12300

First Supplement to USP 36–NF 31 General Information / ⟨1231⟩ Water for Pharmaceutical Purposes 5757

Heat Penetration and Microbiological Challenge—Thecore of the validation activity is confirmation of acceptableheat penetration using temperature measurements and mi- ⟨1231⟩ WATER FORcrobial challenge inactivation. Temperature probes and bio-logical indicators are placed within the load at worst-case PHARMACEUTICAL PURPOSESlocations (e.g., the coldest portions of the loaded chamber).Introduction of the thermocouples must not alter the integ-rity of the container. Biological challenges are placed in con-tainers adjacent to those that contain heat penetrationprobes (or the same unit with external temperature meas- INTRODUCTIONurement).

Proof of cycle efficacy is provided by replicate studies in Water is widely used as a raw material, ingredient, andwhich the BIs perform as expected and the physical meas- solvent in the processing, formulation, and manufacture ofurements correspond to the expected values of time and pharmaceutical products, active pharmaceutical ingredientstemperature or F0. If the microbial and physical measure- (APIs) and intermediates, compendial articles, and analyticalments do not correlate, an investigation is in order, and reagents. This general information chapter provides addi-corrective action must be taken to rectify the discrepancy. tional information about water, its quality attributes that areSamples from the hottest regions of the load are used for not included within a water monograph, processing tech-evaluation of material stability and quality. niques that can be used to improve water quality, and a

Product Quality and Stability Evaluation—Manufactur- description of minimum water quality standards that shoulders must conduct ongoing evaluation of the product’s ability be considered when selecting a water source.to withstand the routine sterilizing conditions. The evalua- This information chapter is not intended to replace ex-tion should encompass the essential quality attributes with isting regulations or guides that already exist to cover USAattention focused on known and potential new impurities and international (ICH or WHO) GMP issues, engineeringand those materials that receive the most heat input. Manu- guides, or other regulatory (FDA, EPA, or WHO) guidancesfacturers also should evaluate appearance, other physical for water. The contents will help users to better understandproperties, and container–closure integrity as a part of this pharmaceutical water issues and some of the microbiologi-effort. For microbiological media, the ability of the media to cal and chemical concerns unique to water. This chapter ismeet growth promotion and other requirements is required not an all-inclusive writing on pharmaceutical waters. It con-as indicated in the appropriate test chapter(s) (e.g., Microbi- tains points that are basic information to be considered,ological Examination of Nonsterile Products: Microbial Enumer- when appropriate, for the processing, holding, and use ofation Tests ⟨61⟩, Microbiological Examination of Nonsterile water. It is the user’s responsibility to assure that pharma-Products: Tests for Specified Microorganisms ⟨62⟩, and Sterility ceutical water and its production meet applicable govern-Tests ⟨71⟩). mental regulations, guidances, and the compendial specifi-

cations for the types of water used in compendial articles.Control of the chemical purity of these waters is impor-Routine Process Control tant and is the main purpose of the monographs in this

compendium. Unlike other official articles, the bulk waterAll sterilization processes should be subject to formalized monographs (Purified Water and Water for Injection) alsopractices that maintain them in a controlled state. The prac- limit how the article can be produced because of the belieftices outlined in Sterilization of Compendial Articles ⟨1229⟩ that the nature and robustness of the purification process isinclude the general requirements appropriate for all steriliza- directly related to the resulting purity. The chemical attrib-tion systems. This is accomplished by a number of related utes listed in these monographs should be considered as apractices that are essential for continued use of the process set of minimum specifications. More stringent specificationsover an extended period of time. The practices include: may be needed for some applications to ensure suitabilitycalibrartion, physical measurements, physicochemical inte- for particular uses. Basic guidance on the appropriate appli-grators, indicators for sterilization, monitoring of bioburden, cations of these waters is found in the monographs and isongoing process control, change control, preventive mainte- further explained in this chapter.nance, periodic reassessment, and training. Control of the microbiological quality of water is impor-The use of parametric release is common in the terminal tant for many of its uses. Most packaged forms of watersterilization of finished product containers. This subject is that have monograph standards are required to be sterileaddressed in Terminally Sterilized Pharmaceutical Products— because some of their intended uses require this attributeParametric Release ⟨1222⟩ for health and safety reasons. USP has determined that aREFERENCES microbial specification for the bulk monographed waters is

Berger, T., & Trupp, K., Validation of Terminal Steriliza- inappropriate, and it has not been included within the mon-tion, chapter in Validation of Pharmaceutical Processes: ographs for these waters. These waters can be used in a3rd edition, edited by J. Agalloco & F. J. Carleton, In- variety of applications, some requiring extreme microbiolog-formaUSA, New York, 2007. ical control and others requiring none. The needed micro-Owens, J., Sterilization of LVPs and SVP’s, chapter in bial specification for a given bulk water depends upon itsMorrissey, R., & Phillips, G.B., Sterilization Technology— use. A single specification for this difficult-to-control attri-A Practical Guide for Manufacturers and Users of Health bute would unnecessarily burden some water users with ir-Care Products, Van Nostrand Reinhold, New York, 1993. relevant specifications and testing. However, some applica-Young, J., Sterilization with Steam under Pressure, chap- tions may require even more careful microbial control toter in Morrissey, R., & Phillips, G.B., Sterilization Technol- avoid the proliferation of microorganisms ubiquitous toogy—A Practical Guide for Manufacturers and Users of water during the purification, storage, and distribution ofHealth Care Products, Van Nostrand Reinhold, New York, this substance. A microbial specification would also be inap-1993. propriate when related to the “utility” or continuous supply

■1S (USP36) nature of this raw material. Microbial specifications are typi-cally assessed by test methods that take at least 48–72 h togenerate results. Because pharmaceutical waters are gener-ally produced by continuous processes and used in productsand manufacturing processes soon after generation, thewater is likely to have been used well before definitive testresults are available. Failure to meet a compendial specifica-

Official from December 1, 2013Copyright (c) 2014 The United States Pharmacopeial Convention. All rights reserved.

