Cardiovascular responses to static handgrip exercise and postexercise ischemia in heart failure with preserved ejection fraction

Abstract

Heart failure with preserved ejection fraction (HFpEF) is characterized by reduced ability to sustain physical activity that may be due partly to disease-related changes in autonomic function that contribute to dysregulated cardiovascular control during muscular contraction. Thus, we used a combination of static handgrip exercise (HG) and postexercise ischemia (PEI) to examine the pressor response to exercise and isolate the skeletal muscle metaboreflex, respectively. Mean arterial pressure (MAP), heart rate (HR), cardiac output (CO), and total peripheral resistance (TPR) were assessed during 2-min of static HG at 30 and 40% of maximum voluntary contraction (MVC) and subsequent PEI in 16 patients with HFpEF and 17 healthy, similarly aged controls. Changes in MAP were lower in patients with HFpEF compared with controls during both 30%MVC (Δ11 ± 7 vs. Δ15 ± 8 mmHg) and 40%MVC (Δ19 ± 14 vs. Δ30 ± 8 mmHg), and a similar pattern of response was evident during PEI (30%MVC: Δ8 ± 5 vs. Δ12 ± 8 mmHg; 40%MVC: Δ13 ± 10 vs. Δ18 ± 9 mmHg) (group effect: P = 0.078 and P = 0.017 at 30% and 40% MVC, respectively). Changes in HR, CO, and TPR did not differ between groups during HG or PEI (P > 0.05). Taken together, these data suggest a reduced pressor response to static muscle contractions in patients with HFpEF compared with similarly aged controls that may be mediated partly by a blunted muscle metaboreflex. These findings support a disease-related dysregulation in neural cardiovascular control that may reduce an ability to sustain physical activity in HFpEF.

NEW & NOTEWORTHY The current investigation has identified a diminution in the exercise-induced rise in arterial blood pressure (BP) that persisted during postexercise ischemia (PEI) in an intensity-dependent manner in patients with heart failure with preserved ejection fraction (HFpEF) compared with older, healthy controls. These findings suggest that the pressor response to exercise is reduced in patients with HFpEF, and this deficit may be mediated, in part, by a blunted muscle metaboreflex, highlighting the consequences of impaired neural cardiovascular control during exercise in this patient group.

INTRODUCTION

Heart failure with preserved ejection fraction (HFpEF) accounts for more than 50% of all heart failure cases, yet current pharmacotherapy has remained unsuccessful at improving survival or quality of life in this patient group (1). A hallmark symptom of HFpEF is severe exercise intolerance, as documented by reductions in six-minute walk test distance (2) and peak oxygen uptake (3), as well as challenges associated with activities of daily living (i.e., difficulty carrying and lifting objects) (4, 5) despite being optimized on guideline-directed pharmacotherapy. Given the link between exercise intolerance and poor prognosis (2, 6), identification of factors that underlie exercise intolerance in patients with HFpEF represents a crucial step toward improved patient care.

Although the importance of both central (3, 7, 8) and peripheral factors (9–11) to exercise intolerance in patients with HFpEF are now recognized, disease-related changes in reflex cardiovascular control during exercise remains poorly understood in this patient group. One such reflex of particular relevance to heart failure pathophysiology is the exercise pressor reflex, which is broadly defined as blood pressure (BP) elevations during exercise induced by signals arising from contracting skeletal muscle (12). One of the most important contributors to the exercise pressor reflex is the muscle metaboreflex, mediated by metabolically sensitive group IV afferent fibers (metaboreceptors) originating in skeletal muscle, which increases efferent sympathetic nervous system activity and BP in an effort to augment perfusion of the exercising skeletal muscle (13–15). During static handgrip exercise (HG), the metaboreflex is primarily responsible for increases in arterial BP, heart rate (HR), cardiac output (CO), and total peripheral vascular resistance in an intensity-dependent manner (16–18). Muscle metaboreflex activation during a period of postexercise ischemia (PEI) following static HG has been shown to be augmented in individuals with type 2 diabetes (19) and treated hypertension (20), but blunted in individuals with obesity (21), never-treated hypertension (22), and obstructive sleep apnea (23), all of which are common comorbidities of HFpEF (24). Given that patients with HFpEF present with a constellation of these comorbidities, it is reasonable to suggest muscle metaboreflex activation may also be altered in this patient group.

To date, few studies have sought to evaluate muscle metaboreflex activation during PEI in patients with HFpEF, and results have been equivocal, with evidence for both blunted (25) and preserved (26, 27) responses. The discrepancy in findings may stem, in part, from differences in exercise modality, as these previous studies have utilized rhythmic HG (26) or static HG performed either for a short-period of time (25) or until exhaustion (27). Importantly, no studies to date have evaluated reflex responses across multiple contraction intensities, which is an important consideration given the evidence of an intensity-dependent effect on metaboreflex activation that is present in other patient populations (20, 28). These differing study designs utilized in recent studies in HFpEF, coupled with a current lack of knowledge regarding responses across varying levels of exercise, leave uncertainty regarding whether disease-related alterations in the pressor response to exercise and muscle metaboreflex activation are present in patients with HFpEF.

With this background in mind, the present study sought to evaluate potential alterations in the pressor response to exercise and isolated muscle metaboreflex activation during a period of PEI in patients with HFpEF compared with healthy, similarly aged controls across multiple exercise intensities. We hypothesized that the cardiovascular responses to static HG exercise at 30% and 40% of maximum voluntary contraction (MVC) would be reduced in patients with HFpEF, and that this deficit would be mediated, in part, by a blunted muscle metaboreflex activation.

