Cortical and spinal responses to short-term strength training and detraining in young and older adults in rectus femoris muscle

Twenty-seven participants volunteered for the study (14 female). The recruitment process and exclusion of participants is shown in Fig. 1. Therefore, the data presented in Table 1 are representative of the 21 (11 young adults (6 female) and 10 older adults (6 female)) volunteers fulfilling all study requirements. All included participants were free from musculoskeletal injury in the lower-limbs for the last 6 months and neurologic illness, were not taking any medications known to affect the nervous system and had no contraindications to TMS, which was assessed via a health questionnaire (Rossi et al. 2011). Before testing, all participants were fully informed of the procedures and possible risks, and each participant provided written informed consent. The Ethical Committee of the University of Jyväskylä provided a statement for the study (857/13.00.04.00/2021) and the study was conducted in accordance with the ethical standards establish in the Declaration of Helsinki (2013).

Participants visited the laboratory on five different testing periods and one familiarization session (Fig. 2A). All participants were instructed to maintain their regular dietary habits up to two hours prior to the testing session, consume 500 ml of water immediately before the test, abstain from consuming caffeine within the 12 h leading up to the examination, and refrain from engaging in strenuous physical activities 48 h preceding each testing session. The study’s initial phase consisted of a familiarization session in which participants were introduced to all instructions and stimulation parameters pertinent to the subsequent testing sessions. This session also served as a preliminary assessment of the LS placement and the determination of TMS intensity for active motor threshold (aMT). Then, testing periods were defined as control testing (Con), pre-training testing (Pre), mid-training testing (Mid), post-training testing (Post) and detraining testing (De) (Fig. 2A). Every testing period was structured the same: A LS session, a TMS session and a one-maximum repetition session (1-RM) conducted within a 7-day period. Sessions for each participant were consistently scheduled at the same time of the day, and there was a 48- to 72-h interval between LS, TMS and 1-RM (Fig. 2B).

To assess responses in RF, participants sat in a custom-built chair with a calibrated load cell (Faculty of Sport and Health Sciences, University of Jyväskylä, Finland) with the hip and knee at 90° flexion and the shank strapped by a non-elastic restraint ~ 2 cm superior to the ankle malleoli. The voltage signal originating from the load cell was calibrated and converted into torque (N·m). All measures were performed on the right (i.e., dominant) leg assessed by self-report of which foot they primarily kick a ball (van Melick et al. 2017).

Every session followed the same structure. Once the participant was secured to the dynamometer, the maximum compound action potential (M-max) was assessed in a relaxed condition (i.e., M-maxpre). As a warm-up, two contractions at ~ 50 and ~ 80% of estimated MVC were performed. Then, two MVC trials were performed 60 s apart (i.e., MVCpre). Verbal encouragement and visual feedback were provided to motivate participants to produce maximal effort and torque was recorded. The reliability of this method was excellent (CV = 4.6%; ICC = 0.987).

In every testing session, visual feedback was provided to the participants to produce the required submaximal torque and then a single LS or TMS stimulus was delivered manually. Contractions at 20 and 60% of MVC were held for 5–8 s, because RF MEP amplitude seemingly increases until 50–75% of MVC (Martin et al. 2006; Oya et al. 2008; Goodall et al. 2009; Škarabot et al. 2019a). Sets of ten stimulations were given per condition and per contraction level as a single block, giving a total of 40 LS and 60 TMS stimulations. To avoid fatigue, 30 s and 45 s rest was given between contractions during 20 and 60% of MVC, respectively, and 60 s and 180 s rest was given between the sets of 10 contractions. At the end of the protocol, M-max (M-maxpost) and MVC (MVCpost) were re-assessed (Fig. 2 C and D).

Muscle activity was recorded using adhesive Ag/AgCl electrodes (30 × 20 mm, BlueSensor N, Ambu, Penang, Malaysia) from RF according to SENIAM guidelines (Hermens et al. 2000). Skin was shaved, abraded with sandpaper, and wiped with alcohol before positioning the electrodes in a bipolar arrangement with a 20 mm center-to-center distance. Impedance was set < 2 kΩ, and the reference electrode was positioned on the patella. EMG electrode positions were marked with a permanent marker over the skin, photographs were taken and the distance from the iliac crest to the middle of the electrode pair was recorded. In addition, during the training period, the marks were redrawn by the research assistant after every training session. EMG data were sampled online at 3000 Hz, amplified (1000 ×) and bandpass filtered (16–1000 Hz; Neurolog System, Digitimer Ltd, UK) using CED Power1401-3 (Cambridge Electronic Design Ltd, Cambridge, UK).

