Impact of High Respiratory Exchange Ratio During Submaximal Exercise on Adverse Clinical Outcome in Heart Failure

Abstract

Background: Oxygen uptake (V̇O₂) at peak workload and anaerobic threshold (AT) workload are often used for grading heart failure (HF) severity and predicting all-cause mortality. The clinical relevance of respiratory exchange ratio (RER) during exercise, however, is unknown.

Methods and Results: We retrospectively studied 295 HF patients (57±15 years, NYHA class I–III) who underwent cardiopulmonary exercise testing. RER was measured at rest; at AT workload; and at peak workload. Peak V̇O₂ had an inverse correlation with RER at AT workload (r=−0.256), but not at rest (r=−0.084) or at peak workload (r=0.090). Using median RER at AT workload, we divided the patients into high RER (≥0.97) and low RER (<0.97) groups. Patients with high RER at AT workload were characterized by older age, lower body mass index, anemia, and advanced NYHA class. After propensity score matching, peak V̇O₂ tended to be lower in the high-RER than in the low-RER group (14.9±4.5 vs. 16.1±5.0 mL/kg/min, P=0.06). On Kaplan-Meier analysis, HF patients with a high RER at AT workload had significantly worse clinical outcomes, including all-cause mortality and rate of readmission due to HF worsening over 3 years (29% vs. 15%, P=0.01).

Conclusions: High RER during submaximal exercise, particularly at AT workload, is associated with poor clinical outcome in HF patients.

Cardiopulmonary exercise testing (CPET) is increasingly being used not only to evaluate exercise capacity but also to predict outcome in patients with heart failure (HF).^1–3 Of the various CPET parameters, peak oxygen uptake (V̇O₂) and ventilatory efficiency (i.e., the minute ventilation [V̇E]/production of carbon dioxide [V̇CO₂] slope or lowest V̇E/V̇CO₂ ratio) are each recognized as a strong predictor of all-cause mortality in HF patients.^4–7 Peak V̇O₂, however, depends in part on the patient’s motivation, and ventilatory efficiency tends to be influenced by primary respiratory disease or pulmonary hypertension, which may limit the clinical interpretation of these parameters in HF patients who cannot perform maximal exercise or who have multiple comorbidities.^8–10 Accordingly, it is clinically relevant to identify another parameter of CPET that can predict the severity and outcome of HF.

Respiratory exchange ratio (RER; the same as respiratory quotient [RQ] in a steady state) is defined as the ratio of V̇CO₂ to V̇O₂, and ranges from 0.7 to 1.0, depending mostly on the whole-body substrate metabolism.^11,12 Because the skeletal muscle is the largest energy source for maintaining physical activity, skeletal muscle energy metabolism is a major determinant of RER. During low- to moderate-intensity exercise, fatty acids are the major fuel for energy production, leading to an RER of nearly 0.7 in the skeletal muscle.¹³ As the intensity increases, the energy utilization shifts from fatty acids to glucose, which results in a higher RER.^14,15

In contrast, during strenuous exercise, RER usually exceeds 1.0 as a result of hyperventilation or increased buffering of blood lactic acid derived from skeletal muscle. Thus, RER does not reflect whole-body substrate metabolism but rather hyperventilation and blood lactate levels in this condition. In HF patients, overactivation of intramuscular ergoreceptors can induce excessive ventilatory response (i.e., hyperventilation),¹⁶ thereby yielding a reduced ventilatory efficiency with a higher RER even during submaximal exercise. It is well known that peak RER >1.1 is an objective criterion of maximal effort,¹⁷ but the clinical relevance of RER during submaximal exercise remains to be elucidated.

The aim of this study was to clarify the characteristics of HF patients who have a higher RER during submaximal exercise and to investigate whether higher RER during submaximal exercise is associated with exercise capacity and long-term clinical outcome in patients with HF.

