Elbow Moment and Forces at the Hands During

8/7/2019 Elbow Moment and Forces at the Hands During

1/5

Elbow M om ent and Forces a t the Hands D uringSwing-Through Axi l lary Crutch Gai tMARC REISMAN,RAY G. BURDETT,SHELDON R. SIMON,and CYNTHIA NORKIN

We investigated swing-through axillary crutch gait (nonweight bearing on the leftlower extremity) to determine the effects of gait speed, crutch length,and handleposition on the forces exerted at the hands and on the moments exerted aboutthe elbow joints. Ten healthy subjects, skilled in swing-through crutch gait, walked1) at three speeds using fitted crtuches, 2) at a fixed speed with four differentcrutch lengths, and 3) at a fixed speed with four different handle positions. Wecollected ground reaction forces that exerted simultaneously on the right crutchand motion data with a force plate and three high-speed movie cameras. Abiomechanical model was developed to calculate the forces exerted at the righthand and the moments exerted about the right elbow joint. Changing gait speedfrom slow to the normal gait of the subject showed statistically significant effects(p < .05) on the forces at the hand. When we changed crutch heights for thesubjects, we found no significant effects on the forces at the subjects' hands.Changing handle position significantly affected the moment at the elbow. Increasing the elbow-flexion angle above 30 degrees by raising the crutch handle 1 to 2in resulted in a 100 percent increase in elbow-extension moment. We found acorrelation of .82 between actual average elbow-flexion angle and elbow-extension moment. Changing gait speed or crutch length did not affect elbow moment.Keywords: Biomechanics, Crutches, Elbow joint, Gait.

Fitting axillary crutches for patients is a com m on procedurefor physical therapists. The scientific basis for criteria used toadjust crutch length andhandle position for a patient, however, has not been studied in detail. Standards proposed formeasuring crutch length have included the following: 77percent of the patient's height,1 height minus 18 in,* 2 heightminus 16 in,3 and one and one-half or one to two fingersbelow the axillary fold to a point 4 in4, 5 or 8 in6 from the sideo f the foot. Suggested standards for measuring hand-pieceposition have varied from designating specific joint p lacementof the elbow angle at 30 degrees of flexion5, 7 to vague descriptions that the elbow should be slightly bent.8, 9 N o n e of thesecriteria has been based on scientific or biomechanical data.

Only recently have analytical studies of crutch gait beenperformed. McBeath et al compared energy requirements ofcrutch walking against normal gait.10 They found that partialweight-bearing crutch gait required 33 percent more energythan normal gait, and that nonweight bearing crutch gait

required 78 percent more energy. Peacock did a myographicanalysis of swing-through crutch gait w ith still photographs.11He described an external flexion moment that acted aroundthe elbow during crutch gait, which is balanced by an extension mo me nt provided by the upper arm m uscles, specificallythe triceps brachii muscle. Wells found that the mechanicalwork during swing-through crutch walking was approximatelythe same as in normal walking but that more of this work wasdone by the upper extremities.12 Because these muscles arenot functionally designed for supporting the body's weight,they fatigued more rapidly than the lower extremity muscles.Shoup et al performed a biomechanical displacement analysisof a swing-through crutch gait to establish criteria for improving crutch design.13 Shoup later used these criteria to developa new type of forearm crutch for children.14 Other investigators have measured the forces acting on crutches and otherambulatory aids.15-17

Although these studies have looked atenergy consum ptionin crutch walking and at the forces acting on the crutches, theforces that acted on the joints of the body have not beenadequately examined. The wrist and elbow joints are heavilyrelied on for support during the swing-through phase ofaxillary crutch gait. Although joint forces at the wrist18 andtorque at the elbow have been measured for isometric contractions,19-21 these forces have not been measured duringcrutch walking. Thepurpose of this study was to show thatvariation of speed, crutch length, and handle position wouldpositively affect the forces exerted by the hands and themo me nt exerted by the elbow joint during nonweight-bearingaxillary crutch walking.

Mr. Reisman is a staff member, Washington University, Department ofPhysical Therapy, Irene Walter Johnson Institute of Rehabilitation, St.Louis,MO 63110 (USA).Dr. Burdett is Assistant Professor, Program in Physical Therapy, School ofHealth Related Professions, University of Pittsburgh, Pittsburgh, PA 15261.Dr. Simon is Director, Gait Analysis Laboratory, Children's Hospital, Boston, MA 02115.