Accessed from 220.225.196.130 by anware31 on Wed Mar 05 06:55:37 EST 2014

5758 ⟨1231⟩ Water for Pharmaceutical Purposes / General Information First Supplement to USP 36–NF 31

tion would require investigating the impact and making a ing ammonia, which in turn can carry over to the finishedpass/fail decision on all product lots between the previous water. Pretreatment unit operations must be designed andsampling’s acceptable test result and a subsequent sam- operated to adequately remove the disinfectant, drinkingpling’s acceptable test result. The technical and logistical water DBPs, and objectionable disinfectant degradants. A se-problems created by a delay in the result of such an analysis rious problem can occur if unit operations designed to re-do not eliminate the user’s need for microbial specifications. move chlorine were, without warning, challenged with chlo-Therefore, such water systems need to be operated and ramine-containing drinking water from a municipality thatmaintained in a controlled manner that requires that the had been mandated to cease use of chlorine disinfection tosystem be validated to provide assurance of operational sta- comply with ever-tightening EPA Drinking Water THM speci-bility and that its microbial attributes be quantitatively fications. The dechlorination process might incompletely re-monitored against established alert and action levels that move the chloramine, which could irreparably damagewould provide an early indication of system control. The downstream unit operations, but also the release of ammo-issues of water system validation and alert/action levels and nia during this process might carry through pretreatmentspecifications are included in this chapter. and prevent the finished water from passing compendial

conductivity specifications. The purification process must bereassessed if the drinking water disinfectant is changed, em-

SOURCE OR FEED WATER CONSIDERATIONS phasizing the need for a good working relationship betweenthe pharmaceutical water manufacturer and the drinking

To ensure adherence to certain minimal chemical and mi- water provider.crobiological quality standards, water used in the produc-tion of drug substances or as source or feed water for thepreparation of the various types of purified waters must Change to read:meet the requirements of the National Primary DrinkingWater Regulations (NPDWR) (40 CFR 141) issued by theU.S. Environmental Protection Agency (EPA) or the drinking TYPES OF WATERwater regulations of the European Union or Japan, or theWHO drinking water guidelines. Limits on the types and

There are many different grades of water used for phar-quantities of certain organic and inorganic contaminants en-maceutical purposes. Several are described in USP mono-sure that the water will contain only small, safe quantities ofgraphs that specify uses, acceptable methods of preparation,potentially objectionable chemical species. Therefore, waterand quality attributes. These waters can be divided into twopretreatment systems will only be challenged to removegeneral types: bulk waters, which are typically produced onsmall quantities of these potentially difficult-to-removesite where they are used; and sterile waters, which are pro-chemicals. Also, control of objectionable chemical contami-duced, packaged, and sterilized to preserve microbial qualitynants at the source-water stage eliminates the need to spe-throughout their packaged shelf life. There are several spe-cifically test for some of them (e.g., trihalomethanes andcialized types of sterile waters, differing in their designatedheavy metals) after the water has been further purified.applications, packaging limitations, and other qualityMicrobiological requirements of drinking water ensure theattributes.absence of coliforms, which, if determined to be of fecal

There are also other types of water for which there are noorigin, may indicate the potential presence of other poten-monographs. These are all bulk waters, with names giventially pathogenic microorganisms and viruses of fecal origin.for descriptive purposes only. Many of these waters are usedMeeting these microbiological requirements does not rulein specific analytical methods. The associated text may notout the presence of other microorganisms, which could bespecify or imply certain quality attributes or modes of prep-considered undesirable if found in a drug substance or for-aration. These nonmonographed waters may not necessarilymulated product.adhere strictly to the stated or implied modes of preparationTo accomplish microbial control, municipal water authori-or attributes. Waters produced by other means or controlledties add disinfectants to drinking water. Chlorine-containingby other test attributes may equally satisfy the intended usesand other oxidizing substances have been used for manyfor these waters. It is the user’s responsibility to ensure thatdecades for this purpose and have generally been consid-such waters, even if produced and controlled exactly asered to be relatively innocuous to humans. However, thesestated, be suitable for their intended use. Wherever the termoxidants can interact with naturally occurring organic mat-“water” is used within these compendia without other de-ter to produce disinfection by-products (DBPs), such asscriptive adjectives or clauses, the intent is that water of notrihalomethanes (THMs, including chloroform, bromodichlo-less purity than Purified Water be used.romethane, and dibromochloromethane) and haloacetic

What follows is a brief description of the various types ofacids (HAAs, including dichloroacetic acid and trichloroace-pharmaceutical waters and their significant uses or attrib-tic acid). The levels of DBPs produced vary with the levelutes. Figure 1 may also be helpful in understanding some ofand type of disinfectant used and the levels and types ofthe various types of waters.organic materials found in the water, which can vary

seasonally.Because high levels of DBPs are considered a health haz- Bulk Monographed Waters and Steam

ard in drinking water, drinking water regulations mandatetheir control to generally accepted nonhazardous levels. The following waters are typically produced in large vol-However, depending on the unit operations used for further ume by a multiple-unit operation water system and distrib-water purification, a small fraction of the DBPs in the start- uted by a piping system for use at the same site. Theseing water may carry over to the finished water. Therefore, particular pharmaceutical waters must meet the quality at-the importance of having minimal levels of DBPs in the tributes as specified in the related monographs.starting water, while achieving effective disinfection, is Purified Water—Purified Water (see the USP monograph)important. is used as an excipient in the production of nonparenteralDBP levels in drinking water can be minimized by using preparations and in other pharmaceutical applications, suchdisinfectants such as ozone, chloramines, or chlorine diox- as cleaning of certain equipment and nonparenteral prod-ide. Like chlorine, their oxidative properties are sufficient to uct-contact components. Unless otherwise specified, Purifieddamage some pretreatment unit operations and must be Water is also to be used for all tests and assays for whichremoved early in the pretreatment process. The complete water is indicated (see General Notices and Requirements).removal of some of these disinfectants can be problematic. Purified Water is also referenced throughout the USP–NF. Re-For example, chloramines may degrade during the disinfec- gardless of the font and letter case used in its spelling,tion process or during pretreatment removal, thereby releas-




Figure 1. Water for pharmaceutical purposes.

water complying with the Purified Water monograph is in- The Purified Water monograph also allows bulk packagingtended. Purified Water must meet the requirements for ionic for commercial use elsewhere. In contrast to Sterile Purifiedand organic chemical purity and must be protected from Water, bulk packaged Purified Water is not required to bemicrobial contamination. The minimal quality of source or sterile. Because there is potential for microbial contamina-feed water for the production of Purified Water is Drinking tion and other quality changes in this bulk packaged non-Water. This source water may be purified using unit opera- sterile water, this form of Purified Water should be preparedtions that include deionization, distillation, ion exchange, re- and stored in a fashion that limits microbial growth and/orverse osmosis, filtration, or other suitable purification proce- is simply used in a timely fashion before microbial prolifera-dures. Purified water systems must be validated to reliably tion renders it unsuitable for its intended use. Also depend-and consistently produce and distribute water of acceptable ing on the material used for packaging, there could be ex-chemical and microbiological quality. Purified water systems tractable compounds leaching into the water from thethat function under ambient conditions are particularly sus- packaging. Although this article is required to meet theceptible to the establishment of tenacious biofilms of micro- same chemical purity limits as the bulk water, packagingorganisms, which can be the source of undesirable levels of extractables will render the packaged water less pure thanviable microorganisms or endotoxins in the effluent water. the bulk water. The nature of these impurities may evenThese systems require frequent sanitization and microbiolog- render the water an inappropriate choice for some applica-ical monitoring to ensure water of appropriate microbiologi- tions. It is the user’s responsibility to ensure fitness for use ofcal quality at the points of use. this packaged article when used in manufacturing, clinical,




or analytical applications where the pure bulk form of the quent processing step is used to remove any codepositedwater is indicated. impurity residues. These Pure Steam applications include but

are not limited to porous load sterilization processes, prod-Water for Injection—Water for Injection (see the USPuct or cleaning solutions heated by direct steam injection,monograph) is used as an excipient in the production ofor humidification of processes where steam injection is usedparenteral and other preparations where product endotoxinto control the humidity inside processing vessels where thecontent must be controlled, and in other pharmaceuticalofficial articles or their in-process forms are exposed. Theapplications, such as cleaning of certain equipment and par-primary intent of using this quality of steam is to ensureenteral product-contact components. The minimum qualitythat official articles or article-contact surfaces exposed to itof source or feed water for the generation of Water for Injec-are not contaminated by residues within the steam. tion is Drinking Water as defined by the U.S. Environmental

