Printer Friendly

Filtration in pharmaceutical industry.

Filtration is one of the most commonly used "Technology". Ask any housewife. But when applications become specialised the term filtration acquires new dimensions and the associated peripherals appear in all shapes and sizes.

All applications have their specific goals and to achieve these, technologists have designed suitable filters and peripherals.

Pharmaceutical industry is one of the prime users of filtration and in the present state it can not survive or operate without filtration.

Filtration as a whole covers space to deep blue sea but in pharmaceutical industry it is mainly applied to:

* Processing and Service Gas

* Liquids

* Injectibles

Filtration of water, gas, processing gases and liquid preparations are important for the final quality of the manufactured products and safety of instruments and equipments but filtration of perennials is most critical where quality assurance overrides, or shall we say should over-ride, economic considerations It is a sad state of affair that it is ignored by not all but many, one of the over-riding reasons could be a lack of understanding by the relevant people involved in the application. The solution is simple. G.M.P. of all the Pharmaceutical Companies dealing in perennials should include in its training programme in exclusive regimen on filtration. Such an exercise will familiarise and train the staff so that they can use the 'Right' filter or filter set-up for the application. Further it will prevent them from being taken for a ride by many a unscrupulous 'Sellers'.

Most of the considerations are straight forward like compatibility of filter medium with the system to be filtered. Among other considerations like pressure differential generated by the filter set-up, Media migration which can Ruin the whole batch, channeling which can have similar results, removal rating deserves the most critical analysis.

Removal Rating

Filters are rated by their efficiency to remove a certain size and above of particulate matter.

What is generally not elaborated is the efficiency factor of a particular filter. Usually we come across the removal rating only as 5 micron or 1 micron etc. Some manufacturers use the word 'Nominal' after the rating like 5 micron - nominal. Unless queried the description, 'Nominal' does not carry much information. Poor customer thinks that a 5 micron filter will stop all the particulate matter of 5 micron and above. This would only be possible if technology existed which would help to produce a filter of the same sized 'Holes'. A 1 micron filter would have all the 'Holes' of |is less than~1 micron, a 5 micron filter would have all the 'Holes' of |is less than~5 micron and so hence & so forth.

However, practically this is not so. In fact there are no Holes. These are channels of a certain dimension created by controlled weaving or packing or wintering correct grade fibre or material. Liquid or gas flows through a path of least resistance and these paths dictate removal rating. Size distribution of these paths dictate efficiency. Narrower the size distribution better is the filter.


Nominal Rating

Nominal rating used by many filter manufacturers is defined by National Fluid Power Association (NFPA) as "an arbitrary micron value assigned by the filter manufacturer, based upon removal of some percentage of all particles of a given size or larger". Usually 'Removal Rating-Nominal' means that 98% by weight of the contaminant above the specified rating is removed and 2% by weight can pass through.

Superficially it is a very fair conclusion. But on critical analysis of the result it is evident that there is an enormous flaw in the procedure.

Relation between the weight of same number of particles of different size is based on cubes of the radii of the size hence ratio of weight between say 1 micron and 40 micron is 1 to 80.

Some manufacturers do not base their removal rating on 98 per cent by weight but instead on 95 or 90 or even lower. One cannot, therefore, assume that a filter with a nominal rating of 10 micron will retain all or most particles 10 micron or larger. An obvious reason seems to be that it make their filters look better and that assigning of absolute rating is impossible on a non-fixed pore filter.

Absolute Rating

The NFPA defines 'Absolute Rating' as: "The diameter of the largest hard spherical particle that will pass through a filter under specified test conditions". It is an indication of the largest opening in the filter. Only integrally bonded filter media qualify for such rating.

The original test and term for the absolute filtration rating was proposed by Dr. D.B. Pall-Chairman of the board of Pall Corporation - in the Mid-1950's. It was considered by the filter panel of SAE Committee A-6 and was adopted with minor changes.

