Biomedical Engineering Reference
In-Depth Information
with stage 3).
SPM
was likewise calculated as the sum of API mass deposited on
each of the stages below the NGI stage at the defi ned
d
50
, in this case from stage 4
to the MOC. It is important to note that the boundary size between
LPM
and
SPM
was varied in the subsequent analysis of the sensitivity of the EDA technique.
In the example illustrated in Fig.
9.15
, the location of the boundary between
LPM
and
SPM
was set close to the
MMAD
at 2.30
μ
m aerodynamic diameter
[i.e.,
LPM =
(
m
1
+ m
2
+ m
3
+ m
4
)] for optimum EDA sensitivity [
1
]. Group deposi-
tion,
MMAD
, and
LPM
/
SPM
ratio at different values of
d
50
calculated from the
component-by-component drug mass deposition (studies 1-3) are summarized in
Table
9.4
and the corresponding correlations presented in Fig.
9.16
.
Mao et al
.
[
11
] found that the correlations (
R
2
-
values) between
LPM/SPM
and
API mass deposition in stage groups 2 and 4 were better compared with the correla-
tion between
LPM/SPM
and API mass in group 3, regardless of the boundary size
selected for the calculation of
LPM/SPM
(Table 9.5). Such an outcome was fully
consistent with predictive analyses [
12
,
13
] and refl ects the introduction of con-
founding associated with the movement of API mass from either group 2 or 4 into
this neighboring grouping. Thus, API mass transfer from either direction (group 2
to 3 or group 4 to 3) is exactly counterbalanced by the change in mass in group 3.
This case study demonstrates the high degree of sensitivity of the
LPM/SPM
ratio
to small variations in MDI-generated aerosol APSDs infl uencing the location of the
MMAD
values.
LPM/SPM
can be readily calculated from CI stage deposition data with
no more diffi culty than is the case in determining API mass-per-stage groupings.
In another series of case studies, Strickland examined real stability data from two
older products no longer in the market in which probable causes for observed mul-
tistage CI-measured APSD shifts had been identifi ed [
14
]. The data from multiple
APSD assessments of product A by ACI over a 24-month stability trial period were
initially analyzed in terms of stage groupings applying to the following components
of the CI system, as would be the case in the US regulatory environment:
Σ
Group 1: non-sizing components including stage 0
Group 2: coarse mass fraction captured on stages 1 and 2
Group 3: fi ne mass faction captured on stages 3-5
Group 4: extra-fi ne mass fraction captured on stages 6, 7, and the backup fi lter.
When the data expressed as mass % of label claim (LC) are evaluated for the
groupings involved with the APSD measurement (groups 2-4), it is evident that no
trend was obtained with the group 2 data, whereas both groups 3 and 4 demon-
strated a decrease in API mass as a function of duration in the trial (Fig.
9.17
).
However, it is important to note that the group 4 stability change would not have
been identifi ed as a specifi cation risk as the specifi cation for this grouping is typically
a one-sided upper limit. This signifi cant stability trend was also detected when the
same data were analyzed in terms of FPM, as would be the case in a European
regulatory environment (Fig.
9.18
).
Search WWH ::
Custom Search