Orthopnea and Tidal Expiratory Flow Limitation in Chronic Heart Failure*

Original Research
Orthopnea and Tidal Expiratory Flow
Limitation in Chronic Heart Failure*
Roberto Torchio, MD; Carlo Gulotta, MD; Pietro Greco-Lucchina, MD;
Alberto Perboni, MD; Luigina Avonto, MD; Heberto Ghezzo, MD; and
Joseph Milic-Emili, MD
Background: Tidal expiratory flow limitation (FL) is common in patients with acute left heart
failure and contributes significantly to orthopnea. Whether tidal FL exists in patients with chronic
heart failure (CHF) remains to be determined.
Purpose: To measure tidal FL and respiratory function in CHF patients and their relationships to
Methods: In 20 CHF patients (mean [ⴞ SD] ejection fraction, 23 ⴞ 8%; mean systolic pulmonary
artery pressure [sPAP], 46 ⴞ 18 mm Hg; mean age, 59 ⴞ 11 years) and 20 control subjects who
were matched for age and gender, we assessed FL, Borg score, spirometry, maximal inspiratory
pressure (PImax), mouth occlusion pressure 100 ms after the onset of inspiratory effort (P0.1), and
breathing pattern in both the sitting and supine positions. The Medical Research Council score
and orthopnea score were also determined.
Results: In the sitting position, tidal FL was absent in all patients and healthy subjects. In CHF
patients, PImax was reduced, and ventilation and P0.1/PImax ratio was increased relative to those
of control subjects. In the supine position, 12 CHF patients had FL and 18 CHF patients claimed
orthopnea with a mean Borg score increasing from 0.5 ⴞ 0.7 in the sitting position to 2.7 ⴞ 1.5
in the supine position in CHF patients. In contrast, orthopnea was absent in all control subjects.
The FL patients were older than the non-FL patients (mean age, 63 ⴞ 8 vs 53 ⴞ 12 years,
respectively; p < 0.03). In shifting from the seated to the supine position, the P0.1/PImax ratio and
the effective inspiratory impedance increased more in CHF patients than in control subjects. The
best predictors of orthopnea in CHF patients were sPAP, supine PImax, and the percentage
change in inspiratory capacity (IC) from the seated to the supine position (r2 ⴝ 0.64; p < 0.001).
Conclusions: In sitting CHF patients, tidal FL is absent but is common supine. Supine FL,
together with increased respiratory impedance and decreased inspiratory muscle force, can elicit
orthopnea, whom independent indicators are sPAP, supine PImax and change in IC percentage.
(CHEST 2006; 130:472– 479)
Key words: chronic heart failure; expiratory tidal flow limitation; orthopnea
Abbreviations: AHF ⫽ acute heart failure; CHF ⫽ chronic heart failure; Dlco ⫽ diffusing capacity of the lung for
carbon monoxide; ERV ⫽ expiratory reserve volume; FEF75 ⫽ forced expiratory flow when 75% of FVC has been
exhaled; FL ⫽ flow limitation; FRC ⫽ functional residual capacity; IC ⫽ inspiratory capacity; MRC ⫽ Medical Research Council; NEP ⫽ negative expiratory pressure; P(A-a)O2 ⫽ alveolar-arterial oxygen pressure difference;
PCWP ⫽ pulmonary capillary wedge pressure; PEEPi ⫽ intrinsic positive end-expiratory pressure; Pimax ⫽ maximal
inspiratory pressure; Pimus ⫽ pressure generated from inspiratory muscles; P0.1 ⫽ mouth occlusion pressure 100 ms
after onset of inspiratory effort; Prs ⫽ relaxation pressure; Raw ⫽ airway resistance; RV ⫽ residual volume;
sPAP ⫽ systolic pulmonary artery pressure; TLC ⫽ total lung capacity; V˙o2max ⫽ maximal oxygen uptake;
Vr ⫽ relaxation volume; Vt ⫽ tidal volume
rthopnea is a major complaint of patients with acute
O heart
failure (AHF). Although its nature is multifactorial, recent studies1,2 have shown that tidal expiratory
flow limitation (FL) is common in supine patients with
acute AHF in whom, by imposing an inspiratory threshold
Downloaded From: http://journal.publications.chestnet.org/ on 10/28/2014
load due to dynamic hyperinflation and intrinsic positive
end-expiratory pressure (PEEPi), contributes significantly
to orthopnea. While in the sitting patients with chronic
heart failure (CHF) tidal FL is absent,3 its prevalence in
patient in the supine position has not as yet been assessed.
Original Research
Accordingly, in seated and supine CHF patients
and in age-matched and sex-matched control subjects we have assessed the following: the prevalence
of tidal FL and its association to dyspnea and
orthopnea. In addition, we have measured the maximal inspiratory pressure (Pimax), blood gas levels at
rest, and the control of breathing.
