“CLINICAL-BFR”: An exploratory blood flow restriction protocol focusing on clinical application

Abstract

Purpose

Traditional blood flow restriction resistance training (BFR-Trad) is a cost-effective and time-efficient exercise modality that offers significant skeletal muscle benefits, including strength and mass gains. Despite employing low loads, BFR-Trad may not be well tolerated by all populations due to its high intensity, pain-associated muscle discomfort, low affectivity, and cardiovascular responses. Consequently, reducing BFR intensity may be an interesting approach to making traditional BFR more clinically applicable. This was approached through the proposal of the present Clinical BFR.

Methods

Twenty-two healthy, untrained subjects (both female and male) aged between 18 and 30 years participated. Subjects completed three sessions of unilateral knee extension exercises randomly assigned to the BFR-Trad, BFR-Clinical_c (cadenced), and BFR-Clinical_a (-autoregulated) exercise protocols.

Results

The BFR-Trad condition induced greater increases in perceived exertion, discomfort, delayed onset muscle soreness (DOMS), and lactate levels compared to the BFR-Clinical conditions (p ​< ​0.05). Muscle concentric velocity showed greater increases in the BFR-Clinical_a condition (p ​< ​0.001), along with higher affectivity ratings for the exercise compared to the BFR-Clinical_c and BFR-Trad conditions (p ​< ​0.05).

Conclusion

Compared to traditional BFR, clinical BFR was scored as less intense, reduced pain-associated discomfort, decreased DOMS, increased affectivity, diminished systolic blood pressure responses, decreased fatigue and muscle activation, and promoted similar glycemic responses. Altogether, our results support the potential use of clinical BFR in clinical settings.

Abbreviation list

1 RM

One-repetition maximum test

AOP

Arterial occlusion pressure

ATP

Adenosine Triphosphate

BFR

Blood flow restriction resistance training

BFR-Trad

Traditional blood flow restriction resistance training

BFR-Clinical_c (cadenced)

Blood flow restriction resistance training cadenced

BFR-Clinical_a (autoregulated)

Blood flow restriction resistance training autoregulated

BMI

Body Mass Index

BP

Blood pressure

DBP

Diastolic blood pressure

DOMS

Delayed onset muscle soreness

EPR

Exercise Pressor Reflex

FFM

Fat-free mass

HR

Heart rate

IPAQ

International Physical Activity Questionnaire

PAR-Q

Physical Activity Readiness Questionnaire

RFD

Rate of force development

RPE

Ratings of perceived exertion

SBP

Systolic blood pressure

sEMG

Surface electromyography

VAS

Visual Analog Scale

1. Introduction

Traditional blood flow restriction resistance training is a cost-effective and time-efficient exercise modality that provides significant skeletal muscle benefits, including strength and mass gains, to a wide range of subjects, from the elderly to patients undergoing rehabilitation.^1,2 Since it employs low load and low exercise volume, it is well-suited for skeletal muscle rehabilitation in healthy individuals, including those recovering from immobilization, knee joint surgery, and arthrosis.3, 4, 5 However, traditional BFR, despite employing low loads, may not be well tolerated by all populations for several reasons: 1) it is intense and generally of maximal nature, 2) it provokes moderate to high acute pain-associated muscle discomfort, 3) it causes reasonably delayed onset muscle soreness (DOMS) responses, 4) subjects usually demonstrates low affectivity to it, and 5) promotes disproportional increases in cardiovascular responses to the low load employed.6, 7, 8

A primary concern regarding traditional BFR is its high-intensity-mediated physiological effects, even when using low loads.⁹ This can be partly attributed to the muscle hypoxia occasioned by vascular restriction, which forces skeletal muscle to seek ATP reinforcements in anaerobic metabolism, including glycolysis.⁹ Enhanced glycolysis coupled to muscle failure (or near-muscle failure) leads to the production and accumulation of hydrogen ions within the muscles, stimulating class III/IV afferents.¹⁰ Activation of these metaboreceptors is associated with a local burning sensation that generates pain-associated discomfort and increased cardiovascular response (via the exercise pressor reflex - EPR).⁷ Collectively, such responses may make it less tolerable to debilitated clinical populations. Consequently, reducing BFR intensity may reduce glycolysis-associated muscle pain/discomfort and cardiovascular responses, making it more clinically applicable.

Therefore, the primary objective of this study was to develop clinically applicable BFR protocols intended to reduce exercise intensity compared to a traditional BFR exercise. Altogether, our results support the potential use of clinical BFR in clinical settings, which can be implemented at the onset of a rehabilitation program or whenever a reduction in exercise intensity may be necessary. The comprehensive approach to investigating muscle performance, muscle physiology, muscle metabolism, and psychological responses may provide health professionals with a deep understanding of the new methodology, facilitating the identification of the population of their interest that fits better with it.

2. Methods

2.1. Study design

This was an experimental cross-sectional study conducted over four weeks. Participants were instructed to visit our laboratory after abstaining from physical exercise and alcoholic beverages for at least 24 ​h. The study comprised sixteen visits: one for baseline assessments, two for familiarization with procedures, one for a one-repetition maximum (1 RM) test, 3 experimental sessions spaced 7 days apart, and subsequent retesting visits.

During the first visit, informed consent was obtained, and participants completed the International Physical Activity Questionnaire (IPAQ, short version) and Physical Activity Readiness Questionnaire (PAR-Q). Anamnesis and baseline measurements were conducted, including weight, height, blood pressure (BP), and heart rate (HR). Additionally, anthropometric measurements, body composition assessment, and determination of arterial occlusion pressure (AOP) were performed.

