Edited by: Stephen D. Patterson, St Mary’s University, United Kingdom
Reviewed by: Moacir Marocolo, Universidade Federal de Juiz de Fora, Brazil; Alan Kacin, University of Ljubljana, Slovenia
This article was submitted to Exercise Physiology, a section of the journal Frontiers in Physiology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
The metabolic stress induced by blood flow restriction (BFR) during resistance training (RT) might maximize muscle growth. However, it is currently unknown whether metabolic stress are associated with muscle hypertrophy after RT protocols with high- or low load. Therefore, the aim of the study was to compare the effect of high load RT (HL-RT), high load BFR (HL-BFR), and low load BFR (LL-BFR) on deoxyhemoglobin concentration [HHb] (proxy marker of metabolic stress), muscle cross-sectional area (CSA), activation, strength, architecture and edema before (T1), after 5 (T2), and 10 weeks (T3) of training with these protocols. Additionally, we analyzed the occurrence of association between muscle deoxygenation and muscle hypertrophy. Thirty young men were selected and each of participants’ legs was allocated to one of the three experimental protocols in a randomized and balanced way according to quartiles of the baseline CSA and leg extension 1-RM values of the dominant leg. The dynamic maximum strength was measured by 1-RM test and vastus lateralis (VL) muscle cross-sectional area CSA echo intensity (CSAecho) and pennation angle (PA) were performed through ultrasound images. The measurement of muscle activation by surface electromyography (EMG) and [HHb] through near-infrared spectroscopy (NIRS) of VL were performed during the training session with relative load obtained after the 1-RM, before (T1), after 5 (T2), and 10 weeks (T3) training. The training total volume (TTV) was greater for HL-RT and HL-BFR compared to LL-BFR. There was no difference in 1-RM, CSA, CSAecho, CSAecho/CSA, and PA increases between protocols. Regarding the magnitude of the EMG, the HL-RT and HL-BFR groups showed higher values than and LL-BFR. On the other hand, [HHb] was higher for HL-BFR and LL-BFR. In conclusion, our results suggest that the addition of BFR to exercise contributes to neuromuscular adaptations only when RT is performed with low-load. Furthermore, we found a significant association between the changes in [HHb] (i.e., metabolic stress) and increases in muscle CSA from T2 to T3 only for the LL-BFR, when muscle edema was attenuated.
Resistance training- (RT) induced changes in muscle strength are partially due to increases in muscle cross-sectional area (CSA) (i.e., muscle hypertrophy) and changes in muscle architecture (e.g., increase in the pennation angle of muscle fibers) (
It has been widely accepted that RT-induced muscle hypertrophy occurs through two primary mechanisms: mechanical tension and metabolic stress (
Our group previously demonstrated that high-load blood flow restriction (HL-BFR) training programs do not produce additive effects compared to traditional HL-RT programs (
Therefore, we compared the effects of HL-RT, HL-BFR, and LL-BFR on muscle [HHb], CSA, activation, strength, architecture, and edema before (T1), after 5 (T2), and 10 weeks (T3) of training. Additionally, we analyzed the association between muscle deoxygenation and muscle hypertrophy. Our hypotheses were (1) BFR would not produce an additive effect on neuromuscular adaptations when mechanical tension is high; (2) high mechanical tension protocols (i.e., HL-RT and HL-BFR) would not present an association between metabolic stress and muscle hypertrophy; (3) low load BFR training would produce similar neuromuscular adaptations to HL-RT; and (4) metabolic stress would have an association with muscle hypertrophy only when mechanical tension is low (i.e., LL-BFR).
