Previous reviews relating to the effects of respiratory muscle training (RMT) after stroke tend to focus on only one type of training (inspiratory or expiratory muscles) and most based the results on poor-quality studies (PEDro score ≤4).

With this systematic review and meta-analysis, we aimed to determine the effects of RMT (inspiratory or expiratory muscle training, or mixed) on exercise tolerance, respiratory muscle function and pulmonary function and also the effects depending on the type of training performed at short- and medium-term in post-stroke.

Databases searched were MEDLINE, PEDro, CINAHL, EMBASE and Web of Science up to the end of April 2020. The quality and risk of bias for each included study was examined by the PEDro scale (including only high-quality studies) and Cochrane Risk of Bias tool.

Nine studies (463 patients) were included. The meta-analysis showed a significant increase in exercise tolerance [4 studies; n = 111; standardized mean difference [SMD] = 0.65 (95% confidence interval 0.27–1.04)]; inspiratory muscle strength [9 studies; n = 344; SMD = 0.65 (0.17–1.13)]; inspiratory muscle endurance [3 studies; n = 81; SMD = 1.19 (0.71–1.66)]; diaphragm thickness [3 studies; n = 79; SMD = 0.9 (0.43–1.37)]; and peak expiratory flow [3 studies; n = 84; SMD = 0.55 (0.03–1.08)] in the short-term. There were no benefits on expiratory muscle strength and pulmonary function variables (forced expiratory volume in 1 s) in the short-term.

The meta-analysis provided moderate-quality evidence that RMT improves exercise tolerance, diaphragm thickness and pulmonary function (i.e., peak expiratory flow) and low-quality evidence for the effects on inspiratory muscle strength and endurance in stroke survivors in the short-term. None of these effects are retained in the medium-term. Combined inspiratory and expiratory muscle training seems to promote greater respiratory changes than inspiratory muscle training alone.