Sterilizing filtration
 |
Objective of the study : Our manufacturing processes provide a 0.22 µm pre-filtration step immediately before the sterilizing filtration stage. These two operations are performed in controlled atmosphere zone, grade A. |
First, filtration devices are assembled, packaged, autoclaved, introduced in aseptic area and installed on the manufacturing line. To exempt autoclaving constraints and to reduce the number of manipulations in the vicinity of the solution sterilization site, it was envisaged to perform the pre-filtration step in controlled atmosphere zone, grade C, using clean but not sterile equipments. We assessed the microbiological impact of this change of practice on the bioburden before sterilizing filtration.
Equipment description : Our pre-filtration devices are mainly consisting of a crankcase, reusable after cleaning and final rinsing with water for injection, and a new filter cartridge, model MILLIDISK ® , filtering surface ranging from 500 to 2 000 sq. depending on the amount of solution processing.
 |
 |
Cartridges are supplied packaged but non sterile. Their bioburden, considered by the manufacturer, MILLIPORE, is 20 CFU to 7,000 sq. of filtering surface. 
Feasability study: We used the following formula to estimate the number of germs that it would be theoretically admitted that a pre-filtration device release in the filtered solution:
C = [V x Y] / n
It has been established on the principle that the number of admitted germ (C, expressed in colony forming unit) was function of the volume of the filtered solution (V, expressed in ml) in which would be released the contaminating germs.
It also took into account the bioburden acceptance limit before sterilizing filtration (Y). It is usually expressed as CFU / 100 ml and must not exceed 10 CFU / 100 ml.
Finally, if the pre-filtration device was intended to be replaced during manufacturing (e.g.: clogging), the number (n) of actually used pre-filters should be integrated into the formula, each providing its own bioburden.
Note: If the process uses 2 pre-filtration serial devices, only the device located immediately before the sterilizing filter will have to be considered.
In our situation, C value has been estimated at 500 CFU per pre-filtration device, while the contribution of the filter cartridge would be 6 CFU (theoretical bacterial contamination for a filtering surface of 2,000 sq.). We have thus assessed that it was feasible to use non autoclaved pre-filtration devices and have developed a protocol study to demonstrate it.
Methodology: This study was to simulate 3 successive passages of one litre of solution each through the pre-filtration device and to enumerate the contaminating germs on all the volume of the 3 collected fractions. The solution was replaced by water for injection. WFI was also controlled before its passage through the device.
Results and interpretation:
The results obtained on the 3 fractions are equivalent and close to 0 CFU. Thus, it can be assessed that almost all of the contaminants present in the pre-filtration device and able to pass through the filter media has been collected.
On the two contaminating germs found in the 2,900 ml:
- 1 CFU could have been brought by the environment or by the filtration device itself (Bacillus spp)
- 1 CFU could have been brought by the water used for the study or for the preparation of the filtration equipment (S. paucimobilis)
The number of the collected bacteria is clearly under our maximum permissible value (C); it is equivalent to the theoretical number of contaminating bacteria that would be delivered by the filter cartridge.
Conclusion and discussion: Under our specific conditions, the performance of pre-filtration with non sterile equipments would have no significant negative impact on the bioburden before sterilizing filtration. The requirements to arrive at this conclusion are:
- The microbiological cleanness of the filter cartridge, which only depends to its manufacturer,
- The effectiveness of internal procedures for equipment preparation which are strongly related to the microbiological quality of the rinsing water and of the environment
The methodology presented has the advantage of being simple and informative; it could contribute to stopping expensive practices with no significant advantage for the product microbiological safety. It was interesting to discuss with the manufacturer of filter cartridges and to assess the impact of our equipment preparation practices on the bioburden.
|
|
Christian Mroz Schering Plough This email address is being protected from spambots. You need JavaScript enabled to view it. |
