Introduction

The investigation into arterial occlusion pressure (AOP) and its effects on both heart rate variability (HRV) and post-occlusion hyperemic responses (PORH) in the microvasculature has seen considerable advancements, especially in elucidating the body’s reactions after blood flow restriction1. For instance, a study aiming to assess HRV indices during high-load and low-load aerobic exercises with and without blood flow restriction (BFR), found that HRV on-kinetics were faster in low-load compared to low-load + BFR and high load, while recovery of HRV indices was delayed in high load compared to both low-load and low-load + BFR2. On the other hand, a study comparing acute autonomic and cardiovascular responses to low and high-load eccentric exercise with and without BFR in 60 men, found no significant differences between groups or interaction effects on cardiovascular variables or autonomic indices; although, an increase in vagal activity was observed during the recovery phase for both low load eccentric + BFR and high load eccentric + BFR, indicating that different loads of eccentric exercise with or without BFR did not cause autonomic or cardiovascular imbalance post-exercise3.

BFR training and ischemic preconditioning (IPC) are widely used in sports4,5. While muscle adaptations in BFR are typically linked to metabolite buildup and a low-oxygen environment6, studies on microcirculatory and sympathetic responses to different body positions7 and AOP are limited8. AOP, often monitored with a Doppler device, typically set between 40 and 80% of individual occlusion pressure values8 with higher AOP used in IPC likely eliciting more pronounced hyperemic responses compared to lower AOP in BFR protocols9. Despite the preliminary studies mentioned above analyzing HRV after AOP in BFR training2,3, research is still lacking on how different AOP levels impact HRV. Previous studies suggested that lower occlusion pressure leads to a more pronounced activation of the sympathetic system, which, in turn, influences HRV10,11. It is therefore expected that particular AOP values can dictate the extent and duration of the hyperemic response, suggesting a relationship between recovery and adaptation processes and the AOP levels employed12.

Post-occlusion hyperemic response (PORH) is a physiological phenomenon characterized by a transient increase in blood flow above baseline levels following the restoration of circulation after an occlusion13. This response is essential for regulating blood flow in tissues based on metabolic demands14, particularly in skeletal muscle during physical activity or flow restriction14,15. The response is governed predominantly by local, endothelium-dependent control within resistance vessels—rather than by direct autonomic commands—with endothelial pathways typically dominating as vessel caliber decreases16. In small arteries and arterioles, endothelium-derived hyperpolarizing factor (EDHF) signaling and activation of Ca2+-activated K+ channels are major mediators of the PORH vasodilation17,18. In human skin specifically, PORH may rely on sensory-nerve–driven axon-reflex contributions alongside endothelial hyperpolarization mechanisms19. Consistent with this, cutaneous PORH shows little or inconsistent involvement of nitric oxide and prostanoids compared with EDHF/sensory mechanisms20,21,22. In skeletal muscle beds, the early hyperemic peak reflects a combination of myogenic relaxation and metabolic vasodilator accumulation (e.g., adenosine), whereas nitric oxide and prostaglandins tend to modulate the later/sustained phase to a lesser degree23,24,25. Additional evidence in humans indicates a role for inwardly rectifying K+ (Kir) channels during reactive hyperemia26. Importantly, while sympathetic/parasympathetic outflow can shape baseline tone, the magnitude and time course of PORH are largely independent of sympathetic control, underscoring its predominantly local origin27,28.

These autonomic responses to occlusion appear to be pressure-dependent10,29,30. Higher AOP levels are associated with reduced HRV, likely due to increased sympathetic nervous system activity under prolonged ischemic conditions10,29,30. In contrast, lower AOP levels may facilitate parasympathetic reactivation, promoting improved HRV and cardiovascular recovery31. Additionally, the PORH response—characterized by a transient surge in blood flow following occlusion release—has been closely linked to the magnitude of AOP, as demonstrated by a study showing that higher occlusion pressures led to more pronounced increases in blood flow during the PORH response, suggesting a dose-dependent relationship32. This hyperemic response reflects the dynamic interplay between sympathetic and parasympathetic regulation, which is crucial for maintaining hemodynamic stability However, while systemic autonomic tone can modulate baseline hemodynamics, PORH itself is elicited predominantly by local endothelial and sensory-nerve mechanisms, with only a limited direct contribution from autonomic outflow to its immediate magnitude and time course17,19.

