Journal of Bodywork & Movement Therapies 23 (2019) 74e81

Contents lists available at ScienceDirect

Journal of Bodywork & Movement Therapies

journal homepage: www.elsevier.com/jbmt

MYOFASCIAL PAIN AND TREATMENT: RANDOMIZED CONTROLLED TRIAL

Immediate biomechanical, systemic, and interoceptive effects of

myofascial release on the thoracic spine: A randomised controlled trial
Ellie Cathcart, Terence McSweeney, Ross Johnston, Hayley Young, Darren J. Edwards*
College of Human & Health Sciences, Swansea University, Singleton Park, Swansea, Wales, SA2 8PP, UK

a r t i c l e i n f o a b s t r a c t

Article history: Background: Myofascial release (MFR) is used to restore tissue extensibility of the fascia tissue and is
Received 20 January 2018 considered to be useful in a number of clinical settings such as low back pain, ankle injuries, fibromy-
Received in revised form algia, and headaches. There is, however, despite the popularity of MFR in manual therapy, little
1 August 2018
consensus on whether it leads to biomechanical, systemic or interoceptive outcomes.
Accepted 25 August 2018
Aims: This study aimed to explore the immediate biomechanical (increased elasticity for increased range
of motion), systemic (local vs. distal areas of pain threshold) and bodily awareness effects (interoception)
Keywords:
of a myofascial release technique on the thoracic spine.
Myofascial release
Thoracic spine
Method: Twelve healthy participants took part in this triple-bind, repeated measures, cross-over design
ROM study, and were randomised into counterbalanced sequences of three conditions; a control, a sham, and
PPT the MFR condition. The outcome measures used were; range of motion (ROM), pain pressure thresholds
Interoception (PPT), and interoceptive sensitivity (IS) to assess biomechanical, systemic, and interoceptive effects of
MFR.
Results: There were significant increases in ROM and PPT (both local and distal) post MFR intervention.
There was also a positive correlation between baseline interoceptive sensitivity and post-MFR ROM and a
negative correlation for baseline interoceptive sensitivity and post-MRF PPT. Interoceptive sensitivity did
increase post-MFR but this was non-significant.
Conclusions: The increase in ROM suggests that the MFR may have caused a biomechanical change in
tissue elasticity creating an increase in tissue flexibility. The increase in both local and distal sites of the
PPT suggest an overall systemic response to the therapy. The correlation between baseline IS and post-
MFR ROM and PPT suggest that IS may be usefully applied as a predictor for ROM and PPT post-MFR.
© 2018 Elsevier Ltd. All rights reserved.

1. Introduction aetiology of more systemic symptoms such as chronic fatigue

(Mastaglia, 2012).
Fascial tissue includes the loose areolar tissue of the superficial In terms of epidemiology, some studies reveal that myofascial
fascia and deeper layers such as the epimysium layer of the muscles pain occurs in 37% and 65% of middle-aged men and women
and those which envelopes the nerves, blood, and lymphatic ves- respectively (Drewes and Jennum, 1995) and up to 85% in the older
sels (Drake et al., 2009). In a more all-encompassing definition, it is population (Podichetty et al., 2003). It has also been estimated to
described as the soft tissue component of the connective tissue occur in an astounding 85% of chronic pain patients and is the
system that permeates the human body (Huijing et al., 2009). foremost diagnosis in musculoskeletal pain patients reported in
Facia is considered to be a source of nociceptive pain (myofascial general practices (Skootsky et al., 1989). From these demographics,
pain) in several musculoskeletal disorders including plantar fasci- it is evident that this myofascial pain creates a significant burden
itis, Dupuytren's contracture, and non-specific low back pain for the medical system.
(Mense et al., 2016). When chronic, it has also been associated with Numerous therapies have been employed in the treatment of
the deteriorating health of a patient and is implicated in the myofascial pain such as varying forms of myofascial release (MFR)
which have been based upon Rolf's structural integration model
and developed by Stecco over the last 30 years (Stecco, 2004). In
* Corresponding author.
this model and similarity with most forms of MFR techniques, the
E-mail address: d.j.edwards@swansea.ac.uk (D.J. Edwards). duration of the stroke or technique on a particular area is usually

https://doi.org/10.1016/j.jbmt.2018.10.006
1360-8592/© 2018 Elsevier Ltd. All rights reserved.
 E. Cathcart et al. / Journal of Bodywork & Movement Therapies 23 (2019) 74e81 75

