Oxygen uptake efficiency slope, a new submaximal parameter in evaluating exercise capacity in chronic heart failure patients Christophe Van Laethem, MSc, Jozef Bartunek, MD, PhD, Marc Goethals, MD, Paul Nellens, MD, Erik Andries, MD, and Marc Vanderheyden, MD Aalst, Belgium
Background
The oxygen uptake efficiency slope (OUES) is a new submaximal parameter which objectively predicts the maximal exercise capacity in children and healthy subjects. However, the usefulness of OUES in adult patients with and without advanced heart failure remains undetermined. The present study investigates the stability and the usefulness of OUES in adult cardiac patients with and without heart failure.
Methods Forty-five patients with advanced heart failure (group A) and 35 patients with ischemic heart disease but normal left ventricular ejection fraction (group B) performed a maximal exercise test. PeakVO2 and percentage of predicted peakVO2 were markers of maximal exercise capacity, whereas OUES, ventilatory anaerobic threshold (VAT), and slope VE/VCO2 were calculated as parameters of submaximal exercise. Results
Group A patients had lower peakVO2 (P ⬍ .001), lower percentage of predicted peakVO2 (P ⫽ .001), lower VAT (P ⬍ .05), steeper slope VE/VCO2 (P ⬍ .001), and lower OUES (P ⬍ .02). Within group A, significant differences were found for VAT, slope VE/VCO2, and OUES (all P ⬍ .01) between patients with peakVO2 above and below 14 mL O2/kg/min. Of all the submaximal parameters, VAT correlated best with peakVO2 (r ⫽.814, P ⬍ .01) followed by OUES/kg (r ⫽ .781, P ⬍ .01), and slope VE/VCO2 (r ⫽ ⫺.492, P ⬍ .001). However, VAT could not be determined in 18 (23%) patients.
Conclusions OUES remains stable over the entire exercise duration and is significantly correlated with peakVO2 in adult cardiac patients with and without impaired LVEF. Therefore, OUES could be helpful to assess exercise performance in advanced heart failure patients unable to perform a maximal exercise test. Further studies are needed to confirm our hypothesis. (Am Heart J 2005;149:175– 80.) Since the original report of Mancini,1 numerous studies have demonstrated that oxygen consumption at peak exercise provides valuable information in the evaluation of heart failure patients.2,3 A consistent finding has been that short-term survival is poor when peakVO2 is ⬍10 mL/kg/min1,4 and that cardiac transplantation can be safely deferred in patients with a peakVO2 ⬎14 mL/kg/min.1–7 However, peakVO2 might be underestimated and might be a less reliable parameter because of reduced patient motivation, selected exercise protocol, and knowledge and skills of the examiner. Because of these pitfalls, submaximal exercise parameters such as the ventilatory anaerobic threshold
From the Cardiovascular Center, Onze Lieve Vrouw Hospital, Aalst, Belgium. Submitted February 19, 2004; accepted July 3, 2004. Reprint requests: Marc Vanderheyden, MD, Cardiovascular Center, Onze Lieve Vrouw Ziekenhuis, Moorselbaan 164, 9300 Aalst, Belgium. E-mail:
[email protected] 0002-8703/$ - see front matter © 2005, Elsevier Inc. All rights reserved. doi:10.1016/j.ahj.2004.07.004
(VAT) have been introduced to evaluate the cardiopulmonary functional reserve.8,9 However, VAT cannot be obtained in 25% to 30 % of patients with heart failure because of severe deconditioning, early onset of acidosis, and the presence of an irregular breathing pattern.10 In their attempt to develop an objective and independent submaximal measure of cardiorespiratory reserve, Baba et al11 introduced the oxygen uptake efficiency slope (OUES) derived from the logarithmic relation between oxygen uptake and minute ventilation during incremental exercise. This new slope represents how effectively the oxygen is extracted by the lungs and used in the periphery. It can be calculated from submaximal exercise data and using a set of cardiopulmonary gas analysis data, without depending on interobserver and intraobserver variability.11 Initially, OUES has been clinically applied in a population of pediatric patients with heart disease11 and in healthy adults.12,13 Finally, scarce data exist regarding OUES in adult patients with heart failure. The present study
American Heart Journal January 2005
176 Van Laethem et al
was set up to evaluate the stability of OUES at different exercise stages in a group of adult cardiac patients with and without left ventricular dysfunction. Furthermore, it investigated the relationship between this parameter and other markers of maximal and submaximal cardiopulmonary exercise capacity.
