Effects of supine ergometer exercise on central and peripheral circulation in patients with ischemic heart disease

Abstract

[Purpose] To clarify the characteristics and safety of circulatory dynamics during supine ergometer exercise in patients with cardiac disease by examining autonomic nervous system activity and lower limb blood flow, and to compare these responses with those observed during seated ergometer exercise. [Participants and Methods] Ten patients with ischemic heart disease performed ergometer exercises in two positions, supine and seated at the anaerobic threshold level intensity. Measurements included lower limb blood flow, autonomic nervous activity, physiological indicators, and perceived exertion at rest and during warm-up, main exercise, and cool-down. [Results] The Sup Ex condition exhibited significantly higher lower limb blood flow and parasympathetic activity than the Sit Ex condition. No significant differences were observed in sympathetic nervous system activity, heart rate, systolic blood pressure, or double product. The heart rate, systolic blood pressure, and double product increased during exercise and decreased during cool-down, along with increased parasympathetic activity. Borg scale scores for leg and dyspnea increased over time but did not differ significantly between positions. [Conclusion] Despite similar cardiac loads and perceived exertion, the supine ergometer exercise resulted in greater parasympathetic activation and lower limb blood flow, suggesting its potential safety and benefits in the rehabilitation of patients with ischemic heart disease.

INTRODUCTION

Exercise therapy is a core strategy in cardiac rehabilitation, aiming to improve exercise capacity, prevent recurrence, and enhance clinical outcomes in patients with cardiovascular disease. Recently, supine ergometer exercise has emerged as a promising alternative. The supine position reduces gravitational effects and is associated with increased venous return and more efficient cardiac output compared with sitting or standing postures^1,2,3). Therefore, heart rate and blood pressure may rise less at comparable exercise intensities, potentially lowering myocardial oxygen demand. Moreover, previous studies have reported that exercise posture influences autonomic nervous system activity. Depending on the intensity, sympathetic and parasympathetic responses vary. In the supine position, excessive sympathetic activation may be suppressed, while a parasympathetic-dominant state is more likely to be maintained^{4, 5)}. These autonomic characteristics are crucial for regulating heart rate and managing cardiovascular risk during exercise. Exercise in the supine position may also promote more efficient blood flow distribution to the lower extremities. Sakai et al. reported increased superficial femoral artery blood flow in the non-exercising leg during single-leg pedaling, suggesting that posture influences systemic blood flow regulation⁶⁾. Similarly, Joyner et al. found that sympathetic activity was lower and forearm blood flow significantly higher during supine exercise than during standing⁷⁾, indicating time-dependent changes in blood flow distribution and autonomic modulation during prolonged exercise. However, most studies on supine ergometer exercise have involved healthy individuals^{4, 8)}, and evidence on its effects on hemodynamics and autonomic activity in patients with cardiac disease remains limited⁹⁾. Furthermore, few studies have examined physiological responses or time-dependent changes during exercise at the anaerobic threshold (AT), a commonly used clinical intensity level^10,11,12). Previous studies comparing supine and seated exercise postures in patients with IHD have primarily evaluated central hemodynamic indicators such as heart rate and blood pressure. However, comprehensive assessments including peripheral circulation, autonomic modulation, and perceived exertion across exercise phases remain limited. This study addresses this gap by examining these parameters under a standardized anaerobic threshold load, providing clinically relevant insights for cardiac rehabilitation. Therefore, in this study, we aimed to clarify the hemodynamic characteristics and safety of supine ergometer exercise in patients with cardiovascular disease by comparing autonomic nervous system activity and lower-limb blood flow during supine and seated ergometer exercise.