METHODS

Ethical Approval

All experimental procedures and protocols were approved by the University of Utah, the Salt Lake City Veterans Affairs, and the University of Texas at Arlington Institutional Review Boards. All aspects of the study conformed to the standards set by the Declaration of Helsinki, except for registration in a database. All experimental procedures were explained in writing and verbally, and written informed consent was obtained from all participants before study participation.

Subjects

Sixteen patients with NYHA Class II-III HFpEF and 17 similarly aged controls control participants volunteered to participate in the study. Patients with HFpEF were recruited either by word of mouth or in the heart failure clinics at the University of Utah Health Sciences Center and the Salt Lake City Veterans Affairs Medical Center (n = 14), as well as at the University of Texas at Arlington (n = 2). Control participants were recruited through the Utah Vascular Research Laboratory at the University of Utah (n = 9) and the Human Neural Cardiovascular Control Laboratory at the University of Texas at Arlington (UTA) (n = 8). Inclusion criteria for patients with HFpEF were based on a clinical diagnosis of HFpEF from a cardiologist, which includes 1) the presence of ≥1 symptom (i.e., paroxysmal nocturnal dyspnea, orthopnea, and dyspnea on exertion) and one sign (i.e., edema and elevation in jugular venous distention) in the previous 12 mo; 2) left ventricular ejection fraction ≥45%; 3) controlled hypertension; and 4) either hospitalization in the previous 12 mo for which heart failure was a major component or an elevated B-type natriuretic peptide (≥100 pg/mL) in the previous 60 days. Exclusion criteria included pacemaker, significant valvular heart disease, atrial fibrillation, end-stage renal disease, type I diabetes, or any orthopedic limitations that would prevent performance of static HG exercise. With recognition of the high prevalence of obesity that characterizes the HFpEF phenotype (29), we enrolled patients with a body mass index (BMI) between 25 and 45 kg/m². All patients were on standard heart failure pharmacotherapy, and no prescribed medications were withheld apart from diuretics, which were discontinued 12 h before the experimental visit. Control participants were free from overt cardiovascular disease, as indicated by a medical health history questionnaire. All participants were nonsmokers, and all female participants were postmenopausal and not currently on hormonal replacement therapy. In addition, all participants self-reported as White.

Cardiovascular Measurements

Beat-to-beat heart rate (HR) and arterial blood pressure (BP) were continuously measured at a sampling rate of 1,000 Hz using an electrocardiogram [Biopac Systems, Goleta, CA (Utah) and Q710, Quintin, Bothell, WA (UTA)] and finger photoplethysmography (Finometer, Finapres Medical Systems, Amsterdam, The Netherlands), respectively. For Finometer measurements, return-to-flow calibrations were performed and physiocal turned off during recording periods. An automated sphygmomanometer [Tango M2, SunTech Medical, Inc., Morrisville, NC (Utah) and Welch Allyn, Skaneateles Falls, NY (UTA)] was used to record resting brachial artery BP on the right arm to validate BP measurements from the Finometer. The changes in BP measured using the Finometer have been demonstrated to provide an accurate estimate of directly measured intra-arterial BP (30, 31). Beat-to-beat pressure waveforms were also used to derive estimates of stroke volume (SV) using Modelflow software, which incorporates age, sex, weight, and height (32). Cardiac output (CO) and total peripheral resistance (TPR) were calculated as a product of SV and HR and as a ratio of mean arterial pressure (MAP) and CO, respectively. SV, CO, and TPR were then indexed to body surface area to derive cardiac stroke volume index (SVI), cardiac index (CI), and total peripheral resistance index (TPRI), respectively. Respiratory movements were monitored using a strain-gauge pneumograph placed around the abdomen (Biopac Systems, Goleta and Pneumotrace, UFI, Morro Bay, CA) to ensure that participants did not inadvertently perform Valsalva maneuvers during HG exercise and PEI. Data collected at both study sites were comparable.

Experimental Protocol

All experiments were performed with the participants in the supine position in a quiet, dimly lit room at an ambient temperature of 22°C–24°C. Participants arrived at the laboratory following an overnight fast and were instructed to avoid caffeine for 12 h and strenuous physical activity and alcohol for 24 h. Prior to the performance of the experimental protocol, each participant was familiarized with all measurements, the equipment, and testing procedures. For HG exercise, participants held a HG dynamometer [TSD 121C, Biopac Systems, Goleta, CA (Utah) and Model 76618, Lafayette Instrument, Lafayette, LA (UTA)] in the right hand with the limb supported on an adjustable table at heart level. Maximum voluntary contraction (MVC) was determined as the highest of three to five maximal efforts each separated by 1 min and was used to calculate relative work rates of 30% and 40% MVC for the experimental protocol. Then, after a quiet rest for 10–15 min, a 5-min baseline recording was performed to determine resting cardiovascular variables. Participants then performed 2 min of static HG exercise at 30% (moderate intensity) followed by 2 min of forearm PEI to isolate muscle metaboreflex activation. PEI was achieved by inflation of a pneumatic blood pressure cuff (D.E. Hokanson, Inc., Bellevue, WA) around the upper arm to suprasystolic pressure (>250 mmHg) 5 s before the end of exercise. Visual feedback regarding the HG force exerted was provided on a computer screen displayed at eye level. In all cases, the 30%MVC trial was performed first followed by 40%MVC (high intensity) and PEI. The HG trials were separated by at least 10 min to allow cardiovascular variables to return to baseline values. Data analyses were performed with investigators blinded to groups.