There are various methods for establishing the absolute rating of a filter. It mostly depends on the filter medium and processing industry. In all cases the filter is 'Challenged' by pumping through it a suspension of readily recognised contaminant (E.G.) glass beads as they are easily recognised from the 'Back-Ground' contamination by their shape or bacterial suspension.

Glass Bead Challenge Test

A suspension of glass beads, all lying within a specific size range, is passed through a filter and effluent collected on an analysis membrane. Membrane is examined microscopically to determine the largest spherical bead that has passed thorough. The size of this particle dictates the removal rating.

Beta Rating System

While absolute ratings are clearly more useful than nominal ratings, a more recent system of rating is the 'Beta Value' it is a ratio and is determined by using the Oklahoma State University 'OSU F-2 filter performance test' the test was originally developed for use on hydraulic and lubricating oil filters and it has been adapted by Pall Corporation.

ACFTD (A.C. Fine Test Dust) is slurried in water and mixed for about six weeks to obtain a stable suspension.

Filter is challenged with this slurry. Upstream challenge and down-stream effluent is continuously monitored by an array of particle counters calibrated for different sizes. Beta value is obtained as a ratio of upstream challenge of a given size and above to downstream recovery of the same given size and above. Some procedural difficulties become extremely important with such tasks. To ensure credibility a count of less than 10 particles is not accepted and therefore careful challenge levels have to be applied so that on one hand the detection limit of counter is not crossed and on the other hand a reasonable count is obtained for downstream.

Beta value is related to percentage efficiency by the equation:
Percentage Efficiency
(Beta Value - 1)/Beta Value X 100

Beta (%)
Value Efficiency

100 99
1000 99.9
5000 99.98
10000 99.99
100,000 99.999

Beta value is a better descriptor for filter efficiency for filter than percentage. Difference between a filter of 99.98% and 99.99% does not register as sharply as when differentiated by beta value of 5000 and 10,000.

It is now evident that 'Nominal' cannot be given an efficiency level because it can be anything from 60 to 95%, but 'Absolute' rating is precise. It is usually taken at a beta value of 5000 to 10,000.

For sterile filters procedures are much more critical and well defined.

FDA defines a sterile filter as "A filter when challenged with bacterial pseudomonas diminuta at a concentration of atleast 1x10 to the power 7 per square centimeter of the effective filter area produces a sterile effluent."

Most of the filter manufacturers provide this assurance but one ought to use his judgement in deciding whether this minimum requirement has an acceptable statistical criterion to meet quality assurance limits of his organisation as dictated by GMP after all quality assurance is statistical in nature. If it were not so than no company which quality assured their products would have an exactly laid out "Product Recall Procedure".

There would be no need to but they all have one. To stress the point lightly: We all indulge in "Safety" when we invite guests! We arrange for enough to cater for all and a plus! Why do we shy away from this exercise in our dealing with human lives and their well beings? Mind Boggles!!

There is no way of actually testing each filter of make sure it shall give a sterile effluent. That would be destructive testing and filter after test would no more be usable.

Therefore integrity tests are developed results of which can be Co-Related with actual challenge test results. Some filter manufacturers use "Bubble Point" and "Pressure Decay Test". In the "Bubble Test". Air pressure is applied to a thoroughly wetted filter. Pressure is slowly and gradually increased. The pressure at which the bubbles start to appear is called the "Bubble Pressure Point" and it is directly related to the removal rating.

Some manufacturers, like Pall use forward flow test. In forward flow test air pressure is applied to a thoroughly pre-wetted filter. System is allowed to stabilize and then volume of air passing through the filter by "Diffusive flow" is measured. This is directly co-related to the actual bacterial challenge test results.

A critical study of these co-relations will convincingly show that co-relation to "forward flow" is better than "Bubble Point". Copy rights prevent me from making an actual comparison and therefore, I shall have to leave it to an inquisitive mind to indulge in such an exercise.

One of the major differences between "Bubble Point" and "Forward Flow" is that "Bubble Point" gives qualitative information regarding the condition of filter but cannot reliably confirm its removal efficiency and diffusive flow measurement gives quantitative information regarding the condition of filter and can reliably confirm its removal rating.

