Reductions in body mass deliver positive gait changes

[Guest guest]

October 2006 Biomechanics Magazine

http://www.biomech.com/showArticle.jhtml?articleID=193200724

Reductions in body mass deliver positive gait changes

Reducing the BMI of overweight and obese children may help normalize

their gait.

By: Maureen Becker, PT, DHsc, Arthur , PhD, PT,

Rothman, EdD, PT, Amy Ng, MSPT, Barker, MSPT, Brady,

MSPT, and Andy Koser, MSPT

Excess weight and obesity may not be infectious diseases, but they

have reached epidemic proportions in the U.S. According to the

surgeon general's 2001 report, excess weight and obesity are

increasing in both genders and among all population groups. In 1999,

an estimated 61% of U.S. adults were overweight or obese and 13% of

children and 14% of adolescents were overweight. Today there are

nearly twice as many overweight children and almost three times as

many overweight adolescents as there were in 1980.1

In addition to a host of health problems associated with excess

weight and obesity, such as heart disease, high blood pressure, and

type 2 diabetes,1 clinicians have identified a spectrum of increased

orthopedic demands closely related to obesity. Orthopedic problems

such as osteoarthritis and a variety of musculoskeletal disorders

are frequently linked to degrees of weight increase and may

ultimately have an effect on ambulation.1,2

Walking is a fundamental locomotor task. Obesity presents a distinct

disadvantage in movement as a large proportion of body weight does

not contribute to performance. Characteristics of walking patterns

have been identified for normal-weight participants. But to date,

little research exists on overweight and obese school-age children.

Hills and found that obese subjects displayed longer cycle

duration, lower cadence, lower relative velocity, and longer stance

period than normal-weight subjects.3 Other temporal differences

included gait asymmetry and greater stride width for the obese, all

pointing to a slower, more tentative normal speed, and problems

encountered when walking at speeds other than normal. Obese subjects

displayed a more flat-footed weight acceptance period in early

stance and a greater external rotation (out-toeing) of the foot at

all phases of the gait cycle.3 McGraw et al found gait deviations in

obese prepubertal boys. They found obese boys spent a significantly

greater percentage of the gait cycle in dual stance.4

Methods

The purpose of this study was to assess the effect of body mass

index (BMI) on the gait of overweight school-age children. We

hypothesized that there would be a change in the gait of overweight

children after an eight-week diet and exercise intervention as a

result of a change in their BMI.

Participants. The 10 children composing the overweight sample were

recruited from the " Defensive Eating/Healthy Living Program, "

designed and conducted by Ruth Sundback, a registered dietitian. The

age requirement for participation in her program was seven to 12

years. The 10 children making up the normal-weight sample were

conveniently recruited from among friends and relatives of the

investigators. The age range of the normal-weight group was chosen

to coincide with that of the overweight group. The subjects were not

matched for gender.

Previous studies have established that children between seven and 12

years of age will exhibit a mature gait pattern. Sutherland5 found

that growth and central nervous maturation control gait changes up

to 3.5 to four years. However, growth alone explains the majority of

changes through the remainder of the growing years.5 Preis and

colleagues found that continuous slight changes in gait are achieved

until the age of five years, when the gait pattern becomes like that

of an adult.6

Body mass index has been widely used to assess weight relative to

height and is calculated by dividing body weight in kilograms by

height in meters squared. BMI has been established as a valid and

reliable measure with which to assess adiposity in children and

adolescents.1,7,8 As children grow, their body fatness changes over

the years. Additionally, girls and boys differ in their body fat

distribution as they mature; the interpretation of BMI for children

and adolescents depends on their age and gender. Therefore, the

Centers for Disease Control and Prevention have established a sex-

and age-specific BMI for children and adolescents aged two to 20

years.9 For the purpose of this study, the overweight sample

included children at or above the 95th percentile of the sex- and

age-specific BMI, and the normal-weight sample included children

below the 95th percentile. The identification of children

as " overweight " and " normal weight " in this study is consistent with

the terminology used by the CDC.

The parents/guardians of the participants signed an informed consent

document and the children gave verbal assent prior to participation

in the study. In addition, the parents/guardians of the participants

were administered a health questionnaire to identify factors that

would exclude a child from this study. These factors included recent

lower limb injuries, frequent falls, inner ear tube implants, use of

corrective footwear such as orthoses, and a leg length discrepancy

of one inch or greater. The study design was approved by the

Institutional Review Board at the College of Staten Island.

