ 12)
Age (years)a 57.4 + 9.08 60.88 + 6.24 55.33 + 5.25
BMI (kg/m2)b 24.82 + 5.46 21.05 + 3.58 18.88 + 2.46
Duration of disease (years) 4.0 + 1.63 9.35 + 6.70 9.5 + 5.6
Smoking (pack-years)a 25.70 + 11.99 33.24 + 6.73 30.96 + 10.18
Haemoglobin (g/dl)a 13.61 + 1.47 13.29 + 1.23 13.40 + 1.67
FEV1 percentage predicted (post-bronchodilator)c 94.30 + 10.91 61.41 + 8.60 39.42 + 5.50
Percentage predicted DLCOc 87.70 + 22.89 58.65 + 18.61 55.00 + 28.43
Percentage predicted KCOc 94.70 + 19.55 73.29 + 21.95 69.17 + 33.81
BODE index scorec 0.6 + 0.84 2.41 + 1.33 4.50 + 0.80
GOLD: Global Initiative for Chronic Obstructive Lung Disease; COPD: chronic obstructive pulmonary disease; BMI: body mass index;
DLCO: diffusion capacity for carbon monoxide; KCO: Krogh’s constant; BODE index: body mass index, airflow obstruction, dyspnoea
and exercise capacity index; FEV1: forced expiratory volume in 1 second.
a
p > 0.05 (not significant).
b
p < 0.001.
c
p < 0.0001.

Table 2. HR response during exercise in patients and

controls.
Patients Controls
(n ¼ 39) (n ¼ 11)
Resting HR (bpm)a 78.62 + 11.9 82.00 + 9.07
Peak HR (bpm)b 115.95 + 13.51 139.00 + 7.29
Heart rate recovery 13.0 + 9.1 23.9 + 5.85
(beats)c
Chronotropic response 44.59 + 13.99 61.15 + 5.14
indexd
HR: heart rate; bpm: beats per minute.
a
p > 0.05 (not significant).
b
p < 0.001.
c
p < 0.0001.
d
p < 0.01.
Figure 1. HRR and CRI according to GOLD categories of
severity; p values of HRR ¼ 0.09 and CRI ¼ 0.08. HRR is
expressed in beats in first minute, CRI is expressed in per-
centage. HRR: heart rate recovery; CRI: chronotropic
(p < 0.0001). Majority of the patients (33, 84.6%) had
response index; GOLD: Global Initiative for Chronic
at least one of the autonomic responses as abnormal. Of Obstructive Lung Disease.
the 39 patients, 23 (58.9%) had both abnormal HRR
and CI, 6 (15.4%) had isolated abnormal HRR, and 4
(10.3%) had isolated CI. Only six patients (15.4%) had patients with severe COPD (p < 0.05). Four of the six
no abnormality in the two exercise responses. subjects who had normal HRR and did not have CI
Figures 3 and 4 show the proportions of subjects belonged to the mild group. Abnormal HRR was
with abnormal HRR and CI across GOLD and BODE observed in 16 of 24 (66.7%) patients in the low and
categories, respectively. The frequency of patients 13 of 15 (86.7%) patients in the high BODE index
with abnormal HRR increased with increasing GOLD groups (p > 0.05). CI was significantly more frequent
severity: 5 of 10 (50%) in mild, 13 of 17 (70%) in in the high BODE index group (14 of 15 patients –
moderate and 11 of 12 (92%) in patients with severe 93.3%) compared to 13 of 24 (54%) patients in the
COPD (0.05 > p < 0.1). While 4 of 10 (40%) in mild low BODE index group (p < 0.01). All the six subjects
COPD had CI, the proportion of CI increased to 12 of who had normal HRR and no CI were in the low
17 (70.5%) in moderate and to 11 of 12 (92%) in BODE index group.
Gupta et al.
Figure 2. HRR and CRI in patients with low and high
BODE index scores (þ indicates p < 0.05, * indicates
p < 0.01). HRR is expressed in beats in first minute, CRI
is expressed in percentage. HRR: heart rate recovery;
CRI: chronotropic response index; BODE: body mass
index, airflow obstruction, dyspnoea and exercise
capacity.
