Dynamic Stability in Elders: Momentum Control in Locomotor ADL

Transcription

1 Journal of Gerontology: MEDICAL SCIENCES Vol. 53A. No. 2. M126-M134 Copyright 1998 by The Cerontological Society of America Dynamic Stability in Elders: Momentum Control in Locomotor ADL Bradley K. Kaya, David E. Krebs, and Patrick O. Riley The Massachusetts General Hospital Biomotion Laboratory, Boston. Background. Momentum must be controlled in stable locomotor activities, including sit-to-stand and gait. The relationship of momentum control and balance maintenance in elders or in a balance-impaired population has not been studied. Although decreased locomotor speed has long been reported among elders, the literature is lacunar concerning the mechanical mechanisms underlying this slowing. The purpose of this study was to describe the whole body and upper body linear and angular momentum for healthy elders during sit-to-stand and gait and compare them to a group of balance-impaired elders who have bilateral vestibular hypofunction (BVH). Methods. Ten elders with BVH were matched to 10 healthy elders aged Linear and angular momentum were calculated for sit-to-stand and for free speed and paced gait. s and 95% confidence intervals were used to compare groups. Results. Elders with BVH used significantly less linear and angular momentum to rise from a chair than healthy elders and showed excessive lateral momentum during gait, despite walking at a slower velocity. Conclusions. Healthy elders limit momentum generation by decreasing gait velocity, apparently because they lack sufficient strength or balance control to safely dissipate the momentum that a faster, less controlled gait engenders. Elders with BVH further limit momentum in locomotor activities to decrease their risk of falling, but are apparently unable to control lateral momentum during gait. Excessive lateral momentum in gait among balance-impaired elders leads to loss of balance, a frequent occurrence in this patient population. A LL human movement involves momentum. Although l\ the neuromuscular component of joint movement is the focus of much research, momentum has not been studied extensively, especially for daily activities. Momentum is the tendency of a moving body to remain in motion unless an opposing force is applied (1). Linear momentum is the product of mass and velocity. Angular momentum involves rotation about an axis and is the product of body inertia and angular velocity. Momentum can be used to generate movement for functional activities. Elderly people may use momentum as a compensatory strategy. For example, a person who lacks the lower extremity strength to rise to standing from a chair may rely on momentum to perform this activity independently. Likewise, in gait, if a person lacks sufficient lower extremity torque-generating capacity, upper body momentum may be used to maintain steady state gait (1-3). Momentum, however, must also be controlled to maintain stability. That is, the individual must have dynamic stability or the ability to control the whole body (WB) position and momentum during gait and other activities in which the center of gravity (CG) is displaced outside the base of support (4-6). Older adults and people with bilateral vestibular hypofunction (BVH) are two populations in which balance impairment can lead to functional limitations. More than one-third of community-living elders will fall or are at risk to fall, with loss of balance and dizziness as prevalent causes (7,8). Because of vestibular system pathology, BVH patients may be unstable during gait or with transitional movements such as sit-to-stand (9). They may report nausea, dizziness, and visual disturbances (blurred vision or oscillopsia) with these activities of daily living (10-12). Because falls tend to occur during activities of daily living (13), it is most appropriate to examine BVH patients during transitional movements. Several investigators have described sit-to-stand biomechanics (6,14-18). Pai and Rogers studied segmental contributions to momentum and how momentum relates to sitto-stand speed (19,20). Although vertical momentum increased as sit-to-stand speed increased, horizontal momentum values remained constant across natural and fast speeds (19-21). Pai and colleagues concluded that horizontal momentum was held constant as a movement control strategy to maintain upright balance (21,22). If alterations in horizontal momentum lead to instability during the transition from sitting to standing, particularly in patients with BVH, it is important to measure the whole body momentum generated during sit-to-stand as well as the momentum present after upright standing has been achieved. An overreliance on momentum may result in the inability to brake the momentum generated, leading to postural instability (21,22). Momentum generated during gait has not been studied extensively. Instability during gait is a significant problem for elders and BVH patients. The relationship of momentum and balance maintenance in a balance-impaired population has not been reported. Momentum analysis of sit-to- M126

