Do low tidal volumes harm people with acute brain injury?
Frustratingly underpowered PROLABI trial yields concerning results
Low tidal volume ventilation became standard care for most patients receiving mechanical ventilation, after its benefits were demonstrated in randomized trials in patients with acute respiratory distress syndrome (ARDS). Large observational series and meta-analyses followed, suggesting strong tendencies toward improved outcomes in patients without ARDS treated with low tidal volumes.
“Low” tidal volumes are generally 6 to 8 ml/kg predicted body weight (commonly indexed to height and biological sex) for patients without ARDS, and even lower (e.g., 4 to 6 ml/kg) for patients with ARDS.
Patients with acute brain injuries due to strokes or trauma were generally excluded from trials testing low tidal volume ventilation, because of potential harms from elevated blood carbon dioxide levels or impairment of cerebral blood flow by high positive end-expiratory pressure (PEEP).
Thus there has been very little evidence to guide ventilator management in patients with acute brain injuries.
A recent randomized trial, although significantly underpowered, will likely create some anxiety in clinicians considering a low tidal volume strategy in acutely brain-injured patients.
The PROLABI Trial
Between 2014 and 2018, 190 adults with acute brain injuries from trauma or stroke were randomized to either lung protective ventilation at 6 mL per kilogram of predicted body weight with 8 cm H2O of PEEP, or what they termed conventional ventilation (≥8 mL/kg tidal volumes and PEEP of 4 cm H2O).
After 28 days, patients in the protective ventilation group had worse outcomes (all statistically significant):
28.9% mortality versus 15.1% mortality in the conventional group.
A higher rate of a composite outcome of death, ventilator, dependency, or ARDS, 61.5% versus 45.3%.
At 6 months, neurologic outcomes were also worse in patients who received protective ventilation, with 1.5 times the risk of dying or being in a persistent vegetative state (P = 0.04).
However, the trial was perhaps fatally underpowered, having been targeted for an enrollment of 524 patients based on power calculations, but terminated early due to funding difficulties.
This substantially raises the possibility that the findings are a type I error due to chance.
Unfortunately, this seems to be the only randomized trial testing low tidal volume ventilation in acute brain injury. The only available evidence previously was a series of prospective observational studies before and after the implementation of low tidal volume ventilation in brain-injured patients. That study did not show an obvious harm signal.
The more granular results do not give clinicians much actionable information. There were no obvious contributors to the observed increase in mortality. The average daily PaCO2 was only about 2 mm higher in the protective ventilation group, and the average daily PEEP was only about 2 cm H2O higher. The average daily intracranial pressure was less than 2 mm higher in the protective ventilation group, and patients in both arms were generally maintained at plateau pressures lower than 30 centimeters H2O.
No randomized trials testing low tidal volume ventilation in acutely brain-injured patients are currently in process, according to clinicaltrials.gov. Although a scientific agenda to study the issue has been proposed, there are no guidelines to advise ventilator strategy in these patients.