Significant inverse correlations were observed
between HRR and age (r ¼ 0.41, p < 0.01) and
pack-years of smoking (r ¼ 0.35, <0.05), while posi-
tive correlations were found between HRR and percent-
age predicted DLCO (r ¼ 0.37, p ¼ 0.05) and percentage
predicted KCO (r ¼ 0.31, p < 0.05). Correlations were
not significant between HRR and post-bronchodilator
FEV1 percentage predicted (p > 0.05). CRI did not show
significant correlations with any of the above variables
(p > 0.05). On multiple logistic regression, KCO remai-
ned a significant, though weak predictor of abnormal
HRR, after adjusting for age, smoking intensity and
post-bronchodilator FEV1. It explained only 22% of the
variability in HRR.
Comparison of clinical characteristics between
patients with abnormal/normal HRR and present/
absent CI is shown in Table 3. Patients with an abnor-
mal HRR were older and had poorer lung function and
lung diffusion parameters as compared to those with a
normal HRR. On the other hand, patients with CI had
higher BODE index scores (p < 0.05) as compared to
those without CI.
Discussion
We found that nearly 85% of normoxemic patients of
COPD have either abnormal HRR or CI or both on
exercise. These tended to occur together and were
observed at all levels of severity, becoming more fre-
quent with higher GOLD stages and significantly
121
increasing with higher BODE index. Patients with
abnormal HRR had greater airways obstruction and
a lower diffusion capacity compared to those with a
normal HRR. After adjusting for age, smoking history
and airways obstruction, only a reduced diffusion
capacity predicted an abnormal HRR, though weakly.
No clinical or lung function parameter predictor of
abnormal CI was identified.
The first minute rapid decline in HR following
exercise-induced tachycardia is largely due to para-
sympathetic restoration and a slower decrease was first
shown to mark an increased risk of mortality by
Schwartz et al.25 This was subsequently confirmed in
other disease states independent of exercise conditions,
peak and resting HR.6–10 Though not specifically
investigated in COPD, an abnormal HRR has been
noted in studies on cardiac responses to exercise in
such patients. Chick et al. observed prolonged tachy-
cardia and delayed recovery after exercise that was
independent of FEV1.11 In a retrospective analysis of
data in a heterogeneous group of patients, including
COPD, Seshadri et al.12 observed that abnormal HRR
was found with increasing frequency as FEV1
decreased. In another retrospective study, Lacasse
et al.13 found an association between abnormal HRR
and increased risk of all-cause mortality among
patients with an FEV1 <50% predicted. Ba et al.26 fol-
lowed the HR decay after maximal cycling exercise to
study metabolic factors affecting the recovery in
patients with COPD and identified greater lactic acid
production and/or hypoxaemia as being associated
with slower recovery. These factors determine the
recovery later on after exercise. They did not study
autonomic mechanisms of recovery that operate almost
exclusively in the first minute. There are no previous
reports of CI in COPD. Multiple definitions and lack
of standardized criteria likely account for the wide
range in reported prevalence of CI in different disease
states.27 The most often used criteria to diagnose CI is a
failure to reach an arbitrary percentage (usually 80%)
of the age-predicted maximal HR or a failure to attain
80% of the HR reserve from rest to peak exercise
(CRI).16,24,27 We defined CI by a strict criteria requir-
ing both these conditions to be met.
Implications for clinical practice and future
research
AD was first reported in advanced COPD with
respiratory failure2 and later abnormalities of HR
variability were also documented in patients without
122 Chronic Respiratory Disease 10(3)

Figure 3. Proportions of subjects with normal and abnormal HRR across GOLD severity and BODE index categories.
HRR: heart rate recovery; BODE: body mass index, airflow obstruction, dyspnoea and exercise capacity; GOLD: Global
Initiative for Chronic Obstructive Lung Disease.

Figure 4. Proportions of subjects in the absence and presence of CI across GOLD severity and BODE index categories.
BODE: body mass index, airflow obstruction, dyspnoea and exercise capacity; CI: chronotropic incompetence; GOLD:
Global Initiative for Chronic Obstructive Lung Disease.

