The effects of isometric resistance training on stretch reflex induced ...

J Appl Physiol 114: 1647–1656, 2013. First published April 11, 2013; doi:10.1152/japplphysiol.00917.2011.

The effects of isometric resistance training on stretch reflex induced tremor in the knee extensor muscles Rade Durbaba,1 Angela Cassidy,2 Francesco Budini,3,4 and Andrea Macaluso3 1

Department of Applied Sciences, Faculty of Health and Life Sciences, Northumbria University, Newcastle, United Kingdom; Strathclyde Institute of Pharmacy and Biomedical Sciences, University of Strathclyde, Glasgow, United Kingdom; 3 Department of Human Movement and Sport Sciences, University of Rome Foro Italico, Rome, Italy; 4Institute for Sport and Health, University College Dublin, Dublin, Ireland 2

Submitted 21 July 2011; accepted in final form 4 April 2013

Durbaba R, Cassidy A, Budini F, Macaluso A. The effects of isometric resistance training on stretch reflex induced tremor in the knee extensor muscles. J Appl Physiol 114: 1647–1656, 2013. First published April 11, 2013; doi:10.1152/japplphysiol.00917.2011.— This study examines the effect of 4 wk of high-intensity isometric resistance training on induced tremor in knee extensor muscles. Fourteen healthy volunteers were assigned to either the training group (n ⫽ 7) or the nontraining control group (n ⫽ 7). Induced tremor was assessed by measuring force fluctuations during anisometric contractions against spring loading, whose compliance was varied to allow for preferential activation of the short or long latency stretch reflex components. Effects of high-intensity isometric resistance training on induced tremor was assessed under two contraction conditions: relative force matching, where the relative level of activity was equal for both pre- and post-training sessions, set at 30% maximum voluntary contraction (MVC), and absolute force matching, where the level of activity was set to 30% pretrained MVC. The training group experienced a 26.5% increase in MVC in contrast to the 0.8% for the control group. For relative force-matching contractions, induced tremor amplitude and frequency did not change in either the training or control group. During absolute force-matching contractions, induced tremor amplitude was decreased by 37.5% and 31.6% for the short and long components, respectively, with no accompanying change in frequency, for the training group. No change in either measure was observed in the control group for absolute force-matching contractions. The results are consistent with high-intensity isometric resistance training induced neural changes leading to increased strength, coupled with realignment of stretch reflex automatic gain compensation to the new maximal force output. Also, previous reported reductions in anisometric tremor following strength training may partly be due to changed stretch reflex behavior. gain realignment; tremor; steadiness; force fluctuations; motor control

force during contractions of the lower limb is important for mobility and steadiness and is known to be impaired in older adults (2, 7, 23, 36, 41, 53, 60, 70, 84, 85, 86). As this impairment is seen for both isometric and anisometric contractions, it is suggested that neural mechanisms such as decreased number of motor units (56, 57) and increased variability of discharge (46) could be responsible. High-intensity resistance training, either isometric or anisometric, is known to restore some of this loss in function (41, 53, 67, 77, 84, 86), as well as leading to an increase in maximal force production. An interesting feature of note was that improvements in steadiness were specific to submaximal anisometric

THE ABILITY TO MAINTAIN STEADY

Address for reprint requests and other correspondence: R. Durbaba, Department of Applied Sciences, Faculty of Health and Life Sciences, Ellison Building, Northumbria University, Newcastle upon Tyne, UK, NE1 8ST (e-mail: [email protected]). http://www.jappl.org

contractions (41, 53, 84, 86). This implies that high-intensity resistance training may be influencing neural circuitry that is associated more with steadiness in anisometric contractions than isometric and thus could implicate some change in behavior to neural circuitry involved in monitoring length, namely the muscle spindle and the stretch reflex pathway. Various studies have shown that self-maintaining force fluctuations, hereafter referred to as induced tremor, occur when contracting anisometrically under appropriate compliant loading, that are not present during isometric contractions. This induced tremor during anisometric contractions has been attributed to the stretch reflex pathway (21, 24, 43, 50, 55, 78, 79, 88). Indeed, Durbaba et al. (21) showed that in the digastric muscle, which is naturally devoid of muscle spindles and hence lacks a stretch reflex, it was not possible to induce tremor. In addition, other studies have indicated a role for the stretch reflex in contributing to “physiological tremor” (9, 24, 31, 34, 40, 72, 80, 81). The question then arises what the link is between the stretch reflex and reductions in force fluctuations following resistance training. As high-intensity resistance training leads to an increase in maximal force production, it is possible that the reductions are linked to the “relative” levels of activity pre- and posttraining and the behavior of the stretch reflex pathway to these different levels of activity. It is known from experimental and modeling studies that the response of the stretch reflex pathway, measured either as reflex torque, electromyogram (EMG) amplitude, or reflex gain, shows a bell-shaped relationship to the background level of activity, with an increase in gain up to 30 – 40% maximum voluntary contraction (MVC), a plateau region between 40 and 50% MVC, followed by a decline for high contraction levels (5, 8, 11, 35, 54, 61, 83). This is known as automatic gain compensation. A similar behavior has been observed for the amplitude of induced tremor during anisometric contractions at the elbow against a compliant load (43, 55). Thus if, following high-intensity training, automatic gain compensation is “realigned” to the new maximal force production, then the response of the stretch reflex pathway at any “relative” level of activity should be the same pre- or post-training. This would imply that reductions in force fluctuations during anisometric contractions seen in the lower limb following training, could be due to a lower response of the stretch reflex pathway associated with a lower “relative” level of activity post-training. The purpose of this study was to examine the effects of 4 wk high-intensity isometric resistance training using a protocol that is known to produce an increase in maximal force production (52) on induced tremor in knee extensors during aniso-

8750-7587/13 Copyright © 2013 the American Physiological Society

1647

1648

Resistance Training and Stretch Reflex Induced Tremor

metric contractions against a compliant load. We hypothesized, for submaximal (ⱕ30% MVC) anisometric contractions, that 1) during a relative force-matching contraction task, where the relative level of activity pre- and post-training are the same (30% MVC), following training there will be no change in induced tremor; and 2) during an absolute force-matching contraction task, where the relative level of activity posttraining will be less than pretraining, training would result in a decrease in the amplitude of induced tremor. We would attribute the lack of change and the reduction in induced tremor for relative and absolute force-matching contraction tasks, respectively, to be due to a realignment of automatic gain compensation for the stretch reflex pathway to the new maximal force production post-training. Preliminary results have been published in abstract form (20). METHODS

