It is well-known that progressive resistance training using loads >65% of maximal loading (one-repetition maximum [1RM]) effectively stimulate gains in maximal muscle strength and muscle mass (37). However, during the last decade, it has been established that the application of blood flow–restricted (BFR) training using low external training loads (~20%–30% 1RM) can also lead to significant gains in maximal skeletal muscle strength and skeletal muscle mass (27,33,34). The unique combination of low muscle–tendon loading and the considerable adaptive training effects suggest that this training modality may possess a considerable rehabilitation potential in clinical populations with limited functional capability and to whom resistance training using high external training loads (>65%) is contraindicative.
In terms of mechanical muscle function, previous BFR training studies have mainly focused on changes in maximal isometric and isokinetic muscle strength (21,33,34). However, from a functional perspective, the ability to rapidly produce muscle force may be of greater importance than maximal muscle strength, as most functional movements typically are performed within a limited time frame (0–200 ms) (2,28), whereas peak muscle force usually occur between 300 and 600 ms after onset of contraction (2). Thus, rapid force capacity (i.e., rate of torque development [RTD]), which is defined as the ability to produce muscle force rapidly, may be a better predictor of various types of daily functional movements compared with maximal muscle strength (8,29,32). However, the effect of BFR training on skeletal muscle rapid force capacity currently remains unknown.
Notably, recent studies have indicated that a single bout of low-load BFR exercise may acutely exert a negative influence on skeletal muscle contractile properties (9,22,31,38), and both short (≤1 h) (9,22) and more prolonged (≥24 h) impairments have been reported (31,38). Yet, it remains unknown whether the longitudinal application of low-load BFR training imposes inhibitory effects on skeletal muscle contractile properties.
Therefore, the purpose of the present study was to investigate the effect of high-frequency low-load BFR training on the magnitude and time course of adaptation in skeletal muscle contractile properties with special emphasis on rapid force capacity (i.e., RTD) in untrained male adults.
On the basis of our earlier findings of an increase in muscle mass and maximal muscle strength, we hypothesized that short-term BFR training would lead to improvements in skeletal muscle RTD. However, because of the documented negative effects of acute BFR exercise on skeletal muscle contractile properties, we expected a delayed longitudinal time course of rapid force adaptation to occur, with more marked gains in RTD manifested at later time points after cessation of training.
In the present study, we report a delayed increase in voluntary muscle mechanical function after BFR training, with the delay potentially being explained by an early decrement in evoked twitch response. To further investigate myocellular factors potentially triggering the apparent early decrement in evoked muscle mechanical function, additional muscle protein analyses on markers related to myocellular Ca2+ exposure and nitrosylation of myocellular proteins were performed, as these factors are known to initiate prolonged decrements in contractile function (13,26,35). Ca2+/calmodulin-dependent kinase II (CaMKII) has been suggested as a Ca2+ sensor (14), both responding acutely and prolonged to changes in Ca2+ amplitude and duration (7,12), whereas annexin A6 seem to be involved in plasma membrane repair initiated by Ca2+ influx (30). Further, S-nitrosylation of proteins involved in the excitation–contraction (EC) coupling may potentially affect contractile function negatively for a prolonged period (5,13).
MATERIALS AND METHODS
Twenty-one healthy male subjects volunteered to participate in the study. Subjects were divided into a BFR training group (n = 12; age = 23 ± 2 yr, height = 181 ± 6 cm, body mass = 82.3 ± 13.7 kg) and a control group (CG) (n = 9; age = 22 ± 3 yr, height = 183 ± 9 cm, body mass = 80.2 ± 11.4 kg). All participants were recreationally active and had not participated in any systematic strength training within a year before the study. The study was approved by the local ethics committee (S-200900070) in accordance with the Declaration of Helsinki, and written informed consent was obtained from the subjects before inclusion. Data from this study have been published previously (27).
Protocol Overview
Contractile knee extensor muscle function was evaluated using isokinetic dynamometry before (Pre) as well as 5 and 12 d after (Post5 and Post12) cessation of training (Fig. 1). Muscle biopsy samples were obtained from the vastus lateralis (VL) muscle before (Pre) and 8 d into the intervention period (Mid8) as well as 3 and 10 d after (Post3 and Post10) cessation of training. All measurements were performed in the experimental (EX) leg, which was chosen by paired randomization between legs. At least 1 wk before testing, subjects visited the laboratory for familiarization with the strength test procedures and for determination of 1RM. Subjects were carefully instructed to maintain their normal daily routines in regard to nutritional intake and physical activity and to refrain from alcohol during the intervention period. In addition, subjects were asked not to participate in moderate to hard physical activity within 48 h before test/sampling procedures, which were conducted at the same time of the day to avoid diurnal variations.
;)
FIGURE 1: Schematic illustration of the study design—BFR (n = 12) and CON (n = 9). White arrows denote individual training sessions. For more details, see Materials and Methods section.
Training Protocol
The training protocol has been described in detail elsewhere (27). In brief, subjects completed 3 wks of training consisting of 23 separate supervised training sessions (cf. Fig. 1). Both groups performed four sets of unilateral knee extensor exercise at 20% of 1-RM with a 30-s break between sets. BFR performed repetitions to concentric failure (cadence = 1.5/1.5 s), whereas CON performed the same number of repetitions and cadence ensuring that the same relative amount of work was performed between the groups (27). Blood flow restriction was obtained with a pneumatic cuff (13.5 cm width) (Delfi Medical) connected to a computerized system (Zimmer A.T.S. 750), which enabled automatic regulation of occlusion pressure. Before each training session, the cuff was placed around subjects' upper thigh, where it was inflated to 100 mm Hg and remained inflated until completion of the fourth exercise set.
Evaluation of Contractile Muscle Properties
Maximal voluntary and evoked skeletal muscle contractile properties were measured in an isokinetic dynamometer (Kinetic Communicator 500H; Chattecx Corp., Hixon, TN) as described in detail previously (2,27). In brief, subjects were placed in a standardized position, which enabled identical positioning before and after the intervention period. The length of the lever arm was noted for calculation of knee extensor torque (N·m).
Evaluation of voluntary contractile function
Subjects performed a standardized 5-min warm-up on an ergometer cycle, followed by a warm-up consisting of submaximal isokinetic contractions in the dynamometer. Afterward, isometric maximal voluntary muscle strength (maximal voluntary contraction [MVC]) of the knee extensors was recorded at 70° knee joint angle (0° = full extension) (2) followed by concentric MVC of the knee extensors (range of motion = 90°–20°) using slow and fast angular knee joint velocities (30/240°·s−1) (1). Before all trials, subjects received instructions to contract as fast and hard as possible, whereas online visual feedback of the exerted force was provided on a computer screen. During each contraction mode, successive attempts were performed until no further increase in peak torque could be reached. Each attempt was separated by 60 s to exclude fatigue. All trials with countermovement contractions were disregarded and repeated. Force and lever arm position signals were recorded at 1 kHz using a 16-bit A/D converter. During later offline analysis, signals were low-pass filtered using a fourth-order zero-lag Butterworth filter at a 15-Hz cutoff frequency. Torque signals were corrected for the influence of gravity (2). RTD and isokinetic peak torque were obtained from the respective isometric/isokinetic trial with the highest torque. Voluntary RTD (vRTD) was derived from the average slope of the torque–time curve (Δtorque/Δtime) in the initial (0–30 and 0–50 ms) and late (0–100 and 0–200 ms) phases of rising isometric muscle force, relative to onset of contraction (2). Preceding reliability assessments (n = 5) revealed acceptable test–retest reliability for the present measurements of maximal knee extensor strength and rapid force capacity (slow conc. KE torque: ICC = 0.94; fast conc. KE torque: ICC = 0.92; KE RFD: ICC = 0.93–0.96).
Evoked muscle contraction properties
Evoked twitch muscle contractions were obtained in the rested state (rT) and potentiated (pT) state (e.g., 2 min post-MVC). All subjects went through a careful preparation of the skin whereupon pairs of percutaneous surface stimulation electrodes (ValuTrode, CF5090, Axelgaard Ltd. Lystrup, Denmark) were placed over the distal (10 cm above patella) and proximal (15 cm below the anterior superior iliac spine) part of musculus rectus femoris. To ensure identical electrode placement in Pre and Post assessments, positions were recorded. Twitch contractions were evoked using 1-ms single-pulse percutaneous stimulations delivered to the relaxed muscle (Model DS7AH, Digitimer Ltd, Welwyn Garden City, Hertfordshire, UK). The maximal twitch response was determined during stepwise increments in current intensity (50 mA) delivered every 60 s until torque leveling-off between two concurrent stimulations (<5 N) (4). The leveling-off current obtained before training was maintained during the remaining test sessions to ensure that approximately the same amount of muscle mass was activated at all time points. The following contractile parameters were determined from the rT and pT force outputs: 1) peak torque (PT), 2) RTD in 0- to 30- and 0- to 50-ms intervals, and 3) RTD and rate of torque relaxation at 50% PT (RTD50% and RTR50%). Twitch stimulations were omitted in CON, as familiarization of these procedures could not be performed in CON because of the acute sickness of the tester.
Muscle Tissue Sampling and Handling
Muscle biopsy samples were obtained from VL during local anesthesia (1% lidocaine; Amgros, Denmark), as described in detail previously (27). In short, muscle biopsies were obtained from a standardized depth in a randomized fashion ~3 cm apart. A piece of muscle was embedded in Tissue-Tec (Sakura Finetek, Netherlands) and frozen in isopentane precooled in liquid nitrogen, whereas another piece was frozen in liquid nitrogen. After handling, samples were stored at −80°C for later analysis.
Immunofluorescence Analysis
Transverse serial sections were fixed (4% Paraformaldehyde) and subsequently blocked (X0909, Dako, Glostrup, Denmark) for 10 min. Sections were incubated for 60 min with primary (1:1000, laminin, Z0097; Dako) and secondary antibodies (1:1000, A11034; Life Technologies, Carlsbad, CA). Lastly, sections were mounted (H5000, Vector, Burlingame, CA), covered with cover glass, and stored protected from light at 5°C. Stainings were visualized using a light microscope (Axio Imager M1, Carl Zeiss [CZ], Oberkochen, Germany) and a high-resolution AxioCam (CZ), and all analyses were performed by a blinded investigator in AxioVision 4.6 (CZ). Myofiber area (MFA) was determined in 371 ± 103 fibers per muscle biopsy.