2 DYNAMIC STABILITY M127 stand and gait may permit better understanding of the mechanisms underlying dynamic instability. Therefore, the purpose of this study is to: (a) describe whole body and upper body linear and angular momentum for healthy elders during sit-to-stand and gait; (b) compare these data to a matched group of elders with bilateral vestibular hypofunction. We hypothesize that linear and angular momentum will be lower in elders with BVH: braking and dissipating momentum should be easier in subjects with intact vestibular systems, and excess momentum is stability-threatening and thus is probably avoided in subject BVH-induced instability. METHODS Subjects Data from 20 subjects 60 years or older, cognitively intact, and independent in sit-to-stand and gait were analyzed for this study. All subjects provided written informed consent. Ten subjects (2 males, 8 females) had BVH but no other neuromusculoskeletal pathology. BVH patients had >3 below normal vestibulo-ocular reflex gains (ratio of slow-phase eye movement response to stimulus magnitude) at all frequencies (.01 to 1.0 Hz) of sinusoidal vertical axis rotation (SVAR) testing, and decreased hot and cold caloric responses measured by electronystagmography (ENG) (10,23). Ten healthy elders (1 male, 9 females) were age, height, and weight matched to the BVH group. There were no significant differences (p > 0.5) in age, height, and weight between the two groups (Table 1). Instrumentation Instrumentation has been described in detail previously (3,17,18,24). Kinematic data were collected by four Selspot II (Partille, Sweden) optoelectronic cameras with a viewing volume of about 2X2X2 meters. The cameras contain detectors that track the Selspot system's infrared light emitting diodes (LEDs), which are embedded in rigid arrays. Arrays were attached to 11 body segments: head, trunk, pelvis, arms (2), thighs (2), shanks (2), and feet (2). Each segment was modeled as a rigid body with six degrees of freedom (three translations and three rotations). Telemetered Rapid Automatic Computerized Kinematic (TRACK, Massachusetts Institute of Technology, Cambridge, MA) software was used to acquire and analyze each segment's kinematic data according to techniques described by Riley et al. (24). Accuracy in determining the LED array positions are ±1 mm for linear displacement and <1 degree for angular displacement (25). Kinetic data were collected with two Kistler (Winterthur, Switzerland) forceplates which have an accuracy equal to ±1% of full scale. Body segment mass, mass centers, and inertial parameters were estimated using regression equations (26,27). Torques, segmental and whole body linear and angular momentum were calculated using Newton-Euler inverse dynamics (15). Both kinematic and kinetic data were sampled at 150 Hz. Procedure All subjects wore shorts, were barefoot, and had the arrays strapped to the body. The protocol has been described in previous studies (2,3,28). Sit-to-stand. Subjects were seated on an armless, backless chair at a height equal to knee height (the distance from the tibiofemoral joint line to the floor). The anterior edge of the chair was 4 cm distal to the greater trochanter. Each foot was on a left and right forceplate with 10 cm between the medial borders of each foot and the ankle positioned in 18 degrees of dorsiflexion. The arms were folded, with hands grasping the elbows and kept against the chest. Subjects were instructed not to use their arms or move their feet from the initial position as they stood from the chair. Each trial was sampled for 7 seconds with 1- to 2-minute rest periods between trials. Gait. Subjects walked on a 10-meter walkway at two different speeds: a free gait (preferred pace) speed and a paced gait speed at a set cadence. For free gait trials, the subject was instructed to "walk at your normal pace, as if you were taking a brisk walk in the park." For paced gait trials, a metronome was set to 120 beats per minute, and the subject was instructed to do the same except to "walk to this beat," such that each beat coincided with a foot strike. Each trial was sampled for 3 seconds with a 1- to 2-minute rest period between trials. Data Analysis Linear momentum and angular momentum were calculated for the whole body and upper body. Linear momentum is the vector product of the body's mass and linear velocity. Angular momentum is the vector product of the body's moment of inertia and angular velocity. WB linear momentum is the linear momentum of the center of gravity (CG). WB angular momentum is the body's angular momentum about the CG. In addition, we analyzed the upper body, composed of the head, arms, trunk, and pelvis segments (HAT). Thus, HAT linear momentum is the composite linear momentum of these five segments. The HAT angular momentum is the combination of upper body seg- Group Healthy elders BVH elders Table 1. Healthy Elder and BVH Elder Group Subject Characteristics Age (yrs) Note: BVH = bilateral vestibular hypofunction. Range Height (m) Range Weight (kg) Range