HRV serves as a valuable non-invasive marker for monitoring these autonomic shifts, helping to elucidate how varying AOP levels modulate cardiovascular control, tissue perfusion, and microcirculatory adaptation in BFR and IPC protocols. For instance, a study33 found that cycles of cuff inflation/deflation altered several spectral HRV indices during stress tests, such as increased very-low-frequency power and changes in detrended fluctuation analysis, indicating modulation of autonomic control linked with occlusion‐related perfusion events. Although established methods exist for assessing microcirculatory function and HRV, the scientific literature still lacks comprehensive comparisons of hyperemic and HRV responses across varying levels of AOP. Analyzing these variables may offer deeper information into the temporal and contextual factors that influence both HRV and post-occlusive hyperemia, particularly during BFR. Therefore, this study aimed to compare the effects of four different AOP levels (40, 80, 100, and 130%) on microcirculatory responses—including resting flow (RF), time to peak (TP), recovery time (TR), biological zero (BZ), and AOP itself—as well as autonomic nervous system activity, assessed via HRV parameters such as the average NN interval (AVNN), standard deviation of NN intervals (SDNN), and the low-frequency to high-frequency ratio (LF/HF).

Material and methods

Study design and setting

A prospective experimental study took place, with participants visiting Provita Medical Center four times between Monday and Thursday, from 9:00 to 11:00 a.m. Initially, anthropometric measurements were obtained using a Tanita MC-580 M P bioimpedance body composition analyzer (Japan, 2022). Subsequently, participants rested in a seated position for 20 min. On subsequent days, volunteers lay supine and underwent daily assessments of HRV and PORH responses at varying applied AOP levels: 40, 80, 100, and 130%. They remained in these positions for 5 min prior to each assessment. Measurement positions adhered to scientific literature standards to ensure the stability of Laser Doppler Flowmeter (LDF) measurements34,35,36. All assessments were conducted by the same two researchers (a physician and a physiotherapist) in a temperature-controlled room (21.2 ± 0.75 °C; humidity: 54.49 ± 2.5%).

Participants

An a priori power analysis was performed utilizing G*Power software. With a projected effect size of f = 0.50, a significance level set at α = 0.05, and a target statistical power of 0.80, the estimated minimum sample size needed was 28 participants. The study ultimately involved 30 participants, thus providing sufficient statistical power for the analyses conducted.

Thirty healthy volunteers, aged 18 to 30, were recruited via convenience sampling. Based on McKay’s participant classification scheme, the group was level 137. All participants were non-smokers, had an ankle-brachial index (ABI) between 0.9 and 1.2, reported no other health issues, and were not on any medications. They were instructed to refrain from vigorous exercise and alcohol for 24 h before each study session and to avoid ergogenic drinks like coffee and cola for 6 hours. All women were examined during the follicular phase of their menstrual cycle. Exclusion criteria included hypertension (≥ 140/95 mmHg), use of steroids or contraceptives, nicotine dependence, or medications affecting systemic hemodynamics (e.g., β-blockers, calcium antagonists, renin-angiotensin system inhibitors). Participants provided written informed consent after receiving detailed information about the study’s risks and potential benefits and were informed of their right to withdraw at any time. The study received ethical approval from the Polish Society of Physiotherapy (ref. no: 3.03.2024), was registered on 18/04/2024, under clinical trial number ISRCTN15418049, and was conducted in accordance with the principles of the Declaration of Helsinki.

Experiments were conducted with 30 participants (19 men, 11 women). Table 1 summarizes their physical characteristics, presenting the mean values (MEAN) and standard deviations (STD) for age, height, body mass, body mass index (BMI), Ankle-Brachial Index (ABI), and AOP at 100%.

Table 1 Description of physical characteristics across examined participants presenting mean value within each group (MEAN), and standard deviation (STD) for Age, Height, Weight, Body Mass Index (BMI) Ankle-Brachial Index (ABI), and Arterial Occlusion Pressure (AOP).
Full size table

Measurement procedures

  1. 1.

    Following a period of rest in a supine position, thigh circumference was assessed using a centimeter tape, while the depth of the femoral artery in the area covered by the cuff was measured via ultrasound. The circumference of the thigh is a key determinant of the pressure needed to achieve arterial occlusion, particularly in the context of BFR exercise. Several factors, including cuff width and body position, modulate the correlation between thigh circumference and AOP. A clear understanding of these interactions is essential for the precise prescription of BFR, thereby ensuring safety and effectiveness across different exercise and physiotherapy protocols38.

