Dry powder inhalers (DPIs) are used for delivering drugs to the airways. In addition to the drug, the formulations often contain a coarse carrier, most commonly alpha lactose monohydrate. The presence of fine lactose particles in the formulation is known to improve the formulation performance. The active site, drug-fines agglomeration and increased cohesion theories have been suggested to explain improved DPI performance upon addition of fine excipient particles. This project aimed to investigate the validity of those theories. The viability of the active sites theory in explaining the improved DPI performance was investigated by studying the impact of loaded drug dose on the in vitro performance for formulation series prepared with coarse carriers with different surface characteristics. The formulations prepared with the rougher lactose carrier were seen to outperform the formulations prepared with the smoother carrier at all drug concentrations. These findings were concluded to be non-compatible with the active sites theory. The impact of addition of lactose fines with different size distributions on powder flow and fluidisation properties and in vitro performance was studied. Powder cohesion increased independent of size distribution of the fines, but did not necessarily correspond to improved performance. Therefore, the increased cohesion theory was concluded not to be the sole explanation for the improvement in DPI performance in the presence of lactose fines. Instead, the increase in performance could be preliminarily attributed to the formation of agglomerated systems. The formation and co-deposition of drug-fines agglomerates, and consequential improvement in the DPI performance was proved using morphologically directed Raman spectroscopy. The project also aimed to develop a universal model for predicting DPI performance based on the lactose properties for a wide range of carriers with different properties. No simple linear correlations between any the lactose properties and the final DPI performance were found. Therefore no single parameter can be used as a universal predictor for DPI performance. To establish more complex relationships, artificial neural networks were used for modelling the importance of different lactose properties in determining DPI performance. The proportion of fine lactose particles (<4.5 μm) was identified as the most important parameter. However, this parameter was capable of explaining only approximately half of the variation seen in the formulation performance. The current study showed that to obtain more accurate predictions for the purposes of quality-by-design approach, also other lactose properties need to be characterised.
Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:564006 |
Date | January 2012 |
Creators | Kinnunen, Hanne |
Contributors | Price, Robert |
Publisher | University of Bath |
Source Sets | Ethos UK |
Detected Language | English |
Type | Electronic Thesis or Dissertation |
Page generated in 0.0107 seconds