Pain Medicine, 2025, 00, 1–10

https://doi.org/10.1093/pm/pnaf001
Advance access publication 11 January 2025
Review Article

Deep brain stimulation and motor cortex stimulation

for central post-stroke pain: a systematic review and
meta-analysis

Downloaded from https://academic.oup.com/painmedicine/advance-article/doi/10.1093/pm/pnaf001/7951886 by guest on 16 March 2025

Siddarth Kannan �,1, Conor S. Gillespie, MPhil2, Jeremy Hanemaaijer3,4,5, John Eraifej, PhD4,5,
Andrew F. Alalade, PhD1,6, Alex Green , PhD4,5
1
School of Medicine, University of Central Lancashire, Preston PR1 7BH, United Kingdom
2
Department of Clinical Neurosciences, University of Cambridge, Cambridge CB2 1PG, United Kingdom
3
Department of Neurosurgery, RadboudUMC, Nijmegen 6525GA, The Netherlands
4
Oxford Functional Neurosurgery Group, John Radcliffe Hospital, Oxford OX39DU, United Kingdom
5
Nuffield Department of Surgical Sciences, University of Oxford, Oxford OX1 2JD, United Kingdom
6
Department of Neurosurgery, Royal Preston Hospital, Preston PR2 9HT, United Kingdom
�Corresponding author: School of Medicine, 135A Adelphi Street, Preston PR1 7BH, United Kingdom. Email: skannan@uclan.ac.uk

Abstract
Introduction: Deep brain stimulation (DBS) and motor cortex stimulation (MCS) are invasive interventions in order to treat various neuropathic
pain syndromes such as central post-stroke pain (CPSP). While each treatment has varying degree of success, comparative analysis has not yet
been performed, and the success rates of these techniques using validated, objective pain scores have not been synthesized.
Methods: A systematic review and meta-analysis was conducted in accordance with PRISMA guidelines. Three databases were searched, and
articles published from January 2000 to October 2024 were included (last search date October 25, 2024). Meta-Analysis was performed using
random effects models. We evaluated the performance of DBS or MCS by assessing studies that reported pain relief using visual analogue
scale (VAS) or numerical rating scale (NRS) scores.
Results: Of the 478 articles identified, 32 were included in the analysis (330 patients—139 DBS and 191 MCS). The improvement in mean VAS
score for patients that underwent DBS post-surgery was 48.6% compared to a score of 53.1% for patients that had MCS. The pooled number
of patients who improved after DBS was 0.62 (95% CI, 0.51–0.71, I2 ¼ 16%). The pooled number of patients who improved after MCS was
0.64 (95% CI, 0.53–0.74, I2 ¼ 40%).
Conclusion: The use of neurosurgical interventions such as DBS and MCS are last-resort treatments for CPSP, with limited studies exploring
and comparing these two techniques. While our study shows that MCS might be a slightly better treatment option, further research would
need to be done to determine the appropriate surgical intervention in the treatment of CPSP.
Keywords: neuromodulation; post-stroke pain; deep-brain stimulation; motor cortex stimulation.

Introduction Currently, pharmacological treatments, such as anticonvul­

sants (eg, gabapentin), selective serotonin reuptake inhibitors
Central post-stroke pain (CPSP) is one of the most challeng­
(SSRIs), and antidepressants like amitriptyline, are commonly
ing and distressing complications that stroke survivors face
used to manage CPSP. However, these medications are often
during recovery, affecting approximately 10%-39% of stroke
associated with significant side effects, especially at higher
patients.1,2 The onset of CPSP typically occurs within 1-
doses, and many patients are unable to tolerate these treat­
3 months following a stroke, with the majority of cases mani­ ments.6 As a result, there is a growing need for alternative
festing symptoms by 6 months.3 This condition is character­ therapeutic options.
ized by chronic pain resulting from damage to the central In recent years, invasive neuromodulation techniques have
nervous system, which severely impacts patients’ quality of emerged as promising alternatives to manage CPSP, with deep
life. The duration of CPSP can be chronic, lasting for months brain stimulation (DBS) and motor cortex stimulation (MCS)
or even years after the initial stroke event. Studies suggest at the forefront of innovative interventions.7 Both techniques
that CPSP affects approximately 8%-35% of stroke patients, involve the application of electrical impulses to specific brain
with pain often persisting long after the stroke.4 The type of areas, aiming to modulate neural circuitry and provide relief
pain in CPSP is typically described as sharp, paroxysmal, and from chronic pain. However, the targeting areas for both
often localized to the hemiplegic side, though some patients interventions are different. In MCS, two electrode leads are
report a more diffuse pain experience.5 Research indicates placed over the motor and sensory cortices, while in DBS sur­
that stroke survivors with chronic pain often exhibit higher gery, deeper located brain structures are targeted, such as the
levels of depression and anxiety, which can exacerbate pain ventral posterolateral (VPL) and ventral posteromedial (VPM)
perception and complicate treatment.2 nuclei of the thalamus, the periventricular and periaqueductal

Received: 30 June 2024. Revised: 22 November 2024. Accepted: 6 January 2025

© The Author(s) 2025. Published by Oxford University Press on behalf of the American Academy of Pain Medicine.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which
permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.
2 Pain Medicine, 2025, Vol. 00, No. 0

grey matter (PVG/PAG), or the rostral anterior cingulate comparability, and outcome (for cohort studies) or Exposure
cortex (ACC).8 (for case–control studies). Each of these criteria includes sev­
While some studies have showcased this potential benefit, eral sub-criteria, with points awarded to studies based on
pooled comparative analysis has not yet been performed and how well they meet each criterion.
the success rates of these techniques using validated, objective
pain scores have not been synthesized. In this systematic
review and meta-analysis, we aim to analyze the effect on Statistical analysis
pain relief offered by MCS and DBS on patients with CPSP Baseline characteristics were presented as descriptive frequen­
using clearly defined outcomes. cies. For meta-analysis, we used random effects models of

