Skip to main content
NCBI home page
European Journal of Physical and Rehabilitation Medicine logoLink to European Journal of Physical and Rehabilitation Medicine
. 2025 Jun 19;61(3):520–531. doi: 10.23736/S1973-9087.25.08870-7

Exploring the therapeutic potential of transcranial direct current stimulation for chronic low back pain: a scoping review

Aminu A IBRAHIM 1, Katayan KLAHAN 2, Kanphajee SORNKAEW 2, Jarugool TRETRILUXANA 3, Sheri P SILFIES 4, Peemongkon WATTANANON 2,*
PMCID: PMC12409359  PMID: 40536666

Abstract

INTRODUCTION

Chronic low back pain (CLBP) is a common disabling condition, inflicting a substantial socioeconomic burden. Given its association with neuroplastic changes, as evidenced by central and peripheral sensitization, neuromodulatory techniques such as transcranial direct current stimulation (tDCS) have emerged as potential treatments. This scoping review aimed to identify and map the existing literature on tDCS studies for CLBP to provide insight into how these studies are conducted, and to address their potential gaps in knowledge.

EVIDENCE ACQUISITION

PubMed, Embase, Web of Science, and Cochrane Library were searched for relevant studies from inception to 23 March 2025. Eligible studies included were those examining tDCS alone or with other interventions in adults with CLBP, regardless of the outcome evaluated and included adults with CLBP. The review was conducted using Arksey and O’Malley’s six-stage framework and was guided by the PRISMA for scoping review framework.

EVIDENCE SYNTHESIS

Of 134 screened records, 26 studies were included: 23 randomized controlled trials and 3 quasi-experimental studies. Half of the studies (50%) had a low risk of bias while one-third (34.6%) showed a high or serious risk of bias. Outcomes evaluated varied and included clinical, biophysical, biomechanical, and psychosocial measures. There was considerable variability in tDCS treatment protocols across studies. The effectiveness of tDCS was inconsistent, particularly for clinical outcomes, with some studies indicating positive effects while others reported no significant effects.

CONCLUSIONS

Overall, this review reveals inconsistent results for tDCS effectiveness in CLBP, likely due to variability in study designs, sample characteristics, treatment protocols, and outcome measures. Future well-designed trials are needed to clarify the therapeutic potential of tDCS for CLBP, particularly in combination with other interventions.

Key words: Low back pain, Transcranial direct current stimulation, Scoping review

European Journal of Physical and Rehabilitation Medicine 2025 June;61(3):520-31

DOI: 10.23736/S1973-9087.25.08870-7

© 2025 THE AUTHORS

SYSTEMATIC REVIEW

(Cite this article as: Ibrahim AA, Klahan K, Sornkaew K, Tretriluxana J, Silfies SP, Wattananon P. Exploring the therapeutic potential of transcranial direct current stimulation for chronic low back pain: a scoping review. Eur J Phys Rehabil Med 2025;61:520-31. DOI: 10.23736/S1973-9087.25.08870-7)

Introduction

Low back pain remains the most common musculoskeletal condition, which occurs across all age groups,1 affecting over half a billion people worldwide.2 It is the leading cause of disability2 and is associated with work absenteeism, and lost productivity thus, inflicting a substantial personal, medical, societal, and economic burden.1, 3, 4 This highlights the need for further research to understand the underlying mechanisms of low back pain and identify optimal treatment strategies.

Contemporary understanding of pain suggests that chronic low back pain (CLBP) is not only linked to peripheral tissue injury or inflammation but also abnormalities in pain processing mechanisms. Emerging evidence suggests that impaired sensorimotor control contributes to recurrent and persistent low back pain.5 Neuroplastic changes in the sensorimotor cortex, along with peripheral musculoskeletal alterations,6-12 indicate that both peripheral and central sensitization are implicated in CLBP.13

Evidence of peripheral sensitization includes a decreased pain threshold, due to nociceptive pathway stimulation, which expands the receptive field in the dorsal horn of the spinal cord and activates associated pain regions in the CNS.14, 15 Central sensitization, on the other hand, is marked by increased activation of cortical and subcortical structures such as the medial prefrontal cortex (mPFC), cingulate cortex, amygdala, and insular lobe, along with decreased activation in the pain-relief areas and altered functional connectivity in pain-associated areas.16 Given the evidence of altered neuroplasticity in patients with CLBP,16, 17 researchers have focused on emerging neuromodulator techniques like transcranial direct current stimulation (tDCS) to restore such alterations by targeting relevant cortical areas of the brain18-26 to improve pain and sensorimotor function.17

As a non-invasive procedure, tDCS is a painless, safe, cost-effective procedure for modulating brain activity.27, 28 It uses a low-intensity current (1-2 mA) applied over the cortex through at least two electrodes (anode and cathode), creating an electric field between them.29 Typically, anodal tDCS increases cortical excitability by inducing membrane depolarization, whereas cathodal tDCS decreases cortical excitability through membrane hyperpolarization.29

