Key points are not available for this paper at this time.
Background: Speech-language therapy (SLT) is standard care for post-stroke aphasia (PSA), and adjunctive transcranial magnetic stimulation (TMS) may further improve language function. Which TMS protocol optimally complements SLT for which language domain at which time point remains unresolved, motivating this network meta-analysis. Objective: To compare TMS protocols combined with SLT across immediate and long-term language outcomes in post-stroke aphasia. Methods: Web of Science, EMBASE, PubMed, and the Cochrane Library were searched from inception to 24 January 2026 for randomized trials of TMS plus SLT in post-stroke aphasia. Bayesian network meta-analyses were run in R 4.4.2 with the gemtc and rjags packages, with outcomes grouped into seven immediate and four long-term language domains and effect sizes reported as standardized mean differences (SMDs) with 95% credible intervals (CrIs). Treatments were ranked by the surface under the cumulative ranking curve (SUCRA), and certainty of evidence was rated using GRADE (Grading of Recommendations Assessment, Development and Evaluation). Results: Across 26 trials enrolling 1,136 patients, three protocols led across the language domains. LF-rTMS+Music+SLT ranked first for immediate aphasia severity (SUCRA = 96.86%), repetition (97.25%), naming (99.69%), and spontaneous speech (95.67%); LF-rTMS+SLT, for writing (87.73%); and Dual-rTMS+SLT, for immediate functional communication (91.49%) and comprehension (97.57%). Dual-rTMS+SLT was also optimal across all four long-term outcomes (overall aphasia severity 95.82%, comprehension 95.06%, naming 99.74%, repetition 97.90%). Heterogeneity was low-to-moderate (moderate-to-high for immediate aphasia severity), and all Bayesian models converged (PSRF < 1.05). Evidence was moderate-to-high for standard comparisons (LF-rTMS+SLT vs SLT) but low to very low for the highest-ranked novel combinations such as LF-rTMS+Music+SLT, driven by imprecision and indirectness. Conclusion: The optimal TMS-plus-SLT protocol depends on which language domain and which time point matters. LF-rTMS+Music+SLT led the immediate domains, while Dual-rTMS+SLT led the long-term domains. Standard adjunctive comparisons (e.g. LF-rTMS+SLT vs SLT) rest on moderate-to-high certainty evidence, but the highest-ranked novel combinations rest on low-certainty evidence and require head-to-head confirmation. A text-heavy research infographic with a large title at the top, two text columns and a row of small network diagrams along the bottom. The layout uses section headings and table grids to present study background, methods, results and conclusion. Text is in English. Title: Transcranial Magnetic Stimulation Combined with Speech-Language Therapy for Post-Stroke Aphasia Recovery: A Network Meta-Analysis. Left column headings and text. Background: Post-stroke aphasia (PSA) affects approximately 30 percent stroke survivors. Speech-language therapy (SLT) is standard care. Optimal TMS plus SLT protocol remains unclear. Methods: Web of Science, EMBASE, PubMed and the Cochrane Library were searched from inception to January 24, 2026. Risk of bias was assessed using Cochrane RoB 2.0. A Bayesian NMA was performed using R 4.4.2 and Stata 18. Conclusion: Optimal TMS plus SLT depends on language domain and time point. Short-term: LF-rTMS plus Music plus SLT is best. Long-term: Dual-rTMS plus SLT is best. Head-to-head trials are needed for confirmation. Right column heading and tables. Results: 26 RCTs (1,136 patients). Two tables are labeled Immediate Outcomes (Post-treatment) and Long-term Outcomes (Follow-up). Each table has columns Language Domain, Most Effective Intervention and SUCRA. Immediate Outcomes (Post-treatment) rows and SUCRA values: Aphasia severity, LF-rTMS plus Music plus SLT, 96.86 percent. Repetition, LF-rTMS plus Music plus SLT, 97.25 percent. Naming, LF-rTMS plus Music plus SLT, 99.69 percent. Spontaneous speech, LF-rTMS plus Music plus SLT, 95.67 percent. Writing, LF-rTMS plus SLT, 87.73 percent. Comprehension, Dual-rTMS plus SLT, 97.57 percent. Functional communication, Dual-rTMS plus SLT, 91.49 percent. Long-term Outcomes (Follow-up) rows and SUCRA values: Aphasia severity, Dual-rTMS plus SLT, 95.82 percent. Repetition, Dual-rTMS plus SLT, 97.90 percent. Naming, Dual-rTMS plus SLT, 99.74 percent. Comprehension, Dual-rTMS plus SLT, 95.06 percent. Along the bottom edge, multiple small network meta-analysis diagrams show circular nodes connected by lines, with some connections drawn thicker than others.Infographic: TMS & Speech Therapy for Aphasia Recovery Post-Stroke. Keywords: stroke, cerebrovascular accident, aphasia, transcranial magnetic stimulation, network meta-analysis
Lin et al. (2026) studied this question.