hypoxemia.28 Although Stewart et al.2 found abnorm-
alities largely in the parasympathetic limb, Stein
et al.28 observed a mixed sympathetic and parasympa-
thetic modulation of HR in COPD. From our labora-
tory studies, we reported that both parasympathetic
and sympathetic limbs of the cardiac autonomic
control were affected even in patients without hypox-
aemia and in mild COPD.3 In the present study, the
observation of an abnormal HRR suggests parasym-
pathetic dysfunction, while the findings of CI raise the
possibility of an impaired sympathetic response. The
present study also confirms our earlier observations
that AD develops early in COPD and becomes more
frequent as the disease advances. This is likely to be
a significant complication as cardiac events may be
responsible for mortality even in mild COPD.1,29
CI has been shown to be associated with carotid
intima–media thickness17 and diastolic dysfunction18
in cardiac patients and may thus be a marker of cardi-
ovascular involvement in COPD. An abnormal HRR
and CI may therefore predict a bad cardiac prognosis
in COPD as these do in other diseases.6–10,14–16 This
appears likely as both indices were associated with
increasing BODE index, a well-recognized predictor
Gupta et al. 123

Table 3. Comparison of clinical characteristics between patients with abnormal/normal HRR and present/absent CI.
HRR CI
Abnormal (n ¼ 29) Normal (n ¼ 10) Present (n ¼ 27) Absent (n ¼ 12)
Age 59.62 + 6.30a 54.40 + 8.04 57.70 + 6.35a 59.58 + 8.62
Duration of disease 8.86 + 6.12b 5.60 + 4.43 7.78 + 5.04 8.58 + 7.62
Pack-years 31.11 + 8.79b 29.13 + 12.12 30.84 + 8.68 30.06 + 11.88
Post-bronchodilator FEV1 58.93 + 21.16a 75.10 + 22.66 58.41 + 21.13 73.58 + 22.53
Percentage predicted DLCO 58.55 + 22.25c 83.60 + 29.18 61.30 + 28.64 73.25 + 18.26
Percentage predicted KCO 72.00 + 23.42a 93.50 + 31.66 73.78 + 30.21 85.92 + 16.13
BODE index score 2.90 + 1.78b 1.70 + 1.70 3.00 + 1.88a 1.67 + 1.30
HRR: heart rate recovery; CI: chronotropic incompetence; DLCO: diffusion capacity for carbon monoxide; KCO: Krogh’s constant;
BODE index: body mass index, airflow obstruction, dyspnoea, and exercise capacity index; FEV1: forced expiratory volume in 1 second
a
p < 0.05.
b
p > 0.05 (not significant).
c
p < 0.01.

of mortality in COPD,21 and with increasing GOLD
stage, a classification based on a reduced FEV1, that
is a marker for cardiovascular mortality.30 Therefore,
abnormal HRR and CI may be useful tools for risk stra-
tification in COPD. This needs to be explored in long-
itudinal studies.
Among the commonly measured clinical and physio-
logical parameters, only a reduced diffusion capacity
predicted an abnormal HRR, though weakly, and none
predicted abnormal CI. It suggests that though AD is
common in COPD, it probably develops independently
of pulmonary involvement and may therefore be looked
upon as an extrapulmonary manifestation.
Possible mechanisms for observed cardiac
responses
The underlying mechanisms of impaired cardiac
autonomic control during exercise are not known. Sus-
tained tachypnoea on exercise in patients with COPD11
may impair parasympathetic recovery, a response
known to be enhanced by slower breathing. Reduced
baroreflex sensitivity in COPD is associated with an
impaired sympathetic response.31 On the other hand,
increased sympathetic activity has been shown at rest
in COPD32,33 that may counter parasympathetic
restoration. However, we did not find any difference
in the resting HR in patients and controls, and also
between patients who had and did not have abnormal
HRR and CI arguing against increased sympathetic
activity at rest. Furthermore, Imai et al.34 have also
shown that excessive sympathetic nervous system
activity does not play a dominant role in immediate
HRR. An incomplete washout of anti-cholinergic drugs
is unlikely to explain the delayed parasympathetic
restoration after exercise as these drugs were stopped
24 hour earlier. Whether there is any subsensitivity
of cardiac muscarinic receptors in COPD that may
explain the abnormal HRR is not known. Furthermore,
as anticholinergic drugs will delay HRR by their phar-
macological action, the cardiac safety of these drugs
may be questioned. Indeed, this has been a subject of
several investigations though the results have been
inconclusive.35 Recently, Kawasaki et al.24 suggested
that a post-synaptic downregulation of b-adrenergic
receptors in the sino-atrial node following sympathetic
activation may be responsible for CI in cardiac
patients. We speculate that similar to COPD, regular
use of LABAs may produce a downregulation of b-
receptors resulting in CI.