Materials and Methods
The study was carried out on 20 stable ambulatory patients (18
men) with congestive heart failure due to cardiomyopathy (6
postischemic patients) without pleural effusions. No patients had
been hospitalized within the 20 days preceding the study. No
patients were current smokers, but nine patients were exsmokers. All patients received therapy with diuretics. Within a
month prior to our study, the Weber class was determined by
cardiopulmonary exercise testing as follows4: Weber class B, 7
patients (maximal oxygen uptake [V˙o2max], between 16.0 and
19.8 mL/kg/min); Weber class C, 10 patients (V˙o2max, between
11.7 and 15.5 mL/kg/min); and Weber class D, 3 patients
(V˙o2max, between 7.1 and 9.8 mL/kg/min). Heart failure was
defined as symptomatic left ventricular dysfunction, with a left
ejection fraction of ⬍ 0.45 documented by bidimensional echocardiography. Patients were excluded from the study if they had
primary pulmonary, neurologic, or myopathic disease. Echocardiographic ejection fraction, systolic pulmonary artery pressure
(sPAP), and heart diameters were measured within the 2 weeks
preceding our study. Twenty healthy subjects (ie, control subjects) matched for sex and age were also studied. All control
subjects were nonsmokers, but nine patients were ex-smokers
(Table 1). The study was approved by the local ethics committee,
and informed consent was obtained from each subject. In the
subjects in the present study, the closing capacity and gas
exchange were assessed in the sitting position, as previously
Chronic dyspnea was scored using the modified Medical
Research Council (MRC) scale based on six increasing grades (0
to 5).5 Seated and supine dyspnea were measured by a modified
Borg scale, ranking the magnitude from 0 (none) to 10 (maximal).6 Twenty minutes of positioning in the decubitus position
was required before assessing the supine Borg score. Orthopnea
was defined as a worsening of the Borg score with the patient in
the supine position.7 Twelve months after this study, it was
possible to retrieve the clinical data of 13 of the 20 patients.
Three patients had died, and 10 patients were still alive.
*From Fisiopatologia Respiratoria (Drs. Torchio, Gulotta, and
Perboni) and Cardiology (Drs. Greco-Lucchina and Avonto),
Ospedale San Luigi Gonzaga, Orbassano, Turin, Italy; and
Meakins-Christies Laboratories (Drs. Ghezzo and Milic-Emili),
McGill University, Montreal, QC, Canada.
The authors have reported to the ACCP that no significant
conflicts of interest exist with any companies/organizations whose
products or services may be discussed in this article.
Manuscript received December 23, 2005; revision accepted
February 13, 2006.
Reproduction of this article is prohibited without written permission
from the American College of Chest Physicians (www.chestjournal.
Correspondence to: Roberto Torchio, MD, Fisiopatologia Respiratoria, Ospedale S Luigi Gonzaga, I-10043 Orbassano, Torino,
Italy; e-mail: [email protected]
DOI: 10.1378/chest.130.2.472
Downloaded From: http://journal.publications.chestnet.org/ on 10/28/2014
Table 1—Anthropometric Characteristics and Baseline
Respiratory Data for Control Subjects and CHF
Patients Measured in the Seated Position*
Sex, No.
Age, yr
BMI, kg/m2
Nonsmokers, No.
Ex-smokers, No.
Ejection fraction, %
sPAP, mm Hg
FEV1, % predicted
FVC, % predicted
FEV1/FVC, % predicted
FEF75, % predicted
TLC, % predicted
FRC, % predicted
IC, % predicted
ERV, % predicted
RV, % predicted
Dlco, % predicted
P(A-a)O2, kPa
MRC score
Borg score
Control Subjects CHF Patients
(n ⫽ 20)
(n ⫽ 20)
59 ⫾ 11
23 ⫾ 8
105 ⫾ 11
103 ⫾ 11
105 ⫾ 7
78 ⫾ 23
97 ⫾ 7
93 ⫾ 13
103 ⫾ 13
103 ⫾ 13
93 ⫾ 11
94 ⫾ 11
2.7 ⫾ 0.5
59 ⫾ 11
26 ⫾ 3
23 ⫾ 8
46 ⫾ 18
313 ⫾ 61
82 ⫾ 19
82 ⫾ 18
102 ⫾ 7
51 ⫾ 24
81 ⫾ 15
78 ⫾ 12
84 ⫾ 20
56 ⫾ 23
88 ⫾ 17
69 ⫾ 21
4.3 ⫾ 1.2
2.2 ⫾ 1
0.5 ⫾ 0.7
⬍ 0.001
⬍ 0.001
⬍ 0.001
⬍ 0.001
⬍ 0.001
⬍ 0.001
⬍ 0.001
⬍ 0.001
⬍ 0.001
*Values are given as the mean ⫾ SD, unless otherwise indicated.
BMI ⫽ body mass index; LVTDV ⫽ left ventricular telediastolic
volume; NS ⫽ not significant.
