- Email: [email protected]
Three-dimensional kinematics and trunk muscle myoelectric activity in the elderly spine – a database compared to young people
Clinical Biomechanics 14 (1999) 389±395
Three-dimensional kinematics and trunk muscle myoelectric activity in the elderly spine ± a database compared to young people Stuart M. McGill *, Vanessa R. Yingling, John P. Peach Occupational Biomechanics Laboratory, Faculty of Applied Health Sciences, Department of Kinesiology, University of Waterloo, Waterloo, Ont., CanadaN2L 3G1 Received 29 May 1998; accepted 8 December 1998
Abstract Objective. To establish a database of lumbar spine kinematics and muscle activation pro®les for a healthy elderly population. Design. Spine motion parameters and muscle activation pro®les of the elderly were identi®ed and quanti®ed in part by comparison with an existing database of younger people. Background. Databases are often used as a benchmark to establish what is ``normal'', or for developing appropriate exercise programs or diagnostic methodologies. Methods. Twelve people (average age 69 yr) performed full range of motion movements about the ¯exion, lateral bending and axial twist axes of the lumbar spine. Fourteen torso muscles (7 per side) were monitored with electromyographic electrodes and the signals normalized to maximum voluntary eorts. Results. The elderly were slower moving, and had a reduced range of motion in full ¯exion and lateral bend but not in axial twist. Furthermore there was more coupled motion evident in the twisting eorts of the elderly. Abdominal muscles appeared to become more active, earlier in the lateral bending movement. Conclusions. Loss of motion is normal with age together with subtle changes in the way the spine moves and groups of muscles are recruited around the torso. Relevance Quanti®cation of ``normal'' changes associated with aging will enable clinicians and scientists to distinguish normal changes from pathology. Ó 1999 Elsevier Science Ltd. All rights reserved. Keywords: Electromyography; Elderly; Lumbar; Database; Kinematics
1. Introduction
2. Methods
Many studies documenting the range of motion in the human lumbar spine have shown a decline with age [1± 4]. Thus, what might be classi®ed as being ``normal'' for one age group will be ``abnormal'' for another. Given the establishment of databases for clinical comparisons, consisting of spine kinematics and muscle activation pro®les (e.g. Refs. [5, 6]) and the apparent dependancy on age, the purpose of this work was to compile a database of three-dimensional dynamic lumbar spine kinematics, and the associated myoelectric responses of 14 trunk muscles in a healthy elderly population.
Twelve subjects participated in this study: seven females (mean age: 69 (SD, 3.5) yr, mean height: 164 (SD, 7.7) cm, mass: 65.3 (SD, 8) kg and ®ve males (age: 68.8 (SD, 5) yr, height: 178.8 (SD, 6.6) cm, mass: 78.3 (SD,11.6) kg). The participants were in good physical condition, with no previous history of disabling low back injury, or recent recurrent pain. All participants signed information and consent forms that were approved by the University of Waterloo Oce of Human Research Ethics Committee. Each participant performed a set of tasks consisting of maximal isometric exertion eorts (MVC), intended to produce the largest amplitudes of myoelectric activity (EMG) from the selected trunk muscles. These maximal eorts provided a basis for normalisation of the EMG
*
Corresponding author. E-mail: [email protected]
0268-0033/99/$ ± see front matter Ó 1999 Elsevier Science Ltd. All rights reserved. PII: S 0 2 6 8 - 0 0 3 3 ( 9 8 ) 0 0 1 1 1 - 9
390
S.M. McGill et al. / Clinical Biomechanics 14 (1999) 389±395
signals. Three basic isometric restraint strategies (developed and described by McGill [7]) were employed. The ®rst strategy consisted of the participant sitting on a padded table as if to perform a bent-knee sit-up. While their feet were held down with a strong Velcro strap and their hands were placed across their chest, the participants were instructed to attempt to ``sit-up'' with maximum eort (to recruit their abdominal muscles). The second strategy consisted of securing the participantÕs pelvis in a restraint jig while in a semi-sit posture with a strap around their back and secured anteriorly. The participant was instructed to extend their torso against the strap in an attempt to maximally recruit their back extensor muscles. During the third task, the participants sat on a padded table with their arms held up in the air, slightly in front and above their head. Maximal eort to adduct and retract the arms was performed to activate the latissimus dorsi muscles. The EMG signals were full-wave recti®ed and low-pass ®ltered (using the identical process for the functional trials) and the highest peak of these processed was taken as the MVC during these strategies for subsequent normalization of signals. The