Hemodynamic Determinants of Dyspnea Improvement in

Acute Decompensated Heart Failure

Amir Solomonica, MD, MPH; Andrew J. Burger, MD; Doron Aronson, MD

Background—Dyspnea relief constitutes a major treatment goal and a key measure of treatment efficacy in decompensated
heart failure. However, there are no data with regard to the relationship between hemodynamic measurements during
treatment and dyspnea improvement.
Methods and Results—We studied 233 patients assigned to right heart catheterization in the Vasodilation in the Management
of Acute Congestive Heart Failure trial. Dyspnea (assessed using a 7-point Likert scale) and hemodynamic parameters
were measured simultaneously at 15 and 30 minutes and 1, 2, 3, 6, and 24 hours. Dyspnea relief was defined as moderate
or marked improvement. There was a time-dependent association between the reductions in pulmonary capillary wedge
pressure (PCWP; 25.4, 24.6, 24.0, 23.5, 23.4, 21.5, and 19.9 mm Hg) and the percentage of patients achieving dyspnea
relief (17.7%, 24.6%, 32.2%, 36.2%, 37.8%, 47.4%, and 66.1%, in the respective time points). Multivariable logistic
generalized estimating equations modeling demonstrated that reductions of both PCWP and mean pulmonary artery
pressure were independently associated with dyspnea relief. Compared with the highest PCWP quartile, the adjusted odds
ratios for dyspnea relief were 0.92 (95% confidence interval [CI], 0.67–1.29), 1.07 (95% CI, 0.75–1.55), and 1.80 (95%
CI, 1.22–2.65) in the third, second, and first PCWP quartiles, respectively (Ptrend=0.003). Compared with the highest
mean pulmonary artery pressure quartile, the adjusted odds ratios for dyspnea relief were 2.0 (95% CI, 1.41–2.82),
2.23 (95% CI, 1.52–3.27), and 2.98 (95% CI, 1.91–4.66) in the third, second, and first mean pulmonary artery pressure
quartiles, respectively (Ptrend<0.0001).
Conclusions—A clinically significant improvement in dyspnea is associated with a reduction in both PCWP and mean
pulmonary artery pressure.  (Circ Heart Fail. 2013;6:53-60.)
Key Words: acute heart failure ◼ dyspnea ◼ pulmonary capillary wedge pressure ◼ pulmonary hypertension
Downloaded from http://ahajournals.org by on November 18, 2018

I n the acute care setting, dyspnea is the most commonly

reported symptom among patients with worsening heart
failure and is the principal cause of hospitalization for patients
with acute decompensated heart failure (ADHF).1,2 Relief
of dyspnea is one of the most important standards by which
efficacy of therapy is ascertained and is a major end point in
clinical trials of investigational agents for the management
of patients with ADHF.3–5 A number of studies and position
articles emphasize the importance of achieving rapid improve-
ment in dyspnea.6,7 However, recent studies have shown that
many patients have only partial relief of dyspnea and conges-
tion during admission, even when guideline-recommended
therapies are fully implemented.8–13 In addition, persistent
dyspnea is associated with higher risk for mortality.11 There
are no clinical characteristics or laboratory measurements
(including B-type natriuretic peptide levels) that are reliably
associated with dyspnea relief.9–11
Clinical Perspective on p 60
The expectation for the improvement of dyspnea with
current vasodilator and diuretic therapy is based on the
assumption that reductions in filling pressures are likely to
translate into a reduction in pulmonary venous congestion,
leading to symptomatic benefit. However, there are no data
with regard to changes in specific hemodynamic parameters
during treatment of ADHF and their relation to dyspnea
improvement. Hence, the relationship between dyspnea and
objective hemodynamic measures remains unclear.
In the present study, we sought to determine the hemody­
namic profile associated with a clinically significant
improvement in dyspnea among patients admitted for ADHF.
To this aim, we used hemodynamic measurements obtained
in patients who were enrolled in the Vasodilation in the
Management of Acute Congestive Heart Failure (VMAC) trial.14
The VMAC hemodynamic substudy included simultaneous
assessments of hemodynamic parameters and dyspnea and
provided an opportunity to examine the relationship between

Received July 7, 2012; accepted November 1, 2012.

From the Department of Cardiology, Rambam Medical Center, Haifa, Israel (A.S., D.A.); Rappaport Faculty of Medicine and Research Institute, Technion,
Israel Institute of Technology, Haifa, Israel (A.J.B.); and the Division of Cardiovascular Disease, University of Cincinnati, Cincinnati, OH (A.J.B.).
The online-only Data Supplement is available at http://circheartfailure.ahajournals.org/lookup/suppl/doi:10.1161/CIRCHEARTFAILURE.
112.970335/-/DC1.
Correspondence to Doron Aronson, MD, Department of Cardiology, Rambam Medical Center, Bat Galim, POB 9602, Haifa 31096, Israel. E-mail:
daronson@tx.technion.ac.il
© 2012 American Heart Association, Inc.
Circ Heart Fail is available at http://circheartfailure.ahajournals.org DOI: 10.1161/CIRCHEARTFAILURE.112.970335

