Teach-in '98 on Step & Stride Length: What people said...

Maybe I am taking my life in my hands after my last contribution wrt step length, but here goes.. I personally think that we are trying to read too much meaning into these definitions. They are secondary variables - descriptors, and therefore to some extent arbitrary. Is step length, by whichever definition, the absolute distance or the resolved distance in the direction of progression? If the former how is the direction of the step defined and if the latter, what is the definition used for direction of progression? As they are descriptors, they need to reflect the attribute(s) we are trying to observe, which are established by our own clinical /research priorities. I think it would be wrong to try and establish any of them as definitive, only that the definition should be clearly stated for the particular protocol. Also going back to the point about variablility, a particular left step and right step can only summate to one value, which ever order they occur in, but consequetive strides will by definition NOT use the same left and right step combination.

So I do not think there are correct definitions, only agreed ones. I think that McNeil's definition is elegant, but I haven't worked out yet if it offers a better definition from a clinical perspective, putting aside the fact that you would need a gait lab to measure it!

As to the meaning of step length, it is fair to say that by whatever definition it is an index of asymmetry and there is nothing else to add to that. It has no additional significance in its own right!

Jeremy Linskell Manager, Gait Analysis Laboratory Co-Ordinator, Electronic Assistive Tehcnology Service Dundee Limb Fitting Centre Dundee, DD5 1AG, Scotland tel +1382-730104, fax +1382-480194 email: j.r.linskell@dth.scot.nhs.uk web: http://www.dundee.ac.uk/orthopaedics/dlfc/gait.htm


Hello all,

Perhaps I'm missing something, but I do not see a great contradiction between Whittle's and Winter's definitions of these distance parameters. Whittle mentions the possibility of different step lengths, but states that stride length must be the same for the left and right side. I believe Winter says the same thing: "stride length... will be equal for left and right limbs if the person is walking in a straight line, even in the presence of marked asymmetry."

Similar definitions are available: "Stride length... can be measured as the length between the heels from one heel strike to the next on the same side. Two step lengths (left plus right) make one stride length. With normal subjects, the two step lengths will be approximately equal, but with certain patients... there will be an asymmetry between the left and right sides." (Vaughn, Davis, O'Connor, Dynamics of Human Gait, 1992) "One complete cycle, from left foot takeoff to left foot takeoff, is defined as a stride. A step is defined as the part of the cycle from the takeoff (or footstrike) of one foot to the takeoff (or footstrike) of the other foot." (Enoka, Neuromechanical Basis of Kinesiology, 2nd ed. 1994)

I believe left and right step length can differ according to the above definitions. I find it easiest to understand the difference between the two by thinking of each side as a repeatable pattern, like a sine wave. For a normal subject, these two waves (left and right) are essentially completely out-of-phase. As soon as the "phase lag" decreases, step length is no longer equal. In Whittle's example, the "waves" are now have no phase difference, and the difference in step length is maximized (one step length is zero). However, stride length cannot be different when comparing right to left because, as you say, "the right and left sides remain joined together."

What I find interesting is a consideration of the relationship between step length discrepancies and timing parameters, such as stance time. Perry states that, "Step refers to the timing between the two limbs" (Gait Analysis, 1992). It would seem that one could vary step length discrepancies and stance time discrepancies independently. However, I would guess that the same gait adaptation (pathological or otherwise) that causes a step length discrepancy also causes a stance time discrepancy. (This is a novice opinion -- has research been conducted in this area?)

Thanks for your time and comments, Mark

Mark D. Geil, Ph.D. Center for Human Movement Studies Department of Health and Performance Sciences Georgia Tech Atlanta, Georgia 30332-0110 Phone: 404-894-9993 Fax: 404-894-7593 email: mark.geil@oip.gatech.edu


Hi Chris

A couple of comments. I think that step lengths can certainly be different lengths. What does not seem to vary is the stride length. It seems that the left stride length must always equal the right stride length unless the person is walking in a curve or circle.

Jan < BRUCKNER@neu.edu >


Hello all:

As a relatively new arrival into the CGA list, I thought I would throw in my two cents into this step vs. stride length discussion:

Question #1. Can the right and left sides of the body really travel different distances and still remain joined together?

Since patients with asymmetrical step length patterns are quite common but, as others have noted, stride length must be the same for the left and right sides, this is an interesting question. It all comes down to the definitions for step and stride length which one chooses to adopt.

