JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY
13, 2017
Listen to this manuscript’s
audio summary by
JACC Editor-in-Chief
Dr. Valentin Fuster.
VOL. 70, NO.
ª 2017 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION ISSN 0735-1097/$36.00
PUBLISHED BY ELSEVIER http://dx.doi.org/10.1016/j.jacc.2017.08.012
THE PRESENT AND FUTURE
STATE-OF-THE-ART REVIEW
Cardiopulmonary Exercise Testing
What Is its Value?
Marco Guazzi, MD, PHD,a Francesco Bandera, MD, PHD,a Cemal Ozemek, PHD,b David Systrom, MD,c,d
Ross Arena, PHDb
ABSTRACT
Compared with traditional exercise tests, cardiopulmonary exercise testing (CPET) provides a thorough assessment of
exercise integrative physiology involving the pulmonary, cardiovascular, muscular, and cellular oxidative systems. Due to
the prognostic ability of key variables, CPET applications in cardiology have grown impressively to include all forms of
exercise intolerance, with a predominant focus on heart failure with reduced or with preserved ejection fraction. As
impaired cardiac output and peripheral oxygen diffusion are the main determinants of the abnormal functional response
in cardiac patients, invasive CPET has gained new popularity, especially for diagnosing early heart failure with preserved
ejection fraction and exercise-induced pulmonary hypertension. The most impactful advance has recently come from the
introduction of CPET combined with echocardiography or CPET imaging, which provides basic information regarding
cardiac and valve morphology and function. This review highlights modern CPET use as a single or combined test that
allows the pathophysiological bases of exercise limitation to be translated, quite easily, into clinical practice.
(J Am Coll Cardiol 2017;70:1618–36) © 2017 by the American College of Cardiology Foundation.
I n cardiopulmonary disorders, exercise intoler-
ance is a major clinical feature from early stages,
and becomes a source of symptoms and the
reason for referral to a physician. Exercise limitation
Despite this attractive evidence base and the clinical
potential of CPET, the question put forth in the title of
the present review (i.e., what is the value of CPET), is
not trivial, and the definitive response requires
is one of the most disabling problems experienced mentioning a few historical notes and passages.
by patients with heart failure (HF) (1). Its quantifica- The idea of a CPET application in cardiology was
tion may be approximated by several methods, but a introduced in the early 1980s by Weber et al. (5), whose
thorough analysis of the organ systems and pathways work allowed the landmark classification of patients
involved in the impaired physiological response is with HF with reduced ejection fraction (HFrEF) based
obtained by exercise gas exchange analysis with on peak oxygen consumption (VO 2), from A (peak VO 2
cardiopulmonary exercise testing (CPET). This > 20 ml $ kg
1 $ min
1 ) to D (peak V O2 < 10 ml $ kg
1 $
technique enables the clinician to scrutinize reasons min
1) through B (peak V O2 <20 to >15 ml $ kg
1 $ min
1)
for dyspnea and fatigue to precisely differentiate and C (peak VO2 < 15 and 10 ml $ kg
1 $ min
1).
cardiac from pulmonary disorders, optimize the A few years later, Mancini et al. (6), in their seminal
decision-making process and outcome prediction, 1991 paper, demonstrated that VO2 measured at peak
and objectively determine targets for therapies (2). exercise stratifies the risk of cardiovascular (CV) death
Furthermore, CPET has become a reproducible (3) at 1 year in ambulatory patients with advanced HF.
and safe technique (4). These remarkable findings were subsequently
From the aUniversity of Milan, Cardiology University Department, Heart Failure Unit, IRCCS Policlinico San Donato, San Donato
Milanese, Milan, Italy; bDepartment of Physical Therapy, Department of Kinesiology and Nutrition, College of Applied Health
Sciences, University of Illinois at Chicago, Chicago, Illinois; cDivision of Pulmonary and Critical Care Medicine, Department of
Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts; and the dHeart and Vascular
Center, Brigham and Women’s Hospital, Boston, Massachusetts. The authors have reported that they have no relationships
relevant to the contents of this paper to disclose.
Manuscript received August 1, 2017; accepted August 2, 2017.
JACC VOL. 70, NO. 13, 2017
SEPTEMBER 26, 2017:1618–36
validated and reproduced by several laboratories (7,8).
This solid evidence perhaps led to a quite paradoxical
static vision of CPET applications for a long period (i.e.,
until the 2000s), with a single parametric approach
focused just on advanced HF.
In the last 15 years, the utility of CPET has been
increasingly recognized by both extending medical in-
terest to the physiological bases of many variables that
were previously under-recognized and by aligning evi-
dence for a multivariable approach, including primarily
abnormalities in ventilation and its control (1). In HF, the
combined use of variables has led to the generation of
algorithms (9) and risk scores (10–12) covering the entire
set of HF stages. This process has been witnessed and
validated by a significant number of official documents
and statements definitively (1,2,13) entering exercise gas
exchange variables as study endpoint in the assessment
of the effects of emerging pharmacological therapies
(14,15) and in interventional trials (16,17). Along with the
main developments in HF, the role of routine CPET in
cardiology has been extended to specific patient pop-
ulations, including those with suspected ischemic heart
disease (18), congenital heart defects (19), valve diseases
(20), hypertrophic cardiomyopathy (21), suspected or
confirmed pulmonary arterial hypertension (PAH) (22),
and left-sided pulmonary hypertension (PH) (23).
In this paper, the modern key applications of CPET
in CV diseases, with primary emphasis on HF, are
discussed, starting from the principles that precipi-
tate reduced exercise performance and impaired
ventilation, and highlighting the most recent de-
velopments on combining exercise invasive hemo-
dynamic and stress echocardiography with gas
exchange evaluation. The large body of evidence on
the established pathophysiological clinical and prog-
nostic impact of CPET-derived variables will be
emphasized, making a continuum of value from
physiological bases to their translation into the
practical applications. Accordingly, CPET is here
proposed as a technique that may provide significant
and synergistic advancements in the process of
precision medicine and phenotyping.
GAS EXCHANGE ANALYSES AND
THE PRINCIPAL BASES FOR
EXERCISE LIMITATION IN HF
OXYGEN TRANSPORT AND USE. The body under
physical stress behaves as a perfect machine that in-
tegrates and harmonizes the functional responses of
multiple organs and pathways. In this process, the
delivery of oxygen (O2 ) to mitochondria is essential to
perform at aerobic capacity (24). Optimal O 2 delivery
depends on a set of elegant biological interactions
Guazzi et al. 1619
Cardiopulmonary Exercise Testing
between the functional components of the O 2 ABBREVIATIONS
transport chain, which requires oxygenation AND ACRONYMS
of the blood in the lung (alveolar diffusion),
CaO2 = arterial oxygen content
normal O 2-carrying capacity of the blood
CO = cardiac output
by adequate cardiac output (CO), yielding
CPET = cardiopulmonary
to its redistribution to working muscles
exercise testing
(delivery or convection) and adequate O2
CvO2 = venous oxygen content
release, diffusion from capillaries to cells,
EOV = exercise oscillatory
and tissue extraction from the blood
ventilation
(Central Illustration).
HF = heart failure
HF represents the typical condition where
HFpEF = heart failure with
most of these pathways exhibit a maladaptive preserved ejection fraction
response, impeding attainment of maximal
HFrEF = heart failure with
VO2 because of the reduction in muscle O 2 reduced ejection fraction
supply simultaneous with increasing de- PETCO2 = end-tidal pressure of
mands for O2. Maximal performance is carbon dioxide
therefore defined by V O2 at peak exercise, PETO2 = end-tidal pressure
conventionally measured as the VO2 averaged of oxygen
over a 20- to 30-s period at maximal effort, tau = time constant
pending attainment of a respiratory exchange VCO2 = volume of carbon
dioxide released
ratio (RER) >1.15 (1).
Measuring peak VO 2 and, especially, the VD = dead space
percentage predicted, normalized to age-, VE = ventilation
sex-, and weight-based normative values, is VO2 = peak oxygen
consumption
the gold standard to objectively assess func-
tional limitations in cardiac patients (1). VT = tidal volume
What are the major pathophysiological reasons for
a low peak VO2 in cardiac patients? These can be
described by analyzing the framework of the Fick
principle (i.e., VO 2 ¼ CO  [CaO 2
CvO 2], where CO is
cardiac output [stroke volume  heart rate], CaO 2 is
arterial oxygen content, CvO 2 is venous oxygen con-
tent, and [CaO 2
CvO2 ] is the arteriovenous [a-v]
difference in O 2).
On the basis of this equation, delivery or convec-
tion and extraction are the 2 physiological processes
that convey O2 use through cellular pathways.
O2 delivery is not only described by CO distribution,
but is also the O 2 content and the mechanisms
involved in O 2 dissociation from hemoglobin (Hb).
O2 content is 1.34 (which corresponds to the ml of O 2
carried by each gram of Hb)  O 2 saturation  Hb
concentration.
Extraction is the net result of the chain of O 2
transport and use, and depends on the ability of O2 to
diffuse from capillaries to cells and on mitochondria
function.
Normal, healthy adults can increase V O2 up to
6-fold during exercise. The relative contribution
of Fick’s equation determinants to VO2 changes is
approximately 1.2-, 2.5-, and 2.5-fold for stroke
volume, heart rate, and CaO 2
CvO 2, respectively.
Landmark pioneering studies (5,25) performed in
patients with HFrEF first demonstrated that low peak
1620 Guazzi et al. JACC VOL. 70, NO. 13, 2017

Cardiopulmonary Exercise Testing SEPTEMBER 26, 2017:1618–36

C E NT R AL IL L U STR AT IO N Determinants of the O 2 Transport and Utilization Chain Framed on the Fick Principle

Guazzi, M. et al. J Am Coll Cardiol. 2017;70(13):1618–36.

Optimal O2 delivery depends on the optimal biological interaction between functional components of the O2 transport chain. This requires oxygenation of the blood in
the lung (alveolar diffusion), normal O2 carrying capacity of the blood by adequate CO; CO redistribution to working muscles (delivery or convection), and efficient O2
release, diffusion from capillaries to cells, and tissue extraction from the blood. The arteriovenous O2 difference reflects the extraction of O2 from mitochondria by
muscle cells. ATP ¼ adenosine triphosphate; CaO2
CvO2 ¼ the arteriovenous difference in oxygen; CO ¼ cardiac output; Hb ¼ hemoglobin; HFpEF ¼ heart failure
with preserved ejection fraction; HFrEF ¼ heart failure with reduced ejection fraction; O2 ¼ oxygen; VO2 ¼ peak oxygen consumption.