A previous systematic review,30 based on two studies, recommended against using primary cortex (M1) tDCS for CLBP pain and disability. A subsequent review with meta-analysis of nine studies, including five pooled studies, found no significant effect of multiple M1 tDCS sessions on pain and disability for CLBP.29 A recent systematic review and meta-analysis of four studies concluded there was no evidence of tDCS effectiveness for pain intensity or depressive symptoms in CLBP patients.31 Although these reviews suggest limited evidence regarding the utility of tDCS for CLBP, they were limited to pain and disability outcomes,29, 30 a small number of included studies,30, 31 and are relatively outdated, having been published over four years ago.29, 30 Considering the emergence of new studies and the potential for tDCS to influence other clinical outcomes such as neuroplasticity in addition to biophysical, biomechanical and broader psychosocial outcomes, it is imperative to map the existing body of literature on tDCS across a broad range of CLBP-related outcomes.

The aim of this review was, therefore, to identify and map the existing tDCS studies on CLBP, to evaluate how these studies are conducted, and to highlight potential gaps in knowledge. The results are not intended to provide evidence to guide clinical practice but to provide insight into the scope and nature of research supporting the utility of tDCS for CLBP.

Evidence acquisition

Study approach

This scoping review followed the methodological framework outlined by Arksey and O’Malley,32 encompassing the following stages: 1) formulating the research question; 2) identifying relevant studies; 3) selecting appropriate studies; 4) charting the data; and 5) collating, summarizing, and reporting the results. The review was reported as per the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) guidelines.33

Identifying research question

What are the characteristics of tDCS protocol used in the studies for CLBP?

What is the available evidence from studies on the effects of tDCS for CLBP?

Identifying relevant studies

A comprehensive search was performed by the second author (PW) using the databases PubMed, Embase, Web of Science, and Cochrane Library. Articles published in English from inception to 23 March 2025 were searched by utilizing key terms (i.e., transcranial direct current stimulation and low back pain) and their synonyms from MeSH database. Reference lists of retrieved articles were manually examined for additional eligible studies. The search string and key search terms used in the review are shown in Table I.

Table I. —Search string and key search terms used in the review.

# of terms Search of term(s)
#1 “Transcranial Direct Current Stimulation”[Mesh] OR “tDCS”[tiab] OR “Transcranial Alternating Current Stimulation”[tiab] OR “Transcranial Random Noise Stimulation”[tiab] OR “Repetitive Transcranial Electrical Stimulation “[tiab]
#2 “Low Back Pain”[Mesh] OR “Back Pain”[Mesh] OR “LBP”[tiab] OR “nonspecific low back pain”[tiab] OR “chronic low back pain”[tiab] OR “chronic nonspecific low back pain”[tiab]
#3 #1 AND #2
Open in a new tab

Selecting appropriate studies

The screening process first involved reviewing titles and abstracts, followed by a full-text review by two independent authors (A.A.I. and K.S.). Any disagreements regarding article selection were resolved through discussion and consensus.

Studies that met the following criteria were included in the review: 1) adults aged 18 years and above with low back pain of ≥3 months duration; 2) randomized controlled trials, quasi-experimental/pre-post studies examining the effects of anodal or cathodal tDCS alone or in combination with other interventions; and 3) Studies evaluating relevant outcomes related to CLBP including clinical, biophysical, biomechanical and psychosocial outcomes. Studies were excluded from the review if they met any of the following criteria: 1) systematic or scoping reviews and case reports; 2) studies involving patients with other chronic pain conditions like fibromyalgia or neuropathic pain; 3) studies evaluating the effects of repetitive transcranial magnetic stimulation (TMS) or surgically implanted brain stimulators; and 4) studies published as conference abstracts, dissertations, or in books.

Data charting

Relevant data (e.g., author, year, country, study design, sample size, intervention, comparator, outcomes, report of adverse events, and main results) consistent with the study objectives were extracted from the included studies to a pre-established spreadsheet template by one reviewer (A.A.I.) and checked by a second review (K.S.).

Risk of bias (quality) assessment

Methodological quality was assessed independently by two reviewers (A.A.I. and K.K.) and disagreements in ratings were resolved between the assessors and an additional reviewer (P.W.) was consulted if necessary. For randomized trials with parallel and cross-over designs, the Cochrane Risk of Bias (RoB) 2 tool34 was used, with risk of bias being categorized as low risk, high risk, and some concerns.34 The overall risk of bias was interpreted as follows:

  • low risk of bias: when all domains were judged to have a low risk of bias;

  • some concerns: when at least one domain raised some concerns but no domain was judged to be at high risk of bias;

  • high risk of bias: when at least one domain was rated as high risk of bias, or multiple domains were judged to have some concerns that substantially lowered the confidence in the result.34