PARTICIPANTS AND METHODS

This single-center study was conducted between February and December 2024 and included 10 male patients (mean age: 77.6 ± 6.5 years; height: 163.5 ± 3.6 cm; weight: 61.5 ± 7.9 kg) diagnosed with IHD (Table 1). The inclusion criteria required that participants were more than 6 months post-onset and had no exacerbation of ischemic or heart failure symptoms, hemodynamic instability, or changes in medication for at least 4 weeks prior to measurement. The required sample size was calculated using G*Power 3.1.9.7, based on an effect size of 0.40, a significance level of 5%, and a statistical power of 80%, resulting in a minimum of nine participants. Exclusion criteria included peripheral neuropathy, history of smoking, orthopedic or neurological disorders, disease exacerbation within the past 6 months, or a determination of unsuitability by the attending physician. Clinical data, including left ventricular ejection fraction, New York Heart Association classification, body composition, medication use, and laboratory results, were obtained from electronic medical records. The study was approved by the Ethics Committee of Teikyo University of Science (Approval No: 24A017), and written informed consent was obtained from all participants. Peak oxygen uptake was assessed using a cardiopulmonary exercise test (CPX), during which the anaerobic threshold (AT) was determined. Experimental sessions consisted of supine (Sup Ex) and seated (Sit Ex) ergometer exercises performed in random order, with at least a 7-day interval between sessions. A supine ergometer (Terasu Ergo III PLUS, TE3PLUS-20/70; Showa Denki Co., Ltd., Osaka, Japan) was used for the supine exercise, and an upright ergometer (Strength Ergo 8; Mitsubishi Electric Engineering Co., Ltd., Tokyo, Japan) was used for the sitting exercise. Exercise intensity in both conditions was set at the AT, (mean: 48.6 ± 13.9 W). Each session included 4 min of rest, 3 min of warm-up, 10 min of steady-state exercise, 3 min of cool-down, and 4 min of recovery rest. Pedaling cadence was maintained at 50–55 rpm. The order of exercise conditions was determined using computer-generated random numbers. Sufficient rest was provided between sessions, and each session began only after participants had returned to resting physiological conditions. Lower-limb blood flow was assessed by locating the right dorsalis pedis artery using a portable ultrasound imaging device (miruco, Nippon Sigmax Co., Ltd., Tokyo, Japan) and measuring blood flow with a laser Doppler flowmeter (ALF21R, Admedic Co., Ltd., Tokyo, Japan) by placing the probe over the artery. Autonomic nervous system activity was evaluated using the MemCalc analysis method (Bonaly Light RF-ECG, EMG Co., Ltd., Tokyo, Japan). Frequency domain analysis of RR intervals was conducted to calculate high-frequency (HF: 0.15–0.4 Hz) and low-frequency (LF: 0.04–0.15 Hz) components. The LF/HF ratio was used as an index of cardiac sympathetic activity. Subjective exercise intensity was assessed using the Borg Scale^{13, 14)}, as recommended in the 2021 Guidelines for Rehabilitation in Patients with Cardiovascular Disease¹⁵⁾. Ratings of lower-limb and dyspnea were recorded every 2 min during each exercise phase. Heart rate was measured using the Dynascope 700 series DS-7100 system (Fukuda Denshi Co., Ltd., Tokyo, Japan), and systolic blood pressure was measured by auscultation. The double product, calculated as heart rate × systolic blood pressure, was used as an index of cardiac workload. Average double product values were calculated for each exercise phase under both Sup Ex and Sit Ex conditions. Statistical analysis was performed for each parameter—lower-limb blood flow, LF/HF ratio, parasympathetic activity, Borg scale ratings for lower-limb and chest fatigue, and double product—across four phases: rest, warm-up, main exercise, and cool down. Measurements in each phase were recorded every 2 minutes, and the mean values were used for analysis. Comparisons between Sup Ex and Sit Ex conditions were conducted using two way repeated measures analysis of variance (ANOVA) in SPSS version 23. For variables with significant interactions, Bonferroni post hoc tests were applied. A p-value <0.05 was considered statistically significant.