participants then performed three dierent movement tasks to quantify and contrast function in the elderly spine. Each task was performed three times consecutively with ample rest time between trials in order to avoid fatigue. The experimental conditions were de®ned as follows: Lateral bending trials: The participants were positioned with their feet shoulder width apart, their knees slightly bent and their arms hanging freely to their side. They were then instructed to bend laterally to their right until full lateral ¯exion. After pausing for a moment at full lateral ¯exion, they rose to an upright stance. This motion was also repeated three times to the left side in a separate trial. Axial twisting trials: The participants were positioned with their feet shoulder width apart, their knees slightly bent and their hands held against their chest to ensure that the upper body moved as a single unit. The participants then rotated (twisted) to their right until full twist was reached. After a brief pause at full twist, the participant returned to neutral stance. Even though the participantÕs pelvis was not immobilised, the experimenter constantly reminded the participant to twist at the waist and not turn at the hips. The motion was then repeated three times to the left side in a separate trial. Flexion trials: The participants were positioned with their feet shoulder width apart, their knees slightly bent and their arms hanging freely to their side. The participantÕs ¯exed their trunk forward in the sagittal plane until full ¯exion was reached. The participants were instructed to pause for a moment at full ¯exion before returning to an upright posture.
Flexion trials were repeated at two velocities, the ®rst chosen by the participant and the second at a predetermined pace. The free velocity trials allowed the participant to perform the movement at a velocity that was perceived as ``comfortable''. During the paced velocity trials, the subject was led by an audio guide signal to maintain a constant velocity of movement. The audio signal was variable in tone to represent dierent levels of the range of motion (RoM) and the timing was controlled by the computer. The pace velocity used was 20°/s. There were two dierent load conditions for the ¯exion trials; a no load condition (0 kg) and a loaded condition under which the participant would hold a load in hands which was comfortable (ranging from 2.2 to 10 kg). The spine kinematic data was collected using a 3 S P A C E I S O T R A K (Polhemus Navigation Sciences, McDonnel Douglas Electronics Company, Colchester, VT, USA). The device consisted of a magnetic source placed over the sacrum and secured with a strap system and an inductive sensor placed on the skin over the spinous process of the 12th thoracic vertebra. The source generated a low frequency magnetic ®eld which induces current in the sensor depending upon its threedimensional orientation. Three-dimensional spine motion was collected at a sampling frequency of 20.5 Hz and normalised relative to a neutral upright standing posture and synchronised with the EMG signals. Raw EMG signals were collected using disposable bipolar Ag±AgCl surface electrodes with a center-to-center spacing of 3 cm over the following muscles: Rectus Abdominis (3 cm lateral to the umbilicus), External Oblique (15 cm lateral to the umbilicus), Internal Oblique (below the external oblique and just superior to the inguinal ligament), Latissimus Dorsi (lateral to the T9 spinous process over the muscle belly), Thoracic Erector Spinae (5 cm lateral to the T9 spinous process), Lumbar Erector Spinae (3 cm lateral to L3 spinous process) and Multi®dus (1 cm lateral to the L5 spinous process). All raw EMG signals were pre®ltered to produce a band width of 10±500 Hz, and ampli®ed with a dierential ampli®er (common mode rejection ratio of 80 dB at 60 Hz) to produce signals of approximately 4V, and A/D converted at 1024 Hz (W A T S C O P E A/D Converter, 12 bit, Northern Digital Inc., Waterloo, Canada). The EMG signals were full-wave recti®ed and lowpass ®ltered (Butterworth ± cuto frequency 3 Hz) and normalised to the maximum electrical activity observed in the MVC trials (previously described). The three trials of each subject were then time-normalised to a common length (from 0% to 100% of each movement task) in order to ensemble average the trials to form the muscle activity pro®les. The peak range of motion, peak velocity and acceleration and mean velocity and acceleration were compared to a database comprised
S.M. McGill et al. / Clinical Biomechanics 14 (1999) 389±395
of younger participants using a Students t-test (P < 0.05). 3. Results The three-dimensional kinematics and muscle activation pro®les forming the data base are shown in Figs. 1±5. Some key features are summarised below together with some discrepancies when compared with a younger database (mean age 21 3.4 yr) [5]. Lumbar kinematics: The elderly group exhibited a signi®cantly lower peak displacement value for full ¯exion (Table 1) lateral bend but not axial twist (Fig. 1). In fact the ¯exion movement (approximately 50°) was approximately 70% of that of the younger group. The