  2. 2.

    HRV data were collected using Polar H10 chest strap sensors, which detect beat-to-beat R-R intervals via ECG-based electrical signal measurement using skin-contact electrodes. The R-R interval data, accurate to within ± 1 ms, were wirelessly transmitted over Bluetooth Low Energy to the HRV Logger software application. This software recorded, timestamped, and stored the data for further HRV analysis, including time-domain, frequency-domain, and non-linear measures. Since HRV is sensitive to posture-related autonomic shifts, the supine position was selected to standardize baseline parasympathetic predominance, which is particularly relevant in the context of BFR research and the study of hyperemic rebound phenomena39.

  3. 3.

    ABI was determined by a skilled physician using a 2D SonoScape P20 B-mode ultrasound scanner (China 20220, linear transducer with a frequency range of 4–18 Hz). To maintain consistency in measurements, the location of the ultrasound probe on the skin was marked following the application of ultrasound gel40.

  4. 4.

    Following skin disinfection of the big toe, the LDF probe was positioned on the plantar skin, aligning its placement with the projection of the nail’s center.

  5. 5.

    The physician monitored the tibial pulse of the dominant leg until complete flow occlusion, placing probes along both arteries at a constant insonation angle of < 60°41. The Doppler window encompassed the entire vessel. Standardizing ankle-brachial index (ABI) measurements is crucial for accurate diagnosis and risk assessment of peripheral arterial disease (PAD) and cardiovascular disease, as well as the safety of BFR33. After 10 min of supine rest, a 13 cm wide pneumatic cuff (Riester ®, Germany) was placed over the dominant thigh’s inguinal region. Cuff pressure was increased from 0 to 100 mmHg, then by 10 mmHg increments until ultrasound no longer detected arterial flow, thus determining individual AOP values. Each pressure level was maintained for 30 s to stabilize blood flow42. Our study employed the standard 5-min occlusion test on the dominant leg43.

  6. 6.

    Post-occlusion responses were evaluated using a PeriFlux System 5000 Laser Doppler Flowmeter (Perimed AB, Järfälla, Sweden), considered a gold-standard device for noninvasive microcirculatory assessment due to its high temporal resolution and reproducibility35. The LDF probe was placed on the plantar skin of the dominant leg, specifically aligned with the center of the nail of the big toe, targeting a skin volume of approximately 1 mm3 at a tissue depth of 2.5 mm.

The PORH protocol involved inducing transient ischemia by inflating a pneumatic cuff (13 cm wide, placed at the inguinal region of the dominant thigh) to various pressure levels expressed as percentages of the participant’s individual AOP, previously determined via Doppler ultrasound44. An automated pneumatic cuff system (Hokanson E20 Rapid Cuff Inflator with AG101 Air Source, Bellevue, WA, USA) was used to apply and precisely maintain the target occlusion pressure. The device automatically inflated the cuff to the desired percentage of AOP (40, 80, 100, or 130%) and released it in a controlled manner to ensure consistent inflation and deflation timing across participants and sessions.

Each occlusion lasted for 5 min, a standardized duration adequate to evoke maximal metabolic and endothelial vasodilatory responses. Baseline resting flow (RF) was measured prior to cuff inflation. Upon rapid cuff deflation, reactive hyperemia ensued, characterized by a transient increase in skin perfusion detected by the LDF at a 32 Hz sampling rate. Parameters extracted from the PORH response included45,46: Rest Flow (RF), The baseline perfusion value prior to occlusion, expressed in perfusion units (PU); Time to Peak (TP): The interval (seconds) from cuff release to the maximum hyperemic perfusion value, reflecting the speed of microvascular reperfusion; Recovery Time (TR): The duration (minutes) required for perfusion to return from peak hyperemia back to baseline resting levels, indicating vascular recovery kinetics; AOP min: The minimum cuff pressure at which resting blood flow begins to decline, marking the threshold for arterial flow restriction; AOP 100%: The cuff pressure corresponding to complete arterial occlusion, confirmed by the absence of detectable arterial flow on Doppler ultrasound; and Biological Zero (BZ): The minimal flow value recorded during full arterial occlusion, representing residual signal or non-flow related noise.

