- Email: [email protected]
Age-related deficit in dynamic stability control after forward falls is affected by muscle strength and tendon stiffness
Available online at www.sciencedirect.com
Journal of Electromyography and Kinesiology 18 (2008) 980–989 www.elsevier.com/locate/jelekin
Age-related deficit in dynamic stability control after forward falls is affected by muscle strength and tendon stiffness Kiros Karamanidis, Adamantios Arampatzis *, Lida Mademli Institute of Biomechanics and Orthopaedics, German Sport University of Cologne, Carl-Diem-Weg 6, 50933 Cologne, Germany Received 15 March 2007; received in revised form 27 April 2007; accepted 27 April 2007
Abstract The purpose of the work was to determine whether the age-related muscle weakness diminishes older adults’ ability to use mechanisms responsible for maintaining dynamic stability after forward falls. Nine older and nine younger adults participated in this study. To analyse the capacities of the leg-extensor muscle–tendon units, all subjects performed isometric maximal voluntary plantarflexion and knee extension contractions on a dynamometer. The elongation of the gastrocnemius medialis and the vastus lateralis tendon and aponeuroses during isometric contraction was examined by ultrasonography. Recovery behaviour was determined after a sudden fall from two forward-inclined lean angles. Compared to older adults, younger adults had higher muscle strength and tendon stiffness. Younger adults created a higher margin of stability compared to older, independent of perturbation intensity. The main mechanism improving the margin of dynamic stability was the increase of the base of support. The results, further, demonstrated that the locomotion strategy employed before touchdown affects the stability of the stance phase and that muscle strength and tendon stiffness contributed significantly to stability control. We concluded that, to reduce the risk of falls, older individuals may benefit from muscle–tendon unit strengthening programs as well as from interventions exercising the mechanisms responsible for dynamic stability. 2007 Elsevier Ltd. All rights reserved. Keywords: Margin of stability; Balance recovery; Base of support; Human
1. Introduction Falls are a major cause of injury among elderly people, including serious hip and wrist fractures (Baker and Harvey, 1985; Cumming and Klineberg, 1994). Approximately one third of adults over 65 years of age fall at least once a year (Blake et al., 1988). Most of these falls occur after a loss of stability in a forward direction such as tripping while walking (Blake et al., 1988). Identifying the mechanisms by which stability deficits related to forward falls occur within the elderly population may be of importance for the development of effective exercise interventions for fall prevention. From a mechanical point of view, the margin of stability during locomotion can be quantified by the position of the extrapolated centre of mass (CM; deter*
Corresponding author. Tel.: +49 221 4982 5620; fax: +49 221 4971598. E-mail address: [email protected] (A. Arampatzis).
1050-6411/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.jelekin.2007.04.003
mined by the position and velocity of the CM) in relation to the base of support (Hof et al., 2005). A loss of stability in the forward direction apparently occurs when the position of the extrapolated CM exceeds the anterior boundary of the base of support (Hof et al., 2005). In such an unstable body position, the neuro-motor system must select and execute postural corrections including mechanisms responsible for the control of dynamic stability in order to prevent any unintended forward fall. Hof (2007) recently presented three mechanisms by which stability may be maintained: (a) by increasing the base of support in relation to the extrapolated CM, (b) by counter-rotating segments around the CM, and (c) by applying an external force, other than the ground reaction force (e.g. grasping). Sudden release from a forward-inclined body position is commonly used to examine humans’ capabilities of regaining stability in the anterior direction (Hsiao and Robinovitch, 1999; Thelen et al., 1997; Wojcik et al., 2001). Such an
K. Karamanidis et al. / Journal of Electromyography and Kinesiology 18 (2008) 980–989
experimental design has the benefit that the magnitude of postural disturbance can be controlled and easily varied (Hsiao and Robinovitch, 1999). When released from an inclined body position it is observed that the inability to regain balance is greater among older adults than among younger adults (Thelen et al., 1997; Wojcik et al., 1999; Wojcik et al., 2001). Several studies (Mackey and Robinovitch, 2006; Pijnappels et al., 2005a,b) have suggested that the deterioration of the musculoskeletal function in the elderly may play a key role for the observed age-related deficits in stability control. Simulation models showed that a lower magnitude (Hsiao and Robinovitch, 1999) and rate (Simoneau and Corbeil, 2005) of moment generation at the ankle joint clearly reduces a person’s capacity to regain stability. Recently, it has been reported (Karamanidis and Arampatzis, 2007) that deficits in balance recovery by stepping in older adults are associated with the lower muscle strength and tendon stiffness in their leg-extensor muscle– tendon units (MTUs). The authors (Karamanidis and Arampatzis, 2007) concluded that the age-related degeneration of the MTUs diminishes older adults’ ability to restore balance with a single-step in the anterior direction. However, a faithful analysis of the dynamic stability has not been included and, therefore, it is difficult to identify the exact mechanisms responsible for the reported findings. Postural correction after a sudden perturbation depends on actions of both the support limb during the push-off phase and the recovery limb during the step execution phase (Karamanidis and Arampatzis, 2007; Pijnappels et al., 2005b). The maximal moments generated at the support limb during the push-off phase (e.g. at the ankle joint up to 200 N m) (Pijnappels et al., 2005b,c) as well as at the recovery limb during the step execution phase (e.g. at the knee joint up to 100 N m) (Wojcik et al., 2001) suggest that the muscle strength and tendon stiffness of the lower extremities may be relevant for performing successful postural corrections after a sudden perturbation. We can conclude that a reasonable challenge of the human system after a sudden perturbation is to perform effective balance responses by selecting motor plans which include mechanisms responsible for maintaining dynamic stability. Although there are several studies in the literature examining balance corrections after anterior postural disturbance (Karamanidis and Arampatzis, 2007; Madigan and Lloyd, 2005; Thelen et al., 1997; Wojcik et al., 1999) we did not find any study determining in a sophisticated way the components of the dynamic stability in younger and older adults after induced forward falls and relating them to the MTUs’ capacities. Combining the above observations, we hypothesised that, compared to younger adults, older adults would show deficits in using mechanisms responsible for maintaining dynamic stability after an induced forward fall, reflecting the lower capacities of their MTUs in the lower extremities. Therefore, the aims of the current work were to examine the control of dynamic stability in younger and older adults after a sudden induced perturbation from a fixed forward-
981
inclined position using a small and a large body lean angle and, hence, varying the intensity of disturbance. Further, we intended to determine whether the age-related degeneration of leg-extensor MTU capacities diminishes older adults’ ability to use mechanisms responsible for maintaining dynamic stability after a sudden perturbation. 2. Materials and methods 2.1. Experimental design The investigation was conducted with nine male older adults aged 60–68 years and nine male younger adults aged 22–32 years. All subjects were experienced endurance runners and performed running commonly 4–7 times per week over the last 10 years and participated regularly in middle- and long-distance running competitions. The training distance at the time of the study ranged from 40 to 100 km per week. Approval was obtained from the university committee for the protection of human subjects and informed consent was given by all subjects. None of the chosen test subjects had a history of neuromuscular or musculoskeletal impairments. The experimental design of the recovery task has been previously reported in detail (Karamanidis and Arampatzis, 2007). Briefly, the subjects were released suddenly without warning from a fixed forward-inclined position (see also: Thelen et al., 1997; Wojcik et al., 2001). The participants were initially maintained in an inclined forward posture by a horizontal inextensible cable, attached at one end to a belt worn by the participants around the pelvis and at the other end to a custom-built pneumatic release system. Subjects were instructed and encouraged to restore balance by taking a single-step after forward fall was initiated. All participants wore a full trunk safety harness suspended from an overhead track that allowed for forward and lateral motion while preventing contact of any body part, other than the feet, with the ground. The angle of the forward lean was controlled by adjusting the lean control cable length until the load cell attached to the cable indicated that it supported a specified percentage of the subject’s body weight (BW). In the first set of trials (small lean angle), the subjects attempted to recover balance following three releases at a lean-control cable load of 23 ± 3% BW. In the second set of trials (large lean angle), three trials at a lean-control cable load of 33 ± 3% BW were carried out. The examined lean-angle cable loads correspond to lean angles (defined as the angle between the vertical in the sagittal plane and the line connecting the CM and the midpoint of the foot) of 19.7 ± 2.4 and 26.0 ± 2.7. Specific lean-angle cable loads were chosen because it has been shown that there is a clear reduction in the ability to recover balance with a single-step from lean-control cable loads of about 20% to loads of more than 30% BW for older individuals (Madigan and Lloyd, 2005; Thelen et al., 1997). 2.2. Measurement of recovery mechanics At the initiation of each trial, the subject stood barefoot on a force plate (60 cm · 90 cm, Kistler, Winterthur, Switzerland). The first step always landed on a second force plate (60 cm · 90 cm, Kistler, Winterthur, Switzerland) mounted in front of the initial force plate. The ground reaction forces were collected at a sampling rate of 1080 Hz. To determine the onset of release, the pneumatic release system triggered a TTL signal at the instant of release which was simultaneously captured as an analogue signal by the Vicon motion capture system (1080 Hz; Model 624, Vicon,
982
K. Karamanidis et al. / Journal of Electromyography and Kinesiology 18 (2008) 980–989
Oxford, United Kingdom). Kinematic data were recorded with 12 Vicon cameras operating at 120 Hz. Thirty-eight reflective markers (radius 14 mm) were used to track the whole body kinematics. The markers defined the left and right foot, left and right lower leg, left and right thigh, pelvis, thorax, left and right upper arm, left and right forearm, left and right hand and the head. The segmental masses and the location of the segment centres of masses were calculated based on the data reported by Dempster et al. (1959). Sagittal angular joint angle kinematics at the knee were analysed for the recovery limb. A completely extended knee was defined as 180 knee joint angle. For each trial, three time points were identified (Fig. 1): (A) subject’s release (determined by the analogue signal which the Vicon motion capture system received from the pneumatic release system), (B) touchdown of the recovery limb (determined by the ground reaction force data – threshold level 20 N) and (C) first minimum at the knee joint angle of the recovery limb after touchdown. Time point (C) indicated termination of downwards motion of the body (i.e. vertical CM velocity almost zero; Fig. 1). Further, recovery stepping task was divided into two phases based on the identified three time points (Fig. 1): (1) phase until touchdown (defined from the subject’s release until touchdown of the recovery limb) and (2) main stance phase (defined from touchdown until the first minimum at the knee joint angle of the recovery limb). For better illustration, the curves in Fig. 1 end 400 ms after touchdown of the recovery limb (stance phase). The margin of stability in the anteroposterior direction was calculated according to Hof et al. (2005) as follows: bx ¼ U x max X CM
B
A
Release
D
C
TD
KJAmin
400 ms after TD
200 180
0.4 KJA
VzCOM
where bx is margin of stability in the anteroposterior direction, Uxmax is anterior boundary of the base of support (i.e. horizontal component of the vertical projection of the toe from the recovery limb to the ground; a zero value represents the position of the toe before release), XCM is extrapolated CM in the anteroposterior V ffiffiffiffiffi Þ, P X CM is horizontal (anteropostedirection ðX CM ¼ P X CM þ pX CM g=l
rior) component of the vertical projection of the CM to the ground, V X CM is horizontal (anteroposterior) CM velocity, g is acceleration of gravity, and l is distance between CM and centre of the ankle joint in the sagittal plane. Margin of stability was determined in the anteroposterior direction because, after a forward fall, both the extrapolated CM as well as the base of support mainly shifts in the anterior direction. To illustrate this, Fig. 2 provides the anteroposterior boundary of the base of support, horizontal component of the vertical projection of the CM to the ground and the position of the extrapolated CM during the recovery task for both a singleand a multiple-stepper. Postural stability is maintained in circumstances where the position of the extrapolated CM is within the base of support (positive values of margin of stability) while stability is lost in cases where the extrapolated CM passes the anterior boundary of the base of support during stance phase (negative values of margin of stability). To examine the conformity between the margin of stability calculated by the model presented by Hof et al. (2005) and the real behaviour of the human system during forward falls we compared the margin of stability at the end of the main stance phase (i.e. positive vs. negative values) with subjects recovery strategy (i.e. single-step vs. multiple-step recovery strategy). A trial was defined as single-step behaviour if only one step was taken or if the anterior displacement of the second step (contralateral limb) did not exceed the anterior displacement of the recovery limb. A multiple-step behaviour was defined when the subject took a second step of the recovery limb of any kind or when the subject took a contralateral step whose anterior displacement exceeded that of the recovery limb (for exact definition see Karamanidis and Arampatzis, 2007). We found that in about 96% of the cases the margin of stability at the end of the main stance phase was in line with the recovery strategy of the participants (negative margin of stability value associated to multiple-step recovery strategy and vice versa).