Question #2. What are the correct definitions of step & stride length?

It seems that the definitions offered by Whittle and Winter are saying the same thing. Also the definitions (courtesy of Mark Geil) by Vaughn, Davis, O'Connor, and Enoka seem to agree with Whittle and Winter. What is correct, of course, is a matter of opinion. Perhaps the definitions of step and stride length offered by R. McNeil Alexander in regards to trunk position are better definitions for use on those animals who do not only use two limbs by which to ambulate.

Question #3. What is the significance of different step lengths in a gait report?

On a somewhat of an aside to this question, one of the more interesting observations I have made in the step lengths of my patients is that lateral forefoot wedging (forefoot valgus wedging) seems to increase step length in many individuals. This effect is more pronounced when combined with the addition of medial wedging of the rearfoot (rearfoot varus wedging). This is a clinical observation that I thought would be of interest to the researchers on this list. However, I don't know if this effect has been tested in a formal study.

BTW, special thanks to Chris for his excellent work here in regards to gait analysis.

Sincerely,

Kevin <Kevkirby@aol.com>

Kevin A. Kirby, D.P.M. Assistant Clinical Professor of Biomechanics California College of Podiatric Medicine

Private Practice: 2626 N Street Sacramento, CA 95816 Voice: (916) 456-4768 Fax: (916) 451-6014


Dear all! I think that the confusion about step lenght is dur to the rather fuzzy definition. As Chris points out, it is necessary that, in order to avoid waling is circles, it is necessary that the left foot moves the same distance as the right foot. However, the standard definition of step lengt (as defined by Whittle, Winter and others) has nothing to do with the distance that the corresponding foot is moving. Instead it only describes where one foot is placed in relation to the other.

The figure below attempts to illustrate this point. The left and right wtep lengths are different, although the left foot and right foot are moving the same distance. The stride length is the sum of the left and right step lengts!

___ ___ ___ |___> ___ |___> Left foot |___> |___> Right foot | | | |<--------------->|<---->| | Left step lenght|Right | | step | | lenght| | | |<---------------------->| | Stride lenght |

Håkan Lanshammar Systems and Control Group, Uppsala University P.O. Box 27, S-751 03 Uppsala, SWEDEN E-mail: hl@SysCon.uu.se, Tel: +46-18-471 30 33, Fax: +46-18-50 36 11 http://www.syscon.uu.se/Personnel/hl/hl.html


Dear Hakan,

If only you'd continued your diagram for one more cycle, you'd see that you would have to make the same trick as me (alternating short and long steps):

___ ___ ___ |___> ___ ___ |___> Left foot |___> |___> |___> Right ___ | | | |___> Need to put it here |<--------------->|<---->| ^ but how long this is! | Left step lenght|Right | | Next right step should be here! | step | | lenght| | | |<---------------------->| | Stride lenght |

Chris -- Dr. Chris Kirtley (Kwok Kei Chi) MD PhD Assistant Professor Department of Rehabilitation Sciences The Hong Kong Polytechnic University


Hello Chris!

I am sorry, but I do not see the problem. I have extended my drawing, and I have not yet started to walk in circles!? - Håkan

__ ___ ___ |___> ___ |___> ___ Left foot |___> |___> |___> Right foot | | | |<--------------->|<---->| | Left step lenght|Right | | step | | lenght|>

Håkan Lanshammar


But isn't that normal limping? Or do I mix stride with step?

Theo C R van der Meer Managing Director

MUSGRAVE SYSTEMS Ltd & Preston Communications Ltd

Redwither Tower, Redwither Business Park Wrexham LL13 9XT United Kingdom Tel: +44 (0)1978 66 44 82 Fax: +44 (0)1978 66 44 83 E-Mail: VDMEER@CIX.CO.UK MUSLABS@AOL.COM MUSRESEARCH@ENTERPRISE.NET


Chris -

Think of a 'step to' gait as opposed to a 'step through' gait - the step length on the side that is 'stepping to' is approximately zero, but one can still move in a straight line forward.

Or just walk holding one knee locked. You will naturally end up with unequal step lengths, but can still walk in a straight line.

As far as unequal stride lengths are concerned, I believe this is only possible (for straight line walking) if one compensates in the following stride - something often seen in the clinic (i.e., the repeating gait 'cycle' consists not one one left and one right step, but of two lefts and two rights.) I hve seen this often in BK amputees, as well as in older individuals. I have no suggestions for nomenclature for a 'two stride cycle'.

