Secondary Logo

Journal Logo

Original Research

Effects of Regular Heel-Raise Training Aimed at the Soleus Muscle on Dynamic Balance Associated With Arm Movement in Elderly Women

Fujiwara, Katsuo; Toyama, Hiroshi; Asai, Hitoshi; Yaguchi, Chie; Irei, Mariko; Naka, Masami; Kaida, Chizuru

Author Information
Journal of Strength and Conditioning Research: September 2011 - Volume 25 - Issue 9 - p 2605-2615
doi: 10.1519/JSC.0b013e3181fb4947
  • Free

Abstract

Introduction

Deterioration of equilibrium function is considered to be a primary cause of falls among the elderly (22). To reduce the risk of falls, improving muscle strength of the legs has been applied as a successful preventive strategy for improving dynamic balance. For dynamic balance, comparatively strong muscle strength is needed compared to static balance (9,31,33,34). However, most studies on balance have been conducted with a very global view and have been mainly concentrated on the stability of balance. Moreover, no study has taken muscle fiber composition into consideration in leg muscle training. In many cases, muscle strength in resistance training, where fast-twitch muscle fibers are numerously recruited, has been shown to be considerably high (>50% of maximum) (18,23). In daily life, the soleus and gastrocnemius are most strongly activated to maintain a standing posture. These activities are approximately <30% of the maximum (30) and activate slow-twitch muscle fibers (18). The number of fibers and mean fiber area in fast-twitch muscle fibers decrease more rapidly with age than those in slow-twitch muscle fibers (26,27). The rate of the decrease is thought to be related to trainability. For example, slow-twitch muscle fibers should have higher trainability than fast-twitch muscle fibers do. Therefore, it is presumably important to execute low-intensity muscle training focused on slow-twitch muscle fibers to improve balance, particularly in the elderly.

In daily life, various arm movements are performed while standing. Many studies have shown that during rapid arm flexion while standing, the postural muscles of the legs and trunk that control standing posture are activated in advance of the focal muscles for rapid movement of the arm (3). It has been presumed that this prior activation of the postural muscles is controlled by a program selected in advance to moderate postural disturbance caused by the arm movement (14,20). The preceding activation of the biceps femoris is less observable in the elderly compared to in young subjects (35,40). In elderly subjects, postural movement during arm movement is smaller at the ankle than at the hip (1,5). These findings probably result from shifts in focus of muscular control from the entire body to the trunk caused by age-related changes, such as deterioration of the leg muscles. To demonstrate this hypothesis, this study was designed to increase leg muscle strength and size through heel-raise training.

Muscle strength is proportional to muscle cross-sectional area, and, in turn, muscle thickness is an important factor related to muscle cross-sectional area (2). Many researchers have used ultrasound to measure muscle thickness (29). Ultrasound predominates for this measurement, in part, because of portability and avoidance of radiation exposure. Using ultrasound, the effects of low-intensity muscle training, in which heel-raise performed 100 times per day for 2 months, have previously been investigated in healthy elderly women (13). The 100 times was decided based on the finding that maximum repetition of the muscle contraction is 60-110 times at 30% of maximum voluntary contraction (4,23). In this muscle load, slow-twitch muscle fibers may be predominantly activated. The training in this study focused on the soleus, which has a high proportion (ca. 90%) of slow-twitch muscle fibers compared to the gastrocnemius (ca. 50% slow-twitch muscle fibers) (24,38). In our previous study (13), significant increases in plantar flexor strength and in soleus thickness have been observed through the low-intensity muscle training, indicating that this is an effective muscle training method for the soleus. It is presumed that the enhancement of plantar flexor strength by such training could induce a change of postural control modality during arm flexion and improve postural stability.

When the arms are bilaterally and rapidly flexed anteriorly, balance is disturbed by the increase in rotational momentum of the body at the ankle joints in the anterior direction. Therefore, the strength of the triceps surae, which is one of the plantar flexors, is important to counteract this postural disturbance and maintain balance (17). In the control of standing posture during arm flexion, the gastrocnemius and soleus act synergistically and have a similar action. However, these muscles are not activated in advance of the focal muscles that rapidly move the arm from the hanging position (7,14), and activation timing of these muscles is yet to be thoroughly investigated. Muscle activities of the erector spinae and biceps femoris reportedly precede focal muscle activity to a greater extent in own-timing tasks compared to in reaction tasks (20,37). These findings suggest that the duration by which activation of the postural muscles precedes that of focal muscles decreases when there is insufficient preparation for postural control. The effect of muscle training of the triceps surae on postural control modality and stability associated with arm flexion presumably differs between these 2 tasks, as the preparation for postural control differs.

Therefore, in this study, elderly subjects underwent regular heel-raise training focusing on the soleus. The effects of this training on the relationships between the changes in plantar flexor strength and muscle thicknesses of the soleus and gastrocnemius (independent variables), and changes in postural control and stability (dependent variables) in a task involving bilateral arm flexion were investigated in own-timing and simple-reaction conditions. It was predicted that plantar flexor strength and soleus thickness would markedly increase as a result of heel-raise training and that ankle movement for postural control would increase during arm flexion in the own-timing condition. Moreover, in the own-timing condition, it was predicted that the electromyogram (EMG) burst in the soleus would initiate earlier and postural stability would improve.