based upon the palpable changes felt underneath the practitioner's in a sensation categorization task (Peterson et al., 2014). In terms of
hands, but generally, this lasts between 120 and 300 s (Adigozali pain, specifically, Pollatos et al. (2012) observed that individuals
et al., 2016). MFR involves a manual application of low amplitude, with higher interoceptive sensitivity (IS) had lower pain thresholds
long duration stretches to the fascia and muscles which will rarely and tolerance, higher pain perceptual experience and higher levels
involve the manipulation of one area for more than 2 min (Schleip, of anxiety. As such, several models of cognition suggest that pain is
2003). modulated based on emotion, attention, and memory of previous
The purpose of MFR is to restore tissue extensibility to con- experience leading to anticipation (Stoeter et al., 2007; Melzack,
nective tissue which has undergone changes to its mechanical 1999).
properties such as loss of normal pliability and viscosity (Barnes, Interoception has been used in general pain threshold studies,
1997). In addition to this, the same authors suggest that MFR in but not specifically in a case where a MFR technique is used. So, the
used to affect putative changes in local inflammatory mediator third hypothesis is that baseline-IS will be negatively correlated
proliferation (via drainage) and mechanical pressure on nerves and with PPT as it has been shown to reduce PPT in the study by Pollatos
circulatory vessels. The efficacy of MFR has been demonstrated in a et al. (2012). It is also hypothesised that there will be a correlation
multitude of conditions including low back pain, ankle injuries, between baseline-IS and ROM, but the direction is unspecified
carpal tunnel syndrome, chronic asthma, headaches, and fibromy- given the lack of specific evidence supporting one direction or the
algia (Tozzi, 2012). However, despite the clinical usefulness of MFR other, so this would be two-tailed.
there is little consensus on what i t does at the biomechanical (e.g., In summary, this study aimed to explore the biomechanical,
cellular elasticity, neuronal), systemic (local vs. distal effects), and systemic, and interoceptive effects of the MFR technique, and
the bodily awareness level (e.g., interoceptive sensitivity). There presents a comprehensive battery of outcome measures that
have been some developments towards a comprehensive model of measure components of the biomechanical (ROM for an elastic
how different biomechanical, cognitive and autonomic nervous effect and PPT for a hypoalgestic effect), systemic (local vs. distal),
system (ANS) pathways interact within a manual therapy context and bodily awareness (interoception) before and after the MFR. It is
(see Bialosky et al., 2009), however, as identified by the authors, anticipated that there will be an increase in ROM and PPT (across
this is by no means complete, and it is not specific to any one type of local and distal areas) after MFR, as well as baseline-IS correlating
manual therapy. with ROM and PPT post-MFR outcomes.
As there is a clear gap in the literature regarding this, this pre-
sent study aims to explore the immediate biomechanical (increased
2. Methods
elasticity), systemic (local vs. distal), and bodily awareness effects
(interoception) of a myofascial release technique on the thoracic
2.1. Participants
spine, to help develop the empirical knowledge in this area further.
From the literature, the biomechanical component refers to the
The recruitment involved a purposive sample of 12 asymp-
plasticity, elasticity, and viscosity of the tissue itself. It has been
tomatic first year Swansea University osteopathic students who
suggested that MFR may cause a change in fascia as it causes an
were invited to participate. The purposive sample of first-year
alteration in the density, tonus, viscosity, and arrangement of fascia
students were recruited as they were more naive to the active in-
via mechanical pressure (Rolf, 1977; Schleip, 2003). Rolf claimed
terventions than the more experienced students of later years (see
that as the ground substance of fascia is a colloidal substance, this
CONSORT flow diagram, Fig. 1). The inclusion criteria for this study
allows it to convert from its dense ‘gel’ state to a more fluid ‘sol’
involved being of the ages between 18 and 55, and female or male.
state (Rolf, 1977). From this, the first hypothesis of the present
The exclusion criteria consisted of any systemic disease and long-
study is that MFR may increase a more fluid state in the facia and
term medications that could alter perceptions of pain, a recent or
therefore increase range of motion (ROM), more so than a sham or
long-term spinal musculoskeletal injury/pathology and any
control.
vigorous exercise or manual therapy two days prior to the study.
From the systemic perspective, MFR not only has a local
neurological response but other systemic responses may be trig-
gered via autonomic reflexes. When stimulated, Ruffini corpuscles 2.2. Research design
(mechanosensitive nerves) have been associated with a decrease in
activity of the sympathetic nervous system of the autonomic ner- The experimental design conducted was a triple-blind, rando-
vous system (ANS), as fascia has high density of free nerve endings mised, sham-controlled, within subjects, crossover study design.
that belong to the sympathetic nervous system (Schleip, 2003).
Likewise, stimulation of the sensory mechanoreceptors has been
2.3. Ethical approval
identified as a cause for the activating the anterior lobe of the hy-
pothalamus, which induces a global overall decrease of sympa-
Ethical approval was obtained through Swansea University
thetic muscle tonus and emotional arousal, as well as a change in
College of Human and Health Science.
local tissue viscosity (Gellhorn, 1967). As there is a reported global
as well as local impact in terms of mechanosensitivity, pain pres-
sure thresholds (PPT) may increase in other areas outside of the 2.4. Examiner repeatability
area the MFR is conducted. This, therefore, is the second hypothesis
of this study where it is predicted that MFR will lead to a local and Interclass correlation coefficients (ICC) were conducted to
systemic increase in PPT and more so than control and sham ensure that there was a high level of examiner reliability. Multiple
conditions. measures were taken at the same time and location to ensure this
The bodily awareness component explored in this present study as described by Fless (1987). The classification system used was as
is that of interoceptive pathways (Craig, 2004). Interoception refers suggested by Shrout and Fleiss (1979), where: >0.75 was deter-
to a set of neuro-anatomical pathways which allow bodily signals to mined as excellent; 0.6e0.75 as good; 0.4e0.59 as fair; and <0.04 as
travel through, to form bodily awareness (Craig, 2004; Garfinkel poor. Fixed raters and random participants were utilised in an
and Critchley, 2013; Garfinkel et al., 2015). Interoception has a analysis of variance using a two-way mixed model for the ROM and
strong impact on cognition and has been shown to alter cognition PPT measures.
76 E. Cathcart et al. / Journal of Bodywork & Movement Therapies 23 (2019) 74e81

Fig. 1. Consort Flow Diagram with three groups and with immediate effects recorded.