The analysis of HRVconsidered the following measures:

  • AVNN, or Average NN Interval [ms], represents the mean duration in milliseconds between consecutive heartbeats. Longer RR intervals correspond to a lower heart rate, while shorter intervals indicate a higher one. The normal average RR interval (AVNN) typically ranges from 800 to 1000 [ms].

  • SDNN, the Standard Deviation of NN Intervals in milliseconds [ms], is the standard deviation of the RR intervals. For healthy adults, SDNN (Total HRV Variability) norms are: low variability below 20 [ms], medium variability between 20 and 50 [ms], and high variability above 50 [ms]47.

  • LF/HF (Low/High Frequency Ratio) [non-arbitrary unit] represents the ratio of low-frequency (LF) power to high-frequency (HF) power. Normative values for LF/HF (sympatho-parasympathetic balance) are generally: < 1 indicates parasympathetic predominance (relaxation, recovery), 1–3 suggests sympatho-parasympathetic balance, and > 3 indicates sympathetic predominance (stress, activation)48.

Statistical analysis

Statistical analyses were performed based on data distribution. Normality was tested using Shapiro–Wilk; normally distributed data were analyzed with repeated measures ANOVA, while non-normal data used the Friedman test. Post-hoc tests identified specific group differences. Significance was set at α = 0.05 with Bonferroni correction to control Type I error.

Effect sizes were calculated using Hedges’ g, categorized as small (0.0–0.2), medium (0.2–0.8), and large (> 0.8). Positive g values indicate a higher mean in the first group, negative values indicate a lower mean, and effect strength was based on the absolute value. Final conclusions considered both adjusted p-values and effect size magnitudes to assess statistical and practical significance. All statistical analyses were performed using GraphPad Prism software (version 10.1; GraphPad Software, San Diego, CA, USA).

Results

Figure 1 shows the comparison of the distributions of RF, BZ, Rhmax, and TP across different AOP levels (40%, 80%, 100%, and 130%). There were no significant differences in RF values across AOP levels (F(3, 87) = 0.27, p = 0.847; ηp2 = 0.00036). The Friedman test revealed statistically significant differences for BZ (p < 0.001), Rhmax (p < 0.001), and TP (p < 0.001) across AOP levels. Figure 1 presents the pairwise comparisons indicating these significant differences, while Supplementary Material 1 provides detailed p-values and Hedges’ g effect sizes for each comparison.

Fig. 1
figure 1

Comparison of the distribution of (a) resting flow (RF) values; (b) biological zero (BZ); (c) peak hyperemia (Rhmax); and (d) time to peak (TP) across different levels of arterial occlusion pressure (AOP) (40%, 80%, 100%, and 130%). *Indicates significant differences in pairwise comparisons.

Full size image

Specifically, BZ values differed significantly between 40 and 80 (p < 0.001, Hedges’ g = 3.72), 40 and 100 (p < 0.001, g = 9.07), 40 and 130 (p < 0.001, g = 9.10), 80 and 100 (p < 0.001, g = 7.37), and 80 and 130 (p < 0.001, g = 7.42). Similarly, significant differences in Rhmax were observed between 40 and 80 (p < 0.001, Hedges’ g = − 6.06), 40 and 100 (p < 0.001, g = − 8.11), 40 and 130 (p < 0.001, g = − 8.36), 80 and 100 (p < 0.001, g = − 5.44), and 80 and 130 (p < 0.001, g = − 5.60). Significant differences in TP were also found between 40 and 80 (p < 0.001, Hedges’ g = − 2.65), 40 and 100 (p < 0.001, g = − 4.64), 40 and 130 (p < 0.001, g = − 5.44), 80 and 100 (p < 0.001, g = − 3.46), and 80 and 130 (p < 0.001, g = − 4.26).

Figure 2 shows the comparison of the distributions of AVNN, SDNN, LF-HF and HR across different AOP levels (40%, 80%, 100%, and 130%). The Friedman test revealed statistically significant differences across AOP levels for AVNN (p < 0.001), SDNN (p < 0.001), LF/HF ratio (p < 0.005), and HR (p < 0.001). Figure 2 presents the pairwise comparisons indicating these significant differences, while Supplementary Material 1 provides detailed p-values and Hedges’ g effect sizes for each comparison.