53
 54  Circ Heart Fail  January 2013
changes in hemodynamic variables and dyspnea improvement
in the setting of ADHF.
Methods
The study population included patients enrolled in the VMAC study,
a randomized, multicenter trial comparing the hemodynamic and
clinical effects of nesiritide (human B-type natriuretic peptide) to ni-
troglycerin in patients with decompensated congestive heart failure
for whom in-patient parenteral vasoactive therapy was considered
appropriate. Patients were recruited into the study between October
1999 and July 2000.14
Patients were included if they had dyspnea at rest attributed to
decompensated heart failure that was severe enough to require hos-
pitalization and intravenous therapy. One of the inclusion criteria of
the VMAC trial was evidence of heart disease, rather than pulmonary
disease, as the primary etiology for the dyspnea. A cardiac etiology
for dyspnea was established by jugular venous distention, paroxysmal
nocturnal dyspnea or 2-pillow orthopnea, chest x-ray film with findings
indicative of heart failure and by estimated or measured elevation of
cardiac filling pressures (pulmonary capillary wedge pressure [PCWP]
≥20 mm Hg in catheterized patients). Exclusion criteria were systolic
blood pressure <90 mm Hg, cardiogenic shock or volume depletion,
any condition that would contraindicate an intravenous vasodilator,
mechanical ventilation, and anticipated survival of <30 to 35 days.
The protocol was performed in conformance with the guidelines
established in the Declaration of Helsinki and was approved by the
institutional review board of the participating hospitals. Written, in-
formed consent was obtained from each patient.
Hemodynamic Evaluation
In the VMAC trial, the randomization was stratified based on the
investigator’s clinical decision, before randomization, to use a right
heart catheter to manage decompensated heart failure (catheterized or
Downloaded from http://ahajournals.org by on November 18, 2018
noncatheterized strata).
According to investigators, their clinical decision to use a right
heart catheter was influenced by >1 clinical factor in many subjects.
The 6 most common reasons for the use of right heart catheter, in
order of decreasing frequency, were as follows: (1) uncertain hemo-
dynamics in 204 subjects; (2) suspected low cardiac output in 131
subjects; (3) to optimize outpatient medications in 110 subjects; (4)
to evaluate a potential cardiac transplant candidate in 41 subjects; (5)
significant renal dysfunction in 40 subjects; and (6) low or unstable
blood pressure in 19 subjects.
In the catheterized group, PCWP and pulmonary artery pressures
were measured at baseline (before the initiation of study drugs), at 15
and 30 minutes, and at 1, 2, and 3 hours. The cardiac output and mean
right atrial pressure were measured at 1 and 3 hours. After 3 hours,
PCWP and mean pulmonary artery pressure (mPAP) were obtained in
catheterized patients at 6, 9, 12, 24, 36, and 48 hours and when study
drug was discontinued (if <48 hours).14,15 Total fluid intake and urine
output were recorded for the first 24 hours after the start of study drug.
Evaluation of Dyspnea
Dyspnea was measured with the use of a self-reported 7-point cate-
gorical Likert scale, ranging from markedly, moderately, or minimally
better; no change; or minimally, moderately, or markedly worse as
compared with the degree of dyspnea present at the start time of study
drug administration.3,14 Use of Likert scales has been validated for
measuring dyspnea in settings other than ADHF and has been used ex-
tensively as an end point in clinical trials of patient with ADHF.3,7,16,17
Dyspnea and global clinical evaluations were assessed at 15 and 30
minutes and at 1, 2, and 3 hours after the start of study drug and repeat-
ed at 6 and 24 hours.14 For the present analysis, clinically significant
relief of dyspnea was defined as a moderate or marked improvement in
dyspnea at each time point, as in previous studies.9–11
Statistical Methods
Continuous variables are presented as means±SDs or medians with
25th and 75th percentiles; categorical variables are presented as
frequencies and percentages. Baseline characteristics of the groups
were compared using unpaired t test or the nonparametric Mann-
Whitney U test for continuous variables and by the χ2 statistic for
noncontinuous variables (or the Fisher exact test, where appropriate).
Comparisons of hemodynamic measurements within patients were
carried out with the Wilcoxon matched–paired rank–sum test. The
relationship between 2 continuous variables was tested by Spearman
rank correlation.
The association between the clinically significant dyspnea im-
provement and hemodynamic variables over time was determined
by fitting longitudinal logistic regression models implemented with
generalized estimating equations with an assumed binomial family
distribution, a logistic link function, and an exchangeable structure of
the correlation matrix.18 In this framework, the generalized estimating
equation model parallels the logistic regression model but also takes
into account the correlation among multiple observations (time points
of dyspnea evaluation and hemodynamic measurements) per patient.
In these models, the dependent variables were the 7 repeated assess-
ments of dyspnea status during the first 24 hours. Results are reported
as odds ratios, together with associated 95% confidence intervals.
The following baseline clinical characteristics were considered in
the multivariate procedure, age, sex, history of diabetes mellitus, pri-
mary etiology of heart failure stratified as ischemic or nonischemic,
systolic blood pressure on admission, baseline epidermal growth fac-
tor receptor, ejection fraction (dichotomized as above or below 25%),
presence of atrial fibrillation, history of implantable cardioverter-de-
fibrillator implantation, and vasoactive therapy assignments (nesirit-
ide or nitroglycerin). The following medications were adjusted for as
dichotomous variables, indicating the use or nonuse of the medication
at study entry, digoxin, β-blockers, angiotensin-converting enzyme
inhibitors or angiotensin II receptor blockers, and spironolactone.
The following hemodynamic variables were also considered in the
multivariable analysis, PCWP, mPAP, right atrial pressure, cardiac
output, cardiac index, and systemic vascular resistance. Because of
the potential nonlinear relationship between continuous hemodynam-
ic variables and the likelihood of dyspnea relief, the relationships be-
tween hemodynamic variables and dyspnea relief were also assessed
with the use of restricted cubic spline transformations with 5 knots
at values based on the Harrell recommended percentiles.19,20 Spline
functions were used to flexibly model the relation between hemo-
dynamic measurements (as continuous variables) and dyspnea relief,
avoiding the need for previous specification of the risk function or the
location of a threshold exposure value. When a nonlinear relationship
was discovered, hemodynamic variables were divided into quartiles
and included in the model as categorical variables.
Differences were considered statistically significant at the 2-sided
P<0.05 level. All of the statistical analyses were performed using the
STATA statistical software version 11.0 (College Station, TX).
Results
Of the total 489 randomized and treated patients, 246 were in
the catheterized stratum. Complete evaluations of early dys-
pnea changes were obtained in 233 of these patients. Thirteen
patients were excluded because they did not have full data on
dyspnea relief.
At 24 hours, none of the patients reported marked worsen-
ing of dyspnea; 1 patient (0.4%) reported moderate worsening
of dyspnea, 2 patients (0.9%) reported mild worsening, and 29
patients (12.4%) reported no change in dyspnea. Forty-seven
(20.2%), 74 (31.8%), and 80 (34.3%) patients reported mild,
moderate, and marked improvement in dyspnea at 24 hours as
compared with baseline. Overall, at 24 hours, clinically sig-
nificant dyspnea relief occurred in 154 patients (66.1%). No
difference in dyspnea reduction occurred between the catheter-
ized and not catheterized strata (66.1% versus 69.6%; P=0.41).
Baseline characteristics of patients with and without
dyspnea relief are presented in Table 1. Patients without
dyspnea relief had a lower mean blood pressure and were
 Solomonica et al   Hemodynamics of Dyspnea Relief   55