2.5. Internal validity and excellent intra-examiner reliability (ICC ¼ 0.84) (Antonaci
et al., 1998).
2.5.1. Blinding
The tentative use of the term ‘triple-blinding’ was used, as (1) 2.6.2. Range of motion via an inclinometer
the participants were blind to the condition, (2) the researcher The Acumar dual digital inclinometer was used to measure
taking measurements (ROM, PPT, and interoception) was blind to ROM. In studies, digital inclinometers are regularly used and have
the condition, and (3) the osteopath delivering the MFR technique been acknowledged as reliable and easy to use. Examples of this
was blind to all measurements. include high inter-rater reliability when measuring movement in
the scapula (ICC>0.892) (Tucker and Ingram, 2012) and thoracic
2.5.2. Randomisation spine (ICC>0.89e0.99) (Lin and Wang, 2015). These are highly
Before the experiment began, participants were assigned to reliable, with even smartphone inclinometer apps having an
their intervention group sequence which was selected on the basis excellent reliability (ICC>0.75) (Charlton et al., 2015).
of six possible sequence combinations in order to balance any order Inclinometers are instruments used to measure the ROM of a
effects; [1, 2, 3]; [1, 3, 2]; [2, 1, 3]; [2, 3, 1]; [3, 1, 2]; [3, 2, 1]. To do this, joint with respect to a particular level or angle and has been pre-
the second researcher randomly allocated each of the participants viously used in several trials assessing cervical, thoracic, lumbar
into one of the six sequences via a computer randomised number spine and median nerve ROM where all authors concluded mod-
generator (e.g., sequence [1, 2, 3] equaled 1) (Urbaniack and Plous, erate to good reliability (ICC ¼ 0.6e0.9) (MacDermid et al., 2015;
2013). MacDermid et al., 2014; Prushansky et al., 2010; Whelan et al.,
2017).
2.6. Materials and dependent measures
2.6.3. Interoceptive sensitivity (IS) via biopac
2.6.1. Pain pressure threshold via an algometer IS was measured through an electrocardiogram (ECG) analysis
PPT measurements were assessed using a digital algometer BioPac which has been used in other studies (e.g., Buttagat, et al.,
(Wagner Force Ten FDX). The algometer was calibrated by the 2008). The current study used the BioPac MP160 version.
manufacturer and a 1 cm2 rubber tip was used for the pressure Interoceptive sensitivity is commonly quantified by measuring a
application. The gauge displays values in kg/cm2 ranging from 0 to person's ability to perceive and accurately report one's heartbeats
5.5 kg/cm2. at rest. Differences in IS are related to differences in pain perception
Pressure algometry is designed to record the smallest mea- (Pollatos et al., 2012). There are a few heartbeat mental-tracking
surement of mechanical stimuli that can be perceived as pain tasks that have been used to index the IS of a participant. One
(Fischer, 1987). Pressure algometry is frequently utilised to quantify way to gain the perceptual accuracy score, and what was used in
whether there are any alterations in the participants’ pain the present study, is to get participants to verbally estimate their
perception following a treatment intervention (McCoss et al., 2017). heartbeats without an exteroceptive aid using intervals of 30, 35,
There is much literature supporting the algometer as a reliable, 40, and 45, 50, 55 and 60 s. These time intervals were separated by
valid and easy method of measuring the hypoalgestic effects of 30 s resting periods, and the participant's verbal estimate was then
manual therapy (ICC ¼ 0.78e0.93) (Ylinen et al., 2007). Other pa- compared to the recorded score via a heart rate monitor (Schandry,
pers have also reported good inter-examiner reliability (ICC ¼ 0.75) 1981; Ehlers and Breuer, 1996). This method is a widely used
 E. Cathcart et al. / Journal of Bodywork & Movement Therapies 23 (2019) 74e81 77

measure for IS, has a good test-retest reliability (Pollatos et al., along the thoracic erector spinae between T6-T12 was identified for
2009) and it has also emerged as the dominant method for the intervention and PPT check-points were marked with a felt-tip
testing IS (Schandry, 1981; Critchley et al., 2004). The participants marker. These check-points were bilateral and marked on the
were not informed about the length of the counting phases nor the tibialis anterior muscle belly half-way up the shin, the thoracic
quality of their performance. The following transformation was erector spinae of the back at the level T10, and the cervical erector
used to calculate IS, as used in other studies: spinae of the neck at the level C7. The inclinometer check-points
jn beats n beats j
et al., 2014).
1  ðn beats actualþn beats perceivedÞ=2 (Mallorquí-Bague were marked on the spinous process of T6 and S2. The data mea-
actual percived
surements were then taken, starting with ECG (interoception) then
2.7. Experimental intervention the inclinometer (ROM) and finally the algometer (PPT).
Once all of the pre-intervention data had been recorded, the first
MFR was applied to the thoracic erector spinae muscles between researcher left the room and the second researcher (a qualified
the levels of T6-T12 for 120 s. MFR involves a manual application of osteopath with over ten years of clinical experience) entered the
low amplitude, long duration stretches to the fascia and muscles, room and performed one of the three randomly allocated in-
usually between 120 and 300 s (Ajimsha et al., 2015). terventions (control, sham, or experimental). The second
researcher then exited the room and the first researcher returned
2.8. Sham intervention to the room, not knowing which intervention was conducted, to re-
measure the same measures and in the same order (interoception,
This was a purposefully disengaged balanced ligamentous ten- ROM, and PPT). The participants were subjected to all three con-
sion technique unilaterally on the ribcage. This involved the prac- ditions in this repeated measure crossover design with a one-week
titioner placing his hands directly on the ribs and resting them period washout period in-between conditions.
there for 2 min (see Fig. 2).
2.11. Data analysis
2.9. Control
A Shapiro-Wilk test was used to confirm that the data were
For this condition, the participants lay supine on the plinth with normally distributed (p > 0.05), thus justifying the use of para-
their head on a pillow for 2 min (see Fig. 2). metric tests. A general linear model, consisting of a repeated-
measures univariate Analysis of Variance (ANOVA) was used
2.10. Procedure comparing the control, sham, and experimental conditions for PPT,
ROM, and IS. The dependent measure (DV) used the change score,
The study began with a brief clinical evaluation performed on i.e., pre-subtracted by post PPT, ROM, and IS scores. In addition to
the participants by the first researcher (the final year student this, comparisons were made between pre and post PPT, ROM, and
osteopath) to confirm no evident pathology in the spine. An area IS measures for all three conditions, using paired samples t-tests.

Fig. 2. Top left, lying supine (control); top right, not active touch (sham); bottom, the myofascial release technique.
78 E. Cathcart et al. / Journal of Bodywork & Movement Therapies 23 (2019) 74e81

Finally, a series of bivariate correlations were conducted correlating t(11) ¼ 1.324, p ¼ 0.213, Cohen's d ¼ 0.139 (small effect size) for pre-
baseline IS with post condition ROM and PPT scores. post sham; t(11) ¼ 1.802, p ¼ 0.09, Cohen's d ¼ 0.17 (small effect
size) for pre-post MFR.
3. Results
3.4.3. PPT left back
3.1. Demographic results Change-PPT left back revealed; F(2) ¼ 0.534, p < 0.001, h2p
¼ 0.534 (large effect size). Pairwise comparisons again revealed
Table 1 shows the demographical data for age, height, weight, that the MFR was significant when compared against the control
and body mass index. As these were the same individuals tested and sham conditions, MFR vs control p < 0.01 (CI ¼ 0.399, 0.808);
over the three condition (repeated measures, crossover design) MFR vs sham p < 0.01 (CI ¼ 0.281, 0.931); control vs. sham,
homogeneity tests were not needed. p ¼ 0.831 (CI ¼ 0.295, 0.360). Pairwise t-tests were conducted for
base and post conditions with the following results; t(11) ¼ 0.747,
3.2. ICC results p ¼ 0.471, Cohen's d ¼ 0.04 (small effect size) for pre-post control;
t(11) ¼ 0.771, p ¼ 0.457, Cohen's d ¼ 0.059 (small effect size) for pre-
The intra-class reliability measures for the PPT scores were post sham; t(11) ¼ 6.084, p < 0.01, Cohen's d ¼ 0.388 (medium
excellent and therefore reliable (see Table 2). effect size) for pre-post MFR.