The second visit involved familiarizing the individuals with unilateral knee extension exercises using a leg extension machine with and without inflatable cuffs. This session aimed to familiarize participants with the 1 RM test, BFR exercise, and scales such as ratings of perceived exertion (RPE), discomfort (CR10 scale), pain (Visual Analog Scale - VAS), and affectivity scale. Sessions were designed to minimize the influence of exercise learning on subsequent tests.¹¹ A draw was conducted to determine the exercise order for the exercise protocol: BFR-Trad, BFR-Clinical_c, BFR-Clinical_a.

During the third visit, participants performed the 1 RM test in both legs to determine maximum concentric strength and establish the protocol's exercise load (30% of 1 RM).

Subsequent visits involved evaluating BP, HR, delayed-onset muscle soreness (DOMS), and lactate levels before and after the exercise protocols (BFR-Trad, BFR-Clinical_c, BFR-Clinical_a). Each exercise protocol occurred during separate sessions, spaced seven days apart to allow for adequate recovery and wash-out periods. Participants rated RPE, discomfort, and affectivity scales during exercises. Post-exercise measurements included blood pressure, heart rate, and lactate levels. All test and exercise sessions were conducted between 1 p.m. and 6 p.m. After completing the experimental procedures, participants returned to the laboratory for 24- and 48-h post-exercise for repeated blood pressure measurements, heart rate, and DOMS. The experimental design, timing of measurements, and tests are summarized in Fig. 1.

Fig. 1.

a) Participants screening; b) Consort flow chart.

2.2. Ethical approval

This study was approved by the Research Ethics Committee of the Federal University of Maranhão (CAAE: 59371522.2.0000.5087), and all research was conducted by the 2013 revision of the Declaration of Helsinki. Written consent forms were obtained from all participants.

2.2. Participants

Out of 38 participants who participated in the first screening, 27 healthy and strength-untrained subjects met all inclusion criteria and agreed to participate in this study. Five participants withdrew from the research during the intervention for personal reasons (n ​= ​5). Thus, 22 healthy, untrained undergraduate students (9 male and 13 female) from the Federal University of Maranhão completed all laboratory visits. Fig. 1. b) shows the recruitment procedures (CONSORT flowchart). All subjects read and signed the informed consent form. The inclusion criteria for the study were: 1) BMI (Body Mass Index) between 18.00 and 24.99 ​kg ​m⁻² for eutrophic subjects, 2) total absence of a history of use of anabolic androgenic steroids, 3) no use of prescription medications (e.g., antidepressants, antibiotics, anti-inflammatories), 4) no pregnant or nursing women, 5) not having used a dietary supplement, ergogenic, or illicit drug in the last two years before the study, 6) not being a smoker or using of alcoholic beverages one week before the test visits and during the study, 7) not having illnesses and pathologies. The characteristics of the participants are presented in Table 1. There were no significant differences between any variables in the baseline.

Table 1.

General characteristics of the sample.

Variables	Subjects
	(n ​= ​22)
Age (years)	23.14 ​± ​3.18
Height (m)	169.3 ​± ​8.81
Body mass (kg)	65.90 ​± ​11.79
BMI (kg·m−2)	22.84 ​± ​2.52
SBP (mmHg)	114.77 ​± ​12.16
DBP (mmHg)	69.00 ​± ​6.71
Heart rate (bpm)	72.59 ​± ​9.43

Data are presented as mean ​± ​standard deviation. BMI: body mass index; SBP: systolic blood pressure; DBP: diastolic blood pressure; kg: kilogram; m²: square meters; m: meters; mmHg: millimeters of mercury; bpm: beats per minute.

2.4. Procedures

2.4.1. Anthropometric measurements and body composition

Total body weight, height, and total circumference (shoulder, chest, waist, abdomen, hip, arms, forearms, thighs, and legs) were measured. In addition, the subjects' body mass index (BMI) was calculated by dividing body weight in kilograms by height in meters squared (kg·m⁻²). Total body weight was measured on an electronic scale (Filizzola PL 50, Filizzola Ltda., Brazil) with a precision of 0.1 ​kg, and height was measured on a fixed stadiometer (Sanny, São Paulo, Brazil) with an accuracy of 0.1 ​cm and a length of 2.20 ​m. A metallic measuring tape (Sanny, São Paulo, Brazil) was used for circumference measurements. All measurements were taken before the first exercise session. According to the manufacturer's user manual, a single-trained ultrasound technician performed the body composition assessment using the Bodymetrix Pro System (Intelametrix Inc., Livermore, CA, USA). Total fat-free mass (FFM) was then estimated by subtracting fat mass from body mass (Total body mass minus total fat mass ​= ​FFM), utilizing the Jackson & Pollock at the following points: 1) subscapular; 2) triceps; 3) chest; 4) middle axillary; 5) supra iliac; 6) abdominal; 7) thigh.

2.4.2. Determination of arterial occlusion pressure (AOP)

A scientific cuff measuring 12.5 ​cm ​× ​84 ​cm (7 ​cm ​× ​52 ​cm inflatable chamber, by Cardiomed, Brazil) was placed around the proximal region of the thigh. Subsequently, a portable Doppler device (MEDMEGA - DV 610V, Brazil) had the tip of its transducer immersed in conductive gel (Mercur®, Brazil) and then positioned on the posterior tibial artery to detect the arterial pulse. Next, the cuff was inflated to 50 ​mmHg and then deflated. Afterward, it was inflated again until the subject's systolic pressure value was obtained, following gradual increases of 10 ​mmHg (pausing for 5 ​s at each point). This procedure was carried out until the auscultatory pulse was interrupted. After interrupting the auscultatory pulse, a 30-s pause was performed to ensure that there would be no return of arterial pulsation.¹² The AOP obtained was the lowest cuff pressure at the end of pulse elimination. The cuff pressure was released immediately after AOP determination. To carry out the BFR training protocol, 50% of this value (restriction pressure) was calculated. All AOP measurements were performed once at the pre (basal) moment.¹² These procedures' re-test intraclass correlation coefficient proved reliable (ICC ​= ​0.91).

2.4.3. Dynamic strength test

In the leg extension exercise, the subjects performed the one-repetition maximum (1 RM) unilateral test to measure their maximum concentric strength in both legs. The starting weight for the maximum test was estimated during the familiarization sessions. After that, under the supervision of a qualified instructor, the individuals performed a specific warm-up exercise with 5–6 repetitions at approximately 30%–40% of the estimated 1 RM, after which they performed two sets of three repetitions with a load corresponding to approximately 80% of 1 RM. The load progressively increased by 5%–10% until the subjects could no longer perform the entire movement correctly, which was recorded as 1 RM. The 1 RM test was considered valid only if the flexion and extension of the knees occurred between 90° of flexion (beginning) and the complete extension of the knee at 0° (final).¹³

2.4.4. Experimental procedures

Before the training week, all subjects were meticulously familiarized with the three exercise protocols BFR-Trad ​× ​BFR-Clinical_c ​× ​BFR-Clinical_a in the leg extension machine. Each exercise condition was performed only once a week, with each protocol separated by seven days. Subjects were instructed to perform each repetition with full extension and flexion of the movement. The subjects were trained in the leg extension chair unilaterally (Performance®, Brazil), with the previously described pneumatic cuff placed in the proximal region of the quadriceps (thigh, unilaterally) and restriction pressure at 50%. Soon after, they performed a warm-up with ten repetitions at 20% of 1 RM, without restriction pressure, resting for 2 ​min. Next, they performed the BFR-Trad protocol, with three sets of leg extension exercise until concentric failure, at 30% of 1 RM, and a 1-min interval between sets. All subjects controlled the speed of the exercise execution in a cadenced manner, under the aid of a metronome at 1.5 ​s for the eccentric phase and 1.5 ​s for the concentric phase. Blood flow restriction was maintained throughout the exercise session and released immediately after the protocol's end, ensuring the training's precision and consistency. The following week, seven days after the end of the BFR-Trad protocol, the subjects were submitted to the BFR-Clinical_c (cadenced) in the contralateral leg and, after another week, the BFR-Clinical_a (autoregulated) in the same leg of BFR-Trad (Fig. 2a).

Fig. 2.

a) Scheme for the leg selection to the acute Blood flow restriction exercise protocol. A leg (thick right arrow) was drawn and selected for the first exercise (Traditional blood flow restriction exercise = BFR-Trad, cadenced). After, the opposite (thick left arrow) leg performed the blood flow restriction exercise, clinical cadence BFR-Clinical_c. Last, the same leg of BFR-Trad (thick right arrow) performed the BFR-Clinical_a, auto-regulated version. b) Blood flow restriction exercise protocol characteristics. BFR-Trad ​= ​Traditional; BFR-Clinical_c ​= ​Clinical, cadenced; BFR-Clinical_a ​= ​Clinical, auto-regulated.

In the BFR-Clinical versions (cadenced and autoregulated), the subjects trained in the same manner as BFR-Trad regarding load (30% of 1 RM), number of repetitions, and cuff pressure (50% of AOP).^14,15 The only two differences were that: 1) In the BFR-Clinical versions, repetitions were not to fail in opposition to the BFR-Trad (Figure below). 2) BFR-Clinical versions consisted of the same number of total repetitions from the BFR-Trad; however, they were divided into six (6 sets, with a 1-min interval between sets. (Fig. 2b). All subjects controlled the exercise speed through a research staff in a cadenced way, with the help of a metronome, at 1.5 ​s for the eccentric phase and 1.5 ​s for the concentric phase. Blood flow restriction was maintained throughout the exercise session and released immediately after the end of the protocol. Seven days after the end of the BFR-Clinical_c protocol, the subjects were subjected to the same exercise conditions, performing the BFR-Clinical_a exercise protocol without using a metronome, allowing subjects to choose their own pace and totaling the same number of total repetitions at the end of the protocols.

