01 96-601 1 /86/0704-Ol73$02.00/0 THEJOURNALOFORTHOPAEDICANDSPORTSPHYSICALTHERAPY Copyright O 1986 by The Orthopaedic and Sports Physical Therapy Sections of the American Physical Therapy Association

Kinematic Effects of Heel Lift Use to Correct Lower Limb Length Differences* WILLIAM D. BANDY, MA, PT, ATC,? WAYNE E. SINNING, PhD*

The effect of the heel lift to correct a limb length difference was studied by electrogoniometry (elgons) in four male subjects with a limb length inequality between 3/~6 inch (0.48 cm) and 3h inch (0.95 cm). Six elgons were attached to bilateral hip, knee, and ankle joints as the subject walked (3 mph) and jogged (6 mph) on the treadmill twice, once with the heel lift and once without. Recordings from the elgons examined maximal flexion and extension during support and swing phase, amplitude of movement (ROM), duration of each movement, and angular velocity of each joint. Within the limitations of the study, the following conclusions were drawn: 1) the addition of a heel lift did not appear to significantly affect biomechanical measures of gait; and 2) insertion of a heel lift did tend to cause more symmetrical movement for the maximum angle of hip extension and ROM of the swing plantarflexion phase of the ankle but more asymmetrical ROM of the swing flexion phase of the knee.

+ Research has shown that a lower limb length

difference (LLD) may be present in 25-93% (de- pending on the source) of the p ~ p u l a t i o n . ~ ~ ~ ~ . ' ~ A LLD that may be considered minor (1b1/2 inch, 1 - 2 cm) and not cause any problems in a nonathlete or sedentary individual, may become a major problem in the active athlete. The constant, re- petitive strides made while training for a distance event may be predisposing to several injuries. Symptoms that have been suggested to be caused by LLD include sciatica as well as pain in the hip, knee, ankle, and iliotibial band.317.9.10.12.13

The treatment for the alleviation of LLD is the insertion of a heel lift equal to the height of the discrepancy and inserted into the shoe of the short leg in order to equalize the lengths of the lower Several authors have sup- ported the theory suggesting that the use of the heel lift ensures biomechanical symmetry of movement which is essential to decrease the chance of injury to an active i n d i ~ i d u a l . ~ . ~ . ' ~ . ~ ~

Other writers have reached conclusions con- trary to the authors previously reviewed. H ~ l t , ~ in an investigation of Swedish industrial and factory workers, found no differences in the incidence of LLD in workers complaining of lumbar spine trou- ble and the entire study population. Fisk and Biagent4 state that "moderate degrees of leg length discrepancy played little if any part in the etiology of back ache." G r ~ s s , ~ in reviewing young adults' perceptions on the functional effects of LLD, found that individuals with less than an inch (2 cm) difference did not consider the discrepancy to be a problem. However, neither investigator studied athletes as a separate group.

Only one investigation analyzed the changes that occur with the insertion of a heel lift2 Inves- tigators hypothesized that the insertion of a heel lift had a positive effect on the running gait; how- ever, two problems arise. First, the study investi- gated the kinetic energy changes occurring in the lower extremity but did not define the actual

'A research project submitted to the Kent State University PERD changes in displacement occurring at the joints of

Graduate Department in partial fulfillment of the requirements for the the body. Second, the study dealt with a LLD of degree of Master of Arts in Physical Education. t Mr. Bandy is presently on the faculty of the University of Central

over 1 inch (2.5 cm) which is very large clinically.14 Arkansas, Little R W ~ , AR. It is suggested from the literature that minor $Dr. Sinning is a professor of Physical Education and Director of the deviations from a normal running gait due to LLD Applied Physiology Research Laboratoryat Kent State Un~versity, Kent, OH. may severely handicap the athlete's ability to per-

173

174 BANDY AND SINNING JOSPT Vol. 7, No. 4

form and be predisposing to further injury.7s10 However, there is little biomechanical evidence as to what these deviations are, what specific changes occur with the addition of a heel lift, and if the addition of the heel lift is indeed advanta- geous to efficient locomotion.