3.3. Change-scores for ROM, PPT and IS 3.4.4. PPT right back
Change-PPT right back revealed; F(2) ¼ 25.487, p < 0.001, h2p
The difference (or change) scores for ROM, PPT, and IS were ¼ 0.699 (large effect). Pairwise comparisons again revealed that the
calculated as post subtracted by pre-scores (see Tables 3e5). From MFR was significant when compared against the control and sham
this, it is interesting to see that for ROM and PPT scores, in most conditions; MFR vs control p < 0.01 (CI ¼ 0.465, 0.929); MFR vs
situations post-MFR Change-scores were positive whilst the control sham p < 0.01 (CI ¼ 0.357, 0.853); control vs. sham, p ¼ 0.378
and sham scores were negative. This indicates that the MFR inter- (CI ¼ 0.311, 0.128). Pairwise t-tests were conducted for base and
vention had a positive effect on PPT and ROM. For IS, all three post conditions with the following results; t(11) ¼ 2.842, p < 0.01,
conditions seemed to increase and more so for MFR which suggests Cohen's d ¼ 0.127 (small effect size) for pre-post control;
that the intervention makes individuals more interoceptively t(11) ¼ 0.921, p ¼ 0.377, Cohen's d ¼ 0.056 (small effect size) for pre-
sensitive. post sham; t(11) ¼ 5.063, p < 0.01, Cohen's d ¼ 0.380 (medium
effect size) for pre-post MFR.
3.4. Inferential statistics
3.4.5. PPT left neck
3.4.1. PPT left leg Change-PPT left neck revealed; F(2) ¼ 5.204, p < 0.01, h2p ¼ 0.321
For the differences scores of PTT, Change-PPT for the left leg (large effect). Pairwise comparisons again revealed that the MFR
revealed; F(2) ¼ 12.398, p < 0.001, h2p ¼ 0.530 (large effect size). was significant when compared against the control and sham
Pairwise comparisons revealed that the MFR was significant when conditions; MFR vs control, p < 0.05 (CI ¼ 0.075, 0.630); MFR vs
compared against the control and sham conditions; MFR vs control, sham, p < 0.05 (CI ¼ 0.02, 0.516); control vs. sham, p ¼ 0.325
p < 0.01 (CI ¼ 0.293, 0.852); MFR vs sham p < 0.01 (CI ¼ 0.142, (CI ¼ 0.299, 0.109). Pairwise t-tests were conducted for base and
0.582); control vs. sham, p ¼ 0.107 (CI ¼ 0.475, 0.054). Pairwise t- post conditions with the following results; t(11) ¼ 0.237,
tests were conducted for base and post conditions with the p ¼ 0.817, Cohen's d ¼ 0.013 (small effect size) for pre-post control;
following results; t(11) ¼ 2.669, p < 0.05, Cohen's d ¼ 0.23 (medium t(11) ¼ 1.413, p ¼ 0.185, Cohen's d ¼ 0.095 (small effect size) for
effect size) for pre-post control; t(11) ¼ 1.351, p ¼ 2.04, Cohen's pre-post sham; t(11) ¼ 0.367, p < 0.01, Cohen's d ¼ 0.271 (medium
d ¼ 0.117 (small effect size) for pre-post sham; t(11) ¼ 6.791, effect size) for pre-post MFR.
p < 0.01, Cohen's d ¼ 0.24 (medium effect size) for pre-post MFR.
3.4.6. PPT right neck
3.4.2. PPT right leg Change-PPT right neck revealed; F(2) ¼ 10.857, p < 0.01, h2p
Change-PPT right leg revealed; F(2) ¼ 7.655, p < 0.01, h2p ¼ 0.410 ¼ 0.497 (large effect). Pairwise comparisons again revealed that the
(large effect size). Pairwise comparisons again revealed that the MFR was significant when compared against the control and sham
MFR was significant when compared against the control and sham conditions; MFR vs control p < 0.01 (CI ¼ 0.160, 0.477); MFR vs
conditions; MFR vs control p < 0.05 (CI ¼ 0.171, 0.626); MFR vs sham p < 0.01 (CI ¼ 0.149, 0.554); control vs. sham, p ¼ 0.705
sham p < 0.01 (CI ¼ 0.122, 0.524); control vs. sham, p ¼ 0.567 (CI ¼ 0.153, 0.219). Pairwise t-tests were conducted for base and
(CI ¼ 0.355, 0.204). Pairwise t-tests were conducted for base and post conditions with the following results; t(11) ¼ 1.059,
post conditions with the following results; t(11) ¼ 2.390, p < 0.05, p ¼ 0.312, Cohen's d ¼ 0.042 (small effect size) for pre-post control;
Cohen's d ¼ 0.16 (small effect size) for pre-post control; t(11) ¼ 0.286, p ¼ 0.780, Cohen's d ¼ 0.016 (small effect size) for
pre-post sham; t(11) ¼ 5.218, p < 0.001, Cohen's d ¼ 0.288 (me-
dium effect size) for pre-post MFR.
Table 1
Demographic data.
3.5. ROM
Measurement Total Mean SD Range