Ms. Norkin is Assistant Professor, Sargent College of Allied Health Professions, Boston University, Boston, MA 0 2215.This article was submitted August 2, 1983; waswith theauthors for revision38 weeks; andwas accepted November 11, 1984.

* 1 in = 2.54 cm.

Volume 65/Number 5, May 1985 601


2/5

M E T H O DS u b j e c t s

Ten healthy women, randomly selected from a group ofphysical therapists and physical therapy students who hadsome skill in performing a swing-through crutch gait, weresubjects in this study. Their average age was 27.6 years. Torule out height as a factor in m easuring forces and m oments,their heights were limited to a range of 63.5 to 66 in. Noneof the subjects reported any history of gait abnorm alities. Weused a form approved by Boston University and the GaitAnalysis Laboratory of Children's Hospital to obtain informed consent from each subject. Approval for this studywas received from the Human Subjects Committee of theUniversity.P r o c e d u r e

For each gait session, the subjects wore shorts, t-shirts, andsneakers. To aid in identifying landmarks on film, smallsquares of black tape with white dots in the center were placedover the right acromion process, the right lateral epicondyleof the hum erus, the dorsum o f the right wrist m idway betweenthe ulnar and radial styloid processes, the right crutch tip, andthe center of the axilla crossbar. The crutches used in thisstudy were standard wooden axillary crutches with rubber-covered axilla crossbars, hand crossbars, and tips. Extra holeswere drilled to allow for 1-in adjustments of the handles. Ateam of physical therapists at the Gait Analysis Laboratorydecided the method for fitting the crutches based on commonly used criteria.3,5,7 Crutch length was measured in thefollowing manner: The subject stood with her arm abductedto 90 degrees. The crutches were measured from a point 2 inbelow the axilla to a point 12 in lateral to the midline of thebody on a line along the tips of the shoes. The handle positionwas adjusted to provide 30 degrees of elbow flexion. Twophysical therapists independently measured the elbow angleby centering a standard 7-in plastic goniometer over the lateralcondyle of the humerus and by using the acromion and radialstyloid processes as references.Crutch gait may involve full, partial, or nonweight-bearingstatus on the part of either leg. For the purposes of this study,swing-through axillary crutch gait was defined as a nonweight-bearing gait with the left foot off the ground. The right foot,therefore, supported all of the force not carried by thecrutches. The "normal" average speed of crutch gait for asubject with this height was determined in a pretest to be 0.73m/sec. We considered that speeds of 0.46 m/sec and 1.12 m/sec deviated enough from the normal to be considered fastand slow speeds. The cadences and step lengths to achievethese speeds were also determined from the pretest. Table 1gives the gait speeds, cadences, and step lengths from thisstudy.

T A B L E 1A v e r a g e S p e e d , C a d e n c e , a n d S t e p L e n g t h o f C r u t c h - W a l k i n gTr ia lsSpeed

SlowNormalFast

C a d e n c e0.46 m/sec 60 steps/min0.73 m/sec 72 steps/min1.12 m/sec 88 steps/min

Step Length1.5 ft/step2.0 ft/step3.5 ft/step

TABLE 2A v e r a g e N o r m a l i z e d F o r c e o n H a n d s D u r i n g C r u t c h S t a n c e o fD i f fe rent Tr ia lsSpeed

Crutch lengthHandle position

Slow(0.46 m/sec).37+2 in.41+2 in.39

+1 in.39+1 in.39

Normal(0.73 m/sec).41Fitted.41Fitted.41

- 1 in.39- 1 in.40

Fast(1.12 m/sec).40- 2 in.40- 2 in.41

To determine the effects of speed, crutch length, and handleposition on the forces at the hands and m oments at the elbow,we asked each subject to walk with the crutches under 11different conditions: 1) at the three different speeds, usingcrutches fitted according to the standard selected; 2) at thenormal speed, using four different crutch lengths (longer by 2in and by 1 in; shorter by 2 in and by 1 in), but with thehandle position set so that a 30-degree flexion angle existedat the elbow; and 3) at the normal speed, using four differenthandle positions (higher by 2 in and 1 in; lower by 2 in and1 in), but with the crutch length the same as the fittedcondition.Speed was controlled by controlling both cadence and steplength. Cadence was set by a m etronome, and step length wascontrolled by marks placed along thefloor.The subjects wereasked to place each step of the foot or crutch near the markers.A practice session was taken for each walk so that the subjectcould become familiar with each speed, crutch length, andhandle position. The walks were initiated from specific spotson thefloorso that the right crutch would land in the m iddleof a force plate after three strides.F o r c e a n d M o t i o n A n a l y s i s