Marcus P. Besser, PhD Assistant Director, Human Performance Laboratory Assistant Professor, Department of Physical Therapy Thomas Jefferson University Philadelphia, Pennsylvania, USA 19107-5233 Phone:215-503-1645 Fax:215-503-3499 email: marcus.besser@mail.tju.edu


Dr. Kirtley et al.

I have been following your discussion on stride/step characteristics and am somewhat confused. This does not really seem to be a tremendously difficult issue.

If we assume that the definitions are according to Winter, Whittle, and others:

1) a STEP depends on the relative positions of the two feet from initial contact of one foot to initial contact of the OTHER foot. It represents the net gain in forward position of the body. This is how it relates to Alexander's definition.

2) a STRIDE depends on the relative position of one foot from initial contact of that foot to the second contact of that same foot and does not consider what the other foot is doing. In fact, there may not even be another foot if the person is hopping or using crutches.

An example of this that has been mentioned during this discussion is a step-to type of gait pattern, a common occurrence in hemiplegic type gait.

If we define the x axis to be in the general direction that the person is walking. (avoiding the strict definition of "direction of progression" for the moment)

then:

In summary, during this step-to gait pattern, the right foot has a step length of 0.5 meters and the left has a step length of 0.0 meters. But both stride lengths are 0.5 meters (0 + .5 = .5 and .5 + 0 = .5).

Using these definitions of step and stride length, the analysis is very clear. One misconception is that the step length is measured from the position of the foot at toe off to the position of the same foot at initial contact. If THIS definition were adopted, stride length (contact to contact of the same foot ) and step length (foot off to contact of the same foot) would refer to approximately the same quantity and a stride length would NOT equal the sum of two steps because the steps would overlap in space. We would be counting the same distance twice as contributing to forward progression of the center of gravity of the body.

George E. Gorton, III, B.S. Director, Gait Analysis Laboratory 516 Carew Street Shriners Hospitals for Children Springfield, Massachusetts 01104

Tel: (413)735-1269 Fax: (413)787-2063 email: ggorton@shrinerspfld.org


Dear all,

OK, OK... I think I got it! I now see that it is just a problem of definition. But what a strange definition this is. On Hakan's diagram above, both legs are advancing an equal distance each stride, yet the right step length is recorded as being much shorter!

So this brings me back to the 3rd of my original questions: what significance have step length measurements in the gait report. Clearly they do not, as many people apparently believe, tell us anything about pathology (e.g. weak push-off) on the "short" side (it's of interest, incidentally, that the last Case of the Week had a "short" step on his normal side). They simply record an assymmetrical curiosity of the gait (that one leg advances further wrt the trunk). What does this mean clinically? What makes a patient walk like this?

Chris


In answer to the question of what makes a patient take a short step with his sound, or normal, side, it is possible that the person feels insecure or has poor control on his affected side. If a person is off balance or unable to control his forward and rotational momentum when standing on the weaker or affected side, they will take a short step witht he sound limb so as to minimize the single leg stance time on the weaker leg. The cause of the short step may be anything from decreased proprioception, ankle, knee or hip weakness, or even decreased pelvic rotation on the affected limb. This is often seen in BK amputees, according to Winter, et al, and the results of our thesis on BK amputee gait. It may also be a result of learned patterns of walking due to an improperly fitted prosthesis during training, or just a distrust of the prosthesis. Just another thought, if a person has decreased hip flexion or knee extension during a step, the step length would be shorter on the affected side, as is seen in people with hemiplegic CP.Just some ideas to play with, let me know what you think.

Kris Dart, MSPT Grand Rapids, MI


Regarding Chris' tendency to walk in circles...

A) I think that the goal of the task is overlooked. For all ambulators, the goal is to get from point A to point B. Hemiplegic patients do this in the best way they know how...maximizing stablity, but miminimzing mobility (hence the altered step length). Chris is not trying to solve the problem in the same way a hemiplegic patient would.

B) I think it would also be useful to look at how the step is occuring. Since preswing is mostly passive, and set up by an adequate step length on the contralateral limb, step length may vary due to variances in momentum of the gait. Clinically, I see preswing as a very active vs. passive process in many hemiplegics.

I am sure other factors play into this, but I prefer to look at gait deviations from a task-oriented view.