Methods

Experimental Approach to the Problem

In this study, the effects of low-intensity training of the leg muscle on postural control and stability in dynamic balance associated with arm movement were investigated. The target muscle was the soleus, which has a high proportion (ca. 90%) of slow-twitch muscle fibers. Subjects performed heel-raise training 100 times per day for 2 months. The effect of this muscle training was evaluated in terms of plantar flexor strength and muscle thicknesses of the soleus and gastrocnemius (independent variables). Postural control modality (dependent variable) was evaluated as the time by which postural muscle was activated with respect to anterior deltoid during arm flexion, and this was taken as an indicator of anticipatory postural control. At the same time, the postural movement pattern during arm flexion was evaluated by measurement of ankle and hip movement angles. In addition, postural stability (dependent variable) was evaluated as the displacement of the center of foot pressure in the anteroposterior direction (CoPy) during arm flexion and as the SD of CoPy fluctuation just after arm movement.

Subjects

Subjects were 26 community-dwelling, healthy, postmenopausal elderly women, who were not performing any regular training or exercise before the study. They were divided into 2 groups: a training group (n = 13; mean age, 69.3 years [SD = 4.4]) and a control group (n = 13; mean age, 69.0 years [SD = 3.5]). All subjects lived within a farming community of a single district in central Japan. Subject health status was assessed by a questionnaire. No restrictions in activities of daily living because of pain, cardiopulmonary, musculoskeletal and neurological impairment were present, and no subjects were using analgesics. In accordance with the Declaration of Helsinki, all subjects provided informed consent by signing a consent form after an explanation of ensuring confidentiality of results, protecting privacy rights, and experimental protocols, which were approved by the institutional ethics committee of Kanazawa University.

Training Protocol

As training for the ankle plantar flexors, subjects in the training group performed 1 set of 100 heel raises per day for 62 days (Figure 1). For each set, subjects were instructed to raise both heels slowly from a standing position at intervals of approximately 2 seconds and to maintain the heel-raise position for approximately 1 second, to avoid a sudden high load on the triceps surae. The heel-raise position was set at the height at which subjects were able to repeat the movement 100 times. Subjects were instructed not to change their usual life style, nutrition, and hydration during this test and training period.

F1-33
Figure 1:
Heel-raise training.

Tests described below were performed before and after the training period in both groups. Each subject was tested at the same time in the morning for 1 hour.

Measurement of Plantar Flexor Strength

Maximal isometric muscle strength of the plantar flexor was measured in a sitting position using an instrument invented by Fujiwara et al. (13) (Figure 2). The right knee and ankle were kept at 90° flexion, and the knee was secured by a belt on the knee. Plantar flexor strength was measured as the force exerted at the first metatarsal head of the right foot. When plantar flexion force was generated, subjects were instructed to generate the force during exhalation to prevent rapid elevation of blood pressure. This measurement was performed twice, and the higher value was used for calculations.

F2-33
Figure 2:
Experimental setup for measurement of plantar flexor strength.

Measurement of Muscle Thickness

Thicknesses of the soleus and gastrocnemius were measured using a real-time B-mode ultrasound scanner (EUB-405B, Hitachi Medico, Tokyo, Japan) with a 3.8-cm, 10-MHz linear array probe. Measurements were performed directly on the screen using electronic calipers with 0.1-mm resolution. To measure the right side, subjects sat on a chair with the right foot on the floor; the ankle at 0° of plantar flexion, and the knee at 90° flexion. To define a measuring target point on the leg, the level of maximum girth of the right leg was first determined using a tape measure. Next, at this level, the midpoint of the width of the gastrocnemius medialis was marked with a red permanent marker as the target point. During ultrasound measurement, the longitudinal axis of the right leg was parallel to the vertical axis of gravity. For ultrasound scanning, the head of the ultrasound probe was coated with coupling gel and applied to the target point. The probe was oriented in the axial plane, perpendicular to the muscle. The ultrasound image was displayed on a computer screen. The thicknesses of the soleus and gastrocnemius were measured under the target point. During scanning, great care was taken to manipulate the probe so the fasciae were parallel and to avoid compressing the dermal surface. Gastrocnemius thickness was defined as the anterior-posterior distance between the midpoints of the fascia posterior to the muscle and the fascia separating the gastrocnemius medialis from the soleus. Soleus thickness was defined as the anterior-posterior distance between the midpoints of the fascia separating the gastrocnemius medialis from the soleus and the fascia anterior to the soleus (Figure 3). The reliability and validity of the ultrasound measurement of thickness of the gastrocnemius and soleus have been previously confirmed (11). The intraclass coefficient for 2 measurements of the muscle thicknesses was extremely high (r = 0.99) (11). Muscle thickness in this study was measured by a single experimenter.

F3-33
Figure 3:
Experimental setup for measurement of muscle thickness of gastrocnemius medialis and soleus (A) and a typical ultrasound image (B). Thicknesses are indicated by white arrows.