391
smaller range of ¯exion motion in the elderly was performed at a slower velocity than the younger group although the peak velocities during lateral bending and axial twist movements were not dierent. However the average velocities over the ®rst 30° of movement, and last 30° of movement were lower in the elderly in both the ¯exion and lateral bending tasks but not during twisting. These results were observed despite the attempt to pace some ¯exion trials. Interestingly, there were no dierences in peak acceleration during any movement task. There were apparent dierences in the amount of motion in the minor axes, or coupled motion, during some movements. For example, the principal motion of ¯exion had very little apparent coupling of motion in the lateral bend or twist axes which is sometimes seen in patients [6]. During twisting there was more apparent coupled motion in the ¯exion axis in the elderly (Fig. 2), however, we do acknowledge the limitations of quantifying coupled motion with the 3 S P A C E [8]. 3.1. Muscle activation The following EMG data are presented normalized to the EMG amplitude measured during maximal eort tasks. It should be noted that maximal exertion eorts in the elderly were probably not as ``maximal'' as with
Fig. 1. Range of motion (mean and standard deviation) of the elderly population in the three planes of motion (compared to the younger data base).
Fig. 2. During axial twisting eorts there was relatively more coupled motion in the ¯exion axis of the elderly, particularly about the ¯exion± extension axis.
*
49
50
50
17
20
12
17
Flex loaded, paced
Flex loaded, no pace
Flex, no load, paced
Latbend, right
Latbend, left
Axial twist, right
Axial twist, left
Denotes dierence (P level).
48
Flex, no load, no pace
(0.78)
(0.50)
(0.0008)
(0.0001)
(0.0001)
(0.0001)
(0.0001)
(0.0001)
16
15
29
29
71
73
73
71
16
9
23
19
47
41
42
38
(0.43)
(0.0542)
(0.569)
(0.0004)
(0.004)
(0.0001)
(0.0001)
(0.0001)
13
14
21
28
62
65
63
66
Y
E
E
Y
Peak velocity (°/s)
Peak displacement (°)
127
154
141
166
490
294
447
381
E
(0.1926)
(0.31)
(0.492)
(0.1313)
(0.089)
(0.79)
(0.15)
(0.287)
93
88
125
133
322
301
340
331
Y
Peak acceleration (°/s2 )
+ve
Table 1 Kinematics for each principle axis of motion for the elderly spine (compared to a young population)
ÿ98
ÿ80
ÿ337
ÿ102
ÿ373
ÿ284
ÿ409
ÿ308
E
ÿve
(0.5266)
(0.7649)
(0.0417)
(0.0545)
(0.235)
(0.08)
(0.17)
(0.74)
ÿ88
ÿ87
ÿ126
ÿ136
ÿ291
ÿ348
ÿ307
ÿ326
Y
9
ÿ6
12
ÿ11
28
26
25
26
E
(0.99)
(0.2047)
(0.026)
(0.0001)
(0.0001)
(0.0001)
(0.0001)
(0.0001)
Average velocity
First 30°
9
ÿ8
15
ÿ17
43
43
43
42
Y
ÿ9
5.2
ÿ12
12
ÿ31
ÿ28
ÿ31
ÿ30
E
Last 30°
(0.616)
(0.0281)
(0.0104)
(0.0002)
(0.0008)
(0.0001)
(0.0001)
(0.0001)
ÿ9
8.7
ÿ18
19
ÿ40
ÿ46
ÿ42
ÿ43
Y
392 S.M. McGill et al. / Clinical Biomechanics 14 (1999) 389±395
S.M. McGill et al. / Clinical Biomechanics 14 (1999) 389±395
younger subjects ± resulting in the larger normalized activation levels. Flexion task: The back extensor muscle activity similar in shape to the younger group, which was characterised by an asymmetrical bimodal curve with an area of diminished activity at full ¯exion and larger muscle activity occurring during the extension phase (Fig. 3). Activity of all muscles generally appeared to be higher, particularly in oblique muscles, however this may have been simply due to the lower normalization eorts of the elderly. Axial twist task: The predominant muscles during twist (Fig. 4), external and internal oblique muscles, the latissimus dorsi and the upper and lower erector spinae, were the same active muscles as in the young group although the elderly activated these muscles sooner. For example, while younger people exhibited peak muscle activation amplitudes within 10% of the point of peak
393
twist (50% of cycle in ®gures) the elderly exhibited peak activity in rectus abdominis external and internal oblique, latissimus dorsi and upper erector spinae at approximately 35% of cycle (well before peak twist was reached at 50%). Interestingly the elderly exhibited lower activation levels in the internal oblique. Lateral bend task: Rectus abdominis and the obliques were activated at a more constant level during lateral bending in the elderly (for example in young external oblique approximately 2±12% MVC compared to a range of 7±13% MVC in the elderly) (Fig. 5). 4. Discussion The purpose of this study was to develop a database of ``elderly'' lumbar spine kinematics and myoelectric activity. Features of the elderly spines were identi®ed
Fig. 3. Mean (and standard deviation) normalized activation of the torso muscles (on the right side of the body) in both young and elderly people during the ¯exion, no load, free paced condition.