Melanie Weller MPT


Thanks very much to everyone contributing to this discussion. I think you'll now agree that there's much more to a short step length than meets the eye. I've just reviewed last year's Cases of the Week, to see if there's any pattern. It seems to me that it's possible for the short step to occur on either the affected or unaffected side. Here's my classification:

In some caes, where it's not clear which is the worst side. Since we have a lot of diplegics, I am going to put my tongue firmly in my cheek and say that one side is arbitrarily "worse"(?) than the other. I will assume that the patient stays longer on the unaffected side, i.e. stance duration is longer on that side. Hopefully someone else out there might help by providing data from their own cases.

Case Pathology Step_R Step_L StDur_R StDur_L Type
6-2-97 FemNerveL 63 57 60 58 b
26-4-97 DiplegiaL 53 43 62 56.5 b
26-6-97 Diplegia?L 44 40 66 61 ?c
4-8-97 Bone-graftR 47 46 67 70 slight c
25-9-97 Diplegia?L 43 46 65.5 61 ?
21-11-97 R Hemi 43 36 61 70 c

Thus, assuming my method of classification is correct, it seems that there are both type 'b's and type 'c's. In other words, the bad side is sometimes left trailing further, and sometimes advanced further wrt the sound side. Why could this be?

I am going to suggest that in type 'b' there is a problem with the "reach" mechanism of the bad leg, whereby the leg is advanced (e.g. hip or knee range of motion limited, as suggested by Kris Dart); and in type 'c' there is inadequate propulsion (push/pull-off) of the bad leg. I am reasoning that it would be sensible not to leave a leg with poor push-off too far behind (as it would be difficult to retrieve it!), and difficult to adequately advance a leg with a poor reach (ipso facto).

Of course, very often, both poor propulsion and poor reach will occur simultaneoulsy, as in the last Case of the Week. What would then be expected to happen would depend on which mechanism is most compromised. In this patient his push-off was poorest, so he used a shorter step on the sound side. Conversely, patient 6-2-97 with a left femoral nerve lesion has a shorter step on the affected side, presumably because she has problems with reach.

Does that make any sense at all, or have I totally lost touch with reality...???

There is a new Case of the Week awaiting to save me if the latter proves true!

Chris -- Dr. Chris Kirtley (Kwok Kei Chi) MD PhD Assistant Professor Department of Rehabilitation Sciences The Hong Kong Polytechnic University


The problem is not one of confusing step and stride length, but rather what "step length" actually means.

Since the two legs are joined together, both legs must travel the same distance if the person is to walk in a straight line. Therefore, the length of each step must always be equal. The term "step length", as it is defined (from contact of the contralateral foot to contact of the foot in question), actually means the offset position of one foot relative to the other (stylised Lanshammer diagram follows!):

- _ - _ - symmetrical steps

-_ -_ -_ still equal step lengths, but R step length now "short"

_- _- _- still equal step lengths, but L step length now "short"

Now, the interesting thing is that the "short" step is not invariably on the affected or worst side. To my way of thinking that tells us something about the pathology in question.

Does everyone understand what I'm getting at? I have the feeling I am on my own here!

Chris -- Dr. Chris Kirtley (Kwok Kei Chi) MD PhD Assistant Professor Department of Rehabilitation Sciences The Hong Kong Polytechnic University


Affected side / non affected side:

It is my impression that the step length is influenced by

This would mean that if the right leg's push off intensity is affected it would reduce right->left step length

or

if the right leg's load acceptance capability is affected it would reduce left->right step length because of some protective pattern.

Am I entirely wrong?

Regards Christian Calame Product Manager Biomechanics Kistler Instrumente AG Winterthur, P.O.Box 304, CH-8408 Winterthur, Switzerland Tel: +41 52 224 11 11, Fax: +41 52 224 14 14 E-Mail: cl@kistler.ch, WWW:http://www.kistler.ch/biomech


Dear Chris:

Have you heard of a term 'affected leg lead' mentioned by Brunstrom to describe a pathological gait pattern in hemiplegic patient? This gait is quite common when a hemiplegic patient reached Brunstrum stage III - IV, whithout enough strength developed in affected leg for proper walking. The patients might not have enough confidence in their affected legs, therefore they not only put more force on their normal legs but also shorten the stance duration of the affected legs. It will make a figure that they can't have enough time to swing the normal leg forward so that the step length is shorter in the normal leg than in the affected leg. The reason of shorter step length in normal leg is of course not limited to 'poor push-off'. What do you think about it?