Instrumentation. Gait data were collected using the GAITRite gait

analysis system, which employs an electronic walkway connected to a

laptop computer using GAITRite Silver Version 2.7 software. The

system requires a mature gait pattern. The walkway contains six

sensor pads embedded in a roll-up carpet to produce an active area 2

feet wide and 12 feet long. As the participant ambulates across the

walkway, the system captures the applied pressure of each footfall

as a function of time. The application software processes the raw

data into footfall patterns and computes temporal and spatial

parameters of gait.10 These parameters include step, cycle, swing,

stance, single and double support time; step length; stride length;

base of support (BOS); toe in/out; distance; ambulation time;

velocity; mean normalized velocity; cadence; and step extremity

ratio. GAITRite computes a functional ambulation performance (FAP)

score using the following parameters: step functions left, step

functions right, differential step extremity ratio, heel-to-heel

dynamic BOS, ambulatory aids, and assistive devices. The FAP score

ranges from zero to 100 points; a healthy adult will score between

95 and 100 points. Although an FAP scoring system has not been

established for children, the children in this study demonstrated a

mature gait pattern.6 The FAP is a clinical test developed to

quantify a sequence of skills fundamental to walking.11 The

reliability and validity of the GAITRite system compared to paper

and pencil- and video-based methods has been established by

McDonough et al.12

In order to establish each participant's BMI, height and weight were

assessed using a Detecto weight scale. Leg length was measured from

the greater trochanter to the floor using a meter stick. The

GAITRite software used this measurement to compute mean normalized

velocity and step extremity ratio.

Intervention. Ruth Sundback's " Defensive Eating/Healthy Living

Program " consists of an eight-week diet and exercise instructional

program. The group using this program met for one hour once a week

(Table 1). This intervention required the involvement of at least

one parent/guardian with the child at each session. The intervention

was for the overweight children only.

Each session consisted of an educational lesson on diet and the

importance of increased physical activity, a cooking project

emphasizing healthy food choices, and exercise. A pedometer was

given to the children so they could monitor their walking. In order

to ensure compliance with this program, a contract was established

between the parent/guardian, child, and dietitian. The children were

required to keep a diet and activity journal to chart their

progress. Moderate exercise was encouraged throughout the week. The

parent/guardian agreed to apply what was learned in each session to

their home life.

Procedure. Body weight, height, and leg length measurements were

taken at the beginning of the study (week one) and at the end of the

study (week eight). For the testing procedure, verbal instructions

were read to the child, explaining exactly what was expected of him

or her. A visual demonstration of the gait assessment by one of the

investigators was also provided.

Chairs were placed six feet in front of the beginning of the mat and

six feet beyond the end of the mat. The participants walked from one

chair to another, allowing each child to establish his/her normal

walking pattern. Participants were required to conduct six passes

with shoes off and socks on. The first four passes were to

familiarize the child with the procedures. Data were recorded for

the fifth and sixth passes. A recorded audio tape signaled the child

to begin each pass ( " Begin walking " ). The procedures were consistent

throughout the study for all participants.

To ensure intratester reliability, all procedures were performed by

the same investigators at each testing session. Testing was

conducted at Zion Lutheran Church on Staten Island, NY.

Statistical analysis. Data for the following temporal and spatial

parameters of gait were analyzed: mean normalized velocity, cycle

time, BOS, double support time, swing time, and step extremity

ratio. FAP scores and BMI were also analyzed. The statistical

analysis was calculated using SPSS Version 10.0 with a general

linear model repeated measures analysis of variance (ANOVA) on a 2

(time) x 2 (group) design. The accepted level of statistical

significance was determined to be P < 0.05.

Results

Initial testing data were collected from the 10 overweight and 10

normal-weight children who had been recruited. Six in each group

completed the study. Four participants in the overweight group,

three boys and one girl, with a mean age of 8.5 years, failed to

complete the study. Four participants in the normal-weight group,

again three boys and one girl, with a mean age of 9.5 years, also

failed to complete the study. One from the overweight and three from

the normal-weight group were excluded after the initial testing for

not meeting the BMI inclusion criteria. Three from the overweight

group and one from the normal-weight group were excluded for failing

to show for the retest.

The overweight participants were four boys and two girls. The normal-

weight participants were one boy and five girls. Mean morphologic

and body composition data for each group are presented in Table 2.

