Currently, a wide variety of exercise training methods are used to improve physical fitness, muscular strength, and endurance for fitness programs and rehabilitation programs. Among them, blood flow restriction (BFR) training was reported to have significant benefits to increase skeletal muscle mass strength (Yasuda et al., 2015). This exercise practice with low-intensity workloads has generated considerable interest in recent years due to its positive effects comparable to traditional protocols with higher intensity resistance training (i.e., ~70-85% of a one-repetition max (1-RM) (Slysz et al., 2016). BFR uses an inflatable cuff or tourniquet placed proximally to the exercising muscle to partially that fully occlusion blood flow in order to stimulate neuromuscular adaptations (Iida et al., 2011). It produces partial restriction of arterial inflow and full occlusion of venous outflow of the extremity to which the external cuff pressure is applied (Loenneke et al., 2012; Partsch and Partsch, 2005). This maneuver elicits health benefits of the normal physiological adaptations to exercise due to accumulated metabolites which increase muscle growth (Yasuda et al., 2015).

BFR training provides health benefits to the young and the elderly as well as individuals with cardiovascular disease (Lowery et al., 2014; Yasuda et al., 2015; Zhao et al., 2021). Moreover, previous research suggested BFR combined with low load resistance training, or cardiovascular endurance training can be utilized as a potential therapeutic alternative in the rehabilitation of patients who may be incapable of performing high load resistance training (e.g., elderly, recovering athletes) (Centner et al., 2019; Mouser et al., 2019). Known beneficial effects are an improvement in VO_2max (de Oliveira et al., 2016; Held et al., 2020) aerobic capacity (Cardoso et al., 2020), carotid arterial compliance (Ozaki et al., 2011), and microvascular filtration capacity of skeletal muscles (Evans et al., 2010). A big advantage of BFR training produces is that it similar benefits of high-intensity resistance exercise even though low-loads are used.

However, even though side effects with BFR were minimal, studies have reported potential negative cardiovascular effects such as an increase in blood clots, ischemia-reperfusion, and muscle damage (Cristina-Oliveira et al., 2020; Loenneke et al., 2011; Patterson et al., 2019). Particularly, among the side effects associated with BFR training is the muscle metaboreflex (MMR)-induced cardiovascular abnormalities. Continued research is required to determine if the exercise pressor reflex is altered. This MMR (a component of the exercise pressor reflex) rases blood pressure (BP) due to an increase in metabolites produced by mismatch between metabolic demand and supply within exercising skeletal muscle. Evidence showed BFR training was positively effective in reducing BP in both normotensive and hypertensive individuals (Araujo et al., 2014; Neto et al., 2015). On the other hand, BP responses were exaggerated in hypertensive individuals, which is partially associated with abnormal metaboreflex (Chant et al., 2018). Furthermore, Araujo and colleagues (Araujo et al., 2014) reported BFR exercise-induced BP response was substantially elevated compared to exercise without BFR. Thus, more research is needed to reveal a protocol which can be used safely without excessive increase in BP during BFR training sessions. However, with regard to negative effects most studies focused on exercising BP responses during lower body mainly BFR exercise. Our study would be relevant to providing exercise professionals, as well as healthy individuals, that the use of BFR during exercise can cause negative impacts to the body. Thus, we examined whether upper body resistance exercise (UBRE) with BFR evoked excessive cardiovascular responses. It was hypothesized that lower body resistance exercise (LBRE) with BFR would result in higher BP responses during exercise and recovery compared to the UBRE with BFR.

A total of 18 apparently healthy sedentary women college students aged between 20 and 30 years were recruited in this study. Subjects were normotensive, non-medicated, non-smokers, and free of cardiovascular diseases that could affect physiological responses to exercise and post-exercise. Subjects were considered to be sedentary since none of them had participated in at least 30 min moderate intensity physical activity on three days per week for at least three months as assessed by health history questionnaire (American College of Sports Medicine, 2021). They are informed of the pretesting guidelines and reviewed the experimental procedures and signed an informed consent form containing the risk and benefits prior to participating in this study. The study was approved by World Health Beauty Institutional Research Board (1-20170113119-AB-N-01-09).

