Dubious Microbial Control Methods in Pharmaceutical Waters
Case Studies on Foreign Substances and Impurities Violations
By Andrew W. Collentro - Water Consulting Specialists, Inc.
The United States Pharmacopeia (USP)i delineates standards for drug preparations
to be bought, sold or dispersed in the United States. Among such standards, are
the Official Monographs for Purified Water and Water for Injection (WFI),
two common, non-packaged, pharmaceutical waters. The chemical and microbial quality
of these waters, which are two of the most widely used excipients, diluents, or
solvents used in pharmaceutical manufacturing, is gauged only by a small number
of generalized parameters. By definition, these critical process parameters define
the minimum product water quality attributes. The necessity of testing for additional
impurities should be evaluated based on specific product and process requirements.
One should not be required to verify the absence of Added Substances of Foreign
Substances and Impurities in final compendial waters if it were not logically
feasible for suspected impurities to be present. Nor should specific contaminant
testing and monitoring be required if an occurrence would not alter the safety,
efficacy, identity, or quality of the final product, particularly when the alleged
impurities are indistinguishable from other ubiquitous compounds.
click the image to enlarge
Figure 1: This USP Purified Water System features a generation system
based on reverse osmosis technology followed by a deionization process for
polishing the water.
Nonetheless, the presence of objectionable impurities in final product waters,
whether present in the incoming feed water, specifically added to the process
stream, or a byproduct of a water treatment or conditioning process, would logically
be impermissible. This would include the presence of anti-microbial agents, bactericides,
preservatives, or other compounds which could mask or control viable microorganisms
in process streams where waters are utilized (e.g. points-of-use in a compendial
water distribution system). The presence of such compounds would likely not be
evident if testing was limited to those critical process parameters outlined in
the monographs. The case studies discussed in this paper will describe the presence
of objectionable materials whose occurrence could only be theorized by microbial
data for operations which were inconsistent with the control methods utilized.
Added Substances
The USP Official Monographs for Purified Water and WFI both stipulate that
they shall contain no Added Substances.ii The definition of an Added Substance
for pharmaceutical waters is not specifically defined by the pharmacopoeia and
has been the subject of debate. However, the General Notices Section of USP, while
not specific, describes an Added Substance as antimicrobial agents, bases, carriers,
coatings, colors, flavors, preservatives, stabilizers, and vehicles which may
be added to pharmaceutical preparations (not Official Substances) to enhance its
stability, usefulness or elegance or to enhance its preparation.iii However, both
USP Purified Water and WFI are Official Substances. They should be free of any
added materials regardless of the requirement of No Added Substances in the monographs.
This specifically does not exclude the employment of chemicals or compounds used
during treatment, preparation, or conditioning preparation in the finished product.
The use of water treatment chemicals and compounds, in accordance with good engineering
practice, has been widely accepted in pharmaceutical water treatment as discussed
elsewhere.iv This includes the use of antimicrobial agents, either intermittently
or continuously, such as chlorinated compounds and ozone. The employment of these
techniques would be permissible, even in storage and distribution system for pharmaceutical
waters provided that the compounds are removed prior to use. Specifically, the
Official Substance refers only to Purified Water or WFI as used, not in a storage
or distribution network.
Foreign Substances and Impurities
While Added Substances have been defined above as substances added to Official
Substances or Official Preparations after they have been produced, Foreign Substances
and Impurities refer to a general requirement for purity that places responsibility
upon the manufacturer of the article. The General Information Chapter (1086) on
Impurities in Official Articles defines Foreign Substances as followsv:
“Foreign Substances, which are introduced by contamination or adulteration, are
not consequences of the synthesis or preparation of compendial articles and thus
cannot be anticipated when monograph tests and assays are selected.”
click the image to enlarge
Figure 2: The generation system shown here is a traditional deionization
based system including regenerable cation and anion ion exchange followed
by a mixed-bed ion-exchange polisher.