Pure Steam is prepared from suitably pretreated sourceProtection Agency (EPA), EU, Japan, or WHO. This sourcewater analogously to either the pretreatment used for Puri-water may be pretreated to render it suitable for subsequentfied Water or Water for Injection. The water is vaporized withdistillation (or whatever other validated process is used ac-suitable mist elimination, and distributed under pressure.cording to the monograph). The finished water must meetThe sources of undesirable contaminants within Pure Steamall of the chemical requirements for Purified Water as well ascould arise from entrained source water droplets, anticorro-an additional bacterial endotoxin specification. Because en-sion steam additives, or residues from the steam productiondotoxins are produced by the kinds of microorganisms thatand distribution system itself. The attributes in the Pureare prone to inhabit water, the equipment and proceduresSteam monograph should detect most of the contaminantsused by the system to purify, store, and distribute Water forthat could arise from these sources. If the official article ex-Injection must be designed to minimize or prevent microbialposed to potential Pure Steam residues is intended for par-contamination as well as remove incoming endotoxins fromenteral use or other applications where the pyrogenic con-the starting water. Water for Injection systems must be vali-tent must be controlled, the Pure Steam must additionallydated to reliably and consistently produce and distributemeet the specification for the Bacterial Endotoxins Test ⟨85⟩.this quality of water.

These purity attributes are measured on the condensate ofThe Water for Injection monograph also allows bulk pack-the article, rather than the article itself. This, of course, im-aging for commercial use. In contrast to Sterile Water forparts great importance to the cleanliness of the Pure SteamInjection, bulk packaged Water for Injection is not required tocondensate generation and collection process because itbe sterile. However, to preclude significant changes in itsmust not adversely impact the quality of the resulting con-microbial and endotoxins content during storage, this formdensed fluid.of Water for Injection should be prepared and stored in a

Other steam attributes not detailed in the monograph, infashion that limits microbial growth and/or is simply used inparticular, the presence of even small quantities of noncon-a timely fashion before microbial proliferation renders it un-densable gases or the existence of a superheated or drysuitable for its intended use. Also depending on the materialstate, may also be important for applications such as sterili-used for packaging, there could be extractable compoundszation. The large release of energy (latent heat of condensa-leaching into the water from the packaging. Although thistion) as water changes from the gaseous to the liquid statearticle is required to meet the same chemical purity limits asis the key to steam’s sterilization efficacy and its efficiency,the bulk water, packaging extractables will render the pack-in general, as a heat transfer agent. If this phase changeaged water less pure than the bulk water. The nature of(condensation) is not allowed to happen because the steamthese impurities may even render the water an inappropri-is extremely hot and in a persistent superheated, dry state,ate choice for some applications. It is the user’s responsibil-then its usefulness could be seriously compromised. Non-ity to ensure fitness for use of this packaged article whencondensable gases in steam tend to stratify or collect in cer-used in manufacturing, clinical, or analytical applicationstain areas of a steam sterilization chamber or its load. Thesewhere the purer bulk form of the water is indicated.surfaces would thereby be at least partially insulated fromWater for Hemodialysis—Water for Hemodialysis (see the the steam condensation phenomenon, preventing themUSP monograph) is used for hemodialysis applications, pri- from experiencing the full energy of the sterilizing condi-marily the dilution of hemodialysis concentrate solutions. It tions. Therefore, control of these kinds of steam attributes,is produced and used on site and is made from EPA Drink- in addition to its chemical purity, may also be important foring Water that has been further purified to reduce chemical certain Pure Steam applications. However, because these ad-and microbiological components. It may be packaged and ditional attributes are use-specific, they are not mentionedstored in unreactive containers that preclude bacterial entry. in the Pure Steam monograph.The term “unreactive containers” implies that the container, Note that less pure “plant steam” may be used for steamespecially its water contact surfaces, is not changed in any sterilization of nonproduct contact nonporous loads, forway by the water, such as by leaching of container-related general cleaning of nonproduct contact equipment, as acompounds into the water or by any chemical reaction or nonproduct contact heat exchange medium, and in all com-corrosion caused by the water. The water contains no patible applications involved in bulk pharmaceutical chemi-added antimicrobials and is not intended for injection. Its cal and API manufacture.attributes include specifications for water conductivity, total Finally, owing to the lethal properties of Pure Steam, mon-organic carbon (or oxidizable substances), microbial limits, itoring of microbial control within a steam system is unnec-and bacterial endotoxins. The water conductivity and total essary. Therefore, microbial analysis of the steam condensateorganic carbon attributes are identical to those established is unnecessary.for Purified Water and Water for Injection; however, instead of

total organic carbon (TOC), the organic content may alter-natively be measured by the test for Oxidizable Substances. Sterile Monographed WatersThe microbial limits attribute for this water is unique amongthe “bulk” water monographs, but is justified on the basis The following monographed waters are packaged formsof this water’s specific application that has microbial content of either Purified Water or Water for Injection that have beenrequirements related to its safe use. The bacterial endotoxins sterilized to preserve their microbiological properties. Theseattribute is likewise established at a level related to its safe waters may have specific intended uses as indicated by theiruse. names and may also have restrictions on packaging configu-

rations related to those uses. In general, these sterile pack-Pure Steam—Pure Steam (see the USP monograph) isaged waters may be used in a variety of applications in lieualso sometimes referred to as “clean steam”. It is usedof the bulk forms of water from which they were derived.where the steam or its condensate would directly contactHowever, there is a marked contrast between the qualityofficial articles or article-contact surfaces, such as duringtests and purities for these bulk versus sterile packaged wa-their preparation, sterilization, or cleaning where no subse-




