Artículo15 - Parkinson

PLOS ONE
RESEARCH ARTICLE
Long-term treatment with transcranial pulsed

electromagnetic fields improves movement
speed and elevates cerebrospinal
erythropoietin in Parkinson’s disease
Bente Rona Jensen ID1,2*, Anne Sofie Bøgh Malling1,2, Sissel Ida Schmidt3,
Morten Meyer1,2,3, Bo Mohr Morberg1, Lene Wermuth1,2
1 Department of Neurology, Odense University Hospital, University of Southern Denmark, Odense,

Denmark, 2 Department of Clinical Research/BRIDGE, University of Southern Denmark, Odense, Denmark,
a1111111111 3 Department of Neurobiology Research, Institute of Molecular Medicine, University of Southern Denmark,
a1111111111 Odense, Denmark
a1111111111
a1111111111 * bente.r.jensen@rsyd.dk, brjensen100@gmail.com
a1111111111
Abstract
OPEN ACCESS
Background
Citation: Jensen BR, Malling ASB, Schmidt SI,
Meyer M, Morberg BM, Wermuth L (2021) Long- Parkinson’s disease is characterized by motor dysfunctions including bradykinesia. In a
term treatment with transcranial pulsed recent study, eight weeks of daily transcranial stimulation with bipolar pulsed electromag-
electromagnetic fields improves movement speed netic fields improved functional rate of force development and decreased inter-hand tremor
and elevates cerebrospinal erythropoietin in
Parkinson’s disease. PLoS ONE 16(4): e0248800.
coherence in patients with mild Parkinson’s disease.
https://doi.org/10.1371/journal.pone.0248800
Editor: J. Lucas McKay, Emory University, UNITED Objective

STATES To investigate the effect of long-term treatment with transcranial bipolar pulsed electromag-
Received: January 25, 2020 netic fields on motor performance in terms of movement speed and on neurotrophic and
Accepted: March 4, 2021 angiogenic factors.
Published: April 28, 2021

Methods
Copyright: © 2021 Jensen et al. This is an open
access article distributed under the terms of the Patients diagnosed with idiopathic Parkinson’s disease had either daily 30-min treatment
Creative Commons Attribution License, which with bipolar (±50 V) transcranial pulsed electromagnetic stimulation (squared pulses, 3ms
permits unrestricted use, distribution, and duration) for three eight-week periods separated by one-week pauses (T-PEMF group) (n =
reproduction in any medium, provided the original
16) or were included in a PD-control group (n = 8). Movement speed was assessed in a six-
author and source are credited.
cycle sit-to-stand task performed on a force plate. Cerebrospinal fluid and venous blood
Data Availability Statement: All relevant data are
were collected and analyzed for erythropoietin and vascular endothelial growth factor.
within the paper and its Supporting Information
files.
Funding: The research group received funding

Results
from The Jascha Foundation, Denmark; Den A.P. Major significant improvement of movement speed compared to the natural development of
Møllerske Støttefond, Copenhagen, Denmark; and
the disease was found (p = 0.001). Thus, task completion time decreased gradually during
Grosserer L.F. Foghts Foundation, Charlottenlund,
Denmark. The authors report no competing the treatment period from 10.10s to 8.23s (p<0.001). The untreated PD-control group did
interest. Thus, the funders had no role in the study not change (p = 0.458). The treated group did not differ statistically from that of a healthy
PLOS ONE | https://doi.org/10.1371/journal.pone.0248800 April 28, 2021 1 / 15

PLOS ONE Long-term T-PEMF treatment in PD
design, data collection and analysis, decision to age matched reference group at completion of treatment. Erythropoietin concentration in
publish, or preparation of the manuscript. the cerebrospinal fluid also increased significantly in the treated group (p = 0.012).
Competing interests: The authors have declared
that no competing interests exist. Conclusion
Long-term treatment with transcranial bipolar pulsed electromagnetic fields increased
movement speed markedly and elevated erythropoietin levels. We hypothesize that treat-
ment with transcranial bipolar pulsed electromagnetic fields improved functional perfor-
mance by increasing dopamine levels in the brain, possibly through erythropoietin induced
neural repair and/or protection of dopaminergic neurons.
Introduction
Parkinson’s disease (PD) is a progressive chronic disorder characterized by motor dysfunc-
tions including bradykinesia, rigidity, akinesia, tremor, and postural instability. Bradykinesia,
i.e. slowness of movements, was perceived as the most prevalent troublesome symptom by
patients with early PD (< 6 years) [1]. The ability to perform fast muscle contractions is
impaired in PD even though no deficits in maximal force occur [2, 3]. Coordination and bal-
ance are also impaired in PD [4, 5]. These specific motor deficits contribute to the slow move-
ment speed in patients with PD, which constitutes a challenge for daily activities. The primary
pathological processes involve degeneration of nigrostriatal dopaminergic neurons leading to
reduced striatal dopamine. Dopamine is the major catecholamine neurotransmitter in the
brain and is highly significant for motor control processes [6–8].
Medical treatment of PD is purely symptomatic as no cure exists. PD has traditionally been
considered a single-region disease in the basal ganglia but has recently been proposed as a sys-
tem-level disorder that includes functional interactions between brain regions such as the
dopamine dependent motor circuit (cortico-striatal-thalamic circuit) and the cognitive circuit
(frontal-striatal circuit) [9]. This has stimulated new research directed towards further under-
standing of the disease, its functional consequences, and innovative treatments.
Electromagnetic stimulation of biological tissue is a non-invasive rapidly emerging tech-
nique. The biophysical effect of pulsed electromagnetic stimulation is to induce ion currents in
the tissue and to depolarize the membrane potential slightly [10]. Electromagnetic stimulation
in animals and in-vitro seems to enhance cellular activity and stimulate growth-related
responses and regeneration [10]. Thus, pulsed electromagnetic fields (PEMF) stimulated nerve
growth and attenuated nerve abnormalities, increased microvascular blood flow and tissue
oxygenation, and increased capillary density [11–14]. Treatment with PEMF may thus delay
disease progression or even induce neuro-repair in PD.
In a recent clinical trial, we applied transcranial PEMF (T-PEMF) in patients with idio-
pathic PD [15–17]. This study focused on non-invasive functional effect measures and was the
first to apply T-PEMF in PD. We found that 30-min daily treatment with T-PEMF over eight
weeks were superior to placebo treatment to increase functional rate of force development dur-
ing a cyclic chair rise task, and that inter-hand coherence was reduced in the least affected
patients with PD [15, 17]. However, structural changes in the brain take time and eight weeks
of treatment is unlikely to be sufficient. Thus, even longer treatment periods are needed to bet-
ter understand the potential of this treatment modality [15, 17].
We speculate on a potential connection between T-PEMF treatment and cerebral levels of
erythropoietin (EPO), vascular endothelial growth factor (VEGF), and dopamine. EPO is

