Characterization of a Novel Symptom

of Advanced Heart Failure: Bendopnea

Jennifer T. Thibodeau, MD, MSC, Aslan T. Turer, MD, MHS, Sarah K. Gualano, MD,
Colby R. Ayers, MS, Mariella Velez-Martinez, MD, Joseph D. Mishkin, MD, Parag C. Patel, MD,
Pradeep P. A. Mammen, MD, David W. Markham, MD, MSC, Benjamin D. Levine, MD,
Mark H. Drazner, MD, MSC
Dallas, Texas
Objectives This study sought to examine the frequency and hemodynamic correlates of shortness of breath when bending
forward, a symptom we have termed bendopnea.
Background Many heart failure patients describe bendopnea such as when putting on their shoes. This symptom has not
previously been characterized.
Methods We conducted a prospective study of 102 subjects with systolic heart failure referred for right-heart catheterization.
Time to onset of bendopnea was measured prior to catheterization. Forty-six subjects also underwent hemodynamic
assessment when sitting and bending. Hemodynamic proles were assigned on the basis of whether pulmonary
capillary wedge pressure (PCWP) was 22 mm Hg and cardiac index (CI) was 2.2 l/min/m
2
.
Results Bendopnea was present in 29 of 102 (28%) subjects with median (25th, 75th percentiles) time to onset of
8 (7, 11) seconds. Subjects with bendopnea had higher supine right atrial pressure (RAP) (p 0.001) and PCWP
(p 0.0004) than those without bendopnea but similar CI (p 0.2). RAP and PCWP increased comparably
in subjects with and without bendopnea when bending, but CI did not change. In those with, versus without,
bendopnea, there was more than a 3-fold higher frequency of a supine hemodynamic prole consisting of elevated
PCWP with low CI (55% vs. 16%, respectively, p < 0.001) but no association with a prole of elevated PCWP with
normal CI (p 0.95).
Conclusions Bendopnea is mediated via a further increase in lling pressures during bending when lling pressures are
already high, particularly if CI is reduced. Awareness of bendopnea should improve noninvasive assessment of
hemodynamics in subjects with heart failure. (J Am Coll Cardiol HF 2014;2:2431) 2014 by the American
College of Cardiology Foundation
Shortness of breath, a dominant symptom in patients with
heart failure, is subclassied on the basis of the activity that
provokes its onset. Dyspnea with exertion, orthopnea and
paroxysmal nocturnal dyspnea, criteria used to dene heart
failure by the Framingham Heart Study in 1971 (1),
continue to be used in routine clinical practice to classify
breathlessness. Recently, we have noticed that many heart
failure patients describe shortness of breath specically when
bending forward, such as when putting on their shoes or
socks. To our knowledge, this symptom has not previously
been characterized. In order to determine the potential
mechanism and clinical implications of this symptom, we
conducted the present prospective study to examine the
frequency and hemodynamic correlates of shortness of
breath when bending forward, or bendopnea, in heart
failure patients with systolic dysfunction who were referred
for right heart catheterization.
Methods
Study design and subject selection. We conducted a single-
center prospective observational study of a convenience sam-
ple of 102 subjects with systolic heart failure who were
referred for right heart catheterization at the University of
See page 32
From the Division of Cardiology, Department of Internal Medicine, University of
Texas Southwestern Medical Center, Dallas, Texas. Dr. Markham has a nancial
relationship with Thoratec. Dr. Drazner has received support from the James M.
Wooten Chair in Cardiology. All other authors have reported that they have no re-
lationships relevant to the contents of this paper to disclose. This study was presented
in part at the American Heart Association Scientic Sessions as part of the Samuel
Levine Young Clinical Investigator Award session on November 3, 2012.
Manuscript received July 19, 2013; accepted July 25, 2013.
JACC: Heart Failure Vol. 2, No. 1, 2014
2014 by the American College of Cardiology Foundation ISSN 2213-1779/$36.00
Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jchf.2013.07.009
Downloaded From: http://heartfailure.onlinejacc.org/ by Marcos Pariona on 03/29/2014
Texas Southwestern Medical Center for clinical indications
between June 2010 and May 2012. Inpatient or outpatient
subjects 18 years of age or older with systolic heart failure,
dened as a left ventricular ejection fraction (LVEF) 40%
within the previous 3 months, were eligible for enrollment. Sub-
jects were excluded if they had undergone cardiac trans-
plantation or required mechanical circulatory support with
an intra-aortic balloon pump or ventricular assist device. The
study protocol was reviewed and approved by the institutional
review board. All subjects gave written informed consent.
Clinical assessment. A history and physical examination
were completed for each subject by an investigator within
6 h prior to right-heart catheterization. Data collected
included demographic data, medical history, cardiac medi-
cations, etiology of cardiomyopathy, peak oxygen con-
sumption, LVEF determined within 3 months of
enrollment, and assessment of functional status by New
York Heart Association (NYHA) classication system (2).
