Cardiac power output is a direct measure of overall cardiac function that integrates both flow- and pressure-generating capacities of the heart. The present study assessed the reproducibility of cardiac power output and other more commonly reported cardiopulmonary exercise variables in patients with chronic heart failure. Metabolic, ventilatory and non-invasive (inert gas re-breathing) central haemodynamic measurements were undertaken at rest and near-maximal exercise of the modified Bruce protocol in 19 patients with stable chronic heart failure. The same procedure was repeated 7 days later to assess reproducibility. Cardiac power output was calculated as the product of cardiac output and mean arterial pressure. Resting central haemodynamic variables demonstrate low CV (coefficient of variation) (ranging from 3.4% for cardiac output and 5.6% for heart rate). The CV for resting metabolic and ventilatory measurements ranged from 8.2% for respiratory exchange ratio and 14.2% for absolute values of oxygen consumption. The CV of anaerobic threshold, peak oxygen consumption, carbon dioxide production and respiratory exchange ratio ranged from 3.8% (for anaerobic threshold) to 6.4% (for relative peak oxygen consumption), with minute ventilation having a CV of 11.1%. Near-maximal exercise cardiac power output and cardiac output had CVs of 4.1 and 2.2%, respectively. Cardiac power output demonstrates good reproducibility suggesting that there is no need for performing more than one cardiopulmonary exercise test. As a direct measure of cardiac function (dysfunction) and an excellent prognostic marker, it is strongly advised in the assessment of patients with chronic heart failure undergoing cardiopulmonary exercise testing.

INTRODUCTION

Cardiopulmonary exercise testing data have been widely used in cardiac patients to assess functional capacity and to determine prognosis. Repeated measurements obtained during cardiopulmonary exercise tests have generally shown good reproducibility in healthy adults [1] and in highly trained athletes [2]. A basic requirement for the clinical application of exercise-derived cardiopulmonary indices is high reproducibility. Particularly in patients with chronic heart failure, high reproducibility is of great importance because even small changes in cardiopulmonary data due to treatment or disease progression may have clinical significance [3].

Several studies have assessed reproducibility of cardiopulmonary exercise variables in patients with chronic heart failure [39]. Reproducibility of V̇O2 peak [peak V̇O2 (oxygen consumption)] has been most frequently analysed. This is unsurprising, as in addition to reflecting physiological reserve of the cardiovascular system, V̇O2 peak plays an important role in risk stratification, management and evaluation of the efficacy of treatment in patients with heart failure.

The importance of evaluating exercise haemodynamics in patients with heart failure has been previously described [1012]. CPO (cardiac power output) is a central haemodynamic measure that demonstrates overall cardiac function and is the product of simultaneously measured QT (cardiac output) and MAP (mean arterial pressure) [13,14]. By incorporating both pressure and flow domains of the cardiovascular system, CPO is a unique, direct and integrative measure of cardiac pumping capability [13,15,16]. Not surprisingly, resting and peak exercise CPO have been shown to be powerful predictors of prognosis and mortality in patients with chronic heart failure [1721]. Recently, our group has shown that CPO differentiates well during cardiac restoration using left ventricular assist devices [22] and emphasizes the benefits of this therapy in terms of organ function at rest and peak exercise [23]. It has also been shown that commonly reported cardiopulmonary exercise-derived indices demonstrate limited capacity in reflecting cardiac pumping capability in patients treated with ventricular assist devices [24].

In contrast with other, well-established cardiopulmonary exercise indices (e.g. V̇O2), reproducibility of CPO has been reported by one study only, using a specific two-stage exercise protocol [25]. However, reproducibility of CPO using a standard incremental exercise test has not yet been demonstrated. Therefore, the purpose of the present study was to assess the reproducibility of CPO in conjunction with other important cardiopulmonary exercise indices. We hypothesize that CPO will demonstrate a high degree of reproducibility in patients with chronic heart failure.

MATERIALS AND METHODS

Patients

The study group consisted of 19 stable chronic heart failure patients with New York Heart Association classes I and II. The patients' demographic and clinical characteristics are presented in Table 1. All patients were symptomatically stable and on the same dose of optimal drug therapy for the preceding 2 months. Exclusion criteria were exercise-induced myocardial ischaemia or uncontrolled arrhythmias, valvular heart disease, congestive heart failure, uncontrolled hypertension [resting SBP (systolic blood pressure) above 160 mmHg, and/or resting DBP (diastolic blood pressure) above 105 mmHg], peripheral vascular disease, chronic obstructive pulmonary disease and orthopaedic or other condition precluding participation in exercise testing. The present study was approved by the Hounslow and Hillingdon NHS Research Ethics Committee. An informed written consent form was obtained from each patient prior the investigation.

