Tidal Volume Lowering by Instrumental Dead Space Reduction in Brain-Injured ARDS Patients: Effects on Respiratory Mechanics, Gas Exchange, and Cerebral Hemodynamics

BACKGROUND: Limiting tidal volume (V(T)), plateau pressure, and driving pressure is essential during the acute respiratory distress syndrome (ARDS), but may be challenging when brain injury coexists due to the risk of hypercapnia. Because lowering dead space enhances CO(2) clearance, we conducted a study to determine whether and to what extent replacing heat and moisture exchangers (HME) with heated humidifiers (HH) facilitate safe V(T) lowering in brain-injured patients with ARDS. METHODS: Brain-injured patients (head trauma or spontaneous cerebral hemorrhage with Glasgow Coma Scale at admission < 9) with mild and moderate ARDS received three ventilatory strategies in a sequential order during continuous paralysis: (1) HME with V(T) to obtain a PaCO(2) within 30–35 mmHg (HME1); (2) HH with V(T) titrated to obtain the same PaCO(2) (HH); and (3) HME1 settings resumed (HME2). Arterial blood gases, static and quasi-static respiratory mechanics, alveolar recruitment by multiple pressure–volume curves, intracranial pressure, cerebral perfusion pressure, mean arterial pressure, and mean flow velocity in the middle cerebral artery by transcranial Doppler were recorded. Dead space was measured and partitioned by volumetric capnography. RESULTS: Eighteen brain-injured patients were studied: 7 (39%) had mild and 11 (61%) had moderate ARDS. At inclusion, median [interquartile range] PaO(2)/FiO(2) was 173 [146–213] and median PEEP was 8 cmH(2)O [5–9]. HH allowed to reduce V(T) by 120 ml [95% CI: 98–144], V(T)/kg predicted body weight by 1.8 ml/kg [95% CI: 1.5–2.1], plateau pressure and driving pressure by 3.7 cmH(2)O [2.9–4.3], without affecting PaCO(2), alveolar recruitment, and oxygenation. This was permitted by lower airway (− 84 ml [95% CI: − 79 to − 89]) and total dead space (− 86 ml [95% CI: − 73 to − 98]). Sixteen patients (89%) showed driving pressure equal or lower than 14 cmH(2)O while on HH, as compared to 7 (39%) and 8 (44%) during HME1 and HME2 (p < 0.001). No changes in mean arterial pressure, cerebral perfusion pressure, intracranial pressure, and middle cerebral artery mean flow velocity were documented during HH. CONCLUSION: The dead space reduction provided by HH allows to safely reduce V(T) without modifying PaCO(2) nor cerebral perfusion. This permits to provide a wider proportion of brain-injured ARDS patients with less injurious ventilation.