Tai chi balance therapy improves slip response

[Guest guest]

January 2007

Biomechanics Magazine

http://www.biomech.com/showArticle.jhtml?articleID=196900593

Tai chi balance therapy improves slip response

Practitioners are testing new and old ways of improving elders'

reactions to ease fear of falling.

By: Strawberry Gatts, PhD

Falls and fall-related injuries are a major concern for older adults

individually and a major health problem for society in general as

the aging population swells. The overall financial cost of falls is

expected to reach $43.8 billion by 2020.1 Because falls and fall-

related injuries represent an increasing burden on the healthcare

system, efforts to design intervention strategies and balance

training for aging populations are increasing nationally. While much

is known about preventing falls, the current need is for research to

identify the interventions that are most effective for complex real

world populations that comprise individuals with different

characteristics or include multiple risk factors.1

Tai chi has proven to be a practical intervention because it has

lowered fall rates in certain groups of seniors,2 maintained balance

and strength gains from other types of training,3 and appears to be

safe and effective for seniors with chronic conditions.4 In their

2004 review, Wang et al4 concluded that although tai chi appeared to

promote balance control, flexibility, and cardiovascular fitness in

older patients, available studies didn't address which mechanisms

profit from tai chi training, making it difficult to draw firm

conclusions about the benefits of the practice. The question of how

tai chi achieves its salubrious effects still remains.

Based on previous research, we chose key neuromuscular and

biomechanical variables shown to contribute to balance recovery to

research how tai chi improves balance in the real world. We

developed a theory of how tai chi may influence these variables, and

then examined these variables during a large/fast slip imposed

during a walk, because most falls occur while walking.5,6 We chose

to examine balance responses in subjects representing an older real

world population with balance impairments.

Study subjects

The age range for participants in this study was 68 to 92 years, and

included subjects at high risk for falling. One subject walked with

a cane before training, had deteriorating mobility due to knee

surgery, and had had both knees replaced. Another had fallen three

times, and each time had been unable to get up without help.

Participants also had undergone back, hip, and knee surgery, had

received joint replacements and metal bone implants, and suffered

from arthritis and bursitis (Table 1). Most participants had been

referred by their medical doctor or physical therapist. We analyzed

data from only one trial because this is most similar to a real

world situation, in which we do not have a chance to do it again.

Twenty-two balance-impaired seniors (Berg score 44 or less) were

randomly divided into tai chi (TC) or control groups. Nineteen

subjects completed the study. TC balance training included

repetitive exercises using TC's motor and biomechanical strategies,

techniques, and postural elements. Control training included

balance/awareness education, stretching, stress reduction, deep

breathing, and axial mobility exercises. (Axial mobility exercises

were developed by Duke University Medical Center to improve postural

alignment, increase range of motion, coordinate relaxed-as opposed

to effortful-movement, relax tight muscles using deep breathing, use

muscle groups with appropriate mechanical advantage, and enhance

participation of appropriate muscle synergies used in activities of

daily living.) Groups trained 1.5 hours a day, five days a week for

three weeks. After post-testing the control group received TC

training.

Subjects were tested before any training, after the first training,

and the control group was tested after both trainings for any change

in response. We also analyzed the data to find whether the groups

showed similar trends in their performance measures after TC

training and in which measures those trends were most apparent. If

significant trends (e.g., muscle onset latency) were found for both

groups after TC training and not for the control group before TC

training, the change in this response mechanism would likely be due

to the TC training.

Our study investigated both neural and biomechanical mechanisms

older subjects use to recover balance in response to a large/fast

walking perturbation and four clinical measures of functional

balance pre- and post-training.7,8

Subjects walked across a force plate triggered to move forward 15 cm

at a speed of 40 cm/sec at heel strike of the right foot.

Perturbation onset was triggered by the initial touch down on the

force plate; i.e., 40 Newtons pressure from the contact foot

activated the force plate to move forward. Muscle responses (surface

electromyograms) from the tibialis anterior and medial

gastrocnemius, whole-body kinematics, center of pressure (COP), and

center of mass (COM) were recorded during balance recovery. Four

commonly used clinical/behavioral measures of functional balance:

timed up and go, functional reach, and single leg and tandem stance

time, were also recorded pre- and post-training.

We found significant improvement after TC training on all clinical

measures (timed up and go [p lesser than or equal to .0001],

functional reach [p lesser than or equal to .0001], single leg

stance time [p lesser than or equal to .0007], tandem stance time [p

lesser than or equal to .0004]). See Table 2 for details. Controls

improved significantly only on functional reach (p lesser than or

equal to .0011).

Neural response mechanisms also improved after TC training. TC

subjects significantly reduced tibialis anterior response time, from

149 plus/minus 45 ms to 98 plus/minus 67 ms (p = .004) and reduced

cocontraction of antagonist muscles from .6 to .0 probability.

Controls reduced tibialis anterior response time, but not

significantly (p lesser than or equal to .138), and reduced

cocontraction from .63 to .5 probability.

