Mike’s HBOT Testimonial: Post Concussion Syndrome –
Mike had been plagued with sleep issues after multiple concussions. He’s been through multiple sleep studies and his sleep quantity and quality is better than it’s been in years since completing 40 hyperbaric oxygen dives. Better sleep = more energy. As far as his chronic pain goes, Mike says [HBOT] has by far been the best solution for his pain. “You don’t realize how much pain you were in until you aren’t in pain anymore.” Watch Mike’s story here.
In just two months and 40 hyperbaric sessions later my life has changed and I no longer have migraines or pain from Fibromyalgia.
Below is a testimonial from Elena Rosario, who was treated in 2016 at Hampton Roads Hyperbaric Therapy. She is the son of another patient who was successfully treated for mild traumatic brain injury. Her son translated her exact words from Spanish to English so that she could share her experiences with HBOT. Please read about her experience!
My name is Elena Rosario,
I’ve had a rough painful life for the last 13 years, since then I’ve been a very sad and a depressed person unable to enjoy most of my adulthood. I’ve have been suffering with fibromyalgia along with other pains throughout my body and I had to stop working because the pain was so great. People did not understand the pain I was going through and I was humiliated and considered weak. My symptoms tormented me on a daily basis and I was never in a good mood. Words couldn’t describe the level of pain that I was in but the thing that comes close to it is having the feeling of having needles pricked into my muscle fibers. It was very hard to stay positive since I was always under pain with painful migraines that made it difficult to concentrate on anything. My skull was always under pain and I could feel the pressure crawling into my eyes and it affected my vision
129 W. Virginia Beach Blvd, Suite 140, Norfolk, VA 23510 757-452-3934 www.hrhyperbaric.com
greatly. I was heavily medicated and for years I had no success in treating my symptoms. I realized how much the medications have not worked and only made me feel worst since it tried to treat one thing but instead I had more problems because of the side effects I was
having. After all the years of being medicated I had to stop using the medications since my body was in very bad shape and not responding well. In just two months and 40 sessions later my life has changed and I no longer have migraines or in such pain. I remember having a difficult time in just trying to concentrate in simple tasks and now things are much more clearer and brighter. Two months ago I had trouble communicating and writing anything but now I’m able to write this testimonial with no trouble whatsoever. My message to the people is that if they suffer from the same medical issue that I went through is that there is hope and to really consider this treatment because it has helped me so much. For the first time in my life I’m no longer being tormented by the dreaded daily aches and pains that have emotionally destroyed me for years and now I’m just able to have such calmness that I no longer have to worry about having the feeling of needles anymore. I have so much lightness on my skull that my eyesight has improved and I no longer have to worry about wearing my glasses. I was so used being heavily sedated and for the first time I can say is that it feels great not being on any medication and to finally be able to enjoy life pain free. I am thankful for all the help I received and feel blessed to have gone through this treatment since its not available everywhere and not many people know about it.
129 W. Virginia Beach Blvd, Suite 140, Norfolk, VA 23510 757-452-3934 www.hrhyperbaric.com
Data Availability Statement: All relevant data are within the paper and its Supporting Information files.
Funding: The study was supported by the research fund of Assaf-Harofeh Medical Center. EB-J and G-S were supported by a grant from the Tauber Family Funds and the Maguy-Glass Chair in Physics of Complex Systems. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
Shai Efrati1,2,3,4*, Haim Golan3,5, Yair Bechor2, Yifat Faran6, Shir Daphna-Tekoah6,7, Gal Sekler8, Gregori Fishlev2,3, Jacob N. Ablin9,3, Jacob Bergan2,3, Olga Volkov3,5, Mony Friedman2,3, Eshel Ben-Jacob1,4,8,10*, Dan Buskila11
1 Research and Development Unit, Assaf Harofeh Medical Center, Zerifin, Israel, 2 The Institute of Hyperbaric Medicine, Assaf Harofeh Medical Center, Zerifin, Israel, 3 Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel, 4 Sagol School of Neuroscience, Tel-Aviv University, Tel-Aviv, Israel, 5 Nuclear Medicine institute, Assaf Harofeh Medical Center, Zerifin, Israel, 6 School of Social Work, Ashkelon Academic College, Ashkelon, Israel, 7 Social Work Department, Kaplan Medical Center, Rehovot, Israel, 8 School of Physics and Astronomy, The Raymond and Beverly Sackler Faculty of Exact Sciences, Tel-Aviv University, Tel-Aviv, Israel, 9 Institute of Rheumatology, Tel Aviv Sourasky medical center Israel, Tel- Aviv, Israel, 10 Center for Theoretical Biological Physics, Rice University, Houston, Texas, United States of America, 11 Department of Medicine H, Soroka Medical Center, BGU University of the Negev, Beer Sheva, Israel
Fibromyalgia Syndrome (FMS) is a persistent and debilitating disorder estimated to impair the quality of life of 2–4% of the population, with 9:1 female-to-male incidence ratio. FMS is an important representative example of central nervous system sensitization and is associ
ated with abnormal brain activity. Key symptoms include chronic widespread pain, allodynia and diffuse tenderness, along with fatigue and sleep disturbance. The syndrome is still elu sive and refractory. The goal of this study was to evaluate the effect of hyperbaric oxygen therapy (HBOT) on symptoms and brain activity in FMS.