In order to determine the presence of bendopnea, each
subject sat in a chair and bent forward at the waist as if
putting on their socks or shoes, while an investigator timed
the duration to the onset of shortness of breath, as stated to
the patient tell me when you feel short of breath. While
some patients reported lightheadedness or fullness of the
head, chest, or abdomen, the symptom that we describe as
bendopnea is specically shortness of breath when bending
forward. This maneuver was performed after consent was
obtained but prior to the subject entering the catheterization
suite. The subject was classied as having bendopnea if they
reported shortness of breath within 30 sec of bending. In
addition, subjects were queried regarding the presence of the
following symptoms within 7 days prior to enrollment:
angina, dyspnea on exertion (if present, the number of
blocks before symptom onset), orthopnea, paroxysmal
nocturnal dyspnea, weight gain, lower extremity swelling,
abdominal fullness or pressure, early satiety, nausea or
vomiting, palpitations, syncope, and debrillator ring.
Physical examination included height, weight, and body
mass index (BMI) on the day of enrollment as recorded in
the clinical chart, measurement of waist and hip circum-
ference, estimation of jugular venous pressure (JVP), and
assessment of the presence of a third heart sound, rales, and
lower extremity edema. Laboratory values from the clinical
chart were recorded if they were within 1 week of enrollment
and included hemoglobin, creatinine, and natriuretic peptide
concentration. Estimated glomerular ltration rate was
determined by the Modication of Diet in Renal Disease
equation (3). Due to a change in assay availability at our
institution, B-type natriuretic peptide concentrations were
measured in the rst 54 subjects, whereas subsequent sub-
jects had N-terminal pro-B-type natriuretic peptide con-
centrations measured.
Hemodynamic assessments. SUPINE MEASUREMENTS. Inter-
ventional cardiologists who were blinded to the results of
the clinical assessment and determination of bendopnea
status performed the right-heart catheterization. All
enrolled subjects underwent
routine right-heart catheteriza-
tion while lying in the supine
position. Beat-by-beat arterial
blood pressure measurements
were taken by using nger pho-
toplethysmography (Finapres,
Ohmeda, Englewood, Colorado),
and heart rate was determined
from a 3-lead electrocardiogram.
Dual pressure transducers were
used to allow simultaneous mea-
surement of right atrial pressure
(RAP) and pulmonary capillary
wedge pressure (PCWP). The
mean pressures were used for the
RAP and PCWP. The PCWP
was conrmed by oxygen satura-
tion in two-thirds of subjects and by waveform inspection in
the remaining cases. All pressure measurements were ob-
tained at end expiration. Cardiac output was obtained using
both the thermodilution and the Fick methods. For the Fick
equation, oxygen consumption was estimated according to the
derived formula programmed into the Xper Flex Cardio
(Philips, Andover, Massachusetts) physiomonitoring system
(4) used in our catheterization suite, as follows: VO
2
(ml/
min) 133 ml/min/m
2
body surface area (m
2
), where
body surface area was calculated according to the formula of
Dubois and Dubois (5): [0.007184 weight (kg)
0.425

height (cm)
0.725
]. Thermodilution cardiac output measure-
ments were obtained at least in triplicate and averaged such
that all included values were within 10% of the mean. Cardiac
index (CI) was calculated as the cardiac output indexed to
body surface area.
MEASUREMENTS WHEN SITTING AND BENDING. A subset of
the cohort who consented to additional measurements
(n 46) also underwent hemodynamic assessment when
sitting upright in a chair and when sitting in a chair while
bending forward at the waist. Location of the right atrial
site was determined under uoroscopy at the beginning
of the catheterization procedure, and the corresponding
skin site was marked for leveling purposes. After comple-
tion of the standard supine right-heart catheterization,
subjects were moved to a chair and repeat assessment of
hemodynamics was performed after releveling of the
pressure transducers and at least 2 min of rest. The subjects
then bent forward at the waist as if tying their shoes
and underwent repeat hemodynamic assessment after rel-
eveling of the pressure transducers and at least 1 min of
bending. The presence of bendopnea was also reassessed
at this time.
Denitions. Subjects were classied into 1 of 4 hemody-
namic proles, as advocated by other investigators (6),
based on the basis of invasively-measured left ventricular
lling pressures and CI: prole A, warm and dry:
Abbreviations
and Acronyms
BMI = body mass index
CI = cardiac index
LVEF = left ventricular
ejection fraction
JVP = jugular venous
pressure
NYHA = New York Heart
Association
PCWP = pulmonary capillary
wedge pressure
PVR = pulmonary vascular
resistance
RAP = right atrial pressure
SVR = systemic vascular
resistance
JACC: Heart Failure Vol. 2, No. 1, 2014 Thibodeau et al.