The purpose of this study was to investigate bilateral changes in range of motion, duration, and angular velocity of the hip, knee, and ankle during the swing and support phase when a heel lift, which is inserted into the shoe of the short limb in order to correct a LLD in an effort to produce symmetrical movements, is removed. Potential changes in gait patterns were examined while walking (3 mph) and jogging (6 mph) on the tread- mill.

METHODS

Subjects

Four male volunteers having a LLD between 3/1~ inch (0.48 cm) and 343 inch (0.95 cm) served as subjects for this study. All subjects had worn a heel lift for more than 1 year and for the pur- poses of this study were defined as "experienced" heel lift users. Descriptive data are presented in Table 1.

The criteria for LLD used in this study are that the subjects present with the following character- istics on the short limb side: I ) low anterior su- perior iliac spine, 2) low posterior superior iliac spine, and 3) low iliac crest. A LLD between 3 /1~

and 343 inch was chosen because this difference is large enough to cause lower extremity asym- metry that can be accurately measured, but not so large that a correction beyond a heel lift is necessary. Subotnick14 suggested that forefoot control is necessary with an orthotic device to correct a difference greater than 1/2 inch (1.3 cm), and a heel lift alone does not supply enough foot control. In order to measure LLD, each subject

was asked to stand erect with his bare feet facing forward, shoulder width apart, knees held in full extension, and hindfoot in the neutral position. The LLDs of the subject were measured with a Pelvic Level (J. A. Preston Corp., 60 Page Rd., Clifton, NJ 07012). The Pelvic Level was posi- tioned on the superior aspect of the subject's iliac crest with the bubble in the leveling device show- ing tilt in the subjects with a LLD. Calibrated blocks were placed beneath the heel of the short side until the crests were level, as indicated by the bubble on the leveling device moving to the center. The width of the blocks needed to level the crests was considered the amount of LLD the subject possessed. The subject was then given a heel lift of Pelite (Alimed Co., 68-70 Harrison Ave., Boston, MA 021 11) material structured to equal- ize the LLD and placed in the shoe of the short limb.

Equipment

Six electrogoniometers (elgons) were attached to the subject at bilateral hip, knee, and ankle joints. An elgon is an instrument which uses elec- trical circuitry and recording devices to make a continuous record of joint movement. Several types of elgons have been designed for use with the different joints. The one used in this investi- gation is similar to the universal elgon described by Adrian et al.' A Wheatstone Bridge was used to balance the circuits and calibrate the output. Data were recorded by a model 1600 Strip Chart Recorder manufactured by the MFE Corp.

PROCEDURES

Data Collection

Subjects reported to the testing area wearing shorts and rubber soled running shoes. All sub- jects had previous experience with the treadmill

TABLE 1 Individual subject data

Weight Height LLD Leg Time of Subject Age heel lift use

%a inch Right 13 months 0.48 cm 3/s inch Left 36 months 0.95 cm l/4 inch Right 18 months 0.64 cm % inch Right 29 months 0.64 cm

JOSPTJanuary 1986 HEEL LIFT FOR LOWER LIMB LENGTH DIFFERENCES 175

\ for slightly more competitive athletes during their

Fig. 1 . Subject on treadmill with six elgons attached to both limbs.

c DIRECTION OF PAPER TRAVEL

HIP

ANKLE

E

F

I

H

v v v v v v K

KNEE

Fig. 2. A typical elgon tracing for one limb taken while running at 6 mph. A downward tracing indicates flexion; an upward tracing, extension. Specific points measured are: A, maximum hip extension; 6, maximum hip flexion; C, maximum knee extension, swing phase; D, maximum knee flexion, support phase; E, maximum knee extension, support phase; F, maxi- mum knee flexion, swing phase; G, ankle plantarflexion at end of swing phase; H, maximum ankle dorsiflexion, support phase; I, maximum ankle plantarflexion, support phase; J, maximum ankle dorsiflexion, swing phase; K, foot contact.

and were given the opportunity to walk and jog at the experimental speeds (3 and 6 mph) until they felt comfortable. These speeds were chosen because they represent a "normal" walking speed and are within a range that might be used when jogging in a physical fitness class (or even used

training runs). Six elgons were attached to both limbs at the

hip, knee, and ankle joints according to proce- dures described by Adrian et al.' (Fig. 1). These instruments were attached with velcro straps and reinforced with adhesive tape. Before testing, the subject stood in an erect position and elgon re- cordings were made to establish a reference po- sition for proper analysis of the recordings while the joints were moving. In order to ensure that no elgon displacement occurred during testing, the reference position of the elgon was recorded again following testing. Adrian et al.' have shown that the elgon measures joint angles with less than 3% error when attached by experienced operators. Each subject walked and jogged on the treadmill twice, once with the heel lift and once without while the recording was being taken.