Torque was sampled at 1000 Hz, amplified by a custom-built amplifier (ForAmps 1 v1.2, University of Jyväskylä, Finland) and converted by a 16-bit A/D board (CED Power1401-3, Cambridge Electronics Design, Cambridge, UK) in combination with Spike2 software (version 6.10, Cambridge Electronic Design, Cambridge, UK).

Transcutaneous electrical stimulation of the femoral nerve (32 mm cathode/anode arrangement; Polar Neurostimulation Electrodes, Espoo, Finland) was performed to elicit M-max in RF (1 ms squared pulse duration; Digitimer DS7AH, Hertfordshire, UK). Electrodes were placed 2 cm apart and positioned at each side of the femoral nerve, located by palpation and identification of the femoral artery (Walker et al. 2016). M-max was elicited by gradually increasing stimulator output intensity until the EMG response plateaued. To ensure a supramaximal response was elicited, this intensity was further increased by 50% and two individual simulations were given (Table 1).

Transcutaneous electrical LS was used to elicit LEPs with a constant-current stimulator (1 ms square pulse duration; Digitimer DS7AH, Hertfordshire, UK) via self-adhesive electrodes (Polar Neurostimulation Electrodes, Espoo, Finland). Originally, the cathode (5 × 9 cm) was centered over the first lumbar vertebra and the anode (circular shape; 3.2 cm diameter) was placed on the midline of the vertebral column ~ 5 cm above the top edge of the cathode as described by Škarabot et al. (2019a).

Potential activation of ventral roots was examined from the onset latency of the LEP of an increasing stimulator intensity (Petersen et al. 2002) up to 25% of the M-max and also tracking LEP amplitude during increasing voluntary contraction while maintaining stimulator output intensity to that which produced a LEP amplitude of 25% of the M-max (Taylor et al. 2002). Should the ventral roots be activated by the stimulation procedures, onset latency would have shortened with an increase in stimulator intensity and LEP amplitude would have been the same during increased voluntary contraction (Petersen et al. 2002; Taylor et al. 2002). Three participants demonstrated no change in LEP amplitude with an increase in voluntary torque during offline analyses, and they were, therefore, removed from further analyses.

Dorsal root activation was assessed via paired LS with a 50 ms time delay, where the second LEP amplitude was compared to the first. Paired stimulation was conducted at rest, with the stimulator output intensity set to produce a LEP equivalent to 25% of the M-max. Evidence of dorsal-root activation would manifest as a decrease in the second LEP compared to the first, attributed to post-activation depression at the motor-neuron pool between the two stimuli (Hofstoetter et al. 2018). If the participant failed any of the tests (i.e., dorsal or ventral stimulation protocols), the electrodes were relocated 1 cm higher, until the participant passed all tests, or the anode was place between the third and fourth thoracic vertebrae. To ensure the placement was the same in all sessions, the distance from the 7th cervical vertebra to the anode (21.7 ± 4.1 cm) and from the bottom of the anode to the top of the cathode (3.7 ± 1.1 cm) (i.e., inter-electrode distance) were taken. All remaining participants showed no sign of the responses described and reported that they found LS to be tolerable. Once the placement was confirmed, stimulator intensity was kept to that which produced a LEP of 25% of the M-max at rest, and this stimulation intensity was used throughout the session (Table 1, Fig. 2C). The reliability of this method is reported in Gomez-Guerrero et al. (2023b) and considered moderate-to-good (ICC: 20% of MVC = 0.632; 60% of MVC = 0.520).