VO2 is tightly related to a limited CO increase and
quite reasonably preserved peripheral O2 use. When
data were stratified according to functional impair-
ment, even Weber class D patients, despite having no
increase in CO, were still able to maintain their
CaO 2
CvO 2 difference within the lower-normal
range (12 to 16 ml/dl). The underlying pathophysi-
ology needs, of course, to be complemented with the
concept that along with reduced CO, several patients
with HF may have anemia and a lower O 2 concen-
tration, thus further impairing O 2 delivery and
challenging the a-v O2 difference, irrespective of the
ability to efficiently extract O 2 (1).
Intriguingly, observations have suggested that
iron deficiency in nonanemic patients with HF is
responsible for worse exercise performance and
exercise gas exchange phenotypes (26).
observations have been confirmed over time by
several studies proving that in HFrEF, the capability
to extract O 2 remains preserved thanks to an “opti-
mized” peripheral blood flow redistribution and high
mitochondrial activity (27–29), and that the cause of
an inadequate increase in CO is predominant, even
though likely not exclusive. An additional set of
remarkable works aimed at assessing how much an
impaired O 2 diffusion may be involved in the low
aerobic performance (30), on the basis of Fick’s law
of diffusion: VO 2 ¼ D  K  (PaO 2
PvO 2), where D is
equal to the O 2 diffusive capacity (D), K is a constant,
which is the ratio between arterial and venous O 2
partial pressure (which is approximately 2), and
(PaO 2
PvO 2) is the capillary mitochondrial differ-
ence, assuming an intracellular partial pressure of O2
close to 0.
The relation of delivery or convection (Fick prin-
These ciple) and diffusion (Fick’s law) can be represented on
a diagram relating VO2 and PvO 2, which are common
to both equations (Figure 1A).
JACC VOL. 70, NO. 13, 2017 Guazzi et al. 1621
SEPTEMBER 26, 2017:1618–36 Cardiopulmonary Exercise Testing

F I G U R E 1 The Relationship Between O 2 Diffusion and Convection

A CONVECTION: VO2 = Q x [CaO2 – CvO2]

DIFFUSION: VO2 = D x k x PvO2
B C
3500
DIFFUSION Fick principle lines
Oxygen Consumption ml/min

3000
4000 Fick law lines
VO2 peak

Leg VO2 (ml/min)

2500

VO2 in ml/min
Peak
3000 A Control
peak VO2 2000

2000 1500
B E
C Peak
1000
D 30 W
1000 30 W
500 30 W
CONVECTION Rest Rest

0 0
0 10 20 30 40 50 60 70 80 90 100 Muscle Venous PO2 (Torr) 0 2 4 6 8 10 12 14 16 18
Muscle Venous PO2, mm Hg CvO2 in ml/dl
Controls HFpEF HFrEF

(A) Scheme of the 2 relationships of delivery or convection (Fick principle) and diffusion (Fick’s law) in the diagram relating VO2 and in venous O2 pressure, which are
common to both equations (30). (B) Relative impairment of convection versus diffusion by 2 different exercise modalities, incremental test at maximum (A and B)
versus local isolated knee extension (C and D) (28). (C) VO2 and PvO2 relationship in a group of patients with HFrEF, HFpEF, and controls, showing a limitation in either
diffusion and convection (33). CaO2 ¼ arterial oxygen content; CvO2 ¼ venous oxygen content; D ¼ the O2 diffusive capacity; HFpEF ¼ heart failure with preserved
ejection fraction; HFrEF ¼ heart failure with reduced ejection fraction; K ¼ a constant (the ratio between arterial and venous O2 partial pressure, which is
approximately 2); O2 ¼ oxygen; PO2 ¼ oxygen pressure; PVO2 ¼ venous oxygen pressure; Q ¼ cardiac output; VO2 ¼ peak oxygen consumption.

As proposed by Wagner (30), the diagram relates
VO 2 to the muscle-venous PO 2 from rest to peak
exercise, with the lines of convection and diffusion
crossing the axes and intercepts, and whose slope
varies according to the muscle-venous PO 2, with an
upper and leftward shift representing the best diffu-
sion, whereas the point of intersection between the 2
lines represents the VO 2 max (i.e., the resultant of
convection and diffusion).
In a set of elegant, controlled experiments per-
formed in HFrEF (28), the relative impairment of
convection versus diffusion was assessed by testing
the same patients with 2 different exercise modal-
ities: an incremental test at maximal exercise (need
for CO increase, central challenge) versus a local
isolated knee extension (need for increase in O2
extraction, peripheral challenge) (Figure 1B). These
experiments were paralleled by the analysis of mus-
cle fiber type, mitochondria volume,
capillary-to-fiber ratio.
Interestingly, skeletal muscle and mitochondria
contributed minimally to O 2 limitation, in agreement
with the concept that mitochondrial
capacity largely exceeds O 2 delivery at peak VO2 (31).
Rather, patients with HF, compared with controls,
exhibited an attenuation of convective and diffusive
components of O 2 transport, both during maximal
exercise and knee extension.
How do these concepts adapt to HF with preserved
ejection fraction (HFpEF) syndrome? Whereas some
pathophysiology overlaps between the 2 HF pheno-
types, there is more uncertainty and unequivocal
definition on the O 2 transport/utilization-limiting
steps involved in the functional response of patients
with HFpEF. These patients are elderly and anemic in
a high rate of cases, and exhibit a decreased total
circulating red blood cell volume in 9 of 10 patients
affected by Hb-based anemia, thus presenting with
impaired blood O 2 carrying capacity (32). In addition,
they may show a defect in O 2 diffusion and lower
(a-v) O 2 (Figure 1C) (33).
COMPLEMENTARY CPET-DERIVED MEASURES OF
AEROBIC EFFICIENCY. Although peak V O 2 describes
the net limitation in exercise capacity, an even more
comprehensive idea of aerobic efficiency is obtained
by measuring the V O2 at the first ventilatory threshold
(V T ). VO2 at first VT reflects the metabolic condition
above which blood lactate and pH start to increase and
and the decrease, respectively, generating isocapnic compen-
satory acidotic buffering by bicarbonate (HCO3
).
There are 3 CPET-derived methods for detecting V O2 at
the first VT : the V-slope (i.e., the point at which the
oxidative incremental volume of carbon dioxide [CO2 ] released
[V CO2 ] becomes higher, as compared with VO2, due to
the additional CO 2 produced by lactic acid buffering);
the end-tidal CO 2 (PET CO2 ) versus the end-tidal O 2
(P ETO 2) method (when a divergent kinetic point of these
variables occurs with P ETO 2 progressively increasing
and the P ETCO2 slightly decreasing) and the minute
ventilation (VE)/VO2 versus VE/V CO2 ratio change in
1622 Guazzi et al. JACC VOL. 70, NO. 13, 2017

Cardiopulmonary Exercise Testing SEPTEMBER 26, 2017:1618–36

F I G U R E 2 VO 2 /WR Relationship and the Phenotypes Encountered in Cardiac Failure

A B
VO2 peak VO2 VO2
(average 30 sec)
shallow

VO2/WR slope
around 10 ml/
min x watt

35 sec WR WR

C D
VO2 VO2

flattening
downsloping

WR WR

(A) In normal individuals, the increase in the VO2/WR relationship is 10 ml/kg/min irrespective of the ramp protocol. Physical deconditioning
and initial cardiac limitation to exercise make this relationship shallow (B). Two patterns have been described in patients with advanced heart
failure: VO2 flattening (C) (i.e., loss of linearity from a certain point of exercise) and downsloping (D), which is tightly related to an acute drop
in cardiac output and blood pressure. VO2 ¼ peak oxygen consumption; WR ¼ work rate.

pattern of increase (identifying the point of contin-
uous increase in VE/VO 2 and stable VE/V CO2 kinetics).
The lack of V O2 at the first VT determination by any of
these methods occurs in around 10% of patients with
HF and carries a strong independent prognostic role
(34). Measuring gas exchange during submaximal
exercise response to constant workload is the gold
standard for studying the early VO2 kinetic time
constant (tau) during exercise or recovery of exercise
O 2 kinetics, which corresponds to 63% of the D V_ O2
from rest to steady state (exercise) or vice versa
(recovery) during a constant workload.
Tau provides additional objective information on
the mechanisms that control the ability to adapt to
and recover from exercise indicative of daily life
activity (35). Along with an impaired tau, VO2 may
typically fail to linearly increase relative to energy
demands as the work rate (WR) is increased (VO 2/WR),
resulting in delayed post-exercise recovery from a
maximal test. During a maximal test, VO 2/WR accu-
rately reflects the extent of aerobically regenerated
adenosine triphosphate. In physiological conditions
and nonobese people, the V O2/WR linearity corre-
sponds to a 10 ml/min increase
independently of the load imposed and slightly
changing according to exercise duration (Figure 2A).
These assumptions, however, are not true in CV dis-
orders, such as HF, and the pattern of V O 2 increase
may change in a shallow downward shift (Figure 2B)
or by VO 2 flattening at a given WR (Figure 2C). There is
then an alarming, rare condition with VO 2 down-
sloping associated with an acute drop in blood pres-
sure and CO (Figure 2D).
The last 2 patterns of V O2 generally match with a
similar O 2 pulse, which is the ratio of VO 2 to heart rate
and reflects the amount of O 2 extracted per heart
beat. The O 2 pulse provides an estimate of left ven-
tricular (LV) stroke-volume changes during exercise,
assuming that C(a-v)O2 is maximal and no anemia or
hypoxia is present. As the relative contribution of
stroke volume to CO is predominant during the initial
and intermediate phases of exercise, the O 2 pulse has
a typical hyperbolic profile, with a rapid rise during
the initial stages of exercise and a slow approach to an
asymptotic value at the end of exercise. Thus, once
V O2/WR changes in linearity, a flattening or even a
downward displacement of O 2 pulse kinetics during
per watt, progressive CPET would very likely reflect a
JACC VOL. 70, NO. 13, 2017
SEPTEMBER 26, 2017:1618–36
cardiogenic, rather than a peripheral vascular perfu-
sion/extraction, limitation to exercise performance.
Finally, the addition of systolic blood pressure to the
peak VO 2, as circulatory power, has been proposed
as having the potential to better investigate the
circulatory impairment (36).
VENTILATORY EFFICIENCY AND LUNG MECHANICS.
Cardiac patients may exhibit a lung mechanical–
related mechanism of exercise limitation, which is
tightly related to restrictive physiology due to
congestion and physical interaction between the
heart and the lung (37). However, detection of lung
mechanical dysfunction is often overlooked because
patients with HF often display a normal breathing
reserve, which is described by the relationship
between exercise VE and maximal breathing capacity,
as estimated by resting maximal voluntary ventila-
tion. Values <15% suggest a ventilatory limitation,
and may help to discriminate between patients with
HF and those with comorbid chronic obstructive
pulmonary disease (COPD) (38).
The impaired breathing reserve would, however,
be insensitive to mechanical ventilatory constraints
caused by differences in lung mechanics during ex-
ercise compared to the maximal voluntary ventilation
maneuver. For this reason, the most recent consensus
documents have introduced the use of flow-volume
loops as standard for the assessment of mechanical
VE limitation in cardiac versus lung disorders (2).
Indeed, flow-volume loops supplement basic breath-
ing mechanics information by identification of expi-
ratory airflow limitation and changes in operational
lung volumes during exercise.
Assessment of VE efficiency during CPET perhaps
provides the most relevant clues for addressing the
pathophysiological changes behind the impaired
exercise performance in HFrEF and HFpEF, especially
when associated with PH and right ventricular (RV)
dysfunction (39,40). Similar reasoning may, in part,
be applied to cases of post-embolic or idiopathic PAH.
An inefficient VE typically translates into an
abnormal rate of increase in the slope of VE increase
versus V CO2 production. This relationship shows a
near-linear increasing pattern that is determined by 3
factors: the amount of CO 2 produced; the physiolog-
ical dead space/tidal volume ratio (VD/V T); and the
arterial carbon dioxide partial pressure (PaCO2). This
relationship can be explained using the modified
alveolar equation:
VE ¼ 863  V CO2 /(Pa CO 2  [1
VD/V T])
For low and moderate intensities of work, the VE
response is tightly regulated by the PaCO2. At higher
work intensities, VE is affected by the increased
amount of VT over V D and the development of lactic
Guazzi et al. 1623
Cardiopulmonary Exercise Testing
acidosis and proton (H þ) production from the
prevailing anaerobic metabolism, which further
increases CO 2 release and the consequent amount of
VE. In HF, 3 different orders of mechanisms mediate
an impaired VE requirement for a given CO 2 produc-
tion: increased waste ventilation (41,42); early
occurrence of decompensated acidosis (41,43); and
abnormal chemoreflex and/or metaboreflex control
(44). Recent findings obtained in HFpEF document a
primary role of increased V D/VT in causing a steep
VE/VCO 2 slope (45). In cases of a comorbid condition
with COPD or COPD in isolation, high V D/V T may
prevail as a functional gas exchange abnormality
sustaining dyspnea sensation and hyperpnea (46).
The oxygen-uptake efficiency slope (OUES) is an
underused variable that reflects the global (pulmo-
nary, CV, and skeletal muscle) functional impairment
by combining VO 2 and VE/V CO2 slope. OUES is calcu-
lated on the logarithmic transformation of VE data in
liters/min (plotted on the x-axis), creating a linear
relationship with V O2 (plotted on the y-axis, where
VO2 ¼ log10 [VE þ b]). Given the tight linear relation-
ship the OUES creates between VE and VO2
throughout a progressive exercise test (47), this
calculation requires only submaximal effort.
Finally, an additional VE pattern at CPET evalua-
tion, which is typically unmasked by gas exchange
analysis, is an exercise oscillatory ventilatory (EOV)
pattern (Figure 3) and consists of a cyclic fluctuation
of VE and expired gas kinetics of variable amplitude,
frequency, and duration detectable in up to 30% of
symptomatic patients with HF (48). The source of this
ventilatory abnormality is still controversial, but its
ominous clinical and prognostic significance is clear
(49,50). At variance with PAH, this pattern seems
typical of HF and pulmonary interstitial congestion,
and a delayed circulatory transit time may explain the
observed differences (51).
APPLICATIONS OF CPET IN
CLINICAL PRACTICE
EVIDENCE IN PREVENTIVE MEDICINE AND REHABILITATION
PROGRAMS. Application of CPET to primary and
secondary prevention is challenging. Interestingly, a
role for CPET-derived data in detecting abnormal gas
exchange pattern phenotypes has been just recently
explored in the general population at CV risk.
An example is the European Exercise (EURO-EX)
population-based trial, whose preliminary data have
highlighted some cases of EOV, typically occurring in
the phenotype of elderly diabetic women (52).
Overall, despite a wealth of persuasive evidence
and numerous statements supporting the utility of
1624 Guazzi et al. JACC VOL. 70, NO. 13, 2017

