Good Cascade Impactor Practices - Good Cascade Impactor Practices, AIM and EDA for Orally Inhaled Products - page 104

Fig. 4.11 Cumulative mass-weighted APSDs considering emitted mass of API entering the size-
fractionating stages of an ACI having an upper size limit in relation to the mass entering the induc-
tion port of this measurement system ( From [ 78 ]— used by permission )

4.4.2 Data Reduction

4.4.2.1 Initial Inspection of the APSD

The APSD by itself, though useful in understanding the behavior of the whole popu-
lation of the aerosol particles that are size-fractionated within the operating limits of
the CI, is awkward as a means of representing OIP performance parameters that are
meaningful both in the product QC and clinical environments.
This difficulty arises because the true APSD is a continuous variable function.
However, data coming from a given CI-based measurement discretizes the APSD
into a multivariable form with increasing complexity as the number of size-
fractionating stages in the CI system as a whole becomes greater. Thus, treating the
APSD as a collection of seven or more independent measurements increases
the overall error in decision-making due to the effect of multiplicity. Furthermore,
the time and resources required to arrive at a decision concerning OIP quality from
such data become more complex than it need be. It therefore becomes necessary to
adopt a data reduction strategy that is appropriate for the application for which the
CI data are required. This approach is at the heart of EDA and will be discussed in
depth in later chapters. However, at this stage when considering good data analysis
practices, it is a good idea to check for the following when inspecting data obtained
from the size-fractionating stages of the CI:

1. More than one mode may be present in the APSD, as processes such as atomiza-
tion of liquids often produce bimodal distributions [ 81 ] as the result of Rayleigh
breakup of the liquid stream via ligament formation into primary and satellite