Verification of the EDA Concept Through an Assessment of Theoretical Failure Modes, Failure Mode Analysis, and Case Studies with Real Data - Good Cascade Impactor Practices, AIM and EDA for Orally Inhaled Products - page 255

fall in SPM , making the effects detectible by a shift in LPM / SPM and also by a
decrease in ISM . Likewise, turbulent de-agglomeration of the aerosol particles
would result in a shift to fi ner sizes, detectible as a decrease in LPM / SPM .
The introduction of increased numbers of larger particles by changes to the way
they are bound to the carrier particles of certain DPIs might conceivably result in a
shift within both LPF and SPF components of the APSD. However, such a change
would be detected by an increase in LPM / SPM . In the extremely unlikely case
where the exact mass transfer to SPF exactly balances the incoming mass to the
LPF component, SPM and therefore ISM would have to increase, again making the
change detectible by EDA.

9.3.4 Summary

The theoretical foundation upon which the metrics associated with EDA are con-
structed appears to be robust, given the outcomes from both the case studies with
real OIPs and from a consideration of aerosol mechanics associated with plausible
scenarios (however remote) where this approach might fail to detect changes in
measured aerosol APSDs. The practical examples that follow in Sect. 9.5 , in which
case studies are examined, are based on full-resolution cascade impactor experi-
ence, because AIM-based measurements are not yet in routine use for product
qualifi cation.
If abbreviated systems are adopted, the underlying physical processes described
in the scenarios for potential EDA failure will not be affected. However, it would be
prudent to conduct an appraisal of the effect of reducing internal wall surfaces for
potential particle loss by deposition in the abbreviated impactor as part of the
method development process, whichever AIM system is chosen. The next chapter
describes the large amount of validation work that has already been undertaken in
support of a variety of AIM-based apparatuses that are available commercially, or
which have been constructed from existing full-resolution CI components.

9.4 Failure Mode Analysis

Glaab et al. have looked at the robustness of the EDA concept by undertaking a
failure mode analysis approach [ 9 ]. They fi rst considered the factors that could con-
tribute to changes in the CI-measured APSD associated with (a) the formulation, (b)
the device, and (c) the raw material of a hypothetical DPI (Fig. 9.6 ).
They then went on to look at causes arising from the manufacturing process
(Fig. 9.7 ).
Looking further down the chain of events leading to the fi nal OIP, they then
focused their attention on potential causes of APSD changes attributable to the ana-
lytics for the API(s) and the measurement process itself (Fig. 9.8 ).