Wen-ling Chen, PT, MS. Lecturer, Physiotherapy Department of National Cheng Kung University D.Phil. student Oxford Orthopaedic Engineering Centre University of Oxford email: wen-ling.chen@sable.ox.ac.uk


Dear Chris

Thanks for your explanation. What is apparent to me now is the rate of progression is altered by asymmetry of step length, even though each step is a similar distance.

Now, the interesting thing is that the "short" step is not invariably on the affected or worst side. To my way of thinking that tells us something about the pathology in question.

The short step is really an indication that the other side is not contributing its fair share to the energy of progression. This may be because the long side is being used for postural bracing of the opposite side. Or may itself be affected, depending on the location on the problem.

Regards

Craig Nevin Anatomical Engineer <CNEVIN@anat.uct.ac.za>


Hi there: we think we finally understand what you are talking about. However, since your definition of step length is different from the conventional, you should go ahead and call your measure something different, and not say step length but mean it your own way. you have confused millions of us the last several weeks ...

:)

regards,

Raymond Chong, Ph.D.


A few comments on the issue of step/stride length follow. I hope that I am not being redundant to earlier messages.

Since the two legs are joined together, both legs must travel the same distance if th eperson is to walk in a straight line. Therefore, the length of each step must always be equal.

Since a "step" is defined as the offset position BETWEEN limbs, it is not the step, but the stride that must be equal to walk in a straight line.

The term "step length", as it is defined (from contact of the contralateral foot to contact of the foot in question), actually means the offset position of one foot relative to the other (stylised Lanshammer diagram follows!):

- _ - _ - symmetrical steps

Yes!

-_ -_ -_ still equal step lengths, but R step length now "short"

The STRIDE lengths are equal, but the R step is short

_- _- _- still equal step lengths, but L step length now "short"

Again, the STRIDE lengths are equal, but the L step is short

Now, the interesting thing is that the "short" step is not invariably on the affected or worst side. To my way of thinking that tells us something about the pathology in question.

There are many mechanisms that can affect the ability of a leg to achieve a "normal" step length. Perry (Gait Analysis, 1992) describes the role of the ankle "rockers" in progression. In a hemiplegic gait, for example, a typical subject may have a plantar flexion contracture on the affected side on clinical examination. If the contracture is flexible, adequate dorsiflexion may still be achieved in stance and the step length of the unaffected side will not be limited. If it is not a flexible contracture, however, at least two options are available. If the patient remains in equinus without a midfoot break, the progression of the tibia over the foot in stance will be restricted and this will shorten the step length on the unaffected limb. If a midfoot break occurs, or hyperextension of the knee joint occurs as the result of an external extensor moment from the equinus in stance, then adequate progression may occur to maintain a near normal step length.

In the same type of subject, we may find a knee flexion contracture. This would limit step length by limiting the forward reach of the swing limb.

Working back from this, if there is a hemiplegic gait with a short step length on the good side, we should look to mechanisms affecting forward progression and weight transfer on on the affected limb (eg - limited hip extension or limited dorsiflexion in stance). (Although there may still be compensatory tightness on the unaffected side.) If the step length is short on the affected side, then we should look to mechanisms of reach (eg - incomplete extension of the knee or inadequate hip flexion).

Part of the difficulty here is in reliably identifying "affected" and "unaffected". Even in a hemiplegic population, the designation may not be clear cut and there may be compensations that cause the "unaffected" limb not to mimic normal joint motions. If we move to a CP diplegic population with this assessment, everything falls apart because both limbs are affected and we can not separate limitations of reach from limitations of forward progression and weight transfer.

What we are left with is that step lengths are indicators of an ability to compensate for limitations of propulsion and advancement. Inadequate strength may limit a person with a polio residual from flexing the hip adequately, but with a caliper type gait, they may still achieve a near normal step length bilaterally using pelvic and trunk rotation. The identification of this gait does not come from looking at the step lengths themselves, but by understanding that the lack of hip flexion and knee extension require a transverse plane compensation to maintain adequate step length for progression.

In a person who has asymmetric step lengths, we know that there is a kinematic deviation that exists that is not compensated for. We can not tell however, from this information alone, what the deviation is, or what side to look for it. Therefore, at least in my own mind, asymmetric, or shortened step lengths are indicators of the presence of a problem, but not diagnostic in their own right.

