Fascia, and the Gap Between the Research and the Sales Pitch

For most of the twentieth century, fascia was the thing you cut through to get to the interesting part. It was the white, slippery, faintly disappointing material that a dissector scraped off a muscle so that the muscle could be seen and named. Within about fifteen years it became the headline on treatment menus, the subject of its own international research congresses, and the explanation of choice for almost anything a pair of hands might do to a body. That is an unusually fast journey for a tissue, and journeys that fast are worth looking at closely.

This page keeps four things apart that usually arrive fused together: what fascia anatomically is, what research has actually found about it, what the bodywork industry says that research means, and what this house does with its hands on a Tuesday afternoon. They are not the same, and the distance between the second and the third is the real story here.

Moonlight sells bodywork, and two of its Experiences name fascia explicitly. Head, Face & Deep Rest describes "the slow release used in fascia work"; Body Reset describes pressure "worked into the fascia and the deeper layers of muscle rather than only across the surface." That is a commercial position, and it gives this house an obvious interest in fascia sounding profound, ancient, and mechanically consequential. A reader should discount what follows accordingly.

The only way to earn that discount back is to publish the parts that do not help. So: the single most damaging finding for the standard "myofascial release" explanation is set out below in full, with its numbers, and it is not hedged. The house takes the view that a customer who understands what she is actually buying is a better customer than one who has been told a flattering story about her collagen.

Fascia is connective tissue: collagen and elastin fibres held in a hydrated ground substance, populated by fibroblasts and other cells. It is arranged in layers. Superficial fascia sits within the subcutaneous tissue, between skin and the deeper structures. Deep fascia includes the tough aponeurotic sheets such as the fascia lata of the thigh and the rectus sheath of the abdomen, the thinner epimysial layers that wrap and divide muscles, and the retinacula that strap tendons down across joints. Between the dense fibrous layers lie thinner layers of loose connective tissue whose job is to let the dense layers glide over one another.

That gliding function is not folklore; it has been quantified. Fede and colleagues, working in Carla Stecco’s group at the University of Padua, measured hyaluronan content across human fascial samples and published the results in the Journal of Anatomy in 2018. Fascia lata carried about 35 micrograms of hyaluronan per gram of tissue and the rectus sheath about 29; fascia adherent to the trapezius and deltoid carried roughly 6; the retinacula around the ankle, where large excursions occur, carried about 90. Sites that need to slide are lubricated more. That is a satisfying, ordinary, mechanical fact about a tissue, arrived at by measuring it.

The property that made fascia conceptually exciting is continuity: the tissue does not stop neatly at the edge of a muscle. It is also, awkwardly, the property that makes fascia hard to define. The Fascia Research Society convened a Fascia Nomenclature Committee and ran a Delphi consensus process; Adstrum, Hedley, Schleip, Stecco and Yucesoy published the result as "Defining the fascial system" in the Journal of Bodywork and Movement Therapies in 2017, proposing that the fascial system is the three-dimensional continuum of soft, collagen-containing, loose and dense fibrous connective tissues permeating the body. Not everyone accepted it. In Clinical Anatomy in July 2026, John Sharkey of the Irish College of Osteopathic Medicine published "Fascia Is Not a System", arguing that under the nomenclature standards of the International Federation of Associations of Anatomists and the Federative International Programme for Anatomical Terminologies, fascia does not meet the criteria for an anatomical system at all, and that calling it one has produced conceptual drift in teaching and research. That disagreement is live, and it is between anatomists, not between science and quackery.

The First International Fascia Research Congress was held on 4–5 October 2007 at the Conference Center of Harvard Medical School in Boston. It was coordinated by Thomas Findley, MD, PhD, and Robert Schleip — both of whom were, and are, practising Rolfers as well as researchers. Something over two hundred people attended, roughly three-quarters of them manual therapists rather than laboratory scientists. Further congresses followed in Amsterdam in 2009, Vancouver in 2012, Reston, Virginia in 2015, Berlin in 2018 and Montreal in 2022. It is worth stating plainly what that founding arrangement means: the modern field of fascia research was convened at the meeting point of manual therapy and laboratory science, by people with a foot in each. That produced good anatomy quickly. It also meant that findings tended to arrive with a treatment rationale already attached to them.

