BACKGROUND: Chronic low back pain (CLBP) involves peripheral nociception, central sensitization, and persistent disability. Transcranial magnetic stimulation (TMS) may reduce pain by modulating cortical excitability and central pain processing, while electroacupuncture (EA) combines acupuncture with electrical stimulation to influence nociceptive pathways and improve function. However, the relative efficacy of these and other physical agent therapies remains clinically uncertain in CLBP management.
OBJECTIVE: To systematically compare the efficacy of different PATs in pain relief and physical function improvement for CLBP patients via network meta-analysis (NMA).
METHODS: Following PRISMA guidelines, PubMed, Embase, the Cochrane Library, and Web of Science were searched from inception through November 2025 for randomized controlled trials. The primary outcomes were VAS/NRS pain scores and ODI disability. A bayesian random-effects NMA was used to estimate mean differences (MDs), 95% credible intervals (CrIs), and SUCRA rankings.
RESULTS: Fifty RCTs involving 2,811 participants evaluated nine interventions: TMS, PMS, ESWT, EA, tDCS, PEMF, TENS, IFC, and PMS+TMS. TMS ranked highest for pain relief (SUCRA=0.98), followed by PMS (0.82) and ESWT (0.71), although four TMS studies had a high risk of performance bias. EA ranked highest for ODI improvement (0.94), followed by ESWT (0.72) and PMS (0.68). PMS+TMS showed no additional benefit over monotherapy.
CONCLUSION: TMS may provide superior pain relief among physical agent therapies for CLBP, yet these results should be interpreted cautiously owing to potential bias in the available evidence. EA shows the greatest benefit for functional recovery, and moderate-to-large analgesic effects are also observed with ESWT and PMS. When choosing physical agent therapies, clinicians ought to weigh these comparative rankings against the methodological limitations of the analysis.
| Discipline Area | Score |
|---|---|
| Physician |  |
Physician rater
Only repetitive transcranial magnetic stimulation, peripheral magnetic stimulation, and extracorporeal shock wave therapy exceeded the pain MCID. Electroacupuncture ranked first for improving disability, but based on two post-surgical trials. Populations are mixed (non-specific pain, stenosis, radiculopathy, post-surgery). There is no GRADE/CINeMA certainty rating, no adverse-event synthesis, and a short follow-up. Most intervention comparisons were indirect.
Physician rater
Physical agents should be used as adjuncts to active exercise. The proposed phenotype-based approach remains hypothetical and unproven. The author’s conclusions are correct: well-designed head-to-head RCTs with standardised protocols and follow-up periods of at least 6 months are essential.
Physician rater
The authors appear more confident in the treatment rankings than the network structure and certainty of evidence justify. The absence of GRADE/CINeMA, the reliance on indirect comparisons, the substantial clinical heterogeneity, and the elevated risk of bias in several studies mean that the conclusion should be interpreted as a hypothesis-generating finding rather than a definitive clinical recommendation.
Physician rater
The authors appear more confident in the treatment rankings than the network structure and certainty of evidence justify. The absence of GRADE/CINeMA, the reliance on indirect comparisons, the substantial clinical heterogeneity, and the elevated risk of bias in several studies mean that the conclusion should be interpreted as a hypothesis-generating finding rather than a definitive clinical recommendation.
, McMaster University