Fig. 2
figure 2

Comparison of the distribution of (a) average NN intervals (AVNN) values; (b) Standard Deviation of NN Intervals (SDNN); (c) Low/High Frequency Ratio (LF-HF); and (d) heart rate (HR) across different levels of arterial occlusion pressure (AOP) (40%, 80%, 100%, and 130%). * indicates significant differences in pairwise comparisons.

Full size image

Significant differences in AVNN were found between 40 and 80 (p < 0.001, Hedges’ g = 3.32), 40 and 100 (p < 0.001, g = 6.01), and 40 and 130 (p < 0.001, g = 6.46). Smaller, yet still significant, differences were also observed between 80 and 100 (p = 0.046, g = 0.60) and between 80 and 130 (p = 0.043, g = 0.65). For SDNN, significant differences occurred between 40 and 80 (p < 0.001, Hedges’ g = − 4.19), 40 and 100 (p < 0.001, g = − 7.88), and 40 and 130 (p < 0.001, g = − 8.46), with smaller but still significant differences found between 80 and 100 (p = 0.046, g = − 1.89) and between 80 and 130 (p = 0.043, g = − 2.00). Significant differences in the LF− HF ratio were found between 40 and 80 (p < 0.001, Hedges’ g = − 5.43), 40 and 100 (p < 0.001, g = − 7.48), and 40 and 130 (p < 0.001, g = − 7.58), as well as between 80 and 100 (p < 0.001, g = − 2.45) and between 80 and 130 (p < 0.001, g = − 2.69). Differences in HR were observed between 40 and 80 (p < 0.001, Hedges’ g = − 0.35), 40 and 100 (p < 0.001, g = − 1.31), 40 and 130 (p < 0.001, g = − 1.47), 80 and 100 (p < 0.001, g = − 0.82), and 80 and 130 (p < 0.001, g = − 0.96), as well as between 100 and 130 (p < 0.010, g = − 0.19).

Discussion

This study shows that varying levels of arterial occlusion pressure (AOP) have distinct effects on both microcirculatory and autonomic responses. While rest flow (RF) remained unchanged across all AOP levels, significant differences were observed in biological zero (BZ), peak hyperemia (Rhmax), time to peak (TP), and heart rate variability (HRV) measures including AVNN, SDNN, LF-HF ratio, and heart rate (HR). The most substantial physiological changes occurred between the lowest pressure (40 mmHg) and higher levels (80–130 mmHg), with large effect sizes across parameters. However, comparisons between 100 and 130% AOP consistently revealed no significant differences, suggesting a plateau or stabilization in microvascular and autonomic responses at higher occlusion pressures.

HRV analysis revealed that increasing AOP significantly affected all examined HRV parameters—AVNN, SDNN, LF-HF ratio, and HR. The most pronounced changes occurred between 40 mmHg and higher AOP levels (80, 100, 130 mmHg), with large effect sizes indicating strong autonomic modulation. However, comparisons between 100 and 130% AOP showed no significant differences across these measures, suggesting a saturation point or plateau in autonomic response at higher occlusion levels. Previous study49 found that greater arterial occlusion pressure in blood flow restriction training increased cardiovascular and metabolic stress, but also mean arterial pressure, indicating high myocardial workload. Moreover, applying higher relative pressures resulted in greatest cardiovascular response in a study comparing different AOP levels50.

The observed saturation point or plateau in autonomic response at higher AOP levels (between 100 and 130% AOP), despite significant changes occurring between 40 mmHg and higher AOP levels (80, 100, 130 mmHg), likely reflects a maximal or near-maximal stimulation of the mechanoreceptors involved in the autonomic response to vascular occlusion51. As AOP increases, baroreceptors and other pressure-sensitive afferent nerve endings within the vasculature are increasingly stimulated4, leading to adjustments in AVNN, SDNN, LF-HF ratio, and HR, indicative of autonomic modulation52. However, these receptors likely have a finite capacity to respond to increasing pressure. Once a sufficiently high pressure was reached (around 100% AOP in this context), further increases in occlusion pressure may not elicit a proportionally greater increase in afferent signaling53. This plateau effect could be due to the receptors reaching their maximal firing rate or the downstream neural pathways becoming saturated in their ability to process and relay the increasing afferent information. Importantly, the pressure at which this saturation occurs is likely more indicative of the diastolic pressure rather than the mean arterial pressure, as diastolic pressure better represents the baseline arterial load sensed by the receptors during the cardiac cycle54.

