Relationship of stage mensuration data to the performance of new and used cascade impactors.

Cascade impaction is a standard test method for characterizing the quality of inhalable drug products. The sizes of the nozzles on each stage of the impactor are the critical dimensions for the performance of the impactor. Compendial reference methods call for periodic measurement of the size of the nozzles on each stage, a procedure known as stage mensuration. There is however currently no guidance on acceptable mensuration criteria. We aim to remedy this situation by providing a sound basis for understanding and using mensuration data, be it for acceptance criteria for new impactors or for the setting of mensuration tolerances for in-use impactors. We first show that multi-nozzle impactor stages behave as if all of the nozzles are equal in size to an effective diameter, , that is composed of the area-mean and areamedian diameters, W* and , calculated directly from the individual nozzle diameters for all nozzles on a given stage (equation 1): W= (W*)(2/3) x (W)(1/3) (1). Hence, the effective diameter provides an intuitive and technically sound basis for setting acceptance criteria for new and in-use impactors. We tabulate these criteria for the Mark II eight-stage Andersen cascade impactor and the Next Generation Pharmaceutical Impactor in a manner similar to the tables of critical impactor dimensions published in EP Supplement 5.1 and in USP 28. For two different impactors or for one impactor measured at two different times (e.g., at manufacture and in use), we find that the D50 values of a given stage are related to the effective diameters by D(50,2)/D(50,1)= (W(2)/W(1))(3/2) (2). Using the stage mensuration data for new, as-manufactured NGIs, we compare the D(50 )values of the first 125 as-manufactured NGIs with those of the archivally calibrated NGI. We further establish that the archivally calibrated NGI has D(50) values within 0.3% of an entirely perfect, hypothetical NGI with all nozzles equal to the nominal nozzle diameters. We also apply the equations to a specific mensurated impactor to show that a used impactor with some nozzles outside of the original manufacturing specifications can have the same aerodynamic performance as a new impactor.