Key words
tDCS - transcranial direct current - deep brain stimulation - chronic pain - neuromodulation
Introduction
Over the past decades, various methods of neuromodulatory stimulation have been established.
Of these, the stimulation of the primary motor cortex (M1) and the dorsolateral prefrontal
cortex (DLPFC) using transcranial direct current stimulation (tDCS) and repetitive
transcranial magnetic stimulation (rTMS) are of particular interest. Both techniques
allow the induction and modulation of neuroplastic changes in the cerebrum through
modification of neuronal activity or resting membrane potential. As the two techniques
primarily permit stimulation of cortical areas, they are less suitable for direct
stimulation of deeper structures, such as the thalamus. However, functional magnetic
resonance imaging (fMRI) and positron emission tomography (PET) studies showed that
multiple deeper structures can be reached indirectly by (superficial or epidural)
stimulation of M1 or DLPFC [1]
[2]
[3].
TMS uses the magnetic field of a coil to induce a brief electrical current in the
brain, while tDCS uses constant direct current delivered via electrodes on the scalp.
The primary effect of tDCS is most likely caused by de - or hyperpolarization of neuronal
membrane potentials. The aftereffects of the stimulation can be influenced by pharmacological
interventions either positively or negatively [4]
[5]. Response to tDCS in patients with depression can be enhanced by co-application
of serotonin reuptake inhibitors [6]
[7]. In M1, D-cycloserine (glycine receptor agonist and thus indirect NMDA receptor
agonist) can prolong tDCS aftereffects from 1 h to 24 h [5]; however, this has not yet been evaluated in pain therapy.
Application of rTMS and tDCS in pain patients
Application of rTMS and tDCS in pain patients
Chronic pain induces a change of the anatomical pain network, primarily comprising
DLPFC, M1, somatosensory cortex, the anterior cingulate cortex, and the thalamus.
In patients with chronic back pain, a progressive decrease in neocortical gray matter
(DLPFC) of 5–11% compared to healthy subjects is observed over the years [8]. Per pain year the loss of gray matter density was approximately 1.3 cm3.
The use of rTMS and tDCS in chronic pain patients has evolved from the concept of
invasive motor cortex stimulation (MCS) using epidurally implanted electrodes in patients
with chronic thalamus pain [9]
[10]
[11]. The MCS effect is explained by inhibition of both nociceptive afferent fibers and
affective-motivational pain components via the medial thalamus and anterior cingulate
cortex. The empirically established stimulation parameter space differs between rTMS
(typically 5 to 20 Hz) and MCS (20 Hz and higher frequencies).
Patients with treatment-resistant pain are typically stimulated for 20 min per session
with 'excitability-enhancing' 10 Hz over M1 [12]
[13]. In line with MCS data, it was shown that most likely not an inhibition but an activation
of M1 via its projection to the thalamus inhibits afferent pain fibers ascending there
[14]. In this way, activation of thalamic nuclei can in turn modify the activity of other
pain-associated structures (e.g., anterior cingulate cortex, periaqueductal gray)
at least temporarily and thus inhibit e.g., pain of primarily spinal origin.
A PubMed search (search terms: rTMS/TBS (theta burst stimulation) and neuropathic/neurogenic
pain) currently identified 68 studies, including 19 placebo-controlled studies with
at least 10 patients with chronic neuropathic pain, receiving active low-frequency
(LF) or high-frequency (HF) rTMS over M1. The analysis included a total of 688 patients
[15]. Stimulation was always applied contralaterally to the pain site. The conclusions
drawn from this analysis are in line with those from other reviews (e.g., [16]
[17]
[18]): (i) LF rTMS over M1 is most likely not effective; (ii) HF rTMS over M1 results
in pain relief of more than 50% only in one-third of the patients; (iii) repeated
HF rTMS sessions (5–10 sessions, 1 session/day) increase the analgesic effect [15]. The effectiveness of one single HF rTMS session usually lasts for a few days and
can be enhanced and extended by repeated stimulation [16]. A most effective protocol (site of stimulation, stimulation frequency, number of
impulses per session and number of sessions) has not been determined yet. A predictive
value of HF rTMS over M1 appears to show a positive correlation with invasive MCS
[14]
[19]
[20]
[21]
[22].