Measurement of Postural Controllability

Measurements of postural controllability were performed while subjects were standing barefoot on a force platform with feet 10 cm apart and parallel. Subjects were instructed to extend their elbows, position their hands on the thigh anterior to the greater trochanter and keep their eyes fixed on a light-emitting diode (LED). Initially, CoPy position was measured for 10 seconds while subjects maintained a quiet standing posture. The trial was repeated 5 times, and the mean value for these trials was adopted as the representative quiet standing position.

Next, bilateral arm flexion trials were commenced in 2 tasks (simple-reaction and own-timing conditions) (Figure 4). In each condition, subjects maintained the CoPy position within a range of ±1 cm of quiet standing for at least 3 seconds while a buzzer rang for 3 seconds. The experimenter then stopped the sound. In the simple-reaction condition, an LED on-signal was presented at 2-4 seconds from cessation of the buzzer sound. In response to the on-signal, subjects initiated bilateral arm flexion as quickly as possible. In the own-timing condition, within 3 seconds of cessation of the buzzer sound, subjects initiated arm flexion at their own pace. In both conditions, subjects were instructed to flex their arms at maximum speed, stop voluntarily when their hands reached shoulder level and maintain the horizontal arm position for 3 seconds. To help subjects understand and become familiar with these tasks, there were 10 practice trials for each condition before the experimental trials. In each condition, the experimental trials were repeated 10 times with a 30-second rest period between trials. A 3-minute seated rest was allowed between the 2 conditions. The CoPy position during arm flexion trials was monitored by the experimenter. Trials in which the CoPy position just before arm flexion was beyond ±1 cm of the subject's representative CoPy position during quiet standing were excluded.

F4-33
Figure 4:
Experimental setup for measurement of postural controllability. Open circles and filled circles indicate landmarks and surface electrodes, respectively.

Apparatus for Evaluating Postural Controllability

A force platform (WA1001, WAMI, Tokyo, Japan) consisting of 3 load cells was used to measure CoPy. The digital CoPy signal was sent to a computer (PC9801BX2, NEC, Tokyo, Japan) via an A/D converter (PIO9045, IO-DATA, Kanazawa, Japan) with a 1,000-Hz sampling rate and 12-bit resolution. The CoPy position was shown as percentage distance from the heels in relation to foot length (%FL). Onset timing of burst activation of postural muscles with respect to focal muscle is influenced by CoPy position just before arm movement (14). To control the initial CoPy position, the buzzer sound was generated by the computer when CoPy was located within ±1.0 cm of the subject's mean quiet standing position. Goshima (15) reported that the SD for CoPy fluctuation during quiet standing for 60 seconds was approximately 0.5 cm for normal subjects (mean age, 20 years). Therefore, the range of CoPy fluctuation during quiet standing was set at 1.0 cm, corresponding to 2SDs.

An LED was positioned 1.5 m in front of the force platform at the eye level of the subjects and was used both as the on and off signal indicating the beginning and end of each trial and as a fixation point.

To measure motion of the trunk and leg in the sagittal plane during arm flexion, a video camera (NV-SX550, Panasonic, Osaka, Japan) was placed 8 m to the left side of the subject. Small reflective markers (15-mm diameter) were placed over the following landmarks on the left side: the spinous processes of C7, greater trochanter midpoint, knee (lateral epicondyle of humerus), lateral malleolus, and fifth metatarsal head. Body images during arm flexion were recorded on videotape in the sagittal view using a video recorder (BR-9000, Victor, Yokohama, Japan) along with a digital display of a video timer (VTG-33, Fora, Tokyo, Japan). The frame rate of the recorder was set at 60 Hz.

Surface electrodes (P-00-S, Ambu, Ballerup, Denmark) were used in a bipolar derivation to record surface EMG activity on the left side of the following muscles: the anterior deltoid, erector spinae at the point 2.5 cm lateral to the midpoint between the spinous processes of L3 and L4, the long head of biceps femoris at the midpoint between the ischial tuberosity and head of the fibula, the gastrocnemius medialis, and the soleus. For the anterior deltoid, gastrocnemius, and soleus, electrodes were positioned at the midpoint of the muscle belly. The electrodes were aligned along the long axis of the muscle with an interelectrode distance of approximately 3 cm. A ground electrode was placed over the lateral malleolus of the tested side. All electrodes were fixed after shaving and cleaning the skin with alcohol. The electrode input impedance was reduced to <5 kΩ. Signals from the electrodes were amplified (×5,000) and band-pass filtered (5-500 Hz) with an EMG amplifier (6R12, NEC-Sanei, Tokyo, Japan).

Arm acceleration was recorded using a miniature unidirectional accelerometer (AS-2GB, Kyowa, Tokyo, Japan), which was taped to the dorsal surface of the left wrist so the axis of sensitivity was along the sagittal plane.

The electrical signals of CoPy, EMGs, and the LED on-signal were recorded on a computer (TYPE2881-6NJ, TBM, New York, NY, USA) via an A/D converter (ADA16-32/2(CB)F, Contec, Osaka, Japan) with a 1000-Hz sampling rate and 16-bit resolution.