For non-randomized studies, the Risk of Bias in Non-randomized Studies of Interventions (ROBINS-1) was used, with risk of bias being categorized as low risk, moderate risk, serious risk, critical risk, and no information.35 The overall risk of bias was interpreted as follows:

  • low risk of bias: when all domains were judged to have a low risk of bias except for concerns about uncontrolled confounding in Domain 1;

  • moderate risk of bias: when one or more domains were judged to have a moderate risk of bias, with no serious or critical risk of bias in any domain;

  • serious risk of bias: when one or more domains were judged as having a serious risk of bias, with no critical risk of bias in any domain, OR when multiple domains had a moderate risk of bias, resulting in an overall serious risk of bias;

  • critical risk of bias: when one or more domains were judged as having a critical risk of bias, OR when multiple domains had a serious risk of bias, resulting in an overall critical risk of bias.35

Collating, summarizing, and reporting the results

A narrative synthesis was conducted to summarize the characteristics of treatment protocols and their effects on the outcomes reported in each included study.

Evidence synthesis

Study selection

Detailed search results across different databases are provided in Supplementary Digital Material 1 (Supplementary Text File 1). The search yielded 167 records. After the removal of duplicates, 134 records were screened by title and abstract. Of these, 29 studies were considered for full-text review for potential eligibility, but three studies were excluded as two were abstracts36, 37 and one used high-definition transcranial infraslow pink-noise stimulation.38 Twenty-six studies were finally included in the narrative synthesis. Figure 1 illustrates the study selection process.

Figure 1.

Figure 1

Open in a new tab

—Flow chart for the study selection process according to PRISMA statement.33

Study characteristics

The included studies consisted of a total of 893 (range 8 to 135) participants with CLBP, with ages ranging between 18 and 75 years. Details of the characteristics of the included studies are fully described in Supplementary Digital Material 2 (Supplementary Table I).18-26, 39-55

Studies were published between 2012 and 2025, with the majority being published between 2019 and 2025 (N.=17, 69%). Most of the studies were conducted in Iran19, 39-42, 56 followed by Australia,24, 25, 43, 44 Thailand,45-47 USA,48-50 Denmark,51, 52 Brazil,18, 53 Germany,20, 21 China,54 Italy,26 and UK.23 The majority of them (N.=23, 88.5%) were RCTs,18-23, 25, 26, 39-43, 45, 47-54, 56 of which sixteen (61.5%) had a parallel design18, 19, 21, 22, 26, 39-43, 45, 47, 50, 53, 54, 56 and seven (26.9%) had a cross-over design.20, 23, 25, 48, 49, 51, 52 Except for two studies that were single-blinded,39, 56 the majority were double-blinded. However, one study had unclear blinding.47 Three studies (11.5%) had a quasi-experimental design.24, 44, 46

Fourteen studies employed tDCS alone compared to sham-tDCS20, 22, 23, 40, 42, 43, 48, 49, 51, 52, 54 or control receiving44 or not receiving tDCS,46 whereas fourteen studies employed tDCS in combination with other interventions such as attention bias modification (ABM),39 cognitive behavioral therapy (CBT),21, 41 exercise,26, 45 pain neuroscience education (PNE),50 peripheral electrical stimulation (PES),18, 24, 25 physical therapy (PT),47 postural training,19 osteopathic manipulative treatment (OMT),53 and sensory-motor training.56 Most studies employed anodal-tCDS for cortical excitation except for four studies.22, 40-42 Of these, three studies22, 40, 42 used both anodal- and cathodal-tDCS for cortical stimulation and cortical inhibition, respectively, while the other study41 used cathodal-tDCS only. Two studies51, 52 used high-definition tDCS.

Eighteen studies used large-sized electrodes with dimensions of 5x7 cm2 (35 cm2),18-22, 24-26, 40-44, 46-48, 51, 52 four studies used moderate-sized electrodes with dimensions of 5x5 cm2 (25 cm2)49, 50, 53 or 4x4 cm2 (16 cm2),39 and two studies54, 56 used small-sized electrodes. One of the small-electrode studies used an electrode of 2x4 cm (8 cm2) anode and a 4x4 cm (16 cm2) cathode,56 whereas the other used a ring electrode with a 0.78 cm2 anode and a 12.5 cm2 cathode.54 For the tDCS technique (i.e., ramp-up/down protocol), one study used 60 s,52 seven studies used 30 s,22, 39, 43, 47, 50, 51, 53 one study used 15 s,45 twelve studies used 10 s,19, 24, 25, 40-42, 44, 46, 48, 49, 54, 56 two studies used 8 s,20, 21 and one study used 5 s.23 However, the ramp-up/down protocol was not reported in one study.26 There was significant variation in current intensity and density used, as the electrode size varied. Nineteen studies used a current intensity of 2 mA, with fourteen of them having a corresponding current density of 0.057 mA/cm2,18-22, 40-42, 45-48, 51, 54 two having 2.54 mA/cm2,52, 54 two having 0.008 mA/cm2,50, 53 and one study having 0.125 mA/cm2.39 Two studies used 1.5 mA with a corresponding current density of 0.188 mA/cm2,56 and 0.043 mA/cm2.43 Four studies used 1 mA with three of them having similar corresponding current density of 0.028 mA/cm2,24, 25, 44 and the other having a current density of 0.04 mA/cm2.49 For the duration of tDCS intervention, two studies applied tDCS for 40 min,48, 49 two studies for 30 min,24, 25 twenty studies for 20 min,18, 19, 21-23, 26, 39-43, 45-47, 50-54, 56 one study for 19 min,44 and another study for 15 min.20 Treatment sessions ranged from single24, 40, 42, 46, 48, 49 to 5022 sessions, with follow-up periods ranging from 3 days25 to 6 months.18, 42