Each patient underwent a spirometric, plethysmographic, and
diffusing capacity of the lung for carbon monoxide (Dlco) study
that was performed in the sitting position. Using a plethysmograph (Autobox 2800; SensorMedics; Yorba Linda, CA), airway
resistance (Raw) was measured at a panting frequency of ⬍ 1 Hz.
Spirometric and plethysmographic volumes were assessed according to European Respiratory Society (ERS).8 Dlco was
measured with a water-sealed spirometer (Biomedin; Padua,
Italy) using helium for alveolar volume measurement.8 Predicted
values for Raw were from Peslin et al,9 and those for Dlco from
the ERS.8
Breathing pattern, mouth occlusion pressure 100 ms after the
onset of inspiratory effort (P0.1), and Pimax were measured
(VMAX 229; SensorMedics) as previously described.10 The Pimax
was measured at residual volume (RV) according to American
Thoracic Society/ERS11 with predicted values from Black and
Tidal FL was assessed with the negative expiratory pressure
(NEP) technique.13 A NEP of ⫺5 cm H2O was applied (Direc/
NEP System 200A; Raytech Instruments; Vancouver, BC, Canada) 0.2 s after the onset of expiration. Flow-volume curves
obtained without and with NEP were superimposed as follows:
patients in whom the expiratory flow with NEP was the same as
the reference flow during part of the whole expiration were
considered to have FL.13
Pao2 and Paco2 were also measured with a blood gas analyzer
(ABL 735; Radiometer; Copenhagen, Denmark); the PAo2 that
was used to compute the alveolar-arterial oxygen pressure difference (P[A-a]O2) was estimated with the following equation:
PAo2 ⫽ [(PB ⫺ 47) ⫻ Fio2)] ⫺ Paco2/R
CHEST / 130 / 2 / AUGUST, 2006
where PB is barometric pressure, Fio2 is the fraction of inspired
O2, and R is the respiratory quotient, which was assumed to be
Statistical Analysis
The data are presented as the mean ⫾ SD. Correlation
coefficients were obtained with the Spearman (␳ value) nonparametric test for MRC score and the Pearson test (r value) for all
other parameters. Where appropriate, paired and unpaired Student t tests were used. Statistical analysis was performed using a
statistical software package (SPSS; SPSS Inc; Chicago, IL).
Table 1 shows the anthropometric characteristics and baseline respiratory data of CHF patients
and control subjects. In control subjects, all respiratory data were within normal limits, the MRC
and resting Borg scores were zero, and tidal FL
was absent.
In the CHF patients, the total lung capacity and
its subdivisions were reduced relative to control
subjects, while the FEV1/FVC ratio was within
normal limits. In CHF patients, the functional
residual capacity (FRC) percent predicted correlated negatively with the telediastolic volume of
the left ventricle (r ⫽ ⫺0.52; p ⬍ 0.02) [Fig 1].
None of the CHF patient experienced FL in the
seated position. In the CHF patients, the mean
MRC score amounted to 2.2 ⫾ 1.0 (moderate
breathlessness) and the mean Borg score was
0.5 ⫾ 0.7, with the two scores being weakly correlated (p ⬍ 0.05).
In CHF patients, resting ventilation and P0.1 were
Figure 1. Relationship between left ventricular telediastolic
volume and FRC. % pred ⫽ percent predicted.
Downloaded From: http://journal.publications.chestnet.org/ on 10/28/2014
high with respect to control subjects (Table 2). Since
Pimax was reduced in CHF patients, the mean
P0.1/Pimax ratio was very high relative to that in
control subjects (5.3 ⫾ 4.2% vs 2.5 ⫾ 0.9%, respectively; p ⬍ 0.01). The Pao2 and Paco2 were lower in
CHF patients than in control subjects, while the
P(A-a)o2 was higher (Table 1).
The MRC score was significantly higher in the
CHF patients than in control subjects. This score
was correlated with P0.1/Pimax ratio and respiratory
rate. The mean MRC score and P0.1/Pimax ratio
were the only significantly different parameters
among CHF patients who died within 12 months and
survivors (1.7 ⫾ 0.8 vs 3.3 ⫾ 1.2 [p ⬍ 0.02]; and
4 ⫾ 2 vs 11 ⫾ 8, respectively [p ⬍ 0.02]).
None of the control subjects experienced FL while
in the supine position. In contrast, 12 CHF patients
experienced FL while in the supine position, reflecting the decrease in supine expiratory reserve volume
(ERV) with consequent decrease in expiratory flow
None of the variables studied varied significantly
among the 12 CHF patients who experienced FL
while in the supine position and in the other 8
patients who did not experience FL while in the
supine position. In contrast, there was a significant
difference in mean age between these two groups
(63 ⫾ 8 years vs 53 ⫾ 12 years, respectively;
p ⬍ 0.03) but not in Weber class. The percentage of
the tidal volume that entailed FL also correlated
with age (r ⫽ 0.527; p ⬍ 0.02). It should be noted,
however, that among the 12 FL patients 58% were
ex-smokers, whereas in the non-FL group only 2 of 8
patients (25%) were ex-smokers.