Participants. With approval of the Ethics Committee of the University of Strathclyde, 14 individuals [aged 28.5 ⫾ 6.8 yr (mean ⫾ SD); body mass 74.6 ⫾ 19.1 kg; stature 1.76 ⫾ 0.11 m], 9 males and 5 females, were matched for maximal isometric force of their quadriceps muscle and randomly assigned to either training or control group. Volunteers provided written informed consent and were instructed to maintain their usual levels of physical activity throughout the duration of the study. After completing a familiarization session on a separate day, all participants were tested before the onset of training (week 0) and then the assessment was repeated after training (week 4). Recording apparatus. Knee extension torque of the dominant leg was recorded using a dynamometer (Biodex System 3, Biodex Medical System, Shirley, NY). The participants were seated comfortably in the dynamometer chair, with their trunk erect and fastened by two crossing belts, with the angle at the hip and knee being set to 90°. The torque motor of the Biodex was placed behind the seated subject. The lever arm of the torque motor was locked in a vertical position, pointing toward the ground and was linked to the subject to allow for either isometric or anisometric contractions. The linkage was attached to the subject via a noncompliant cuff placed around the lower leg, with the cuff’s bottom edge about 1 cm above the malleolus. With this arrangement the only rotation that could occur was at the knee of the subject. The sensitivity of the torque transducer was 139 N·m·V⫺1, and since the length of the lever arm was set to 0.5 m for all subjects, the force sensitivity was 278 N/V. The force output from the Biodex was accessed and two signals extracted. The first was the DC force output, which provided a feedback to the subjects on a computer screen to assist them in maintaining the requested levels of contraction. The second signal was an AC coupled version of the output, to allow for the amplification of irregularities in force during the steady state period of the contraction. Both signals were low-pass filtered at 100 Hz, while the AC-coupled version of the force was also high-pass filtered at 0.5 Hz. The two force signals were A/D converted, using the Cambridge Electronic Design (CED) Power1401 system, and captured at a sampling rate of 2000 Hz using the CED Spike2 package. Data files were stored on computer for off-line analysis. Force output from an individual was assessed isometrically and anisometrically. Isometric recordings were achieved by contracting against the dynamometer lever arm through an inextensible rod. To achieve anisometric recordings, a spring of appropriate stiffness was inserted between the cuff and the dynamometer lever arm, with the inextensible rod being shortened to maintain a 90° knee angle. Under anisometric conditions, the frequency at which oscillation might be expected to occur is very much dependent upon the spring stiffness and the total inertia of the moving parts. The resonant frequency (fr) of a spring-mass system rotating about a pivot point can be determined by the equation

•

Durbaba R et al.

f r ⫽ 共1 ⁄ 2␲兲 ⫻ 共k ⫻ d2 ⁄ IT兲0.5

(1)

where k and IT are the spring stiffness and total inertia of the system, respectively, and d is the distance from the pivot to the point of attachment of the spring (⫽ 38 cm; knee joint to middle of cuff). Two springs of different stiffness were used, 5.35 N/mm and 11.06 N/mm. The total inertia can be expressed as IT ⫽ IL ⫹ ISp

(2)

where IL and ISp are the inertial components for the lower limb below the knee and spring, respectively. The simplest model for the leg is to assume that it is made up of two cylinders representing the shank and the foot. Thus IL ⫽ IShank ⫹ IFoot

(3)

where IShank is the inertia of the shank about the knee joint, and IFoot the inertia of the foot about the ankle joint but also taking into account that the foot rotates about the knee joint. The equations for IShank and IFoot are IShank ⫽ 共MS ⁄ 12兲 ⫻ 共3SR2 ⫹ 4SL2兲

IFoot ⫽ 关共MF ⁄ 12兲 ⫻ 共

FL2

⫹

3FR2

(4)

兲兴 ⫹ MF ⫻ 关共FL ⁄ 2兲 ⫺ SR兴 ⫹ M F ⫻ 共 S L ⫹ F R兲 2 2

(5)

where MS, MF, SL, FL, SR, and FR are the masses, lengths, and radii of the shank and the foot, respectively. With the exception of SR, estimates of mean values for these measures could be computed based on simple scaling factors related to whole body mass or length (49, 87). For whole body mass or length, we used the mean values for the individuals recruited for this study. An estimate of the mean shank radius was made from measures taken from cadavers. Using these mean values, we computed mean inertial values for the shank and foot of 0.22 and 0.28 kg·m2, respectively, giving the total mean inertial component due to the leg below the knee of 0.50 kg·m2. The inertia due to the spring was modeled as a thick-walled hollow cylinder and took into account how its point of attachment relates to the rotation about the knee joint. The equation for ISp is ISp ⫽ 共MSp ⁄ 12兲 ⫻ 共3共R2o ⫹ Ri2兲 ⫹ 4L2兲 ⫹ 共MSp ⫻ d2兲

(6)

where MSp, Ro, Ri, and L are the mass, outer and inner radii, and length of the spring, values for which were obtained by weighing and measuring the springs. The inertial components due to the 5.35 N/mm and 11.06 N/mm springs were calculated to be 0.08 and 0.04 kg·m2, respectively. Thus, by combining appropriate inertial components and spring stiffnesses together, it was computed that the resonant frequency of the spring-mass system associated with the 5.35 N/mm and 11.06 N/mm springs would be 5.85 and 8.72 Hz, respectively. In addition to the spring-mass element, it is necessary to consider the contribution of the whole oscillating system in determining the final expected oscillation frequencies. To do this, the model developed by Durbaba et al. (21) was modified to represent the behavior about the knee. The above-computed frequencies for the spring mass system were incorporated into the model, along with a second-order critical damped element relating EMG to force output from the muscle and muscle spindle afferent feedback. Only Ia afferent feedback is modeled, as this is generally accepted as being the principal sensory feedback related to induced tremor (24, 31, 43, 50, 55). A feature of the model is the neural loop delay related to the stretch reflex. As we are only considering Ia afferent feedback, then only neural delays related to the short and long latency components need be considered, as the medium latency component is thought to be predominantly due to secondary (group II) muscle spindle afferent input (74, 75). Ideally, the model should include the short and long components together, in terms of synaptic gain and their relative interaction at the level of the alpha motoneurone. However, these are not known. Therefore the two components were modeled separately as though they “dominated” the reflex contribution to the induced tremor for each of the springs used.

