GAIT ANALYSIS A C O M PARISO N BETW EEN O B SER VA TIO N A L AN A LYSIS AND TE M P O R A L DISTANCE M EASUREM ENTS. ■ R iley M M C SP , M Sc Phys (Wits) Dept o f Physiotherapy, Johannesburg H ospital an d Univesity o f the W itw atersrand ■8 G oodm an M , PhD (W its) Dept o f Physiotherapy, Johannesburg H ospital and Univesity o f the W itw atersrand B Fritz V U M B B C h , FCP(SA), PhD (M ed) Dept o f N eurology, University o f W itw atersrand INTRODUCTION ff patients suffering from neurological deficits resulting from diseases such as stroke or P arkin son's disease are asked w hat is the m ost im portant aspect o f their rehabilitation, the m ajority of them will reply that independ ent w alking is their main goal. For the elderly the ability to w alk independently, or w ith m inimal assistance, to the toilet m ay m ean the difference betw een being able to live at hom e or having to go into a care centre. For young patients the ability to be independently am bulant allow s them to participate again in activities with their friends and fam ilies and often to return to w ork or studies'1. As gait retraining is such an im portant p art o f rehabilitation, accurate evaluation and m easurem ent o f gait is crucial for prov­ ing the effectiveness o f treatm ent. W hile m any com puter-based system s are now available for the collection o f data related to w alking patterns^, the m ajority of physiotherapy departm ents are unable to ju stify the cost o f such equipm ent and subjective evalu- ^ A B S T R A C T ^ The purpose of this study was to compare observational analysis of gait to six temporal distance measurements in order to rate the accuracy of the observational analysis. Ten hemiparetic and ten parkinsonian patients were asked to walk along a paper walkway with ink pads attached to their normal footwear. Measure­ ments of velocity, cadence, step length, stride length, base width and foot angle were taken. Ten normal subjects were also evaluated on the paper walkway to give normal values as a baseline for comparison with the hemiparetic and parkinsonian patient’s measurements. Observational analysis was recorded on a gait assessment form and a video recording was made of each patient. Comparison was made between the results recorded on the gait assessment form and the objective data. Observational analysis was found to be fairly reliable for the assessment of some gait parameters but as no accurate data are produced it cannot be used to give scientific proof of the effectiveness of treatment. Step length was the most difficult parameter to evaluate observationally in the hemiparetic patients, whereas cadence, foot angle and base width were the most difficult in the parkinsonian patients. As velocity was an easy value to record objectively it should be used in all gait assessments. V_______________________ ____________________ J ation is usually the m ain form o f gait evaluation. The question arises as to how accurately w e can ju d ge im provem en t or the degree o f im pairm ent in our p atient's gait? W ith this question in m ind, a study w as undertaken in which six tem poral distance m easurem ents w ere analysed objectively from ink footprints. O bservational analysis of the sam e six tem­ poral distance factors w as then com pared to the objective data to show w hether accuracy could be achieved observationally. METHOD S u b je c ts • Norm al Subjects T en norm al subjects w ere selected from the staff and students w orking a t the Johannesburg H ospital. Eight fem ale and two m ale subjects w ere selected with ages ranging from 20-74 years (m ean 45.8 years). The age range o f the norm al subjects w as m atch ed as closely as possible to that o f the hem iparetic patients. • H em iparatic subjects Ten hem iparetic subjects w ere selected from the w ards, clinics and physiotherapy d epartm ent of the Jo h an n esb u rg H ospital, these being the first ten patients w ho m et the inclu sion criteria. There w ere seven fem ale patients and three m ale patients. Ages ranged from 21-73 years (m ean 48.5 years). • Parkinsonian Subjects T en parkinsonian patients w ere selected fro m