RF remained unchanged across all AOP levels, indicating that baseline microvascular perfusion is not significantly influenced by moderate variations in external pressure. In contrast, BZ, Rhmax, and TP all showed significant differences between conditions, particularly between lower (40 mmHg) and higher AOP levels. BZ, Rhmax, and TP values plateaued at 100% and 130% AOP, with no significant differences observed between these two levels, suggesting a ceiling effect in vascular occlusion-induced microvascular suppression and subsequent reactive hyperemia. For instance, a previous study found that blood flow response to low-load exercise with and without blood flow restriction is pressure-dependent, with higher pressures reducing resting hyperemia55.

BZ likely reflects tissue compression effects and capillary signal attenuation, which intensify with higher pressures56. The marked changes in Rhmax and TP reflect enhanced PORH, driven by the accumulation of metabolic vasodilators during occlusion and their sudden release upon reperfusion57. The longer TP and higher Rhmax at greater AOP levels suggest a more profound ischemic challenge and stronger rebound vasodilation. However, the absence of further increase beyond 100% AOP implies maximal vasodilatory capacity may have been reached56. These findings align with previous studies indicating that the PORH response is pressure- and duration-dependent55,58 but may saturate once the ischemic stimulus exceeds physiological thresholds.

This study had some limitations that should be acknowledged. First, the sample size and demographic characteristics may limit the generalizability of findings to broader or clinical populations. Additionally, due to recruitment limitations, participant inclusion was unbalanced in favor of men, which should be acknowledged as a study limitation. Second, while multiple AOP levels were tested, the discrete pressure increments (40, 80, 100, 130 mmHg) may not capture subtler thresholds of physiological change or individual variability in pressure tolerance. Third, the design prevents long-term inferences about adaptation to repeated occlusion exposures, such as those encountered BFR training. Additionally, although HRV and microvascular measures were assessed, the study did not include biochemical or neurohormonal markers that could provide deeper insight into the mechanisms underpinning the observed plateau effects at higher AOP levels.

The results of this study suggests that BFR training at pressures between 80 and 100% AOP elicits significant changes in both microcirculatory (BZ, Rhmax, TP) and autonomic (AVNN, SDNN, LF-HF ratio, HR) responses compared to low pressure (40 mmHg), suggesting that moderate to high occlusion pressures are effective in stimulating physiological adaptations relevant to BFR. However, no further significant differences were observed between 100 and 130% AOP across these measures, indicating a plateau in response. This suggests that applying occlusion pressures above 100% AOP may not provide additional benefit in terms of vascular or autonomic stimulation. Additionally, resting flow remained stable across all AOP levels, indicating that baseline perfusion is maintained even at higher pressures.

While this study focuses on BFR training, the findings may also be relevant to clinical settings such as orthopedic surgery, where arterial occlusion cuffs are applied at high pressures for extended periods. Our results suggest that microvascular and autonomic responses reach a plateau around 100% AOP, indicating a limit to physiological adaptation. In surgical cases, patients are often unprepared for such occlusion and reperfusion stresses, which can lead to meaningful hemodynamic changes after cuff release. Understanding the saturation of these responses may help explain some of the cardiovascular impacts observed post-surgery. Additionally, the stable resting flow at moderate pressures suggests that baseline perfusion is maintained during occlusion, but longer or higher pressure applications could exceed this capacity. These findings might inform safer cuff pressure use during surgery and highlight the potential value of preconditioning strategies to improve patient tolerance to vascular occlusion.

Conclusions

The findings show that increasing AOP produces significant changes in Rhmax, TP, BZ, and HRV indices (AVNN, SDNN, LF-HF ratio, and HR)—particularly when comparing low pressure (40 mmHg) to moderate and high pressures (80–100% AOP). However, responses plateaued between 100 and 130% AOP, suggesting a saturation point in both microvascular and autonomic modulation. These results highlight a pressure-dependent relationship up to a physiological ceiling, beyond which further occlusion does not elicit additional response. The results provide new findings into the dose–response characteristics of vascular and autonomic systems under occlusive stress and offers suggestions for optimizing AOP in blood flow restriction applications.