ters. These quality tests and specifications for sterile pack- lowing is a description of several of these nonmonographedaged waters have diverged from those of bulk waters to waters as cited in various locations within these compendia.accommodate a wide variety of packaging types, properties, Drinking Water—This type of water can be referred tovolumes, and uses. As a result, the inorganic and organic as Potable Water (meaning drinkable or fit to drink), Na-impurity specifications and levels of the bulk and sterile tional Primary Drinking Water, Primary Drinking Water, orpackaged forms of water are not equivalent as their name National Drinking Water. Except where a singular drinkingsimilarities imply. The packaging materials and elastomeric water specification is stated (such as the NPDWR [U.S. Envi-closures are the primary sources of these impurities, which ronmental Protection Agency’s National Primary Drinkingtend to increase over these packaged articles’ shelf lives. Water Regulations as cited in 40 CFR Part 141]), this waterTherefore, due consideration must be given to the chemical must comply with the quality attributes of either thepurity suitability at the time of use of the sterile packaged NPDWR, or the drinking water regulations of the Europeanforms of water when used in manufacturing, analytical, and Union or Japan, or the WHO Drinking Water Guidelines. Itcleaning applications in lieu of the bulk waters from which may be derived from a variety of sources including a publicthese waters were derived. It is the user’s responsibility to water utility, a private water supply (e.g., a well), or a com-ensure fitness for use of these sterile packaged waters in bination of these sources. Drinking Water may be used in thethese applications. Nevertheless, for the applications dis- early stages of cleaning pharmaceutical manufacturingcussed below for each sterile packaged water, their respec- equipment and product-contact components. Drinkingtive purities and packaging restrictions generally render Water is also the minimum quality of water that should bethem suitable by definition. used for the preparation of official substances and other

Sterile Purified Water—Sterile Purified Water (see the USP bulk pharmaceutical ingredients. Where compatible with themonograph) is Purified Water, packaged and rendered ster- processes, the allowed contaminant levels in Drinking Waterile. It is used in the preparation of nonparenteral com- are generally considered safe for use for official substancespendial dosage forms or in analytical applications requiring and other drug substances. Where required by the process-Purified Water where access to a validated Purified Water sys- ing of the materials to achieve their required final purity,tem is not practical, where only a relatively small quantity is higher qualities of water may be needed for these manufac-needed, where Sterile Purified Water is required, or where turing steps, perhaps even as pure as Water for Injection orbulk packaged Purified Water is not suitably microbiologically Purified Water. Such higher purity waters, however, mightcontrolled. require only selected attributes to be of higher purity than

Drinking Water (see Figure 2). Drinking Water is the pre-Sterile Water for Injection—Sterile Water for Injectionscribed source or feed water for the production of bulk(see the USP monograph) is Water for Injection packagedmonographed pharmaceutical waters. The use of Drinkingand rendered sterile. It is used for extemporaneous prescrip-Water specifications establishes a reasonable set of maxi-tion compounding and as a sterile diluent for parenteralmum allowable levels of chemical and microbiological con-products. It may also be used for other applications wheretaminants with which a water purification system will bebulk Water for Injection or Purified Water is indicated butchallenged. As seasonal variations in the quality attributes ofwhere access to a validated water system is either not prac-the Drinking Water supply can occur, due consideration totical or where only a relatively small quantity is needed.its synthetic and cleaning uses must be given. The process-Sterile Water for Injection is packaged in single-dose contain-ing steps in the production of pharmaceutical waters musters not larger than 1 L in size.be designed to accommodate this variability.Bacteriostatic Water for Injection—Bacteriostatic Water

Hot Purified Water—This water is used in the prepara-for Injection (see the USP monograph) is sterile Water fortion instructions for USP–NF articles and is clearly intendedInjection to which has been added one or more suitable an-to be Purified Water that has been heated to an unspecifiedtimicrobial preservatives. It is intended to be used as a dilu-temperature in order to enhance solubilization of other in-ent in the preparation of parenteral products, most typicallygredients. There is no upper temperature limit for the waterfor multi-dose products that require repeated content with-(other than being less than 100°), but for each monographdrawals. It may be packaged in single-dose or multiple-dosethere is an implied lower limit below which the desiredcontainers not larger than 30 mL.solubilization effect would not occur.Sterile Water for Irrigation—Sterile Water for Irrigation

(see the USP monograph) is Water for Injection packagedand sterilized in single-dose containers of larger than 1 L in Nonmonographed Analytical Waterssize that allows rapid delivery of its contents. It need notmeet the requirement under small-volume injections in the Both General Notices and Requirements and the introduc-general test chapter Particulate Matter in Injections ⟨788⟩. It tory section to Reagents, Indicators, and Solutions clearlymay also be used in other applications that do not have state that where the term “water”, without qualification orparticulate matter specifications, where bulk Water for Injec- other specification, is indicated for use in analyses, the qual-tion or Purified Water is indicated but where access to a vali- ity of water shall be Purified Water. However, numerous suchdated water system is not practical, or where somewhat qualifications do exist. Some of these qualifications involvelarger quantities than are provided as Sterile Water for Injec- methods of preparation, ranging from specifying the pri-tion are needed. mary purification step to specifying additional purification.

Other qualifications call for specific attributes to be met thatSterile Water for Inhalation—Sterile Water for Inhalationmight otherwise interfere with analytical processes. In most(see the USP monograph) is Water for Injection that is pack-of these latter cases, the required attribute is not specificallyaged and rendered sterile and is intended for use in inhala-tested. Rather, a further “purification process” is specifiedtors and in the preparation of inhalation solutions. It carriesthat ostensibly allows the water to adequately meet this re-a less stringent specification for bacterial endotoxins thanquired attribute.Sterile Water for Injection and therefore is not suitable for

However, preparation instructions for many reagents wereparenteral applications.carried forward from the innovator’s laboratories to the orig-inally introduced monograph for a particular USP–NF articleNonmonographed Manufacturing Waters or general test chapter. The quality of the reagent waterdescribed in these tests may reflect the water quality desig-

In addition to the bulk monographed waters described nation of the innovator’s laboratory. These specific waterabove, nonmonographed waters can also be used in phar- designations may have originated without the innovator’smaceutical processing steps such as cleaning, synthetic awareness of the requirement for Purified Water in USP–NFsteps, or a starting material for further purification. The fol- tests. Regardless of the original reason for the creation of

these numerous special analytical waters, it is possible that




Figure 2. Selection of water for pharmaceutical purposes.

the attributes of these special waters could now be met by cited uses of this water imply a need for a particular puritythe basic preparation steps and current specifications of Pu- attribute that can only be derived by distillation, waterrified Water. In some cases, however, some of the cited post- meeting the requirements for Purified Water derived by otherprocessing steps are still necessary to reliably achieve the means of purification could be equally suitable where Dis-required attributes. tilled Water is specified.