the w ards, clin­ ics and physiotherapy d epartm ent o f the Jo hannesb u rg H ospital, these being the first ten patients w ho m et the inclu sion criteria and w ho w ere w illing to participate in the study. T here w ere seven fem ale patients and three m ale patients. A ges ranged from 59-84 years £mean_69.3 years). All patients w ere taking[L-Dopa in the form o f ̂ i n e m e t j • Inclusion Criteria * To understand w h at w as required d uring testing. * The ability to w alk at least ten m etres three tim es using any type of w alking aid (except parallel bars) and requiring assistance from no m ore than one person. • Testing (Data Collection) * Each patient w as allow ed to read (or had read to him ) an explanation o f the project and w as asked if he understood w h at w as to take place and w hether he w as happy to take p art in the trial. * Ink pads w ere attached to the p atients' shoes, red on the right and blu e on the left, w ith a triangular pad at the toe and a square pad at the heel. Each p atien t w as then asked to w alk at his m ost com fortable speed along a p ap er w alk­ w ay placed over a m easured fourteen m etre length o f cor­ ridor leaving ink footprints from which m easurem ents w ere later taken. M easurem ents w ere taken over the central ten m etres, w ith the first two m etres being allow ed for accelera­ tion and the last two m etres for d eceleration. .Data from six temporal distance m easurem ents w ere collected. The m eas­ urem ents started w ith the first heel strike over the starting m arker and ended w ith the first heel strike over the finishing marker. The patient w ore his usual footw ear. * The p atient's gait w as evaluated by observational analysis recorded on a gait assessm ent form . O bservation al analysis SA J o u r n a l P h y s io th e r a p y , V o l 5 2 N o 2 M a y 1996 P a g e 2 7 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) o f the six tem poral d istance m easurem ents w as rated as b eing either less than norm al, norm al or m ore than norm al. The observational analysis w as considered to b e accurate w h en these values w ere seen to correspond with variations from norm al show n by the temporal distance m easure­ m ents. N orm al values w ere obtained from the li terature and from the ten norm al subjects evaluated (Table I). * A V ideo record ing w as taken as each patient w alked to­ w ards the cam era. This w as then k e p ts o thatfu tu re view ing could confirm the observational data recorded on the gait assessm ent form . * A com parison w as m ade of observational analysis evalu­ ation and tem poral d istance m easurem ents o f the hemi- p aretic and p arkinsonian subjects. DATA ANALYSIS V elocity - or speed o f w alking - w as m easured in m etres per second and w as calculated b y d ividing the ten m etre data collec­ tion section by the time taken in second s to w alk along it. C adence - steps per m inute - w as calculated by counting the num ber o f steps taken over the ten m etre data collection section, dividing by the tim e taken in seconds to covex-this-section-and then m ultiplying by 60 to bring this to pteps per minute. / Step len gth is the d istance from the initial contact o f one foot to the initial con tact o f the successive step o f the opposite foot. R ight step length and left step length w ere m easured. (Fig 1). Stride len gth is the d istance from initial contact o f one foot to the next initial contact o f the sam e foot. It therefore includes both stance and sw ing phases. (Fig 1). Base w idth at heel is the w idth betw een the heel m arkers on each foot. M easurem ents w ere taken from the centre point o f the heel squares to the ed ge of the w alkw ay and the sm aller distance w as subtracted from the larger. Base w idth at toe w as calculated in the sam e m anner using the apices o f the triangular pads as reference points for m easurem ents. (Fig 2). Foot angle - refers to the am ount o f toe-out or toe-in of each foot. (Fig 3). RESULTS Fig 4 show s that observational analysis w as m ost accurate in the hem iparetic patients, w ith seven or m ore patients being accu ­ rately assessed for velocity (7 patients), cadence (8 patients), stride length (9 patients), foot angle (9 patients) and b ase w id th (9 patients). O b s e rv a tio n a l a n a ly sis am o n g the p a rk in so n ian p atien ts show ed accuracy for velocity (8 patients), step length (9 patients), and stride length (9 patients). Tables I, II and III show the objective m easurem ents obtained for the norm al, hem iparetic and parkinsonian subjects and are referred to in the discussion. DISCUSSION____________________ V elocity - f^tormal~walking velocity h^s been described by R ob in ett and V on^Tan^I sH^ping: approxim ately! 