Reasons becomes evident when methods of co-relations are critically analysed.

FDA definition for a "Sterile Filter" does dictate the minimum challenge level of pseudomonas diminuta but can not quantitify flow rate. If same filter was challenged with same challenge level at different flow rates, results can be different. Further - Test conditions like pressure, viscosity, pH, temperature and pressure fluctuations can all have influence on the filter's capacity capture or retain bacteria.

Validation results provided by a manufacturer are under ideal conditions and they should all be taken with "A Pinch of Salt". Such ideal conditions are hardly met in actual applications. Therefore validation results co-related to the minimum challenge level as dictated by FDA have no safety margin at all. Some manufacturers like Pall do take this in consideration and they do provide such safety margins and their co-relations are based on results of actual bacterial challenges of upto 1x10 raised to the power of 13.

Another important consideration is whether co-relation span contains "User's Limits". Guide Published by one of the filter manufacturer for co-relation between "Bubble Point" and actual sterility of effluent starts at 58 PSI and ends at 48 PSI. Users limit of 45 PSI is not covered.

A further critical look at FDA definition will reveal is "Limitation" when mycoplasma and viruses and bacteriophages are considered.

The range of "Sterile Filters" start at 0.3 micron - The size of pseudomonas diminuta. Paper published by Glen Howard and Ronnie Bubberstein in journal of parental drug association of July/August 1978 states that studies of challenge by pseudomonas diminuta on 0.2 micron rated filter by different manufacturers revealed that 100% sterility was not obtained in some cases. The culprits are usually naturally occuring water-borns bacteria of mycoplasma class. These bacteria have no cell walls and they maintain their integrity by a very thin and pliable plasma membrane. Under certain conditions they "Squeeze" their way through a filter medium of upto 0.2 micron rating. Rather than using a finer filter which may generate flow rate problems it is most economical to use a filter with zeta potential. Most of these sub-0.3 micron Bacteria, bacteriophages, viruses and pyrogens are negatively charged and a positive zeta potential on media surface of the filter will capture them by electro-static force. As the channels are not blocked by such capture pressure differential is not generated and filter provides its normal useful life.

It should be amply clear that manufacture of a "Good" filter does involve vast amount of technology, research, complete understanding of applications, official requirements and excellent G.M.P. manufacturers who take the trouble to provide "Absolute Performance" naturally want their clients to know about it and membership of accreditation agencies is one way of doing just that.

ISO 9001 is one such accreditation and their inspectors ruthlessly maintain "Credibility" of their agency so that membership can be a matter of pride. They evolve and update their "Legislature" continuously. ISO 9001 members usually deal with member companies for their raw materials and when not available they manufacture their own raw materials! This helps them to maintain their membership spotlessly. Such companies dread "Substandard" and its occurrence is highly unlikely. Pall is an example in the field of filtration.

It is difficult to grasp the whole of filtration technology by the responsible personnel in an organisation using filtration. Filter manufacturers or their trained agents provide full technical service to the end user so that he can with confidence buy an appropriate filter set-up for his application at the most economical price. But it is well to understand the involvement for such sizing exercises. Otherwise how would he understand and identify the "Rights" and the "Wrongs". Anyway sales person is always on the look out for "Gullibles".

But to be fair to the sale person, one must consider the sure sling shot he faces. Actually its two in one. Namely how much? and how long it will last!!

How much is easy but how long is impossible. Poor chap doesn't know the magnitude of challenge, parameters of flow rates, operative differential pressure, fluctuations in pressures, temperature variations etc. etc. All he can do is give a good FIB and close the sales.


1. Principles of filtration by Pall process filtration limited

2. Journal of perentral drug association - July/August 1978
COPYRIGHT 1993 Economic and Industrial Publications
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Abdulhusein, Mansoor
Publication:Economic Review
Date:Aug 1, 1993
Previous Article:Pharmaceutical industry in Pakistan.
Next Article:SIZA challenges and dreams.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters