Exhaled Acetone Concentration Is Related to Hemodynamic Severity in Patients With Non-Ischemic Chronic Heart Failure

Abstract

Background: We hypothesized that exhaled acetone concentration (EAC), reflecting altered blood ketone body metabolism and increased acetone exhaust because of pulmonary congestion in heart failure (HF), would correlate with hemodynamic parameters in patients with non-ischemic chronic HF.

Methods and Results: We prospectively enrolled 102 non-ischemic HF patients with New York Heart Association (NYHA) class I–III. Exhaled breath was collected after an overnight fast. Echocardiography and cardiac catheterization were performed in all patients. We also enrolled 17 control patients without HF. EAC in the HF patients was significantly higher than that in the control patients (median EAC; 0.53 vs. 0.38 ppm, P=0.012). EAC positively correlated with blood total ketone bodies (r=0.454, P<0.001), NYHA class (r=0.489, P<0.001), and plasma B-type natriuretic peptide (r=0.316, P=0.001). Right heart catheterization revealed that EAC significantly correlated with pulmonary capillary wedge pressure (PCWP, r=0.377, P<0.001). Receiver-operating characteristic analysis revealed that EAC >1.05 ppm was associated with PCWP ≥18 mmHg (area under the curve [AUC] 0.726, sensitivity 50%, specificity 89%). EAC was shown to be a comparable diagnostic biomarker for HF to BNP (AUC 0.760, sensitivity 80%, specificity 70%).

Conclusions: EAC may be a novel noninvasive biomarker that correlates hemodynamic severity in non-ischemic chronic HF. (Circ J 2016; 80: 1178–1186)

Heart failure (HF) is a global problem with a prevalence of more than 23 million worldwide.¹ Survival estimates of HF are poor, being 50% at 5 years and 10% at 10 years after diagnosis.^2,3 For the management of HF, several blood biomarkers have been proposed.⁴ As a representative example, plasma B-type natriuretic peptide (BNP) or NT-proBNP is the most widely used blood biomarker of HF and is recommended in guidelines for the diagnosis and management of HF.⁵ However, either of those biomarkers requires blood collection, which can be accompanied by pain and may cause several complications. The development of a less invasive means to detect and monitor HF as an alternative or additional method to blood markers of HF in clinical practice is crucial.

Impaired tissue metabolism is one of the key pathophysiologic hallmarks related to HF.⁶ The heart utilizes glucose, lactate, pyruvate, fatty acids, amino acids, and ketone bodies as energy sources.⁷ Cardiac energy metabolism mainly depends on fatty acids and glucose, and ketone bodies are normally a minor substrate for the intact myocardium.⁸ In HF, however, energy metabolism is altered; the failing myocardium exhibits impaired fatty acid and glucose metabolism, resulting in increased transportation of ketone bodies from the liver.^9,10 Additionally, ketone utilization by skeletal muscle is impaired in patients with HF as a clinical feature of a debilitating systemic condition.¹⁰ For these reasons, blood concentrations of ketone bodies are elevated in HF.

Acetone, a volatile component of ketone bodies produced by decarboxylation of acetoacetate, is easily and accurately measured in the exhaled breath. Exhaled acetone concentration (EAC) has been reported to reflect the concentration of blood ketone bodies and to be increased in patients with congestive HF.^11–14 In addition, alveolar-capillary barrier dysfunction and increased intravascular pressures because of pulmonary congestion may cause increased release of some exhaled breath substances.^15,16

To date, the relationship between EAC and hemodynamic parameters is unknown. We hypothesized that EAC could be a noninvasive biomarker related to hemodynamic severity as measured by cardiac catheterization in patients with chronic HF.

Methods

Study Population

The study group comprised 312 consecutive patients who underwent elective cardiac catheterization with the diagnosis of HF, judged by at least 2 independent cardiologists with 2 major criteria or 1 major criterion in conjunction with 2 minor Framingham criteria, from December 2014 to September 2015.¹⁷ Of them, after excluding patients who were confirmed to have ischemic heart disease by coronary angiography, 242 clinically stable non-ischemic HF patients with New York Heart Association (NYHA) class I–III were prospectively included in this study. Patients with factors that may influence ketone body metabolism were excluded: diabetes mellitus, hypercholesterolemia, chronic hepatitis, end-stage renal failure, pregnancy and postpartum period, and history of long-term administration of corticosteroids, tricyclic antidepressants, tetracyclic antidepressants, or selective serotonin reuptake inhibitors. Patients who were suspected of having infectious lung diseases were also excluded because of the possibility of infection spread during exhaled gas collection. Patients who were unable to breathe as instructed were also excluded because of difficulty in collecting a proper exhaled breath. To avoid possible alteration of ketone body metabolism related to myocardial ischemia, patients with ischemic heart disease were also excluded. Finally, 102 patients were enrolled and were assigned to the HF group. As the control group 17 patients without HF, who were admitted to the same hospital during this study period, were also enrolled. Detailed information on the study population in the HF group and control group are shown in Figure 1. Baseline data of sex, age, body mass index, NYHA class, etiology of HF, past and preexisting medical history, physical findings, and current medication were collected at the time of enrollment in this study. The study protocol was approved by the institutional ethics committee of the National Cerebral and Cardiovascular Center. Written informed consent was given by all study patients.

Figure 1.

Study population for evaluation of exhaled acetone concentration as a marker of heart failure.

Blood Tests and Breath Analysis

All patients were given the regular daily hospital diet with an average total energy count of 1,600 kcal/day (245 g carbohydrates, 60 g protein, 40 g fat). Blood sampling and breath analysis were performed in the early morning after at least a 12-h fast during the hospitalization. Blood was analyzed for hemoglobin, creatinine, sodium, total bilirubin, albumin, BNP, glucose, hemoglobin A1c, total cholesterol, and total ketone bodies (TKB). Breath was collected and maintained in a breath-sampling bag (Collection Bag, Laboratory for Expiration Biochemistry, Nourishment Metabolism Co, Ltd, Nara, Japan). Within 48 h the collected breath was transferred to a gas-tight glass syringe, and 2.5 ml of the total was injected into a gas analysis device (Biogas Acetone Analyzer, BAS-2000, Mitleben R&D Assoc, Osaka, Japan) to measure breath acetone concentration. In this system, acetone gas is stable in the collecting bag over 48 h. Breath concentration of acetone was calculated by subtracting the acetone concentration of ambient air around the patient using an identical sampling bag for the collected breath.

Echocardiography

Echocardiographic examination of all patients was performed by at least 2 experienced technicians in a blinded manner. Left ventricular end-diastolic and end-systolic diameters were obtained from the parasternal long-axis view. Left ventricular ejection fraction (LVEF) and left atrial volume index were measured by modified biplane Simpson method according to the recommendations of the American Society of Echocardiography.¹⁸

Cardiac Catheterization

Coronary angiography was performed via the radial or femoral artery to exclude significant coronary stenosis. Right heart catheterization was performed through femoral or cervical vein access using a 7Fr thermodilution catheter (Goodman Co, Nagoya, Japan) in all patients in the HF group. Mean pulmonary capillary wedge pressure (PCWP), mean pulmonary artery pressure (PA), mean right atrial pressure (RA), and cardiac output (CO) were measured. CO was determined by the direct Fick principle.

Statistical Analysis

Data were analyzed using the Statistical Package for Social Sciences version 22.0 (SPSS Inc, Chicago, IL, USA). All quantitative data are expressed as mean±SD. The statistical significance of differences was analyzed using Student’s t-test for parametric continuous variables, and the Mann-Whitney U-test for nonparametric continuous variables. Categorical variables were compared using Chi-square test or Fisher’s exact test. Correlations were analyzed using Spearman’s correlation analysis for variables. The receiver-operating characteristic (ROC) curve was constructed to determine the cut-off values of EAC and BNP. P<0.05 was considered statistically significant.

Results

Baseline Characteristics of the Study Group (Table 1)

The HF group showed higher serum creatinine and plasma BNP levels, left ventricular dimensions, and left atrial volume index, and lower diastolic blood pressure, heart rate, plasma glucose level, serum total cholesterol level, and LVEF. Angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers, β-blockers, mineralocorticoid receptor antagonists, and loop diuretics were more frequently used in the HF group than in the control group.

Table 1. Baseline Characteristics of Study Population for Evaluation of EAC as a Marker of Heart Failure

	Heart failure (n=102)	Control (n=17)	P value
Male	57 (56)	5 (29)	0.044
Age, years	63.6±15.8	61.9±13.9	0.427
BMI, kg/m2	17.6±3.1	18.6±3.4	0.408
NYHA class, I/II/III	10/75/17	–	–
Etiology
Valvular	65 (64)	–	–
NICM	33 (32)	–	–
Other	3 (3)	–	–
Past history
Hypertension	35 (34)	8 (47)	0.313
Smoking history	38 (37)	6 (35)	0.877
COPD	2 (2)	0 (0)	0.562
Physical findings
SBP, mmHg	111±17	116±13	0.119
DBP, mmHg	63±12	70±12	0.024
HR, beats/min	68±12	76±14	0.019
O2 saturation, %	98±1	97±1	0.188
Medications
ACEI/ARB	55 (54)	4 (24)	0.021
β-blockers	44 (43)	2 (12)	0.014
MRA	31 (30)	0 (0)	0.008
Loop diuretics	46 (45)	0 (0)	<0.001
Laboratory findings
Hemoglobin, g/dl	13.3±1.7	13.6±1.3	0.826
CRE, mg/dl	0.9±0.3	0.7±0.3	0.007
Sodium, mEq/L	140±4	141±1	0.580
T-Bil, mg/dl	0.8±0.5	0.7±0.3	0.255
Alb, g/dl	4.2±0.4	4.4±0.3	0.067
BNP, pg/ml, median (IQR)	141 (71–252)	16 (13–35)	<0.001
Glucose, mg/dl	93±16	100±8	0.007
HbA1c, %	5.7±0.4	5.8±0.3	0.703
T-Chol, mg/dl	180±32	207±22	0.001
TKB, μmol/L, median (IQR)	62 (35–106)	–	–
Echocardiography
LVEF, %	50.1±15.3	60.0±2.0	0.023
LVDd, mm	54.8±9.6	46.6±4.1	0.001
LVDs, mm	38.4±10.5	27.8±4.0	<0.001
LAVI, ml/m2	86.6±83.4	33.4±8.5	<0.001
Hemodynamic data
Mean PCWP, mmHg	12.7±6.4	–	–
Mean PA, mmHg	20.1±8.1	–	–
Mean RA, mmHg	5.1±3.8	–	–
CI, L·min−1·m−2	2.77±0.74	–	–

All values are shown as mean±SD, median (IQR), or n (%). ACEI, angiotensin-converting enzyme inhibitor; Alb, albumin; ARB, angiotensin II receptor blocker; BMI, body mass index; BNP, B-type natriuretic peptide; CI, cardiac index; COPD, chronic obstructive pulmonary disease; CRE, creatinine; DBP, diastolic blood pressure; EAC, exhaled acetone concentration; HbA1c, hemoglobin A1c; HR, heart rate; IQR, interquartile range; LAVI, left atrial volume index; LVDd, left ventricular end-diastolic diameter; LVDs, left ventricular end-systolic diameter; LVEF, left ventricular ejection fraction; MRA, mineralocorticoid receptor antagonist; NICM, non-ischemic cardiomyopathy; NYHA, New York Heart Association; PA, pulmonary artery pressure; PCWP, pulmonary capillary wedge pressure; RA, right atrial pressure; SBP, systolic blood pressure; T-Bil, total bilirubin; T-Chol, total cholesterol; TKB, total ketone bodies.

EAC in Patients With and Without HF

Median (interquartile range) EAC in the HF group (n=102) and the control group (n=17) was 0.53 ppm (0.33–0.88 ppm) and 0.38 ppm (0.33–0.46 ppm), respectively. As shown in Figure 2, the HF group had significantly higher EAC compared with the control group (P=0.012).

Figure 2.

Box-and-whisker graph of exhaled acetone concentrations (EAC) in the heart failure group and control group. EAC was significantly increased in the heart failure group, compared with the control group (P=0.012). The bottom and top of each box show the 1st and 3rd quartiles, and the band inside the box is the median. The ends of the whiskers denote the 1.5 interquartile range of the lower and upper quartiles. Colored circles are outliers.

EAC and Blood TKB

EAC positively correlated with blood TKB (r=0.454, P<0.001; Figure 3), indicating that the increase in concentration of breath acetone, a volatile component of ketone bodies, was reflecting increase in blood ketone bodies in HF patients.

Figure 3.

Exhaled acetone concentration and blood total ketone body level were positively correlated in the present study (r=0.454, P<0.001).

Correlation of EAC With HF Severity

Median (interquartile range) EAC in HF patients with NYHA I, II, and III was 0.31 (0.22–0.52), 0.49 (0.31–0.85), and 0.91 (0.61–1.50) ppm, respectively. There was a significant positive relationship of EAC (r=0.489, P<0.001) with NYHA class (Figure 4A). EAC was positively correlated with mean PCWP (r=0.377, P<0.001, Figure 5A). However, EAC had no significant correlation with cardiac index (r=−0.125, P=0.225, Figure 5B).

Figure 4.

Exhaled acetone concentration (EAC) according to New York Heart Association (NYHA) class and B-type natriuretic peptide (BNP). (A) EAC positively correlated with NYHA class (r=0.489, P<0.001). Bottom and top of the box show 1st and 3rd quartiles, and the band inside the box is the median. The ends of the whiskers denote the 1.5 interquartile range of the lower and upper quartiles. Colored circles represent outliers. (B) EAC positively correlated with plasma BNP level (r=0.316, P=0.001).

Figure 5.

Correlation of exhaled acetone concentration (EAC) and B-type natriuretic peptide (BNP) with hemodynamic parameters. (A) EAC significantly correlated with pulmonary capillary wedge pressure (PCWP), but not with cardiac index (B). Plasma BNP level significantly correlated with both PCWP (C) and cardiac index (D).

Comparison of Baseline Characteristics Based on EAC

As a result of the ROC analysis for PCWP ≥18 mmHg, a cut-off EAC value of 1.05 ppm was determined, with the area under the curve (AUC) of 0.726 (95% confidence interval [CI], 0.567–0.886; P=0.019; Figure 6). We divided patients based on EAC of 1.05 ppm: low EAC (minimum to maximum; 0.13–1.00 ppm; n=87) and high EAC (1.09–2.22 ppm; n=15). As shown in Table 2, there were no significant differences between the groups in sex, age, body mass index, HF etiology, past history, and physical findings. However, the high EAC group showed significantly higher NYHA class (P=0.001), serum creatinine level (P=0.004), serum total bilirubin level (P=0.006), plasma BNP level (P=0.001), blood TKB level (P=0.008), and left atrial volume index (P=0.043). On the other hand, no significant difference was observed in LVEF. With regard to hemodynamic parameters, mean PCWP (P<0.001), mean PA (P=0.003), and mean RA (P=0.005) were higher in the high EAC group, whereas the cardiac index was not significantly different between the groups. Use of mineralocorticoid receptor antagonists (P=0.037) and loop diuretics (P=0.018) was higher in the high EAC group. These findings suggest that higher EAC was associated with worse HF symptomatic severity and volume overload, rather than cardiac contractile dysfunction itself.

Figure 6.

Receiver-operating characteristic analysis for pulmonary capillary wedge pressure (PCWP) ≥18 mmHg. The area under the curve (AUC) is 0.726 (95% confidence interval [CI], 0.567–0.886; P=0.019) of exhaled acetone concentration (EAC) for PCWP ≥18 mmHg. EAC threshold of 1.05 ppm maximized the sensitivity and specificity of PCWP ≥18 mmHg diagnosis (sensitivity 50%, specificity 89%). AUC is 0.760 (95% CI, 0.606–0.915; P=0.007) for BNP and PCWP ≥18 mmHg. BNP threshold of 204 pg/ml maximized the sensitivity and specificity of PCWP ≥18 mmHg diagnosis (sensitivity 80% and specificity 70%).

Table 2. Comparison of Baseline Characteristics of Study Patients Based on EAC

	EAC ≥1.05 ppm (High EAC) (n=15)	EAC <1.05 ppm (Low EAC) (n=87)	P value
Male	8 (53)	49 (56)	0.830
Age, years	69.0±9.4	62.7±16.6	0.186
BMI, kg/m2	17.2±3.2	17.7±3.1	0.342
NYHA class, I/II/III	0/8/7	10/67/10	0.001
Etiology
Valvular	11 (73)	54 (62)	0.404
NICM	4 (27)	29 (33)	0.612
Other	0 (0)	3 (3)	0.468
Past history
Hypertension	5 (33)	30 (35)	0.931
Smoking history	6 (40)	32 (37)	0.813
COPD	1 (7)	1 (1)	0.157
Physical findings
SBP, mmHg	111±20	111±16.3	0.467
DBP, mmHg	63±8	63±12	0.788
HR, beats/min	71±20	67±10	0.424
O2 saturation, %	97±1	98±1	0.819
Medications
ACEI/ARB	6 (40)	49 (56)	0.244
β-blocker	7 (47)	37 (43)	0.766
MRA	8 (53)	23 (26)	0.037
Loop diuretic	11 (73)	35 (40)	0.018
Laboratory findings
Hemoglobin, g/dl	13.0±1.6	13.4±1.7	0.438
CRE, mg/dl	1.1±0.4	0.9±0.3	0.004
Sodium, mEq/L	140±3	141±4	0.292
T-Bil, mg/dl	1.4±0.8	0.8±0.4	0.006
Alb, g/dl	4.2±0.5	4.2±0.4	0.913
BNP, pg/ml, median (IQR)	272 (215–478)	121 (60–230)	0.001
Glucose, mg/dl	91±7	93±17	0.760
HbA1c, %	5.8±0.6	5.7±0.4	0.135
T-Chol, mg/dl	175±39	181±30	0.674
TKB, μmol/L, median (IQR)	110 (66–216)	50 (34–101)	0.008
Echocardiography
LVEF, %	44.2±16.5	51.1±14.9	0.090
LVDd, mm	55.2±10.6	54.7±9.4	0.603
LVDs, mm	41.5±9.4	37.9±10.6	0.125
LAVI, ml/m2	112±106	82.0±78.6	0.043
Hemodynamic data
Mean PCWP, mmHg	19.5±9.9	11.2±4.7	<0.001
Mean PA, mmHg	27.3±13.5	18.6±5.8	0.003
Mean RA, mmHg	8.3±6.0	4.4±2.9	0.005
CI, L·min−1·m−2	2.43±0.67	2.84±0.74	0.084

All values are shown as mean±SD, median (IQR), or n (%). Abbreviations as in Table 1.

Comparison of EAC and BNP as Biomarkers of HF

To test the usefulness of EAC as a marker of HF, we compared EAC with BNP, a widely used blood biomarker for the diagnosis and prognostic evaluation of chronic HF. EAC positively correlated with plasma BNP level (r=0.316, P=0.001, Figure 4B). The plasma BNP level, as with EAC, positively correlated with mean PCWP (r=0.494, P<0.001, Figure 5C). However, unlike EAC, the plasma BNP level showed a significant correlation with cardiac index (r=−0.478, P<0.001, Figure 5D).

In addition, we analyzed the diagnostic accuracy of EAC for high mean PCWP ≥18 mmHg, in comparison with plasma BNP level. ROC analysis for PCWP ≥18 mmHg revealed an AUC of 0.726 (95% CI, 0.567–0.886; P=0.019) for EAC. An EAC threshold of 1.05 ppm maximized the sensitivity and specificity for PCWP ≥18 mmHg (sensitivity 50%, specificity 89%). Meanwhile, the AUC for BNP with PCWP ≥18 mmHg was 0.760 (95% CI, 0.606–0.915; P=0.007; Figure 6). BNP threshold of 204 pg/ml maximized the sensitivity and specificity for PCWP ≥18 mmHg (sensitivity 80%, specificity 70%). These results suggest that EAC had comparable diagnostic power to detect high PCWP as the plasma BNP level.

Discussion

This study demonstrated for the first time a positive correlation of EAC and hemodynamic parameters in patients with non-ischemic chronic HF. EAC greater than 1.05 ppm was associated with increased risk of HF with pulmonary and systemic hemodynamic congestion.

An increase in blood ketone bodies in patients with HF, known as HF ketosis, has been reported previously.⁹ Several mechanisms underlying HF ketosis were suggested. Sympathetic nerve activity and consequently impaired metabolism are considered to play a vital role in the pathogenesis of HF.^6,19,20 Elevation of plasma catecholamine levels because of sympathetic activation in HF may release free fatty acids into the systemic circulation, subsequently producing ketone bodies in the liver.^21,22 Elevated insulin resistance related to HF, which is caused by neurohormonal activation including sympathetic activity, also contributes to ketone body metabolism by reducing the availability of glucose as a systemic energy source,²³ even though overtly diabetic patients were excluded from this study. As a substitute energy source, ketone bodies might be produced in the liver and utilized in the heart and other organs.²⁴ Although we did not show direct evidence of activation of catecholamines or insulin resistance in this study, it is plausible that neurohormonal activation in HF would lead to increased ketone body production, resulting in elevated EAC.

In this study, we revealed by right heart catheterization that higher EAC was associated with systemic and pulmonary congestion rather than cardiac pump function. Experimental studies have previously shown that myocardial metabolism of ketone bodies and fatty acids is altered by myocardial pressure overload.^25,26 Elevated blood ketone bodies would be consumed by the myocardium; that is, there is elevated energy expenditure of ketone bodies in the setting of HF.²⁴ In addition, impaired utilization of skeletal muscle ketone bodies because of the debilitating heart condition may contribute to the increased serum ketone body concentration used as a myocardial fuel source.¹⁰ Thus, myocardial pressure overload, represented by high PCWP, might induce altered myocardial ketone body metabolism and augmented cardiac energy expenditure, subsequently leading to a compensatory increase in the biosynthesis of ketone bodies. In this study, increased serum creatinine level, increased total bilirubin level, and higher use of loop diuretics and mineralocorticoid receptor antagonists were observed in the high EAC group. These findings are possibly caused by systemic volume overload and treatment for fluid retention. Our study also suggested that cardiac index was not significantly associated with EAC, consistent with a previous report that congestive status, not cardiac index, correlated with blood ketone bodies.⁹ Moreover, we found no significant difference in EAC between HF patients with preserved and reduced EF (data not shown), suggesting that EAC is not affected by cardiac contractility.

Pulmonary congestion and lung capillary injury play an important role in HF. Inflammatory damage from pulmonary congestion alters the permeability of the lung.¹⁶ Thus, the pulmonary capillary hydrostatic pressure threshold for pulmonary edema decreases, causing alveolar flooding, and may increase the release of acetone into the airway.²⁵ In this study population, although EAC showed a positive relationship with blood TKB level and PCWP, the blood TKB level did not show a significant direct relationship with PCWP. PCWP, an indicator of pulmonary congestion, might be associated with EAC through the blood-gas barrier, in addition to the increased blood ketone body level in HF.

Exhaled breath contains thousands of molecules. Owing to the improvement of mass spectrometry and gas chromatography mass spectrometry instruments, many unique substances can be identified in exhaled breath,²⁶ and several kinds of exhaled breath are potential sources of biomarkers in cardiovascular diseases.²⁷ For example, exhaled nitric oxide produced during exercise reflects vascular endothelial dysfunction and has been associated with increased mortality in chronic HF.²⁸ In patients with HF, EAC has been previously reported as a biomarker associated with plasma BNP level and NYHA class.^12–14 Our current study demonstrated that EAC also correlated with hemodynamic parameters measured by cardiac catheterization in patients with chronic HF.

Cardiac biomarkers are essential as a clinical guide for the management of HF.^5,29–32 Serial measurement of the plasma BNP level is useful to optimize the use of cardiovascular drugs such as β-blockers and to confirm clinical improvement of HF.³³ However, many cardiac biomarkers such as BNP need blood sampling, with the accompanying pain and possible complications such as faintness, hematomas and nerve injury. Breath analysis, on the other hand, is noninvasive and convenient, simply requiring the collection of a breath sample, as we have previously reported in a case of HF illustrating the usefulness of EAC in clinical practice.³⁴ This present study demonstrated that EAC may have equivalent clinical value to BNP for estimating PCWP in chronic HF. Although BNP has limited usefulness in assessing PCWP in critically ill patients, it has been reported to have a significant correlation with PCWP in chronic HF.^35,36 For diagnosis and evaluation of the severity of HF, exhaled breath analysis has the possibility of being a noninvasive alternative to BNP measurement. In this study, because of the short duration of follow-up, we could not demonstrate whether EAC might be a candidate prognostic marker in HF patients. Future long-term observation is anticipated.

Study Limitations

The main limitation of this study was the small number of subjects. We were unable to confirm the association of EAC with functional capacity because of the small number of patients who underwent cardiopulmonary exercise testing. Second, the background disease of chronic HF in this study was disproportionate; 66% of all cases were of valvular origin, which is not in accord with the normal distribution of chronic HF. This is because we perform a large number of routine cardiac catheterizations before valve surgery as preoperative evaluation in patients with valvular heart disease. Parameters including LVEF, left ventricular dimensions, and left atrial volume index may be influenced by valvular heart disease such as severe mitral regurgitation. Third, a large number of HF patients may have been excluded in this study because ketone body metabolism can be influenced by systemic metabolic impairments other than HF.³⁷ Diabetes mellitus is a common illness that affects ketone metabolism; in our study 24% of excluded patients were diabetic.

In conclusion, breath analysis of exhaled acetone may be considered a useful noninvasive method of evaluating congestive hemodynamics in patients with non-ischemic chronic HF.

Acknowledgments

This work was supported by the Japanese Society for the Promotion of Science Kakenhi Grant No. 15K19401 (T.Y.); Kakenhi Grant No. 2511176 (N.J.); Kakenhi Grant No. 26253037 (A. Simouchi); the Intramural Research Fund of the National Cerebral and Cardiovascular Research Center 25-2-1 (A. Simouchi); and the Center of Innovation, Science and Technology based Radical Innovation and Entrepreneurship Program, Japan (A. Simouchi).

We express special thanks to Ms Yoshiko Kokusho and Dr Makoto Sawano in the Saitama Medical University for their helpful assistance.

Disclosures

None.

References

• 1.    Roger  VL. Epidemiology of heart failure. Circ Res 2013; 113: 646–659.
• 2.    Mosterd  A,  Cost  B,  Hoes  AW,  de Bruijne  MC,  Deckers  JW,  Hofman  A, et al. The prognosis of heart failure in the general population: The Rotterdam Study. Eur Heart J 2001; 22: 1318–1327.
• 3.    MacIntyre  K,  Capewell  S,  Stewart  S,  Chalmers  JW,  Boyd  J,  Finlayson  A, et al. Evidence of improving prognosis in heart failure: Trends in case fatality in 66 547 patients hospitalized between 1986 and 1995. Circulation 2000; 102: 1126–1131.
• 4.    Braunwald  E. Biomarkers in heart failure. N Engl J Med 2008; 358: 2148–2159.
• 5.    Tang  WH,  Francis  GS,  Morrow  DA,  Newby  LK,  Cannon  CP,  Jesse  RL, et al; National Academy of Clinical Biochemistry Laboratory Medicine. National Academy of Clinical Biochemistry Laboratory Medicine practice guidelines: Clinical utilization of cardiac biomarker testing in heart failure. Circulation 2007; 116: e99–e109, doi:10.1161/CIRCULATIONAHA.107.185267.
• 6.    Doehner  W,  Frenneaux  M,  Anker  SD. Metabolic impairment in heart failure: The myocardial and systemic perspective. J Am Coll Cardiol 2014; 64: 1388–1400.
• 7.    Ashrafian  H,  Frenneaux  MP,  Opie  LH. Metabolic mechanisms in heart failure. Circulation 2007; 116: 434–448.
• 8.    van Bilsen  M,  van Nieuwenhoven  FA,  van der Vusse  GJ. Metabolic remodelling of the failing heart: Beneficial or detrimental? Cardiovasc Res 2009; 81: 420–428.
• 9.    Lommi  J,  Kupari  M,  Koskinen  P,  Naveri  H,  Leinonen  H,  Pulkki  K, et al. Blood ketone bodies in congestive heart failure. J Am Coll Cardiol 1996; 28: 665–672.
• 10.    Janardhan  A,  Chen  J,  Crawford  PA. Altered systemic ketone body metabolism in advanced heart failure. Tex Heart Inst J 2011; 38: 533–538.
• 11.    Galassetti  PR,  Novak  B,  Nemet  D,  Rose-Gottron  C,  Cooper  DM,  Meinardi  S, et al. Breath ethanol and acetone as indicators of serum glucose levels: An initial report. Diabetes Technol Ther 2005; 7: 115–123.
• 12.    Kupari  M,  Lommi  J,  Ventila  M,  Karjalainen  U. Breath acetone in congestive heart failure. Am J Cardiol 1995; 76: 1076–1078.
• 13.    Marcondes-Braga  FG,  Gutz  IG,  Batista  GL,  Saldiva  PH,  Ayub-Ferreira  SM,  Issa  VS, et al. Exhaled acetone as a new biomaker of heart failure severity. Chest 2012; 142: 457–466.
• 14.    Samara  MA,  Tang  WH,  Cikach  F Jr,  Gul  Z,  Tranchito  L,  Paschke  KM, et al. Single exhaled breath metabolomic analysis identifies unique breathprint in patients with acute decompensated heart failure. J Am Coll Cardiol 2013; 61: 1463–1464.
• 15.    Guazzi  M. Alveolar-capillary membrane dysfunction in heart failure: Evidence of a pathophysiologic role. Chest 2003; 124: 1090–1102.
• 16.    Pappas  LK,  Giannopoulos  G,  Loukides  S,  Gavrielatos  G,  Athanasopoulou  E,  Alexanian  IP, et al. Exhaled breath condensate in acute and chronic heart failure: New insights into the role of lung injury and barrier dysfunction. Am J Respir Crit Care Med 2014; 190: 342–345.
• 17.    McKee  PA,  Castelli  WP,  McNamara  PM,  Kannel  WB. The natural history of congestive heart failure: The Framingham study. N Engl J Med 1971; 285: 1441–1446.
• 18.    Lang  RM,  Bierig  M,  Devereux  RB,  Flachskampf  FA,  Foster  E,  Pellikka  PA, et al. Recommendations for chamber quantification: A report from the American Society of Echocardiography’s Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 2005; 18: 1440–1463.
• 19.    Eisenhofer  G,  Friberg  P,  Rundqvist  B,  Quyyumi  AA,  Lambert  G,  Kaye  DM, et al. Cardiac sympathetic nerve function in congestive heart failure. Circulation 1996; 93: 1667–1676.
• 20.    Grassi  G,  Seravalle  G,  Cattaneo  BM,  Lanfranchi  A,  Vailati  S,  Giannattasio  C, et al. Sympathetic activation and loss of reflex sympathetic control in mild congestive heart failure. Circulation 1995; 92: 3206–3211.