Downloaded from https://academic.oup.com/painmedicine/advance-article/doi/10.1093/pm/pnaf001/7951886 by guest on 16 March 2025

variables and endpoints, with pooled proportions used for
Methods the reduction of pain scores using VAS as a continuous out­
come measure. We evaluated the performance of DBS or
Search strategy and selection criteria
MCS by assessing studies that reported pain relief using VAS
We conducted this systematic review and meta-analysis or NRS scores. Pain improvement was defined as a reduction
according to the Preferred Reporting Items for Systematic
of ≥30% on the VAS or NRS score. This is considered clini­
Reviews and Meta-Analyses (PRISMA) Guidelines.9
cally significant, as previous studies have indicated that
We searched PubMed, Embase and Medline database of
reductions in pain scores of around 30%-40% are needed to
systematic reviews for full-text articles published in English
reflect clinically useful improvements.12,13 The 30% cut-off
(Search date October 25, 2024). Search terms used a combi­
nation of the terms “Central Post-Stroke Pain,” “Deep Brain for defining improvement was consistently applied across all
Stimulation,” and “Motor Cortex Stimulation,” and their studies included in this analysis.
associated synonyms. The full search strategy for all data­ The total number of patients in each study, along with the
bases can be found in Supplementary Tables S1–S3. The number of patients who experienced pain improvement of
Population, Intervention, Comparator, Outcome, Study ≥30%, was extracted. These data were then pooled by calcu­
Design (PICOS) criteria was used (Supplementary Table S4). lating proportions representing the percentage of patients
Furthermore, excluded reviews and the reference list of experiencing clinically significant pain relief. The pooled pro­
retrieved articles were cross-referenced for enriching and portions were aggregated across studies using random effects
completing the included database. We included studies of models to provide an overall estimate. A 95% confidence
adults (≥18 years) that specifically mentioned the use of interval (CI) was calculated to account for study variability
either DBS or MCS for the treatment of CPSP. We excluded and quantify the uncertainty around the estimated propor­
studies that reported exclusively pediatric populations, and tions. This approach allowed us to evaluate the clinical effec­
studies that examined other forms of neuropathic pain such tiveness of the interventions across different studies,
as trigeminal neuralgia, diabetic and peripheral neuropathy. providing a measure of precision for the pooled estimates and
We excluded studies that were conference abstracts or if the reflecting the range within which the true proportion is
primary language was not English. expected to lie 95% of the time.
Two reviewers (S.K. and C.S.G.) independently screened We carried out an additional sensitivity analysis by selec­
titles, abstracts and full texts to include articles. If reviewers tively removing studies at high risk of bias, then re-running
failed to reach consensus, a third author was sought for the meta-analysis.
clarification. Data analysis of descriptive statistics was performed using
the software Statistical Package for the Social Sciences (ver­
Data extraction
sion 27; IBM; Armonk; NY). R statistics (Rstudio Version
Data extraction was completed by two authors independently 4.0.1) was used to perform a meta-analysis and create figures,
(S.K. and C.S.G.). The following data were extracted from forest, and funnel plots (ggplot2 and meta-packages).
included studies: Year published, journal, type of study For each random effects model, we tested heterogeneity
(Randomized Control Trial [RCT] or observational study),
using the maximum restricted likelihood estimator.
single/multi center, number of patients with CPSP, and num­
Prevalence was calculated using pooled proportions methods
ber of these patients that underwent either MCS or DBS,
using the inverse variance method. The I2 statistic was used
number of patients that saw an improvement in pain, mean
to quantify the percentage of total variation across studies
postoperative visual analogue scale (VAS) or numerical rating
scale (NRS) scores and mean follow up time. The VAS or that is due to heterogeneity rather than chance.14 I2 values
NRS score was used as primary outcome metric in this study, were interpreted as follows: 0%-25%: Low heterogeneity,
offering a validated and standardized measure of pain inten­ 26%-50%: Moderate heterogeneity, >50%: Significant het­
sity; the scale in each study was measured in 0-10. Any study erogeneity.15 Heterogeneity was considered significant when
that did not report individual patient NRS/VAS improvement I2 > 50% and the P value < 0.1. In cases of significant heter­
scores were excluded. ogeneity, further investigation was conducted to explore the
sources of variability among studies. Publication bias was
Risk of bias assessment assessed using Egger’s test and by inspection of funnel plots.
Risk of bias assessment was completed by two reviewers
independently (S.K. and C.S.G.). Retrospective studies were
classified according to the Newcastle-Ottowa Scale Sensitivity analysis
(NOS).10,11 NOS is a tool used to assess the quality of non- Further analysis was performed by including studies with a
randomized studies, particularly cohort and case–control minimum of only five patients in order to assess if there was
studies. Evaluation is based on three broad criteria: selection, any impact on the overall results.
Pain Medicine, 2025, Vol. 00, No. 0 3

Results who have shown an improvement out of each cohort is pre­

Study details sented in Figure 3. The pooled proportion of patients whose
After removal of duplicates, 97 studies were identified. After pain scores improved after DBS was 0.62 (95% CI, 0.51-
full-text assessment, 37 full-text studies were assessed for 0.71, I2¼16%).
inclusion and were finally included, shown in Figure 1 Further analysis was conducted based on the stimulation
(Supplementary Table S5). In total, 32 studies were included target. Majority of the studies (10/16) targeted the PVG and
in the meta-analysis, after removing five case reports. VPL thalamic nucleus in 108 patients (Figure 4A). Across
these studies, two electrodes are placed, one each in the PVG
Baseline characteristics and VPL unilaterally. The placement of the electrodes is con­

Downloaded from https://academic.oup.com/painmedicine/advance-article/doi/10.1093/pm/pnaf001/7951886 by guest on 16 March 2025

The baseline characteristics of included studies are summar­ tralateral to the pain affected regions.
ized in Table 1 with a total number of 330 patients. The most The remaining studies targeted the anterior cingulate cor­
common country of published studies was the United tex (ACC) in 11 patients, and the posterior limb of the inter­
Kingdom (26.3%%, n ¼ 10). Among the 32 studies included, nal capsule (PLIC) in 10 patients. Six patients in the
16 studies explored the effects of DBS and 16 studies the out­ centromedian–parafascicular nucleus were also included.30
come of MCS. Mean follow up for DBS studies was No significant conclusion regarding the optimal site for pain
18.7 months. For MCS studies, seven studies reported a reduction could be made due to the fewer number of studies
follow-up time, with a mean of 22.3 months. and smaller sample size (Figure 4B).