Immunoblotting Analysis
VL muscle tissue was put in ice-cold homogenizing buffer (pH 7.8; 40 mM Tris/base, 200 mM sucrose, 1 mM EDTA, 10 mM sodium azide, and protease inhibitor mini tablets; Roche, Indianapolis, IN) and homogenized gently in a glass Teflon homogenizer. Homogenate was frozen at −80°C for later analysis. Standardized volumes of homogenate were loaded per well (10–20 μg) and separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (10%) (Bio-Rad [BR], Hercules, CA) at 180 V in running buffer (162–0734, BR) for 30 min. Samples were transferred to polyvinylidene difluoride membranes (162–0177, BR) at 100 V in 4°C transfer buffer (162–0734, BR) for 60–90 min. Membranes were blocked for 30 min in 5% nonfat milk (170–6404, BR) and incubated with the following primary antibodies: CaMKII (1:1000; 611292, BD Biosciences, Franklin Lakes, NJ), phos-CaMKII (Thr286) (1:1000;MAI-047, Thermo Scientific [TS], Waltham, MA), annexin A6 (1:500; HPA009650, Sigma-Aldrich [SA], St. Louis, MO), or S-nitroso-cysteine (SNO-CYS) (1:1000; N5411, SA) at 4°C overnight. Membranes were washed in tris-buffered saline/0.05% Tween (28358, TS/170-6531, BR) and then incubated 30 min with secondary antibody (1:5000; GE Healthcare, Sweden). Membranes were visualized using SuperSignal Substrate (34075, TS) and ChemiDoc XRS+ (BR). Blots were evaluated using Image Lab (BR). SNO-CYS antibody generally recognizes S-nitrosylated proteins; our analysis revealed one abundant band at ~36 kDa, and this band was evaluated (for full images of representative blots, see Figure, Supplemental Digital Content 1, Representative blots for annexin A6, https://links.lww.com/MSS/A849; see Figure, Supplemental Digital Content 2, Representative blots for CaMKII immunoblotting, https://links.lww.com/MSS/A850; see Figure, Supplemental Digital Content 3, Representative blots for SNO-CYS immunoblotting, https://links.lww.com/MSS/A851). All analyses were normalized to total protein content quantified by SYPRO Stain (BR).
Rating of Perceived Pain
Subjects were asked to rate their level of perceived pain in their thigh musculature on a 0- to 100-mm visual analog scale (VAS) before cuff inflation (Pre) as well as immediate and 1 min upon cuff release (Post0 and Post1). VAS scoring of zero was designated as “no pain,” whereas 100 mm was denoted as “worst imaginable pain.” A statically significant VAS change of ≥12 mm has previously been reported to represent a significant important change in pain perception (Kelly, 2001), and thus a similar change (≥12 mm) was used in the present study to denote a significant meaningful change in VAS scoring.
Statistical Analysis
Assumptions of variance homogeneity and Gaussian distribution were verified, and data with a non-Gaussian distribution were transformed accordingly. A mixed linear model was used to assess changes in parameters with subject-ID as a random effect and time and group as fixed effects (23,27). Exceptions were VAS (ordinal data) and annexin A6 (non-Gaussian); these were analyzed using Wilcoxon signed rank/sum testing. Associations between relevant parameters were evaluated with Pearson's product–moment correlation analysis. All analyses were performed using STATA 12.1 (StataCorp, US). The α-level was set at 0.05. Values are presented as means ± SD, except for VAS data presented as median, 25th–75th percentiles.
RESULTS
Two subjects from the BFR group and one from the CON group left the project prematurely because of circumstances unrelated to the study intervention or experimental procedures.
Skeletal Muscle Contractile Properties
Isokinetic maximal voluntary contraction
Maximal isokinetic knee extensor strength at slow angular velocity (30°·s−1) showed a group–time interaction (P < 0.05). It remained unchanged in both groups at Post5 but increased 8%–9% in BFR subjects (P ≤ 0.01) while remaining unchanged in CON when evaluated Post12 (Fig. 2). A main time effect (P < 0.01) was observed for knee extensor strength obtained at fast angular velocity (240°·s−1); further analysis revealed a 7.7% decrease at Post5 (P ≤ 0.01) for BFR, whereas no changes were observed in CON (Fig. 2).
FIGURE 2: Maximal isokinetic knee extensor muscle strength obtained at slow (30°·s−1) (top panel) and fast (240°·s−1) (bottom panel) knee joint angular velocity, Pre as well as 5 and 12 d posttraining test data are displayed for BFR (left columns) and CON (right columns). Pre to Post differences: *P ≤ 0.01; **P ≤ 0.001. Post5 to Post 12 differences: †P ≤ 0.001. BFR, n = 10; CON, n = 8.
Voluntary rate of force development
Knee extensor voluntary contractile RTD (vRTD) showed time–group interactions for all time intervals (0–30, 0–50, 0–100, and 0–200 ms) (P < 0.05). vRTD remained unchanged at Post5 in BFR for all time intervals examined except for the 0–200 ms interval, where vRTD increased 10.5% in BFR (P ≤ 0.01) (Fig. 3). At Post12, vRTD increased compared with Pre and Post5 in BFR subjects during the initial phase of rising muscle force (0–30 and 0–50 ms) as well as in the later vRTD phase (0–100 ms), corresponding to relative increases of 9%–20% (P ≤ 0.05–0.001) (Fig. 3), whereas RTD in the 0–200 ms time interval remained increased at Post12 (P < 0.001). By contrast, vRTD remained unaltered in CON at all time points.
FIGURE 3: Contractile RTD measured in the 0- to 30-, 0- to 50-, 0- to 100-, and 0- to 200-ms time intervals (t = 0 denoting onset of contraction). Pre as well as 5 and 12 d posttraining test data are shown for BFR (left columns) and CON (right columns). Pre to Post differences: *P ≤ 0.01; **P ≤ 0.001. Post5 to Post12 differences: §P < 0.05; §§P ≤ 0.01; Time–group: ¤P < 0.05. BFR, n = 10; CON, n = 8.
Evoked twitch properties
Resting twitch peak torque decreased 18.7% from baseline to Post5 (P ≤ 0.001), while returning to baseline level at Post12 after BFR training (Table 1). Likewise, decreases of 21%–23% in rT-RTD were observed between Pre and Post5 in the 0–30 and 0–50 ms time intervals, while returning to baseline at Post12. At Post5, resting twitch RTD and RTR at 50% of PT were reduced and increased, respectively (P ≤ 0.001), while showing no differences at Post12 relative to baseline. Relative resting twitch RTD in the 0–30 and 0–50 ms time intervals remained unchanged Pre to Post5, whereas an increase was observed from Post5 to Post12 (P ≤ 0.05) (Table 1).