Methods

This was a single-center and retrospective cohort study. We searched the database of the Exercise Testing Laboratory at Hokkaido University Hospital with respect to all HF patients studied between January 2009 and March 2016 who underwent CPET. All participants had a history of ≥1 hospital admission due to HF worsening diagnosed according to the American College of Cardiology Foundation/American Heart Association Task Force on Practice guidelines¹⁸ by 2 or more cardiologists. Patients with New York Heart Association (NYHA) functional class IV requiring inotropic agents or mechanical circulatory support, or difficulty in determination of the anaerobic threshold (AT) because of oscillatory ventilation or other reasons were excluded. Patients who were unable to undergo CPET due to peripheral artery disease, respiratory disease, stroke, or orthopedic disease were also excluded. This study was approved by the Medical Ethics Committee of Hokkaido University Hospital in accordance with the ethics principles described in the Declaration of Helsinki (2013 revised version).

CPET

All HF patients performed symptom-limited CPET using an upright electromechanical bicycle ergometer (Aerobike 75XLII; Combi Wellness, Tokyo, Japan) with a ramp protocol as described previously.^19,20 Briefly, CPET started with a 3-min rest on the ergometer followed by a 3-min warm-up, and full exercise with 5–20-W increments every 1 min. The exercise protocol was set to achieve peak exercise in 10 min. All CPET was completed in accordance with guidelines of the American College of Cardiology and the American Heart Association.¹⁷

Each patient’s V̇O₂, V̇CO₂, and V̇E were acquired using a breath-by-breath method and averaged at 10-s intervals using an expired gas analyzer (Aeromonitor AE-300S; Minato Medical Science, Osaka, Japan). Peak V̇O₂ was defined as V̇O₂ attained at maximum exercise, and AT was determined using the V-slope method²¹ by at least 2 CPET experts. The ratio of V̇E to V̇CO₂ was calculated each minute during exercise; the lowest minute sample was taken as the lowest V̇E/V̇CO₂ ratio, as previously descrived.⁷ RER was measured at 3 different time points: at rest; at AT workload; and at peak workload during the incremental exercise. Respiratory rate (RR) and end-tidal CO₂ (ETCO₂) were also evaluated.

Other Clinical Variables and Outcomes

We reviewed all patients’ medical records to evaluate demographic data including age, gender, body mass index (BMI), causes of HF, NYHA functional class, medication, and comorbidities. Echocardiographic parameters and laboratory data were acquired ≤30 days before or after CPET. Left ventricular end-diastolic diameter (LVEDD) and LV ejection fraction (LVEF) were determined on echocardiography. All patients underwent measurement of estimated glomerular filtration rate (eGFR), hemoglobin, and plasma brain natriuretic peptide (BNP). eGFR was calculated from serum creatinine and age using the Japanese equation:²² eGFR=194×(serum creatinine, mg/dL)^−1.094×(age, years)^−0.287×(0.739 if female). Plasma BNP concentration was measured using chemiluminescence immunoassay in an automated analyzer (Architect; Abbott Japan, Tokyo, Japan). Patients were followed up for composite endpoints, including all-cause death and readmission due to worsening HF via medical chart review for up to 3 years or to the first event, whichever came first.

Statistical Analysis

We used unpaired Student’s t-test or Mann-Whitney U-test to compare continuous variables, summarized as mean±SD or median (IQR) due to non-normal distribution, as appropriate. Categorical variables are presented as numbers or percentages, and were compared on chi-squared test. The association between peak V̇O₂ and RER was examined using Spearman’s correlation coefficients. Based on median RER at AT workload, we divided the patients into 2 groups: high RER and low RER.

To adjust for significant differences in baseline characteristics between the high-RER and low-RER groups, we conducted propensity score matching by fitting a logistic regression model based on the following variables: age, gender, and NYHA functional class. The discrimination capability of this model was evaluated using c-statistics (0.678). The propensity score matching was performed in a 1-to-1 manner with a caliper width of 0.25, and the 2 matched groups were compared for clinical characteristics and outcomes. We conducted Kaplan-Meier analysis with log-rank test to assess the rates of all-cause death and readmission due to worsening HF. To ensure the adjustment of variables that significantly differ between 2 groups, we also performed multivariate Cox proportional hazard analysis to determine significant predictors of all cardiac death and HF hospitalization in all 295 patients. Variables that differed significantly at P<0.05 between high- and low-AT RER groups, and which have been reported as independent predictor of adverse clinical outcome, including age, BMI, NYHA functional class III, LVEF, β-blocker, hemoglobin, peak V̇O₂, lowest V̇E/V̇CO₂, and high RER at AT workload, were entered into the multivariate Cox proportional hazard analysis. All analyses were performed using JMP 13.1.0 (SAS Institute, Cary, NC, USA). P<0.05 was considered significant.