Effect of MCS and DBS on VAS pain relief Sensitivity analysis

The analysis of Sixteen MCS studies involving a total of 195 In DBS, 12 studies met the criteria for sensitivity analysis
patients,1,6,16–29 revealed that 0.62 (95% CI: [0.53–0.74], I2 compared to thirteen MCS studies. Removing four stud­
¼ 40%) experienced pain improvement after undergoing ies35,36,43,45 studies reduced the improvement in pain relief
MCS (Figure 2). post-DBS implantation to 0.59 (95% C1: 0.48–0.69,
Among the sixteen DBS studies included that could be P ¼ 0.24) from 0.624 (95% CI: 0.51–0.71, P ¼ 0.27). This
pooled for the meta-analysis,30–45 the number of patients was similar for the MCS cohort where the removal of three

Figure 1. PRISMA Flow diagram, of study selection for inclusion in this review and meta-analysis.
4 Pain Medicine, 2025, Vol. 00, No. 0

Table 1. Baseline characteristics. Discussion

Characteristic N (%) This systematic review and meta-analysis is the first to pool
the effect of MCS and DBS in patients with CPSP. By assess­
Country of origin
UK 10 (26.3%)
ing the 32 studies, we found that MCS has an improvement
France 5 (13.1%) in pain in 64.3% (123/191) of the patients and 62.5% (87/
USA 4 (10.5%) 139) in patients receiving DBS. Patients that underwent MCS
Germany 3 (7.9%) had a mean VAS improvement of 53.1% compared to 48.6%
Japan 3 (7.9%) in patients that underwent DBS.
Italy 2 (5.3%) Several theories about CPSP have been proposed. It is

Downloaded from https://academic.oup.com/painmedicine/advance-article/doi/10.1093/pm/pnaf001/7951886 by guest on 16 March 2025

China 2 (5.3%)
Poland 2 (5.3%) believed that CPSP could be caused by an imbalance between
Russia 1 (2.6%) the paleospinothalamic (affective-emotional) and neospino­
Belgium 1 (2.6%) thalamic (sensory-discriminative) pathways. Evidence indi­
Netherlands 1 (2.6%) cates that impaired spinothalamic tract function is related to
Canada 1 (2.6%) the pathogenesis of CPSP.46 It has been hypothesized that
South Korea 1 (2.6%)
MCS decreases thalamic hyperactivity by inducing cortico­
Switzerland 1 (2.6%)
Study design thalamic connections.47,48 However, the role of N-methyl-D-
Retrospective 33 (100%) aspartate (NMDA) receptors and GABAergic interneurons
Total number of patients 330 has also been proposed as an interesting aspect of the poten­
Total number of improved patients 210 tial mechanism in MCS.3,49,50 Repetitive Transcranial
DBS Magnetic Stimulation (rTMS) of the motor cortex has shown
Number of patients 139
Number of patients showed improvement in 87 (62.5%)
restoration of intracortical inhibition in neuropathic pain
pain relief patients, where the degree of pain relief correlates with the
Mean VAS score improvement post-surgery 48.6% (±13.42%) amount of restoration of inhibition.51 In addition, activation
Mean follow up time in months (Standard 18.67 (±13.52) of the endogenous opioid descending system and alterations
deviation) in the limbic system are described as potential mechanisms in
MCS MCS.52,53
Patients 191
Number of patients showed improvement in 123 (64.3%)
As in MCS, the precise pain-relieving effect of DBS remains
pain relief incompletely elucidated. The main difference between DBS
Mean VAS score improvement post-surgery 53.17% (±9.27%) and MCS is that the effect ensured through stimulation in
Mean follow-up time in months (Standard 22.3 (±10.1) DBS varies depending on the targeted brain area. Pain
deviation) improvement in VPL/VPM thalamus DBS could be caused by
the alteration of the balance of excitatory and inhibitory neu­
rotransmitters within the pain pathways,54,55 which partly
studies,1,22,28 reduced the improvement in pain relief to 0.63
aligns with the potential underlying mechanisms of MCS.
(95% C1: 0.52–0.72, P < 0.05) from 0.64 (95% CI: 0.53–
Stimulating the PVG and PAG will result in releasing endoge­
0.74, P ¼ 0.05) (Supplementary Figures S1 and S2). nous opioid peptides with a decrease in activity of nociceptive
signal-transmitting neurons.54,55 An alteration of the emo­
tional and cognitive aspects of chronic pain is suggested
Egger’s test when stimulating brain targets of the limbic system, such as
The potential for publication bias in this meta-analysis was the ACC.54,55 The lack of insight in the underlying mecha­
assessed using a funnel plot and Egger’s regression test. The nisms of DBS and MCS, as well as the pathophysiology of
funnel plot demonstrated a symmetrical distribution of study CPSP, illustrates the complexity of this pain syndrome and
effect sizes around the central line, indicating no obvious vis­ the need for a multidimensional approach in pain modulation
ual signs of asymmetry (Figure 5). To statistically evaluate therapies.
this, Egger’s test was performed, yielding a non-significant While the literature is limited on the direct comparison of
result (t ¼ 0.81, df ¼ 17, P ¼ 0.4297). This P value, which is DBS vs MCS on CPSP, several studies have scrutinized its
well above the conventional threshold of 0.05, suggests no efficacy individually. A study by Owen et al.,41 examined the
statistically significant asymmetry in the funnel plot. effect of DBS on 47 patients with CPSP and found a mean
Consequently, these findings provide no strong evidence of improvement in VAS score of 59%. Studies have found vary­
publication bias within the studies included in this analysis. ing results with mean VAS pain relief of between 38.1% and
Thus, it can be reasonably concluded that the results of this 68.4%.33,36,56 However, determinant factors should be taken
meta-analysis are unlikely to be influenced by publication into consideration, such as heterogeneity in terms of targeted
areas: PVG and VPL were the most common sites with two
bias. Four studies were excluded from the Egger’s test due to
studies targeting the ACC and one study each targeting the
missing standard error.
PLIC. Studies looking at the effect of MCS are limited in
comparison with DBS. A study by Zhang et al.29 looked at
the effect of MCS on 16 patients with CPSP and found a
Risk of bias mean improvement in VAS score of 42.3%. Similar to
The Risk of bias for retrospective cohort studies, using the patients who underwent DBS surgery, MCS has been shown
Newcastle-Ottawa Scale. The mean score for all studies was to improve VAS scores by 40%-63.8%.57,27
7.5 (out of a total maximum score of 9), and 5 studies were A study by Nandi et al.,58 analyzed the use of MCS and
classified as high risk of bias (Figure 6). VPL/PVG DBS on patients with CPSP. This study concluded
Pain Medicine, 2025, Vol. 00, No. 0 5

Downloaded from https://academic.oup.com/painmedicine/advance-article/doi/10.1093/pm/pnaf001/7951886 by guest on 16 March 2025

Figure 2. Forest plot showcasing VAS improvement after MCS using and random effects models.