0.0 140
[m/s]
[Degrees]
160
120 100
-0.4
80 Main stance phase
60 -100
Phase until touchdown
-50
Stance phase
0
5
-0.8 100
[%] Fig. 1. Analysed time points and time intervals, and angular motion at the knee joint (KJA) and vertical centre of mass velocity (VzCOM) of one subject. A: subjects release; B: touchdown (TD) of the recovery limb; C: first minimum at the KJA of the recovery limb after touchdown (KJAmin); D: 400 ms after touchdown. Phase until touchdown: from release until TD (x-axis: normalized from 100% to 0%); Stance phase: from TD until 400 ms after TD (x-axis: normalized from 0% to 100%); Main stance phase: from TD until KJAmin. A completely extended knee was defined as 180 knee joint angles. Note that time point KJAmin provided termination of downwards motion of the body (i.e. vertical COM velocity almost zero).
2.3. Measurement of muscle strength and tendon elongation To examine muscle strength potential and tendon stiffness of the leg-extensor MTU’s all subjects performed isometric maximal voluntary ankle plantarflexion and knee extension contractions on a dynamometer (Biodex Medical Systems. Inc., Shirley, NY, USA). The method for calculating the resultant joint moments has been previously described in detailed (Arampatzis et al., 2004; Arampatzis et al., 2005a). Axis misalignment between dynamometer and ankle or knee joint during contraction was taken into consideration using 12 Vicon cameras operating at 120 Hz. The effect of moments arising from antagonistic muscles during the ankle plantarflexion and knee extension efforts on the resultant joint moment was taken into account using the method also described previously (Baratta et al., 1988; Mademli et al., 2004). Therefore, in the following text maximal knee and ankle joint moment refer to the maximal joint moment values where the effect of the joint axis alteration relative to the dynamometer axis, gravitational effects and the effect of antagonist moment on the measured moment were taken into account.
K. Karamanidis et al. / Journal of Electromyography and Kinesiology 18 (2008) 980–989 TD
Release
1.6 1 .6
1.6
boundary BS projection CM extrapolated XCM
[m]
1.2
0.8
0.8
0.4
0.4
a 0
200
400
600
800
1000
KJA min
boundary BS projection CM extrapolated XCM
margin of stability
1.2
margin of stability
0.0
TD
Release
KJA min
983
b
0.0
1200
0
200
400
600
800
1000
[ms]
[ms]
Fig. 2. Anteroposterior boundary of the base of support (boundary BS), horizontal component of the vertical projection of the centre of mass to the ground (projection CM) and the position of the extrapolated centre of mass (extrapolated XCM) during the recovery task of a single- (a) and multiplestepper (b). Margin of stability is the instantaneous difference between the boundary BS and the extrapolated XCM. Postural stability is maintained in circumstances where the extrapolated XCM is within the boundary BS ((a) positive values of margin of stability) while stability is lost in cases where the extrapolated XCM passes the anterior boundary BS during stance phase ((b) negative values of margin of stability). The zero value in the vertical axis represents the initial position (before release) of the horizontal component of the vertical projection of the toe from the recovery limb to the ground. TD, touchdown; KJAmin, first minimum at the knee joint angle of the recovery limb after TD.
The exact procedure for the measurement of the elongation of the tendon and aponeurosis during contraction has also been described earlier (Arampatzis et al., 2005b; Stafilidis et al., 2005). Briefly, to measure the elongation of the triceps surae and quadriceps femoris tendon and aponeurosis during contraction, the gastrocnemius medialis and the vastus lateralis and their distal aponeurosis were visualized by ultrasonography (Aloka SSD 4000, Tokyo, Japan, 43 Hz). The effect of inevitable joint angular rotation on the measured elongation of the gastrocnemius medialis or vastus lateralis tendon and aponeurosis during contraction was taken into account by capturing the motion of the tendon and aponeurosis from the corresponding muscle during a passive (relaxed condition) rotation of the ankle or knee joint (Arampatzis et al., 2005b; Stafilidis et al., 2005). The strain values of the triceps surae and quadriceps femoris tendon and aponeurosis during MVC were calculated at every 100 N and 200 N and at maximal calculated tendon force respectively; corresponding tendon moment arms were taken from Maganaris et al. (1998), and Herzog and Read (1993). To compare the stiffness of triceps surae and quadriceps femoris tendon and aponeurosis between older and younger adults we calculated the slope of the calculated tendon force versus strain of the tendon and aponeurosis (normalized stiffness) between 50% and 100% of the maximal tendon force by means of linear regression. 2.4. Statistics A t-test for independent samples was used in order to check for differences in isometric maximal voluntary ankle plantarflexion moment and knee extension moment, stiffness and strain values (every 100 N and 200 N for the triceps surae and quadriceps femoris MTUs, respectively) of the triceps surae and quadriceps femoris tendon and aponeurosis between older and younger adults. For the analysis of the recovery task the mean values from the three trials were utilised for the comparison between age (younger vs. older adults) and lean angle (small vs. large lean angle) groups. A two-factor (age · lean angle) analysis of variance
(ANOVA) was used in order to detect group differences in the analysed parameters of the dynamic stability. The level of significance was set at a = 0.05. When a significant age-by-lean angle interaction was present a post hoc test (Bonferroni) was conducted in order to determine where these differences occurred. To examine relationships between parameters of dynamic stability we used the Pearson correlation coefficient. Further, we used Pearson correlation coefficient and a multiple regression analysis to identify the relationship between leg-extensor MTU’s capacities and margin of stability at the end of the main stance phase for both, the small (n = 18) and large (n = 18) lean angle. All results in the figures are presented as mean and standard error of mean (mean and SEM), whereas in the text and tables they are expressed as mean and standard deviation (mean and SD).
3. Result 3.1. Muscle strength and tendon stiffness Older adults showed significantly (P < 0.05) lower maximal isometric ankle plantarflexion and knee extension moments (Table 1). At the triceps surae MTU, the strain Table 1 Anthropometric data, maximal joint moments and normalized tendonaponeurosis stiffness of the triceps surae (TS) and quadriceps femoris (QF) muscle–tendon units for the older and younger adults (means ± SD)
Age [yr] Body mass [kg] Body height [cm]* Momentmax. ankle joint [N m/kg]* StiffnessTS tendon [kN/Strain] Momentmax. knee joint [N m/kg]* StiffnessQF tendon [kN/Strain]* *
Old (n = 9)
Young (n = 9)
64 ± 3 75 ± 7 176 ± 3 1.25 ± 0.32 25.95 ± 10.48 2.00 ± 0.41 46.12 ± 15.47
27 ± 4 73 ± 6 181 ± 4 1.78 ± 0.35 27.00 ± 11.40 2.51 ± 0.48 57.87 ± 10.55
Statistically significant age effect (P < 0.05).
984
K. Karamanidis et al. / Journal of Electromyography and Kinesiology 18 (2008) 980–989
values of the tendon and aponeurosis for a given tendon force (every 100 N) showed no significant differences (P > 0.05) between the two groups (Fig. 3). At the quadriceps femoris MTU, however, the strain values of the tendon and aponeurosis for a given tendon force (every 200 N) was significantly higher (P < 0.05) for the older compared to the younger adults (Fig. 3). In the same manner, the normalized tendon stiffness showed statistically significant (P < 0.05) lower values for the older adults at the quadriceps femoris MTU whereas no significant (P > 0.05) differences between younger and older adults were found at the triceps surae MTU (Table 1). Body height was significantly lower (P < 0.05) for the older compared to the younger adults (on average about 5 cm; Table 1).
gin of stability at the instants of touchdown for the large compared to the small lean angle (Table 2 and Fig. 4). The comparison between older and younger adults revealed significant (P < 0.05) differences in the anterior boundary of the base of support with lower values for the older adults (Table 2). There were no significant (P > 0.05) differences between older and younger adults in the horizontal component of the vertical projection of the CM to the ground, horizontal CM velocity, or the term pffiffiffiffiffiffiffi g=l and, therefore, on the position of the extrapolated CM at touchdown (Table 2). As a result of the above findings (reduced boundary of the base of support and similar position of the extrapolated CM) the older compared to the younger adults showed a significantly (P < 0.05) lower margin of stability at touchdown (Fig. 4).
3.2. Stability at touchdown 3.3. Stability at the end of the main stance phase The duration until touchdown was significantly (P < 0.05) lower for the large compared to the small lean angle (Table 2). The boundary of the base of support at touchdown, the horizontal component of the vertical projection of the CMp toffiffiffiffiffiffiffi the ground, the horizontal CM velocity and the term g=l at the instants of touchdown were significantly (P < 0.05) higher for the large compared to the small lean angle (Table 2). The knee joint angle showed significantly (P < 0.05) smaller values (more flexed knee joint) for the large compared to the small lean angle at touchdown (old and young adults at large lean angle 137.5 ± 6.2 and 136.3 ± 10.4, respectively; at small lean angle 148.1 ± 8.6 and 139.2 ± 8.7, respectively). There was a significant correlation (r = 0.588, P < 0.001, n = 36) between knee angle and l (distance between CM and centre of the ankle joint) at touchdown. This shows that the pffiffiffiffiffiffiffi higher value for the term g=l at touchdown for the large compared to the small lean angle was partly related to the knee joint angle (more flexed knee joint for the large lean angle). Further, we found a significantly (P < 0.05) more anterior position of the extrapolated CM and a lower mar-
For the large compared to the small lean angle we found a significantly (P < 0.05) more anterior boundary of the base of support, more horizontal component of the vertical pffiffiffiffiffiffiffi projection of the CM to the ground, higher term g=l and a more anterior position of the extrapolated CM at the end of the main stance phase (Table 3). Further, the knee joint angle at the end of the main stance phase showed significantly (P < 0.05) lower values (more flexed knee joint) for the large compared to the small lean angle (Table 4). As for the instants of touchdown, we found a significant correlation between knee angle and l at the end of the main stance phase (r = 0.665, P < p 0.001, ffiffiffiffiffiffiffi n = 36), showing that the higher value for the term g=l for the large lean angle compared to the small lean angle was partly related to the more flexed knee joint for the large lean angle. The margin of stability at the end of the main stance phase was not significantly (P > 0.05) different between the small and large lean angles (Fig. 4). The comparison between older and younger adults revealed a significantly (P < 0.05) lower boundary of the base of support, higher horizontal CM
Triceps suraeMTU
8
Old (n = 9) Young (n = 9)
8
6
Strain [%]
Quadriceps femorisMTU
10
Old (n = 9) Young (n = 9)
6
*
4
*
*
4 2
*
** * *
*
2
a
0 0
500
1000
Tendon force [N]
1500
2000
b
0 0
1000
2000
3000
4000
Tendon force [N]
Fig. 3. Strain values at every 100 N and 200 N and at maximal calculated tendon force of the triceps surae (a) and quadriceps femoris (b) tendon and aponeurosis during isometric maximal voluntary contraction, respectively (means ± SEM). MTU, muscle–tendon unit; *statistically significant age effect (P < 0.05).
