BACKGROUND: The aim of this study was to investigate the effects of different delivery circuit configurations, nebulizer positions, CPAP levels, and gas flow on the amount of aerosol bronchodilator delivered during simulated spontaneous breathing in an in vitro model. METHODS: A pneumatic lung simulator was connected to 5 different circuits for aerosol delivery, 2 delivering CPAP through a high-flow generator tested at 30, 60, and 90 L/min supplementary flow and 5, 10, and 15 cm H 2 O CPAP and 3 with no CPAP: a T-piece configuration with one extremity closed with a cap, a T-piece configuration without cap and nebulizer positioned proximally, and a T-piece configuration without cap and nebulizer positioned distally. Albuterol was collected with a filter, and the percentage amount delivered was measured by infrared spectrophotometry. RESULTS: Configurations with continuous high-flow CPAP delivered higher percentage amounts of albuterol compared with the configurations without CPAP (9.1 ؎ 6.0% vs 6.2 ؎ 2.8%, P ؍ .03). Among configurations without CPAP, the best performance was obtained with a T-piece with one extremity closed with a cap. In CPAP configurations, the highest delivery (13.8 ؎ 4.4%) was obtained with the nebulizer placed proximal to the lung simulator, independent of flow. CPAP at 15 cm H 2 O resulted in the highest albuterol delivery (P ؍ .02). CONCLUSIONS: Based on our in vitro study, without CPAP, a T-piece with a cap at one extremity maximizes albuterol delivery. During high-flow CPAP, the nebulizer should always be placed proximal to the patient, after the T-piece, using the highest CPAP clinically indicated.