Figure 1
Methods Study population The study population consisted of 80 consecutive patients who visited the outpatient clinic between January and July 2003 prior to their enrollment in a cardiac rehabilitation program. All patients performed a maximal cardiopulmonary exercise test with a respiratory exchange ratio (RERmax) ⬎1.10 and were free of exercise-limiting comorbidities, such as cerebrovascular disease, musculoskeletal impairment, or vascular disease of the lower extremities. Patients were divided into 2 groups according to the presence of LV dysfunction. Group A consisted of 45 patients with LV dysfunction, characterized by a left ventricular ejection fraction (LVEF) ⬍40%, whereas group B included 35 patients suffering from ischemic heart disease (IHD) with an LVEF ⬎50%.14 The oxygen uptake efficiency slope was calculated in each patient and compared between both groups.
Exercise testing Subjects were exercised on a computer-driven cyclo-ergometer (Marquette Case 8000, Marquette Electronics, Milwaukee, Wisc) using a ramp protocol starting at 20 watts with gradual increase of 10 watts every minute. The 12-lead electrocardiogram and heart rate were recorded continuously during the test. Cuff blood pressure was measured every 2 minutes of the exercise test with a manual manometer. Subjects were exercised to their self-determined maximal capacity or until the physician stopped the test because of significant symptoms, such as chest pain or dizziness, potentially dangerous arrhythmias or ST-segment deviations, or marked systolic hypotension or hypertension.
Respiratory gas measurements Continuous respiratory gas measurements were obtained by using a Medical Graphics Cardiopulmonary Exercise System (Medical Graphics, Minneapolis, Minn). The oxygen consumption (VO2), CO2 production (VCO2), minute ventilation (VE), tidal volume, respiratory rate, and mixed expiratory CO2 concentration were continuously measured on a breathby-breath basis. In addition several derived variables, such as the respiratory exchange ratio and the ventilatory equivalent for oxygen (VE/VO2) and carbon dioxide (VE/VCO2), were calculated. PeakVO2 was expressed as the highest attained VO2 during the final 30 seconds of exercise. Ventilatory anaerobic threshold was defined as the level of oxygen uptake during exercise when one of the following occurred: 1) an increase in VE/VO2 without a simultaneously increase in VE/ VCO2, or 2) the disappearance of the linear relationship between VCO2 and VO2 (using the V-slope method). 15 VE/ VCO2-slope was determined by linear regression analysis of the relation between VE and VCO2 during exercise, with data obtained over the complete duration of the exercise test (in-
The relationship between VO2 and VE during incremental exercise in a 52-year old male patient with congestive heart failure. A, Semilog A-axis (OUES). B, Linear X-axis (VO2/VE).
cluding respiratory compensation).16 Flow meters and gas analyzers were calibrated for accuracy and linearity with a syringe of known volume and with precisely analyzed gas mixtures on a daily basis.
Measurement of OUES The relationship between oxygen uptake and ventilation volume is expressed by the OUES. This index is best described by a single-segment logarithmic curve-fitting model using the following equation VO2 ⫽ a ⫻ log VE ⫹ b, in which the constant a represents the rate of increase in VO2 in response to an increase in VE. This constant is called OUES. A steeper slope or higher OUES represents a more efficient oxygen uptake, whereas a more shallow slope or lower OUES represents a higher amount of ventilation required for any given oxygen uptake (Figure 1). In order to evaluate the linearity and as a consequence the usefulness of OUES during a submaximal exercise test, OUES was also calculated from data taken from the first 50% (OUES50) and 75% (OUES75) of the entire maximal exercise duration. Additionally, OUES50 and OUES75 were correlated to peakVO2.