In the supine position, the Borg score increased
markedly in CHF patients (ie, orthopnea7), while in
the control subjects it remained zero (Table 2). This
was associated with increased P0.1 and especially
increased P0.1/Pimax ratio, with the latter reflecting
in part the concurrent decrease in Pimax. The mean
effective inspiratory impedance14 also increased
from 5.3 ⫾ 1.8 to 7.6 ⫾ 2.7 cm H2O/L/s (p ⬍ 0.001),
whereas the slight increase observed in control subjects was not significant.
Supine Pimax was significantly reduced in both
CHF patients (p ⬍ 0.001) and control subjects
(p ⬍ 0.05). Moreover, in CHF patients with FL the
mean Pimax reduction was highly significant
(70 ⫾ 30 to 55 ⫾ 27 cm H2O; p ⬍ 0.001).
The supine Borg score correlated with the change
in inspiratory capacity (⌬IC) [r ⫽ ⫺0.59; p ⬍ 0.01]
(Fig 2), supine Pimax (r ⫽ ⫺0.58; p ⬍ 0.01), sPAP
(r ⫽ 0.53; p ⬍ 0.02) [ Fig 3], P0.1/Pimax ratio
Original Research
Table 2—Cardiac and Respiratory Data for 20 Control Subjects and 20 CHF Patients Measured in the Seated and
Supine Positions*
Control Subjects
(n ⫽ 20)
CHF Patients
(n ⫽ 20)
FEV1/FVC ratio, %
FEF75, L/s
V˙e, L/min
Vt, L
Respiratory rate, breaths/min
Pimax, % predicted
P0.1/ Pimax, %
P0.1/(Vt/tI), cm H2O/L/s
MRC score
⌬Borg score
3.07 ⫾ 0.79
1.06 ⫾ 0.46
4.06 ⫾ 1.1
3.29 ⫾ 0.9
80 ⫾ 5
1.36 ⫾ 0.38
10.2 ⫾ 2.7
0.72 ⫾ 0.18
14.2 ⫾ 4.0
87 ⫾ 23
2.5 ⫾ 0.9
5.5 ⫾ 2.0
3.49 ⫾ 0.74†
0.74 ⫾ 0.32†
4.01 ⫾ 1.1
3.19 ⫾ 0.82
80 ⫾ 9
1.33 ⫾ 0.39
10.4 ⫾ 3.5
0.82 ⫾ 0.17¶
13.2 ⫾ 4.0¶
75 ⫾ 29¶
3.6 ⫾ 1.5†
6.2 ⫾ 1.7
2.50 ⫾ 0.68‡
0.62 ⫾ 0.28㛳
3.16 ⫾ 0.85㛳
2.49 ⫾ 0.64㛳
78 ⫾ 5
0.84 ⫾ 0.32㛳
12.9 ⫾ 5.0‡
0.72 ⫾ 0.23
17.9 ⫾ 5.1‡
69 ⫾ 30‡
5.3 ⫾ 4.2㛳
5.3 ⫾ 1.8
2.2 ⫾ 1.0
0.5 ⫾ 0.8
2.73 ⫾ 0.82§
0.28 ⫾ 0.19†
3.00 ⫾ 0.80¶
2.30 ⫾ 0.62†
75 ⫾ 6§
0.73 ⫾ 0.27¶
11.8 ⫾ 3.0¶
0.68 ⫾ 0.13
17.6 ⫾ 5.8
58 ⫾ 28†
8.2 ⫾ 5.9†
7.6 ⫾ 2.7†
2.7 ⫾ 1.1
*Values are given as the mean ⫾ SD, unless otherwise indicated. P0.1/(VT/tI) ⫽ pulmonary impedance. V˙e ⫽ minute ventilation; NFL ⫽ negative
flow limitation.
†p ⬍ 0.001 (sitting vs supine position; paired t test).
‡p ⬍ 0.05 (CHF patients vs control subjects; unpaired t test).
§p ⬍ 0.005 (sitting vs supine position; paired t test).
㛳p ⬍ 0.005 (CHF patients vs control subjects; unpaired t test).
¶p ⬍ 0.05 (sitting vs supine position; paired t test).
Borg score ⫽ 2.61 ⫹ 0.033 sPAP
(r ⫽ ⫺0.48; p ⬍ 0.003), inspiratory time/total breath
cycle time ratio (r ⫽ ⫺0.50; p ⬍ 0.024), and supine
ERV (r ⫽ ⫺0.461; p ⬍ 0.04).
Stepwise multivariate regression analysis selected
all of the above parameters as significant independent contributors to Borg score, reflecting its multifactorial nature. However, the best predictors were
supine Pimax (cm H2O), sPAP (mm Hg), and ⌬IC,
as follows:
where r ⫽ 0.80, r2 ⫽ 0.64, and p ⬍ 0.001. In FL
patients, the supine Borg score was correlated with
the ⌬IC (r ⫽ ⫺0.67; p ⬍ 0.017), whereas in non-FL
patients this was not the case.