Data and Statistical Analyses

Resting measurements were averaged over 5 min at baseline and averaged in 30-s bins during HG exercise and PEI. The change values were calculated as the difference between the averaged resting value and the average value during the last 30 s of the HG exercise or PEI condition. Statistics were performed using SigmaPlot 11 (Systat Software Inc., Point Richmond, CA). Normality was confirmed using the Shapiro–Wilk test. Student’s unpaired t tests were used to identify group differences between patients with HFpEF and control participants at baseline. All outcome variables were assessed with two-way ANOVA with repeated measures in which group (HFpEF vs. control) and time (rest vs. 30-s bins over time as well as change values from baseline during HG and PEI) were performed to determine responses during each exercise intensity. We also performed subset analyses of the pressor responses during HG and PEI in patients with HFpEF and control participants who had similar MVC (n = 9/group, 5 females in each group). In case of a significant interaction, the Tukey method was used for α adjustment and post hoc analysis. Given a direct association between grip strength and the pressor response to static HG exercise (33), correlational analyses were performed to determine the relationship between MVC and changes in MAP during the last 30 s of exercise at both exercise intensities. Data are presented as means ± standard deviation (SD). Statistical significance was established at P < 0.05.

RESULTS

Participant Characteristics

Baseline participant characteristics are provided in Table 1. Participants were similar in age and height, while bodyweight, body mass index, and body surface area were greater and MVC was lower in patients with HFpEF than in control participants (P < 0.05). Disease-specific characteristics and medications for patients with HFpEF are provided in Table 2. Cardiac morphology and function measures, and the presence of comorbidities were consistent with the HFpEF phenotype.

Table 1.

Participant characteristics

	HFpEF	Control	P Value
n (male/female)	16 (9/7)	17 (8/9)
Age, yr	69 ± 9	67 ± 8	0.513
Height, cm	168.9 ± 9.0	169.0 ± 8.7	0.979
Weight, kg	95.2 ± 20.2	74.8 ± 13.5	0.002
Body mass index, kg/m2	33.3 ± 6.2	26.1 ± 3.6	<0.001
Body surface area, m2	2.04 ± 0.24	1.85 ± 0.19	0.014
MVC, kg	19 ± 6	27 ± 13	0.033

Data are means ± SD. HFpEF, heart failure with preserved ejection fraction; MVC, maximum voluntary contraction. Student’s unpaired t test was performed.

Table 2.

Patient disease-specific characteristics and medications

	HFpEF
Echocardiogram
Ejection fraction, %	60 ± 7
LV IVSD, cm	1.1 ± 0.2
LV PWD, cm	1.0 ± 0.2
LV ID diastole, cm	4.8 ± 0.7
LV ID systole, cm	3.4 ± 0.7
Peak E wave, m/s	0.85 ± 0.18
Peak A wave, m/s	0.83 ± 0.30
E/A ratio	1.13 ± 0.35
Medial E/E′ ratio	7.31 ± 7.58
Lateral E/E′ ratio	7.84 ± 6.06
Mitral E wave deceleration time, ms	213 ± 56
NYHA Functional Classification
NYHA class II	12
NYHA class III	4
Comorbidities
Type 2 diabetes	3
Coronary artery disease	5
Obstructive sleep apnea	7
Chronic obstructive pulmonary disease	3
Chronic kidney disease	1
Medications
β-Blocker	4
Angiotensin-converting enzyme inhibitor	2
Angiotensin receptor inhibitor	4
Statin	6
Diuretic	14
Aldosterone antagonist	7
Anticoagulant	7
Aspirin	6
Sodium-glucose cotransportor-2 inhibitor	0

Data are means ± SD or number of cases (n = 16). A wave, peak velocity of late transmitral flow; E wave, peak velocity of early diastolic transmitral flow; E′, peak velocity of early diastolic mitral annular motion; HFpEF, heart failure with preserved ejection fraction; ID, internal dimension; IVSD, interventricular septum thickness at end-diastole; LV, left ventricle; NYHA, New York heart association; PWD, posterior wall thickness.

Static Handgrip Exercise and Postexercise Ischemia

Cardiovascular responses at rest and during the last 30 s of exercise and PEI for both 30% and 40% MVC trials are provided in Tables 3 and 4. For both trials, averaged (30-s) responses for MAP are provided in Fig. 1, while averaged responses for CO and TPR are provided in Fig. 2. Resting cardiovascular variables were not different between groups (P > 0.05), except for a slightly higher resting SVI in patients with HFpEF at the 30%MVC trial (P < 0.05) (Table 3).

Table 3.

Cardiovascular responses at rest and during static handgrip exercise (last 30 s) and postexercise ischemia (last 30 s) for the 30%MVC trial in patients with heart failure with preserved ejection fraction and similarly aged control participants