Table 1. Participant characteristics.

n		10
Sex		Men
Age, years		77.6 ± 6.5
Height, cm		163.5 ± 3.6
Weight, kg		61.5 ± 7.9
Body mass index, kg/m2		23.1 ± 3.0
Underlying disease
	Myocardial infarction, n, %	4 (40)
	Angina pectoris, n, %	6 (60)
LVEF, %		65.0 ± 9.4
NYHA ClassⅠ/Ⅱ		3/7
NTpro-BNP, pg/mL		122.4 ± 145.6
Peak V̇O2, mL/kg/min		20.0 ± 4.3
WR, W		48.6 ± 13.9
SMI		17.3 ± 1.2
β-blocker, n, %		6 (60)

Data are presented as means ± SD. LVEF: left ventricular ejection fraction; NYHA: New York Heart Association; WR: work rate; SMI: skeletal muscle mass index; Peak V̇O₂: peak oxygen uptake.

RESULTS

Table 2 summarizes the effects of Sup Ex and Sit Ex ergometer exercise on each parameter in patients with IHD. For lower-limb blood flow, a two-way repeated measures ANOVA showed a significant main effect of posture (p<0.05, η²=0.64), with Sup Ex producing higher values than Sit Ex. Time also had a significant effect (p<0.05, η²=0.84), with values during the main exercise phases exceeding those at rest. A significant interaction between posture and time was detected (p<0.05, η²=0.48). For the LF/HF ratio, neither posture (p=0.234, η²=0.15) nor time (p=0.153, η²=0.19) had a significant effect, and no significant interaction was observed (p=0.199, η²=0.12). For HF power, posture had a significant effect, with higher values in Sup Ex (p<0.05, η²=0.38). Although HF power declined progressively during exercise, neither the main effect of time (p=0.222, η²=0.14) nor the interaction (p=0.905, η²=0.03) reached significance. For heart rate, posture had no significant effect (p=0.17, η²=0.19), but time was significant (p<0.05, η²=0.94), with higher values during the main exercise than at rest. No significant interaction was found (p=0.053, η²=0.74). For systolic blood pressure, posture was not significant (p=0.44, η²=0.07), while time was significant (p<0.001, η²=0.60), with higher values during the main exercise than at rest. The interaction was not significant (p=0.70, η²=0.05). For the double product, posture was not significant (p=0.19, η²=0.17), but time showed a strong effect (p<0.05, η²=0.98), with values increasing during exercise and peaking in the main phase. The interaction was not significant (p=0.64, η²=0.04). For leg fatigue, posture was not significant (p=0.17, η²=0.20), whereas time was significant (p<0.05, η²=0.70), with greater fatigue during the main exercise than at rest. No significant interaction was found (p=0.51, η²=0.08). For dyspnea, posture had no significant effect (p=0.12, η²=0.25), but time was significant (p<0.05, η²=0.80), with higher values during the main exercise than at rest. The interaction was not significant (p=0.411, η²=0.40).

Table 2. Effects of Sup Ex and Sit Ex on various physiological parameters.

	Sup Ex				Sit Ex
	Rest	Warm-up	Main exercise	Cool-down	Rest	Warm-up	Main exercise	Cool-down
Lower limb blood flow (mL/min/100 g)	2.7 ± 1.0	15.8 ± 4.6	22.7 ± 3.2*	15.5 ± 6.2	1.9 ± 1.2	10.8 ± 6.7	11.2 ± 3.5*	11.2 ± 7.1
LF/HF (Hz)	3.4 ± 2.0	2.4 ± 1.8	3.4 ± 2.5	3.2 ± 1.7	11.8 ± 19.8	9.8 ± 18.2	10.3 ± 20.2	13.4 ± 21.0
HF (Hz)	80.7 ± 72.4	70.6 ± 60.9	61.9 ± 67.0	42.1 ± 46.5	43.3 ±27.6	43.7 ± 58.5	22.5 ± 31.4	22.7 ± 20.1
Heart rate (bpm)	67.5 ± 8.4	79.4 ± 10.7	92.9 ± 16.0*	84.5 ± 15.3	70.9 ± 13.6	84.0 ± 16.6	96.3 ± 20.1*	86.4 ± 20.1
Systolic blood pressure (mmHg)	125.0 ± 6.0	131.6 ± 5.7	143.3 ± 13.0*	134.9 ± 12.3	123.9 ± 4.5	134.4 ± 6.5	146.1 ± 13.3*	134.5 ± 6.0
Double product	8,433.6 ± 1,140.3	10,454.7 ± 1,536.0	13,380.0 ± 3,239.2*	11,514.6 ± 1,991.8	8,785.3 ± 1,7336.1	1,1287.6 ± 2,328.2	14,176.6 ± 3,633.5*	11,654.4 ± 2,944.6
Leg fatigue	7.4 ± 1.7	9.3 ± 1.6	12.4 ± 1.8*	10.8 ± 1.9	7.5 ± 1.6	9.1 ± 2.0	11.5 ± 2.1*	10.2 ± 1.9
Dyspnea	7.3 ±1.3	9.3 ± 1.9	11.6 ± 2.7*	10.1 ± 2.3	7.3 ± 1.3	9.0 ± 1.6	10.9 ± 2.4*	9.1 ± 2.1

Sup Ex and Sit Ex under AT load. Values are expressed as mean ± standard deviation. Lower limb blood flow: main effect of posture, p<0.05; main effect of time, p<0.05; interaction between posture and time, p<0.05. *p<0.05 indicates a significant difference between the main exercise and rest. LF/HF: main effect of posture, not significant (ns); main effect of time, not significant (ns); interaction between posture and time, not significant (ns). HF: main effect of posture, p<0.05; main effect of time, not significant (ns); interaction between posture and time, not significant (ns). Heart rate: main effect of posture, not significant (ns); main effect of time, p<0.05; interaction between posture and time, not significant (ns). *p<0.05 indicates a significant difference between the main exercise and rest. systolic blood pressure: main effect of posture, not significant (ns); main effect of time, p<0.05; interaction between posture and time, not significant (ns). *p<0.05 indicates a significant difference between the main exercise and rest. double product: main effect of posture, not significant (ns); main effect of time, p<0.05; interaction between posture and time, not significant (ns). *p<0.05 indicates a significant difference between the main exercise and rest. Leg Fatigue: main effect of posture, not significant (ns); main effect of time, p<0.05; interaction between posture and time, not significant (ns). *p<0.05 indicates a significant difference between the main exercise and rest. Dyspnea: main effect of posture, not significant (ns); main effect of time, p<0.05; interaction between posture and time, not significant (ns). *p<0.05 indicates a significant difference between the main exercise and rest.

DISCUSSION

In this study, we aimed to clarify the characteristics of circulatory dynamics by examining the relationship between autonomic nervous system activity and lower-limb blood flow during supine ergometer exercise. The Sup Ex resulted in significantly higher lower-limb blood flow (p<0.05) and HF power (p<0.05) than those in the Sit Ex. In contrast, no significant differences were observed between postures in LF/HF ratio, heart rate, systolic blood pressure, double product, or subjective measures of exertion (leg and dyspnea). In both exercise positions, physiological and perceptual responses increased significantly over time (p<0.05), indicating comparable trends across conditions. In the supine position, hydrostatic pressure is more evenly distributed, reducing the load on peripheral vessels and increasing venous return^{16, 17)}. This promotes cardiac output and enhances arterial blood flow¹⁷⁾. Reduced vascular pressure helps maintains vascular compliance and decreases vascular resistance, thereby increasing shear stress on the endothelium. This shear stress likely stimulates vasodilation via nitric oxide production and improves the arterial flow¹⁸⁾. The supine posture may also enhance muscle pump efficiency and optimize blood flow regulation during exercise. In contrast, the seated posture can promote venous pooling due to gravity, making the muscle pump more critical for maintaining venous return¹⁶⁾. However, in the supine position, venous pooling is minimized, enabling sufficient venous return even with minimal muscle contraction. This may facilitate smooth circulation and improve arterial delivery to the lower limbs.

