Increased muscle force does not induce greater stretch-induced damage to calf muscles during work-matched heel drop exercise

Twelve recreationally-active volunteers (age 21 ± 2.8 years, mass 71.9 ± 13.5 kg, height 178.2 ± 11.1 cm; six females) provided written informed consent to participate in the study. Participants were excluded if they had any pre-existing lower limb injuries in the past six months and/or had undertaken eccentric training exercises specific to the calf muscle group within six months before the experiment. The protocol was approved by the local university ethics committee and conducted according to the Declaration of Helsinki.

The exercise consisted of single-leg eccentric heel drops performed on the edge of a wooden step as previously described (Alfredson et al. 1998). In brief, standing upright with the knee fully extended, BW on the forefoot, and the ankle joint in full plantar flexion, the ankle extensors were eccentrically loaded by having the participant lower the heel beneath the level of the forefoot. The non-test leg returned the participant to the start position such that the test leg performed only eccentric contractions. A metronome controlled exercise speed and frequency, while video feedback ensured that the ankle range of motion was consistent.

In a randomised within-subject design, participants performed one bout of eccentric exercise with each leg as part of either a low- or high-load condition. The low-load condition consisted of heel drops performed with the participant’s BW. The high-load condition consisted of performing heels drops with BW while wearing a weight vest with 30% BW. This method of load standardisation has been used previously to set relative training/exercise loads for plantar flexor muscles in basic science/cross sectional studies (Pincheira et al. 2021a) and clinical/interventional (Alfredson et al. 1998; Mafi et al. 2001) studies. This added BW was selected based on previous studies (Pincheira et al. 2021a) and allows to maximise the exercise load of the triceps surae in an ecologically sound manner, while avoiding the interference of weight-vest-induced fatigue (e.g., shoulder/neck muscles) on exercise performance and movement kinematics. Further, this high-load condition largely exceeds the heavy-load prescribed in common heel drop protocols used for calf muscle training (Alfredson et al. 1998), and is similar to the increase in plantar flexor torque seen when transitioning from walking to jogging (Schache et al. 2011). Each condition was completed approximately two weeks apart. The exercise in each condition consisted of sets of ~ 20 heel drops separated by a 2-min rest period. The total work performed for each condition was matched between legs to standardise the training volume (see below for details).

To analyse the effect of load on MG EIMD, total twitch torque, MG soreness scores, MG passive stiffness and MG active fascicle length at maximum twitch torque were measured before (PRE), two hours (2H) and 48 h (48HR) after each bout of exercise. A significant change in total twitch torque and/or MG soreness was considered as evidence of EIMD (Hoffman et al. 2014; Pincheira et al. 2018). Twitch torques elicited by supramaximal electrical stimulation were used to assess force loss from the contractile tissue, avoiding any influence of peripheral muscle fatigue during post-exercise testing (Guilhem et al. 2016; Carroll et al. 2016). MG fascicle length changes during plantar flexion twitches and changes in passive stiffness were collected as complimentary muscle damage outcomes. MG stretch, triceps surae electromyographic (EMG) activity (MG, lateral gastrocnemius (LG) and soleus (SOL)), MG MTU stretch and ankle range of motion were collected to characterize lower leg neuromechanical behaviour during the exercise bouts.

Total twitch torque was determined using a modified version of a method (presented previously by our group) used to estimate the maximum torque occurring on the length-tension relationship of the gastrocnemius muscle (Hoffman et al. 2012, 2014; Pincheira et al. 2018, 2021a). Total twitch torque was calculated as the mean of peak plantar flexion twitch torques obtained across the ankle range of motion (15° plantar flexion to maximal dorsiflexion). Plantar flexion twitch torques were electrically-evoked by stimulating the tibial nerve percutaneously at the popliteal fossa with a constant-current stimulus (three square wave pulses, 20 ms inter-stimulus interval, 500 μs pulse width; DS7AH, Digitimer, Welwyn Garden City, UK). In brief, with the participants lying in the prone position and their foot firmly attached to a dynamometer footplate (Humac Norm, CSMi, Stoughton, USA), supramaximal stimuli (120% of the maximal twitch torque achieved at 15° of dorsiflexion) were applied at several different joint angles across the ankle range of motion (7 to 11 joint angles depending on the full range of motion of the participant) (Hoffman et al. 2014, 2016; Pincheira et al. 2018, 2021a). A single pulse was used 3–5 s prior to the triplet to minimize any thixotropic effect (Proske et al. 1993). Measurements were made for each twitch, at each joint angle, in a randomized fashion.

Self-reported muscle soreness was assessed using an algometer (Wagner Instruments, Riverside, USA). With the probe head of the algometer placed perpendicular to the mid-belly of the MG, the applied force was gradually increased until 2.5 kg of pressure was achieved. The participant then reported their perceived soreness of the muscle using a 10-point visual analogue scale (0: no soreness; 10: worst soreness ever felt).