2.4.5. Assessment of ratings of perceived exertion (RPE)

The Borg RPE scale, a modified scale comprising 6 to 20 points, adding 15 points, has been used to measure the perceived exertion or level of physical exertion during exercise.^16,17 A rating of 6 on the scale represents no effort or “very easy,” and a rating of 20 represents maximum effort or “exhausting”.¹⁸ The RPE scale was applied during the test's familiarization week before, during, and after each exercise series was performed. All subjects received verbal guidance on the classification of RPE, as determined above. For the RPE classification, the subjects were instructed to evaluate their perception of exertion to answer how heavy and intense the exercise felt.¹⁹

2.4.6. Discomfort assessment (CR-10)

Like the RPE, the subjects' discomfort scale (CR10) was applied to identify the perceived discomfort after exercise. First, they were performed in the test familiarization week, before, during, and after each exercise series. All subjects were instructed to describe their worst experiences of discomfort. “The referred maximum discomfort is your primary reference point compared to your worst pain experience.” If, for example, it is not your worst discomfort experienced, with a pain level higher than 10, the subjects could report a higher number, such as 11 or 12. If it is higher (1.5 times higher), it could report 15.¹⁹

2.4.7. Evaluation of delayed onset muscle soreness (DOMS)

The subjects' delayed onset muscle soreness was measured using the visual analog scale (VAS) before and after performing the exercises. The subjects were instructed to classify the pain felt in a line from 0 to 10 ​cm, represented by two primary descriptors: 0 “no pain” and 10 “unbearable pain,” after performing palpation of the rectus femoris muscle (primary muscle exercised in the study) and knee flexion with the help of an experienced evaluator. The VAS is considered reliable and can be used to evaluate pain in other ways. The subjects were assessed pre-, 24 h, and 48 h after exercise.²⁰

2.4.8. Assessment of perceived affectivity

The subjects were instructed to respond to the affectivity scale to define the level of satisfaction and pleasure in the exercise. The responses were registered in the familiarization, before and after each series during the exercise sessions, according to Hardy and Rejewski.²¹ When performing exercises, it is common for individuals to change their perception of affectivity, be able to feel good or bad during the execution, and reflect on their feelings about a particular exercise. The affectivity scale concerns a bipolar scale with 11 points ranging from +5 to −5, generally used to measure affectivity responses, such as feelings of pleasure and displeasure, during the exercise performance. The scale is represented by a numeric sequence: −5 ​= ​very bad; −3 ​= ​evil; −1 ​= ​fairly bad; 0 ​= ​neutral; +1 = fairly well; +3 ​= ​good; and +5 ​= ​very good. All subjects received standardized guidelines regarding their use.²¹

2.4.9. Lactate levels

Lactate levels, a measure of internal training intensity,²² were determined by measuring blood lactate before, 5, and 10 ​min after the end of the exercise session (with the cuffs deflated immediately after the exercise). Lactate was measured in blood samples extracted from the fingertips, using a lancet (Accu-Chek Active, Roche®, Brazil) and lactate strips (Accusport BM-Lactate), which were immediately analyzed in a portable lactometer (Accutrend Plus, Roche, Brazil) with proven scientific validation.²³ The lactacidemia analysis was used as a biochemical indicator of exercise intensity.

2.4.10. Glucose levels

Subjects were instructed to keep their pattern meal before the exercise tests. They were euglycemic in resting conditions, and at the moment of exercise, they were in a post-prandial state for at least 2 ​h. Blood glucose was measured immediately through blood collected from the fingertip using strips and a digital glucometer (ACCU-CHEK Performa, Roche).²⁴ Blood was collected at the following times: pre-exercise, immediately after exercise, 10 ​min after exercise, 30 ​min after exercise, 45 ​min after exercise, and 60 ​min after exercise.

2.4.11. Determination of average speed

To determine the execution speed of each repetition during the BFR-Trad, BFR-Clinical_c, and BFR-Clinical_a protocols, a linear displacement system was used (Tendo Weightlifting Analyzer System®, Slovak Republic). This system has a cable that, once attached to the lever of the leg extension chair, allows the measurement of its displacement through the appropriate software (Tendo Weightlifting analyzer®, 3.6.15). Once the weight lifted at each repetition is known, the equipment automatically measures the average, peak velocity, and peak power.²⁵

2.4.12. Determination of muscle activation

Surface electromyography (sEMG) was performed in a subset of four individuals during the BFR-Trad, BFR-Clinical_c, but not BFR-Clinical_a protocol, to determine muscle activation. Double differential sEMG electrodes (Lessa Electrodes, Brazil) were placed on the vastus lateralis. Before placing the EMG electrodes, the skin was shaved and cleansed with an alcohol preparation pad. The EMG electrode for the vastus lateralis was placed approximately 3–5 ​cm superior to the patella at an oblique angle just lateral to the midline of the muscle, parallel to the direction of the muscle fibers, following the procedures by Lauver et al.²⁶ The distance from the patella to the sEMG electrode was recorded during the first experimental condition and replicated for each subsequent condition.²⁶ The reference electrode was placed on the head of the fibula, and electromyography was measured during the exercise. The EMG signal was obtained using Miograph software (Miotec, Brazil), with an acquisition rate of 2 000 ​Hz, and later analyzed using the Miotool from Miotec software. Signal processing included filtering with a Butterworth band-pass filter at 20–500 ​Hz cutoff frequency. An electro-goniometer (Miotec, Brazil) was placed in the knee joint, in the same leg where the electrodes were placed to measure joint angle (°), according to Jenkins et al.²⁷ The electro-goniometer signal was low-pass filtered (zero-phase shift fourth-order Butterworth filter) with a 15 ​Hz cutoff, and the subsequent analysis was performed on the filtered signals.²⁷ The EMG values were calculated from the signal epochs corresponding to the 90° range of motion occurring between 80° and 165° of leg extension (165° ​= ​full extension). Only the concentric portion of each repetition was used to analyze the EMG signal. The EMG values from the first repetition were not used, and all subsequent repetitions were considered. Excluding the first repetition, the mean value of the following three repetitions was normalized to 100% values, and the subsequent repetitions were expressed in terms of the initial percentage values for each exercise protocol.²⁸ The total number of repetitions was also normalized in percentage to facilitate comparison between protocols, as Burd et al.²⁸