394
S.M. McGill et al. / Clinical Biomechanics 14 (1999) 389±395
Fig. 4. Mean (and standard deviation) normalized activation of the right side torso muscles during axial twisting to the right. Elderly subjects appear to activate the internal oblique to lower levels than the younger subjects.
with comparison to a database comprised of younger healthy participants. It should be noted that the database illustrates the average curves for the healthy elderly population. However, the ensemble average process is good for illustrating mean patterns but buries the random variability between subjects. However, this variance is indicated with an increase in the standard deviation about the mean (shown over the movement history in Figs. 3±5). EMG normalization to a maximal eort was chosen given its physiological basis ± these levels were the maximum that the subjects were voluntarily willing to perform at that time. We strongly suspect that the elderly were capable of higher maximal levels but that ``reserved'' eorts resulted from either the fear of injury or wisdom to simply conserve themselves. Nonetheless, while this strategy resulted in overall higher normalized activation levels, this was chosen over a standard sub-
maximal task where a feel of physiological maximum would have been completely lost. In summary, features of the ``elderly'' database were observed with comparison to a database of ``younger'' participants, speci®cally: a decrease in the range of motion for ¯exion and lateral bending movements (but not axial twist); a reduced velocity of all these same movements; reserved eorts to produce maximal contractions; and some muscles appeared to be activated earlier in the motion (particularly the abdominals in lateral bending) in the elderly. The observations of earlier activation and increased co-contraction suggests that elderly people may be seeking greater stabilization either for general balance or for actual spine stabilization, or both. It is hoped that by quanti®cation of the ``normal'' changes associated with aging that low back pathologies will be distinguishable from normal aging conditions.
S.M. McGill et al. / Clinical Biomechanics 14 (1999) 389±395
395
Fig. 5. Mean (and standard deviation) normalized activation of the right side torso muscles during lateral bending to the left.
References [1] Hilton RC, Ball J, Benn RT. In-vitro mobility of the lumbar spine. Ann Rheum Dis 1979;38:378±383. [2] Hilton RC. Systematic studies of spinal mobility and SchormlÕs nodes. In: Jayson MIV, editor. The Lumbar Spine and Back Pain. 2nd ed. Pitman, London, 1980;115±134. [3] Taylor J, Twomey L. Sagittal and horizontal plane movement of the human lumbar vertebral column in cadavers and in living. Rheumatol Rehab 1980;19:223±232. [4] Tanz SS. Motion of the lumbar spine: a roentgenolic study. Am J Roentgenol 1953;69:399±412.
[5] Peach JP, Sutarno C, McGill SM. 3D Kinematics and trunk muscle myoelectric activity of the asymptomatic young lumbar spine ± A data base. Arch Phys Med Rehab 1998;79(6):663±669. [6] Peach JP, McGill SM. Kinematics and trunk muscle myoelectric activity in the chronic low back pain patient. Phys Ther, submitted. [7] McGill SM. Electromyographic activity of the abdominal and low back musculature during the generation of isometric and dynamic axial trunk torque: Implications for lumbar mechanics. J. Orthop Res. 1991;9(1):91±103. [8] McGill SM, Cholewicki J, Peach JP. Methodological considerations for using inductive sensors (3 - S P A C E I S O T R A K ) to monitor 3-D orthopaedic joint motion. Clin Biomech 1997;12:190±194.