George E. Gorton, III, B.S. Director, Gait Analysis Laboratory 516 Carew Street Shriners Hospitals for Children Springfield, Massachusetts 01104

Tel: (413)735-1269 Fax: (413)787-2063 email: ggorton@shrinerspfld.org


My field of interest is not in biomechanics but in mantracking. I see your point and offer this to the discussion.

Step length has little to do with navigation (making a turn) but one of maintaining balance. How a subject adjust one's step to counter a pathology is individualized and terrain driven. The adjustment seems to be merely a tactic to avoid falling or slipping.

Del Morris <dlmorris@concentric.net>


Many thanks for all your comments on the step length issue. I finally feel vindicated in my stubborness!

George E. Gorton of Shriners in Boston said:

Working back from this, if there is a hemiplegic gait with a short step length on the good side, we should look to mechanisms affecting forward progression and weight transfer on on the affected limb (eg - limited hip extension or limited dorsiflexion in stance). (Although there may still be compensatory tightness on the unaffected side.) If the step length is short on the affected side, then we should look to mechanisms of reach (eg - incomplete extension of the knee or inadequate hip flexion).

This is a similar idea to the one I proposed - the only difference being that I thought the sound side short step indicated a problem with push-off on the affected limb, reasoning that the patient would be reluctant to trail it too far behind as they would have difficulty "retrieving" it if push-off was weak. I agree that it could also indicate a problem with weight acceptance/balance on that side too. Del Morris' comment about it being a tactic to avoid falling is relevant here. Another obvious cause would be pain on that side (antalgic).

Now, Raymond Chong said that I should:

"go ahead and call your measure something different, and not say step length"

I'm tempted to do this, as I think a word like "offset" is more appropriate, although "step length" is so old and so frequently used that it's going to pretty hard to avoid. As I see it, the problem with it is that it makes us concentrate on the distance between the feet in double support (the definition of step length) and we then forget about the distance the foot travels in swing. I hope I have convinced everyone now that the sum of these two (which I had mistakenly understood as being "step length") is always equal on both sides. Note that this has nothing to do with stride length (which I still find people mentioning) since we are only talking about one leg.

So, I'm tempted to offer the following:

Step Offset = Right step length - Left step length

to make it completely clear. A positive Step Offset would indicate a short left step (left leg trails), i.e. problem with right push-off/weight acceptance or left reach, and negative Step Offset the opposite.

Does that help, or just confuse more?

Chris -- Dr. Chris Kirtley (Kwok Kei Chi) MD PhD Assistant Professor Department of Rehabilitation Sciences The Hong Kong Polytechnic University


The reason for shortened swing on the unaffected limb of a hemiplegic is the rapid transfer of weight to the "good leg". The affected side is less stable in single limb support and shortening the swing of the unaffected side reduces the time spent in single limb support on the affected limb. Unfortunately, this also increases the proportion of time the "good leg" spends in support and frequently leads to overuse phenomena.

Please contact me with questions.

Dr. Nicholas Sol The Walking Clinic, PC Colorado Springs, CO <nksl@market1.com>


Very interesting discussion, although somewhat collateral. Having followed this thread silently for some time I must now put in my 2 cents worth. For the past two years I have been studying stride length in children with cerebral palsy. There is a pervasive assumption that by performing hamstring lengthenings to correct crouch gait, stride length is also improved: it is assumed that the hamstring restrict swing and therefore stride length. However, stride length improvements following the successful correction of crouch gait are difficult to predict with any accuracy. When the hamstrings are lengthened to correct crouch, some individuals improve their stride length greatly and others remain the same, or improve at a rate consistent with increased leg length due to growth, while other get slightly shorter strides.

This data and the following data was presented at the 1997 ASME Summer Bioengineering Conference:

Stride length (cm) = -0.130 + 1.165(leg length) + 0.637(hip arc) - 2.082(first double support period) + 1.267(contralateral ankle push-off energy) + 0.647(ankle push-off energy)

p <.0454, R2 = .739, n = 54 sides chosen randomly

Where leg length in in cm, hip arc is in degrees, first double support period is in % of gait cycle, ankle push-off energy is in joules (area under to A2 power generation burst in late stance). mean age 11.8 ± 4.2 years, range 20.5 to 4.7 years Diagnosis: cerebral palsy