Strengths and limitations
Abnormal HRR and CI are well-known predictors of
adverse cardiac prognosis. Although in recent years,
there has been an increasing recognition of a signifi-
cant cardiovascular risk in patients with COPD, the
prevalence and predictors of these markers and their
association with severity have not been evaluated in
these patients. This is the first study to investigate these
aspects of COPD. The study has identified markers that
may be useful in risk stratification in COPD. Similar to
their application in patients with coronary artery dis-
ease, abnormal HRR and CI may identify patients with
COPD with increased risk of cardiac mortality. In view
of lack of standard definitions and methods of comput-
ing CI,27 we used a stricter definition as pointed out
above to lend greater assurance to our observations.
124
The study is limited by being gender-specific; the
reasons for non-inclusion of female subjects were
explained above. Smoking itself may be a factor in
causing AD.36 Comparison of responses in patients
with non-smoking COPD is therefore required to
determine the contribution of COPD itself. This was
not part of our study design. It may, however, be
pointed out that all the patients in the present study
were ex-smokers. The only way autonomic balance
can be directly evaluated is by measuring the HR
variability. If we had also measured HR variability
during and after exercise, it would provide direct evi-
dence of autonomic imbalance on exercise. Finally,
the clinical implications are only hypothetical. The
present study was cross-sectional in design and, there-
fore, outcome measures such as mortality or arrhyth-
mias could not be built into the protocol. These need
confirmation in a longitudinal study.
Factors such as exercise levels, medication, cardio-
vascular fitness, neuromuscular limitations and exer-
cise modality may also modulate the chronotropic
response and thus the observed CI may not necessa-
rily represent sympathetic dysfunction. It is important
to take into account the patient’s level of effort and
reasons for terminating the exercise test before con-
cluding that a patient has CI.27 The respiratory
exchange ratio is a more definitive and objective mea-
sure of physiological level of effort and may have pro-
vided a more definite confirmation of maximal
exercise. Symptom-limited exercise also, however,
ensures reasonably that maximal levels are reached.
Thus, it is likely that inadequate metabolic activity
was the cause of a reduced chronotropic response.
Another important factor is the method of exercise.
A treadmill exercise may allow achievement of higher
HRs than a bicycle ergometry.37 Several drugs such as
beta agonists and theophyllines can confound the HR
responses to exercise. However, inhaled beta agonists
were withdrawn 24 hours before the test and none of
the patients were on theophyllines or any other cardi-
oactive drugs. Previous exercise habits and neuromus-
cular limitations may modulate the HR response to
exercise. These were not objectively recorded. Venti-
latory limitation may cause a premature stoppage of
exercise and thus prevent the target HR from being
achieved. It is now well established that dyspnoea
on exercise in patients with COPD is related to
dynamic hyperinflation and is the main reason for ter-
mination of exercise.38 In most of our patients, dys-
pnoea was not the reason for stopping exercise.
Thus, it can be inferred that exercise was not stopped
Chronic Respiratory Disease 10(3)
due to ventilator limitations and lung mechanics
alterations during exercise. In view of the above lim-
itations, our observations on CI may be considered as
preliminary and require confirmation.
Conclusions
To conclude, abnormal HRR and CI are observed dur-
ing exercise in a majority of patients with COPD and
tend to occur together. These occur even in early
stages of the disease and increase in frequency with
increasing GOLD-defined severity and the BODE
index. Clinical and lung function parameters are poor
predictors of these phenomena, with only a reduced
diffusion capacity weakly predicting an abnormal
HRR and no factor predicting CI. Therefore, their
occurrence cannot be inferred and these need to be
experimentally documented. As these indices of AD
are established markers of poor cardiac prognosis,
these measurements may find application as tools for
risk stratification in COPD. The mechanisms underly-
ing the pathogenesis of these abnormalities are not
known and can only be speculated. The long-term sig-
nificance of these abnormal responses needs to be
evaluated in longitudinal studies.