This passage suggests that Foreign Substance are often the result of product contamination
and not byproducts of production of the article. In general, water treatment techniques,
which purify, yet in many cases alter the nature of the impurities in a theoretically
pure water system, would not be classified as Foreign Substances. As an example,
ultraviolet light used for decomposition of organic material may produce new,
smaller molecular weight organic compounds, not found in the feed water and specifically
present due to the type of treatment. These organic compounds would be a consequence
of treatment, generally qualified by the monograph requirement for compliance
with the USP Test Chapter for Total Organic Carbon , and not considered a Foreign
Substance.
A more resonant definition of Foreign Substances and Impurities can be found in
the General Notices Section of USP and states the following:
“Tests for the presence of foreign substances and impurities are provided to
limit such substances to amounts that are unobjectionable under conditions in
which the article is customarily employed. While one of the primary objectives
of the Pharmacopeia is to assure the user of official articles of their identity,
strength, quality, and impurity, it is manifestly impossible to include in each
monograph a test for every impurity, contaminant, or adulterant that might be
present, including microbial contamination. These may arise from a change in the
source of material or from a change in processing, or may be introduced from extraneous
sources. Tests suitable for detecting such occurrences, the presence of which
is inconsistent with applicable manufacturing, practice or good pharmaceutical
practice, should be employed in addition to the tests provided in the individual
monograph.”
Official Monographs the minimum requirements. Additional testing would certainly
be required if any materials were present that could possible alter the identity
of the product. If suspected, this specifically outlines additional testing requirements
even if by unofficial methods; those which may not be defined by an Official USP
Test Chapter. For example, if a chemical that was not naturally-occurring in the
feed water was used in a specific treatment step for pharmaceutical water generation
(anti-scalents, polymers, flocculants are good examples), specific tests must
be performed to ensure the absence of these specific compounds in the official
article.
In addition, The General Notices referenced section places ownership on
defining quantifying those compounds present deemed to be objectionable on the
manufacturer of the article. Pharmaceutical waters are not pure substances; in
fact they are quite impure when compared to high purity waters used for certain
semiconductor, microelectronics, or analytical purposes. It would be shortsighted
to think that there are not specific compounds found in pharmaceutical waters
whose presence increases the possibility, no matter how improbable, of compromising
product efficacy (e.g. certain microbial species).
Finally, the passage defines a Foreign Substance or Impurity not only as that
introduced by contamination or adulteration, but those present regardless of the
origin of said compounds. The production of compendial water is unique when compared
to other Official Substances. The starting material is not only different at every
production site, but can be very dynamic for a specific system even over a short
operating period. For compendial waters, the only limit placed on the source material
is that it must meet the water quality attributes outlined in the EPA National
Primary Drinking Water Regulations, comparable local doctrine, or World Health
Organization (WHO) standards,. All are very broad standards and certainly do not
encompass all impurities which may be deleterious to an article. It is entirely
possible that changes in feed water (source material) quality can lead to changes
in product water quality, the detection of which may go unnoticed by relying solely
on testing of the product to the monograph requirements. This emphasizes the importance
of periodic feed water testing and intra-component monitoring throughout a pharmaceutical
water generation system.
Foreign Substances and Impurities can also originate from extraneous sources such
as leachates from piping system or any other water-contact surface. Chemical injection
systems, such as those described above should also be free from objectionable
impurities. Nitrogen, ozone or other dissolved gases that may be added to a process
should not contain impurities which could alter the safety, efficacy, purity,
or identity of a product. The purity of any additive should be carefully scrutinized,
especially if they are employed where there is not downstream treatment technology
for removal of the contaminant (e.g. nitrogen quality for a blanketed WFI storage
tank).
Above all else, the most concerning origin of Foreign Substances and Impurities
are those which go undetected and untreated and those which are by-products of
treatment processes whose occurrence could not be readily predicted. The following
case studies will review specific instances where extraneous compounds, whose
presence would not specifically be detected by compliance with USP Purified or
WFI Official Monograph test requirements and whose presence would not be
readily predicted, yet certainly could be classified as Foreign Substances and
Impurities.