Three-dimensional kinematics and trunk muscle myoelectric activity in the elderly spine ± a database compared to young people Stuart M. McGill *, Vanessa R. Yingling, John P. Peach Occupational Biomechanics Laboratory, Faculty of Applied Health Sciences, Department of Kinesiology, University of Waterloo, Waterloo, Ont., CanadaN2L 3G1 Received 29 May 1998; accepted 8 December 1998
Abstract Objective. To establish a database of lumbar spine kinematics and muscle activation pro®les for a healthy elderly population. Design. Spine motion parameters and muscle activation pro®les of the elderly were identi®ed and quanti®ed in part by comparison with an existing database of younger people. Background. Databases are often used as a benchmark to establish what is ``normal'', or for developing appropriate exercise programs or diagnostic methodologies. Methods. Twelve people (average age 69 yr) performed full range of motion movements about the ¯exion, lateral bending and axial twist axes of the lumbar spine. Fourteen torso muscles (7 per side) were monitored with electromyographic electrodes and the signals normalized to maximum voluntary eorts. Results. The elderly were slower moving, and had a reduced range of motion in full ¯exion and lateral bend but not in axial twist. Furthermore there was more coupled motion evident in the twisting eorts of the elderly. Abdominal muscles appeared to become more active, earlier in the lateral bending movement. Conclusions. Loss of motion is normal with age together with subtle changes in the way the spine moves and groups of muscles are recruited around the torso. Relevance Quanti®cation of ``normal'' changes associated with aging will enable clinicians and scientists to distinguish normal changes from pathology. Ó 1999 Elsevier Science Ltd. All rights reserved. Keywords: Electromyography; Elderly; Lumbar; Database; Kinematics
1. Introduction
2. Methods
Many studies documenting the range of motion in the human lumbar spine have shown a decline with age [1± 4]. Thus, what might be classi®ed as being ``normal'' for one age group will be ``abnormal'' for another. Given the establishment of databases for clinical comparisons, consisting of spine kinematics and muscle activation pro®les (e.g. Refs. [5, 6]) and the apparent dependancy on age, the purpose of this work was to compile a database of three-dimensional dynamic lumbar spine kinematics, and the associated myoelectric responses of 14 trunk muscles in a healthy elderly population.
Twelve subjects participated in this study: seven females (mean age: 69 (SD, 3.5) yr, mean height: 164 (SD, 7.7) cm, mass: 65.3 (SD, 8) kg and ®ve males (age: 68.8 (SD, 5) yr, height: 178.8 (SD, 6.6) cm, mass: 78.3 (SD,11.6) kg). The participants were in good physical condition, with no previous history of disabling low back injury, or recent recurrent pain. All participants signed information and consent forms that were approved by the University of Waterloo Oce of Human Research Ethics Committee. Each participant performed a set of tasks consisting of maximal isometric exertion eorts (MVC), intended to produce the largest amplitudes of myoelectric activity (EMG) from the selected trunk muscles. These maximal eorts provided a basis for normalisation of the EMG
*
Corresponding author. E-mail: [email protected]
0268-0033/99/$ ± see front matter Ó 1999 Elsevier Science Ltd. All rights reserved. PII: S 0 2 6 8 - 0 0 3 3 ( 9 8 ) 0 0 1 1 1 - 9
390
S.M. McGill et al. / Clinical Biomechanics 14 (1999) 389±395
signals. Three basic isometric restraint strategies (developed and described by McGill [7]) were employed. The ®rst strategy consisted of the participant sitting on a padded table as if to perform a bent-knee sit-up. While their feet were held down with a strong Velcro strap and their hands were placed across their chest, the participants were instructed to attempt to ``sit-up'' with maximum eort (to recruit their abdominal muscles). The second strategy consisted of securing the participantÕs pelvis in a restraint jig while in a semi-sit posture with a strap around their back and secured anteriorly. The participant was instructed to extend their torso against the strap in an attempt to maximally recruit their back extensor muscles. During the third task, the participants sat on a padded table with their arms held up in the air, slightly in front and above their head. Maximal eort to adduct and retract the arms was performed to activate the latissimus dorsi muscles. The EMG signals were full-wave recti®ed and low-pass ®ltered (using the identical process for the functional trials) and the highest peak of these processed was taken as the MVC during these strategies for subsequent normalization of signals. The participants then performed three dierent movement tasks to quantify and contrast function in the elderly spine. Each task was performed