In general, training can be instituted to enhance muscle endurance or strength. These types of training require different strategies and have distinct physiological responses. In healthy subjects, respiratory muscle activity is characterized by development of low pressure during the entire life span of a subject. The pressure generated by the inspiratory muscles is only ±5 cmH2O (5 % of maximum inspiratory pressure) to generate a tidal volume of 500 ml, when respiratory compliance is 100 ml/cmH2O. At first sight, training of the respiratory muscle should, therefore, be designed to improve endurance. Indeed, in patients who are difficult to wean from the ventilator, progressive weaning trials (T-tube or low pressure support) are frequently instituted as training stimulus. Although reasonable from a physiological perspective, it has never been proven that this strategy indeed improves respiratory muscle endurance.
In very few circumstances is high inspiratory pressure required for prolonged periods of time and therefore strength training of the diaphragm seems of limited relevance. It has been demonstrated, however, that respiratory effort sensation depends on the maximal inspiratory pressure [
42]. In healthy subjects, pharmacological induction of inspiratory muscle weakness increases respiratory effort sensation for the same workload [
42], confirming the importance of adequate strength beyond that strictly required to generate tidal volume. It has been shown that inspiratory muscle strength training (IMST) improves whole body exercise performance, in particular in less fit subjects [
43]. Also, in patients with chronic obstructive pulmonary disease (COPD), IMST improves inspiratory muscle strength and total body exercise and reduces dyspnea sensation [
44]. Only three randomized studies have reported the effectiveness of IMST in ventilated ICU patients. In the trial by Cader et al. [
45], 41 ventilated ICU patients with respiratory muscle weakness were randomized between inspiratory threshold loading and no training intervention. Training consisted of an inspiratory load of 30 % maximum inspiratory pressure for 5 minutes, twice a day, 7 days a week throughout the weaning period. Maximum inspiratory pressure was significantly increased in the training group (15 ± 3 to 25 ± 4 cmH
2O) but not in the control group (15 ± 2 to 18 ± 2 cmH
2O). The study was underpowered for clinically relevant endpoints, although weaning time was reduced in the training group. In another study, Martin and colleagues [
46] randomized 69 ventilator-bound patients (mean duration of ventilation at inclusion ±44 days) to IMST or sham training added to endurance training. Strength training consisted of four sets of 6–10 breaths per day with 2 minutes rest between each set. Loading was individualized and set at a level that could just be tolerated by the patient. IMST significantly improved maximum inspiratory pressure (end of training 54 cmH
2O in intervention group and 45 cmH
2O in sham group,
P < 0.05). Also, successful weaning at day 28 after inclusion was more likely in the intervention group compared with sham (71 versus 47 %,
P < 0.05). No adverse events of IMST were reported in these two trials. Finally, Condessa et al. [
47] randomized ventilated patients to inspiratory strength training on top of usual care versus usual care only. Each training session consisted of five sets of loaded breaths (40 % maximum inspiratory pressure), twice a day, 7 days a week. In this study, IMST significantly increased maximum inspiratory pressure but did not affect weaning time.
In conclusion, inspiratory muscle training is feasible and appears safe in patients with respiratory muscle weakness who are difficult to wean from the ventilator, which is supported by a recent systematic review [
48]. Studies in other patient categories, including COPD, indicate that IMST improves outcome. In our opinion, it is reasonable to add IMST to endurance training in stable, difficult-to-wean patients with confirmed respiratory muscle weakness. Future studies are needed to determine the optimal training protocol and appropriate timing for initiation of IMST.