J Appl Physiol • doi:10.1152/japplphysiol.00917.2011 • www.jappl.org


The mean latency for the short component was taken as 25 ms and that for the long component as 78 ms (62). Thus for the whole oscillating system, the expected frequencies of oscillation were 5.6 and 5.0 Hz for the short and long components when using the 5.35 N/mm spring and 8.4 and 5.3 Hz for the 11.06 N/mm spring. Experimental protocol. A recording session consisted of two parts. At the start of the session the strength of the knee extensors of the dominant leg of an individual was assessed by asking them to voluntarily contract to their maximum effort (maximum voluntary contraction; MVC) and hold there for 3 s, before returning to the rest position. During the MVC, participants followed their performance on a computer screen and were verbally encouraged to achieve a maximum, in an attempt to exceed the previous force recording. MVC was calculated as the largest 1 s average reached within any single force recording. Three attempts were performed, separated by 5 min, and the greatest of the three attempts was chosen as MVC. After the last MVC participants rested for 20 min prior to making the tremor recordings. During this part of the recording session volunteers were asked to contract anisometrically, against a compliant spring, to a particular level of force and to hold at this level for 20 s before returning to rest. Both the target force and the actual force were displayed on a computer screen to provide feedback to the subjects. Individuals repeated the contractions four times with a 2 min resting period occurring between contractions to minimize the effects of fatigue. Thus two sets of anisometric contraction were recorded, one for each spring, in a random order, with a 5 min rest period between sets. For the pretraining session (week 0) the level of force required during anisometric contractions was 30% of their MVC. In the post-trained recording session (week 4), in addition to contracting to their new 30% MVC (referred to as “relative force matching”) subjects also performed contractions at a force level equivalent to 30% of their “pretraining” MVC (old 30% MVC force, referred to as “absolute force matching”). Strength training. The training protocol used in this study was identical to that used by Macaluso et al. (52). The protocol was chosen as it has been observed to be effective in increasing MVC force output by 20% following 4 wk of training. Participants took part in a 4-wk period of isometric resistance training involving the knee extensor muscles of the dominant leg, three times a week, for about 45 min per session. Each training session consisted of a total of 15 sets, 10 repetitions each (5 s contraction, 3 s rest). Each subject started with 2 sets at 20% MVC, followed by 2 sets at 40% MVC, 4 sets at 60% MVC, 3 sets at 80% MVC (only 8 repetitions each), 2 sets at 40% MVC and finally 2 sets at 20% MVC. The interval time between each set was 15 s at 20% MVC, 30 s at 40% MVC, 2 min at 60% MVC, and 3 min at 80% MVC. The resting period at the end of each set was 15 s after 20% MVC, 1 min after 40% MVC and 3 min after 60% MVC and 80% MVC. In the subsequent training weeks, the load for each set was increased by 5% each week, to maintain a high training intensity. Data analysis. All force signals were converted off-line into Newtons by multiplying the voltage by the force sensitivity (see above). For anisometric contractions, the A/C coupled force trace was further dividing by the appropriate stiffness to obtain displacement in millimeters. The steady state period of each of the four 20 s contractions of a set were combined and the autospectra of the displacement signal were constructed by fast Fourier transformation (FFT) using in-built functions of the CED Spike2 analysis package. Each spectra ranged from DC to 15 Hz, with a resolution of 0.488 Hz, which for simplicity is considered to be 0.5 Hz. Frequencies above 15 Hz were not required as they had no relevance to this study. Also, the high-pass filtering renders the DC to 1 Hz components meaningless so these were not plotted. Spectra were plotted on linear amplitude scales. The computed spectra were then inspected to determine the amplitude and frequency of any principal rhythmic components that might have been present, which reveal themselves as a peak about a frequency of oscillation. If a peak was present, then its amplitude was expressed as

•

Durbaba R et al.

1649

the root mean squared amplitude (ARMS), where ARMS is √(PT/2) and PT is the total power within the peak, with the peak being defined as the bin where maximum power occurred ⫾3 bins. This approach was used to take into account that with the application of a windowing function, such as a Hanning window, to counteract the sharp start/end tapering effects of data sections used for the FFT analysis there is a loss in power of the signal at the expected frequency. Also, this is accompanied by a spread of the signal into one or more adjacent bins around the expected. Thus one would underestimate the amplitude by simply taking the value at the maximum power of the peak. Frequency of oscillation (fo) of the induced tremor was taken to be the frequency of the bin that corresponded to the maximum power of the peak. Statistical analysis. All data were normally distributed in terms of kurtosis and skewness (all values ⬍2). Statistical comparisons of the parameters (MVC, ARMS, fo) between the two groups (training and control) across time (week 0 and week 4) were carried out using two-way repeated-measures ANOVA, followed by post hoc Student’s paired and unpaired t-tests where appropriate. Comparisons of fo between the predicted values from the model and the experimental data were carried out by single sample t-test. Statistical significance level was set at P ⱕ 0.05 and all statistical procedures performed using SPSS software. Unless otherwise specified, data were presented as mean ⫾ standard error. RESULTS

There were no significant differences in any of the variables measured between the two groups at week 0 (P ⬎ 0.05). No significant changes were observed in any of the anthropometric measurements at any time point (P ⬎ 0.05). Compliance of the exercise group with the exercise protocol was 100% and no injuries related to training occurred. Effect of training on MVC force. The ANOVA for MVC force showed a significant effect of time and a significant group by time interaction (F ⫽ 21.2, P ⬍ 0.001). The post hoc analysis showed that at week 0 there were no differences in MVC force between the control and training groups. Following the intervention, there was a significant increase across time in the training group by 26.5% (P ⬍ 0.01, 534 ⫾ 75 N at week 0, 676 ⫾ 82 N at week 4), compared with the 0.8% increase in the control group (P ⬎ 0.05, 589 ⫾ 42 N at week 0, 593 ⫾ 47 N at week 4). Stretch reflex induced tremor in a control subject. Fig. 1, A and B, show 1 s segments of displacement recordings from a control participant during anisometric contraction at 30% MVC, with the 5.35 N/mm and 11.06 N/mm springs, respectively. Visual inspection shows no difference in the amplitude or the frequency of oscillation between week 0 and week 4. Figure 1, C and D, show the amplitude autospectra of the displacement recordings from Fig. 1, A and B. A clear peak is seen centered at about 4.5 Hz and 8 Hz in Fig. 1, C and D, respectively, and which show no change between weeks 0 and 4. For this individual, the observed frequencies of oscillation are lower than the expected values for the two springs. Quantitative analysis of the ARMS of the peaks revealed no change between weeks 0 and 4 (370 and 362 ␮m for Fig. 1C and 210 and 203 ␮m for Fig. 1D). A similar lack of change was observed in the other six control subjects. Effect of strength training on stretch reflex induced tremor. Figure 2, A and B, shows 1-s segments of displacement recordings from a training participant during anisometric contractions with the 11.06 N/mm spring for the absolute and relative force-matching protocols, respectively. Prior to the training intervention, this individual showed oscillations in


1650


Displacement (mm)

A

Durbaba R et al.

B

1

0

1

0

-1

-1 0. 0

0.5

1. 0

0 .0

1 .0

Time (s)

C 0.4

D 0.4 week 0 week 4

0.2

0.2

0.0

0.0 0

5

10

15

0

Frequency (Hz)

displacement with peak-to-peak amplitude of ⬃0.6 mm and a frequency of about 8 Hz. Following the training intervention, when the subject was asked to do anisometric contractions under the absolute force-matching protocol, there is a reduction in the amplitude of the oscillation but no change in the frequency of oscillation (Fig. 2A). This reduction in amplitude is clearly seen in Fig. 2C, which shows the amplitude autospectra for the records in Fig. 2A. There is a 55% reduction in ARMS with no change in frequency. In contrast, with the relative force-matching protocol there was no change in amplitude of oscillation (Fig. 2, B and D). As in the case of

Displacement (mm)

5

15

absolute force-matching, during relative force-matching the frequency of oscillation did not change. When this subject was asked to contract using the 5.35 N/mm spring, the same behavior was observed, though the oscillations occurred at a lower frequency. A similar pattern was observed in the other six training subjects. Tremor amplitude. Figure 3 shows a comparison of the ARMS of the induced tremor for control and training groups between week 0 and week 4. The top and bottom rows in Fig. 3 relate to 5.35 N/mm and 11.06 N/mm springs, respectively. For the subjects in the control group, no consistent change in

B 0.4

0.0

0.0

-0.4

-0.4 0. 0

0. 5

1. 0

0 .0

0. 5

Time (s)

1. 0

Time (s)

C 0.3 Amplitude (mm)

10

Frequency (Hz)

A 0.4

Fig. 2. A and B: 1 s segments of week 0 (black line) and week 4 (gray line) anisometric records for the 11.06 N/mm spring for absolute and relative force-matching contraction tasks, respectively, of a training group subject. C and D: autospectra relating to full records from A and B. See Fig. 1 for symbol legend.

0 .5

Time (s)

Amplitude (mm)

Fig. 1. A and B: 1 s segments of week 0 (black line) and week 4 (gray line) anisometric records for 5.35 N/mm and 11.06 N/mm springs, respectively, of a control subject. C and D: autospectra relating to full records from A and B; week 0 is represented by a black line, and week 4 is represented by a gray line.