3 M128 KAYA ETAL. ment angular momenta around a point midway between the hip joints. All momentum values were normalized to the subject's mass. WB and HAT linear momenta were analyzed in three planes: sagittal (anteriorposterior), vertical (updown), and frontal (mediolateral). WB and HAT angular momenta were analyzed in the same planes, each acting on the axis perpendicular to the plane. Sit-to-stand. The duration of the sit-to-stand movement was recorded as the cycle time. Cycle time is defined as the interval from the time when the AP CG velocity deflects from zero in the positive (forward) direction to the time when the vertical CG position reaches its maximum. Linear and angular momenta were analyzed at the time of maximum vertical ground reaction force (VGRF)* maximum WB vertical linear momentum, and the end of rise, which occurred when the vertical linear momentum became negligible and the subject had reached erect standing as determined by maximum vertical CG displacement and by viewing the computer-generated wire model (Figure 1). Maximum HAT AP linear and angular momentum were recorded. Net hip, knee, and back torques in the sagittal plane were also recorded at the time of maximal VGRF, a time which closely approximates the liftoff time (29) and the time of peak torque development. Torques were normalized to body weight. The average of two trials was statistically analyzed for each subject. Two BVH subjects and one healthy subject had only one trial included in the analysis because body segments were not visible throughout the sit-to-stand movement for the second trial. Gait. The maximum WB and HAT linear and angular momenta in three planes for one full gait cycle were recorded. Each subject had one gait trial analyzed. The cycle time and AP average CG velocity were also recorded. Statistical analyses were done with Statistical Analysis System (SAS version 6.04) software. Descriptive statistics consisted of means and standard deviations for dependent variables. s and 95% confidence intervals of dependent variables were used to determine if the two groups' values were from different populations (30,31). RESULTS Sit-to-Stand Key events related to momentum generated during the sit-to-stand of a BVH subject are shown in Figure 1. Table 2 summarizes momentum values during five events of the sit-to-stand movement. BVH elders generally had lower values of linear and angular momentum for all events. The timing of the events was calculated relative to the time of maximum VGRF. The two events that occur prior to maximum VGRF maximum HAT sagittal linear momentum and maximum HAT sagittal angular momentum have negative time differences. Two events were significantly different between the groups. The BVH group's maximum HAT angular momentum occurred earlier, while the end of rise was later than that of the healthy group (Table 2). There was no significant difference in sit-to-stand cycle time between the two groups. Torque values taken at the time of maximal VGRF differed between groups. Healthy subjects had significantly higher knee extension torque, whereas BVH subjects had higher hip flexion torque (Table 3). Healthy subjects also had significantly greater maximum VGRF than the BVH subjects. Gait Figure 2 shows linear and angular momentum data from a paced gait trial of a single BVH subject. Significant between-group differences were found for sagittal linear momentum in both free and paced gait (Table 4). BVH sub- \ WB CG \ GRF Figure 1: Computer-generated model of events during sit-to-stand of a BVH subject. A = initiation of movement defined by positive deflection of AP center of gravity (CG) velocity (time = 1.07 s). B = maximum HAT (upper body) sagittal angular momentum (1.41 s). C = maximum HAT sagittal linear momentum (1.44 s). D = maximum WB (whole body) vertical ground reaction force (1.66 s). E = maximum WB vertical linear momentum (1.95 s). F = end of rise (2.51 s). GRF = ground reaction force as measured by the forceplates; + represents the whole body center of gravity.