The setup of the equipment, electronic systems, and computer programs for computing the elbow moment and theforces at the hands was developed at the Gait A nalysis Laboratory and Children's Hospital, Medical Center, Boston. Thebody m otion, in addition to ground reaction forces acting onthe crutch tip in the vertical, anterior-posterior, and medial-lateral directions, were collected simultaneously by an Advanced Mechanical Technology force platform and by threehigh-speed Photosonics 16-mm movie cameras positionedorthogonally on three sides of the walkway. The right and leftcameras were 12 ft away from the center of the force plate,and the front camera was 29 ft away. After the films weredeveloped, they were analyzed on a Vanguard Motion Analyzer || with a Graf Pen Sonic Digitizer# to collect the two-dimensional coordinates of specific points on each film. Thethree-dimensional coordinates o f the wrist, elbow, and shoulder joints; crutch tip; and center of the axilla crosspiece werethen calculated from these two-dimensional coordinates. Simon et al have described this force and motion analysis systemin more detail.22

Advanced Mechanical Technology, Inc, 141 California St, Newton, MA02158. Instrumentation Marketing Corp, 820 S Mariposa St, Burbank, CA 91506.1 ft =.3048 m.| | Vanguard Instrument Co, Melville, NY 11747.# Science Accessories Corp, Southport, CT 064 90.

602 PHYSICAL THERAPY


3/5

RESEARCHB i o m e c h a n i c a l M o d e l

We developed a biomechanical model of crutch walking tocalculate the forces acting on the hands and the momentexerted by the elbow extensors. The assumptions used in thismodel were 1) the m ass and acceleration of the crutch, forearm, and hand could be neglected during crutch stance; 2)forces on the crutch occur only at the tip, handle, and crutchtop; 3) the force at the crutch top is perpendicular to thesagittal plane; and 4 ) no twisting m oments existed that wereexerted on the crutch from the wrist (Figure).By neglecting the mass and acceleration of the crutch duringcrutch stance, the forces acting on the crutch from the handwere calculated from Newton's laws of equilibrium. Thedirection of the axis of the elbow joint was assumed to beperpendicular to a plane formed by the shoulder-, elbow-, andwrist-joint centers. This axis is not, in general, perpendicularto the sagittal plane during crutch walking. Therefore, thecomponent of the resultant force within this plane was calculated. This force was multiplied by the perpendicular distance between the elbow-joint center and the action line ofthis component to get the external moment exerted by thecrutch about theflexion-extensionaxis of the joint. By neglecting the mass and acceleration of the forearm duringcrutch stance, the internal moment exerted by the elbowextensors was equal to this external moment exerted by thecrutch. The resultant force was divided by body weight andwas averaged over the crutch-stance time to give the averageresultant normalized force at the hands. Elbow-extensionmoments were normalized by dividing by body weight andforearm length and were averaged over either the crutch-stance phase or the entire gait cycle. We thought average forceand average moment were better indicators of muscle effortthan peak force or moment, which may be exerted for onlyan instant. The actual elbow angle at each instant in timeduring crutch stance was also calculated and was used toobtain the average angle of the elbow during crutch stance.D a t a A n a l y s is

Six analyses of variance (ANOVAs) with repeated measureswere performed to determine the effect of gait speed, crutchlength, and handle position on the resultant normalized forceat the hand and on the normalized mom ent at the elbow. Thesignificance level was set at p < .05. Individual comparisonsof m eans were made w hen appropriate by using the Neuman-Keuls multiple comparison test. A Pearson product-momentcorrelation coefficient was used to determine the correlationbetween elbow torque and elbow-flexion mom ent.RESULTS

Table 2 shows the average resultant force exerted on thecrutch by the hands during crutch stance as a function ofspeed, crutch length, and handle position . Analysis of varianceshowed the following results. 1) There was a statisticallysignificant difference (p < .05) between the average forceexerted at the hands during slow walking and normal walking.This force difference, however, is only 3.8 percent of bodyweight smaller than the force at normal speed. Althoughstatistically significant, this difference may not represent anyvaluable clinical difference. 2) We found no significant differ-

F i g u r e . Forces exerted on the right crutch from the ground(Rx,Ry,Rz), the hand (Hx,Hy,Hz), and the body (Bx).