Results

Patient Characteristics

Of the 350 patients who underwent CPET, 295 patients were analyzed, after exclusion of 55 patients with non-defined AT point (Figure 1). The characteristics of the 295 patients before propensity score matching are listed in Table 1. Mean age was 57±15 years; 77% were men; and 18% had ischemic heart disease. NYHA functional class III was noted in 23%, and hypertension was identified as the most common comorbidity (37%), followed by diabetes (27%). Mean LVEF was 36±12%. Most of the patients were taking β-blockers (89%) and angiotensin-converting enzyme inhibitors or angiotensin receptor blockers (89%). Median RER at rest, at AT workload, and at peak workload was 0.92 (IQR, 0.86–0.98), 0.97 (IQR, 0.92–1.02), and 1.23 (IQR, 1.17–1.33), respectively.

Figure 1.

Patient selection. AT, anaerobic threshold; NYHA, New York Heart Association; RER, respiratory exchange ratio.

Table 1. Heart Failure Patient Characteristics

	All patients	RER at AT workload		P-value
		<0.97 (Low RER)	≥0.97 (High RER)
n	295	148	147
Demographic factors
Age (years)	57±15	55±15	60±15	0.006
Male	227 (77)	114 (78)	112 (76)	0.784
BMI (kg/m2)	23.1±4.2	23.8±4.2	22.5±4.1	0.007
Cause of heart failure
Ischemic heart disease	54 (18)	21 (14)	33 (22)	0.072
Valvular heart disease	40 (14)	20 (14)	20 (14)	<0.999
Hypertensive heart disease	32 (11)	15 (11)	16 (11)	<0.999
Dilated cardiomyopathy	91 (31)	52 (35)	39 (27)	0.130
NYHA functional class (I/II/III)	41/185/69	23/101/24	18/84/45	0.014
Medical history
Hypertension	108 (37)	51 (35)	56 (38)	0.629
Diabetes mellitus	78 (27)	34 (23)	44 (30)	0.188
Echocardiographic parameters
LVEDD (mm)	61.2±10.7	61.6±11.4	61.5±10.6	0.929
LVEF (%)	36±12	37±13	35±12	0.413
Medication
β-blocker	259 (89)	134 (92)	123 (85)	0.043
ACEI or ARB	259 (89)	131 (90)	126 (87)	0.360
Laboratory test
eGFR (mL/min/1.73 m2)	62.8±26.3	63.1±21.6	62.3±30.1	0.821
Hemoglobin (g/dL)	13.2±1.8	13.6±1.6	12.9±1.9	0.001
Plasma BNP (pg/mL)	206 (92–416)	212 (92–417)	192 (93–414)	0.541
CPET
Peak V̇O2 (mL/kg/min)	15.6±4.8	16.6±5.0	14.5±4.4	<0.001
Peak load (W)	83±35	90±35	76±33	<0.001
Lowest V̇E/V̇CO2 ratio	35.5±7.1	34.4±6.8	36.6±7.3	0.010
V̇O2 at AT workload (mL/kg/min)	9.9±2.5	10.1±2.4	9.7±2.7	0.180
V̇CO2 at AT workload (mL/kg/min)	9.6±2.5	9.2±2.2	10.1±2.7	0.002
V̇E at AT workload (L/min)	22.2±5.6	21.5±4.9	22.8±6.1	0.036
RR at AT workload (breaths/min)	23±5	22±5	24±5	<0.001
ETCO2 at AT workload (%)	5.3±0.8	5.4±0.7	5.2±0.9	0.018
AT load (W)	41±19	43±16	39±21	0.037
RER at rest	0.92 (0.86–0.98)	0.88 (0.84–0.93)	0.96 (0.91–1.01)	<0.001
RER at AT workload	0.97 (0.92–1.02)	0.92 (0.87–0.95)	1.02 (0.99–1.07)	<0.001
RER at peak workload	1.23 (1.17–1.33)	1.20 (1.13–1.26)	1.28 (1.20–1.36)	<0.001

Data given as mean±SD, n (%) or median (IQR). ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; AT, anaerobic threshold; BMI, body mass index; BNP, brain natriuretic peptide; CPET, cardiopulmonary exercise testing; eGFR, estimated glomerular filtration rate; ETCO₂, end-tidal carbon dioxide; LVEDD, left ventricular end-diastolic diameter; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association; RER, respiratory exchange ratio; RR, respiratory rate; V̇CO₂, carbon dioxide production; V̇E, minute ventilation; V̇O₂, oxygen uptake.