•

D 0.3

0.2

0.2

0.1

0.1

0.0

0.0 0

5

10

15

0

Frequency (Hz)


5

10

Frequency (Hz)

15

Week 0 RMS Amplitude (µm)

D

450

450

B 450

400

400

350

*

Durbaba R et al.

450

C 450

450

400

400

400

400

350

350

350

350

350

300

300

300

300

300

300

250

250

250

250

250

250

200

200

200

200

200

200

150

150

150

150

150

150

100

100

100

100

100

100

50

50

50

50

50

50

0

0

0

0

0

0

300

300

E 300

300

F 300

300

250

250

250

250

250

250

200

200

200

200

200

200

150

150

150

150

150

150

100

100

100

100

100

100

50

50

50

50

50

50

0

0

0

0

0

0

ARMS was observed from week 0 to week 4 (filled circles, Fig. 3, A and D). Within the training group, for the absolute force-matching condition all subjects showed a clear decrease in ARMS at week 4 compared with week 0 for either spring (filled circles, Fig. 3, B and E), while for the relative forcematching condition no consistent change in ARMS was observed (filled circles, Fig. 3, C and F). In the absolute force-matching condition, ANOVA showed a significant effect of time and a significant group by time interaction in induced tremor ARMS for either spring used (F ⫽ 5.34 for the 5.35 N/mm spring; F ⫽ 5.58 for the 11.06 N/mm spring; P ⬍ 0.05). Post hoc analysis revealed that at week 0 there was no difference in ARMS between control and training groups (open circles on left-hand axes of Fig. 3). For the control group, there was no significant difference between week 0 and week 4 for either spring (open circles, Fig. 3, A and D). In the training group, the mean induced tremor ARMS reduced from 181 ⫾ 26 ␮m at week 0 to 124 ⫾ 23 ␮m at week 4 (a 31.6% reduction; open circles, Fig. 3B) and from 112 ⫾ 30 ␮m at week 0 to 70 ⫾ 13 ␮m at week 4 (a 37.5% reduction; open circles, Fig. 3E) for the 5.35 N/mm and 11.06 N/mm springs, respectively. In the case of the 5.35 N/mm spring this reduction was significant (P ⬍ 0.01), while for the 11.06 N/mm spring the reduction did not quite reach significance (P ⫽ 0.068) despite the greater relative reduction in mean induced tremor ARMS. In the case of the relative force-matching task, there were no significant differences in ARMS in both groups for either spring.



A

•



1651

Fig. 3. Root mean squared amplitude (ARMS) values at week 0 and week 4 for the (top) 5.35 N/mm and (bottom) 11.06 N/mm springs. A and D: control group. B and E: training group absolute force-matching condition. C and F: training group relative force-matching condition. Filled circles with solid lines represent individual subjects in each group. Open circles with dashed line represent mean values. *Significantly different between week 0 and week 4 at the P ⬍ 0.01 level.

Tremor frequency. Figure 4 shows a comparison of the frequency of oscillation of the induced tremor for control and training groups between week 0 and week 4. As in Fig. 3, the top and bottom rows in Fig. 4 relate to 5.35 N/mm and 11.06 N/mm springs, respectively. No consistent change in frequency was observed from week 0 to week 4 for the subjects in the control group or for those in the training group in either the absolute or relative force-matching conditions. ANOVA showed no significant effect of time and no significant group by time interaction in induced tremor frequency for either spring used, and this is reflected in Fig. 4 which shows the mean values for tremor frequency for each group and contraction task (open circles). This lack of change in induced tremor frequency is consistent with a lack of change in the dynamics of the system being tested. Owing to the lack of difference between groups for each spring or contraction task, it was decided to combine the observed values between groups. Thus for the 5.35 N/mm spring the mean frequency of oscillation was 5.1 ⫾ 0.1 Hz, while for the 11.06 N/mm spring it was 8.0 ⫾ 0.1 Hz. When these values were compared with the predicted values from the model, it was observed that in the case of the 5.35 N/mm spring, the computed mean value differs significantly from that expected when the short latency pathway dominates (P ⬍ 0.01; predicted ⫽ 5.6 Hz) but not when the long latency pathway dominated the oscillations (P ⬎ 0.05; predicted ⫽ 5.0 Hz). This suggests that for the 5.35 N/mm spring, the oscillations were being generated with a strong influence from the


Week 0 Frequency (Hz)

A

D Week 0 Frequency (Hz)

Fig. 4. Frequency of oscillation values at week 0 and week 4 for the (top) 5.35 N/mm and (bottom) 11.06 N/mm springs. A and D: control group. B and E: training group absolute forcematching condition. C and F: training group relative force-matching condition. Filled circles with solid lines represent individual subjects in each group. Open circles with dashed line represent mean values.

10

10

8

B

•

Durbaba R et al.

10

10

8

8

6

6

4

C

10

10

8

8

8

6

6

6

6

4

4

4

4

4

2

2

2

2

2

2

0

0

0

0

0

0

10

10

8

E

10

10

8

8

6

6

4

F

10

10

8

8

8

6

6

6

6

4

4

4

4

4

2

2

2

2

2

2

0

0

0

0

0

0

long latency stretch reflex pathway. For the 11.06 N/mm spring, the computed mean value was significantly different from the expected values from the model for both the short and long latency pathway, P ⬍ 0.05 (predicted ⫽ 8.4 Hz) and P ⬍⬍ 0.001 (predicted ⫽ 5.3 Hz), respectively. DISCUSSION

The main finding of this study was that 4 wk of highintensity isometric resistance training lead to a contraction task specific reduction in induced tremor amplitude during anisometric recordings in healthy young individuals, which accompanied the increased level of maximal voluntary contraction. The 26.5% increase in maximal voluntary strength observed in the training group is similar to that found by others over an equivalent period of training (12, 44, 52, 58, 68). It has to be noted that direct comparison between studies is difficult, since many variables, such as the intensity, the number of sets and repetitions of a given exercise, the muscle groups exercised, and the subjects’ group (e.g., age, sex), must be considered. We chose an isometric resistance training program that has previously shown to be effective in increasing maximal isometric voluntary strength by 20% after 4 wk of training (52). Given the period of training, 4 wk, it seems likely that this increase can be attributed mostly to neural adaptations of the trained muscle, rather than structural changes to the trained muscle itself (33, 51, 59, 71), although the evidence is far from conclusive (28). Neural adaptations include, among other fac-