Single TMS pulses were delivered using a MagStim 200² magnetic stimulator (MagStim Co., Ltd., Whitland, UK) connected to a concave double-cone coil positioned over the left cortical hemisphere for RF with a posterior-to-anterior current orientation. The hotspot was defined at rest as the position eliciting the largest visible MEP recorded by EMG using the same intensity (approx. 50–70% stimulator output). Once the hotspot was found, the coil position was marked with a permanent marker on the scalp to maintain the same position throughout the protocol. Active motor threshold (aMT) was determined by increasing stimulator intensity in 5% steps, starting at 30% of the stimulator output. Thereafter, stimulator intensity was decreased in steps of 1% until clear MEPs (> 100 μV) were elicited in three out of five stimulations during unilateral isometric contractions of the right limb at 10% of MVC. Sets of ten single TMS stimulations were delivered in a random order for each of the assigned conditions (i.e., 120, 140 and 160% aMT) during unilateral isometric contractions at 20% and 60% of MVC (Fig. 2D). The reliability of these methods is reported in Gomez-Guerrero et al. (2023b) and considered good-to-excellent (ICC: 20% of MVC = 0.821–0.861; 60% of MVC = 0.901–0.941).

All participants performed a bilateral concentric knee extension (David 200, David Health Solutions Ltd, Helsinki, Finland) one-repetition maximum (1-RM) test during the 5 test periods (Fig. 2A). First, each participant went through anthropometric analysis (Inbody 770, Inbody Co. Ltd, Seoul, Korea). Then, a 5 min cycling (1 kg load at 70 rpm) warm-up was performed followed by a series of submaximal warm-up sets (6 repetitions at an estimated 10-RM load, 3 repetitions at an estimated 6-RM load, 1-repetition at an estimated 3-RM load). Thereafter, single repetitions were performed until the participant could no longer lift the load from the beginning knee angle of ~ 85° to the required knee angle (≥ 170° knee angle), by visual inspection. The last successfully lifted load was recorded as the participant’s 1-RM and used to prescribe the load for the first and 4th week of training. Four-to-eight attempts where needed to calculate 1-RM with 1.25 kg precision. Verbal encouragement was provided to motivate participants to produce a maximal effort. 3 minutes rest were provided between attempts. The reliability of this method was excellent (CV = 8.4%; ICC = 0.991).

Over the course of the 7 weeks of strength training, participants engaged in a total of 13 supervised sessions of conventional strength training. Mid-training testing was conducted after seven training sessions. Training sessions were conducted twice-a-week, with at least a 48-h break between sessions. The strength-training program was created following the guidelines provided by Fragala et al. (2019). The training program may be considered whole-body, targeting both upper- and lower-limbs, although we acknowledge that there were no dedicated abdominal or lower back exercises. Nevertheless, one or two exercises per muscle group were performed with a total volume of eight sets per muscle group for the lower-limbs and back/biceps and three sets for chest/triceps (Fragala et al. 2019). Each training session consisted of five different exercises for the upper- and lower-limbs: leg press, knee extension, bicep curl, smith-machine bench press and chest-supported seated row, in that order during normal training sessions. During testing sessions (Pre, Mid, Post), the order was: knee extension, leg press, smith-machine bench press, bicep curl. This training program closely resembles the most potent program for older adults identified in a meta-analysis (Borde et al. 2015). During the last set of the last session of the week, participants performed the maximum number of repetitions for each exercise to adjust either the volume or intensity (according to the estimated %RM) for the following week, so they could perform at least 8 repetitions.

All training sessions started with a warm-up, which consisted of 5 min of cycling and dynamic mobility exercises. During the initial training session, knee extension 1-RM testing was conducted. Subsequently, a 3–5 RM test was performed for the remaining exercise to determine and prescribe the training load. The rest of the sessions consisted of five (knee extension and bicep curl) and three sets (leg press, smith-machine bench press and chest-supported seated row) of 8–10 repetitions at 75–80% of 1-RM. The participants were asked to perform a 2 s-controlled eccentric phase, with no isometric phase and fast concentric phase.

A 4-week detraining period followed the strength-training period. Participants were allowed to maintain their normal physical activity (i.e., cycling, walking, running) during the whole intervention, but strength training was terminated during the detraining period.

Offline analyses were performed with Spike2 software (version 6.10, Cambridge Electronic Design, Cambridge, UK) to manually obtain M-max amplitude and MVC torque. The other outcome measures were analyzed by a customized MATLAB script (version R2020b, The MathWorks, Inc., Natick, USA). Peak-to-peak amplitude of LEPs and MEPs were analyzed automatically between latencies-of-interest following LS or TMS, respectively. SP duration was defined, as the time from the stimulator artifact to the return of voluntary EMG (Damron et al. 2008). Torque was averaged over the 100 ms before the stimulator artifact (Škarabot et al. 2019b). LEP and MEP amplitude is represented as relative to M-max.