Data analysis

The analysis of double support time yielded a significant main

effect of group (F[1,10] = 20.6, P < 0.05). Analysis found no

significant main effect of time (F[1,10] = 0.02, P = n.s.) and no

interaction of time and group (F[1,10] = 0.43, P = n.s.).

The analysis of BMI yielded a significant main effect of group (F

[1,10] = 13.6, P < 0.05), but no main effect of time (F[1,10] =

0.36, P = n.s.). No interaction of time and group (F[1,10] = 3.2, P

= 0.104) was found, though there was a trend.

Analysis of the average of the left and right swing yielded a

significant main effect of group (F[1,10] = 28.8, P < 0.05) but no

main effect of time (F[1,10] = 0.01, P = n.s.) and no interaction of

time and group (F[1,10] = 0.03, P = n.s.).

The analysis of BOS showed a trend in group similar to that of

double support time (F[1,10] = 3.3, P = 0.10), although this effect

was not significant. No main effect of time and no interaction of

time and group were noted.

The analysis of cycle time yielded no main effect of group (F[1,10]

= 0.71, P = n.s.) or time (F[1,10] = 0.76, P = n.s.). While the

interaction of time and group was not significant, it did

demonstrate a trend (F[1,10] = 3.6, P = 0.09).

The analysis of mean normalized velocity, step extremity ratio, and

FAP yielded no significance for main effect of time, main effect of

group, or interaction of time and group.

Summaries of the analyzed data are presented in Tables 3 and 4.

Discussion

The purpose of this study was to examine the effect of BMI change on

various gait characteristics of overweight school-age children

before and after a diet and exercise intervention; we hypothesized

that there would be a change. Of the gait characteristics examined,

changes in double support time, BMI, and swing time were found to be

statistically significant. A trend was found in BOS and cycle time.

Normalized velocity, step extremity ratio, and FAP demonstrated no

statistically significant difference.

Differences in double support time between the overweight and normal-

weight groups were found to be statistically significant. The

overweight group spent a greater percentage of time in double

support than the normal-weight group for both trials. Hills and

3 found that obese subjects displayed a consistently higher

double stance period at normal, fast, and slow walking speeds.

McGraw et al4 concluded that obese boys spent a significantly

greater percentage of the gait cycle in dual stance. Weight-bearing

stability during the gait cycle is identifiable from the

relationship between single and double stance. Individuals with

reduced stability display a lengthened double stance period and a

shortened single stance period.3 These findings suggest that

overweight children spend a longer time in double support in order

to maintain stability during the gait cycle. Overweight children may

also require increased muscle strength to move the lower limbs

forward against the forces of gravity. Overweight children may

require more balance control and lower extremity muscle strength

than non-overweight children in order to accommodate for the

increased body mass. This would appear to account for the need for

overweight children to spend more time in double support during gait.

There was a significant difference in BMI between the two groups at

week one of the study. Although the interaction of time and group

was not significant, it demonstrated a trend. At week eight the BMI

of the overweight group did decrease, though the change was not

statistically significant. We speculate that a longer period for the

intervention would have shown a greater decrease in BMI.9 Further

research, such as a repeat of this study, should include a longer

intervention time (16 weeks) to determine if BMI continues to

decrease in the overweight group.

Differences in swing time were found to be statistically significant

between the overweight and normal-weight groups. The overweight

group demonstrated a decreased percentage of the gait cycle in swing

compared to the normal-weight group. Excess noncontributory weight

may cause mediolateral instability, which may in turn displace the

center of gravity beyond the BOS during the swing phase. Our

findings suggest that the overweight group spent less time in swing

because of difficulty balancing on a single limb.

Hills and found no difference in the swing phase of normal

and obese children at normal walking speed, but did find a decreased

swing phase for obese children at both slow and fast walking

speeds.3 Although our study analyzed only preferred walking speeds,

the findings in regard to decreased swing phase are consistent with

those of Hills and at slow and fast walking speeds. The

percentage of swing phase for the overweight children in this study

is inconsistent with that for the normal walking speeds found by

Hills and . This may be due to differences in the BMI of the

two overweight groups, the BMI of the overweight participants in

this study being greater than the BMI of the obese subjects in the

study by Hills and . Spyropoulos et al found that the

percentage of swing was decreased for obese men compared with that

for nonobese men.2 Since school-age children have mature gait

patterns,5 their percentage of swing phase is comparable to that of

adult. Therefore, the decreased percentage of swing phase among the

overweight participants in this study is consistent with that of

Spyropoulos et al.