Table 1
Patients' demographic and clinical characteristics (n=19)

ACE, angiotensin-converting enzyme; BMI, body mass index; BSA, body surface area; LVEF, left ventricular ejection fraction; ARBs, angiotensin II receptors blockers.

Parameter
Age (years) 62±11 
Men/women (n15/4 
Weight (kg) 79.6±17 
Height (cm) 172±10 
BMI (kg/m226.6±4.0 
BSA (m21.92±0.25 
Cause of heart failure (ischaemic/non-ischaemic) (n13/6 
LVEF (%) 29±12 
Medication (n 
 ACE inhibitors 12 
 β-Blockers 16 
 Diuretics 
 ARBs antagonists 
 Digoxin 
 Anti-arrhythmics 
 Aspirin 
 Warfarin 
 Spironolactone 
Parameter
Age (years) 62±11 
Men/women (n15/4 
Weight (kg) 79.6±17 
Height (cm) 172±10 
BMI (kg/m226.6±4.0 
BSA (m21.92±0.25 
Cause of heart failure (ischaemic/non-ischaemic) (n13/6 
LVEF (%) 29±12 
Medication (n 
 ACE inhibitors 12 
 β-Blockers 16 
 Diuretics 
 ARBs antagonists 
 Digoxin 
 Anti-arrhythmics 
 Aspirin 
 Warfarin 
 Spironolactone 

Study protocol and measurements

In order to assess reproducibility of CPO and selected cardiopulmonary exercise variables, patients visited the Cardiac Rehabilitation Department of Hillingdon Hospital twice within 1 week of each other. All equipments were calibrated prior to each exercise test according to the manufacturer's recommendations. Measurements were performed at the same time of day under identical conditions and medication. Patients who were not familiar with the use of treadmill were asked to visit the laboratory a few days prior to their first test to complete a brief familiarization protocol. This was not a maximal exercise test but familiarization with walking on a treadmill at different speeds.

All patients attended the clinic at least 2 h after having any food or drink and did not consume alcohol or caffeine containing beverages before the tests. The tests were performed at an ambient room temperature maintained at approx. 20°C.

System configurations

Gas-exchange measurements were recorded on a breath-by-breath basis using the Cardio O2 system (Medical Graphics), whereas QT was estimated using inert gas re-breathing methodology (Innocor; Innovision), as described below. The Innocor respiratory valve unit, with bacterial filter, was inserted into the pneumotach of the Medical Graphics system. Both systems were attached to the subject's face mask, where the pneumotach of the Medical Graphics system was closer to the subject's mouth. Such configuration increased dead space and the distance between the subject's mouth and the Innocor gas sensors. The volume calibration was performed after which the two systems were placed in series. The Innocor respiratory valve unit was supported with the Innocor metal arm attached to the monitor trolley. Prior to measurements, all patients were familiarized with the face mask and a learning session of the Innocor QT procedure was performed.

Resting measurements

With the patient seated and following 5 min of acclimatization, arterial blood pressure was assessed from the brachial artery by cuff sphygmomanometry. Expired air was collected for online measurement of ventilation and expired gas concentrations. QT was measured using the inert gas re-breathing methodology while the patient was still in a seated position, as previously described [2628]. Briefly, an oxygen-enriched mixture of an inert soluble gas (0.5% nitrous oxide) and an inert insoluble gas (0.1% sulfur hexafluoride) from a 4-litre pre-filled anaesthesia bag was used. Photo-acoustic analysers measure gas concentrations over a 5-breath interval. Nitrous oxide concentration decreases during the re-breathing manoeuvre at a rate proportional to pulmonary blood flow allowing estimation of QT. Three to four respiratory cycles were needed to obtain a value for nitrous oxide washout.

Exercise measurements

Patients then performed a modified Bruce protocol on a motor driven treadmill (Cardio Control). During the test, ventilation and expired gases were collected and analysed as described at rest. Arterial blood pressure and Borg scale recordings for dyspnoea and fatigue were performed during the last 30 s of each exercise stage. At near-maximal exercise [defined as at least one of the following: RER (respiratory exchange ratio) >1.05, the absence of a rise in V̇O2 with further increase in exercise intensity, Borg scale >17], QT was measured again using the re-breathing technique described at rest and blood pressure was recorded. Patients were instructed to give an approx. 30-s warning before they felt they would end the exercise so that a final QT re-breathing measurement was obtained at near-maximal exercise. Patients were verbally encouraged to continue exercise test for as long as they were able, despite termination of QT measurement. The time between near-maximal QT measurement and termination of exercise test was recorded. V̇O2 peak was defined as the average V̇O2 during the last minute of exercise. Anaerobic threshold was identified by the V-slope method [29].

Calculations and statistical analysis

The CPO was calculated from the product of QT and MAP using the following equation [20]:

 
formula

where K is the conversion factor (2.22×10−3), QT is measured in l/min and MAP is in mmHg. Near-maximal exercise arterial-venous oxygen difference, expressed in ml of O2/100 ml of blood, was calculated as the ratio between V̇O2 peak and peak QT.