Questionnaire on Heel-raise Training

Using a questionnaire, subjects in the training group were asked to report the place, time and day of training, occurrence of training-associated injuries or problems, and difficulties with the training.

Evaluating Parameters

All data were analyzed blindly to the condition and training status. Two video images in the sagittal plane at 150 milliseconds before the start of arm movement and at the end were printed using motion analyzing software (Frame-DIAS, DKH, Tokyo, Japan). Each of the printed video image was set on the digitizer (K-519mr2, Logitec, Tokyo, Japan), where the x- and y-coordinates of reflective markers on the video image were identified at 0.1-mm resolution. Previous experiments have shown in healthy young and elderly subjects that the movement angle of the knee between these 2 time points (the start and end of arm movement) is slight (1). Therefore, because the leg could be considered a functional unit, the postural movement pattern was analyzed in terms of ankle movement angle (leg movement relative to the foot) and hip movement angle (trunk movement relative to the leg). Thus, ankle angle (greater trochanter-lateral malleoulus-fifth metatarsal head) and hip angle (C7 spinous process-greater trochanter-lateral malleolus) were measured from x- and y-coordinates of each reflective marker on the video image. The differences in each angle between the 2 time points were calculated and defined as the ankle and hip movement angles. A movement angle of 0° indicated that the angles calculated before and after arm movement were identical. When the ankle and hip angles indicated dorsiflexion and flexion, respectively, the movement angles were considered positive.

Subsequent analyses were performed using BIMUTAS-II software (Kissei Comtec, Matsumoto, Japan). The time course of EMG bursts of the focal and postural muscles in each trial was analyzed as described below. To exclude electrocardiographic and movement artifacts, all EMGs and the LED on-signal were 40-Hz high-pass filtered using the seventh-order Butterworth method and then full-wave rectified. The background activity of the anterior deltoid was extremely small because the subjects' arms were hanging by their sides with elbows extended. Therefore, the burst onset of the anterior deltoid was identified by visual inspection of EMGs. In the simple-reaction condition, the time difference between the LED signal onset and burst onset of the anterior deltoid was measured as the anterior deltoid reaction time.

Next, mean amplitude and SD for the background activity of postural muscles was calculated for the period from −300 to −150 milliseconds with respect to the burst onset of the anterior deltoid. Visual inspection identified that the envelope line of each postural muscle activity deviated more than the mean + 2SDs from the background activity for at least 50 milliseconds, within −150 to +100 milliseconds with respect to burst onset of the anterior deltoid. The burst onset of the postural muscle was defined as the time point at which the EMG wave included in the envelope line first deviated more than the mean + 2SDs from the background activity. The burst onset timing of the postural muscle was negative when the onset of the postural muscle preceded that of the anterior deltoid.

Mean CoPy positions were calculated from −300 to −150 milliseconds with respect to burst onset of the anterior deltoid and from 0 to +150 milliseconds with respect to the end of arm movement. The difference of these mean positions was defined as the CoPy displacement. The end of arm movement was specified as the end point of second burst activity of the anterior deltoid, referring to video images and the acceleration curve of the wrist. The end point was defined as the end of the second EMG burst wave included in the envelope line that first deviated lower than the mean + 2SDs just before 2 seconds of the arm lowering. Furthermore, CoPy fluctuation was calculated as SD of CoPy for 2 seconds after the end of arm movement. For calculation of CoPy fluctuation, CoPy data that were resampled at 20 Hz and smoothed using the weighted 5-point moving average method were used.

Statistical Analyses

Shapiro-Wilks and Levene's tests confirmed that all data satisfied the assumption of normality and equal variance. Two-way analysis of variance (ANOVA) was used to assess the effects of group (training and control) and experiment date (before and after the training period) on physical characteristics, plantar flexor strength, muscle thicknesses of the soleus and gastrocnemius, and anterior deltoid reaction times. When a significant interaction between these effects was shown, a paired t-test and a nonpaired t-test were used to assess the differences between experiment dates and between groups, respectively. Four-way ANOVA was used to assess the effects of group, experiment date, condition (simple-reaction and own-timing) and muscle on onset timing of postural muscles (erector spinae, biceps femoris, gastrocnemius medialis, and soleus). When a significant main effect of a muscle was shown, a post-hoc multiple-comparison analysis using Tukey honestly significant difference (HSD) test was used to assess the differences among muscles. A 1-tailed t-test was used to assess whether burst onset of postural muscles significantly differed from that of anterior deltoid. A 3-way ANOVA was used to assess the effects of group, experiment date, and condition on ankle and hip movement angles, CoPy displacement, and fluctuation. The significance level was set at p ≤ 0.05. All statistical analyses were performed using SPSS 14.0J (SPSS Japan, Tokyo, Japan).

Results

On height, weight, and FL, no significant main effects and interaction in the group and experiment date were found (Table 1).

T1-33
Table 1:
Means and SDs of height, weight, and foot length.