TABLE 1: Evoked twitch contractile parameters for BFR group.
Potentiated twitch properties showed a similar response because peak torque, RTD (0–30 and 0–50 ms), and RTD at 50% PT demonstrated decreases at Post5 (P ≤ 0.001), whereas RTD at 50% of PT increased (P ≤ 0.001), and these parameters returned to baseline levels at Post12. Relative pT-RTD (0–30 and 0–50 ms) intervals decreased by 9.2% and 12.3% between Pre and Post5 (P ≤ 0.01; P ≤ 0.05), while returning to baseline levels at Post12 (Table 1).
Muscle Fiber Cross-Sectional Area
Data for mean MFA revealed a time–group interaction (P < 0.05). An increase from Pre to Mid8, Post3, and Post10 in BFR emerged (P < 0.001) (Fig. 4). In CON, a transient increase in MFA was seen from Pre to Mid8 (P < 0.01), returning back to baseline at Post3.
FIGURE 4: Quantification of mean fiber cross-sectional area and proteins related to exercise-induced stress in BFR and CON subjects before (Pre) and after 8 d (Mid8) as well as 3 and 10 d postexercise (Post3 and Post10). Mean fiber cross-sectional area (upper left), annexin A6 (upper right), CaMKII (lower left), and SNO-CYS (lower right). Annexin A6, CaMKII, and SNO-CYS data presented as percentage of baseline values. *Significantly different from baseline (P < 0.05). Time–group interaction: ¤P < 0.05. Representative immunoblots are depicted above each graph for the specific protein targets. WB—BFR, n = 10 (Mid8 = 9); CON, n = 8 (Mid8 = 7). MFA—BFR, n = 10 (Mid8 = 9); CON, n = 7.
Protein Quantification
A tendency to a main time effect for CaMKII expression emerged (P = 0.0792). Further analysis revealed an increase in CaMKII content in BFR at Post3 (57%) and Post10 (71%) (P < 0.05), while remaining unchanged in CON (Fig. 4). We also attempted to quantify phos-CaMKII (Thr286) but were unable to gather reliable data. For SNO-CYS, a time–group interaction emerged (P < 0.05), whereas an increase of 44% was observed for CON at MID8 (P < 0.01) (Fig. 4). Annexin A6 remained unchanged in both groups.
Rating of Perceived Pain
Median VAS score data are depicted in Fig. 5. No difference was found in VAS scoring between groups immediate before exercise sessions, whereas an overall mean difference was observed immediately and 2 min after cessation of the fourth exercise set (BFR/CON Post0: 37, 31–44/1, 1–3; Post2: 5, 5–8/1, 0–1) (P < 0.001). Because VAS scoring recorded immediately after CON exercise was initially low (first week: 3, 2–4) and remained low during the remaining intervention (second and third week: ≤5), further analysis was conducted in the BFR group only. In BFR, a marked increase and decline in VAS scoring during the first set emerged, when Post0 (56, 29–67) was compared with Pre (0, 0–2) and Post2 (5, 1–17), respectively (P ≤ 0.01). The difference between these time points remained ≥20 and ≥19, respectively; thus, no extra analyses were performed. For VAS scores obtained immediately after exercise (Post0), a significant change exceeding the cutoff (12 mm) was observed between the first and the third training sessions (ΔVAS: 13; P < 0.05), and this difference remained elevated (>12) during the remainder of the intervention period (ΔVAS: 15–36) except between the first and fifth session (<12).
FIGURE 5: Number of repetitions and VAS scoring during the time course of low-load BFR training. Data for CON group, error bars, and markings of statistical significance have been omitted for clarity. Curves represent mean values. BFR, n = 10.
Correlations
After BFR training, a positive relationship emerged between the relative change in isometric MVC and MFA (Pre-to-Post10/12: r = 0.70, P = 0.02), whereas a tendency toward a positive relationship was observed between the relative change (Post5 vs Post12) in voluntary and evoked RTD (0- to 50-ms interval) (r = 0.60, P = 0.06).
DISCUSSION
The present study is the first to examine the change in rapid muscle force characteristics (vRTD) in response to BFR low-load resistance training. Notably, gains in rapid force capacity were observed 12 d after cessation of high-frequent BFR training (Post12), whereas no changes could be detected when evaluated 5 d after training (Post5). Likewise, BFR training led to increases in slow-speed dynamic (isokinetic) knee extensor strength in the late phase of posttraining recovery (Post12) while absent in the early recovery phase (Post5). No changes in mechanical muscle function were observed when work-matched training was performed in free-flow condition (CON).
Similar to previous BFR training studies (21,25,33,34), we report increases in isometric (~11%) and slow velocity isokinetic MVC (~9%). It should be recognized, however, that isometric and slow-velocity isokinetic peak torques are usually achieved at ≥300 ms after contraction onset (2). By contrast, numerous functional movements (i.e., recovery of postural stability during tripping) have to be performed within a short time frame (0–200 ms) (2,28), suggesting that the successful outcome of such movements requires a certain level of force within a limited time frame. Thus, the present findings of improved vRTD (11%–20%) 12 d after 3 wk of high-frequency BFR training (Post12) may be of functional importance, especially to individuals demonstrating reduced levels of RTD such as frail elderly individuals and clinical patient populations.
In support of our initial hypothesis, delayed increases were observed in majority of the voluntary contractile properties (isokinetic MVC and vRTD) after BFR training. The majority of the voluntary contractile parameters remained unchanged 5 d after training (e.g., vRTD 0–30, 0–50, and 0–100 ms and 30°·s−1 MVC), whereas small (~7%–10%) albeit statistically significant gains were observed in isometric MVC and RTD (0–200 ms) at Post5. By contrast, fast-velocity (240°·s−1) isokinetic strength declined at Post5. Likewise, all evoked contractile parameters were negatively affected when evaluated at Post5. Together, these observations seem to indicate a pattern of RTD dependence, as high velocity isokinetic peak torque (240°·s−1) and early-phase vRTD (within 0–100 ms) were obtained in a narrow time interval relative to force onset (<200 ms) (1,2). Conversely, parameters showing increases in the early phase after training (vRTD 0–200 ms, isometric MVC; Post5) were obtained ≥200 ms after the onset of contraction. This pattern was supported by a parallel decline in evoked contractile properties; for example, reduced peak twitch torque and tRTD (≤70 ms from force onset), including decreased relative RTD of potentiated twitches. The increase in late phase vRTD (0–200 ms) observed at Post5 may be related to the increase in isometric MVC, as these parameters have shown to correlate (r = 0.89) (4).
The discrepancy in response between voluntary contractile parameters is intriguing and may be primarily explained by physiological adaptations taking place between baseline and Post5 and/or this time-dependent factor. In relation to the former, neural adaptations cannot be excluded to contribute to the present study data. Previous studies have shown no changes in either central activation or surface EMG amplitude after 8 and 12 wk of low-load BFR training (21,25), although reports of enhanced neuromuscular activity after BFR training also exist (24). Nevertheless, we suggest that a main part of the increase in isometric MVC at Post5 would largely be explained by the substantial degree of myofiber hypertrophy (~35%) observed at this time point.
It is noteworthy that substantial elevations in rapid force capacity as well as in slow and fast velocity muscle strength emerged between the early (Post5) and the late (Post12) posttraining test sessions despite the fact that no exercise stimuli was applied between these time points. Accordingly, these changes must rely on a delayed recovery effect that might involve modulatory changes in intrinsic force capacity of myofibers and/or altered neuromuscular properties of the CNS.
Unfortunately, central (CNS) parameters were not assessed in the present study. Cook et al. (9) recently reported a lack of change in central activation and surface EMG activity during MVC immediately (30 s) after an acute bout of BFR exercise, whereas a decline in both voluntary and evoked muscle contractile performance was observed. In addition, decreases in central activation and neuromuscular activity immediately after an acute bout of BFR exercise have also been reported (19). Conversely, using transcranial magnet stimulation signs of elevated excitability in corticospinal pathways were recently demonstrated to occur acutely after a single BFR exercise session (6), suggesting that BFR training may elicit longitudinal adaptations in neuromuscular function. Specifically addressing the aspect of long-term neural adaptation to BFR training, 5 wk of low-load BFR training increased EMG amplitude (24), whereas contrary 8 and 12 wk of low-load BFR training did not evoke changes in central activation or surface EMG activity (21,25). Together, these somewhat conflicting observations indicate that BFR exercise may have positive, negative, or no effects on central activation and efferent neuromuscular activity during maximal voluntary contraction, possibly as a result of varying exercise protocols. In this perspective, it is unclear whether central fatigue could be involved in the apparent depression of mechanical muscle function in the early phase after BFR exercise (Post5). To the best of our knowledge, no data exist to support long lasting central fatigue (>24 h) in response to BFR exercise or in the general literature. Thus, central fatigue seems unlikely to have affected voluntary contractile function evaluated 5 d after cessation of training in the present study.
In addition to the absence of gains in voluntary RTD parameters in the early phase after BFR training (day 5), evoked twitch peak torque and twitch RTD were negatively affected (19%–23%), indicating that intrinsic myofiber contractility was impaired. This further suggests that the nonresponsiveness in voluntary contractile properties observed 5 d after training, at least in part, could be contributed to a peripheral origin. The evoked twitch torque to voluntary isometric MVC ratios decreased at Post5 and showed a strong tendency (P = 0.054–0.057) to remain reduced at Post12, whereas a similar pattern emerged when RTD was evaluated (data not shown). Hence, evoked contractile properties seem to be more severely affected than voluntary force capacity, especially when evaluated during the early phase of posttraining recovery (Post5). Interestingly, similar results have been observed previously, showing an increase in maximal voluntary peak torque (8%) in parallel with a decrease in resting twitch torque (−21%) ~48 h after 8 wk of elbow flexor BFR training, whereas changes in potentiated twitch peak torque (−4%) failed to reach significance (25). Altogether, these data suggest that BFR training may result in a transient decrement in intrinsic function of the peripheral contractile apparatus when evaluated in the early phase after training.
The present discrepancy in the time course of adaptation in voluntary and evoked contractile properties might be a result of training-induced muscle hypertrophy and/or impaired rate sensitive sites in EC coupling.
Several time/rate sensitive sites in the activation chain of the contractile apparatus could potentially have influenced the voluntary contractile capacity obtained at Post5. However, as evoked contractile parameters appeared more heavily affected than voluntary parameters, sites downstream to the neuromuscular junction are likely to play an essential role. In support hereof, relative RTD of resting and potentiated evoked twitches increased with BFR training from Pre and Post5 to Post12 and Post5 to Post12, respectively, all together pointing toward an improvement in one or more rate-dependent segments of EC coupling (e.g., sarcoplasmic reticulum [SR] Ca2+ release). Furthermore, a tendency to a positive relationship in the relative change (Post5 vs Post12) of voluntary and evoked RTD (0- to 50-ms interval) emerged (r = 0.60, P = 0.06), indicating that peripheral structural modifications taking place between these time points are candidates to explain the observed gain in vRTD in this phase of posttraining recovery.
Evoked twitch contractile properties could be negatively affected by inherent myocellular factors such as transient changes in fiber type distribution (Myosin Heavy Chain isoform content) and/or failure in EC coupling, including impairments in SR Ca2+ release, and/or be directly reflecting impairments in the contractile machinery. However, the former parameter is not likely to have played a significant role in the present study, as the present BFR training regime does not seem to stimulate fiber type transformations (27).
Lasting impairments (>24 h) in contractile function seem to be observed only in response to direct mechanical insult to the contractile apparatus (10,36) and/or in the presence of low-frequency fatigue (LFF) (3,18). However, mechanical disruption of the cytoskeletal apparatus as sometimes seen with highly intense eccentric exercise (10,36) is unlikely to have influenced the present results, as both the magnitude of mechanical loading of the contractile apparatus and the total training volume were low. In support of this notion, we and others have found no or only minor indications of ultrastructural myocellular damage after acute or longitudinal BFR exercise (11,27). It should be recognized, however, that marked elevations in indirect indicators of myocellular damage have been reported after acute BFR exercise performed to voluntary failure; that is, ≥20-fold rise in plasma CK and persistent (days) decline in isometric force producing capacity (31,38), whereas BFR exercise performed submaximally does not seem to result in prolonged impairments in contractile muscle function (22).
LFF is normally attributed to muscular factors and has largely been associated with severely fatiguing exercise and unaccustomed eccentric exercise (3,18). Notably, LFF was recently observed after an acute bout of BFR knee extensor exercise in recreationally active young subjects (cf. attenuated 20/100 Hz ratio) (31). LFF is thought to be caused by impairments in EC coupling, mainly associated with attenuated Ca2+ release (3,39). This may be initiated by (i) focal elevations of intracellular Ca2+ stimulating calpain-induced degradation of myogenic proteins, triggering attenuation of SR Ca2+ release (26,35), or (ii) formation of reactive oxygen and/or nitrogen species that can initiate structural modifications of SR Ca2+ release channels through oxidation and/or nitrosylation (13).
To investigate the potential involvement of these myocellular factors, muscle protein analyses related to Ca2+ exposure and nitrosylation of myogenic proteins were performed in the present study. CaMKII, which is thought to be an important Ca2+ sensor (14), was 60%–70% elevated 3 and 10 d after cessation of BFR training, whereas a somewhat comparable albeit nonsignificant change (48%) was observed at Post3 in controls. Thus, the observed variation (although invariant) in CaMKII expression with free-flow training raises some uncertainty as to whether the observed change in CaMKII expression with BFR training can be directly linked to the concurrent delay in early-phase contractile function that was noted in this intervention group.
Consequently, BFR training may induce a somewhat higher degree of myocellular Ca2+ exposure than work-matched free-flow training. This could be a result of Ca2+ transients of larger amplitude and/or longer duration with BFR training compared with free-flow conditions (12), which in turn may be ascribed to a higher contractile/metabolic demand of BFR exercise (i.e., contraction to failure, limited O2 availability [16], and accumulation of metabolites [33,34]). Furthermore, increases in intracellular [Ca2+] could potentially result from increased sarcolemma permeability, as previously indicated after acute BFR training (38). However, Ca2+ influx as a result of increased sarcolemma permeability due to; for example, membrane-related microruptures (cf. no signs of ultrastructural cytoskeletal damage), would activate sarcolemma resealing involving annexin A6 (30). The present lack of change in annexin A6 expression with BFR training suggests that sarcolemma microruptures were probably not a major factor or that the degree of microdamage was insufficient of inducing changes in annexin A6 expression.
Higher Ca2+ exposure of myocellular components in BFR subjects could potentially initiate calpain-induced proteolysis of proteins involved in Ca2+ release (15,26,35), which in turn could have caused the observed declines in acute and subacute evoked twitch responses. Interestingly, in terms of evoked twitch properties, we observed a comparable time course and similar magnitude of decrement in tRTD and tRTR after cessation of BFR training (Post3 and Post10), indicating that the mechanisms involved in SR Ca2+ release and uptake might be overlapping. Ca2+ has been shown to negatively influence sarcoplasmic/endoplasmic reticulum Ca2+ ATPase (SERCA) in skeletal muscle (15). Thus, the Ca2+-induced proteolysis of proteins involved in EC coupling could potentially be a plausible common denominator for the prolonged decline in evoked tRTD and tRTR observed after BFR training.
However, the present data related to the potentially elevated myocellular Ca2+ exposure with BFR training compared with free-flow training do not seem unambiguous. Thus, future studies should explore the potential association between BFR training, intracellular Ca2+ transients, and intrinsic myocellular contractility.
An increase in nitrosylated proteins (SNO-CYS) was observed in the controls at Mid8, whereas BFR subjects remained unaffected. Nitric oxide–induced nitrosylation has been attributed to skeletal muscle fatigue, negatively affecting Ca2+ regulating proteins such as the ryanodine receptor and SERCA (5,13). As mechanical muscle function was not assessed during the intervention, it is impossible to establish a link between the increase in SNO-CYS and muscle function here. This SNO-CYS expression pattern was unexpected, as BFR exercise involve several parameters (e.g., hypoxia, reperfusion) thought to be prerequisites for nitric oxide production (20).
The initial perceived pain (VAS) score during the first bout of the present BFR exercise protocol of ~57 (100 = worst imaginable pain) may represent a practical concern, as an exercise regime requiring such high pain tolerance may mostly be applicable to highly motivated individuals. The increase in VAS is likely a result of local accumulation of metabolites (33,34), which provide pain feedback through group III and IV afferents. Importantly, the VAS score showed a consistent decline (~50%) during the 3-wk BFR training, leveling-off during the final week (VAS ~30.0). This may reflect increased pain tolerance and/or metabolic adaptation, as local metabolite accumulation would be expected to remain largely unaltered throughout the intervention period. Importantly, subjects reported low VAS scores (mean < 6) 1 min after exercise; thus, although high during exercise, the perception of pain subsided rapidly upon cuff release, indicating that the afferent pain response is indeed related to the local buildup of metabolites.