• 21.    Paolisso  G,  Gambardella  A,  Galzerano  D,  D’Amore  A,  Rubino  P,  Verza  M, et al. Total-body and myocardial substrate oxidation in congestive heart failure. Metabolism 1994; 43: 174–179.
• 22.    Lommi  J,  Kupari  M,  Yki-Jarvinen  H. Free fatty acid kinetics and oxidation in congestive heart failure. Am J Cardiol 1998; 81: 45–50.
• 23.    Swan  JW,  Anker  SD,  Walton  C,  Godsland  IF,  Clark  AL,  Leyva  F, et al. Insulin resistance in chronic heart failure: Relation to severity and etiology of heart failure. J Am Coll Cardiol 1997; 30: 527–532.
• 24.    Du  Z,  Shen  A,  Huang  Y,  Su  L,  Lai  W,  Wang  P, et al. 1H-NMR-based metabolic analysis of human serum reveals novel markers of myocardial energy expenditure in heart failure patients. PLoS One 2014; 9: e88102, doi:10.1371/journal.pone.0088102.
• 25.    Guazzi  M,  Phillips  SA,  Arena  R,  Lavie  CJ. Endothelial dysfunction and lung capillary injury in cardiovascular diseases. Prog Cardiovasc Dis 2015; 57: 454–462.
• 26.    Dweik  RA,  Amann  A. Exhaled breath analysis: The new frontier in medical testing. J Breath Res 2008; 2: 030301, doi:10.1088/1752-7163/2/3/030301.
• 27.    Cikach  FS Jr,  Dweik  RA. Cardiovascular biomarkers in exhaled breath. Prog Cardiovasc Dis 2012; 55: 34–43.
• 28.    Katz  SD,  Hryniewicz  K,  Hriljac  I,  Balidemaj  K,  Dimayuga  C,  Hudaihed  A, et al. Vascular endothelial dysfunction and mortality risk in patients with chronic heart failure. Circulation 2005; 111: 310–314.
• 29.    Hanatani  S,  Izumiya  Y,  Takashio  S,  Kimura  Y,  Araki  S,  Rokutanda  T, et al. Circulating thrombospondin-2 reflects disease severity and predicts outcome of heart failure with reduced ejection fraction. Circ J 2014; 78: 903–910.
• 30.    Do e  Z,  Fukumoto  Y,  Sugimura  K,  Miura  Y,  Tatebe  S,  Yamamoto  S, et al. Rho-kinase activation in patients with heart failure. Circ J 2013; 77: 2542–2550.
• 31.    Neves  FM,  Meneses  GC,  Sousa  NE,  Menezes  RR,  Parahyba  MC,  Martins  AM, et al. Syndecan-1 in acute decompensated heart failure: Association with renal function and mortality. Circ J 2015; 79: 1511–1519.
• 32.    Ueda  T,  Kawakami  R,  Nishida  T,  Onoue  K,  Soeda  T,  Okayama  S, et al. Plasma renin activity is a strong and independent prognostic indicator in patients with acute decompensated heart failure treated with renin-angiotensin system inhibitors. Circ J 2015; 79: 1307–1314.
• 33.    Kawai  K,  Hata  K,  Takaoka  H,  Kawai  H,  Yokoyama  M. Plasma brain natriuretic peptide as a novel therapeutic indicator in idiopathic dilated cardiomyopathy during beta-blocker therapy: A potential of hormone-guided treatment. Am Heart J 2001; 141: 925–932.
• 34.    Yokokawa  T,  Sugano  Y,  Shimouchi  A,  Shibata  A,  Nakayama  T,  Ohara  T, et al. A case of acute decompensated heart failure evaluated by series of exhaled acetone concentrations as noninvasive biomarker of heart failure severity. Int J Cardiol 2015; 204: 112–113.
• 35.    Forfia  PR,  Watkins  SP,  Rame  JE,  Stewart  KJ,  Shapiro  EP. Relationship between B-type natriuretic peptides and pulmonary capillary wedge pressure in the intensive care unit. J Am Coll Cardiol 2005; 45: 1667–1671.
• 36.    Parsonage  WA,  Galbraith  AJ,  Koerbin  GL,  Potter  JM. Value of B-type natriuretic peptide for identifying significantly elevated pulmonary artery wedge pressure in patients treated for established chronic heart failure secondary to ischemic or idiopathic dilated cardiomyopathy. Am J Cardiol 2005; 95: 883–885.
• 37.    Li  W,  Liu  Y,  Lu  X,  Huang  Y,  Liu  Y,  Cheng  S, et al. A cross-sectional study of breath acetone based on diabetic metabolic disorders. J Breath Res 2015; 9: 016005, doi:10.1088/1752-7155/9/1/016005.