The correlation of a bubble-point pressure level with an (LRV) organism retention of 1 × 10⁷ cfu/cm² EFA was established for the B. diminuta organism cultivated in a prescribed manner (31). This relationship need not necessarily be the same for B. diminuta cultivated to a different size or for other organisms having a different size or size distribution. The more recent recognition that certain organisms, B. diminuta not included, may diminish in size after exposure for a period of time to given drug preparations renders uncertain that a filter found "sterilizing" in one application will necessarily be sterilizing in another (23). No pore-size rating in itself can ensure a sterile effluent. Ralstonia pickettii and Berkholderia cepacia in particular have been studied for their size decrease after undergoing exposure to certain drug preparations.

The point being made is that the filter manufacturer's bubble point was established within a protocol of specified operational steps. Deviations from these may impugn conclusions not so specifically arrived at. The testing is usually outsourced to testing services, intending bias-free results. Correlating the bubble-point values of the various pore-sized filters with the LRVs required for the various filter types using organisms of interest is not a standard protocol. In any case, meeting the manufacturer's bubble point does not exempt the filtration process from the requirement of being validated.

To repeat, the pore-size rating is not assigned on the basis of an actual pore size measurement but according to bubble-point values that have been found to be characteristic of membranes suitable for sterilizing applications. That this identification is warranted is confirmed by a follow-up microbiological assay demonstrating that the membrane does indeed withstand the classic 1 × 10⁷ cfu/cm² B. diminuta confrontation.

An appropriate bacterial challenge is performed to determine the membrane's suitability for the "sterilizing" grade identification before bestowing the 0.2/0.22-μm pore-size rating. The testing is used by each of the individual filter manufacturers using a program based on an American Society for Testing and Materials procedure (27). However, there is not an industry-wide standard for the specifics of the organism challenge test. For example, the organism content of the bacterial test suspension is calculated to furnish the target 1 × 10⁷ cfu challenge per square centimeter of filter surface, but the total organism load is dispensed in from 2 to 20 L of (usually) saline lactose broth. Interestingly, there is reason to believe that the concentration of the organism suspension may influence the sterility of the effluent (1, 44, 45).

All the 0.2/0.22-μm rated membranes of commerce do meet FDA's definition of "sterilizing" membrane, but all do not retain B. diminuta organisms to the same extent. The relative influences of such parameters as bacterial density, volume of test solution, test duration, applied pressure differential, and so forth, have largely not been investigated. Therefore, the distinctions found among different membranes are of an unknown significance. They may reflect the variations in the nonstandard testing.

Consequently, the filters qualified by their fabricators as being suitable for sterilizing trials are labeled and purveyed either as 0.2 or 0.22-μm rated, depending upon their manufacturer. This "pore size" label is affixed to the membranes that exhibit the bubble point, identifying it as a "sterilizing filter." The label signifies that the membrane type did meet FDA's definition of sterilizing filter in the B. diminuta microbial challenge test performed by the filter manufacturer and that it, therefore, has the potential to perform similarly in other filtrations. Whether it would in fact do so would require validation of the filtration process wherein it was used.