PEEP therapy is necessary to prevent lung injury due to repetitive opening and collapse of alveoli, or atelectrauma. PEEP may improve outcomes, but only in patients with moderate or severe ARDS by the current definition, suggesting a preferential effect based on lung compliance [7, 43]. As mentioned earlier, increases in PEEP have the potential to reduce CBF and raise ICP, although the effects demonstrated in the limited available studies are variable and mild. In a small clinical series of TBI patients with ICP < 20 mm Hg, an increase in PEEP from 0–15 cm H₂O did not increase ICP [34]. Yet another study suggested that effect on ICP is based on alveolar recruitment. Mascia, et al. studied the effects of increased PEEP in 12 severely brain-injured patients. They found that when PEEP resulted in alveolar recruitment, ICP was unchanged; however, when increased PEEP failed to result in alveolar recruitment, ICP was increased. They hypothesized this was due to alveolar hyperinflation and resultant hypercapnia [28]. A similar study found that increasing PEEP from 0 to 12 mm Hg in mixed neurocritical care patients resulted in decreased CPP and MAP in patients with normal lung compliance, but had no effect on these parameters in patients with poor lung compliance [44].

In a small study of nine TBI patients with polytrauma with hypoxia and elevated ICP, incremental increases in PEEP from 0 to 21 mm Hg improved oxygenation (PaO₂/FiO₂ increased from 128 to 245), while ICP only marginally increased (27–29 mm Hg) and CPP remained stable [45]. A study in mixed neurocritical care patients with raised ICP (18 mm Hg), PEEP up to 15 mm Hg did not significantly alter ICP or CPP [33].

High levels of PEEP for short periods of time, or recruitment maneuvers, as a salvage therapy for refractory hypoxemia have been shown to improve oxygenation [46]. However, in one study, the tradeoff for improved oxygenation in TBI patients was a reduction in jugular venous saturation, concerning for cerebral ischemia, despite an improvement in PaO₂ [47].

Given the paucity and heterogeneity of studies, firm conclusions regarding the use of PEEP in TBI are difficult. Modest amounts of PEEP (5–10 mm Hg) are unlikely to be associated with detrimental intracranial effects. High PEEP (>10 mm Hg) should be avoided with mild ARDS and patients exhibiting minimal improvement of PaO₂/FiO₂ with increasing PEEP, but can be cautiously applied in patients with severe hypoxemic respiratory failure . As PEEP is increased, ICP and CPP should be closely monitored. Decreases in MAP secondary to PEEP should be addressed with fluid loading or vasopressors to prevent resultant decreases in CPP. Few studies have assessed the effect of PEEP >15 mm Hg; so extreme caution should be exercised in applying this strategy.

As mentioned previously, hypoxemia is associated with worsened outcomes in the brain-injured patient [22–24]. At a minimum, FiO₂ should be set to avoid hypoxia (PaO₂ < 60 mm Hg). Hyperoxia may also be detrimental, though a clear threshold at which negative effects ensue has yet to be defined [48–50]. Based on the available data, it is reasonable to target a PaO₂ of 80–100 mm Hg in the TBI population to provide some “buffer” against transient insults that may cause hypoxemia while avoiding the possible detrimental effects of hyperoxia as well.

APRV is a pressure-limited, time-cycled mode of ventilation that allows a patient to breathe spontaneously during the application of continuous positive airway pressure [51]. The mode maintains a high pressure for the majority of the respiratory cycle with periodic releases to a low pressure. It allows a very high mean airway pressure to be achieved without high peak pressures which may result in barotrauma. A single case report by Marik and colleagues describes the successful use of this mode in the setting of increased ICP. No significant detrimental effects on ICP or CPP were observed and oxygenation improved significantly [51]. Clarke, et al. reported a similar experience with inverse ratio pressure control ventilation in nine patients with head injury [52]. If inverse ratio ventilation or APRV is utilized in the setting of TBI, it should be done so in the setting of hypoxemic respiratory failure refractory to conventional ventilator settings by a practitioner experienced with the modality. The effects on PCO₂, CPP and ICP should be monitored closely.

Prone positioning recruits collapsed lung regions and has lowered mortality in severe ARDS(PaO₂/FiO₂ < 100) particularly when added to tidal volumes <8 mL/kg IBW [53]. However, proning TBI patients with elevated ICP can raise intrathoracic and intraabdominal pressures, increasing ICP. In an observational study of 29 neurocritical care patients with ICP monitors, prone positioning improved oxygenation at the expense of higher mean ICP and more frequent ICP elevation >20 mm Hg [54]. As with high PEEP, prone positioning has to be undertaken with caution.

In multiple studies, inhaled nitric oxide has been shown to improve oxygenation in ARDS, but without improving outcomes or mortality [55]. Several case reports have described the use of inhaled nitric oxide in the setting of ARDS associated with TBI [56–59]. Some have postulated that nitric oxide may have beneficial effects on both the lung and injured brain in the setting of ARDS in TBI [60]. Further study is required to determine if inhaled nitric oxide truly has a role in the management of TBI complicated by ARDS.