Thirty young men volunteered to participate in the present study (age: 22 ± 3 years; body mass: 72.7 ± 10.7; kg; height: 178 ± 5 cm; BMI: 22.81 ± 2.99 kg⋅m2. Inclusion criteria were: (
The present randomized controlled trial used a prospective, single-group, intra-subject design in which each leg of the subjects was exposed to one of three experimental protocols. Before the commencement of the experimental protocol, participants engaged into two familiarization sessions to get acquainted with the training protocol and testing procedures. Familiarization sessions were interspaced by 72 h. Seventy-two hours after the last familiarization session, leg extension 1-RM test was performed. Ninety-six hours after the 1-RM test, vastus lateralis (VL) muscle CSA, echo intensity (CSAecho), pennation angle (PA) were assessed by ultrasonography (US). Each of the participants’ legs was allocated to one of the three experimental protocols in a randomized and balanced way according to baseline CSA and leg extension 1-RM values of the dominant leg. In short, participants’ legs were divided into quartiles according to muscle CSA and 1-RM values; afterward, legs within each quartile were randomly allocated into the three training protocols: (1) high-load resistance training (HL-RT); (2) high-load resistance training with blood flow restriction (HL-BFR); and (3) low-load resistance training with blood flow restriction (HL-BFR). Importantly, traditional LL-BFR studies have maintained blood flow restricted (i.e., cuff inflated) throughout training sessions. However, maintaining BFR during an exercise session is very uncomfortable and painful (
The described dependent variables (i.e., CSA, CSAecho/CSA, PA, and 1-RM) were assessed before the experimental protocol (T1), after 5 (T2), and 10 weeks (T3) of the commencement of the experimental period (
Schematic of study protocol. RE, resistance exercise; 1-RM, one-repetition maximum; EMG, electromyographic; NIRS, near-infrared spectroscopy.
Unilateral quadriceps maximum dynamic strength was assessed using the 1-RM test on a leg-extension machine (Effort NKR; Nakagym, São Paulo, Brazil), according to the procedures described elsewhere (
Before the commencement of the training protocol, blood-flow restriction pressure was determined as follows. Participants in the HL-BFR and LL-BFR protocols were asked to rest comfortably in supine position. A vascular Doppler probe (DV-600; Marted, Ribeirão Preto, São Paulo, Brazil) was placed over the tibial artery to capture its auscultatory pulse. For the determination of blood pressure (mmHg) necessary for complete vascular occlusion (pulse elimination pressure), a standard blood-pressure cuff [175 mm (width) 920 mm (length)] was wrapped around the participant’s thigh at the inguinal fold region and then inflated up to the point at which the auscultatory pulse was interrupted (
A B-mode ultrasound (US) with a 7.5-MHz linear-array probe (Mysono U6 EX; Samsung-Medison, Gangwon-do, South Korea) was used to capture images in the axial plane of the VL muscle after the participants had laid supine for 20 min to allow for fluid distribution before the assessments (
Muscle CSA previously delimited was analyzed using a Fast Fourier transformation to identify the frequency spectrum of the pixel intensity over the VL CSA (CSAecho). This analysis results in a histogram of grayscale shades (0 = black and 256 = white), where any abnormality (e.g., edema-induced muscle swelling, probably due to muscle damage) results in higher echo intensity value (increased white areas), while intact muscle mass presents low echo intensity (i.e., dark areas) (
PA of the VL was measured using the B-mode ultrasound at the thigh mid-point. The PA was defined as the angle between the fascicle and the deep aponeurosis of the VL muscle (
Muscle cross section area (CSA), CSA echo intensity (CSAecho), CSAecho to CSA ratio, pennation angle (PA) and maximum dynamic strength test (1-RM) at baseline (T1), after 5 (T2), and 10 weeks (T3) for high-load resistance training (HL-RT), high-load resistance training with blood flow restriction (HL-BFR) and low-load strength training with blood flow restriction (LL-BFR).