Data Analysis

From the elgon recordings (Fig. 2), information was obtained from each joint. Measurements in- cluded maximal flexion and extension (mean an- gles) during support phase and swing phase of the knee and ankle. Maximal flexion is the joint angle at the termination of flexion and beginning of extension; maximal extension is the converse. The swing phase is the position of the stride during which foot swings free of support. The support phase is when the foot is on the surface and begins with heel strike and ends with push off. Maximal flexion and extension of the hip was also examined.

In addition, the amplitude of movement (or total range of motion) which is the difference between maximum flexion and maximum extension was measured. The following movements were ex- amined for the knee and ankle: 1) swing exten- sion, 2) support flexion, 3) support extension, and 4) swing flexion. Only the total amplitude of move- ment for flexion and extension were examined for the hip.

Also, the duration of each movement, which is time required to move from one maximum angle to the other, was computed for the hip, knee, and ankle from the horizontal travel speed of the re- cording paper. Angular velocity was also com- puted. This computation gave an average value for complete movement since no allowance is made for the positive and negative accelerations at the beginning and end of the movements.

Statistical analysis utilized the analysis of vari-

BANDY AND SINNING JOSPT Vol. 7, No. 4

Degrees

o f

Motion

Short Limb

-------- Long Limb

Without With

Use o f Heel L i f t

Fig. 3. Interaction effect of lift x limb for angle of maximum extension of the hip. All differences between the short limb and long limb were significant. However, differences for each limb with and without the lift were not significant.

ance (ANOVA) with repeated measures. The Tu- key HSD Test was used for post hoc analysis. Significance was accepted at p < 0.05.

RESULTS

No significant differences were found due ex- clusively to the use of the heel lift for any de- pendent variables. However, there were interac- tions which were found to be significant. The interaction between heel lift (with and without) and limb (short and long) was significant for the angle of maximum hip extension (F = 25.31, p < 0.01), the range of motion of the swing plantarflexion phase of the ankle (F = 14.31, p < 0.03), and the range of motion of the swing flexion phase of the knee (F = 180.24, p < 0.001). These interactions are graphed in Figures 3 through 5.

DISCUSSION

Although no significant differences were found due exclusively to the use of the heel lift for any

of the measures in the study, significant interac- tions were found. These interactions indicated trends toward symmetry of movement with the insertion of the heel lift for the angle of maximum hip extension and the range of motion of the swing plantarflexion phase of the ankle. The interactions indicated a trend toward asymmetry of movement with the insertion of a heel lift upon range of motion of the swing flexion phase of the knee. No other measurements related to the experimental question ihvestigating biomechanical changes due to the insertion of a heel lift to correct LLD were significant.

These results differ from the only other investi- gation on the kinematic changes in gait due to LLD by DeLacerda and Wickoff .2 The difference may be due to several reasons. The study by DeLacerda and Wickoff used only one female subject and based any conclusions from the in- vestigation on that single subject. In addition, the

JOSPTJanuary 1986 HEEL LIFT FOR LOWER LIMB LENGTH DIFFERENCES

Short Limb

-- - - - - - - Long Limb

Without With

Use o f H e e l L i f t

Fig. 4. Interaction effect of lift x limb for range of motion of swing plantarflexion phase of the ankle. Differences between short limb with lift and long limb no heel lift, and between short limb no lift and long limb no lift were significant.

subject had a LLD of 3.18 cm (1.25 inches) while all the LLDs in the present study were'less than 1 cm (Y2 inch). Last, DeLacerda and Wickoff measured kinematic changes at a walking speed (3 rnph) while the present investigation also incor- porated a faster speed (6 mph).