All statistical analysis was carried out using SPSS version 13.0. Prior to statistical analysis, data were checked for univariate and multivariate outliers using standard Z-distribution cut-offs and Mahalanobis distance tests respectively. Normality of distribution was assessed using a Kolmogorov–Smirnov test. To test differences in measured variables between test I and test II, a Student's t test for paired samples was used. Statistical significance was indicated if P<0.05. To assess the reproducibility of CPO and selected cardiopulmonary variables, the CV (coefficient of variation) and Pearson's correlation coefficient (r) were calculated. For each variable, the CV was defined as the (within-person S.D./within-person mean)×100%. A CV of ≤6% was considered as an indicator of good, 6–10% acceptable, and >10% poor reproducibility. Additionally, Bland–Altman analysis was used to demonstrate between-test agreement of cardiopulmonary exercise test variables measured at peak exercise [30]. All data are presented as means±S.D. unless otherwise indicated.

RESULTS

All subjects completed each exercise test without any contraindications. Thirteen patients stopped the first exercise test due to fatigue and six patients stopped due to dyspnoea. Four patients reported a different limiting symptom at the end of the second test. Time between QT measurement and termination of exercise test was 16±8 for test 1 and 22±11 min for test 2 (P>0.05).

There were no significant differences in resting metabolic, ventilatory and central haemodynamic variables between the two tests (Table 2). Central haemodynamic variables demonstrated low CVs and a higher coefficient of correlation (Table 2).

Table 2
Reproducibility of resting measurements in patients with chronic heart failure

HR, heart rate.

ParameterTest ITest IIP valuerCV (%)
Metabolics      
V̇O2 (ml/min) 304±98 320±100 0.45 0.69 14.2 
V̇O2 (ml/kg of body weight per min) 3.9±0.9 4.0±0.9 0.68 0.43 11.9 
V̇CO2 (ml/min) 262±87 268±94 0.66 0.83 10.3 
V̇E (l/min) 12.4±4.2 12.0±4.7 0.53 0.75 14.4 
 RER 0.80±0.12 0.80±0.10 0.95 0.44 8.2 
Haemodynamics      
QT (l/min) 4.0±1.0 3.9±0.9 0.63 0.99 3.4 
 HR (beats/min) 62±12 60±9 0.24 0.83 5.6 
 SBP (mmHg) 118±11 114±10 0.13 0.55 5.3 
 DBP (mmHg) 77±11 75±10 0.31 0.87 4.3 
 MAP (mmHg) 94±11 91±9 0.15 0.79 4.5 
 CPO (W) 0.83±0.24 0.79±0.19 0.15 0.96 5.4 
ParameterTest ITest IIP valuerCV (%)
Metabolics      
V̇O2 (ml/min) 304±98 320±100 0.45 0.69 14.2 
V̇O2 (ml/kg of body weight per min) 3.9±0.9 4.0±0.9 0.68 0.43 11.9 
V̇CO2 (ml/min) 262±87 268±94 0.66 0.83 10.3 
V̇E (l/min) 12.4±4.2 12.0±4.7 0.53 0.75 14.4 
 RER 0.80±0.12 0.80±0.10 0.95 0.44 8.2 
Haemodynamics      
QT (l/min) 4.0±1.0 3.9±0.9 0.63 0.99 3.4 
 HR (beats/min) 62±12 60±9 0.24 0.83 5.6 
 SBP (mmHg) 118±11 114±10 0.13 0.55 5.3 
 DBP (mmHg) 77±11 75±10 0.31 0.87 4.3 
 MAP (mmHg) 94±11 91±9 0.15 0.79 4.5 
 CPO (W) 0.83±0.24 0.79±0.19 0.15 0.96 5.4 

At anaerobic threshold point there was no significant difference in V̇O2, V̇CO2 (carbon dioxide production), V̇E (minute ventilation) and RER between the two tests (P>0.05).

Exercise time, V̇O2 and V̇E increased significantly from test I to test II (P<0.05). Calculated near-maximal exercise arterial-venous oxygen difference increased by 4% (from 16.0±1.2 to 16.7±1.4 ml O2/100 ml of blood; P<0.05). The coefficient of correlation was high and the CV was low between the two tests for all measured variables except for V̇E (Table 3). The Bland–Altman analysis revealed that the mean difference between the first and second test for exercise time was 61 s and the limits of agreement were −183 to 61 s. The mean difference between the first and the second test for V̇O2 peak was −92 ml/min or −1.3 ml/kg of body weight per min and the limits of agreement were −394 to 210 ml/min or −5.3 to 2.7 ml/kg of body weight per min.

Table 3
Reproducibility of cardiopulmonary exercise testing measurements in patients with chronic heart failure

AT, anaerobic threshold; HR, heart rate.

