OBJECTIVE. The purpose of this work was to evaluate the effects of administering either heliox or air oxygen in combination with nasal continuous positive airway pressure in infants with refractory bronchiolitis. PATIENT AND METHODS. We conducted a prospective, interventional, single-center, crossover study in a teaching hospital including infants 1 month to 2 years of age, consecutively admitted to the PICU from February 2004 to February 2005 for treatment of severe acute bronchiolitis unresponsive to therapy. Patients with a clinical score (Modified Wood's Clinical Asthma Score) of >5, arterial oxygen saturation of <92%, or transcutaneous CO2 pressure of >50 mmHg despite supportive therapy, nebulized L-epinephrine, and heliox therapy through a nonrebreathing reservoir face mask were eligible. During the study period, 40 infants with bronchiolitis were admitted to the PICU; 12 fulfilled inclusion criteria. A predetermined balanced sequential allocation to either 30 minutes of treatment with nasal continuous positive airway pressure with heliox or to air-oxygen nasal continuous positive airway pressure was performed. Measurements were taken at baseline and after 30 minutes of each treatment. RESULTS. Baseline mean values were as follows: nasal continuous positive airway pressure of 7.2 cmH2O; clinical score of 7.7 points; transcutaneous CO2 pressure of 61.6 mmHg; and arterial oxygen saturation of 88.6%, with the fraction of inspired oxygen at 35.4%. Clinical score, transcutaneous CO2 pressure, and arterial oxygen saturation improved during the study time with both heliox-nasal continuous positive airway pressure and air-oxygen-nasal continuous positive airway pressure: after 1 hour, the clinical score fell 1.7 points, transcutaneous CO2 pressure decreased 8.2 mmHg, and arterial oxygen saturation increased by 7.7%. Improvement in clinical score was double with heliox-nasal continuous positive airway pressure compared with the air-oxygen-nasal continuous positive airway pressure (2.12 vs 1.08 points), and the fall in the transcutaneous CO2 pressure was greater with heliox-nasal continuous positive airway pressure compared with air-oxygen-nasal continuous positive airway pressure (9.7 vs 5.4 mm Hg). There was no difference in arterial oxygen saturation between groups. No patients required endotracheal intubation. No adverse effects attributable to either of the study interventions were detected. CONCLUSIONS. Nasal continuous positive airway pressure improves the clinical score and the CO2 elimination of infants with refractory bronchiolitis. These positive effects are significantly enhanced when nasal continuous positive airway pressure is combined with heliox instead of air oxygen. Both techniques are noninvasive, seem safe, and may reduce the need for endotracheal intubation.