Variable | Time | HL-RT | HL-BFR | LL-BFR |
---|---|---|---|---|
CSA (cm2) | T1 | 22.3 ± 6.7 | 21.8 ± 4.0 | 21.4 ± 5.8 |
T2∗ | 23.4 ± 7.0 | 22.9 ± 4.3 | 22.6 ± 6.2 | |
T3∗† | 24.5 ± 7.4 | 24.2 ± 4.7 | 23.7 ± 6.7 | |
CSAEcho (AU) | T1 | 19.5 ± 5.1 | 19.5 ± 5.5 | 21.0 ± 7.0 |
T2∗ | 38.0 ± 9.0 | 38.2 ± 9.7 | 40.0 ± 11.0 | |
T3† | 17.6 ± 3.3 | 19. ± 5.5 | 17.6 ± 4.1 | |
CSAEcho (AU)/CSA (cm2) | T1 | 1.0 ± 0.4 | 0.9 ± 0.3 | 1.0 ± 0.4 |
T2∗ | 1.7 ± 0.6 | 1.7 ± 0.4 | 2.0 ± 0.9 | |
T3∗† | 0.7 ± 0.2 | 0.8 ± 0.3 | 0.8 ± 0.2 | |
PA (°) | T1 | 15.0 ± 2.6 | 15.4 ± 2.0 | 14.6 ± 2 |
T2∗ | 16.0 ± 2.8 | 16.3 ± 2.0 | 15.5 ± 2.0 | |
T3∗† | 16.4 ± 2.9 | 17.0 ± 2.1 | 16.0 ± 2.1 | |
1-RM (kg) | T1 | 46.6 ± 11.5 | 46.4 ± 10.8 | 46.5 ± 11.7 |
T2∗ | 55.7 ± 10.8 | 56.5 ± 10.4 | 51.3 ± 17.7 | |
T3∗† | 63.8 ± 10.9 | 62.1 ± 18.1 | 55.2 ± 23.2 |
An angular potentiometer was placed on the right knee of the individuals with its center of rotation aligned with the lateral intercondylar line of the knee joint to determine knee angular excursion and, therefore, the concentric and eccentric phases of the lift. Full extension was defined as “zero degree.” The concentric phase was defined from the maximum to the minimum value of the knee flexion angle, while the eccentric phase was defined from the minimum to the maximum value of the knee flexion angle. The frequency of acquisition was set at 1000 Hz in the A/D converter of the EMG unit described below, which synchronized data acquisition from the angular potentiometer and both the EMG system and the NIRs device. The signal from an external trigger was split and sent to both EMG and NIRs A/D converters to align the data in time.
Muscle activation of the VL muscle, assessed by the amplitude of the electromyographic (EMG) signal (EMG832C; EMG System do Brazil, São José dos Campos, Brazil), was determined at T1, T2, and T3. Before electrode placement, the skin area was shaved, abraded and cleaned with an isopropyl alcohol pad to reduce skin impedance before electrode placement (
A continuous dual-wavelength near-infrared spectroscopy apparatus (NIRS; Oxymon, Artinis Medical Systems, Arnhem, the Netherlands) was used to monitor changes in muscle oxygenation during the HL-RT, HL-BFR, and LL-BFR protocols in T1, T2, and T3 assessments. Data was collected at a frequency of 25 Hz. The system uses a modified Beer-Lambert law to analyze the changes in light absorbed at wave lengths of 761 and 844 nm, estimating concentrations of deoxygenated hemoglobin ([HHb]), which has been considered an indicator of metabolic stress (
Initially, thickness of the subcutaneous fat layer at the site of NIRs optodes (i.e., an emitter and a detector) placement was assessed by ultrasound (HL: 0.83 ± 0.18 cm, HL-BFR: 0.91 ± 0.22 cm, and LL-BFR: 0.87 ± 0.20 cm) to set the value of laser penetration depth (
Data were extracted from the NIRS device using Oxisoft (3.0.X; Artinis Medical Systems B.V, Arhem, Netherlands) and a customized script analyzed the data off-line. Raw data was filtered with a moving average algorithm over a 2 s period. Then, [HHb] resting values were obtained during the last 5 min of a 15 min rest period, in which individuals remained still and as relaxed as possible prior to exercise commencement (
Training protocols were performed unilaterally using a conventional leg-extension machine, twice a week for 10 weeks. The HL-RT and HL-BFR protocols performed 3 sets of 10 repetitions with a load corresponding to 80% 1-RM (HL-RT: T1–T2 = 37.2 ± 9.2 kg and T2–T3 = 44.5 ± 8.6 kg; HL-BFR: T1–T2 = 37.1 ± 8.7 kg and T2–T3 = 45.2 ± 8.4 kg), while the LL-BFR protocol performed 3 sets of 20 repetitions with 20% 1-RM (LL-BFR: T1–T2 = 9.3 ± 2.3 kg and T2–T3 = 10.2 ± 3.5 kg). A 1 min rest period was granted between sets for all of the protocols. After the fifth week (10th session), 1-RM was re-assessed to adjust training load. From week 6 (T2–T3) onwards, the number of sets was increased to four, for all of the participants. The cuff pressure used during the BFR protocols was set at 60% of occlusion pressure in the resting condition. The cuff pressure remained inflated during the exercise and deflated during the rest periods. The average pressure used throughout the training protocol was 81.85 ± 4.45 mmHg.