K. Karamanidis et al. / Journal of Electromyography and Kinesiology 18 (2008) 980–989 Table 2 Duration until touchdown, anterior boundary of the base of support (Boundary BS), horizontal component of the vertical projection of the centre of mass to the ground (Projection pffiffiffiffiffiffiffi CM), horizontal centre of mass velocity (Horizontal VCM), term g=l (l: distance between the centre of mass and centre of the ankle joint; g: acceleration of gravity) and position of the extrapolated centre of mass (Extrapolated XCM) at the instants of touchdown for the different groups (means ± SD)
Duration until touchdown [ms]^ Boundary BS [cm]*, ^ Projection CM [cm]^ Horizontal VCM [m/s]^ Term pffiffiffiffiffiffiffi g=l½1=s? Extrapolated XCM [cm]^
Large lean angle
Small lean angle
Large lean angle
Old (n = 9)
Young (n = 9)
Old (n = 9)
Old (n = 9)
Young (n = 9)
Old (n = 9)
523 ± 31
535 ± 68
472 ± 26
469 ± 36
85.0 ± 14.6
92.4 ± 15.7
91.2 ± 15.3
110.6 ± 13.3
65.4 ± 5.1
63.1 ± 14.6
78.3 ± 5.8
82.8 ± 12.0
83.3 ± 13.1
89.6 ± 16.3
90.2 ± 13.9
106.8 ± 14.2
0.89 ± 0.35
0.69 ± 0.28
1.25 ± 0.38
0.59 ± 0.28
53.4 ± 3.2
50.3 ± 11.1
67.8 ± 3.4
66.1 ± 8.3
3.55 ± 0.11
3.63 ± 0.13
3.69 ± 0.08
3.76 ± 0.15
1.09 ± 0.13
1.06 ± 0.27
1.36 ± 0.13
1.27 ± 0.21
90.8 ± 12.0
82.1 ± 20.8
3.49 ± 0.07
3.49 ± 0.07
3.64 ± 0.06
3.59 ± 0.12
84.5 ± 5.6
80.6 ± 18.2
105.0 ± 5.8
Young (n = 9)
101.5 ± 13.0
Statistically significant age effect (P < 0.05). Statistically significant lean angle effect (P < 0.05).
Margin of stability [cm]
40
Small lean angle
Small lean angle
Old adults (n=9)
Young adults (n=9)
Boundary BS [cm]*, ^ Projection CM [cm]^ Horizontal VCM [m/s]*, + Term pffiffiffiffiffiffiffi g=l½1=s? Extrapolated XCM [cm]*, ^
0
98.4 ± 16.8
End main stance phase was defined as the first minimum at the knee joint angle of the recovery limb after touchdown (i.e. termination of downwards motion of the body). * Statistically significant age effect (P < 0.05). ^ Statistically significant lean angle effect (P < 0.05). + Statistically significant age-by-lean angle interaction (P < 0.05). The post hoc analysis revealed significantly (P < 0.05) higher Horizontal VCMvalues for the older adults at the large lean angle compared to all other groups (younger adults at the large and small lean angle, and older adults at the small lean angle).
-20 -40
40
112.3 ± 9.7
Young (n = 9)
Table 4 Minimum of the knee joint angle of the recovery limb after touchdown (Knee anglemin), duration of the main stance phase (Time to knee anglemin) and average values (over main stance phase) of the horizontal ground reaction force for the different groups (means ± SD)
20
i
-60
Margin of stability [cm]
Table 3 Anterior boundary of the base of support (Boundary BS), horizontal component of the vertical projection of the centre of mass to the ground (Projection pffiffiffiffiffiffiffi CM), horizontal centre of mass velocity (Horizontal VCM), the term g=l (l: distance between the centre of mass and the centre of the ankle joint; g: acceleration of gravity) and position of the extrapolated centre of mass (extrapolated XCM) at the end of the main stance phase for the different groups (means ± SD)
Small lean angle
* ^
985
Small lean angle
Large lean angle
Old (n = 9)
Old (n = 9)
Young (n = 9)
Young (n = 9)
Knee anglemin []*, 131.8 ± 14.0 119.3 ± 10.1 126.3 ± 10.0 110.1 ± 8.1 ^
Large lean angle
Large lean angle
Old adults (n=9)
Young adults (n=9)
20 0
-40
ii TD
END
bTD =
*, ⊥
173 ± 74
193 ± 58
123 ± 63
224 ± 56
2.02 ± 1.53 2.72 ± 0.61 1.71 ± 2.26 3.16 ± 0.56
Main stance phase was defined from touchdown until minimum knee joint angle of the recovery limb. A completely extended knee was defined as 180 knee joint angle. Posterior-directed forces were defined as negative force values. * Statistically significant age effect (P < 0.05). ^ Statistically significant lean angle effect (P < 0.05).
-20
-60
Time to knee anglemin [ms]* Average forcehorizontal [N/kg]*
TD
END
bEND = *
Fig. 4. Individual and mean (SEM) margin of stability values (b) for the older (left) and younger (right) adults at the instants of touchdown (TD) and at the end of the main stance phase (END) for the small (i) and large (ii) lean angle. *Statistically significant age effect (P < 0.05). ^Statistically significant lean angle effect (P < 0.05).
velocity, a more anterior position of the extrapolated CM and a less flexed knee joint at the end of the main stance phase for the older adults (Tables 3 and 4). As a result of the above findings, the older compared to the younger adults showed a significantly (P < 0.05) lower margin of stability at the end of the main stance phase (Fig. 4). There was a significant (P < 0.05) age-by-lean angle interaction
K. Karamanidis et al. / Journal of Electromyography and Kinesiology 18 (2008) 980–989
a
Small lean angle 75
Margin of stabilityEND [cm]
for the horizontal CM velocity at the end of the main stance phase (Table 3). The post hoc analysis revealed a significantly (P < 0.05) higher horizontal CM velocity at the end of the main stance phase for the older adults at the large lean angle compared to all other groups (younger adults at small and large lean angle, and older adults at small lean angle; Table 3). Fig. 5 shows the average curves of the anteroposterior horizontal ground reaction force for the younger and older adults at the small and large lean angle. Concerning the duration of the main stance phase and the average anteroposterior horizontal ground reaction force (average values over the main stance phase), there were no significant (P > 0.05) differences between the small and large lean angles (Table 4). The comparison between older and younger adults revealed significant (P < 0.05) differences in the duration of the main stance phase and average anteroposterior ground reaction force (average values over the main stance phase) with lower values for the older compared to younger adults (Table 4). Margin of stability at the end of the main stance was significantly related to margin of stability at touchdown for both, the small (r = 0.954, P < 0.001) and large (r = 0.957, P < 0.001) lean angle (Fig. 6). Concerning the small lean angle, the margin of stability at the end of the
50
Old (n=9) Young (n=9)
25 0 -25 -50 2
r = 0.910 , P<0.001 -75 -40
b Margin of stabilityEND [cm]
986
50
40
Old (n=9) Young (n=9)
25 0 -25 -50 2
r = 0.916 , P<0.001
-75
Anteroposterior Force
-20
0
20
40
Fig. 6. Relationship between margin of stability at the end of the main stance phase (END) and at the instants of touchdown (TD) for the small (a) and large lean angle (b).
0 -2 -4 -6
Old (n=9) Young (n=9)
Small lean angle
2 0 -2 -4 -6 -8
20
Margin of stabilityTD [cm]
2
-8
0
Large lean angle 75
-40
[N/kg]
-20
Large lean angle 0
100
200
300
main stance phase and leg-extensor MTUs’ capacities showed significant correlation coefficients as follows (Fig. 7): for the triceps surae and quadriceps femoris muscle strength 0.495 (P = 0.037) and 0.522 (P = 0.026), respectively, and for the quadriceps femoris tendon stiffness 0.429 (P = 0.048). The correlation coefficients between margin of stability at the end of the main stance phase and leg-extensor MTUs’ capacities for the large lean angle were 0.663 (P = 0.003, triceps surae muscle strength), 0.587 (P = 0.010, quadriceps femoris muscle strength) and 0.538 (P = 0.021, quadriceps femoris tendon stiffness; Fig. 7). The multiple regression procedure showed a relationship between the leg-extensor MTUs’ capacities (triceps surae and quadriceps femoris muscle strength, and quadriceps femoris tendon stiffness) and the margin of stability at the end of the main stance phase with R = 0.588 for the small and R = 0.742 for large lean angle.
400
Stance phase [ms] Fig. 5. Average curves of the anteroposterior ground reaction force for the older and younger adults at the small and large lean angle (means and SEM). The vertical lines represent the end of the main stance (first minimum at the knee joint of the recovery limb after touchdown) of the examined groups. For better illustration, the anteroposterior ground reaction force curves are provided over the main stance (400 ms after touchdown of the recovery limb).
4. Discussion The present study aimed to examine the control of dynamic stability in younger and older adults after a sudden induced perturbation from a fixed forward-inclined position using a small (23 ± 3% BW) and a large (33 ± 3% BW) body lean angle. We found that the older
Margin of stabilityEND [cm]
Margin of stabilityEND [cm]
K. Karamanidis et al. / Journal of Electromyography and Kinesiology 18 (2008) 980–989
987
Small lean angle
60 40 20 0 -20 -40 -60
r = 0.495, P = 0.037
r = 0.429, P = 0.048
r = 0.522, P = 0.026
Large lean angle
60 40 20 0 -20 -40 -60
r = 0.587, P = 0.01
r = 0.663, P = 0.003 0.5 1.0 1.5 2.0 2.5 3.0
Mmax, Ankle [Nm/kg]
1.0 1.5 2.0 2.5 3.0 3.5 4.0
Mmax, Knee [Nm/kg]
r = 0.538, P = 0.021 15
30
45
60
75
90
StiffnessQF [kN/Strain]
Fig. 7. Relationship between margin of stability at the end of the main stance phase and muscle–tendon unit capacities (maximal isometric voluntary ankle plantarflexion moment Mmax, Ankle and knee extension moment Mmax, Knee, and normalized tendon-aponeurosis stiffness at the quadriceps femoris StiffnessQF) for the small (top, n = 18) and large (bottom, n = 18) lean angle.