Figure 3. Forest plot showcasing VAS improvement after DBS using random effects models.

that while MCS offers better pain relief, this varies between patients. As previously mentioned, the site of DBS insertion is
patients and is inconsistent in the long-term outcome.58 key and could have played an important role in the heteroge­
Another study by Katayama et al.,59 found that a greater pro­ neity of the population.
portion of MCS patients experienced pain relief compared to
ventral caudalis (VC) DBS patient. Both studies indicated Clinical and research implications
that DBS is a simpler procedure and generally better tolerated Our results have several implications for research and clinical
in patients. practise. The slightly better success of MCS, solely based on
Similar results were also found in studies comparing MCS pain scores, could aid clinicians in determining its appropri­
vs DBS in other forms of neuropathic pain. Son et al.,45 ate use. However, multiple factors should be taken into
directly analyzed MCS and DBS in the same eight patients account before utilizing MCS over DBS as a last-resort treat­
with chronic intractable neuropathic pain. MCS was success­ ment for CPSP, including the risk profile, side effects, and
ful in reducing pain in 6/8 compared to 2/8 in DBS.45 patients’ medical history.
While our results vary slightly with the current literature The practical implications, such as treatment cost, could
on mean VAS score, this could be attributed to various fac­ play a crucial role in determining the most suitable interven­
tors such as a larger total population of 330 pooled into the tion, given the similar success-rates. While there are currently
analysis, with other studies varying between 6 and 47 no studies directly comparing the cost-effectiveness of DBS
6 Pain Medicine, 2025, Vol. 00, No. 0

Downloaded from https://academic.oup.com/painmedicine/advance-article/doi/10.1093/pm/pnaf001/7951886 by guest on 16 March 2025

Figure 4. (A) Forest plot showcasing VAS improvement in studies targeting the PVG/VPL using random effects models. (B) Forest plot showcasing VAS
improvement in various other targets using random effects models.

Figure 5. Funnel plot assessing publication bias.

and MCS for neuropathic pain, a cost analysis study by total costs for both interventions, including the first year of
Zaghi et al.60 indicated that MCS incurs significant initial intensive follow-up.
expenses, estimated at $42 000.00. After 1-year of follow-up, Follow-up care is essential for optimizing treatment out­
with monthly visits to assess the parameters and configura­ comes in both interventions. Although the nature and inten­
tions, the total cost of treatment is estimated to be around sity of follow-up can vary significantly within patients. The
$45 600.00.60 In comparison, an analysis by Bishay et al.61 “trial and error” approach to programming different config­
of DBS costs across various disorders, neuropathic pain not urations and parameters in patients treated with MCS or
included, estimated an inflation- and currency-adjusted mean DBS is time consuming and patient specific. Therefore, novel
cost of $40 942.85 ± $17 987.43 for total DBS surgery. The research on investigating connectivity-based predictive mod­
initial cost of DBS treatment increases to $47 632.22 ± els could potentially address these challenges. A personalized
$23 067.08 after 1-year of follow-up.61 Regardless, this is approach that optimizes configuration and parameters to tar­
not a direct in-depth cost-effectiveness analysis between MCS get specific brain networks may improve the future applica­
and DBS for CPSP. It provides an impression of the estimated tion of MCS or DBS in the treatment of CPSP.
Pain Medicine, 2025, Vol. 00, No. 0 7

Downloaded from https://academic.oup.com/painmedicine/advance-article/doi/10.1093/pm/pnaf001/7951886 by guest on 16 March 2025

Figure 6. Risk of bias using the Newcastle–Ottowa scale.

Generally, DBS is considered more invasive than MCS. of NIBS on CPSP, a recent meta-analysis on sensory function
This procedure involves the implantation of electrodes into recovery in stroke patients showed that both tDCS and rTMS
deeper brain regions, while MCS requires electrode place­ significantly outperformed control conditions. NIBS was ben­
ment on the surface of the dura mater. Nevertheless, a cra­ eficial in the acute and subacute phases of stroke, while a
niotomy has been carried out over the Rolandic region for moderate effect was observed in chronic stroke patients.69
appropriate placement in MCS, whereas a burr hole Although rTMS currently has been described as a predic­
approach is used in DBS surgery. tive factor for MCS, the application of NIBS treatments could
Although MCS and DBS are distinct procedures, adverse potentially being integrated in the treatment of CPSP after
events appear to be rare and manageable. Potential adverse warrant future research. Moreover, the integration of indi­
effects following DBS and MCS surgery include seizures, vidualized treatment protocols is likely to play a crucial role
hematomas, infections, headaches and hardware malfunc­ in the future of non-invasive neuromodulation.70 Current
tion. In addition, some complications are procedure specific, research suggests that the same stimulation parameters may
such as epidural fibrosis, electrode migration, and effusion not yield uniform effects across different individuals due to
formation are associated with MCS surgery.62 In DBS, the variations in brain anatomy and neurophysiology.71
side effects are more location dependent, such as paraesthe­ Personalized approaches, such as adjusting stimulation inten­
sias, muscle spasms, and phosphenes. Side effects could also sity and targeting specific brain regions based on individual
occur at accustomed therapeutic voltages, the electrode leads patient profiles, could enhance treatment outcomes and mini­
therefore should be repositioned into a slightly altered loca­ mize side effects.72
tion.63 The aforementioned findings highlight that while DBS Though we analyze the efficacy of DBS and MCS on
and MCS are effective therapies for chronic neuropathic patient outcome there are certain aspects that need to be
pain, careful and accurate post-surgical management is not addressed. The success of MCS could be due to fewer studies
only required for optimizing result, but also cost-effectiveness present on this topic and the impact of information bias
and minimizing the risk of adverse events. would need to be considered. Various factors impacting pain
In contrast to the invasive nature of DBS and MCS, non- relief such as severity and location of stroke, age, and social
invasive brain stimulation (NIBS) techniques have been inves­ habits could have an impact on the overall outcome.73,74
tigated to adjust the excitability of specific functional brain Studies have also shown that coping strategies such as social
regions.64,65 The most prevalent utilized NIBS techniques in support can influence the effect of pain relief.75,76 A cross-
clinical setting are transcranial magnetic stimulation (TMS) sectional study found that access to a trusted healthcare pro­
and transcranial direct current stimulation (tDCS). fessional, living with pain for ≥10 years and polypharmacy
Research indicates that rTMS can effectively reduce neuro­ had a significant effect on the amount of pain relief.77
pathic pain. A study demonstrated that rTMS targeting the
motor cortex resulted in significant pain relief for patients Limitations
suffering from refractory neuropathic pain.66 The mechanism This study has several limitations. Firstly, all studies included
described by which rTMS alleviates pain is thought to involve were retrospective, precluding pooled analysis of prospective
modulation of cortical excitability and restoration of normal studies. In addition, the region of the brain where DBS was
brain function in pain processing pathways.67 A clinical trial performed is heterogeneous and could have impacted the
found that anodal tDCS applied to the motor cortex signifi­ effect of pain relief. In MCS, the variety of surgical
cantly ameliorated chronic pain and reduced intracortical approaches over time and the exact location of the electrode
inhibition, suggesting a potential mechanism for its analgesic leads could influence the level of pain relief due to the lack of
effects.68 While there is no study directly comparing the effect a standardized protocol. In this meta-analysis, only pain
8 Pain Medicine, 2025, Vol. 00, No. 0

