European Journal of Heart Failure 2008 10(1):85-88; doi:10.1016/j.ejheart.2007.10.010
© 2008 European Society of Cardiology
Combining the ventilatory response to exercise and peak oxygen consumption is no better than peak oxygen consumption alone in predicting mortality in chronic heart failure
Lee Inglea,
Klaus K. Wittec,*,
John G.F. Clelandb and
Andrew L. Clarkb
a Carnegie Research Institute, Leeds Metropolitan University Leeds, UK
b Academic Department of Cardiology, University of Hull Hull, UK
c Department of Cardiology, Leeds General Infirmary Leeds, UK
* Corresponding author. Leeds General Infirmary, Great George Street, Leeds, LS1 3EX, United Kingdom. Tel.: +44 113 2787206. E-mail address: klauswitte{at}hotmail.com (K. K. Witte).
 |
Abstract
|
|---|
Background: A low peak oxygen uptake (pV
O2) and steep V
E/V
CO2 slope are
independently associated with a worse prognosis in patients
with chronic heart failure (CHF). We wished to confirm whether
combining these variables as a ratio would lead to a more accurate
predictor of prognosis than using either alone.
Methods: 388 CHF patients completed a treadmill-based cardiopulmonary exercise test (CPET) to volitional exhaustion using a modified Bruce protocol.
Results: 212 CHF patients completed the CPET with a peak RER 
1.0. Of these, 48 patients died and one was transplanted during follow-up. In surviving patients, the median follow-up period was 42 months (IQR 34–49 months). The ratio VE/VCO2 slope/pVO2 was calculated for each individual and its ability to predict outcome compared with other variables. The Cox multivariable survival analysis showed that pVO2 was the strongest independent predictor of mortality in CHF patients.
Conclusion: Our study shows that the composite variable VE/VCO2 slope/pVO2 is a less effective prognosticator than pVO2 alone in patients with CHF.
Key Words: Exercise testing • Cardiopulmonary exercise testing • Prognosis
Received June 28, 2007; Revised September 21, 2007; Accepted October 18, 2007
|
1. Introduction
|
|---|
The cardinal symptom of chronic heart failure (CHF) is exercise
intolerance, usually due to breathlessness and fatigue.
[1] This
can be objectively assessed as a reduction in peak oxygen consumption
(pV
O2) during cardiopulmonary exercise testing (CPET).
[2] Patients
also have an increased ventilatory response throughout exercise
as shown by an increase in the slope relating ventilation to
carbon dioxide production (V
E/V
CO2 slope).
[3-5] Rather than
examining each of these two variables separately, a combination,
presented as a ratio, might be better at predicting outcome
in patients with a range of heart failure than either alone.
In 100 CHF patients, the ratio V
E/V
CO2 slope/pV
O2 related to
prognosis more closely than either variable alone.
[6] However,
many exercise variables are highly related, and deviations of
single data-points can change the order of prognostic information
particularly when one variable is a composite of two others
included in a multivariate model. We wished therefore to assess,
in a larger unselected population of ambulatory CHF patients
followed for a longer period of time, with a larger number of
endpoints, whether combining pV
O2 and V
E/V
CO2 slope as a ratio
leads to more accurate prediction of prognosis than using either
alone.
|
2. Methods
|
|---|
The Hull and East Riding Ethics Committee approved the study,
and all patients provided informed consent for participation.
We reviewed baseline exercise tests of 388 CHF patients under
follow-up at a community heart failure centre. All patients
were symptomatic with a left ventricular ejection fraction <45%,
but had been stable without exacerbation, hospital admission
or medication change for the preceding 3 months. We did not
include patients with neurological conditions, limiting anginal
pain or pulmonary disease (FEV
1 <80% of predicted).
All subjects underwent a treadmill-based symptom-limited cardiopulmonary exercise test (CPET) with metabolic gas exchange (Jaeger Oxycon Delta, Viasys, USA) using an incremental Bruce protocol modified by the addition of a ‘stage 0’ at onset consisting of 3 min of exercise at 1.61 km/h (1 mile/h) with a 5% gradient. A peak respiratory exchange ratio (RER), (VCO2/VO2) of 1.0 was taken to indicate maximal effort [7]. Peak oxygen uptake (pVO2) was calculated as the average VO2 for the last 30 s of exercise. The VE/VCO2 slope (full) was calculated by linear regression by analysing breath-by-breath values obtained throughout the full test from all data points.[8] The anaerobic threshold (AT) was calculated using the VO2/VCO2 slope method, [9] The VE/VCO2 slope to pVO2 ratio (VE/VCO2/pVO2) was calculated for each patient. Patients were excluded from the survival analysis if they completed the CPET with a peak RER (pRER) <1.0. Follow-up was censored on 1st July 2006.
