The BMG
The take-home message is crystal clear :
- if you intend to stretch the fascia lata, even only about 1% of its initial length, you should at least apply a load of 9075 Newtons (which would be equivalent to hang a 925 kg mass at its distal end!)
- the same argument would lead you to load the plantar fascia with a force equavalent to 852 kg (that's pretty precise!), and if you're lucky it would lengthen by 1 %.
In other words, if you held out hope about having a mechanical action with your hands on that kind of tissue, you'd better prepare to mourn for having let such a silly idea cross your manual therapist's mind, because it doesn't stand a single second when confronted to the evidence : fasciae's distortions are out of reach of human hands. Period!
These figures and conclusions are taken from a 2008 study by Chaudhry et al published in the Journal of the American Osteopathic Association (JAOA), "Three-Dimensional Mathematical Model for Déformation of the Human Fasciae in Manuel Therapy"
Let's dig on…
First and foremost, Chaudhry's study is about a mathematical model, which is a sort of simulation destined to mimic the mechanical behaviour of a specific material, in this case the fascia. Try to have a look at the study, it's totally out of reach of 99.9 % of any of us, me included (in spite of my mechanical engineer background), and with all due respect I would bet it's been out of reach of JAOA's reviewers too.
A mathematical model is only useful if it does reflect faithfully and relatively accurately the real-life situation it is supposed to mimic. In the present case, it is supposed to give distortions values of the fascia, given different loading hypothesis. If a model is reliable, it bears the advantage of running a bunch of simulations and produce loads of data without having to run real-life experiments anymore.
As René Thom, a Fields medal awarded french mathematician, said in an article named "modélisation et scientificité", a model is only useful if its outcomes match with the real-life measurements obtained from the real-life situations the mathematical model is supposed to simulate.
For an unknown reason, it looks like the JAOA authors didn't check the verification box this time. The verification step is absolutely necessary in order to assess any model's usefulness (Thom suggests to leave aside the term "validity" when talking about a model, he explains why in the original article, the idea behind this is that some qualitative models may prove to be useful even if not quantitatively accurate).
To build their model, the authors took real-life data extracted from an original research lead in 1964!!! Who knows why, in 2008, when many more recent and accurate experimental data were available, they decide to stick to the 1964 study about the mechanical behaviour of the plantar fascia and - again to my surprise - a 1931 study, cited in the 1964 one, about the fascia lata.
A quick literature review leads to many recent and reliable in-vivo and in-vitro studies, for example this one, by Gefen, about the distortions of the plantar fascia, and this one about the abdominal fascia. Another article published in the Journal of Biomechanics the same year Chaudhry et al. got their simulation accepted in the JAOA, provides a verified computerized model of the mechanical behaviour of the plantar fascia, based on the in-vivo data obtained by Gefen, and not the 1964 study…
A quick literature review leads to many recent and reliable in-vivo and in-vitro studies, for example this one, by Gefen, about the distortions of the plantar fascia, and this one about the abdominal fascia. Another article published in the Journal of Biomechanics the same year Chaudhry et al. got their simulation accepted in the JAOA, provides a verified computerized model of the mechanical behaviour of the plantar fascia, based on the in-vivo data obtained by Gefen, and not the 1964 study…
Because of the relatively low strength and high flexibility of plantar fascia, the gauge was especially designed for this study
Wright & Rennels, 1964
Let's have a closer look at the 1964 study in which Chaudhry and colleagues took the major part of the experimental data to build their mathematical model.
The experimental mechanism is relatively classic: they removed the plantar fascia from fresh cadavers, put them in a device destined to run a tensile test and traced the stress/strain diagrams.
A first surprise awaits the reader, as soon as the "materials and methods" section : "Because of the relatively low strength and high flexibility of plantar fascia, the gauge was especially designed for this study ". This statement is quite at the opposite of the undeformable fascia described by Chaudhry.
Not less surprising, the data analysis reveals that the tensile tests all ended in failure of the fasciae (failure is the obvious endpoint of most tensile tests), and that the failures generally occurred for a tension strength of 200 lb, which corresponds to ca 880 N, or 90 kg…
… that's about 10 times less than the 852 kg announced by the JAOA study!
Interestingly, just before failure, the plantar fascia specimen reached a relative stretching about 5%…
… Chaudhry et al arithmetics still predicted that a force equivalent to 852 kg would be required to observe a 1% lengthening!
The experimental mechanism is relatively classic: they removed the plantar fascia from fresh cadavers, put them in a device destined to run a tensile test and traced the stress/strain diagrams.
A first surprise awaits the reader, as soon as the "materials and methods" section : "Because of the relatively low strength and high flexibility of plantar fascia, the gauge was especially designed for this study ". This statement is quite at the opposite of the undeformable fascia described by Chaudhry.
Not less surprising, the data analysis reveals that the tensile tests all ended in failure of the fasciae (failure is the obvious endpoint of most tensile tests), and that the failures generally occurred for a tension strength of 200 lb, which corresponds to ca 880 N, or 90 kg…
… that's about 10 times less than the 852 kg announced by the JAOA study!
Interestingly, just before failure, the plantar fascia specimen reached a relative stretching about 5%…
… Chaudhry et al arithmetics still predicted that a force equivalent to 852 kg would be required to observe a 1% lengthening!
There must be a couple of flaws somewhere…
I'm totally confident in the good faith of the authors and their genuine commitment to provide a reliable, and probably ambitious, piece of research. I may be wrong but there should be a couple of flaws that can be hard to find in the maze of these 3 pages containing complex formulas and equations.
Wherever they may be, there are obviously major flaws that led to a mathematical model that is totally divergent from the real-life data. And that sole fact is sufficient to discard the model first, and more obviously the values it is supposed to generate.