compared to the younger adults showed clear deficits in controlling stability by a protective stepping response reflecting their lesser ability to use mechanisms responsible for maintaining dynamic stability. Compared to the younger adults, the older adults demonstrated in the first phase (from release until touchdown of the recovery limb) a lower increase of the base of support while falling for both lean angles. Moreover, we could confirm that the age-related deficits in the control of dynamic stability were partly associated with the reduced muscle strength and tendon stiffness of the leg-extensor MTUs in the elderly. The mechanism responsible for the higher margin of stability at touchdown for the younger compared to the older adults was the more anterior boundary of the base of support for the younger adults (Table 2 and Fig. 4). The components of the extrapolated CM at touchdown (horizontal CM velocity, vertical pffiffiffiffiffiffiffi projection of the CM to the ground and the term g=l) were similar between younger and older adults. This demonstrates that the younger compared to the older adults benefited more from the ability to change the base of support and, therefore, created a more stable body position at touchdown (greater margin of stability values, Fig. 4) after forward falls. Moreover, the above findings were present for both the small and large lean angles, showing that the greater skill in using mechanisms responsible for dynamic stability (i.e. greater increase of the base of support in relation to the extrapolated CM) for the younger adults was consistent across perturbation intensities. The more anterior boundary of the base of support for the young adults, when coupled with the essen-
tially invariant duration until touchdown between age groups, indicates that the source of the age-related changes in margin of stability at touchdown was the velocity with which the boundary of the base of support was shifted after release. The rise in body lean angle clearly affected the components of the extrapolated CM at touchdown (higher horizontal CM velocity and more anterior vertical projection of the CM to the ground, Table 2) leading to a lower margin of stability at touchdown for the large compared to the small lean angle (Fig. 4). However, it is important to notice that the younger as wellp asffiffiffiffiffiffiffi the older adults demonstrated an increase in the term g=l from the small to the large lean angle through a more flexed knee joint angle at touchdown. The mean differences in l (5 cm) between small and large lean angle and the effect of this parameter on the position of the extrapolated CM were too low to counteract the effect of the increased velocity and vertical projection of the CM on the position of the extrapolated CM at touchdown due to the rise in lean angle. As at the beginning of the stance phase, the younger compared to the older adults showed a higher margin of stability at the end of the main stance phase (i.e. termination of downwards motion of the body) for both lean angles (Fig. 4). The components responsible for the agerelated effect on margin of stability at the end of the main stance phase were the more anterior boundary of the base of support and the lower values of the position of the extrapolated CM for the younger adults (Table 3). The values of the extrapolated CM at touchdown were similar
988
K. Karamanidis et al. / Journal of Electromyography and Kinesiology 18 (2008) 980–989
between younger and older adults. Therefore, the lower values of the extrapolated CM at the end of the main stance phase illustrate that the younger adults were able to change the components of the extrapolated CM during stance independent of body lean angle. During the main stance phase, the younger compared to the older adults showed a higher time duration and greater average horizontal ground reaction force (Table 4 and Fig. 5). In this way, young adults were able to reduce horizontal CM velocity and in consequence the position of the extrapolated CM during the main stance phase to a greater degree than the older adults. In the literature (Karamanidis and Arampatzis, 2006; Narici et al., 2005; Savelberg and Meijer, 2004), it is well accepted that the aging process is associated with a degeneration of the musculoskeletal system. In accordance with those findings, the older in comparison to the younger adults examined in our study had lower triceps surae and quadriceps femoris muscle strength, and an increase in strain of the quadriceps femoris tendon and aponeurosis at a given tendon force (Fig. 3). In order to examine whether the observed differences in margin of stability at the end of the main stance phase between older and younger adults were associated to the age-related MTU weakness we used a multiple regression analysis including the variables of the MTU’s showing an age affect. We found that triceps surae and quadriceps femoris muscle strength, and quadriceps femoris tendon stiffness contributed from 35% to 55% to the margin of stability at the end of the main stance phase (R = 0.588 for the small lean angle; R = 0.742 for the large lean angle), confirming that legextensor MTU capacities partly affect the control of dynamic stability after falls. The strong relationship (r2 = 0.910 and 0.916 for the small and large lean angles, respectively, Fig. 6) between the margin of stability at touchdown and at the end of the main stance phase showed that the locomotion strategy employed before touchdown, limits the postural corrections during the stance phase after a forward fall. During the phase until touchdown, the main postural correction of the human system which aimed to improve the margin of dynamic stability was the increase of the base of support. It seems that the central nervous system senses the state of the dynamic stability (i.e. the boundary of the base of support in relation to the position of the extrapolated CM) of the musculoskeletal system before and after release and attempts to adjust its motor commands in an appropriate manner for a successful balance correction. However, the exact degree of the dynamic stability (i.e. margin of stability) at touchdown is not known for the human system during the swing phase of the recovery limb. Therefore, we can argue that the main mechanism used to regain dynamic stability after a loss of balance in the anterior direction (i.e. increased base of support) is mediated before touchdown by motor plans which aim to keep the extrapolated CM inside the boundary of the base of support.
Model predictions (Wu et al., 2007) have shown that the muscle strength of the leg affects minimal step length required for stability after a forward balance loss (the lower the muscle strength, the greater the minimal step length required) because the muscles have to generate mechanical work during the following stance phase, absorbing the mechanical energy of the body. Therefore, it might be suggested that some older adults should move from a stable body position at touchdown to an unstable position during the following stance phase due to their lower MTU capacities. However, our results in Fig. 4 demonstrated that almost all younger and older participants (except one younger and one older adult) having a stable body position at touchdown (positive margin of stability) were able to remain within the base of support during the following stance phase. Therefore, it can be confirmed that, given a stable position at touchdown, the MTU capacities of the participants were sufficient to maintain dynamic stability during the following stance phase. In conclusion, our findings demonstrate that the main mechanism by which humans regain balance after forward falls is the increase of the base of support during the swing phase of the recovery limb in a feedforward control manner. Older adults have deficits in using mechanisms responsible for the control of dynamic stability (i.e. lower increase of the base of support in relation to the extrapolated CM) after a sudden perturbation, leading to a higher risk of falls. Moreover, we have been able to confirm that these age-related deficits in stability control are partly associated with the lower muscle strength and tendon stiffness in their leg-extensor MTUs. Combining the above findings, therefore, it is reasonable to propose that, to reduce risk of falls, older individuals may benefit from leg-extensor-MTU strengthening programs as well as from interventions exercising the mechanisms responsible for dynamic stability. Acknowledgements We thank Martin Ku¨sel-Feldker, Thomas Fo¨rster and their teams for the technical assistance and their active support through the entire project. References Arampatzis A, Karamanidis K, DeMonte D, Stafilidis S, Morey-Klapsing G, Bru¨ggemann G-P. Differences between measured and resultant joint moments during voluntary and artificially elicited isometric knee extension contractions. Clin Biomech 2004;19:277–83. Arampatzis A, Morey-Klapsing G, Karamanidis K, DeMonte D, Stafilidis S, Bru¨ggemann G-P. Differences between measured and resultant joint moments during isometric contractions at the ankle joint. J Biomech 2005a;38:885–92. Arampatzis A, Stafilidis S, DeMonte G, Karamanidis K, Morey-Klapsing G, Bru¨ggemann G-P. Strain and elongation of the human gastrocnemius tendon and aponeurosis during maximal plantarflexion effort. J Biomech 2005b;38:833–41. Baker SP, Harvey AH. Fall injuries in the elderly. Clin Geriatr Med 1985;1:501–12.
K. Karamanidis et al. / Journal of Electromyography and Kinesiology 18 (2008) 980–989 Baratta R, Solomonow M, Zhou BH, Letson D, Chuinard R, D’Ambrosia R. Muscular coactivation. The role of the antagonist musculature in maintaining knee stability. Am J Sports Med 1988;16:113–22. Blake AJ, Morgan K, Bendall MJ, Dallosso H, Ebrahim SB, Arie TH, et al.. Falls by elderly people at home: prevalence and associated factors. Age Ageing 1988;17:365–72. Cumming RG, Klineberg RJ. Fall frequency and characteristics and the risk of hip fractures. J Am Geriatr Soc 1994;42:774–8. Dempster WT, Gabel WC, Felts WJ. The anthropometry of the manual work space for the seated subject. Am J Phys Anthropol 1959;17:289–317. Herzog W, Read LJ. Lines of action and moment arms of the major forcecarrying structures crossing the human knee joint. J Anat 1993;182:213–30. Hof AL. The equations of motion for a standing human reveal three mechanisms for balance. J Biomech 2007;40:451–7. Hof AL, Gazendam MG, Sinke WE. The condition for dynamic stability. J Biomech 2005;38:1–8. Hsiao ET, Robinovitch SN. Biomechanical influences on balance recovery by stepping. J Biomech 1999;32:1099–106. Karamanidis K, Arampatzis A. Mechanical and morphological properties of human quadriceps femoris and triceps surae muscle–tendon unit in relation to aging and running. J Biomech 2006;39:406–17. Karamanidis K, Arampatzis A. Age-related degeneration in leg-extensor muscle–tendon units decreases recovery performance after a forward fall: compensation with running experience. Eur J Appl Physiol 2007;99:73–85. Mackey DC, Robinovitch SN. Mechanisms underlying age-related differences in ability to recover balance with the ankle strategy. Gait Posture 2006;23:59–68. Mademli L, Arampatzis A, Morey-Klapsing G, Bruggemann GP. Effect of ankle joint position and electrode placement on the estimation of the antagonistic moment during maximal plantarflexion. J Electromyogr Kinesiol 2004;14:591–7. Madigan ML, Lloyd EM. Age-related differences in peak joint torques during the support phase of single-step recovery from a forward fall. J Gerontol A Biol Sci Med Sci 2005;60:910–4. Maganaris CN, Baltzopoulos V, Sargeant AJ. Changes in Achilles tendon moment arm from rest to maximum isometric plantarflexion: in vivo observations in man. J Physiol 1998;510:977–85. Narici MV, Maganaris C, Reeves N. Myotendinous alterations and effects of resistive loading in old age. Scand J Med Sci Sports 2005;15:392–401. Pijnappels M, Bobbert MF, van Dieen JH. Control of support limb muscles in recovery after tripping in young and older subjects. Exp Brain Res 2005a;160:326–33. Pijnappels M, Bobbert MF, van Dieen JH. Push-off reactions in recovery after tripping discriminate young subjects, older non-fallers and older fallers. Gait Posture 2005b;21:388–94. Pijnappels M, Bobbert MF, van Dieen JH. How early reactions in the support limb contribute to balance recovery after tripping. J Biomech 2005c;38:627–34. Savelberg HH, Meijer K. The effect of age and joint angle on the proportionality of extensor and flexor strength at the knee joint. J Gerontol A Biol Sci Med Sci 2004;59:1120–8. Simoneau M, Corbeil P. The effect of time to peak ankle torque on balance stability boundary: experimental validation of a biomechanical model. Exp Brain Res 2005;165:217–28.
989
Stafilidis S, Karamanidis K, Morey-Klapsing G, DeMonte G, Bru¨ggemann G-P, Arampatzis A. Strain and elongation of the vastus lateralis aponeurosis and tendon in vivo during maximal isometric contraction. Eur J Appl Physiol 2005;94:317–22. Thelen DG, Wojcik LA, Schultz AB, Ashton-Miller JA, Alexander NB. Age differences in using a rapid step to regain balance during a forward fall. J Gerontol A Biol Sci Med Sci 1997;52:M8–M13. Wojcik LA, Thelen DG, Schultz AB, Ashton-Miller JA, Alexander NB. Age and gender differences in single-step recovery from a forward fall. J Gerontol A Biol Sci Med Sci 1999;54:M44–50. Wojcik LA, Thelen DG, Schultz AB, Ashton-Miller JA, Alexander NB. Age and gender differences in peak lower extremity joint torques and ranges of motion used during single-step balance recovery from a forward fall. J Biomech 2001;34:67–73. Wu M, Ji L, Jin D, Pai YC. Minimal step length necessary for recovery of forward balance loss with a single step. J Biomech 2007;40: 1559–66. Kiros Karamanidis, received his PhD at the German Sport University of Cologne in 2006. His main research interests are in the field of adaptation of aging muscles and its effect on gait mechanics focusing on the prevention of falls in the elderly.
Adamantios Arampatzis received his PhD at the German Sport University of Cologne in 1995. He is the head of the research group focusing on the neuromechanics of the human musculoskeletal system at the Institute of Biomechanics and Orthopaedics from the German Sport University Cologne. Among his research interests are the adaptation potential of the human system to physical activity and the influence of the neuromechanical capacity of the musculoskeletal system on motor task behaviour during daily and sport activities.
Lida Mademli graduated in sport science from the Aristotle University of Thessaloniki (Greece) in 2002. The same year she moved to Cologne (Germany), where she joined the institute of Biomechanics and Orthopaedics at the German Sport University of Cologne. She is enjoying a grant for making her PhD thesis on the agerelated effects of fatigue on the neuromechanical properties of the muscle–tendon unit and the postural stability after sudden perturbations.
Journal of Electromyography and Kinesiology 18 (2008) 980–989 www.elsevier.com/locate/jelekin
Age-related deficit in dynamic stability control after forward falls is affected by muscle strength and tendon stiffness Kiros Karamanidis, Adamantios Arampatzis *, Lida Mademli Institute of Biomechanics and Orthopaedics, German Sport University of Cologne, Carl-Diem-Weg 6, 50933 Cologne, Germany Received 15 March 2007; received in revised form 27 April 2007; accepted 27 April 2007
Abstract The purpose of the work was to determine whether the age-related muscle weakness diminishes older adults’ ability to use mechanisms responsible for maintaining dynamic stability after forward falls. Nine older and nine younger adults participated in this study. To analyse the capacities of the leg-extensor muscle–tendon units, all subjects performed isometric maximal voluntary plantarflexion and knee extension contractions on a dynamometer. The elongation of the gastrocnemius medialis and the vastus lateralis tendon and aponeuroses during isometric contraction was examined by ultrasonography. Recovery behaviour was determined after a sudden fall from two forward-inclined lean angles. Compared to older adults, younger adults had higher muscle strength and tendon stiffness. Younger adults created a higher margin of stability compared to older, independent of perturbation intensity. The main mechanism improving the margin of dynamic stability was the increase of the base of support. The results, further, demonstrated that the locomotion strategy employed before touchdown affects the stability of the stance phase and that muscle strength and tendon stiffness contributed significantly to stability control. We concluded that, to reduce the risk of falls, older individuals may benefit from muscle–tendon unit strengthening programs as well as from interventions exercising the mechanisms responsible for dynamic stability. 2007 Elsevier Ltd. All rights reserved. Keywords: Margin of stability; Balance recovery; Base of support; Human
1. Introduction Falls are a major cause of injury among elderly people, including serious hip and wrist fractures (Baker and Harvey, 1985; Cumming and Klineberg, 1994). Approximately one third of adults over 65 years of age fall at least once a year (Blake et al., 1988). Most of these falls occur after a loss of stability in a forward direction such as tripping while walking (Blake et al., 1988). Identifying the mechanisms by which stability deficits related to forward falls occur within the elderly population may be of importance for the development of effective exercise interventions for fall prevention. From a mechanical point of view, the margin of stability during locomotion can be quantified by the position of the extrapolated centre of mass (CM; deter*
Corresponding author. Tel.: +49 221 4982 5620; fax: +49 221 4971598. E-mail address: [email protected] (A. Arampatzis).
1050-6411/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.jelekin.2007.04.003
mined by the position and velocity of the CM) in relation to the base of support (Hof et al., 2005). A loss of stability in the forward direction apparently occurs when the position of the extrapolated CM exceeds the anterior boundary of the base of support (Hof et al., 2005). In such an unstable body position, the neuro-motor system must select and execute postural corrections including mechanisms responsible for the control of dynamic stability in order to prevent any unintended forward fall. Hof (2007) recently presented three mechanisms by which stability may be maintained: (a) by increasing the base of support in relation to the extrapolated CM, (b) by counter-rotating segments around the CM, and (c) by applying an external force, other than the ground reaction force (e.g. grasping). Sudden release from a forward-inclined body position is commonly used to examine humans’ capabilities of regaining stability in the anterior direction (Hsiao and Robinovitch, 1999; Thelen et al., 1997; Wojcik et al., 2001). Such an
K. Karamanidis et al. / Journal of Electromyography and Kinesiology 18 (2008) 980–989
experimental design has the benefit that the magnitude of postural disturbance can be controlled and easily varied (Hsiao and Robinovitch, 1999). When released from an inclined body position it is observed that the inability to regain balance is greater among older adults than among younger adults (Thelen et al., 1997; Wojcik et al., 1999; Wojcik et al., 2001). Several studies (Mackey and Robinovitch, 2006; Pijnappels et al., 2005a,b) have suggested that the deterioration of the musculoskeletal function in the elderly may play a key role for the observed age-related deficits in stability control. Simulation models showed that a lower magnitude (Hsiao and Robinovitch, 1999) and rate (Simoneau and Corbeil, 2005) of moment generation at the ankle joint clearly reduces a person’s capacity to regain stability. Recently, it has been reported (Karamanidis and Arampatzis, 2007) that deficits in balance recovery by stepping in older adults are associated with the lower muscle strength and tendon stiffness in their leg-extensor muscle– tendon units (MTUs). The authors (Karamanidis and Arampatzis, 2007) concluded that the age-related degeneration of the MTUs diminishes older adults’ ability to restore balance with a single-step in the anterior direction. However, a faithful analysis of the dynamic stability has not been included and, therefore, it is difficult to identify the exact mechanisms responsible for the reported findings. Postural correction after a sudden perturbation depends on actions of both the support limb during the push-off phase and the recovery limb during the step execution phase (Karamanidis and Arampatzis, 2007; Pijnappels et al., 2005b). The maximal moments generated at the support limb during the push-off phase (e.g. at the ankle joint up to 200 N m) (Pijnappels et al., 2005b,c) as well as at the recovery limb during the step execution phase (e.g. at the knee joint up to 100 N m) (Wojcik et al., 2001) suggest that the muscle strength and tendon stiffness of the lower extremities may be relevant for performing successful postural corrections after a sudden perturbation. We can conclude that a reasonable challenge of the human system after a sudden perturbation is to perform effective balance responses by selecting motor plans which include mechanisms responsible for maintaining dynamic stability. Although there are several studies in the literature examining balance corrections after anterior postural disturbance (Karamanidis and Arampatzis, 2007; Madigan and Lloyd, 2005; Thelen et al., 1997; Wojcik et al., 1999) we did not find any study determining in a sophisticated way the components of the dynamic stability in younger and older adults after induced forward falls and relating them to the MTUs’ capacities. Combining the above observations, we hypothesised that, compared to younger adults, older adults would show deficits in using mechanisms responsible for maintaining dynamic stability after an induced forward fall, reflecting the lower capacities of their MTUs in the lower extremities. Therefore, the aims of the current work were to examine the control of dynamic stability in younger and older adults after a sudden induced perturbation from a fixed forward-
981
inclined position using a small and a large body lean angle and, hence, varying the intensity of disturbance. Further, we intended to determine whether the age-related degeneration of leg-extensor MTU capacities diminishes older adults’ ability to use mechanisms responsible for maintaining dynamic stability after a sudden perturbation. 2. Materials and methods 2.1. Experimental design The investigation was conducted with nine male older adults aged 60–68 years and nine male younger adults aged 22–32 years. All subjects were experienced endurance runners and performed running commonly 4–7 times per week over the last 10 years and participated regularly in middle- and long-distance running competitions. The training distance at the time of the study ranged from 40 to 100 km per week. Approval was obtained from the university committee for the protection of human subjects and informed consent was given by all subjects. None of the chosen test subjects had a history of neuromuscular or musculoskeletal impairments. The experimental design of the recovery task has been previously reported in detail (Karamanidis and Arampatzis, 2007). Briefly, the subjects were released suddenly without warning from a fixed forward-inclined position (see also: Thelen et al., 1997; Wojcik et al., 2001). The participants were initially maintained in an inclined forward posture by a horizontal inextensible cable, attached at one end to a belt worn by the participants around the pelvis and at the other end to a custom-built pneumatic release system. Subjects were instructed and encouraged to restore balance by taking a single-step after forward fall was initiated. All participants wore a full trunk safety harness suspended from an overhead track that allowed for forward and lateral motion while preventing contact of any body part, other than the feet, with the ground. The angle of the forward lean was controlled by adjusting the lean control cable length until the load cell attached to the cable indicated that it supported a specified percentage of the subject’s body weight (BW). In the first set of trials (small lean angle), the subjects attempted to recover balance following three releases at a lean-control cable load of 23 ± 3% BW. In the second set of trials (large lean angle), three trials at a lean-control cable load of 33 ± 3% BW were carried out. The examined lean-angle cable loads correspond to lean angles (defined as the angle between the vertical in the sagittal plane and the line connecting the CM and the midpoint of the foot) of 19.7 ± 2.4 and 26.0 ± 2.7. Specific lean-angle cable loads were chosen because it has been shown that there is a clear reduction in the ability to recover balance with a single-step from lean-control cable loads of about 20% to loads of more than 30% BW for older individuals (Madigan and Lloyd, 2005; Thelen et al., 1997). 2.2. Measurement of recovery mechanics At the initiation of each trial, the subject stood barefoot on a force plate (60 cm · 90 cm, Kistler, Winterthur, Switzerland). The first step always landed on a second force plate (60 cm · 90 cm, Kistler, Winterthur, Switzerland) mounted in front of the initial force plate. The ground reaction forces were collected at a sampling rate of 1080 Hz. To determine the onset of release, the pneumatic release system triggered a TTL signal at the instant of release which was simultaneously captured as an analogue signal by the Vicon motion capture system (1080 Hz; Model 624, Vicon,
982
K. Karamanidis et al. / Journal of Electromyography and Kinesiology 18 (2008) 980–989
Oxford, United Kingdom). Kinematic data were recorded with 12 Vicon cameras operating at 120 Hz. Thirty-eight reflective markers (radius 14 mm) were used to track the whole body kinematics. The markers defined the left and right foot, left and right lower leg, left and right thigh, pelvis, thorax, left and right upper arm, left and right forearm, left and right hand and the head. The segmental masses and the location of the segment centres of masses were calculated based on the data reported by Dempster et al. (1959). Sagittal angular joint angle kinematics at the knee were analysed for the recovery limb. A completely extended knee was defined as 180 knee joint angle. For each trial, three time points were identified (Fig. 1): (A) subject’s release (determined by the analogue signal which the Vicon motion capture system received from the pneumatic release system), (B) touchdown of the recovery limb (determined by the ground reaction force data – threshold level 20 N) and (C) first minimum at the knee joint angle of the recovery limb after touchdown. Time point (C) indicated termination of downwards motion of the body (i.e. vertical CM velocity almost zero; Fig. 1). Further, recovery stepping task was divided into two phases based on the identified three time points (Fig. 1): (1) phase until touchdown (defined from the subject’s release until touchdown of the recovery limb) and (2) main stance phase (defined from touchdown until the first minimum at the knee joint angle of the recovery limb). For better illustration, the curves in Fig. 1 end 400 ms after touchdown of the recovery limb (stance phase). The margin of stability in the anteroposterior direction was calculated according to Hof et al. (2005) as follows: bx ¼ U x max X CM
B
A
Release
D
C
TD
KJAmin
400 ms after TD
200 180
0.4 KJA
VzCOM
where bx is margin of stability in the anteroposterior direction, Uxmax is anterior boundary of the base of support (i.e. horizontal component of the vertical projection of the toe from the recovery limb to the ground; a zero value represents the position of the toe before release), XCM is extrapolated CM in the anteroposterior V ffiffiffiffiffi Þ, P X CM is horizontal (anteropostedirection ðX CM ¼ P X CM þ pX CM g=l
rior) component of the vertical projection of the CM to the ground, V X CM is horizontal (anteroposterior) CM velocity, g is acceleration of gravity, and l is distance between CM and centre of the ankle joint in the sagittal plane. Margin of stability was determined in the anteroposterior direction because, after a forward fall, both the extrapolated CM as well as the base of support mainly shifts in the anterior direction. To illustrate this, Fig. 2 provides the anteroposterior boundary of the base of support, horizontal component of the vertical projection of the CM to the ground and the position of the extrapolated CM during the recovery task for both a singleand a multiple-stepper. Postural stability is maintained in circumstances where the position of the extrapolated CM is within the base of support (positive values of margin of stability) while stability is lost in cases where the extrapolated CM passes the anterior boundary of the base of support during stance phase (negative values of margin of stability). To examine the conformity between the margin of stability calculated by the model presented by Hof et al. (2005) and the real behaviour of the human system during forward falls we compared the margin of stability at the end of the main stance phase (i.e. positive vs. negative values) with subjects recovery strategy (i.e. single-step vs. multiple-step recovery strategy). A trial was defined as single-step behaviour if only one step was taken or if the anterior displacement of the second step (contralateral limb) did not exceed the anterior displacement of the recovery limb. A multiple-step behaviour was defined when the subject took a second step of the recovery limb of any kind or when the subject took a contralateral step whose anterior displacement exceeded that of the recovery limb (for exact definition see Karamanidis and Arampatzis, 2007). We found that in about 96% of the cases the margin of stability at the end of the main stance phase was in line with the recovery strategy of the participants (negative margin of stability value associated to multiple-step recovery strategy and vice versa).