Cardiopulmonary Exercise Testing SEPTEMBER 26, 2017:1618–36

F I G U R E 3 9-Plot CPET Report of a Case of HFrEF With EOV

1.5 80 1.5 1.5

60
1.0 1.0
VO2 (mL/min)

1.0

VCO2 (L/min)

VO2 (L/min)
VE (L/min)
40

0.5 0.5 0.5

20

0.0 0 0.0 0.0

0 10 20 30 40 0 100 200 300 0 100 200 300
Work Time (sec) Time (sec)

120 10 100
80
100 8 80
60
80
HR (bpm)

VE (L/min)

6 60

VE/VCO2
VE/VO2
60 40
4 40
40
20
2 20
20

0 0 0 0
0.0 0.4 0.8 1.2 0.0 0.4 0.8 1.2 0 100 200 300
VO2 (L/min) VCO2 Time (sec)

100 0.5 2.0 60 140 40

0.4
90 1.5 30

PETCO2 (mm Hg)

HR (beats/min)

PETO2 (mm Hg)

40 120
0.3
VO2/HR

VT (L)

RR

80 1.0 20
0.2
20 100
70 0.5 10
0.1

60 0.0 0.0 0 80 0
0 50 100 150 200 250 0 20 40 60 80 0 100 200 300
Time (sec) VE (L/min) Time (sec)

EOV is a specific VE abnormal phenotype occurring in approximately 30% of patients with mid-to-late manifestations of chronic heart failure, characterized by
cyclic fluctuation of VE and expired gas kinetics of variable amplitude, frequency, and duration. Because oscillatory manifestations may occur even in normal
subjects, criteria for an abnormal definition are an oscillatory pattern at rest that persists for $60% of the exercise test at an amplitude of $15% of the average
resting value (2). CPET ¼ cardiopulmonary exercise testing; EOV ¼ exercise oscillatory ventilation; HR ¼ heart rate; PETCO2 ¼ partial pressure of end-tidal carbon
dioxide; PETO2 ¼ partial pressure of end-tidal oxygen; RR ¼ respiratory rate; VCO2 ¼ volume of carbon dioxide released; VE ¼ ventilation; VO2 ¼ peak oxygen
consumption; VT ¼ tidal volume.

CPET in prevention and in early HF stages, practi- (but not exclusive) purpose of evaluating signs and
tioners have not generally adopted a portfolio of symptoms of coronary insufficiency (53). However,
variables that provides a 3-dimensional view of despite observations pointing out the high sensitivity
cardiorespiratory fitness with diagnostic and prog- and specificity of gas exchange analysis in detecting
nostic applicability, and an effective means to eval- suspected myocardial ischemia (53,54), few of them
uate therapeutic benefits. recognize a CPET role.
In the realm of patient care, exercise testing Implementation of rehabilitation and exercise
without the simultaneous collection of expired gases training (ET) programs is a Class I (55) to IIa (56) guide-
has seemingly taken on a static role, with the singular line recommendation. In this context, implementation
JACC VOL. 70, NO. 13, 2017
SEPTEMBER 26, 2017:1618–36
of CPET is 3-fold: 1) to plan correct exercise intensity
level or domain; 2) to assess benefits of exercise
prescription by monitoring gas exchange variables and
their related phenotypes; and 3) to identify subjects
who, despite adherence to the program, are non- or poor
responders to this multilevel intervention.
Although guidelines recommend the application of
objective exercise prescriptions using CPET data (57),
it is commonplace for programs without CPET capa-
bilities or with limited resources to establish exercise
intensity on the basis of resting heart rate (e.g.,
exercising heart rate threshold set 20 or 30 beats/min
above the resting heart rate). This simple method has
been criticized and demonstrated to be inadequate by
many. For example, Reed et al. (58) reported that only
26% and 38% of participants were exercising at the
recommended exercise intensity of 40% to 60% of
heart rate reserve (confirmed by CPET) when using
resting heart rate þ 20 and 30 beats/min, respectively.
Patients exercising at 30 beats/min or above their
resting heart rates experienced significantly greater
increases in 6-min walk distances compared
with patients exercising at 20 to 29 beats/min
above resting heart rate, suggesting that the imple-
mentation of CPET as a standard tool in cardiac
rehabilitation programs can really help to optimize
health-related outcomes.
Accordingly, the recent joint European Association
for Cardiovascular Prevention & Rehabilitation
(EACPR)/American Association of Cardiovascular and
Pulmonary Rehabilitation Statement provided
directions on the identification of ET intensity
domains by using constant WR exercise tests, looking
at the different kinetic responses to V O2 (57), which,
although it may be impractical in most cases, remains
the most accurate physiology-based approach. Briefly,
VO 2 kinetics during a constant WR exercise reflect 3
phases of adaptation of the organ systems and factors
involved in the alveolar-to-cell O2 coupling. The 3
phases are identified as: phase I, or cardiodynamic,
during which the increase in V O 2 is mediated by the
immediate increase in CO and pulmonary blood flow at
the start of exercise; phase II, or cell respiration,
reflecting the decreased O 2 content (muscle extrac-
tion) and increased venous CO2 content secondary to
increased cell respiration, as well as a further increase
in CO; and phase III, or the eventual steady-state
phase, during which an equilibrium is reached be-
tween O 2 extraction and the CO 2 production rate. If the
WR is above the subject’s first VT , the rate of increase
during phase III is not steady and correlates tightly
with the magnitude of lactate increase.
This background applies and allows for the precise
identification of 4 exercise training domains whose
Guazzi et al. 1625
Cardiopulmonary Exercise Testing
intensity is based on the physiology of O 2 uptake
kinetics: light to moderate; moderate to high;
high to severe; and severe to extreme. The light-to-
moderate– and moderate-to-high–intensity programs
are performed by continuous exercise, a condition
that is not sustainable for high-to-severe and severe-
to-extreme protocols, requiring an interval exercise
approach. Intensity domains of light to moderate
encompass the corresponding WR that engenders a
VO2 steady-state value below the corresponding first
VT . During this constant workload range and in this
domain, a VO 2 steady state is attained relatively
rapidly following the onset of exercise, and lactate is
not produced. For this reason, exercise can be very
well tolerated and is generally sustainable for long
periods of time (30 to 40 min) with only a mild sense
of fatigue and breathlessness. The moderate-to-high–
intensity domain corresponds to workloads between
the first and second VT s.
An improvement in functional capacity is the most
immediate and objective result of an effective ET
program. Although functional capacity is governed by
an integrated response of multiple organ systems that
are wholly or partly involved in this response, benefits
of ET interventions are generally quantified as changes
in VO 2. The physiological background behind exercise
capacity response and the progressive familiarization
of the cardiac rehabilitation personnel with the gas
exchange analysis technique points, however, to a
step forward that is a more in-depth analysis and
comprehensive interpretation of exercise intervention
trials. Indeed, attention to the determinants of exer-
cise improvement that may add to peak V O2 , and
potentially result in even more remarkable demon-
stration of efficacy, seems quite timely.
One example of this reasoning is the unexplored,
but perhaps intriguing, question of how much ET may
improve VO2 kinetics (tau and V O2/WR linearity),
rather than peak VO2 . Specifically, looking at favor-
able modifications that ET may induce in patterns of
VO2/WR (i.e., shallow, typical of deconditioned car-
diac patients, flattening, or downsloping associated
with acute drop in blood pressure and CO) (Figure 2).
Thus, an analysis of the effects of different exercise
training programs may be of value when considering
the potential to modify these abnormal patterns, even
when changes in peak V O2 per se are not remarkable.
DIAGNOSTIC ROLE OF CPET IN HF. As the severity of
CV disease advances, the role of CPET in the diag-
nostic setting of HF is not completely recognized. The
2013 American College of Cardiology (ACC)/American
Heart Association (AHA) guidelines for the manage-
ment of HF highlight the subjective nature of the ACC
Foundation/AHA stages of HF and the New York
1626 Guazzi et al.
Cardiopulmonary Exercise Testing
Heart Association functional classification system.
These are primarily based on patient interviews per-
taining to the “severity and triggers of dyspnea and
fatigue, presence of chest pain, exercise capacity as
well as physical and sexual activity” (59), and are
then followed by objective procedures for character-
ization of LV structure/function and evaluation of
diagnostic blood markers. Despite evidence of CPET
as a strong predictor of future HF development, its
consideration as a primary diagnostic tool is
cautioned, mainly due to the manifestations common
to other chronic conditions (e.g., PAH, congenital
heart defects, and interstitial lung disease) (60).
However, it seems more appropriate to use CPET
variables to guide the diagnosis of secondary patho-
physiological consequences of HF, irrespective of
etiology. Assessment of exercise ventilatory patterns
is particularly attractive in identifying left-sided PH
due to its physiological link with ventilation-
perfusion matching. Identification of left-sided PH is
particularly important, given that it is prevalent in
roughly 30% to 50% patients with overt HF and
carries a poor prognosis (61). The VE/V CO2 slope, the
PETCO 2, and the presence of EOV hold a high level of
prognostic utility (9). In a study of parallel evaluation
of pulmonary pressure, a VE/V CO2 slope of </$36 was
found to be the strongest predictor (odds ratio [OR]:
12.1; p < 0.001) of a pulmonary artery pressure
(PAP) $40 mm Hg, followed by peak exercise PETCO2
(</$36 mm Hg; OR: 3.8; p < 0.001), and presence of
EOV (OR: 3.2; p < 0.001) in patients with HFrEF. More
compelling was the greater predictive utility when
all measures were considered together (OR: 16.7;
p < 0.001) (9). There have been significantly fewer
studies evaluating the relationship between disease
severity and CPET variables in patients with HFpEF,
although most recent observations have shed new
light on VE efficiency (45,62). Considering that the
VE/V CO2 relationship possesses prognostic value at
submaximal levels of effort, this observation in-
creases the feasibility of implementing CPET for
prognostic purposes, particularly in an elderly popu-
lation that is likely not accustomed to exercise or may
be hesitant to provide maximal effort.
Collectively, the aforementioned CPET indexes
should not be thought of as markers to identify
specific pathophysiological conditions associated
with HF, but rather for use of objective CPET data to
help guide characterization of the HF disease state,
which is staged, in part, by subjective determinants.
Interpreting CPET outcomes in this manner helps
confirming HF and identifying secondary issues and
may provide a clear objective in deciding the best
course of patient treatment.
JACC VOL. 70, NO. 13, 2017
SEPTEMBER 26, 2017:1618–36
MONITORING THERAPEUTIC EFFICACY WITH CPET. The
ACC/AHA stages of HF provide a comprehensive
outline of patient characteristics that define the stage
of HF, as well as respective therapies and their asso-
ciated goals. Many, if not all, therapeutic goals aim to
mitigate HF-related symptoms (e.g., dyspnea at rest
and/or upon exertion, fatigue), reduce hospital
readmission rates, and improve survival. The acute
efficacy of a number of therapeutic regimens are
typically verified by standard follow-up visits that
assess the patient’s resting blood pressure, weight,
blood markers, and patient-reported changes in
symptom severity/presence. This system of patient
management is generally considered to be adequate
in most clinical settings, and therefore leaves many
practitioners with a lack of clarity as to the added
value of CPET. However, an important consideration
is that follow-up visit examinations are done at rest,
thus providing no information on symptoms and
on hemodynamic measures during physical exertion
(when symptoms typically become evident).
Furthermore, the main aim of any intervention is to
improve prognosis, limit morbidity, and increase the
quantity of years a patient is able to live a higher
quality of life, all of which are markers associated
with high peak V O2 values and lower VE/VCO2 slopes.
Furthermore, the current ACC/AHA guidelines
recognize the therapeutic benefits of various phar-
macological interventions to improve cardiorespira-
tory fitness. Angiotensin-converting enzyme (ACE)
inhibitors or angiotensin II receptor blockers (ARBs)
significantly improve peak V O2 on their own and have
led to greater enhancements when taken together
(63). These improvements also translate to reductions
in the VE/VCO2 slope with ACE inhibitor therapy;
however, there are conflicting outcomes with ARB
therapy (64). The effects of b-blockers on exercise
performance have been extensively studied, and
CPET application has garnered relevant information,
in terms of influence of different types of b-blockers
on gas exchange (65). Moreover, decreases in PAP,
with concomitant decreases in the VE/V CO 2 slope (66)
and some reversal of EOV (67), have been demon-
strated with sildenafil therapy in patients with
HFrEF. By contrast, CPET outcomes in response to
ACE inhibitor, ARB, or sildenafil therapy in patients
with HFpEF have not been as favorable (68–70). The
same seems to be true for ivabradine (71).
Given the associated improvements in key prog-
nostic CPET variables with the initiation of pharma-
cological therapy in certain patient populations, it
may be interesting to explore the feasibility of
titrating medications based on CPET “responders” or
“nonresponders” in future clinical trials. Peak V O 2 is
JACC VOL. 70, NO. 13, 2017
SEPTEMBER 26, 2017:1618–36
an endpoint in trials of cardiac resynchronization
therapy, LV assist device implantation, and valve
surgical or percutaneous treatments, even though
CPET-derived variables are not standardized criteria
for these indications (2).
The frequency at which a patient should perform
CPET evaluation to monitor the effectiveness of
interventions is not well-defined. Certainly, in clini-
cally stable patients, CPET should be considered at
2- to 4-year intervals, whereas in patients with signs
and symptoms, the test should occur in a time frame
that has been reported to cause significant improve-
ments in the variable of interest (2).
PROGNOSTIC UTILITY OF CPET. Among the exten-
sive list of criteria for determining candidates for
transplantation, CPET variables are recognized to
help refine the selection process. At time of Mancini
et al.’s investigation (6), demonstrating a higher
1-year survival rate in HF patients with peak V O 2
>14 ml $ kg
1 $ min
1 compared with patients with
peak VO 2 #14 ml$kg
1 $min
1 , and a similar 1-year
survival rate of transplanted patients compared with
those above the 14 ml$kg
1 $min
1 threshold,
b-blocker therapy was not part of routine clinical
management of patients with HF, as it became in the
early 2000s. Despite the marginal effects of
b-blockers on peak VO2, survival rates were signifi-
cantly improved, and therefore prompted reconsid-
eration of Mancini et al.’s proposed cutoff for
transplantation. Consequently, a threshold of
10 ml$kg
1 $min
1 was proposed to be the optimal
prognostic threshold for transplantation consider-
ation for b -blocked patients with HFrEF (72). In
addition to these landmark studies, a continuing
wealth of evidence has fortified the prognostic
strength of peak V O2, so much so that the Interna-
tional Society for Heart Lung Transplantation pro-
vides a Class I, Level of Evidence: B recommendation
for inclusion of maximal CPET (RER > 1.05) to guide
transplant listing, with peak V O 2 #14 ml$kg
1 $min
1
(Class I, Level of Evidence: B) in the absence of a
b-blocker and a peak VO2 #12 ml$kg
1$min
1 (Class I,
Level of Evidence: B) in its presence (73). However, an
equally impressive and convincing body of published
reports identifies ventilatory efficiency (e.g., the
VE/V CO2 slope and EOV) as a more powerful prog-
nostic index than peak V O2 (74). Despite the growing
body of evidence highlighting the greater prognostic
utility of the VE/V CO2 slope, current ACC/AHA guide-
lines recognize peak VO2 as the sole CPET outcome for
transplant consideration. Similarly, the International
Society for Heart Lung Transplantation recommended
that the VE/V CO2 slope be reserved (slope >35; Class
IIb, Level of Evidence: C) for CPET performances with
Guazzi et al. 1627
Cardiopulmonary Exercise Testing
a submaximal effort (RER <1.05). Both sets of rec-
ommendations are also similar in that they recom-
mend the use of one of the CPET parameters, even
though there is strong evidence for the increased
discriminatory power of the VE/V CO 2 slope and other
variables considered together (2,10,75).
M u l t i p a r a m e t r i c C P E T d a t a t a b l e . It is evident
that across the years, there has been a surge of CPET
variables, making it difficult for clinicians to identify
and interpret measures of interest. To ameliorate this
concern, early efforts by Wasserman (24) (Figure 3)
provided 9 visually friendly panel plots incorporating
the CPET variables of interest in a single-page report.
A potential disadvantage of this method, however,
was the lack of guidance or references to variable
thresholds pertinent to CPET clinical application.
Accordingly, Arena et al. (76) developed an early
iteration of the figures by proposing color-coded
interpretive tables applied to different diseases. An
example is shown in Table 1 (for HF), where the list of
CPET variables (the VE/V CO 2 slope, peak V O2, EOV,
resting P ETCO2) was separated into primary (e.g., the
VE/VCO 2 slope and peak V O2) and secondary (e.g., EOV
and resting PET CO2 ) CPET measures of interest, along
with respective categories of risk. Furthermore,
instructions were provided to help with guiding the
interpretation of multivariable CPET results. This
approach took valuable initial steps toward
condensing a large amount of CPET-related informa-
tion into a single table that could be used by clinicians
in patient management. Many modifications were
made and subsequently integrated and presented in
the 2012 EACPR/AHA Scientific Statement (1). The
color-coded tables now provide striking visual clues
to facilitate the interpretation of CPET. Normal re-
sponses are identified in green, whereas intensifying
abnormalities are highlighted in yellow, orange, and
red. Expanded clinical recommendations for the
overall severity of the patient’s condition are pro-
vided, based on the frequency of color occurrence in
the table. Of note, the decision to include the
respective CPET variables in the outcome tables was
based on the available evidence, summarized in
Table 2, for the predictive application of each measure
independently, but more importantly, collectively
across populations. Moreover, these tables were
developed with the objective of being applicable and
relevant to the qualifications for CPET (e.g., prog-
nosis, therapeutic efficacy, and diagnosis) in order
to promote their use. As such, the AHA, and EACPR
have made an effort to increase the visibility of a
multivariable paradigm that implements visual clues
to trigger appropriate actions. Because no studies
have yet tested the feasibility of this model’s
1628 Guazzi et al. JACC VOL. 70, NO. 13, 2017