Since cocontraction reflects increased expenditure of energy used to

regain balance and slows down restoration of joint/segment angular

trajectories used to maintain the gait cycle,9 balance responses

would likely be faster and cause less tripping if this response were

eliminated. Tai chi and the control training were similar in their

ability to reduce reversed activation (i.e., antagonist medial

gastrocnemius activated before agonist tibialis anterior). Reversed

activation was 6.23 times less likely after TC (p lesser than or

equal to .103) and 6.16 times less after control training (p lesser

than or equal to .282).

The COP was measured under the perturbed stance foot during single

leg stance. Before TC training, our subjects typically interrupted

the forward movement of the swing leg by dropping it to the ground

behind the body when reacting to a perturbation. This strategy was

used in order to support the backward fall of the body, which lagged

behind the forward movement of the slipping foot. The unexpected

drop of the swing leg was coded as a trip. TC enhanced balance

recovery strategies by significantly reducing the number of trip

responses used to recover from the slip (p < .005). TC also

significantly reduced swing leg cross-step medial distance (the

distance the swing leg moves in a medial direction [cross-step] from

toe-off to heel strike off the plate after the perturbation) (p

< .038), resulting in a straighter pathway of the swing leg, and

increased use of a swing-leg heel strike (as opposed to a flatfoot

or toe-strike touch-down) (p < .001) used to step off the force

plate and continue the gait cycle. Whole-body COM anteroposterior

(sagittal plane) total path (the length of the curved COM

trajectory, not the straight-line displacement from start to end)

significantly increased after TC (p lesser than or equal to .017).

COM anteroposterior velocity at perturbation onset was not

significantly different from pre- to post-testing. This led us to

conclude that changes were not due to changes in walking speed.

Tripping and medial cross-stepping were significantly reduced and

use of a heel strike significantly increased after TC training.

Whole-body COM anteroposterior path significantly increased after

training for the intervention group, while controls reduced their

path. This increase could be due to greater flexibility in the right

hip joint (all perturbations were to the right foot), which allowed

a longer step. This change was not due to previous experience on the

force plate because control subjects decreased their path at post-

testing.

Because most subjects fell backward at baseline and curtailed the

fall by dropping the contralateral swing leg briefly to the ground

(coded as a trip), we analyzed COP data from the last 250 ms during

which all subjects were in single right leg stance as they prepared

to step off the plate. Even in this small amount of time, a trend (p

lesser than or equal to .07) was found. After TC training subjects

reduced their COP anteroposterior path (the length of the curved COP

trajectory, not the straight line displacement from start to end)

and COP anteroposterior maximum velocity while controls increased

theirs (path: p lesser than or equal to .066; velocity: p lesser

than or equal to .069). This means that COP movement under the

single stance foot was more stable for those who received TC

training because the point (COP) representing the aggregate of the

pressure under the stance foot moved less and moved at a slower

velocity. A trend, though not significant, was also found for COM-

COP anteroposterior angular separation at right heel strike. This is

the sagittal plane angle formed between the vertical projection of

the whole-body COM to the ground and the line connecting the whole-

body COM to the COP of the step foot. After TC subjects increased

their separation (p lesser than or equal to .067) while controls

decreased their separation. This indicated that TC subjects were

taking longer steps after training and controls were taking shorter

steps.

Seniors with balance problems often use cautious gait, a

conservative strategy which is characterized by decreased COM motion

and velocity, and reduced anteroposterior distance between the COM

and COP positions. Since a decrease in the separation of the COM and

COP indicates a reduction in mechanical loading on the joints of the

supporting limb, TC training may have increased the ability to load

the joint of the supporting limb.

We hypothesized that tai chi would improve balance control via

improvements in neural response mechanisms controlling the agonist

(tibialis anterior) and antagonist (medial gastrocnemius) muscles in

the perturbed single stance leg. We also hypothesized that tai chi

would improve balance control via changes in the biomechanics of the

musculoskeletal system controlling the stepping strategy of the

swing leg, flexibility of the COM, and stability of the COP. In

answer to the question of how tai chi improves balance in the real

world, our results indicated that a focused and intensified TC

balance program was capable of quickly-within three weeks-improving

balance responses by significantly enhancing neural response

mechanisms in an at-risk older population with multiple impairments.

It also showed that impaired seniors could significantly increase

anteroposterior movement of the COM during a slip rather than

limiting COM movement (as controls did). Flexible control over large

COM movement has been found in dancers and athletes who have good

balance, and it has been recommended that patients with balance

decrements be trained in the use of proper stepping strategies to

abort falls.2 The likelihood the TC training would translate to

balance improvement in the real world was supported by our results

from four clinical tests of functional balance. Scores for the timed

up and go, single leg stance time, tandem stance time, and

functional reach all improved significantly after the TC program.

