0710_GF_ONdrugDelivery Pulmonary April 2012.indd

After the Montreal Protocol banned the use of chlorofluorocarbon (CFC) propellants in metereddose inhalers (MDIs), the pharmaceutical industry now has provided suitable alternatives, one of which is the dry-powder inhaler (DPI). A DPI is a device in which small doses of drug powder are delivered to the lung. The inhaled dose consists of drug particles attached to carrier particles. The drug particles are typically smaller than 6 μm in diameter and the carrier particles are 20-200 μm in diameter. Carrier particles make the powder “flowable” and manageable inside the inhaler and prevent the powder from sticking. However, carrier particles have too great an inertia and tend to impact on the back of the throat rather than go down into the lung. Therefore, drug particles must de-agglomerate (break away) from their carrier particle before inhalation. The de-agglomeration step is one of the most challenging aspects of DPI design. Powder engineering provides many opportunities to tune the physical properties of individual particles, example manufacturing processes include spray drying, mechanofusion and processing in supercritical media. However, the large number of parameters governing the interaction of the particles with the device (and with each other) makes the product development task very complex. Simulation tools provide the unique ability to predict the effects of individual forces on the delivery and to separate them from the data scatter inevitably connected with real-world powder experiments. Computational fluid dynamics (CFD) is the most common simulation tool used for DPI development because it can model the flow of air and suspended particles through the device and into patient anatomy or other geometry. To give a brief introduction to the CFD process, the domain of interest is broken down into smaller control volumes, called cells, collectively termed a grid (or mesh). Specifying the material properties of the fluid and boundary conditions (for example, inlet velocity) on external boundaries of the domain completes the problem description. The fluid flow equations are then solved iteratively to calculate the flow patterns. Additional equations can be solved to include the effects of turbulence, species transport, reactions and heat transfer, amongst others. The motion of a discrete (particle) phase can also be calculated; particles may be passive tracers that follow the flow or there can be two-way coupling between the fluid-flow and particulate phase. The rest of this article describes two situations where CFD is currently being used in the development of DPIs. The first section summarises a flow analysis of the Twincer DPI. This device was originally designed to deliver colistimethate sodium to cystic fibrosis patients, replacing nebulizer therapy with a faster and more effective delivery method. Building on its initial success, the Twincer is currently being redesigned to deliver many more drugs, such as antibiotics for tuberculosis therapy. The second half of the article summarises our recent development efforts in the area of particle modelling, with a focus on enhancing our ability to predict de-agglomeration more accurately. Computational fluid dynamics (CFD) is a simulation tool used for modelling powder flow through inhalers to allow optimisation both of device design and drug powder. Here, Ralf Kröger, Consulting Senior CFD Engineer, ANSYS Germany GmbH; Marc Horner, Lead Technical Services Engineer, Healthcare, ANSYS, Inc; Robert Woolhouse, Senior CFD Engineer, ANSYS UK, Ltd; Michael Becker, PhD Student, and Herbert Wachtel, Senior Principal Scientist, both of Boehringer Ingelheim Pharma GmbH & Co KG; and Anne De Boer, Research Leader Inhalation, University of Groningen, Groningen, The Netherlands, describe in detail how CFD was used in the optimisation of the TwincerTM dry-powder inhaler, extensions of current particle modelling physics to account for particle-particle interactions, and the benefits of this approach in product development. COMPUTATIONAL MODELLING FOR DRY-POWDER INHALERS