Noninvasive Ventilation with Helmet Reduces Endotracheal Intubation and Mortality Compared to Face Mask in Patients with ARDS

EBM Focus - Volume 11, Issue 25

Read the full EBM Focus and earn CME credit.

Reference - JAMA 2016 Jun 14;315(22):2435 (level 1 [likely reliable] evidence)

  • Although many patients with acute respiratory distress syndrome (ARDS) require some form of mechanical ventilation, there is limited data suggesting the methods of delivery of noninvasive positive pressure ventilation might influence patient outcomes.
  • Compared to a standard face mask, a helmet interface significantly reduced intubation rates and mortality in 83 patients with ARDS requiring noninvasive ventilation for ≥ 8 hours.
  • While oxygen saturation rates were similar during treatment, the helmet group had higher median sustained positive end-expiratory pressure and a lower fraction of inspired oxygen, suggesting less air leakage with the helmet may have increase the efficacy of noninvasive ventilation.

Mechanical ventilation is often required to support adequate gas exchange in patients with ARDS. While some patients can be managed with noninvasive positive pressure ventilation, endotracheal intubation with invasive mechanical ventilation is often necessary when noninvasive strategies fail to provide adequate oxygenation (Crit Care 2013 Nov 11;17(6):R269). However, endotracheal intubation of critically ill patients is associated with a high rate of complications (Crit Care 2015 Jun 17;19:258). Limited data suggests that the mode of oxygen delivery and the airway pressure may influence the outcomes of noninvasive ventilation in patients with acute hypoxemic respiratory failure (Respir Care 2004 Mar;49(3):270, N Engl J Med 2015 Jun 4;372(23):2185). To determine if noninvasive ventilation delivered by a helmet interface could reduce air leak and improve intubation rates, 83 patients with ARDS (median age 60 years) requiring noninvasive ventilation via face mask for ≥ 8 hours were randomized to continued noninvasive ventilation via either a helmet or a face mask. A standard protocol was used for titration of noninvasive ventilation in both groups and intubation decisions were made by a clinical intensive care team based on predefined criteria.

This trial planned to enroll 206 patients, but the trial was terminated early after the first planned interim analysis when the predefined stopping rule for efficacy of the helmet interface was reached. Compared to the face mask group during treatment, patients in the helmet group had a higher median sustained positive end-expiratory pressure, lower fraction of inspired oxygen, and greater decrease in respiratory rate. Both groups had similar oxygen saturations. The intubation rate was 18.2% with the helmet vs. 61.5% with the face mask (p< 0.001, NNT 3). The most common reason for intubation was neurological deterioration in the helmet group and respiratory failure in the face mask group. The helmet interface was also associated with reduced in-hospital mortality (27.3% vs. 48.7%, p = 0.04, NNT 5) and 90-day mortality (34.1% vs. 56.4%, = 0.02, NNT 5). While the length of intensive care unit stay was significantly reduced with the helmet interface, there were no significant differences in the length of hospitalization or in the rate of skin ulcerations.

The results of this trial suggest that noninvasive ventilation utilizing a helmet is more effective than ventilation via a face mask at reducing the need for intubation and mortality due to ARDS. The inclusion of patients requiring 8 hours of noninvasive ventilation ensured that the trial evaluated patients at high risk of intubation, as confirmed by the 61.5% intubation rate in the face mask control group. While the interventions could not be blinded, the a priori definition of the intubation criteria helps reduce the risk of bias. The success of the helmet interface is likely related to less air leak allowing for the significantly higher levels of positive end-expiratory pressure observed. Although this trial is fairly small, these results suggest that increased availability and proper use of helmet interfaces could substantially impact the care of patients with ARDS.

For more information see the Noninvasive positive pressure ventilation (NPPV) in adults topic in DynaMed.