For clinical outcomes, pain intensity was evaluated using the Numerical Pain Rating Scale (NPRS),18, 20, 25, 49-51, 54 Defense and Veterans Pain Rating Scale (DVRS),22, 48, 49 and Visual Analogue Scale (VAS);19, 21, 23, 26, 43, 47, 51-53 pain intensity and interference using the Brief Pain Inventory (BPI);39 perceived pain interference using the Wes Haven-Yale Multidimensional Pain Inventory’s General Activity Subscale (WHY-MPI-C);22 pain sensitivity/central sensitization level using the NPRS to evaluate perceived electrical pain threshold on the dorsum of hand and thermal pain threshold on the forearm,20 pressure pain thresholds (PPT),25, 49, 51, 52 Schober test,25 conditioned pain modulation (CPM),51, 52 and temporal summation of pain (TSP);51, 52 and sensory and affective aspects of pain using McGill Pain Questionnaire (MPQ);18 disability was evaluated using the Oswestry Disability Index (ODI),21, 48 modified ODI,45 Roland-Morris Disability Questionnaire (RMDQ),18, 22, 23, 26, 39, 42, 43, 52, 53 and Funktionsfragebogen Hannover Functional Ability Questionnaire (FfBH);21 movement control was evaluated using movement control impairment test battery;45 and perceived level of recovery using Global Rating of Change Scale (GRCS).18

For biophysical outcomes, cortical excitation-related outcomes were evaluated using TMS including motor evoked potentials (MEPs),44 peak-to-peak MEP amplitude,46 cortical silent period (CSP),24, 46 map volume and discrete map peaks (motor cortical organization),24, 25 and cortical topography (of lumbar multifidus [LM] and erector spinae [ES]);45 muscle activation (of LM) using real-time ultrasound imaging,45 and muscle activity (of back muscles) using root-mean-square difference (RMSD) from surface electromyography (EMG);54 and regulatory autonomic nervous system (ANS) function including heart rate variability (HRV) components and respiratory sinus arrhythmia (RSA) from electrocardiogram data.49

For biomechanical outcomes, postural stability was evaluated using the Biodex Balance System (BBS);19, 40, 42 balance using Berg Balance Scale;19 and gait speed using 4-meter walk test.47

For psychosocial outcomes, quality of life (QOL) was evaluated using the EuroQuol-5 Dimension 5 level (EQ-5D-5L),26 or EuroQuol-5 Dimension 5 level 3-level (EQ-5D-5L),53 RAND 36-item health survey,21 36-item short-form health survey (SF-36),45 World Health Organization Quality of Life (WHOQOL);47 psychological well-being using 9-item Patient Health Questionnaire (PHQ-9),26 anxiety using Depression Anxiety Stress Scales (DASS),39 Hospital Anxiety and Depression Scale (HADS)21, 23 and 7-item Generalized Anxiety Disorder scale;22 anxiety state using Spielberger State and Trait Anxiety Inventory;51, 52 pain-related anxiety using Pain Anxiety Symptoms Scale (PASS);41 depression using Beck Depression Inventory (BDI),52 DASS,39 HADS,21, 23 and PHQ-9;22 affective state using Positive and Negative Affective Schedule;51, 52 pain catastrophizing using Pain Catastrophizing Scale (PCS);43, 50, 52 pain related fear and avoidance using 20-item Positive and Negative Affective Schedule (PANAS-20);22 fear of movement using Fear Avoidance Beliefs Questionnaire (FABQ);21 kinesiophobia using Tampa Scale for Kinesiophobia (TSK);41, 50 pain acceptance using 8-item Chronic Pain Acceptance Questionnaire (CPAQ-8);22 and executive function (inhibitory control and set shifting) using Comprehensive Trail Making Test – Second Edition (CTMT2) and Stroop Color Word Test (SCWT).50

Fourteen studies evaluated adverse events/side effects of tDCS, with seven studies using a researcher-developed questionnaire,19, 26, 40-42, 49 or an evaluation form,53 and seven relying on passively collected patient’s reports.18, 23, 45, 46, 51, 52, 54 However, twelve studies did not report adverse events/side effects20-22, 24, 25, 39, 43, 44, 47, 48, 50, 56 (Supplementary Digital Material 3: Supplementary Table II).18-26, 39-55

The main findings of the included studies are summarized in Supplementary Table II.