Unique contributions of the study

This study is different from other TC studies in several ways. To

date, TC interventions have not included balance-impaired older

adults who have had surgical interventions to their backs, hips,

and/or knees, in addition to arthritis and bursitis. Since it is

likely that any intervention for balance-impaired older populations

would include individuals who had had such surgeries, we included

these subjects.

Second, previous TC studies relied heavily on standing balance

measures,3,4,10-12 whereas our study tested responses during a

large/fast perturbation while walking. This difference is important,

since most falls do not occur while standing.5,6 Third, TC research

has not looked at differences in balance responses in the same

subjects given two types of training (controls were first given

balance/awareness education, stress reduction, deep breathing, and

axial mobility exercises, and after post-testing were given TC

training). This allowed us to examine differences in response

mechanisms before and after two types of theoretically similar

training in the same subjects. Fourth, previous TC research3,4,10-12

focused on performing TC postural sequences, whereas our

intervention focused on TC's key training principles (motor and

biomechanical techniques, strategies, and postural elements) used to

develop dynamic balance and master individual postures. Fifth, prior

studies have not examined a short-term (three weeks) intensified

training that can fit easily within clinical settings.

The difference between our study and previous studies10,13 that may

have led to strong training effects is that we focused on the

underlying balance principles involved in each form (posture) rather

than on linking a series of different postures. In addition, this

training was more intense than most others, with sessions occurring

five times per week for 1.5 hours/day over the three weeks of the

intervention.

Recommendations for future research

Using larger groups of people and replicating the training at other

labs would be informative. Repeating the study in other impaired

populations would also be useful. Employing TC training principles

to develop robotic or other mechanical training aids and protocols

that could assist those at various stages of balance and mobility

dysfunction is another fertile area for future research.

Other areas that need clarification include examining the type of TC

intervention that is most appropriate for various populations with

balance deficits. Not all styles of TC nor all TC training methods

are fit for older adults and impaired populations. The particular

postures and training methods selected for a protocol will

dramatically affect participants' safety, learning time, and the

resultant behavior. The frequency, intensity, and duration of TC

interventions have varied widely in previous studies.4 Therefore, a

minimum dosage, such as we explored in our study, needs to be

examined further. Lastly, even though improvement was found with

only three weeks of training, because human physiology is plastic,

we conclude that maintenance programs are necessary and need to be

investigated.

Strawberry Gatts, PhD, is designing and testing treadmill training

protocols for preventing trip-related falls in older adults at the

University of Illinois at Chicago College of Applied Health

Sciences, Musculoskeletal Biomechanics Lab.

References

s J. Falls among older adults-risk factors and prevention

strategies. National Council on Aging Falls Free: Promoting a

National Falls Prevention Action Plan. 2004;3-18.

Wolf SL, Barnhart HX, Ellison GL, Coogler CE. The effect of Tai Chi

Quan and computerized balance training on postural stability in

older subjects. Atlanta FICSIT Group. Frailty and injuries:

cooperative studies on intervention techniques. Phys Ther 1997;77

(4):371-381.

Wolfson L, Whipple R, Derby C, et al. Balance and strength training

in older adults: Intervention gains and Tai Chi maintenance. J Am

Geriatr Soc 1996;44(5):498-506.

Wang C, Collet JP, Lau J. The effect of Tai Chi on health outcomes

in patients with chronic conditions: a systematic review. Arch

Intern Med 2004;164(5):493-501.

Gabell A, Simons MA, Nayak US. Falls in the healthy elderly:

predisposing causes. Ergonomics 1985;28(7):965-975.

Nevitt MC, Cummings SR, Hudes ES. Risk factors for injurious falls:

A prospective study. J Gerontol 1991;46(5):164-170.

Gatts SK, Woollacott MH. How Tai Chi improves balance: Biomechanics

of recovery to a walking slip in impaired seniors. Gait Posture, e-

published April 29, 2006 (ahead of print).

Gatts SK, Woollacott MH. Neural mechanisms underlying balance

improvement with short term Tai Chi training. Aging Clin Exp Res

2006;18(1):7-19.

Tang PF, Woollacott MH. Inefficient postural responses to unexpected

slips during walking in older adults. J Gerontol A Biol Sci Med Sci

1998;53(6):M471-480.

Wolf SL, Barnhart HX, Kutner NG, et al. Reducing frailty and falls

in older persons: an investigation of Tai Chi and computerized

balance training. J Am Geriatr Soc 1996;44(5):489-497.

Yan JH. Tai Chi practice improves senior citizens' balance and arm

movement control. J Aging Phys Activity 1998;6:271-284.

Lan C, Lai J, Chen S, Wong M. 12-month Tai Chi training in the

elderly: its effect on health fitness. Med Sci Sports Exerc 1998;30

(3):345-351.

Wolf SL, Sattin RW, Kutner M, et al. Intense tai chi exercise

training and fall occurrences in older, transitionally frail adults:

a randomized, controlled trial. J Am Geriatr Soc 2003;51(12):1693-

1701.