Users are not obligated to employ specific and perhaps Freshly Distilled Water—Also called “recently distilledarchaically generated forms of analytical water where alter- water”, it is produced in a similar fashion to Distilled Waternatives with equal or better quality, availability, or analytical and should be used shortly after its generation. This impliesperformance may exist. The consistency and reliability for the need to avoid endotoxin contamination as well as anyproducing these alternative analytical waters should be veri- other adventitious forms of contamination from the air orfied as producing the desired attributes. In addition, any containers that could arise with prolonged storage. It is usedalternative analytical water must be evaluated on an applica- for preparing solutions for subcutaneous test animal injec-tion-by-application basis by the user to ensure its suitability. tions as well as for a reagent solvent in tests for which thereFollowing is a summary of the various types of appears to be no particularly high water purity needed thatnonmonographed analytical waters that are cited in the could be ascribable to being “freshly distilled”. In the testUSP–NF. animal use, the term “freshly distilled” and its testing use

Distilled Water—This water is produced by vaporizing imply a chemical, endotoxin, and microbiological purity thatliquid water and condensing it in a purer state. It is used could be equally satisfied by Water for Injection (although noprimarily as a solvent for reagent preparation, but it is also reference is made to these chemical, endotoxin, or microbialspecified in the execution of other aspects of tests, such as attributes or specific protection from recontamination). Forfor rinsing an analyte, transferring a test material as a slurry, nonanimal uses, water meeting the requirements for Purifiedas a calibration standard or analytical blank, and for test Water derived by other means of purification and/or storageapparatus cleaning. It is also cited as the starting water to periods could be equally suitable where “recently distilledbe used for making High-Purity Water. Because none of the water” or Freshly Distilled Water is specified.




Deionized Water—This water is produced by an ion-ex- Carbon Dioxide-Free Water—The introductory portionchange process in which the contaminating ions are re- of the Reagents, Indicators, and Solutions section defines thisplaced with either H+ or OH– ions. Similarly to Distilled water as Purified Water that has been vigorously boiled for atWater, Deionized Water is used primarily as a solvent for rea- least 5 minutes, then cooled and protected from absorptiongent preparation, but it is also specified in the execution of of atmospheric carbon dioxide. Because the absorption ofother aspects of tests, such as for transferring an analyte carbon dioxide tends to drive down the water pH, most ofwithin a test procedure, as a calibration standard or analyti- the uses of Carbon Dioxide-Free Water are either associatedcal blank, and for test apparatus cleaning. Also, none of the as a solvent in pH-related or pH-sensitive determinations orcited uses of this water imply any needed purity attribute as a solvent in carbonate-sensitive reagents or determina-that can only be achieved by deionization. Therefore, water tions. Another use of this water is for certain optical rotationmeeting the requirements for Purified Water that is derived and color and clarity of solution tests. Although it is possibleby other means of purification could be equally suitable that this water is indicated for these tests simply because ofwhere Deionized Water is specified. its purity, it is also possible that the pH effects of carbon

dioxide-containing water could interfere with the results ofFreshly Deionized Water—This water is prepared in athese tests. A third plausible reason that this water is indi-similar fashion to Deionized Water, although as the namecated is that outgassing air bubbles might interfere withsuggests, it is to be used shortly after its production. Thisthese photometric-type tests. The boiled water preparationimplies the need to avoid any adventitious contaminationapproach will also greatly reduce the concentrations ofthat could occur upon storage. This water is indicated formany other dissolved gases along with carbon dioxide.use as a reagent solvent as well as for cleaning. Due to theTherefore, in some of the applications for Carbon Dioxide-nature of the testing, Purified Water could be a reasonableFree Water, it could be the inadvertent deaeration effect thatalternative for these applications.actually renders this water suitable. In addition to boiling,Deionized Distilled Water—This water is produced by deionization is perhaps an even more efficient process fordeionizing (see Deionized Water) Distilled Water. This water is removing dissolved carbon dioxide (by drawing the dis-used as a reagent in a liquid chromatography test that re- solved gas equilibrium toward the ionized state with subse-quires a high purity. Because of the importance of this high quent removal by the ion-exchange resins). If the startingpurity, water that barely meets the requirements for Purified Purified Water is prepared by an efficient deionization pro-Water may not be acceptable. High-Purity Water (see below) cess and protected after deionization from exposure to at-could be a reasonable alternative for this water. mospheric air, water that is carbon dioxide-free can be ef-

Filtered Water—This water is Purified Water that has fectively made without the application of heat. However,been filtered to remove particles that could interfere with this deionization process does not deaerate the water, so ifthe analysis where this water is specified. It is sometimes Purified Water prepared by deionization is considered as aused synonymously with Particle-Free Water and Ultra-Filtered substitute water in a test requiring Carbon Dioxide-FreeWater and is cited in some monographs and general chap- Water, the user must verify that it is not actually water akinters as well as in Reagents. Depending on its location, it is to Deaerated Water (discussed below) that is needed for thevariously defined as water that has been passed through test. As indicated in High-Purity Water, even brief contactfilters rated as 1.2-µm, 0.22-µm, or 0.2-µm; or unspecified with the atmosphere can allow small amounts of carbonpore size. Although the water names and the filter pore dioxide to dissolve, ionize, and significantly degrade thesizes used to produce these waters are inconsistently de- conductivity and lower the pH. If the analytical use requiresfined, the use of 0.2-µm pore size filtered Purified Water the water to remain as pH-neutral and as carbon dioxide-should be universally acceptable for all applications where free as possible, even the analysis should be protected fromParticle-Free Water, Filtered Water, or Ultra-Filtered Water are atmospheric exposure. However, in most applications, at-specified. mospheric exposure during testing does not significantly af-

■High-Purity Water—This water may be prepared by fect its suitability in the test.deionizing previously distilled water, and then filtering it Ammonia- and Carbon Dioxide-Free Water—As impliedthrough a 0.45-µm rated membrane. This water must have by the name, this water should be prepared by approachesan in-line conductivity of not greater than 0.15 µS/cm (not compatible with those mentioned for both Ammonia-Freeless than 6.67 Megohm-cm) at 25°. If the water of this pu- Water and Carbon Dioxide-Free Water. Because the carbonrity contacts the atmosphere even briefly as it is being used dioxide-free attribute requires post-production protectionor drawn from its purification system, its conductivity will from the atmosphere, it is appropriate to first render theimmediately increase, by as much as about 1.0 µS/cm, as water ammonia-free using the High-Purity Water process fol-atmospheric carbon dioxide dissolves in the water and equil- lowed by the boiling and carbon dioxide-protected coolingibrates to hydrogen and bicarbonate ions. Therefore, if the process. The High-Purity Water deionization process for cre-analytical use requires that water conductivity remains as ating Ammonia-Free Water will also remove the ions gener-low as possible or the bicarbonate/carbon dioxide levels be ated from dissolved carbon dioxide and ultimately, byas low as possible, its use should be protected from atmos- forced equilibration to the ionized state, all the dissolvedpheric exposure. This water is used as a reagent, as a sol- carbon dioxide. Therefore, depending on its use, an accept-vent for reagent preparation, and for test apparatus cleaning able procedure for making Ammonia- and Carbon Dioxide-where less pure waters would not perform acceptably. How- Free Water could be to transfer and collect High-Purity Waterever, if a user’s routinely available Purified Water is filtered in a carbon dioxide intrusion-protected container.and meets or exceeds the conductivity specifications of Deaerated Water—This water is Purified Water that hasHigh-Purity Water, it could be used in lieu of High-Purity been treated to reduce the content of dissolved air by “suit-Water.■1S (USP36) able means”. In the Reagents section, approaches for boil-