Case Study #1 – Point-of-use Preparation Prior to Sampling
The FDA Guide to Inspections of High-Purity Water Systemsvii discusses how samples
obtained during the validation of sampling procedures should be as representative
as possible. Specifically mentioned is the use of hoses during sampling if hoses
are employed when water is drawn and also flushing prior to sampling if the use-point
is flushed when water is drawn. By extension, preparation of use-points with an
anti-microbial or other agent prior to sampling would only be appropriate if an
identical procedure were used prior to actual use of the water. The use of isopropyl
alcohol (IPA) or similar compounds for preparing a sample point is commonplace,
whilst it is seldom, if ever, used during actual water consumption.
click the image to enlarge
Figure 3: The USP purified water system shown here is similar to
the one shown in Figure 1. The storage and tribution system is sanitary
polypropylene construction with zero dead-leg type diaphragm valves at all
use-points.
Consider the USP Purified Water System shown below in Figure 1. The generation
system is based on reverse osmosis (RO) technology followed by a deionization
(DI) process for polishing the water such that it exceeds the quality required
to pass the USP Conductivity Test.viii The distribution loop was fabricated of
solvent welded polyvinylchloride (PVC) and solvent welded ball type valves were
used at use-points. Overall, the system was not considered sanitary and materials
of construction and mechanical design were not in accordance with current Good
Manufacturing Practices (cGMPs). Chemical sanitization was performed on an annual
basis and the use-points were sprayed excessively with a 20% IPA solution prior
to taking grab sample. Interestingly, given the infrequent sanitizations and substandard
design, the total viable bacteria results for samples taken from use-points consistently
were reported as less than 1 colony forming unit (cfu) in a 1 ml sample. Conductivity
and total organic carbon (TOC) measurements taken on-line were within specification.
At one point the TOC monitor was taken off-line for maintenance and grab samples
for TOC were taken at the use-points. Due to concern of possible contamination
of the TOC samples from the IPA spray, its use was temporarily suspended. Within
days of sampling with this modified procedure, all total viable bacteria counts
were reported as too numerous to count (TNTC). A daily chemical sanitization plan
was initiated to maintain the system within acceptable microbial levels.
From the data generated, it was apparent that the IPA had a profound effect masking
the actual microbial activity within the system. It was determined that the use
of IPA, prior to water being drawn for operations could lead to potential contamination
of the product (i.e. USP Purified Water) and was a potential violation of the
USP Purified Water Official Monograph requiring the absence of Added Substances.
Although the IPA spraying procedure was only used during sampling, it became apparent
that its use concealed a systemic microbial problem. Interestingly, no published
position on the use of alcohol or any other agents for sampling point preparation
is available from industry regulators.
Case Study #2 – Presence of Feed Water Residual Disinfectant in USP Purified Water
The generation system shown in Figure 2 is a traditional deionization (DI) based
system including regenerable cation and anion ion-exchange followed by a mixed
bed ion-exchange polisher. The storage and distribution system design employed
sanitary 316L Stainless Steel and was hot water sanitized every three months at
80°C for two hours. All use point valves were sanitary zero dead-leg style
diaphragm valves. There were six point-of-use drops in manufacturing areas and
one that was hard piped to an autoclave. Each use-point was sampled every other
day for total viable bacteria. The use-point valves were flushed for 30 seconds
prior to sampling and were not prepared with IPA or any other compound.
The system was design to produce USP Purified Water. Conductivity and TOC were
measured in-line with periodic grab samples taken from use-points. Total viable
bacteria samples were taken and reported in cfu/100 ml of sample. Results were
typically reported less than 1 cfu / 100 ml and always below the Action Level
of 100 cfu / 100 ml. The system had produced water within quality levels in excess
of fifteen years.
