Cerebro-pulmonary interactions during the application of low levels of positive end-expiratory pressure

OBJECTIVE: In patients with severe brain injury and acute lung injury the use of positive end-expiratory pressure (PEEP) is limited by conflicting results on its effect on intracranial pressure. We hypothesised that the occurrence of alveolar hyperinflation during the application of PEEP would lead to an increase in PaCO(2) responsible for a rise in intracranial pressure. DESIGN: Prospective interventional study. SETTING: Intensive Care Unit of University Hospitals. PATIENTS AND PARTICIPANTS: Twelve severely brain-injured patients with acute lung injury and intracranial pressure higher than applied PEEP. INTERVENTIONS: 5 and 10 cmH(2)O of PEEP was randomly applied. MEASUREMENTS AND RESULTS: In all patients intracranial pressure, flow velocity by transcranial Doppler of middle cerebral artery, and jugular oxygen saturation were recorded. Static volume-pressure curves of the respiratory system were obtained, recruited volume and elastance calculated to classify patients as recruiters and non-recruiters. In recruiters (= 6 patients), elastance decreased (P<0.01) and PaO(2) increased (P<0.005), while in non-recruiters (= 6 patients) elastance and PaCO(2) significantly increased (P<0.001). Intracranial pressure, Doppler flow velocity, and jugular saturation remained constant in recruiters but significantly increased (P<0.0001) in non-recruiters. A significant correlation was found between changes in intracranial pressure and elastance (r(2) = 0.8 P<0.0001) and between changes in PaCO(2) and intracranial pressure (P<0.001, r(2) = 0.4) and elastance (P<0.001, r(2) = 0.4), respectively. CONCLUSIONS: When PEEP induced alveolar hyperinflation leading to a significant increase in PaCO(2), intracranial pressure significantly increased, whereas when PEEP caused alveolar recruitment intracranial pressure did not change.