Attempts to include reverse osmosis (RO) as a production method in the Ph. Eur. Monograph Water for Injections (WFI) have been ongoing for over 15 years. However, up to now these efforts have failed, due in particular to concerns of regulators with regard to microbial safety. In the last few years, discussions of the topic have resumed and become more intense. Now, after a positive review of current scientific data, the Working Group of the European Pharmacopeia on Water for Pharmaceuticals has also included RO in the draft of the revised WFI monograph. In an interview with our author Ruven Brandes, Head of Technology and Compliance Support, Technical QS, WDT, we talked about the background and consequences of these important new developments.
Why has it taken so long to include reverse osmosis in the WFI monograph? Were the available data actually that sparse or were the regulators’ fundamental misgivings simply too strong?
From the semiconductor industry we have known for decades that it is possible to use classic water-treatment techniques (membrane technology and ion exchange) to obtain water of much higher quality than WFI. About 20 years ago, reverse osmosis was approved for a time in the USA as a final stage of treatment. However, experience gathered with microbiological quality at that time was very negative. As a consequence, further development was slowed.
Added to this were the rather conservative approach of the pharmaceutical industry and a representative of the British regulators in the EDQM Water Working Group who was vehemently opposed to membrane technology. This gentleman has since retired and - lo and behold - suddenly things are happening. Insiders predicted this would happen years ago.
Another strong contributing factor is that the equipment used for the membrane techniques can now be hot-water sanitised. Microbiological contamination can be reliably eliminated in this way.
What are the advantages of membrane technology?
Generally, investments and operating costs are significantly lower. One way to save a truly large amount of money is to produce and distribute only one single water quality grade - in other words, only WFI - in a production unit and to do so economically using membrane technology, even putting it into cold storage. Of course, this may call for a higher monitoring frequency, but the overall savings here are far greater. It is only slightly more expensive than purified water.
Why must reverse osmosis be combined with another method (e.g., deionisation, ultrafiltration) for the production of WFI?
It is the combination of different techniques that brings success. Each one contributes to achieving the water quality (see Figure 1).
|Softening||Protects the reverse osmosis from lime deposits|
Reverse Osmosis (RO)
|Primary reduction stage for salts, TOC, bacteria and endotoxins (but not 100%, since RO modules lack integrity (leakage))|
|CO2 removal and reduction of micromolecular organic carbon com-pounds (which even distillation does not retain). You can also use membrane degasification to remove oxygen.|
|Fine salt removal to achieve lowest conductivity|
|Final integral barrier for endotoxins and germs|
Figure 1: Water-treatment method
What do you think of the monograph’s new requirement for regular measurement of TOC: Does it make sense or is it overly bureaucratic?
I definitely think the TOC measurement makes sense. Aside from temperature, the organic carbon content is a crucial factor for the tendency towards contamination. At a TOC < 10 ppb no germ growth is possible. The actual problem is that the present limit value of 500 ppb is far too high. Many drinking water sources in this part of the world have already reached this value. I think that it will be lowered to < 50 ppb within the next few years. Incidentally, this value corresponds to values found in practice, which are about 25 ppb, depending on the quality of the system. Values well below 25 ppb can also be achieved by strict monitoring and compliance with quality features of a system.
In practice this change would mean that cold storage of WFI is also possible. Does this conflict with Annex 1 of the EU-GMP Guide?
Not at all! Annex I of the EU GMP Guide says: [...] Water treatment plants and distribution systems should be designed, constructed and maintained so as to ensure a reliable source of water of an appropriate quality. They should not be operated beyond their designed capacity. Water for injections should be produced, stored and distributed in a manner which prevents microbial growth, for example by constant circulation at a temperature above 70°C. [...]
I visualise the standard for the future as having WFI production using only membrane technology with double-pass RO, membrane degasification, EDI, UF, supplemented where applicable by UV equipment for TOC reduction. Membrane degasification with N2 vacuum operation for oxygen reduction. Two completely separate distribution systems for hot and cold WFI (with ozone), insofar as there will still be a need for hot WFI in the production process. Otherwise, everything should be stored cold, perhaps operating with ozone. However, ozone should not always be the first choice, since it requires substantial qualification and validation efforts. Additional negative aspects are represented by maintenance of the ozone equipment and the present legal grey area regarding biocide regulations.
Will no one use distillation to produce WFI in the future?
Perhaps not in the distant future. But this will take time, since production by distillation is often described in manufacturing specifications. Making changes in the status quo will be difficult. However, membrane technology could become established fairly quickly in new production facilities with a large WFI requirement.
I think that for the time being there will not be much change with existing systems. The change in paradigms must be accepted first. Afterwards, the advantages of immediate savings will con-tribute to changing attitudes.
But membrane technology is still a source of concern. Subjects in discussion are particularly topics relating to biofilm, possible fibre fracture or defective seals, hygiene during maintenance or module exchange, module testing and membrane ageing.
This is and will continue to be a very exciting area.
This interview was conducted with:
WDT eG, Garbsen, Germany