Table 1.  Baseline Characteristics of Patients With and Without Dyspnea Relief at 24 Hours
Dyspnea Relief
All Patients
Characteristics (n = 233) No (n = 79) Yes (n = 154) P  Value
Age, y 60±14 60±14 61±14 0.46
Female sex 60 (26) 18 (23) 42 (27) 0.46
Body mass index, kg/m 2
29±11 27±15 29±7 0.67
Ischemic etiology of heart failure 126 (54) 41 (52) 85 (55) 0.63
Left ventricular ejection fraction, % 25±13 24±14 25±12 0.60
Diabetes mellitus 108 (46) 35 (44) 73 (47) 0.65
Atrial fibrillation 81 (35) 30 (38) 51 (33) 0.46
AICD implantation 71 (31) 22 (28) 49 (32) 0.53
Baseline heart rate, beats/min 84±15 83±14 84 (16) 0.41
Systolic blood pressure, mm Hg 118±20 115±19 120±20 0.08
Mean blood pressure, mm Hg 83±14 80±13 85±15 0.01
Baseline creatinine, mg/dL 1.4 (1.0–1.9) 1.5 (1.1–1.9) 1.4 (1.0–1.9) 0.82
Estimated GFR, mL/min per 1.73 m2 51 (35–74) 61 (35–70) 52 (34–74) 0.99
Randomized to nesiritide 174 (75) 61 (72) 113 (73) 0.52
Cardiac medications
 Digoxin 112 (48) 40 (51) 72 (47) 0.58
  ACE inhibitors 156 (67) 46 (52) 110 (71) 0.04
  Angiotensin II receptor blockers 17 (7) 4 (5) 13 (8) 0.35
  β-Blockers 46 (20) 16 (20) 30 (20) 0.89
 Spironolactone 67 (29) 22 (28) 45 (29) 0.83
  Calcium channel blockers 29 (12) 10 (13) 19 (12) 0.94
GFR indicates glomerular filtration rate; and ACE, angiotensin-converting enzyme. Values are presented as n (%), mean±SD, or as median
(interquartile range).
Downloaded from http://ahajournals.org by on November 18, 2018

less likely to be treated with angiotensin-converting enzyme
inhibitors.
Hemodynamic measurements in patients with and without
dyspnea relief at baseline and at 24 hours are shown in Table
2. There was no significant difference between the 2 groups
with regard to baseline hemodynamic measurements. Both
groups had reductions in PCWP and mPAP and an increase in
cardiac index. However, at 24 hours, PCWP and mPAP were
significantly lower in patients with dyspnea relief as compared
with patients without dyspnea relief (Table 2). There was a
moderate correlation between PCWP and mPAP (r=0.68;
P<0.0001).
Figure 1 demonstrates a time-dependent association between
the reductions in PCWP and the percentage of patients achiev-
ing clinically significant dyspnea relief, reflecting the increas-
ing number of patients with normal or near normal PCWP.
When cubic spline regression was used to explore unadjusted
nonlinear relationships between hemodynamic variables and