2.5. Statistical analysis

A power analysis used the effect size for delayed onset muscle soreness (${\eta }_p^2$ ​= ​0.29), with an alpha value of 5%. Our sample size achieved 99% of the power size determined by GPower 3.1 software. The data normality was verified using the Shapiro-Wilk test. The estimated sphericity was verified according to Mauchly's W test, and the Greenhouse–Geisser correction was used when necessary. Therefore, data are presented as means ​± ​standard deviations. Three conditions (BFR-Trad ​× ​BFR-Clinical_c ​× ​BFR-Clinical_a) ​× ​5 times (rest, immediately post-exercise, 15 ​min post-exercise, 24 and 48 ​h post-exercise) repeated measures analysis of variance (ANOVA) with the Bonferroni adjustment for multiple comparisons was used to compare hemodynamics. Three conditions (BFR-Trad ​× ​BFR-Clinical_c ​× ​BFR-Clinical_a) ​× ​6 times (rest, immediately post-exercise, 10-, 30-, 45-, and 60-min post-exercise) repeated measures ANOVA with the Bonferroni adjustment for multiple comparisons was used to compare glycemia. A 3-condition (BFR-Trad ​× ​BFR-Clinical_c ​× ​BFR-Clinical_a) ​× ​3 times (rest, 5 and 10 ​min post-exercise) repeated measures ANOVA with the Bonferroni adjustment for multiple comparisons was used to compare lactate. To verify the difference between conditions immediately post-exercise on late-onset muscle pain (DOMS), affectivity scale, muscle discomfort, RPE, EMG, and speed (power) were used as a repeated-measure ANOVA. The partial eta squared (${\eta }_p^2$) was reported for time. Statistical significance was set at p ​< ​0.05. The data were analyzed using the Statistical Package for Social Sciences 29.0 (SPSS Inc., Chicago, IL., USA).

3. Results

Twenty-two subjects who underwent three exercise conditions at different times (BFR-Trad ​× ​BFR-Clinical_c ​× ​BFR-Clinical_a) were evaluated. All subjects completed all tests, measurements, and exercise sessions without interruptions. There were no significant differences between any variables at baseline. The volunteers' physical activity level (via IPAQ - short version) was classified as low, and the fat percentage remained within baseline values without significant changes. The total AOP was (210 ​± ​23.90) mmHg, and the pressure corresponding to 50% of the AOP used during the exercises was (105 ​± ​11.95) mmHg.

Fig. 3 presents the comparison between BFR exercise conditions in hemodynamic assessment (systolic blood pressure [SBP], diastolic blood pressure [DBP], and heart rate [HR]).

Fig. 3.

Cardiovascular parameters.

When examining systolic blood pressure (SBP) (Fig. 3), a notable condition ​× ​time interaction (F ​= ​2.186, p ​= ​0.031, ${\eta }_p^2$ ​= ​0.09) was observed. The BFR-Trad condition showed a higher systolic blood pressure immediately post-exercise ([133.9 ​± ​18.8] mmHg) compared to BFR-Clinical_c ([126.2 ​± ​17.4] mmHg, p ​= ​0.002) and BFR-Clinical_a ([125.8 ​± ​17.4] mmHg) (p ​= ​0.002). However, no statistical differences were found between the conditions BFR-Clinical_c versus BFR-Clinical_a (p ​= ​1.00). For diastolic blood pressure (DBP), a main effect of condition (F ​= ​4.035, p ​= ​0.025, ${\eta }_p^2$ ​= ​0.16) and a main effect of time (F ​= ​3.739, p ​= ​0.008, ${\eta }_p^2$ ​= ​0.15) were observed, with higher DBP immediately post-exercise for all conditions analyzed. Still, no interaction was observed (p ​> ​0.05). For heart rate (HR), there was a main effect of time with a significant increase immediately post-exercise for all conditions (p ​> ​0.05) but without significant interaction.

Fig. 4 compares the BFR exercise conditions in blood lactate concentration and glycemia. For lactate concentration, there was a condition ​× ​time interaction (F ​= ​14.663, p ​< ​0.001, ${\eta }_p^2$ ​= ​0.41) with higher values in the BFR-Trad, at times: post5' ​= ​[6.3 ​± ​1.3] mmol/dl and post-10' ​= ​[5.9 ​± ​1.3] mmol/dLl in relation to the BFR-Clinical_c (moment: post5' ​= ​[4.6 ​± ​1.1] mmol/dl and Post-10' ​= ​[4.4 ​± ​1.1] mmol/dl, p ​< ​0.001) and BFR-Clinical_a (post-5' ​= ​[4.9 ​± ​1.2] mmol/dl and post-10' ​= ​[4.5 ​± ​1.0] mmol/dl, p ​< ​0.001). For glycemia, there was a main effect of time (F ​= ​5.350, p ​= ​0.001, ${\eta }_p^2$ ​= ​0.35) with lower values post-exercise for all conditions analyzed, but no interaction was observed.

Fig. 4.

Comparison of the Blood flow restriction exercise conditions regarding blood lactate concentration and glycemia.

For blood glucose, there was a main effect of time with a significant decrease after exercise (F ​= ​5.350, p ​= ​0.001, ${\eta }_p^2$ ​= ​0.35), but no interaction was observed (F ​= ​0.600, p ​= ​0.810, ${\eta }_p^2$ ​= ​0.06).