When considering normal children only ankle push-off energy and leg length were significant predictors of stride length. (p < .007; R2 = .832, n = 15) It seems that the assumption that the tight hamstrings are shortening stride may not be entirely correct for most individuals. We seem to spend a fair amount of our time considering the affected limb in looking for reasons for stride length limitations, and this may hold the primary reason for some individuals. However, my data suggests that ankle push-off energy on the contralateral side and first double support time are very important to side length on the ipsilateral side. We consider the amount the swing limb moves forward to be the primary indicator of stride length but we don't often consider the movement of the trunk forward. It is this inverted pendulum period, which advances the trunk, that makes the pendulum (swing) period possible. The ankle push-off energy is a prime accelerator of the center of mass in the forward direction (see Kepple, et al, (1997) Relative contributions of the lower extremity joint moments to forward progression and support during gait, Gait and Posture 6, pp1-8), and appears from my data to have a relatively minor, although perhaps still important, effect on the swing limb. The length of first double support time is also important. If the contralateral ankle provides the acceleration for the trunk, the ipsilateral side must provide a stable base in order that the trunk move forward. Individuals who take a long time to get into single limb stance generally have shorter strides. (Notice that this is negative in the regression equation, meaning that decreased first double support leads to longer strides) Leg length, contralateral ankle push-off energy and first double support time account for something like 70% of the variance in stride length. It seems to reason that clinicians should consider these factors first before moving to more obscure reasons for inadequate stride length. Are we ready to move beyond definitions and begin to discuss the mechanisms of step and stride length, hoping to eventually tease out ways to improve and/or normalize them in our pathological patients?

Michael Orendurff, MS Gait Lab Portland Shriners Hospital MSO@shcc.org


Chris,

Your comments are somewhat confusing since "push off" does not involve much muscle activity. Preswing is primarily driven by momentum from an adequate step length on the contralateral limb. Inadequate push off is more an issue of inadequate step length on the opposite limb.....and this correction works clinically in most instances.

Melanie Weller MPT, ATC Sentara CarePlex Rehab Center Hampton, VA <MWellerPT@aol.com>


Chris (and others),

This is a slighltly belated reply to a post of yours from last week. Chronic lower back pain and other myofascial pain syndromes represent a hugh problem worldwide. Since some form of physical asymmetry is not uncommon in society (foot or hand size variation, leg length difference, etc), a problem may initiate when adaptions during gait occur to match this uneven form to the flat, level, generally unyielding surfaces (concrete, tile, wood floors etc.) on which we live and walk. While the adaptation visualized in a gait lab may seem to be minimal and inconsequential, when viewed as an event which is repeated over millions of cycles/year, the effect can be better understood from the prespective of it being a repetitive strain type injury. The circular walking pattern you described is common, but continually corrected by the patient through visual cues. These motions are so subtle that they are often appear meaningless. This may be an asymmetry in arm swing, slight trunk bend to one side but not the other, asymmetry in pelvic rotation, etc. However, the long term effect creates postural changes which are so gradual in their development, that they are often mistaken as primary pathology. The unilateral muscle tightness, for instance, classically seen in a physical therapy exam of a chronic lower back pain patient, is an example of how long term compensatation for this asymmetry manifests itself. Stretching and other forms of care of often successful in the short term, but recurrence has been shown to occur in approximately 70% of those who experience an initial attack. When this tightness is understood as the RESULT, rather than the cause of the problem, success of longterm outcome improves significantly. I have reported that in chronic postural pain patients previously considered at or near medical endpoiont, 77% demonstrate 50-100% improvement over a two year period, when they underwent video and in shoe pressure analysis of their gait with the information obtained used specifically to evaluate and prescribe podiatric type custom foot orthotics. A colleague of mine, Nick Sol, a podiatrist in Colorado Springs, Colorado has found similiar response in failed back syndrome patients. While not a single cause of LBP or other postural problems, it is rather a common form of "a perpetuating factor" as described by Simons & Travel's text on Trigger Points and referred myofascial pain syndromes. To those who suffer from this type of problem, appropriate care is often a godsent to them. It also provides an innovative use for gait lab technology with the custom foot orthotic design based on the analysis capable of reducing overall health care costs for chronic lower back and other myofascial pain syndromes. I would be happy to discuss this concept further.