Figure 2. Relationship between supine Borg score and ⌬IC
from the seated to the supine position.
Figure 3. Relationship between supine Borg score and sPAP.
Downloaded From: http://journal.publications.chestnet.org/ on 10/28/2014
⫺ 0.019 Pimax ⫺ 0.042 ⌬IC
CHEST / 130 / 2 / AUGUST, 2006
The new findings of this study are as follows: (1) in
CHF patients, tidal FL is absent while sitting but is
common while in the supine position; (2) the patients who experienced FL in the supine position
were significantly older than those who did not; (3) in
the supine position, almost all patients (18 of 20
patients) exhibited orthopnea, the nature of which
appears to be multifactorial; but its best predictors,
selected by multiple regression analysis, were sPAP,
supine Pimax, and ⌬IC from the seated to the supine
In line with most previous reports,3,15,16 our CHF
patients exhibited a reduction of total lung capacity,
FRC, and RV but had a normal FEV1/FVC ratio.
The FRC reduction was correlated with cardiomegaly, since there was a significant correlation of left
end-diastolic volume to FRC (Fig 1). This finding
could partially explain the difference between our
data and those of Yap et al17 and Hart et al,18 who
found no reduction of FRC in their CHF patients.
Eight of 10 patients in the study by Yap et al17 and 5
of 10 patients in the study by Hart et al18 had CHF
due to coronary artery disease or hypertension,
whereas most of our patients had idiopathic cardiomyopathy.
As previously reported,3,16,19 –21 Pimax was also
reduced. The nature of the inspiratory muscle weakness is multifactorial, including the reduction of
respiratory muscle blood flow, hypoxia, oxidative
stress, disuse, medication, systemic inflammation,
and nutritional depletion. The enlarged chest wall
due to cardiomegaly and increased intrathoracic
blood volume should also contribute to reduced
Pimax because the pressure generated from inspiratory muscles (Pimus) should decrease due to length
tension and geometric factors.22 This uncoupling of
the lung and chest wall with a concurrent decrease in
the relaxation volume (Vr) of the respiratory system,
as reflected by the decreased FRC, should also
decrease the Pimax measured at RV because of the
decreased contribution of the relaxation pressure
(Prs) of the respiratory system to Pimax. In fact, as in
the following equation:
Pimax ⫽ Pimus ⫺ Prs
it follows that at RV Pimax decreases as a result of
both decreased Pimus and Prs. Altered diaphragmatic position can also modify its length-tension
relationships and contribute to reduced Pimax.23
Ventilation and Neuromuscular Inspiratory Drive:
Resting ventilation in CHF patients was high relative
Downloaded From: http://journal.publications.chestnet.org/ on 10/28/2014
to control subjects, resulting in reduced mean Paco2
(4.9 ⫾ 0.4 mm Hg vs 5.2 ⫾ 0.3 mm Hg, respectively;
p ⬍ 0.03). In patients with reduced Pimax, P0.1
underestimates the inspiratory neural drive. Accordingly, the P0.1/Pimax ratio is used instead as an index
of neuromuscular inspiratory drive normalized for
muscle strength. In agreement with most previous
studies,3,16,19,20 but not all,21 we found an increase in
P0.1/Pimax ratio in CHF patients relative to control
subjects (5.3 ⫾ 4.2 vs 2.5 ⫾ 0.9, respectively;
p ⬍ 0.01). Bruschi et al24 found increased P0.1/Pimax
ratio in CHF patients who are at risk for nocturnal
Cheynes-Stokes respiration and showed that this
parameter, together with respiratory frequency, is
a useful predictor of Cheynes-Stokes respiration.
Reduced levels of Paco2 and increased pH were
previously found in patients with high pulmonary
capillary wedge pressure (PCWP) because the stimulation of pulmonary receptors, as a result of raised
PCWP, heightens the central neural drive.25 However, we found no significant correlation of sPAP to
P0.1 or P0.1/Pimax ratio. This may be due to the fact
that most of our patients had elevated levels of sPAP,
which may mask any correlation between sPAP and
other variables. Recently, Meyer et al26 found respiratory muscle weakness in patients with idiopathic
pulmonary hypertension and found P0.1 to be significantly higher than in those patients than in control
subjects. In their patients, the P0.1/Pimax ratio
showed significant correlation with the total ventilation-carbon dioxide slope during exercise. This observation confirms that increased ventilation and
increased respiratory drive are hallmarks of heart
failure. Hyperventilation during exercise, whether or
not it was associated with periodic breathing, is now
considered to be an independent risk factor for
mortality and morbidity in CHF patients.27
Tidal Flow Limitation: In the study by Duguet et
al,1 9 of 12 patients with acute left ventricular failure
exhibited tidal expiratory FL when in the supine
position, and all of these 9 patients claimed to have
orthopnea. By contrast, only one of the three subjects who denied having orthopnea exhibited tidal
FL while in the supine position. Recently, Boni et al2
showed that in patients with both AHF and CHF
supine FL is relatively common and that it can be
reversed by therapy with diuretics. Age is a major
factor promoting FL, because with increasing age
the maximal expiratory flows at low lung volume
decrease progressively due to gas trapping.28 De
Bischopp et al29 recently studied apparently healthy
subjects aged from 66 to 88 years and found that
Original Research
⬎ 30% had FL even in the seated position. The
patients studied by Duguet et al1 and Boni et al2
were quite old (mean age, 78 ⫾ 15 years and 77 ⫹ 7
years, respectively).
