Particulate contamination in parenterals: current issues.

Particulate matter may be described as insoluble material(s) inevitably present in injectable solutions. Processing conditions for parenteral solutions are designed to substantially minimize the absolute amounts of particolate present down to around 0.1 micron diameter. It is technically relatively easy and, therefore, inexpensive to remove most particulate above 100 microns in diameter. Currently it is extremely difficult, and correspondingly expensive, to remove material in the 1 micron and submicron ranges. Pharmaceutical injectable products, made available at a reasonable cost, must inevitably contain some particulate over a range of sizes, from the molecular level to the visible. An unfiltered solution could contain, at random, various identities (with differing morphologies) of particulate. Filtration affects the size, and numbers, of particulate present but, generally, not the randomness of the identities. An exception would be in situations where clearly definable particle species such as bacteria or intact starch grains are totally removed during the filtration process. A parenteral solution may legitimately contain low levels of random numbers of random species of particulate without necessarily being "contaminated" and, therefore, unacceptable for use. However, it could be argued that a solution containing a dominant particulate species is contaminated by that species. The particle size distribution of a random number/random identity system is characteristic and can be separated from a contaminating species situation. Simple statistical methodology for making this critical distinction will be reviewed. National compendial limits for particulate in injectable fluids are compared, and attention is drawn to the scientific background underlying limits published in the USP.