At one point, the distribution system was modified to include additional drops
for a manufacturing expansion project. A change control was generated and an intense
30 day sampling program was initiated for the new drops, the existing use-points,
and at various points throughout the generation system. During the enhanced testing
phase for the generation system, the presence of combined (total) chlorine was
identified in components downstream of the activated carbon unit, the unit operation
specifically design for total chlorine removal. Further testing revealed the presence
of chloramines compounds, a residual disinfectant added to the water system by
the local municipality, at the outlet of the generation system and even at points-of-use.
Chloramines are relatively stable compounds that can be difficult to remove by
traditional methods.ix
It was confirmed that high concentration of chloramines (greater than 2.0 ppm)
were present in the feed water and that the existing activated carbon unit was
undersized for the application. The system was retrofitted with three additional
activated carbon units, such that two units were operated in series at any particular
time. Upon system modification, the absence of any residual disinfectant at points-of-use
was confirmed and total viable bacteria counts exceeded action levels on a regular
basis.
The hot water sanitization sequence of the storage and distribution system was
increased to every other day and a weekly chemical sanitization of the generation,
storage, and distribution systems was initiated to further inhibit the development
of bacteria and microorganisms. In addition, the replacement frequency of the
0.2 micron filters in the distribution loop was increased from every six months
to every month. The system was re-validated to reflect all of the operating and
process changes that were implemented.
The presence of chloramines, a chemical anti-microbial agent used in municipal
water treatment, at use-points is a clear violation of the Foreign Substances
and Impurities instruction and could also be considered an Added Substance. In
this application, there was likely an impact on product quality and stability.
Case Study #3 – Presence of System Generated Disinfectant in USP Purified Water
The USP Purified Water System shown in Figure 3 is similar to that in Figure 1;
an RO and DI based generation with a chemically sanitized storage and distribution
system. Like the system shown in Figure 1, this system was only sanitized on an
annual basis with a 1% solution of hydrogen peroxide and peracetic acid. The storage
and distribution system is sanitary polypropylene construction with zero dead-leg
type diaphragm valves at all use-points. There are only four use-point drops and
only one is hard piped with a minimal dead-leg to a glassware washer. Each point
is sampled every other day and the drops are not prepared (sprayed) with IPA prior
to sampling. However, each sample point is flushed for 15 minutes prior to taking
a sample. A sample volume of 1000 ml was historically taken and total viable bacteria
levels always were reported less than 1 cfu per 1000 ml.
During an annual preventive maintenance it was determined that the ultraviolet
(UV) light in the distribution system was incorrectly equipped with 185 nm wavelength
UV bulbs. The original process was based on 254nm wavelength bulbs designed for
microbial control. Immediately after the 185 nm were replaced with the 254 nm
bulbs, total viable bacteria samples were returned indicating the presence of
viable organisms in a 1000 ml sample. Within days, results were reported as TNTC.
Initially it was suspected that some form of contamination occurred during the
annual maintenance work on the system. However, after another chemical sanitization
was performed, total viable bacteria levels were again reported as TNTC within
a few days.
After an exhaustive investigation, the UV bulbs were replaced with 185 nm bulbs
in an effort to recreate the historical performance of the system. Immediately,
the system returned to a state of complete microbial control with total viable
bacteria counts reported less than 1 cfu in 1000 ml samples from all use-points.
Further laboratory testing confirmed, as suspected, a very powerful oxidative
species with a limited half-life. Upon review of the UV design, it was determined
that hydroxyl radicals generated by the UV dose of 90,000 microWatt–second / square
cm dose at 185 nm wavelength were acting as an anti-microbial agent by oxidizing
any viable microbial compounds in the system.