Autonomic modulation may also contribute to the observed differences. In healthy individuals, sympathetic activation is attenuated in the supine position, while parasympathetic activity increases^{4, 5)}. This may be due to increased hydrostatic pressure on baroreceptors in the carotid sinus and aortic arch, enhancing baroreflex sensitivity¹⁴⁾, as well as stimulation of atrial stretch receptors by increased venous return¹⁷⁾. In contrast, venous pooling in the seated position may reduce venous return and suppress baroreflex sensitivity, promoting sympathetic dominance¹⁹⁾. In patients with IHD, elevated baseline sympathetic tone may be counterbalanced by enhanced parasympathetic activation in the supine position, explaining the lack of significant posture-related differences observed. Sympathetic activation during exercise induces vasoconstriction and raises systolic blood pressure¹⁹⁾, but these effects may be partly attenuated in the supine posture. In the supine position, increased cardiac output due to greater venous return may limit the rise in systolic blood pressure. Heart rate is regulated by both sympathetic and parasympathetic nervous system activity, with sympathetic dominance contributing to its increase over time²⁰⁾. The double product, an index of myocardial oxygen demand, rises with increasing exercise load²¹⁾. The double product increased in both conditions; however, no significant difference was observed between postures, suggesting a comparable myocardial load. In the seated position, greater venous pooling may require increased sympathetic regulation of cardiac output, potentially increasing heart rate and systolic blood pressure. Both groups exercised at the anaerobic threshold (AT), a safe and effective intensity for improving aerobic capacity and endurance in patients with cardiac disease²²⁾. Posture did not significantly affect outcomes, indicating that exercise load was the primary driver of physiological responses. Differences in ventilatory mechanics between postures, along with the influence of IHD on autonomic regulation and peripheral circulation, may have reduced posture-related differences. Although seated AT-level ergometer exercise is well established, its supine counterpart has been little studied. In this study, no adverse events occurred in either posture, suggesting that AT-level supine exercise can be safely performed under appropriate supervision. The lack of posture-related differences in Borg scores suggests that perceived exertion was not dependent on posture. The absence of posture-related differences in Borg scale scores indicates that subjective exertion was not directly influenced by posture. While fatigue scores increased over time, similar ratings between postures suggest that both conditions imposed an equivalent exercise load at the AT threshold.

These findings show that, despite similar cardiac load and perceived exertion, supine ergometer exercise elicited greater parasympathetic activation and lower-limb blood flow than seated exercise, with no adverse events in either posture, supporting its potential safety in IHD rehabilitation. Increases in HF power and limb blood flow in the supine position suggest parasympathetic predominance may promote vasodilation and improve oxygen delivery without increasing cardiac workload. Similar improvements in stroke volume and baroreflex sensitivity have been observed in healthy individuals^{4, 5, 14, 17)}. In contrast, IHD patients often have reduced HRV and HF power, reflecting impaired autonomic function²³⁾, with β-blockers and cardiac dysfunction blunting heart rate and blood pressure responses. Under these conditions, peripheral circulatory and autonomic mechanisms become critical, and the present results indicate that supine exercise can enhance these functions without adding central hemodynamic stress. This study has several limitations. First, only acute effects were examined, and long-term adaptations and clinical outcomes associated with prolonged supine and seated ergometer training should be clarified in future research. Second, differences in clinical background (e.g., disease severity, comorbidities, and medication use) may have influenced the results and should be considered in future studies. Stratified analyses based on patient characteristics are essential for developing individualized exercise prescriptions. Third, this study focused on the AT load; however, comparing physiological responses across different exercise intensities is important for optimizing posture-specific exercise strategies. The use of beta-blockers among participants may also have influenced autonomic and hemodynamic responses. Since beta-blockers attenuate sympathetic nervous system activity and blunt heart rate responses, they may have contributed to an underestimation of physiological differences between exercise postures. Furthermore, differences in beta-blocker type, dosage, and timing of administration could have introduced inter-individual variability. These findings suggest that, despite comparable cardiac load and perceived exertion, supine ergometer exercise elicits greater parasympathetic activation and lower-limb blood flow. This indicates its potential utility as a safe and beneficial modality in the rehabilitation of patients with IHD.

Funding

This study was supported by the JSPS KAKENHI (grant number: JP21K12803).

Conflict of interest

All authors declare that there are no potential conflicts of interest.