Conflicting of Interests
The authors declared no conflicts of interest.
Funding
This research received no specific grant from any funding
agency in the public, commercial or not-for-profit sectors.
References
1. Hansell AL, Walk JA and Soriano JB. What do chronic
obstructive pulmonary disease patients die from? A
multiple cause coding analysis. Eur Respir J 2003;
22: 809–814.
2. Stewart AG, Waterhouse JC and Howard P. Cardiovas-
cular autonomic nerve function in patients with hypox-
aemic chronic obstructive pulmonary disease. Eur
Respir J 1991; 4: 1207–1214.
3. Chhabra SK and De S. Cardiovascular autonomic neu-
ropathy in chronic obstructive pulmonary disease.
Respir Med 2005; 99: 126–133.
4. Curtis BM and O’Keefe JHJr. Autonomic tone as a car-
diovascular risk factor: the dangers of chronic fight or
flight. Mayo Clin Proc 2002; 77: 45–54.
5. Kannankeril PJ, Le FK, Kadish AH, et al. Parasympa-
thetic effects on heart rate recovery after exercise.
J Investig Med 2004; 52: 394–401.
Gupta et al.
6. Cole C, Blackstone E, Pashkow F, et al. Heart rate
recovery immediately after exercise as a predictor of
mortality. N Engl J Med 1999; 341: 1351–1357.
7. Nishime EO, Cole CR, Blackstone EH, et al. Heart rate
recovery and treadmill exercise score as predictors of
mortality in patients referred for exercise ECG. JAMA
2000; 284: 1392–1398.
8. Chen MS, Blackstone EH, Pothier CE, et al. Heart rate
recovery and impact of myocardial revascularization on
long term mortality. Circulation 2004; 110: 2851–2857.
9. Jouven X, Empana J, Schwartz P, et al. Heart rate pro-
file during exercise as a predictor of sudden death. N
Engl J Med 2005; 352: 1951–1958.
10. Mishra A, Mishra C, Mohanty RR, et al. Heart rate
recovery as predictor of mortality with or without
revascularization. Indian Heart J 2009; 61: 74–79.
11. Chick TW, Cagle TG, Vegas FA, et al. Recovery of gas
exchange variables and heart rate after maximal exer-
cise in COPD. Chest 1990; 97: 276–279.
12. Seshadri N, Gildea TR, McCarthy K, et al. Association
of an abnormal exercise heart rate recovery with pul-
monary function abnormalities. Chest 2004; 125:
1286–1291.
13. Lacasse M, Maltais F, Poirier P, et al. Post-exercise
heart rate recovery and mortality in chronic obstructive
pulmonary disease. Respir Med 2005; 99: 877–886.
14. Lauer MS, Okin PM, Larson MG, et al. Impaired heart
rate response to graded exercise: prognostic implica-
tions of chronotropic incompetence in the Framingham
heart study. Circulation 1996; 93: 1520–1526.
15. Lauer MS, Francis GS, Okin PM, et al. Impaired
chronotropic response to exercise stress testing as a
predictor of mortality. JAMA 1999; 281: 524–529.
16. Jae SY, Fernhall B, Heffernan KS, et al. Chronotropic
response to exercise testing is associated with carotid
atherosclerosis in healthy middle-aged men. Eur Heart
J 2006; 27: 954–959.
17. De Sutter J, Van de Veire N and Elegeert I. Chronotro-
pic incompetence: are the carotid arteries to blame?
Eur Heart J 2006; 27: 897–898.
18. Phan TT, Shivu GN, Abozguia K, et al. Impaired heart
rate recovery and chronotropic incompetence in
patients with heart failure with preserved ejection frac-
tion. Circ Heart Fail 2010; 3: 29–34.
19. Armstrong HF, Gonzalez-Costello J, Jorde UP, et al. The
effect of lung volume reduction surgery on chronotropic
incompetence. Respir Med 2012; 106: 1389–1395.