The mean training period in the training group was 59 days (SD = 6). Most subjects performed the training, while at home in the kitchen (69%) or in the living room (38%). Training was performed during the morning (85%), afternoon (31%), and at night (46%). The training period was divided into 3 periods (early: 1-20 days, middle: 21-40 days, later: 41-62 days). Of the 13 subjects, 6 (46%) felt that the training was easy to perform throughout the entire training period. Difficulty performing the training was reported by only 6 subjects (46%) in the early training period and 1 subject (8%) in the middle training period. No subjects reported difficulties lasting throughout the entire training period, and no injuries or disorders were reported.

Plantar flexor strength and muscle thicknesses are shown in Table 2. Significant interactions between group and experiment date were found in plantar flexor strength (F[1, 24] = 37.1, p < 0.001) and soleus thickness (F[1, 24] = 13.7, p < 0.01). No significant differences in these values were found between the training and control groups before the training period. In the training group, plantar flexor strength (by 32% [SD = 20]) and soleus thickness were significantly greater after the training period (t[12] > 3.8, p < 0.01). No significant main effects and interaction in the group and experiment date were found in gastrocnemius thickness.

T2-33
Table 2:
Mean and SDs of plantar flexor strength and muscle thicknesses of gastrocnemius and soleus.

A significant interaction between group and experiment date was found with regard to anterior deltoid reaction time (F[1, 24] = 8.0, p < 0.01). No significant differences in anterior deltoid reaction time before the training period were found between training and control groups (training group: 183 milliseconds [SD = 28]; control group: 174 milliseconds [SD = 21]). In the training group, anterior deltoid reaction time was significantly shorter after the training period (time difference between before and after the training period, training group: −17 milliseconds [SD = 25]; control group: 9 milliseconds [SD = 22], t[12] = 2.4, p < 0.05).

Onset timings of postural muscles are shown in Figure 5. Significant differences were found with onset timing of postural muscles for condition (F[1, 96] = 235.8, p < 0.001) and muscle (F[3, 96] = 57.2, p < 0.001). For onset timing of postural muscles, significant interactions were found between group and experiment date (F[1, 96] = 6.9, p < 0.05), and also among group, experiment date and condition (F[1, 96] = 6.7, p < 0.05). Before the training period, no significant differences in onset timing of any postural muscles in both conditions were found between the training and control groups. In the training group, erector spinae onset timing in the simple-reaction condition and soleus onset timing in the own-timing condition were significantly earlier after the training period (t[12] > 2.6, p < 0.05). Onset timing of all postural muscles was earlier in the own-timing condition than in the simple-reaction condition (t[12] > 2.3, p < 0.05). Onset timing was significantly earlier in the following order: erector spinae < biceps femoris < gastrocnemius = soleus (p < 0.001). Onset timings of erector spinae in both conditions and biceps femoris in the own-timing condition were significantly earlier (t[12] < −2.3, p < 0.05), and those of the biceps femoris, gastrocnemius, and soleus in the simple-reaction condition were significantly later compared to the onset of the anterior deltoid (t[12] > 3.21, p < 0.01).

F5-33
Figure 5:
Means and SDs of onset timing of the erector spinae (ES), biceps femoris (BF), gastrocnemius (GcM), and soleus (SoL) with respect to the anterior deltoid in simple-reaction (A) and own-timing (B) conditions. *p < 0.05, **p < 0.01: significant difference in onset timing before and after the training period. ***p < 0.001: significant difference in onset timing between muscles. Significant difference between burst onsets of AD and those of each postural muscle. Onset timings of all postural muscles were earlier in the own-timing condition than the simple-reaction condition (p < 0.05).

The CoPy displacement and fluctuation are shown in Table 3. No significant main effects and interactions in group, experiment date, and condition were found on CoPy displacement. However, in CoPy fluctuation, a significant interaction among group, experiment date and condition was found (F[1, 24] = 5.2, p < 0.05). Before the training period, no significant differences in CoPy fluctuation in both conditions were found between training and control groups. In the training group, CoPy fluctuation in the own-timing condition was significantly smaller after the training period (t[12] = 3.5, p < 0.01).

T3-33
Table 3:
Means and SDs of displacement and fluctuation of CoPy.*

Ankle and hip movement angles are shown in Table 4. With ankle movement angle, a significant interaction between group and experiment date was found (F[1, 24] = 4.2, p < 0.05). Before the training period, no significant differences in ankle movement angle in both conditions were found between training and control groups. In the training group, ankle movement angle in the own-timing condition was significantly larger toward plantar flexional direction after the training period (t[12] = 4.1, p < 0.01). A significant main effect of condition was found on hip movement angle (F[1, 24] = 20.5 p < 0.001); the angle was significantly larger toward the extensional direction in the own-timing condition than in the simple-reaction condition.

T4-33
Table 4:
Mean and SDs of movement angle of ankle and hip.*†

Discussion

Subjects in this study were representative of the general elderly population in Japan (1,11), and there were no significant differences in all measurements between training and control groups before training. As the result of heel-raise training, plantar flexor strength, and soleus thickness markedly increased, postural control modality and stability improved in the own-timing condition.