Effects of tDCS on clinical outcomes

Pain

Four studies on anodal-26, 43, 48 or cathodal-tDCS22 reported a significant reduction in pain intensity compared to sham-tDCS after intervention,26, 43, 48 at 4 weeks follow-up,26 and 6 weeks follow-up.22 Five studies on combined anodal-tDCS with other interventions including ABM,39 PES,18, 25 PNE,50 postural training,19 and PT47 reported significant reduction in pain intensity compared to sham-tDCS post-intervention,18, 19, 25, 39, 47, 50 and at 3 days,25 1 week, 2 weeks, 3 weeks,47 4 weeks,19, 39, 47 and 12 weeks follow-up.18 In one of the studies, anodal-tDCS plus OMT and sham-tDCS plus OMT resulted in clinically significant pain reduction compared to sham-tDCS plus sham OMT post-intervention,53 whereas in another study anodal-tDCS plus ABM was superior to no treatment.39 Also, one study reported a significant reduction in pain interference in favor of cathodal-tDCS over sham-tDCS.22 However, four studies reported no significant reduction in pain intensity between anodal-tDCS and sham-tDCS.21, 23, 51, 52 Additionally, another study reported no significant improvement in sensory and affective aspects of pain with anodal-tDCS plus PE or anodal-tDCS alone compared to sham, except for a clinically significant reduction in the affective pain aspect with anodal-tDCS at 12-week follow-up.18

Disability

Five studies reported a significant reduction in disability in favor of cathodal-tDCS,22 cathodal-tDCS plus CBT,41 anodal-tDCS,43 or anodal-tDCS plus ABM,39 or anodal-tDCS plus PES18 compared to sham-tDCS after intervention,18, 39, 41, 43 at 4 weeks39, 41 and 6 weeks22 follow-up. Additionally, anodal-tDCS plus ABM was reported to be superior to no treatment.39 In contrast, seven studies on anodal-tDCS alone23, 48, 51 or combined with CBT20, 21 or exercise26, 45 found no significant reduction in disability compared to sham-tDCS. Moreover, another study53 reported no significant difference in disability between anodal-tDCS plus OMT, sham-tDCS plus OMT, and sham-tDCS plus sham OMT post-intervention.

Pain sensitivity/central sensitization

One study25 reported significant improvement in central sensitization measured with the Schober test and PPT in favor of anodal-tDCS plus PES compared to sham treatment after intervention. However, two studies51, 52 found no significant differences between anodal-tDCS and sham-tDCS in PPT, CPM, and TSP. One study20 found no significant alterations in electrical and thermal pain perceptions either within or between anodal- or cathodal-tDCS and sham-tDCS. Additionally, another study49 found no significant changes in RSA or other frequency-domain HRV components between real-tDCS and sham-tDCS.

Movement control

One study45 on anodal-tDCS plus MCE versus sham-tDCS reported no significant changes in movement control post-intervention.

Perceived level of recovery

One study18 showed significant improvement in global perceived effect in favor of anodal-tDCS plus PES compared to sham-tDCS at 12- and 24-week follow-ups.

Effects of tDCS on biophysical outcomes

Cortical excitation

One study24 reported a significant increase in cortical excitability including increased map volume and discrete map peaks and reduced CSP in favor of combined anodal-tDCS plus PES in CLBP patients compared to pain-free controls. Significant fewer discrete peaks difference in both ES and LM in favor of anodal-tDCS plus MCE compared to sham-tDCS were also reported in another study.45 However, while tDCS plus MCE in CLBP patients decreased MEP amplitude (P2P), no significant difference was found compared to pain-free controls.46 Moreover, tDCS plus MCE did result in significant differences in CSP and LM activation.46 Similarly, in another study,44 though a single application of tDCS decreased MEPs, no significant difference was found between CLBP patients and pain-free controls.

Muscle activation/activity

There were no significant changes in LM activation in the trial comparing anodal-tDCS plus MCE and sham-tDCS.45 Similarly, in a quasi-experimental study,46 anodal-tDCS in CLBP patients did not result in significant changes in LM activation compared to pain-free controls. For muscle activity, one study54 reported no significant difference between anodal-tDCS and sham-tDCS.