1652

tors, improvements in agonist surface EMG amplitude, which might be related to either increased motor unit recruitment or firing rate of single motor units (51), with concurrent reductions in the activity of antagonist muscles (32), an increased corticospinal excitability (10), although some studies reported an unexpected decrease in corticospinal excitability (6, 42, 47), or reduced presynaptic inhibition of spinal reflexes (39, 45). The possibility that tremor could arise due to the stretch reflex has been studied previously (21, 43, 50, 55, 71, 78, 79, 88). These studies have tended to emphasize the contribution played by the spinal, or short latency, stretch reflex, with little consideration to the long latency pathway. The modeling described in the Methods section indicates that the long latency pathway does have the potential to induce instability, with a tendency toward oscillations having a lower frequency than for the short latency component. As both components of the stretch reflex can be seen in muscles that activate knee extensors (62), it seems likely that these frequencies are derived as a result of the individual responsiveness of the two pathways and their relative interaction. As we do not know what these interactions and their values are, our modeling was based on predicting expected frequency values with each component dominating the reflex behavior of the system. Compared with the expected values from the modeling study, there is a suggestion that one pathway may dominate the other for each particular spring. Thus, for the 11.06 N/mm spring, the observed mean frequency of induced tremor (8.0 Hz) though not



matching the expected value for the short latency pathway, is nevertheless at the lower limit of the generally accepted range of frequency values (8 –12 Hz) for tremor linked to short latency stretch reflex activity (21, 43, 50, 55, 71, 78, 79, 88). Therefore it seems likely that in our setup the short latency pathway may be the dominant pathway for inducing tremor when using the 11.06 N/mm spring, though a contribution from the long latency pathway cannot be excluded. For the 5.35 N/mm spring, the long latency pathway appears to dominate with there being no significant difference between the mean observed frequency and the expected value from the modeling for a dominant effect via the long latency pathway. This agrees with previous experimental and modeling studies relating to possible role for long loop reflex being able to generate tremor at 4 – 6 Hz (29, 66, 78, 79, 80, 88). There is some evidence to suggest that the long latency reflex might be involved in the generation of tremor in Parkinson’s disease or drug-induced Parkinsonism, through increased loop gain (25, 82). Also, Deuschl et al. (14) found that Essential Tremor patients could be subdivided according to long loop reflex behavior, with one group having increased long loop reflex amplitudes. This group of patients exhibited a lower tremor frequency than the group whose long loop reflex was normal. Therefore using springs of appropriate stiffness to mimic tremor frequencies associated with pathological tremors could be a useful tool for assessment of associated neural pathways. It is important to recognize that there are limitations with any modeling relating to the choice of values used for the parameters. In our model the critical calculation relates to that of lower leg inertia. With the exception of the shank radius, to determine appropriate mass, length, and radii of the shank and foot we used a mean value for scaling factors from published data (49, 87). If we consider the upper and lower limits of the published scaling factors and taking the ⫾5% variation for shank radius, then it is possible to calculate that the potential range of values of total inertia for the lower limb is between 0.48 and 0.51 kg·m2. This translates itself to a ⫾1.5% variation in either the resonant frequency of the spring-mass system alone or expected frequency of the whole oscillating system. This degree of variation does not affect our conclusions relating the oscillation frequency to a dominant reflex component. Previous studies have examined the effect of strength training on steadiness of knee extensor contractions (3, 41, 53, 84, 86). All these studies reported that strength training resulted in a reduction in force fluctuations when subjects contracted anisometrically but not during isometric contractions. As induced tremor has been reported to occur during anisometric contractions and not during isometric contractions (21, 50, 55), this reported difference of the effect of strength training on the anisometric and isometric contracted states would be consistent with the idea that the reductions might be mediated via the stretch reflex pathway. Hortobagyi et al. (41) found a reduction in force fluctuations of 31% and 30% for eccentric and concentric absolute force-matching anisometric submaximal contractions. Subsequent studies were all carried out using relative force-matching eccentric and concentric anisometric submaximal contractions (53, 84, 86), and they also found a reduction in force fluctuations. This is in contrast to the present study where no change was observed during relative force-matching contractions. Though the previous studies examined training on elderly adults, while in our study the subjects were young,

•

Durbaba R et al.

1653

it seems unlikely that age could account for the difference observed during a relative force-matching task, as similar neural and muscular changes have been reported for young and older individuals following resistance training (2, 28, 51). Instead, differences may be related to some methodological aspects. First, different forms of anisometric contractions were used, namely isokinetic concentric and eccentric contractions in the previous studies (53, 84, 86) as opposed to the “steadystate” contraction against a compliant spring in this study. During a lengthening contraction, stretch reflex activity for both the short and long latency component is decreased compared with a constant load contraction, while for a shortening contraction, the short latency component shows no change compared with constant load contraction and the long latency pathway is increased (17, 19, 63, 65, 69). It is possible that these sensory feedback differences coupled to the different motor commands that occur for the different forms of anisometric contraction (15, 64) may be influenced by training in different ways and could in part explain the difference between the current study and previous studies (53, 84, 86) during relative force-matching tasks. Further work is needed to determine if this is the case. Second, length of training; 4 wk used in our study while previous studies used either 10 wk (53) or 16 wk (84, 86). Periods of training ⬎ 8 wk have been shown to lead to significant muscular changes, particularly hypertrophy (2, 28, 30, 51). It is possible that these muscular changes, interacting with reported neural changes, might in part also explain the difference between the current study and previous studies (53, 84, 86) during relative force-matching tasks. It is well known that the response of the stretch reflex varies as the level of muscle activity increases. Initially there is an increase, followed by a plateau region between 40 and 50% MVC with a reduction at higher contraction levels (5, 8, 11, 35, 54, 61, 83). This is the concept known as automatic gain compensation. Our observations, that following 4 wk of highintensity isometric resistance training, induced tremor amplitude did not change during the relative force-matching task but was reduced during the absolute force-matching task could be explained in terms of “realignment” of the automatic gain compensation of the stretch reflex to the new maximal force output that typically accompanies high-intensity isometric resistance training. With such a realignment, the response of the stretch reflex pathway at a particular “relative” level of activity should be the same pre- or posttraining. Thus for a relative force-matching task, where the relative levels of activity are the same pre- or post-training, the lack of reduction we observed would be consistent with no change in stretch reflex response. For the absolute force-matching task, where the post-training activity is at a lower relative level to that pretraining, then the reduction in induced tremor amplitude can be explained by the stretch reflex response being reduced to that required for the lower level of relative activity. Previous studies have examined the effect of resistance training on the behavior of the stretch reflex pathway, as assessed by the H-reflex response. Most of these studies have been done by assessing the H-reflex response at rest (i.e., zero contraction) and, despite post-training increases in MVC ranging from 18 to 46% for training ranging from 3 to 16 wk, observed no change in H-reflex behavior (1, 13, 18, 22, 26, 27, 39, 73, 76). This is perhaps not surprising given that comparing the rest state pre- and post-training is an extreme example of