Ammonia-Free Water—Functionally, this water must ing, cooling (similar to Carbon Dioxide-Free Water but with-have a negligible ammonia concentration to avoid interfer- out the atmospheric carbon dioxide protection), andence in tests sensitive to ammonia. It has been equated with sonication are given as applicable for test uses other thanHigh-Purity Water that has a significantly tighter Stage 1 (see dissolution and drug release testing. Although DeaeratedWater Conductivity ⟨645⟩) conductivity specification than Pu- Water is not mentioned by name in Dissolution ⟨711⟩, sug-rified Water because of the latter’s allowance for a minimal gested methods for deaerating dissolution media (whichlevel of ammonium among other ions. However, if the may be water) include warming to 41°, vacuum filteringuser’s Purified Water were filtered and met or exceeded the through a 0.45-µm rated membrane, and vigorously stirringconductivity specifications of High-Purity Water, it would the filtrate while maintaining the vacuum. This chapter spe-contain negligible ammonia or other ions and could be cifically indicates that other validated approaches may beused in lieu of High-Purity Water. used. In other monographs that also do not mention Deaer-




ated Water by name, degassing of water and other reagents ograph is the temperature of “hot” water specified; so in allis accomplished by sparging with helium. Deaerated Water is the other cases, the water temperature is less important, butused in both dissolution testing as well as liquid chromatog- should be high enough to achieve the desirable effect. In allraphy applications where outgassing could either interfere cases, the chemical quality of the water is implied to be thatwith the analysis itself or cause erroneous results due to in- of Purified Water.accurate volumetric withdrawals. Applications where ambi-ent temperature water is used for reagent preparation, but

VALIDATION AND QUALIFICATION OFthe tests are performed at elevated temperatures, are candi-dates for outgassing effects. If outgassing could interfere WATER PURIFICATION, STORAGE, ANDwith test performance, including chromatographic flow, col- DISTRIBUTION SYSTEMSorimetric or photometric measurements, or volumetric accu-racy, then Deaerated Water should probably be used, Establishing the dependability of pharmaceutical waterwhether called for in the analysis or not. The above deaera- purification, storage, and distribution systems requires antion approaches might not render the water “gas-free”. At appropriate period of monitoring and observation. Ordinar-best, they reduce the dissolved gas concentrations so that ily, few problems are encountered in maintaining the chem-outgassing caused by temperature changes is not likely. ical purity of Purified Water and Water for Injection. Neverthe-

Recently Boiled Water—This water may include recently less, the advent of using conductivity and TOC to defineor freshly boiled water (with or without mention of cooling chemical purity has allowed the user to more quantitativelyin the title), but cooling prior to use is clearly intended. assess the water’s chemical purity and its variability as aOccasionally it is necessary to use when hot. Recently Boiled function of routine pretreatment system maintenance andWater is specified because it is used in a pH-related test or regeneration. Even the presence of such unit operations ascarbonate-sensitive reagent, in an oxygen-sensitive test or heat exchangers and use point hoses can compromise thereagent, or in a test where outgassing could interfere with chemical quality of water within and delivered from an oth-the analysis, such as specific gravity or an appearance test. erwise well-controlled water system. Therefore, an assess-

ment of the consistency of the water’s chemical purity overOxygen-Free Water—The preparation of this water istime must be part of the validation program. However, evennot specifically described in the compendia. Neither is therewith the most well controlled chemical quality, it is oftenan oxygen specification or analysis mentioned. However, allmore difficult to consistently meet established microbiologi-uses involve analyses of materials that could be sensitive tocal quality criteria owing to phenomena occurring duringoxidation by atmospheric oxygen. Procedures for the re-and after chemical purification. A typical program involvesmoval of dissolved oxygen from solvents, although not nec-intensive daily sampling and testing of major process pointsessarily water, are mentioned in Polarography ⟨801⟩ andfor at least one month after operational criteria have beenSpectrophotometry and Light-Scattering ⟨851⟩. These proce-established for each unit operation, point of use, and sam-dures involve simple sparging of the liquid with an inert gaspling point.such as nitrogen or helium, followed by inert gas blanketing

An overlooked aspect of water system validation is theto prevent oxygen reabsorption. The sparging times citeddelivery of the water to its actual location of use. If thisrange from 5 to 15 minutes to an unspecified period. Sometransfer process from the distribution system outlets to thePurified Water and Water for Injection systems produce waterwater use locations (usually with hoses) is defined as outsidethat is maintained in a hot state and that is inert gas blan-the water system, then this transfer process still needs to beketed during its preparation and storage and distribution.validated to not adversely affect the quality of the water toAlthough oxygen is poorly soluble in hot water, such waterthe extent it becomes unfit for use. Because routine micro-may not be oxygen-free. Whatever procedure is used forbial monitoring is performed for the same transfer processremoving oxygen should be verified as reliably producingand components (e.g., hoses and heat exchangers) as thatwater that is fit for use.of routine water use (see Sampling Considerations), there isWater for BET—This water is also referred to as LAL rea- some logic to including this water transfer process withingent water. This is often Water for Injection, which may have the distribution system validation.been sterilized. It is free from a level of endotoxin that Validation is the process whereby substantiation to a highwould yield any detectable reaction or interference with the level of assurance that a specific process will consistentlyLimulus Amoebocyte Lysate reagent used in the Bacterial En- produce a product conforming to an established set of qual-dotoxins Test ⟨85⟩. ity attributes is acquired and documented. Prior to and dur-

Organic-Free Water—This water is defined by Residual ing the very early stages of validation, the critical processSolvents ⟨467⟩ as producing no significantly interfering gas parameters and their operating ranges are established. Achromatography peaks. Referenced monographs specify us- validation program qualifies and documents the design, in-ing this water as the solvent for the preparation of standard stallation, operation, and performance of equipment. It be-and test solutions for the Residual solvents test. gins when the system is defined and moves through several