A reduction in expiratory flow reserve plays a
pivotal role in determining tidal flow limitation.
Similarly, a reduction of the ERV plays an important
role. In fact, when a subject breathes tidally at low
volumes (ie, with a low ERV), the tidal volume
changes occur over a portion of the maximal flowvolume curve with low maximal flows. Since the
ERV is normally reduced while in the supine position,22 the prevalence of tidal FL in the supine
position is generally higher than that in the sitting
Orthopnea: Tidal FL in the supine position is
associated with orthopnea in COPD patients,7 obese
patients,30 and goiter patients.31 Almost all of our
CHF patients (18 of 20 patients) claimed to have
orthopnea,7 the critical trigger for which is commonly attributed to increased pulmonary venous
return while in the supine position. This is supported
by the significant correlation between PCWP and
orthopnea.32 We found a significant correlation between supine Borg score and sPAP (r ⫽ 0.53;
p ⬍ 0.02) [Fig 3]. Furthermore, the two CHF patients who denied orthopnea were younger (48 and
57 years) and had relatively low values of sPAP
(25 mm Hg and 35 mm Hg). The several following
mechanisms25,33,34 have been suggested to link the
increase in central blood volume in CHF patients to
orthopnea: (1) stimulation of the juxtacapillary receptors and increased vagal afferentation due to
congestion and edema; (2) increased inspiratory
loading; and (3) impaired Pimax. Increased vagal
afferentation may lead to hyperventilation and cause
breathlessness. Minute ventilation was higher than
that in control subjects, both in the sitting and supine
position, and the P0.1/Pimax ratio increased significantly in the supine position in both CHF patients
and control subjects (p ⬍ 0.001), but in CHF patients the mean increase was much more pronounced (62 ⫾ 50% vs 44 ⫾ 28%, respectively). The
postural increase in respiratory impedance
(45 ⫾ 33% vs 20 ⫾ 35%, respectively; p ⬍ 0.03) was
also higher in CHF patients. This implies increased
inspiratory loading and could be due in part to the
increased Raw and respiratory system elastance
caused by the reduction in FRC.
In our 12 patients who had FL while in the supine
position, the FEV1/FVC ratio (p ⬍ 0.001), forced
expiratory flow when 75% of FVC has been exhaled
(FEF75) [p ⬍ 0.05], and Pimax (p ⬍ 0.001) decreased significantly when shifting from the sitting to
Downloaded From: http://journal.publications.chestnet.org/ on 10/28/2014
the supine position, while such changes were not
significant in the 8 non-FL patients.
The increase in mean Borg score with posture was
higher in these subjects with respect to non-FL
patients (from 0.5 ⫾ 0.7 to 3.0 ⫾ 1.4 vs 0.5 ⫾ 0.8 to
2.1 ⫾ 1.5, respectively), and the supine Borg score
was negatively correlated with the ⌬IC, as previously
described.2 In our study, ⌬IC also correlated with
left ventricular telediastolic volume (r ⫽ ⫺0.457;
p ⬍ 0.05). In adopting the supine position, there is
normally little or no change in vital capacity, while
ERV is substantially reduced with a concurrent
increase of IC.22 The latter is caused by a reduction
of Vr due to gravity.22 In our CHF patients, the
supine ERV was very small (ie, close to RV), probably due to cardiomegaly and increased intrathoracic
liquid volume.31 In our patients, the supine Borg
score correlated with supine ERV (r ⫽ ⫺0.46;
p ⬍ 0.04). Such an association is also found in
patients with morbid obesity,30 and it has been
suggested that, in these patients, the supine Vr is
located below RV. This implies the presence of
PEEPi at RV. This phenomenon was probably also
present in our FL subjects and may explain the
paradoxical reduction of Pimax observed in these
patients (see below). Tidal FL seems to be a factor
contributing to orthopnea in CHF patients, but the
finding that supine Borg score correlated also with
sPAP and supine Pimax suggest a multifactorial
origin for this symptom.