Methods and Findings
A prospective, active control, crossover clinical trial. Patients were randomly assigned to treated and crossover groups: The treated group patients were evaluated at baseline and after HBOT. Patients in the crossover-control group were evaluated three times: baseline, after a control period of no treatment, and after HBOT. Evaluations consisted of physical ex amination, including tender point count and pain threshold, extensive evaluation of quality of life, and single photon emission computed tomography (SPECT) imaging for evaluation of brain activity. The HBOT protocol comprised 40 sessions, 5 days/week, 90 minutes, 100% oxygen at 2ATA. Sixty female patients were included, aged 21–67 years and diag nosed with FMS at least 2 years earlier. HBOT in both groups led to significant amelioration of all FMS symptoms, with significant improvement in life quality. Analysis of SPECT
PLOS ONE | DOI:10.1371/journal.pone.0127012 May 26, 2015 1 / 25
Hyperbaric Oxygen Therapy Can Diminish Fibromyalgia Syndrome
imaging revealed rectification of the abnormal brain activity: decrease of the hyperactivity
mainly in the posterior region and elevation of the reduced activity mainly in frontal areas.
No improvement in any of the parameters was observed following the control period.
Conclusions
The study provides evidence that HBOT can improve the symptoms and life quality of FMS
patients. Moreover, it shows that HBOT can induce neuroplasticity and significantly rectify
abnormal brain activity in pain related areas of FMS patients.
Trial Registration
ClinicalTrials.gov NCT01827683
Introduction
Fibromyalgia Syndrome (FMS) is a persistent and debilitating disorder estimated to impair the
quality of life of 2–4% of the population, with 9:1 female-to-male incidence ratio. FMS is the
second most common disorder, after osteoarthritis, observed by rheumatologists [1]. The de
fining symptoms of FMS include chronic widespread pain, intense pain in response to tactile
pressure (allodynia), prolonged muscle spasms, weakness in the limbs, nerve pain, muscle
twitching, palpitations and diffuse tenderness, along with fatigue, sleep disturbance and cogni
tive impairments. These impairments include problems with short- and long- term memory,
short-term memory consolidation, impaired speed of information processing, reduced atten
tion span and limited multi-tasking performance. FMS is a persistent disorder with symptoms
that have a devastating effect on people’s lives, including limited ability to engage in everyday
activities, limited ability to maintain outside work and difficulties to maintain normal relation
ships with family, friends and employers [2]. These limitations can lead to the occurrence of
anxiety and depression in many FMS patients.
Challenging syndrome
FMS is not completely understood, in part because there is no evidence of a single event that
“causes” fibromyalgia. Rather, many physical and/or emotional stressors may trigger or aggra
vate symptoms. Those have included certain infections, such as a viral illness or Lyme disease,
as well as emotional or physical trauma [3, 4]
Establishing proper diagnostic criteria is also a challenge [5, 6]. The American College of
Rheumatology (ACR) introduced the first fibromyalgia classification in 1990 [7]. Over time,
those criteria invoked both conceptual and practical objections [6]. For example, many physi
cians did not know how to evaluate the tender points [6]. Another reservation had to do with
the fact that important features such as fatigue and cognitive symptoms were not included in
the 1990 criteria. Some questioned the validity of defining fibromyalgia as a unique syndrome
because of the overlap between its symptoms and those of other conditions such as chronic fa
tigue syndrome [8]. To resolve the difficulties associated with the classification and diagnosis
of FMS, Wolfe et al. [6] proposed new, simple practical criteria that do not require tender point
examination and still classify correctly almost 90% of the cases diagnosed by the 1990 ACR
classification criteria.
PLOS ONE | DOI:10.1371/journal.pone.0127012 May 26, 2015 2 / 25
Hyperbaric Oxygen Therapy Can Diminish Fibromyalgia Syndrome
As with many other syndromes, there is no efficient cure for FMS and no agreed upon
treatment – the suggested treatment depends on the classification of choice. Those who regard
FMS as a neurological disorder advocate pharmacotherapy. All current treatments, such as pre
scribed medications, aerobic exercises and cognitive behavioral therapies, consist of symptom
management [1, 9, 10]. Integrated programs based on these treatments have been shown to al
leviate pain and some other symptoms but with limited effectiveness [10].
Association with changes in brain activity
The level of pain sensation is determined by the relevant sensors recording at the location of
the pain and by the processing of that information in the brain. Comparison between SPECT
imaging of FMS patients and healthy subjects revealed elevated activity in the somatosensory
cortex and reduced activity in the frontal, cingulate, medial temporal and cerebellar cortices
[11, 12]. These results are in agreement with earlier studies based on fMRI imaging [13]. Other
fMRI studies found that depressive symptoms were associated with the pain response in areas
of the brain that participate in interpretation and assignment of the pain sensation, but not in
areas involved in sensory processing of the input signal [14]. These findings might indicate
that the amplified pain sensation in FMS patients is largely associated with higher level process
ing of information in the brain. However, there is an ongoing controversy, in which many
rheumatologists take the opposite stand on this issue. As we explain in the discussion, our find
ings—that the pain amelioration in those patients who responded to the HBOT treatments
goes hand-in-hand with changes in brain activity—provide important validation to the idea
that in many of FMS patients the syndrome is associated with abnormal pain processing in the
brain. This is opposed to the stand shared by other rheumatologists, according to which FMS is
a sort of peripheral small fiber inflammation [15]. It is likely that the latter is the cause of FMS
in some patients. However, a claim that it is the only cause stands in contradiction to a wide
body of literature. For example, it fails to explain why FMS appears in many patients following
a traumatic brain injury.
