HYPOTHESIS: Albuterol delivered during noninvasive positive-pressure ventilation is affected by use of a nebulizer or metered-dose inhaler (MDI) and whether the leak port is in the hose or the mask. METHODS: A lung model that simulated spontaneous breathing at 20 breaths/min was used. A bi-level positive-airway-pressure ventilator (Respironics S/T30) was set for an inspiratory positive airway pressure of 15 cm H(2)O and an expiratory positive airway pressure of 5 cm H(2)O. The tidal volume delivered was 0.4 L. Two mask types were used: one in which the leak port was incorporated into the circuit, and another in which the leak port was incorporated into the mask. The nebulizer was filled with 4 mL, which contained 5 mg of albuterol, connected via a T-piece directly to the mask, and operated at 8 L/min for 15 min. For the MDI studies, a spacer was placed between the mask and the circuit, and an MDI was actuated into the spacer, either synchronized with the initiation of inhalation or during the exhalation phase (4 actuations separated by \G 15 s in each case). Albuterol was washed from the filter and measured with ultraviolet spectrophotometry. RESULTS: With the nebulizer, significantly more albuterol was delivered to the filter when the leak port was in the circuit (p = 0.001). Significantly more albuterol was delivered with the nebulizer than with the MDI (p < 0.001). The efficiency of albuterol delivery (percent delivered) was similar for nebulizer and MDI with the leak port in the circuit (p = 0.57), but better with the MDI with the leak port in the mask (p = 0.001). Albuterol delivery was significantly less when the MDI was actuated during exhalation (p = 0.001). CONCLUSIONS: Albuterol delivery with noninvasive positive-pressure ventilation was affected by the type of aerosol delivery device, by the location of the leak port, and by actuating the MDI at the proper time in the respiratory cycle.