0.0 140
[m/s]
[Degrees]
160
120 100
-0.4
80 Main stance phase
60 -100
Phase until touchdown
-50
Stance phase
0
5
-0.8 100
[%] Fig. 1. Analysed time points and time intervals, and angular motion at the knee joint (KJA) and vertical centre of mass velocity (VzCOM) of one subject. A: subjects release; B: touchdown (TD) of the recovery limb; C: first minimum at the KJA of the recovery limb after touchdown (KJAmin); D: 400 ms after touchdown. Phase until touchdown: from release until TD (x-axis: normalized from 100% to 0%); Stance phase: from TD until 400 ms after TD (x-axis: normalized from 0% to 100%); Main stance phase: from TD until KJAmin. A completely extended knee was defined as 180 knee joint angles. Note that time point KJAmin provided termination of downwards motion of the body (i.e. vertical COM velocity almost zero).
2.3. Measurement of muscle strength and tendon elongation To examine muscle strength potential and tendon stiffness of the leg-extensor MTU’s all subjects performed isometric maximal voluntary ankle plantarflexion and knee extension contractions on a dynamometer (Biodex Medical Systems. Inc., Shirley, NY, USA). The method for calculating the resultant joint moments has been previously described in detailed (Arampatzis et al., 2004; Arampatzis et al., 2005a). Axis misalignment between dynamometer and ankle or knee joint during contraction was taken into consideration using 12 Vicon cameras operating at 120 Hz. The effect of moments arising from antagonistic muscles during the ankle plantarflexion and knee extension efforts on the resultant joint moment was taken into account using the method also described previously (Baratta et al., 1988; Mademli et al., 2004). Therefore, in the following text maximal knee and ankle joint moment refer to the maximal joint moment values where the effect of the joint axis alteration relative to the dynamometer axis, gravitational effects and the effect of antagonist moment on the measured moment were taken into account.
K. Karamanidis et al. / Journal of Electromyography and Kinesiology 18 (2008) 980–989 TD
Release
1.6 1 .6
1.6
boundary BS projection CM extrapolated XCM
[m]
1.2
0.8
0.8
0.4
0.4
a 0
200
400
600
800
1000
KJA min
boundary BS projection CM extrapolated XCM
margin of stability
1.2
margin of stability
0.0
TD
Release
KJA min
983
b
0.0
1200
0
200
400
600
800
1000
[ms]
[ms]
Fig. 2. Anteroposterior boundary of the base of support (boundary BS), horizontal component of the vertical projection of the centre of mass to the ground (projection CM) and the position of the extrapolated centre of mass (extrapolated XCM) during the recovery task of a single- (a) and multiplestepper (b). Margin of stability is the instantaneous difference between the boundary BS and the extrapolated XCM. Postural stability is maintained in circumstances where the extrapolated XCM is within the boundary BS ((a) positive values of margin of stability) while stability is lost in cases where the extrapolated XCM passes the anterior boundary BS during stance phase ((b) negative values of margin of stability). The zero value in the vertical axis represents the initial position (before release) of the horizontal component of the vertical projection of the toe from the recovery limb to the ground. TD, touchdown; KJAmin, first minimum at the knee joint angle of the recovery limb after TD.
The exact procedure for the measurement of the elongation of the tendon and aponeurosis during contraction has also been described earlier (Arampatzis et al., 2005b; Stafilidis et al., 2005). Briefly, to measure the elongation of the triceps surae and quadriceps femoris tendon and aponeurosis during contraction, the gastrocnemius medialis and the vastus lateralis and their distal aponeurosis were visualized by ultrasonography (Aloka SSD 4000, Tokyo, Japan, 43 Hz). The effect of inevitable joint angular rotation on the measured elongation of the gastrocnemius medialis or vastus lateralis tendon and aponeurosis during contraction was taken into account by capturing the motion of the tendon and aponeurosis from the corresponding muscle during a passive (relaxed condition) rotation of the ankle or knee joint (Arampatzis et al., 2005b; Stafilidis et al., 2005). The strain values of the triceps surae and quadriceps femoris tendon and aponeurosis during MVC were calculated at every 100 N and 200 N and at maximal calculated tendon force respectively; corresponding tendon moment arms were taken from Maganaris et al. (1998), and Herzog and Read (1993). To compare the stiffness of triceps surae and quadriceps femoris tendon and aponeurosis between older and younger adults we calculated the slope of the calculated tendon force versus strain of the tendon and aponeurosis (normalized stiffness) between 50% and 100% of the maximal tendon force by means of linear regression. 2.4. Statistics A t-test for independent samples was used in order to check for differences in isometric maximal voluntary ankle plantarflexion moment and knee extension moment, stiffness and strain values (every 100 N and 200 N for the triceps surae and quadriceps femoris MTUs, respectively) of the triceps surae and quadriceps femoris tendon and aponeurosis between older and younger adults. For the analysis of the recovery task the mean values from the three trials were utilised for the comparison between age (younger vs. older adults) and lean angle (small vs. large lean angle) groups. A two-factor (age · lean angle) analysis of variance
(ANOVA) was used in order to detect group differences in the analysed parameters of the dynamic stability. The level of significance was set at a = 0.05. When a significant age-by-lean angle interaction was present a post hoc test (Bonferroni) was conducted in order to determine where these differences occurred. To examine relationships between parameters of dynamic stability we used the Pearson correlation coefficient. Further, we used Pearson correlation coefficient and a multiple regression analysis to identify the relationship between leg-extensor MTU’s capacities and margin of stability at the end of the main stance phase for both, the small (n = 18) and large (n = 18) lean angle. All results in the figures are presented as mean and standard error of mean (mean and SEM), whereas in the text and tables they are expressed as mean and standard deviation (mean and SD).
3. Result 3.1. Muscle strength and tendon stiffness Older adults showed significantly (P < 0.05) lower maximal isometric ankle plantarflexion and knee extension moments (Table 1). At the triceps surae MTU, the strain Table 1 Anthropometric data, maximal joint moments and normalized tendonaponeurosis stiffness of the triceps surae (TS) and quadriceps femoris (QF) muscle–tendon units for the older and younger adults (means ± SD)
Age [yr] Body mass [kg] Body height [cm]* Momentmax. ankle joint [N m/kg]* StiffnessTS tendon [kN/Strain] Momentmax. knee joint [N m/kg]* StiffnessQF tendon [kN/Strain]* *
Old (n = 9)
Young (n = 9)
64 ± 3 75 ± 7 176 ± 3 1.25 ± 0.32 25.95 ± 10.48 2.00 ± 0.41 46.12 ± 15.47
27 ± 4 73 ± 6 181 ± 4 1.78 ± 0.35 27.00 ± 11.40 2.51 ± 0.48 57.87 ± 10.55
Statistically significant age effect (P < 0.05).
984
K. Karamanidis et al. / Journal of Electromyography and Kinesiology 18 (2008) 980–989
values of the tendon and aponeurosis for a given tendon force (every 100 N) showed no significant differences (P > 0.05) between the two groups (Fig. 3). At the quadriceps femoris MTU, however, the strain values of the tendon and aponeurosis for a given tendon force (every 200 N) was significantly higher (P < 0.05) for the older compared to the younger adults (Fig. 3). In the same manner, the normalized tendon stiffness showed statistically significant (P < 0.05) lower values for the older adults at the quadriceps femoris MTU whereas no significant (P > 0.05) differences between younger and older adults were found at the triceps surae MTU (Table 1). Body height was significantly lower (P < 0.05) for the older compared to the younger adults (on average about 5 cm; Table 1).
gin of stability at the instants of touchdown for the large compared to the small lean angle (Table 2 and Fig. 4). The comparison between older and younger adults revealed significant (P < 0.05) differences in the anterior boundary of the base of support with lower values for the older adults (Table 2). There were no significant (P > 0.05) differences between older and younger adults in the horizontal component of the vertical projection of the CM to the ground, horizontal CM velocity, or the term pffiffiffiffiffiffiffi g=l and, therefore, on the position of the extrapolated CM at touchdown (Table 2). As a result of the above findings (reduced boundary of the base of support and similar position of the extrapolated CM) the older compared to the younger adults showed a significantly (P < 0.05) lower margin of stability at touchdown (Fig. 4).
3.2. Stability at touchdown 3.3. Stability at the end of the main stance phase The duration until touchdown was significantly (P < 0.05) lower for the large compared to the small lean angle (Table 2). The boundary of the base of support at touchdown, the horizontal component of the vertical projection of the CMp toffiffiffiffiffiffiffi the ground, the horizontal CM velocity and the term g=l at the instants of touchdown were significantly (P < 0.05) higher for the large compared to the small lean angle (Table 2). The knee joint angle showed significantly (P < 0.05) smaller values (more flexed knee joint) for the large compared to the small lean angle at touchdown (old and young adults at large lean angle 137.5 ± 6.2 and 136.3 ± 10.4, respectively; at small lean angle 148.1 ± 8.6 and 139.2 ± 8.7, respectively). There was a significant correlation (r = 0.588, P < 0.001, n = 36) between knee angle and l (distance between CM and centre of the ankle joint) at touchdown. This shows that the pffiffiffiffiffiffiffi higher value for the term g=l at touchdown for the large compared to the small lean angle was partly related to the knee joint angle (more flexed knee joint for the large lean angle). Further, we found a significantly (P < 0.05) more anterior position of the extrapolated CM and a lower mar-
For the large compared to the small lean angle we found a significantly (P < 0.05) more anterior boundary of the base of support, more horizontal component of the vertical pffiffiffiffiffiffiffi projection of the CM to the ground, higher term g=l and a more anterior position of the extrapolated CM at the end of the main stance phase (Table 3). Further, the knee joint angle at the end of the main stance phase showed significantly (P < 0.05) lower values (more flexed knee joint) for the large compared to the small lean angle (Table 4). As for the instants of touchdown, we found a significant correlation between knee angle and l at the end of the main stance phase (r = 0.665, P < p 0.001, ffiffiffiffiffiffiffi n = 36), showing that the higher value for the term g=l for the large lean angle compared to the small lean angle was partly related to the more flexed knee joint for the large lean angle. The margin of stability at the end of the main stance phase was not significantly (P > 0.05) different between the small and large lean angles (Fig. 4). The comparison between older and younger adults revealed a significantly (P < 0.05) lower boundary of the base of support, higher horizontal CM
Triceps suraeMTU
8
Old (n = 9) Young (n = 9)
8
6
Strain [%]
Quadriceps femorisMTU
10
Old (n = 9) Young (n = 9)
6
*
4
*
*
4 2
*
** * *
*
2
a
0 0
500
1000
Tendon force [N]
1500
2000
b
0 0
1000
2000
3000
4000
Tendon force [N]
Fig. 3. Strain values at every 100 N and 200 N and at maximal calculated tendon force of the triceps surae (a) and quadriceps femoris (b) tendon and aponeurosis during isometric maximal voluntary contraction, respectively (means ± SEM). MTU, muscle–tendon unit; *statistically significant age effect (P < 0.05).