Results are reported as median±inter-quartile range (IQR). Categorical data are presented as percentages. Cox regression models were developed to predict candidate univariate predictors of mortality using all baseline variables. All variables in Table 1 (pRER 1.0) were entered as potential candidate predictors of mortality. We checked for collinearity between candidate univariate predictors by calculating Pearson's correlation coefficients using a cut-off of r>0.3. Model building was based on backwards elimination (P-value for entry was <0.05; P-value for removal >0.1). We then constructed a Cox regression model to calculate hazard ratios with 95% CI. SPSS (version 13.0) was used to analyze the data. An arbitrary level of 5% statistical significance was used throughout (two-tailed).
|
3. Results
|
|---|
Table 1 shows data for all patients dichotomised by those achieving
a pRER more or less than 1.0. Clinical profile and functional
characteristics were similar between groups. From an initial
population of 388, 212 patients completed the CPET with a pRER
1.0, of whom 48 patients died and one was transplanted representing
an event rate of 22.9%. In surviving patients, the median follow-up
time was 42 months (IQR 34-49 months). In CHF patients, Kaplan-Meier
survival curves showing cumulative survival according to pV
O2,
V
E/V
CO2 slope (full), and V
E/V
CO2 slope/pV
O2 are reported in
Figs. 1-3
. Six variables were significantly associated with
mortality on univariate Cox analysis. Each was included in the
final multivariable Cox model (
Table 2), and each was an independent
predictor of mortality.
However, when we ran a correlation analysis to identify collinearity
between variables we were forced to eliminate the following;
V
E/V
CO2 slope/pV
O2, peak heart rate, V
E/V
CO2 slope (full), V
E/V
CO2@AT,
and anaerobic threshold (
r>0.3;
P<0.001). Peak oxygen
uptake remained in the model as it had the highest
2 value (46.2).
Thus, the strongest independent predictor of events was pV
O2 (in CPETs with a pRER
1.0). V
E/V
CO2 slope/pV
O2 was the next
strongest predictor (
Table 3). We also ran a sub-group analysis
on beta-blocked patients only, V
E/V
CO2 slope/pV
O2 was removed
from the final multivariable model (loss in Chi-Square=4.2)
at Step 1, and pV
O2 remained the most potent prognosticator
(HR=0.763;
P=0.002; 95% CI=0.640-0.909).
|
4. Conclusions
|
|---|
Our findings suggest that pV
O2 is a more effective prognosticator
than V
E/V
CO2 slope (full), V
E/V
CO2@AT, or the composite variable
V
E/V
CO2 slope/pV
O2 in CHF patients. This finding is in conflict
with that of Guazzi et al.
[6] who reported on 100 CHF patients
(mean age 59 years). Over their follow-up period of 26±19
months, 21 patients died from cardiovascular causes. Cox multivariable
analysis revealed that the V
E/V
CO2 slope was a stronger predictor
of mortality than pV
O2; however, the composite V
E/V
CO2 slope/pV
O2 ratio more effectively predicted outcome than either variable
alone. Our findings are in a larger and older group of patients
with a longer follow-up period (42±28 months vs 26±19
months). It should be noted that our primary end-point was all-cause
mortality while Guazzi included cardiovascular deaths only.
In addition, beta-blocker use in our patients was higher (68%
vs 30%). However, although the increased ventilation seen in
CHF is related to sympathetic activity,
[10] there is little
influence of long term beta-blockade on the V
E/V
CO2 slope
[11] and these agents also do not significantly alter pV
O2.
[11-13] Beta-blockers do however reduce mortality and for a given pV
O2,
a patient taking a beta-blocker has a lower overall mortality
than a patient not taking one.
It has been suggested that pVO2 might be less powerful a predictor of outcome in patients with mild-moderate heart failure.[14] However, in our study more than 75% of patients had NYHA class I/II symptoms, and median LVEF was 34% (29-40), yet pVO2 remained the most potent prognostic indicator. Our findings thereby also disagree with studies that have argued that an elevated VE/VCO2 slope is a more powerful outcome measure than pVO2.[15,16]
In conclusion, our findings indicate that the composite variable VE/VCO2 slope/pVO2 is a less effective prognosticator than pVO2 alone in patients with CHF. The VE/VCO2 slope/pVO2 ratio is however a better predictor of mortality than other well-researched univariate predictors including VE/VCO2 slope (full), and VE/VCO2 slope@AT.
|
5. Limitations
|
|---|
Only just over half of our patients achieved a pRER >1.0
on our modified Bruce protocol. The use of this protocol on
a treadmill may be too challenging for our patients, and alternative
modes/methods of exercise testing such as a more gentle ramping
treadmill protocol or cycle ergometry, may yield greater success.
Despite a greater number of endpoints in our study, a major limitation of such an analysis remains that the variables gained from an exercise test are closely related, and deviations in single data-points can lead to changes in ranking of prognostic ability, particularly when one of the variables included in the model is a composite of two others.
|
References
|
|---|
- Clark A.L., Sparrow J.L. Coats AJS. Muscle fatigue and dyspnoea in chronic heart failure: two sides of the same coin? Eur Heart J (1995) 16:49–52.[Free Full Text]
- Clark A.L., Poole-Wilson P.A., Coats A.J.S. Exercise limitation in chronic heart failure: central role of the periphery. J Am Coll Cardiol (1996) 28:1092–1102.[Abstract]
- Buller N.P., Poole-Wilson P.A. Mechanism of the increased ventilatory response to exercise in patients with chronic heart failure. Br Heart J (1990) 63:281–283.[Abstract/Free Full Text]
- Davies S.W., Emery T.M., Watling M.I., Wannamethee G., Lipkin D.P. A critical threshold of exercise capacity in the ventilatory response to exercise in chronic heart failure. Br Heart J (1991) 65:179–183.[Abstract/Free Full Text]
- Witte K.K.A., Clark A.L. Is the elevated slope relating ventilation to carbon dioxide production in chronic heart failure a consequence of slow metabolic gas kinetics? Eur J Heart Fail (2002) 4(4):469–472.[Abstract/Free Full Text]
- Guazzi M., De Vita S., Cardano P., Barlera S., Guazzi M.D. Normalization for peak oxygen uptake increases the prognostic power of the ventilatory response to exercise in patients with chronic heart failure. Am Heart J (2003) 146:542–548.[CrossRef][Web of Science][Medline]
- Mezzani A., Corra U., Bosimini E., Giordano A., Giannuzzi P. Contribution of peak respiratory exchange ratio to peak VO2 prognostic reliability in patients with chronic heart failure and severely reduced exercise capacity. Am Heart J (2003) 145:1102–1107.[CrossRef][Web of Science][Medline]
- Ingle L, Goode K, Carroll S, et al. Prognostic value of the VE/VCO2 slope calculated from different time intervals in patients with suspected heart failure. Int J Cardiol (2007) 12;118(3):350–355.
- Beaver W.L, Wasserman K, Whipp B.J. A new method for detecting the anaerobic threshold by gas exchange. J Appl Physiol (1986) 60:2020–2027.[Abstract/Free Full Text]
- Australia/New Zealand Heart Failure Research Collaborative Group. Randomised, placebo-controlled trial of carvedilol in patients with congestive heart failure due to ischaemic heart disease. Lancet (1997) 349:375–380.[CrossRef][Web of Science][Medline]
- Witte K.K., Thackray S.D., Nikitin N.P., Cleland J.G., Clark A.L. The effects of
- and β-blockade on ventilatory responses to exercise in chronic heart failure. Heart (2003) 89:1169–1173.[Abstract/Free Full Text] - Guazzi M., Agostoni P., Matturri M, Pontone G, Guazzi MD. Pulmonary function, cardiac function, and exercise capacity in a follow-up of patients with congestive heart failure treated with carvedilol. Am Heart J (1999) 138:460–467.[CrossRef][Web of Science][Medline]
- Wolk R, Johnson BD, Somers VK, et al. Effects of beta-blocker therapy on ventilatory responses to exercise in patients with heart failure. J Card Fail (2005) 11:333–339.[CrossRef][Web of Science][Medline]
- Meyers J, Gullestad L. The role of exercise testing and gas exchange measurements in the prognostic assessment of patients with heart failure. Curr Opin Cardiol (1998) 13:145–155.[Web of Science][Medline]
- Robbins M, Francis G, Pashkow FJ, et al. Ventilatory and heart rate response to exercise: better predictors of heart failure mortality than peak oxygen consumption. Circulation (1999) 100:2411–2417.[Abstract/Free Full Text]
- Kleber FX, Vietzke G, Wernecke KD, et al. Impairment of ventilatory efficiency in heart failure. Circulation (2000) 101:2803–2809.[Abstract/Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
Top
Abstract
1. Introduction
