Throat exercises may ease sleep apnoea; lifting doesn't

Obstructive sleep apnoea is driven in part by what the muscles of the upper airway do during sleep, so training those muscles is a plausible way to reduce its severity. That premise underpins oropharyngeal training — exercises for the mouth and throat — and respiratory muscle training, while some have also proposed general resistance training. Scattered small trials have tested each approach, but it has been unclear whether any of them reliably reduces the apnoea-hypopnoea index, the standard measure of severity. This review pooled the randomised evidence within each modality, and also asked whether daytime sleepiness improves alongside.

The authors searched PubMed, EBSCO and Scopus for randomised controlled trials published between 1 January 2015 and 31 January 2025. From 1805 records identified, 13 RCTs with 429 participants in total met eligibility. Trials were grouped by modality — resistance training, oropharyngeal training and respiratory muscle training — and pooled separately using random-effects meta-analyses. Severity was measured with the apnoea-hypopnoea index (AHI) and daytime sleepiness with the Epworth Sleepiness Scale (ESS).

Oropharyngeal training reduced AHI by a mean of 7.55 events per hour (95% CI -13.39 to -1.71), and respiratory muscle training reduced it by 3.73 events per hour (95% CI -7.44 to -0.03), an effect that only just reached statistical significance. Resistance training showed no significant change (mean difference -5.88 events per hour, with a confidence interval spanning both benefit and harm). Across modalities, the Epworth Sleepiness Scale showed no consistent, clinically meaningful improvement. Heterogeneity was substantial for resistance-training trials on AHI and for oropharyngeal trials on sleepiness, reflecting variable protocols, durations and participants.

These results support oropharyngeal and respiratory muscle training as adjunctive options for reducing the physiological severity of obstructive sleep apnoea, while complicating the idea that any form of exercise helps equally. The gap between the objective measure (fewer breathing events) and the symptom measure (sleepiness that did not consistently improve) is a notable open question. The authors highlight the need for standardised protocols, longer follow-up and outcomes that matter to patients. Whether resistance training could contribute in some combination or population remains untested by this evidence.

This is a reason to think that targeted training of the throat and breathing muscles can shift the objective severity of sleep apnoea, whereas general strength work alone showed no such effect in these trials. It is equally a reason for realistic expectations: a lower apnoea-hypopnoea index did not consistently translate into feeling less sleepy during the day. The trials were small and varied, so the size of any benefit for a given person is uncertain. Anyone managing sleep apnoea would sensibly treat these approaches as adjuncts under clinical guidance rather than replacements for their existing care.