Studies of brain metabolism using single-voxel magnetic resonance spectroscopy (1H-MRS)
found abnormalities within the hippocampal complex in patients with fibromyalgia [16, 17].
Since the hippocampus plays crucial roles in maintenance of cognitive functions, sleep regula
tion and pain perception, it was suggested to associate the hippocampal metabolic dysfunction
with these symptoms in FMS patients.
The evidence suggests that the pain in fibromyalgia results primarily from abnormalities in
pain processing pathways, which may be described as the “volume” of the neurons set too high,
and these hyper-excitability of pain processing pathways and under-activity of inhibitory pain
pathways in the brain result in pain experience in the affected individual. Since some of the
neuro-chemical abnormalities that occur in fibromyalgia can also regulate mood, sleep and en
ergy, it might explain why mood, sleep and fatigue problems are commonly co-morbid with
fibromyalgia.
Looking for a solution – Hyperbaric oxygen therapy (HBOT)
Clearly, new methods should be examined in order to provide sustained relief to FMS patients.
Our study was motivated by the idea that hyperbaric oxygen therapy (HBOT) can rectify ab
normal brain function underlying the symptoms of FMS patients. The hypothesis is based on
new trials demonstrating that HBOT can induce neuroplasticity that leads to repair of chroni
cally impaired brain functions and improved quality of life in post-stroke patients and mild
Traumatic Brain Injury (mTBI) patients with prolonged post concussion syndrome (PCS),
even years after the brain insult [18–20] (see Discussion section for more details). As explained
PLOS ONE | DOI:10.1371/journal.pone.0127012 May 26, 2015 3 / 25
Hyperbaric Oxygen Therapy Can Diminish Fibromyalgia Syndrome
in the discussion it is plausible that increasing oxygen concentration by HBOT can change the
brain metabolism and glial function to rectify the FMS-associated brain abnormal activity. It
has already been demonstrated that exposure to hyperbaric oxygen induces significant anti-in
flammatory effect in different conditions and pathologies [21–24]. As such, it was also demon
strated that repetitive HBOT may attenuate pain by reducing production of glial cells
inflammatory mediators [25, 26].
About a decade ago, Yildiz et al. (2004) [27] found a significant reduction in the number
and threshold of tender points following HBOT. The effect of HBOT was not restricted to
FMS. Similar improvements following HBOT were reported in complex regional pain syn
Regional cerebral blood flow change analysis was conducted by fusing pre- and post-treat
ment studies that were normalized to median brain activity. SPECT images were reoriented
into Talairach space using NeuroGam (Segami Corporation) for identification of Brodmann
cortical areas and in order to compute the mean perfusion in each Brodmann area (BA). All
SPECT analyses were done while blinded to the laboratory and clinical data.
Changes, average changes and normalized average changes. Changes in perfusion in all
Brodmann areas for each subject were determined by calculating the percentage difference be
tween post-period and pre/baseline-period divided by the pre/baseline-period perfusion. The
relative change, Rchange(i,n) of Brodmann area (n) for patient (i), is defined as:
Rchange ið Þ¼ ; n ½PostAði; nÞ PreAði; nÞ
½PreAði; nÞ
Where PostA(i,n) and PreA(i,n) represent the normalized activity of the nth Brodmann area
at the end point and start point of the assessment period (either treatment or control) respec
tively. Note that when multiplied by 100, Rchange(i,n) is the percent difference.
An averaged relative change,
< Rchange > ðnÞ ¼< Rchangeði; nÞ > i
was calculated for each Brodmann area for each group according to study phase (control and
treatment periods of the crossover group and treatment period of the treated group).
Response group. To inspect the association between changes in the brain activity accord
ing to SPECT imaging and changes in the syndrome severity, we divided the 48 patients into
two subgroups according to their response to the treatment. More specifically, we use the
changes in the number of tender points and the level of threshold pressure as classifiers. The 41
patients which exhibited improvements in these parameters were classified as responders
(physiologically improved), and were assigned to a response group. The other 7 patients were
classified as non responders and were assigned to a non response group.
Significance index
Brain activity is signified by variations between the different brain’s locations, and these varia
tions change over time according to the tasks performed. These inherent spatiotemporal varia
tions are reflected by high variance in the brain activity at each Brodmann area, as measured by
SPECT imaging. The statistical challenge imposed by the SPECT imaging is the low signal-to
noise ratio: that the magnitude of the non arbitrary changes in the brain activity (following
treatment) in most of the Brodmann area are comparable to the magnitude of the arbitrary
change related to the inherent person-to-person and time variations that are not related to the
treatment.
To meet the challenge, we introduced a significance index Iσ(n) to substantiate the compari
son between the changes in brain activity in the response group during treatment and those in
PLOS ONE | DOI:10.1371/journal.pone.0127012 May 26, 2015 7 / 25
Hyperbaric Oxygen Therapy Can Diminish Fibromyalgia Syndrome
the crossover group during the control period. We defined Iσ (n)as:
IsðnÞ¼fPCðnÞ ½1 PRðnÞ g1=2
where PC(n) is the p-value of the change in SPECT measurements (calculated in two-tailed t
test) for the post control vs. pre control period of the patients in the crossover group. Similarly,
PR(n) is the p-value of the change in SPECT measurements for the post treatment vs. pre treat
ment period of the patients in the response group (the responders). The rationale for the new
index is that lower values of PR(n), hence higher values of [1-PR(n)], correspond to higher sig
nificance of the changes during treatment. On the other hand, higher values of PC(n) imply
that the changes during control are likely to vary arbitrarily prior to treatment. Hence, consis
tent changes measured during treatment are more significant. The significance index is defined
such that both contributions are included. We tested other putative definitions of the signifi
cance index—for example, {[PC(n)]/[PR(n)]}
1/2 that represents the ratio between the significance of the changes during treatment vs. the
changes during control—and obtained similar results.