Means subjects Minimum Maximum For ROM, a repeated measures univariate ANOVA with differ-
ence scores (post minus pre) and with the three conditions (control,
(Years) 12 23.08 7.25 18 41
Height (CM) 12 178.17 11.83 158 198 sham and MFR) demonstrated; (F(2) ¼ 18.969, p < 0.01,h2p ¼ 0.633)
Weight (KG) 12 73.33 14.05 50 105 a significant difference in Change-ROM between the three condi-
BMI 12 22.84 2.02 19.5 26.70 tions (large effect). Pairwise comparisons demonstrated that MFR
SD¼Standard Deviation; Age ¼ years; Weight ¼ kilograms; Height ¼ Centimetres; was significant when compared against sham and control; MFR vs
BMI¼ Body Mass Index. Male (N ¼ 7), Female (N ¼ 5). Total N ¼ 12. control p < 0.01 (CI ¼ 3.98, 9.02), MFR vs sham p < 0.01 (CI ¼ 3.55,
 E. Cathcart et al. / Journal of Bodywork & Movement Therapies 23 (2019) 74e81 79

Table 2
Intra-rater reliability PPT.

Interclass 95% Confidence interval Level of reliability p

Correlation Lower Bound Upper Bound

Pre-intervention
Cervical spine 0.986 0.951 0.996 Excellent <0.001
Pre-intervention
Back 0.986 0.944 0.996 Excellent <0.001
Pre-intervention
Leg 0.986 0.956 0.997 Excellent <0.001

Note: Shrout and Fleiss (1979) classification reliability>0.75, excellent; 0.6e0.75, good; 0.4e0.59, fair; and <0.4, poor.

Table 3 3.6. Interoceptive sensitivity (IS)

Mean, standard deviation (SD) and standard error (SE) of PPT change scores for each
condition. Positive numbers indicate an increase in the measure post condition.
A repeated measures univariate ANOVA with difference scores
Study Condition N Mean SD SE Range (min-max) (post minus pre) and with the three conditions (control, sham and
ConChgePPTLL 12 0.32 0.42 0.12 1.25 to 0.11 MFR), was used to compare the differences in change of IS,
ShamChgePPTLL 12 0.11 0.29 0.08 0.55 to 0.28 demonstrating; (F(2) ¼ 0.413, p ¼ 0.66,h2p ¼ 0.036) a small but non-
MFRChgePPTLL 12 0.24 0.12 0.04 0.04 to 0.42 significant increase in IS for the MFR condition compared to the
ConChgePPTRL 12 0.21 0.31 0.89 0.86 to 0.20
other conditions (small effect). Pairwise comparisons demon-
ShamChgePPTRL 12 0.14 0.36 0.10 0.89 to 0.24
MFRChgePPTRL 12 0.18 0.35 0.10 0.58 to 0.90
strated; MFR vs control, p ¼ 0.5212 (CI ¼ 0.064, 0.119), MFR vs
ConChgePPTLB 12 0.06 0.27 0.08 0.59 to 0.34 sham, p ¼ 0.495 (CI ¼ 0.061, 0.118); control vs. sham, p ¼ 0.957
ShamChgePPTLB 12 0.09 0.41 0.12 0.77 to 0.63 (CI ¼ 0.049, 0.046). Pairwise t-tests were conducted for base and
MFRChgePPTLB 12 0.52 0.29 0.08 0.165 to 1.12 post conditions with the following results, indicating that these
ConChgePPTRB 12 0.18 0.22 0.06 0.620 to 0.09
were non-significant changes for all three conditions;
ShamChgePPTRB 12 0.09 0.33 0.09 0.75 to 0.33
MFRChgePPTRB 12 0.52 0.32 0.09 0.07 to 1.25 t(11) ¼ 0.049, p ¼ 0.962, for pre-post control; t(11) ¼ 0.160,
ConChgePPTLN 12 0.02 0.22 0.06 0.34 to 0.45 p ¼ 0.876, for pre-post sham; t(11) ¼ 0.707, p ¼ 0.494, for pre-post
ShamChgePPTLN 12 0.11 0.27 0.08 0.27 to 0.75 MFR.
MFRChgePPTLN 12 0.37 0.38 0.11 0.08 to 1.25
A series of bivariate correlations were also conducted where
ConChgePPTRN 12 0.05 0.17 0.05 0.24 to 0.33
ShamChgePPTRN 12 0.02 0.23 0.07 0.31 to 0.54
baseline (pre-condition) IS were associated with post ROM mea-
MFRChgePPTRN 12 0.37 0.25 0.07 0.06 to 0.50 sures. The relationship between pre-interoception and post MFR
was significant; r ¼ 0.596, p < 0.05. The relationship between pre-
Note: ConChge ¼ control change; ShamChge ¼ sham change; MFRchge ¼ MFR
change; PPT ¼ pain pressure threshold; LL ¼ left leg; RL ¼ right leg; LN ¼ left neck; interoception and post-sham was non-significant; r ¼ 0.369,
RN ¼ right neck; LB ¼ left back; RB ¼ right back. p ¼ 0.238. The relationship between pre-interoception and post-
control was also non-significant; r ¼ 0.079, p ¼ 0.806.
In addition to this, another series of bivariate correlations were
Table 4 conducted, where baseline (pre-condition) IS were associated with
Mean, standard deviation (SD) and standard error (SE) of ROM change scores for
post PPT measures. The association between baseline (pre-condi-
each condition. Positive numbers indicate an increase in the measure post condition.
tion) IS and post-MFR-cervical spine right was non-significant, but
Study Condition N Mean SD SE Range (min-max) close to a border-line significance and negatively correlated;
ConChgeROM 12 0.75 2.34 .66 5 to 2 r ¼ 0.464, p ¼ 0.065. The relationship between baseline (pre-
ShamChgeROM 12 0.75 2.77 0.79 4 to 2 condition) interoception and post-MFR-cervical spine left was
MFRChgeROM 12 5.72 3.13 0.90 1 to 12 broadline negatively correlated; r ¼ 0.492, p ¼ 0.052. The rela-
Note: ConChge ¼ control change; ShamChge ¼ sham change; MFRchge ¼ MFR tionship between baseline (pre-condition) interoception and post-
change; ROM ¼ range of motion. sham-tibialis anterior, was also of borderline significance and
negatively correlated; r ¼ 0.491, p ¼ 0.053.

