Comparative Efficacy of Non-invasive Brain Stimulation for Pain, Anxiety, and Depression: A Systematic Review and Network Meta-analysis

Although non-invasive brain stimulation (NIBS) has been used to manage chronic musculoskeletal pain (CMP) and related emotional symptoms, the comparative efficacy of different NIBS techniques remains unclear. Therefore, this network meta-analysis (NMA) compared the effects of NIBS interventions on pain intensity, anxiety, and depressive symptoms in CMP. We searched seven databases from inception to 7 October 2025, including PubMed, the Cochrane Library, Web of Science, Embase, MEDLINE, PsycINFO, and CINAHL, and included only English-language randomized controlled trials evaluating NIBS for CMP that reported pain, anxiety, or depression outcomes. Effect sizes were synthesized using a frequentist NMA and expressed as standardized mean differences (SMDs) or mean differences (MDs) with 95% confidence intervals (CIs). Risk of bias was assessed with the Cochrane Risk of Bias 2.0 tool. A total of 50 randomized controlled trials were included, involving 2,203 participants with an average age ranging from 31 to 75 years; approximately 82% of the participants were women. The NMA showed that: (1) in the overall CMP population, low-frequency repetitive transcranial magnetic stimulation (LF-rTMS) targeting the dorsolateral prefrontal cortex (DLPFC) (SMD = − 1.19, 95% CI − 1.86 to − 0.52), transcranial random noise stimulation (tRNS) targeting the primary motor cortex (M1) (SMD = − 1.66, 95% CI − 3.05 to − 0.28), and intermittent theta-burst stimulation (iTBS) targeting the DLPFC (SMD = − 0.78, 95% CI − 1.63 to 0.08) ranked first for pain, anxiety, and depression, respectively. Sensitivity analyses suggested that LF-rTMS targeting the DLPFC may be the most robust option for CMP with comorbid emotional symptoms; (2) in fibromyalgia, LF-rTMS targeting the DLPFC ranked first for pain (SMD = − 1.19, 95% CI − 1.99 to − 0.39) and depression (SMD = − 0.63, 95% CI − 0.98 to − 0.27), while tRNS targeting M1 (SMD = − 1.66, 95% CI − 2.99 to − 0.34) ranked first for anxiety; after excluding sparse-evidence interventions, transcranial direct current stimulation (tDCS) targeting M1 became the preferred option for anxiety; (3)in non-fibromyalgia populations, cranial electrotherapy stimulation (CES) (SMD = − 1.39, 95% CI − 2.41 to − 0.37) ranked first for pain, while iTBS targeting the DLPFC (MD = − 0.79, 95% CI − 1.48 to − 0.10) showed potential benefit for depression, although supporting evidence was limited. Different non-invasive brain stimulation modalities may confer outcome-specific benefits in chronic musculoskeletal pain. Overall, low-frequency repetitive transcranial magnetic stimulation targeting the dorsolateral prefrontal cortex showed relatively robust effects across outcomes in chronic musculoskeletal pain and on pain and depression in fibromyalgia; however, further high-quality clinical trials are needed. CRD420251131206.