Statistical Analysis
SPSS software (version 19, IBM Inc.) was used for the statistical analyses. Continuous data is
expressed as means ± standard deviations (STD). For each dependent measure, an analysis of
variance was performed according to the time-point of data collection (before vs. after HBOT)
and according to the associated group (treated vs. crossover) as independent measures. Addi
tionally, repeated one-way analysis of variance was computed using the three time-points of
data collection for the crossover group. When relevant, post hoc comparisons were used as is
reported in the results section. Categorical data is expressed in numbers and percentages and
compared by chi-square test. With regards to dolorimeter thresholds analysis, an average of
thresholds was calculated for each patient, and this average was used in the ANOVA model.
Sample size was based on the assumption that exposure to the Dolorimeter evaluation (at
baseline) without any additional training might induce up to 8% (0.06 Dolorimeter change)
improvement in the second Dolorimeter evaluation (following HBOT), based on Yildiz et al.
[27]). A threshold of tender sites was selected as a criterion for sample size since this was the
smaller anticipant effect. The sample size was calculated to provide 80% power to show that
HBOT induces at least 87% improvement on Dolorimeter threshold of the tender sites. This
was based on a power analysis using the normal approximation for the binomial, with one
sided Alpha = 0.05. Note that it is based on a cross over design without sequence effect.
Registration
The study was officially registered in ClinicalTrials.gov, Identifier: NCT01827683, after pa
tients enrolment started due to technical delay. The authors confirm that all ongoing and relat
ed trials for HBOT in fibromyalgia are registered.
Results
The study was conducted between May 2010 and December 2012. Sixty female patients signed
a written informed consent. Eight patients were excluded before the hyperbaric oxygen treat
ment and additional four patients were excluded during treatment.
PLOS ONE | DOI:10.1371/journal.pone.0127012 May 26, 2015 8 / 25
Table 1. Demographic of patients’ characteristics.
Hyperbaric Oxygen Therapy Can Diminish Fibromyalgia Syndrome
Treated Group Crossover Group p Value (n = 24) (n = 26)
Age (years) 50.4±10.9 48.1±11.1 0.677 Years of education 17.1±3.5 14.8.±3.0 0.019 Duration of fibromyalgia (years) 6.75±5.9 6.2±5.1 0.735 Number of children 2.38±1.21 2.95±1.43 0.156 Marital status: Married 21 (87.5%) 18 (69.2%) 0.239 Single 1 (4.1%) 5 (19.2%)
Divorce 2 (8.3%) 1 (3.8%)
Widow 0 (0%) 1 (3.8%)
Life style: Secular 19 (79.2%) 17 (65.3%) 0.662 Traditional 4 (16.6%) 6 (23.1%)
Religious 1 (4.1%) 2 (7.6%)
Place of born: Israel 20 (83.3%) 18 (69.2%) 0.297 USSR 0 (0%) 2 (7.6%)
else 4 (8.3%) 6 (23%)
Economic status: Very bad 0 (0%) 1 (3.8%) 0.77 Bad 2 (8.3%) 2 (7.6%)
Medium 16 (66.7%) 18 (69.2%)
Very good 6 (25%) 5 (19.2%)
Work 16 (66.7%) 17 (77.3%) 0.425 Body Mass Index (kg/m2) 26.9±5.8 27.2±4.7 0.849 Diabetes Mellitus 1 (4.1%) 2 (7.6%) 0.55 Dyslipidemia 9 (37.5%) 10 (38.5%) 0.859 Hypertension 6 (25%) 5 (19.2%) 0.671
doi:10.1371/journal.pone.0127012.t001
Pre-study exclusions
Seven patients refused to enter the hyperbaric chamber before the beginning of the control/
treatment period (3 in the crossover group and 4 in the treated group). One patient was exclud
ed in the crossover group during the control period.
In-study exclusions. Four patients decided to drop out during the treatment protocol due
to dizziness, claustrophobia and inability to adjust by “ear pumping” to the hyperbaric condi
tion (2 in the crossover group and 2 in the treated group).
Accordingly, 48 patients (24 in the treated group and 24 in crossover group) were included
in the final analysis (Fig 1). All patients were females of ages 21–67, and the time elapsed from
the FMS diagnosis to the study recruitment was 2–22 years with mean of 6.5 years.
Baseline characteristics. Patients’ characteristics are summarized in Table 1. As seen from
this table, there was no significant difference in the included measures between the two groups.
The Effect on Pain
Tender point evaluation. The effect of the hyperbaric oxygen treatment on the patients’
pain, as assessed by the change in the dolorimeter threshold of the tender points (see Methods)
is summarized in Fig 2 and in Table 2. Fig 2A shows the treatment effect on the dolorimeter
thresholds and Fig 2B shows the effects on the number of tender points. It is transparent in the
figure that the two groups had very close mean scores at baseline for both measures (within the
standard error). It is also transparent that the HBOT treatments of both groups led to
PLOS ONE | DOI:10.1371/journal.pone.0127012 May 26, 2015 9 / 25
Hyperbaric Oxygen Therapy Can Diminish Fibromyalgia Syndrome
Fig 2. The HBOT effects on tender points. A) The effect on dolorimeter threshold. For both groups, the threshold level tripled after treatment (about 1.5, red bars, vs. about 0.5, blue bars). B) The effect on the number of tender points. The treatment led to significant reduction in the number of tender points in both groups: by a factor of 2 in the treated group and by a factor of 3 in the crossover group.
doi:10.1371/journal.pone.0127012.g002
statistically significant improvements in the mean scores of both the dolorimeter thresholds
and of the number of tender points.
As seen in Fig 2 and detailed in Table 2, the dolorimeter threshold score significantly im
proved following HBOT in the treated group (mean change 1.11±0.79, p < 0.001) and in the
crossover group after HBOT (mean change 1.29±0.76, p < 0.001). Effect sizes were large: the
Cohen’s d measures were 1.3 and 1.68, respectively. The number of tender points was signifi
cantly reduced following HBOT in the treated group (mean change 8.46±5.36, p < 0.001) and
in the crossover group after HBOT (mean change 11.54±4.93, p < 0.001). The effect sizes were
large: Cohen’s D measures were 1.5 and 2.24, respectively.
As expected, no improvement was noticed in the crossover group following the control peri
od, neither in the dolorimeter thresholds nor in the point count. It can be seen that the cross
over group had the same general score at baseline and after the control period. This value
seems higher than the score of the treated group at baseline – 0.65 vs. 0.55, and the post-HBOT
dolorimeter thresholds score of the treated group seems lower than that of the crossover
group – 1.65 vs. 1.85
Examining the relative changes. There is a high patient-to-patient variability in the
dolorimeter thresholds. The magnitude of the change in a dolorimeter threshold has different
implications for patients at low or high base levels. Hence, we inspected the effect of the HBOT
on the relative change, i.e., the change relative to the base value. We calculated, for each person,
the relative change in the dolorimeter threshold for each period (control and HBOT for the
crossover group and HBOT for the treated group). In Fig 3A we show the mean relative
changes in dolorimeter threshold for the crossover group following the control period and fol
lowing HBOT, and for the treated group following HBOT. We note that calculating the mean
of the relative changes is more informative than calculating the changes in the mean values, es
pecially for small groups with high patient-to-patient variability. Looking at the relative
changes elucidates the improvements after the HBOT period vs. the control period of the cross
over group and the baseline for the treated group. The same analysis was conducted for the
number of tender points. In Fig 3B we show the mean relative changes in the number of tender
PLOS ONE | DOI:10.1371/journal.pone.0127012 May 26, 2015 10 / 25
Hyperbaric Oxygen Therapy Can Diminish Fibromyalgia Syndrome
Table 2. Summary of the results of the tender points evaluation, physical function assessment, symptoms and quality of life questionnaires.
P1- p values for comparison before and after HBOT in the treated group (paired t test).
P2- p values for comparison before and after the control period in the crossover group (paired t test).
P3- p values for comparison after the control period before and after HBOT in the crossover group (paired t test).
P4- p values for comparison of the treated group after HBOT and the crossover after the control period (independent sample t test). * Data is presented as mean± standard deviation
doi:10.1371/journal.pone.0127012.t002
points for the crossover group following the control period and following HBOT, and for the
treated group following HBOT. For the control group, we also compared between the relative
changes during the control + treatment periods (the combined period) and during the treat
ment period and found them statistically equal (S1 File).
Scatter plot analysis of the dolorimeter threshold. In Fig 4, we show a scatter plot of the
relative changes in dolorimeter threshold as a function of baseline. The results illustrate the dif
ferences between the control period of the crossover group and the post HBOT of both groups.
Fig 3. Assessments of the mean relative changes in the pain level. A) The mean relative change and standard errors in the dolorimeter thresholds for the crossover group following the control period (green) and following HBOT (blue), and for the treated group following HBOT (red). B) The mean relative changes and standard errors in the number of tender points for the crossover group following the control period (green) and following HBOT (blue), and for the treated group following HBOT (red).
doi:10.1371/journal.pone.0127012.g003
PLOS ONE | DOI:10.1371/journal.pone.0127012 May 26, 2015 11 / 25
Hyperbaric Oxygen Therapy Can Diminish Fibromyalgia Syndrome
Fig 4. Scatter plot of the individual relative changes in the dolorimeter threshold. The figure shows the
relative change in all patients (y-axis in unit change) as a function of the baseline value. For the treated group,
each patient is represented by a single red dot. The relative change is the change during HBOT and the
baseline value is the value before treatment. For the crossover group, each patient is represented by two
dots: a green dot represents the relative change during the control period, with the baseline being the value
before the control. A blue dot represents the relative change during treatment, with the baseline value being
the value before treatment (which is also the value at the end of the control period). The green line represents
the mean relative change in the crossover group following the control period and the green dashed lines
represent the ±1std from the mean.
doi:10.1371/journal.pone.0127012.g004
Notably, apart from 6 patients (3 from the crossover group and 3 from the treated group), all
others showed significant improvement following the treatment. Note that, in general, the
higher the baseline threshold the smaller the improvement.
The Effects on Physical Functions, Psychological Distress and Quality of
Life
The HBOT effects on the physical functions, the psychological distress and the quality of life
are detailed in Table 2.