ParameterTest ITest IIP valuerCV (%)
Metabolics      
V̇O2 (ml/min) 1792±827 1884±767 0.04 0.99 5.7 
V̇O2 (kg of body weight per min) 21.7±7.0 23.0±5.8 0.03 0.97 6.4 
V̇O2 at AT (kg of body weight per min) 13.7±3.5 14.2±3.4 0.14 0.97 3.8 
V̇CO2 (ml/min) 1953±251 2147±337 0.11 0.98 5.5 
V̇E (l/min) 53.4±16.9 60.2±21.7 0.02 0.90 11.1 
 RER 1.09±0.04 1.14±0.8 0.13 0.91 4.2 
 Exercise time (s) 684±191 745±190 0.00 0.95 7.1 
 Borg score 16.8±2.8 17.6±2.1 0.18 0.91 3.2 
Haemodynamics      
QT (l/min) 11.2±5.7 11.3±5.6 0.84 0.99 2.2 
 HR (beats/min) 116±31 118±26 0.42 0.96 4.4 
 SBP (mmHg) 148±27 146±25 0.67 0.93 4.5 
 DBP (mmHg) 74±13 78±11 0.23 0.90 7.5 
 MAP (mmHg) 104±17 106±15 0.51 0.98 6.8 
 CPO (W) 2.68±1.53 2.69±1.52 0.63 0.96 4.1 
ParameterTest ITest IIP valuerCV (%)
Metabolics      
V̇O2 (ml/min) 1792±827 1884±767 0.04 0.99 5.7 
V̇O2 (kg of body weight per min) 21.7±7.0 23.0±5.8 0.03 0.97 6.4 
V̇O2 at AT (kg of body weight per min) 13.7±3.5 14.2±3.4 0.14 0.97 3.8 
V̇CO2 (ml/min) 1953±251 2147±337 0.11 0.98 5.5 
V̇E (l/min) 53.4±16.9 60.2±21.7 0.02 0.90 11.1 
 RER 1.09±0.04 1.14±0.8 0.13 0.91 4.2 
 Exercise time (s) 684±191 745±190 0.00 0.95 7.1 
 Borg score 16.8±2.8 17.6±2.1 0.18 0.91 3.2 
Haemodynamics      
QT (l/min) 11.2±5.7 11.3±5.6 0.84 0.99 2.2 
 HR (beats/min) 116±31 118±26 0.42 0.96 4.4 
 SBP (mmHg) 148±27 146±25 0.67 0.93 4.5 
 DBP (mmHg) 74±13 78±11 0.23 0.90 7.5 
 MAP (mmHg) 104±17 106±15 0.51 0.98 6.8 
 CPO (W) 2.68±1.53 2.69±1.52 0.63 0.96 4.1 

There was no significant difference between QT, arterial blood pressures and near-maximal exercise CPO in the two tests. For all measured central haemodynamic variables, there was a high coefficient of correlation and a low CV (Table 3).

The Bland–Altman analysis reveals that for near- maximal exercise CPO, the mean difference between the two tests was −0.02 W and the limits of agreement were −0.14 to 0.10 W. The mean difference between the first and second test for QT was −0.02 litre/min and the limits of agreement were −0.40 to 0.36 litre/min, whereas for MAP, the mean difference between the two tests was −2.3 mmHg and the limits of agreement were −7.8 to 6.1 mmHg. Additionally, the mean difference for SBPs and DBPs between the two tests was 1.9 and −4.2 mmHg. The limits of agreement were −11.2 to 8.8 mmHg for SBP and −3.8 to 7.7 mmHg for DBP.

DISCUSSION

In the present study, we evaluated the reproducibility of CPO, a direct measure of cardiac function and pumping capability [13,14], and other important cardiopulmonary exercise parameters in patients with chronic heart failure. The major findings suggest that (i) CPO measured at rest and near-maximal exercise demonstrates good reproducibility and that (ii) exercise duration, V̇CO2 and V̇E also demonstrate low CV.

A common method of assuring a reproducible response to exercise is to have the patient perform two exercise tests on separate days, at the same time of the day, and a test is considered reproducible if peak oxygen uptake is within ±10% on both days [31]. Reproducible physiological responses to exercise are obtainable over extended periods of time in patients with stable chronic heart failure [32]. In the light of this finding and according to recent reproducibility studies [4,5], the present study was designed to assess reproducibility of CPO in patients with stable mild chronic heart failure performing two exercise tests within 1 week of each other.

CPO is emerging as an important physiological variable in the assessment of cardiac function and prognosis in heart failure patients. Therefore the findings of the present study are discussed from the following perspectives: (i) exercise physiology, (ii) prognostic utility and (iii) feasibility of measurement.