In this case, the simulation errors take on dramatic proportions (never seen in my short but nevertheless actual experience as an industrial engineer graduate). In brief, the formulas predict that it would be necessary to put a load higher than 8000 N just to stretch the plantar fascia by about 1%, when the real-life data tell us, quite bluntly that :
1- about 100 N would do the work
2- no plantar fascia exist anymore beyond 880 N of traction
The second point is by far the most comical : it is physically impossible to reach the recommended 8000 N for the simple reason that the plantar fascia would have broken way before.
In this case the simulated and actual values diverge from a factor 80.
Just to give you an idea of how huge this divergence is, imagine a reckoning that is supposed to give a rather accurate estimation of your body weight, given other body data. If you weight 70 kg, an equation system that goes wrong by a factor 80 would predict à theoretical body weight of about 5600 kg !!!
Wherever they may be, there are obviously major flaws that led to a mathematical model that is totally divergent from the real-life data. And that sole fact is sufficient to discard the model first, and more obviously the values it is supposed to generate.
In this case, the simulation errors take on dramatic proportions (never seen in my short but nevertheless actual experience as an industrial engineer graduate). In brief, the formulas predict that it would be necessary to put a load higher than 8000 N just to stretch the plantar fascia by about 1%, when the real-life data tell us, quite bluntly that :
1- about 100 N would do the work
2- no plantar fascia exist anymore beyond 880 N of traction
The second point is by far the most comical : it is physically impossible to reach the recommended 8000 N for the simple reason that the plantar fascia would have broken way before.
In this case the simulated and actual values diverge from a factor 80.
Just to give you an idea of how huge this divergence is, imagine a reckoning that is supposed to give a rather accurate estimation of your body weight, given other body data. If you weight 70 kg, an equation system that goes wrong by a factor 80 would predict à theoretical body weight of about 5600 kg !!!
I've taken enough of your reading time but the rationale would be the same for the data and simulations of the force needed to stretch the facia lata.
You'll find hereby a copy of the Force/elongation diagram of the 1931 study cited by the 1964 study that has been used by Chaudhry.
Here again, failure would occur, with a corresponding stretch of about 10 %, long before the predicted values displayed in the JAOA would be reached.
You'll find hereby a copy of the Force/elongation diagram of the 1931 study cited by the 1964 study that has been used by Chaudhry.
Here again, failure would occur, with a corresponding stretch of about 10 %, long before the predicted values displayed in the JAOA would be reached.
Actual order of magnitude of dense fascia distortions
There are many recent and reliable studies that have been conducted in order to better understand and measure the mechanical properties of dense fascial tissue, particularly the plantar fascia and the fascia lata. Some studies include in-vivo or in-vitro experimentations, other are actually coherent computer modelling compliant with real-life measurements.
Values vary between the different studies but just to keep in mind an ordrer of magnitude of the dense fascia, it requires a force about 100 to 200 N (which correspond to the weight of a 10 to 20 kg mass) to get a stretch of about 1 % (see the figures at the end of this blog post).
This is totally consistent with the evidence that every tissue in the body undergoes stretches and distortions during daily activities, may it be the skin, the muscle, the fascia or the bone. The order of magnitude of these distortions is compliant with the most probable loading conditions the body will be confronted to, even taking into account a safety factor preventing those distortions and stresses to be sufficiently distant from the failure values.
For those who want to dig further and get a comprehensive update about the facial tissue, the BJSM published two years ago Consensus Statement available in Open Access (thanks to Valery Belan (AUS) for the tip ;-).
Values vary between the different studies but just to keep in mind an ordrer of magnitude of the dense fascia, it requires a force about 100 to 200 N (which correspond to the weight of a 10 to 20 kg mass) to get a stretch of about 1 % (see the figures at the end of this blog post).
This is totally consistent with the evidence that every tissue in the body undergoes stretches and distortions during daily activities, may it be the skin, the muscle, the fascia or the bone. The order of magnitude of these distortions is compliant with the most probable loading conditions the body will be confronted to, even taking into account a safety factor preventing those distortions and stresses to be sufficiently distant from the failure values.
For those who want to dig further and get a comprehensive update about the facial tissue, the BJSM published two years ago Consensus Statement available in Open Access (thanks to Valery Belan (AUS) for the tip ;-).
Conclusion
But we know that living tissue distort during daily activities and the amount of tension or pressure needed to provoke these deformations are not at all as huge and out of reach as predicted by Chaudhry et al.
Technically, if one would carry on with this research theme, it would be interresting to assess the mechanical behaviour of the fascial (and other) tissues in a situation of 3 points flexion, which figures better the real-life situation than the axial tensile test.
To conclude, the major pitfall in manual therapy, may it be physio, chiro, osteo or whatever, are the explanatory models: the models we use to teach students what to do, to explain patients what we do and to convince ourselves that it works.
As you can see in the comment in the image attached that says "the key is to affect fascial neurons", manual therapies are a perfect breeding ground for bullshit theories, and we are sometimes tempted to restore some sort of equilibrium by incarnating the opposite excess, which is unfortunately totally counter-productive by reinforcing the beliefs we're trying to change, and digging a little deeper the pit that separates the informed from the uniformed practitioners.
I'm not an absolute skeptic, some kind of magic may still happen in a unpredictable way during a manual therapy session. Narratives, fictions and analogies may still be useful when trying to transmit some sort of know-how, as they are useful in the therapeutic encounter to foster change and engage in better comprehension of the factors involved in a problematic situation.
But these narratives, this kind of mental imagery, is not intended to substitute for physical and rigorous understanding of the mechanisms that underlie our activities.
Keep it pure, but keep it plausible
MG
A Blog post by Marco Gabutti
F L X 