Table 2.  Hemodynamic Measurements in Patients With and Without Dyspnea Relief
Baseline Measurements, Dyspnea Relief 24-h Measurements, Dyspnea Relief
Hemodynamic Variables No (n=79) Yes (n=154) P Value No (n=79) Yes (n=154) P Value
Right atrial pressure, mm Hg 15±8 14±6 0.23 12±6‡ 10±5‡ 0.09
Pulmonary capillary wedge pressure, mm Hg 29±6 27±6 0.79 22±7‡ 19±6‡ 0.003
Cardiac output, L/min 4.3±1.7 4.4±1.6 0.71 4.7±1.6* 4.7±1.3† 0.93
Cardiac index, L/min/m2 2.2±0.8 2.2±0.8 0.15 2.4±0.7 2.4±0.6† 0.91
Stroke volume index, mL/m2 27±6 27±10 0.92 29±9 30±10† 0.67
Systemic vascular resistance, dynes·s/cm5 1369±580 1463±608 0.27 1238±505* 1235±421‡ 0.97
Mean pulmonary artery pressure, mm Hg 40±9 38±8 0.15 34±10‡ 30±8‡ 0.006
Pulmonary vascular resistance, WU 3.1±2.3 2.9±2.2 0.66 2.3±1.7 2.7±1.7 0.30
Transpulmonary gradient, mm Hg 11±6 11±6 0.63 13±6 11±6 0.11
Within group changes (baseline vs 24 h). WU indicates Wood units.
*P<0.05.
†P<0.01.
‡P<0.001.
 56  Circ Heart Fail  January 2013
Figure 1.  Relationship between time-dependent changes in
Downloaded from http://ahajournals.org by on November 18, 2018
hemodynamic variables and dyspnea improvement. A, Pulmo-
nary capillary wedge pressure (PCWP; SE) at each time point
is presented as red squares with error bars. The percentage
of patients with clinically significant improvement in dyspnea
at each time point is displayed by the cyan bars. B, Mean
pulmonary artery pressure (mPAP; SE) at each time point is
presented as red squares with error bars. The percentage of
patients with clinically significant improvement in dyspnea at
each time point is displayed by the cyan bars. P values for the
association of PCWP and mPAP with dyspnea improvement
were obtained from logistic generalized estimating equations
modeling.
dyspnea relief, we observed that the probability of dyspnea
relief increased progressively with decreasing PCWP, but the
slope of the smoothing spline was steeper for PCWP <20 mm
Hg (Figure 2A). For PCWP change (ΔPCWP), we observed
a monotonic increase in the probability of dyspnea relief for
any decrease in PCWP from baseline value (Figure 2B).
Analysis of the relationship between dyspnea and mPAP
demonstrated an inverse linear association between the prob-
ability of dyspnea relief and absolute mPAP (Figure 2C).
Similar to ΔPCWP, mPAP change (ΔmPAP) demonstrated a
monotonic increase in the probability of dyspnea relief for any
decrease in mPAP from baseline value (Figure 2D).
Univariable and multivariable logistic generalized estimat-
ing equation regression models of the association of baseline
characteristics and hemodynamic variables with relief of
dyspnea were constructed. In these models, hemodynamic
variables were described both as absolute values (Table 3,
model 1) and as the change from baseline (Table 3, model 2).
In a univariable analysis, the only nonhemodynamic variable
associated with dyspnea relief was the use of angiotensin-
converting enzyme inhibitors. In multivariable analyses, only
PCWP and mPAP (analyzed either as absolute or as change
from baseline) were independent determinants of greater like-
lihood of dyspnea relief. See the online-only Data Supplement
for a model in which baseline PCWP and mPAP were forced
into model 2.
To further demonstrate the importance of both PCWP and
mPAP reduction, the patients were categorized into 4 groups
based on whether PCWP and mPAP were above or below
median values. Figure 3 shows that the likelihood of achieving
dyspnea relief was higher when both PCWP and mPAP were
below median than when only PCWP or mPAP were below
median.
The amount of fluids removed during the first 24 hours was
not significantly associated with dyspnea relief (P=0.52). In
addition, a poor correlation was demonstrated between the
amount of fluid removed during the first 24 hours and the
magnitude of PCWP reduction during the same time (r=0.06;
P=0.35).
Discussion
Based on invasive hemodynamic measurements, the results of
the current study indicate that the short-term improvement of
dyspnea during therapy with vasodilators and diuretics depends
on 2 hemodynamic variables, PCWP and mPAP. Both the
absolute level and the magnitude of reduction determine the
likelihood of dyspnea improvement. Improvements of other
hemodynamic variables, such as cardiac index and systemic
vascular resistance, were not associated with dyspnea relief.
Alleviation of dyspnea has been a key measure of efficacy
for vasodilator therapy in recent ADHF trials.3,4 Dyspnea is
also important for the clinician and constitutes a major treat-
ment goal in ADHF. However, although few clinical variables
have been associated with dyspnea relief,9,10 the hemodynamic
determinants of dyspnea improvement in the setting of ADHF
have not been reported previously.
Recent studies have shown that rapid relief of dyspnea is
frequently not achieved in ADHF patients. In recent ADHF
trials, moderately or marked improvement in dyspnea was
achieved in only 25% to 65% of patients within 24 hours.9,10,12,13
These studies demonstrate that contemporary ADHF therapy
is suboptimal not only with respect to outcomes,21 but also
with respect to symptom relief.22 Therefore, it is important to
understand the reasons for the failure to improve dyspnea.
The present study demonstrates a clear time dependency
of dyspnea relief that is strongly related to the hemodynamic
response over time. Only a minority of patients achieved a
meaningful reduction in PCWP and mPAP within several
hours after initiation of therapy. Thus, the failure to achieve
early dyspnea relief in a large proportion of patients, as dem-
onstrated in recent studies,9–13 is likely related, at least in part,
to the failure to achieve a meaningful reduction in PCWP and
mPAP within a short time frame.
In this context, a recent consensus article on the assessment
of dyspnea in clinical trials recommended hourly assessment
of dyspnea for the first 3 hours followed by an assessment at 6
hours and every 6 hours thereafter until 24 hours.17 The present
results suggest that achieving a sufficient reduction in key
 Solomonica et al   Hemodynamics of Dyspnea Relief   57

