Association Between Aortic Stiffness and Exercise Tolerance in Patients at the Risk Stage of Heart Failure

Abstract

Background: The number of patients with heart failure (HF) has increased, and it is crucial to prevent the development of HF in patients at risk of HF. The present study aimed to risk stratify patients in Stage A and B HF based on associations between exercise-induced changes in aortic stiffness and exercise tolerance.

Methods and Results: Patients in Stage A and B HF who performed a cardiopulmonary exercise test were enrolled in the study (n=106; median age 65.0 years [interquartile range 52.8–73.0 years]). Exercise tolerance was examined by the percentage of predicted peak oxygen consumption (%V̇O₂peak). The ascending aortic pressure waveform was estimated non-invasively. Aortic stiffness was assessed using the augmentation index (AIx) and reflection magnitude (RM). Multivariable regression analysis showed that AIx measured both before and after exercise was significantly associated with %V̇O₂peak (β=−0.221 [P=0.049] and β=−0.342 [P=0.003], respectively). When participants were divided into %V̇O₂peak subgroups using a cut-off value of 60%, RM decreased immediately after exercise and remained lower 5 min after exercise in the group with preserved exercise tolerance, but recovered to baseline levels 5 min after exercise in the group with reduced exercise tolerance.

Conclusions: Exercise-induced increases in aortic stiffness were associated with exercise tolerance in patients at risk of HF, suggesting that exercise-induced changes in aortic stiffness may be useful to stratify high-risk patients.

The increasing number of patients with heart failure (HF) has become a social problem worldwide.^1–3 In Japan, which is considered an aging society, the number of patients with HF with preserved ejection fraction (HFpEF) is increasing.^4,5 Medications with beneficial effects on prognosis and cardiovascular events in patients with HF with reduced ejection fraction (HFrEF) or HF with mildly reduced ejection fraction (HFmrEF) have been administered clinically.^6–12 Recently, sodium-glucose cotransporter 2 inhibitors were shown to reduce cardiovascular death or worsening HF in patients with HFpEF.^13,14 However, there are few medications to reduce mortality or clinical events in patients with HFpEF compared with HFrEF and HFmrEF. Therefore, it is important to prevent the development of HF, especially HFpEF, in patients at risk of HF with cardiovascular risk factors (Stage A) or with organic cardiovascular disease but no history of HF (Stage B).^15,16 To this end, it is critical to establish a reliable method for stratifying patients in Stages A and B at high risk of developing HFpEF.

Hypertension (HT) is a leading cause of HFpEF¹⁷ because it increases the afterload and leads to the development of left ventricular (LV) hypertrophy and diastolic dysfunction. Therefore, investigating LV afterload may help risk stratify patients in HF Stages A and B.

Central blood pressure (CBP) reflects the LV afterload more directly than brachial blood pressure (BP) and has been found to be closely related to cardiovascular events.^18,19 CBP can be measured non-invasively from brachial BP based on the generalized transfer function.²⁰ Aortic stiffness parameters, such as the augmentation index (AIx) and reflection magnitude (RM), can be measured using the CBP waveform.

Conversely, we focused on exercise tolerance as an objective risk of patients in Stages A and B, because reduced exercise tolerance is related to poor prognosis in patients with advanced HF.^1,21,22 Exercise tolerance is determined by several factors, such as respiratory, cardiac, skeletal muscle, and autonomic nerve function, as well as the pulmonary and peripheral circulation and anemia.²³ We hypothesized that the characteristics of HT, especially elevated CBP, an essential factor causing HFpEF, are determinants of exercise tolerance. Furthermore, analyzing the characteristics of CBP and its association with exercise tolerance could prove useful for risk stratifying patients in HF Stages A and B. To test this hypothesis, we examined the relationship between exercise-induced changes in the CBP waveform and exercise tolerance in patients in HF Stages A and B.

Methods

The study was approved by the Ethics Committee of Nagoya City University and was performed in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from all participants prior to enrollment. All data used in the analysis were anonymized.

Study Participants and Protocol

The present study investigated patients in HF Stages A and B,¹⁵ with cardiovascular risk factors or cardiovascular diseases but no history of HF, respectively. All participants underwent a cardiopulmonary exercise test (CPX) between 2019 and 2021 at Nagoya City University Hospital. Patients with frequent arrhythmia, including atrial fibrillation, pacemaker implantation, pulmonary HT, or a history of congenital heart or aortic disease were excluded from the study.

Patients taking antihypertensive medications or with systolic BP (SBP) ≥140 mmHg and/or diastolic BP (DBP) ≥90 mmHg were defined as having HT.²⁴ Patients taking lipid-lowering medications or with high-density lipoprotein cholesterol <40 mg/dL, low-density lipoprotein cholesterol ≥140 mg/dL, or triglyceride ≥150 mg/dL were defined as having dyslipidemia.²⁵ Patients taking blood glucose-lowering medications or with a fasting plasma glucose concentration ≥126 mg/dL and HbA1c ≥6.5% were defined as having diabetes.²⁶ Patients with stenosis ≥50% in at least 1 coronary vessel or a clinical history of percutaneous coronary intervention were defined as having coronary artery disease (CAD).²⁷ Patients with abnormal kidney structure or persistent abnormalities in kidney function (e.g., glomerular filtration rate [GFR] <60 mL/min/1.73 m² or albuminuria ≥30 mg/24 h) for >3 months were defined as having chronic kidney disease.²⁸

Ordinary brachial SBP and DBP were measured in patients in a seated position using a validated oscillometric technique. Fasting blood samples were collected from the antecubital vein for laboratory measurements. Echocardiographic examinations were performed to evaluate cardiac function. Blood tests and echocardiographic examinations were performed within 2 weeks before or after CPX.

CBP Waveform Analysis

The CBP waveform was obtained by transforming the brachial artery pressure waveform using a non-invasive device (SphygmoCor XCEL; AtCor Medical, Sydney, NSW, Australia). The reliability of this device has been validated previously by comparison with CBP measured using a pressure wire.²⁹

Central pulse pressure (CPP) was defined as the pressure difference between the maximum and minimum pressures on the CBP waveform (Figure A). The augmentation pressure (AP) was calculated as the pressure difference between the first systolic shoulder and peak pressure of the CBP waveform. The AIx was calculated as the ratio of AP to CPP.

Figure.

(A) Representative image of the central blood pressure (CBP) waveform at baseline. (B,C) Summary images of CBP waveforms before (red lines), immediately after (blue lines), and 5 min after (green lines) exercise in groups with preserved (B) and reduced (C) exercise tolerance (percentage predicted peak oxygen consumption [%V̇O₂peak] ≥60% and <60%, respectively). Solid lines indicate the aortic pressure waveform. Dashed and dotted lines represent the forward and backward wave components, respectively. AIx, augmentation index; AP, augmentation pressure; CPP, central pulse pressure; Pb, peak pressure of the backward wave; Pf, peak pressure of the forward wave; RM, reflection magnitude; Tbc, corrected time to peak pressure in backward waves divided by the interval in each phase; Tfc, corrected time to peak pressure in forward waves divided by the interval in each phase.

The forward and backward wave components of the CBP waveform were also obtained (Figure A). The amplitude of the peak pressure in the forward and backward waves (Pf and Pb, respectively) and the time to peak pressure in the forward and backward waves (Tf and Tb, respectively) were measured. The Pb/Pf ratio was calculated using the RM. To correct for the influence of heart rate (HR), we corrected the Tf and Tb based on the interval and defined Tf/ as the corrected Tf (Tfc) and Tb/ as the corrected Tb (Tbc) in each phase. AIx and RM were used as indices of aortic stiffness.^30,31

CPX Protocol

CPX was performed using an electrocardiographic monitored ergometer (STB-3200; CAT EYE, Osaka, Japan) equipped with an expiratory gas analysis machine (Minato Medical Science Co., Ltd., Osaka, Japan) to measure peak oxygen consumption (V̇O₂). The exercise load was performed using the ramp protocol.³² To determine aerobic capacity, we encouraged all patients to exercise up to the symptom limit as indicated by a respiratory exchange ratio (RER) greater than 1.1.³² Peak V̇O₂ was defined as the highest V̇O₂ value, and the aerobic threshold (AT) was evaluated by the V-slope method.²³

Exercise tolerance was examined based on the percentage of the predicted peak V̇O₂ (%V̇O₂peak) because the standard value of peak V̇O₂ depends on age and sex.³² In this study, %V̇O₂peak was calculated as the measured value divided by the standard value based on the Japanese standard value of the respiratory and circulatory indices during exercise.³³ Participants were divided into 2 subgroups, one with preserved exercise tolerance (pET; %V̇O₂peak ≥60%) and the other with reduced exercise tolerance (rET; %V̇O₂peak <60%), according to the 2021 Japanese Circulation Society guidelines on rehabilitation in patients with cardiovascular disease.³² These guidelines define %V̇O₂peak ≥80% as normal exercise tolerance, %V̇O₂peak between 60% and <80% as mildly reduced exercise tolerance, %V̇O₂peak between 40% and <60% as moderately reduced exercise tolerance, and %V̇O₂peak <40% as severely reduced exercise tolerance.

We aimed to detect patients with clearly reduced exercise tolerance and defined the rET group as those patients with a moderate or greater reduction in %V̇O₂peak <60%. Therefore, the pET group included patients with mildly reduced exercise tolerance based on the guideline.

Each CBP waveform was obtained at baseline, immediately after exercise, and 5 min after exercise, and all parameters described above were measured.

Cardiac function was examined by trained sonographers using an echocardiographic system (Vivid E9; GE Healthcare, Chicago, IL, USA), as described previously.³⁴ LV ejection fraction was measured using the biplane disk summation method. In the apical 4-chamber view, the peak velocity of the transmitral inflow during early and late diastole (E and A, respectively) and mitral annular movement at both the septal and lateral annular corners of the mitral annulus (septal e′ and lateral e′, respectively) were recorded, with E/A and averaged E/e′ calculated as indices of LV diastolic function.

Statistical Analysis

Data were analyzed using IBM SPSS version 23 (IBM Corp., Chicago, IL, USA). Data with a normal distribution, as determined using the Shapiro-Wilk test, are presented as numbers with percentages or as the mean±SD. Data that were not normally distributed are presented as the median with interquartile range. Correlations among continuous variables were analyzed using Pearson’s correlation coefficients (r). The 2 groups (rET and pET) were compared using Student’s t-test for parametric continuous variables, the Mann-Whitney U test for non-parametric continuous variables, and the χ² test for categorical variables. Multivariable regression analyses were performed to determine whether there was a significant correlation between exercise tolerance and AIx or other CBP parameters. Logistic regression analyses were used to investigate possible associations between abnormal aortic parameters and reduced exercise tolerance. Receiver operating characteristic (ROC) curve analysis was used to determine cut-off values to predict rET. The best cut-off value was defined as the point with the highest sum of sensitivity and specificity (Youden index). The dichotomous variable of sex was assigned a value of 0 for female and 1 for male. Statistical significance was set at P<0.05.

Results

In all, 106 patients were enrolled in the study. Of these patients, 15 presented with significant ST-T changes during the CPX, indicative of myocardial ischemia.

The baseline characteristics of the enrolled patients are presented in Table 1. The median patient age was 65 years, and 59.4% of patients were men. The proportion of patients with HT, dyslipidemia, and diabetes was 44.3%, 42.5%, and 15.1%, respectively; approximately one-third of patients had a history of CAD.

Table 1. Characteristics of Study Participants (n=106)

Age (years)	65.0 [52.8–73.0]
Male sex	63 (59.4)
Body mass index (kg/m2)	22.7 [21.0–25.5]
Brachial SBP (mmHg)	138±19
Brachial DBP (mmHg)	81±13
Heart rate (beats/min)	77±14
Chronic kidney disease	28 (28)
Moderate or severe heart valve disease	14 (13.2)
Mitral regurgitation	6 (5.7)
Aortic regurgitation	5 (4.7)
Aortic stenosis	3 (2.8)
Chronic obstructive pulmonary disease	3 (2.8)
Moderate or severe obesity, including SAS	1 (0.9)
Coronary artery disease	38 (35.8)
Cardiovascular risk factors
Hypertension	47 (44.3)
Dyslipidemia	45 (42.5)
Diabetes	16 (15.1)
Laboratory data
Hemoglobin (g/dL)	13.6 [12.8–14.3]
HbA1c (%)	6.0 [5.5–6.5]
LDL-C (mg/dL)	98 [78–122]
HDL-C (mg/dL)	54 [43–66]
Triglycerides (mg/dL)	118 [87–165]
Creatinine (mg/dL)	0.82 [0.67–0.97]
eGFR (mL/min/1.73 m2)	68.0±18.4
BNP (pg/dL)	47.2 [25.7–89.4]
Medications
Calcium channel blocker	25 (23.6)
ACEI/ARB	38 (35.8)
Diuretics	10 (9.4)
β-blocker	23 (21.7)
Statin	42 (39.6)
Hypoglycemic drug	14 (13.2)
Echocardiographic parameters
LVEF (%)	65.3 [56.3–69.4]
No. participants with reduced (<50%) LVEF	14 (13.2)
E/A	0.82 [0.70–1.10]
E/e′	9.9 [7.7–13.6]
Central hemodynamics parameters
Central SBP (mmHg)	126.0 [111.0–135.3]
Central DBP (mmHg)	82±13
CPP (mmHg)	40 [33.0–48.0]
AIx (%)	26.4±15.4
Pf (mmHg)	31.0 [26.0–37.0]
Pb (mmHg)	17.8 [14.4–21.5]
RM (%)	57 [51.5–66.5]
Tfc (ms)	143.6±22.7
Tbc (ms)	280.7 [261.7–292.4]
Cardiopulmonary exercise test
Peak V̇O2 (mL/kg/min)	19.3 [15.1–22.7]
%V̇O2peak (%)	78.6±19.2
Aerobic threshold (mL/kg/min)	11.5 [9.9–13.6]
Peak RER (V̇CO2/V̇O2)	1.23 [1.15–1.29]
V̇E vs. V̇CO2 slope	29.4 [26.2–34.7]

Data are presented as the mean±SD, median [interquartile range], or n (%). ACEI, angiotensin-converting enzyme inhibitor; AIx, augmentation index; ARB, angiotensin receptor blocker; BNP, B-type natriuretic peptide; CPP, central pulse pressure; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; LVEF, left ventricular ejection fraction; Pb, peak pressure of the backward wave; Pf, peak pressure of the forward wave; RER, respiratory exchange ratio; RM, reflection magnitude; SAS, sleep apnea syndrome; SBP, systolic blood pressure; Tbc, corrected time to peak pressure in backward waves divided by the interval in each phase; Tfc, corrected time to peak pressure in forward waves divided by the interval in each phase; V̇CO₂, carbon dioxide production; V̇E, minute ventilation; V̇O₂, oxygen consumption; %V̇O₂peak, percentage of predicted peak V̇O₂.

Initially, we analyzed the correlation between %V̇O₂peak and parameters obtained at baseline (Table 2). Age, body mass index (BMI), CAD, HT, diabetes, E/e′, CPP, AIx, RM, Tfc, and Tbc were inversely correlated with the %V̇O₂peak, whereas male sex and E/A were positively correlated with %V̇O₂peak.

Table 2. Correlation Between %V̇O₂peak and Parameters Obtained at Baseline in All Patients (n=106)

	r	P value
Age	−0.310	0.001
Male sex	0.241	<0.05
Body mass index	−0.267	0.006
Brachial SBP	−0.120	0.222
Brachial DBP	0.070	0.479
Heart rate	−0.174	0.074
Chronic kidney disease	0.123	0.226
Coronary artery disease	−0.268	<0.01
Cardiovascular risk factors
Hypertension	−0.260	<0.01
Dyslipidemia	−0.190	0.051
Diabetes	−0.215	<0.05
Echocardiographic parameters
LVEF	0.004	0.967
E/A	0.376	<0.001
E/e′	−0.397	<0.001
Central hemodynamics parameters
Central SBP	−0.169	0.083
Central DBP	0.046	0.642
CPP	−0.313	0.001
AIx	−0.388	<0.001
RM	−0.286	0.003
Tfc	−0.210	0.031
Tbc	−0.231	0.018
Cardiopulmonary exercise test
Peak V̇O2	0.934	<0.001
%V̇O2peak	–	–
Aerobic threshold	0.645	<0.001
Peak RER (V̇CO2/V̇O2)	0.196	0.044
V̇E vs. V̇CO2 slope	−0.273	0.005

Abbreviations as in Table 1.

Next, we examined independent determinants of %V̇O₂peak. Table 3 summarizes the results of the multivariable regression analysis. Multicollinearity was observed between AIx and CPP, RM, and Tfc (Supplementary Table 1). Therefore, these parameters were assessed separately in further multivariable regression analyses. Age, sex, BMI, CAD, HT, diabetes, E/A and E/e′ were used as adjustment variables because they were associated with %V̇O₂peak. AIx values at each time point showed a significant independent correlation with %V̇O₂peak. RM values also showed significant independent correlations with %V̇O₂peak immediately and 5 min after exercise, but not at baseline. In contrast, there were no significant correlations between %V̇O₂peak and CPP, Tfc, or Tbc at any time point in any of the statistical models. We also evaluated associations between AIx and the other CPX parameters, such as peak V̇O₂, AT, and minute ventilation (V̇E) vs. carbon dioxide production (V̇CO₂) slope. Although peak V̇O₂ and AT were inversely correlated with AIx at baseline (r=−0.413 [P<0.001] and r=−0.274 [P=0.005], respectively), V̇E vs. V̇CO₂ slope was not (r=0.187; P=0.060). The multivariable regression analysis did not show a significant independent correlation between AIx and peak V̇O₂ or AT at any time point (Supplementary Table 2).

Table 3. Multivariable Regression Analyses Showing Correlations of %V̇O₂peak With the AIx, CPP, RM, and Tfc and Tbc (n=106)

	Model 1			Model 2			Model 3
	R2	β	P value	R2	β	P value	R2	β	P value
Analysis using AIx
At baseline	0.251	−0.221	0.049	0.336	−0.261	0.015	0.342	−0.281	0.014
Immediately after exercise	0.262	−0.342	0.003	0.347	−0.302	0.005	0.400	−0.456	<0.001
5 min after exercise	0.254	−0.232	0.047	0.320	−0.221	0.047	0.375	−0.349	0.002
Analysis using CPP
At baseline	0.244	−0.155	0.113	0.314	−0.155	0.103	0.320	−0.120	0.121
Immediately after exercise	0.219	0.009	0.927	0.296	−0.003	0.971	0.298	−0.006	0.950
5 min after exercise	0.228	−0.061	0.550	0.298	−0.054	0.581	0.305	−0.070	0.494
Analysis using RM
At baseline	0.225	−0.093	0.365	0.344	−0.143	0.201	0.304	−0.104	0.317
Immediately after exercise	0.278	−0.246	0.012	0.391	−0.268	0.020	0.340	−0.211	0.031
5 min after exercise	0.258	−0.232	0.037	0.380	−0.261	0.022	0.355	−0.294	0.007
Analysis using Tfc
At baseline	0.225	−0.081	0.369	0.333	−0.073	0.480	0.307	−0.107	0.235
Immediately after exercise	0.251	−0.153	0.094	0.370	−0.174	0.081	0.323	−0.144	0.119
5 min after exercise	0.229	−0.067	0.444	0.351	−0.142	0.160	0.320	−0.141	0.120
Analysis using Tbc
At baseline	0.264	−0.211	0.015	0.351	−0.149	0.116	0.327	−0.174	0.044
Immediately after exercise	0.233	−0.061	0.486	0.350	−0.085	0.368	0.315	−0.103	0.239
5 min after exercise	0.225	0.006	0.947	0.333	0.019	0.843	0.302	−0.007	0.934

Listed variables were analyzed separately in each regression model. Model 1 was adjusted for age, sex, and body mass index. Model 2 was further adjusted for hypertension, diabetes, and coronary artery disease, in addition to the variable adjusted for in Model 1. Model 3 was further adjusted for E/A and E/e′, in addition to the variable adjusted for in Model 1. Abbreviations as in Table 1.

Because AIx and RM were found to be determinants of %V̇O₂peak, we next examined how physiological parameters change during exercise at different exercise tolerance levels. Participants were divided into 2 subgroups according to a cut-off value of %V̇O₂peak of 60%. Consequently, 88 patients were classified into the pET group (%V̇O₂peak ≥60%) and the remaining 18 patients were classified into the rET group (%V̇O₂peak <60%). Table 4 summarizes the clinical characteristics and physiological parameters from the exercise stress test in the pET and rET groups. The prevalence of CAD and HT was higher, accompanied by an increased frequency of medications, in the rET than pET group. In addition, the differences in the Doppler indices suggested the presence of LV diastolic dysfunction in the rET group. At baseline, there were no differences in any parameters, except for AIx, between the 2 groups. At each time point, AIx was significantly lower in the pET than rET group (baseline: 25±15% vs. 33±16% [P=0.045]; immediately after exercise: 7±23% vs. 25±24% [P=0.006]; 5 min after exercise: 15±19% vs. 25±27% [P=0.049]).

Table 4. Characteristics of Groups With pET and rET (%V̇O₂peak ≥60% and <60%, Respectively) at Baseline and After Cardiopulmonary Exercise Testing

	pET (n=88)	rET (n=18)	P value
Age (years)	65 [50–73]	65 [60–76]	0.266
Male sex	54 (61)	8 (44)	0.188
Body mass index (kg/m2)	22.5 [21.0–25.2]	25.3 [22.0–30.6]	0.076
Brachial SBP (mmHg)	138±18	139±25	0.849
Brachial DBP (mmHg)	82±12	78±16	0.084
Heart rate (beats/min)	76±13	81±18	0.097
Chronic kidney disease	21 (26)	6 (35)	0.219
Moderate or severe heart valve disease	11 (12.5)	3 (3.4)	0.613
Mitral regurgitation	5 (5.7)	1 (5.6)	0.983
Aortic regurgitation	4 (4.5)	1 (5.6)	0.823
Aortic stenosis	2 (2.3)	1 (5.6)	0.424
Chronic obstructive pulmonary disease	2 (2.3)	1 (5.6)	0.449
Moderate or severe obesity, SAS	0 (0)	1 (5.6)	0.340
Coronary artery disease	28 (32)	10 (56)	<0.001
Cardiovascular risk factors
Hypertension	34 (39)	13 (72)	0.029
Dyslipidemia	34 (39)	11 (61)	0.313
Diabetes	10 (11)	6 (33)	0.116
Laboratory data
Hemoglobin (g/dL)	13.7 [13.0–14.4]	12.1 [11.4–13.9]	0.008
HbA1c (%)	5.9 [5.5–6.2]	6.2 [6.0–6.8]	0.069
LDL-C (mg/dL)	101 [79–127]	97 [77–107]	0.359
HDL-C (mg/dL)	59 [47–72]	42 [38–48]	0.001
Triglycerides (mg/dL)	120 [84–164]	110 [92–159]	0.892
Creatinine (mg/dL)	0.80 [0.67–0.97]	0.83 [0.64–0.97]	0.829
eGFR (mL/min/1.73 m2)	68.9±16.9	64.0±24.5	0.323
BNP (pg/dL)	39.2 [23.1–70.7]	118.8 [68.9–310.3]	0.002
Medications
β-blocker	16 (18)	6 (33)	0.289
ACEI/ARB	28 (32)	12 (67)	0.016
Calcium channel blocker	20 (23)	6 (33)	0.537
Diuretics	2 (2)	7 (39)	0.007
Statin	31 (35)	11 (61)	0.088
Hypoglycemic drug	9 (10)	5 (28)	0.196
Echocardiographic parameters
LVEF (%)	65.3 [56.6–69.4]	64.1 [56.2–69.2]	0.536
No. participants with reduced (<50%) LVEF	12 (13.6)	2 (11.1)	0.477
E/A	0.96±0.39	0.75±0.19	0.002
E/e′	9.3 [7.4–13.4]	11.3 [9.2–14.2]	0.052
Cardiopulmonary exercise test
Peak V̇O2 (mL/kg/min)	20.6 [16.6–23.2]	12.6 [11.3–13.7]	<0.001
%V̇O2peak (%)	85.0 [69.5–96.0]	50.5 [48.3–54.0]	<0.001
Aerobic threshold (mL/kg/min)	11.9 [10.5–13.9]	9.2 [8.3–9.8]	<0.001
Peak RER (V̇CO2/V̇O2)	1.23±0.10	1.21±0.13	0.081
V̇E vs. V̇CO2 slope	29.3 [26.4–34.2]	33.7 [25.7–38.6]	0.438
Pressure parameters at baseline
Central SBP (mmHg)	124±16	126±23	0.684
Central DBP (mmHg)	83±12	80±17	0.450
CPP (mmHg)	41±10	46±16	0.073
AIx (%)	25±15	33±16	0.045
Pf (mmHg)	31±7	34±12	0.545
Pb (mmHg)	18±5	21±7	0.421
RM (%)	58±10	63±11	0.081
Tfc (ms)	143±23	149±23	0.292
Tbc (ms)	278±26	282±28	0.487
Immediately after exercise
Central SBP (mmHg)	140±23	133±27	0.305
Central DBP (mmHg)	92±17	84±18	0.056
CPP (mmHg)	47±15	49±17	0.616
AIx (%)	7±23	25±24	0.006
Pf (mmHg)	42±12	39±13	0.389
Pb (mmHg)	21±7	21±8	0.849
RM (%)	49±8	55±13	0.092
Tfc (ms)	139±22	147±21	0.199
Tbc (ms)	293±26	297±22	0.522
5 min after exercise
Central SBP (mmHg)	122±17	120±23	0.787
Central DBP (mmHg)	83±14	77±15	0.138
CPP (mmHg)	39±12	43±19	0.411
AIx (%)	15±19	25±27	0.049
Pf (mmHg)	33±8	32±12	0.896
Pb (mmHg)	17±6	19±10	0.366
RM (%)	52±9	61±14	0.021
Tfc (ms)	144±19	145±26	0.929
Tbc (ms)	296±26	284±33	0.096

Data are presented as the mean±SD, median [interquartile range], or n (%). pET, preserved exercise tolerance; rET, reduced exercise tolerance. Other abbreviations as in Table 1.

Next, changes in physiological parameters during the exercise test were evaluated within each group (Table 5). In the pET group, AIx decreased significantly from baseline immediately after exercise (25±15 vs. 7±23 mmHg, respectively; P<0.001), and this decrease persisted 5 min after exercise (15±19 mmHg; P<0.001 vs. baseline).

Table 5. Changes of Physiological Parameters, Including Central Hemodynamics, Before and After Cardiopulmonary Exercise Testing in Groups With pET and rET (%V̇O₂peak ≥60% and <60%, Respectively)

	At baseline	Immediately after exercise	5 min after exercise
pET group (n=88)
Brachial SBP (mmHg)	138±18	165±28*	139±20
Brachial DBP (mmHg)	82±12	88±15*	80±13***
Heart rate (beats/min)	76±13	102±20*	94±18*
Central hemodynamic parameters
Central SBP (mmHg)	124±16	140±23*	122±17***
Central DBP (mmHg)	83±12	92±17*	83±14
CPP (mmHg)	41±10	47±15*	39±12***
AIx (%)	25±15	7±23*	15±19*
Pf (mmHg)	31±7	42±12*	33±8
Pb (mmHg)	18±5	21±7*	17±6***
RM (%)	58±10	49±8*	52±9*
Tfc (ms)	143±23	139±22	144±20
Tbc (ms)	278±26	293±26*	296±26*
rET group (n=18)
Brachial SBP (mmHg)	139±25	152±30***	134±25
Brachial DBP (mmHg)	78±16	81±17	75±14
Heart rate (beats/min)	81±18	94±25*	87±25
Central hemodynamic parameters
Central SBP (mmHg)	126±23	133±27	120±23
Central DBP (mmHg)	80±17	84±18	77±15
CPP (mmHg)	46±16	49±17	43±19
AIx (%)	33±16	25±24***	25±27***
Pf (mmHg)	34±12	39±13	32±12
Pb (mmHg)	21±7	21±8	19±10
RM (%)	63±11	55±13***	61±14
Tfc (ms)	149±23	147±21	145±26
Tbc (ms)	282±28	297±22***	284±33

Data are presented as the mean±SD. *P<0.001, **P<0.01, ***P<0.05 compared with baseline. Abbreviations as in Tables 1,4.

Looking at changes in factors affecting AIx, we found significantly decreased RM and prolonged Tbc immediately and 5 min after exercise compared with baseline. Meanwhile, CPP increased significantly from baseline immediately after exercise (41±10 vs. 47±15 mmHg; P<0.001), with a modest decrease 5 min after exercise (39±12 mmHg; P=0.035 vs. baseline).

In the rET group, AIx decreased from baseline immediately after exercise (33±16% vs. 25±24%; P=0.012), with significantly decreased RM and prolonged Tbc compared with baseline. A modest decrease in AIx persisted 5 min after exercise (25±27%; P=0.044 vs. baseline), but RM and Tbc had returned to baseline levels.

There was no significant change in CPP after exercise.

Table 6 summarizes the results of the logistic regression analysis of the correlation between exercise intolerance, defined as %V̇O₂peak <60%, and physiological parameters during the exercise test. The results of ROC curve analyses to determine cut-off points for different parameters are shown in Supplementary Figure. Abnormal AIx and RM at each time point were significantly associated with reduced exercise tolerance.

Table 6. Logistic Regression Analyses Showing Possible Associations With rET (%V̇O₂peak <60%; n=106)

	Odds ratio	95% CI	P value
Abnormal AIx
At baseline	3.526	1.228–10.128	0.019
Immediately after exercise	6.364	1.979–20.465	0.002
5 min after exercise	5.615	1.876–16.807	0.002
Abnormal RM
At baseline	2.986	1.049–8.495	0.040
Immediately after exercise	3.321	1.112–9.922	0.032
5 min after exercise	5.571	1.809–17.157	0.003

For each factor, the cut-off value to predict rET determined by receiver operating characteristic (ROC) curve analysis was used. The best cut-off value was defined as the point with the highest sum of sensitivity and specificity (Youden index). Abnormal AIx and RM are shown in the Supplementary Figure. CI, confidence interval. Other abbreviations as in Tables 1,4.

Discussion

The main findings of the present study are as follows. First, AIx was significantly associated with %V̇O₂peak before and after exercise, after adjusting for age, sex, BMI, E/A, and E/e′, but there was no association between %V̇O₂peak and CPP before and after exercise. Second, in the pET group, the AIx decreased significantly immediately after exercise, and this persisted up to 5 min after exercise, whereas in the rET group AIx was modestly decreased immediately after exercise and 5 min later. Third, RM was significantly lower in the pET than rET group 5 min after exercise, although baseline levels were similar in both groups. Both Pf and Pb increased significantly in the pET group, whereas RM decreased after exercise; these responses were blunted in the rET group. In contrast, although Tfc did not change after exercise in either group, Tbc increased significantly in the pET group but only modestly increased in the rET group. These findings are in line with the hypothesis that aortic stiffness before and after exercise is associated with exercise tolerance in individuals at risk of HF.

Various studies have reported that non-invasively measured arterial stiffness is increased in patients with HFpEF compared with healthy controls;^35,36 other studies have shown that arterial stiffness in patients with HFpEF is associated with decreased exercise tolerance.³⁷ The data of the present study are consistent with previous findings showing a correlation between exercise tolerance and arterial stiffness at rest; however, our study may be the first to report that the correlation between AIx and exercise tolerance is stronger during exercise than at rest in patients at risk of HF. Because HR may be an independent predictor of AIx, increased HR after exercise could be responsible for this result.³⁸

Increased arterial stiffness leads to decreased DBP and increased SBP, which, in turn, increases pulse pressure (PP).^39,40 Increased PP reduces coronary perfusion because DBP regulates the coronary circulation.⁴¹ Exercise tolerance decreases when coronary perfusion is compromised in individuals with CAD, including the approximately 40% of patients in the present study. Coronary perfusion is not sufficient to maintain myocardial performance, especially in patients with CAD, because of their low ischemic threshold during exercise.⁴² This could reduce exercise tolerance.

Increases in arterial pressure and blood flow with exercise may cause vascular stiffening that is different from that at rest, which could emphasize the inverse relationship between arterial stiffness and exercise tolerance. In the pET group, a significant decrease in AIx was observed immediately after exercise. RM was also decreased at this time, which could be attributed to an increase in Pf. The increase in Pf may reflect an increase in cardiac output, because central SBP increased immediately after exercise, along with an increase in brachial SBP. Tbc was also significantly prolonged, which may reflect increased vascular compliance due to vasodilation. Furthermore, these changes continued until 5 min after exercise, at which point the decrease in RM was attributed to changes in Pb, suggesting that the vasodilator effects persisted after exercise. Conversely, the involvement of the parasympathetic nervous system was not clear as far as HR was concerned.

The rET group also showed a decrease in AIx and RM immediately after exercise. However, there were no significant changes in Pf or Pb. Tbc was also prolonged, but this change was statistically weaker than in the pET group. Thus, AIx and RM immediately after exercise in the rET group were higher than in the pET group. After 5 min, AIx in the rET group continued to decrease, but was significantly higher than in the pET group, and both RM and Tbc were no longer significantly different from baseline. These changes are summarized in Figure, which shows that the rET group showed less change during exercise than the pET group. These results suggest that differences in post-exercise cardiac function and vascular reactivity between the pET and rET groups may have contributed to the significant association between postexercise RM and %V̇O₂peak. However, because AIx was affected not only by RM, but also by Tbc, AIx changed more dynamically after exercise in the pET group, whereas this change was less pronounced in the rET group, suggesting that AIx after exercise is a stronger determinant of %V̇O₂peak. In addition, our study suggests that changes in AIx after exercise may have additional value in the risk stratification of patients with HFpEF.

Both AIx and RM focus on backward waves, and the correlation between them is strong (baseline: r=0.751, P<0.001; immediately after exercise: r=0.752, P<0.001; 5 min after exercise: r=0.836, P<0.001). As shown in Figure, CPP can approximate peak afterload. The CPP value is the sum of Pb and the BP at a point of descending Pb, but not the sum of Pb and Pf. Because RM is the ratio of BP of different time phases, the AIx (AP/CPP) may be a more accurate parameter for the backward wave component in the afterload. Increased arterial wave reflection and stiffness increase the systolic load on the heart, limit cardiac output during exercise, and may reduce peak V̇O₂. AIx and backward wave pressure have been reported to be higher in an HFpEF group than a control group after exercise.³⁶ Although the patients enrolled in the present study did not have a history of HF, non-invasively measured aortic stiffness during exercise could be important for categorizing the decline in exercise tolerance and the risk stage of HFpEF.

This study has several limitations. First, CBP measurement during exercise is needed to precisely assess the influence of exercise; however, it is difficult to accurately measure CBP during exercise because of the difficulty in sensing the pulse. Instead of CBP at peak exercise, we used CBP immediately after exercise. Second, various cardiovascular diseases, such as CAD, valvular disease, and cardiomyopathy, were included in this study; thus, the results should be interpreted carefully. Third, we could not study the influence of smoking, one of the most important coronary risk factors, because of missing data in the medical records. Fourth, we fixed the recovery phase to 5 min based on a previous study.⁴³ However, the ability to recovery could differ between patients and protocols. Because a recent study defined the appropriate recovery phase based on falling V̇O₂,⁴⁴ this could have affected the results. Fifth, in addition to LV ejection fraction, E/A, and E/e′, peak tricuspid annular plane systolic excursion/systolic pulmonary artery pressure (right ventricle-pulmonary artery coupling) by stress echocardiography and estimated aortic arch pulse wave velocity/global longitudinal strain (ventricular-arterial coupling) have been reported to be significantly associated with peak V̇O₂ for patients with early stages of HF and those with HFpEF, but we did not conduct stress echocardiography or analyze these parameters.^45,46 In the future, these parameters should be evaluated to detect patients at high risk of developing HF. Sixth, although physical activity was associated with exercise tolerance, we could not assess its influence on the relationship between exercise tolerance and aortic stiffness.⁴⁴

Conclusions

AIx measured before and after exercise was significantly associated with exercise tolerance. A more significant and continuous decrease in AIx after exercise was observed only in the pET group, suggesting that associations between changes in aortic stiffness owing to exercise and exercise tolerance may be used to risk stratify patients in HF Stages A and B.

Acknowledgment

The authors thank Editage (www.editage.com) for English language editing a draft of this paper.

Sources of Funding

This research did not receive any grants from funding agencies in the public, commercial, or not-for-profit sectors.

Disclosures

Y.S. is a member of Circulation Journal’s Editorial Team. The remaining authors have no conflicts of interest to disclose.

IRB Information

The Institutional Review Board of Nagoya City University Hospital approved this study (Study ID 60-18-205).

Data Availability

All analyzable data collected for this study will be shared upon reasonable request, for any purpose, to the corresponding author as Excel files via email for 5 years after publication.

Supplementary Files

Please find supplementary file(s);

https://doi.org/10.1253/circj.CJ-22-0772

References

• 1.   Okura Y, Ramadan MM, Ohno Y, Mitsuma W, Tanaka K, Ito M, et al. Impending epidemic: Future projection of heart failure in Japan to the year 2055. Circ J 2008; 72: 489–491.
• 2.   Virani SS, Alonso A, Aparicio HJ, Benjamin EJ, Bittencourt MS, Callaway CW, et al. Heart disease and stroke statistics – 2021 update: A report from the American Heart Association. Circulation 2021; 143: e254–e743, doi:10.1161/CIR.0000000000000950.
• 3.   Vasan RS, Larson MG, Benjamin EJ, Evans JC, Reiss CK, Levy D. Congestive heart failure in subjects with normal versus reduced left ventricular ejection fraction: Prevalence and mortality in a population-based cohort. J Am Coll Cardiol 1999; 33: 1948–1955.
• 4.   Shiba N, Nochioka K, Miura M, Kohno H, Shimokawa H; CHART-2 Investigators. Trend of Westernization of etiology and clinical characteristics of heart failure patients in Japan: First report from the CHART-2 study. Circ J 2011; 75: 823–833.
• 5.   Nagai T, Yoshikawa T, Saito Y, Takeishi Y, Yamamoto K, Ogawa H, et al. Clinical characteristics, management, and outcomes of Japanese patients hospitalized for heart failure with preserved ejection fraction: A report from the Japanese Heart Failure Syndrome With Preserved Ejection Fraction (JASPER) registry. Circ J 2018; 82: 1534–1545.
• 6.   Consensus Trial Study Group. Effects of enalapril on mortality in severe congestive heart failure: Results of the cooperative north Scandinavian enalapril survival study (CONSENSUS). N Engl J Med 1987; 316: 1429–1435.
• 7.   SOLVD Investigators; Yusuf S, Pitt B, Davis CE, Hood WB, Cohn JN. Effect of enalapril on mortality and the development of heart failure in asymptomatic patients with reduced left ventricular ejection fractions. N Engl J Med 1992; 327: 685–691.
• 8.   The Cardiac Insufficiency bisoprolol Study II (CIBIS-II): A randomised trial. Lancet 1999; 353: 9–13.
• 9.   Effect of metoprolol CR/XL in chronic heart failure: Metoprolol CR/XL randomised intervention trial in congestive heart failure (MERIT-HF). Lancet 1999; 353: 2001–2007.
• 10.   Packer M, Coats AJ, Fowler MB, Katus HA, Krum H, Mohacsi P, et al. Effect of carvedilol on survival in severe chronic heart failure. N Engl J Med 2001; 344: 1651–1658.
• 11.   Pitt B, Remme W, Zannad F, Neaton J, Martinez F, Roniker B, et al. Eplerenone, a selective aldosterone blocker, in patients with left ventricular dysfunction after myocardial infarction. N Engl J Med 2003; 348: 1309–1321.
• 12.   Zannad F, McMurray JJ, Krum H, van Veldhuisen DJ, Swedberg K, Shi H, et al. Eplerenone in patients with systolic heart failure and mild symptoms. N Engl J Med 2011; 364: 11–21.
• 13.   Anker SD, Butler J, Filippatos G, Ferreira JP, Bocchi E, Böhm M, et al. Empagliflozin in heart failure with a preserved ejection fraction. N Engl J Med 2021; 385: 1451–1461.
• 14.   Solomon SD, McMurray JJV, Claggett B, de Boer RA, DeMets D, Hernandez AF, et al. Dapagliflozin in heart failure with mildly reduced or preserved ejection fraction. N Engl J Med 2022; 387: 1089–1098.
• 15.   Writing Committee Members; 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. Circulation 2013; 128: e240–e327, doi:10.1161/CIR.0b013e31829e8776.
• 16.   Tsutsui H, Isobe M, Ito H, Ito H, Okumura K, Ono M, et al; Japanese Circulation Society and the Japanese Heart Failure Society Joint Working Group. JCS 2017/JHFS 2017 guideline on diagnosis and treatment of acute and chronic heart failure: Digest version. Circ J 2019; 83: 2084–2184.
• 17.   Messerli FH, Rimoldi SF, Bangalore S. The transition from hypertension to heart failure: Contemporary update. JACC Heart Fail 2017; 5: 543–551.
• 18.   Roman MJ, Devereux RB, Kizer JR, Okin PM, Lee ET, Wang W, et al. High central pulse pressure is independently associated with adverse cardiovascular outcome the strong heart study. J Am Coll Cardiol 2009; 54: 1730–1734.
• 19.   Roman MJ, Devereux RB, Kizer JR, Lee ET, Galloway JM, Ali T, et al. Central pressure more strongly relates to vascular disease and outcome than does brachial pressure: The Strong Heart Study. Hypertension 2007; 50: 197–203.
• 20.   McEniery CM, Cockcroft JR, Roman MJ, Franklin SS, Wilkinson IB. Central blood pressure: Current evidence and clinical importance. Eur Heart J 2014; 35: 1719–1725.
• 21.   Myers J, Gullestad L, Vagelos R, Do D, Bellin D, Ross H, et al. Cardiopulmonary exercise testing and prognosis in severe heart failure: 14 mL/kg/min revisited. Am Heart J 2000; 139: 78–84.
• 22.   Francis DP, Shamim W, Davies LC, Piepoli MF, Ponikowski P, Anker SD, et al. Cardiopulmonary exercise testing for prognosis in chronic heart failure: Continuous and independent prognostic value from VE/VCO(2)slope and peak VO(2). Eur Heart J 2000; 21: 154–161.
• 23.   Adachi H. Cardiopulmonary exercise test. Int Heart J 2017; 58: 654–665.
• 24.   Umemura S, Arima H, Arima S, Asayama K, Dohi Y, Hirooka Y, et al. The Japanese Society of Hypertension guidelines for the management of hypertension (JSH 2019). Hypertens Res 2019; 42: 1235–1481.
• 25.   Kinoshita M, Yokote K, Arai H, Iida M, Ishigaki Y, Ishibashi S, et al. Japan Atherosclerosis Society (JAS) guidelines for prevention of atherosclerotic cardiovascular diseases 2017. J Atheroscler Thromb 2018; 25: 846–984.
• 26.   Araki E, Goto A, Kondo T, Noda M, Noto H, Origasa H, et al. Japanese clinical practice guideline for diabetes 2019. Diabetol Int 2020; 11: 165–223.
• 27.   Weber T, Auer J, O’Rourke MF, Kvas E, Lassnig E, Berent R, et al. Arterial stiffness, wave reflections, and the risk of coronary artery disease. Circulation 2004; 109: 184–189.
• 28.   Chen TK, Knicely DH, Grams ME. Chronic kidney disease diagnosis and management: A review. JAMA 2019; 322: 1294–1304.
• 29.   Shoji T, Nakagomi A, Okada S, Ohno Y, Kobayashi Y. Invasive validation of a novel brachial cuff-based oscillometric device (SphygmoCor XCEL) for measuring central blood pressure. J Hypertens 2017; 35: 69–75.
• 30.   Woodman RJ, Kingwell BA, Beilin LJ, Hamilton SE, Dart AM, Watts GF. Assessment of central and peripheral arterial stiffness: Studies indicating the need to use a combination of techniques. Am J Hypertens 2005; 18: 249–260.
• 31.   Chester RC, Gornbein JA, Hundley WG, Srikanthan P, Watson KE, Horwich T. Reflection magnitude, a measure of arterial stiffness, predicts incident heart failure in men but not women: Multi-Ethnic Study of Atherosclerosis (MESA). J Card Fail 2017; 23: 353–362.
• 32.   Makita S, Yasu T, Akashi YJ, Adachi H, Izawa H, Ishihara S, et al. JCS/JACR 2021 guideline on rehabilitation in patients with cardiovascular disease. Circ J 2022; 87: 155–235.
• 33.   Standard levels of respiratory circulation parameters during exercise: 1990 report. Jpn Circ J 1992; 56(Suppl 5): 1514–1523 (in Japanese).
• 34.   Yamamoto J, Wakami K, Muto K, Kikuchi S, Goto T, Fukuta H, et al. Verification of echocardiographic assessment of left ventricular diastolic dysfunction in patients with preserved left ventricular ejection fraction using the American Society of Echocardiography and European Association of Cardiovascular Imaging. Circ Rep 2019; 1: 525–530.
• 35.   Fujiwara K, Shimada K, Nishitani-Yokoyama M, Kunimoto M, Matsubara T, Matsumori R, et al. Arterial stiffness index and exercise tolerance in patients undergoing cardiac rehabilitation. Int Heart J 2021; 62: 230–237.
• 36.   Reddy YNV, Andersen MJ, Obokata M, Koepp KE, Kane GC, Melenovsky V, et al. Arterial stiffening with exercise in patients with heart failure and preserved ejection fraction. J Am Coll Cardiol 2017; 70: 136–148.
• 37.   Alves AJ, Oliveira NL, Lopes S, Ruescas-Nicolau MA, Teixeira M, Oliveira J, et al. Arterial stiffness is related to impaired exercise capacity in patients with coronary artery disease and history of myocardial infarction. Heart Lung Circ 2019; 28: 1614–1621.
• 38.   Papaioannou TG, Vlachopoulos CV, Alexopoulos NA, Dima I, Pietri PG, Protogerou AD, et al. The effect of heart rate on wave reflections may be determined by the level of aortic stiffness: Clinical and technical implications. Am J Hypertens 2008; 21: 334–340.
• 39.   Weber T, Wassertheurer S, O’Rourke MF, Haiden A, Zweiker R, Rammer M, et al. Pulsatile hemodynamics in patients with exertional dyspnea: Potentially of value in the diagnostic evaluation of suspected heart failure with preserved ejection fraction. J Am Coll Cardiol 2013; 61: 1874–1883.
• 40.   Mitchell GF, Moyé LA, Braunwald E, Rouleau JL, Bernstein V, Geltman EM, et al. Sphygmomanometrically determined pulse pressure is a powerful independent predictor of recurrent events after myocardial infarction in patients with impaired left ventricular function. SAVE Investigators. Survival and ventricular enlargement. Circulation 1997; 96: 4254–4260.
• 41.   Torjesen AA, Sigurðsson S, Westenberg JJ, Gotal JD, Bell V, Aspelund T, et al. Pulse pressure relation to aortic and left ventricular structure in the Age, Gene/Environment Susceptibility (AGES)–Reykjavik Study. Hypertension 2014; 64: 756–761.
• 42.   Kingwell BA, Waddell TK, Medley TL, Cameron JD, Dart AM. Large artery stiffness predicts ischemic threshold in patients with coronary artery disease. J Am Coll Cardiol 2002; 40: 773–779.
• 43.   Nayor M, Shah RV, Miller PE, Blodgett JB, Tanguay M, Pico AR, et al. Metabolic architecture of acute exercise response in middle-aged adults in the community. Circulation 2020; 142: 1905–1924.
• 44.   Nayor M, Chernofsky A, Spartano NL, Tanguay M, Blodgett JB, Murthy VL, et al. Physical activity and fitness in the community: The Framingham Heart Study. Eur Heart J 2021; 42: 4565–4575.
• 45.   Pugliese NR, De Biase N, Conte L, Gargani L, Mazzola M, Fabiani I, et al. Cardiac reserve and exercise capacity: Insights from combined cardiopulmonary and exercise echocardiography stress testing. J Am Soc Echocardiogr 2021; 34: 38–50.
• 46.   Pugliese NR, Balletti A, Armenia S, De Biase N, Faita F, Mengozzi A, et al. Ventricular-arterial coupling derived from proximal aortic stiffness and aerobic capacity across the heart failure spectrum. JACC Cardiovasc Imaging 2022; 15: 1545–1559.