20. GOLD strategy for diagnosis, management and pre-
vention of chronic obstructive pulmonary disease.
Executive summary update 2009. Available at: www.
goldcopd.org. (2009, accessed 29 October 2010).
125
21. Celli BR, Cote CG, Marin JM, et al. The body-mass
index, airflow obstruction, dyspnoea, and exercise
capacity index in chronic obstructive pulmonary dis-
ease. New Engl J Med 2004; 350: 1005–1012.
22. Miller MR, Hankinson J, Brusasco V, et al. Standardi-
sation of spirometry. Eur Respir J 2005; 26: 319–338.
23. ATS Committee on Proficiency Standards for Clinical
Pulmonary Function Laboratories. ATS Statement:
guidelines for the six minute walk test. Am J Respir
Crit Care Med 2002; 166: 111–117.
24. Kawasaki T, Kaimoto S, Sakatani T, et al. Chronotro-
pic incompetence and autonomic dysfunction in
patients without structural heart disease. Europace
2010; 12: 561–566.
25. Schwartz PJ, La Rovere MT and Vanoli E. Autonomic
nervous system and sudden cardiac death: experimen-
tal basis and clinical observations for post-myocardial
infarction risk stratification. Circulation 1992; 85:
177–191.
26. Ba A, Delliaux S, Bregeon F, et al. Post-exercise heart
rate recovery in healthy, obese, and COPD subjects:
relationships with blood lactic acid and PaO2 levels.
Clin Res Cardiol 2009; 98: 52–58.
27. Brubaker PH and Kitzman DW. Chronotropic incom-
petence: causes, consequences, and management. Cir-
culation 2011; 123: 1010–1020.
28. Stein PK, Nelson P, Rottman JN, et al. Heart rate
variability reflects severity of COPD in PiZ
alpha1-antitrypsin deficiency. Chest 1998; 113:
327–333.
29. Sin DD, Anthonisen NR, Soriano JB, et al. Mortality in
COPD: role of co-morbidities. Eur Respir J 2006; 28:
1245–1257.
30. Sin DD, Wu L and Man SF. The relationship between
reduced lung function and cardiovascular mortality: a
population-based study and a systematic review of the
literature. Chest 2005; 127: 1952–1959.
31. Costes F, Roche F, Pichot V, et al. Influence of exer-
cise training on cardiac baroreflex sensitivity in
patients with COPD. Eur Respir J 2004; 23: 396–401.
32. Sakamaki F, Satoh T, Nagaya N, et al. Abnormality of
left ventricular sympathetic nervous function assessed
by (123)I-metaiodobenzylguanidine imaging in patients
with COPD. Chest 1999; 116: 1575–1581.
33. Heindl S, Lehnert M, Criee CP, et al. Marked sympa-
thetic activation in patients with chronic respiratory fail-
ure. Am J Respir Crit Care Med 2001; 164: 597–601.
34. Imai K, Sato H, Hori M, et al. Vagally mediated heart
rate recovery after exercise is accelerated in athletes
but blunted in patients with chronic heart failure. J
Am Coll Cardiol 1994; 24: 1529–1535.
126
35. Hilleman DE, Malesker MA, Morrow LE, et al. A sys-
tematic review of the cardiovascular risk of inhaled
anticholinergics in patients with COPD. Int J Chron
Obstruct Pulmon Dis 2009; 4: 253–263.
36. Kotamaki M. Smoking induced differences in auto-
nomic responses in military pilot candidates. Clin
Autonomic Res 1995; 5: 31–36.
Chronic Respiratory Disease 10(3)
37. Hsia D, Casaburi R, Pradhan A, et al. Physiological
responses to linear treadmill and cycle ergometer exer-
cise in COPD. Eur Respir J 2009; 34: 605–615.
38. O’Donnell DE, Banzett RB, Carrieri-Kohlman V, et al.
Pathophysiology of dyspnea in chronic obstructive
pulmonary disease: a roundtable. Proc Am Thorac Soc
2007; 4: 145–168.
Copyright of Chronic Respiratory Disease is the property of Sage Publications, Ltd. and its
content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email
articles for individual use.