three times consecutively with ample rest time between trials in order to avoid fatigue. The experimental conditions were de®ned as follows: Lateral bending trials: The participants were positioned with their feet shoulder width apart, their knees slightly bent and their arms hanging freely to their side. They were then instructed to bend laterally to their right until full lateral ¯exion. After pausing for a moment at full lateral ¯exion, they rose to an upright stance. This motion was also repeated three times to the left side in a separate trial. Axial twisting trials: The participants were positioned with their feet shoulder width apart, their knees slightly bent and their hands held against their chest to ensure that the upper body moved as a single unit. The participants then rotated (twisted) to their right until full twist was reached. After a brief pause at full twist, the participant returned to neutral stance. Even though the participantÕs pelvis was not immobilised, the experimenter constantly reminded the participant to twist at the waist and not turn at the hips. The motion was then repeated three times to the left side in a separate trial. Flexion trials: The participants were positioned with their feet shoulder width apart, their knees slightly bent and their arms hanging freely to their side. The participantÕs ¯exed their trunk forward in the sagittal plane until full ¯exion was reached. The participants were instructed to pause for a moment at full ¯exion before returning to an upright posture.
Flexion trials were repeated at two velocities, the ®rst chosen by the participant and the second at a predetermined pace. The free velocity trials allowed the participant to perform the movement at a velocity that was perceived as ``comfortable''. During the paced velocity trials, the subject was led by an audio guide signal to maintain a constant velocity of movement. The audio signal was variable in tone to represent dierent levels of the range of motion (RoM) and the timing was controlled by the computer. The pace velocity used was 20°/s. There were two dierent load conditions for the ¯exion trials; a no load condition (0 kg) and a loaded condition under which the participant would hold a load in hands which was comfortable (ranging from 2.2 to 10 kg). The spine kinematic data was collected using a 3 S P A C E I S O T R A K (Polhemus Navigation Sciences, McDonnel Douglas Electronics Company, Colchester, VT, USA). The device consisted of a magnetic source placed over the sacrum and secured with a strap system and an inductive sensor placed on the skin over the spinous process of the 12th thoracic vertebra. The source generated a low frequency magnetic ®eld which induces current in the sensor depending upon its threedimensional orientation. Three-dimensional spine motion was collected at a sampling frequency of 20.5 Hz and normalised relative to a neutral upright standing posture and synchronised with the EMG signals. Raw EMG signals were collected using disposable bipolar Ag±AgCl surface electrodes with a center-to-center spacing of 3 cm over the following muscles: Rectus Abdominis (3 cm lateral to the umbilicus), External Oblique (15 cm lateral to the umbilicus), Internal Oblique (below the external oblique and just superior to the inguinal ligament), Latissimus Dorsi (lateral to the T9 spinous process over the muscle belly), Thoracic Erector Spinae (5 cm lateral to the T9 spinous process), Lumbar Erector Spinae (3 cm lateral to L3 spinous process) and Multi®dus (1 cm lateral to the L5 spinous process). All raw EMG signals were pre®ltered to produce a band width of 10±500 Hz, and ampli®ed with a dierential ampli®er (common mode rejection ratio of 80 dB at 60 Hz) to produce signals of approximately 4V, and A/D converted at 1024 Hz (W A T S C O P E A/D Converter, 12 bit, Northern Digital Inc., Waterloo, Canada). The EMG signals were full-wave recti®ed and lowpass ®ltered (Butterworth ± cuto frequency 3 Hz) and normalised to the maximum electrical activity observed in the MVC trials (previously described). The three trials of each subject were then time-normalised to a common length (from 0% to 100% of each movement task) in order to ensemble average the trials to form the muscle activity pro®les. The peak range of motion, peak velocity and acceleration and mean velocity and acceleration were compared to a database comprised
S.M. McGill et al. / Clinical Biomechanics 14 (1999) 389±395
of younger participants using a Students t-test (P < 0.05). 3. Results The three-dimensional kinematics and muscle activation pro®les forming the data base are shown in Figs. 1±5. Some key features are summarised below together with some discrepancies when compared with a younger database (mean age 21 3.4 yr) [5]. Lumbar kinematics: The elderly group exhibited a signi®cantly lower peak displacement value for full ¯exion (Table 1) lateral bend but not axial twist (Fig. 1). In fact the ¯exion movement (approximately 50°) was approximately 70% of that of the younger group. The
