Future Directions for the AIM and EDA Concepts - Good Cascade Impactor Practices, AIM and EDA for Orally Inhaled Products - page 395

that has Petri dish collection surfaces. Currently, there is very little evidence to
quantify the effect of such differences on OIP aerosol APSD measurement.]
Computer simulations could be used to understand the differences between spe-
cifi c [CI- and AIM-based] platforms.
9. As an intermediate step before going to an actual AIM apparatus, could you do
a consolidated extract from a group of stages?
Yes, you could do EDA from the full-resolution CI data, and you could pool the
material from several stages by washing them together. Although, while you
are still in development, you may not know which stages to group yet.
Furthermore, from the time and resource perspective, it would not be much
different than collecting and measuring stages individually, and then adding up
the numbers.
10. The 1998 draft FDA guidance for MDIs and DPIs requires a ±15 % specifi cation
on the mass balance from CI measurements. How will it be done with EDA?
With EDA, besides the LPM / SPM ratio, the impactor-sized mass ( ISM ) needs
to be controlled. We are not discussing in detail the performance of the
ISM portion of the EDA approach because it will be the same as with either a
full-resolution CI or an AIM-based apparatus. The ISM includes much of the
mass balance but not all (e.g., the non-sizable portions [that deposit in the
induction port and pre-separator, if used] are not included). However, the deliv-
ered dose uniformity (DDU) test, which is also required for these products,
does control for the total emitted dose and in a much more accurate way than
current CI measurements with multistage impactors. The IPAC-RS position has
always been that mass balance should be used as a “system suitability” indica-
tion, not as a specifi cation. But your actual regulatory requirement needs to be
discussed with the FDA [or appropriate regulatory agency]. We have also pre-
sented a poster at Drug Delivery to the Lungs 21 [ 5 ] about a dual-use DDU/
APSD apparatus. If such an apparatus is ever developed, it could provide a
single measurement that would replace both the current DDU and CI tests.
11. The air fl ow [rates] recommended for ACIs are 60 and 28.3 L/min. If we want
to match the full-stage ACI at 15 L/min, how can we do that? How would AIM
measure at low fl ow rates?
In response to the above question, a participant (at the symposium referred
to at the beginning of this section) indicated that he would like to see calibra-
tion data fi rst. He acknowledged that the full-resolution ACI can measure at
15 L/min but had not seen the calibration data. [Such data are unavailable to the
best knowledge of the coeditors, although Garmise and Hickey in 2008 [ 6 ]
published calibration data for stages −0, −1, and −2 (see Chap. 2 ) at this fl ow
rate in the context of using the CI for sampling aerosols from nasal inhaled
products.] However, the archival calibration data for the NGI at 15 L/min were
published in 2004 [ 7 ]. This speaker requested that before we go forward with
the ACI, we need reliable [calibration] data. Currently, the [inhaler testing com-
munity is] relying on old data from the manufacturer's operating manual for
this CI. The ACI used to be made of aluminum, but it erodes. Newer ACIs are
made of steel, and we now need new [archival calibration] data.