Five toed windlass

Are they? Who says?

 

Came across this paper. No pdf online unfortunately.


Clinical Orthopaedics & Related Research:

December 2001 - Volume 393 - Issue - pp 326-334


Contributions of Active and Passive Toe Flexion to Forefoot Loading


Hamel, Andrew J. PhD*,**; Donahue, Seth W. PhD‡; Sharkey, Neil A. PhD*,†,‡

Abstract


Toe flexion during terminal stance has an active component contributed by the muscles that flex the toes and a passive component contributed by the plantar fascia. This study examined the relative importance of these two mechanisms in maintaining proper force sharing between the toes and forefoot. Thirteen nonpaired cadaver feet were tested in a dynamic gait stimulator, which reproduces the kinematics and kinetics of the foot, ankle, and tibia by applying physiologic muscle forces and proximal tibial kinematics. The distribution of plantar pressure beneath the foot was measured at the terminal stance phase of gait under normal extrinsic muscle activity with an intact plantar fascia, in the absence of extrinsic toe flexor activity (no flexor hallucis longus or flexor digitorum longus) with an intact plantar fascia, and after complete fasciotomy with normal extrinsic toe flexor activity. In the absence of the toe flexor muscles or after plantar fasciotomy the contact area decreased beneath the toes and contact force shifted from the toes to the metatarsal heads. In addition, pressure distribution beneath the metatarsal heads after fasciotomy shifted laterally and posteriorly, indicating that the plantar fascia enables more efficient force transmission through the high gear axis during locomotion. The plantar fascia enables the toes to provide plantar-directed force and bear high loads during push-off.

 

I took this one just now, have a look here...
I would think digit five isn't Dflexing the same amount as the hallux.

Never thought about it from this angle - I like the idea, makes sense.

 

Thanks Athol, would have been great if they had transected each digital slip in turn.


I been thinking about the lesser digit dorsiflexion test, i.e. performing a Jacks test on each of the digits in turn. Each time we dorsiflex a digit we exert both an internal moment through the plantar-fascia to its attachment at the medial tubercle of the calcaneus and an external moment through the plantarflexion of the metatarsal head on the ground.

So, if we take the internal moments first, effectively the dorsiflexion of the digit is trying to pull the calcaneus forward (or distally) through its attachment to the calcaneus, it can only do this if the force it produces are higher than the forces trying to hold the calcaneus in its original position. But because of the relationship between the medial tubercle of the calcaneus to the subtalar joint axis, this internal moment is usually going to be a supination moment about the subtalar axis. To try to simplify, I'm ignoring the posterior compression of the rays for now, Eric.

If we think about the external moments acting at the subtalar joint, due to plantarflexion of each metatarsal head in response to digital dorsiflexion, they will vary from toe to toe, depending on the subtalar axial position. Think about how dorsiflexing each toe sequentially might shift the centre of pressure location, relative to the axial position.

So, the net moment generated by the dorsiflexion of each digit via the windlass mechanism is the sum of the internal and external moments. It's all windlass mechanism, but sometimes dorsflexing a digit in isolation might result in a net pronation moment about the STJ axis, and sometimes it will result in a net supination moment in isolation.

Try the digital dorsiflexion test for yourself. Then think about how these forces act when the digits are working as they might do during gait.

Try dorsiflexing the hallux in isolation, then the hallux + 2nd toe, then hallux + 2nd + 3rd, then hallux + 2nd + 3rd + 4th. Then do it the other way: dorsiflex the 4th toe in isolation, then 4th + 3rd, then 4th + 3rd + 2nd, then 4th +3rd +2nd + 1st toes. Does it matter which order the digits are dorsiflexed in terms of kinematic effect?

 

This is where I think digital and metatarsal lengths are important. Lets assume that the contact period for all the toes is the same and they all load at the same time (this probably isn't true but we'll avoid phasic activity for now). Given this, the shorter the ray + the shorter the digit then the greater rotational excursion the digit has to undergo in order to maintain contact with the floor.

Now, given that the dorsiflexion of each toe might differentially influence the external moment acting about the subtalar joint axis via depression of the metatarsal heads, and that we know from pressure studies that the loading and unloading of the metatarsal heads is phasic during gait, once the metatarsal head belonging to a certain ray leaves the ground then the ground reaction force adding to the external moment about the STJ axis from that metatarsal head becomes zero. However, the lever arm for the external moment from that ray segment is probably then extended to the apex of the toe until it then leaves the ground. At which time the toe/ ray in question cannot add anything to the external moment via it's windlassing, it can however, continue to add to the internal moment by virtue of it's relative extension. Hope that makes sense.