MG passive stiffness was determined as the slope of a curve (k) fitted with an exponential expression to the relationship between passive MG fascicle length and passive plantar flexion torque (Hoffman et al. 2014, 2016; Pincheira et al. 2018). MG fascicle length was recorded during passive ankle rotations at a constant velocity (10 deg/s) through the full range of ankle motion, performed with the dynamometer described above. MG fascicles were captured using B-mode ultrasound (Echoblaster 128, LV7.5/60/96 40 mm transducer, 6 MHz, ~ 110 fps; Telemed, Vilnius, Lithuania). The flat-shaped transducer was firmly strapped to the MG using an elastic bandage. The location of the transducer relative to the skin was marked with an indelible marker for consistent placement in subsequent testing sessions. MG muscle fascicle length during passive ankle rotation was estimated offline using custom-written Matlab (The MathWorks, Natick, USA) scripts (Farris and Lichtwark 2016). Plantar flexion torque and ankle joint position were measured by the dynamometer and collected at a sampling rate of 2 kHz using an AD board (Micro 1401-3; Cambridge Electronic Design, Cambridge, UK).

MG fascicle length changes during plantar flexion twitches were measured using the ultrasonography methods described above. Ultrasound and torque measurements were synchronised via a continuous 5 V signal sent to the AD board from the ultrasound beamformer whenever fascicle images were being recorded. This method allowed active fascicle length to be quantified at the instant of peak plantar flexion torque during the elicited twitch. MG fascicle length was estimated automatically at the time of peak twitch torque utilising the Simple Muscle Architecture Analysis Software (Seynnes and Cronin 2020).

To analyse MG neuromechanical behaviour during the exercise bouts, lower limb kinematics, MG fascicle length and triceps surae muscle activation were measured synchronously. Lower limb kinematics were measured using an eight-camera optoelectronic motion capture system (200 Hz sampling frequency; Qualysis, Gothenburg, Sweden). Single reflective spherical markers were placed at anatomical landmarks on both legs as previously described (Hoffman et al. 2014, 2016). Labelled marker data were then exported to OpenSim, where a modified generic model (Arnold et al. 2010) scaled to each participant was used to estimate ankle range of motion (ROM) and MG MTU length in a process described elsewhere (Schache et al. 2013).

MG fascicle length changes during the heel drops were recorded using the ultrasonography methods described above. The same custom-written Matlab scripts mentioned above (Farris and Lichtwark 2016) were used to track MG fascicle length during the exercise bouts. Ten cycles of heel drops (taken from the beginning and end of each exercise bout) were analysed. The MTU-level stretch phase of each cycle was determined using the 3D position of a reflective marker located on participant’s calcaneus (Fig. 1). Absolute stretch amplitude (i.e., difference between maximum and minimum length) was calculated within the stretch phase.

Fig. 1figure 1

Neuromechanical measurements during the exercise bouts. Neuromechanical parameters of the medial gastrocnemius MG (fascicle length, muscle–tendon unit length (MTU), electromyography (EMG)) were estimated within the eccentric phase of the heel drops (shaded area within the blue dotted lines), defined by the vertical displacement of a marker placed on the heel. Ultrasound images of the change in fascicle lengths during one heel drop cycle are presented in panels (a) and (b)

Surface EMG signals from LG, MG and SOL were also recorded during the heel drops. Bipolar electrodes (2 cm inter-electrode distance) were placed over the muscle bellies according to well-known anatomical guidelines (Hermens et al. 1999), after appropriate skin preparation (shaving and alcohol cleansing). EMG signals were sampled at a frequency of 2 kHz and bandwidth-filtered at 10–500 Hz (Neurolog NL900D, Digitimer, Hertforsdire, UK). For EMG normalisation purposes, in the same session that heel drops were performed, maximal voluntary isometric contractions (MVIC) were recorded. For MVIC recording, the participants laid in the prone position with their foot firmly strapped to the dynamometer mentioned above. Three MVICs (separated by 3 min of rest) were exerted with the ankle at neutral position (i.e., the sole of the foot perpendicular to the tibia). The highest MVIC value was used as a reference for EMG normalisation.

For EMG activity estimation, MG, LG and SOL EMG root mean square (RMS) activity (50 ms window) were calculated and averaged over the same 10 cycles where MG fascicle stretch was estimated. EMG signals were then normalised to the RMS EMG obtained during the MVICs. The RMS EMG normalised values were used for further analyses.

Before the statistical analysis, all variables were tested for normality using the Shapiro–Wilk test. Deviations from sphericity were corrected using the Greenhouse–Geisser method. To analyse the effect of load on the markers of muscle damage, two-way repeated-measures ANOVAs were performed to determine the effect of time (PRE, 2HR, 48HR) and load (low-load, high-load) on total twitch torque, soreness score, MG passive stiffness and active fascicle length at maximum twitch torque. Pre-planned post-hoc comparisons were made using Tukey’s test when a significant interaction or main effect was found. To analyse the effect of load on MG neuromechanical behaviour during the heel drops, paired t-tests were used to compare MG absolute stretch, MG MTU absolute stretch, ankle ROM and triceps surae EMG (MG, LG and SOL) between the high- and low-load conditions. Significant differences were established at p ≤ 0.05. The Šidák correction was applied for multiple comparisons. Data are reported as means and standard deviations. Effect sizes are reported as partial eta-squared (ηp2).

留言 (0)

沒有登入
gif