Issue link: https://www.e-digitaleditions.com/i/1032386
18 OctOber 2018 Inhalation deviation (RSD) of less than 3% and a recovery above 95%. Before performing the impaction analysis, the blends were stored for a minimum of two weeks in open plastic bottles at 45% relative humidity (RH) and room temperature (RT, 20-23°C). Investigation of the aerodynamic particle size distribu- tion was performed with the Next Generation Impac- tor (NGI ® ) (Copley Scientific, Nottingham, United Kingdom), according to the European Pharmacopoeia 9.0) utilizing two commercially available inhaler devices (one reservoir-based device (Novolizer) and one capsule-based device (Cyclohaler ® )). Flow rates were adjusted to ensure a 4 kPa pressure drop over the devices, according to the Ph. Eur. (the flow rate for the 4 kPa pressure drop of the Novolizer was 78.3 L/min and for the Cyclohaler was 100 L/min). BUD and SBS content were quantified by RP-HPLC. Data were eval- uated with CITDAS 3.0 software (Copley Scientific, Nottingham, United Kingdom). e FPF below 5 µm (of emitted dose) was calculated from the resulting aero- dynamic particle size distribution. All impaction tests were done in triplicate and measured at constant condi- tions (21°C and 45% RH). In the second part of the study, two more blends, consist- ing of a new batch of Parteck M DPI and BUD or SBS, respectively, were prepared. e mixing process was car- ried out as described for the first adhesive mixtures in the Turbula tumble blender. e two powder blends were placed in open plastic bottles at room temperature in three different desiccators with 0%, 45% and 75% RH (Figure 1). e aerodynamic behavior of the ordered mixtures was tested directly after they were blended, as well as after 1 week and 1, 2, 3, 6 and 14 months under each storage condition, by using the Novolizer. Geometric particle size distribution of the APIs and the carrier was measured by laser light diffraction (HELOS, Sympatec GmbH, Clausthal-Zellerfeld, Germany) using an R1 lens for the APIs and an R5 lens for the carrier. The powder was dispersed at 3.0 bar b y t h e RO D O S s y s t e m ( S y m p a t e c G m b H , Clausthal-Zellerfeld, Germany). e data was evalu- butamol sulphate (SBS) as a hydrophilic example—was examined with respect to aerodynamic performance during storage. Material and methods is study was carried out with Parteck M DPI (Merck KGaA, Darmstadt, Germany) as a mannitol carrier material. Micronized budesonide (BUD, mean particle size: 1.45 ± 0.03 µm; Farmabios SpA, Cropello Cairoli, Italy) and micronized salbutamol sulphate (SBS, mean particle size: 1.61 ± 0.03 µm; Lusochimica SpA, Peschi- era Borromoo, Italy) were selected as model APIs. To reach a typical market product dose of 200 µg (Novo- pulmon ® 200 µg Novolizer ® , MEDA Pharma GmbH & Co. KG or (Cyclocaps ® 200 µg Salbutamol, PB Pharma GmbH), the API content was set to 2 wt% for all blends. In the first part of the study, two blends were prepared with a Turbula low shear tumble blender (Type T2C, Willy A. Bachhofen AG Maschinenfabrik, Muttenz, Switzerland) and two blends were prepared with a Pico- mix high shear mixer (Hosokawa Alpine, Augsburg, Germany). BUD or SBS, respectively, were weighed into the mixing vessel using a double-weighing method. For both blender vessels, a filling volume of 54% was achieved. Each preparation in the Turbula tumble blender was blended three times and each blend in the high shear mixer was blended once. e rotation speed for the low shear tumble blender was 42 rpm with a blending time of 5 minutes compared to the high shear mixer which used 500 rpm and 30 seconds of blending time. To destroy potential blend-agglomerates, a sieving step (with a 710 µm sieve) was introduced. e blends were sieved after every 5 minutes for the preparations in the low shear tumble blender and after 30 seconds for preparations in the high shear mixer. Ten samples (8-12 mg) of each powder mixture were randomly selected for homogeneity testing and API content was analyzed by reverse phase high perfor- mance liquid chromatography (RP-HPLC). Blends were judged to be homogeneous at a relative standard Figure 1 Schematic drawing of the storage stability method 2% BUD 2% SBS RT 0% RH RT 45% RH RT 75% RH