SPSS software (version 26.0, SPSS Inc., Chicago, USA) was used for all statistical methods. Means and standard deviation (SD) were calculated and reported throughout. Normality of the data was tested with the Shapiro–Wilk test and confirmed by a z-score with an acceptance of + 2 to − 2 (e.g., skewness score/skewness score_SE and kurtosis score/kurtosis score_SE) and Q-plots for visualization. Data that did not fulfill those requirements were Log₁₀ transformed, which then fulfilled the requirements for normality. A two-way repeated measures ANOVA (5 Time × 2 Group) was employed to assess most outcome variables (MVC, 1-RM, skeletal muscle mass, M-max, aMT, and silent periods of LEPs at 25% of the M-max and MEPs at 120, 140, 160% aMT) during contractions at 20 and 60% of MVC. When assumptions of sphericity were violated, Greenhouse–Geisser corrections were used. Post-hoc Bonferroni adjustments were used when significant main effects were found. To investigate the influence of strength training on the TMS- and LS-induced MEP/LEP amplitude, and to accommodate for missing data points and baseline variability, we employed a Linear Mixed Model (LMM) (Wilkinson et al. 2023). This model served as a robust framework for analyzing our data considering both fixed and random effects simultaneously. Cortico-spinal (MEPs at 120, 140, 160% aMT) and spinal (LEPs at 25% of the M-max) excitability at 20 and 60% of MVC were assessed using the LMM. The model included time (Con, Pre, Mid, Post, and De) and age group (young and older) as main effects and an interaction between age group (young and older) and time with participants as the random effect within the model. Bonferroni adjustments were used when significant main effects were found. Reliability, based on ICCs was categorized as poor (ICC < 0.5), moderate (ICC: > 0.5–< 0.75), good (ICC: > 0.75–< 0.9) and excellent (ICC: > 0.9) (Koo and Li 2016).Data are presented in the Tables by mean and SD, and in the results section by mean difference (MD), effect sizes are represented as partial eta-squared values (η_p² = small: 0.01, medium: 0.06, large: 0.14) for the factors of the ANOVA and post-hoc effect sizes reported as Hedge’s g (g = small: < 0.3, medium: 0.3–0.8, large: > 0.8). Αlpha was set at 0.05.

Main effects for Group were observed for 1-RM (F _(1,19) = 15.94, p = 0.001, η_p² = 0.46), MVC (F _(1,19) = 9.60, p = 0.006, η_p² = 0.34), M-max (F _(1,19) = 20.86, p < 0.001, η_p² = 0.53), aMT (F _(1,19) = 11.75, p = 0.038, η_p² = 0.21), MEP amplitude during 60% of MVC with 120% aMT (F _(1,19) = 4.65 p = 0.044) and 140% aMT (F _(1,19) = 4.62 p = 0.045), MEP silent period during 20% of MVC with 120% aMT (F _(1,19) = 13.96, p = 0.001, η_p² = 0.42), LEP silent period duration during 20% of MVC (F _(1,19) = 5.60, p = 0.029, η_p² = 0.229), MEP silent period during 60% of MVC with 120% aMT (F _(1,19) = 23.39, p < 0.001, η_p² = 0.650), and LEP silent period duration during 60% of MVC (F _(1,19) = 23.39, p < 0.001, η_p² = 0.552).

During the first measurement session (i.e., control), young adults were stronger than older adults, and had a higher M-max and lower aMT (Table 1 and Fig. 3). Further, during control, MEP amplitude at 120% and 140% aMT was greater in the older group during 60% of MVC (Fig. 4D and E). Silent period duration was longer in older adults during both 20% of MVC (99 ± 15 ms versus 117 ± 18 ms, p = 0.027) and 60% of MVC (94 ± 12 ms versus 121 ± 20 ms, p = 0.001) when stimulated at 120% aMT, and during 60% of MVC from LS (62 ± 7 ms versus 82 ± 20 ms, p = 0.006) (see supplementary material).