Physical function assessments. The FIQ score significantly improved following HBOT in
the treated group (mean change 1.31±0.99, p < 0.001) and in the crossover group after HBOT
(mean change 1.02±0.92, p = 0.05). The effect sizes were large and medium: Cohen’s D
PLOS ONE | DOI:10.1371/journal.pone.0127012 May 26, 2015 12 / 25
Hyperbaric Oxygen Therapy Can Diminish Fibromyalgia Syndrome
Fig 5. Assessments of the mean relative changes in the FIQ, SCL-90 and the SF-36 scores. The figures show the mean relative changes and standard errors in the three measures for the crossover group following the control period (green) and following HBOT (blue), and for the treated group following HBOT (red). A) Mean relative changes and standard errors in physical function assessed by the FIQ score. B) Mean relative changes in and standard errors in the psychological distress assessed by the SCL-90 score. c) Mean relative changes and standard errors in the quality of life assessed by the SF-36 score.
doi:10.1371/journal.pone.0127012.g005
measures were 1.29 and 0.64, respectively. As expected, there was no improvement in the FIQ
score in the crossover group following the control period.
Psychological distress. The SCL-90 score significantly improved following HBOT in the
treated group (mean change 1.10±0.79, p < 0.01) and in the crossover group after HBOT
(mean change 1.29±0.76, p = 0.05). The effect sizes were medium: Cohen’s D measures were
0.66 and 0.60, respectively. As expected, there was no improvement in the SCL-90 score in the
crossover group following the control period.
Quality of life assessments. The SF-36 score significantly improved following HBOT in
the treated group (mean change 0.34±0.33, p < 0.01) and in the crossover group after HBOT
(mean change 0.23±0.39, p = 0.05). The effect sizes were large medium: Cohen’s D measures
were 1.0 and 0.58, respectively. As expected, there was no improvement in the SF-36 score in
the crossover group following the control period.
Examining the relative changes. Similar to the pain related scores, there is also a high pa
tient-to-patient variability in the FIQ, SCL-90 and the SF-36 scores. Hence, we also inspected
the effect of the HBOT on the relative changes in these scores. The results shown in Fig 5 reveal
significant improvements in all scores following treatment for both groups. In S1 File we show
a comparison between the relative changes in FIQ, SCL-90 and SF-36, during the combined
and the treatment periods for the patients in the crossover group (see definition in the effect on
pain section above).
SPECT assessments of changes in brain activity
Motivation. As mentioned in the introduction, earlier studies compared SPECT images of
FMS patients to those of healthy subjects. The studies revealed a notable difference in brain ac
tivity between the two groups. In particular, they found that FMS is associated with elevated ac
tivity in the somatosensory cortex and reduced activity in the frontal, cingulate, medial
temporal and cerebellar cortices [11, 12]. These results provide an excellent independent con
trol reference to which changes in brain activity following HBOT should be compared to.
Within group and between groups comparison. The crossover affords two types of com
parison: 1. within group—between the changes in FMS symptoms and in brain activity during
the control period and during the treatment period in the same patients (of the crossover
group). 2. between groups – between the changes during treatment in patients of the crossover
group vs. patients of the treated group. Even more persuasive was the correspondence we
found between the brain areas whose activity increased/decreased following the HBOT sessions
PLOS ONE | DOI:10.1371/journal.pone.0127012 May 26, 2015 13 / 25
Hyperbaric Oxygen Therapy Can Diminish Fibromyalgia Syndrome
and the brain areas that were shown in previous studies to have reduced/enhanced activity in
FMS patients relative to normal subjects. In order to attain greater validity, symptom assess
ment and SPECT analysis were done by blinded evaluations and evaluators: the tests of the
FMS state were done by computerized validated methods and the SPECT analysis was blind to
patients’ participation in treated/crossover group.
Association. Brain SPECT imaging was performed and evaluated for all patients. The pa
tients in the treated group had two SPECT imagings (pre- and post-treatment) and the patients
in the control group had three SPECT imagings (pre- and post-control period, and post-treat
ment). One patient from the control group missed the post-control SPECT imaging (hence we
have 23 results for SPECT assessed brain activity during the control period). In S2 File we pres
ent detailed results of SPECT imaging for all Brodmann areas (BAs) of all the tested patients.
NeuroGam software, used to normalize and average the SPECT measurements into Brodmann
areas, excludes small volume BAs from the available data in order to avoid inconsistent results.
Therefore, the following BAs were not assessed in this study: Bilateral 1, 2, 3, 12, 26, 29, 30, 33,
34, 35, 41, 42, 43, 48, 52.
Association vs. correlation. We specifically use the term “association” rather than “corre
lation” since direct mathematical correlations between the physiological changes and the
changes in brain activity are ill defined—there is no one-to-one correspondence between the
Brodmann areas and the physiological functions, as each physiological function can be per
formed by locations spread over several Brodmann areas and vice versa. We would like to em
phasize that even in the cases that correlation can be defined and computed, correlations do
not reveal causality. Moreover, from biological perspective, the changes in the brain activity are
expected to cause physiological changes that in turn can lead to additional changes in the brain
activity. Therefore, our aim was to show correspondence, rather than mathematical correla
tions, between the changes in the brain activity and the physiological changes.
BA histogram of mean relative changes. To summarize and assess the results, we con
structed histograms of the mean relative changes, <Rchange>(n), for each Brodmann area (n).
To construct the results shown in Fig 6, we calculated, for each patient (i), the relative change
in the SPECT measured brain activity, Rchange(i,n), during each phase of the trial (see Methods
section). Then we calculated the average changes, <Rchange>(n), for the 41 patients (out of 48)
from the treated group and the crossover group that showed significant improvement in the
FMS symptoms following HBOT (the response group mentioned in the method section) and
ordered the results from the most reduced to the most elevated activity. The changes in the
BAs of te response group following HBOT were compared with those of the patients in the
crossover group during the control period.