February 2014:2431 Characterization of Bendopnea
25
Downloaded From: http://heartfailure.onlinejacc.org/ by Marcos Pariona on 03/29/2014
CI >2.2 l/min/m
2
, PCWP <22 mm Hg; prole B, warm
and wet: CI >2.2 l/min/m
2
, PCWP 22 mm Hg; prole
C, cold and wet: CI 2.2 l/min/m
2
, PCWP 22 mm Hg;
and prole L, cold and dry: CI 2.2 l/min/m
2
,
PCWP <22 mm Hg.
Statistical analysis. We estimated that 30% of enrolled
subjects would have bendopnea. Given this proportion, a
total of 100 subjects would allow a power of 90% to detect a
5 mm Hg change in RAP and PCWP, assuming a standard
deviation of 7 (7); and a power of over 95% to detect a 0.5
l/min/m
2
change in CI, assuming a standard deviation of
0.6 (7). Descriptive statistics are reported as median (25th,
75th percentiles) or as numbers (%), as appropriate. The
chi-square test (categorical variables) and the Wilcoxon rank
sum test (continuous variables) were used for comparison
of clinical characteristics of the bendopnea and the no-
bendopnea groups. In cases of low cell counts, the Fishers
exact test was used for comparison of the categorical variables.
For the primary outcome of differences in hemodynamic
parameters when supine, the Wilcoxon rank sumtest was used
to compare the median values between subjects with
and without bendopnea. The Wilcoxon rank sum test was
also used to compare the median values of the hemodynamic
parameters of those with and without bendopnea at each
position and to assess comparisons in the change in hemo-
dynamics across supine, sitting, and bending positions. Sta-
tistical signicance was set at 0.05, and all tests were 2-tailed.
Statistical analyses were conducted using SAS version
9.2 software (SAS Institute, Inc., Cary, North Carolina).
Results
Clinical characteristics. Bendopnea was present in 29 of
102 (28%) subjects. In those with bendopnea, the median
(25th, 75th percentiles) time to symptom onset was 8 (7, 11)
seconds. The clinical characteristics of subjects with and
without bendopnea are summarized in Table 1. Most subjects
were white, male, 60 to 65 years of age, and NYHAfunctional
classes III and IV, with an LVEFof approximately 20%. Most
subjects were taking angiotensin-converting enzyme in-
hibitors or angiotensin receptor blockers, beta-blockers, and
diuretics. Subjects with bendopnea were more likely to have
other symptoms of decompensated heart failure, including
angina, dyspnea on exertion, orthopnea, paroxysmal nocturnal
dyspnea, and abdominal fullness or pressure. BMI was higher
in subjects with bendopnea than in those without, but there
were no differences in waist or hip circumferences and waist/
hip ratios between these 2 groups. On physical examination,
the only sign associated with bendopnea was an elevated JVP.
There were no differences between subjects with and without
bendopnea with respect to renal function, hemoglobin, and
natriuretic peptide concentrations.
Supine hemodynamic ndings. Supine hemodynamics for
subjects with and without bendopnea are summarized in
Table 2. Subjects with bendopnea had higher RAP
and PCWP and lower pulmonary artery oxygen saturation
and CI calculated using the Fick equation. There was no
difference in the thermodilution CI between subjects with and
without bendopnea (p 0.17). Additionally, there were no
difference in mean arterial pressure, heart rate, pulmonary
vascular resistance (PVR), or systemic vascular resistance
(SVR) between those with and without bendopnea.
Changes in hemodynamics with position. Of the 46 pa-
tients who underwent repeat hemodynamic assessment when
bending, 16 (35%) had bendopnea and 30 (65%) did not
(Online Table 1). Median RAP in subjects with bendopnea
(compared to those without bendopnea) when sitting was
10 (6, 14) mm Hg versus 4 (2, 9) mm Hg, respectively (p
0.02), and 19 (10, 26) mmHg versus 12 (7, 18) mmHg when
bending, respectively (p 0.07) (Fig. 1). This increase in
RAPwith bending was signicant in both groups (p 0.0001
for both), but there was no difference in the amount of in-
crease between these groups [11 (2, 15) mm Hg vs. 7 (3, 10)
mmHg, respectively, p 0.24]. Subjects with bendopnea had
higher median PCWP when sitting than those without
[27 (19, 30) mm Hg vs. 13 (8, 22) mm Hg, respectively, p
0.003] and also when bending [35 (29, 41) mmHg vs. 22 (16,
35) mm Hg, respectively, p 0.01]. As with RAP, PCWP
increased when bending in subjects with and without bend-
opnea (p 0.0001 for both), but there was no difference in the
amount of increase between these groups (p 0.90). In
contrast to the above ndings, thermodilution CI did not
differ between those with and those without bendopnea
when sitting or bending (p 0.34 and p 0.42, respec-
tively), nor did thermodilution CI change when bending
either in subjects with or without bendopnea (p 0.40).