1654


relative force matching. In addition, it is difficult to directly relate H-reflex behavior at rest with that which occurs during a submaximal contraction, due to the different states of depolarization and cortical drive to ␣-motoneurones, as well as automatic gain compensation of the stretch reflex behavior. Indeed, Aagaard et al. (1) questioned the usefulness of relating training effects on H-reflex behavior assessed at rest to muscle activation. Only three studies have looked at H-reflex characteristics during submaximal contractions pre- and post-training and offer some evidence of an adaptation in stretch reflex behavior during submaximal contractions following resistance training (13, 39, 45). Lagerquist et al. (45) found that 5 wk of MVC isometric training of plantar flexors of the ankle lead to a 15.3% increase in MVC of the trained leg, which was accompanied by a 35% increase in H-reflex amplitude post-training during a 10% maximal EMG relative force-matching task. This was considered to be consistent with an increase in spinal excitability of the stretch reflex pathway following training and potentially mediated via a reduction in presynaptic inhibition of Ia synapses on to ␣-motoneurones. However, though the H-reflex amplitude was increased, the input/output characteristic of the system, reflex gain, was not changed as the slope of the ascending limb of the M-H recruitment curve was unaltered post-training. Del Balso and Cafarelli (13) similarly reported that 4 wk isometric resistance training of ankle plantar flexors induced a 20% increase in MVC, but during a 20% MVC relative force-matching task they did not observe a change in reflex, which in their study was computed as the ratio of Hslope and Mslope, obtained from standard H-reflex and M-wave response/stimulus intensity curves. The lack of a change in reflex gain seen in both studies would indicate that the reflex input is increased in proportion to changes in behavior of motor output following resistance training. Thus the overall behavior of the reflex pathway pre- and post-training could be the same at a particular relative level of background activity. Reflex gain has been shown to be an important factor in determining tremor amplitude, with an increase or decrease leading to increased or decreased tremor output, respectively (78, 79, 88). Our observation of a lack of change in induced tremor amplitude during relative force-matching tasks following resistance training would be consistent with the idea that there has been no change in reflex gain. The observations by Langerquist et al. (45) and Del Balso and Cafarelli (13) that reflex gain does not change for relative force-matching tasks following training relates only to low-level contractions, and therefore further studies relating reflex gain to training effects covering a wider range of submaximal contraction levels are needed to confirm their observations and whether it can be applied to all levels of activity. The only study to have examined H-reflex behavior during submaximal absolute and relative force-matching contractions following resistance training is that by Holtermann et al. (39). Following training, they found for a low-level submaximal relative force-matching contraction (20% MVC) that the H-reflex amplitude was increased by 17.4%. When the response was measured for the corresponding absolute force-matching contraction (17% of post-training MVC) the increase was only 14.8%. Also, they examined the H-reflex response at a higher level of submaximal contraction. For a 60% MVC relative force-matching task the H-reflex amplitude increased by 14.7%, while a larger increase (18.5%) was observed with the

•

Durbaba R et al.

corresponding absolute force-matching task (51% of posttraining MVC). The increases, irrespective of the level or the type of contraction task, were consistent with those seen by Langerquist et al. (45), and were in part attributed to decreased presynaptic inhibition following resistance training. The difference in the increase in H-reflex amplitude between relative and absolute force-matching tasks at either level of contraction observed by Holtermann et al. (39) may be explained by automatic gain compensation. If the reported profile of automatic gain compensation for the stretch reflex is realigned to the new MVC output following training, then for contractions ⬍40% MVC a smaller overall reflex output would be observed during an absolute force-matching contraction when the relative level of activity is less than for an relative force-matching task. For contractions ⬎50% MVC a larger reflex output would be observed for absolute force-matching contractions despite the decreased level of activity. Thus it is suggestive that the reduction in the amplitude of induced tremor we observe during low-level absolute force-matching contractions following resistance training could be explained by automatic gain compensation leading to a lower stretch reflex output. The idea that automatic gain compensation may be realigned to the new MVC output following resistance training is a novel concept though might not be unexpected. Capaday and Stein (5) showed that the reported profile for stretch reflex automatic gain compensation is dependent upon the recruitment of progressively larger motor units. As motor unit recruitment order is not changed by resistance training (16), then it might be that the profile of stretch reflex automatic gain compensation is likewise unaltered. Investigation of stretch reflex automatic gain compensation pre- and post-training is required over the full range of submaximal contraction levels to determine if this is the case. Resistance training has been shown to restore some of the loss in function in stability in the lower limb of elderly individuals, with improvements in steadiness specific to submaximal anisometric contractions (41, 53, 84, 86). On the basis of our present observations it is a reasonable assumption that these observed improvements may in part be due to reduced stretch reflex activity that accompanies a lower level of muscle activity during an absolute force-matching contraction posttraining. Resistance training has also been shown to be effective in individuals who have neuropathological tremor. Bilodeau et al. (4) found in individuals with Essential Tremor that 4 wk of high-intensity strength training lead to a 13% increase in maximal index finger abduction force, which was accompanied by a reduction of isometric force fluctuations during absolute force-matching tasks only, where the relative force level was smaller post-training compared with pretraining. This decrease appears unlikely to be associated to the stretch reflex due to the minimal role it plays in generating instability during an isometric contraction (21, 50, 55). However, there is evidence to indicate that the stretch reflex may be a contributing factor to the isometric tremor of individuals with Essential Tremor (37, 38, 48). Thus it is possible that some of the loss in tremor reported by Bilodeau et al. (4) could be due to a reduction in stretch reflex response, though further studies will need to be carried out on individuals with Essential Tremor to determine if this is the case.



In conclusion, the present study has shown that high-intensity isometric resistance training leads to a reduction in induced tremor that is contraction task dependent. This reduction can be related to a realignment of the stretch reflex automatic gain compensation in relation to the new maximal force output. This could in part explain the reported usefulness of resistance training as a rehabilitation tool for individuals with instability problems associated to aging. ACKNOWLEDGMENTS The authors thank Chris Mercer for his help in carrying out some of the experimental measures. DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the author(s). AUTHOR CONTRIBUTIONS Author contributions: R.D. and A.M. conception and design of research; R.D., A.C., F.B., and A.M. analyzed data; R.D., A.C., F.B., and A.M. interpreted results of experiments; R.D., A.C., and F.B. prepared figures; R.D. and A.M. drafted manuscript; R.D., A.C., F.B., and A.M. edited and revised manuscript; R.D., A.C., F.B., and A.M. approved final version of manuscript; A.C. and F.B. performed experiments. REFERENCES 1. Aagard P, Simonsen EB, Andersen JL, Magnusson P, Dyhre-Poulsen P. Neural adaptation to resistance training: Changes in evoked V-wave and H-reflex responses. J Appl Physiol 92: 2309 –2318, 2002. 2. Barry BK, Carson RG. The consequences of resistance training for movement control in older adults. J Gerontol Med Sci 59A: 730 –754, 2004. 3. Beck TW, Defreitas JM, Stock MS, Dillon MA. Effects of resistance training on force steadiness and common drive. Muscle Nerve 43: 245– 250, 2011. 4. Bilodeau M, Keen DA, Sweeney PJ, Shields RW, Enoka RM. Strength training can improve steadiness in persons with essential tremor. Muscle Nerve 23: 771–778, 2000. 5. Capaday C, Stein RB. A method for simulating the reflex output of a motoneuron pool. J Neurosci Methods 21: 91–104, 1987. 6. Carroll TJ, Riek S, Carson RG. The sites of neural adaptation induced by resistance training in humans. J Physiol 544: 641–652, 2002. 7. Carville SF, Perry MC, Rutherford OM, Smith ICH, Newham DJ. Steadiness of quadriceps contractions in young and older adults with and without a history of falling. Eur J Appl Physiol 100: 527–533, 2007. 8. Cathers I, O’Dwyer N, Neilson P. Variation of magnitude and timing of wrist flexor stretch reflex across the full range of voluntary activation. Exp Brain Res 157: 324 –335, 2004. 9. Christakos CN, Papadimitriou NA, Erimaki S. Parallel neuronal mechanisms underlying physiological force tremor in steady muscle contraction of humans. J Neurophysiol 95: 53–66, 2006. 10. Classen J, Liepert J, Wise SP, Hallett M, Cohen LG. Rapid plasticity of human cortical movement representation induced by practice. J Neurophysiol 79: 1117–23, 1998. 11. Cronin NJ, Peltonen J, Ishikawa M, Komi PV, Avela J, Sinkjaer T, Voigt M. Effects of contraction intensity on muscle fascicle and stretch reflex behavior in the human triceps surae. J Appl Physiol 105: 226 –232, 2008. 12. Darcus HD, Salter N. The effect of repeated muscular exertion on muscle strength. J Physiol 129: 325–336, 1955. 13. Del Balso C, Cafarelli E. Adaptations in the activation of human skeletal muscle induced by short-term isometric resistance training. J Appl Physiol 103: 402–411, 2007. 14. Deuschl G, Lücking CH, Schenck E. Essential tremor: Electrophysiological and pharmacological evidence for a subdivision. J Neurol Neurosurg Psychiatry 50: 1435–1441, 1987. 15. Duchateau J, Enoka RM. Neural control of shortening and lengthening contractions: Influence of task constraints. J Physiol 586: 5853–5864, 2008. 16. Duchateau J, Semmler JG, Enoka RM. Training adaptations in the behavior of human motor units. J Appl Physiol 101: 1766 –1775, 2006.