				P Value
	Rest	HG	PEI	Group	Time	Interaction
MAP, mmHg
HFpEF	86 ± 6	97 ± 11	94 ± 7	0.080	<0.001	0.154
Control	89 ± 9	104 ± 12	101 ± 9
SBP, mmHg
HFpEF	124 ± 10	140 ± 14	138 ± 15	0.010	<0.001	0.193
Control	133 ± 13	155 ± 19	153 ± 16
DBP, mmHg
HFpEF	68 ± 9	76 ± 14	72 ± 9	0.661	<0.001	0.220
Control	67 ± 12	78 ± 13	75 ± 11
HR, beats/min
HFpEF	62 ± 14	70 ± 16	65 ± 16	0.567	<0.001	0.953
Control	60 ± 12	67 ± 13	62 ± 11
SV, mL
HFpEF	88 ± 21	88 ± 24	90 ± 23	0.759	0.945	0.577
Control	92 ± 22	91 ± 26	90 ± 19
CO, L/min
HFpEF	5.4 ± 1.6	6.0 ± 1.9	5.8 ± 1.9	0.929	0.001	0.515
Control	5.4 ± 1.6	6.1 ± 2.0	5.5 ± 1.3
TPR, mmHg/L/min
HFpEF	17.5 ± 5.8	18.0 ± 7.2	17.9 ± 6.2	0.713	0.212	0.626
Control	17.7 ± 5.3	18.8 ± 7.5	19.3 ± 5.2
SVI, mL/m2
HFpEF	43 ± 8†	42 ± 9	44 ± 9	0.054	0.868	0.610
Control	49 ± 9	49 ± 11	49 ± 9
CI, L/min/m2
HFpEF	2.6 ± 0.7	2.9 ± 0.7	2.8 ± 0.7	0.246	0.001	0.519
Control	2.9 ± 0.7	3.3 ± 0.8	3.0 ± 0.6
TPRI, mmHg/L/min/m2
HFpEF	8.9 ± 4.0	9.2 ± 4.6	9.1 ± 4.0	0.141	0.846	0.948
Control	9.8 ± 3.4	10.4 ± 4.5	10.6 ± 3.4

Data are means ± SD; n = 16 patients with HFpEF and 17 control participants. A two-way ANOVA with repeated measures in which group (HFpEF vs. control) and time (rest vs. time) was performed. CI, cardiac index; CO, cardiac output; DBP, diastolic blood pressure; HFpEF, heart failure with preserved ejection fraction; HG, handgrip exercise; HR, heart failure; MAP, mean arterial pressure; MVC, maximum voluntary contraction; PEI, postexercise ischemia; SBP, systolic blood pressure; SV, stroke volume; SVI, stroke volume index; TPR, total peripheral resistance; TPRI, total peripheral resistance index. †P < 0.05 vs. control participants.

Table 4.

Cardiovascular responses at rest and during static handgrip exercise (last 30 s) and postexercise ischemia (last 30 s) for the 40%MVC trial in patients with heart failure with preserved ejection fraction and similarly aged control participants

				P Value
	Rest	HG	PEI	Group	Time	Interaction
MAP, mmHg
HFpEF	88 ± 8	107 ± 14*†	101 ± 10*	0.025	<0.001	0.011
Control	91 ± 10	120 ± 13*	109 ± 13*
SBP, mmHg
HFpEF	126 ± 11	157 ± 21	148 ± 28	0.019	<0.001	0.114
Control	135 ± 17	177 ± 26	165 ± 23
DBP, mmHg
HFpEF	69 ± 11	82 ± 17*	77 ± 10*	0.231	<0.001	0.004
Control	68 ± 11	92 ± 10*	81 ± 12*#
HR, beats/min
HFpEF	62 ± 14	70 ± 17*	64 ± 17	0.636	<0.001	0.016
Control	60 ± 13	77 ± 13*	66 ± 11#
SV, mL
HFpEF	90 ± 24	83 ± 29	90 ± 29	0.927	<0.001	0.513
Control	91 ± 23	81 ± 26	93 ± 27
CO, L/min
HFpEF	5.5 ± 1.7	5.7 ± 2.0	5.8 ± 2.2	0.652	0.023	0.320
Control	5.5 ± 1.8	6.3 ± 2.5	6.2 ± 2.1
TPR, mmHg/L/min
HFpEF	17.5 ± 6.5	22.1 ± 12.0	21.1 ± 11.7	0.858	0.001	0.422
Control	18.3 ± 6.8	21.9 ± 8.7	19.0 ± 4.7
SVI, mL/m2
HFpEF	44 ± 9	41 ± 13	44 ± 12	0.183	<0.001	0.310
Control	49 ± 10	43 ± 10	50 ± 10
CI, L/min/m2
HFpEF	2.7 ± 0.6	2.8 ± 0.9	2.8 ± 0.9	0.120	0.021	0.306
Control	3.0 ± 0.8	3.4 ± 1.0	3.3 ± 0.9
TPRI, mmHg/L/min/m2
HFpEF	8.9 ± 4.2	11.3 ± 7.4	10.9 ± 7.4	0.597	0.255	0.838
Control	10.1 ± 4.3	12.3 ± 5.6	10.5 ± 3.2

Data are means ± SD; n = 16 patients with HFpEF and 17 control participants. A two-way ANOVA with repeated measures in which group (HFpEF vs. control) and time (rest vs. time) was performed. CI, cardiac index; CO, cardiac output; DBP, diastolic blood pressure; HFpEF, heart failure with preserved ejection fraction; HG, handgrip exercise; HR, heart failure; MAP, mean arterial pressure; MVC, maximum voluntary contraction; PEI, postexercise ischemia; SBP, systolic blood pressure; SV, stroke volume; SVI, stroke volume index; TPR, total peripheral resistance; TPRI, total peripheral resistance index. *P < 0.05 vs. rest; #P < 0.05 vs. HG; †P < 0.05 vs. control participants.

Figure 1.

Thirty-second bins (A and C) of mean arterial pressure (MAP) and its absolute change values (B and D) during static handgrip exercise (HG) at 30% and 40% of maximum voluntary contraction (MVC) and postexercise ischemia (PEI) in patients with heart failure with preserved ejection fraction (HFpEF; n = 16; black circles) and similarly aged control participants (control; n = 17; open circles). Data are means ± SD. A two-way ANOVA with repeated measures in which group (HFpEF vs. control) and time (rest vs. 30-second bins over time as well as change values from baseline during HG and PEI) was performed. *P < 0.05 vs. rest. †P < 0.05 vs. control participants.