391
smaller range of ¯exion motion in the elderly was performed at a slower velocity than the younger group although the peak velocities during lateral bending and axial twist movements were not dierent. However the average velocities over the ®rst 30° of movement, and last 30° of movement were lower in the elderly in both the ¯exion and lateral bending tasks but not during twisting. These results were observed despite the attempt to pace some ¯exion trials. Interestingly, there were no dierences in peak acceleration during any movement task. There were apparent dierences in the amount of motion in the minor axes, or coupled motion, during some movements. For example, the principal motion of ¯exion had very little apparent coupling of motion in the lateral bend or twist axes which is sometimes seen in patients [6]. During twisting there was more apparent coupled motion in the ¯exion axis in the elderly (Fig. 2), however, we do acknowledge the limitations of quantifying coupled motion with the 3 S P A C E [8]. 3.1. Muscle activation The following EMG data are presented normalized to the EMG amplitude measured during maximal eort tasks. It should be noted that maximal exertion eorts in the elderly were probably not as ``maximal'' as with
Fig. 1. Range of motion (mean and standard deviation) of the elderly population in the three planes of motion (compared to the younger data base).
Fig. 2. During axial twisting eorts there was relatively more coupled motion in the ¯exion axis of the elderly, particularly about the ¯exion± extension axis.
*
49
50
50
17
20
12
17
Flex loaded, paced
Flex loaded, no pace
Flex, no load, paced
Latbend, right
Latbend, left
Axial twist, right
Axial twist, left
Denotes dierence (P level).
48
Flex, no load, no pace
(0.78)
(0.50)
(0.0008)
(0.0001)
(0.0001)
(0.0001)
(0.0001)
(0.0001)
16
15
29
29
71
73
73
71
16
9
23
19
47
41
42
38
(0.43)
(0.0542)
(0.569)
(0.0004)
(0.004)
(0.0001)
(0.0001)
(0.0001)
13
14
21
28
62
65
63
66
Y
E
E
Y
Peak velocity (°/s)
Peak displacement (°)
127
154
141
166
490
294
447
381
E
(0.1926)
(0.31)
(0.492)
(0.1313)
(0.089)
(0.79)
(0.15)
(0.287)
93
88
125
133
322
301
340
331
Y
Peak acceleration (°/s2 )
+ve
Table 1 Kinematics for each principle axis of motion for the elderly spine (compared to a young population)
ÿ98
ÿ80
ÿ337
ÿ102
ÿ373
ÿ284
ÿ409
ÿ308
E
ÿve
(0.5266)
(0.7649)
(0.0417)
(0.0545)
(0.235)
(0.08)
(0.17)
(0.74)
ÿ88
ÿ87
ÿ126
ÿ136
ÿ291
ÿ348
ÿ307
ÿ326
Y
9
ÿ6
12
ÿ11
28
26
25
26
E
(0.99)
(0.2047)
(0.026)
(0.0001)
(0.0001)
(0.0001)
(0.0001)
(0.0001)
Average velocity
First 30°
9
ÿ8
15
ÿ17
43
43
43
42
Y
ÿ9
5.2
ÿ12
12
ÿ31
ÿ28
ÿ31
ÿ30
E
Last 30°
(0.616)
(0.0281)
(0.0104)
(0.0002)
(0.0008)
(0.0001)
(0.0001)
(0.0001)
ÿ9
8.7
ÿ18
19
ÿ40
ÿ46
ÿ42
ÿ43
Y
392 S.M. McGill et al. / Clinical Biomechanics 14 (1999) 389±395
S.M. McGill et al. / Clinical Biomechanics 14 (1999) 389±395
younger subjects ± resulting in the larger normalized activation levels. Flexion task: The back extensor muscle activity similar in shape to the younger group, which was characterised by an asymmetrical bimodal curve with an area of diminished activity at full ¯exion and larger muscle activity occurring during the extension phase (Fig. 3). Activity of all muscles generally appeared to be higher, particularly in oblique muscles, however this may have been simply due to the lower normalization eorts of the elderly. Axial twist task: The predominant muscles during twist (Fig. 4), external and internal oblique muscles, the latissimus dorsi and the upper and lower erector spinae, were the same active muscles as in the young group although the elderly activated these muscles sooner. For example, while younger people exhibited peak muscle activation amplitudes within 10% of the point of peak
393
twist (50% of cycle in ®gures) the elderly exhibited peak activity in rectus abdominis external and internal oblique, latissimus dorsi and upper erector spinae at approximately 35% of cycle (well before peak twist was reached at 50%). Interestingly the elderly exhibited lower activation levels in the internal oblique. Lateral bend task: Rectus abdominis and the obliques were activated at a more constant level during lateral bending in the elderly (for example in young external oblique approximately 2±12% MVC compared to a range of 7±13% MVC in the elderly) (Fig. 5). 4. Discussion The purpose of this study was to develop a database of ``elderly'' lumbar spine kinematics and myoelectric activity. Features of the elderly spines were identi®ed
Fig. 3. Mean (and standard deviation) normalized activation of the torso muscles (on the right side of the body) in both young and elderly people during the ¯exion, no load, free paced condition.
