Figure 1 presents the flow of studies through the review. Authors of all the Libraries included studies were contacted to clarify interpretation and or extraction of data and all authors responded to the queries. There were no disagreements regarding

eligibility or the extracted data, so arbitration by the third author was not required. All of the studies (n = 3) reported the effects of inspiratory muscle training on inspiratory muscle strength as measured by maximal inspiratory pressure. Two studies reported data about weaning success (Cader et al 2010, Martin et al 2011), two studies find more reported data on weaning duration (Cader et al 2010, Caruso et al 2005), and three studies reported survival data (Cader et al 2010, Caruso et al 2005, Martin et al 2011). Therefore, the effect of inspiratory muscle training was examined using meta-analysis for four outcomes: inspiratory muscle strength, weaning success, weaning duration, and survival. Only one study reported data about reintubation (Caruso et al 2005) and tracheostomy (Cader et al 2010) and so these outcomes could not be meta-analysed. No studies reported inspiratory muscle endurance, the duration of unassisted breathing periods, and

length of stay in the intensive care unit and hospital. The quality of the included studies is outlined in Table 1 and a summary of the studies is presented in Table 2. Quality: The mean PEDro score of the included studies was 6. In all studies, randomisation was carried out correctly and group data and between-group comparisons were reported adequately. No study blinded participants or therapists, found but one study ( Martin et al 2011) blinded assessors. MLN8237 molecular weight Participants: There were 150 participants across the three studies. The mean age of participants across the three studies ranged from 65 to 83 years, and 50% were male. The reasons for mechanical ventilation included

respiratory, surgical, cardiovascular, other medical, trauma, sepsis, and decreased level of consciousness. One study ( Cader et al 2010) excluded patients who were tracheostomised, one study ( Martin et al 2011) included only tracheostomised patients, and it is unknown whether participants in the other study were ventilated via tracheostomy or endotracheal tube. APACHE II scores ranging from 20 to 24 were reported in two of the studies ( Caruso et al 2005, Cader et al 2010) and SAPS II score was reported in one study ( Martin et al 2011). In all three studies, the mean duration of ventilation before inspiratory muscle training commenced was reported and varied greatly between 1 ( Caruso et al 2005) and 45 days ( Martin et al 2011). Prior to initiation of training, the mean maximal inspiratory pressure of the participants, measured at residual volume, ranged from 15 to 51 cmH2O among the included studies. No study reported the maximal inspiratory pressures as a percentage of the predicted values.