89 m etres per m inu te (1.48 m etres per second) for m en and 74 m etres p e r minute Y l.23 m etres per second) for w o m en. Tftis w as confirm ed by m easurem enfs'fFom fine norm al subjects w hich ranged from 12.5 to 1.56 m etres per second at com fortable w alking speed (Table I). For this study a velocity o f over 1 m etre/sec w as considered norm al. T im in g b y stopw a tcli_ s h o w e cL th a t^ l the hem ipareticPy7 [patients w alked at velocities o f less th a n l m etre per second (TjEble II). T he velocity o f seven patients w as correctly evaluated obser- v ationally, w hilst three w ere thought to show normal velocity. >1 0 CO 10-1 o 9 -o CO R - TJ L. (11 U 7 - CO 3 f 6 - (0 > Q) 5 - 0) .O 4 -tft O 3 - .0 - m Q_ 1 - O u - l O Velocity Step length T e m p o ra l dis ta n c e m e a s u re m e n ts L I Hemiparetic patients I H Parkinsonian patients Figure 4: Com parison of observational analysis with _________ tem poral distance m easu rem en ts_________ B la d s y 2 8 M e i 1996 SA T y d s k rif F is io te r a p ie , D e e 1 5 2 N o 2 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) It is interesting to note that one you ng stroke patient, w hose gait w as rated as norm al by the p h y s io th e r a p is t, fe lt th a t sh e w a lk e d too slow ly to k e e ^ up w ith her friends. R obinette and V ond ran also found that goals set within rehabili ta tion departm ents often w ere not ade­ quate for a patient to return to his usual daily environm ent. This show s how im portant it is for the physiotherapist to plan together with the patient the goal o f treatm ent. This m ight be w alking from the bed room to the bathroom in an unspecified length of time, achieving su ffi­ cient velocity to cross the road safely at traffic lights, or having to negotiate stairs and consid­ erable distances in order to be on time for a lecture. The parkinsonian patients w ere m ostly in the upper age range, a factor w hich could in­ fluence velocity, how ever a significant d iffer­ ence in the velocities o f the norm al subjects of varying ages tested w as n ot found.fO nly twcT ,̂of the parkinsonian patients walked faster than i one m etre per second (T^ble [IT). The velocity o f eight patients was correctly evaluated obser­ vationally. Six o f the ten patients stated that th e ir w a l k i n g v a r ie d a t d if f e r e n t tim e s throughout the day according to w hen their m ed ication w as taken. This w ould present problem s if several sets o f m easurem ents were to be com pared over a time span as the effects of the m edication w ould have to be taken into account. This w as n ot a problem in this study as only one set o f m easurem ents w as recorded with the parkinsonian patients. As velocity is an easy m easurem ent to obtain objectively, and has been show n by Brandstater et «/4 to correlate w ith im proved m otor function in hem iplegic patients, it w ould be preferable to use objective velocity m easurem ents w ith all gait assessm ents. C ad ence - Carr and.Shepherd"’ relate that fa^ u lts w alking norm ally tak^ approxim ately 100 steps per m inu te". | For this study 90-100 steps per m inute was taken as being norm al. In the ten norm al su b ­ jects measured in our departm ent, cadence was found to be slightly higher than this at a com ­ fortable w alking speed, varying betw een 132_______________ a n d ll 4 s t e p s p e r m in u t e ( T a b le I)]C ad en ce w asfo u n d t o b e lo w e i^ considered to be norm al. Step length proved very difficult to ' than 100 steps