scores have been considered, whereas in pain research, neuropathic pain. Clin Neurophysiol. 2008;119(5):993-1001.
Quality of Life (QoL) and medication use is at least as impor­ https://doi.org/10.1016/J.CLINPH.2007.12.022
tant. We also excluded full-text papers not available in 07. Galafassi GZ, Simm Pires de Aguiar PH, Simm RF, et al.
Neuromodulation for medically refractory neuropathic pain: spi­
English, restricting paper eligibility.
nal cord stimulation, deep brain stimulation, motor cortex stimu­
lation, and posterior insula stimulation. World Neurosurg.
Conclusion 2021;146:246-260.
08. Farrell SM, Green A, Aziz T. The current state of deep brain stimu­
The use of neurosurgical interventions, such as DBS and lation for chronic pain and its context in other forms of neuromo­
MCS, are a propitious field for the treatment of CPSP, with dulation. Brain Sci. 2018;8(8):158-164. https://doi.org/10.3390/

Downloaded from https://academic.oup.com/painmedicine/advance-article/doi/10.1093/pm/pnaf001/7951886 by guest on 16 March 2025

limited studies exploring and comparing these two BRAINSCI8080158
techniques. 09. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020
While our study suggests a modest improvement in pain statement: an updated guideline for reporting systematic reviews.
scores with MCS over DBS in patients suffering from CPSP, BMJ. 2021;372:n71. https://doi.org/10.1136/BMJ.N71
10. Ottawa Hospital Research Institute. 2021. Available: https://
these findings should be viewed as preliminary. Pain scores
www.ohri.ca/programs/clinical_epidemiology/oxford.asp.
alone is insufficient to establish MCS as a definitively supe­ Accessed October 15, 2024.
rior or non-inferior treatment option compared to DBS. 11. Sterne JAC, Savovi�c J, Page MJ, et al. RoB 2: a revised tool for
Further factors such as cost-effectiveness in pain treatment, assessing risk of bias in randomised trials. BMJ. 2019;366:4898.
long-term efficacy and multi-dimensional functional out­ https://doi.org/10.1136/BMJ.L4898
comes would need to be assessed in order to determine the 12. Bodian CA, Freedman G, Hossain S, Eisenkraft JB, Beilin Y. The
appropriate surgical neuromodulation for CPSP. visual analog scale for pain: clinical significance in postoperative
patients. Anesthesiology. 2001;95(6):1356-1361. https://doi.org/
10.1097/00000542-200112000-00013
Supplementary material 13. Myles PS, Myles DB, Galagher W, et al. Measuring acute postoper­
ative pain using the visual analog scale: the minimal clinically
Supplementary material is available at Pain Medicine online.
important difference and patient acceptable symptom state. Br J
Anaesth. 2017;118(3):424-429.
Funding 14. Von Hippel PT. The heterogeneity statistic I2 can be biased in
small meta-analyses. BMC Med Res Methodol. 2015;15:35-38.
This research received no external funding. https://doi.org/10.1186/S12874-015-0024-Z/FIGURES/4
15. Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Measuring
Conflicts of interest: The authors declare no conflicts of
inconsistency in meta-analyses. BMJ. 2003;327(7414):557-560.
interest. https://doi.org/10.1136/BMJ.327.7414.557
16. Fagundes-Pereyra WJ, Teixeira MJ, Reyns N, et al. Motor cortex
electric stimulation for the treatment of neuropathic pain. Arq
Data availability Neuropsiquiatr. 2010;68(6):923-929. https://doi.org/10.1590/
The original contributions presented in the study are included S0004-282X2010000600018
in the article, further inquiries can be directed to the corre­ 17. Guo S, Zhang X, Tao W, Zhu H, Hu Y. Long-term follow-up of
sponding author. motor cortex stimulation on central poststroke pain in thalamic
and extrathalamic stroke. Pain Pract. 2022;22(7):610-620. https://
doi.org/10.1111/PAPR.13137
References 18. Henssen DJHA, Kurt E, Van Cappellen Van Walsum AM, et al.
Long-term effect of motor cortex stimulation in patients suffering
01. Delavall�ee M, Abu-Serieh B, De Tourchaninoff M, Raftopoulos C. from chronic neuropathic pain: an observational study. PLoS One.
Subdural motor cortex stimulation for central and peripheral neu­ 2018;13(1):e0191774. https://doi.org/10.1371/JOURNAL.PONE.
ropathic pain: a long-term follow-up study in a series of eight 0191774
patients. Neurosurgery. 2008;63(1):101-105. https://doi.org/10. 19. Isagulyan ED, Tomsky AA, Dekopov AV, et al. Results of motor
1227/01.NEU.0000335076.24481.B6 cortex stimulation in the treatment of chronic pain syndromes. Zh
02. Gandolfi M, Donisi V, Battista S, et al. Health-related quality of Vopr Neirokhir Im N N Burdenko. 2015;79(6):46-60. https://doi.
life and psychological features in post-stroke patients with chronic org/10.17116/NEIRO201579646-60
pain: a cross-sectional study in the neuro-rehabilitation context of 20. Maarrawi J, Peyron R, Mertens P, et al. Brain opioid receptor den­
care. Int J Environ Res Public Health. 2021;18(6):3089. https:// sity predicts motor cortex stimulation efficacy for chronic pain.
doi.org/10.3390/IJERPH18063089 Pain. 2013;154(11):2563-2568. https://doi.org/10.1016/J.PAIN.
03. DosSantos MF, Ferreira N, Toback RL, Carvalho AC, DaSilva AF. 2013.07.042
Potential mechanisms supporting the value of motor cortex stimula­ 21. Nguyen JP, Lefaucher JP, Le Guerinel C, et al. Motor cortex stimu­
tion to treat chronic pain syndromes. Front Neurosci. 2016;10:18. lation in the treatment of central and neuropathic pain. Arch Med
https://doi.org/10.3389/FNINS.2016.00018/BIBTEX Res. 2000;31(3):263-265.
04. Bo Z, Jian Y, Yan L, et al. Pharmacotherapies for central post- 22. Nguyen JP, Velasco F, Brugi�eres P, et al. Treatment of chronic neu­
stroke pain: a systematic review and network meta-analysis. Oxid ropathic pain by motor cortex stimulation: results of a bicentric
Med Cell Longev. 2022;2022:3511385. https://doi.org/10.1155/ controlled crossover trial. Brain Stimul. 2008;1(2):89-96. https://
2022/3511385 doi.org/10.1016/J.BRS.2008.03.007
€
05. Şahin-Onat Ş, Unsal-Delialio� €
glu S, Kulaklı F, Ozel S. The effects of 23. Nuti C, Peyron R, Garcia-Larrea L, et al. Motor cortex stimulation
central post-stroke pain on quality of life and depression in for refractory neuropathic pain: four year outcome and predictors
patients with stroke. J Phys Ther Sci. 2016;28(1):96-101. https:// of efficacy. Pain. 2005;118(1-2):43-52. https://doi.org/10.1016/J.
doi.org/10.1589/JPTS.28.96 PAIN.2005.07.020
06. Hosomi K, Saitoh Y, Kishima H, et al. Electrical stimulation of pri­ 24. Rasche D, Ruppolt M, Stippich C, Unterberg A, Tronnier VM.
mary motor cortex within the central sulcus for intractable Motor cortex stimulation for long-term relief of chronic
Pain Medicine, 2025, Vol. 00, No. 0 9