All subjects reported to the laboratory for the baseline measurement and BFR exercise. The experimental protocol consisted of measurements of cardiovascular responses during and after the UBRE and LBRE with or without BFR. Each subject completed workouts with or without BP cuff inflated on their arms or legs. All exercise tests were performed at the same time of day to reduce diurnal variation. They were asked to refrain from consuming alcohol and strenuous physical activity for 48 h, and caffeine for three hours prior to each test. In all experiments, each exercise was measured over a total of 4 visits. Each protocol was separated by three days. Resting BP was measured at least twice in a seated position via a sphygmomanometer and pressure cuff. For the relative exercise intensities for the resistance exercise used in this study, the one-repetition maximum (RM) test was conducted to determine the maximal weight each subject can perform with one repetition. The following day, subject completed four bouts of UBRE and LBRE at a 20% workload of their predetermined 1RM with and without BFR. The body composition, including height, weight, body mass index, fat mass, and percent body fat was measured using a body composition analyzer (MC190-EM; Tanita,Tokyo, Japan) (Table 1).

The resistance exercise program consisted of a three minutes warm-up exercise on cycle ergometer with low intensity, followed by exercise for the upper and lower limbs (squat and arm curl). The trial was performed with loads corresponding to the 20% of 1 RM without BFR. Under resting conditions, the maximum number of repetitions was examined several times to obtain exact values for one-repetition maximum (1RM). If they were able to perform more than 10 repetitions, then a heavier weight was provided. Whenever the maximum number of repetitions was ≤10, the weight was considered submaximal weight and the 1RM was calculated as follows: estimated 1RM (kg) = submaximal weight (kg)/(102.78-2.78×maximum number of repetitions)/100 (Shimizu et al., 2016). Determination of 1-RM involved performing 5-10 repetitions with a light load (40-60% of estimated 1RM). After 3 minutes of rest, subjects performed 3 to 5 repetitions with increased load (50 to 70% of estimated 1-RM). After resting for another 3 minutes, the exercise was repeated 2 to 3 times with a load of 60 to 80% of the estimated 1-RM. After these submaximal repetitions, 1-RM was determined within 5 repetitions, including 3 minutes of rest. All repetitions were performed at the same movement speed and range of motion were identical between measurements. Final weight was recorded as the participant's 1-RM (Lowery et al., 2014). If the subject failed to complete 5 repetitions, 1-RM was estimated using Epley's formula (1-RM = load [kg] × (1 + [0.033 × number of repetitions]) (Shimizu et al., 2016).

The exercise intervention was repeated 10 times at the speed of 1 second with 4 sets. Rest times between sets used during acute resistance training with BFR are generally short and BFR is maintained. A study on acute resistance exercise used rest periods of 150 seconds (Loenneke et al., 2010), but was not found to increase metabolic responses. The resting time between sets was 30 seconds. The BP was measured at rest, and during exercise and recovery using an automated Omron JPN 500 device (Omron Healthcare, Japan). All subjects were instructed to adhere to their normal living and dietary routines throughout the study.

Although muscle adaptation was achieved by applying the same pressure to each individual in a prior study, it was found that higher BFR pressure can increase cardiovascular response (Mattocks et al., 2017). We employed the cuff pressure of 80% arterial occlusive pressure during low-load resistance training. This training induced hypertrophic and strength responses comparable to traditional high-load training (Laurentino et al., 2012). The cuff was placed in the most proximal part of both legs and arms using pneumatic and stretchable bands size (50 mm for lower limbs and 40 mm for upper limbs, respectively) (KAATSU Global Inc., USA). The arterial occlusion pressure was equal to the systolic pressure in the upper limbs and twice the systolic pressure in the lower limbs (Iida et al., 2011). The BFR was applied immediately before the squat set and used only during the exercise, releasing during each rest interval.

Stroke volume (SV) and heart rate (HR) were obtained continuously via impedance cardiography (Physioflow, Manatec Biomedical, Paris, France) from the rest throughout the recovery. This non-invasive device measures real-time cardiac output (CO) data and has been used to quantify cardiac parameters in healthy subjects (Dillon et al., 2021; Lee et al., 2015; Tordi et al., 2004). For the SV measurement, two electrodes were placed on the supraclavicular fossa at the base of the left side of the neck, two electrocardiography (ECG) electrodes used for recoding single lead ECG, and two electrodes were placed at the xiphoid process. The Physioflow is a thoracic bioimpedance technique that detects the change in impedance by injecting a high frequency (75 kHz) alternating electrical current of low magnitude (1.8 mA peak to peak) via two skin electrodes positioned on the neck and another two positioned on xiphoid process (Bougault et al., 2005). By detecting the difference of thoracic impedance over time, this device noninvasively measures the SV (Charloux et al., 2000). CO was calculated by the formula (HR x SV x BSA). HR is measured from the ECG, SV is SV index (i.e., SV/BSA), and BSA is body surface area. It has well known that this transthoracic bioimpedance technique is valid against the direct Fick method at rest and during exercise (Charloux et al., 2000; Richard et al., 2001). Mean arterial pressure (MAP) was calculated using the formula: MAP = [(SBP-DBP) x 1/3)] + DBP. Total vascular conductance (TVC) was calculated as CO/MAP.

Data analysis was performed using SPSS PC+ for Windows (version 26.0). Based on Cohen’s D power equation, in order to maintain the effect size of 0.95 and power of 0.8, a minimum 18 participants was needed. Data are expressed as the mean ± standard deviation (SD). Mean values of SBP, DBP, MAP, HR, SV, CO, and TVC for each 30 seconds interval at rest and during exercise and recovery used for comparison between two resistance exercises with and without BFR. A 2 x 2 x 6 repeated measures ANOVA was used to test the effects of resistance exercise groups (UBRE and LBRE) on cardiovascular response across condition (with and without BFR) and across time (Rest, last minute of exercise, R10, R20, R30, and R40). Statistical significance was set at p < 0.05.

Table 2 indicates the comparison of cardiovascular responses during UBRE and recovery according to BFR and non-BFR. There were significant interactions between time and group effects in both DBP and MAP. BFR resulted in greater DBP and MAP only during exercise compared to without BFR. SV was significantly increased at only 10 min recovery compared to the exercise. HR and CO were significantly decreased at 10 min recovery from exercise and this reduction remained during 20 min, 30 min, and 40 min recovery. There was significant interaction effect in TPR. BFR had greater TPR during exercise compared to the without BFR.

Table 3 indicates the comparison of cardiovascular responses at LBRE and recovery according to BFR and non-BFR. SBP significantly decreased from exercise throughout the recovery in both conditions. There were significant interactions between time and group effects in both DBP and MAP. BFR resulted in greater DBP and MAP only during exercise compared to without BFR. SV was significantly increased at 10 min, 20 min, 30 min and 40 min recovery compared to the exercise. There were significant interactions between time and group effects for in both HR and CO. HR and CO were significantly decreased at 10 min recovery from exercise and this reduction remained at 20 and 40 min of recovery. There was significant interaction effect in TPR. BFR had greater TPR during exercise compared to the without BFR. There were no differences in TPR in both conditions.

Table 4 indicates comparison of cardiovascular responses according to UBRE, LBRE, and BFR during recovery. SBP significantly decreased from exercise throughout the recovery in both conditions. There were significant interactions between time and group effects in both DBP and MAP. LBRE with BFR resulted in greater DBP and MAP only during exercise compared to UBRE with BFR. There were significant interactions for HR, SV, and CO. There were significantly differences in HR and SV between UBRE and LBRE with BFR and at 10 min, 20 min, and 30 min recovery. CO was significantly higher during LBRE with BFR and recovery compared to the UBRE. There was significant interaction effect in TPR. BFR had greater TPR during exercise compared to the without BFR. There were no differences in TPR in both conditions. There was significant interaction in TPR between LBRE with BFR. TPR was significantly decreased in LBRE with BFR compared to the UBRE with BFR and this reduction remained throughout the recovery.

BFR stimulates the exercise pressor reflex (EPR) in response to tissue hypoxia, and there are concerns about how reduced blood flow affects cardiovascular function (Spranger et al., 2015). BFR is an exercise method with a low load that may induce muscle hypertrophy. Since it artificially reduces the amount of blood returning to the heart, subject's cardiovascular response must be monitored. When the working muscles need more oxygen, the central nervous system responds by attenuating the parasympathetic nerve activation and facilitating the sympathetic nerve activation (Smith and Fernhall, 2011). This causes the body to increase HR to try to pump more blood to deoxygenated tissues, which increases the cardiac load and the heart requires more amount of oxygen to the myocardium (Smith and Fernhall, 2011).