ence (p > .05) among the average force values as a functionof crutch length or handle position.The forces at the hand create moments at the elbow jointthat must be balanced internally by the triceps brachii muscleacting on its lever arm. These normalized mom ents are shownin Table 3. The average elbow moment changed very littlewith walking speed and indicated that elbow-extension muscleeffort did not vary much with speed of walking. The averageelbow moment exerted during crutch stance also did not varysignificantly with crutch length. We found a significant variation, however, in elbow moment with changes in handleposition. The two higher handle positions resulted in abouttwice as much moment as the fitted position or the two lowerpositions.Volume 65/Number 5, May 1985 603


4/5

T A B L E 3A v e r a g e N o r m a l i ze d E l b o w - E x t e n s io n M o m e n t D u r in g C r u t c hS t a n c e ( % )Speed

Crutch lengthHan dle position

Slow(0.46 m/sec)4.14+2 in5.15+2 in8.20

+1 in5.63+1 in8.12

Normal(0.73 m/sec)4.19Fitted4.19Fitted4.19

- 1 in5.24- 1 in4.51

F a s t(1.12 m/sec)4.52- 2 in5.30- 2 in3.86

The effect of changing handle positions on the elbow anglecan also be seen in Table 4. At the fitted position, the elbowangle averaged 31 degrees among all the subjects before crutchwalking. Raising the crutch handle resulted in an increasedelbow angle of abou t 11 degrees per inch ; lowering the han dleresulted in a decrease in elbow flexion of about 8 or 9 degreesper inch. The actual, average flexion angle during crutchstance was, in general, much smaller than the angle measuredbefore gait, but this difference was not as great for the fittedposition and the lower handle positions. The correlationbetween ac tual, average elbow-flexion ang le and elbow-extension mom ent was high (.82).DISCUSSION

Several compensations that could account for the lack ofsignificant differences in forces at the hands existed in thisstudy. The shoulder girdle complex or the stance leg, througheccentric contractions, may absorb some of the energy causedby increases in speed or changes in crutch length or handleposition. The shoulder girdle muscles may also help to keepthe c enter of gravity at a relatively co nstan t level during crutchstance, which would decrease forces on the crutches and onthe han ds. Another way that the ce nter of gravity may be keptat a relatively constant level is by adjusting the abductionangle of the crutches from the body. When the crutches arelong, they can be placed further from the body. All of theabove occurred in this study.

The elbow-joint moment is the product of the force at thehands times the perpendicular distance from the elbow jointto the force (Figure). This perpe ndicu lar distance is a functionof the elbow angle; in general, increasing the flexion anglewill increase the perpendicular distance. The force at thehands w as approximately th e sam e for each crutch length andeach gait speed, but the elbow angle was readjusted to 30degrees of flexion after each length change. Therefore, theresult of no significant difference in elbow moments with

T A B L E 4E f fe c t o f C r u t c h H a n d l e P o s it io n o n E l b o w A n g l e a n d M o m e n t

Position

2 in high1 in highFitted1 in lo w2 in lo w

Flexion Ang le (%)Pregait

5342312214

Dur ing Ga i t2623161311

Extens ion Moment (%)

8.28.124.194.513.86

crutch-length changes and speed changes is not surprising.Energy consumption has been shown in other studies toincrease with speed of crutch walking.23 The present studyindicates this increase in energy is probably caused by increased effort by muscles crossing other joints, possibly theshoulder girdle muscles and the muscles of the stance leg,rather than the elbow-joint muscles.The difference in elbow mo me nt tha t occurred with changesin handle position is almost entirely caused by an increase inthe lever-arm distance between the elbow joint and the actionline of the force on the hands. For the three, lower handlepositions, the actual elbow angles were very similar, eventhough the initial angles were different. As a result, theaverageextension moments were also very similar. At the higherhandle positions, the subjects used m uch larger elbow flexionangles during gait, and, therefore, the elbow-joint momentswere much larger also. By using some relatively quick andinexpensive method of measuring elbow angle during gait,such as videotape or electrogoniometry, the elbow-joint moment could be estimated from the strong linear relationshipbetween actual elbow angle and joint moment without theuse of a force platform.

Cl inical Impl icat ionsThe results of this study indicate that the 30-degree restingelbow-flexion angle often used for fitting axillary crutches hasa good biomechanical basis. If a therapist is going to deviatefrom this standard, it should be in the direction of loweringthe handle slightly to decrease elbow flexion.Crutches shouldbe made with fine enough adjustments at the handle to allowfor proper fitting because increments of 1 in can make signif-icant differences in elbow moment. An increase in elbow

moment may result in a significant increase in the amo unt offorce that the elbow extensors must exert. Muscle fatigue may,therefore, become one of the limiting factors in crutch walking.