In the training group, plantar flexor strength was increased by 32% after heel-raise training. Furthermore, all training subjects showed an increase in soleus thickness. Gastrocnemius thickness also tended to increase, although this difference did not reach a significant level. In the control group, no significant changes in those values were found between before and after the training period. These results suggest that the low-intensity training performed in this study was also effective in strengthening the triceps surae, particularly the soleus.

For calf-muscle training at more strength (50-70% of maximum) than set in this study, the increase ratio of muscle strength after 7 weeks of training was reported to be approximately 40%. On the contrary, no changes in muscle volume of the midthigh and upper arm were found (8,28). Thus, effects of muscle training and deterioration rate with age may be different between slow- and fast-twitch muscle fibers. Performing the heel-raise movement 100 times per day was relatively easy for half of all subjects in this study. The period of time during which training difficulty was subjectively experienced was the first 20 days. This suggests that the muscle load of the triceps surae may be <30% of maximum voluntary contraction (4). In repetitive muscle contraction, blood flow is a necessary factor for training slow-twitch muscle fibers because contraction of the fibers is closely related to aerobic metabolism (6,32). Such a low load predominantly activates slow-twitch muscle fibers (18). The significant enlargement of the soleus thickness in this study suggests that the low-intensity muscle training is especially effective in the soleus, because the trainability of slow-twitch muscle fibers would be higher than fast-twitch muscle fibers in the elderly.

In the control group, no significant differences in all measurements on postural control properties and stability were found between before and after the training period. Therefore, the undermentioned training effects would be caused by strengthening of plantar flexors, mainly the soleus. The effects on postural controllability should be different between simple-reaction and own-timing conditions.

In elderly subjects, postural movement angle during arm movement has been reported to be smaller at the ankle than at the hip (1,5). As described in the Introduction, these findings probably result from shifts in focus of muscular control from the entire body to the trunk because of age-related changes, such as deterioration of the leg muscles. In this study, ankle movement angle before the training period was not significantly different between the own-timing and simple-reaction conditions. However, hip movement angle was significantly larger toward the extensional direction in the own-timing condition than the simple-reaction condition, which would relate to deterioration of the leg muscles. After the training, the ankle movement angle was significantly larger toward plantar flexion in the own-timing condition. This increase may function as a backward shift of the center of gravity to prevent the postural disturbance associated with arm movement (7). Postural movements can be classified as either hip or ankle strategies (21). The present results suggest that the increase in strength of the triceps surae is related to an increase in the contribution of the lower limbs to postural control in the elderly. The increase in plantar flexor strength presumably enabled elderly subjects to effectively apply plantar flexion torque to the floor during arm flexion, thus altering postural movements of the ankle.

After the training in this study, the difference in activation timing of postural muscles with respect to the anterior deltoid became significantly earlier only for the soleus in the own-timing condition. Thus, heel-raise training was confirmed to have a significant effect on the soleus. Training of the transversus abdominis in patients with low back pain reportedly leads to earlier activation of the muscle during unilateral arm movement (39). One rationale for this change might be a lowering of the threshold for motor unit recruitment (16). These results suggest that training that focuses on muscles associated with deterioration of postural control could improve the controllability. This verification requires further investigation by examining different muscles for training. On the contrary, the activities of focal and postural muscles are considered to be closely synergistic, and controlled in the brain (10,36). However, the temporal relationship of 2 muscles is known to change according to the internal and external environment (7,14). The present results suggest that the temporal component, which constitutes the postural synergy, changes through the training of postural muscles.

The CoPy fluctuation after arm movement decreased after the training in the own-timing condition. This indicates that standing posture stability is improved by the muscle training. Stability of standing posture is reportedly influenced by the strength of muscles around the ankle joint (12). Moreover, postural responsiveness to an externally induced postural perturbation is reportedly related to the generation of torque around the ankle joint (19). These previous studies indicate the importance of muscle strength around the ankle joint to maintain postural equilibrium in a static and dynamic situation. One of the reported benefits of adopting the ankle strategy is that relatively smaller joint movements are required to make the intended shifts in the position of the center of mass (21). In this study, the ability to maintain postural equilibrium was presumably improved using an ankle strategy as a postural movement when the plantar flexors were strengthened.

In the simple-reaction condition, a significant shortening of the anterior deltoid reaction time and acceleration of activation onset timing of only the erector spinae were recognized after the training. Thus, the training in this study accelerated the activation timing of both the anterior deltoid and postural muscles with respect to the response stimulus. The effects of heel-raise training were possibly exerted not only on muscle plantar flexors strength but also integration function and fast information processing throughout the central nervous system (25). Prior activation of postural muscles with respect to the focal muscle has been reported to be clearly observed in the muscles that act to maintain equilibrium (10). In the simple-reaction condition, preparation for postural control is presumably insufficient because decreasing the reaction time was a main outcome. Because of this insufficiency, the effect of training appeared not in the soleus but in the erector spinae, which would be a target muscle for postural control in the own-timing condition.