Table 5
Mean, standard deviation (SD) and standard error (SE) of IS change scores for each 4. Discussion
condition. Positive numbers indicate an increase in the measure post condition.

Study Condition N Mean SD SE Range (min-max) This study explored the immediate effects of MFR on the
thoracic spine with three areas of interest, that being; (1) Biome-
ConChgeIS 12 68.00 42.91 12.39 15 to 161
ShamChageIS 12 70.08 56.39 16.28 6 to 177
chanical; (2) Systemic; and (3) Interoceptive effects. ROM was
MFRChgeIS 12 118.33 90.77 26.20 11 to 275 recorded for Biomechanical, where it was hypothesised that if MFR
led to greater elasticity of the facia at the biomechanical level, then
Note: ConChge ¼ control change; ShamChge ¼ sham change; MFRchge ¼ MFR
change; ROM ¼ range of motion; IS ¼ interoceptive sensitivity. this would increase ROM. PPT was recorded for the systemic
component where it was hypothesised that if it was shown that
there were increases in PPT in other areas outside of the MFR
application (thoracic spine area) then this was evidence of MFR
9.45); control vs. sham, p ¼ 1.00 (CI ¼ 0.256, 2.556). Pairwise t- having a systemic effect across the body. In this case, PPT were
tests were conducted between base and post conditions with the recorded at the cervical, thoracic, and tibialis anterior sites. Finally,
following results; t(11) ¼ 1.11, p ¼ 0.29, Cohen's d ¼ 0.09 (small ef- it was hypothesised that baseline IS would correlate with post-MFR
fect size), for pre-post control; t(11) ¼ 0.94, p ¼ 0.37, Cohen's ROM (no predicted direction) as well as post-MFR PPT (negatively).
d ¼ 0.06 (small effect size), for pre-post sham; t(11) ¼ 6.35, p < 0.01, Of these, the results showed that there were biomechanical
Cohen's d ¼ 0.55 (large effect size) for pre-post MFR. effects where ROM increased (significantly) for the MFR condition
80 E. Cathcart et al. / Journal of Bodywork & Movement Therapies 23 (2019) 74e81

but not for sham or control conditions. It also showed that there cognition can alter pain perception through categorization studies.
were systemic effects where PPT increased for other areas outside In addition to this many other cognition studies have explored
of the locally applied MFR. Finally, baseline IS positively correlated perceptual biases and contextual bias effects (e.g., Edwards, 2017;
with post-ROM and negatively with PPT. Edwards and Wood, 2016; Edwards et al., 2012a, 2012b), including
For the biomechanical effects, the greater elasticity in the facia, contextual behavioral psychology (Edwards et al., 2017). Future
demonstrated through an increase in ROM, suggests that perhaps studies could utilise some of these theories to better understand
the MFR stimulated the facia in way as described by Schleip (2003), any expectation bias.
leading to change in viscosity and density, and as Rolf (1977) sug- In terms of clinical practice, as these findings demonstrate that
gests, this allows it to transform into a more fluid state. MFR can be effective in increasing both local and distal ROM and
PPT after the MFR also increased both locally and distally, but PPT, this can potentially help practitioners to better understand the
not for the sham and control conditions, indicating that MFR caused efficacy of MFR treatment in restoring tissue extensibility to con-
a systemic effect. This non-local neurological systemic responses nective tissue and returning it to normal pliability and viscosity
may be triggered via autonomic reflexes. For example, when both locally and distally as well as reducing pain. In addition to this,
stimulated, Ruffini corpuscles (mechanoreceptors) have been as baseline IS correlated with both ROM and PPT, this suggests that
associated with a decrease in activity of the sympathetic nervous clinicians may be able to use this as a predictor of treatment out-
system of the ANS, as fascia has a high density of free nerve endings comes in the diagnosis stage. However, these suggestions are made
that belong to the sympathetic nervous system (Schleip, 2003). with caution as clinical trials would need to follow in order to make
Likewise, stimulation of the sensory mechanoreceptors has been concrete implications of these findings to clinical practice.
identified to activate the anterior lobe of the hypothalamus, which There were some limitations to this study, firstly, asymptomatic
induces a global overall decrease of sympathetic muscle tonus and participants were used and not clinical patients, so the degree to
emotional arousal, as well as a change in local tissue viscosity which these results will generalise to a clinical population is un-
(Gellhorn, 1967). So, there may be a connection between this certain until a clinical trial is conducted. Secondly, this was a con-
complex ANS system, reduction in the sympathetic arousal, and an venience sample of only 12 participants, so a larger study with a full
increase in systemic PPT. However, a combination of further brain sample size calculated through a power analysis should follow this
imaging and ECG studies need to be conducted to provide further study to ensure the consistency of these results. Thirdly, it is also
evidence for this interaction specifically. recognised that having several hypotheses does increase the
There was also an IS effect where IS correlated positively with chance of finding a false positive (a type one error). Fourth, in the
ROM and negatively with PPT. This may be due to the stimulation of control and sham conditions the participants were in the supine
the sensory interstitial mechanoreceptors which may have acti- position, whilst in the experimental intervention (MFR) the par-
vated the anterior lobe of the hypothalamus, thus inducing a global ticipants were in the prone position. It is acknowledged that posi-
overall decrease of sympathetic muscle tonus and emotional tion may have influence the results to some degree. The fact that
arousal (Gellhorn, 1967). Therefore, a higher IS may have amplified this was a repeated measures study was not deemed a limitation as
the hypothalamus reaction and further decreased the sympathetic the order of condition sequences were randomised with a one-
response leading to a more relaxed parasympathetic response, thus week washout period to balance any learning effects, but as with
allowing for greater ROM. all cross-over designs it is acknowledged that learning effects do
The connection between mechanosensitive nerves and the ANS occur.
system seem to be extremely complex. For example, within the In summary, this work produced some interesting findings in
interoceptive pathways, the mid insula has ongoing communica- relation to how MFR can affect PPT and ROM outcomes as well as
tion with the amygdala regarding emotional memories and the how IS relates to ROM and PPT. Several accounts were given
stimulus salience, as well as with the hypothalamus in terms of the through the biomechanical, systemic and interoceptive systems.
current state of the ANS (Craig, 2008). The higher IS scores leading This work may help further develop more unified models of oste-
to lower PPT and tolerance, higher pain perceptual experience opathy such as the ‘comprehensive pathway’ model as described by
demonstrated by Pollatos et al. (2012), may explain why there was a Bialosky et al. (2009). Further work could now explore additional
negative correlation between IS and PPT, whilst expectation bias of aspects of the various pathways, with additional cognitive model-
the intervention may have accounted for the positive correlation ling and neuroscience data. One specific aspect for further work
between IS and ROM. For example, pain modulation and effec- could be to explore how other more specific ANS measures such as
tiveness of intervention have been suggested to relate to emotion, heart rate variability is influenced by MFR.
memory, attention, experience, and expectations (e.g., Stoeter et al.,
2007; Melzack, 1999). So, the higher emotional, and attentional Conflicts of interest
sensitivity could have led to an increased expectation bias that the
intervention would cause a positive effect in increasing ROM. The authors report no conflict of interest.
However, again, many more studies are needed to be conducted
utilising psychology to understand the relationship between psy- Acknowledgments
chological bias, osteopathy, and the complex interoceptive
pathway. These psychological biases have been noted in other None.
studies (e.g., Whelan et al., 2017) and were noted again here with
global increases in some places between baseline and post- References
condition scores for control, sham, and MFR. Therefore, it seems
important to understand any placebo effects that might be caused. Adigozali, H., Shadmehr, A., Ebrahimi, E., Rezasoltani, A., Naderi, F., 2016. Reliability
of assessment of upper trapezius morphology, its mechanical properties and
One way of doing this is through the Neuromatrix model (Melzack,
blood flow in female patients with myofascial pain syndrome using ultraso-
1999). This model suggests that sensory experience, perceptions, nography. J. Bodyw. Mov. Ther. 21 (1), 35e40.
and expectations can all influence subjective experience, so these Ajimsha, M.S., Al-Mudahka, N.R., Al-Madzhar, J.A., 2015. Effectiveness of myofascial
may lead to the placebo effect. release: systematic review of randomized controlled trials. J. Bodyw. Mov. Ther.
19 (1), 102e112.
Placebo effect can also be explained by cognitive psychology Antonaci, F., Sand, T., Lucas, G., 1998. Pressure algometry in healthy subjects: inter-
where, for example, Peterson et al. (2014) demonstrated how examiner variability. Scand. J. Rehabil. Med. 30 (1), 3e8.
 E. Cathcart et al. / Journal of Bodywork & Movement Therapies 23 (2019) 74e81 81