A C

B D
Downloaded from http://ahajournals.org by on November 18, 2018

Figure 2.  Spline function graph of the unadjusted relationship between absolute pulmonary capillary wedge pressure (PCWP; A), PCWP
change (B), absolute mean pulmonary artery pressure (mPAP; C), mPAP change (D), and the probability of dyspnea relief (orange line)
during the first 24 hours. The blue area represents the pointwise 95% confidence interval. Negative numbers on the x-axis in B and D
indicate a reduction in PCWP and in mPAP vs baseline values, respectively.

hemodynamic parameters within several hours after initiation of
therapy may not be attainable in many patients. For example, a
reduction in PCWP and mPAP that would translate into dyspnea
improvement may be particularly difficult in patients with
extremely elevated initial filling pressures, mitral regurgitation
and severe diastolic dysfunction, or diuretic resistance. Thus,
dyspnea evaluation at longer time points after initiation of
therapy should be considered in future clinical trials.
Dyspnea is one of the principal symptoms of heart failure and
is often found with pulmonary hypertension regardless of the
underlying cause. In patients with heart failure, pulmonary vas-
cular resistance is an important determinant of exercise capac-
ity, which contributes to dyspnea on exertion.23,24 The present
study demonstrates that the probability of a clinically signifi-
cant improvement in dyspnea was highest when both PCWP
and mPAP were reduced. Thus, the severity of pulmonary
hypertension may also be an important determinant of dyspnea
at rest in patients with ADHF. These findings also indicate that
pulmonary vascular hemodynamics may have an important role
in determining the clinical response to treatment in ADHF.
Previous studies reported a discrepancy between weight
loss and symptom improvement or outcome in patients
with ADHF.25 In the Initiation Management Pre-discharge
Assessment of Carvedilol Therapy in Heart Failure registry,
the degree of weight loss during hospitalization for ADHF
was not associated with the degree of improvement in dyspnea
at rest.26 The MEASURE-AHF registry27 and the PROTECT
pilot study9 also reported that changes in body weight were not
related to changes in symptoms. By contrast, in the Efficacy
of Vasopressin Antagonism in Heart Failure Outcome Study
With Tolvaptan (EVEREST) trial, reductions in body weight
in response to tolvaptan on day 1 were accompanied by sig-
nificant improvements in patient-assessed dyspnea.13
The discordance between early clinical improvement and
fluid loss becomes clearer in view of the poor correlation of the
later with the reduction in PCWP. There are several potential
explanations for this finding. The effect of vasodilator therapy
leads to a reduction of PCWP, which is independent of fluid
removal. Furthermore, changes in filling pressures may occur
independent of fluid loss because of sympathetically mediated
changes in venous capacitance that lead to rapid shifts in effec-
tive circulatory volume.28 Finally, even an effective removal of
volume at 24 hours may not be sufficient to produce a signifi-
cant reduction in filling pressures in some patients because of
reabsorption of peripheral edema into the vascular space with
minimal reduction in intravascular volume.
 58  Circ Heart Fail  January 2013

Table 3.  Unadjusted and Adjusted GEE Logistic Regression Analysis of Determinants of Clinically Significant Dyspnea Relief
Variable Unadjusted OR (95% CI) P Value P Trend Adjusted (95% CI) P Value P Trend
Model 1*
ACE inhibitor 1.79 (1.02–3.16) 0.044 — —
PCWP, mm Hg
  Q1 (≤19 mm Hg) 2.96 (2.13–4.12) <0.0001 <0.0001 1.80 (1.22–2.65) 0.003 0.003
  Q2 (20 to 22 mm Hg) 1.53 (1.10–2.13) 0.01 1.07 (0.75–1.55) 0.70
  Q3 (23 to 28 mm Hg) 1.19 (0.88–1.61) 0.26 0.92 (0.67–1.29) 0.63
  Q4 (≥29 mm Hg) 1.0 (Referent) — 1.0 (Referent) —
mPAP, mm Hg
  Q1 (≤29 mm Hg) 4.19 (2.85–6.16) <0.0001 <0.0001 2.98 (1.91–4.66) <0.0001 <0.0001
  Q2 (30 to 34 mm Hg) 2.47 (1.74–3.52) <0.0001 2.23 (1.52–3.27) <0.0001
  Q3 (35 to 40 mm Hg) 2.04 (1.46–2.83) <0.0001 1.99 (1.41–2.82) <0.0001
  Q4 (≥41 mm Hg) 1.0 (Referent) — 1.0 <0.0001
Model 2†
ACE inhibitor 1.79 (1.02–3.16) 0.044 — —
ΔPCWP (mm Hg)
  Q1 (≤0 mm Hg) 1.0 (Referent) — <0.0001 1.0 (Referent) — <0.0001
  Q2 (–1 to –4 mm Hg) 0.98 (0.76–1.27) 0.90 0.94 (0.72–1.23) 0.65
  Q3 (–5 to –8 mm Hg) 1.90 (1.44–2.51) <0.0001 1.66 (1.23–2.23) <0.0001
  Q4 (at or more than –9 mm Hg) 2.79 (2.08–3.75) <0.0001 2.01 (1.43–2.81) <0.0001
ΔmPAP (mm Hg)
  Q1 (≤0 mm Hg) 1.0 (Referent) — <0.0001 1.0 (Referent) — <0.0001
  Q2 (–1 to –3 mm Hg) 1.53 (1.17–2.02) 0.002 1.51 (1.15–2.00) 0.003
  Q3 (–4 to –7 mm Hg) 1.36 (1.01–1.83) 0.04 1.21 (0.89–1.66) 0.22
Downloaded from http://ahajournals.org by on November 18, 2018