present in the central nervous system, has a multimodal neuroprotective profile and has a key
role in neuroprotection and neuro repair [18, 19]. EPO releasing cells such as Er-NPC pro-
moted neural repair and recovery of function in a mouse model of spinal cord traumatic
injury, and EPO improved memory and spatial learning in a rat model of Alzheimer’s disease
[20, 21]. Signs of neuroprotective effects of EPO were also found in an animal model of PD
[22, 23]. VEGF mediates angiogenesis, neural migration and neuroprotection [24, 25] and,
along with EPO may be a potential therapy for certain neurological disorders. The literature
mainly comprises animal and in-vitro studies, however, and our knowledge in humans is
limited.
The present study, aimed to test the hypothesis that long-term treatment with T-PEMF
would improve motor performance (in terms of movement speed) and stimulate production
of neuroprotective and angiogenetic compounds (EPO and VEGF) in the brain in patients
with PD.
Materials and methods

Study design
The study was a prospective interventional treatment study with parallel group design and par-
tially blinded outcome assessment. Treatment outcomes were compared to natural develop-
ment of the disease and to healthy controls. The study was approved by the Regional Scientific
Ethical Committees for Southern Denmark (S-20160106) and the Danish Medicines Agency
(CIV-16-02-014744) and was conducted in accordance with the Declaration of Helsinki. All
participants gave written informed consent prior to participation.
Participants
Patients diagnosed with idiopathic PD, according to the UK Brain Bank criteria, were
recruited from the clinic. Inclusion criteria were a Mini-Mental State Examination (MMSE)
score >22, age >18 years, ability to give written informed consent, and to obtain certification
in the use of the T-PEMF device. Exclusion criteria were any known neuromuscular or neuro-
logical diseases other than PD that might interfere with motor function; psychopathological
treatment of other conditions than depression; substance abuse; active medical implants; preg-
nancy; current or previous cancer in the head, brain, or neck region; leukemia; autoimmune
disease; epilepsy; and open scalp wounds. Treatment compliance >80% was required to be
included in the analyses. Of the 24 patients recruited, 16 patients were included in an active
T-PEMF treatment group and eight patients were included in a PD-control group (no
T-PEMF treatment). The PD-control group was included in the study to indicate the time
course of the typical development of PD. The patients were pharmacologically well-treated,
and the medication remained unchanged for at least six weeks prior to and throughout the
intervention period (T-PEMF group). Disease severity was assessed using the Unified Parkin-
son Disease Rating Scale (UPDRS) and used to describe the study population [26]. In addition,
healthy reference values (n = 39) based on data from an age and sex matched healthy cohort
were extracted without any knowledge of their other characteristics [27].
Eight participants from the active group in the clinical trial (NCT02125032) (8 weeks treat-
ment) were enrolled in the present study (3x8 weeks treatment) (T-PEMF group: n = 6, PD-
control group: n = 2). The eight participants from the previous clinical trial were recruited
without any knowledge regarding prior results from the clinical trial. The distribution of the
eight participants between the T-PEMF group and the PD-control group was by convenience.
The average time from last treatment in the 8-week intervention to enrolment in the present
long-term intervention was 25 months (range 14–32 months), so no carry over effect from the

8-week intervention was expected. No results presented in the present study have been
reported previously.
Treatment intervention
In the intervention group, daily T-PEMF was applied for three eight-week periods, separated
by one-week pauses. Treatment with T-PEMF has not been conducted for such a long period
previously, and the one-week pause between treatments periods were required by the Danish
Medicines Agency to permit the study. Thus, the total treatment intervention period was 26
weeks. The treatment was home-based, and each daily session lasted 30 min. T-PEMF was
applied through seven coils placed in a helmet-like shape, with one coil in the central occipital
region, one in each frontal-parietal region, and two bilaterally in the temporal region (anterior
and posterior) (Re5 NTS Parkinson Treatment System, Re5, Frederiksberg, Denmark). The
seven coils were connected to an external pulse generator that generated bipolar, squared
pulses (±50 V, 3 ms, 50 Hz) to initiate rapid current changes in the coils, giving rise to rapidly
changing, time-dependent electromagnetic fields. An electromagnetic field penetrates through
electrically insulated tissue (e.g. the skull) and induces a driving force on charged particles and
thereby electrical currents in the brain. The peak E-field intensity has been estimated to
approximately 2.5 mV/cm near the coil [10], and the spatial distribution is highly dependent
on the steep gradient of the applied bipolar pulsed electromagnetic field.
Treatment initiation and duration were controlled by a pre-programmed chip-card
inserted into the pulse generator. Date and time of each treatment were stored on the chip
card and used to assess treatment compliance. The participants were in contact with the inves-
tigators once a week (telephone or e-mail). They were also given a hotline telephone number
so they could contact the investigators at any time if they experienced adverse effects or had
treatment-related questions.
Effect measures
Movement speed. The primary outcome was movement speed, which was assessed in a
six-cycle sit-to-stand (STS) task as described previously [15, 28]. In short, the participants sat
on a chair with their arms folded at chest height. Seat height was 120% tibia length. The knee
angle was 90–100 degrees and the feet were placed on a force plate (AMTI, 120x90 cm,
BP6001200, USA) with an inter-feet distance corresponding to shoulder width. The criteria for
an approved trial were fully extended knees in the standing position and back-to-backrest con-
tact in the sitting position. Familiarization was performed before measurements. The research-
ers performing the examinations (week 17 and week 27) were blinded to prior results of the
STS task.
The participants were asked to perform the task as fast as possible. Three approved STS tri-
als were performed with rest between trials to avoid fatigue. If the last trial was the fastest an
additional trial was performed to ensure maximal performance. Force plate data were sampled
(1000Hz), A/D converted (16-bit, Data Translation Inc., USA) and filtered with a zero phase-
lag, second order 25 Hz low-pass Butterworth filter. The average of the data from the two fast-
est approved trials is reported. The STS-task was performed three times: just before initiation
of the treatment, within one day after completion of the second treatment period and within
one day after completion of the third treatment period. The measurements were performed
with the patients in self-reported “on”-state and at approximately the same time of the day.
STS completion time was calculated as the time from the first vertical ground reaction force
peak of the 1st repetition to the 1st vertical ground reaction force peak of the 6th repetition,
thus reflecting the time for five repetitions. Functional rate of force development (RFDSTS-up)

was calculated as the average slope of the increasing force of the 1st force peak and decrease
gran
en
(RFDSTS-down) was the decreasing force of the 2nd force peak in the 2nd-6th repetition, within
the time interval corresponding to 30–70% of the exerted peak force. Results are reported in
dando
bodyweight per second (BW/s). The exact time while sitting and standing was not measured,
and we used instead the time interval from the first to the second force peak and from the sec-
e ond to the first force peak, to largely represent the standing and the sitting time, respectively.
Cerebrospinal fluid (CSF) and blood samples. Eight patients from the T-PEMF group
Loginat
underwent lumbar puncture just before treatment and within one day after treatment comple-
tion. The patients were consecutively recruited while excluding patients taking anticoagulants.
Lumbar puncture was performed using a 22-gauge spinal needle. CSF (5 ml) was collected
with polypropylene syringes in an intervertebral lumbar space and subsequently stored within
30-min at—80˚C. Venous blood samples were collected from 13 patients (eight in the
T-PEMF group, five in the PD-control group) before treatment and within one day after the
treatment completion. The patients were consecutively recruited (convenience sub-sample).
Blood samples were collected from the antecubital vein and centrifuged at 2000g for 10
min. The supernatants were stored in CRYO-S tubes and stored within 30-min at -80˚C. CSF
and blood samples were analyzed for EPO and VEGF using the MSD U-PLEX Human EPO
Assay (#K151VXK) / V-PLEX Angiogenesis Panel 1 Human Kit (#K15190D) (Mesoscale Dis-
covery, Rockville, MD, USA) and a SECTOR Imager 6000 (Mesoscale Discovery) plate reader
ht
according to the manufacturer’s instructions. Data were analyzed using MSD Discovery
Workbench software. Analyses were performed in duplicate and reported as average values.
The laboratory technician was blinded to sample identity.
además, subdivicta valorar
Statistical analysis
se
parten
variables confusosten
Statistics were performed in SAS 9.4 (USA). Normality tests were performed (Shapiro-Wilk). ④
it gre
A linear mixed model with parallel group approach was used to assess treatment effects on STS
performance between PD groups (T-PEMF group vs PD-control group). The model used was:
Performance outcome = treatment time treatment×time, performed with unstructured covari-
ance. Dependent performance outcome; STS completion time, RFDSTS-up, RFDSTS-down, stand-
ing time and sitting time. Independent factors; time (week 0, week 17 and week 27) and
sadatEn
treatment (T-PEMF and no T-PEMF). Unstructured covariance was used to allow inhomoge-
neous variance across the treatment period. For hypothesis test a significance level of
alpha = 0.05 was used. Post hoc paired two-sample t-test (two-tailed) between time points N
(week 0 versus week 27) was performed for STS performance means to assess differences from
baseline to post-treatment in the two PD groups. Significant values were corrected for multiple
testing according to Bonferroni.
To analyze the effect of T-PEMF treatment (T-PEMF group) on CSF-EPO and CSF-VEGF
a two-sample T-test (week 0 versus week 27) was used (two-tailed). To assess the effect of
T-PEMF treatment (T-PEMF group versus PD-control group) on plasma-EPO and plasma-
VEGF a repeated measure ANOVA was used.
inment
Comparisons between the T-PEMF group and the healthy reference group at specific time
points were performed using unpaired two-sample t-test (two-tailed).
Baseline comparisons between PD groups on participant characteristics were performed by
chi-square (sex), Mann-Whitney U-test (Mini-Mental State Examination Score and Hoehn
and Yahr Score, not normally distributed) and unpaired two-sample t-test (age, height, body
weight, UPDRS, Levodopa equivalent dose, disease duration).
Finally, correlations between UPDRS score at baseline and absolute as well as relative
change in STS completion time were calculated (Pearson).
*

Statistics are reported in accordance with general guidance on statistical reporting (SAMPL
guidelines).
Results
Participants
In the T-PEMF group, one patient withdrew, and one was excluded from the study due to
medical conditions unrelated to the T-PEMF treatment. Baseline characteristics for the PD
groups are shown in Table 1. We found no significant between-group differences in age,
height, bodyweight, UPDRS-total, UPDRS-motor, UPDRS-bradykinesia, MMSE score, levo-
dopa equivalent dose or disease duration. Average treatment compliance across the interven-
tion period (3x8 weeks) was 97.6% (SD 2.8) and no participants were excluded due to low
compliance. The healthy reference group had an age of 66.8 years (SD 6.5), bodyweight of 76.6
kg (SD10.8) and height of 1.74 m (SD 0.09).
Sit-to-stand/movement speed
The time course of the STS completion time differed significantly between the T-PEMF group
and the PD-control group (p = 0.001). The T-PEMF group completed the STS-task in 10.10s
(SD 2.41) at week 0 and decreased significantly to 8.23s (SD 1.78) post treatment (p<0.001)
(Fig 1A). The PD-control group did not change significantly across time (p = 0.458). The
healthy reference group completed the STS-task in 8.64s (SD 1.61). Completion time was sig-
nificantly longer for the PD group than for the healthy reference group at baseline (p = 0.015).
However, no between group (T-PEMF group vs. healthy reference group) difference in com-
pletion time was found post treatment (p = 0.429). Relative changes for the two PD groups are
shown in Fig 1B. The relative difference between the two PD groups increased gradually across
Table 1. Characteristics of the participants with Parkinson’s disease (PD) in the T-PEMF intervention group and the PD-control group.
T-PEMF, N = 14 PD-Control, N = 8 P-values
Sex (female/male) 5/9 2/6 0.604a
Age (years) 66.6 (7.6) 70.5 (6.3) 0.240b
Height (cm) 172 (8.7) 172 (6.6) 0.925b
Weight (kg) 73.5 (11.4) 73.3 (10.5) 0.960b
UPDRS-total 46.4 (22.2) 35.3 (5.1) 0.099c
UPDRS-motor 27.8 (15.0) 21.1 (3.4) 0.137c
UPDRS-bradykinesia 14.6 (8.1) 12.3 (3.2) 0.351c
Hoehn-Yahr score 2.0 (2.0–2.9) 2.3 (2.0–2.5) 0.904 d
MMSE 29.5 (28.0–30.0) 30.0 (28.8–30.0) 0.527d
Levodopa equivalent dose (mg/day) 498.3 (259.1) 469.4 (148.6) 0.777b
Disease duration (years) 6.2 (3.5) 5.3 (3.7) 0.581b
a
Chi-square test
b
T-test, equal variance
c
T-test, unequal variance (Cochran-Cox)
d
Mann-Whitney U-test
Unified Parkinson’s disease Rating Scale total score (UPDRS-total, range 0–199), motor score (UPDRS-motor, range 0–108) and bradykinesia score (UPDRS-
bradykinesia, range 0–36). Mini-Mental State Examination (MMSE, range 0–30).
Disease duration is the time period from diagnosis to study inclusion. Data are means (SD), except for MMSE and Hoehn-Yahr where medians and inter-quartile ranges
are reported.
https://doi.org/10.1371/journal.pone.0248800.t001