In a previous work,3 sPAP was correlated to
P(A-a)O2 and to the closing volume/vital capacity
ratio. In most of the patients (13 of 20 patients), CC
exceeded FRC (ie, during tidal breathing there was a
cyclic opening and closing of peripheral airways with
a concurrent maldistribution of ventilation, with
decreased Pao2 and increased P[A-a]O2). Supine
positioning that reduces FRC could enhance this
phenomenon, contributing to the worsening of orthopnea.
Inspiratory Load and Muscular Force: The overall
increase in inspiratory load is reflected by the significant mean increase in effective inspiratory impedance with change in posture in CHF patients compared to that in control subjects (7.6 ⫾ 2.7 vs
6.2 ⫾ 1.7, respectively; p ⫽ 0.05) indicating that P0.1
was sufficient not only to overcome the increased
breathing load but also to induce hyperventilation.
The decrease in Pimax in the supine position
contributed to the disproportionate increase in P0.1/
Pimax ratio, reflecting increased effort in the face of
weakened muscles with a concurrent increase in
Borg score (Table 2). In fact, Pimax in the supine
position together with sPAP and ⌬IC were the most
significant independent contributors to Borg score
CHEST / 130 / 2 / AUGUST, 2006
(equation 1). In FL patients, the decreases in Pimax
and the increases in Borg score with change in
posture were greater than in non-FL subjects.
In stable CHF patients, Nava et al35 found a
strong correlation between orthopnea and increased
diaphragmatic effort in the supine position. The
reduction of diaphragmatic effort through assisted
noninvasive ventilation correlated significantly with
the reduction in orthopnea. Tidal FL was not assessed, but dynamic PEEPi was increased in the
supine position with a concurrent increase in resistance and transdiaphragmatic pressure.
The effect of posture on Pimax in healthy subjects
is controversial. In two previous studies,22,36 no
differences between Pimax measured in the upright
and supine position were found, while in another
study37 Pimax was significantly lower in the supine
position. There is no study comparing Pimax in the
upright and supine positions in CHF patients. We
found a slight reduction in Pimax in control subjects
and in CHF patients. The percentage change in
posture was similar in the two groups. The reduction
in mean Pimax was significantly higher in the 12 FL
subjects compare to non-FL subjects (⫺24 ⫹ 16 vs
⫺6 ⫹ 18%, respectively). This was probably due to
the fact that in the supine patient Vr was below RV,
and hence PEEPi was present at RV. Since PEEPi
corresponds to positive Prs (equation 2), Pimax
should be less than Pimus in the supine position. In
contrast, Prs is normally negative at RV, and hence
Pimax is higher than Pimus. It should be noted that
altered diaphragmatic position23 and chest wall geometry22 may also contribute to changes in Pimax
with posture.
In conclusion, the present data show that, as a
result of a marked decrease in ERV in the supine
position, most CHF patients also exhibit tidal FL
when supine with concurrent inspiratory threshold
loading. This, together with increased respiratory
resistance and elastance, and decreased inspiratory
muscle force, elicits orthopnea.
1 Duguet A, Tantucci C, Lozinguez O, et al. Expiratory flow
limitation as a determinant of orthopnea in acute left heart
failure. J Am Coll Cardiol 2000; 35:690 –700
2 Boni E, Bezzi M, Carminati M, et al. Expiratory flow
limitation is associated with orthopnea and reversed by
vasodilators and diuretics in left heart failure. Chest 2005;
128:1050 –1057
3 Torchio R, Gulotta C, Greco-Lucchina P, et al. Closing
capacity and gas exchange in chronic heart failure. Chest 129;
5:1330 –1336
4 Weber KT, Kinasewitz GT, Janicki JS. Oxygen utilization and
ventilation during exercise in patients with chronic cardiac
failure. Circulation 1982; 65:1213–1223
5 Bestall JC, Paul EA, Garrod R, et al. Usefulness of the
medical research council (MRC) dyspnea scale as a measure
Downloaded From: http://journal.publications.chestnet.org/ on 10/28/2014
of disability in patients with chronic obstructive lung disease.
Thorax 1999; 54:581–586
Borg G. Psychophysical basis of perceived exertion. Med Sci
Sports Exerc 1982; 14:377–381
Eltayara L, Ghezzo H, Milic-Emili J. Orthopnea and tidal
expiratory flow limitation in patients with stable COPD.
Chest 2001; 119:99 –104
Quanjer PhH, Tammeling JE, Cotes OF, et al. Lung volumes
and forced ventilatory flows: Report Working Party Standardization of Lung Function Tests; European Coal and Steel
Community. Eur Respir J 1993; 16:5– 40
Peslin R. Resistance. In: Milic-Emili J, Lucangelo U, Pesenti
A, et al, eds. Basics of respiratory mechanics and artificial
ventilation. Milano, Italy: Springer-Verlag, 1999; 37–57
Whitelaw WA, Derenne JP, Milic-Emili J. Occlusion pressure
as a measure of respiratory center output in conscious man.
Respir Physiol 1975; 23:181–199
American Thoracic Society/European Respiratory Society.