Statistics All results are expressed as mean ⫾ SD. Student t test, analysis of variance, Mann-Whitney U test, and the Spearman correlation coefficient were used for appropriate comparisons. The relation between peakVO2 and submaximal variables and the correlation between OUES values at different exercise intensities were assessed by linear regression analy-
American Heart Journal Volume 149, Number 1
Van Laethem et al 177
Table I. Clinical and demographic characteristics of patients with (Group A) and without (Group B) LV dysfunction
Number (n) Age (y) Sex (% male) IHD (%) BMI LVEF (%) -blocker user (%) ACE-I user (%)
Group A
Group B
45 64 ⫾ 6 89 ⫾ 5 82 ⫾ 6 25.0 ⫾ 3.9 23 ⫾ 9 84 93
35 58 ⫾ 10* 89 ⫾ 5 100* 26.0 ⫾ 4.0 63 ⫾ 9* 83 89
IHD, Ischemic heart disease; BMI, body mass index; LVEF, left ventricular ejection fraction; ACE, angiotensin-converting enzyme. *P ⬍ .05
Table II. Exercise parameters of patients with (Group A) and without (Group B) LV dysfunction Parameter Number (n) Peak RER PeakVO2 % of Predicted peakVO2 VAT Slope VE/VCO2 PeakVE/VCO2 PeakVE/VO2 OUES50 OUES75 OUES OUES/kg
Group A
Group B
45 1.19 ⫾ 0.08 13.6 ⫾ 4.4 54 ⫾ 16 10.6 ⫾ 3.3 38.3 ⫾ 9.8 41.9 ⫾ 7.2 52.1 ⫾ 13.2 1302 ⫾ 474 1279 ⫾ 469 1234 ⫾ 459 17.8 ⫾ 6.2
35 1.22 ⫾ 0.10 22.0 ⫾ 3.8* 73 ⫾ 14* 14.0 ⫾ 2.1* 29.5 ⫾ 5.0* 32.8 ⫾ 4.7* 40.6 ⫾ 5.6* 1777 ⫾ 445* 1790 ⫾ 375* 1820 ⫾ 396* 22.3 ⫾ 5.1*
*P ⬍ .05
sis. Differences in OUES at different levels of exercise were assessed by analysis of variance. Statistical significance was set at a 2-tailed probability level of ⬍.05.
Results
Table III. Exercise parameters for group A1 (⬍14 mLO2/ kg䡠min) and group A2 (⬎14 mLO2/kg䡠min) within LVdysfunction group
Baseline clinical and demographic characteristics Characteristics of patients with and without LV dysfunction are detailed in Table I. By definition group A patients had a lower EF (P ⬍ .01) and a higher New York Heart Association (NYHA) class (P ⬍ .01). No difference in age, body mass index, or sex was noted between the groups. The incidence of ischemic heart disease was significantly higher in group B patients (P ⬍ .01). No difference was noted between both groups in the percentage of patients taking angiotensin-converting enzyme inhibitors or -blockers. All individuals in the study exercised to exhaustion and stopped due to fatigue or acute shortness of breath. They all achieved a RER ⱖ1.10, indicating sufficient metabolic stress. Maximal heart rate and peak RER were similar between both groups. However, group A patients had a significantly lower peakVO2 (P ⬍ .01) and percentage of predicted peakVO2 (P ⬍ .01) compared to group B patients. Submaximal exercise parameters such as VAT (P ⬍ .01) and slope VE/ VCO2 (P ⬍ .01) were significantly different in group A patients compared to group B patients (Table II). Within group A, we divided our patients into 2 subgroups. One group consisted out of 22 patients with a peakVO2 ⬍ 14 mL O2/kg/min (group A1); while the other group of 23 patients had a peakVO2 ⱖ 14mL O2/kg/min (group A2). Group A1 had significant lower values for peakVO2, percentage of predicted peakVO2, and VAT. Slope VE/VCO2 and peakVE/VCO2 were significantly higher in this group (Table III).