4 DYNAMIC STABILITY M129 Table 2. Comparison of Sit-to-Stand Momentum Between Healthy Elder and BVH Elder Groups Variable Cycle time (s) Event: max VGRF Time max VGRF (s) HAT sag lm (ms)* HAT sag am (m 2 rads) Event: max HAT sag lm Time max HAT sag lm - time max VGRF (s) Max HAT sag lm (ms) Event: max HAT sag am Time max HAT sag am - time max VGRF (s)* Max HAT sag am (m 2 rads)* Event: max WB vert lm Time max WB vert lm - time max VGRF (s) WB sag lm (ms) Max WB vert lm (ms)* Abs val WB lat lm (ms) WB sag am (m 2 rads) Event: end of rise Time end of rise - time max VGRF(s)* WB sag lm (ms)* Abs val WB lat lm (ms) WB sag am (m 2 rads) Healthy Elders BVH Elders % CI Notes: max VGRF = maximum vertical ground reaction force, lm = linear momentum, am = angular momentum, sag = sagittal, vert = vertical, lat = lateral, abs val = absolute value, m = meters, s = seconds, r = radians. Significant; outside the 95% confidence interval. Variable Table 3. Comparison of Sit-to-Stand Joint Torques and Maximum Vertical Ground Reaction Force (max VGRF) at the Time of max VGRF Between Healthy Elder and BVH Elder Groups Knee extension torque (m X 100%BW)* Hip flexion torque (m X 100%BW)* Back extension torque (m X 100%BW) Max VGRF (%BW)* Notes: m = meters, BW = body weight. Significant; outside the 95% confidence interval Healthy Elders BVH Elders % CI jects generated less WB and HAT sagittal linear momentum. BVH subjects also had significantly less WB and HAT vertical linear momentum at both free and paced gait. Lateral linear momentum represented the only linear momentum values that were higher in the BVH group, but it was significant only for WB and HAT lateral linear momentum to the left during free gait (Table 4). HAT sagittal angular momentum was significantly lower in the BVH group in both free and paced gait (Table 5). HAT lateral angular momentum was significantly higher for BVH subjects to the left in free gait and to the right in paced gait. Healthy subjects had significantly greater transverse angular momentum to the left in free gait only. BVH subjects walked significantly more slowly than healthy subjects (free gait, 64.8 and 98.6 cms, 95% confidence interval (CI): ; and in paced gait, 83.0 and cms, 95% CI: , in BVH and healthy elders, respectively). Gait velocity in both groups significantly increased from free to paced gait trials. Cycle time in free gait was also significantly different between the BVH group and healthy group (BVH = 1.35 s, Healthy = 1.15 s, 95% CI: ). There was no difference in the cycle time at paced gait (BVH =1.06 s, Healthy =1.04 s, 95% CI: ). Both groups' cycle time significantly decreased from free to paced gait trials. DISCUSSION Whole body linear and angular momentum during two common locomotor activities of daily living (ADLs) are reported in this study for the first time. Momentum control

5 M130 KAYA ETAL. Lateral Linear Momentum (ms) Vertical Linear Momentum (ms) Sagittal Linear Momentum (ms) i i P p \" Lateral Angular Momentum m 2 rad8 ) S s 8 \ ) I Transverse Angular Momentum Sagittal Angular Momentum (m 2 rads ) (m 2 rads ) 1 \ f 1 I ) 5 X di 1 Figure 2: WB and HAT linear and angular momentum versus raw time (% gait cycle) in three planes for paced gait of a single BVH subject. Solid line indicates WB momentum. Dashed line indicates HAT momentum. The time of each heel strike is marked by a vertical line and is labeled for each foot. Negative values for sagittal linear momentum (upper left record) represent the forward direction. Otherwise, positive values equal maximum momentum.

6 DYNAMIC STABILITY M131 Table 4. Free and Paced Speed Gait Linear Momentum (kg*ms normalized to kg body mass) Comparison Between Healthy Elder and BVH Elder Groups Free Speed Max WB sag Im* Max HAT sag lm* Max WB lat lm (L)* Max WB lat lm (R) Max HAT lat lm (L)* Max HAT lat Im (R) Max WB vert lm* Max HAT vert lm* Paced Speed Max WB sag lm* Max HAT sag lm* Max WB lat lm (L) Max WB lat lm (R) Max HAT lat lm (L) Max HAT lat lm (R) Max WB vert Im* Max HAT vert lm* f Healthy Elders f Notes: lm = linear momentum, sag = sagittal, lat = lateral, vert = vertical, (L) = left, (R) = right, indicates significant difference between groups; outside the 95% confidence interval. "("Indicates significant difference between free and paced speeds within the group. Free Speed Max WB sag am Max HAT sag am* Max WB lat am (L) Max WB lat am (R) Max HAT lat am (L)* Max HAT lat am (R) Max WB transv am (L)* Max HAT transv am (R) Max WB transv am (L)* Max HAT transvam (R) Paced Speed Max WB sag am Max HAT sag am* Max WB lat am (L) Max WB lat am (R) Max HAT lat am (L)* Max HAT lat am (R) Max WB transv am (L)* Max HAT transv am (R) Max WB transv am (L)* Max HAT transvam (R) BVH Elders Table 5. Free and Paced Speed Gait Angular Momentum (m 2 rads) Comparison Between Healthy Elder and BVH Elder Groups Healthy Elders f Notes: am = angular momentum, sag = sagittal, lat = lateral, transv = transverse, (L) = left, (R) = right. Indicates significant between BVH and healthy elders; outside the 95% confidence interval. ("Indicates significant difference between free and paced speeds within the group. BVH Elders % CI % CI _

