Issue link: http://www.e-digitaleditions.com/i/830814
A novel aerodynamic sizing method for pharmaceutical aerosols using image-based analysis of settling velocities Abstract This article discusses a novel method to estimate aerody- namic particle size distributions (APSDs) of pharma- ceutical aerosols through direct measurement of particle settling velocities using image-based analysis and parti- cle tracking techniques. This simple, optical method provides accurate and fast measurements (approxi- mately 1 minute) with few sources of bias due to specific device design choices or operating conditions. A proof-of- concept for the method is demonstrated by measuring APSDs for widely available commercial dry powder inhalers (DPIs), then comparing the results with previ- ously published data from cascade impactors (CIs) and the Aerodynamic Particle Sizer (APS). Introduction Measurements of aerodynamic particle size distribu- tions (APSDs) from orally inhaled products (OIPs) are important for estimating the inhaled drug dose and are a key component for development and vali- dation of drug/inhaler combinations. Currently, the only APSD estimation method approved for regula- tory purposes in both US and European pharma- copeias 1, 2 is based on the collection of particle frac- tions using a cascade impactor (CI) or a liquid impinger, followed by chemical analysis (e.g., high performance liquid chromatography) of drug weight within these fractions. This approach has several advantages including estimation of aerodynamic (rather than geometrical) diameters, chemical speci- ficity that distinguishes drug-containing particles from excipient particles and full sampling of the whole dose rather than a limited sample. Despite such advantages, CI-based analysis remains not only relatively expensive but also very time-consuming. Thus, this method is often impractical for applica- tions where high throughput measurements are important, such as product development and, partic- ularly, extensive stability studies. Furthermore, CI- based results are known to yield pronounced vari- ability across different CI designs. 3, 4 Notably, one anticipates approximately a 30% difference in the measured mass median aerodynamic diameter (MMAD) from a Next Generation Impactor™ (NGI) (TSI, Incorporated, St. Paul, MN, US) com- pared to an Andersen Cascade Impactor (ACI). 4 Such large differences are somewhat surprising in light of the high reproducibility of each single impactor (i.e., generally around 2-3% in MMAD estimation). Although an in-depth discussion detail- ing the sources of variability in OIP characterization is beyond the scope of the present article and has been discussed elsewhere 3, 5 it should be emphasized that CI-based methods are extremely sensitive to small changes in device design (e.g., nozzle diameter) and operation conditions (e.g., flow rate). The search for faster OIP characterization methods has led to the adaptation of several real-time tech- niques for this purpose. 6 However, among the avail- able methods, the Aerodynamic Particle Sizer ® (APS) (TSI Incorporated, St. Paul, MN, US) is the only apparatus that directly measures aerodynamic parti- cle diameters. 7 Despite its advantages, the APS remains a relatively complicated device that requires complex optics, in conjunction with sample dilution, to avoid particle coincidence errors. Other sources of bias, such as droplet distortion and phantom parti- cles, should also be considered. 5 Furthermore, such high-end devices are relatively costly and often remain beyond the reach of early, cash-strapped, startup companies that are in the initial phases of developing new inhalation devices. This article presents a simple, fast (i.e., approximately 1 minute per measurement) and cost-effective method Rami Fishler, PhD Department of Biomedical Engineering, Technion—Israel Institute of Technology Josué Sznitman, Dr Sc Department of Biomedical Engineering, Technion—Israel Institute of Technology Inhalation JUNE 2017 21