  Q4 (at or more than –8 mm Hg) 3.30 (2.43–4.49) <0.0001 2.38 (1.68–3.37) <0.0001
mPAP indicates mean pulmonary artery pressure; PCWP, pulmonary capillary wedge pressure; ACE, angiotensin-converting enzyme; OR, odds ratio; CI, confidence
interval; and GEE, generalized estimating equations.
*GEE logistic model is based on absolute hemodynamic measurements.
†GEE logistic model is based on the changes in hemodynamic parameters compared with baseline.

Although the preset study demonstrates the importance of
hemodynamic improvement for dyspnea relief in ADHF, the
mechanisms of dyspnea in heart failure and pulmonary hyper-
tension remain incompletely understood.29 Pulmonary con-
gestion reduces the compliance of the lung and increases the
work of breathing.30 Several studies have demonstrated that
the sensation of dyspnea is also related to reduced inspiratory
muscle strength.31,32 It has also been postulated that vascular
distension and interstitial edema may directly stimulate pul-
monary vascular nerve endings and result in dyspnea.29,33

Figure 3.  Dyspnea improvement stratified by the

achieved level of pulmonary capillary wedge pres-
sure (PCWP) and mean pulmonary artery pressure
(mPAP). Error bars represent 95% confidence
intervals. Cutoff values for PCWP and mPAP are
based on the median values.
 Solomonica et al   Hemodynamics of Dyspnea Relief   59
Study Limitations
Our study has some limitations that merit emphasis. Although
mPAP is directly measured, PCWP is an indirect measure of
left ventricular filling. PCWP may not accurately reflect left-
side filling pressures, especially when proper wedge position
is not routinely confirmed by measuring oxygen saturation.34
Thus, difficulties in obtaining a proper wedge position may
have weakened the association between PCWP and dyspnea.
No data were available with regard to specific conditions
that might have contributed to the lack of hemodynamic
response to vasodilators and diuretics (eg, severe mitral regur-
gitation). In addition, the short-term improvement in dyspnea
does not necessarily translate into long-term clinical response.
Conclusions
The present study of acute hemodynamic changes in the set-
ting of ADHF suggests that a meaningful improvement in dys-
pnea depends on the ability to reduce both PCWP and mPAP.
The failure to achieve early dyspnea relief in clinical trials is
likely related, at least in part, to the failure to achieve a rapid
reduction in these hemodynamic parameters.
Disclosures
None.
References
1. Gheorghiade M, Pang PS. Acute heart failure syndromes. J Am Coll Car-
diol. 2009;53:557–573.
2. Adams KF Jr, Fonarow GC, Emerman CL, LeJemtel TH, Costanzo MR,
Downloaded from http://ahajournals.org by on November 18, 2018
Abraham WT, Berkowitz RL, Galvao M, Horton DP; ADHERE Scientif-
ic Advisory Committee and Investigators. Characteristics and outcomes
of patients hospitalized for heart failure in the United States: rationale,
design, and preliminary observations from the first 100,000 cases in the
Acute Decompensated Heart Failure National Registry (ADHERE). Am
Heart J. 2005;149:209–216.
3. O’Connor CM, Starling RC, Hernandez AF, Armstrong PW, Dickstein
K, Hasselblad V, Heizer GM, Komajda M, Massie BM, McMurray JJ,
Nieminen MS, Reist CJ, Rouleau JL, Swedberg K, Adams KF Jr, Anker
SD, Atar D, Battler A, Botero R, Bohidar NR, Butler J, Clausell N, Cor-
balán R, Costanzo MR, Dahlstrom U, Deckelbaum LI, Diaz R, Dunlap
ME, Ezekowitz JA, Feldman D, Felker GM, Fonarow GC, Gennevois
D, Gottlieb SS, Hill JA, Hollander JE, Howlett JG, Hudson MP, Kociol
RD, Krum H, Laucevicius A, Levy WC, Méndez GF, Metra M, Mittal S,
Oh BH, Pereira NL, Ponikowski P, Tang WH, Wilson WH, Tanomsup
S, Teerlink JR, Triposkiadis F, Troughton RW, Voors AA, Whellan DJ,
Zannad F, Califf RM. Effect of nesiritide in patients with acute decom-
pensated heart failure. N Engl J Med. 2011;365:32–43.
4. Teerlink JR, Metra M, Felker GM, Ponikowski P, Voors AA, Weatherley
BD, Marmor A, Katz A, Grzybowski J, Unemori E, Teichman SL, Cotter
G. Relaxin for the treatment of patients with acute heart failure (Pre-
RELAX-AHF): a multicentre, randomised, placebo-controlled, parallel-
group, dose-finding phase IIb study. Lancet. 2009;373:1429–1439.
5. West RL, Hernandez AF, O’Connor CM, Starling RC, Califf RM.
A review of dyspnea in acute heart failure syndromes. Am Heart J.
2010;160:209–214.
6. Colucci WS, Elkayam U, Horton DP, Abraham WT, Bourge RC,
Johnson AD, Wagoner LE, Givertz MM, Liang CS, Neibaur M, Haught
WH, LeJemtel TH. Intravenous nesiritide, a natriuretic peptide, in the
treatment of decompensated congestive heart failure: Nesiritide Study
Group. N Engl J Med. 2000;343:246–253.
7. Allen LA, Hernandez AF, O’Connor CM, Felker GM. End points for
clinical trials in acute heart failure syndromes. J Am Coll Cardiol.
2009;53:2248–2258.
8. Yancy CW, Lopatin M, Stevenson LW, De Marco T, Fonarow GC;