Regulatory autonomic nervous system function

One study49 reported no significant changes in RSA or other frequency-domain HRV components with anodal-tDCS compared to sham-tDCS post-intervention, although anodal-tDCS showed a significant increase in the standard deviation of normal-to-normal R-R intervals (SDNN), which is a measure of ANS balance.49

Effects of tDCS on biomechanical outcomes

Postural stability

Three studies reported significant improvement in postural stability in favor of anodal-tDCS or cathodal-tDCS40 or anodal-tDCS plus postural training19 compared to sham-tDCS immediately19, 40, 42 and 24 hours post-intervention,40, 42 and at 1 week40, 42 and 4 weeks follow-up.19

Balance

One study19 evaluating postural stability also showed significant balance improvement in favor of anodal-tDCS plus postural training compared to sham-tDCS, post-intervention and at 4 weeks follow-up.

Gait speed

One study47 that assessed gait speed showed no significant difference between anodal-tDCS plus physical therapy and sham-tDCS.

Effects of tDCS on psychosocial outcomes

Quality of life (QOL)

Five studies evaluating QOL reported no significant difference between anodal-tDCS plus CBT,21 exercise,26, 45 physical therapy,47 or OMT53compared to sham-tDCS. Additionally, sham-tDCS plus OMT was not superior to sham-tDCS.53

Psychological well being

One study26 on anodal-tDCS plus exercise found significant improvement in psychological well-being compared to sham-tDCS at 4 weeks follow-up.

Anxiety

One study39 reported a significant reduction in anxiety with anodal-tDCS plus ABM compared to sham-tDCS or no treatment after intervention and at 4 weeks follow-up. Also, another study41 found cathodal-tDCS plus CBT to significantly reduce pain-related anxiety compared to sham-tDCS or CBT only immediately post-intervention and at 4 weeks follow-up. However, four studies found no significant difference in anxiety between anodal- and sham-tDCS23, 51, 52 or cathodal-tDCS.22

Depression

Two studies reported that anodal-tDCS plus ABM or cathodal-tDCS alone significantly reduced depression compared to sham-tDCS or no treatment post-intervention,39 at 4 weeks39 and 6 weeks22 follow-up. However, three studies found no significant difference in depression between anodal-tDCS23, 52 or anodal-tDCS plus CBT and sham-tDCS.21

Pain catastrophizing

One study50 evaluating pain catastrophizing reported a significant improvement with anodal-tDCS plus PNE compared to sham-tDCS plus PNE post-intervention.50 In contrast, two studies43, 52 found no significant difference in pain catastrophizing between anodal-tDCS and sham-tDCS.

Pain and movement-related fear and avoidance

One study41 reported a significant reduction in fear of movement with both anodal- and cathodal-tDCS compared to sham-tDCS after intervention and at 4 weeks follow-up. In contrast, two studies found no significant difference in fear of movement between anodal-tDCS plus CBT21 or PNE50 and sham-tDCS. Also, another study reported no significant improvement in pain-related fear and avoidance in either cathodal-tDCS or sham-tDCS.22

Pain acceptance

One study22 reported significant improvement in pain acceptance in favor of cathodal-tDCS compared to sham-tDCS at 6 weeks of follow-up.22

Executive function

One study50 found significant improvements in inhibitory control and attention interference in favor of anodal-tDCS + PNE compared to sham-tDCS post-intervention.50

Adverse events/side effects

Among the fourteen studies evaluating adverse events/side effects, mild or minor effects such as burning or warming sensation,18, 45, 51 difficulty concentrating,18, 26, 52, 53 dizziness,23, 54 drowsiness,53 headaches,18, 19, 23, 26, 52, 53 increased fatigue,52, 53 mood change,26 nausea,18, 53 pain,18, 53 skin redness,18, 26, 53 sleepiness,18, 26, 49, 52 and tingling and/or itching at the stimulation site18, 19, 26, 40-42, 45, 46, 49, 51, 53, 54 were commonly reported. In general, these effects were equally distributed between the active and sham stimulation groups.

Risk of bias assessment

Overall, thirteen studies (50%) had a low risk of bias, while nine (34.6%) showed a high or serious risk of bias. Out of the sixteen RCTs with a parallel design (61.5%), three were categorized as having a high risk of bias, two as having some concerns, and eleven as having a low risk of bias (Figure 2).

Figure 2.

Figure 2

Open in a new tab

—Risk of bias assessment score for the included parallel randomized trials.

Bias due to missing outcome data and bias in selection of the reported results were the primary methodological weaknesses identified. Of the seven RCTs with a cross-over design (Figure 3), three were rated as having a high risk of bias, two as having some concerns, and the other two as having a low risk of bias. A high risk of bias rating was prevalent with the domain ‘bias in measurement of the outcome’. As for non-randomized studies, all the studies were rated as having a serious risk of bias, with failure to account for confounding variables being the most common attribution bias (Figure 4).

Figure 3.

Figure 3

Open in a new tab

—Risk of bias assessment score for the included cross-over randomized trials.

Figure 4.

Figure 4

Open in a new tab

—Risk of bias assessment score for the included non-randomized studies.