Methodological considerations
A potential limitation of the present study was the use of an identical absolute cuff pressure across all individuals. Hence, the absolute pressure involved (100 mm Hg) may have resulted in some variation in arterial inflow during exercise (17). Little is known about the physiological effect of variations in arterial inflow with BFR exercise, and consequently we cannot dismiss that the potential variability in arterial inflow among BFR subjects may have influenced the present results to some degree. In addition, the work matching between the BFR and the CON groups assumes that both groups have similar endurance capacity, which was not tested and thus remains unknown. However, the observation of similar levels of maximal KE strength and similar fiber type composition (type I vs II fiber type proportion) in the two intervention groups (27) does not indicate that a substantial difference in endurance capacity should exist between BFR and CON subject. Because of logistic problems (acute sickness of the tester), evoked muscle twitch recordings were not obtained in the present CON subjects, for which reason it was not possible to determine whether CON subjects had a systemic decline in evoked contractile properties similar to that observed after BFR training. However, this scenario may not be highly likely, as no changes were observed for any of the voluntary muscle strength parameters assessed at any time point in CON.
CONCLUSION
Mechanical muscle function was substantially improved in subjects exposed to high-frequency (twice a day) low-load BFR resistance training, whereas no response was seen in control subjects performing a work-matched exercise protocol in free-flow conditions. However, gains in rapid force capacity (RTD) and isokinetic MVC (slow velocity) were not manifested before 12 d after cessation of the final exercise session, whereas the remaining were unaltered or even negatively affected in the early phase (5 d) of training recovery. The delayed increase in contractile muscle function was suggested to be mainly of peripheral origin, as evoked contractile muscle twitch properties showed a similar pattern of suppression in the early (5 d) followed by subsequent normalization in the late (12 d) phase of training recovery.
The authors address their appreciation to the volunteer subjects who participated in this study. In addition, they thank the Danish Ministry of Culture, the Region of Southern Denmark, the Danish Council for Independent Research, and the foundation of A.P. Møller and his wife Chastine Mc Kinney Møller for their financial support. All authors declare no conflict of interest. The results of this study do not constitute endorsement of the techniques used by the American College of Sports Medicine.
The contributions of the authors were as follows: conception and design of the study—J. N., U. F., and P. A.; collection, analysis, and interpretation of data—J. N., U. F., T. P., R. B., T. N., C. S., and P. A.; drafting the article or revising it critically for important intellectual content—J. N., U. F., T. P., R. B., T. N., C. S., and P. A. All authors have approved the final version of the manuscript.
REFERENCES
1. Aagaard P, Andersen JL. Correlation between contractile strength and myosin heavy chain isoform composition in human skeletal muscle.
Med Sci Sports Exerc. 1998;30(8):1217–22.
2. Aagaard P, Simonsen EB, Andersen JL, Magnusson P, Dyhre-Poulsen P. Increased rate of force development and neural drive of human skeletal muscle following resistance training.
J Appl Physiol (1985). 2002;93(4):1318–26.
3. Allen DG, Lamb GD, Westerblad H. Skeletal muscle fatigue: cellular mechanisms.
Physiol Rev. 2008;88(1):287–332.
4. Andersen LL, Aagaard P. Influence of maximal muscle strength and intrinsic muscle contractile properties on contractile rate of force development.
Eur J Appl Physiol. 2006;96(1):46–52.
5. Bellinger AM, Reiken S, Dura M, et al. Remodeling of ryanodine receptor complex causes “leaky” channels: a molecular mechanism for decreased exercise capacity.
Proc Natl Acad Sci U S A. 2008;105(6):2198–202.
6. Brandner CR, Warmington SA, Kidgell DJ. Corticomotor excitability is increased following an acute bout of blood flow restriction resistance exercise.
Front Hum Neurosci. 2015;9:652.
7. Chin ER. Role of Ca
2+/calmodulin-dependent kinases in skeletal muscle plasticity.
J Appl Physiol (1985). 2005;99(2):414–23.
8. Clark DJ, Manini TM, Fielding RA, Patten C. Neuromuscular determinants of maximum walking speed in well-functioning older adults.
Exp Gerontol. 2013;48(3):358–63.
9. Cook SB, Murphy BG, Labarbera KE. Neuromuscular function after a bout of low-load blood flow–restricted exercise.
Med Sci Sports Exerc. 2013;45(1):67–74.
10. Crameri RM, Aagaard P, Qvortrup K, Langberg H, Olesen J, Kjaer M. Myofibre damage in human skeletal muscle: effects of electrical stimulation versus voluntary contraction.
J Physiol. 2007;583(Pt 1):365–80.
11. Cumming KT, Paulsen G, Wernbom M, Ugelstad I, Raastad T. Acute response and subcellular movement of HSP27, αB-crystallin and HSP70 in human skeletal muscle after blood-flow-restricted low-load resistance exercise.
Acta Physiol (Oxf). 2014;211(4):634–46.
12. De Koninck P, Schulman H. Sensitivity of CaM kinase II to the frequency of Ca2+ oscillations.
Science. 1998;279(5348):227–30.
13. Dutka TL, Mollica JP, Posterino GS, Lamb GD. Modulation of contractile apparatus Ca2+ sensitivity and disruption of excitation-contraction coupling by S-nitrosoglutathione in rat muscle fibres.
J Physiol. 2011;589(Pt 9):2181–96.
14. Eilers W, Gevers W, van Overbeek D, et al. Muscle-type specific autophosphorylation of CaMKII isoforms after paced contractions.
Biomed Res Int. 2014;2014:943806.