In summary, the following training-related effects were demonstrated: (a) an increase in plantar flexor strength and in thickness of the soleus, (b) a shortening of the anterior deltoid reaction time in response to visual stimulation in the simple-reaction condition, (c) earlier activation of the erector spinae in the simple-reaction condition and the soleus in the own-timing condition, and (d) an increase in ankle movement in the own-timing condition and a decrease of CoPy fluctuation.

Practical Applications

In many previous studies on muscle training, muscle fiber composition has not been taken into consideration, and the strength in resistance muscle training for the elderly was considerably high (>50% of maximum, where fast-twitch muscle fibers are numerously recruited). Postural muscles contain an abundance of slow-twitch muscle fibers, and in this study low-intensity (<30% of maximum) muscle training was executed, focused on the soleus (which contains many slow-twitch muscle fibers and is a main postural muscle). The training intensity was set based on 100 repetitions of the movement. It was shown that with 100 heel raises a day for 2 months, planter flexion strength and soleus thickness within the triceps surae increased. Moreover, the properties of soleus activation and ankle movement in dynamic balance associated with bilateral arm movements were improved along with postural stability. Consequently, low-intensity heel-raise muscle training focused on the soleus improved postural control modality and stability that are effectively contributed to by the leg muscle. This training can be performed safely and easily by the elderly without any special equipment. These findings will be useful for fitness instructors and physical therapists to help the elderly perform low-intensity muscle training in any settings for dynamic balance improvement. It is possible that the trainability of muscles with a high proportion of slow-twitch muscle fibers would be higher than that with an abundance of fast-twitch muscle fibers. The effect of low-intensity muscle training will also be applicable to other postural muscles of the thigh and trunk in the elderly.

Acknowledgments

This study was supported by a Grant-in-Aid for Health Promotion for the Elderly from Kanazawa City Government.