Central haemodynamics and CPO demonstrate low CV and good reproducibility. An increase in exercise duration of 61 s, and consequent increase in metabolic and ventilatory responses, suggest that patients were able to demonstrate greater effort during the second test. These findings are likely to be explained by ‘learning effect’, previously demonstrated in heart failure patients [7], and also perhaps due to day-to-day variability in physiological variables described earlier [33]. Increased V̇O2 peak on the second test was due to a 4% increase in arterial–venous oxygen difference as QT was not different between the two tests. This is an important finding because it suggests that cardiac function and pumping capability are less affected by exercise test ‘learning effect’ and physiological day-to-day variation observed in gas exchange measurements. The assessment of CPO does not require performing more than one cardiopulmonary exercise test.

The present results demonstrate, for the first time, good reproducibility of resting CPO in chronic heart failure patients, having a CV of 5% and correlation coefficient of 0.96. This is an important finding as in contrast with resting metabolic and ventilatory variables, resting CPO may have a practical implication in the detection of therapy-derived changes in cardiac function and prognosis in patients with heart failure [17,18]. QT measured at rest by the inert gas re-breathing methodology (a component of CPO) is also highly reproducible, in agreement with previous reports [26,34], and demonstrates a CV below 10% in heart failure patients. In contrast with central haemodynamics, resting gas exchange measurements demonstrated greater CV. There are many uncontrollable factors that could affect reproducibility of resting metabolic measurements and most are beyond the control of the study. Although all patients were instructed to eat nothing at least 2 h prior to the exercise tests and not to consume alcohol or caffeine-containing beverages before the tests, there is no guarantee that this was indeed the case. Other causes may include variations in the amount of sleep, degree of activity during the previous day, dietary salt load and fluid intake, and emotional state.

The results further reveal good reproducibility of near-maximal exercise CPO demonstrated by a low CV (4.1%), high correlation coefficient (r=0.96), small mean difference (−0.02 W) and narrow limits of agreement (−0.14 to 0.10 W) between repeated measures. This should be unsurprising as both components of CPO (i.e. QT and MAP) demonstrate good reproducibility at peak exercise. To date, only one study has reported the reproducibility of exercise cardiac power in patients with heart failure [25]. In that study, patients performed a treadmill exercise test in two stages with a 40 min resting period between the two stages. The first stage was an incremental test to determine the maximum V̇CO2, and a second maximum workload stage was used to make QT measurements. Unlike that study, we deliberately selected a protocol which is commonly used in clinical assessment. Using different methodology, Cooke et al. [25] reported a CV for CPO of 9.1%. Although the authors [25] questioned the reproducibility of arterial blood pressure measurement by cuff manometry during exercise, they did not report repeated blood pressure measurements, or reproducibility analysis of this variable. However, the good reproducibility and low CV for blood pressure found in the present study has been reported elsewhere [3,7].

Unlike CPO, reproducibility of well-established cardiopulmonary exercise test variables and particularly V̇O2 are well documented in single- and multi-site studies [39]. The present study suggests that exercise performance and gas exchange measurements demonstrate an acceptable level of reproducibility with the CV <10%.

Some investigators have suggested that due to substantial variability and significant differences between repeated measurements of V̇O2 peak and exercise time more than one test should be performed at baseline for clinical trial purposes [7]. Conversely, Russell et al. [9] performed a series of five baseline maximal treadmill tests and found that although there was a significant improvement in exercise time, there was no significant change in V̇O2 peak. Similarly, as in the present study, others also reported high reproducibility for duplicate measurements of cardiopulmonary exercise testing variables in patients with heart failure. These studies report a CV for V̇O2 peak, RER, anaerobic threshold, V̇CO2 and exercise time between 3% and 9% [3,6,8]. In the largest trial (398 heart failure patients), Bensimhon et al. [5] reported V̇O2 peak and anaerobic threshold to be unchanged from test 1 to test 2, with the CVs of 6.6 and 7.8% [5]. In contrast, mean exercise time increased significantly from test 1 to test 2 by 4%. Most recently, in a multi-site clinical trial setting Keteyian et al. [4] assessed reproducibility of V̇O2 peak and anaerobic threshold and reported CVs of 5.9 and 6.8%. The present study confirms good reproducibility of V̇O2 peak and anaerobic threshold. The difference of 5.7% (1.3 ml/kg of body weight per min) is unlikely to have practical and clinical implications and is likely to be an indication that patients were able to tolerate greater effort during the second test as a consequence of ‘learning effect’.

It has been shown that both resting and near-maximal exercise CPO are excellent predictors of prognosis in patients with heart failure [1721]. Also CPO obtained at near-maximal exercise is a direct measure of overall cardiac function and pumping capability of the heart [13,14]. The findings of the present study suggest that both resting and near-maximal exercise CPOs demonstrate good reproducibility. Therefore, it is reasonable to suggest that there is no need for performing more than one cardiopulmonary exercise test to assess CPO in patients with stable chronic heart failure.