There are three limitations of the design of this study. The first is that the number of subjects is small. A greater number of subjects would be expected to reduce the variability and increase the probability of finding significant differences. This is especially significant in regard to the cases where post hoc tests did not support the signifi- cant F ratios for interaction. Second, even though all four subjects were young, active males with a LLD who had worn a heel lift for over 1 year, they differed in their activity level. Two of the subjects were actively engaged in a running program jog- ging 3-5 miles, 4-5 times a week. The other two subjects were active in sports such as biking,

softball, and soccer and were not involved in a consistent jogging program. Further research uti- lizing a larger number of subjects and a more specific population (trained subjects) may lead to different results.

A third limitation of the present investigation was that all the measurements were taken in a sagittal plane of the lower extremity. No attempt was made to measure rotation of the long axis of the limb or movements in the transverse plane which occur during gait such as tibia1 rotation and pronation-supination of the foot. While very few significant differences were obtained in this study, further investigations incorporating the rotary components of movement may lead to interesting findings.

CONCLUSION

1) The addition of a heel lift to the short limb did not appear to significantly affect the biome-

BANDY AND SINNING JOSPT Vol. 7, No. 4

Short Limb

-- - - - - - - Long Limb

Without H i t h

Use of Heel L i f t

Fig. 5. Interaction effect of lift x limb for the range of motion of swing flexion phase of the knee. All differences between short limb and long limb were significant. However, differences for each limb with and without the lift were not significant.

chanical measures of maximal flexion and exten- vidualized to each subject and that frequent mon- sion during support and swing phase, range of itoring of the subject's response is mandatory. motion, duration of movement, and angular veloc- ity of the hip, knee, and ankle while walking (3 in p~~~r~t~'~,";~,"~~s and Don

their assistance

mph) and jogging (6 mph) on the treadmill. 2) Insertion of a heel lift did tend to cause more REFERENCES ~ - - ~ - - -

symmetrical movement for the angle Of 1. Adrian M. Tipton M. Karpovich YP: Electrogoniometry Manual. hip extension and range of motion of the swing Springfield, MA: Springfield College, Physiological Research Lab ~lantarflexion phase of the ankle, but more asym- oratow (Mimwraphed). 1965 metrical range of motion of the swing flexion 2. De~a&rda ~ ~ , ~ i c k o f f 0: Effect of lower extremity asymmetry on

the kinematics of gait. J Orthop Sports Phys Ther 3:17-20, 1982 phase of the knee. 3. DuPlessis MP: Stress fractures and overuse syndrome. S Afr Med

Implications J 58:670. 1980

4. Fisk JW, Biagent MC: Clinical and radiological assessment of leg lenqth. N Z Med J 81:477-480, 1975

It would appear reasonable to employ a heel lift 5. Gross RH: Leg length discrepancy, how much is too much? Orthopedics 1 :307-310

On a trial basis for the treatment of symptoms 6. Hult L: The Munkfors investigation. Acta Orthop Scand (Suppl16) possibly related to LLD, but it is important to 24:30-31,1954

Iealize that all subjects do not respond'to the use 7. James SL, Bates BT, Osternig LR: Injuries to runners. Am J Sports Med 6:40-50,1978

Of a heel lift in the same way. it is 8. Klein K, Buckley J: Asymmetries of growth in the pelvis and legs imperative that treatment with a heel lift be indi- of growing children. Am Correct Ther J 22:53-55, 1968

JOSPTJanuary 1986 HEEL LIFT FOR LOWER LIMB LENGTH DIFFERENCES 179

9. Klein K: Progression in pelvic tilt in adolescent boys from elemen- 13. Subotnick SI: The short leg syndrome. J Am Podiatry Assoc tary through high school. Arch Phys Med Rehabil54:57-58, 1973 66:720-723,1976

10. Mann RA, Baxtel D, Lutter L: Running symposium. Foot Ankle . 14. Subotnick SI: Limb length discrepancies of the lower extremity (the 1 :190-224,1981 short leg syndrome). J Orthop Sports Phys Ther 3:ll-16,1981

11. Pearson WM: Progressive structural study of school children. J Am Osteopath Assoc 51 :155-157, 1951

12. Redler I: Clinical significance of minor inequalities in leg length. New Orleans Med Surg J 104:308-312, 1952