Figure 2.

Thirty-second bins of cardiac output (CO; A and C) and total peripheral resistance (TPR; B and D) during static handgrip exercise (HG) at 30% and 40% of maximum voluntary contraction (MVC) and postexercise ischemia (PEI) in patients with heart failure with preserved ejection fraction (HFpEF; n = 16; black circles) and similarly aged control participants (control; n = 17; open circles). Data are means ± SD. A two-way ANOVA with repeated measures in which group (HFpEF vs. control) and time (rest vs. 30-s bins over time) was performed.

During the 30%MVC trial, MAP was increased from rest in both groups (time effect: P < 0.05), as expressed in absolute values (Table 3 and Fig. 1A) and change values (HFpEF: Δ11 ± 7 mmHg for HG and Δ8 ± 5 mmHg for PEI; Control: Δ15 ± 8 mmHg for HG and Δ12 ± 8 mmHg for PEI) (Fig. 1B) but was overall qualitatively lower in patients with HFpEF (group effect: P = 0.048 for Fig. 1A and P = 0.078 for Fig. 1B). Similarly, systolic BP (SBP) and diastolic BP (DBP) increased from rest in both groups (time effect: P < 0.05), but SBP was overall lower in patients with HFpEF (group effect: P = 0.010) (Table 3). CO and CI increased from rest in both groups (time effect: P < 0.05), as expressed in absolute values (Table 3 and Fig. 2A) or changes from rest (HFpEF: Δ1.1 ± 1.8 L/min or Δ0.3 ± 0.5 L/min/m² for HG and Δ0.5 ± 0.8 L/min or Δ0.2 ± 0.4 L/min/m² for PEI; Control: Δ0.7 ± 0.9 L/min or Δ0.3 ± 0.5 L/min/m² for HG and Δ0.1 ± 1.0 L/min or Δ0.1 ± 0.5 L/min/m² for PEI), with no group differences (group effect: P < 0.05). HR also increased from rest in both groups (time effect: P < 0.001) (Table 3), whereas SV, SVI, TPR, and TPRI remained unchanged (time effect: P > 0.05), and these were all not different between groups (group effect: P > 0.05), as expressed in absolute values (Table 3 and Fig. 2B) or changes from rest for both TPR and TPRI (HFpEF: Δ0.5 ± 3.3 mmHg/L/min or Δ0.3 ± 1.7 mmHg/L/min/m² for HG and Δ0.4 ± 2.2 mmHg/L/min or Δ0.2 ± 1.2 mmHg/L/min/m² for PEI; Control: Δ1.1 ± 4.9 mmHg/L/min or Δ0.6 ± 2.8 mmHg/L/min/m² for HG and Δ0.6 ± 3.0 mmHg/L/min or Δ0.8 ± 1.6 mmHg/L/min/m² for PEI).

During the 40%MVC trial, MAP increased from rest in both groups and was overall lower in patients with HFpEF compared with control participants (group effect: P < 0.05), as expressed in absolute values (Table 4 and Fig. 1C) and change values (HFpEF: Δ19 ± 14 mmHg for HG and Δ13 ± 10 mmHg for PEI; Control: Δ30 ± 8 mmHg for HG and Δ18 ± 9 mmHg for PEI) (Fig. 1D). SBP and DBP also increased from rest in both groups (time effect: P < 0.05), but SBP was overall lower in patients with HFpEF (group effect: P = 0.019) (Table 4). SV, SVI, and HR increased during exercise in both groups, and HR remained elevated during PEI only in the control participants (time effect: P < 0.05) (Table 4). CO and CI increased from rest in both groups (time effect: P < 0.05), as expressed in absolute values (Table 4 and Fig. 2C) or changes from rest (HFpEF: Δ0.2 ± 1.3 L/min or Δ0.1 ± 0.7 L/min/m² for HG and Δ0.3 ± 1.1 L/min or Δ0.1 ± 0.5 L/min/m² for PEI; Control: Δ0.8 ± 1.3 L/min or Δ0.4 ± 0.6 L/min/m² for HG and Δ0.6 ± 1.0 L/min or Δ0.3 ± 0.5 L/min/m² for PEI), with no group differences (group effect: P > 0.05). TPR increased similarly from rest in both groups (time effect: P < 0.05), as expressed in absolute values (Table 4 and Fig. 2D) or changes from rest (HFpEF: Δ4.6 ± 9.4 mmHg/L/min for HG and Δ3.6 ± 7.7 mmHg/L/min for PEI; Control: Δ3.7 ± 4.9 mmHg/L/min for HG and Δ0.8 ± 3.5 mmHg/L/min for PEI). Conversely, when accounting for body surface area, TPRI did not change from rest when expressed in absolute values (time effect: P = 0.255) (Table 4), but increased in both groups when expressed as change values from rest (HFpEF: Δ2.4 ± 5.1 mmHg/L/min/m² for HG and Δ1.9 ± 4.3 mmHg/L/min/m² for PEI; Control: Δ2.1 ± 2.9 mmHg/L/min/m² for HG and Δ0.4 ± 2.1 mmHg/L/min/m² for PEI) (time effect: P = 0.025).

In addition, when the exercise-induced increase in MAP was compared between patients and controls with a similar MVC (see participant characteristics in Supplemental Tables S1 and S2), patients with HFpEF still exhibited blunted exercise pressor responses compared with control participants during static HG exercise performed at both 30% and 40%MVC (HFpEF: Δ11 ± 6 mmHg and Δ16 ± 15 mmHg; Control: Δ19 ± 7 mmHg and Δ35 ± 7 mmHg; group effect: P < 0.05), and this remained lower in patients with HFpEF during PEI following both exercise intensities (HFpEF: Δ7 ± 4 mmHg and Δ9 ± 8 mmHg Control: Δ14 ± 7 mmHg and Δ19 ± 9 mmHg; group effect: P < 0.05).