Oxygen uptake efficiency slope In the entire study population, OUES could easily be calculated and ranged from 313 to 2528 (1467 ⫾
Number (n) Age (y) BMI LVEF (%) PeakVO2 % of predicted peakVO2 VAT Slope VE/VCO2 PeakVE/VCO2 PeakVE/VO2 OUES OUES/kg
Group A1
Group A2
22 65 ⫾ 4 26.8 ⫾ 3.7 20 ⫾ 5 11.0 ⫾ 2.3 46 ⫾ 7 8.1 ⫾ 1.7 41.4 ⫾ 9.9 44.5 ⫾ 7.1 54.6 ⫾ 11.9 977 ⫾ 300 12.3 ⫾ 3.9
23 63 ⫾ 6 23.9 ⫾ 3.6 25 ⫾ 11 18.2 ⫾ 2.8* 70 ⫾ 12* 13.0 ⫾ 2.7* 33.3 ⫾ 5.2* 39.4 ⫾ 6.5* 49.8 ⫾ 14.3 1561 ⫾ 405* 21.4 ⫾ 4.5*
*P ⬍ .01
481). In all individuals, the OUES50 and OUES75 did not differ significantly from OUES. OUES50 slightly overestimated OUES with 2.98% ⫾ 3.14%, while OUES75 overestimated OUES by as much as 2.23% ⫾ 2.15%. Figure 2 shows the values obtained at 50%, 75%, and 100% of exercise duration for VO2, slope VE/VCO2, and normalized OUES for body weight. There was a significant influence of the foreshortened exercise duration on the calculated values for peakVO2 and slope VE/VCO2, however OUES/kg remained stable during the second part of the exercise test. Similar to other submaximal exercise parameters OUES (P ⬍ .001) as well as OUES normalized for body weight (P ⬍ .001) were significantly lower in group A compared to group B. Within group A, OUES and OUES/kg were significantly lower in the patients with a peakVO2⬍14ml/kg/min (group A1) (Table III).
American Heart Journal January 2005
178 Van Laethem et al
Figure 2
Effects of foreshortened exercise duration on peakVO2, slope VE/VCO2, and normalized OUES for body weight. Calculated values for peakVO2 and slope VE/VCO2 change significantly, while OUES/kg remains stable during the second part of exercise.
Table IV. Correlation between submaximal exercise parameters and peakVO2 Submaximal variable VAT OUES OUES/kg SlopeVE/VCO2
Correlation with peakVO2 r r r r
⫽ ⫽ ⫽ ⫽
0.814 0.684 0.781 0.457
P ⬍.001 ⬍.001 ⬍.001 ⬍.005
Submaximal exercise parameters and peakVO2 From all submaximal exercise parameters VAT correlated best with peakVO2 (r ⫽ 0.814, P ⬎ .01) (Table IV). However, it could not be determined in 23% of our patients (11 in group A and 7 in group B). OUES highly correlated with peakVO2 (r ⫽ 0.684, P ⬍ .001), a correlation which even became more significant when OUES was normalized for body weight (r ⫽ 0.781, P ⬍ .001) (Figure 3). These correlations were not only present in the whole study population but also in both groups of patients with and without LV dysfunction. A significant correlation between peakVO2 and OUES (r ⫽ 0.751, P ⬍ .001) and peakVO2 and OUES/kg (r ⫽ 0.861, P ⬍ .001) was noted in the heart failure subgroup as well.
