Use of simulated inhaled volumes, flow rates and flow rate ramps to evaluate in vitro dry powder inhaler performance

Abstract

Dry Powder Inhalers (DPIs) utilize a patient's inspiratory maneuver to entrain and deaggregate a powder forming an inhalation aerosol. United States Pharmacopoeia (USP) DPI methods replicate the peak rate a patient might achieve, but do not account for how this peak is approached. This research developed a simple, robust system to simulate different rates of rise (ramps) to a final peak inspiratory flow rate. Ramps were programmed to reach 30, 60, 90 and 120 L/min over 100 ms, 500 ms, 1, 2 and 3 s. Using this computerized system, it was demonstrated that peak flow rate, the rate at which that flow rate was achieved and carrier particle size, influenced powder emptying from three passive DPIs. Generally, powder emptying increased as ramp steepness increased---this being most evident at 30 L/min. Greater emptying was usually facilitated by larger carrier particles using all ramps and peak flows. The USP method of instantaneous peak flow rate generation produced more emptying than even the steepest ramp, and may therefore overestimate powder emptying during use. While maximized powder emptying is a prerequisite of optimized dose delivery, a meaningful determination of Fine Particle Fraction (FPF), the fraction of the aerosolized output likely to reach the lungs, is more critical to therapeutic outcome. A semi-automated system was developed permitting ramped air flow through a DPI while maintaining a relatively constant air flow through an inertial impactor. This system suggested that a steeper rate of rise in air flow elicited a higher FPF compared to a shallower ramp at test flow rates of 30 and 60 Umin. FPF determined by the USP method was higher than the ramp method at these flow rates. There were no significant differences in FPF when a 2 and 4 L inhaled volume were compared. This research highlights the limitations of two existing compendial tests when applied to therapeutic aerosols generated from DPIs, and provides a basis and apparatus for evaluating in vitro DPI performance more realistically. It also illustrates the potential dangers in using in vitro tests to infer in vivo performance in the absence of valid correlations.