Correlational Analyses for Grip Strength

The MAP responses during the last 30 s of exercise at 30%MVC were not correlated with MVC in patients with HFpEF (r = 0.19, P = 0.474), control participants (r = −0.11, P = 0.680), or across the whole cohort (r = 0.07, P = 0.694). Likewise, the MAP responses during the last 30 s of exercise at 40%MVC were not correlated with MVC in patients with HFpEF (r = 0.35, P = 0.178), control participants (r = −0.36, P = 0.157), or across the whole cohort (r = 0.14, P = 0.455).

DISCUSSION

In the current investigation, we found that, during static HG, the increase in arterial blood pressure was diminished in an exercise intensity-dependent manner in patients with HFpEF compared with healthy, similarly aged controls. This blunted pressor response persisted when PEI was performed to isolate muscle metaboreflex activation, suggesting a disease-related dysfunction of the exercise pressor reflex in patients with HFpEF. Notably, changes in CO and TPR were similar between patients with HFpEF and controls, precluding a delineation of the contributions from central and peripheral hemodynamics. These data indicate that the pressor response to static muscle contractions is impaired in patients with HFpEF, particularly at a higher exercise intensity, and this impaired response may be due, in part, to a blunted muscle metaboreflex activation. The intensity-dependent nature of these responses highlights the importance of assessing neural cardiovascular control across multiple exercise intensities to identify disease-related changes that may not be detected at lower levels of metaboreflex activation. Taken together, these findings provide novel evidence supporting the presence of dysregulation in neural cardiovascular control during static muscle contractions in HFpEF that may contribute to exercise intolerance in this patient group.

Cardiovascular Responses during Handgrip Exercise

The current investigation has identified a marked diminution in the pressor response to static HG in patients with HFpEF compared with healthy, similarly aged controls (Fig. 1). The blunted exercise pressor response in patients with HFpEF was unrelated to potential group differences in muscular strength given the lack of correlation between exercise-induced increase in arterial blood pressure and MVC. Importantly, the blunted exercise pressor response was also evident when a subset of patients with HFpEF and control participants who had similar grip strength were investigated. These new findings build upon recent work from our group and others examining the regulation of arterial blood pressure during exercise in patients with HFpEF. Using a more dynamic (static intermittent) HG exercise model, our group recently identified similar increases in arterial blood pressure across a range of exercise intensities between patients with HFpEF and hypertensive controls (34) and also observed a similar pressor response during isolated lower limb (knee-extensor) exercise (35) compared with similarly aged controls. Although the reasons for a lack of difference in blood pressure responses during dynamic exercise in these previous studies is not immediately apparent, it is likely that the rise in intramuscular pressure in the exercising limb was insufficient to impede muscle perfusion to a degree that fully engages the muscle metaboreflex (36). Nevertheless, these studies combined demonstrate that the pressor response to exercise is clearly not exaggerated in patients with HFpEF when examined using a more dynamic exercise modality. In addition, there is evidence to suggest that static muscular contraction elicits a more pronounced pressor response compared with that of dynamic muscular contraction despite comparable exercise intensities and duration (37, 38). Thus, these findings suggest that factors contributing to the pressor response to dynamic versus static muscular contraction may be different while also highlighting the importance of considering exercise modality (i.e., dynamic vs. static exercise) in data interpretation.

Specific to the static HG exercise modality, although Sarma et al. (27) recently failed to identify group differences in MAP at the end of 40% MVC HG exercise performed until exhaustion the peak changes in MAP were qualitatively lower in patients with HFpEF (ΔMAP: ∼54 mmHg) compared with similarly aged controls (ΔMAP: ∼81 mmHg). Using a protocol more similar to the present study, Moriwaki et al. (25) identified a similar pressor response to 30% MVC static HG exercise between patients with heart failure with reduced ejection fraction (HFrEF) and HFpEF, though group differences in grip strength and age, two factors known to influence the pressor response to static HG exercise (33, 39, 40), may have confounded the results of this previous study. The present study therefore builds upon this important earlier work, providing new evidence for an attenuation in the exercise-induced rise in arterial blood pressure, which was exercise intensity-dependent and independent of handgrip strength, in patients with HFpEF compared with healthy, similarly aged controls that adds to a nascent body of literature exploring neural cardiovascular control during exercise in this patient group.

The increase in arterial blood pressure during static HG exercise is achieved through the combined effect of changes in cardiac output and total peripheral resistance. Interestingly, while the changes in cardiac output and total peripheral resistance were not significantly different between groups in the present study, possibly due partly to high variability in these data, it appears that the increases in blood pressure were achieved differently in patients with HFpEF and similarly aged, healthy controls during two exercise intensities. During the 30%MVC trial, the changes in cardiac output and total peripheral resistance appeared qualitatively higher and lower, respectively, in patients with HFpEF compared with similarly aged, healthy controls, which were evident during both static HG exercise and PEI. This observation suggests that an augmented rise in cardiac output was insufficient to offset a smaller rise in total peripheral resistance in patients with HFpEF during moderate-intensity static HG exercise. Interestingly, during the 40%MVC trial, the changes in cardiac output and total peripheral resistance appeared qualitatively lower and higher, respectively, in patients with HFpEF compared with similarly aged, healthy controls, which were evident during both static HG exercise and PEI. This observation suggests that an attenuated rise in cardiac output was not sufficiently compensated by an augmented increase in total peripheral resistance in patients with HFpEF during high-intensity static HG exercise. Conversely, Moriwaki et al. (25) also did not observe any differences in cardiac output or total peripheral resistance responses in patients with HFpEF versus HFrEF, whereas Roberto et al. (26) and Sarma et al. (27) did not examine hemodynamic determinants of blood pressure. Taken together, while our initial observations highlight the potential complexity in the interactive influence of these hemodynamic variables that determine the BP responses during increasing exercise intensities in patients with HFpEF, there still remains uncertainty concerning this aspect of physiology, thus warranting additional studies in this patient group.