7 M132 KAYA ETAL. differed significantly between a group of healthy elders and elders with gait instability from bilateral vestibular hypofunction. Sit-to-Stand Restraint of horizontal linear momentum generated during sit-to-stand is critical to balance maintenance after standing is achieved (22). The BVH subjects maintained stability during sit-to-stand through less reliance on momentum in both the horizontal and vertical planes. They consistently generated less momentum than healthy subjects during sit-to-stand and controlled that momentum at critical points when dynamic balance was required (Table 2). These data suggest that excessive momentum is avoided by unstable people with BVH when rising from a chair. Momentum in the sagittal plane was significantly lower for the BVH group. The BVH group attained less maximal HAT momentum prior to liftoff. Liftoff is an unstable phase of sit-to-stand marked by the transfer of the body CG position from the large, stable base of support of the chair to a smaller, unstable base over the feet. BVH elders had significantly lower HAT sagittal linear momentum at liftoff. Similarly, at the end of rise, BVH subjects had significantly lower WB sagittal linear momentum. Once standing is achieved, BVH elders were actually more stable in the sagittal plane than healthy elders. Thus, apparently to avoid vestibulopathy-induced instability when rising from a chair, BVH elders generate less sagittal plane momentum and brake this momentum to lower values compared to healthy elders. Backward rotation of the HAT and shank, eccentric use of trunk and hip extensors and ankle plantarflexors are mechanisms to brake forward momentum (6,22,32). BVH subjects also generated significantly less WB vertical linear momentum during sit-to-stand by employing lower knee extension torque and ground reaction force (Table 3). Hutchinson (33) proposed that hip and knee torques transfer horizontal momentum to vertical momentum during the momentum transfer phase (the phase from liftoff from the chair to maximum ankle dorsiflexion). Lower knee extension torque and less sagittal plane momentum generated from the previous phases are consistent with the lower vertical linear momentum demonstrated by the BVH group. Although the BVH elders had greater hip flexion torques at liftoff, this may result from transfer of the CG forward over the feet BOS to increase static stability. Balance-impaired individuals may not use horizontal momentum to drive the sit-to-stand movement forward and vertically, but instead employ an alternative mechanism of placing the CG over the feet prior to liftoff to successfully rise from a chair (6). Less reliance on momentum will require more lower extremity strength for this patient population. The BVH elders apparently did not have difficulty controlling WB lateral linear momentum, an indication that stability in the frontal plane was not a problem during chair rise. This is an important finding in a balance-impaired population because excessive lateral linear momentum at the end of rise has been shown to create loss of balance, which may require taking a corrective step to the side to regain balance (29). One explanation for why excessive momentum was not generated by the BVH group is that they may not have been stressed excessively by this task. All subjects were otherwise healthy and had no significant strength deficits that would require them to rely heavily on compensatory momentum to rise from the chair. In addition, the chair height may not have been sufficiently low to challenge lower extremity strength. The lowest knee moments occur when rising from a chair at a height at or greater than knee level (34). Future studies should examine BVH subjects during sit-to-stand at chair heights below knee height because greater reliance on momentum may be required. Gait BVH subjects generate less maximum WB and HAT linear momentum in the sagittal and vertical planes during gait. Propulsion in the sagittal plane is governed by: (a) the forward motion of