For 1-RM, main effects for Time (F _(2.3,42.9) = 28.29, p < 0.001, η_p² = 0.60) and Time*Group interaction (F _(2.3,42.9) = 11.06, p < 0.001, η_p² = 0.38) were observed. Post-hoc comparisons showed that young adults increased from Pre to Post (p < 0.001) and then decreased from Post to De (p = 0.011, Fig. 3A). Older adults did not increase statistically Pre to Post but did Mid to Post (p = 0.027) and they also decreased Post to De (p = 0.012).

MVC demonstrated a significant main effect for Time (F _(4,76) = 10.13, p < 0.001, η_p² = 0.35). Post-hoc analysis showed that young adults increased significantly Mid to Post (p = 0.024) and older increased significantly Pre to Post (p = 0.014, Fig. 3B).

Skeletal muscle mass demonstrated a significant main effect for Time (F _(2.5,47.8) = 3.16, p < 0.001, η_p² = 0.323). Here, only young adults increased Pre to Post (p = 0.009) and then decreased Post to De (p < 0.001, Fig. 3C).

Significant main effects for Time and Time*Group interaction were observed for MEP amplitude during 20% of MVC at 120% aMT (Time: F _(4,1021) = 3.09, p = 0.015; Time*Group: F _(4,1021) = 4.10, p = 0.003), 140% aMT (Time: F _(4,1021) = 4.89, p = 0.001; Time*Group: F _(4,1021) = 14.44, p < 0.001), 160% aMT (Time: F _(4,1021) = 8.12, p < 0.001; Time*Group: F _(4,1021) = 4.10, p = 0.003). In the young adults, significant increases occurred Pre to Post with 140% aMT (p = 0.023) and Pre to Mid at 160% aMT (p = 0.005). In older adults, significant decreases were observed Pre to Post at 140% (p < 0.001) and 160% (p < 0.001) aMT (Figure, Fig. 4B and C).

Significant main effects for Time and Time*Group interaction were observed for MEP amplitude during 60% of MVC at 120% aMT (Time: F _(4,1021) = 4.24, p = 0.002; Time*Group: F _(4,1021) = 10.53, p < 0.001), 140% aMT (Time: F _(4,1021) = 7.97, p < 0.001; Time*Group: F _(4,1021) = 13.69, p < 0.001), 160% aMT (Time: F _(4,1021) = 13.50, p = 0.002; Time*Group: F _(4,1021) = 14.08, p < 0.001). Post-hoc comparisons showed that only older adults decreased Pre to Post with all stimulation intensities (p < 0.001, Fig. 4D–F).

Significant main effects for Time (F _(4,1021) = 3.09, p = 0.015) and Time*Group interaction (F _(4,1021) = 4.10, p = 0.003) were observed for LEP amplitude during 20% of MVC. Young adults significantly increased Pre to Post (p < 0.001) and subsequently decreased Post to De (p < 0.001). Also, in the young adults, there was a significant decrease from Con to Pre (p = 0.022). In older adults, a significant decrease occurred Pre to Post (p < 0.001) (Table 2).

Significant main effects for Time (F _{(4, 1021)} = 8.45, p < 0.001) and Time*Group interaction (F _{(4, 1021)} = 6.66, p < 0.001) were LEP amplitude during 60% of MVC. Post hoc showed that young significantly decreased Con to Pre (p < 0.001), further decreased Pre to Mid (p = 0.023), and then increased Mid to Post (p < 0.001) (Table 2).

An interesting observation was the consistent decrease in MEP excitability in the older group, independent of the contraction intensity. These changes became apparent as early as three weeks into the training. Our results differ from those reported by Christie and Kamen (2014) who reported that two weeks of training (six training sessions) did not induce significant changes in MEP amplitude in the m.tibialis anterior. The authors noted decreases of 4–6% (n.s) in MEP amplitude in the older adults. The magnitude of those results was similar to our results (− 7 to 8%) after 3 weeks/6 sessions of strength training but ours further decreased (to – 12 to 21%) after 7 weeks/ 13 sessions of strength training. Therefore, cortico-spinal adaptation in older adults seems to require more training duration than in young adults.