Fig 1. Sit-to-stand completion time. Absolute (A) and relative (B) completion time over the study period for the group receiving T-PEMF treatment (T-PEMF; green,
solid line with triangles) and the PD control group (PD CON; red, solid line with circles). A: ⇤ Significantly different performance course over time (P<0.05) between
T-PEMF and PD CON. Healthy reference level is indicated (Healthy REF; grey, dotted line). B: The relative change in a T-PEMF group from our previous study by
Malling et al. after 8 weeks of treatment is indicated (Previous T-PEMF; black, dotted line with x’s) [15].
https://doi.org/10.1371/journal.pone.0248800.g001
the intervention period and was 21% points at endpoint. Data from our recent eight-week
T-PEMF treatment study has been added to Fig 1B for comparison [15].
Subdivision of the STS completion time showed no between-group differences regarding rate
of force development although a non-significant trend was seen (Fig 2A). However, the time
periods largely representing standing and sitting time, respectively, were significant shorter in
the T-PEMF group (Fig 2B and 2C). A significant moderate negative correlation was found
between absolute changes in STS completion time and UPDRS-total score in the T-PEMF group
and a tendency was found for the UPDRS-motor score (Fig 3A and 3C) [29]. No association was
found between UPDRS scores and the relative change of completion time (Fig 3B and 3D).
Biomarkers
EPO and VEGF concentrations in CSF and in venous blood plasma are presented in Fig 4.
CSF-EPO concentrations increased significantly in response to the T-PEMF treatment inter-
vention (all patients increased (n = 8)). CSF-VEGF concentration increased in five out of six
patients included in the analysis (non-significant).
Discussion
The present study is the first to document that T-PEMF treatment applied for three eight-week
periods improves movement speed significantly when compared with the untreated PD

Fig 2. Subdivision of the sit-to-stand movement. Rate of force development during rising from a chair (A, upper) and lowering to a chair (A, lower), standing time (B)
and sitting time (C) for the group receiving T-PEMF treatment (T-PEMF; green, solid line with triangles) and the PD control group (PD CON; red, solid line with
circles). ⇤ Significant between-group difference in performance over time (P<0.05).
control group. Thus, the characteristic slowness of movements was markedly reduced. More-
over, enhanced levels of CSF-EPO in response to the treatment intervention were found. A
possible causal mechanistic link between T-PEMF, CSF-EPO and movement speed are
suggested.
Several animal models have shown a neuroprotective action of EPO in neurodegenerative
diseases when injected intravenously or infused directly into the brain. [e.g. 22, 30]. When sys-
temic (intravenous) administration of EPO and intrastriatal injected EPO were compared in a
PD animal model, the blood-brain barrier appeared to limit the passage of EPO into the brain
as only low levels of EPO crossed despite a high intravenous dose [18, 31, 32]. Systemic admin-
istration of EPO in patients with PD is thus unlikely to be appropriate. Nevertheless, improved
autonomic function and cognition were found in patients with PD when assessed clinically
with the Non-Motor Symptom Scale after five weeks of intravenous EPO administration twice
a week [33]. Infusion of EPO or neuroprotective EPO-like compounds directly into the
human brain is a theoretical possibility, but its invasive character and expected inhomoge-
neous spatial distribution in brain tissue make this less realistic. To overcome these challenges,
treatments that upregulate the intrinsic production of EPO in the brain would be preferable.

Fig 3. Association between disease severity and change in sit-to-stand completion time from week 0 to week 27 in the T-PEMF group. Pearson correlations
between UPDRS total score and the absolute (A) and relative (B) change of completion time, and between UPDRS motor score and the absolute (C) and relative (D)
change of completion time.
We found that long-term treatment of the brain with T-PEMF increased CSF levels of EPO,
indicating enhanced levels of locally produced intracerebral EPO. An animal model of PD
showed that intrastriatal administered EPO has specifically protected dopaminergic neurons
from cell death and improved motor performance [34] and thus, we cautiously hypothesize
that T-PEMF treatment has contributed to neural repair and protection of the dopaminergic
neurons in the present study. The present study is the first to assess long-term treatment effect
of T-PEMF and to our knowledge, the first to measure CSF levels of EPO in patients with PD.
As expected, the STS completion time at baseline was longer in the two PD groups than in
the age-matched healthy reference group indicating bradykinesia. Our T-PEMF group
(UPDRS-total: 46(SD 22), UPDRS-motor: 28(SD 15) medication in LED: 498 mg/day (SD 25))
performed 17% slower than the healthy reference group, while in a previous study a PD group
(on-medication, UPDRS-total: 58(SD 18), UPDRS-motor: 35(SD 10), medication in LED: 758
mg/day (SD 412)) performed 35% slower than a healthy reference group [28, 35]. Despite the
pharmacologically well medicated condition in our PD group, we found that the T-PEMF
treatment intervention improved the completion time in the STS-task by 19% to a level