Cardiopulmonary Exercise Testing SEPTEMBER 26, 2017:1618–36

T A B L E 1 Clinical Stratification for Patients With HF

Primary CPET Variables

VE/VCO2 Slope Peak VO2 EOV PETCO2

Ventilatory Class I Ventilatory Class A Not Present Resting PETCO2

$33.0 mm Hg
VE/VCO2 slope <30.0 Peak VO2 >20.0 ml$kg
1$min
1 3–8 mm Hg increase during ET
Ventilatory Class II Ventilatory Class B

VE/VCO2 slope 30.0–35.9 Peak VO2 ¼ 16.0–20.0 ml$kg
1$min
1

Ventilatory Class III Ventilatory Class C Present Resting PETCO2
<33.0 mm Hg
VE/VCO2 slope 36.0–44.9 Peak VO2 ¼ 10.0–15.9 ml$kg
1$min
1 <3 mm Hg increase during ET
Ventilatory Class IV Ventilatory Class D

VE/VCO2 slope $45.0 Peak <10.0 ml$kg
1$min
1

Standard ET Variables

Hemodynamics ECG HRR

Rise in systolic BP during ET
Flat systolic BP response during ET
Drop in systolic BP during ET
No sustained arrhythmias, ectopic foci, and/or ST-segment >12 beats at 1 min recovery
changes during ET and/or in recovery
Altered rhythm, ectopic foci, and or ST-segment changes during #12 beats at 1 min recovery
ET and/or in recovery: did not lead to test termination
Altered rhythm, ectopic foci, and/or ST-segment changes during
ET and/or in recovery: led to test termination

Patient Reason for Test Termination

Lower extremity muscle fatigue Angina Dyspnea

Interpretation

 All variables in green: excellent prognosis in the next 1–4 years ($90% event-free)
B Maintain medical management and retest in 4 years

 Greater number of CPET and standard ET variables in red/yellow/orange indicative of progressively worse prognosis.
B All CPET variables in red: risk for major adverse event extremely high in next 1–4 years (>50%)

 Greater number of CPET and standard ET variables in red/yellow/orange indicative of increasing HF disease severity.
B All CPET variables in red: expected significantly diminished cardiac output, elevated neurohormones, higher potential for secondary PH.

 Greater number of CPET and standard ET variables in red/yellow/orange warrants strong consideration of more aggressive medical
management and surgical options.

Example of a color-coded table for clinical stratification that applies to patients with either HFrEF or HFpEF. A list of evidence-based CPET variables (the VE/VCO2 slope, peak VO2,
EOV, resting PETCO2) are separated into primary (e.g., the VE/VCO2 slope and peak VO2) and secondary (e.g., EOV and resting PETCO2) measures of interest, along with respective
color-code categorizations of risk (2).
BP ¼ blood pressure; CPET ¼ cardiopulmonary exercise testing; ECG ¼ electrocardiogram; EOV ¼ exercise oscillatory ventilation; ET ¼ exercise testing; HF ¼ heart failure;
HFrEF ¼ heart failure with reduced ejection fraction; HFpEF ¼ heart failure with preserved ejection fraction; HRR ¼ heart rate recovery; PETCO2 ¼ partial pressure of end-tidal
carbon dioxide; PH ¼ pulmonary hypertension; VE/VCO2 ¼ minute ventilation/carbon dioxide production; VO2 ¼ oxygen consumption.