the body; (b) the heel, ankle, and forefoot that act as rockers to allow the body to advance over the extended knee; (c) the forward momentum generated by swing of the contralateral leg; and (d) pushoff by the stance leg (1,35). Any of these could be limited to produce a reduction of momentum in the sagittal plane. Sagittal momentum of the trunk, pelvis, and leg are decelerated by back muscles (erector spinae, multifidus, rotatores, quadratus lumborum) and hip muscles (gluteus maximus, gluteus medius, adductor magnus) as the leg enters the stance phase (1,36). Bilateral lumbar erector spinae muscles restrict angular displacement of the trunk in the forward direction at heelstrike (37). A hip extensor moment keeps the trunk and pelvis from vaulting forward at heel contact (38). Limiting gait sagittal plane momentum apparently avoids becoming unstable as the body advances forward soon after initial contact. Lower values of vertical and sagittal linear momentum in the BVH group can be attributed to their slower gait velocity. Increasing gait speed will increase vertical momentum (39); significant increase in vertical linear momentum did occur in the healthy group when increasing from free to paced gait (Table 4). Greater maximum WB and HAT linear momenta in the frontal plane by the BVH group may indicate less momentum control than exhibited in the sagittal and vertical planes (Table 4). The upper body of BVH subjects also had greater lateral angular momentum during free and paced gait than control subjects (Table 5). Linear and angular momenta in the frontal plane result from the lateral translation of the CG over the stance leg and rotation of the body laterally to overcome the inertia of the body mass as the opposite leg prepares to be unloaded (40). In clinical settings, the gait of BVH patients can be assessed by walking along a line, noting how much the body deviates laterally from that line. Typically, patients with ataxia deviate to the side more than people without balance deficits (41). Excessive lateral linear momentum leads to loss of balance. To shift from bipedal to unipedal stance, lateral linear momentum must be reduced before the CG travels beyond the base of support, or loss of balance may occur (42,43). Excessive lateral linear momentum could similarly be the contributive factor in ataxia during gait.

8 DYNAMIC STABILITY M133 Less control of frontal plane momentum may be due to decreased hip and trunk muscular control. Peak lateral linear and angular momentum occurred between the loading and midstance phases for both groups. Hip abductors (gluteus maximus, gluteus medius, tensor fascia latae), hip extensors (gluteus maximus, medial hamstrings), and erector spinae muscles are all active during this period (1). According to Winter et al. (44), the hip abductors and medial acceleration of the hip joint are responsible for controlling the medial-lateral imbalance of the HAT during single limb stance. Thorstensson et al. (37) suggest the erector spinae muscles contralateral to the side of the stance leg brake lateral movement of the trunk in the frontal plane during gait. In the dynamic transition from bipedal to single limb stance, Rogers and Pai (43) reported a propulsive phase generates lateral linear momentum to displace the center of mass over the stance leg. Activation of the flexed limb's adductor magnus and stance limb's gluteus medius muscles occurs just prior to the peak force in the propulsive phase to brake the lateral linear momentum which is at its peak at this time. Altered timing and magnitude of activity of the adductor magnus and gluteus medius muscles could be related to the higher values of lateral momentum seen in BVH subjects. Future studies should more clearly relate excessive lateral momentum with loss of balance during gait and investigate the reasons for greater frontal plane momentum exhibited by BVH subjects. According to studies done thus far, a person with BVH prefers a more stable gait pattern by walking at a slower speed with a longer cycle time, increased time spent in double support, and less CG vertical excursion (9,28,45). The temporal variables in this study are consistent with this gait pattern. The BVH subjects walked significantly slower than not only the matched healthy subjects, but slower than BVH subjects in other studies. The BVH group's average CG velocity was 64.8 cmsec for free gait compared to a range of cmsec in other studies (9,12,28,45,23). Other studies involving BVH subjects include a broader range of age groups, but even our healthy elderly subjects had a slower average