Fig 4. Erythropoietin (EPO) and vascular endothelial growth factor (VEGF). Group mean (SD) of erythropoietin
and vascular endothelial growth factor concentration in cerebrospinal fluid (CSF) and plasma in patients with
Parkinson’s disease who were either treated with transcranial pulsed electromagnetic field treatment (T-PEMF, green
lines) or were in the PD-control group (red lines). Individual values are displayed.
comparable to the healthy reference group. Sub-division of the STS completion time into
movement components indicated that the reduced STS completion time was explained by
shorter times throughout the movement cycle.
We recently reported a positive effect of daily T-PEMF treatment for eight weeks on func-
tional rapid force production in patients with mild PD. This may reflect an increased thalamo-
cortical input to the motor cortex or increased excitability of the motor cortex, which might be
due to higher levels of dopamine [15]. The mechanistic link between T-PEMF, CSF-EPO, and
movement speed needs further investigation. However, as improved STS completion time
after T-PEMF treatment was accompanied by increased CSF-EPO, we speculate on a potential
connection between T-PEMF and cerebral levels of EPO, VEGF, and dopamine. EPO and
VEGF have a common transcription factor, hypoxia inducible factor-1α (HIF-1α), which is
upregulated in hypoxic conditions. At normal oxygen levels, HIF-1α is continuously and rap-
idly degraded. Oxygen-independent regulation of HIF-1α also occurs, however, and in nor-
moxic conditions, increased endogenous nitric oxide seems to reduce degradation and thus

stabilizes the HIF-1α level [36]. PEMF can induce nitric oxide mediated cerebral arteriolar
dilation, leading to increased microvascular blood flow and tissue oxygenation in healthy rats
[37]. It is thus plausible that our T-PEMF treatment up-regulated HIF-1α through increased
production of nitric oxide, resulting in up-regulation of EPO and VEGF. HIF-1α also regulates
tyrosine hydroxylase (TH), which is rate-limiting in dopamine synthesis [38]. TH levels are
reduced in the striatum of post mortem PD brains and may contribute to the lower striatal
dopamine production in patients with PD, although compensatory mechanisms seem to
increase the activity of the remaining TH [39]. T-PEMF may thus improve functional perfor-
mance by increasing the dopamine level in the brain mediated by EPO induced protection of
dopaminergic neurons, via increased TH enzyme activity. However, this does not exclude
additional influence of other growth factor mechanisms.
The present treatment effect on STS completion time was 3–4 times stronger after the lon-
ger treatment period (3x 8 weeks) than found in our previous study where the patients were
treated for eight weeks only, possibly because this was sufficient time to induce structural
changes in the brain while eight weeks were not [15]. We did apply to the Danish Medicines
Agency for a longer treatment period before the eight-week study. However, at this point we
were not allowed to treat the patients for longer than eight weeks due to lack of knowledge
regarding adverse events. Our previous study showed no statistical difference between the
treated group and the placebo group regarding reported adverse events. The present study sup-
ported this conclusion.
Furthermore, our previous study showed that an eight-week treatment period was benefi-
cial for patients with mild PD whereas a significant treatment effect was lacking among the
more severely affected participants [15]. However, the present study where the patients
received the same treatment (30 min/day) but for 3x8 weeks showed a beneficial treatment
effect for the entire study group including mild as well as more severely affected PD patients.
Methodological considerations
Whereas pharmacological treatment of PD is associated with significant adverse effects,
T-PEMF showed no significant adverse effects either over eight weeks [40] or over three peri-
ods of eight weeks as in the present study. T-PEMF was applied at home, and we found high
compliance both in our previous eight-week study (97.9%) [16] and in the present 3x8-week
study (97.6%). Home treatment is a major advantage as daily treatment in the clinic seems to
be unrealistic. We observed unchanged plasma EPO concentrations despite increased CSF lev-
els. This supports the results of animal studies showing that the blood brain barrier signifi-
cantly limits the passage of EPO. Plasma concentration of EPO can thus not be used as a
biomarker for CSF-EPO.
Perspectives
The present study contributed with new knowledge regarding the effect of long-term T-PEMF
treatment on movement speed and specific CSF biomarkers. To further understand the clinical
perspective as well as the physiological mechanisms of action of the new treatment modality
there is a need for knowledge on other outcome parameters such as e.g. Parkinson specific
tremor, fine motor function and motor control, in future studies performed with the same
T-PEMF intervention. Also, knowledge regarding duration of treatment effect in the follow-
up period as well as the optimal treatment period would be valuable from a clinical perspective.
Unfortunately, systematic investigation of the duration of treatment effect post treatment was
beyond the scope of the present study, but a patient performed writing tests during and after
the treatment period independently of the researchers. The patient’s handwriting ability

improved significantly after the treatment period compared to baseline and this lasted approxi-
mately three months after treatment completion, indicating a possible long-term effect [41].
As the treatment effect is expected to level off over time, a longer treatment period, or perhaps
life-long treatment may be considered for patients with PD. Finally, the effect of T-PEMF on
other neurodegenerative diseases could also be relevant to study.
Limitations
We included a PD-control group to allow comparison with the natural progression of the dis-
ease, but we did not have a PD-placebo group. As our previous study (RCT-trial) showed posi-
tive effects among patients with mild PD after eight-weeks of treatment, we decided for ethical
reasons not to include a placebo PD group in the current study with longer treatment period,
although this is a study limitation. However, since all outcome data were objectively measured,
we do not expect this to be a major limitation.
A specific sample size calculation was not performed as previous data on the effect of long-
term treatment did not exist. Our sample size considerations were based on previous experi-
ence with measurements of performance during the STS-task. The test-retest reliability of the
STS-task in PD patients is substantial [42]. However, a limitation is that only a subgroup of the
PD patients was included in the biomarker assessment, which increases the risk of type II
error. Finally, the participants were recruited by convenience, largely based on consecutive
recruitment from clinic (possible risk of allocation bias). Generalization should therefore be
performed with caution.
Conclusions
In patients with Parkinson’s disease, daily treatment with bipolar transcranial pulsed electro-
magnetic fields (T-PEMF) for three periods of eight weeks appeared to strengthen the natural
protective/compensatory response in the human brain.
Motor performance in terms of movement speed was improved markedly compared to the
natural progression seen in the PD-control group and CSF levels of erythropoietin increased
in the T-PEMF group. Post treatment values of the STS completion time corresponded to the
level of an age matched healthy reference group. These results suggest that T-PEMF has a
neuro-restorative and/or a neuroprotective effect and that T-PEMF treatment is a potential
innovative neuroprotective therapy in PD. However, the results do not exclude additional
influence of other growth factor mechanisms.
Supporting information
S1 Table. STS completion time (s).
(DOCX)
S2 Table. STS completion time (% of baseline (week 0)).
(DOCX)
S1 Data.
(XLSX)
S2 Data.
(XLSX)
S3 Data.
(XLSX)