implementation in a clinical setting, as well as its
effect on patient management, future studies are
encouraged to do so.
REAPPRAISAL OF INVASIVE CPET
Invasive CPET (iCPET) allows better characterization
of the hemodynamic reasons for exercise limitations,
precisely dissecting central and peripheral mecha-
nisms. As this approach has the potential to accu-
rately phenotype cardiopulmonary disorders, it has
been reappraised since its initial use in early 1980s as
a routine approach in more
Measuring pulmonary hemodynamics, LV filling
pressures, Fick CO, and a-v difference in O2, become
essential in cases of unexplained exercise-induced
dyspnea (78) and in the evaluation of exercise-
induced PH of either idiopathic or secondary origin (79).
A recent analysis documented that iCPET allows
to considerably shorten the time lag from manifes-
tations of unexplained exercise-induced dyspnea to
an organ-specific diagnosis. Conditions that can be
ultimately diagnosed with iCPET are HFpEF, mito-
chondrial myopathy, and confirmation of decondi-
tioning (78).
Pulmonary hemodynamic measurements during
exercise, especially of PAP and pulmonary arterial
wedge pressure (PAWP), have incremental prognostic
value compared with evaluation at rest (79).
Borlaug et al. (80) first reported the potential to
laboratories (77). diagnose HFpEF in symptomatic euvolemic patients
with normal levels of B-type natriuretic peptide and
without clear signs and symptoms of HF at rest. In
one-half of cases, looking at the PAWP and end-
diastolic LV pressure increase, the observed increase
in PAWP during exercise was diagnostic for HFpEF.
JACC VOL. 70, NO. 13, 2017
SEPTEMBER 26, 2017:1618–36
These data were reproduced by others, showing
that in patients with normal biventricular function, an
increased mean PCWP >20 mm Hg at peak exercise, in
the absence of elevations in pulmonary vascular
resistance, suggests exercise-induced HFpEF (81).
Patients with early stages of PAH and still normal
cardiopulmonary hemodynamics at rest demonstrate
increased mean PAP >30 mm Hg and pulmonary
vascular resistance >80 to 120 dyne $ s $ cm5 on
iCPET (82).
Exertional intolerance may be associated with pre-
load–dependent limitations to stroke volume and CO
(83). In this patient population, failure to augment
right atrial pressure on exercise is observed despite
abnormally decreased CO. Finally, impaired systemic
O 2 extraction indicates left-to-right shunting or a
mitochondrial myopathy. An algorithm for diag-
nosing these conditions was recently proposed by
Huang et al. (78).
Accurate assessment of the pressure-flow relation-
ship during exercise by plotting mean PAP versus CO
provides robust indication of abnormalities in RV to
pulmonary circulation coupling (79). A mean PAP/CO
relationship >3 mm Hg/l/min is reflective of a PH
response, which combines with an elevated VE/VCO 2
slope (39). Even more, occurrence of RV-to-pulmonary
circulation (PC) uncoupling is responsible for a
delayed V O2 on kinetics during early exercise (84).
The most relevant limitations to this approach are
its invasive nature and technical inaccuracies in
obtaining reliable pulmonary hemodynamic tracings
during high respiratory frequencies, despite aver-
aging of repeated measures.
CPET IMAGING: A NEW FRONTIER
CPET imaging is a quite recent and valuable testing
modality, which is receiving attention for its potential
to combine exercise physiological data with noninva-
sive recordings of cardiac function by measures of
chamber volumes and geometry, valvular status, and
systolic and diastolic function, including the evalua-
tion of left atrium (LA) function (85). In addition, the
assessment of cardiac “functional reserve” by CPET
imaging is greatly improved by the study of the path-
ophysiological response of the pulmonary circulation
to exercise, whose clinical implications appear com-
plementary to and synergistic with the information
obtained with iCPET (86,87). A typical example of this
concept is the identification of a basic role of RV to
pulmonary circulation uncoupling in determining the
flattening of the D VO 2/D WR relationship (88).
On a methodological basis, there is not yet stan-
dardization, but the semirecumbent position is
Guazzi et al. 1629
Cardiopulmonary Exercise Testing
T A B L E 2 Class Recommendation and Level of Evidence for CPET Applications in HF
Level of
Classification Evidence
Prognostic Recommendations
CPET should be performed in patients with HFrEF being Class I A
considered for heart transplantation or mechanical device
implantation. However, CPET provides robust prognostic
information in all patients with HFrEF and therefore is also
recommended in those not being considered for end-stage
surgical management.
Primary CPET variables, including the VE/VCO2 slope, peak VO2,
and EOV, are strong predictors of adverse events. The
response of these 3 CPET variables is recommended to form
the basis of the prognostic assessment in patients with
HFrEF. A combination of a VE/VCO2 slope $45.0, peak
VO2 #10 ml$kg
1$min
1, and the presence of EOV carries a
particularly poor prognosis.
Secondary CPET variables, including PETCO2, O2 pulse, systolic Class IIa B
blood pressure, and the ECG response to exercise, are
additional predictors of adverse events and can be useful
during the prognostic assessment in patients with HFrEF. A
resting PETCO2 <33 mm Hg, a rise in PETCO2 <3 mm Hg during
exercise, a drop in systolic blood pressure during exercise,
and/or ECG abnormalities warranting termination of exercise
indicate a worse prognosis.
Prognostic value in patients with HFpEF is showing initial Class IIa B
promise, in particular, the VE/VCO2 slope, peak VO2, and EOV.
These variables can be useful in providing prognostic
information also in patients with HFpEF.
Use of a multivariate model composed of key CPET variables and Class IIa B
thresholds, as illustrated in Figure 4, can be useful in
improving prognostic resolution in patients with HFrEF in
comparison with variables assessed independently.
Recommendations for Gauging Therapeutic Efficacy
Primary CPET variables, including the VE/VCO2 slope and peak Class I A
VO2, are responsive to numerous pharmacological, surgical,
and exercise interventions in patients with HFrEF. As such,
assessment of the change in the VE/VCO2 slope and peak VO2
is recommended when gauging therapeutic efficacy in
patients with HFrEF in both clinical and research settings.
Reversal of EOV also may occur with pharmacological and Class IIa B
exercise interventions in patients with HFrEF. Therefore,
assessing for reversal of EOV as a gauge of therapeutic
efficacy can be useful in both clinical and research settings.
Primary CPET variables, including the VE/VCO2 slope and peak Class IIa C
VO2, may be responsive to pharmacological, surgical, and
exercise interventions in patients with HFpEF. As such, the
VE/VCO2 slope and peak VO2 can be useful as core variables
when gauging therapeutic efficacy in patients with HFpEF in
both clinical and research settings.
Diagnostic Recommendations
CPET variables reflecting ventilatory efficiency, including the Class IIb B
VE/VCO2 slope, EOV, and PETCO2 at rest and during exercise,
may be considered in detecting left-sided PH in patients with
HFrEF and HFpEF. If all 3 of these variables are in their
respective red zones, as illustrated in Table 1, the suspicion
that the patient has left-sided PH may be reasonable. A
primary indication of CPET for diagnostic purposes cannot be
recommended at this time. Diagnostic assessments may be
considered in patients with HF and suspicion of PH.
Abbreviations as in Table 1.
suggested for a better Doppler evaluation of param-
eters that are strictly dependent on workload (i.e.,
systolic pulmonary pressure, transvalvular gradients,
LV outflow gradients) and decline quickly during the
first seconds of recovery. Feasibility is strictly
dependent on the acoustic window and its variability
1630 Guazzi et al.
Cardiopulmonary Exercise Testing
during exercise caused by breath, body motion, and
heart movements.
Incremental ramps at low increasing workload (8 to
15 W/min) are preferable for fully acquiring images
every 2 min. Loop storage of adequate duration (3 to 5
beats) is required in order to perform the averaging
of measures, especially for Doppler parameters,
accounting for physiological respiratory variations.
Figure 4 summarizes the main parameters provided
by CPET imaging, pointing out their applications to
specific cardiac diseases, with some illustrative cases
of HFrEF, HFpEF, mitral valve insufficiency, and
hypertrophic cardiomyopathy.
HF WITH REDUCED EJECTION FRACTION. There are
multiple applications of CPET imaging in HF. First, a
good discrimination between circulatory rather
deconditioning limitation is signaled by the inability
to reach an exercise stroke volume >50 ml/m 2 (89).
The importance of measuring CO reserve (90,91) in
parallel with LV compliance and filling pressures is
well-established (92). Furthermore, estimation of O 2
peripheral extraction by CPET imaging using the
echocardiographic-derived Fick equation is an im-
mediate method to explore the role of the periphery
(90). Assessment of functional mitral regurgitation
(MR) is of basic relevance as a marker of severity and
adverse outcome (93), which, in early stages, may be
detectable just during exercise (20). Exercise-induced
MR determines the same extent of exercise limitation
observed in patients with severe MR at rest. Dynamic
PH is a further limiting mechanism, also induced by
MR, and is associated with specific ventilatory phe-
notypes of gas exchange, such as EOV (20). In
advanced HFrEF, assessment of exercise-induced RV
contractile reserve may be investigated by CPET
imaging. In a study of 97 patients with advanced
HFrEF, RV exercise contractile reserve and RV-to-PC
coupling response to maximal exercise were
analyzed through the relationships of systolic pul-
monary arterial pressure (sPAP) to tricuspid annular
peak systolic excursion (TAPSE) and sPAP to CO using
stress echocardiography and CPET. Categorization
into 3 groups based on TAPSE at rest $16 mm (group A,
n ¼ 60) and patients with TAPSE at rest <16 mm,
who were further divided into 2 subgroups (group B,
n ¼ 19; group C, TAPSE <15.5 mm, n ¼ 18) according to
whether their respective median TAPSE was higher or
lower than 15.5 mm at peak exercise. Group B, at
variance with group C, showed an upward shift of the
TAPSE versus sPAP relationship and some degree of
favorable coupling adaptation during exercise. Thus,
severely impaired RV function at rest may still be
associated with the capacity to improve RV-to-PC
coupling during exertion in a proportion of the
JACC VOL. 70, NO. 13, 2017
SEPTEMBER 26, 2017:1618–36
patients with HFrEF (87). Interestingly, the worst
RV-to-PC coupling pattern was associated with the
highest rate of exercise ventilation inefficiency. This
analysis appears useful for unmasking different right
heart phenotypes not detectable with rest evaluation.
Most recent findings suggest that impaired LA
dynamics, estimated by LA strain response during
maximal exercise and during the recovery phase,
plays a central hemodynamic role in triggering RV to
PC uncoupling and ventilatory inefficiency, especially
when LV systolic function is reduced (85).
HF WITH PRESERVED EJECTION FRACTION. HFpEF
is a syndrome whose diagnostic process and limited
therapeutic options remain major challenges (94).
The expanding body of published reports on CPET
imaging in this syndrome is based on several points:
to uncover unexplained dyspnea and suspected
HFpEF; to confirm HFpEF diagnosis in subjects not
meeting the criteria required at rest; and to stratify
CV risk once a true diagnosis is established.
Recently, the American Society of Echocardiogra-
phy/European Association of Cardiovascular Imaging
recommendations on diastolic function evaluation
advocated the use of the exercise E/e0 ratio to detect
diastolic dysfunction (95) and, for the past several
years, changes in E/e 0 were taken as a reference to
interpret unexplained dyspnea and suspected HFpEF
(96,97). In a study by Nedelikovic et al. (98), per-
formed in 87 patients with dyspnea unexplained by
exercise and hypertension, gas exchange and Doppler
analysis led to the ultimate diagnosis of HFpEF in the
minority of patients presenting with a combination
of E/e 0 >15 at peak exercise, along with an elevated
VE/V CO2 slope.
A parallel number of observations have focused on
how exercise E/e0 may add to the clinical follow-up of
established HFpEF, especially looking at the associ-
ated changes in pulmonary pressure during effort
(99,100). Interestingly, CPET imaging with immedi-
ately post-exercise echocardiographic assessment has
documented that in about 1 of 3 patients with HFpEF,
exercise elevation of LV filling pressure (E/e 0 EXE >13)
can be absent, but associated with reduced contrac-
tile reserve and ventriculoarterial coupling, as well as
with a mild degree of functional impairment (101).
However, a similar approach has shown that peak
E/e 0 >15 occurred in 9% of hypertensive patients
with exertional dyspnea and normal resting LV
function, being associated with lower peak VO2 and
significantly impaired ventilator efficiency (higher
VE/V CO2 ), compared with patients with normal
exercise E/e 0 (98).
The intrinsic value of exercise echocardiography
has been recently confirmed in an elegant
JACC VOL. 70, NO. 13, 2017 Guazzi et al. 1631
SEPTEMBER 26, 2017:1618–36 Cardiopulmonary Exercise Testing