Discussion

This study aims to identify and map the existing literature on tDCS for CLBP. We have included twenty-six studies with considerable variability in terms of tDCS parameters employed, outcomes evaluated, treatment efficacy, and follow-up periods. Regarding the effects of tDCS, there are inconsistent results supporting its utility either alone or priming with other interventions across clinical, biophysical, biomechanical, and psychosocial outcomes.

From the study characteristics point of view, the majority of the studies used large electrodes (5x7 cm2) whereas only a few utilized moderate or small electrodes. The size of the electrodes directly influences the current density, which might potentially influence the efficacy of the stimulation. Ramp-up and ramp-down protocols varied, with most studies using shorter duration (10 s), which may potentially influence the stimulation’s tolerability. For example, studies using a higher protocol such as 30 s51 or 60 s52 ramp-up and ramp-down, reported more adverse events associated with tDCS compared to those using a 10 s protocol.19, 40-42, 46, 54 Regarding current intensity, most of the studies used 2 mA, which is within the recommended range (used 1-2 mA). Stimulation durations also varied, predominantly set at 20 min, but with some studies extending to 40 min. Similarly, the frequency of the treatment varied widely, with session counts ranging from a single session to as many as 50 sessions, which might have contributed to the mixed results observed. Furthermore, follow-up periods were generally short, spanning from just 3 days to 6 months. The lack of longer follow-up periods undermines the sustainability of tDCS over time.

The findings that the majority of the included studies examined the effects of tDCS alone compared to sham stimulation is critical to elucidate its efficacy without the influence of placebo effects. Conversely, other studies applied tDCS priming with other interventions as an integrative approach to enhance outcomes by simultaneously targeting multiple interconnected cognitive, physical, and neurophysiological pathways. The M1 is the predominant targeted area using anodal stimulation (Supplementary Table II). Although the efficacy of stimulation of different brain areas is unlikely to be similar, four studies applied cathodal-tDCS over the left DLPFC22, 40-42 which raises intriguing questions about the potential benefits of targeted inhibition.

We observed more inconsistent results concerning the effects of tDCS on clinical outcomes. For pain and disability, some studies reported significant improvement following anodal-tDCS18, 26, 43, 48 or cathodal-tDCS22, 39, 41 as a stand-alone or in combination with other interventions such as CBT,41 PES,18, 25 PNE,50 OMT,53 physical therapy,47 and postural training,19 with improvements lasting for up to 12 weeks for pain18 and 6 weeks for disability.22 Other studies, however, found no significant improvements either with tDCS alone23, 48, 51, 52 or in combination with other interventions,20, 21, 26, 45 akin to the findings of a previous systematic review.29 As for central sensitization and perceived recovery levels, limited evidence emerged due to the small number of studies.18, 25 Although we did not perform a formal analysis in this review, the positive results regarding pain and disability outcomes may be due to the inclusion of studies published from 2019 onwards, in contrast to the previous review, which included studies up to 2019.29

As for biophysical outcomes, while two studies reported significant improvements in cortical excitation parameters (i.e. MEP amplitude, map volume, and discrete map peaks CSP) in favor of combined anodal-tDCS and PES24 or MCE46 in CLBP patients compared to pain-free controls, the quasi-experimental design of these studies limits the strength of the evidence. Additionally, except for one trial that applied anodal-tDCS alone for 6 sessions,54 most studies evaluating biophysical outcomes24, 46 were limited to a single application, which may be insufficient to produce a substantial effect.

In examining the effects of tDCS on biomechanical outcomes, cathodal-tDCS alone22 or anodal-tDCS combined with postural training19 were shown to be beneficial for improving postural stability22 and balance19 in the short-term compared to sham stimulation, but not gait speed.47 These improvements may be attributed to tDCS’s ability to enhance neural activity in key cognitive and postural regions of the brain. By modulating the excitability of the primary motor cortex and other associated areas, tDCS can enhance communication between brain regions responsible for motor control and coordination.40

Regarding the effects of tDCS on psychosocial outcomes, although anodal tDCS – whether used alone or combined with CBT,37 OMT,53 or physical therapy,47 – improved QOL, none of these interventions demonstrated superiority compared to sham stimulation. However, combined anodal-tDCS and exercise26 or ABM,39 and combined cathodal-tDCS and CBT41 were superior to sham stimulation in improving psychological well-being and reducing anxiety, respectively. Similarly, anodal-tDCS plus ABM39 or cathodal-tDCS alone22 improved depression better than sham stimulation. Interestingly, both anodal- and cathodal-tDCS were shown to reduce fear of movement compared to sham stimulation.41 While anodal-tDCS alone was as effective as sham stimulation for pain catastrophizing,43 combining it with PNE yielded superior results versus sham-tDCS plus PNE.50 This combination also enhanced inhibitory control and attention interference more than sham stimulation,50 demonstrating PNE’s additive benefits for both pain behavior-related and cognitive-related outcomes. Furthermore, cathodal-tDCS was superior to sham stimulation in enhancing pain acceptance,22 indicating a greater willingness to accept pain and engage in daily activities despite experiencing it. The mechanisms behind the effects of tDCS on these psychosocial outcomes may involve the cortical excitation of targeted brain regions, leading to either enhanced or reduced activity that affects cognitive and emotional processing. Additionally, the incorporation of ABM, exercise, and CBT could further augment these effects. However, due to the limited number of studies and conflicting findings, the effects of tDCS on psychosocial outcomes remain inconclusive.

