Diaphragm dysfunction: weaning perspective

Weaning is the process of successfully liberating the patient from mechanical ventilation. The majority of patients will separate from the ventilator after a successful spontaneous breathing trial (SBT). In a minority of patients, weaning can be challenging and prolonged. Finding the cause of weaning difficulty is crucial to minimize the rates of extubation failure and prolonged ventilation. Diaphragm dysfunction (DD) has been described as a separate entity responsible for weaning failure with an incidence of 23–80%. It has also been associated with difficult weaning, prolonged intensive care unit (ICU) stay and mechanical ventilation, and increased ICU and hospital mortality. Sepsis, shock, and ventilator induced diaphragm dysfunction are important risk factors of DD. Diaphragm dysfunction has several mechanisms. Disuse atrophy and microstructural changes of the diaphragm have been described as the two cardinal pathophysiologic features. Establishing the diagnosis of DD can be complex in critically ill patients. Bilateral anterior magnetic phrenic stimulation is widely considered as the gold standard but is only available in large research centers with limited availability. Ultrasonography of the diaphragm is a promising tool given its wide availability, affordability, and non-invasive nature. Ultrasound is operator dependent, however and it does not provide continuous monitoring capabilities. The diaphragm thickening fraction (DTF) can be calculated from measuring the end-expiratory and end-inspiratory diaphragm thickness at the bedside. It correlates well with transdiaphragmatic pressure. Electromyography of the diaphragm may overcome the limitation of ultrasound by offering a continuous assessment of the diaphragmatic electrical activity, but it requires the placement of a specialized nasogastric tube. Management of DD is better approached by implementing a preventive and a curative strategy. From animal studies, allowing for spontaneous breathing on mechanical ventilation may prevent the problem. The degree of the recommended patient effort and ventilator assistance to achieve optimal balance between diaphragmatic loading and unloading are yet to be defined. Monitoring DTF while finding the optimal ventilator support level can be useful in this context. Another modality to prevent DD is diaphragm pacing applied through a transvenous phrenic nerve pacing system. Animal studies in pigs showed that this modality resulted in less diaphragm atrophy when pacing was synchronized with ventilation. There is an ongoing study to assess the role of diaphragm pacing to recondition and strengthen the diaphragm in difficult to wean mechanically ventilated patients (Clinicaltrials.gov NCT03107949). Once diaphragm dysfunction is established, no specific treatments exist at this time. Other causes of weaning failure like cardiac dysfunction have to be excluded and treated. Improving respiratory load and respiratory muscle weakness imbalance is also crucial. While it appears to improve inspiratory muscle strength parameters, inspiratory muscle training has not consistently shown improvements in weaning success. Levosemindan showed some benefit in improving diaphragm contractility and efficiency in healthy volunteers but was later found to increase likelihood of weaning failure in septic patients. Anabolic steroids were not found to be effective in treating diaphragm dysfunction in several studies. More evidence is needed before recommending non-invasive ventilation post-extubation in all DD patients.