There were no differences in SVR or PVR between those
with and those without bendopnea at all positions, and there
were no changes in SVR or PVR with position change in
either those with or without bendopnea (p > 0.05 for
all).There was 100% agreement in bendopnea status
assessed at study enrollment and during catheterization.
Mean arterial pressure, heart rate, SVR, PVR, trans-
pulmonary gradient (mean PAP PCWP) and transmural
gradient (PCWP RAP) (810) were similar in subjects
with and without bendopnea at each position and did not
signicantly increase or decrease with position change
(data not shown).
Distribution of hemodynamic proles, stratied by
bendopnea. As shown in Figure 2, subjects with bend-
opnea were predominantly hemodynamic prole C,
whereas subjects without bendopnea were predominantly
proles A and L. To ensure that the increase in frequency
of hemodynamic prole C and not prole B in those
with bendopnea was not due to a higher PCWP in the
former group, we compared the median PCWP values of
these 2 proles. Overall, the median PCWP in subjects
who were prole B and C were similar [24 (22, 26) mm Hg
and 25 (23, 28) mm Hg, respectively, p 0.2]. Addi-
tionally, the median PCWPs were similar in prole B and
C subjects with bendopnea [26 (24, 26) mm Hg and 25
(23, 29) mm Hg, respectively, p 1.0] and without
Thibodeau et al. JACC: Heart Failure Vol. 2, No. 1, 2014
Characterization of Bendopnea February 2014:2431
26
Downloaded From: http://heartfailure.onlinejacc.org/ by Marcos Pariona on 03/29/2014
Table 1 Clinical Characteristics of Subjects With and Without Bendopnea
Characteristic
Subjects With Bendopnea
(n 29)
Subjects Without Bendopnea
(n 73) p Value
Age yrs 58 (50, 65) 64 (54, 68) 0.07
Male 20 (69%) 57 (78%) 0.33
Race 0.90
Caucasian 20 (69%) 52 (71%)
Black 8 (28%) 17 (23%)
Hispanic 1 (3%) 3 (4%)
Medical history
COPD 2 (7%) 8 (11%) 0.72
Atrial brillation 6 (21%) 18 (25%) 0.67
Diabetes 12 (41%) 29 (40%) 0.88
Hypertension 12 (41%) 47 (64%) 0.03
Hyperlipidemia 16 (55%) 43 (59%) 0.73
Smoking 15 (52%) 39 (53%) 0.88
Severe mitral regurgitation 3 (10%) 9 (12%) 0.78
Severe tricuspid regurgitation 2 (7%) 10 (14%) 0.50
Current medications
Angiotensin-converting enzyme inhibitor 15 (52%) 47 (64%) 0.24
Amiodarone 6 (21%) 23 (32%) 0.28
Angiotensin receptor blocker 6 (21%) 7 (10%) 0.19
Aldosterone antagonist 15 (52%) 24 (33%) 0.08
Aspirin 17 (59%) 54 (74%) 0.13
Beta-blocker 26 (90%) 58 (80%) 0.22
Digoxin 15 (52%) 29 (40%) 0.27
Diuretic 23 (79%) 64 (88%) 0.35
HMG-CoA reductase inhibitor 12 (41%) 53 (73%) 0.003
Long-acting nitrate or hydralazine 7 (24%) 21 (29%) 0.64
Inotropic therapy 6 (21%) 8 (11%) 0.21
Ischemic cause of cardiomyopathy 12 (41%) 42 (58%) 0.14
LVEF, % 21 (19, 25) 20 (15, 28) 0.55
Peak oxygen consumption,
ml/kg/min (n 43)
14.5 (11.3, 15.4) 12.8 (12.0, 16.4) 0.88
NYHA class I 0 (0%) 4 (6%) 0.58
NYHA class II 2 (7%) 12 (16%) 0.34
NYHA class III 16 (55%) 48 (66%) 0.32
NYHA class IV 11 (38%) 9 (12%) 0.003
Symptoms in the preceding 7 days of
Angina 13 (45%) 10 (14%) 0.001
Dyspnea on exertion 29 (100%) 58 (80%) 0.005
Number of blocks before dyspnea 0.5 (0.5, 1.0) 1.0 (0.5, 3.0) 0.003
Orthopnea 18 (62%) 21 (29%) 0.002
Paroxysmal nocturnal dyspnea 17 (59%) 17 (23%) 0.001
Weight gain 14 (48%) 15 (21%) 0.001
Lower extremity swelling 8 (28%) 11 (15%) 0.14
Abdominal fullness or pressure 23 (79%) 24 (33%) <0.001
Early satiety 20 (69%) 36 (49%) 0.07
Nausea/vomiting 10 (35%) 14 (19%) 0.10
Palpitations 12 (41%) 13 (18%) 0.01
Syncope 3 (10%) 3 (4%) 0.35
Near syncope 10 (35%) 7 (10%) 0.006
Continued on next page
JACC: Heart Failure Vol. 2, No. 1, 2014 Thibodeau et al.
February 2014:2431 Characterization of Bendopnea
27
Downloaded From: http://heartfailure.onlinejacc.org/ by Marcos Pariona on 03/29/2014
bendopnea [24 (22, 25) mm Hg and 27 (23, 28) mm Hg,
respectively, p 0.1]. The median increase in the PCWP
with bending was 9 (2, 11) mm Hg in prole B subjects and
11 (6, 16) mm Hg in prole C subjects (p 0.28).
Discussion
Our study is the initial characterization of a novel symptom
of heart failure, shortness of breath with bending forward,
which we have termed bendopnea. Bendopnea was present
in a sizable minority (approximately one-third) of the study
cohort, was reproducible when reassessed later the same day,
and was associated with hemodynamic derangements; spe-
cically, elevated right ventricular and left ventricular lling
pressures. Additionally, when subjects were categorized
into 1 of 4 hemodynamic proles on the basis of elevation of
left ventricular lling pressures and adequacy of cardiac
perfusion (6), we found that a prole of high PCWP with a
low CI was signicantly more common in those with
bendopnea. Finally, by measuring ventricular lling
Table 2 Supine Hemodynamic Findings for Subjects With and Without Bendopnea
Characteristic
Subjects With Bendopnea
(n 29)
Subjects Without Bendopnea
(n 73) p Value
RAP, mm Hg 11 (7, 15) 5 (3, 10) 0.001
RVSP, mm Hg 49 (42, 66) 42 (25, 53) 0.004
PASP, mm Hg 48 (42, 62) 42 (26, 52) 0.006
PADP, mm Hg 24 (20, 30) 20 (12, 25) 0.003
Mean PAP, mm Hg 33 (30, 44) 30 (19, 36) 0.003
PCWP, mm Hg 23 (20, 26) 19 (8, 23) 0.0004
Transpulmonary gradient, mm Hg 10 (8, 16) 11 (7, 14) 0.55
Transmural gradient, mm Hg 12 (10, 15) 10 (4, 15) 0.05
PA saturation, % 57 (48, 59) 61 (54, 65) 0.003
CI by Thermodilution method, l/min/m
2
2.1 (1.8, 2.6) 2.3 (2.0, 2.5) 0.17
CI by Fick method, l/min/m
2
2.0 (1.8, 2.2) 2.4 (2.1, 2.6) 0.001
MAP, mm Hg 82 (74, 94) 84 (76, 93) 0.59
Heart rate, beats/min 76 (66, 87) 76 (69, 84) 0.84
PVR by thermodilution method, Woods units 2.4 (1.8, 3.9) 2.7 (1.6, 3.5) 0.41
PVR by Fick method, Woods units 2.4 (1.9, 4.3) 2.2 (1.7, 3.3) 0.19
SVR by thermodilution method, dyne/s/cm
5
1,319 (1,115, 1,555) 1,396 (1,134, 1,615) 0.62
SVR by Fick method, dyne/s/cm
5
1,433 (1,243, 1,624) 1,327 (1,099, 1,558) 0.18
Values are median (25th, 75th percentiles).
CI cardiac index; PADP pulmonary artery diastolic pressure; PA pulmonary artery; PAP pulmonary artery pressure; PASP pulmonary
artery systolic pressure; PCWP pulmonary capillary wedge pressure; PVR pulmonary vascular resistance; RAP right atrial pressure;
RVSP right ventricular systolic pressure; SVR systemic vascular resistance.
Table 1 Continued
Characteristic
Subjects With Bendopnea
(n 29)
Subjects Without Bendopnea
(n 73) p Value
Physical examination ndings
BMI, kg/m
2
30 (28, 34) 27 (24, 31) 0.01
Waist circumference, in 44 (38, 47) 40 (35, 44) 0.17
Hip circumference, in 42 (37, 48) 39 (37, 44) 0.34
Waist/hip ratio 1.02 (0.94, 1.05) 0.98 (0.93, 1.03) 0.29
Jugular venous pressure, cm 10 (7, 14) 7 (5, 9) 0.01
Third heart sound 1 (4%) 1 (1%) 0.50
Rales 5 (17%) 8 (11%) 0.51
Lower extremity or dependent edema 8 (28%) 12 (16%) 0.20
Laboratory values
Hemoglobin level, g/dl 12.8 (11.1, 14.6) 12.8 (11.5, 13.6) 0.60
Creatinine level, mg/dl 1.2 (1.0, 1.6) 1.4 (1.1, 1.7) 0.40
eGFR level, ml/min/1.73 m
2
55 (40, 84) 55 (40, 74) 0.69
BNP level, pg/ml (n 54) 389 (281, 680) 395 (159, 676) 0.68
NT-proBNP level, pg/ml (n 30) 2,719 (1,142, 4,008) 2,536 (1,346, 8,359) 0.56
Values are median (25th, 75th percentiles) or n (%).
BMI body mass index; BNP B-type natriuretic peptide; COPD chronic obstructive pulmonary disease; eGFR estimated glomerular ltration
rate; LVEF left ventricular ejection fraction; NT-proBNP N-terminal pro-B-type natriuretic peptide; NYHA New York Heart Association.
Thibodeau et al. JACC: Heart Failure Vol. 2, No. 1, 2014
Characterization of Bendopnea February 2014:2431
28
Downloaded From: http://heartfailure.onlinejacc.org/ by Marcos Pariona on 03/29/2014
pressures and CI in subjects when they bent forward, we
found that bending led to an increase in ventricular lling
pressures but no reduction in the cardiac index.
Previous studies have assessed the association of heart
failure signs and symptoms with invasively measured he-
modynamics. Orthopnea has been shown to be the symptom
best correlated with elevated left ventricular lling pres-
sures measured in the supine position (6,11,12). In the
ESCAPE (Evaluation Study of Congestive Heart Failure
and Pulmonary Artery Catheterization Effectiveness),
orthopnea within the last week was associated strongly
with very high PCWPs (e.g., 28 to 30 mm Hg) (11). In our
current study, bendopnea (conrmed present on the day of
the right heart catheterization) was also associated with
elevated PCWP when measured in the supine and sitting
positions and when subjects bent forward. Furthermore,
orthopnea was 2-fold more common in those with bend-
opnea, a nding consistent with the hypothesis that both
symptoms have the same underlying pathophysiological
mechanism (i.e., elevated left ventricular lling pressures).
As with symptoms, signs detected on physical examination
also have been shown to be associated with elevated left
ventricular lling pressures. In the ESCAPE trial, elevated
JVP was the only physical examination nding associated
with elevated PCWP (11). This nding is consistent with
the present study where elevated JVP was the only physical
examination nding associated with bendopnea (Table 1).
One of the objectives of the present study was to deter-
mine the pathophysiological basis of bendopnea. First, we
wanted to conrm that bendopnea was not simply related to
body habitus (i.e., increased abdominal girth). Although
BMI was higher in those with bendopnea, we found there
were no differences in waist circumference or waist/hip
Figure 2
Distribution of Hemodynamic Proles,
Stratied by Bendopnea
When comparing subjects with bendopnea to those without bendopnea, there was
an increase in the frequency of prole C in the former (55% vs. 16%, respectively,
p < 0.0001) and a trend toward an increase in the frequency of prole A (p 0.07)
and in prole L (p 0.048) in the latter. There was no signicant difference in the
frequency of prole B (p 0.95) in this comparison. The number of subjects
without bendopnea in prole A was 26, 13 in B, 11 in C, and 21 in D; the number
of subjects with bendopnea in prole A was 5, 5 in B, 16 in C, and 3 in L.
Figure 1
Changes in Hemodynamic Parameters From
Sitting to Bending, Stratied by Bendopnea
(A) Right atrial pressure and (B) pulmonary capillary wedge pressure were higher
while sitting in subjects with bendopnea than in those without bendopnea (n 46;
n 16 with bendopnea, n 30 with no bendopnea; p < 0.05 for all). Pressures
increased signicantly from sitting to bending in those with or without bendopnea
(p 0.0001 for all). However, there was no signicant difference in the amount of
increase between those with and without bendopnea (p 0.24 for RAP, p 0.90
for PCWP). (C) Cardiac index by thermodilution was similar between those with and
without bendopnea either when sitting or bending (p 34 and p 0.42,
respectively) and did not change from sitting to bending in those with (p 0.38) or
without bendopnea (p 0.40).
JACC: Heart Failure Vol. 2, No. 1, 2014 Thibodeau et al.
February 2014:2431 Characterization of Bendopnea
29
Downloaded From: http://heartfailure.onlinejacc.org/ by Marcos Pariona on 03/29/2014
ratio in those with bendopnea, suggesting that increased
abdominal girth was not the primary cause of bendopnea.
Second, we wanted to determine whether bendopnea was
related to elevated ventricular lling pressures or to a
reduction in CI during bending. We found that subjects
with bendopnea had higher ventricular lling pressures than
those without, both in the supine and sitting positions.
Furthermore, there was a signicant increase in both right
and left ventricular lling pressures during bending (Fig. 1),
consistent with an increase in intrathoracic pressure. We
interpret these data to suggest that during bending,
increased intrathoracic pressure leads to a further increase in
ventricular lling pressures, and subjects with bendopnea
(who start with higher lling pressures) are more likely to
reach a threshold pressure necessary to induce shortness of
breath. We hypothesize that bendopnea is likely caused by
elevation in the left-sided lling pressure, or PCWP, as has
been demonstrated with other manifestations of shortness of
breath in patients with heart failure (1315). However,
because right- and left-sided ventricular lling pressures
often track together in patients with chronic heart failure
(1315), we cannot exclude with certainty the fact that
further elevation of RAP also contributed to bendopnea. In
contrast to the above-mentioned changes in lling pressures,
we found that the CI as measured by the thermodilution
method did not change signicantly with bending, sug-
gesting that a fall in CI with bending was not the cause of
bendopnea. Nevertheless, a low CI appears to have some
contributory role in this symptom given that hemodynamic
prole C (elevated PCWP and low CI) but not prole B
(elevated PCWP and adequate CI) was associated with
bendopnea, even though the PCWP among those with
prole C was no higher than those with prole B. However,
we recognize that the number of subjects with prole B was
relatively small in our study (n 18) and that it may have
been underpowered to detect an association with bendopnea.
Clinical implications of bendopnea. Assessing hemody-
namics by history and physical examination is important both
for prognostic and therapeutic purposes. Evidence of
congestion has been shown to portend poor prognosis
(1621). For example, in a retrospective analysis of the
SOLVD (Studies Of Left Ventricular Dysfunction) treat-
ment trial, an elevated JVP was associated with a signicantly
increased risk of all-cause mortality, hospitalization for heart
failure, and death from pump failure (17). Additionally, in
those with advanced heart failure, lack of congestion at hos-
pital discharge (11) or at a follow-up clinic visit after hospital
discharge (22) was associated with improved outcomes. Thus,
the prognostic utility of bendopnea should be assessed. In
particular, bendopnea may have utility for physicians who are
not adept at estimating jugular venous pressure.
Study limitations. The number of subjects enrolled was
relatively small, particularly when assessing hemodynamics
during postural changes. Nevertheless, we were able to
document specic hemodynamic changes that occur with
bending. We conrmed the PCWP with oxygen saturation
in the supine but not sitting and bending positions. We
estimated oxygen consumption rather than directly
measuring it. However, having subjects breathe into a
metabolic cart might have inuenced their perception of
bendopnea. The thermodilution method was used to mea-
sure cardiac output, and this approach can be inaccurate in
low-output states or with severe tricuspid regurgitation or
irregular rhythms (23). However, determination of cardiac
output by thermodilution is used in clinical practice and
forms the basis of routine medical decision making. Addi-
tionally, the thermodilution method was used in the Na-
tional Heart, Lung, and Blood Institute-sponsored
ESCAPE trial (7). Another limitation was that assessment
of other symptoms of heart failure was by subject recall over
the previous week and not determined on the day of their
right-heart catheterization; thus, comparison of other
symptoms of heart failure with bendopnea is limited.
However, this was not the main focus of our study. Some
patients with bendopnea did not have elevated PCWP,
suggesting that factors other than elevated left ventricular
lling pressures can cause this symptom. Nevertheless, our
data show that in many patients, bendopnea is a marker of
an elevated PCWP. We enrolled a convenience sample of
patients that was based on the availability of the investigator
and the cardiac catheterization laboratory to enroll research
patients. However, the convenience sampling was due
neither to the severity of illness nor the presence or absence
of bendopnea. Our study cohort consisted of patients who
were referred for right-heart catheterization, many of whom
had advanced heart failure. As such, additional studies are
needed to determine the prevalence of bendopnea among
the general heart failure population.
Conclusions
Bendopnea, a newly described symptom in patients with
heart failure, is mediated via a further increase in ventricular
lling pressures during bending in subjects whose sitting
ventricular lling pressures are already high, particularly in
patients with low CI. Awareness of this symptom by phy-
sicians should improve their noninvasive assessment of he-
modynamics in patients with heart failure.
Reprint requests and correspondence: Dr. Mark H. Drazner,
University of Texas Southwestern Medical Center, 5323 Harry
Hines Boulevard, Dallas, Texas 75390-9047. E-mail: mark.
drazner@utsouthwestern.edu.
REFERENCES
1. McKee PA, Castelli WP, McNamara PM, Kannel WB. The natural
history of congestive heart failure: the Framingham study. N Engl J
Med 1971;285:14416.
2. Criteria Committee of the New York Heart Association. Nomenclature
and Criteria for Diagnosis of Diseases of the Heart and Great Vessels.
9th edition. Boston, MA: Little, Brown; 1994.
Thibodeau et al. JACC: Heart Failure Vol. 2, No. 1, 2014
Characterization of Bendopnea February 2014:2431
30
Downloaded From: http://heartfailure.onlinejacc.org/ by Marcos Pariona on 03/29/2014
3. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more
accurate method to estimate glomerular ltration rate from serum
creatinine: a new prediction equation. modication of diet in Renal
Disease Study Group. Ann Intern Med 1999;130:46170.
4. Philips Healthcare. Xper Flex Cardio Physiomonitoring System: In-
structions for use. Andover, MA: 2011.
5. Dubois D, Dubois E. Clinical calorimetry. A formula to estimate the
approximate surface area if height and weight be known. Arch Intern
Med 1916:81371.
6. Nohria A, Tsang SW, Fang JC, et al. Clinical assessment identies
hemodynamic proles that predict outcomes in patients admitted with
heart failure. J Am Coll Cardiol 2003;41:1797804.
7. Binanay C, Califf RM, Hasselblad V, et al. Evaluation study of
congestive heart failure and pulmonary artery catheterization effec-
tiveness: the ESCAPE trial. JAMA 2005;294:162533.
8. Tyberg JV, Taichman GC, Smith ER, Douglas NW, Smiseth OA,
Keon WJ. The relationship between pericardial pressure and right atrial
pressure: an intraoperative study. Circulation 1986;73:42832.
9. Belenkie I, Kieser TM, Sas R, Smith ER, Tyberg JV. Evidence for left
ventricular constraint during open heart surgery. Can J Cardiol 2002;
18:9519.
10. Dauterman K, Pak PH, Maughan WL, et al. Contribution of external
forces to left ventricular diastolic pressure. Implications for the clinical
use of the Starling law. Ann Intern Med 1995;122:73742.
11. Drazner MH, Hellkamp AS, Leier CV, et al. Value of clinician
assessment of hemodynamics in advanced heart failure: the ESCAPE
trial. Circ Heart Fail 2008;1:1707.
12. Solomonica A, Burger AJ, Aronson D. Hemodynamic determinants of
dyspnea improvement in acute decompensated heart failure. Circ Heart
Fail 2013;6:5360.
13. Drazner MH, Hamilton MA, Fonarow G, Creaser J, Flavell C,
Stevenson LW. Relationship between right and left-sided lling
pressures in 1000 patients with advanced heart failure. J Heart Lung
Transplant 1999;18:112632.
14. Campbell P, Drazner MH, Kato M, et al. Mismatch of right- and left-
sided lling pressures in chronic heart failure. J Card Fail 2011;17:
5618.
15. Drazner MH, Brown RN, Kaiser PA, et al. Relationship of right- and left-
sided lling pressures in patients with advanced heart failure: a 14-year
multi-institutional analysis. J Heart Lung Transplant 2012;31:6772.
16. Morley D, Brozena SC. Assessing risk by hemodynamic prole in
patients awaiting cardiac transplantation. Am J Cardiol 1994;73:
37983.
17. Drazner MH, Rame JE, Stevenson LW, Dries DL. Prognostic
importance of elevated jugular venous pressure and a third heart sound in
patients with heart failure. N Engl J Med 2001;345:57481.
18. Unverferth DV, Magorien RD, Moeschberger ML, Baker PB,
Fetters JK, Leier CV. Factors inuencing the one-year mortality of
dilated cardiomyopathy. Am J Cardiol 1984;54:14752.
19. Ekman I, Cleland JG, Swedberg K, Charlesworth A, Metra M, Poole-
Wilson PA. Symptoms in patients with heart failure are prognostic
predictors: insights from COMET. J Card Fail 2005;11:28892.
20. Aaronson KD, Schwartz JS, Chen TM, Wong KL, Goin JE,
Mancini DM. Development and prospective validation of a clinical
index to predict survival in ambulatory patients referred for cardiac
transplant evaluation. Circulation 1997;95:26607.
21. Gheorghiade M, Gattis WA, OConnor CM, et al. Effects of tol-
vaptan, a vasopressin antagonist, in patients hospitalized with wors-
ening heart failure: a randomized controlled trial. JAMA 2004;291:
196371.
22. Lucas C, Johnson W, Hamilton MA, et al. Freedom from congestion
predicts good survival despite previous class IV symptoms of heart
failure. Am Heart J 2000;140:8407.
23. Nishimura RA, Carabello BA. Hemodynamics in the cardiac cathe-
terization laboratory of the 21st century. Circulation 2012;125:
213850.
Key Words: dyspnea
-
heart failure
-
hemodynamics.
APPENDIX
For a supplemental table, please see the online version of this paper.
JACC: Heart Failure Vol. 2, No. 1, 2014 Thibodeau et al.
February 2014:2431 Characterization of Bendopnea
31
Downloaded From: http://heartfailure.onlinejacc.org/ by Marcos Pariona on 03/29/2014