Fig. 5 compares BFR exercise conditions regarding perceived affectivity and discomfort and ratings of perceived exertion.

Fig. 5.

Perceived affectivity, discomfort, and perceived exertion ratings.

Significantly, there was a difference between conditions for affectivity post-exercise (F ​= ​7.082, p ​= ​0.002, ${\eta }_p^2$ ​= ​0.25). This difference was particularly pronounced between the BFR-Trad condition (0.14 ​± ​2.5) and the BFR-Clinical_a condition (1.32 ​± ​2.6, p ​= ​0.004), underscoring the impact of the BFR exercise conditions on affectivity. However, no significant difference was observed between BFR-Clinical_c conditions (p ​= ​0.512). In the perception of muscle discomfort of subjects during exercise, there was a tendency to a significant difference (F ​= ​3.036, p ​= ​0.059, ${\eta }_p^2$ ​= ​0.13) with higher values observed in the BFR-Trad condition (6.5 ​± ​2.4) ​× ​BFR-Clinical_c (5.8 ​± ​2.0) ​× ​BFR-Clinical_a (5.3 ​± ​2.5). Of particular note, in the RPE evaluated immediately post-exercise, there was a significant difference between conditions (F ​= ​5.430, p ​= ​0.008, ${\eta }_p^2$ ​= ​0.20). This difference was marked by a higher increase in effort perception in the BFR-Trad condition (16.9 ​± ​1.8) compared to BFR-Clinical_c (15.4 ​± ​2.5, p ​= ​0.002) and BFR-Clinical_a (15.2 ​± ​2.8, p ​= ​0.029), further highlighting the impact of the BFR exercise conditions on perceived exertion.

Fig. 6 compares the BFR exercise conditions at the average speed (power). For average speed, there was significant difference between conditions (F ​= ​9.723, p ​< ​0.001, ${\eta }_p^2$ ​= ​0.32) with higher increase in speed in the BFR-Clinical_a ([0.49 ​± ​0.06] m·s⁻¹) compared to BFR-Trad conditions ([0.43 ​± ​0.05] m·s⁻¹, p ​= ​0.011) and BFR-Clinical_c ([0.44 ​± ​0.04] m·s⁻¹, p ​= ​0.005).

Fig. 6.

Average speed, exercise density, and delayed onset muscle soreness (DOMS).

In late-onset muscle pain, represented by pain scale or analog visual scale-EVA, there was a condition ​× ​time interaction (F ​= ​8.443, p ​< ​0.001, ${\eta }_p^2$ ​= ​0.29). Post-Hoc analysis showed a greater perception of muscle pain post-24 ​h between BFR-Trad and BFR-Clinical_c (p ​= ​0.007) and BFR-Clinical_a (p ​= ​0.005) and post-48 ​h between BFR-Trad and BFR-Clinical_c (p ​= ​0.001) and BFR-Clinical_a (p ​< ​0.001), but no significant differences were found between Clinical_c and Clinical_a condition (p ​> ​0.05).

When analyzing performance, there was a significant difference between groups for exercise density (F ​= ​108.828, p ​< ​0.001, ${\eta }_p^2$ ​= ​0.85) with higher values in the BFR-trad compared to BFR-Clinical_c (p ​< ​0.001) and BFR-Clinical_a (p ​< ​0.001). Still, there was no difference between the two BFR-Clinical (p ​= ​0.634). There was no significant difference between groups for total weight lifted (F ​= ​0.655, p ​= ​0.252, ${\eta }_p^2$ ​= ​0.03).

Fig. 7 compares the BFR exercise conditions at muscle activation using sEMG. Muscle activation behavior was similar between BFR-Trad and BFR-Clinical_c during the first half of the exercise. After that, BFR-Trad demonstrated enhanced muscle activation compared to BFR-Clinical_c, and such differences were more evident at the end of the exercise. Considering the final exercise at 100% repetitions, BFR-Trad demonstrated significantly increased muscle activation compared to the BFR-Clinical_c.

Fig. 7.

Surface Electromyography (sEMG) muscle activation during an entire session of traditional Blood flow restriction (BFR-Trad) or Clinical, cadenced (Clinical_c). The onset compares the end of exercise at 100% of repetitions.

4. Discussion

In the present study, we propose an adapted BFR exercise potentially suitable for clinical settings, termed “Clinical-BFR.” On the one hand, we recognize that traditional BFR is a highly effective and cost-effective exercise model for increasing muscle mass and strength in various populations. On the other hand, traditional BFR is perceived as high-intensity exercise,⁸ substantially increasing the sense of effort and pain,⁶ causing considerable discomfort,⁸ and enhancing the muscle metaboreflex and cardiovascular responses.⁷ Such heightened intensity and perceived effort may not be ideal for individuals with weak deconditioning, frailty, or the sarcopenic elderly. These populations often exhibit intolerance to high-intensity exercise, increased sensitivity to pain and discomfort, and disproportionate increases in cardiovascular responses to exercise.^7,10 Although our Clinical BFR version has a similar external workload to the traditional one, shorter sets and increased total rest intervals are perceived as less intense, have lower exercise density, reduce pain-associated discomfort, increase affectivity, decrease delayed onset muscle soreness (DOMS), and diminish systolic blood pressure responses. Based on these findings, we suggest that the Clinical-BFR protocol may be more suitable for supra-cited individuals as well as suited to clinical settings.