To quantify the results shown in Fig 6 and illustrate the statistical significance, we also cal
culated the Pearson correlations for the following four combinations. 1. The correlations be
tween the vectors of the mean relative changes for the response group and the vectors for the
crossover group during the control period. 2. The correlations between the mean relative
changes during treatment for the group of 41 responders and those for the group of 7 non re
sponders. 3. The correlations between the mean relative changes during treatment for the
whole response group and those for the responders from the treated group. 4. The correlations
between the mean relative changes during treatment for the whole response group and those
for the responders from the crossover group. The correlations for the four combinations were
found to be -0.25, -0.05, 0.77 and 0.68, respectively.
Normalized BA histogram of mean relative changes. In Fig 7A we show a histogram
similar to the aforementioned one, but in which we normalized the mean relative changes of
each BA (n) by its corresponding significance index Iσ(n) as is defined and explained in the
Methods section. To better scrutinize the effect of the normalization, we constructed
PLOS ONE | DOI:10.1371/journal.pone.0127012 May 26, 2015 14 / 25
Hyperbaric Oxygen Therapy Can Diminish Fibromyalgia Syndrome
Fig 6. BA histogram of mean relative changes. The figure shows the histogram as is explained in the text. The Y-axis shows the mean relative change <Rchange>(n) for the Brodmann area indicated in the X-axis. The results for the patients of the response group after the HBOT period are colored from light blue (BA with the strongest activity reduction) to light red (BA with the highest activity elevation). The green bars correspond to the mean relative changes in the patients of the crossover group following the control period.
doi:10.1371/journal.pone.0127012.g006
2-dimentional scatter plots of the significance index vs. the normalized relative changes. In Fig
7B we show the results for the patients in the response group following the HBOT period; in
Fig 7C we show the results for the patients in the crossover group following the control period.
Comparison between the two scatter plots reveals that, following treatment, the Brodmann
areas that show large changes in brain activity also have high significance factors (see Figs 6
and 7B). In contrast, comparing Fig 7C and 7A reveals that, following the control period, the
significance index is low for Brodmann areas that exhibit big changes in brain activity. The cor
relations for the four combinations mentioned above, calculated for the normalized mean
changes, were found to be -0.28, -0.09, 0.66 and 0.61, respectively.
Assessment of the results
The results in Fig 7 reveal several distinct Brodmann areas with significant normalized changes
in the brain activity following the HBOT period. More specifically, in the response group, 10
BAs showed above +0.6 normalized mean changes (hyper-perfusion) and 5 BAs showed below
-0.6 normalized mean changes (hypo-perfusion) following the HBOT period. In contrast, the
normalized mean changes in brain activity for all BAs are scattered within the (-0.6 — +0.6)
range following the control period of the patients in the crossover group. In addition, the scat
ter of the normalized mean changes after HBOT fits a distinct funnel shape distribution (Fig
7B) that is significantly different from the distribution after the control period (Fig 7C). In Fig
8 we show a projection of the aforementioned findings on the brain maps. For clarification, we
used the same color code as in Fig 7.
The results revealed that following the HBOT period, improved patients (responders) ex
hibit elevated activity of BAs in the frontal lobe (25L+R, 10L+R, 47R, 45R, 11R, 9R, 8R) and in
BA 38L, and reduced activity of BAs in the posterior brain (7L+R, 37L, 36L, 17L). As
PLOS ONE | DOI:10.1371/journal.pone.0127012 May 26, 2015 15 / 25
Hyperbaric Oxygen Therapy Can Diminish Fibromyalgia Syndrome
Fig 7. The effect of significance index normalization. A) Normalized BA histogram of mean relative changes. The figure is similar to Fig 6 but the Y-axis is for the normalized values, that is for Iσ(n)* <Rchange>(n) and not for <Rchange>(n) that are used in Fig 6. The BAs within the rectangles are the ones with normalized mean relative changes smaller than -0.6 or larger than +0.6. B) The two dimensional scatter plot Iσ(n) vs. Iσ(n)* <Rchange>(n) for the patients of the response group following the HBOT period. C) Similar scatter plot for the patients in the crossover group following the control period. The color code in (B) and (C) is the same as in (A). The funnel shaped black curve is a fit of the results in (B) to a reciprocal Lorentzian curve: f(x) = {Xmax- γ*[π*(γ2+x2)]-1} with Xmax = 0.95, γ = 0.335.
doi:10.1371/journal.pone.0127012.g007
mentioned before, earlier studies showed that FMS patients have reduced brain activity in BAs
in the frontal cortex and elevated activity in the posterior brain 11, 12. We found that, after treat
ment, BAs in the posterior brain show reduced activity and BAs in the frontal cortex show ele
PLOS ONE | DOI:10.1371/journal.pone.0127012 May 26, 2015 16 / 25
Hyperbaric Oxygen Therapy Can Diminish Fibromyalgia Syndrome
Fig 8. Projection of the significant changes on the brain maps. The figure shows the results of the normalized mean changes as projected on the brain maps, left BAs (A) and right BAs (B). We colored the BAs that show significant changes in activity using the same color code as in Figs 6 and 7 – from light blue (BA with the strongest activity reduction) to light red (BA with the highest activity elevation).
doi:10.1371/journal.pone.0127012.g008
leads to beneficial changes in the brain activity of specific BAs known to have abnormal activity
in these patients.