When the patients were divided into low-RER (<0.97; n=148) and high-RER (≥0.97; n=147) groups according to median RER at AT workload (0.97), we observed 3 significant between-group differences: older age, lower BMI, and higher percentage of NYHA class III were found in the high-RER group compared with the low-RER group. Moreover, the high-RER patients had a significantly lower prevalence of β-blocker use and a higher prevalence of anemia compared with the low-RER patients, whereas LVEF and renal function were not significantly different between the groups.

Of the CPET variables, peak V̇O₂ and V̇O₂ at AT workload were significantly lower and the lowest V̇E/V̇CO₂ ratio was significantly higher in the high-RER group compared with the low-RER group. Interestingly, the high-RER patients had significantly higher V̇CO₂, V̇E, and RR, and significantly lower ETCO₂ at AT workload. Consistent with the high RER at AT workload, RER at rest and at peak workload were also higher in the high-RER group than in the low-RER group. In contrast, there was an inverse correlation between peak V̇O₂ and RER at AT workload (r=−0.256, P<0.001), but not at rest (r=−0.084, P=0.153) or at peak workload (r=0.090, P=0.127; Figure 2).

Figure 2.

(A) Respiratory exchange ratio (RER) at rest, (B) at anaerobic threshold (AT) workload, and (C) at peak workload vs. peak oxygen uptake (V̇O₂).

High RER at AT Workload Predicts Adverse Clinical HF Outcome

To adjust for baseline characteristics that can affect clinical outcomes in HF, we performed propensity score matching with age, gender, and NYHA functional class. After the matching, all baseline characteristics except for V̇CO₂ and V̇E at AT workload, and RER at rest, at AT workload, and at peak workload were well balanced with no significant differences between the 2 groups (Table 2). Nevertheless, peak V̇O₂ tended to be lower, and RR at AT workload and the incidence of anemia tended to be higher in the high-RER group than in the low-RER group. A representative graphs of the V-slope method to determine the AT point and time trend of V̇E/V̇O₂, V̇E/V̇CO₂, RER and workload are presented in Figure 3. During the median follow-up of 1,030 days (IQR, 380–1,095 days), the combined clinical events occurred in 47 patients (22%), consisting of 4 all-cause deaths and 43 readmissions due to worsening HF (20%). On Kaplan-Meier analysis, the high-RER group had a higher risk for combined events (all-cause death and HF-related readmission) or readmission due to worsening HF compared with the low-RER group (29% vs. 15%, 28% vs. 14%, respectively, P<0.05; Figure 4). In contrast, when the patients were divided based on median RER at rest (0.92) and at peak workload (1.23), there were no significant difference in clinical events between the high-RER and low-RER groups (Figure S1). This was also confirmed on multivariate Cox proportional hazard analysis, demonstrating that high RER at AT workload remained an independent predictor of adverse outcome, together with age, LVEF, and lowest V̇E/V̇CO₂ ratio (Table S1).