As stated before, a primary concern regarding traditional BFR is its high-intensity-mediated physiological effects, even when employing low loads.²⁹ This is mainly because, during BFR exercise, vascular restriction causes muscle hypoxia, which forces skeletal muscle to seek ATP reinforcements in anaerobic metabolism, including glycolysis.⁹ Enhanced glycolysis coupled to muscle failure (or near-muscle failure) leads to the production and accumulation of hydrogen ions within the muscles, stimulating class III/IV receptors.¹⁰ Activation of these muscle metaboreceptors is associated with a local burning sensation that generates pain-associated discomfort and increases cardiovascular responses, via the exercise pressor reflex (EPR).⁷ Consequently, reducing BFR intensity may reduce glycolysis-associated muscle pain/discomfort and cardiovascular responses, making it more clinically applicable.

To accomplish these goals, we utilized the strategy to reduce BFR intensity and glycolytic flux by using shorter sets and increasing the total rest interval in Clinical BFR. This was achieved while maintaining traditional BFR essential exercise variables, such as load, total repetitions, and cuff pressure (Fig. 1). These adjustments primarily aimed to decrease exercise intensity through reduced exercise density(Figs. 6),²³ besides reducing muscle metabolic stress (i.e., lactate, Fig. 4). Indeed, as reported by the subjects, these modifications led them to perceive Clinical-BFR as less intense than traditional BFR (Fig. 5). As discussed below, BFR exercise intensity is closely linked to the discomfort associated with BFR.

Discomfort is often an underestimated aspect of BFR.⁸ While BFR discomfort may be tolerated by healthy and young individuals, particularly when motivated,⁸ it can be challenging for fragile or weak individuals, especially over multiple BFR sessions.³⁰ In this study, muscle discomfort was reduced in the clinical BFR conditions compared to the traditional conditions (Fig. 5). This can be strictly related to the diminished exercise intensity, which reduced glycolytic flux and metabolite accumulation, thus diminishing the activation of class III/IV muscle afferents. A favorable aspect of reducing discomfort associated with BFR is optimizing exercise adherence, potentially enhancing long-term benefits of chronic exercise.³¹ In line with this, the subjects' affectivity related to satisfaction and pleasure showed higher scores following the "Clinical-BFR" protocols than the traditional BFR. At least psychologically, these enhancements may favor exercise adherence and motivation to begin and progress in a BFR rehabilitation program. Interestingly, the Clinical BFR self-regulated version achieved significantly higher affectivity scores than the Clinical BFR cadence-controlled version. The only difference between these two versions is that subjects were allowed to set their exercise velocity for each repetition in the self-regulated one. It is unclear why this BFR condition was superior in this variable; metabolically speaking, the two conditions were very similar (Fig. 4). Maybe not having a person controlling their cadence or having the freedom to set their cadence may be more pleasant in this version. Regardless, increased affectivity and the lack of staff control over their cadence may make an easier-to-implement version more suitable for hospitals or clinical settings.

A BFR “Call for Concern” exists in the literature regarding the exaggerated cardiovascular responses associated with the low load employed in this exercise method, particularly for non-healthy subjects.¹⁰ These cardiovascular concerns are primarily related to the excessive accumulation of hydrogen ions in the occluded muscle during BFR, which can intensify muscle metaboreflex and cardiovascular responses, and increase cardiovascular risk in individuals with unhealthy conditions.^10,32 Considering this concern, we evaluated whether the clinical BFR versions differed from the traditional ones from a cardiovascular perspective. As expected, traditional BFR, a high-intensity exercise, increased cardiovascular responses (Fig. 3). The "Clinical BFR" protocols achieved significantly lower systolic blood pressure levels than traditional BFR. This may be attributed again to reduced exercise intensity and lower accumulation of metabolites from glycolysis, which stimulates class III/IV metaboreceptors, as well as the submaximal nature of the Clinical BFR. Considering that acute cardiovascular events often correlate with increased myocardial oxygen consumption,³³ a possible reduction in systolic blood pressure during exercise may enhance exercise safety, especially in populations at cardiovascular risk.

One of the main goals of a BFR rehabilitation program is to enhance muscle function; in this context, performance evaluation is crucial for understanding the various protocols, particularly the auto-regulated Clinical BFR versus the clinically cadenced BFR. The results show that when subjects were allowed to perform at a different concentric velocity than the cadenced one (1.5 ​s for concentric and eccentric contractions in the cadenced version), they chose to perform at a significantly higher velocity. A potential explanation for this is that increased affectivity may be indirectly related to relief from discomfort during ischemic conditions and may disinhibit motor units through feedback on the muscles, thus potentially facilitating strength development.^34,35 Especially in the elderly, where there is often a deficit in the rate of force development (RFD),³⁶ performing faster concentric movements could lead to the recruitment of faster motoneurons, potentially promoting neural and muscle fiber adaptations.³⁷ However, such adaptations await further investigation.