Finally, it is important to mention feasibility of QT and CPO measurement during exercise. Monitoring QT at rest and particularly during exercise is not a straightforward procedure [12]. Gold standard methods such as direct Fick and thermodilution are invasive, expensive, require specialist expertise, and are associated with inherent risks. Several non-invasive methods for QT assessment have been developed, such as ultrasound techniques, impedance cardiography, bioreactance and pulse contour analysis. However, validity and reliability of these techniques during exercise have often been questioned since they are sensitive to motion artefacts. On the other hand, re-breathing (CO2 and inert gas) methods are, however, frequently used in both healthy and clinical populations, including heart failure patients. Re-breathing methods, however, require patients to learn the re-breathing technique, are effort dependent and provide discontinuous, single-point measurement. Also from our experience approx. 10–15% of patients are unable to perform the re-breathing technique, and this seems to be associated with severity of disease. Nonetheless, while waiting for an ‘ideal’ method to be developed, re-breathing techniques are appropriate for estimating QT and CPO.

The major limitation of the present study is inclusion of patients with mild heart failure only, with well-preserved exercise capacity. It is possible that those with more advanced stages of heart failure with significantly lower levels of maximal exercise capacity may demonstrate an altered response of cardiac function to increasing exercise levels.

In conclusion, CPO demonstrates good reproducibility suggesting that there is no need for performing more than one cardiopulmonary exercise test. As a direct measure of cardiac function (dysfunction) and an excellent prognostic marker, it is strongly advised in the assessment of patients with chronic heart failure undergoing cardiopulmonary exercise testing. Future research is warranted to evaluate reproducibility of CPO in multi-site protocols.

FUNDING

This work was funded by the Research Centre for Health at Buckinghamshire New University.

We thank all of the patients for taking part in this study.

Abbreviations

     
  • CPO

    cardiac power output

  •  
  • CV

    coefficient of variance

  •  
  • DBP

    diastolic blood pressure

  •  
  • MAP

    mean arterial pressure

  •  
  • QT

    cardiac output

  •  
  • RER

    respiratory exchange ratio

  •  
  • SBP

    systolic blood pressure

  •  
  • V̇CO2

    carbon dioxide production

  •  
  • V̇E

    minute ventilation

  •  
  • V̇O2

    oxygen consumption

  •  
  • V̇O2

    peak V̇O2

AUTHOR CONTRIBUTION

Djordje Jakovljevic designed and conducted the study, and analysed data under the supervision of David Brodie and Richard Grocott-Mason. Petar Seferovic and Michael Trenell provided intellectual input. The study was conceived by David Nunan and Gay Donovan. All authors co-wrote the paper.

References

References
1
Skinner
 
J. S.
Wilmore
 
K. M.
Jaskolska
 
A.
Jaskolski
 
A.
Daw
 
W. E.
Rice
 
T.
Gagnon
 
J.
Leon
 
A. S.
Wilmore
 
J. H.
Rao
 
D. C.
, et al 
Reproducibility of maximal exercise test data in the heritage family study
Med. Sci. Sports Exercise
1999
, vol. 
31
 (pg. 
1623
-
1628
)
2
Weston
 
S. B.
Gabbett
 
T. J.
 
Reproducibility of ventilation of thresholds in trained cyclists during ramp cycle exercise
J. Sci. Med. Sport
2001
, vol. 
4
 (pg. 
357
-
366
)
3
Meyer
 
K.
Westbrook
 
S.
Schwaibold
 
M.
Hajric
 
R.
Peters
 
K.
Roskamm
 
H.
 
Short-term reproducibility of cardiopulmonary measurements during exercise testing in patients with severe chronic heart failure
Am. Heart J.
1997
, vol. 
134
 (pg. 
20
-
26
)
4
Keteyian
 
S. J.
Brawner
 
C. A.
Ehrman
 
J. K.
Ivanhoe
 
R.
Boehmer
 
J. P.
Abraham
 
W. T.
 
Reproducibility of peak oxygen uptake and other cardiopulmonary exercise parameters: implications for clinical trials and clinical practice
Chest
2010
, vol. 
138
 (pg. 
950
-
955
)
5
Bensimhon
 
D. R.
Leifer
 
E. S.
Ellis
 
S. J.
Fleg
 
J. L.
Keteyian
 
S. J.
Pina
 
I. L.
Kitzman
 
D. W.
McKelvie
 
R. S.
Kraus
 
W. E.
Forman
 
D. E.
, et al 
Reproducibility of peak oxygen uptake and other cardiopulmonary exercise testing parameters in patients with heart failure (from the heart failure and a controlled trial investigating outcomes of exercise traiNing)
Am. J. Cardiol.
2008
, vol. 
102
 (pg. 
712
-
717
)
6
Corrà
 
U.
Mezzani
 
A.
Bosimini
 
E.
Giannuzzi
 
P.
 
Prognostic value of time-related changes of cardiopulmonary exercise testing indices in stable chronic heart failure: a pragmatic and operative scheme
Eur. J. Cardiovasc. Prev. Rehabil.
2006
, vol. 
13
 (pg. 
186
-
192
)
7
Elborn
 
J. S.
Stanford
 
C. F.
Nicholls
 
D. P.
 
Reproducibility of cardiopulmonary parameters during exercise in patients with chronic cardiac failure. The need for a preliminary test
Eur. Heart J.
1990
, vol. 
11
 (pg. 
75
-
81
)
8
Marburger
 
C. T.
Brubaker
 
P. H.
Pollock
 
W. E.
Morgan
 
T. M.
Kitzman
 
D. W.
 
Reproducibility of cardiopulmonary exercise testing in elderly patients with congestive heart failure
Am. J. Cardiol.
1998
, vol. 
82
 (pg. 
905
-
909
)
9
Russell
 
S. D.
McNeer
 
F. R.
Beere
 
P. A.
Logan
 
L. J.
Higginbotham
 
M. B.
 
Improvement in the mechanical efficiency of walking: an explanation for the ‘placebo effect’ seen during repeated exercise testing of patients with heart failure
Am. Heart J.
1998
, vol. 
135
 (pg. 
107
-
114
)
10
Corra
 
U.
Mezzani
 
A.
Giordano
 
A.
Bosimini
 
E.
Giannuzzi
 
P.
 
Exercise haemodynamic variables rather than ventilatory efficiency indexes contribute to risk assessment in chronic heart failure patients treated with carvedilol
Eur. Heart J.
2009
, vol. 
30
 (pg. 
3000
-
3006
)
11
Chomsky
 
D. B.
Lang.
 
C. C.
Rayos
 
G. H.
Shyr
 
Y.
Yeoh
 
T.
Pierson
 
R.
Davis
 
S.
Wilson
 
J.
 
Hemodynamic exercise testing: a valuable tool in the selection of cardiac transplantation candidates
Circulation
1996
, vol. 
94
 (pg. 
3176
-
3183
)
12
Lang
 
C. L.
Agostoni
 
P.
Mancini
 
D. M.
 
Prognostic significance and measurement of exercise-derived haemodynamic variables in patients with heart failure
J. Card. Failure
2007
, vol. 
13
 (pg. 
672
-
679
)
13
Tan
 
L. B.
 
Cardiac pumping capability and prognosis in heart failure
Lancet
1986
, vol. 
13
 (pg. 
1360
-
1363
)
14
Cotter
 
G.
Williams
 
S. G.
Vered
 
Z.
Tan
 
L. B.
 
Role of cardiac power in heart failure
Cur. Opin. Cardiol.
2003
, vol. 
18
 (pg. 
215
-
222
)
15
Cohen-Solal
 
A.
Beauvais
 
F.
Tan
 
L. B.
 
Peak exercise responses in heart failure: back to basics
Eur. Heart J.
2009
, vol. 
30
 (pg. 
2962
-
2964
)
16
Tan
 
L. B.
Williams
 
S. G.
Tan
 
D. K.
Cohen-Solal
 
A.
 
So many definitions of heart failure: are they all universally valid? A critical appraisal
Expert Rev. Cardiovasc. Ther.
2010
, vol. 
8
 (pg. 
217
-
228
)
17
Mendoza
 
D. D.
Cooper
 
H. A.
Panza
 
J. A.
 
Cardiac power output predicts mortality across a broad spectrum of patients with acute cardiac disease
Am. Heart J.
2007
, vol. 
153
 (pg. 
366
-
370
)
18
Fincke
 
R.
Hochman
 
J. S.
Lowe
 
A. M.
Menon
 
V.
Slater
 
J. N.
Web
 
J. G.
LeJemtel
 
T. H.
Cotter
 
G.
 
Cardiac power is the strongest hemodynamic correlate of mortality in cardiogenic shock: a report from the shock trial registry
J. Am. Coll. Cardiol.
2004
, vol. 
44
 (pg. 
340
-
348
)
19
Roul
 
G.
Moulichon
 
M. E.
Bareiss
 
P.
Gries
 
P.
Koegler
 
A.
Sacrez
 
J.
Germain
 
P.
Mossard
 
J. M.
Sacrez
 
A.
 
Prognostic factors of chronic heart failure in NYHA class II or III: value of invasive exercise haemodynamic data
Eur. Heart J.
1995
, vol. 
16
 (pg. 
1387
-
1398
)
20
Williams
 
S. G.
Cooke
 
G. A.
Wright
 
D. J.
Parsons
 
W. J.
Riley
 
R. L.
Marshall
 
P.
Tan
 
L. B.
 
Peak exercise cardiac power output: a indicator of cardiac function strongly predictive of prognosis in chronic heart failure
Eur. Heart J.
2001
, vol. 
222
 (pg. 
1496
-
1503
)
21
Lang
 
C. C.
Karlin
 
P.
Haythe
 
J.
Lim
 
T. K.
Mancini
 
D. M.
 
Peak cardiac power output, measured noninvasively, is a powerful predictor of outcome in chronic heart failure
Circ. Heart Failure
2009
, vol. 
2
 (pg. 
33
-
38
)
22
Jakovljevic
 
D. G.
George
 
R. S.
Donovan
 
G.
Nunan
 
D.
Henderson
 
K.
Bougard
 
R. S.
Yacoub
 
M. H.
Birks
 
E. J.
Brodie
 
D. A.
 
Comparison of cardiac power output and exercise performance in patients with left ventricular assist devices, explanted (recovered) patients, and those with moderate to severe heart failure
Am. J. Cardiol.
2010
, vol. 
105
 (pg. 
1780
-
1785
)
23
Jakovljevic
 
D. G.
George
 
R. S.
Donovan
 
G.
Nunan
 
D.
Bougard
 
R. S.
Yacoub
 
M. H.
Birks
 
E. J.
Brodie
 
D. A.
 
The impact of acute reduction of continuous-flow left ventricular assist device support on cardiac and exercise performance
Heart
2010
, vol. 
96
 (pg. 
1390
-
1395
)
24
Jakovljevic
 
D. G.
Birks
 
E. J.
George
 
R. S.
Trenell
 
M. I.
Seferovic
 
P. M.
Yacoub
 
M. H.
Brodie
 
D. A.
 
Relationship between peak cardiac pumping capability and selected exercise-derived prognostic indicators in patients treated with left ventricular assist devices
Eur. J. Heart Failure
2011
, vol. 
13
 (pg. 
992
-
999
)
25
Cooke
 
G. A.
Marshall
 
P.
Al-Timman
 
J. K.
Riley
 
R.
Hainsworth
 
R.
Tan
 
L. B.
 
Physiological cardiac reserve: development of a non-invasive method and first estimates in man
Heart
1998
, vol. 
79
 (pg. 
289
-
294
)
26
Agostoni
 
P.
Cattadori
 
G.
Apostolo
 
A.
Contini
 
M.
Palermo
 
P.
Marenzi
 
G.
Wasserman
 
K.
 
Noninvasive measurement of cardiac output during exercise by inert gas rebreathing technique: a new tool for heart failure evaluation
J. Am. Coll. Cardiol.
2005
, vol. 
46
 (pg. 
1779
-
1781
)
27
Jakovljevic
 
D. G.
Nunan
 
D.
Donovan
 
G.
Hodges
 
L. D.
Sandercock
 
G. R.
Brodie
 
D. A.
 
Comparison of cardiac output determined by different rebreathing methods at rest and at peak exercise
Eur. J. Appl. Physiol.
2008
, vol. 
102
 (pg. 
593
-
599
)
28
Lang
 
C.
Karlin
 
P.
Haythe
 
J.
Tsao
 
L.
Mancini
 
D. M.
 
Ease of noninvasive measurement of cardiac output coupled with peak VO2 determination at rest and during exercise in patients with heart failure
Am. J. Cardiol.
2007
, vol. 
99
 (pg. 
404
-
405
)
29
Beaver
 
W. L.
Wasserman
 
K.
Whipp
 
B. J.
 
A new method for detecting anaerobic threshold by gas exchange
J. Appl. Physiol.
1986
, vol. 
60
 (pg. 
2020
-
2027
)
30
Bland
 
M. J.
Altman
 
D. G.
 
Statistical methods for assessing agreement between two methods of clinical measurements
Lancet
1986
, vol. 
i
 (pg. 
307
-
310
)
31
Myers
 
J.
Froelicher
 
V. F.
 
Optimizing the exercise tests for pharmacological investigations
Circulation
1990
, vol. 
82
 (pg. 
1849
-
1856
)
32
Janicki
 
J. S.
Gupta
 
S.
Ferris
 
S. T.
McElroy
 
P. A.
 
Long-term reproducibility of respiratory gas exchange measurements during exercise in patients with stable cardiac failure
Chest
1990
, vol. 
97
 (pg. 
12
-
17
)
33
Armstrong
 
L. E.
Costill
 
D. L.
 
Variability of respiration and metabolism: responses to sub maximal cycling and running
Res. Q. Exercise Sport
1985
, vol. 
56
 (pg. 
93
-
96
)
34
Christensen
 
P.
Clemensen
 
P.
Anderson
 
P. K.
Henneberg
 
S. W.
 
Thermodilution versus inert gas rebreathing for estimation of effective pulmonary blood flow
Crit. Care Med.
2000
, vol. 
28
 (pg. 
51
-
56
)