K. Karamanidis et al. / Journal of Electromyography and Kinesiology 18 (2008) 980–989 Table 2 Duration until touchdown, anterior boundary of the base of support (Boundary BS), horizontal component of the vertical projection of the centre of mass to the ground (Projection pffiffiffiffiffiffiffi CM), horizontal centre of mass velocity (Horizontal VCM), term g=l (l: distance between the centre of mass and centre of the ankle joint; g: acceleration of gravity) and position of the extrapolated centre of mass (Extrapolated XCM) at the instants of touchdown for the different groups (means ± SD)
Duration until touchdown [ms]^ Boundary BS [cm]*, ^ Projection CM [cm]^ Horizontal VCM [m/s]^ Term pffiffiffiffiffiffiffi g=l½1=s? Extrapolated XCM [cm]^
Large lean angle
Small lean angle
Large lean angle
Old (n = 9)
Young (n = 9)
Old (n = 9)
Old (n = 9)
Young (n = 9)
Old (n = 9)
523 ± 31
535 ± 68
472 ± 26
469 ± 36
85.0 ± 14.6
92.4 ± 15.7
91.2 ± 15.3
110.6 ± 13.3
65.4 ± 5.1
63.1 ± 14.6
78.3 ± 5.8
82.8 ± 12.0
83.3 ± 13.1
89.6 ± 16.3
90.2 ± 13.9
106.8 ± 14.2
0.89 ± 0.35
0.69 ± 0.28
1.25 ± 0.38
0.59 ± 0.28
53.4 ± 3.2
50.3 ± 11.1
67.8 ± 3.4
66.1 ± 8.3
3.55 ± 0.11
3.63 ± 0.13
3.69 ± 0.08
3.76 ± 0.15
1.09 ± 0.13
1.06 ± 0.27
1.36 ± 0.13
1.27 ± 0.21
90.8 ± 12.0
82.1 ± 20.8
3.49 ± 0.07
3.49 ± 0.07
3.64 ± 0.06
3.59 ± 0.12
84.5 ± 5.6
80.6 ± 18.2
105.0 ± 5.8
Young (n = 9)
101.5 ± 13.0
Statistically significant age effect (P < 0.05). Statistically significant lean angle effect (P < 0.05).
Margin of stability [cm]
40
Small lean angle
Small lean angle
Old adults (n=9)
Young adults (n=9)
Boundary BS [cm]*, ^ Projection CM [cm]^ Horizontal VCM [m/s]*, + Term pffiffiffiffiffiffiffi g=l½1=s? Extrapolated XCM [cm]*, ^
0
98.4 ± 16.8
End main stance phase was defined as the first minimum at the knee joint angle of the recovery limb after touchdown (i.e. termination of downwards motion of the body). * Statistically significant age effect (P < 0.05). ^ Statistically significant lean angle effect (P < 0.05). + Statistically significant age-by-lean angle interaction (P < 0.05). The post hoc analysis revealed significantly (P < 0.05) higher Horizontal VCMvalues for the older adults at the large lean angle compared to all other groups (younger adults at the large and small lean angle, and older adults at the small lean angle).
-20 -40
40
112.3 ± 9.7
Young (n = 9)
Table 4 Minimum of the knee joint angle of the recovery limb after touchdown (Knee anglemin), duration of the main stance phase (Time to knee anglemin) and average values (over main stance phase) of the horizontal ground reaction force for the different groups (means ± SD)
20
i
-60
Margin of stability [cm]
Table 3 Anterior boundary of the base of support (Boundary BS), horizontal component of the vertical projection of the centre of mass to the ground (Projection pffiffiffiffiffiffiffi CM), horizontal centre of mass velocity (Horizontal VCM), the term g=l (l: distance between the centre of mass and the centre of the ankle joint; g: acceleration of gravity) and position of the extrapolated centre of mass (extrapolated XCM) at the end of the main stance phase for the different groups (means ± SD)
Small lean angle
* ^
985
Small lean angle
Large lean angle
Old (n = 9)
Old (n = 9)
Young (n = 9)
Young (n = 9)
Knee anglemin []*, 131.8 ± 14.0 119.3 ± 10.1 126.3 ± 10.0 110.1 ± 8.1 ^
Large lean angle
Large lean angle
Old adults (n=9)
Young adults (n=9)
20 0
-40
ii TD
END
bTD =
*, ⊥
173 ± 74
193 ± 58
123 ± 63
224 ± 56
2.02 ± 1.53 2.72 ± 0.61 1.71 ± 2.26 3.16 ± 0.56
Main stance phase was defined from touchdown until minimum knee joint angle of the recovery limb. A completely extended knee was defined as 180 knee joint angle. Posterior-directed forces were defined as negative force values. * Statistically significant age effect (P < 0.05). ^ Statistically significant lean angle effect (P < 0.05).
-20
-60
Time to knee anglemin [ms]* Average forcehorizontal [N/kg]*
TD
END
bEND = *
Fig. 4. Individual and mean (SEM) margin of stability values (b) for the older (left) and younger (right) adults at the instants of touchdown (TD) and at the end of the main stance phase (END) for the small (i) and large (ii) lean angle. *Statistically significant age effect (P < 0.05). ^Statistically significant lean angle effect (P < 0.05).
velocity, a more anterior position of the extrapolated CM and a less flexed knee joint at the end of the main stance phase for the older adults (Tables 3 and 4). As a result of the above findings, the older compared to the younger adults showed a significantly (P < 0.05) lower margin of stability at the end of the main stance phase (Fig. 4). There was a significant (P < 0.05) age-by-lean angle interaction
K. Karamanidis et al. / Journal of Electromyography and Kinesiology 18 (2008) 980–989
a
Small lean angle 75
Margin of stabilityEND [cm]
for the horizontal CM velocity at the end of the main stance phase (Table 3). The post hoc analysis revealed a significantly (P < 0.05) higher horizontal CM velocity at the end of the main stance phase for the older adults at the large lean angle compared to all other groups (younger adults at small and large lean angle, and older adults at small lean angle; Table 3). Fig. 5 shows the average curves of the anteroposterior horizontal ground reaction force for the younger and older adults at the small and large lean angle. Concerning the duration of the main stance phase and the average anteroposterior horizontal ground reaction force (average values over the main stance phase), there were no significant (P > 0.05) differences between the small and large lean angles (Table 4). The comparison between older and younger adults revealed significant (P < 0.05) differences in the duration of the main stance phase and average anteroposterior ground reaction force (average values over the main stance phase) with lower values for the older compared to younger adults (Table 4). Margin of stability at the end of the main stance was significantly related to margin of stability at touchdown for both, the small (r = 0.954, P < 0.001) and large (r = 0.957, P < 0.001) lean angle (Fig. 6). Concerning the small lean angle, the margin of stability at the end of the
50
Old (n=9) Young (n=9)
25 0 -25 -50 2
r = 0.910 , P<0.001 -75 -40
b Margin of stabilityEND [cm]
986
50
40
Old (n=9) Young (n=9)
25 0 -25 -50 2
r = 0.916 , P<0.001
-75
Anteroposterior Force
-20
0
20
40
Fig. 6. Relationship between margin of stability at the end of the main stance phase (END) and at the instants of touchdown (TD) for the small (a) and large lean angle (b).
0 -2 -4 -6
Old (n=9) Young (n=9)
Small lean angle
2 0 -2 -4 -6 -8
20
Margin of stabilityTD [cm]
2
-8
0
Large lean angle 75
-40
[N/kg]
-20
Large lean angle 0
100
200
300
main stance phase and leg-extensor MTUs’ capacities showed significant correlation coefficients as follows (Fig. 7): for the triceps surae and quadriceps femoris muscle strength 0.495 (P = 0.037) and 0.522 (P = 0.026), respectively, and for the quadriceps femoris tendon stiffness 0.429 (P = 0.048). The correlation coefficients between margin of stability at the end of the main stance phase and leg-extensor MTUs’ capacities for the large lean angle were 0.663 (P = 0.003, triceps surae muscle strength), 0.587 (P = 0.010, quadriceps femoris muscle strength) and 0.538 (P = 0.021, quadriceps femoris tendon stiffness; Fig. 7). The multiple regression procedure showed a relationship between the leg-extensor MTUs’ capacities (triceps surae and quadriceps femoris muscle strength, and quadriceps femoris tendon stiffness) and the margin of stability at the end of the main stance phase with R = 0.588 for the small and R = 0.742 for large lean angle.
400
Stance phase [ms] Fig. 5. Average curves of the anteroposterior ground reaction force for the older and younger adults at the small and large lean angle (means and SEM). The vertical lines represent the end of the main stance (first minimum at the knee joint of the recovery limb after touchdown) of the examined groups. For better illustration, the anteroposterior ground reaction force curves are provided over the main stance (400 ms after touchdown of the recovery limb).
4. Discussion The present study aimed to examine the control of dynamic stability in younger and older adults after a sudden induced perturbation from a fixed forward-inclined position using a small (23 ± 3% BW) and a large (33 ± 3% BW) body lean angle. We found that the older
Margin of stabilityEND [cm]
Margin of stabilityEND [cm]
K. Karamanidis et al. / Journal of Electromyography and Kinesiology 18 (2008) 980–989
987
Small lean angle
60 40 20 0 -20 -40 -60
r = 0.495, P = 0.037
r = 0.429, P = 0.048
r = 0.522, P = 0.026
Large lean angle
60 40 20 0 -20 -40 -60
r = 0.587, P = 0.01
r = 0.663, P = 0.003 0.5 1.0 1.5 2.0 2.5 3.0
Mmax, Ankle [Nm/kg]
1.0 1.5 2.0 2.5 3.0 3.5 4.0
Mmax, Knee [Nm/kg]
r = 0.538, P = 0.021 15
30
45
60
75
90
StiffnessQF [kN/Strain]
Fig. 7. Relationship between margin of stability at the end of the main stance phase and muscle–tendon unit capacities (maximal isometric voluntary ankle plantarflexion moment Mmax, Ankle and knee extension moment Mmax, Knee, and normalized tendon-aponeurosis stiffness at the quadriceps femoris StiffnessQF) for the small (top, n = 18) and large (bottom, n = 18) lean angle.
