The majority of the studies that have reported the effects of NMBA-induced paralysis were conducted in very short periods of time (<24 h) during anaesthesia (i.e. in “normal lungs”) with muscle paralysis and mechanical ventilation without PEEP [19–21]. Tokics and colleagues demonstrated that patients displayed atelectasis. Shunt was placed in gravity-dependent lung regions that corresponded to the atelectatic areas. There were considerable V˙/Q˙ ventilation-to-perfusion mismatches that resulted in ventilation primarily of the ventral lung regions and perfusion of the dorsal lung regions [20]. Hedenstierna and colleagues demonstrated that due to the reduction of the cross-sectional chest area and the cephalic ascension of the diaphragm, the thoracic volume was reduced. This reduction was accompanied by a decrease in the functional residual capacity (FRC) that was attributed (at least in part) to the loss of muscular tone [21, 22] and was thought to be a major cause of post-operative hypoxaemia. However, few studies have reported the effects of anaesthesia and NMBA-induced paralysis on lung mechanics. In healthy subjects, Brismar and colleagues found no further reduction in FRC or modification of the elastic properties of the lung or the chest wall following the addition of paralysis to anaesthesia [23]. In the intensive care setting, Conti and colleagues conducted a study of 13 patients affected by diseases involving both lungs and the chest wall who required mechanical ventilation for acute respiratory failure and were heavily sedated (Ramsay scores of 5) and failed to demonstrate any further modification of either the chest wall or lung elastance following paralysis. CT scans and analysis of the pressure-volume curves revealed that derecruitment was reduced after the application of a PEEP of 10 cmH₂O [21, 24]. In summary, no data support any additional deleterious role of NMBAs in the onset of atelectasis in sedated ARDS patients.

The results of the three available RCTs about the use of NMBAs during the early phase of ARDS indicate that the effect of the treatment on oxygenation becomes significant after 24 h. Moreover, in the ACURASYS trial, the Kaplan-Meier survival curves did not separate until 18 days of treatment. These observations together with the available knowledge regarding the use of NMBAs for shorter periods in patients with healthy lungs raise hypotheses regarding the manners in which NMBAs could be beneficial (Fig. 12.2) during the acute phase of severe ARDS. Several mechanisms may be involved and are most likely interrelated. Slutsky [25] proposed a summary of the effects of NMBAs that included the following main points:

The beneficial effects of NMBAs in the most hypoxaemic patients in the initial phase of ARDS may also be due to the prevention of the deleterious role of spontaneous breathing (SB). In the early phase of ARDS, the high respiratory drive that results from hypoxaemia may result in elevated and uncontrolled transpulmonary pressure (TPP) with over-distension, particularly in the dependant vertebro-diaphragmatic zones, which results in major ventilator-induced lung injury (VILI) [30]. In these cases, the maintenance of a plateau pressure below 30 cmH₂O might not be sufficient to ensure protective ventilation, particularly in patients with the most severe forms of ARDS, as suggested by recent experimental work from Yoshida et al. [31]. In a rabbit model of ARDS, these authors demonstrated that the preservation of ventilatory effort lacked the same effects on oxygenation, pulmonary ventilation and lung injury depending on the severity of the ARDS (i.e. mild or severe). The preservation of SB induced improvements in pulmonary ventilation and oxygenation in mild ARDS subjects but resulted in increased TPP and VILI in a group of animals with severe ARDS. These deteriorations of the ventilatory and histological parameters were prevented by abolishing SB via the administration of NMBAs in the severe ARDS group.

In summary and in contrast to what occurs in patients with mild to moderate ARDS, in the initial phase of severe ARDS, the prevention of VILI and the optimization of alveolar recruitment appear to be based on controlled protective ventilation and the use of NMBAs and the consequent abolition of spontaneous ventilatory effort.

NMBAs have been dispraised due to a supposed association with ICUAW.

ICUAW is defined by a generalized muscle weakness that develops during the course of an ICU admission and for which no cause other than the acute illness or its treatment can be identified [32]. ICUAW is present among approximately 60% of ARDS patients at the time of awakening [33] and 36% at the time of hospital discharge [34]. ICUAW is associated with prolonged ICU and hospital stays, prolonged mechanical ventilation duration and increased ICU and hospital mortalities [35–38]. Risk factors have been discussed widely in the literature, and independent risk factors, such as female sex, multiple organ dysfunctions (≥2), the duration of mechanical ventilation, hyperglycaemia [39] and the administration of corticosteroids [36], have been identified.

The prolonged use of neuromuscular blocking agents (more than 48 h) [40, 41], their simultaneous administration with corticosteroids [42] and steroid NMBAs may contribute to ICUAW [43, 44], but there is actually no evidence that non-steroid NMBAs increase the risk of ICUAW when used for short durations and without the simultaneous administration of corticosteroids [32]. In a meta-analysis [17] of RCTs evaluating the effects of the use of NMBAs during ARDS, the incidence of ICU-acquired paresis (assessed via clinical evaluations of quadriparesis or by the Medical Research Council (MRC) score at day 28 or at ICU discharge) was not higher among the treated patients than in the control group.

The continuous use of NMBA infusion highlights the problem of inadequate sedation. Hardin et al. [45] demonstrated that patients receiving NMBAs were awake for 22% of the sleep period over a time span of 24 h. Neuromonitoring via continuous electroencephalography or with a device such as the bispectral index (BIS, Aspect Medical Systems, Natick, MA, USA) can reduce the risk of consciousness in paralysed patients [46]. However, studies evaluating the reliabilities of such devices in the monitoring of the level of sedation of paralysed patients have found contradictory results [47, 48].