Table 2. Patient Characteristics After Propensity Score Matching

	RER at AT workload		P-value
	<0.97 (Low RER)	≥0.97 (High RER)
n	106	106
Demographic factors
Age (years)	59±15	58±15	0.838
Male	82 (77)	78 (74)	0.632
BMI (kg/m2)	23.2±4.2	22.8±4.3	0.495
Cause of heart failure
Ischemic heart disease	16 (15)	26 (25)	0.120
Valvular heart disease	19 (18)	15 (14)	0.575
Hypertensive heart disease	12 (11)	12 (11)	>0.999
Dilated cardiomyopathy	34 (32)	25 (24)	0.220
NYHA functional class (I/II/III)	14/71/21	13/73/20	0.956
Medical history
Hypertension	42 (40)	43 (41)	>0.999
Diabetes mellitus	28 (26)	32 (30)	0.648
Echocardiographic parameters
LVEDD (mm)	61.6±11.9	61.6±10.1	0.954
LVEF (%)	37±13	35±12	0.472
Medication
β-blocker	97 (92)	94 (89)	0.482
ACEI or ARB	97 (92)	95 (90)	0.632
Laboratory test
eGFR (mL/min/1.73 m2)	61.5±19.2	61.7±31.7	0.972
Hemoglobin (g/dL)	13.4±1.7	13.0±1.8	0.097
Plasma BNP (pg/mL)	208 (100–440)	182 (92–356)	0.220
CPET
Peak V̇O2 (mL/kg/min)	16.1±5.0	14.9±4.5	0.063
Peak load (W)	83±33	80±35	0.238
Lowest V̇E/V̇CO2 ratio	35.6±7.1	35.7±6.3	0.870
V̇O2 at AT workload (mL/kg/min)	10.1±2.3	9.9±2.8	0.580
V̇CO2 at AT workload (mL/kg/min)	9.2±2.2	10.2±2.9	0.004
V̇E at AT workload (L/min)	21.2±4.4	22.8±6.3	0.033
RR at AT workload (breaths/min)	22±5	23±5	0.061
ETCO2 at AT workload (%)	5.3±0.7	5.2±0.7	0.469
AT load (W)	41±16	41±23	0.876
RER at rest	0.89 (0.85–0.93)	0.95 (0.91–1.01)	<0.001
RER at AT workload	0.92 (0.88–0.95)	1.02 (0.99–1.07)	<0.001
RER at peak workload	1.18 (1.12–1.25)	1.29 (1.21–1.37)	<0.001

Data given as mean±SD, n (%) or median (IQR). Abbreviations as in Table 1.

Figure 3.

Representative graphs of (A,C) the V-slope method and (B,D) time trend of minute ventilation/oxygen uptake (V̇E/V̇O₂), minute ventilation/carbon dioxide production (V̇E/V̇CO₂), respiratory exchange ratio (RER), and workload for (A,B) a low-RER patient and (C,D) a high-RER patient. AT, anaerobic threshold; RC, respiratory compensation.

Figure 4.

Kaplan-Meier survival curves for (A) composite events composed of all-cause death and hospital readmission due to worsening heart failure (HF) and (B) event of hospital readmission due to worsening HF in HF patients with low or high respiratory exchange ratio (RER) divided according to median RER (0.97) at anaerobic threshold (AT) workload.

Discussion

The present study has shown, in patients with HF, that RER during submaximum exercise, particularly at AT workload, is inversely correlated with peak V̇O₂, and that HF patients with higher RER are characterized by older age, lower BMI, anemia, and advanced NYHA functional class. After the adjustment for age, gender, and NYHA functional class on propensity score matching, high RER at AT workload was an independent predictor of adverse clinical outcome including all-cause mortality and readmission due to worsening HF over 3 years. To our knowledge, this is the first report to show the clinical relevance of RER during submaximal exercise in HF patients.

During aerobic exercise (i.e., exercise at or below the AT workload) with steady-state respiratory gas exchange, RER is the same as RQ, reflecting whole-body metabolism and substrate utilization.¹² In the present study, however, RER was measured during incremental exercise with a ramp protocol when hemodynamics, autonomic and hormonal balance are dynamically changed. Accordingly, in that situation, RER is not necessarily equal to RQ, and other factors such as hyperventilation may influence RER even during aerobic exercise.

In the present study, the RER of HF patients at AT workload was inversely correlated with peak V̇O₂, and high RER was an independent predictor of all-cause death and hospital readmission due to HF worsening. Because the V̇O₂ at AT workload was similar between the low-RER and high-RER groups after propensity score matching, the increased RER at this workload may be attributable to an overproduction of CO₂. We previously demonstrated that a reduction of muscle pH is initiated even at moderate-intensity workload in patients with HF,²³ suggesting an effect of exaggerated ventilatory response to exercise through the overactivation of ergoreceptors and/or fiber-type switches in the working muscles.^24,25 In the present study, we observed increased RR and decreased ETCO₂ at AT workload in the high-RER group. This is indicative of the presence of hyperventilation during submaximal exercise, which may contribute to the increased RER at AT workload. Indeed, reduction of arterial partial pressure of CO₂ (PaCO₂), a feature of hyperventilation, during submaximal exercise has been reported in symptomatic HF patients.²⁶