ATS/ERS statement on respiratory muscle testing. Am J
Respir Crit Care Med 2002; 166:518 – 624
Black LF, Hyatt RE. Maximal respiratory pressures: normal
values and relationship to age and sex. Am Rev Respir Dis
1969; 99:696 –702
Koulouris NG, Valta P, Lavoie A, et al. A simple method to
detect expiratory flow limitation during spontaneous breathing. Eur Respir J 1995; 8:306 –313
Hussain SNA, Pardy RL, Dempsey JA. Mechanical impedance as determinant of inspiratory neural drive during exercise in humans. J Appl Physiol 1985; 59:365–375
Collins JV, Clark TJK, Brown J. Airway function in healthy
subjects and patients with left heart disease. Clin Sci Mol
Med 1975; 49:217–228
Ambrosino, Opasich C, Crotti P, et al. Breathing pattern,
ventilatory drive and respiratory muscle strength in patients
with chronic heart failure. Eur Respir J 1994; 7:17–22
Yap JCH, Moore DM, Cleland JGF, et al. Effect of supine
posture on respiratory mechanics in chronic left ventricular
failure. Am J Respir Crit Care Med 2000; 162:1285–1291
Hart N, Kearney MT, Pride NB, et al. Inspiratory muscle load
and capacity in chronic heart failure. Thorax 2004; 59:477–
Meyer FJ, Borst MM, Zugck C, et al. Respiratory muscle
dysfunction in congestive heart failure: clinical correlation
and prognostic significance. Circulation 2001; 103:2153–2158
McParland C, Krishnan B, Wang Y, et al. Inspiratory muscle
weakness and dyspnoea in chronic heart failure. Am Rev
Respir Dis 1992; 148:467– 472
Witt C, Borges AC, Haake H, et al. Respiratory muscle
weakness and normal ventilatory drive in dilatative cardiomyopathy. Eur Heart J 1997; 18:1322–1328
Agostoni E, Hyatt RE. Static behavior of the respiratory
system. In: Handbook of physiology: the respiratory system;
mechanics of breathing. Bethesda MD: American Physiological Society, 1986; 113–130
Caruana L, Petrie MC, McMurray JJ, et al. Altered diaphragm position and function in patients with chronic heart
failure. Eur J Heart Fail 2000; 3:183–187
Bruschi C, Fanfulla F, Traversi E, et al. Identification of
chronic heart failure patients at risk of Cheyne-Stokes respiration. Monaldi Arch Chest Dis 1999; 54:319 –324
Lorenzi-Filho G, Azevedo ER, Parker JD, et al. Relationship
of carbon dioxide tension in arterial blood to pulmonary
wedge pressure in heart failure. Eur Respir J 2002; 19:37– 40
Meyer FJ, Lossnitzer D, Kristen AV, et al. Respiratory muscle
dysfunction in idiopathic pulmonary arterial hypertension.
Eur Respir J 2005; 25:125–130
Original Research
27 Bradley TD, Floras JS. Sleep apnea and heart failure: part II.
Central sleep apnea. Circulation 2003; 107:1822–1826
28 Knudson RJ. Physiology of the aging lung. In: Crystal RG,
West JB, Weibel FR, et al, eds. The lung: scientific foundations. New York, NY: Raven Press Ltd, 1991; 1749
29 De Bisschop C, Marty ML, Tessier JF, et al. Expiratory flow
limitation and obstruction in the elderly. Eur Respir J 2005;
26:594 – 601
30 Ferretti A, Giampiccolo P, Cavalli A, et al. Expiratory flow
limitation and orthopnea in massively obese subjects. Chest
2001; 119:1401–1408
31 Torchio R, Gulotta C, Perboni A, et al. Orthopnea and tidal
expiratory flow limitation in patients with euthyroid goiter.
Chest 2003; 124:133–140
32 Mancini DM, La Manca J, Donchez L, et al. The sensation of
dyspnea during exercise is not determined by the work of
Downloaded From: http://journal.publications.chestnet.org/ on 10/28/2014
breathing in patients with heart failure. J Am Coll Cardiol
1996; 28:391–395
Paintal A. Mechanism of stimulation of type J pulmonary
receptors. J Physiol 1969; 203:511–532
Bradley TD, Holloway RM, McLaughlin PR, et al. Cardiac
output response to continuous positive airway pressure in congestive heart failure. Am Rev Respir Dis 1992; 145:377–382
Nava S, Larovere MT, Fanfulla F, et al. Orthopnea and
inspiratory effort in chronic heart failure patients. Respir Med
2003; 97:647– 653
Yap YCH, Watson RA, Gilbey S, et al. Effects of posture on
respiratory mechanics in obesity. J Appl Physiol 1995; 79:
1199 –1205
Koulouris N, Mulvey DA, Laroche CM, et al. The effect of
posture and abdominal binding on respiratory pressures. Eur
Respir J 1989; 2:961–965
CHEST / 130 / 2 / AUGUST, 2006