per m inute in all except one of the hem iparetic j evaluate observationally in the hem iparetic patients. This was I patients (Table II) and norm al or higher than normal in all the surprising as hem iplegic g ait is so obviou sly asym m etrical in Iparkinsonian patient s i l a b l e HDTThis m easurem ent is also easily obtained objectively and as accuracy o f observational analysis w as poor with the parkinsonian patients it would be better to use objective m easurem ents. In norm al subjects cadence relates to velocity in that velocity is increased b y increasing eith er stride length or cadence and is usually produced by increasing both. This did not apply to the p arkinsonian p atients in this study. In the ten patients tested velocity w as relatively low , with a com paratively high cadence. The patient w ith the slow est velocity o f 0.21 m etres per second had the highest cadence o f 125 steps per m inute (Table III). Step L en gth - Step lengths of 0.5 m etres and above were appearance. H ow ever, although it is easy to see that tly? time spent w eight bearing on the affected limb is alm ost alw ays red uced, st§p~ lengtfTmay be decreased on either side and it w asfou rid th a tfiv e o f the patients took a longer step w ith the affected lim b w hile the other five did so with the unaffected limb. The difficulty in assessing step length by observation has also been described by G audet et al (1990) . The parkinsonian patients, w ith their m ore sym m etrical gait, w ere easier to evaluate observationally and could b e seen to be taking shorter steps than norm al. Strid e L en gth - T his m easurem ent w as accurately evaluated in b o th hem iparetic and parkinsonian patients, w ith nine ou t of Table II: Gait m easu rem ents of hemi Da retie Datients Patient No Velocity Cadence R Step length LStep length R Stride Length L Stride Length Bose Heel Base Tae Fool Angle ° R L 1 0.8 97 514 535 1 1 0 5 1 108 114 151 6 6 2 0.14 35 189 257 4 54 453 196 23 9 10 20 3 0.24 48 1 60 31 5 4 87 4 8 2 115 202 22 5 4 0.7 102 4 1 3 46 3 8 7 4 8 7 0 134 2 2 5 19 9 5 0.36 61 38 9 46 9 8 5 8 86 0 227 27 2 9 6 6 0.9 92 611 492 1 103 1 125 80 2 2 6 32 13 7 0.57 79 42 9 461 894 88 8 113 163 13 3 8 . 0 . 2 9 61 260 238. 501 49 8 164 217 3 13 9 0.62 75 431 565 983 98 7 23 8 3 4 2 6 24 10 0.55 80 43 5 48 2 908 9 1 0 139 161 3 6 Table III: Gait m easu rem ents of parkinsonian patients Patient No Velocity Cadence R Step Leangth LStep Length R Stride Length L Stride Length Base Heel Bose Toe Foot Angle ° R L 1 0.21 125 70 91 172 173 8 0 181 13 20 2 0.79 100 553 50 7 1 0 6 6 1 0 6 6 44 103 4 16 3 1.03 118 646 544 1 198 1 1 9 0 8 6 82 •3 1 4 0.74 121 38 0 341 725 721 54 114 9 7 5 1.18 106 743 69 9 1 2 1 1 1 4 4 2 82 158 6 12 6 0.81 118 402 44 3 8 5 6 84 5 51 65 4 02 7 0.93 96 642 4 9 5 1 1 3 8 1 137 136 2 27 8 15 8 0.71 99 471 49 6 9 7 0 9 6 6 88 107 10 2 9 0.47 100 252 26 8 527 5 2 4 57 132 17 18 10 0.74 93 462 4 3 5 90 3 897 6 40 9 -4 Table I: Gait m easurem ents of ten norm al subjects Subject No Velocity (m/sec) Cadence Step Lenth Stride Length Base Heel Base Toe Foot Angle ° R L 1 1.41 127 6 6 6 1 33 2 84 1 06 3 3 2 1.45 123 714 1 4 2 8 117 181 6 9 3 1.32 119 6 6 6 1 33 2 59 107 9 2 4 1.25 121 625 1 25 0 -6 61 13 7 5 1.25 127 588 1 176 0 -6 -2 2 6 1.29 132 58 8 1 176 -48 71 2 5 7 1.36 123 66 6 1 3 3 2 30 62 6 3 8 1.56 132 714 1 4 2 8 20 84 10 3 9 1.36 114 714 1 4 2 8 39 10 6 8 6 10 1.36 123 66 6 1 3 3 2 81 91 2 1 SA J o u r n a l P h y s io th e r a p y , V o l 5 2 N o 2 M a y 1996 P a g e 2 9 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) P eople w ho stick to conventional w isdom never really get anyw here in life. T h a t’s because conventional w is d o m o fte n s a y s it c a n ’t b e d o n e , s im p ly b e c a u s e it h a s n ’t b e e n d o n e before. But a t S ta n d a rd B ank, w e believe in c h a lle n g in g old b e lie fs . N o t to be different. But b e c ause it often pays to look at things a little unconventionally. T h a t ’s w h y w e ’v e d e s ig n e d Financial S ervices for Professionals. A ra nge of I t ’s m y b u s in e s s to b e p r o fe s s io n a l. T h a t ’s w h y I d