Regards, Howard J. Dananberg, DPM, FACFOAM The Walking Clinic Bedford, New Hampshire howiedbpg@aol.com


Many thanks for all your comments. I was particularly interested to read:

"push off" does not involve much muscle activity. Preswing is primarily driven by momentum from an adequate step length on the contralateral limb (Melanie Weller)

my data suggests that ankle push-off energy on the contralateral side and first double support time are very important to side length on the ipsilateral side... The ankle push-off energy is a prime accelerator of the center of mass in the forward direction (see Kepple, et al, (1997) (Michael Orendurff)

It seems to me that we now need to discuss an issue which I think is central to the whole of clinical gait analysis: the origin of propulsive power and the role of ankle push-off.

Ever since David Winter demonstrated the overwhelming importance of plantarflexor A2 power in the gait cycle, this issue has been controversial. The South Californian camp of Jackie Perry and David Sutherland are still, as far as I understand, sceptical about the role of A2 in swing leg propulsion. Meanwhile, many people (I think wrongly) attribute trunk progression to A2. I think it's time we sorted this one out.

To my mind there's no doubt about the importance of A2 - it's really more surprising that it wasn't anticipated earlier, since the ankle joint moment and kinematics had been known for a long time. I suspect this sprang from a clinging to an Aristotelian belief that the hip was the cause of swing leg propulsion.

That said, it seems to me that At Hof's paper of 1993 ("Calf muscle work and segment energy changes in human treadmill walking", J. Electromyogr. Kinesiol. 2: 203 - 216) nicely showed that the purpose of the A2 power burst is confined to propulsion of the swing leg. In normal walking, the trunk seems to require virtually no energy for progression - it "goes along for the ride" in At Hof's words. Admittedly, things may be very different in pathological gaits.

So how do we fit these findings together? I find the more I look into this most fundamental issue, the less I understand!

Dr. Chris Kirtley (Kwok Kei Chi) MD PhD Assistant Professor Department of Rehabilitation Sciences The Hong Kong Polytechnic University


Dear All;

Thanks to Chris for starting such a splendid discussion, and to all of you for running with it!

First of all, an underlying assumption: We seek to quantify gait.

We seek relevant measurements of gait that will help us to classify different gaits and evaluate changes in gait using different appliances or over periods of time. Measurements of gait can be useful in the same way that body temperature or blood pressure measurements are useful: they allow comparisons with normative data, they enable monitoring of changes, often by plotting graphs, and they aid communication with people not present at the time of measurement.

Given that assumption, and acknowledging (while not addressing) Jeremy Linskell's first observations about details of procedure, I offer some thoughts on how we might explain stride length and step length to people new to the concepts.

Walking is a cyclic process where the same, or very similar, movements are repeated over and over again. We can describe the walking process by describing a representative cycle.

We can choose any event in the walking cycle to define the end of one cycle and the beginning of the next. The most common definition is initial contact of one foot or the other, but final contact, mid-swing, and other identifiable events in the gait-cycle have also been used.

Walking, or any other locomotor activity, is a process of moving through space over time, so whatever event we choose to define beginnings and endings of cycles, the two fundamental variables that describe gait-cycle dimensions are distance and time.

The distance moved in the direction of progression during one gait-cycle has traditionally, and quite consistently, been called the stride length. Stride length can be visualized as the distance between two successive footprints of the same foot -- either foot -- but other definitions are certainly possible. Stride width may be left for another discussion!

The time duration of a gait-cycle has traditionally been called stride duration. Stride length divided by stride duration would be average velocity.

In alternating bipedal locomotion, which is what we usually mean when we say walking, each gait-cycle, or stride, consists of two steps, one where the moving right foot swings forward (that is, in the direction of progression) relative to the stationary left foot ("right step") and one where the left foot moves forward relative to the stationary right foot ("left step"). In a normal gait pattern, we expect the right foot to swing about as far in front of the left foot as the left foot swings in front of the right. This would be a symmetrical gait.

Walking by no means requires symmetry, however: the right and left step lengths can be very different, but two successive step lengths always add up to one gait-cycle length, hence one stride length. Marcus Besser described a step length of zero, and step length on one side or the other can even be negative, as in a severe hemiplegic gait where one foot is dragged behind the other and never quite makes it up even with the foot in front.

Conclusion: For a person walking in a straight line, right and left stride lengths are (on the average) always the same, because either stride length consists of one right step plus one left step. The only difference between a right stride and a left stride is the order in which the two steps occur.

All the best!

Larry Lamoreux

Shriners Hospitals for Children Northern California Sacramento, CA 95817 USA Voice: 916-453-2280 E_mail: LLamoreux@AOL.com


Back to Teach-in