394
S.M. McGill et al. / Clinical Biomechanics 14 (1999) 389±395
Fig. 4. Mean (and standard deviation) normalized activation of the right side torso muscles during axial twisting to the right. Elderly subjects appear to activate the internal oblique to lower levels than the younger subjects.
with comparison to a database comprised of younger healthy participants. It should be noted that the database illustrates the average curves for the healthy elderly population. However, the ensemble average process is good for illustrating mean patterns but buries the random variability between subjects. However, this variance is indicated with an increase in the standard deviation about the mean (shown over the movement history in Figs. 3±5). EMG normalization to a maximal eort was chosen given its physiological basis ± these levels were the maximum that the subjects were voluntarily willing to perform at that time. We strongly suspect that the elderly were capable of higher maximal levels but that ``reserved'' eorts resulted from either the fear of injury or wisdom to simply conserve themselves. Nonetheless, while this strategy resulted in overall higher normalized activation levels, this was chosen over a standard sub-
maximal task where a feel of physiological maximum would have been completely lost. In summary, features of the ``elderly'' database were observed with comparison to a database of ``younger'' participants, speci®cally: a decrease in the range of motion for ¯exion and lateral bending movements (but not axial twist); a reduced velocity of all these same movements; reserved eorts to produce maximal contractions; and some muscles appeared to be activated earlier in the motion (particularly the abdominals in lateral bending) in the elderly. The observations of earlier activation and increased co-contraction suggests that elderly people may be seeking greater stabilization either for general balance or for actual spine stabilization, or both. It is hoped that by quanti®cation of the ``normal'' changes associated with aging that low back pathologies will be distinguishable from normal aging conditions.
S.M. McGill et al. / Clinical Biomechanics 14 (1999) 389±395
395
Fig. 5. Mean (and standard deviation) normalized activation of the right side torso muscles during lateral bending to the left.
References [1] Hilton RC, Ball J, Benn RT. In-vitro mobility of the lumbar spine. Ann Rheum Dis 1979;38:378±383. [2] Hilton RC. Systematic studies of spinal mobility and SchormlÕs nodes. In: Jayson MIV, editor. The Lumbar Spine and Back Pain. 2nd ed. Pitman, London, 1980;115±134. [3] Taylor J, Twomey L. Sagittal and horizontal plane movement of the human lumbar vertebral column in cadavers and in living. Rheumatol Rehab 1980;19:223±232. [4] Tanz SS. Motion of the lumbar spine: a roentgenolic study. Am J Roentgenol 1953;69:399±412.
[5] Peach JP, Sutarno C, McGill SM. 3D Kinematics and trunk muscle myoelectric activity of the asymptomatic young lumbar spine ± A data base. Arch Phys Med Rehab 1998;79(6):663±669. [6] Peach JP, McGill SM. Kinematics and trunk muscle myoelectric activity in the chronic low back pain patient. Phys Ther, submitted. [7] McGill SM. Electromyographic activity of the abdominal and low back musculature during the generation of isometric and dynamic axial trunk torque: Implications for lumbar mechanics. J. Orthop Res. 1991;9(1):91±103. [8] McGill SM, Cholewicki J, Peach JP. Methodological considerations for using inductive sensors (3 - S P A C E I S O T R A K ) to monitor 3-D orthopaedic joint motion. Clin Biomech 1997;12:190±194.