neuropathic pain: a 10 year experience. Pain. 2006;121(1- 2004;21(1):31-39. https://doi.org/10.1097/00004691-200401000-

2):43-52. https://doi.org/10.1016/J.PAIN.2005.12.006 00005
25. Sokal P, Harat M, Malukiewicz A, Kiec M, � Swito�nska M, 40. Nowacki A, Zhang D, Barlatey S, et al. Deep brain stimulation of the
Jabło�nska R. Effectiveness of tonic and burst motor cortex stimu­ central lateral and ventral posterior thalamus for central poststroke
lation in chronic neuropathic pain. J Pain Res. 2019;12 pain syndrome: preliminary experience. Neuromodulation. 2023;26
(6580):1863-1869. (8):1747-1756. https://doi.org/10.1016/J.NEUROM.2022.09.005
26. Sokal P, Harat M, Zieli� nski P, Furtak J, Paczkowski D, Rusinek 41. Owen SLF, Green AL, Nandi DD, Bittar RG, Wang S, Aziz TZ.
M. Motor cortex stimulation in patients with chronic central pain. Deep brain stimulation for neuropathic pain. Acta Neurochir
Adv Clin Exp Med. 2015;24(2):289-296. https://doi.org/10. Suppl. 2007;97(Pt 2):111-116. https://doi.org/10.1007/978-3-
17219/ACEM/40452 211-33081-4_13/COVER

Downloaded from https://academic.oup.com/painmedicine/advance-article/doi/10.1093/pm/pnaf001/7951886 by guest on 16 March 2025