ADHERE Scientific Advisory Committee and Investigators. Clinical
presentation, management, and in-hospital outcomes of patients admitted
with acute decompensated heart failure with preserved systolic function:
a report from the Acute Decompensated Heart Failure National Registry
(ADHERE) Database. J Am Coll Cardiol. 2006;47:76–84.
9. Metra M, Cleland JG, Weatherley BD, Dittrich HC, Givertz MM, Massie
BM, O’Connor CM, Ponikowski P, Teerlink JR, Voors AA, Cotter G.
Dyspnoea in patients with acute heart failure: an analysis of its clinical
course, determinants, and relationship to 60-day outcomes in the PRO-
TECT pilot study. Eur J Heart Fail. 2010;12:499–507.
10. Metra M, Teerlink JR, Felker GM, Greenberg BH, Filippatos G,
Ponikowski P, Teichman SL, Unemori E, Voors AA, Weatherley BD,
Cotter G. Dyspnoea and worsening heart failure in patients with acute
heart failure: results from the Pre-RELAX-AHF study. Eur J Heart Fail.
2010;12:1130–1139.
11. Metra M, O’Connor CM, Davison BA, Cleland JG, Ponikowski P, Teer-
link JR, Voors AA, Givertz MM, Mansoor GA, Bloomfield DM, Jia G,
DeLucca P, Massie B, Dittrich H, Cotter G. Early dyspnoea relief in
acute heart failure: prevalence, association with mortality, and effect of
rolofylline in the PROTECT Study. Eur Heart J. 2011;32:1519–1534.
12. Gheorghiade M, Konstam MA, Burnett JC Jr, Grinfeld L, Maggioni AP,
Swedberg K, Udelson JE, Zannad F, Cook T, Ouyang J, Zimmer C, Or-
landi C; Efficacy of Vasopressin Antagonism in Heart Failure Outcome
Study With Tolvaptan (EVEREST) Investigators. Short-term clinical
effects of tolvaptan, an oral vasopressin antagonist, in patients hospi-
talized for heart failure: the EVEREST Clinical Status Trials. JAMA.
2007;297:1332–1343.
13. Pang PS, Konstam MA, Krasa HB, Swedberg K, Zannad F, Blair JE,
Zimmer C, Teerlink JR, Maggioni AP, Burnett JC Jr, Grinfeld L, Ouyang
J, Udelson JE, Gheorghiade M; Efficacy of Vasopressin Antagonism in
Heart Failure Outcome Study with Tolvaptan (EVEREST) Investigators.
Effects of tolvaptan on dyspnoea relief from the EVEREST trials. Eur
Heart J. 2009;30:2233–2240.
14. Publication Committee for the VMAC Investigators. Intravenous nesirit-
ide vs nitroglycerin for treatment of decompensated congestive heart
failure: a randomized controlled trial. JAMA. 2002;287:1531–1540.
15. Aronson D, Eitan A, Dragu R, Burger AJ. Relationship between reactive
pulmonary hypertension and mortality in patients with acute decompen-
sated heart failure. Circ Heart Fail. 2011;4:644–650.
16. Teerlink JR. Dyspnea as an end point in clinical trials of thera-
pies for acute decompensated heart failure. Am Heart J. 2003;145(2
suppl):S26–S33.
17. Pang PS, Cleland JG, Teerlink JR, Collins SP, Lindsell CJ, Sopko G,
Peacock WF, Fonarow GC, Aldeen AZ, Kirk JD, Storrow AB, Tavares M,
Mebazaa A, Roland E, Massie BM, Maisel AS, Komajda M, Filippatos
G, Gheorghiade M; Acute Heart Failure Syndromes International Work-
ing Group. A proposal to standardize dyspnoea measurement in clinical
trials of acute heart failure syndromes: the need for a uniform approach.
Eur Heart J. 2008;29:816–824.
18. Twisk JWR. Applied Multilevel Analysis: A Practical Guide. Cambridge,
United Kingdom; Cambridge University Press; 2006.
19. Marrie RA, Dawson NV, Garland A. Quantile regression and restricted
cubic splines are useful for exploring relationships between continuous
variables. J Clin Epidemiol. 2009;62:511–7.e1.
20. Harrell FE. Regression Modeling Strategies: With Applications to Lin-
ear Models, Logistic Regression, and Survival Analysis. New York, NY:
Springer; 2001.
21. Aronson D, Krum H. Novel therapies in acute and chronic heart failure.
Pharmacol Ther. 2012;135:1–17.
22. Mebazaa A, Pang PS, Tavares M, Collins SP, Storrow AB, Laribi S,
Andre S, Mark Courtney D, Hasa J, Spinar J, Masip J, Frank Peacock W,
Sliwa K, Gayat E, Filippatos G, Cleland JG, Gheorghiade M. The impact
of early standard therapy on dyspnoea in patients with acute heart failure:
the URGENT-dyspnoea study. Eur Heart J. 2010;31:832–841.
23. Butler J, Chomsky DB, Wilson JR. Pulmonary hypertension and ex-
ercise intolerance in patients with heart failure. J Am Coll Cardiol.
1999;34:1802–1806.
24. Lewis GD, Shah RV, Pappagianopolas PP, Systrom DM, Semigran MJ.
Determinants of ventilatory efficiency in heart failure: the role of right
ventricular performance and pulmonary vascular tone. Circ Heart Fail.
2008;1:227–233.
25. Cotter G, Metra M, Milo-Cotter O, Dittrich HC, Gheorghiade M. Fluid
overload in acute heart failure–re-distribution and other mechanisms be-
yond fluid accumulation. Eur J Heart Fail. 2008;10:165–169.
26. O’Connor CM, Stough WG, Gallup DS, Hasselblad V, Gheorghiade M.
Demographics, clinical characteristics, and outcomes of patients hospi-
talized for decompensated heart failure: observations from the IMPACT-
HF registry. J Card Fail. 2005;11:200–205.
 60  Circ Heart Fail  January 2013