Furthermore, the interaction, and within-group changes of LEP amplitude parallel those of MEP amplitude; older adults showing a reduction in LEP amplitude at 20% of MVC. In addition, LEP amplitude increased in the young group from pre- to post-training at 20% of MVC and then decreased back to baseline after detraining. No clear or systematic changes were observed in either group during 60% of MVC trials, and the observed fluctuations may be due to the relatively high typical error/reliability values of this method (Gomez-Guerrero et al. 2023b). Nevertheless, one previous study investigating short-term strength training effects (Ansdell et al. 2020) observed no changes in MEP nor LEP amplitude at a group level; where large inter-individual differences apparent with approximately half of the group increasing and half decreasing amplitude after 12 sessions of 4 sets of 6–8 back squat repetitions. In contrast, Lundbye-Jensen et al. (2005) demonstrated decreased cortico-spinal excitability in untrained healthy young adults after thirteen training sessions spread over 4 weeks. This effect was observed at several higher TMS stimulator output intensities (160–220% rMT), similar to our differences observed at 140 and 160% aMT. The authors discussed that those changes could potentially be at subcortical levels through changes in spinal motor-neuron firing rate and/or intrinsic firing properties, although this was not specifically tested. In support, Vila-Chã et al. (2012) and Aagaard et al. (2002) observed spinal adaptations, through better modulation of inhibitory pathways, after 3 weeks and 14 weeks of strength training in younger adults. Thus, in the present study, the older group adapted to the training by reducing their MEP amplitude down to the level of the young and these adaptations could be at a spinal level.

Conversely, small magnitude but statistically significant increases in MEP excitability occurred in the young group after strength training, as has been previously reported (Goodwill et al. 2012; Kidgell et al. 2017; Weier et al. 2012). Goodwill et al. (2012) and Weier et al. (2012) found that a short-term training intervention, twelve sessions, produced an increase in MEP amplitude of RF when measured at 10% of MVC. Those results are in line with our results at 20% of MVC. However, and importantly for our interpretation, MEP excitability assessed at 60% of MVC did not show significant changes in the young. Strength training and maximal strength has been proposed as a specific skill (Buckner et al. 2017), and 12 sessions of arm flexion–extension visuomotor tracking skill training (Lundbye-Jensen et al. 2005) along with 12 sessions of 3 s concentric and 4 s eccentric tempo-controlled bicep curl strength training (Leung et al. 2017) has been shown to increase MEP amplitude after four weeks. Since the participants were required to hold the force level constant prior to stimulation (~ 2 s), it may be that lower force levels challenge the sensorimotor system to a greater extent than higher contraction levels, as previously evident in force steadiness tasks (Laidlaw et al. 2000). Therefore, we propose that the statistically significant but small magnitude changes in excitability in the young observed only during 20% of MVC trials reflect the sensorimotor integration needed for force steadiness, a so-called ‘skill element’ of strength training.

Our results showed higher aMT in older adults compared to younger adults, which is an indicator of cortico-spinal excitability (Pascual-Leone et al. 1995; Wassermann 2002). Should this reflect a decline in cortico-spinal excitability with age, as interpreted in previous studies (Bashir et al. 2014; Cirillo et al. 2011), this would directly conflict the MEP amplitude data of the present study. The aging process may lead to reduced activation of cortico-spinal neurons or disrupted synchronization among these neurons leading to a cancelation phase (Pitcher et al. 2003; Magistris et al. 1998). Notably, despite the impact of strength training and subsequent detraining on MEP and LEP amplitudes, aMT remained unchanged across interventions and age groups suggesting a discrepancy between the measures as an indicator of excitability. Previous studies have discussed (Wassermann 2002; Hassanlouei et al. 2017) that caution is advised in interpreting aMT due to factors such as a reduction in motor cortex size (Marner et al. 2003; Salat et al. 2004) and increase in skull thickness (Lillie et al. 2016) with age that potentially increases the coil-to-cortex distance, meaning a requirement for higher intensities for action potential generation. It may be that the between-group differences in aMT of the present study is due to cortex size or skull thickness rather than cortico-spinal excitability per se. While our study did not directly address these factors, our results underscore the need for further investigation to identify the precise mechanisms.