S4 Data.
(XLSX)
Acknowledgments
The authors would like to thank Meaghan E. Spedden for her contribution to the data acquisi-
tion and Ole Gredal for his reflections and discussion regarding T-PEMF treatment in PD.
The technical assistance of Dorte Lyholmer and Ulla Munk with CSF and blood sample analy-
sis are gratefully acknowledged. Thank you to Claire Gudex for language proofing.
Author Contributions
Conceptualization: Bente Rona Jensen, Anne Sofie Bøgh Malling, Morten Meyer, Bo Mohr
Morberg, Lene Wermuth.
Data curation: Bente Rona Jensen, Anne Sofie Bøgh Malling, Lene Wermuth.
Formal analysis: Bente Rona Jensen, Anne Sofie Bøgh Malling, Sissel Ida Schmidt, Morten
Meyer.
Funding acquisition: Bente Rona Jensen.
Investigation: Bente Rona Jensen, Bo Mohr Morberg, Lene Wermuth.
Methodology: Bente Rona Jensen, Anne Sofie Bøgh Malling, Sissel Ida Schmidt.
Project administration: Bente Rona Jensen, Lene Wermuth.
Resources: Bente Rona Jensen, Lene Wermuth.
Software: Anne Sofie Bøgh Malling.
Supervision: Bente Rona Jensen.
Validation: Bente Rona Jensen, Anne Sofie Bøgh Malling, Morten Meyer, Bo Mohr Morberg.
Writing – original draft: Bente Rona Jensen.
Writing – review & editing: Anne Sofie Bøgh Malling, Sissel Ida Schmidt, Morten Meyer, Bo
Mohr Morberg, Lene Wermuth.
References
1. Politis M, Wu K, Molloy S, P GB, Chaudhuri KR, Piccini P. Parkinson’s disease symptoms: the patient’s
perspective. Mov Disord. 2010; 25(11):1646–51. https://doi.org/10.1002/mds.23135 PMID: 20629164.
2. Rose MH, Løkkegaard A, Sonne-Holm S, Jensen BR. Isometric Tremor Irregularity, Torque Steadiness
and Rate of Force Development in Parkinson’s Disease. Motor Control. 2013; 17:203–16. https://doi.
org/10.1123/mcj.17.2.203 PMID: 23155115
3. Hammond KG, Pfeiffer RF, LeDoux MS, Schilling BK. Neuromuscular rate of force development deficit
in Parkinson disease. Clin Biomech (Bristol, Avon). 2017; 45:14–8. https://doi.org/10.1016/j.
clinbiomech.2017.04.003 PMID: 28432901.
4. Swinnen SP, Van LL, Verschueren S, Peeters G, Dom R, De WW. Interlimb coordination deficits in
patients with Parkinson’s disease during the production of two-joint oscillations in the sagittal plane.
Mov Disord. 1997; 12(6):958–68. https://doi.org/10.1002/mds.870120619 PMID: 9399221
5. Crouse JJ, Phillips JR, Jahanshahi M, Moustafa AA. Postural instability and falls in Parkinson’s disease.
Rev Neurosci. 2016; 27(5):549–55. Epub 2016/03/12. https://doi.org/10.1515/revneuro-2016-0002
PMID: 26966928.
6. Petzinger GM, Holschneider DP, Fisher BE, McEwen S, Kintz N, Halliday M, et al. The Effects of Exer-
cise on Dopamine Neurotransmission in Parkinson’s Disease: Targeting Neuroplasticity to Modulate
Basal Ganglia Circuitry. Brain Plast. 2015; 1(1):29–39. https://doi.org/10.3233/bpl-150021 PMID:
26512345; PubMed Central PMCID: PMC4621077.