TBS (grouped stimulation with 3 pulses at 20-ms intervals, repetitive with 5 Hz) has
either an inhibitory (cTBS, continuous) or facilitatory (iTBS, intermittent with 8-second
pause after stimulations of 2 s each) action [23]. According to current data, the shorter, 'excitability-increasing' iTBS protocol
is apparently not capable of inducing an analgesic effect. In addition, the responsiveness
of varies types of neuropathic pain was not further differentiated [24]
[25]
[26]
[27]. Stimulation of DLPFC is based on the concept that, besides chronic pain, a positive
effect is exerted on the depression component (e.g., [15]
[28]).
In chronic pain, tDCS is less effective compared to HF rTMS. A PubMed search [search
terms: tDCS AND (pain or migraine)] identified 269 studies, including 62 clinical
studies with 1426 patients [29]. Central neuropathic pain conditions (including central pain after stroke and traumatic
spinal injury), peripheral neuropathic pain, musculoskeletal pain (including fibromyalgia
and myofascial pain); migraine, orofacial pain, chronic back pain (of lower back)
and abdominal or pelvic pain were included. Application of stimulation was either
to the contralateral M1 (anodal) or, in pain without side preferences, the dominant
hemisphere of the left DLPFC or the primary visual cortex in migraine (cathodal) [30]
[31]. In migraine, a cathodal inhibition of hyperexcitability of the primary visual cortex
is assumed.
Most studies used a stimulation intensity of 1 - 2 mA and an electrode size of 35 cm².
Session duration ranged from 10 min to 20 min/day and sessions were repeated on up
to 20 consecutive days. The most common protocol consisted of anodal stimulation over
M1 for 20 min on five consecutive days. The analgesic effect lasted for 2 to 6 weeks
[32]
[33]
[34]
[35]
[36]
[37].
The evidence grade attributed to direct current stimulation is considered lower compared
to that of rTMS, primarily due to the higher numbers of patients in rTMS studies [14]
[15]: grade A for rTMS in the treatment of chronic neuropathic pain of the head or upper
extremities and grade B for tDCS in the treatment of fibromyalgia and neuropathic
pain of the lower extremities (an evidence grade A for tDCS cannot be established
due to insufficient number of studies). Only one placebo-controlled study directly
compared the analgesic effect of the two approaches [38]. In patients with neuropathic pain, however, rTMS (3 stimulations/day) relieved
the pain associated with lumbosacral radiculopathy, whereas anodal tDCS did not.
Invasive Pain Management
One disadvantage of the non-invasive techniques is the lack of persistent response
to treatment. If the duration of pain relief achieved by tDCS and rTMS is too short
or if no adequate pain control can be achieved, the next step is to evaluate minimally
invasive extracranial surgical interventions. Here, spinal cord stimulation (SCS)
and peripheral nerve field stimulation (sPNS) are of special interest. In patients
with failed back surgery syndrome (FBSS) and peripheral neuropathic pain, SCS was
used with good response. SCS treatment of pain syndromes after incomplete spinal cord
injuries is less successfull. In regional pain conditions, sPNS plays an increasingly
important role and recent attempts of treating trigeminal pain with sPNS have been
successful [39]. Should these surgical procedures fail, two invasive brain stimulation methods are
available as a last resort.