Cardiovascular Responses during Postexercise Ischemia

Although the overall cardiovascular response to static HG exercise involves a complex autonomic response that includes input from central command, the exercise pressor reflex, and the arterial baroreflex, the skeletal muscle metaboreflex may be isolated experimentally through the PEI technique. Pioneered by the seminal studies of Alam and Smirk (41), circulatory arrest of the exercising limb at the cessation of exercise prevents removal of metabolites produced during muscle contraction and, therefore, provides an opportunity to evaluate the metabolic component of the exercise pressor reflex without input from central command or the muscle mechanoreflex. The expectation for potential disease-related changes in the muscle metaboreflex in the present study stems from the so-called “muscle hypothesis” of heart failure (42, 43), a theory proposing that abnormalities in sensory reflex activity in skeletal muscle may contribute, in part, to the exercise limitations in patients with heart failure. Thus, developing a better understanding of this aspect of HFpEF pathophysiology may be particularly relevant to the characterization of exercise intolerance in this patient group.

The overall pattern of the arterial blood pressure response during PEI was consistent with that observed during exercise (Fig. 1), demonstrating a blunted pressor response in patients with HFpEF compared with healthy, similarly aged controls following both 30% (Fig. 1, A and B) and 40%MVC (Fig. 1, C and D). The blunted pressor response to PEI was also evident in subset analyses that compared patients with HFpEF and control participants who had similar grip strength. Although our finding of a blunted muscle metaboreflex in patients with HFpEF differs from that of recent studies that also examined this aspect of physiology in patients with HFpEF (25–27), there are significant differences in methodology between the former and present studies that preclude a direct comparison. Nevertheless, through inclusion of multiple exercise intensities, a more conventional experimental paradigm, and comparison of patients with HFpEF to healthy, similarly aged controls, the present study extends these previous findings, reporting for the first time that muscle metaboreflex is blunted in this patient population.

Mechanisms of Cardiovascular Responses during Exercise and PEI in HFpEF

Identifying the mechanisms underlying the blunted pressor responses during exercise and PEI in patients with HFpEF, which may include disease-related changes in afferent, central, or efferent signaling components of the reflex arc, is well beyond the scope of the present study. However, it is worth noting that Weiss et al. (44) recently failed to identify metabolic differences, assessed via high resolution ³¹P magnetic resonance spectroscopy, during muscular contractions in patients with HFpEF compared with healthy, similarly aged, participants, suggesting that the degree of metabolites emanating from the exercising muscle may not be diminished in HFpEF. Together with this previous work, completion of exercise at similar relative exercise intensities between groups in the present study makes it unlikely that differences in metabolic perturbations accounted for the observed blunted pressor response in patients with HFpEF. However, it is possible that disease-related changes in sensitivity of metabolically sensitive group IV skeletal afferent fibers partially explain the observed intensity-dependent reduction in pressor responses in the HFpEF group. Indeed, animal models of HF have documented a blunted exercise pressor response that is attributed, in part, to reductions in the expression and function of transient receptor potential vanilloid 1 (TRPV1) (45, 46) and acid sensing channel subtype 3 (ASIC₃) (47) metaboreceptor subtypes, though this has not been examined in human HF. In addition, there is initial evidence that the efferent component of the exercise pressor and muscle metabolic reflexes may be altered in patients with HFpEF. Indeed, we have recently identified a blunted pressor response to 40% MVC HG exercise in a patient with HFpEF compared with a healthy, similarly aged control despite similar increases in muscle sympathetic nerve activity (48), which appears to suggest an attenuation in end-organ (i.e., α-adrenergic) receptor sensitivity or responsiveness in patients with HFpEF (49). Furthermore, while the similar increases in muscle sympathetic nerve activity in an intensity-dependent manner suggest no disease-related changes in the central integration of neural signals, it remains possible that the blunted metaboreceptor sensitivity is offset by an accentuated mechanoreceptor contribution to muscle sympathetic nerve activity as a compensatory mechanism (45, 50, 51). Although the present study does not have information related to muscle sympathetic nerve activity due to its technically challenging nature, further studies are certainly warranted to explore these aspects of autonomic dysfunction during exercise and in response to isolated stimulation of mechano- and metaboreceptors.