A trend in BOS was found between the overweight and normal-weight

groups. The overweight group displayed a greater BOS throughout this

study. This may be due to a decreased ability of those in the

overweight group to maintain their balance and stability during

gait. A wider BOS may also result from an increased amount of

adipose tissue between the thighs. These findings are consistent

with those of Spyropoulos et al.2 Hills and 3 also found that

obese children exhibited a wider base of support at normal walking

speeds.

At the beginning of this study, both groups demonstrated similar

cycle times; at the end, the overweight group exhibited an increased

cycle time compared to that of the normal-weight group. The similar

cycle times found at the beginning of the study were inconsistent

with previous studies, which have found that obese subjects

displayed longer cycle duration than the nonobese subjects.2,3 We

speculate that the overweight participants did not walk at their

preferred walking speed at week one because they were self-conscious

about their weight and being observed. At week eight, they were more

comfortable with the investigators and testing conditions and walked

at their preferred cycle time.

No statistical significance

Mean normalized velocity, step extremity ratio, and FAP score

demonstrated no statistical significance. Mean normalized velocity

is the ratio of velocity to leg length. At week one, the overweight

group demonstrated a higher mean normalized velocity than the normal-

weight group. However, at week eight, the overweight group

demonstrated a lower mean normalized velocity than the normal-weight

group. In the overweight group, the leg length did not show any

significant change between weeks one and eight. Therefore, the

decrease in mean normalized velocity occurred due to a decrease in

velocity. The investigators expected the overweight participants to

show an increase in mean normalized velocity due to a decrease in

their BMI. As with cycle time, the investigators speculate that the

overweight participants may not have walked at their preferred

walking speed at week one, but increased comfort with the

investigators and testing conditions at week eight was reflected in

their preferred walking speed and therefore in their normalized

velocity.

Step extremity ratio is the ratio of step length to leg length.

There was no significant change in either step length or leg length.

The investigators had expected an increase in step length and

therefore step extremity ratio with a decrease in BMI among the

overweight participants.

FAP score increased for the overweight participants from week one to

week eight, but the increase was not significant. We expected this

increase because of the decrease in BMI. We speculate that with a

longer intervention the FAP score would continue to increase as BMI

continued to decrease. The overweight group was expected to have a

lower FAP score than they actually demonstrated at the start of the

intervention; the excessive weight of this group should have had a

more adverse affect on gait.2,3 However, they may have developed

compensatory mechanisms for improving their ambulation, which would

explain their higher initial FAP scores.

Diet, exercise, and weight loss

The intervention used in this study, the " Defensive Eating/Healthy

Living Program, " consisted of a nutritional/behavioral modification

lesson, diet, and exercise for weight loss in school-age children.

Five studies contrasting diet with diet and exercise varied with

respect to diet and exercise methods, but all demonstrated better

changes in weight and fitness for diet and exercise groups than for

diet-only groups.13 A meta-analysis of 41 intervention studies to

reduce obesity in children concluded that interventions that

incorporated knowledge, behavioral, and exercise components were

more likely to result in weight loss than those that did not

incorporate these strategies.14 The " Defensive Eating/Healthy Living

Program " required the participation of at least one parent/guardian.

Involvement of at least one parent has been found to produce better

weight loss outcomes than interventions that did not involve

parents.13 Following recommendations from the surgeon general's

report, the program encouraged increased daily physical activity,

such as walking, and decreased sedentary activities, such as playing

computer games and watching television.1

The intervention used in this study consisted of one-hour weekly

sessions over an eight-week period and no quantifiable monitoring of

diet and exercise during the course of the intervention. As

evidenced in this study, there was a slight decrease in BMI from

week one to week eight in the overweight group. A longer

intervention and/or an increased frequency of sessions may have

produced a greater decrease in BMI. Other studies that incorporated

a longer study period and/or increased frequency of the intervention

have demonstrated greater decreases in weight.13

Study limitations

Although the children were encouraged to maintain a diet and

activity journal to chart their progress, it was difficult to

accurately assess their compliance with the program. A verbal

assessment of each child's activity and diet made at each weekly

session, in lieu of a written assessment, was the only monitoring of

their progress. There is a large amount of literature demonstrating

the problems of obtaining valid measures of activity.13 In addition,

no weekly weight loss goals were established, nor were any weight

measurements taken over the course of the program.

