Inspiratory muscle training in critically ill adults with a tracheostomy: A prospective, single-centre observational study.

Inspiratory muscle training (IMT) could improve outcomes, but implementation in clinical practice is poor, with patients at high risk often being excluded in trials. The aim of this study was to assess feasibility and safety of implementing standardised IMT in critically ill adults with a tracheostomy. A prospective, single-centre observational study was performed in a tertiary, mixed intensive care unit (ICU) using a standardised IMT protocol with a threshold device. IMT was performed with two sets of six repetitions at ≥30% of maximal inspiratory pressure (MIP) or ≥9 cmH2O. Primary outcome was feasibility and safety of IMT using a priori defined criteria. Secondary outcomes included a descriptive analysis of functional changes from IMT initiation to ICU discharge. From January to July 2019, 57 patients (age: 62 ± 14 years, 35% female) were eligible for IMT, of whom 31 (54%) were tracheostomised due to weaning failure, 19 (33%) with reduced consciousness and seven (12%) with dysphagia. Fifty-one patients passed safety screening, of whom 42 (82%) completed at least one IMT session. A total of 160 of 203 (79%) sessions were attempted and 152 (75%) performed with a median of three sessions per patient. IMT was terminated in eight (5%) sessions due to a priori defined intolerance reaction: minimal resistance too high (3 [38%]), respiratory rate >35 breaths/min (2 [25%]), patient refusal (1 [13%]), increase of blood pressure (1 [13%]), and agitation (1 [13%]) without further consequences. There were no adverse events. The median difference from IMT initiation to ICU discharge of MIP was 6 cmH2O (95% confidence interval: 3.5 to 10.5). MIP was moderately correlated with the ability to follow commands (r = 0.395). IMT was feasible and safe in critically ill patients with a tracheostomy, including those who could not follow commands. Although MIP improved during the ICU stay, no causal relationship to IMT can be inferred. Future randomised controlled trials should be performed to explore outcomes in this under-researched population.