 

Simon, not sure i'm following you here...

Follow you so far

this is where i'm lost, what do you refer to as 'relative extension' creating an internal moment?

 

Ken, if the toe remains extended (dorsiflexed) after it has left the ground, then by virtue of the increase in tension and/ or shortening of the distance between the origin and insertion of the plantar fascial slip of that digit (windlass effect, if you like), the toe, by virtue of its position, will still be exerting an internal moment via the plantar fascia on the medial tubercle of the calcaneus, but it will no longer be exerting any influence on the external moment acting about the STJ axis.

Recall that I said that the net moment each ray brings to the windlass party is the sum of the internal and external moment, and that I said the internal moment was probably/ usually supination via the medial tubercle attachment. Thus, when the toe leaves the ground, if it remains dorsiflexed, it will most likely only ever produce supination moment at the STJ axis via it's windlass. When the toe and metatarsal head where on the ground, if the external moment it produced was an STJ pronation moment by virtue of the metatarsal heads position relative to the STJ axis and this external moment was greater than the internal STJ supination moment that this ray was producing, then up until the point when the toe left the ground, this ray would have been adding to the net pronation moment and retarding resupination. Once free from the ground, this digit will now only add to the supination moment, so long as the digit stays dorsiflexed. Take a look at toe position during late propulsion / early swing.

 

It makes sense I think. The digits would provide force through an internal plantarflexion moment and reverse windlass as the plantar fascia and tendons release the stored elastic strain energy.

Although the study attached seems to think the 1st MTP joint is not great at returning this energy.

A few paragraphs attached below:

Recent research has indicated that the role of energy absorption at the MTP joint is wasteful in terms of maximizing athletic performance for sprint running tasks (Stefanyshyn & Nigg, 1997).
The MTP joint produces negligible energy during the late stance phase to propel the body forward. Thus, any stored strain energy during early stance is either
dissipated or used in performing kinematic patterns, which do not contribute to the primary movement goal. It has been proposed that minimizing the loss of
this energy is important because it would then be available to enhance performance if it would not have been spent unnecessarily (Stefanyshyn & Nigg, 2000a). In fact, it was reported that increasing the bending stiffness of midsole materials decreased the
energy absorption and peak extension of the MTP joint, which resulted in improved jumping performance (Stefanyshyn & Nigg, 2000b)

 

I'd like to revise that further: the shorter the ray + the shorter the digit + the smaller the radius of the metatarsal head (It's actually the distance from the centre of rotation of the digit about its metatarsal head to the insertion of the fascial slip) then the greater the rotational excursion the digit has to undergo in order to maintain contact with the floor.

 

with you now!

They are indeed probably in most cases in a DF position, sometimes even all the way through to initial heel contact. In some cases I was thinking it was due to lack of force of the tibial anterior muscle en then the secondary DF muscles came in to help. This brings me actually to a point I was looking for an answer on how to create a better (more inverted) heel contact. Don't want to get the topic sidetracked here though...

 

So the question really becomes, do the toes dorsiflex by the same amounts and at the same time during gait and do the metatarsal heads and toes leave the ground at the same time?

We know the answers to part of that, but not all.

 

an extra something to think about could be that certain toes DF even a bit more as the GRF underneath the metatarsal head stops
so this would bring us to
1) do they dorsiflex the same amount during propulsion,
2) when does each metatarsal head/toe leave the ground
(3) does the toe dorsiflex further after GRF stops at the metatarsal head)

 

Sorry, this is a bit off thread but the above is very interesting. Almost without exception, the best sprinters I have seen (club level) all have high passive dorsiflexion stiffness at the 1st MPJ. This is invariably not a structural problem but one of tightness of the plantar structures which seems to be mechanically beneficial. Never could work out the chicken and the egg!

Kenva, regarding more inverted initial contact, the only way I can think to do it is to make the tib ant more mechanically efficient - ie increasing the distance from the tib ant insertions to the STJ axis, although, regardless of how mechanically efficient it is, the CNS probabaly regulates the initial contact frontal plane alignment.

I see a lot of race walkers and the inverted heel with maximum dorsiflexion of the ankle on the straight leg is the holy grail. They are actively trying to acheive supination at the sub talar joint and when the power walkers convert to proper race walking technique, the tib ant invariably becomes overused.

