• Air blood gases at the end of each study period (one with HME, one with HH)
  

• Respiratory parameters including volumes, pressures, P0.1, a comfort score, capnometry and oxygen saturation at the end of each study period (one with HME, one with HH)
  

Objective: Few data are reported in the field of noninvasive ventilation (NIV) regarding humidification devices. It was previously suggested that standard heat and moisture exchangers (HME) had adverse gasometrical and clinical outcomes in acute respiratory failure (ARF) of chronic respiratory failure patients (CRF). This study was performed to evaluate respiratory parameters and arterial blood gases (ABG) of patients during NIV with small dead space HME compared to heated humidifier (HH).

Design: Prospective randomized cross-over study. Setting: A 16-bed medical intensive care unit (ICU) and a 14-bed medico surgical ICU.

Patients: Patients receiving NIV for ARF in the context of CRF or hypoxic ARF and patients with persistent weaning failure receiving NIV just after extubation.

Measurements: HME and HH were randomly compared during 2 NIV periods of 30 minutes separated by a 20 to 30 minutes period of spontaneous breathing with oxygen. ABG were collected at baseline and at the end of each period. As well as respiratory parameters including volumes, pressures, P0.1, a comfort score, capnometry and oxygen saturation. Two sets of patients were successively studied whether they have a flex tube added to the ventilatory circuit or not.