gait velocity compared to other studies done on elders' gait. For free gait, our healthy elders walked at an average velocity of cmsec compared to a range of cmsec in other studies (46,47,48). Because of the limited size of the camera viewing volume (2 3 m 3 ), only subjects walking at slower speeds had all body segments visible for an entire gait cycle and were thus suitable for full-cycle WB momentum analysis. The difference in gait velocity of the subjects used in our sample compared to other studies may represent a limitation in external validity. Both our normal and BVH subjects probably represent the slowest walkers in their respective populations, which then may reflect lower momentum generated during gait. This should not, however, affect the relative differences found between groups. An additional consideration when interpreting the results of this study is that only maximum momentum values were statistically analyzed. Analysis of maximum momentum (i.e., one point in time) does not explain all variability in momentum during the gait cycle. Future studies could analyze in more detail how momentum changes through the phases of gait, particularly in ataxic gait. Elders with BVH tend to be unstable with transitional movements and tend to limit the use of momentum during ADLs. For a balance-impaired population, it is important to maintain adequate lower extremity range of motion and strength such that they need not rely on momentum to rise from a chair. Inadequate strength and balance deficits are a disabling combination. These data clearly indicated that balance-impaired elders with BVH have excessive lateral linear and angular momentum during gait. The slower gait velocity of this patient group may explain differences in momentum in the sagittal and transverse planes compared to healthy elders, but does not explain the excessive frontal plane momentum. Excessive lateral momentum is a clear marker of frontal plane instability during gait, and gait training can help improve dynamic stability (6,23). Further research is needed to determine if momentum control during locomotor ADL can be remedied by rehabilitation or other interventions in elders. ACKNOWLEDGMENTS This project was supported by the National Institute on Aging, NIH Grants P50 AG11669, R01 AG11255 and R01 AG12561, and the National Institute on Disability and Rehabilitation Research (Grant H133G30041). The authors wish to thank Rita Popat, Dr. Jose Ramirez, Kathleen Price, and Niyom Luepongsak for assistance with data analysis. Bradley K. Kaya is now with the Straub Clinic and Hospital, 800 South King Street, Honolulu, HI Address correspondence to Dr. David E. Krebs, Professor and Director, MGH Biomotion Laboratory, 101 Merrimac Street, Boston, MA krebs@helix.mgh.harvard.edu REFERENCES 1. Perry J. Gait Analysis, Normal and Pathological Function. Thorofare, NJ: SLACK; Jevsevar DS, Riley PO, Hodge WA, Krebs DE. Knee kinematics and kinetics during locomotor activities of daily living in subjects with knee arthroplasty and in healthy control subjects. Phys Ther. 1993; 73: Krebs DE, Wong D, Jevsevar D, Riley PO, Hodge WA. Trunk kinematics during locomotor activities. Phys Then 1992;72: Winter DA. Biomechanics and Motor Control of Human Movement. New York: John Wiley; 1990: Zachazewski JE, Riley PO, Krebs DE. Biomechanical analysis of body mass transfer during stair ascent and descent of healthy subjects. J Rehabil Res Dev. 1993; 30:412^ Schenkman M, Berger RA, Riley PO, Mann RW, Hodge WA. Whole body movements during rising to standing from sitting. Phys Ther. 1990;70: Perry BC. Falls among the elderly: a review of methods and conclusions of epidemiologic studies. J Am Geriatric Soc. 1982;30: Zubenstein LZ, Josephson KR, Robbins AS. Falls in the nursing home. Ann Intern Med. 1994; 121-M2-A Krebs DE, Lockert J. Vestibulopathy and gait. In: Spivak BS, ed. Mobility and Gait. New York: Marcel Dekker; 1995: Gill-Body KM, Krebs DE, Parker SW, Riley PO. Physical therapy management of peripheral vestibular dysfunction: two clinical case reports. Phys Ther. 1994;74: Baloh RW, Jacobson K, Honrubia V. Idiopathic bilateral vestibulopathy. Neurology. 1989;39: Azen RW. Head angular displacement and acceleration during functional activities. MGH Institute of Health Professions masters degree thesis. Boston: Heitman DK, Gossman MR, Shaddeau SA, Jackson JR. Balance performance and step width in noninstitutionalized, elderly, female fallers and non-fallers. Phys Ther. 1989;69:

9 M134 KAYA ETAL. 14. Nuzik S, Lamb R, VanSant A, Hirt S. Sit-to-stand movement pattern: a kinematic study. Phys Ther. 1986;66: Hutchinson EB, Riley PO, Krebs DE. A dynamic analysis of the joint forces and torques during rising from a chair. IEEE Trans Rehabil Eng. 1994;2: Wheeler J, Woodward C, Ucovich RL, Perry J, Walker JM. Rising from a chair: influence of age and chair design. Phys Ther. 1985,65: Ikeda ER, Schenkman ML, Riley PO, Hodge WA. Influence of age on dynamics of rising from a chair. Phys Ther. 1991;71: Riley PO, Schenkman M, Mann RW, Hodge WA. Mechanics of constrained chair rise. J Biomech ;24: Pai YC, Rogers MW. Segmental contributions to total body momentum in sit-to-stand. Med Sci Sports Exerc ;23: Pai YC, Rogers MW. Control of body mass transfer as a function of speed of ascent in sit-to-stand. Med Sci Sports Exerc. 1990,22: Pai YC, Naughton BJ, Chang RW, Rogers M. Control of body centre of mass momentum during sit-to-stand among young and elderly adults. Gait Posture. 1994; 2: Pai YC, Lee WA. Effect of a terminal constraint on control of balance during sit-to-stand. J Mot Behav. 1994;26: Krebs DE, Gill-Body KM, Riley PO, Parker SW. Double-blind, placebo-controlled trial of rehabilitation of bilateral vestibular hypofunction: preliminary report. Otolaryngol Head Neck Surg. 1993; 109: Riley PO, Mann RW, Hodge WA. Modelling of the biomechanics of posture and balance. J Biomech. 1990;23: Antonsson E, Mann RW. Automatic 6-D.O.F. kinematic trajectory acquisition and analysis. ASME J Dynam Syst Meas Control. 1989; 111: McConville JT, Clauser CE, Churchill TD, Cuzzi J, Kaleps I. Anthropometric relationships of body and body segment moments of inertia. Technical Report No. ARAMRL-TR , Air Force Aerospace Medical Research Lab Young JW, Chandler RF, Snow LL, Robinette KM, Zehner GF, Loftberg MS. Anthropometric and mass distribution characteristics of the adult female. Technical Report No. FAA-AM-83-16, FAA Civil Aeromedical Institute Tucker CA, Ramirez J, Riley PO, Krebs DE. Dynamic stability and center of gravity control in normal and vestibulopathic gait. (In review). 29. Riley PO, Krebs DE, Popat R. Biomedical analysis of failed sit-tostand. IEEE Trans Rehab Eng. 5: , Goodman SN, Berlin JA. The use of predicted confidence intervals when planning experiments and the misuse of power when interpreting results. Ann Intern Med. 1994; 121: Henderson AR. Chemistry with confidence: should clinical chemistry require confidence intervals for analytical and other data? Clin Chem. 1993;39: Pai YC, Rogers MW. Speed variation and resultant joint torques during sit-to-stand. Arch Phys Med Rehabil ;72: Hutchinson E. Estimation of kinetics using a three-dimensional model of the human body. Masters thesis. Boston: Massachusetts Institute of Technology, Rodosky MW, Andriachhi TP, Andersson GBJ. The influence of chair height on lower limb mechanics during rising. J Orthop Res. 1989; 7: Dillingham TR, Lehmann JF, Price R. Effect of lower limb on body propulsion. Arch Phys Med Rehabil. 1992;73: Carlson H, Thorstensson A. Control of the human trunk during locomotion. Acta Physiol Scand. 1982; 114:14A (Abstr). 37. Thorstensson A, Carlson H, Zomlefer MR, Nilsson J. Lumbar back muscle activity in relation to trunk movements during locomotion in man. Acta Physiol Scand. 1982; 116: Prince F, Winter DA, Stergiou P, Walt SE. Anticipatory control of upper body balance during human locomotion. Gait Posture. 1994; 2: Smidt GL, ed. Gait Rehabilitation. New York: Churchill-Livingstone; 1990: Rogers MW, Pai YC. Dynamic transitions in stance support accompanying leg flexion movements in man. Exp Brain Res. 1990,81: Graybiel A, Fregly AR. A new quantitative ataxia test battery. Acta Otolaryngol. 1966;61: Hanke TA, Rogers MW. Reliability of ground reaction force measurements during dynamic transitions from bipedal to single-limb stance in healthy adults. Phys Ther. 1992;72: Rogers MW, Pai YC. Patterns of muscle activation accompanying transitions in stance during rapid leg flexion. J Electromyog Kinesiol. 1993;3: Winter DA, MacKinnon CD, Ruder GK, Wieman C. An integrated EMGbiomechanical model of upper body balance and posture during human gait. Prog Brain Res. 1993;97: Lockert JD. Mediolateral interfoot distance in vestibulopathic gait. MGH Institute of Health Professions masters degree thesis. Boston: Ostrosky KM, VanSwearingen JM, Burdett RG, Gee Z. A comparison of gait characteristics in young and old subjects. Phys Ther. 1994;74: Hageman PA, Blanke DJ. Comparison of gait of young women and elderly women. Phys Ther. 1986;66: Blanke DJ, Hageman PA. Comparison of gait of young men and elderly men. Phys Ther. 1989;69: Received December 16, 1996 Accepted November 5, 1997