In addition, at higher contraction intensities in the present study, the older group showed greater MEP amplitude than the younger group at baseline. Further, Hassanlouei et al. (2017) showed that individuals engaged in higher physical activity (> 10,000 steps/day) demonstrated lower MEP amplitude in m.vastus lateralis than the ones with low physical activity (< 10,000 steps/day), independent of age. Moreover, cast immobilization has been shown to increase cortico-spinal excitability, when measured at 120% rMT (Roberts et al. 2007). Both studies discuss that modulation of different inhibitory pathways at the cortical level could modify cortico-spinal excitability due to the lack of exercise. These data suggest that better trained muscles for gross force production, remembering that older adults are generally less physically active than young (Martin et al. 2014), are characterized by lower cortico-spinal excitability responses to TMS.

Our results showed that neither strength training nor detraining affected MEP or LEP cSP duration. This is somewhat unexpected as meta-analyses have shown reductions in cSP duration following strength training (Kidgell et al. 2017; Mason et al. 2019), at least in young adults. Nevertheless, within these meta-analyses there have been studies showing no changes in cSP, thus, our data is not without precedent. For example, 12 strength training sessions of 4 sets of 6–8 repetitions with 80% 1-RM using 3 s concentric and 4 s eccentric tempo-controlled contractions led to no changes in biceps brachii cSP in healthy young adults (Kidgell et al. 2011).

At baseline, our results showed that MEP SP at 120% aMT and LEP SP were significantly longer for the older group independently of the contraction intensity used. cSP is an indication of intracortical inhibition (Inghilleri et al. 1993) mediated by Gamma-aminobutyric acid (GABA) inhibitors, particularly involving the activity of GABA_B receptors (Siebner et al. 1998). Consequently, prolonged cSP indicates greater GABA_B activity and longer intracortical inhibition in the older group. These results contradict previous findings, where SP durations were reported shorter (Christie and Kamen 2014; Sale and Semmler 2005) or not different (Fujiyama et al. 2012) comparing younger and older adults at baseline. However, it should be noted that MEPs were either of similar amplitude (Christie and Kamen 2014) or smaller (Sale and Semmler 2005) than the younger adults in those previous studies, which contrasts the higher MEP and LEP amplitudes for the older adults here. Given the correlation between cSP and MEP amplitude (Orth and Rothwell 2004), it is plausible that normalization of cSP to MEP amplitude in the older group might have led to an interpretation of increased inhibition in older adults, due to the decreased MEP size and no changes in cSP in the older adults.

Moreover, the present study showed decreased MEP amplitude following strength training while the SP duration from cortical and spinal stimulation remained unchanged. Therefore, normalizing the SP to MEP or LEP amplitude, would modify the interpretation of excitatory and inhibitory processes influencing the observed outcomes. Thus, the observed decrease in MEP/LEP amplitude and the conserved SP may indicate greater contribution of cortical and/or spinal inhibition in older adults after training, which may improve movement efficiency and result in increased strength.

This study is the first to provide evidence of cortical and spinal excitability and inhibition adaptations to a 7-week strength training intervention in young and older adults. In addition, it also provides information from a detraining period, which strengthens inferences that can be drawn from the causality of the intervention. Furthermore, cortico-spinal responses were recorded during different contraction intensities. Clearer between-group differences (at baseline) were observable at 60% of MVC compared to 20% of MVC, and this finding could direct future studies comparing differences between groups. In addition, the detraining period provides support that the intervention caused the observed alterations in the outcome measures and helps to identify the mechanisms of improved strength. The young increased and decreased both strength and muscle mass concomitantly, suggesting that morphologic adaptations were a large factor in the strength increase. Conversely, the older adults maintained both strength and the altered MEP/LEP amplitude after detraining suggesting that neural adaptations predominantly underpinned the strength gain.

As a limitation, the strength-training program was performed dynamically and mainly bilaterally. Thereby, the unilateral isometric test was non-specific and could have influenced the ability to identify neural adaptations. TMS paired-pulse paradigms (i.e., SICI, LICI, ICF), peripheral stimulation paradigms (H-reflex) and/or paired H-reflex -TMS (cortical recurrent inhibition) were not measured in this study because an increased number of contractions per session would have increased the risk of fatigue. This could have provided more specific information about how strength training modulates cortical and spinal inhibitory process in young and older adults alongside cortico-spinal and spinal excitability.