Two things made the timing possible. One was a reframing. Helene Langevin and Jason Yandow, publishing in The Anatomical Record in 2002, mapped acupuncture points and meridians against intermuscular and intramuscular connective tissue planes and reported that more than eighty per cent of points, and the majority of meridian pathways, lay along such planes. Whatever one concludes from that, it put connective tissue at the centre of a question people were already arguing about. The other was imaging. Ultrasound elastography made it possible to watch tissue layers move in a living person. Langevin and colleagues used it in a study of 121 subjects published in BMC Musculoskeletal Disorders in 2011: thoracolumbar fascia shear strain during passive trunk flexion averaged 56.4 per cent in people with chronic low back pain against 70.2 per cent in those without — about twenty per cent lower.

That study is genuinely interesting and it is also cross-sectional. It shows an association at one moment in time. Langevin and her co-authors said so themselves, writing that the reduction "may be due to abnormal trunk movement patterns and/or intrinsic connective tissue pathology" — which is a careful way of saying that nobody knows whether stiffer gliding causes the pain, follows from moving differently because of the pain, or arrives alongside both. Marketing copy has not generally preserved that sentence. And the older claim that anatomists deliberately discarded fascia for centuries, which circulates as the field’s origin myth, could not be traced here to any primary source; what can be documented is narrower and duller — Tesarz and colleagues open their 2011 study by noting that the available data on thoracolumbar fascia innervation were "inconsistent and partly contradictory". Under-described is not the same as suppressed.

In October 2005, in a paper carried by the 2006 volume of Medical Hypotheses, Schleip and colleagues proposed that intramuscular connective tissue — the perimysium in particular — might contain enough myofibroblasts, cells with smooth-muscle-like contractile behaviour, to contract actively and influence passive muscle stiffness. Fourteen years later the same group tested it. In Frontiers in Physiology in 2019, Schleip, Gabbiani, Wilke, Hinz, Klingler and others reported immunohistochemical staining across 31 human donors and 20 animals, finding the highest myofibroblast density in human lumbar fascia, and mechanographic force measurements on isolated rat fascia showing contraction in response to fetal bovine serum, a thromboxane analogue, TGF-beta 1 and mepyramine, and relaxation in response to a Rho kinase inhibitor.

So fascia does contract. The sentence worth reading twice is the one those authors wrote about what it amounts to: applying the registered forces hypothetically to human lumbar tissue "predicts a potential impact below the threshold for mechanical spinal stability but strong enough to possibly alter motoneuronal coordination." That is the field’s own most committed researchers reporting that the contraction they found is too weak to brace a spine and is interesting mainly as a signal to the nervous system. It is a finding about communication, not about scaffolding.

The sensory side is stronger. Tesarz, Hoheisel, Wiedenhöfer and Mense, in Neuroscience in 2011, quantified CGRP- and substance-P-containing free nerve endings in rat thoracolumbar fascia and found it densely innervated, with the presumed-nociceptive substance-P endings located exclusively in the subcutaneous tissue and the outer fascial layer; their human work was preliminary and non-quantitative, which they said. Schilder and colleagues, in Pain in 2014, injected hypertonic saline under ultrasound guidance into the erector spinae muscle, the thoracolumbar fascia and the overlying subcutis of twelve healthy volunteers: the fascia injections produced longer-lasting and more widely radiating pain than either the muscle or the subcutis. Twelve people is twelve people. But fascia being a pain-relevant, richly innervated tissue is about as well supported as anything in this essay.

One number should be handled with tongs. The figure of 250 million nerve endings in the body-wide fascial network, from which the popular line "fascia is our richest sensory organ" descends, circulates everywhere, and the accompanying ratio against muscle is quoted variously as three times and as six times depending on who is quoting. It could not be traced here to a primary counting study. It appears to be an estimate made by advocates of the field and then repeated as though it were a measurement. The underlying point — that fascia is sensorily important and was long treated as inert packing — is defensible. The number is not, and this page will not use it.

The standard explanation offered for myofascial release is that sustained manual pressure plastically deforms fascia: the tissue was short, tight or stuck, and the practitioner’s hands lengthen it. In 2008, in the Journal of the American Osteopathic Association, Hans Chaudhry, Robert Schleip, Zhiming Ji, Bruce Bukiet, Miriam Maney and Thomas Findley built a three-dimensional finite-deformation model to find out how much force that would take. For the fascia lata, producing one per cent compression and one per cent shear required a normal load on the order of 9,075 newtons — roughly 925 kilograms — together with a tangential force of about 4,515 newtons. For the plantar fascia the figures were about 8,359 newtons and 4,158 newtons. Their conclusion, in their own words: the palpable sensations of tissue release reported by manual therapists "cannot be due to deformations produced in the firm tissues of plantar fascia and fascia lata."

Note who wrote that. Schleip and Findley are the two people who convened the 2007 Harvard congress; Schleip trained as a Rolfer and has spent his career arguing for fascia’s importance. The most damaging single result for the mechanical story was produced by the founders of the field, published in an osteopathic journal, and has been available for eighteen years. This is not a case of outside sceptics attacking a practice. It is a case of a practice being told something inconvenient by its own laboratory and, for the most part, not passing the message on to its customers.

The finding deserves its caveats stated fairly, because overstating it would be the same error in the opposite direction. It is a mathematical model, calibrated on one volunteer’s tissue and on prior experimental data, not a measurement of what occurs during a session. And its own results distinguish between tissues: superficial nasal fascia, a softer tissue, did deform substantially under achievable force, which is why the authors allow that palpable release "could result from deformation in softer tissues". So the defensible statement is not "hands cannot change fascia." It is narrower and more useful: the dense fibrous sheets that marketing invokes — the iliotibial tract, the plantar fascia, the thoracolumbar sheet — cannot plausibly be plastically deformed by a human being pressing on them.

Schleip had in fact proposed the alternative five years before the modelling paper. In "Fascial plasticity — a new neurobiological explanation", published in two parts in the Journal of Bodywork and Movement Therapies in 2003, he argued that the changes practitioners feel under their hands are better explained by the nervous system than by tissue mechanics: fascia is densely supplied with mechanoreceptors, stimulating them alters sympathetic tone, and altered tone changes muscle tension and local fluid dynamics. The honest version of the fascia story was published by its leading advocate at the very start, and the industry built its marketing on the version he had already set aside.

If the tissue is not being stretched out, the plausible mechanisms are neural, reflexive, circulatory, perceptual and interoceptive. The most-cited framework is the one Joel Bialosky, Mark Bishop, Donald Price, Michael Robinson and Steven George published in Manual Therapy in 2009: a mechanical force from manual therapy initiates a cascade of peripheral and central nervous system responses, and those responses — not the mechanical event — are what produce the outcome. It should be said clearly that the authors presented this as a model to direct future research, not as a demonstrated pathway. It organises the question well. It has not closed it.

One strand is unusually well established and rarely mentioned in bodywork marketing, probably because it is about pleasure rather than repair. Human hairy skin carries slow, unmyelinated C-tactile afferents alongside the fast myelinated fibres. Håkan Olausson and colleagues showed in Nature Neuroscience in 2002 that activating them produced a faint pleasant sensation and lit up the insular region rather than primary somatosensory cortex. Line Löken, Johan Wessberg, India Morrison, Francis McGlone and Olausson followed in Nature Neuroscience in 2009: these fibres fired most vigorously at brushing velocities of one to ten centimetres per second, and those were precisely the velocities subjects rated most pleasant. Slowness in good bodywork is not an aesthetic affectation. It sits at the response peak of a specific population of nerve fibres.

Interoception — the sense of the body’s own internal condition, mapped by A. D. Craig in Nature Reviews Neuroscience in 2002 and routed through the insula — supplies the rest of the account. Most people spend most of a working day with no representation whatsoever of their own upper back. An hour of specific, slow, sustained input to that region changes what is available to notice about it. That is a real change in perception. It requires no collagen to have been lengthened.

Two further strands are worth naming with their limits attached. Local biology: Lisa Berrueta, Helene Langevin and colleagues reported in the Journal of Cellular Physiology in 2016 that stretching reduced inflammatory lesion thickness and neutrophil count and raised resolvin D1 concentrations — in rats, with chemically induced inflammation, under a stretching protocol, not in humans receiving massage. Fluid dynamics: Schleip and colleagues, in the Journal of Bodywork and Movement Therapies in 2012, found that isometric stretching of mouse and porcine fascia produced a temporary increase in stiffness after rest, driven by a super-compensation in matrix hydration rather than by cellular contraction. Stretching a fascia, in that preparation, made it stiffer for a while. The tissue is more interesting than the slogan.

"It releases toxins." This claim names no substance. Searching the biomedical literature for a study that defines, measures or tests a toxin released by massage returns nothing usable. Where practitioners do name something, it is usually lactic acid — and the one direct test points the other way. Wiltshire and colleagues, in Medicine & Science in Sports & Exercise in 2010, measured forearm blood flow and venous lactate after strenuous handgrip exercise in twelve subjects under three conditions, and found that massage reduced blood flow and lactate efflux compared with simply resting. Their title says it: massage impairs postexercise muscle blood flow and lactic acid removal. That is one small study with a specific protocol, and it does not make massage harmful. It does mean the toxin claim has no support and at least one contrary result.

"It breaks up adhesions." Adhesions are real. Dense fibrous bands do form after surgery, injury and inflammation, and they can be serious enough to require surgical division. Whether hands can lyse them in a living body has not been shown. The trials that exist come substantially from the developers of the techniques concerned, are small, and are not blinded. There is a more careful version of this idea in the research literature and it is worth knowing: Piero Pavan, Antonio Stecco, Robert Stern and Carla Stecco, in Current Pain and Headache Reports in 2014, distinguish densification — a change in the viscosity of the loose connective tissue between fascial layers, which they propose is comparatively easily reversible — from fibrosis, a change in the fibrous layers themselves following trauma, surgery, diabetes or ageing. That is a proposed distinction in a review paper. It is not a demonstration that hands reverse densification, and it should not be quoted as though it were.

"It realigns the body" or "corrects posture." The Chaudhry figures above are the relevant physics, and this page makes no such claim on behalf of anything. A fourth claim deserves a mention because it shows the shape of the whole problem: massage is routinely said to work by lowering cortisol. Christopher Moyer and colleagues, in the Journal of Bodywork and Movement Therapies in 2011, ran a meta-analysis and found between-groups effect sizes on cortisol almost all small, between 0.05 and 0.30, and statistically indistinguishable from zero. Their conclusion is the exact shape of the fascia problem: the cortisol mechanism "cannot be the cause of MT’s well-established and statistically larger beneficial effects on anxiety, depression, and pain." The effects were better established than the explanation given for them. That is not a reason to stop. It is a reason to stop explaining.

Two systematic reviews of randomised trials of myofascial release reached opposite verdicts, and naming them side by side is more instructive than picking one. M. S. Ajimsha, Noora Al-Mudahka and J. A. Al-Madzhar, in the Journal of Bodywork and Movement Therapies in 2015, concluded that the literature was mixed in quality and results but "encouraging", and that myofascial release "is emerging as a strategy with a solid evidence base and tremendous potential." Katri Laimi, Mikhail Saltychev and colleagues at Turku University Hospital, in Clinical Rehabilitation in 2018, screened 513 records, found eight relevant randomised trials covering 457 participants across lateral epicondylitis, fibromyalgia, low back pain and heel pain, judged risk of bias low in three and high in five, found that effect sizes did not reach the minimal clinically important difference in the low back pain or fibromyalgia trials, and concluded that "current evidence on myofascial release therapy is not sufficient to warrant this treatment in chronic musculoskeletal pain."

One asymmetry is worth stating. Ajimsha has published randomised trials of myofascial release himself — in lateral epicondylitis in 2011 and in plantar heel pain in 2014 — and writes from the Myofascial Therapy and Research Foundation. Those are two of the conditions Laimi’s review covers. That does not make his reading wrong; investigators are often the people who know a literature best. It does mean a reader should know which of the two reviews was written by someone with work inside the evidence base being judged.

Manual therapy is also unusually hard to test. Puhl, Reinhart, Doan and Vernon reviewed 25 placebo-controlled trials of lumbar and pelvic spinal manipulation in The Spine Journal in 2017 and reported that imperfect placebos are ubiquitous: the commonest sham was a manipulation setup with no thrust, exactly one small pilot used a genuinely indistinguishable placebo, only eight trials reported on whether blinding had succeeded, and risk of bias was high or unclear in every included study. The problem is structural rather than careless. The hands are the intervention, the recipient can feel what is being done to her, and expectation, warmth and unhurried attention are delivered along with the pressure whether or not anyone intends them to be.

And the symmetrical caution, which this page is as obliged to state as the negative one. None of the above shows that bodywork is useless. "No good evidence for the proposed mechanism" is a statement about explanations. "No good evidence of effect" is a statement about outcomes, and it is exactly what small, short-follow-up, hard-to-blind trials produce whether or not an effect exists. Absence of evidence in an underpowered literature is weak evidence of absence. One more example of how narrow real findings are: Jan Wilke and colleagues, in Archives of Physical Medicine and Rehabilitation in 2016, reviewed 62 cadaveric dissection studies against the six myofascial meridians proposed by Thomas Myers and found strong anatomical evidence for three of them, partial evidence for two, and none at all for the superficial front line — while stating that whether these continuities matter functionally is "the most urgent task of future research." Anatomical continuity is not therapeutic transmission, and the reviewers said so themselves.

Nobody in this argument disputes the experience. The shoulder line does feel different afterwards. The jaw does unclench. Breathing does drop lower in the body and stay there for a while. What is unsupported is a specific narrative about why — that a sheet of collagen was short and is now longer, that something stuck came unstuck, that a blockage was cleared. Withdrawing that narrative does not withdraw the experience. It leaves the experience standing without a story bolted to it.

The better candidate explanations are, for once, more interesting than the marketing one. Motor tone changes. Nociceptive input into the region changes. Autonomic state shifts. And attention relocates into a part of the body that had not been represented in awareness for several weeks. That last one is not a consolation prize. It is arguably the main event, and it is the only one of the four that a person can verify from the inside while it is happening.

The context is part of the intervention too: warmth, quiet, an hour with no demands in it, and being touched by someone who is paying attention. Clinical trials are obliged to treat all of that as noise to be subtracted, because their question is whether a specific technique adds anything on top of it. A person lying on a table is under no such obligation. She is receiving the whole thing, and the whole thing is what changes how she feels. Trials and afternoons are asking different questions, and it is a category error to let one answer the other.

Read literally, "the slow release used in fascia work" on the Head, Face & Deep Rest page means this: broad, sustained pressure held longer than a stroke, at a depth chosen with you, applied slowly to regions where that pace is comfortable — the shoulder line, the base of the skull, the neck and arms. On the Body Reset page, pressure "worked into the fascia and the deeper layers" means the practitioner is working below the surface rather than gliding across it. Those are descriptions of technique and pace. They are not descriptions of a mechanism, and nothing on this site should be read as claiming one.

What it explicitly does not mean: nothing is being torn, melted, broken up or realigned. No adhesion is being lysed. No toxin is leaving. No posture is being corrected. No condition is being treated, and none is being diagnosed — nobody here is qualified to do either, and the word fascia on a menu does not change that.

In Japan this matters legally as well as editorially. Anma, massage and shiatsu — あん摩マッサージ指圧 — is a licensed occupation under Act No. 217 of 1947, and the word massage is therefore not a neutral description here. Moonlight’s bodywork is offered as relaxation and personal care. It is not offered as 治療, not as licensed anma-massage-shiatsu practice, and not as a substitute for seeing someone qualified. That boundary is not a disclaimer appended to the bottom of a page. It is the reason this page was written the way it was.

And the thing worth saying last: knowing that the mechanism story is weaker than advertised does not make the hour less worth having. It makes what is on offer legible. Time, attention, skill, pressure applied slowly and unhurriedly, and a practitioner who will not tell you an impressive story about your own connective tissue. A house that sells fascia work and publishes the Chaudhry numbers is, on balance, a safer house to be touched by than one that does not.

Sustained firm pressure is not appropriate for everyone, and pain is not a measure of benefit. The belief that deeper hurts more and therefore works better has nothing behind it, and bracing against pressure tends to raise muscle tone rather than lower it — which is the opposite of what the hour is for. If pressure is too much, say so. It will change immediately, and saying so is not an interruption of the work; it is part of it.

There are circumstances in which firm bodywork is better discussed with a doctor before it is booked with anybody, this house included. Stated as general information and not as a description of any reader: anticoagulant medication or a bleeding disorder; recent surgery or fracture; an active skin infection, rash or open wound in the area; acute inflammation or swelling; known osteoporosis; a history of deep vein thrombosis; pregnancy; and any pain that is new, severe, unexplained, spreading, or that wakes you at night. Nobody here can tell you whether any of these apply to you. That is precisely the point of listing them rather than assessing them.

If what you are actually looking for is treatment of a diagnosed condition, or an answer to a symptom that worries you, the right person is a licensed clinician and not this house. That is not modesty and it is not a legal formula. It follows from everything above: a practice whose own founders published the arithmetic showing its standard explanation cannot be right has no business positioning itself as the place you go to get something fixed.