In the next section we mention that the amelioration consequent to HBOT led to a signifi
cant decrease in the intake of pain medications by the patients. In principle, part of the ob
served changes in the SPECT imaging may be associated with the changes in the intake of pain
medication. While this possibility cannot be ruled out, we deem it unlikely. First, we note that
the patients have been taking pain medication for a long time (years). The intake of the drugs
eased the pain but did not reverse the condition, while HBOT did reverse the condition. Also,
the changes in the brain activity as detected by the SPECT coincided with improvement of the
FMS symptoms, so much so that most of the patients could reduce or stop altogether the intake
of pain medications. In other words, the plausible causal chain is that the changes in brain ac
tivity were induced by the HBOT, these changes alleviated the FMS symptoms and eased the
pain, leading to a diminished need for pain medication.
PLOS ONE | DOI:10.1371/journal.pone.0127012 May 26, 2015 17 / 25
Hyperbaric Oxygen Therapy Can Diminish Fibromyalgia Syndrome
Changes in intake of pain related medications
The amelioration of pain consequent to HBOT led to a significant decrease in the level of anal
gesic medications intake by the patients in both groups. More specifically, 9 patients from the
treated group were on chronic daily medication with analgesic drugs (5 were taking two differ
ent drugs and 4 were taking one) before HBOT. After the HBOT, 3 patients got completely off
medication, 3 out of the 5 continued with two drugs, and 3 out of the 4 continued with one
drug, p = 0.02. In the crossover group, 12 patients were on chronic daily medication of analge
sic drugs before HBOT (2 on two drugs and 10 on one drug). All of them continued taking the
medications during the control period. Consequent to the HBOT period, 5 patients stopped
taking drugs altogether and all other 7 patients took one drug, p = 0.02. With regard to chronic
use of antidepressants, in the treated group, the 7 patients that were chronically treated before
HBOT continued with their medications at the end of the treatment. In the crossover group, of
the 12 patients treated with antidepressants at baseline and during the control period, 8 contin
ued with their medications after the HBOT treatments, p = 0.04.
Safety and side effects
Five patients decided to stop the HBOT due to dizziness, claustrophobia and inability to adjust
ear pressure by “ear pumping”. Thirteen patients had mild barotrauma that resolved spontane
ously and did not prevent them from completing the treatment protocol.
Noticeably, 14 patients (29%) reported an increase in the pain/sensation during the first 10–
20 session. However, at the end of the HBOT period, all of these patients experience significant
amelioration of pain and improvements in the different evaluated parameters in this study as
compared to baseline.
Discussion
We presented a prospective active control, clinical trial of evaluating the effect of HBOT on fe
male patients of ages 21–67 with chronic FMS. The time elapsed from FMS diagnosis to study
recruitment was 2–22 years (mean 6.5 years). A crossover approach was adopted in order to
overcome the HBOT inherent sham control problem (see discussion further below). The par
ticipants were randomly divided into two groups. One, the treated group, received two months
of HBOT; the other, the control group, was not treated during those two months and received
treatment in the following two months. The advantage of the crossover approach is the option
for a triple comparison – between treatments in two groups, between treatment and no treat
ment in the same group, and between treatment and no treatment in different groups.
The changes in all measures (pain threshold, number of tender points, FIQ, SCL-90 and SF
36) were assessed by detailed computerized evaluations and were compared to changes in
brain activity obtained by SPECT imaging. The HBOT in both groups led to similar significant
improvements. No significant changes were detected during the non-treatment period in the
crossover group. These results are in agreement with earlier findings by Yildiz et al. [27]. Anal
ysis of brain imaging showed significantly increased neuronal activity after a two-month period
of HBOT, compared to the control period.
Brain functionality
What makes the results particularly convincing is the good correspondence between the physi
ological improvements and the changes in brain functionality as detected by the SPECT scans,
as well as the good agreement with the abnormal brain activity of FMS patients. As presented
in the introduction, comparison between brain activities of healthy subjects and FMS patients,
PLOS ONE | DOI:10.1371/journal.pone.0127012 May 26, 2015 18 / 25
Hyperbaric Oxygen Therapy Can Diminish Fibromyalgia Syndrome
assessed by SPECT imaging, revealed higher activity in the somatosensory cortex and reduced
activity in the frontal, cingulate, medial temporal and cerebellar cortices in FMS patients [11,
12]. We also mentioned that these results are in agreement with earlier studies based on fMRI
imaging [13]. The specially devised analyses of the HBOT imaging revealed that the improve
ments in the syndrome status went hand-in-hand with changes in the patterns of brain activity
towards those of healthy subjects. More specifically, for the response patients, HBOT sessions
led to reduction in brain activity in the somatosensory cortex and enhancement of the brain ac
tivity in the frontal, cingulate, medial temporal and cerebellar cortices.
HBOT can rectify abnormal brain activity
Levels of pain sensations are determined by the sensory recording and higher level information
processing (interpretation) in the brain. Evidence from previous studies suggests that the pain
in fibromyalgia results primarily from abnormality in the function of pain processing path
ways. In simple terms, it may be described as hyper-excitability of pain processing pathways
and under-activity of inhibitory pain pathways in the brain, resulting in the affected individual
experiencing pain. In the present study we found that HBOT can rectify chronically abnormal
brain activity – decrease the activity of hyperactive regions (mainly posterior regions) and in
crease the activity of underactive regions (mainly frontal areas), in good agreement with the
current knowledge regarding the brain’s response to pain.
More specifically, brain areas that are activated in response to pain are S1, S2 (BA 1, 2 and