Given that the ventilatory efficiency retains a high level of prognostic discrimination when measured at submaximal load,²⁷ a high RER during submaximal exercise may also be an important marker for muscular acidosis and a predictor of HF outcome. In terms of clinical applications, resting RER is unstable in a short period and thus requires 24-h measurement for assessment of the whole-body metabolic rate.^28,29 In addition, we observed that resting RER was not associated with adverse events over a period of 3 years, suggesting that its predictive value is relatively low. In contrast, RER at AT workload is reproducible, and is more closely linked to the peak V̇O₂ than resting RER, and is thus associated with clinical outcome in HF patients.

Study Limitations

There are some study limitations that should be acknowledged. First, we did not directly measure the whole body or skeletal muscle substrate metabolism; nor was a tissue analysis of skeletal muscle conducted; therefore, RER during exercise with a ramp protocol should be carefully interpreted. Second, it remains uncertain whether RER at AT workload will improve prognostic ability when added to well-established indicators such as peak V̇O₂ and ventilatory efficiency. Finally, this was a retrospective study. A prospective study with a larger sample size is needed.

Conclusions

High RER during submaximal exercise, particularly at AT workload, predicts adverse clinical outcome in patients with HF. RER during submaximal exercise could be a novel prognostic marker for HF patients and could be advantageous for HF patients who cannot achieve maximal exercise due to severely reduced physical activity or comorbidity such as orthopedic disease.

Acknowledgments

This study was partly supported by a grant from the Center of Innovation Program from Japan Science and Technology Agency.

Disclosures

The authors declare no conflicts of interest.

Supplementary Files

Supplementary File 1

Figure S1. Kaplan-Meier survival curves for (A,C) composite events composed of all-cause death and hospital readmission due to worsening heart failure (HF), and (B,D) event of hospital readmission due to worsening HF in HF patients with low or high respiratory exchange ratio (RER) according to median RER (A,B) at rest (0.92) and (C,D) at peak workload (1.23).

Table S1. Multivariate indicators of all-cause death and rehospitalization due to worsening HF