Encouraging late stance supination by improving Windlass function causes the load on the tib ant to be reduced when trying to acheive supination at initil contact.

I realise this doesn't add much to this thread - apologies

 

Ken and Simon:

I agree that digits 1 and 5 probably dorsiflex at slightly different times during gait and with different degrees. However, adjacent toes probably dorsiflex more at the same time and the same degree. It would be interesting to know how much they all differ in dorsiflexion timing and magnitude.

 

Athol:

Running and walking biomechanics are very different from each other so please don't assume that the running-related research you present here necessarily easily translates to the walking gait pattern.For example, during propulsion during walking, the center of mass of the body is lowering but during propulsion in running, the center of mass of the body is raising.

In addition, I don't believe that the propulsion phase in walking occurs at the correct frequency (i.e. time period) to allow substantial elastic strain energy to be stored and released with each step, while during running, jumping and hopping, storage and release of elastic strain energy is very important.

 

In my paper on the windlass I described the effect of the proximal push and movement of the proximal bones. As the STJ moves from pronation to supination the talar head moves posteriorly relative to the anterior process of the calcaneus. When the tension in the fascia increases and the windlass moves, the head of the phalanx will push the metatarsal head proximally, which will.... push the talar head proximally. This is only true for meatarsal 1,2, and 3. The proximal push on the the 4th and 5th will push he calcaneus proximally while tension in the plantar fascia will pull the calcaneus anteriorly. So, for metatarsals, 4 and 5 you would get a plantar flexion moment of the lateral column, but you would probably not get supination of the STJ.

Eric

 

I agree. Perhaps someone with access to 3D kinematic analysis could put markers on each of the metatarsals and digits and report the data. I'm somewhat surprised that this has seemingly not been done. Perhaps there are methodological difficulties which make it difficult- size of markers, proximity of markers etc.?

 

Hi Kevin,

I agree that the biomechanics of walking and running are obviously not the same.

Maybe a bit off topic but...

I don't agree that the storage and release of elastic strain energy is less important during walking. It may happen in a different manner (economical storage and
release of elastic energy) but it does contribute to elastic recoil by a catapult action in walking through the action on the gastroc/soleus musculo-tendinous unit which will have an affect on the plantar fascia function.

See attached paper (about walking this time!)

Kind regards,
Athol

 

Athol:

Interesting paper, thanks! I think this paper admits that elastic strain energy is different during walking and running and, upon initial skimming through this paper, and previous reading of R. McNeill Alexander's works on the economy of running and walking, I still believe that the elastic strain energy storage and recoil in running is much greater than that in walking. However, I like the analogy of the "catapult" but, honestly, wonder how much mechanical significance this "catapult" really plays in storage and release of energy during human walking. I am still skeptical it is really that signficant but am keeping my mind open. Sounds like this will require much further research.

 

In the foot skeleton I keep by my desk at home (a real one...not a plastic one), I estimate the sagittal plane distance from the center of the 1st metatarsophalangel joint (MPJ) to the estimated plantar sesamoids to be 14 mm. For the lesser MPJs, here are the measurements I made from the center of the MPJs to their plantar aspects when held in anatomic position relative to the ground:

2nd MPJ = 8 mm
3rd MPJ = 7 mm
4th MPJ = 7 mm
5th MPJ = 6 mm.

There are a number of variables at work here that will influence the effect of the "windlass" of all the MPJs on subtalar joint (STJ) moments, assuming an intact plantar fascia:

1. Medial or lateral position of plantar metatarsal head relative to STJ axis (i.e. determines whether MPJ windlass will produce supination or pronation moments across STJ).

2. Perpendicular distance of plantar metatarsal head from STJ axis (i.e. determines length of supination/pronation moment arm to STJ).

3. Distance from plantar fascia to longitudinal arch height at each individual metatarsal ray (i.e. determines length of moment arm for plantar fascia to cause an arch-raising, or forefoot plantarflexion, moment at each individual metatarsal ray).

4. Perpendicular distance from plantar fascia to center of rotation of MPJ at plantar MPJ (i.e. determines not only the length of moment arm for plantar fascia to cause MPJ plantarflexion moment but also the amount of shortening that will occur from distal metatarsal head to plantar calcaneus for every degree of digital dorsiflexion).

5. Tensile stiffness of each slip of the plantar fascia to each proximal phalanx base (i.e. determines amount of stretch, or strain, that will occur in each slip of the plantar fascia for a given magnitude of tensile loading force).

Obviously, this becomes a very complex problem that would probably require a signficant investment of time (and much more knowledge than I have) in front of a computer doing a finite element model in combination with forward dynamics model to determine how changes in foot geometry may affect gait and STJ moments.

However, for now, with these five factors listed above, along with my measurements from a foot skeleton, we can probably make a few generalizations:

A. The first MPJ windlass has, by far, the greatest potential to cause STJ supination moments due not only to its most medial position in the forefoot, which allows it the greatest potential to create STJ supination moments, but also due the medial band of the plantar fascia having the longest moment arm for producing longitudinal-arch raising moments, due to the first ray segment having the highest longitudinal arch height. I estimate that for a normal STJ axis position, the first MPJ windlass has at least 50 times the supination potential than does the 2nd MPJ windlass, due mostly to its much longer supination moment arm to the STJ axis and the much larger joint-sesamoid diameter.

B. The 5th MPJ windlass, even though it is the most lateral MPJ, has a very weak pronation moment effect due to its very small MPJ diameter and the nearly flat arch height of the lateral longitudinal arch.

C. The 2nd MPJ windlass has the most potential to cause a longitudinal arch raising effect of all the lesser MPJ windlasses, but generally contributes minimal STJ moments due to the fact that it tends to be very close to the STJ axis than the 3rd, 4th and 5th MPJs, thereby decreasing the potential of the 2nd MPJ to cause either STJ pronation or STJ supination moments. The 2nd MPJ windlass though does contribute significantly to forefoot plantarflexion moments.

D. The 3rd and 4th MPJ windlass, being more lateral to the STJ axis than the 2nd MPJ windlass probably exert signficantly more STJ pronation moment than does the 2nd MPJ windlass, but again, probably only a fraction of the STJ moment that is produced by the 1st MPJ windlass. Of course, this depends very much on longitudinal arch height and STJ axis spatial location.

Sounds like a good paper which easily could be published one day.

 

Here's a study I'd love to see performed using a cadaver gait simulator: intact fascia, then transect each digital slip sequentially starting at 1st through to 5th, then transect each digital slip sequentially from 5th through to 1st.

I agree wholeheartedly with Kevin that this is a complex problem. I think small steps may open the doors to enlightenment- personally I can't believe that the dynamic range of motion and phasic relationships of digital dorsiflexion during gait is unknown in 2012. If I had the kit, I'd be running that tomorrow. That's a nudge BTW: Ken, Athol and anyone else who may be looking for a research project...

 

Simon:

Don't forget the contributions of the plantar intrinsics, which are not currently being "activated" in any of the world's "dynamic cadaver gait simulators". Luke Kelly and his colleagues, I believe, are making significant progress toward helping us to all appreciate just how important these small muscles may be in affecting the dynamics of the plantar fascia and of the function of the foot and lower extremity as a whole.

 

Sorry to sound thick , but what is a cadever gait simulator ?
Tim

 

It's a jig that you can put a leg in and it "pulls the tendons" to make the leg walk. As Kevin rightly points out, while these machines can apply the correct phasic forces via the the extrinsics, they cannot "fire" the intrinsics.

Here you go Timmy: http://www.youtube.com/watch?v=iEZBfiA63kI

 

I agree. I also think that the newsletter which you sent me early this year fits well with the work Luke is doing. For those that remember, earlier this year when I was talking about the inability to tip-toe rise using only the gastroc/ soleus, one of the points I was trying to get to is that in order to perform this activity, foot stabilization is probably required by increasing midfoot stiffness via the plantar intrinsics. Not sure if I got to that at the time, since I think we got detoured.

 

And the degree of digital dorsiflexion. In the land of the hallux rigidus, the lesser toes are kings. But what happens in the land of the lesser toe deformity?

 

Also, notice that we rarely need to discuss the condition of Lesser MPJ Limitus/Rigidus...there is a good mechanical reason for that.

 

Agreed, yet we do commonly see clawing of the lesser toes. What influence might this have on their contribution to windlass function?

 

Stainsby would have argued that claw toes are due to the windlass on the lessor toes not working properly.

 

I might argue that regardless of cause or effect, in the presence of claw toes the windlass effect from the lesser toes during gait is somewhat altered. Dare I say, diminished.

 

In the cavus foot, claw toes are likely caused by insufficient plantar fascial tensile force due to the higher arch shape of the foot. Higher arched feet results in less plantar fascial tension force, less MPJ plantarflexion moment from the plantar fascia and more liklihood that the MPJ dorsiflexion moments from the long digital flexors will cause claw toes.