o n ’t b e lie v e in m y th s . fin a n c ia l s e r v ic e s c a t e r in g fo r y o u r p e r s o n a l a n d y o u r p r a c t ic e n e e d s . O ur A chieverP lan, P ro-E lite, M e d -E lite and PrestigePlan p a ckages c a te r to all your personal needs. And products like o u r P ro fe s s io n a l R e v o lv in g C r e d it, P r o fe s s io n a l P r a c t ic e B o n d , E q u ity F i n a n c i a l S e r v i c e s f o r P r o f e s s i o n a l s Purchasing Loan and Stannic Finance are there to m e e t all y o u r p ra c tic e needs. W e will also tailor a com prehensive assurance p a c k a g e for all y o u r p e rs o n a l n e e d s . To find out m ore, call us to ll-fre e on 080- 01-21000 - or visit your nearest S ta n d a rd B a n k b ra n c h m a n a g e r. Y o u ’ ll s o o n s e e w h y it p a y s to lo o k a t t h in g s f r o m a d i f f e r e n t perspective. A s s ure a s e g g s are e g g s . W ith u s y o u c a n g o s o m u c h fu rth er. R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) ten patients being correctly evaluated in each case. Stride lengths o f one m etre and over w ere considered as norm al. Being a com bination o f tw o successive step lengths, stride length is directly affected b y the step length m easurem ents but does not show an asym m etry o f gait as does step length. This m ade it a m uch easier gait param eter to ju d g e observationally and nine o f the patients in each group w ere correctly assessed as having either norm al or decreased stride lengths. As the asym m e­ try o f gait is not show n this is n ot such a valuable m easurem ent as step length in the evaluation o f hem iplegic gait. Foo t A n g le - V ariations betw een right and left foot angles of up to five degrees w ere considered as norm al. Foot angle w as accurately observed in nine o f the ten hem iparetic patients but w as only accurately observed in four o f the ten parkinsonian patients. Foot angle w as increased on the affected side in all except one o f the hem iparetic patients and foot angle w as observed accu­ rately in all patients show ing this increase. There w as great vari­ ability in the foot angle o f the parkinsonian patients, w ith three p atients show ing toeing-in o f the foot. This variable gait pattern m ade foot angle m u ch m ore d ifficult to evaluate observationally. B ase W id th - N o descriptions o f norm al b ase w id th w ere found in the literature althou gh this m easurem ent has been described as being useful by G au det et o f ’. There w as a large d iscrepancy betw een the base w idth o f the hem iparetic patients and the parkinsonian patients, w ith m uch narrow er b ase w idths b eing observed in the latter. Base widths of betw een 100 m m and 145 m m at the heel and betw een 100 m m and 170 m m at the toe w ere considered w ithin norm al limits. Base w id th w as correctly assessed observationally in nine of the ten hem iparetic patients and w as easy to see due to the slow ness o f gait. T he m ore rapid gait pattern o f the parkinsonian patients m ade accurate observation m uch m ore difficult and a decreased base w id th w as n ot alw ays noticeable, although nine o f the patients w ere found to have a decreased b ase w id th w hen objective m easurem ents w ere analysed. V id eo R eco rd in g - A vid eo recording w as m ade o f each patient w alking tow ards the cam era. This w as found to be very useful as a record fo r observational analysis. CONCLUSION Tem poral distance param eters o f velocity and stride length can b e judged observationally in b o th hem iparetic and parkinsonian patients w ith reasonable accuracy. H ow ever, as observational analysis can only provide estim ates o f w h eth er the param eters are norm al or show an increase or d ecrease from norm al, this type of analysis is not sufficient for research purposes. M easurem ents o f step length, foot angle and b ase w idth may b e estim ated reliably in som e patients, b u t for accuracy objective m easurem ents should be obtained. This can be done by collecting ink footprints on a length o f paper w alkw ay and then analysing the m easurem ents. A lthough this