References

1. Ai, J, Fujiwara, K, Asai, H, Koshida, K, Tomita, K, Kawahara, N, and Tomita, H. Postural movement during bilateral arm flexion and deformity of lumbar vertebral bodies related to chronic low back pain in elderly women. Health Behav Sci 6: 9-19, 2007.
2. Bandholm, T, Sonne-Holm, S, Thomsen, C, Bencke, J, Pedersen, SA, and Jensen, BR. Calf muscle volume estimates: Implications for botulinum toxin treatment? Pediatr Neurol 37: 263-269, 2007.
3. Belen'kii, VE, Gurfinkel, VS, and Pal'tsev, EI. Control elements of voluntary movements. Biofizika 12: 135-141, 1967.
4. Berger, RA and Hardage, B. Effect of maximum loads for each of ten repetitions on strength improvement. Res Q 38: 715-718, 1967.
5. Bleuse, S, Cassim, F, Blatt, JL, Labyt, E, Derambure, P, Guieu, JD, and Defebvre, L. Effect of age on anticipatory postural adjustments in unilateral arm movement. Gait Posture 24: 203-210, 2006.
6. Bonde-Petersen, F and Robertson, CH Jr. Blood flow in “red” and “white” calf muscles in cats during isometric and isotonic exercise. Acta Physiol Scand 112: 243-251, 1981.
7. Bouisset, S and Zattara, M. Anticipatory postural adjustments and dynamic asymmetry of voluntary movement. In: Stance and Motion: Facts and Concepts. Gurfinkel, VS, Ioffe, ME, Massion, J, and Roll, JP, eds. New York, NY: Plenum Press, 1988. pp. 177-183.
8. Brandon, LJ, Sharon, BF, Boyette, LW, Anderson, KA, and Stiles, R. Effects of resistive training on strength gains in older adults. J Nutr Health Aging 1: 114-119, 1997.
9. Cao, ZB, Maeda, A, Shima, N, Kurata, H, and Nishizono, H. The effect of a 12-week combined exercise intervention program on physical performance and gait kinematics in community-dwelling elderly women. J Physiol Anthropol 26: 325-332, 2007.
10. Cordo, PJ and Nashner, LM. Properties of postural adjustments associated with rapid arm movements. J Neurophysiol 47: 287-302, 1982.
11. Fujiwara, K, Asai, H, Toyama, H, Kunita, K, Yaguchi, C, Kiyota, N, Tomita, H, and Jacobs, JV. Changes in muscle thickness of gastrocnemius and soleus associated with age and sex. Aging Clin Exp Res 22: 24-30, 2010.
12. Fujiwara, K, Ikegami, H, Okada, M, and Koyama, Y. Contribution of age and muscle strength of lower limbs to steadiness and stability in standing posture. J Anthrop Soc Nippon 90: 385-400, 1982 (in Japanese with English abstract).
13. Fujiwara, K, Toyama, H, Asai, H, Maeda, K, and Yaguchi, C. Regular heel-raise training focused on the soleus for the elderly: Evaluation of muscle thickness by ultrasound. J Physiol Anthropol 29: 23-28, 2010.
14. Fujiwara, K, Toyama, H, and Kunita, K. Anticipatory activation of postural muscles associated with bilateral arm flexion in subjects with different quiet standing position. Gait Posture 17: 254-263, 2003.
15. Goshima, K. Studies on items and normal ranges in stabilometry. Equilibrium Res 45: 368-387, 1986 (in Japanese).
16. Griffin, L and Cafarelli, E. Resistance training: Cortical, spinal, and motor unit adaptations. Can J Appl Physiol 30: 328-340, 2005.
17. Gurfinkel, VS, Lipshits, MI, Mori, S, and Popov, KE. Postural reactions to the controlled sinusoidal displacement of the supporting platform. Agressologie 17: 71-76, 1976.
18. Henneman, E, Somjen, G, and Carpenter, DO. Excitability and inhibitability of motoneurons of different sizes. J Neurophysiol 28: 599-620, 1965.
19. Hess, JA, Woollacott, M, and Shivitz, N. Ankle force and rate of force production increase following high intensity strength training in frail older adults. Aging Clin Exp Res 18: 107-115, 2006.
20. Horak, FB, Esselman, P, Anderson, ME, and Lynch, MK. The effects of movement velocity, mass displaced, and task certainty on associated postural adjustments made by normal and hemiplegic individuals. J Neurol Neurosurg Psychiatry 47: 1020-1028, 1984.
21. Horak, FB and Nashner, LM. Central programming of postural movements: Adaptation to altered support-surface. J Neurophysiol 55: 1369-1381, 1986.
22. Horak, FB, Shupert, CL, and Mirka, A. Components of postural dyscontrol in the elderly: A review. Neurobiol Aging 10: 727-738, 1989.
23. Ikai, M, Ishii, M, and Nakamura, J. Muscle endurance (II). J Health Phys Ed Recrea 15: 281-287, 1965 (in Japanese).
24. Johnson, MA, Polgar, J, Weightman, D, and Appleton, D. Data on the distribution of fibre types in thirty-six human muscles. An autopsy study. J Neurol Sci 18: 111-129, 1973.
25. Kalapotharakos, VI, Michalopoulos, MM, Strimpakos, N, Diamantopoulos, K, and Tokmakidis, SP. Functional and neuromotor performance in older adults: Effect of 12 wks of aerobic exercise. Am J Phys Med Rehabil 85: 61-67, 2006.
26. Lee, WS, Cheung, WH, Qin, L, Tang, N, and Leung KS. Age-associated decrease of type IIA/B human skeletal muscle fibers. Clin Othop Relat Res 450: 231-237, 2006.
27. Lexell, J. Human aging, muscle mass, and fiber type composition. J Gerontol A Biol Sci Med Sci 50: 11-16, 1995.
28. Moritani, T and deVries, HA. Potential for gross muscle hypertrophy in older men. J Gerontol 35: 672-682, 1980.
29. Narici, M and Cerretelli, P. Changes in human muscle architecture in disuse atrophy evaluated by ultrasound imaging. J Gravit Physiol 51: P73-P74, 1998.
30. Okada, M. An electromyographic estimation of the relative muscular load in different human postures. J Hum Ergol 1: 75-93, 1973.
31. Persch, LN, Ugrinowitsch, C, Pereira, G, and Rodacki, AL. Strength training improves fall-related gait kinematics in the elderly: a randomized controlled trial. Clin Biomech (Bristol, Avon) 24: 819-825, 2009.
32. Petrofsky, JS, Phillips, CA, Sawka, MN, Hanpeter, D, and Strafford, D. Blood flow and metabolism during isometric contractions in cat skeletal muscle. J Appl Physiol 50: 493-502, 1981.
33. Pijnappels, M, van der Burg, PJ, Reeves, ND, and van Dieën, JH. Identification of elderly fallers by muscle strength measures. Eur J Appl Physiol 102: 585-592, 2008.
34. Ribeiro, F, Teixeira, F, Brochado, G, and Oliveira, J. Impact of low cost strength training of dorsi- and plantar flexors on balance and functional mobility in institutionalized elderly people. Geriatr Gerontol Int 9: 75-80, 2009.
35. Rogers, MW, Kukulka, CG, and Soderberg, GL. Age-related changes in postural responses preceding rapid self-paced and reaction time arm movements. J Gerontol 47: M159-M165, 1992.
36. Shen, X, Fujiwara, K, and Tomita, H. Anticipatory postural control associated with bilateral arm flexion and event-related potential in a Kanji Stroop-like task. Clin Neurophysiol 120: 827-833, 2009.
37. Slijper, H, Latash, ML, and Mordkoff, JT. Anticipatory postural adjustments under simple and choice reaction time conditions. Brain Res 924: 184-197, 2002.
38. Trappe, SW, Trappe, TA, Lee, GA, and Costill, DL. Calf muscle strength in humans. Int J Sports Med 22: 186-191, 2001.
39. Tsao, H and Hodges, PW. Persistence of improvements in postural strategies following motor control training in people with recurrent low back pain. J Electromyogr Kinesiol 18: 559-567, 2008.
40. Woollacott, MH and Manchester, DL. Anticipatory postural adjustments in older adults: Are changes in response characteristics due to changes in strategy? J Gerontol 48: M64-M70, 1993.
Keywords:

muscle strength; muscle thickness; postural control; standing stability; electromyogram

Copyright © 2011 by the National Strength & Conditioning Association.