After visual inspection, the area under the curve (AUC) analysis for EMG and [HHb] were performed using the trapezium rule (GraphPad Prism, GraphPad Software, San Diego, CA, United States) in order to characterize the magnitude of the response and the changes over time. AUC analyses were calculated using the time point immediately before (Pre) and changes in EMG and [HHb] in the 1st, 2nd, and 3rd sets. The 1-RM, CSA, CSAecho, CSAecho/CSA, PA, EMG, and [HHb] data were analyzed using mixed models having training protocol and time as fixed factors, and subjects as random factor. Only TTV was analyzed with a one-way repeated measures model having training protocol as a fixed factor and subjects as a random factor. In case of significant values of
TTV (sets × repetitions × load [kg]) in the LL-BFR (11733.0 ± 3204.9 kg) was lower than HL-RT (24546.0 ± 5329.0 kg,
1-RM values increased, similarly, and significantly for the HL-RT, HL-BFR, and LL-BFR groups from T1 to T2 (main time effect
The HL-RT, HL-BFR, and LL-BFR protocols groups showed significant and similar increases in muscle CSA from T1 to T2 (main time effect,
CSAecho analysis (
In relation to the PA (
EMG amplitude AUC was greater during HL-RT and HL-BFR than LL-BFR (main protocol effect,
Normalized electromyographic (EMG) root mean square (RMS) values from the resistance training session (Pre [Before the beginning of the exercise], set 1 [S1], set 2 [S2], and set 3 [S3]) at baseline (T1,
The [HHb] AUC was significantly lower for all protocols in T3 compared to T2 (main time effect,
Normalized deoxygenated hemoglobin ([HHb]) values from the resistance training session (Pre [Before the beginning of the exercise], set 1 [S1], set 2 [S2], and set 3 [S3]) at baseline (T1,
There was a significant correlation between [HHb] and CSA area only at T2–T3 for the LL-BFR protocol (
Correlations between changes in deoxyhemoglobin concentrations ([HHb]) and changes in muscle cross-sectional area (CSA) for high-load resistance training (HL-RT), high-load resistance training with blood flow restriction (HL-BFR), low-load blood flow restriction (LL-BFR) and all groups together.
Variable | CSA (%) (T1–T2) | CSA (%) (T2–T3) | CSA (%) (T1–T3) | |
---|---|---|---|---|
HL-RT [HHb] (T1–T2) | 0.30 | |||
0.19 | ||||
HL-RT [HHb] (T2–T3) | 0.14 | |||
0.56 | ||||
HL-RT [HHb] (T1–T3) | 0.03 | |||
0.90 | ||||
HL-BFR [HHb] (T1–T2) | 0.136 | |||
0.580 | ||||
HL-BFR [HHb] (T2–T3) | 0.43 | |||
0.08 | ||||
HL-BFR [HHb] (T1–T3) | 0.05 | |||
0.86 | ||||
LL-BFR [HHb] (T1–T2) | 0.10 | |||
0.67 | ||||
LL-BFR [HHb] (T2–T3) | 0.71 | |||
0.0008∗ | ||||
LL-BFR [HHb] (T1–T3) | 0.36 | |||
0.22 | ||||
All protocols [HHb] (T1-T2) | 0.05 | |||
0.71 | ||||
All protocols [HHb] (T2–T3) | 0.09 | |||
0.23 | ||||
All protocols [HHb] (T1–T3) | 0.05 | |||
0.75 |
We aimed to compare the effects of HL-RT, HL-BFR, and LL-BFR on muscle deoxygenation (HHb), CSA, activation, strength, architecture, and edema before (T1), after 5 (T2), and 10 weeks (T3) of training. Additionally, we analyzed the association between [HHb] and muscle hypertrophy. Regarding our four hypotheses, we confirmed that: (1) BFR did not produce an additive effect on muscle hypertrophy when mechanical tension is high; (2) high mechanical tension protocols did not produce a significant correlation between metabolic stress and muscle hypertrophy (i.e., HL-RT and HL-BFR); (3) LL-BFR produced similar neuromuscular adaptations to HL-RT; and (4) metabolic stress has a positive and significant association with muscle hypertrophy only when mechanical tension is low (i.e., LL-BFR). Thus, mechanical tension and metabolic stress seem to share the variance of the muscle hypertrophy response under high mechanical tension protocols, while metabolic stress seems to be the main mechanism responsible for muscle hypertrophy when mechanical tension is low.