is a rather time consum ing procedure, the data produced is reliable enough for research purposes and does n ot require expensive equipm ent. This is im portant in Sou th Africa today as research into effectiveness of treatm ent is essential b u t lim ited funds do not allow the purchase o f expensive equipm ent. T his m ethod o f data collection has been described in two articles b y H olden et af'^ . As velocity is the easiest m easurem ent to obtain objectively and has been show n by Brandsta ter efflZ4 and H olden et al to correlate w ith function in hem iparetic patients, this m easurem ent should b e included in all gait assessm ents. REFERENCES 1. Riley M C. The evaluation of tem poral distance gait m easurem ents in patients with pyram idal and extrapyram idal lesions. M Sc Research report 1995, U niversity of the W itw atersrand. 2. Turnbull G I, Wall J C. The developm ent of a system for the clinical assessm ent of gait following a stroke. P hysiother 1985:71;294-300. 3. Robinett C S, Vondran M A. Functional ambulation velocity and dis­ tance requirements in rural and urban com m unities: A clinical report. Phys T her 1988:68;1371-1373. 4. Brandsta ter M E, De Bruin H , Gow land C et al. H em iplegic gait: analysis of temporal variables. A rch P hys M ed R ehabil 1983:64;583-587. 5. C arr G, Shepherd R B. A m otor relearning p ro g ram m e for stroke. 2nd Ed. 1987, London, W illiam Heineman. 6. G audet G, G oodm an R, Landry M et al. M easurem ent of step length and step w idth: A com parison of videotape and d irect measurem ents. P hysiother Canada 1990:40; 12-15. 7. Holden M K, Gill K M, M agliozzi M R et al. Clinical gait assessm ent in the neurologically impaired: reliability and meaningfulness: P hys Ther 1984:64;35-40. 8. H olden M K, Gill K M, M agliozzi M R. Gait assessm ent for neurologi­ cally impaired patients: Standards for outcom e assessm ent: P hys Ther 1986:55;1530-1539. IBITAH The International B obath Instru ctors/ Tu tors A ssociation for A d ult Hem iplegia (IBITAH) w as founded in 1984 and is an inter­ national organisation o f therapists w h o have developed a special interest in the assessm ent and treatm ent o f ad u lt persons w ho have suffered a stroke or sim ilar neurological im pairm ent. IBI­ TA H unites physiotherapists and occu pational therapists w orld ­ w ide and at p resent represents m ore than 170 therapists in 20 countries. The objectives o f IBITA H are: • To im prove and spread skills in the assessm ent and treatm ent o f persons w ith hem iplegia and other allied neurological con­ ditions • To prom ote and further know led ge o f the B obath concept related to adult hem iplegia and other allied neurological con­ ditions • To increase the availability o f IB IT A H recognised courses w orldw ide • To m aintain and im prove the standards o f IB ITA H recognised courses • T o im p ro v e and s ta n d a rd is e the tra in in g o f in s tru cto rs throughout the world • To organise regular m eetings and edu cational p ro gram s and to facilitate the exchange o f ideas b etw een m em bers. The Bobath concept is a problem solving approach to the assessm ent and treatm ent o f individuals w ith d istu rbances of tone, m ovem ent and function due to a lesion o f the central nerv­ ous system . It is nam ed for its originators, D rs Berta and Karel Bobath. IBITA H is responsible for setting the regulations, stru cture and content o f basic and advanced courses in the assessm ent and treatm ent o f ad ult hem iplegia and related cond itions, as w ell as stipulating the process for training and q ualifying instructors. There are three levels o f instructor - instructor, ad vanced course instructor and senior instructor. In Sou th A frica there are at present one senior instructor, one instructor and two instructors in training. IBITA H recognised courses offered in Sou th Africa are also recognised for certification by the Sou th A frican N eurode- velopm ental Therapy Association. ■ (Information taken from IBITAH brochure) SA J o u r n a l P h y s io th e r a p y , V o l 5 2 N o 2 M a y 1996 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. )