7. Lange KW, Mecklinger L, Walitza S, Becker G, Gerlach M, Naumann M, et al. Brain dopamine and kine-
matics of graphomotor functions. Hum Mov Sci. 2006; 25(4–5):492–509. Epub 2006/07/25. https://doi.
org/10.1016/j.humov.2006.05.006 PMID: 16859791.
8. Contreras-Vidal JL, Stelmach GE. Effects of Parkinsonism on motor control. Life Sci. 1996; 58(3):165–
76. Epub 1996/01/01. https://doi.org/10.1016/0024-3205(95)02237-6 PMID: 9499157.
9. Caligiore D, Helmich RC, Hallett M, Moustafa AA, Timmermann L, Toni I, et al. Parkinson’s disease as
a system-level disorder. NPJ Parkinsons Dis. 2016; 2:16025. Epub 2017/07/21. https://doi.org/10.1038/
npjparkd.2016.25 PMID: 28725705; PubMed Central PMCID: PMC5516580.
10. Rahbek UL, Tritsaris K, Dissing S. Interaction of Low-frequency, Pulsed Electromagnetic Fields with
Living Tissue: Biochemical Responses and Clinical Results. Oral Bioscience and Medicine. 2005; 2
(1):1–12.
11. Hei WH, Byun SH, Kim JS, Kim S, Seo YK, Park JC, et al. Effects of electromagnetic field (PEMF) expo-
sure at different frequency and duration on the peripheral nerve regeneration: in vitro and in vivo study.
Int J Neurosci. 2016; 126(8):739–48. https://doi.org/10.3109/00207454.2015.1054032 PMID:
26010211.
12. Urnukhsaikhan E, Cho H, Mishig-Ochir T, Seo YK, Park JK. Pulsed electromagnetic fields promote sur-
vival and neuronal differentiation of human BM-MSCs. Life Sci. 2016; 151:130–8. https://doi.org/10.
1016/j.lfs.2016.02.066 PMID: 26898125.
13. Lei T, Jing D, Xie K, Jiang M, Li F, Cai J, et al. Therapeutic effects of 15 Hz pulsed electromagnetic field
on diabetic peripheral neuropathy in streptozotocin-treated rats. PLoS One. 2013; 8(4):e61414. Epub
2013/05/03. https://doi.org/10.1371/journal.pone.0061414 PMID: 23637830; PubMed Central PMCID:
PMC3630223.
14. Pan Y, Dong Y, Hou W, Ji Z, Zhi K, Yin Z, et al. Effects of PEMF on microcirculation and angiogenesis
in a model of acute hindlimb ischemia in diabetic rats. Bioelectromagnetics. 2013; 34(3):180–8. Epub
2012/09/07. https://doi.org/10.1002/bem.21755 PMID: 22952041.
15. Malling ASB, Morberg BM, Wermuth L, Gredal O, Bech P, Jensen BR. Effect of transcranial pulsed
electromagnetic fields (T-PEMF) on functional rate of force development and movement speed in per-
sons with Parkinson’s disease: A randomized clinical trial. PLoS One. 2018; 13(9):e0204478. https://
doi.org/10.1371/journal.pone.0204478 PMID: 30252895; PubMed Central PMCID: PMC6155540.
16. Morberg BM, Malling AS, Jensen BR, Gredal O, Bech P, Wermuth L. Parkinson’s disease and transcra-
nial pulsed electromagnetic fields: A randomized clinical trial. Mov Disord. 2017; 32(4):625–6. https://
doi.org/10.1002/mds.26927 PMID: 28116825.
17. Malling ASB, Morberg BM, Wermuth L, Gredal O, Bech P, Jensen BR. The effect of 8 weeks of treat-
ment with transcranial pulsed electromagnetic fields on hand tremor and inter-hand coherence in per-
sons with Parkinson’s disease. J Neuroeng Rehabil. 2019; 16(1):19. Epub 2019/02/02. https://doi.org/
10.1186/s12984-019-0491-2 PMID: 30704504.
18. Noguchi CT, Asavaritikrai P, Teng R, Jia Y. Role of erythropoietin in the brain. Crit Rev Oncol Hematol.
2007; 64(2):159–71. Epub 2007/05/08. https://doi.org/10.1016/j.critrevonc.2007.03.001 PMID:
19. Byts N, Siren AL. Erythropoietin: a multimodal neuroprotective agent. Exp Transl Stroke Med. 2009;
1:4. Epub 2010/02/10. https://doi.org/10.1186/2040-7378-1-4 PMID: 20142991; PubMed Central
PMCID: PMC2816866.
20. Carelli S, Giallongo T, Gombalova Z, Merli D, Di Giulio AM, Gorio A. EPO-releasing neural precursor
cells promote axonal regeneration and recovery of function in spinal cord traumatic injury. Restor Neurol
Neurosci. 2017; 35(6):583–99. Epub 2017/11/25. https://doi.org/10.3233/RNN-170750 PMID:
21. Hamidi G, Arabpour Z, Shabrang M, Rashidi B, Alaei H, Sharifi MR, et al. Erythropoietin improves spa-
tial learning and memory in streptozotocin model of dementia. Pathophysiology. 2013; 20(2):153–8.
Epub 2013/05/07. https://doi.org/10.1016/j.pathophys.2013.01.001 PMID: 23642847.
22. Carelli S, Giallongo T, Viaggi C, Latorre E, Gombalova Z, Raspa A, et al. Recovery from experimental
parkinsonism by intrastriatal application of erythropoietin or EPO-releasing neural precursors. Neuro-
pharmacology. 2017; 119:76–90. Epub 2017/04/05. https://doi.org/10.1016/j.neuropharm.2017.03.035
PMID: 28373075.
23. Qi C, Xu M, Gan J, Yang X, Wu N, Song L, et al. Erythropoietin improves neurobehavior by reducing
dopaminergic neuron loss in a 6hydroxydopamineinduced rat model. Int J Mol Med. 2014; 34(2):440–
50. Epub 2014/06/19. https://doi.org/10.3892/ijmm.2014.1810 PMID: 24939444; PubMed Central
PMCID: PMC4094589.
24. Shim JW, Madsen JR. VEGF Signaling in Neurological Disorders. Int J Mol Sci. 2018; 19(1). Epub
2018/01/18. https://doi.org/10.3390/ijms19010275 PMID: 29342116; PubMed Central PMCID:
PMC5796221.