F I G U R E 4 Example of CPET Imaging Applied to Different Cardiac Diseases

The figure shows the semirecumbent approach (center), allowing for simultaneous echocardiography and expired-gas analysis. The yellow table on the right lists the
main parameters provided by exercise echocardiography; the disease-specific tables indicate prognostic parameters to collect during the test. The blue table on the
left lists the main CPET parameters; the “red flags” table (left) points out CPET pathological patterns with prognostic impact. The 4 panels in blue depict cases of iCPET
with the main CPET (left blue frame) and echocardiographic (right red frames) findings. (Upper left) A case of HFrEF due to post-ischemic LV dysfunction with a very
abnormal functional phenotype; CPET frames show pronounced exercise oscillatory ventilation and severe impairment of ventilatory efficiency; echocardiographic
frames show end-diastolic and end-systolic apical 4-chamber views with apical aneurysmatic evolution and severely reduced ejection fraction, transmitral pulsed wave
Doppler, and annular TDI displaying grade III diastolic dysfunction and transtricuspid continuous wave Doppler demonstrating severe pulmonary hypertension at rest
and during exercise. (Upper right) A case of HCM with preserved functional phenotype, despite the evidence of severe left ventricular outflow obstruction; CPET
frames show normal ventilatory response; echocardiographic frames show left ventricular outflow tract continuous wave Doppler demonstrating severe rest dynamic
obstruction, enhanced by the Valsalva maneuver and by post-exercise period (when afterload is decreasing, but adrenergic inotropic drive is still present). (Bottom
right) A case of HFpEF with typical abnormal functional phenotype characterized by chronotropic incompetence; CPET frames show the flattening of the DVO2/DWR
relationship, expression of blunted cardiac output during exercise, and the reduced chronotropic response related to VO2 consumption; echocardiographic frames show
transmitral pulsed wave Doppler and annular TDI displaying rest grade II diastolic dysfunction with worsening during exercise, transtricuspid continuous wave
Doppler demonstrating a borderline rest gradient with a severe exercise-induced increase, and left ventricular outflow tract pulsed wave Doppler evidencing a slight
increase of flow during exercise, consistent with abnormal stroke volume reserve. (Bottom left) A case of moderate left ventricular systolic dysfunction characterized by
exercise-induced mitral regurgitation with mild exercise oscillatory ventilation. CPET frames show mild exercise oscillatory ventilation with preserved ventilatory
efficiency; echocardiographic frames show rest and exercise midsystolic color Doppler apical 4-chamber views with severe exercise-induced mitral regurgitation and
transtricuspid continuous wave Doppler demonstrating normal rest gradient with abnormal exercise-induced increase, reflecting dynamic pulmonary hypertension.
AT ¼ anaerobic threshold; AVA ¼ aortic valve area; CO2 ¼ carbon dioxide; EDA ¼ end-diastolic area; EDV ¼ end-diastolic volume; ERO ¼ effective regurgitant orifice;
ESA ¼ end-systolic area; ESV ¼ end-systolic volume; EXE ¼ exercise; FAC ¼ fractional area change; GLS ¼ global longitudinal strain; HCM ¼ hypertrophic cardio-
myopathy; iCPET ¼ invasive cardiopulmonary exercise testing; iEDV ¼ indexed end-diastolic volume; iESV ¼ indexed end-systolic volume; LVEF ¼ left ventricular
ejection fraction; MPG ¼ mean pressure gradient; MG ¼ mean gradient; MR ¼ mitral regurgitation; PG ¼ peak gradient; RegVol ¼ mitral regurgitant volume;
RER ¼ respiratory exchange ratio; RV ¼ right ventricular; SPAP ¼ systolic pulmonary artery pressure; SV ¼ systolic volume; TAPSE ¼ tricuspid annular plane systolic
excursion; TDI ¼ tissue Doppler imaging; WMSI ¼ wall motion score index; other abbreviations as in Figures 1 to 3.
1632 Guazzi et al.
Cardiopulmonary Exercise Testing
simultaneous invasive and echocardiographic study
reporting increased sensitivity, but compromised
specificity of guideline-recommended use of the
exercise E/e0 ratio (77), proposing a screening role of
exercise echocardiography aimed at ruling out the
presence of HF.
VALVULAR DISEASES. Exercise echocardiography is
extensively used for the evaluation of valvular dis-
eases, as recently stated in the European Association
of Cardiovascular Imaging/American Society of
Echocardiography recommendations (95). Specific
protocols and target parameters have been estab-
lished according to the valve disease and the clinical
question (102). The main goal of physical stress is to
measure the “valve reserve” (i.e., the entity of
exercise-induced worsening of valve disease) and the
direct pathophysiological consequences (i.e.,
exercise-induced PH, LV transient systolic dysfunc-
tion, increase in transvalvular gradients). The main
indication for testing patients with valvular disease is
an unclear clinical presentation. Considering the
advanced age of people with valvular disease and
their high prevalence of dyspnea (103), it is quite
common to experience some difficulties in under-
standing the origin of self-reported dyspnea. In the
decision process for surgical treatment, what appears
pivotal is the recognition of how much symptoms are
related to the valve disease or reflective of other
conditions, in order to optimize surgical timing. Of
note, CPET imaging was recently used in patients
with rheumatic mitral stenosis, providing evidence
that functional limitation has a complex origin.
Restrictive lung function, chronotropic incompe-
tence, limited stroke volume reserve, and peripheral
factors equally contributed to reduce ability to exer-
cise (104). Interestingly, patients with attenuated
functional responses showed the expected exercise-
induced PH; however, this was not the main
limiting factor. A similar combined approach has been
used in functional MR to demonstrate that exercise-
induced severe MR limits patients with HFrEF,
mainly causing exercise PH. The rationale for using
CPET imaging in valvular disease is strong; however,
systematic data are still lacking and future in-
vestigations are desirable for defining surgical risk,
especially in the assessment of complex conditions,
such as paradoxical low-flow–low-gradient severe
aortic stenosis (105).
HYPERTROPHIC CARDIOMYOPATHY. Hypertrophic
cardiomyopathy (HCM) is commonly associated with
exercise intolerance, and CPET analysis is helpful for
improving prognostic stratification (106). Mechanisms
underlying abnormal gas exchange during exercise
JACC VOL. 70, NO. 13, 2017
SEPTEMBER 26, 2017:1618–36
can differ according to the predominant echocardio-
graphic phenotype (107). CPET imaging allows the
identification of LV outflow obstruction and diastolic
dysfunction assessment during exercise, as well as
objective measurement of functional limitations.
In 156 patients with HCM evaluated with both CPET
and noninvasive hemodynamic assessment, a high
prevalence of exercise intolerance and ventilatory
inefficiency has been reported (108). Impaired peak
V O2 (<80% of predicted) was associated with peak
cardiac index, age, male sex, and RV end-diastolic
area, whereas ventilatory inefficiency (VE/VCO2 >34)
was related to E/e 0 and to indexed LA volume. Both
functional parameters were predictive of increased
risk of major events in the short-term follow-up.
The role of diastolic dysfunction as a limiting factor
of functional response has been reinforced by a recent
paper using iCPET imaging in 197 patients with HCM
(109). Multivariate analysis showed that LV obstruc-
tion, LA dilation, and especially rest or latent
diastolic dysfunction were the main determinants of
exercise intolerance.
According to the European guidelines (110), CPET
has a Class IIa indication in patients with HCM,
irrespective of symptoms reported, and a Class Ia
indication when consistent symptoms are present.
Simultaneous exercise echocardiography has the
advantage of evaluating LV outflow tract obstruction
and weighing all potential causes of exercise intoler-
ance, such as diastolic dysfunction, dynamic MR, and
PH. An additive value of CPET imaging has been
suggested for monitoring therapeutic response to
septal reduction, looking at the linearity of the D VO 2/
DWR relationship, which can be flattened in case of
severe obstruction and normalize after septal reduc-
tion (111).
SUSPECTED OR CONFIRMED PH. CPET imaging and
exercise echocardiography have independent, spe-
cific roles in the diagnostic work-up, functional
evaluation, and clinical management of patients with
confirmed PH, irrespective of disease origin (95).
Nonetheless, the combination of both tests in a
single-step evaluation has the advantage of confirm-
ing suspected PH, making an early diagnosis, identi-
fying secondary causes of PH, and improving overall
patient characterization. Systolic pulmonary pressure
during exercise is one of the most important variables
to collect during exercise echocardiography in several
clinical settings because it reflects the main mecha-
nism of effort intolerance. Nonetheless, in the
context of Group 1 or PAH, the role of exercise echo-
cardiography in the diagnostic work-up is still
debated and it is not recommended for diagnosis and
clinical purposes (112).
JACC VOL. 70, NO. 13, 2017 Guazzi et al. 1633
SEPTEMBER 26, 2017:1618–36 Cardiopulmonary Exercise Testing

The role of CPET imaging has been validated in
several conditions for determining exercise-induced
PH, however no prospective studies have tested the
incremental value of this combined approach in the
diagnostic work-up of PAH.
CONCLUSIONS
CPET now has a definitive place in the armamen-
tarium of the practicing clinician for the evaluation of
cardiopulmonary disorders, primarily HF. It provides
a thorough assessment of the integrative multiorgan
physiological response to exercise. A revival of
invasive CPET and introduction of CPET imaging have
now extended the amount of pathophysiological
and clinical information, providing new insights in
systemic and pulmonary hemodynamics, along with
direct knowledge of cardiac, valve, and functional
data.
ADDRESS FOR CORRESPONDENCE: Dr. Marco Guazzi,
University of Milano, Department of Biomedical Sciences
for Health, Heart Failure Unit, University Cardiology
Department, IRCCS Policlinico San Donato, Piazza E.
Malan 2, 20097 San Donato Milanese, Milan, Italy.
E-mail: marco.guazzi@unimi.it.

