Injurious lung ventilation increases mortality in ICU, one of the key aims in ECMO therefore is to cease injurious lung ventilation. However various approaches to lung ventilation on ECMO are currently used. The purpose however must remain to minimise ventilator induced lung injury and analysing the likely costs of high levels of sedation, associated immobility and lack of spontaneous breathing over a prolonged period. Principally the ventilator is set appropriate to the condition and the fresh gas flow used to titrate to the desired CO2.
Common ventilation pattern used in the early stages of the VV support
- Relatively high PEEP 10-15 cmH2O
- Low-levels of pressure control ventilation with driving pressures of 10 cmH2O or less
- Low respiratory rate 5-10 /min
- Inspiratory time may be increased beyond 1 second
- FiO2 should be ≤ 0.6 with adequate VV ECMO support
In individual patients an adjustment to the above pattern may be required e.g. lowering the PEEP if a bronchopleural fistula is the main indication for ECMO. In exceptional cases there may be more reliance on either oxygenation or ventilation through the lung. Proning on ECMO may therefore be considered to improve lung function or facilitate lung recovery.
Resolution of radiological changes on CXR and changes in compliance are key factors in judging recovery of lung function. It is useful to document serially measurements.
Note: compliance should only be calculated in paralysed patients.
Once patients enter the recovery phase (increasing compliance >15-20), spontaneous breathing may carefully be allowed. Keeping in mind that beyond breathing effort, PEEP and pressure support add to the generated transpulmonary pressures. Respirator settings therefore need to be reviewed. A conscious decision about the permission of spontaneous should be made.
Controlling respiratory drive
It is crucial to primarily attempt to control respiratory drive aggressively with the sweep gas flow rather than sedation. In the first instance increasing sweep gas flow to pH >7.4 and further to an alkalotic range pH >7.5 should be trialled to minimise respiratory effort. In the occasional patient, controlling respiratory drive by lowering CO2 may not be successful. Careful assessment of the work of breathing, the transpulmonary pressure (may include oesophageal pressure monitoring) is to be undertaken and balanced against the sedation requirement.