1. With MCS, continuous stimulation of M1 in the area of the cortical representation
of the pain region is applied. To this end, one or more electrodes are placed in most
cases epidurally, but rarely also subdurally, either via burr-hole trepanation or
via craniotomy. Neuronavigation and functional magnetic resonance imaging can be used
to increase precision of electrode positioning. With neither a standardized technique
to perform the procedure nor uniform stimulation parameters available, the comparability
of clinical studies is limited [40]. In the meantime, a consensus has been established that prior to implantation of
a pulse generator, the stimulation effect should be tested over a period of several
days. However, here again there is no standardization with regard to trial period
duration, stimulation parameters to be tested and need for placebo stimulation [41]
[42]. A longer trial period may lead to an increase of infectious complications. MCS
was used in numerous central and peripheral pain syndromes with variable success [43]. Best and most reproducible are the results with trigeminal neuropathic pain [41]
[44]. Likewise, good results were achieved with pain after spinal cord injury and phantom
pain, whereas the success rates in patients with central pain and plexus injuries
appear to be lower [45]. Except for trigeminal neuropathic pain, it is difficult to draw final conclusions
about suitable indications for MCS due to the limited comparability of the studies
available.
2. Deep brain stimulation (DBS) has also been used as an invasive pain therapy to
treat central and peripheral pain syndromes, using various target regions (sensory
thalamus, periaqueductal gray/periventricular zone, anterior cingulate cortex). Apart
from the target area selected, the surgical technique does not differ from functional
stereotactic procedures performed for movement disorders. Implantation of the pulse
generator is performed based on the results of a trial period extending over several
days, for which no standards have yet been defined, as in MCS [46]. Good results have been achieved especially in patients with peripheral neuropathic
pain or failed back surgery syndrome (FBSS). Pain control achieved with DBS in patients
with central pain—both after stroke and spinal injury—appears to be less favorable
[47]. However, here again there is a lack of prospective, randomized trials which would
allow a conclusive evaluation of the indications. From the few studies comparing MCS
with DBS in the treatment of various pain syndromes, it appeared that MCS was clinically
superior to DBS [40].
MCS and DBS are techniques which should only be used if conservative and less invasive
surgical interventions have failed. Given the complexity of establishing an indication
and their comparative rarity, these surgical procedures should only be performed at
specialized centers.
Summary and outlook
The existing clinical phase III studies with chronic pain patients suggest a potential
usefulness of non-invasive transcranial stimulation. The published pain relief rates
of up to 60% are in line with those from pharmacotherapy-evaluating studies; however,
it should be noted that the patients recruited for non-invasive transcranial stimulation
studies were otherwise drug- or treatment-resistant pain patients, e.g., patients
with fibromyalgia or chronic neuropathic pain of the head and upper extremities [15]
[29]. tDCS offers special feasibility advantages compared to rTMS, as the devices are
affordable and can be used at home. After the signature of the mandatory medical informed
consent and instructions on how to use the device, non-invasive transcranial stimulation
can be applied under supervision of medically experienced technical staff (e.g., MTAs)
in the further course of treatment. In clinical practice, trial stimulation would
be applied on three consecutive days; in case no clinical improvement is observed,
the (currently still) compassionate use of non-invasive transcranial stimulation in
these patients would be stopped.
Placebo rate is high in rTMS and tDCS studies, a phenomenon generally noted in pain
studies. Complete blinding of the skin receptor stimulation during rTMS is difficult
and some authors postulate that it can only be achieved by electrical costimulation.
The tingling sensation underneath the tDCS electrodes decreases over time; the so-called
fade-in/fade-out placebo protocol applies stimulation only at the beginning and at
the end of the stimulation session. With current intensities of 2 mA instead of 1
mA this becomes even more difficult because the active stimulation group experiences
light tingling during the entire stimulation period.
In case of failure of non-invasive techniques, MCS is a good therapeutic option, especially
in patients with trigeminal neuropathic pain. MCS can also achieve good results in
patients with pain after spinal cord injury and phantom limb pain. DBS is a treatment
option of lesser importance according to current data. Definite conclusions about
which indications should be regarded as promising or less promising for MCS or DBS
cannot be drawn due to insufficient data.