27. Allen LA, Metra M, Milo-Cotter O, Filippatos G, Reisin LH, Bensim-
hon DR, Gronda EG, Colombo P, Felker GM, Cas LD, Kremastinos DT,
O’Connor CM, Cotter G, Davison BA, Dittrich HC, Velazquez EJ. Im-
provements in signs and symptoms during hospitalization for acute heart
failure 11:31 am follow different patterns and depend on the measure-
ment scales used: an international, prospective registry to evaluate the
evolution of measures of disease severity in acute heart failure (MEA-
SURE-AHF). J Card Fail. 2008;14:777–784.
28. Fallick C, Sobotka PA, Dunlap ME. Sympathetically mediated changes
in capacitance: redistribution of the venous reservoir as a cause of de-
compensation. Circ Heart Fail. 2011;4:669–675.
29. Parshall MB, Schwartzstein RM, Adams L, Banzett RB, Manning HL,
Bourbeau J, Calverley PM, Gift AG, Harver A, Lareau SC, Mahler DA,
Meek PM, O’Donnell DE; American Thoracic Society Committee on
Dyspnea. An official American Thoracic Society statement: update on
the mechanisms, assessment, and management of dyspnea. Am J Respir
Crit Care Med. 2012;185:435–452.
30. Gehlbach BK, Geppert E. The pulmonary manifestations of left heart
failure. Chest. 2004;125:669–682.
31. McParland C, Krishnan B, Wang Y, Gallagher CG. Inspiratory muscle
weakness and dyspnea in chronic heart failure. Am Rev Respir Dis.
1992;146:467–472.
32. Mancini DM, Henson D, LaManca J, Levine S. Respiratory muscle func-
tion and dyspnea in patients with chronic congestive heart failure. Circu-
lation. 1992;86:909–918.
33. Snashall PD, Chung KF. Airway obstruction and bronchial hyperrespon-
siveness in left ventricular failure and mitral stenosis. Am Rev Respir Dis.
1991;144:945–956.
34. Rich S, Rabinovitch M. Diagnosis and treatment of secondary (non-
category 1) pulmonary hypertension. Circulation. 2008;118:2190–2199.

CLINICAL PERSPECTIVE
Relief of dyspnea constitutes a major treatment goal in acute heart failure and an important end point in clinical trials. How-
ever, recent studies have shown that rapid relief of dyspnea is frequently not achieved in patients with decompensated heart
failure. Previous studies failed to identify clinical characteristics or laboratory tests that reliably predict dyspnea relief. In
the present study we examined the relationship between hemodynamic changes and dyspnea relief (defined as moderate or
marked improvement) using frequent measurements of hemodynamic parameters and simultaneous dyspnea assessments
in patients with acute heart failure who were treated with vasodilators and diuretics. We observed a clear time dependency
of dyspnea relief that was strongly related to the improvement in hemodynamic parameters. Specifically, dyspnea relief
depended on 2 hemodynamic variables, pulmonary capillary wedge pressure and mean pulmonary artery pressure. Both the
absolute level and the magnitude of reductions of these hemodynamic variables were associated with the likelihood of dys-
pnea improvement. The likelihood of achieving dyspnea relief was particularly high when both pulmonary capillary wedge
pressure and mean pulmonary artery pressure were effectively reduced. These results suggest that the failure to achieve early
dyspnea relief in clinical practice and in clinical trials is likely related, at least in part, to the failure to achieve a rapid reduc-
tion in these hemodynamic parameters.
Downloaded from http://ahajournals.org by on November 18, 2018