Discussion Although peakVO2 is considered as a gold standard for stratification of heart failure patients, its use is limited by conditioning status, patient motivation, skeletal muscle structure, pulmonary function, and hemoglobin level. Because of all these limitations, some patients will not achieve their anaerobic threshold during exercise and peakVO2 can not be accurately assessed. To overcome these limitations, submaximal metabolic pa-
rameters have been proposed as surrogates to estimate exercise capacity. The most frequently used is VAT, defined as the VO2 at which CO2 production increases disproportionately to aerobic metabolism. As demonstrated in our and previous studies,17 this index significantly correlates with peakVO2 and is able to identify a subset of patients at high risk for early death from congestive heart failure. However, there are various methods to calculate this index, making it subject to high interobserver and intraobserver variability and lower reproducibility.18 –22 Baba et al11 introduced OUES, a novel index derived from the logarithmic relation between oxygen uptake and minute ventilation during incremental exercise. It is best described by a single exponential function. The physiologic basis of the index is the development of metabolic acidosis, the physiologic dead space, and the arterial CO2 partial pressure. Therefore, OUES incorporates in a single index cardiovascular and peripheral factors that determine oxygen uptake as well as pulmonary factors that influence ventilatory response to exercise. In our study, we extrapolated the observations of Baba et al to a group of adult patients with and without LV dysfunction. OUES remains constant during an incremental exercise test with strong correlation of OUES values obtained at 50%, 75%, and 100% of exercise duration. The OUES differed by ⬍3% when it was calculated from data collected during the first 75% of the exercise test compared with the OUES derived from data from the entire exercise duration. The linear relationship between submaximal and maximal OUES allows its use even in patients with LV dysfunction unable to perform a maximal exercise test. In this regard, we noticed a strong positive correlation between OUES and peakVO2, which even increased when OUES was normalized for body weight. Thus, taken together OUES may be particularly interesting in the evaluation
American Heart Journal Volume 149, Number 1
Figure 3
Van Laethem et al 179
formance might be helpful to differentiate cardiac patients with and without LV dysfunction and could play a role as a prognostic effort-independent marker in the evaluation of heart failure patients. However, this study was not properly designed to test this hypothesis and further prospective trials that validate a prediction equation between OUES and peakVO2 are needed.
Limitations Following limitations should be considered. First, the study population was quite small. Second, we did not assess the reproducibility of the OUES in this study. However, recent studies found a good reproducibility of this variable in a pediatric as well as in a healthy adult population..11–13 Third, we did not assess the prognostic value of OUES in our population. Recently, VAT and slope VE/VCO2 were suggested as being 2 independent prognostic variables in predicting survival and hospitalization in congestive heart failure.17 The prognostic value of OUES remains to be determined. Finally, Hollenberg et al12 demonstrated in a healthy population that OUES is influenced by sex and lean body mass. We did not measure lean body mass in our study.
Conclusions
Top, linear relationship between OUES/kg and peakVO2. Bottom, linear relationship between OUES and peakVO2 for all patients (with and without LV dysfunction).
of heart failure patients who are unable to perform a maximal exercise test. Although the correlation between VAT and peakVO2 was higher in our study, VAT is difficult to obtain in 25% of heart failure patients, which makes this index less useful in daily clinical practice. Therefore, OUES provides an objective index of cardiopulmonary function that is more easily attained than VAT in all tested subjects and less influenced by subjective motivational factors than is peakVO2. Finally, we noticed a lower OUES in patients with LV dysfunction compared to age-matched cardiac patients with normal LV function. The earlier onset of lactate acidosis during exercise, the excessive hyperventilation, and the higher physiologic pulmonary dead space ventilation all account for the depressed OUES observed in heart failure patients. Based upon these preliminary results, we propose that this novel index of cardiopulmonary exercise per-
This study investigated the usefulness of OUES in the evaluation of patients with and without LV dysfunction. In summary, our findings demonstrated that OUES is a stable parameter that can be used in the evaluation of adult cardiac patients, without any significant intra-individual difference in OUES values. In addition, a significant correlation between OUES and peakVO2 was noticed in our patients. Furthermore, OUES was significantly lower in patients with LV dysfunction compared to patients with normal LV function. This suggests that OUES may be a useful parameter in the evaluation of exercise capacity of patients unable to perform a maximal exercise test. However, prospective studies in a larger population are needed to further validate our preliminary results and to determine the prognostic significance of this submaximal exercise parameter.
References 1. Mancini DM, Eisen H, Kussmaul W, et al. Value of peak exercise oxygen consumption for optimal timing of cardiac transplantation in ambulatory patients with heart failure. Circulation 1991;83: 778 – 86. 2. Myers J, Gullestad L. The role of exercise testing and gas exchange measurement in the prognostic assessment of patients with heart failure. Curr Opin Cardiol 1998;13:145–55. 3. Roul G, Moulichon M-E, Bareiss P, et al. Exercise peak VO2 determination in chronic heart failure: is it still of value? Eur Heart J 1994;15:495–502.