REFERENCES

1. Guazzi M, Adams V, Conraads V, et al. EACPR/
AHA Scientific Statement. Clinical recommenda-
tions for cardiopulmonary exercise testing data
assessment in specific patient populations. Circu-
lation 2012;126:2261–74.
2. Guazzi M, Arena R, Halle M, Piepoli MF, Myers J,
Lavie CJ. 2016 focused update: clinical recom-
mendations for cardiopulmonary exercise testing
data assessment in specific patient populations.
Circulation 2016;133:e694–711.
3. Barron A, Dhutia N, Mayet J, Hughes AD,
Francis DP, Wensel R. Test-retest repeatability of
cardiopulmonary exercise test variables in patients
with cardiac or respiratory disease. Eur J Prev
Cardiol 2014;21:445–53.
4. Skalski J, Allison TG, Miller TD. The safety of
cardiopulmonary exercise testing in a population
with high-risk cardiovascular diseases. Circulation
2012;126:2465–72.
5. Weber KT, Kinasewitz GT, Janicki JS,
Fishman AP. Oxygen utilization and ventilation
during exercise in patients with chronic cardiac
failure. Circulation 1982;65:1213–23.
6. Mancini DM, Eisen H, Kussmaul W, Mull R,
Edmunds LH Jr., Wilson JR. Value of peak exercise
oxygen consumption for optimal timing of cardiac
transplantation in ambulatory patients with heart
failure. Circulation 1991;83:778–86.
7. Osada N, Chaitman BR, Miller LW, et al.
Cardiopulmonary exercise testing identifies low
risk patients with heart failure and severely
impaired exercise capacity considered for
heart transplantation. J Am Coll Cardiol 1998;31:
577–82.
8. Chomsky DB, Lang CC, Rayos GH, et al.
Hemodynamic exercise testing. A valuable tool in
the selection of cardiac transplantation candi-
dates. Circulation 1996;94:3176–83.
9. Lim HS, Theodosiou M. Exercise ventilatory
parameters for the diagnosis of reactive pulmo-
nary hypertension in patients with heart failure.
J Card Fail 2014;20:650–7.
10. Agostoni P, Corrà U, Cattadori G, et al., MECKI
Score Research Group. Metabolic exercise test
data combined with cardiac and kidney indexes,
the MECKI score: a multiparametric approach to
heart failure prognosis. Int J Cardiol 2013;167:
2710–8.
11. Levy WC, Aaronson KD, Dardas TF, Williams P,
Haythe J, Mancini D. Prognostic impact of the
addition of peak oxygen consumption to the
Seattle Heart Failure Model in a transplant referral
population. J Heart Lung Transplant 2012;31:
817–24.
12. Lund LH, Aaronson KD, Mancini DM. Validation
of peak exercise oxygen consumption and the
Heart Failure Survival Score for serial risk stratifi-
cation in advanced heart failure. Am J Cardiol
2005;95:734–41.
13. Balady GJ, Arena R, Sietsema K, et al., Council
on Epidemiology and Prevention, Council on
Peripheral Vascular Disease, Interdisciplinary
Council on Quality of Care and Outcomes
Research, American Heart Association Exercise,
Cardiac Rehabilitation, and Prevention Committee
of the Council on Clinical Cardiology. Clinician’s
guide to cardiopulmonary exercise testing in
adults: a scientific statement from the American
Heart Association. Circulation 2010;122:191–225.
14. Redfield MM, Chen HH, Borlaug BA, et al.,
RELAX Trial. Effect of phosphodiesterase-5 inhi-
bition on exercise capacity and clinical status in
heart failure with preserved ejection fraction: a
randomized clinical trial. JAMA 2013;309:1268–77.
15. Zamani P, Rawat D, Shiva-Kumar P, et al.
Effect of inorganic nitrate on exercise capacity in
heart failure with preserved ejection fraction.
Circulation 2015;131:371–80; discussion 380.
16. Swank AM, Horton J, Fleg JL, et al. Modest
increase in peak VO2 is related to better clinical
outcomes in chronic heart failure patients: results
from heart failure and a controlled trial to inves-
tigate outcomes of exercise training. Circ Heart
Fail 2012;5:579–85.
17. Dunlay SM, Allison TG, Pereira NL. Changes in
cardiopulmonary exercise testing parameters
following continuous flow left ventricular assist
device implantation and heart transplantation.
J Card Fail 2014;20:548–54.
18. Chaudhry S, Kumar N, Behbahani H, et al.
Abnormal heart-rate response during cardiopul-
monary exercise testing identifies cardiac
dysfunction in symptomatic patients with
non-obstructive coronary artery disease. Int J
Cardiol 2017;228:114–21.
19. Barber NJ, Ako EO, Kowalik GT, et al. Magnetic
resonance-augmented cardiopulmonary exercise
testing: comprehensively assessing exercise
intolerance in children with cardiovascular disease.
Circ Cardiovasc Imaging 2016;9:e005282.
20. Bandera F, Generati G, Pellegrino M, et al.
Mitral regurgitation in heart failure: insights from
CPET combined with exercise echocardiography.
Eur Heart J Cardiovasc Imaging 2017;18:296–303.
21. Coats CJ, Rantell K, Bartnik A, et al. Cardio-
pulmonary exercise testing and prognosis in
hypertrophic cardiomyopathy. Circ Heart Fail
2015;8:1022–31.
22. Charalampopoulos A, Gibbs JS, Davies RJ,
et al. Exercise physiological responses to drug
treatments in chronic thromboembolic pulmonary
hypertension. J Appl Physiol (1985) 2016;121:
623–8.
23. Guazzi M, Cahalin LP, Arena R. Cardiopulmo-
nary exercise testing as a diagnostic tool for the
detection of left-sided pulmonary hypertension in
heart failure. J Card Fail 2013;19:461–7.
24. Wasserman K. Coupling of external to cellular
respiration during exercise: the wisdom of the
body revisited. Am J Physiol 1994;266:E519–39.
25. Sullivan MJ, Knight JD, Higginbotham MB,
Cobb FR. Relation between central and peripheral
hemodynamics during exercise in patients with
chronic heart failure. Muscle blood flow is reduced
with maintenance of arterial perfusion pressure.
Circulation 1989;80:769–81.
26. Okonko DO, Mandal AK, Missouris CG, Poole-
Wilson PA. Disordered iron homeostasis in chronic
heart failure: prevalence, predictors, and relation
to anemia, exercise capacity, and survival. J Am
Coll Cardiol 2011;58:1241–51.
27. Agostoni P, Wasserman K, Perego GB, et al.
Oxygen transport to muscle during exercise in
chronic congestive heart failure secondary to
idiopathic dilated cardiomyopathy. Am J Cardiol
1997;79:1120–4.
28. Esposito F, Mathieu-Costello O, Shabetai R,
Wagner PD, Richardson RS. Limited maximal
exercise capacity in patients with chronic heart
1634 Guazzi et al.
Cardiopulmonary Exercise Testing
failure: partitioning the contributors. J Am Coll
Cardiol 2010;55:1945–54.
29. Stringer WW, Hansen JE, Wasserman K. Car-
diac output estimated noninvasively from oxygen
uptake during exercise. J Appl Physiol (1985)
1997;82:908–12.
30. Wagner PD. Muscle O2 transport and O2
dependent control of metabolism. Med Sci Sports
Exerc 1995;27:47–53.
31. Boushel R, Gnaiger E, Calbet JA, et al. Muscle
mitochondrial capacity exceeds maximal oxygen
delivery in humans. Mitochondrion 2011;11:303–7.
32. Abramov D, Cohen RS, Katz SD, Mancini D,
Maurer MS. Comparison of blood volume charac-
teristics in anemic patients with low versus
preserved left ventricular ejection fractions. Am J
Cardiol 2008;102:1069–72.
33. Dhakal BP, Malhotra R, Murphy RM, et al.
Mechanisms of exercise intolerance in heart failure
with preserved ejection fraction: the role of
abnormal peripheral oxygen extraction. Circ Heart
Fail 2015;8:286–94.
34. Agostoni P, Corrà U, Cattadori G, et al., MECKI
Score Research Group. Prognostic value of inde-
terminable anaerobic threshold in heart failure.
Circ Heart Fail 2013;6:977–87.
35. Poole DC, Hirai DM, Copp SW, Musch TI.
Muscle oxygen transport and utilization in heart
failure: implications for exercise (in)tolerance. Am
J Physiol Heart Circ Physiol 2012;302:H1050–63.
36. Cohen-Solal A, Tabet JY, Logeart D,
Bourgoin P, Tokmakova M, Dahan M. A non-
invasively determined surrogate of cardiac power
(’circulatory power’) at peak exercise is a powerful
prognostic factor in chronic heart failure. Eur
Heart J 2002;23:806–14.
37. Agostoni P, Pellegrino R, Conca C, Rodarte JR,
Brusasco V. Exercise hyperpnea in chronic heart
failure: relationships to lung stiffness and expira-
tory flow limitation. J Appl Physiol (1985) 2002;
92:1409–16.
38. Barron A, Francis DP, Mayet J, et al. Oxygen
uptake efficiency slope and breathing reserve, not
anaerobic threshold, discriminate between
patients with cardiovascular disease over chronic
obstructive pulmonary disease. J Am Coll Cardiol
HF 2016;4:252–61.
39. Lewis GD, Murphy RM, Shah RV, et al. Pul-
monary vascular response patterns during exercise
in left ventricular systolic dysfunction predict ex-
ercise capacity and outcomes. Circ Heart Fail 2011;
4:276–85.
40. Guazzi M, Dixon D, Labate V, et al. RV con-
tractile function and its coupling to pulmonary
circulation in heart failure with preserved ejection
fraction: stratification of clinical phenotypes and
outcomes. J Am Coll Cardiol Img 2017 Apr 7
[E-pub ahead of print].
41. Guazzi M, Reina G, Tumminello G, Guazzi MD.
Exercise ventilation inefficiency and cardiovascular
mortality in heart failure: the critical independent
prognostic value of the arterial CO2 partial pres-
sure. Eur Heart J 2005;26:472–80.
42. Kee K, Stuart-Andrews C, Ellis MJ, et al.
Increased dead space ventilation mediates
reduced exercise capacity in systolic heart failure.
Am J Respir Crit Care Med 2016;193:1292–300.
43. Hachamovitch R, Brown HV, Rubin SA. Respi-
ratory and circulatory analysis of CO2 output
during exercise in chronic heart failure. Circulation
1991;84:605–12.
44. Olson TP, Joyner MJ, Johnson BD. Influence of
locomotor muscle metaboreceptor stimulation on
the ventilatory response to exercise in heart
failure. Circ Heart Fail 2010;3:212–9.
45. Van Iterson EH, Johnson BD, Borlaug BA,
Olson TP. Physiologic dead space and arterial
carbon dioxide contribution to exercise ventilatory
inefficiency in patients with reduced or preserved
ejection fraction. Eur J Heart Fail 2017. In press.
46. Elbehairy AF, Ciavaglia CE, Webb KA, et al.,
Canadian Respiratory Research Network. Pulmo-
nary gas exchange abnormalities in mild chronic
obstructive pulmonary disease. Implications for
dyspnea and exercise intolerance. Am J Respir Crit
Care Med 2015;191:1384–94.
47. Van Laethem C, Bartunek J, Goethals M,
Nellens P, Andries E, Vanderheyden M. Oxygen
uptake efficiency slope, a new submaximal
parameter in evaluating exercise capacity in
chronic heart failure patients. Am Heart J 2005;
149:175–80.
48. Guazzi M. Treating exercise oscillatory venti-
lation in heart failure: the detail that may matter.
Eur Respir J 2012;40:1075–7.
49. Guazzi M, Raimondo R, Vicenzi M, et al.
Exercise oscillatory ventilation may predict sudden
cardiac death in heart failure patients. J Am Coll
Cardiol 2007;50:299–308.
50. Sun XG, Hansen JE, Beshai JF, Wasserman K.
Oscillatory breathing and exercise gas exchange
abnormalities prognosticate early mortality and
morbidity in heart failure. J Am Coll Cardiol 2010;
55:1814–23.
51. Vicenzi M, Deboeck G, Faoro V, Loison J,
Vachiery JL, Naeije R. Exercise oscillatory ventila-
tion in heart failure and in pulmonary arterial
hypertension. Int J Cardiol 2016;202:736–40.
52. Guazzi M, Arena R, Pellegrino M, et al. Preva-
lence and characterization of exercise oscillatory
ventilation in apparently healthy individuals at
variable risk for cardiovascular disease: a sub-
analysis of the EURO-EX trial. Eur J Prev Cardiol
2016;23:328–34.
53. Belardinelli R, Lacalaprice F, Carle F, et al.
Exercise-induced myocardial ischaemia detected
by cardiopulmonary exercise testing. Eur Heart J
2003;24:1304–13.
54. Popovic D, Martic D, Djordjevic T, et al.
Oxygen consumption and carbon-dioxide recovery
kinetics in the prediction of coronary artery dis-
ease severity and outcome. Int J Cardiol 2017 Jun
28 [E-pub ahead of print].
55. Balady GJ, Ades PA, Bittner VA, et al. Referral,
enrollment, and delivery of cardiac rehabilitation/
secondary prevention programs at clinical centers
and beyond: a presidential advisory from the
American Heart Association. Circulation 2011;124:
2951–60.
56. Ponikowski P, Voors AA, Anker SD, et al. 2016
ESC guidelines for the diagnosis and treatment of
JACC VOL. 70, NO. 13, 2017
SEPTEMBER 26, 2017:1618–36
acute and chronic heart failure: the Task Force for
the diagnosis and treatment of acute and chronic
heart failure of the European Society of Cardiology
(ESC). Eur J Heart Fail 2016;18:891–975.
57. Mezzani A, Hamm LF, Jones AM, et al. Aerobic
exercise intensity assessment and prescription in
cardiac rehabilitation: a joint position statement of
the European Association for Cardiovascular
Prevention and Rehabilitation, the American
Association of Cardiovascular and Pulmonary
Rehabilitation and the Canadian Association of