25. Carmeliet P, Storkebaum E. Vascular and neuronal effects of VEGF in the nervous system: implications
for neurological disorders. Semin Cell Dev Biol. 2002; 13(1):39–53. Epub 2002/04/24. https://doi.org/
10.1006/scdb.2001.0290 PMID: 11969370.
26. Fahn S, Elton RL, Committee. MotUD. Unified Parkinson’s Disease Rating Scale. In: Fahn S, Marsden
CD, Calne DB, Goldstein M, editors. Recent development in Parkinson’s disease. 2 . Florham Park,
NJ: Macmillan Health Care Information; 1987. p. 153–64.
27. Spedden ME, Malling ASB, Andersen KK, Jensen BR. Association Between Gross-Motor and Execu-
tive Function Depends on Age and Motor Task Complexity. Dev Neuropsychol. 2017; 42(7–8):495–
506. https://doi.org/10.1080/87565641.2017.1399129 PMID: 29161178.
28. Malling AS, Jensen BR. Motor intensive anti-gravity training improves performance in dynamic balance
related tasks in persons with Parkinson’s disease. Gait & posture. 2016; 43:141–7. https://doi.org/10.
1016/j.gaitpost.2015.09.013 PMID: 26444077.
29. Hinkle DE, Wiersma W, Jurs SG. Applied Satistics for the Behavioral Sciences. 5th ed. Boston:
Houghton Mifflin; 2003.
30. Yang LZ, Li RL, Wu B, Luo J, Chen Z, Tan XPY, et al. Point of View about Erythropoietin in Treatment of
Central Nervous System Diseases. Journal of Alzheimers Disease & Parkinsonism. 2017; 7(2):1–4.
https://doi.org/10.4172/2161-0460.1000320
31. Xue YQ, Ma BF, Zhao LR, Tatom JB, Li B, Jiang LX, et al. AAV9-mediated erythropoietin gene delivery
into the brain protects nigral dopaminergic neurons in a rat model of Parkinson’s disease. Gene Ther.
2010; 17(1):83–94. Epub 2009/09/04. https://doi.org/10.1038/gt.2009.113 PMID: 19727138.
32. Widl K, Brettschneider J, Schattauer D, Sussmuth S, Huber R, Ludolph AC, et al. Erythropoietin in cere-
brospinal fluid: age-related reference values and relevance in neurological disease. Neurochem Res.
2007; 32(7):1163–8. Epub 2007/04/03. https://doi.org/10.1007/s11064-007-9286-0 PMID: 17401663.
33. Jang W, Park J, Shin KJ, Kim JS, Kim JS, Youn J, et al. Safety and efficacy of recombinant human
erythropoietin treatment of non-motor symptoms in Parkinson’s disease. Journal of the neurological sci-
ences. 2014; 337(1–2):47–54. Epub 2013/12/03. https://doi.org/10.1016/j.jns.2013.11.015 PMID:
24289887.
34. Xue YQ, Zhao LR, Guo WP, Duan WM. Intrastriatal administration of erythropoietin protects dopaminer-
gic neurons and improves neurobehavioral outcome in a rat model of Parkinson’s disease. Neurosci-
ence. 2007; 146(3):1245–58. Epub 2007/03/17. https://doi.org/10.1016/j.neuroscience.2007.02.004
PMID: 17363174.
35. Rose MH, Lokkegaard A, Sonne-Holm S, Jensen BR. Improved Clinical Status, Quality of Life, and
Walking Capacity in Parkinson’s Disease After Body Weight-Supported High-Intensity Locomotor Train-
ing. ArchPhysMedRehabil. 2012. S0003-9993(12)01186-0 [pii]; https://doi.org/10.1016/j.apmr.2012.11.
025 PMID: 23187043
36. Movafagh S, Crook S, Vo K. Regulation of hypoxia-inducible factor-1a by reactive oxygen species: new
developments in an old debate. J Cell Biochem. 2015; 116(5):696–703. Epub 2014/12/30. https://doi.
org/10.1002/jcb.25074 PMID: 25546605.
37. Bragin DE, Statom GL, Hagberg S, Nemoto EM. Increases in microvascular perfusion and tissue oxy-
genation via pulsed electromagnetic fields in the healthy rat brain. J Neurosurg. 2015; 122(5):1239–47.
https://doi.org/10.3171/2014.8.JNS132083 PMID: 25343187.
38. Lim J, Kim HI, Bang Y, Seol W, Choi HS, Choi HJ. Hypoxia-inducible factor-1alpha upregulates tyrosine
hydroxylase and dopamine transporter by nuclear receptor ERRgamma in SH-SY5Y cells. Neuroreport.
2015; 26(6):380–6. Epub 2015/03/26. https://doi.org/10.1097/WNR.0000000000000356 PMID:
25807177.
39. Nagatsu T, Nagatsu I. Tyrosine hydroxylase (TH), its cofactor tetrahydrobiopterin (BH4), other catechol-
amine-related enzymes, and their human genes in relation to the drug and gene therapies of Parkin-
son’s disease (PD): historical overview and future prospects. J Neural Transm (Vienna). 2016; 123
(11):1255–78. Epub 2016/08/06. https://doi.org/10.1007/s00702-016-1596-4 PMID: 27491309.
40. Morberg BM, Malling AS, Jensen BR, Gredal O, Bech P, Wermuth L. Effects of transcranial pulsed elec-
tromagnetic field stimulation on quality of life in Parkinson’s disease. Eur J Neurol. 2018; 25(7):963–
e74. https://doi.org/10.1111/ene.13637 PMID: 29573167.
41. Jensen BR, Malling ASB, Morberg BM, Gredal O, Bech P, Wermuth L. Effects of Long-Term Treatment
with T-PEMF on Forearm Muscle Activation and Motor Function in Parkinson’s Disease. Case Rep
Neurol. 2018; 10(2):242–51. Epub 2018/10/05. https://doi.org/10.1159/000492486 PMID: 30283322;
PubMed Central PMCID: PMC6167712.
42. Rose MH. Rehabilitation in Parkinson’s Disease (PhD thesis): University of Copenhagen, Department
of Nutrition, Exercise and Sports; 2013.

Artículo15 - Parkinson

Uploaded by

Copyright:

Available Formats

Artículo15 - Parkinson

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Artículo15 - Parkinson

Uploaded by

Copyright:

Available Formats

PLOS ONE

Long-term treatment with transcranial pulsed

1 Department of Neurology, Odense University Hospital, University of Southern Denmark, Odense,

Editor: J. Lucas McKay, Emory University, UNITED Objective

Published: April 28, 2021

Funding: The research group received funding

PLOS ONE | https://doi.org/10.1371/journal.pone.0248800 April 28, 2021 1 / 15

PLOS ONE | https://doi.org/10.1371/journal.pone.0248800 April 28, 2021 2 / 15

Materials and methods

PLOS ONE | https://doi.org/10.1371/journal.pone.0248800 April 28, 2021 3 / 15

PLOS ONE | https://doi.org/10.1371/journal.pone.0248800 April 28, 2021 4 / 15

PLOS ONE | https://doi.org/10.1371/journal.pone.0248800 April 28, 2021 5 / 15

PLOS ONE | https://doi.org/10.1371/journal.pone.0248800 April 28, 2021 6 / 15

PLOS ONE | https://doi.org/10.1371/journal.pone.0248800 April 28, 2021 7 / 15

PLOS ONE | https://doi.org/10.1371/journal.pone.0248800 April 28, 2021 8 / 15

PLOS ONE | https://doi.org/10.1371/journal.pone.0248800 April 28, 2021 9 / 15

PLOS ONE | https://doi.org/10.1371/journal.pone.0248800 April 28, 2021 10 / 15

PLOS ONE | https://doi.org/10.1371/journal.pone.0248800 April 28, 2021 11 / 15

PLOS ONE | https://doi.org/10.1371/journal.pone.0248800 April 28, 2021 12 / 15

PLOS ONE | https://doi.org/10.1371/journal.pone.0248800 April 28, 2021 13 / 15

PLOS ONE | https://doi.org/10.1371/journal.pone.0248800 April 28, 2021 14 / 15

PLOS ONE | https://doi.org/10.1371/journal.pone.0248800 April 28, 2021 15 / 15

You might also like