Lead-Free Water—This water is used as a transferring stages: installation qualification (IQ), operational qualifica-diluent for an analyte in a Lead ⟨251⟩ test. Although no tion (OQ), and performance qualification (PQ). A graphicalspecific instructions are given for its preparation, it must not representation of a typical water system validation life cyclecontain any detectable lead. Purified Water should be a suit- is shown in Figure 3.able substitute for this water. A validation plan for a water system typically includes the

following steps: (1) establishing standards for quality attrib-Chloride-Free Water—This water is specified as the sol-utes of the finished water and the source water; (2) definingvent for use in an assay that contains a reactant that preci-suitable unit operations and their operating parameters forpitates in the presence of chloride. Although no specificachieving the desired finished water quality attributes frompreparation instructions are given for this water, its ratherthe available source water; (3) selecting piping, equipment,obvious attribute is having a very low chloride level in ordercontrols, and monitoring technologies; (4) developing an IQto be unreactive with this chloride sensitive reactant. Purifiedstage consisting of instrument calibrations, inspections toWater could be used for this water but should be tested toverify that the drawings accurately depict the final configur-ensure that it is unreactive.ation of the water system and, where necessary, special testsHot Water—The uses of this water include solvents forto verify that the installation meets the design requirements;achieving or enhancing reagent solubilization, restoring the(5) developing an OQ stage consisting of tests and inspec-original volume of boiled or hot solutions, rinsing insolubletions to verify that the equipment, system alerts, and con-analytes free of hot water soluble impurities, solvents fortrols are operating reliably and that appropriate alert andreagent recrystallization, apparatus cleaning, and as a solu-action levels are established (this phase of qualification maybility attribute for various USP–NF articles. In only one mon-




Figure 3. Water system validation life cycle.

overlap with aspects of the next step); and (6) developing a ity attributes of both waters differ only in the presence of aprospective PQ stage to confirm the appropriateness of criti- bacterial endotoxin requirement for Water for Injection andcal process parameter operating ranges (during this phase in their methods of preparation, at least at the last stage ofof validation, alert and action levels for key quality attributes preparation. The similarities in the quality attributes provideand operating parameters are verified); (7) assuring the ade- considerable common ground in the design of water sys-quacy of ongoing control procedures, e.g., sanitization fre- tems to meet either requirement. The critical difference isquency; (8) supplementing a validation maintenance pro- the degree of control of the system and the final purificationgram (also called continuous validation life cycle) that steps needed to ensure bacterial and bacterial endotoxinincludes a mechanism to control changes to the water sys- removal.tem and establishes and carries out scheduled preventive Production of pharmaceutical water employs sequentialmaintenance including recalibration of instruments (in addi- unit operations (processing steps) that address specific watertion, validation maintenance includes a monitoring program quality attributes and protect the operation of subsequentfor critical process parameters and a corrective action pro- treatment steps. A typical evaluation process to select angram); (9) instituting a schedule for periodic review of the appropriate water quality for a particular pharmaceuticalsystem performance and requalification; and (10) complet- purpose is shown in the decision tree in Figure 2. This dia-ing protocols and documenting Steps 1 through 9. gram may be used to assist in defining requirements for

specific water uses and in the selection of unit operations.The final unit operation used to produce Water for Injection

PURIFIED WATER AND WATER FOR is limited to distillation or other processes equivalent or su-INJECTION SYSTEMS perior to distillation in the removal of chemical impurities as

well as microorganisms and their components. DistillationThe design, installation, and operation of systems to pro- has a long history of reliable performance and can be vali-

duce Purified Water and Water for Injection include similar dated as a unit operation for the production of Water forcomponents, control techniques, and procedures. The qual- Injection, but other technologies or combinations of technol-




ogies can be validated as being equivalently effective. Other sizing to minimize excessively frequent or infrequent back-technologies, such as ultrafiltration following another chemi- washing or cartridge filter replacement.cal purification process, may be suitable in the productionof Water for Injection if they can be shown through valida- Activated Carbontion to be as effective and reliable as distillation. The adventof new materials for older technologies, such as reverse os- Granular activated carbon beds adsorb low molecularmosis and ultrafiltration, that allow intermittent or continu- weight organic material and oxidizing additives, such asous operation at elevated, microbial temperatures, show chlorine and chloramine compounds, removing them frompromise for a valid use in producing Water for Injection. the water. They are used to achieve certain quality attributesThe validation plan should be designed to establish the and to protect against reaction with downstream stainlesssuitability of the system and to provide a thorough under- steel surfaces, resins, and membranes. The chief operatingstanding of the purification mechanism, range of operating concerns regarding activated carbon beds include the pro-conditions, required pretreatment, and the most likely pensity to support bacteria growth, the potential for hydrau-modes of failure. It is also necessary to demonstrate the ef- lic channeling, the organic adsorption capacity, appropriatefectiveness of the monitoring scheme and to establish the water flow rates and contact time, the inability to bedocumentation and qualification requirements for the sys- regenerated in situ, and the shedding of bacteria, endotox-tem’s validation maintenance. Trials conducted in a pilot in- ins, organic chemicals, and fine carbon particles. Controlstallation can be valuable in defining the operating parame- measures may involve monitoring water flow rates and dif-ters and the expected water quality and in identifying failure ferential pressures, sanitizing with hot water or steam, back-modes. However, qualification of the specific unit operation washing, testing for adsorption capacity, and frequent re-can only be performed as part of the validation of the in- placement of the carbon bed. If the activated carbon bed isstalled operational system. The selection of specific unit op- intended for organic reduction, it may also be appropriateerations and design characteristics for a water system should to monitor influent and effluent TOC. It is important to notetake into account the quality of the feed water, the technol- that the use of steam for carbon bed sanitization is oftenogy chosen for subsequent processing steps, the extent and incompletely effective due to steam channeling rather thancomplexity of the water distribution system, and the appro- even permeation through the bed. This phenomenon canpriate compendial requirements. For example, in the design usually be avoided by using hot water sanitization. It is alsoof a system for Water for Injection, the final process (distilla- important to note that microbial biofilm development ontion or whatever other validated process is used according the surface of the granular carbon particles (as well as onto the monograph) must have effective bacterial endotoxin other particles such as found in deionizer beds and evenreduction capability and must be validated. multimedia beds) can cause adjacent bed granules to “stick”

together. When large masses of granules are agglomeratedin this fashion, normal backwashing and bed fluidizationUNIT OPERATIONS CONCERNSflow parameters may not be sufficient to disperse them,leading to ineffective removal of trapped debris, loose bi-The following is a brief description of selected unit opera-ofilm, and penetration of microbial controlling conditionstions and the operation and validation concerns associated(as well as regenerant chemicals as in the case of agglomer-with them. Not all unit operations are discussed, nor are allated deionizer resins). Alternative technologies to activatedpotential problems addressed. The purpose is to highlightcarbon beds can be used in order to avoid their microbialissues that focus on the design, installation, operation, main-problems, such as disinfectant-neutralizing chemical addi-tenance, and monitoring parameters that facilitate watertives and regenerable organic scavenging devices. However,system validation.these alternatives do not function by the same mechanismsas activated carbon, may not be as effective at removing

Prefiltration disinfectants and some organics, and have a different set ofoperating concerns and control measures that may be

The purpose of prefiltration—also referred to as initial, nearly as troublesome as activated carbon beds.coarse, or depth filtration—is to remove solid contaminantsdown to a size of 7–10 µm from the incoming source water

Additivessupply and protect downstream system components fromparticulates that can inhibit equipment performance and