Cardiac Rehabilitation. Eur J Prev Cardiol 2013;20:
442–67.
58. Reed JL, Blais AZ, Keast ML, Pipe AL, Reid RD.
Performance of fixed heart rate increment targets
of 20 vs 30 beats per minute for exercise reha-
bilitation prescription in outpatients with heart
failure. Can J Cardiol 2017;33:777–84.
59. Yancy CW, Jessup M, Bozkurt B, et al. 2013
ACCF/AHA guideline for the management of heart
failure: a report of the American College of
Cardiology Foundation/American Heart Associa-
tion Task Force on Practice Guidelines. J Am Coll
Cardiol 2013;62:e147–239.
60. Arena R, Guazzi M, Cahalin LP, Myers J.
Revisiting cardiopulmonary exercise testing
applications in heart failure: aligning evidence
with clinical practice. Exerc Sport Sci Rev 2014;42:
153–60.
61. Guazzi M, Naeije R. Pulmonary hypertension in
heart failure: pathophysiology, pathobiology, and
emerging clinical perspectives. J Am Coll Cardiol
2017;69:1718–34.
62. Klaassen SHC, Liu LCY, Hummel YM, et al.
Clinical and hemodynamic correlates and prog-
nostic value of VE/VCO2 slope in patients with
heart failure with preserved ejection fraction and
pulmonary hypertension. J Card Fail 2017 Jul 20
[E-pub ahead of print].
63. Bhatia V, Bhatia R, Mathew B. Angiotensin
receptor blockers in congestive heart failure:
evidence, concerns, and controversies. Cardiol Rev
2005;13:297–303.
64. Guazzi M, Arena R. The impact of pharmaco-
therapy on the cardiopulmonary exercise test
response in patients with heart failure: a mini
review. Curr Vasc Pharmacol 2009;7:557–69.
65. Agostoni P, Apostolo A, Cattadori G, et al.
Effects of beta-blockers on ventilation efficiency
in heart failure. Am Heart J 2010;159:1067–73.
66. Guazzi M, Myers J, Peberdy MA, Bensimhon D,
Chase P, Arena R. Ventilatory efficiency and
dyspnea on exertion improvements are related to
reduced pulmonary pressure in heart failure
patients receiving sildenafil. Int J Cardiol 2010;
144:410–2.
67. Guazzi M, Vicenzi M, Arena R. Phosphodies-
terase 5 inhibition with sildenafil reverses exercise
oscillatory breathing in chronic heart failure: a
long-term cardiopulmonary exercise testing
placebo-controlled study. Eur J Heart Fail 2012;
14:82–90.
68. Hussain I, Mohammed SF, Forfia PR, et al.
Impaired right ventricular-pulmonary arterial
coupling and effect of sildenafil in heart failure
with preserved ejection fraction: an ancillary
analysis from the Phosphodiesterase-5 Inhibition
JACC VOL. 70, NO. 13, 2017
SEPTEMBER 26, 2017:1618–36
to Improve Clinical Status And Exercise Capacity in
Diastolic Heart Failure (RELAX) trial. Circ Heart Fail
2016;9:e002729.
69. Kitzman DW, Hundley WG, Brubaker PH, et al.
A randomized double-blind trial of enalapril in
older patients with heart failure and preserved
ejection fraction: effects on exercise tolerance and
arterial distensibility. Circ Heart Fail 2010;3:
477–85.
70. Parthasarathy HK, Pieske B, Weisskopf M,
et al. A randomized, double-blind, placebo-
controlled study to determine the effects of
valsartan on exercise time in patients with symp-
tomatic heart failure with preserved ejection
fraction. Eur J Heart Fail 2009;11:980–9.
71. Pal N, Sivaswamy N, Mahmod M, et al. Effect
of selective heart rate slowing in heart failure with
preserved ejection fraction. Circulation 2015;132:
1719–25.
72. O’Neill JO, Young JB, Pothier CE, Lauer MS.
Peak oxygen consumption as a predictor of death
in patients with heart failure receiving beta-
blockers. Circulation 2005;111:2313–8.
73. Mehra MR, Canter CE, Hannan MM, et al.,
International Society for Heart Lung Trans-
plantation (ISHLT) Infectious Diseases, Pediatric
and Heart Failure and Transplantation Councils.
The 2016 International Society for Heart Lung
Transplantation listing criteria for heart trans-
plantation: a 10-year update. J Heart Lung
Transplant 2016;35:1–23.
74. Arena R, Myers J, Abella J, et al. Development
of a ventilatory classification system in patients
with heart failure. Circulation 2007;115:2410–7.
75. Guazzi M, Boracchi P, Arena R, et al. Devel-
opment of a cardiopulmonary exercise prognostic
score for optimizing risk stratification in heart
failure: the (P)e(R)i(O)dic (B)reathing during
(E)xercise (PROBE) study. J Card Fail 2010;16:
799–805.
76. Arena R, Myers J, Guazzi M. Cardiopulmonary
exercise testing is a core assessment for patients
with heart failure. Congest Heart Fail 2011;17:
115–9.
77. Obokata M, Kane GC, Reddy YN, Olson TP,
Melenovsky V, Borlaug BA. Role of diastolic stress
testing in the evaluation for heart failure with
preserved ejection fraction: a simultaneous
invasive-echocardiographic study. Circulation
2017;135:825–38.
78. Huang W, Resch S, Oliveira RK, Cockrill BA,
Systrom DM, Waxman AB. Invasive cardiopulmo-
nary exercise testing in the evaluation of unex-
plained dyspnea: insights from a multidisciplinary
dyspnea center. Eur J Prev Cardiol 2017;24:
1190–9.
79. Lewis GD, Bossone E, Naeije R, et al. Pulmo-
nary vascular hemodynamic response to exercise
in cardiopulmonary diseases. Circulation 2013;128:
1470–9.
80. Borlaug BA, Nishimura RA, Sorajja P, Lam CS,
Redfield MM. Exercise hemodynamics enhance
diagnosis of early heart failure with preserved
ejection fraction. Circ Heart Fail 2010;3:588–95.
81. Dorfs S, Zeh W, Hochholzer W, et al. Pulmo-
nary capillary wedge pressure during exercise and
long-term mortality in patients with suspected
heart failure with preserved ejection fraction. Eur
Heart J 2014;35:3103–12.
82. Berry NC, Manyoo A, Oldham WM, et al.
Protocol for exercise hemodynamic assessment:
performing an invasive cardiopulmonary exercise
test in clinical practice. Pulm Circ 2015;5:610–8.
83. Fu Q, Vangundy TB, Galbreath MM, et al.
Cardiac origins of the postural orthostatic tachy-
cardia syndrome. J Am Coll Cardiol 2010;55:
2858–68.
84. Chatterjee NA, Murphy RM, Malhotra R, et al.
Prolonged mean VO2 response time in systolic
heart failure: an indicator of impaired right
ventricular-pulmonary vascular function. Circ
Heart Fail 2013;6:499–507.
85. Sugimoto T, Bandera F, Generati G,
Alfonzetti E, Bussadori C, Guazzi M. Left atrial
function dynamics during exercise in heart failure:
pathophysiological implications on the right heart
and exercise ventilation inefficiency. J Am Coll
Cardiol Img 2016 Dec 12 [E-pub ahead of print].
86. Claessen G, La Gerche A. Pulmonary vascular
function during exercise: progressing toward
routine clinical use. Circ Cardiovasc Imaging 2017;
10:006326.
87. Guazzi M, Villani S, Generati G, et al. Right
ventricular contractile reserve and pulmonary
circulation uncoupling during exercise challenge in
heart failure: pathophysiology and clinical phe-
notypes. J Am Coll Cardiol HF 2016;4:625–35.
88. Bandera F, Generati G, Pellegrino M, et al.
Role of right ventricle and dynamic pulmonary
hypertension on determining DVO2/DWork Rate
flattening: insights from cardiopulmonary exercise
test combined with exercise echocardiography.
Circ Heart Fail 2014;7:782–90.
89. Rozenbaum Z, Khoury S, Aviram G, et al.
Discriminating circulatory problems from decon-
ditioning: echocardiographic and cardiopulmonary
exercise test analysis. Chest 2017;151:431–40.
90. Shimiaie J, Sherez J, Aviram G, et al.
Determinants of effort intolerance in patients with
heart failure: combined echocardiography and
cardiopulmonary stress protocol. J Am Coll Cardiol
HF 2015;3:803–14.
91. Hummel YM, Bugatti S, Damman K, et al.
Functional and hemodynamic cardiac de-
terminants of exercise capacity in patients with
systolic heart failure. Am J Cardiol 2012;110:
1336–41.
92. Podolec P, Rubís P, Tomkiewicz-Pajak L,
Kopec G, Tracz W. Usefulness of the evaluation of
left ventricular diastolic function changes during
stress echocardiography in predicting exercise
capacity in patients with ischemic heart failure.
J Am Soc Echocardiogr 2008;21:834–40.
93. Grigioni F, Enriquez-Sarano M, Zehr KJ,
Bailey KR, Tajik AJ. Ischemic mitral regurgitation:
long-term outcome and prognostic implications
with quantitative Doppler assessment. Circulation
2001;103:1759–64.
94. Borlaug BA. Is HFpEF one disease or many?
J Am Coll Cardiol 2016;67:671–3.
95. Lancellotti P, Pellikka PA, Budts W, et al. The
clinical use of stress echocardiography in non-
Guazzi et al. 1635
Cardiopulmonary Exercise Testing
ischaemic heart disease: recommendations from
the European Association of Cardiovascular
Imaging and the American Society of Echocardi-
ography. Eur Heart J Cardiovasc Imaging 2016;17:
1191–229.
96. Meluzin J, Sitar J, Kristek J, et al. The role
of exercise echocardiography in the diagnostics
of heart failure with normal left ventricular
ejection fraction. Eur J Echocardiogr 2011;12:
591–602.
97. Ha JW, Oh JK, Pellikka PA, et al. Diastolic
stress echocardiography: a novel noninvasive
diagnostic test for diastolic dysfunction using su-
pine bicycle exercise Doppler echocardiography.
J Am Soc Echocardiogr 2005;18:63–8.
98. Nedeljkovic I, Banovic M, Stepanovic J, et al.
The combined exercise stress echocardiography
and cardiopulmonary exercise test for identifica-
tion of masked heart failure with preserved ejec-
tion fraction in patients with hypertension. Eur J
Prev Cardiol 2016;23:71–7.
99. Donal E, Lund LH, Oger E, et al., KaRen
Investigators. Value of exercise echocardiography
in heart failure with preserved ejection fraction: a
substudy from the KaRen study. Eur Heart J
Cardiovasc Imaging 2016;17:106–13.
100. Shim CY, Kim SA, Choi D, et al. Clinical
outcomes of exercise-induced pulmonary hyper-
tension in subjects with preserved left ventricular
ejection fraction: implication of an increase in left
ventricular filling pressure during exercise. Heart
2011;97:1417–24.
101. Kosmala W, Rojek A, Przewlocka-Kosmala M,
Mysiak A, Karolko B, Marwick TH. Contributions of
nondiastolic factors to exercise intolerance in
heart failure with preserved ejection fraction. J Am
Coll Cardiol 2016;67:659–70.
102. Garbi M, Chambers J, Vannan MA,
Lancellotti P. Valve stress echocardiography: a
practical guide for referral, procedure, reporting,
and clinical implementation of results from the
HAVEC Group. J Am Coll Cardiol Img 2015;8:
724–36.
103. Pedersen F, Mehlsen J, Raymond I, Atar D,
Skjoldborg US, Hildebrandt PR. Evaluation of
dyspnoea in a sample of elderly subjects recruited
from general practice. Int J Clin Pract 2007;61:
1481–91.
104. Laufer-Perl M, Gura Y, Shimiaie J, et al.
Mechanisms of effort intolerance in patients with
rheumatic mitral stenosis: combined echocardi-
ography and cardiopulmonary stress protocol.
J Am Coll Cardiol Img 2017;10:622–33.
105. Bandera F, Generati G, Pellegrino M, et al.
Paradoxical low flow/low gradient aortic stenosis:
can cardiopulmonary exercise test help in identi-
fying it? Int J Cardiol 2016;203:37–9.
106. Magri D, Limongelli G, Re F, et al. Cardio-
pulmonary exercise test and sudden cardiac death
risk in hypertrophic cardiomyopathy. Heart 2016;
102:602–9.
107. Maron BJ, Maron MS. Hypertrophic cardio-
myopathy. Lancet 2013;381:242–55.
108. Finocchiaro G, Haddad F, Knowles JW,
et al. Cardiopulmonary responses and prognosis
in hypertrophic cardiomyopathy: a potential role
1636 Guazzi et al.
Cardiopulmonary Exercise Testing
for comprehensive noninvasive hemodynamic
assessment. J Am Coll Cardiol HF 2015;3:
408–18.
109. Re F, Zachara E, Avella A, et al. Dissecting
functional impairment in hypertrophic cardiomy-
opathy by dynamic assessment of diastolic
reserve and outflow obstruction: a combined
cardiopulmonary-echocardiographic study. Int J
Cardiol 2017;227:743–50.
110. Elliott PM, Anastasakis A, Borger MA, et al.
2014 ESC guidelines on diagnosis and management
of hypertrophic cardiomyopathy: the Task Force for
the Diagnosis and Management of Hypertrophic
Cardiomyopathy of the European Society of Cardi-
ology (ESC). Eur Heart J 2014;35:2733–79.
111. Bandera F, Generati G, Pellegrino M, Secchi F,
Menicanti L, Guazzi M. Exercise gas exchange
analysis in obstructive hypertrophic cardiomyop-
athy before and after myectomy (cardiopulmonary
exercise test combined with exercise-
echocardiography in HCM). Int J Cardiol 2015;
178:282–3.
JACC VOL. 70, NO. 13, 2017
SEPTEMBER 26, 2017:1618–36
112. Galiè N, Humbert M, Vachiery JL, et al. 2015
ESC/ERS guidelines for the diagnosis and
treatment of pulmonary hypertension: the Joint
Task Force for the Diagnosis and Treatment of
Pulmonary Hypertension of the European Society
of Cardiology (ESC) and the European Respiratory
Society (ERS). Eur Heart J 2016;37:67–119.
KEY WORDS exercise, gas exchange
analysis, heart failure, oxygen consumption