Nineteen RCTs were included in the original review and six were identified in the update. Sixteen trials (10 from the original review and six from the update review) compared IMT to sham IMT and were included in the update review (n=426). Sample sizes ranged from 14 to 112.
Only three of the 16 trials described the randomisation process. Eleven studies were double-blind. Fourteen studies reported on participant withdrawal. Thirteen studies reported that treatment groups were comparable at baseline. Twelve studies used ITT.
Inspiratory muscle strength: There were significant improvements in PImax (WMD 11.6cm H2O, 95% CI 8.7 to 14.4), PImax % predicted (WMD 23.2%, 95% CI 11.3 to 35.1) and peak inspiratory flow rate (WMD 12.6L/min, 95% CI 9.7 to 15.6) that favoured participants in the IMT group compared with the sham IMT group.
Inspiratory muscle endurance: There were significant improvements in respiratory muscle endurance time (WMD 4.4 min, 95% CI 0.7 to 8.2) and maximal inspiratory threshold load (WMD 1.4kPa, 95% CI 0.8 to 1.9) that favoured participants in the IMT group.
Exercise capacity: There were significant improvements in maximum exercise minute ventilation (4.9L/min, 95% CI -8.2 to -1.7), Borg score for respiratory effort (1.8, 95% CI -2.4 to -1.2) and six-minute walk test (32.1m, 95% CI 11.6 to 52.7) that favoured participants in the IMT group. There were no significant differences between groups for maximal oxygen consumption and work rate maximum.
Significant improvements that favoured participants in the IMT group in transitional dyspnoea index focal score and quality of life were reported.
The authors reported that all sensitivity analyses resulted in the same conclusions for overall effect; no further details were given.
There was significant statistical heterogeneity between the trials for most outcomes assessed.