27. Tanei T, Kajita Y, Noda H, et al. Efficacy of motor cortex stimula­ 42. Owen SLF, Green AL, Stein JF, Aziz TZ. Deep brain stimulation
tion for intractable central neuropathic pain: comparison of stimu­ for the alleviation of post-stroke neuropathic pain. Pain. 2006;120
lation parameters between post-stroke pain and other central pain. (1-2):202-206. https://doi.org/10.1016/J.PAIN.2005.09.035
Neurol Med Chir (Tokyo). 2011;51(1):8-14. https://doi.org/10. 43. Pereira EAC, Green AL, Bradley KM, et al. Regional cerebral per­
2176/NMC.51.8 fusion differences between periventricular grey, thalamic and dual
28. Yamamoto T, Katayama Y, Obuchi T, et al. Recording of cortico­ target deep brain stimulation for chronic neuropathic pain.
spinal evoked potential for optimum placement of motor cortex Stereotact Funct Neurosurg. 2007;85(4):175-183. https://doi.org/
stimulation electrodes in the treatment of post-stroke pain. Neurol 10.1159/000101296
Med Chir (Tokyo). 2007;47(9):409-414. https://doi.org/10.2176/ 44. Rasche D, Rinaldi PC, Young RF, Tronnier VM. Deep brain stim­
NMC.47.409 ulation for the treatment of various chronic pain syndromes.
29. Zhang X, Zhu H, Tao W, Li Y, Hu Y. Motor cortex stimulation Neurosurg Focus. 2006;21(6):E8-8. https://doi.org/10.3171/FOC.
therapy for relief of central post-stroke pain: a retrospective study 2006.21.6.10
with neuropathic pain symptom inventory. Stereotact Funct 45. Son BC, Kim DR, Kim HS, Lee SW. Simultaneous trial of deep
Neurosurg. 2018;96(4):239-243. https://doi.org/10.1159/00049 brain and motor cortex stimulation in chronic intractable neuro­
2056 pathic pain. Stereotact Funct Neurosurg. 2014;92(4):218-226.
30. Abdallat M, Saryyeva A, Blahak C, et al. Centromedian-parafas­ https://doi.org/10.1159/000362933
cicular and somatosensory thalamic deep brain stimulation for 46. Oh HS, Seo WS. A comprehensive review of central post-stroke
treatment of chronic neuropathic pain: a contemporary series of pain. Pain Manag Nurs. 2015;16(5):804-818. https://doi.org/10.
40 patients. Biomedicines. 2021;9(7):731-750. https://doi.org/10. 1016/J.PMN.2015.03.002
3390/BIOMEDICINES9070731 47. Tsubokawa T, Katayama Y, Yamamoto T, Hirayama T, Koyama
31. Boccard SGJ, Fitzgerald JJ, Pereira EAC, et al. Targeting the affec­ S. Chronic motor cortex stimulation in patients with thalamic
tive component of chronic pain: a case series of deep brain stimula­ pain. J Neurosurg. 1993;78(3):393-401. https://doi.org/10.3171/
tion of the anterior cingulate cortex. Neurosurgery. 2014;74 JNS.1993.78.3.0393
(6):628-635. https://doi.org/10.1227/NEU.0000000000000321 48. Tsubokawa T, Katayama Y, Yamamoto T, Hirayama T, Koyama
32. Boccard SGJ, Pereira EAC, Moir L, Aziz TZ, Green AL. Long- S. Treatment of thalamic pain by chronic motor cortex stimula­
term outcomes of deep brain stimulation for neuropathic pain. tion. Pacing Clin Electrophysiol. 1991;14(1):131-134. https://doi.
Neurosurgery. 2013;72(2):221-230; discussion 231. https://doi. org/10.1111/J.1540-8159.1991.TB04058.X
org/10.1227/NEU.0B013E31827B97D6 49. Ambriz-Tututi M, S� anchez-Gonz� alez V, Drucker-Col�ın R.
33. Gray AM, Pounds-Cornish E, Eccles FJR, Aziz TZ, Green AL, Transcranial magnetic stimulation reduces nociceptive threshold
Scott RB. Deep brain stimulation as a treatment for neuropathic in rats. J Neurosci Res. 2012;90(5):1085-1095. https://doi.org/10.
pain: a longitudinal study addressing neuropsychological out­ 1002/JNR.22785
comes. J Pain. 2014;15(3):283-292. https://doi.org/10.1016/J. 50. Liebetanz D, Nitsche MA, Tergau F, Paulus W. Pharmacological
JPAIN.2013.11.003 approach to the mechanisms of transcranial DC-stimulation-
34. Hamani C, Schwalb JM, Rezai AR, Dostrovsky JO, Davis KD, induced after-effects of human motor cortex excitability. Brain.
Lozano AM. Deep brain stimulation for chronic neuropathic pain: 2002;125(Pt 10):2238-2247. https://doi.org/10.1093/BRAIN/
Long-term outcome and the incidence of insertional effect. Pain. AWF238
2006;125(1-2):188-196. https://doi.org/10.1016/J.PAIN.2006.05. 51. Lefaucheur JP, Drouot X, M�enard-Lefaucheur I, Keravel Y,
019 Nguyen JP. Motor cortex rTMS restores defective intracortical
35. Hunsche S, Sauner D, Runge MJR, et al. Tractography-guided inhibition in chronic neuropathic pain. Neurology. 2006;67
stimulation of somatosensory fibers for thalamic pain relief. (9):1568-1574. https://doi.org/10.1212/01.WNL.0000242731.
Stereotact Funct Neurosurg. 2013;91(5):328-334. https://doi.org/ 10074.3C
10.1159/000350024 52. Garcia-Larrea L, Maarrawi J, Peyron R, et al. On the relation
36. Kim JP, Chang WS, Park YS, Chang JW. Impact of ventralis cauda­ between sensory deafferentation, pain and thalamic activity in
lis deep brain stimulation combined with stereotactic bilateral cin­ Wallenberg’s syndrome: a PET-scan study before and after motor
gulotomy for treatment of post-stroke pain. Stereotact Funct cortex stimulation. Eur J Pain. 2006;10(8):677-688. https://doi.
Neurosurg. 2012;90(1):9-15. https://doi.org/10.1159/000330382 org/10.1016/J.EJPAIN.2005.10.008
37. Levi V, Cordella R, D'Ammando A, et al. Dorsal anterior cingulate 53. Peyron R, Faillenot I, Mertens P, Laurent B, Garcia-Larrea L.
cortex (ACC) deep brain stimulation (DBS): a promising surgical Motor cortex stimulation in neuropathic pain. Correlations
option for the treatment of refractory thalamic pain syndrome (TPS). between analgesic effect and hemodynamic changes in the brain. A
Acta Neurochir (Wien). 2019;161(8):1579-1588. https://doi.org/10. PET study. Neuroimage. 2007;34(1):310-321. https://doi.org/10.
1007/S00701-019-03975-5/TABLES/4 1016/J.NEUROIMAGE.2006.08.037
38. Nandi D, Aziz T, Carter H, Stein J. Thalamic field potentials in 54. Alamri A, Pereira EAC. Deep brain stimulation for chronic pain.
chronic central pain treated by periventricular gray stimulation—a Neurosurg Clin N Am. 2022;33(3):311-321.
series of eight cases. Pain. 2003;101(1-2):97-107. https://doi.org/ 55. Shaheen N, Shaheen A, Elgendy A, et al. Deep brain stimulation
10.1016/S0304-3959(02)00277-4 for chronic pain: a systematic review and meta-analysis. Front
39. Nandi D, Aziz TZ. Deep brain stimulation in the management of Hum Neurosci. 2023;17:1297894. https://doi.org/10.3389/
neuropathic pain and multiple sclerosis tremor. J Clin Neurophysiol. FNHUM.2023.1297894/FULL
10 Pain Medicine, 2025, Vol. 00, No. 0