Not matching for gender and not assessing for puberty were found to

be limitations of this study. During adolescence, a redistribution

of fat from the periphery to the trunk occurs in both boys and

girls. However, the extremity loss has been reported to be greater

in boys than in girls. As a result, men are likely to have a more

central body fat distribution while women are likely to have a more

peripheral fat distribution. This differential in distribution of

body fat in boys and girls should be considered when assessing the

effects of a weight loss program on reduction in body fat.13 Weight

loss programs should take into account the onset of puberty to

address this issue. The difference in fat distribution between boys

and girls during puberty will also have an effect on gait because of

the greater adiposity of pubertal girls' thighs.

Conclusion

Alterations in BMI had an effect on specific gait characteristics of

overweight school-age children. Results of this study demonstrate

that the percentage of time in double support was increased in

relation to the normal-weight children, and the percentage of time

in swing was decreased for the overweight group. Base of support was

increased and cycle time was slower for the overweight group. Future

research studies are needed, incorporating a longer and more

intensive diet and exercise intervention to establish a greater

decrease in BMI. Larger sample sizes are recommended for

generalization to the population. Collecting data periodically

throughout the course of the study may help identify patterns of

change.

Maureen Becker, PT, DHsc, is the academic coordinator for clinical

education; Arthur , PhD, PT, FAPTA, is director for research;

and Rothman, EdD, PT, is director, all in the physical

therapy department at the College of Staten Island of the City

University of New York. Amy Ng, MSPT, is a physical therapist at

South Nassau Communities Hospital in Long Island. Barker,

MSPT, works for Volunteers of America evaluating and treating

preschool children with developmental delay in New York City.

Brady, MSPT, is working at Staten Island University Hospital. Andy

Koser, MSPT, is working for Health South in New Jersey. All four

were students attending the College of Staten Island of the City

University of New York and completed this research as part of their

master's degree requirements.

Acknowledgements

We would like to thank L. Freedland, PhD, Ruth Sundback, RD,

and Lightcap for their help with this study.

References

1. The Surgeon General's Call to Action to Prevent and Decrease

Overweight and Obesity, 2001.

www.surgeongeneral.gov/topics/obesity/calltoaction/CalltoAction.pdf.

2. Spyropoulos P, Pisciotta JC, Pavlou KN, et al. Biomechanical gait

analysis in obese men. Arch Phys Med Rehabil 1991;72(13):1065-1070.

3. Hills AP, AW. Gait characteristics of obese children. Arch

Phys Med Rehabil 1991;72(60):403-407.

4. McGraw B, McClenaghan BA, HG, et al. Gait and postural

stability in obese and nonobese prepubertal boys. Arch Phys Med

Rehabil 2000;81(4):484-489.

5. Sutherland D. The development of mature gait (review paper). Gait

Posture 1997;6(2):163-170.

6. Preis S, Klemms A, Muller K. Gait analysis by measuring ground

reaction forces in children: changes to an adaptive gait pattern

between the ages of one and five years. Dev Med Child Neurol 1997;39

(4):228-233.

7. Dietz WH, Bellizzi MC. Introduction: the use of body mass index

to assess obesity in children. Am J Clin Nutr 1999;70(1):123S-5S.

8. Mei Z, Grummer-Strawn LM, Pietrobelli A, et al. Validity of body

mass index compared with other body-composition screening indexes

for the assessment of body fatness in children and adolescents. Am J

Clin Nutr 2002;75(6):978-985.

9. CDC Body Mass Index. www.cdc.gov/nccdphp/dnpa/bmi/bmi-for-

age.htm, accessed 7/22/02.

10. Gaitrite gait analysis. www.gaitrite.com/overview.htm,

www.gaitrite.com/spatial.htm, www.gaitrite.com/time.htm, accessed

3/18/02.

11. AJ. Functional ambulation profile. Phys Ther 1974;54

(10):1059-1065.

12. McDonough AL, Batavia M, Chen FC, et al. The validity and

reliability of the GAITRite system's measurements: a preliminary

evaluation. Arch Phys Med Rehabil 2001;82(3):419-425.

13. Epstein LH, KJ, Myers MD. Exercise in treating obesity

in children and adolescents. Med Sci Sports Exerc 1996;28(4):428-435.

14. Jerum A, Melnyk BM. Effectiveness of interventions to prevent

obesity and obesity-related complications in children and

adolescents. Pediatr Nurs 2001;27(6):606-610.