While the process for using UV light to minimize microbial proliferation in compendial
water system is well accepted, the concern is the use of high doses of 185 nm
light and the resulting hydroxyl radicals that are formed at this point in the
system. Although these radicals have half-lives of various durations depending
on their structure, the presence of these compounds at points-of-use would be
inappropriate. Since no destruction or removal method for these compounds exist
in this system (other than time), there is absolutely no verification of their
absence at use-points and in most cases such as this, no method to confirm the
effect of their presence on the efficacy of the product.
The application of 185 nm UV light, often oversized, has become increasingly popular
in distribution loops as a microbial control method. Advantages include that it
is a relatively inexpensive process, there is excellent microbial control, and
intervals between sanitizations can be extended, and maintenance is minimized.
However, the presence of hydroxyl radicals or any other oxidative compounds at
points-of-use would be considered a Foreign Substance by the definitions above.
Since half-lives of these compounds are relatively brief, testing for their presence
would be very difficult.
Ozone and Added or Foreign Substances
The use of ozone in the pharmaceutical water and related industries has historically
been limited to storage and distribution systems for USP Purified Water. As a
relatively inexpensive and extremely effective sanitization method, its use has
likely been restricted for several reasons. First is the concern that ozonation
produces unwanted by-products and that ozone itself is considered an Added Substance.
However, as mentioned above, in the synthesis of pharmaceutical waters, by-product
are inevitable and acceptable as long as the critical quality parameter levels
are not exceeded and the compounds are not considered to be Foreign Substances
or Impurities. In addition, the ozone is generally injected continuously into
a storage tank, destroyed continuously in the loop by 254 nm UV light, and its
absence confirmed by on-line monitoring.
The second reason that has driven the market away from ozonated systems was the
concern regarding regulatory acceptance. However illogical, this perception is
likely a result of negative analysis given to ozone systems in the FDA Guide to
Inspection of High-Purity Water Systems.x Thermally sanitized systems had become
commonplace in the industry.
Lastly, ozone had never been used in WFI applications due to the restriction on
the method of manufacture to distillation and reverse osmosis (terminal purification
technique). This historically excluded all WFI generation system from using ozone.
With the recent change in the USP WFI Monograph allowing for alternate methods
of producing WFI, including ambient temperature operations, the use of ozone in
WFI systems may become a reality.
Conclusions
The three case studies above detail situations where the presence of anti-microbial
agents masked the presence of viable bacteria at points-of-use. In the later two
instances the presence of these compounds was more difficult to identify. If the
microbial activity in a system is not consistent with the design expectations,
considering operation and maintenance, there is likely additional, concealed control
by a method that violates the compendial requirements, as the USP General Notices
Section requirements are as binding as the USP Official Monograph requirements.
The use of chemicals, UV light, ozone and other compounds are ubiquitous techniques
used to purify and maintain quality for compendial water treatment systems. As
a result of their use, in specific situations, byproducts will be formed. The
use and presence of anti-microbial agents and processes must be carefully evaluated
to demonstrate that the water delivered to use-points does not contain anything
that masks the presence of bacteria.
United States Pharmacopeia (USP29) and the National Formulary (NF24), United
States Pharmacopeial Convention, Inc. January 1, 2006. ii USP29, pp. 2262, 2265.
iii USP29, p. 6. iv Zoccolante, G. and Collentro, W.V. “Defining and Added Substance
in Pharmaceutical Water”, Ultrapure Water (11) 2, pp. 34-39 (March 1996). v USP29,
pp.2920-2921. vi USP28, pp. 6-7. vii United States Food & Drug Administration,
“Guide to Inspection of High Purity Water Systems” (1993). viii USP29, pp. 2653-2654.
ix Collentro, A.W. “Chloramines and Their Impact on Pharmaceutical Water System
Design” Ultrapure Water pp. 29-34, (March 2001). x United States Food & Drug Administration,
“Guide to Inspection of High Purity Water Systems” (1993).
Talkback! Pharmaceutical Processing is pleased to provide you an opportunity to share your opinions on any of the news stories or articles on our site. We reserve the right to edit/remove comments.
From a vision to a vial in 12 months -
Process development and... 