Barnes, M., 1997. The basic science of myofascial release: morphologic change in flexion and extension. BMC Muscoskel. Disord. 16 (1), 121.
connective tissue. J. Bodyw. Mov. Ther. 1 (4), 231e238. MacDermid, J.C., Arumugam, V., Vincent, J.I., Carroll, K.L., 2014. The reliability and
Bialosky, J.E., Bishop, M.D., Price, D.D., Robinson, M.E., George, S.Z., 2009. The validity of the computerized double inclinometer in measuring lumbar
mechanisms of manual therapy in the treatment of musculoskeletal pain: a mobility. Open Orthop. J. 8, 355.
comprehensive model. Man. Ther. 14 (5), 531e538. Mallorquí-Bague
, N., Garfinkel, S.N., Engels, M., Eccles, J.A., Pailhez, G., Bulbena, A.,
Buttagat, V., Eungpinichpong, W., Chatchawan, U., Kharmwan, S., 2011. The imme- Critchley, H.D., 2014. Neuroimaging and psychophysiological investigation of
diate effects of traditional Thai massage on heart rate variability and stress- the link between anxiety, enhanced affective reactivity and interoception in
related parameters in patients with back pain associated with myofascial people with joint hypermobility. Front. Psychol. 5, 1162.
trigger points. J. Bodyw. Mov. Ther. 15 (1), 15e23. McCoss, C.A., Johnston, R., Edwards, D.J., Millward, C., 2017. Preliminary evidence of
Charlton, P.C., Mentiplay, B.F., Pua, Y.H., Clark, R.A., 2015. Reliability and concurrent Regional Interdependent Inhibition, using a ‘Diaphragm Release’to specifically
validity of a Smartphone, bubble inclinometer and motion analysis system for induce an immediate hypoalgesic effect in the cervical spine. J. Bodyw. Mov.
measurement of hip joint range of motion. J. Sci. Med. Sport 18 (3), 262e267. Ther. 21 (2), 362e374.
Craig, A.D., 2004. Human feelings: why are some more aware than others? Trends Melzack, R., 1999. From the gate to the neuromatrix. Pain 82, S121eS126.
Cognit. Sci. 8 (6), 239e241. Mastaglia, F., 2012. The relationship between muscle pain and fatigue. Neuro-
Craig, A.D., 2008. Interoception and emotion: a neuroanatomical perspective. muscul. Disord. 22 (3), S178eS180.
Handbook of emotions 3 (602), 272e288. Mense, S., Hoheisel, U., 2016. Evidence for the existence of nociceptors in rat
€
Critchley, H., Wiens, S., Rotshtein, P., Ohman, A., Dolan, R., 2004. Neural systems thoracolumbar fascia. J. Bodyw. Mov. Ther. 20 (3), 623e628.
supporting interoceptive awareness. Nat. Neurosci. 7 (2), 189e195. Petersen, S., Schroijen, M., Moelders, C., Zenker, S., Van den Bergh, O., 2014. Cate-
Drake, R., Vogel, A., Mitchell, A., 2009. Gray's Anatomy for Students. Elsevier, gorical interoception: perceptual organization of sensations from inside. Psy-
Philadelphia: Churchill Livingstone. chol. Sci. 25 (5), 1059e1066.
Drewes, A., Jennum, P., 1995. Epidemiology of myofascial pain, low back pain, Podichetty, V., Mazanec, D., Biscup, R., 2003. Chronic non-malignant musculoskel-
morning stiffness and sleep-related complaints in the general population. etal pain in older adults: clinical issues and opioid intervention. Postgrad. Med.
J. Muscoskel. Pain 3 (1), 68. 79 (937), 627e633.
Edwards, D.J., 2017. Unsupervised categorization with a child sample: category Pollatos, O., Traut-Mattausch, E., Schandry, R., 2009. Differential effects of anxiety
cohesion development. Eur. J. Dev. Psychol. 14 (1), 75e86. and depression on interoceptive accuracy. Depress. Anxiety 26 (2), 167e173.
Edwards, D.J., Pothos, E.M., Perlman, A., 2012a. Relational Vs. absolute representa- Pollatos, O., Füsto€s, J., Critchley, H.D., 2012. On the generalised embodiment of pain:
tion in categorization. Am. J. Psychol. 125 (4), 481e497. how interoceptive sensitivity modulates cutaneous pain perception. Pain 153
Edwards, D.J., Perlman, A., Reed, P., 2012b. Unsupervised Categorization in a sample (8), 1680e1686.
of children with autism spectrum disorders. Res. Dev. Disabil. 33 (4), Prushansky, T., Deryi, O., Jabarreen, B., 2010. Reproducibility and validity of digital
1264e1269. inclinometry for measuring cervical range of motion in normal subjects.