compared to the younger adults showed clear deficits in controlling stability by a protective stepping response reflecting their lesser ability to use mechanisms responsible for maintaining dynamic stability. Compared to the younger adults, the older adults demonstrated in the first phase (from release until touchdown of the recovery limb) a lower increase of the base of support while falling for both lean angles. Moreover, we could confirm that the age-related deficits in the control of dynamic stability were partly associated with the reduced muscle strength and tendon stiffness of the leg-extensor MTUs in the elderly. The mechanism responsible for the higher margin of stability at touchdown for the younger compared to the older adults was the more anterior boundary of the base of support for the younger adults (Table 2 and Fig. 4). The components of the extrapolated CM at touchdown (horizontal CM velocity, vertical pffiffiffiffiffiffiffi projection of the CM to the ground and the term g=l) were similar between younger and older adults. This demonstrates that the younger compared to the older adults benefited more from the ability to change the base of support and, therefore, created a more stable body position at touchdown (greater margin of stability values, Fig. 4) after forward falls. Moreover, the above findings were present for both the small and large lean angles, showing that the greater skill in using mechanisms responsible for dynamic stability (i.e. greater increase of the base of support in relation to the extrapolated CM) for the younger adults was consistent across perturbation intensities. The more anterior boundary of the base of support for the young adults, when coupled with the essen-
tially invariant duration until touchdown between age groups, indicates that the source of the age-related changes in margin of stability at touchdown was the velocity with which the boundary of the base of support was shifted after release. The rise in body lean angle clearly affected the components of the extrapolated CM at touchdown (higher horizontal CM velocity and more anterior vertical projection of the CM to the ground, Table 2) leading to a lower margin of stability at touchdown for the large compared to the small lean angle (Fig. 4). However, it is important to notice that the younger as wellp asffiffiffiffiffiffiffi the older adults demonstrated an increase in the term g=l from the small to the large lean angle through a more flexed knee joint angle at touchdown. The mean differences in l (5 cm) between small and large lean angle and the effect of this parameter on the position of the extrapolated CM were too low to counteract the effect of the increased velocity and vertical projection of the CM on the position of the extrapolated CM at touchdown due to the rise in lean angle. As at the beginning of the stance phase, the younger compared to the older adults showed a higher margin of stability at the end of the main stance phase (i.e. termination of downwards motion of the body) for both lean angles (Fig. 4). The components responsible for the agerelated effect on margin of stability at the end of the main stance phase were the more anterior boundary of the base of support and the lower values of the position of the extrapolated CM for the younger adults (Table 3). The values of the extrapolated CM at touchdown were similar
988
K. Karamanidis et al. / Journal of Electromyography and Kinesiology 18 (2008) 980–989
between younger and older adults. Therefore, the lower values of the extrapolated CM at the end of the main stance phase illustrate that the younger adults were able to change the components of the extrapolated CM during stance independent of body lean angle. During the main stance phase, the younger compared to the older adults showed a higher time duration and greater average horizontal ground reaction force (Table 4 and Fig. 5). In this way, young adults were able to reduce horizontal CM velocity and in consequence the position of the extrapolated CM during the main stance phase to a greater degree than the older adults. In the literature (Karamanidis and Arampatzis, 2006; Narici et al., 2005; Savelberg and Meijer, 2004), it is well accepted that the aging process is associated with a degeneration of the musculoskeletal system. In accordance with those findings, the older in comparison to the younger adults examined in our study had lower triceps surae and quadriceps femoris muscle strength, and an increase in strain of the quadriceps femoris tendon and aponeurosis at a given tendon force (Fig. 3). In order to examine whether the observed differences in margin of stability at the end of the main stance phase between older and younger adults were associated to the age-related MTU weakness we used a multiple regression analysis including the variables of the MTU’s showing an age affect. We found that triceps surae and quadriceps femoris muscle strength, and quadriceps femoris tendon stiffness contributed from 35% to 55% to the margin of stability at the end of the main stance phase (R = 0.588 for the small lean angle; R = 0.742 for the large lean angle), confirming that legextensor MTU capacities partly affect the control of dynamic stability after falls. The strong relationship (r2 = 0.910 and 0.916 for the small and large lean angles, respectively, Fig. 6) between the margin of stability at touchdown and at the end of the main stance phase showed that the locomotion strategy employed before touchdown, limits the postural corrections during the stance phase after a forward fall. During the phase until touchdown, the main postural correction of the human system which aimed to improve the margin of dynamic stability was the increase of the base of support. It seems that the central nervous system senses the state of the dynamic stability (i.e. the boundary of the base of support in relation to the position of the extrapolated CM) of the musculoskeletal system before and after release and attempts to adjust its motor commands in an appropriate manner for a successful balance correction. However, the exact degree of the dynamic stability (i.e. margin of stability) at touchdown is not known for the human system during the swing phase of the recovery limb. Therefore, we can argue that the main mechanism used to regain dynamic stability after a loss of balance in the anterior direction (i.e. increased base of support) is mediated before touchdown by motor plans which aim to keep the extrapolated CM inside the boundary of the base of support.
Model predictions (Wu et al., 2007) have shown that the muscle strength of the leg affects minimal step length required for stability after a forward balance loss (the lower the muscle strength, the greater the minimal step length required) because the muscles have to generate mechanical work during the following stance phase, absorbing the mechanical energy of the body. Therefore, it might be suggested that some older adults should move from a stable body position at touchdown to an unstable position during the following stance phase due to their lower MTU capacities. However, our results in Fig. 4 demonstrated that almost all younger and older participants (except one younger and one older adult) having a stable body position at touchdown (positive margin of stability) were able to remain within the base of support during the following stance phase. Therefore, it can be confirmed that, given a stable position at touchdown, the MTU capacities of the participants were sufficient to maintain dynamic stability during the following stance phase. In conclusion, our findings demonstrate that the main mechanism by which humans regain balance after forward falls is the increase of the base of support during the swing phase of the recovery limb in a feedforward control manner. Older adults have deficits in using mechanisms responsible for the control of dynamic stability (i.e. lower increase of the base of support in relation to the extrapolated CM) after a sudden perturbation, leading to a higher risk of falls. Moreover, we have been able to confirm that these age-related deficits in stability control are partly associated with the lower muscle strength and tendon stiffness in their leg-extensor MTUs. Combining the above findings, therefore, it is reasonable to propose that, to reduce risk of falls, older individuals may benefit from leg-extensor-MTU strengthening programs as well as from interventions exercising the mechanisms responsible for dynamic stability. Acknowledgements We thank Martin Ku¨sel-Feldker, Thomas Fo¨rster and their teams for the technical assistance and their active support through the entire project. References Arampatzis A, Karamanidis K, DeMonte D, Stafilidis S, Morey-Klapsing G, Bru¨ggemann G-P. Differences between measured and resultant joint moments during voluntary and artificially elicited isometric knee extension contractions. Clin Biomech 2004;19:277–83. Arampatzis A, Morey-Klapsing G, Karamanidis K, DeMonte D, Stafilidis S, Bru¨ggemann G-P. Differences between measured and resultant joint moments during isometric contractions at the ankle joint. J Biomech 2005a;38:885–92. Arampatzis A, Stafilidis S, DeMonte G, Karamanidis K, Morey-Klapsing G, Bru¨ggemann G-P. Strain and elongation of the human gastrocnemius tendon and aponeurosis during maximal plantarflexion effort. J Biomech 2005b;38:833–41. Baker SP, Harvey AH. Fall injuries in the elderly. Clin Geriatr Med 1985;1:501–12.
K. Karamanidis et al. / Journal of Electromyography and Kinesiology 18 (2008) 980–989 Baratta R, Solomonow M, Zhou BH, Letson D, Chuinard R, D’Ambrosia R. Muscular coactivation. The role of the antagonist musculature in maintaining knee stability. Am J Sports Med 1988;16:113–22. Blake AJ, Morgan K, Bendall MJ, Dallosso H, Ebrahim SB, Arie TH, et al.. Falls by elderly people at home: prevalence and associated factors. Age Ageing 1988;17:365–72. Cumming RG, Klineberg RJ. Fall frequency and characteristics and the risk of hip fractures. J Am Geriatr Soc 1994;42:774–8. Dempster WT, Gabel WC, Felts WJ. The anthropometry of the manual work space for the seated subject. Am J Phys Anthropol 1959;17:289–317. Herzog W, Read LJ. Lines of action and moment arms of the major forcecarrying structures crossing the human knee joint. J Anat 1993;182:213–30. Hof AL. The equations of motion for a standing human reveal three mechanisms for balance. J Biomech 2007;40:451–7. Hof AL, Gazendam MG, Sinke WE. The condition for dynamic stability. J Biomech 2005;38:1–8. Hsiao ET, Robinovitch SN. Biomechanical influences on balance recovery by stepping. J Biomech 1999;32:1099–106. Karamanidis K, Arampatzis A. Mechanical and morphological properties of human quadriceps femoris and triceps surae muscle–tendon unit in relation to aging and running. J Biomech 2006;39:406–17. Karamanidis K, Arampatzis A. Age-related degeneration in leg-extensor muscle–tendon units decreases recovery performance after a forward fall: compensation with running experience. Eur J Appl Physiol 2007;99:73–85. Mackey DC, Robinovitch SN. Mechanisms underlying age-related differences in ability to recover balance with the ankle strategy. Gait Posture 2006;23:59–68. Mademli L, Arampatzis A, Morey-Klapsing G, Bruggemann GP. Effect of ankle joint position and electrode placement on the estimation of the antagonistic moment during maximal plantarflexion. J Electromyogr Kinesiol 2004;14:591–7. Madigan ML, Lloyd EM. Age-related differences in peak joint torques during the support phase of single-step recovery from a forward fall. J Gerontol A Biol Sci Med Sci 2005;60:910–4. Maganaris CN, Baltzopoulos V, Sargeant AJ. Changes in Achilles tendon moment arm from rest to maximum isometric plantarflexion: in vivo observations in man. J Physiol 1998;510:977–85. Narici MV, Maganaris C, Reeves N. Myotendinous alterations and effects of resistive loading in old age. Scand J Med Sci Sports 2005;15:392–401. Pijnappels M, Bobbert MF, van Dieen JH. Control of support limb muscles in recovery after tripping in young and older subjects. Exp Brain Res 2005a;160:326–33. Pijnappels M, Bobbert MF, van Dieen JH. Push-off reactions in recovery after tripping discriminate young subjects, older non-fallers and older fallers. Gait Posture 2005b;21:388–94. Pijnappels M, Bobbert MF, van Dieen JH. How early reactions in the support limb contribute to balance recovery after tripping. J Biomech 2005c;38:627–34. Savelberg HH, Meijer K. The effect of age and joint angle on the proportionality of extensor and flexor strength at the knee joint. J Gerontol A Biol Sci Med Sci 2004;59:1120–8. Simoneau M, Corbeil P. The effect of time to peak ankle torque on balance stability boundary: experimental validation of a biomechanical model. Exp Brain Res 2005;165:217–28.
989
Stafilidis S, Karamanidis K, Morey-Klapsing G, DeMonte G, Bru¨ggemann G-P, Arampatzis A. Strain and elongation of the vastus lateralis aponeurosis and tendon in vivo during maximal isometric contraction. Eur J Appl Physiol 2005;94:317–22. Thelen DG, Wojcik LA, Schultz AB, Ashton-Miller JA, Alexander NB. Age differences in using a rapid step to regain balance during a forward fall. J Gerontol A Biol Sci Med Sci 1997;52:M8–M13. Wojcik LA, Thelen DG, Schultz AB, Ashton-Miller JA, Alexander NB. Age and gender differences in single-step recovery from a forward fall. J Gerontol A Biol Sci Med Sci 1999;54:M44–50. Wojcik LA, Thelen DG, Schultz AB, Ashton-Miller JA, Alexander NB. Age and gender differences in peak lower extremity joint torques and ranges of motion used during single-step balance recovery from a forward fall. J Biomech 2001;34:67–73. Wu M, Ji L, Jin D, Pai YC. Minimal step length necessary for recovery of forward balance loss with a single step. J Biomech 2007;40: 1559–66. Kiros Karamanidis, received his PhD at the German Sport University of Cologne in 2006. His main research interests are in the field of adaptation of aging muscles and its effect on gait mechanics focusing on the prevention of falls in the elderly.
Adamantios Arampatzis received his PhD at the German Sport University of Cologne in 1995. He is the head of the research group focusing on the neuromechanics of the human musculoskeletal system at the Institute of Biomechanics and Orthopaedics from the German Sport University Cologne. Among his research interests are the adaptation potential of the human system to physical activity and the influence of the neuromechanical capacity of the musculoskeletal system on motor task behaviour during daily and sport activities.
Lida Mademli graduated in sport science from the Aristotle University of Thessaloniki (Greece) in 2002. The same year she moved to Cologne (Germany), where she joined the institute of Biomechanics and Orthopaedics at the German Sport University of Cologne. She is enjoying a grant for making her PhD thesis on the agerelated effects of fatigue on the neuromechanical properties of the muscle–tendon unit and the postural stability after sudden perturbations.