•

Durbaba R et al.

1655

17. Duclay J, Martin A. Evoked H-reflex and V-wave responses during maximal isometric, concentric, and eccentric muscle contraction. J Neurophysiol 94: 3555–3562, 2005. 18. Duclay J, Martin A, Robbe A, Pousson M. Spinal reflex plasticity during maximal dynamic contractions after eccentric exercise. Med Sci Sports Exerc 40: 722–734, 2008. 19. Duclay J, Pasquet B, Martin A, Duchateau J. Specific modulation of corticospinal and spinal excitabilities during maximal voluntary isometric, shortening and lengthening contractions in synergist muscles. J Physiol 589: 2901–2916, 2011. 20. Durbaba R, Cassidy A, Budini F, Macaluso A. The effects of isometric strength training on stretch reflex-induced tremor in humans. Proc Physiol Soc 11: C90, 2008. 21. Durbaba R, Taylor A, Manu CA, Buonajuti M. Stretch reflex instability compared in three different human muscles. Exp Brain Res 163: 295–305, 2005. 22. Ekblom MMN. Improvements in dynamic plantar flexor strength after resistance training are associated with increased voluntary activation and V-to-M ratio. J Appl Physiol 109: 19 –26, 2010. 23. Enoka RM, Christou EA, Hunter SK, Kornatz KW, Semmler JG, Taylor AM, Tracy BL. Mechanisms that contribute to differences in motor performance between young and old adults. J Electromyogr Kinesiol 13: 1–12, 2003. 24. Erimaki S, Christakos CN. Coherent motor unit rhythms in the 6 –10 Hz range during time-varying voluntary muscle contractions: Neural mechanism and relation to rhythmical motor control. J Neurophysiol 99: 473– 483, 2008. 25. Filion M, Tremblay L, Bédard PJ. Abnormal influences of passive limb movement on the activity of globus pallidus neurons in parkinsonian monkeys. Brain Res 444: 165–176, 1985. 26. Fimland MS, Helgerud J, Gruber M, Leivish G, Hoff J. Functional maximal strength training induces neural transfer to single-joint tasks. Eur J Appl Physiol 107: 21–29, 2009. 27. Fimland MS, Helgerud J, Solstad GM, Iversen VM, Leivish G, Hoff J. Neural adaptation underlying cross-education after unilateral strength training. Eur J Appl Physiol 107: 723–730, 2009. 28. Folland JP, Williams AG. The adaptations to strength training: Morphological and neurological contributions to increased strength. Sports Med 37: 145–168, 2007. 29. Gottlieb S, Lippold OCJ. The 4 – 6 Hz tremor during sustained contraction in normal human subjects. J Physiol 336: 499 –509, 1983. 30. Guilhem G, Cornu C, Guével A. Neuromuscular and muscle-tendon system adaptations to isotonic and isokinetic eccentric exercise. Ann Phys Rehabil Med 53: 319 –341, 2010. 31. Hagbarth KE, Young RR. Participation of the stretch reflex in human physiological tremor. Brain 102: 509 –526, 1979. 32. Häkkinen K, Kallinen M, Izquierdo M, Jokelainen K, Lassila H, Malkia E, Kraemer WJ, Newton RU, Alen M. Changes in agonistantagonist EMG, muscle CSA, and force during strength training in middle-aged and older people. J Appl Physiol 84: 1341–1349, 1998. 33. Häkkinen K, Kraemer WJ, Newton RU, Alen M. Changes in electromyographic activity, muscle fibre and force production characteristics during heavy resistance/power strength training in middle-aged and older men and women. Acta Physiol Scand 171: 51–62, 2001. 34. Halliday AM, Redfearn JW. Finger tremor in tabetic patients and its bearing on the mechanism producing the rhythm of physiological tremor. J Neurol Neurosurg Psychiatry 21: 101–108, 1958. 35. Heckman CJ. Computer simulations of the effects of different synaptic input systems on the steady-state input-output structure of the motoneuron pool. J Neurophysiol 71: 1727–1739, 1994. 36. Hernandez ME, Goldberg A, Alexander NB. Decreased muscle strength relates to self-reported stooping, crouching, or kneeling difficulty in older adults. Phys Ther 90: 67–74, 2010. 37. Héroux ME, Pari G, Norman KE. The effect of inertial loading on wrist postural tremor in essential tremor. Clin Neurophysiol 120: 1020 –1029, 2009. 38. Héroux ME, Pari G, Norman KE. The effect of contraction intensity on force fluctuations and motor unit entrainment in individuals with essential tremor. Clin Neurophysiol 121: 233–239, 2010. 39. Holtermann A, Roeleveld K, Engstrøm M, Sand T. Enhanced H-reflex with resistance training is related to increased rate of force development. Eur J Appl Physiol 101: 301–312, 2007.