Despite the inconsistent results regarding the usefulness of tDCS for CLBP, it appears safe as mild or minor adverse events/side effects were generally reported even though only 54.2% of the included studies evaluated such effects. The most commonly reported tDCS side effects were mild tingling and/or itching at the stimulation site. However, since there were fewer serious side effects were reported, our review is in harmony with the literature suggesting that tDCS is a safe intervention when applied within the limits of parameters.57

With regard to overall quality rating, half of the studies (50%) had a low risk of bias, while approximately one-third of them (34.6%) had a high or serious risk. Notably, the most common methodological weaknesses included outcome measurement bias in crossover RCTs and confounding bias in non-randomized studies. As expected, quasi-experimental studies exhibited higher risk due to their lack of randomization. These shortcomings highlight areas for improvement in future studies.

Our review identifies several knowledge gaps which need to be addressed in future studies. Although most tDCS parameters employed in the included studies were within the recommended limits, it is worth noting that parameters such as electrode size and placement, current intensity and density, duration and frequency of stimulation, and the use of stimulation alone or priming with other intervention may influence tDCS efficacy.58, 59 For example, studies employing multiple tDCS sessions22, 53, 55 generally show better results than those employing a single session20, 46, 49 or fewer sessions,21, 52 implying that a single session may be insufficient to target hotspot areas of the brain. In another vein, studies priming tDCS with other interventions, such as PNE (a modern neuroscience education)50 and CBT41 demonstrated better clinical and psychological outcomes as these interventions target pain-related brain regions, similarly to tDCS. Analogously, studies combining tDCS with sensory stimulation interventions such as SMT55 or ABM39 showed greater effects, highlighting their synergistic potential with tDCS. However, studies on combined interventions are limited. Furthermore, there has been a dearth of studies evaluating long-term effects, with the longest follow-up being 12 weeks in one study.18

The shortcomings of existing studies underscore the need for further robust RCTs to clarify the utility of tDCS in CLBP. Stimulation protocol should be standardized and longer follow-ups should be incorporated. The effects of varying frequencies of tDCS can be compared to determine the optimal dosage. We suggest that additional studies explore the effects of combining tDCS with other interventions to determine the most effective approach for meeting patients’ needs. Moreover, future RCTs should evaluate clinical outcomes like cortical excitation to better understand the cortical mechanisms underlying CLBP as most studies evaluating this outcome used a quasi-experimental design.24, 44, 46 Although the methodology reporting of trials seems to have improved over time, with most recent studies in the present review exhibiting a low risk of bias, further improvements are needed.

Limitations of the study

Our review is limited by the small sample size and number of included studies. Moreover, the lack of formal analysis like meta-analysis, is an obvious limitation, although heterogeneity was generally observed among the included studies. Further, since this review was not intended to guide clinical practice, future rigorous reviews with formal analyses are desirable to elucidate the efficacy of tDCS in CLBP, particularly, as new studies continue to emerge.

Conclusions

Available studies on tDCS for CLBP show significant variation in tDCS protocols and outcomes, as well as mixed results regarding its effectiveness, particularly for clinical outcomes. Differences in treatment protocols, study design, sample characteristics, and outcome measures may have contributed to these inconsistencies. However, since tDCS is a safe intervention with potential therapeutic benefits, further well-designed RCTs, accounting for the limitations of current studies, are needed to elucidate its effectiveness. Despite this review’s limitations, its findings can inform future reviews with formal analysis.

Supplementary Digital Material 1

Supplementary Text File 1

Search strategy

Supplementary Digital Material 2

Supplementary Table I

Characteristics of the included studies (N.=26).18-26, 39-55

Supplementary Digital Material 3

Supplementary Table II

Main findings of the included studies (N.=26).18-26, 39-55

Footnotes

Conflicts of interest: The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.

References

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Text File 1

Search strategy

Supplementary_Digital_Material1.pdf (13.4KB, pdf)

Supplementary Table I

Characteristics of the included studies (N.=26).18-26, 39-55

Supplementary_Digital_Material2.pdf (269.4KB, pdf)

Supplementary Table II

Main findings of the included studies (N.=26).18-26, 39-55

Supplementary_Digital_Material3.pdf (170KB, pdf)

Articles from European Journal of Physical and Rehabilitation Medicine are provided here courtesy of Edizioni Minerva Medica S.p.A.

RESOURCES