Lower arched feet will have increased plantar fascial tensile force, which allows greater MPJ plantarflexion moments to be generated by the combined mechanical action of Achilles tendon and plantar fascial tension forces. Normal to lower arched feet better ensure good digital purchase force and helps prevent digital deformities, as long as the longitudinal arch isn't too low.

No plantar fascia due to plantar plate tear?...digital deformities at the affected digits due to lack of adequate MPJ plantarflexion moments from the plantar fascia which helps maintain proper digital purchase and sagittal plane alignment.

No plantar fascia due to plantar fascial rupture/total fasciotomy?...digital deformities across the board due to lack of adequate MPJ plantarflexion moments from the plantar fascia which helps maintain proper digital purchase and sagittal plane alignment.

Another good paper that needs to be written.

 

Eric, are you saying that the internal moment about the STJ axis from the 4th and 5th toes due to windlass function will always be pronatory due to retrograde compression at the metatarsal bases, despite the pull of the fascia at the medial tubercle of the calcaneus due to dorsiflexion of these digits? Or, is it the sum of internal "push" and "pull"?

 

I was going to post a thought experiment in which we transected the plantar fascial slips from toes 2-5 but left the slip to the hallux intact. I'd thought we might end up with a cavus foot. Does a cavus foot cause clawed toes, or do clawed toes cause a cavus foot? Further to my response to Craig: in the presence of supinatory windlass function from the hallux, without sufficient counter pronatory windlass force from the lateral digits, should the foot become supinated; viz. it becomes cavus?

Y'all following... ya' see how good this thread has become now?

 

Hi everyone

Great discussion!! I'm still trying to get my head around all the concepts, but I have an article to add for your reading pleasure. An interesting concept from this paper relates to the "pre-loading" of the plantar fascia during late swing. They also describe nicely the phasic strain in the fascia as gait velocity increases.

one questions of my own...

Is there a potential role of lateral force transmission from the intrinsics to the plantar fascia? If so can tension in the PF be directly modified by the activity in the intriniscs? what effect may this have on the lesser toes?


Anyway, I'll rejoin the discussion when I have my head around it

 

I'm not sure that the fascia to the 4th and 5th toes has any effect at the STJ. You could argue that it does not cross the joint. Whereas, the retrograde force from toes 1, 2, and 3 pushes proximally on the talar head, which is on the other side of the STJ from the fascial insertion on the calcaneus, so the fascial forces from 1,2, and 3 does cross the joint.

This is just part of the effect. If he windlass does plantarflex the metatarsals, this may shift the location of the center of pressrue of ground reaction force and this could create a change in moment at the STJ.


Eric

 

Sorry, here's the article...

 

in the sideline of the discussion

Robin,

according to this carravagi study

 

Simon:

Cutting the plantar fascia to the lesser digits would tend to lower the lateral and central longitudinal arch, but would also tend to supinate the foot.

In 1885, Newton Shaffer described a high-arched foot, which he called the non-deforming clubfoot, which was taught to us as being a "Shaffer foot", where a cavus foot deformity is associated with claw toe deformities. I have seen many of these feet over the years and believe the higher longitudinal arch morphology precedes the development of the clawtoes since I have seen clawtoes gradually develop over the years in these feet.

My hypothesis is that the reduced plantar fascial tension from the higher arch foot structure reduces the magnitude of MPJ plantarflexion moment which then, over time, allows the MPJ dorsiflexion moments from the long digital flexors to overpower the MPJ plantarflexion moment from the plantar fascia, causing clawtoes to gradually develop. This theory of clawtoe development in cavus foot deformity makes the most biomechanical sense to me.

 

Yep, the intrinsic mechanism which tunes / adjust the stiffness of the arch are the plantar intrinsic muscles.

 

Which brings me back to my question.

We have already suggested that the intrinsics can "indirectly" reduce the tension on the plantar fascia when activated, by applying a plantar-flexion moment to the mid-foot.

Is it possible that these muscles may be able able to "directly" reduce strain in the plantar fascia via attachments into the fascial slips along the length of the arch. These muscles are generally bi-pennate and have a relatively small fibre length and hence idf they had a direct communication with the fascia they could possibly exert a proximal "lifting" force on the fascia. This would would directly reduce the longitudinal strain in the fascia. Maybe Eric or Kevin could comment on whether these muscles are discreet units, or whether they communicate with the fascia at all. I suspect there would be considerable amount of connection allowing for lateral force transmission.



I hope this makes sense?