15. French JP, Quindry JC, Falk DJ, et al. Ischemia-reperfusion-induced calpain activation and SERCA2a degradation are attenuated by exercise training and calpain inhibition.
Am J Physiol Heart Circ Physiol. 2006;290(1):H128–36.
16. Ganesan G, Cotter JA, Reuland W, Cerussi AE, Tromberg BJ, Galassetti P. Effect of blood flow restriction on tissue oxygenation during knee extension.
Med Sci Sports Exerc. 2015;47(1):185–93.
17. Hunt JE, Stodart C, Ferguson RA. The influence of participant characteristics on the relationship between cuff pressure and level of blood flow restriction.
Eur J Appl Physiol. 2016;116(7):1421–32.
18. Jones DA, Newham DJ, Torgan C. Mechanical influences on long-lasting human muscle fatigue and delayed-onset pain.
J Physiol. 1989;412:415–27.
19. Karabulut M, Cramer JT, Abe T, Sato Y, Bemben MG. Neuromuscular fatigue following low-intensity dynamic exercise with externally applied vascular restriction.
J Electromyogr Kinesiol. 2010;20(3):440–7.
20. Korthuis RJ, Granger DN, Townsley MI, Taylor AE. The role of oxygen-derived free radicals in ischemia-induced increases in canine skeletal muscle vascular permeability.
Circ Res. 1985;57(4):599–609.
21. Kubo K, Komuro T, Ishiguro N, et al. Effects of low-load resistance training with vascular occlusion on the mechanical properties of muscle and tendon.
J Appl Biomech. 2006;22(2):112–9.
22. Loenneke JP, Thiebaud RS, Fahs CA, Rossow LM, Abe T, Bemben MG. Blood flow restriction does not result in prolonged decrements in torque.
Eur J Appl Physiol. 2013;113(4):923–31.
23. Malcata RM, Hopkins WG, Pearson SN. Tracking career performance of successful triathletes.
Med Sci Sports Exerc. 2014;46(6):1227–34.
24. Manimmanakorn A, Manimmanakorn N, Taylor R, et al. Effects of resistance training combined with vascular occlusion or hypoxia on neuromuscular function in athletes.
Eur J Appl Physiol. 2013;113(7):1767–74.
25. Moore DR, Burgomaster KA, Schofield LM, Gibala MJ, Sale DG, Phillips SM. Neuromuscular adaptations in human muscle following low intensity resistance training with vascular occlusion.
Eur J Appl Physiol. 2004;92(4–5):399–406.
26. Murphy RM, Dutka TL, Horvath D, Bell JR, Delbridge LM, Lamb GD. Ca
2+-dependent proteolysis of junctophilin-1 and junctophilin-2 in skeletal and cardiac muscle.
J Physiol. 2013;591(Pt 3):719–29.
27. Nielsen JL, Aagaard P, Bech RD, et al. Proliferation of myogenic stem cells in human skeletal muscle in response to low-load resistance training with blood flow restriction.
J Physiol. 2012;590(Pt 17):4351–61.
28. Pijnappels M, Bobbert MF, van Dieën JH. How early reactions in the support limb contribute to balance recovery after tripping.
J Biomech. 2005;38(3):627–34.
29. Pijnappels M, van der Burg PJ, Reeves ND, van Dieën JH. Identification of elderly fallers by muscle strength measures.
Eur J Appl Physiol. 2008;102(5):585–92.
30. Potez S, Luginbühl M, Monastyrskaya K, Hostettler A, Draeger A, Babiychuk EB. Tailored protection against plasmalemmal injury by annexins with different Ca
2+ sensitivities.
J Biol Chem. 2011;286(20):17982–91.
31. Sieljacks P, Matzon A, Wernbom M, Ringgaard S, Vissing K, Overgaard K. Muscle damage and repeated bout effect following blood flow restricted exercise.
Eur J Appl Physiol. 2016;116(3):513–25.
32. Suetta C, Aagaard P, Rosted A, et al. Training-induced changes in muscle CSA, muscle strength, EMG, and rate of force development in elderly subjects after long-term unilateral disuse.
J Appl Physiol (1985). 2004;97(5):1954–61.
33. Takada S, Okita K, Suga T, et al. Low-intensity exercise can increase muscle mass and strength proportionally to enhanced metabolic stress under ischemic conditions.
J Appl Physiol (1985). 2012;113(2):199–205.
34. Takarada Y, Takazawa H, Sato Y, Takebayashi S, Tanaka Y, Ishii N. Effects of resistance exercise combined with moderate vascular occlusion on muscular function in humans.
J Appl Physiol (1985). 2000;88(6):2097–106.
35. Verburg E, Dutka TL, Lamb GD. Long-lasting muscle fatigue: partial disruption of excitation-contraction coupling by elevated cytosolic Ca2+ concentration during contractions.
Am J Physiol Cell Physiol. 2006;290(4):C1199–208.
36. Warren GL, Ingalls CP, Lowe DA, Armstrong RB. Excitation–contraction uncoupling: major role in contraction-induced muscle injury.
Exerc Sport Sci Rev. 2001;29(2):82–7.
37. Wernbom M, Augustsson J, Thomeé R. The influence of frequency, intensity, volume and mode of strength training on whole muscle cross-sectional area in humans.
Sports Med. 2007;37(3):225–64.
38. Wernbom M, Paulsen G, Nilsen TS, Hisdal J, Raastad T. Contractile function and sarcolemmal permeability after acute low-load resistance exercise with blood flow restriction.
Eur J Appl Physiol. 2012;112(6):2051–63.
39. Westerblad H, Bruton JD, Allen DG, Lännergren J. Functional significance of Ca2+ in long-lasting fatigue of skeletal muscle.
Eur J Appl Physiol. 2000;83(2–3):166–74.