CONCLUSIONSThe force exerted at the hands did not vary greatly withcrutch length, handle position, or speed of walking. Themoment exerted at the elbow joint also did not vary withcrutch length or speed, but handle position did have a significant effect on elbow-extension mom ent. Increasing the rest

ing flexionangle above 30 degrees by raising the crutch handle1 or 2 in resulted in increasing the elbow moment by almost100 percent, but decreasing the elbow-flexion angle by lowering the crutch handle did n ot significantly change the elbow-extension moment. We also found a high positive correlationbetween average elbow angle and elbow-extension moment.Elbow-extensor muscle strength and endurance is not theonly physical factor that affects the ability of someone to useaxillary crutches effectively. Other factors, such as shouldergirdle muscle strength, may be more critical. Further examination of the biom echanics of the shoulder joint during crutchgait may elucidate how forces are kept constant at the elbow

and hand during moderate changes of crutch height, handleheight, and gait speed.604 PHYSICAL THERAPY


5/5

REFERENCES1. Determin ing the length of cru tches. Modern Hospital 15:332,19202. Stuar t L: M eth o d o f m easu rem en t for walking appliances. Canadian Journalof Occupat ional Therapy 32:87-88 ,19653. Olms ted L: Cru tch wa lking. Am J Nurs 45:28- 35,19454. Lowman EW, Rusk HA: Self-help devices: Crutch prescript ions and me asurement. Postgrad Med 31:303-305,19625. Deaver GG: What every physician shou ld know about the teach ing ofcru tch walking. JAMA 142:470-472,19506. Childs T F : An analysis of the swing-through crutch gai t . Phys Ther 44:804-807, 19647. Murray M P: Gai t as a to ta l pa t te r n o f m o vem en t . A m J P h ys M ed 46: 229-

233, 19678. Ed i toria l: Cru tch ma stery. Am J Surg 7 3:404-420,19479. Blodgett M L: The ar t o f cru tch walking. Occupat ional Therapy and Rehab i li tat ion 25:27-32 ,194610 . Mc Beath A A, Bahrke M , Balke B: Efficiency of assisted am bulation determined by oxygen consumption measurement. J Bone Jo in t Surg [Am]56:994-1000,197411 . Peacock B: A myographic and photographic study of walking with cru tches.Physiotherapy 52:264,196812 . Wells RP: The kinematics and energy variations of swing-through crutchgai t . J Biomech 12:579-585,197813 . Shou p TE , Fletcher LS, Me rril l BR: Biomechanics o f crutch locomo tion. JBiomech 7:11-19,1974

14. Shoup T E : Design and test ing of a ch i ld 's c ru tch wi th conservat ive energystorage. Journal o f Mechan ical Design 102:672-676,198015. Seireg A H, Murray M P, Scho lz RC: Method for record ing the t ime, ma gnitude, and orientation of forces applied to wa lking sticks. Am J Phys M ed47:307-314,196816 . Murray M P, Seireg A H, Scho lz RC: A survey of the t ime , magni tude, andorientation of forces a pplied to walking sticks by disabled m e n . A m J P hysM ed 48: 1 -13, 196917 . Baxter M L, Allington RO, Koepke G H : Weight d istribution variables in theuse of cru tches and canes. Phys Ther 49:360-365,196918. Youm Y, F lat A E : Kinematics of the wrist. Clin Orthop 149:21-32,198019. Amis A A, Dow son D, Wr ight V: E lbow jo in t force pred ictions for somestrenuous isometric act ions. J Biomech 13:765-775,198020 . Provins KA, Sa l ter N: Maxim um torque e xer ted about the elbow jo in t . JAppl Physiol 7:393-398,195521 . Amis AA, Jughs S, M i l ler JH, et a l : E lbow jo in t forces in pat ien ts wi thrheumatoid arthritis. Rheumatology and Rehabilitation 18:230-234,197922 . Simon S R, Nuzzo RM , Koskinen M F: A com prehensive cl in ical system forfour dimensional motion a nalysis. Bulletin of the Hospital for Joint Diseases38(1):41-43,197723 . Ghosh AK, T ibarewala DN, Dasgupta SR: Metabo l ic cost of walking atd i fferent speeds wi th axi l lary cru tches. Ergonomics 23:571-577,198 0

Volum e 65/Num ber 5, May 1985 605

Documents

Elbow Moment and Forces at the Hands During