In this study, the cardiovascular response according to BFR showed an interaction between DBP, MAP, and CO of UBRE and LBRE during exercise and recovery (p < .05). SBP showed no difference in both UBRE and LBRE during exercise and recovery regardless of BFR, but significantly decreased after 10 minutes of exercise. This result may have been attributed to the post-exercise hypotensive effect (PHE) effect. The reason for this is the decrease in CO that is not completely explained by increased systemic peripheral vascular resistance (Maior et al., 2015). In this study, DBP decreased according to BFR (19.1%, 22.2%) and non-BFR (17.0%, 17.4%) 10 minutes after UBRE and LBRE recovery, respectively. Previously, similar results were also suggested by Maior et al. (2015) and Neto et al. (2015).

Recently, interest in the exercise pressor reflex (EPR) of skeletal muscle has been demanded as a physiological phenomenon requiring attention in BFR exercise (Spranger et al., 2015). In exercise physiology, it was suggested that the EPR determines the cardiovascular response when performing physical activity such as BFR exercise. This response is characterized by hypersensitivity of the sympathetic nervous system, which may explain the increase in HR and BP (Mitchell, 2017). Metaboreflex and muscle mechanoreflex are regulators of EPR, which are both increased during BFR exercise by skeletal muscle cuff pressure (Renzi et al., 2010).

Most previous studies showed no change in SBP for at least 10 minutes immediately after LBRE using BFR (Tomschi et al., 2018). The increase in SBP during exercise is thought to be from EPR, which occurs even under conditions of BFR due to the stimulation of mechanical reflexes (vascular compression) and metabolic reflexes (metabolite accumulation) known as Groups III and IV (Kaufman et al., 1984). However, it is very difficult to compare changes in SBP because the degree of change in SBP can vary depending on the exercise parameters (e.g., intensity, volume, and cuff pressure and width).

In this study, DBP significantly decreased during recovery after LBRE regardless of BFR, and returned to resting levels after 10 minutes of recovery. Figueroa and Vicil (2011) reported that SBP and DBP increased immediately after LBRE regardless of BFR, but returned to the resting level during 30 minutes of recovery. Elevated MAP after exercise was significantly higher in the BFR group compared to the non-BFR group. CO increased similarly between the two sessions, but SV did not change and HR increased during exercise with BFR. It is believed this result may provide circulatory difficulties to people with cardiovascular disease.

In this study, cardiovascular responses between UBRE and LBRE of BFR showed significant differences in DBP, MAP, SV, HR, CO, and TPR (p < .05). This result shows that a significant difference in SBP after UBRE with BFR. This supports the study by Tomschi et al. (2018) who reported no change, suggesting that a change in SBP is not dependent on intensity, rest interval, or ischemia. Additionally, the significant decrease in DBP at 10 min after BFR may be due to decreased TPR and BP (Coote, 2010). In this study, the BFR group showed a significant increase in HR compared to the non-BFR group in UBRE and LBRE. In particular, the difference that induced higher HR during LBRE compared to UBRE could be attributed to the baroreflex (Spranger et al., 2015). Therefore, low-intensity LBRE (20-40% 1-RM) with BFR is thought to be related to an acute increase in cardiovascular response.

In this study, healthy sedentary women college students aged between 20 and 30 were recruited. Most studies have not stratified men and women for analysis or simply have omitted women completely. In fact, females are often underrepresented in terms of scientific evaluation, largely attributed to the dynamic hormonal fluctuations of the menstrual cycle (Hunter, 2016), which may alter their responses or increase variability to exercises used in the studies. Our study should consider potential limitations. The current study included healthy and physically active individuals. The study cannot be applicable to other populations such as older adults, sedentary populations, and patients. Another limitation is the fact that this study did not evaluate the effects of the BFR during exercise over longer durations.

In conclusion, in resistance exercise with BFR, CO increased due to increase in HR and SV during exercise, and blood pressure response was stable due to a significant decrease in TPR during exercise. That could be the cause of the action of metabolic receptors due to BFR during resistance exercise. Therefore, it was judged that low-intensity resistance exercise with BFR had a positive effect on the hemodynamic response of the cardiovascular system during exercise and recovery. It will be necessary to conduct a long-term training study to confirm the differences in cardiovascular responses of UBRE and LBRE exercise according to BFR in the future. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.