1656


40. Hömberg V, Hefter H, Reiners K, Fruend HJ. Differential effects of changes in mechanical limb properties on physiological and pathological tremor. J Neurol Neurosurg Psychiatry 50: 568 –579, 1987. 41. Hortobágyi T, Tunnel D, Moody J, Beam S, DeVita P. Low- or high-intensity strength training partially restores impaired quadriceps force accuracy and steadiness in aged adults. J Gerontol A Biol Sci Med Sci 56: B38 –B47, 2001. 42. Jensen JL, Marstrand PC, Nielsen JB. Motor skill training and strength training are associated with different plastic changes in the central nervous system. J Appl Physiol 99: 1558 –1568, 2005. 43. Joyce GC, Rack PMH. The effects of load and force on tremor at the normal human elbow joint. J Physiol 240: 375–396, 1974. 44. Komi PV, Viitasalo JT, Rauramaa R, Vihko V. Effect of isometric strength training of mechanical, electrical, and metabolic aspects of muscle function. Eur J Appl Physiol Occup Physiol 40: 45–55, 1978. 45. Lagerquist O, Zehr EP, Docherty D. Increased spinal reflex excitability is not associated with neural plasticity underlying the cross-education effect. J Appl Physiol 100: 83–90, 2006. 46. Laidlaw DH, Bilodeau M, Enoka RM. Steadiness is reduced and motor unit discharge is more variable in old adults. Muscle Nerve 23: 600 –612, 2000. 47. Lee M, Gandevia SC, Carroll TJ. Short-term strength training does not change cortical voluntary activation. Med Sci Sports Exerc 41: 1452–1460, 2009. 48. Lee RG, Stein RB. Resetting of tremor by mechanical perturbations: A comparison of essential tremor and parkinsonian tremor. Ann Neurol 10: 523–531, 1981. 49. de Leva P. Adjustments to Zatsiorsky-Seluyanov’s segment inertia parameters. J Biomech 29: 1223–1230, 1996. 50. Lippold OCJ. Oscillation in the stretch reflex arc and the origin of the rhythmical 8 –12c/s component of physiological tremor. J Physiol 206: 359 –382, 1970. 51. Macaluso A, De Vito G. Muscle strength, power and adaptations to resistance training in older people. Eur J Appl Physiol 91: 450 –472, 2004. 52. Macaluso A, De Vito G, Felici F, Nimmo MA. Electromyogram changes during sustained contraction after resistance training in women in their 3rd and 8th decades. Eur J Appl Physiol 82: 418 –424, 2000. 53. Manini TM, Clark BC, Tracy BL, Burke J, Ploutz-Snyder L. Resistance and functional training reduces knee extensor position fluctuations in functionally limited older adults. Eur J Appl Physiol 95: 436 –446, 2005. 54. Matthews PBC. Observations on the automatic compensation of reflex gain on varying the pre-existing level of motor discharge in man. J Physiol 374: 73–90, 1986. 55. Matthews PBC, Muir RB. Comparison of electromyogram spectra with force spectra during human elbow tremor. J Physiol 302: 427–441, 1980. 56. McComas AJ. Invited review: Motor unit estimation: Methods, results, and present status. Muscle Nerve 14: 585–597, 1991. 57. McComas AJ, Galea V, de Bruin H. Motor unit populations in healthy and diseased muscles. Phys Ther 73: 868 –877, 1993. 58. McDonagh MJ, Hayward CM, Davies CT. Isometric training in human elbow flexor muscles: The effects on voluntary and electrically evoked forces. J Bone Joint Surg Br 65: 355–358, 1983. 59. Moritani T, deVries HA. Potential for gross muscle hypertrophy in older men. J Gerontol 35: 672–682, 1980. 60. Moxley Scarborough D, Krebs DE, Harris BA. Quadriceps muscle strength and dynamic stability in elderly persons. Gait Posture 10: 10 –20, 1999. 61. Mrachacz-Kersting N, Sinkjaer T. Reflex and non-reflex torque responses to stretch of the human knee extensors. Exp Brain Res 151: 72–81, 2003. 62. Mrachacz-Kersting N, Grey MJ, Sinkjaer T. Evidence for a supraspinal contribution to the human quadriceps long-latency stretch reflex. Exp Brain Res 168: 529 –540, 2006. 63. Nakazawa K, Yamamoto SI, Yano H. Short- and long-latency reflex responses during different motor tasks in elbow flexor muscles. Exp Brain Res 116: 20 –28, 1997. 64. Nielsen JB. Sensorimotor integration at spinal level as a basis for muscle coordination during voluntary movement in humans. J Appl Physiol 96: 1961–1967, 2004.

•

Durbaba R et al.

65. Nordland MM, Thorstensson A, Cresswell AG. Variations in the soleus H-reflex as a function of activation during controlled lengthening and shortening actions. Brain Res 952: 301–307, 2002. 66. Noth J, Matthews HR, Friedemann HH. Long latency reflex force of human finger muscles in response to imposed sinusoidal movements. Exp Brain Res 55: 317–324, 1984. 67. Orr R, Raymond J, Fiatarone Singh M. Efficacy of progressive resistance training on balance performance in older adults: A systematic review of randomized controlled trials. Sports Med 38: 317–343, 2008. 68. Porter MM, Vandervoort AA. Standing strength training of the ankle plantar and dorsiflexors in older women, using concentric and eccentric contractions. Eur J Appl Physiol Occup Physiol 76: 62–68, 1997. 69. Romanò C, Schieppati M. Reflex excitability of human soleus motoneurones during voluntary shortening or lengthening contractions. J Physiol 390: 271–284, 1987. 70. Sadeghi H, Prince F, Zabjek KF, Sadeghi S, Labelle H. Knee flexors/ extensors in gait of elderly and young able-bodied men (II). Knee 9: 55–63, 2002. 71. Sale DG. Neural adaptation to resistance training. Med Sci Sports Exerc 20: S135–S145, 1980. 72. Sanes JN. Absence of enhanced physiological tremor in patients without muscle or cutaneous afferents. J Neurol Neurosurg Psychiatry 48: 645– 649, 1985. 73. Scaglioni G, Ferri A, Minetti AE, Martin A, Van Hoecke J, Capodaglio P, Sartorio A, Narici MV. Plantar flexor activation capacity and H reflex in older adults: Adaptations to strength training. J Appl Physiol 92: 2292–2302, 2002. 74. Schieppati M, Nardone A. Medium-latency stretch reflexes of foot and leg muscles analyzed by cooling the lower limb in standing humans. J Physiol 503: 691–698, 1997. 75. Schieppati M, Nardone A, Siliotto R, Grasso M. Early and late stretch responses of human foot muscles induced by perturbation of stance. Exp Brain Res 105: 411–422, 1995. 76. Schubert M, Beck S, Taube W, Amtage F, Faist M, Gruber M. Balance training and ballistic strength training are associated with task-specific corticospinal adaptations. Eur J Neurosci. 27, 2007–2018, 2008. 77. Steib S, Schoene D, Pfeifer K. Dose-response relationship of resistance training in older adults: A meta-analysis. Med Sci Sports Exerc 42: 902–914, 2010. 78. Stein RB, Oguztöreli MN. Tremor and other oscillations in neuromuscular systems. Biol Cybern 22: 147–157, 1976. 79. Stein RB, Oguztöreli MN. Reflex involvement in the generation and control of tremor and clonus, in Progress in Clinical Neurophysiology, edited by Desmedt JE, Basel, Switzerland: Karger, 1978, pp. 28 –50. 80. Stiles RN. Frequency and displacement amplitude relation for normal hand tremor. J Appl Physiol 40: 44 –54, 1976. 81. Stiles RN. Mechanical and neural feedback factors in postural hand tremor of normal subjects. J Neurophysiol 44: 40 –59, 1981. 82. Tatton WG, Lee RG. Evidence for abnormal long-loop reflexes in rigid Parkinsonian patients. Brain Res 100: 671–676, 1975. 83. Toft E, Sinkjaer T, Andreassen S. Mechanical and electromyographic responses to stretch of the human anterior tibial muscle at different levels of contraction. Exp Brain Res 74: 213–219, 1989. 84. Tracy BL, Byrnes WC, Enoka RM. Strength training reduces force fluctuations during anisometric contractions of the quadriceps femoris muscles in old adults. J Appl Physiol 96: 1530 –1540, 2004. 85. Tracy BL, Enoka RM. Older adults are less steady during submaximal isometric contractions with the knee extensor muscles. J Appl Physiol 92: 1004 –1012, 2002. 86. Tracy BL, Enoka RM. Steadiness training with light loads in the knee extensors of elderly adults. Med Sci Sports Exerc 38: 735–745, 2006. 87. Winter DA. Biomechanics and Motor Control of Human Movement. Hoboken, NJ: John Wiley, 2009. 88. Zhang D, Poignet P, Bo AP, Ang WT. Exploring peripheral mechanism of tremor on neuromusculoskeletal model: A general simulation study. IEEE Trans Biomed Eng 56: 2359 –2369, 2009.
