Comparative effectiveness of dysphagia rehabilitation in head and neck cancer: A network meta-analysis.

Abstract

ObjectiveTo compare rehabilitation interventions for dysphagia after head and neck cancer treatment using network meta-analysis.Data sourcesPubMed/MEDLINE, EMBASE, Cochrane Library, Scopus, and Web of Science were searched from inception to 21 July 2026.Review methodsRandomised controlled trials evaluating swallowing exercise, neuromuscular electrical stimulation combined with exercise (NMES+ex), speech therapy combined with exercise (Speech+ex) or telerehabilitation were included. Random-effects network meta-analyses estimated Hedges' g standardised mean differences (SMDs) and treatment rankings across four outcome domains. Confidence was assessed using the Confidence in Network Meta-Analysis framework.ResultsTwenty-one trials were included in the systematic review; 18 trials involving 1139 participants contributed to connected networks. Exercise improved clinician-rated swallowing compared with usual care (SMD 0.98, 95% confidence interval 0.28 to 1.69) and instrumental swallowing compared with telerehabilitation (SMD 1.24, 0.07 to 2.42). No other comparisons were statistically significant. Speech+ex ranked highest for clinician-rated outcomes, exercise for instrumental outcomes and NMES+ex for patient-reported quality of life and symptom/handicap outcomes. In sensitivity analyses separating exercise types, prophylactic exercise ranked highest for instrumental and quality-of-life outcomes. Confidence ranged from moderate to very low and was very low for all patient-reported comparisons.ConclusionRehabilitation effects varied by outcome domain, with no consistently superior intervention. Exercise showed benefits for selected clinician-rated and instrumental outcomes, but sparse networks, imprecision and inconsistency limit certainty. These findings support individualised intervention selection and adequately powered trials using standardised outcomes.