56. Franzini A, Messina G, Levi V, et al. Deep brain stimulation of the 67. Pan LJ, Zhu HQ, Zhang XA, Wang XQ. The mechanism and
posterior limb of the internal capsule in the treatment of central effect of repetitive transcranial magnetic stimulation for post-
poststroke neuropathic pain of the lower limb: case series with stroke pain. Front Mol Neurosci. 2022;15:1091402. https://doi.
long-term follow-up and literature review. J Neurosurg. 2020;133 org/10.3389/FNMOL.2022.1091402
(3):830-838. https://doi.org/10.3171/2019.5.JNS19227 68. Antal A, Terney D, K€ uhnl S, Paulus W. Anodal transcranial direct
57. N€ussel M, Hamperl M, Maslarova A, et al. Burst motor cortex current stimulation of the motor cortex ameliorates chronic pain
stimulation evokes sustained suppression of thalamic stroke pain: a and reduces short intracortical inhibition. J Pain Symptom Manag.
narrative review and single-case overview. Pain Ther. 2021;10 2010;39(5):890-903. https://doi.org/10.1016/J.JPAINSYMMAN.
(1):101-114. https://doi.org/10.1007/S40122-020-00221-0 2009.09.023
58. Nandi D, Smith H, Owen S, Joint C, Stein J, Aziz T. Peri-ventricu­ 69. Chen G, Wu M, Chen J, et al. Non-invasive brain stimulation

Downloaded from https://academic.oup.com/painmedicine/advance-article/doi/10.1093/pm/pnaf001/7951886 by guest on 16 March 2025

lar grey stimulation versus motor cortex stimulation for post effectively improves post-stroke sensory impairment: a systematic
stroke neuropathic pain. J Clin Neurosci. 2002;9(5):557-561. review and meta-analysis. J Neural Transm (Vienna). 2023;130
https://doi.org/10.1054/jocn.2001.1042 (10):1219-1230. https://doi.org/10.1007/S00702-023-02674-X/
59. Katayama Y, Yamamoto T, Kobayashi K, Kasai M, Oshima H, FIGURES/5
Fukaya C. Motor cortex stimulation for post-stroke pain: compar­ 70. Lefaucheur JP, M�enard-Lefaucheur I, Goujon C, Keravel Y,
ison of spinal cord and thalamic stimulation. Stereotact Funct Nguyen JP. Predictive value of rTMS in the identification of res­
Neurosurg. 2001;77(1-4):183-186. https://doi.org/10.1159/ ponders to epidural motor cortex stimulation therapy for pain. J
000064618 Pain. 2011;12(10):1102-1111. https://doi.org/10.1016/J.JPAIN.
60. Zaghi S, Heine N, Fregni F. Brain stimulation for the treatment of
2011.05.004
pain: a review of costs, clinical effects, and mechanisms of treat­
71. Wu YH, Yu J, Hong LR, Luo BY. Neuromodulatory therapies for
ment for three different central neuromodulatory approaches. J
patients with prolonged disorders of consciousness. Chin Med J
Pain Manag. 2009;2(3):339. Accessed November 3, 2024. https://
(Engl). 2021;134(7):765-776. https://doi.org/10.1097/CM9.
pmc.ncbi.nlm.nih.gov/articles/PMC2888303/
0000000000001377
61. Bishay AE, Lyons AT, Koester SW, et al. Global economic evalua­
72. Cash RFH, Zalesky A. Personalized and circuit-based transcranial
tion of the reported costs of deep brain stimulation. Stereotact
magnetic stimulation: evidence, controversies, and opportunities.
Funct Neurosurg. 2024;102(4):257-274. https://doi.org/10.1159/
Biol Psychiatry. 2024;95(6):510-522.
000537865
73. Harno H, Haapaniemi E, Putaala J, et al. Central poststroke pain
62. Ramos-Fresnedo A, Perez-Vega C, Domingo RA, Cheshire WP,
Middlebrooks EH, Grewal SS. Motor cortex stimulation for pain: in young ischemic stroke survivors in the Helsinki Young Stroke
a narrative review of indications, techniques, and outcomes. Registry. Neurology. 2014;83(13):1147-1154. https://doi.org/10.
Neuromodulation. 2022;25(2):211-221. 1212/WNL.0000000000000818
63. Szymoniuk M, Chin JH, Domagalski Ł, Biszewski M, J� o�zwik K, 74. O'Donnell MJ, Diener H-C, Sacco RL, Panju AA, Vinisko R,
Kamieniak P. Brain stimulation for chronic pain management: a Yusuf S; PRoFESS Investigators. Chronic pain syndromes after
narrative review of analgesic mechanisms and clinical evidence. ischemic stroke: PRoFESS trial. Stroke. 2013;44(5):1238-1243.
Neurosurg Rev. 2023;46(1):127. https://doi.org/10.1007/S10143- https://doi.org/10.1161/STROKEAHA.111.671008
023-02032-1 75. Revert�e-Villarroya S, Su~ ner-Soler R, Font-Mayolas S, et al.
64. Kemps H, Gervois P, Br^ one B, Lemmens R, Bronckaers A. Non- Influence of pain and discomfort in stroke patients on coping strat­
invasive brain stimulation as therapeutic approach for ischemic egies and changes in behavior and lifestyle. Brain Sci. 2021;11
stroke: Insights into the (sub)cellular mechanisms. Pharmacol (6):804-820. https://doi.org/10.3390/BRAINSCI11060804
Ther. 2022;235:108160. 76. Yu Y, Hu J, Efird JT, Mccoy TP. Social support, coping strategies
65. Tang A, Thickbroom G, Rodger J. Repetitive transcranial mag­ and health-related quality of life among primary caregivers of
netic stimulation of the brain. Neuroscientist. 2015;23(1):82-94. stroke survivors in China. J Clin Nurs. 2013;22(15-
https://doi.org/10.1177/1073858415618897 16):2160-2171. https://doi.org/10.1111/JOCN.12251
66. Tsai YY, Wu WT, Han DS, et al. Application of repetitive trans­ 77. Zerriouh M, De Clifford-Faug�ere G, Nguena Nguefack HL, et al.
cranial magnetic stimulation in neuropathic pain: a narrative Pain relief and associated factors: a cross-sectional observational
review. Life (Basel). 2023;13(2):258. https://doi.org/10.3390/ web-based study in a Quebec cohort of persons living with chronic
LIFE13020258 pain. Front Pain Res (Lausanne). 2024;5:1306479.

© The Author(s) 2025. Published by Oxford University Press on behalf of the American Academy of Pain Medicine.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits
unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.
Pain Medicine, 2025, 00, 1–10
https://doi.org/10.1093/pm/pnaf001
Review Article