Regarding muscle strength, previous studies have reported similar increases in muscle strength between HL-RT and LL-BFR protocols (
Usually, studies have reported that low-load RT promotes a small or even no increase in the muscle CSA (
Importantly, muscle hypertrophy was accompanied by increases in pennation angle (PA). These increases occurred in the early stages of training [T2, 5 weeks (10 training sessions)] and were even greater at T3 [10 weeks (20 training sessions)], as shown in other studies (
It has been suggested that muscle strength gains and hypertrophy after a RT period, with or without BFR, are associated with increases in the ability to activate the motor unit pool (
It has been suggested that [HHb] is a proxy marker of metabolic stress (
Comparing the protocols, those performed with BFR (i.e., HL-BFR and LL-BFR) showed higher deoxygenation compared to HL-RT. Interestingly, notwithstanding HL-BFR combine the highest levels of metabolic stress and mechanical tension, the muscle hypertrophy and strength gains were similar to protocols with lower levels of metabolic stress (i.e., HL-RT) or mechanical tension (i.e., LL-BFR). These results may suggest that muscle protein synthesis may reach maximal values when training with high intensities (e.g., 80% 1-RM – high mechanical tension) and the metabolic stress does not seem to produce additive effects to muscle hypertrophy. On the other hand, in low-load protocols (e.g., 30% 1-RM) the metabolic stress induced by BFR seems to fully activate the muscle protein synthesis machinery. This is supported by the strong association between changes in [HHb] and changes in muscle CSA only to LL-BFR group (
We show greater muscle deoxygenation for HL-BFR and LL-BFR and higher muscle activation for HL-RT and HL-BFR, but with similar neuromuscular adaptations between these protocols in untrained men. Although one may rightly suggest that these findings may not be directly extended to trained individuals, to the best of our knowledge, there is no empirical evidence suggesting that the activation of specific hypertrophy triggers may change as a function of the training status.
A high and significant correlation between [HHb] and muscle hypertrophy were showed for LL-BFR. However, these findings should be viewed with caution, as other NIRS parameters, such as deoxyhemoglobin (HbO2), total hemoglobin (HBT) and hemoglobin difference (Hbdiff) were not analyzed in the present study. In addition, the agreement between [HHb] and other parameters related to metabolic stress (e.g., phosphocreatine, inorganic phosphate, muscle pH, and lactate) should be analyzed in future studies to confirm our findings. On the other hand, the HHb is highly correlated with blood lactate (
We also suggest that future studies should investigate the effects of secondary hypertrophy-related mechanisms triggered by metabolic stress (e.g., increased fast-twitch fiber recruitment, local hormone, cell swelling, and the production of reactive oxygen species) on muscle hypertrophy to further elucidate these complex mechanisms.
It is also important to consider that the findings reported herein should be confirmed in other BFR training protocols, as it has been demonstrated that metabolic stress may change as a function of the occlusion pressure and training load (
BFR training pressure was determined in a resting state, which may be considered as an inherited limitation of the method, as pressure fluctuates during the eccentric and concentric phases of the lifts.
Muscle deoxygenation seems to play an important role on neuromuscular adaptations when RT is performed with low-load, as it produces similar neuromuscular adaptions to high-load protocols, despite the lower (TTV) (∼53%). Corroborating this suggestion, there was a significant association between the changes in [HHb] and increases in muscle CSA from T2 to T3, when muscle edema was attenuated. We propose that the level of metabolic stress would not influence the magnitude of muscle hypertrophy, as well as changes in muscle strength and architecture, when RT is performed in high-loads.
Our results suggest that the addition of BFR to exercise contributes to neuromuscular adaptations only when RT is performed with low-load. Furthermore, we found a significant association between the changes in [HHb] (i.e., metabolic stress) and increases in muscle CSA from T2 to T3 only for the LL-BFR, when muscle edema was attenuated.
This study was carried out in accordance with the recommendations of ethics committee of Federal University of São Carlos (UFSCar), number 42359015.5.0000.5504 with written informed consent from all subjects. All subjects gave written informed consent in accordance with the Declaration of Helsinki. The protocol was approved by the ethics committee of Federal University of São Carlos (UFSCar).
CAL had the original idea of the study and the final study design was developed by CAL, CU, TMPCB, and AB-S. Participants were recruited, trained and assessed at the Federal University of São Carlos, by TMPCB, RMO, SDS, and JGB. TMPCB, CAL, and CU performed data analyses and statistical procedures and wrote the first version of the manuscript. All authors participated in the interpretation of the data, contributed to the revision of the manuscript, and approved the content of the final version.
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.
Special thanks to the students who volunteered to participate in the experiments.