180 Van Laethem et al
4. Opasich C, Pinna GD, Bobbio M, et al. Peak exercise consumption in chronic heart failure: toward efficient use in the individual patient. J Am Coll Cardiol 1998;31:766 –75. 5. Stelken AM, Younis LT, Jennison LH, et al. Prognostic value of cardiopulmonary exercise testing using percent achieved of predicted peak oxygen uptake for patients with ischemic and dilated cardiomyopathy. J Am Coll Cardiol 1996;27:345–52. 6. Myers J, Gullestad L, Vagelos R, et al. Clinical, hemodynamic and cardiopulmonary exercise test determinants of survival in patient referred for evaluation of heart failure. Ann Int Med 1998;129: 286 –93. 7. Pina IL. Optimal candidates for heart transplantation: is 14 the magic number? J Am Coll Cardiol 1995;26:436 –7. 8. Weber KL, Kinsewitz GT, Janicki AP. Oxygen utilization and ventilation during exercise in patients with chronic cardiac failure. Circulation 1982;65:1213–23. 9. Beaver WL, Wasserman K, Whipp BJ. A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol 1986;60: 2020 –7. 10. Cohen-Solal A, Benessiano J, Himbert D, et al. Ventilatory threshold during exercise in patients with mild to moderate heart failure: determination, relation with lactate threshold and reproducibility. Int J Cardiol 1991;30:321–7. 11. Baba R, Nagashima M, Goto M, et al. Oxygen uptake efficiency slope: a new index of cardiorespiratory functional reserve derived from the relation between oxygen uptake and minute ventilation during incremental exercise. J Am Coll Cardiol 1996;28:1567– 72. 12. Hollenberg M, Tager IB. Oxygen uptake efficiency slope: An index of exercise performance and cardiopulmonary reserve requiring only submaximal exercise. J Am Coll Cardiol 2000;36:194 –201.
American Heart Journal January 2005
13. Baba R, Kubo N, Morotome Y, et al. Reproducibility of the oxygen uptake efficiency slope in normal healthy subjects. J Sports Med Phys Fitness 1999;39:202– 6. 14. Marantz PR, Tobin JN, Wassertheil-Smoller S, et al. The relationship between left ventricular systolic function and congestive heart failure diagnosed by clinical criteria. Circulation 1988;77:607–12. 15. Wasserman K, Beaver WL, Whipp BJ. Gas exchange theory and the lactic acidosis (anaerobic) treshold. Circulation 1990;81(Suppl II):II-14 –30. 16. Tabet JY, Thabut G, Hainaut P, et al. Critical appraisal of the prognostic value of the VE/VCO2 slope in chronic heart failure. Eur J Heart Fail 2003;2(Suppl):67. 17. Gitt AK, Wasserman K, Kilkowski C, et al. Exercise anaerobic threshold and ventilatory efficiency identify heart failure patients for high risk of early death. Circulation 2002;106:3079 – 84. 18. Cohen-Solal A, Benessiano J, Himbert D, et al. Ventilatory threshold during exercise in patients with mild to moderate chronic heart failure: determination, relation with lactate threshold and reproducibility. Int J Cardiol 1991;30:321–27. 19. Elborn JS, Stanford CF, Nicholls DP. Reproducibility of cardiopulmonary parameters during exercise in in patients with chronic cardiac failure. Eur Heart J 1990;11:75– 81. 20. Simonton CA, Higginbotham MB, Cobb FR. The ventilatory threshold: quantitative analysis of reproducibility and relation to arterial lactate concentration in normal subjects and in patients with chronic congestive heart failure. Am J Cardiol 1988;62:100 –7. 21. Katz SD, Berkowitz R, Le Jemtel TH. Anaerobic threshold detection in patients with congestive heart failure. Am J Cardiol 1992;69: 1565–9. 22. Miyagi K, Asanoi H, Ishizaka S, et al. Limited value of anaerobic threshold for assessing functional capacity in patients with heart failure. Clin Cardiol 1993;16:133–7.
