Inspiratory muscle training facilitates weaning from mechanical ventilation among patients in the intensive care unit: a systematic review

Question: Does inspiratory muscle training improve inspiratory muscle strength in adults receiving mechanical ventilation? Does it improve the duration or success of weaning? Does it affect length of stay, reintubation, tracheostomy, survival, or the need for post-extubation non-invasive ventilation? Is it tolerable and does it cause adverse events?

Design: Systematic review of randomised trials.

Participants: Adults receiving mechanical ventilation.

Intervention: Inspiratory muscle training versus sham or no inspiratory muscle training.

Outcome measures: Data were extracted regarding: inspiratory muscle strength and endurance; the rapid shallow breathing index; weaning success and duration; duration of mechanical ventilation; reintubation; tracheostomy; length of stay; use of non-invasive ventilation after extubation; survival; readmission; tolerability and adverse events.

Results: Ten studies involving 394 participants were included. Heterogeneity within some meta-analyses was high. Random-effects meta-analyses showed that the training significantly improved maximal inspiratory pressure (MD 7 cmH2O, 95% CI 5 to 9), the rapid shallow breathing index (MD 15 breaths/min/l, 95% CI 8 to 23) and weaning success (RR 1.34, 95% CI 1.02 to 1.76). Although only assessed in individual studies, significant benefits were also reported for the time spent on non-invasive ventilation after weaning (MD 16 hours, 95% CI 13 to 18), length of stay in the intensive care unit (MD 4.5 days, 95% CI 3.6 to 5.4) and length of stay in hospital (MD 4.4 days, 95% CI 3.4 to 5.5). Weaning duration decreased in the subgroup of patients with known weaning difficulty. The other outcomes weren't significantly affected or weren't measured.

Conclusion: Inspiratory muscle training for selected patients in the intensive care unit facilitates weaning, with potential reductions in length of stay and the duration of non-invasive ventilatory support after extubation. The heterogeneity among the results suggests that the effects of inspiratory muscle training may vary; this perhaps depends on factors such as the components of usual care or the patient's characteristics.