Perspectives

The observed diminution in the exercise pressor response and muscle metaboreflex in patients with HFpEF may have far-reaching consequences on cardiovascular responses during exercise due to the highly integrative nature of autonomic control. In particular, the metabolic component of the exercise pressor reflex serves as an “error signal” to facilitate increases in blood flow to the contracting skeletal muscle (52–54). Thus, an appropriate activation of the muscle metaboreflex is needed to elicit an increase in blood pressure to maintain adequate perfusion of the contracting skeletal muscle (17). Conversely, a blunted pressor response due to diminished muscle metaboreflex responsiveness in the present study may lead to insufficient muscle perfusion within the contracting skeletal muscles, thereby contributing to premature muscle fatigue and exercise intolerance in patients with HFpEF (17). In support of this notion, previous studies have documented central (i.e., cardiac function) and peripheral (i.e., peripheral vasculature and skeletal muscle function) limitations that manifest as attenuated increases in cardiac output and/or oxygen extraction/utilization and contribute significantly to reduced exercise capacity during cycling exercise in patients with HFpEF (55–57). Furthermore, patients with HFpEF exhibited blunted skeletal muscle blood flow responses during small muscle mass exercise, which is an exercise modality that is not limited by cardiac output (34, 35, 57, 58). Importantly, recent work from Weavil et al. (58) has documented a greater susceptibility to premature fatigue due, in part, to diminished skeletal muscle blood flow during small muscle mass exercise in patients with HFpEF. Thus, given the extreme limitations to exercise and even daily activities in patients with HFpEF, additional studies using a larger sample size to further investigate neural cardiovascular control during exercise as a potential mediator of exercise limitations are warranted in this patient group.

Experimental Considerations

Although the specific role of antihypertensive medications (e.g., mineralocorticoid receptor antagonists and β blockers) is an important consideration for interpretation of our findings given their potential to modulate cardiovascular responses to exercise (59–62), there is compelling data suggesting that drugs prescribed for blood pressure management do not affect reflex responses. Indeed, Chant et al. (63) recently reported similar increases in arterial blood pressure during upright cycling exercise and muscle metaboreflex activation between patients with controlled, uncontrolled, or untreated hypertension, suggesting that antihypertensive pharmacotherapy is not effective in mitigating the exercise-induced rise in arterial blood pressure in this patient population. Our group has also documented similar pressor responses during rhythmic handgrip exercise in patients with hypertension who withheld use of antihypertensive medications for two weeks and those who continued standard antihypertensive therapy (64). Nonetheless, this has not been evaluated in patients with HFpEF and may therefore represent a confounding factor in the present study. We also recognize that obesity may contribute to the blunted pressor responses in patients with HFpEF. Although previous work from our group failed to identify differences in blood pressure between obese and nonobese patients with HFpEF during dynamic HG exercise (65), the impact of obesity on the pressor response to static HG exercise and PEI has not been examined in patients with HFpEF. Although the Modelflow method has been documented to reliably track changes in stroke volume and cardiac output (66–68), we acknowledge the limitations of this noninvasive methodology and recognize that alternate “gold-standard” methods (cardiac magnetic resonance imaging, Doppler echocardiography, and thermodilution) may more accurately quantify cardiac hemodynamics. It is also acknowledged that while we utilized what is considered a conventional experimental protocol (2-min static handgrip, 2-min PEI) for assessing pressor responses to exercise and isolated muscle metaboreflex activation, additional studies evaluating varied exercise intensities, modalities, and duration are needed to reach consensus on this aspect of HFpEF pathophysiology. Handgrip strength has been shown to be an independent predictor of six-minute walk test distance and peak oxygen uptake in patients with coronary heart disease (69, 70) and chronic obstructive pulmonary disease (71), suggesting that handgrip strength may provide an indirect insight into disease-related changes in exercise capacity. Importantly, while the present study observed no relationship between the magnitude of pressor responses and handgrip strength possibly due to small sample size, future studies should explore whether other measures of exercise capacity are related to the magnitude of cardiovascular and autonomic responses to small muscle mass exercise in patients with HFpEF. Finally, we acknowledge that the phenotypic and pathophysiological heterogeneity in HFpEF (i.e., comorbidities and medications) likely contributes to variability in the neural cardiovascular control during exercise, due in part to disease-related alterations in skeletal muscle structure and function (72–74) and the resultant variations in specific force (i.e., force per cross-sectional area) (75) and fatigability (44).

Conclusions

The current investigation has identified a reduced pressor response to static HG exercise that may be mediated, in part, by a blunted muscle metaboreflex activation in patients with HFpEF compared with healthy, similarly aged controls. Furthermore, the blunted pressure response to static HG exercise was exercise intensity-dependent, and the group differences in the pressor response were unrelated to handgrip strength. Together, these findings demonstrate a disease-related dysregulation in neural cardiovascular control that may contribute significantly to exercise intolerance in this patient group.

SUPPLEMENTAL DATA

Supplemental Tables S1 and S2: https://doi.org/10.6084/m9.figshare.22626724.

DATA AVAILABILITY

Data will be made available upon reasonable request.

GRANTS

This work was funded in part by the National Institutes of Health Grants T32 HL139451 (to K.B.), 1R15NR016826-01 (to M.J.H.), and HL162856 (to D.W.W.); the US Department of Veterans Affairs Grants CX002152 (to D.W.W.) and IK2RX003670 (to K.B.); and the American Heart Association Grant 18POST33960192 (to K.B.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

K.B., H.L.C., M.J.H., P.J.F., and D.W.W. conceived and designed research; K.B., H.L.C., S.M.R., J.R.V., and J.K.A. performed experiments; K.B., H.L.C., and J.R.V. analyzed data; K.B., M.J.H., J.D.T., J.J.R., P.J.F., and D.W.W. interpreted results of experiments; K.B. prepared figures; K.B. and D.W.W. drafted manuscript; K.B., H.L.C., S.M.R., J.R.V., J.K.A., M.J.H., J.D.T., J.J.R., P.J.F., and D.W.W. edited and revised manuscript; K.B., H.L.C., S.M.R., J.R.V., J.K.A., M.J.H., J.D.T., J.J.R., P.J.F., and D.W.W. approved final version of manuscript.