Edwards, D.J., Wood, R., 2016. Unsupervised categorization with individuals diag- Physiother. Res. Int. 15 (1), 42e48.
nosed as having moderate traumatic brain injury: over-selective responding. Rolf, I., 1977. Rolfing: the Integration of Human Structures. Dennis Landman, Santa
Brain Inj. 30 (13e14), 1576e1580. Monica.
Edwards, D.J., McEnteggart, C., Barnes-Holmes, Y., Lowe, R., Evans, N., Vilardaga, R., Schandry, R., 1981. Heart beat perception and emotional experience. Psychophysi-
2017. The impact of mindfulness and perspective-taking on implicit associa- ology 18 (4), 483e488.
tions toward the elderly: a relational frame theory account. Mindfulness 8 (6), Schleip, R., 2003. Fascial plasticity - a new neurobiological explanation: Part 1.
1615e1622. J. Bodyw. Mov. Ther. 7 (1), 11e19.
Ehlers, A., Breuer, P., 1996. How good are patients with panic disorder at perceiving Shrout, P.E., Fleiss, J.L., 1979. Intraclass correlations: uses in assessing rater reli-
their heartbeats? Biol. Psychol. 42 (1), 165e182. ability. Psychol. Bull. 86, 420e428.
Fischer, A.A., 1987. Pressure algometry over normal muscles. Standard values, val- Skootsky, S., Jaeger, B., Oye, R., 1989. Prevalence of myofascial pain in general in-
idity and reproducibility of pressure threshold. Pain 30 (1), 115e126. ternal medicine practice. West. J. Med. 151 (2), 157.
Fleiss, J.L., 1987. The design and analysis of clinical experiments. In: Computers & Stecco, L., 2004. Fascial Manipulation for Musculoskeletal Pain. Piccin, Padova.
Mathematics with Applications, vol. 14. Wiley Classics Library, New York. Stoeter, P., Bauermann, T., Nickel, R., Corluka, L., Gawehn, J., Vucurevic, G., et al.,
Garfinkel, S.N., Critchley, H.D., 2013. Interoception, emotion and brain: new insights 2007. Cerebral activation in patients with somatoform pain disorder exposed to
link internal physiology to social behaviour. Commentary on: “Anterior insular pain and stress: an fMRI study. Neuroimage 36 (2), 418e430.
cortex mediates bodily sensibility and social anxiety” by Terasawa et al.(2012). Tozzi, P., 2012. Selected fascial aspects of osteopathic practice. J. Bodyw. Mov. Ther.
Soc. Cognit. Affect Neurosci. 8 (3), 231e234. 16 (4), 503e519.
Garfinkel, S.N., Seth, A.K., Barrett, A.B., Suzuki, K., Critchley, H.D., 2015. Knowing Tucker, W.S., Ingram, R.L., 2012. Reliability and validity of measuring scapular up-
your own heart: distinguishing interoceptive accuracy from interoceptive ward rotation using an electrical inclinometer. J. Electromyogr. Kinesiol. 22 (3),
awareness. Biol. Psychol. 104, 65e74. 419e423.
Gellhorn, E., 1967. Principles of Autonomicesomatic Integration: Physiological Basis Urbaniak, G.C., Plous, S., 2013. Research Randomizer [Computer software]. Retrieved
and Psychological and Clinical Implications. University of Minnesota Press, on September 22, 2017, from, Version 4.0. http://www.randomizer.org.
Minneapolis, MN. Whelan, G., Johnston, R., Millward, C., Edwards, D.J., 2017. The immediate effect of
Huijing, P., Langevin, H., 2009. Communicating about fascia: history, pitfalls and osteopathic cervical spine mobilization on median nerve mechanosensitivity: a
recommendations. International Journal of Therapeutic Massage and Bodywork triple-blind, randomized, placebo-controlled trial. To be published in Journal of
2 (4), 3e8. Bodywork & Movement Therapies. Preprint] 2018 [cited 8 January 2017].
Lin, H.Y., Wang, T.J., 2015. Reliability of dual digital inclinometer in measuring Available from: http://www.bodyworkmovementtherapies.com/article/S1360-
thoracic kyphosis angle and thoracic mobility in healthy adults. Physiotherapy 8592(17)30109-2/fulltext https://doi.org/10.1016/j.jbmt.2017.05.009.
101, e879ee880. Ylinen, J., Nyka €nen, M., Kautiainen, H., Ha €kkinen, A., 2007. Evaluation of repeat-
MacDermid, J.C., Arumugam, V., Vincent, J.I., Payne, K.L., So, A.K., 2015. Reliability of ability of pressure algometry on the neck muscles for clinical use. Man. Ther. 12
three land marking methods for dual inclinometry measurements of lumbar (2), 192e197.