A fundamental question in our research was how to improve traditional BFR for use in clinics and hospitals. As part of this modification, exercise intensity was reduced in the Clinical BFR versions, resulting in favorable cardiovascular, metabolic, and psychological responses. However, reducing exercise intensity may also impact skeletal muscle responses. To gain information about muscle activation, surface EMG analysis was performed in a subset of four subjects following the cadenced traditional BFR or the cadenced Clinical BFR. The results show that both protocols present similar muscle activation responses from the beginning to the first half of the exercise. After that, muscle activation in the traditional BFR increased beyond that in the Clinical BFR, and by the end of the exercise, a statistically significant increase in muscle activation was observed in the traditional BFR (Fig. 7). Since the load and repetitions employed in both protocols were matched, the increased amplitude of the EMG signal was probably derived from the accumulation of fatigue and muscle metabolites in the muscle fibers, increasing the recruitment of fibers.⁹ Regarding the Clinical BFR, such decreased muscle activation may not be harmful but rather favorable for muscle adaptations in deconditioned, frail, weak, and sarcopenic elderly individuals, especially at the beginning of their rehabilitation program. In this regard, Kaczor et al.³⁸ investigated the effects of exercise intensity on oxidative stress markers in MDX mice, which exhibit inherited dystrophin mutations that weaken their skeletal muscle fibers. In this study, low-intensity exercise did not elicit positive responses regarding malondialdehyde and protein carbonyls in wild-type animals. On the other hand, in MDX mice, which exhibit increased oxidative stress markers in muscle fibers, the same low-intensity exercise was beneficial, decreasing these markers, especially in the fast-twitch fibers, which are crucial for high-force production. A possible explanation for this observation is related to the different exercise intensity thresholds, which are decreased in pathological versus nonpathological MDX mice.³⁹ Thus, it is possible that in many clinical conditions, low-intensity Clinical BFR exercise may be a better fit than traditional BFR.

Exaggerated delayed-onset muscle soreness (DOMS), a type of pain that typically appears one to two days after intense or unaccustomed exercise,³⁵ may represent a barrier to exercise in frail or pain-sensitive individuals. In the present study, DOMS was measured to evaluate this important aspect of exercise. Despite similar external workloads, the clinical versions showed reduced DOMS from 24 to 48 ​h after exercise compared to the traditional one. A possible explanation for the reduced DOMS in the Clinical BFR versions despite the same mechanical load and volume is that non-fatigued muscle fibers conserve the capacity to absorb energy compared to a fatigued one, and absorbing energy makes the muscle fiber less susceptible to muscle damage,⁴⁰ and consequently, DOMS.⁴¹ Decreased DOMS may be valuable in individuals who are frail or pain-sensitive, especially at the beginning of a rehabilitation program.

Finally, deconditioned and frail individuals, as well as the sarcopenic elderly, may experience loss of muscle and physical activity, which can impair glucose homeostasis. BFR, on the other hand, can enhance glucose uptake in skeletal muscle through muscle contraction, which is partly mediated by increased calcium levels and hypoxia.⁴² Across all BFR versions, blood glucose levels significantly reduced immediately after exercise. Blood glucose rapidly returned to baseline during recovery, showing no significant differences among the BFR versions in healthy subjects.

5. Conclusion

Traditional BFR is known for its cost-effectiveness and time-efficiency but may not be well tolerated by all populations for several reasons: 1) it is intense and of maximal nature, 2) it provokes moderate to high acute pain-associated discomfort, 3) it causes reasonably delayed onset muscle soreness (DOMS) responses, 4) subjects usually demonstrates low affectivity to it, and 5) promotes disproportional increases in cardiovascular responses to the low load employed. We demonstrated that clinical BFR was 1) less intense, 2) reduced pain-associated discomfort, 3) decreased DOMS, 4) increased affectivity, 5) diminished systolic blood pressure responses, 6) decreased fatigue and muscle activation, and 7) promoted similar blood glucose responses compared to traditional BFR. These results support the potential use of clinical BFR in clinical settings. It is essential to note that this version is not very easy but rather slightly less intense than the traditional BFR version, as the perceived rates of exertion and lactate levels indicate. Additionally, it was unilaterally tested, so it is important to note that the responses should be increased if performed bilaterally; however, they are likely still diminished in the Clinical version compared to the traditional one. Finally, one possible application for clinical BFR is to implement it at the onset of the rehabilitation program or whenever a reduction in exercise intensity may be necessary. Additionally, it can be used in conjunction with traditional BFR to address various individual needs. Clinical BFR methodology was tested on healthy subjects as an exploratory consideration for the skeletal muscle rehabilitation of deconditioned weak, frailty, sarcopenic elderly, and fragile individuals. Abe et al.⁴³ developed a similar strategy for using blood flow restriction during walking conditions in healthy subjects, which was then successfully applied to the elderly.⁴³

CRediT authorship contribution statement

Fabricio Eduardo Rossi: Writing – review & editing, Formal analysis, Conceptualization. Kassiana de Araújo Pessôa: Writing – review & editing, Methodology. Zhi Xia: Writing – review & editing, Conceptualization. Jakob Lauver: Writing – review & editing. Daniel Newmire: Writing – review & editing. Jason Michael Cholewa: Writing – review & editing. Steven Machek: Writing – review & editing. Mario Alves de Siqueira Filho: Writing – review & editing. Fabio Santos Lira: Writing – review & editing. Antonio Herbert Lancha-Jr: Writing – review & editing. Emilia Zawieja: Writing – review & editing. Nelo Eidy Zanchi: Writing – review & editing, Methodology, Conceptualization.

Ethics approval and consent to participate

This study was approved by the Research Ethics Committee from Federal University of Maranhão (CAAE: 59371522.2.0000.5087), and all research was conducted by the 2013 revision of the Declaration of Helsinki.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Funding

This research was supported by Zhejiang Provincial Natural Science Foundation of China under Grant No. LY23C110001 and the Basic Scientific Research Project of Wenzhou City under Grant No. Y20220209.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.