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-18-0103

References

• 1.   Arena R, Myers J, Abella J, Pinkstaff S, Brubaker P, Kitzman DW, et al. Cardiopulmonary exercise testing is equally prognostic in young, middle-aged and older individuals diagnosed with heart failure. Int J Cardiol 2011; 151: 278–283.
• 2.   Goto Y. Exercise capacity: Just a powerful prognostic predictor, or a potential therapeutic target in patients with chronic heart failure? Circ J 2015; 79: 2547–2548.
• 3.   Keteyian SJ, Patel M, Kraus WE, Brawner CA, McConnell TR, Pina IL, et al. Variables measured during cardiopulmonary exercise testing as predictors of mortality in chronic systolic heart failure. J Am Coll Cardiol 2016; 67: 780–789.
• 4.   Arena R, Myers J, Aslam SS, Varughese EB, Peberdy MA. Peak VO2 and VE/VCO2 slope in patients with heart failure: A prognostic comparison. Am Heart J 2004; 147: 354–360.
• 5.   Magri D, Re F, Limongelli G, Agostoni P, Zachara E, Correale M, et al. Heart failure progression in hypertrophic cardiomyopathy: Possible insights from cardiopulmonary exercise testing. Circ J 2016; 80: 2204–2211.
• 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: 778–786.
• 7.   Myers J, Arena R, Oliveira RB, Bensimhon D, Hsu L, Chase P, et al. The lowest VE/VCO2 ratio during exercise as a predictor of outcomes in patients with heart failure. J Card Fail 2009; 15: 756–762.
• 8.   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.
• 9.   Nakanishi M, Takaki H, Kumasaka R, Arakawa T, Noguchi T, Sugimachi M, et al. Targeting of high peak respiratory exchange ratio is safe and enhances the prognostic power of peak oxygen uptake for heart failure patients. Circ J 2014; 78: 2268–2275.
• 10.   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 pressure. Eur Heart J 2005; 26: 472–480.
• 11.   Koike A, Wasserman K, McKenzie DK, Zanconato S, Weiler-Ravell D. Evidence that diffusion limitation determines oxygen uptake kinetics during exercise in humans. J Clin Invest 1990; 86: 1698–1706.
• 12.   Wasserman K, Beaver WL, Whipp BJ. Gas exchange theory and the lactic acidosis (anaerobic) threshold. Circulation 1990; 81(1 Suppl): II14–II30.
• 13.   Ruby BC, Robergs RA. Gender differences in substrate utilisation during exercise. Sports Med 1994; 17: 393–410.
• 14.   Clark AL, Poole-Wilson PA, Coats AJ. Exercise limitation in chronic heart failure: Central role of the periphery. J Am Coll Cardiol 1996; 28: 1092–1102.
• 15.   Okita K, Yonezawa K, Nishijima H, Hanada A, Ohtsubo M, Kohya T, et al. Skeletal muscle metabolism limits exercise capacity in patients with chronic heart failure. Circulation 1998; 98: 1886–1891.
• 16.   Piepoli M, Clark AL, Volterrani M, Adamopoulos S, Sleight P, Coats AJ. Contribution of muscle afferents to the hemodynamic, autonomic, and ventilatory responses to exercise in patients with chronic heart failure: Effects of physical training. Circulation 1996; 93: 940–952.
• 17.   Balady GJ, Arena R, Sietsema K, Myers J, Coke L, Fletcher GF, et al. Clinician’s guide to cardiopulmonary exercise testing in adults: A scientific statement from the American Heart Association. Circulation 2010; 122: 191–225.
• 18.   Yancy CW, Jessup M, Bozkurt B, Butler J, Casey DE Jr, Drazner MH, et al. 2013 ACCF/AHA guideline for the management of heart failure: A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2013; 62: e147–e239.
• 19.   Yokota T, Kinugawa S, Yamato M, Hirabayashi K, Suga T, Takada S, et al. Systemic oxidative stress is associated with lower aerobic capacity and impaired skeletal muscle energy metabolism in patients with metabolic syndrome. Diabetes Care 2013; 36: 1341–1346.
• 20.   Fukushima A, Kinugawa S, Homma T, Masaki Y, Furihata T, Yokota T, et al. Serum brain-derived neurotropic factor level predicts adverse clinical outcomes in patients with heart failure. J Card Fail 2015; 21: 300–306.
• 21.   Beaver WL, Wasserman K, Whipp BJ. A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol (1985) 1986; 60: 2020–2027.
• 22.   Matsuo S, Imai E, Horio M, Yasuda Y, Tomita K, Nitta K, et al. Revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis 2009; 53: 982–992.
• 23.   Okita K, Yonezawa K, Nishijima H, Hanada A, Nagai T, Murakami T, et al. Muscle high-energy metabolites and metabolic capacity in patients with heart failure. Med Sci Sports Exerc 2001; 33: 442–448.
• 24.   Scott AC, Wensel R, Davos CH, Georgiadou P, Kemp M, Hooper J, et al. Skeletal muscle reflex in heart failure patients: Role of hydrogen. Circulation 2003; 107: 300–306.
• 25.   Takada S, Masaki Y, Kinugawa S, Matsumoto J, Furihata T, Mizushima W, et al. Dipeptidyl peptidase-4 inhibitor improved exercise capacity and mitochondrial biogenesis in mice with heart failure via activation of glucagon-like peptide-1 receptor signalling. Cardiovasc Res 2016; 111: 338–347.
• 26.   Woods PR, Olson TP, Frantz RP, Johnson BD. Causes of breathing inefficiency during exercise in heart failure. J Card Fail 2010; 16: 835–842.
• 27.   Koike A, Itoh H, Kato M, Sawada H, Aizawa T, Fu LT, et al. Prognostic power of ventilatory responses during submaximal exercise in patients with chronic heart disease. Chest 2002; 121: 1581–1588.
• 28.   Toth MJ, Gottlieb SS, Fisher ML, Poehlman ET. Daily energy requirements in heart failure patients. Metabolism 1997; 46: 1294–1298.
• 29.   Ferrannini E. The theoretical bases of indirect calorimetry: A review. Metabolism 1988; 37: 287–301.