Chemical additives are used in water systems (a) to con-shorten its effective life. This coarse filtration technologytrol microorganisms by use of sanitants such as chlorineutilizes primarily sieving effects for particle capture and acompounds and ozone, (b) to enhance the removal of sus-depth of filtration medium that has a high “dirt load” ca-pended solids by use of flocculating agents, (c) to removepacity. Such filtration units are available in a wide range ofchlorine compounds, (d) to avoid scaling on reverse osmosisdesigns and for various applications. Removal efficienciesmembranes, and (e) to adjust pH for more effective removaland capacities differ significantly, from granular bed filtersof carbonate and ammonia compounds by reverse osmosis.such as multimedia or sand for larger water systems, toThese additives do not constitute “added substances” asdepth cartridges for smaller water systems. Unit and systemlong as they are either removed by subsequent processingconfigurations vary widely in type of filtering media and lo-steps or are otherwise absent from the finished water. Con-cation in the process. Granular or cartridge prefilters are of-trol of additives to ensure a continuously effective concen-ten situated at or near the head of the water pretreatmenttration and subsequent monitoring to ensure their removalsystem prior to unit operations designed to remove theshould be designed into the system and included in thesource water disinfectants. This location, however, does notmonitoring program.preclude the need for periodic microbial control because bi-

ofilm can still proliferate, although at a slower rate in thepresence of source water disinfectants. Design and opera- Organic Scavengerstional issues that may impact performance of depth filtersinclude channeling of the filtering media, blockage from silt, Organic scavenging devices use macroreticular weakly ba-microbial growth, and filtering-media loss during improper sic anion-exchange resins capable of removing organic ma-backwashing. Control measures involve pressure and flow terial and endotoxins from the water. They can be regener-monitoring during use and backwashing, sanitizing, and re- ated with appropriate biocidal caustic brine solutions.placing filtering media. An important design concern is siz- Operating concerns are associated with organic scavenginging of the filter to prevent channeling or media loss result- capacity; particulate, chemical and microbiological foulinging from inappropriate water flow rates as well as proper of the reactive resin surface; flow rate; regeneration fre-




quency; and shedding of resin fragments. Control measures hydrogen and hydroxide ions. This permits continuous re-include TOC testing of influent and effluent, backwashing, generation of the resin without the need for regenerant ad-monitoring hydraulic performance, and using downstream ditives. However, unlike conventional deionization, CEDIfilters to remove resin fines. units must start with water that is already partially purified

because they generally cannot produce Purified Water qualitywhen starting with the heavier ion load of unpurified sourceSofteners water.

Concerns for all forms of deionization units include micro-Water softeners may be located either upstream or down- bial and endotoxin control, chemical additive impact onstream of disinfectant removal units. They utilize sodium- resins and membranes, and loss, degradation, and fouling ofbased cation-exchange resins to remove water-hardness resin. Issues of concern specific to DI units include regenera-ions, such as calcium and magnesium, that could foul or tion frequency and completeness, channeling caused by bi-interfere with the performance of downstream processing ofilm agglomeration of resin particles, organic leaching fromequipment such as reverse osmosis membranes, deionization new resins, complete resin separation for mixed bed regen-devices, and distillation units. Water softeners can also be eration, and mixing air contamination (mixed beds). Controlused to remove other lower affinity cations, such as the am- measures vary but typically include recirculation loops, efflu-monium ion, that may be released from chloramine disinfec- ent microbial control by UV light, conductivity monitoring,tants commonly used in drinking water and which might resin testing, microporous filtration of mixing air, microbialotherwise carryover through other downstream unit opera- monitoring, frequent regeneration to minimize and controltions. If ammonium removal is one of its purposes, the sof- microorganism growth, sizing the equipment for suitabletener must be located downstream of the disinfectant re- water flow and contact time, and use of elevated tempera-moval operation, which itself may liberate ammonium from tures. Internal distributor and regeneration piping for mixedneutralized chloramine disinfectants. Water softener resin bed units should be configured to ensure that regenerationbeds are regenerated with concentrated sodium chloride so- chemicals contact all internal bed and piping surfaces andlution (brine). Concerns include microorganism proliferation, resins. Rechargeable canisters can be the source of contami-channeling caused by biofilm agglomeration of resin parti- nation and should be carefully monitored. Full knowledge ofcles, appropriate water flow rates and contact time, ion-ex- previous resin use, minimum storage time between regener-change capacity, organic and particulate resin fouling, or- ation and use, and appropriate sanitizing procedures areganic leaching from new resins, fracture of the resin beads, critical factors ensuring proper performance.resin degradation by excessively chlorinated water, and con-tamination from the brine solution used for regeneration.Control measures involve recirculation of water during peri- Reverse Osmosisods of low water use, periodic sanitization of the resin andbrine system, use of microbial control devices (e.g., UV light Reverse osmosis (RO) units employ semipermeable mem-and chlorine), locating the unit upstream of the disinfectant branes. The “pores” of RO membranes are actually interseg-removal step (if used only for softening), appropriate regen- mental spaces among the polymer molecules. They are bigeration frequency, effluent chemical monitoring (e.g., hard- enough for permeation of water molecules, but too small toness ions and possibly ammonium), and downstream filtra- permit passage of hydrated chemical ions. However, manytion to remove resin fines. If a softener is used for factors including pH, temperature, and differential pressureammonium removal from chloramine-containing source across the membrane affect the selectivity of this perme-water, then capacity, contact time, resin surface fouling, pH, ation. With the proper controls, RO membranes can achieveand regeneration frequency are very important. chemical, microbial, and endotoxin quality improvement.

The process streams consist of supply water, product water(permeate), and wastewater (reject). Depending on sourceDeionization water, pretreatment and system configuration variations andchemical additives may be necessary to achieve desired per-Deionization (DI), and continuous electrodeionization formance and reliability.(CEDI) are effective methods of improving the chemical A major factor affecting RO performance is the permeatequality attributes of water by removing cations and anions. recovery rate, that is, the amount of the water passingDI systems have charged resins that require periodic regen- through the membrane compared to the amount rejected.eration with an acid and base. Typically, cationic resins are This is influenced by the several factors, but most signifi-regenerated with either hydrochloric or sulfuric acid, which cantly by the pump pressure. Recoveries of 75% are typical,replace the captured positive ions with hydrogen ions. An- and can accomplish a 1–2 log purification of most impuri-ionic resins are regenerated with sodium or potassium hy- ties. For most feed waters, this is usually not enough todroxide, which replace captured negative ions with hydrox- meet Purified Water conductivity specifications. A secondide ions. Because free endotoxin is negatively charged, there pass of this permeate water through another RO stage usu-is some removal of endotoxin achieved by the anionic resin. ally achieves the necessary permeate purity if other factorsBoth regenerant chemicals are biocidal and offer a measure such as pH and temperature have been appropriately ad-of microbial control. The system can be designed so that justed and the ammonia from chloraminated source waterthe cation and anion resins are in separate or “twin” beds has been previously removed. Increasing recoveries withor they can be mixed together to form a mixed bed. Twin higher pressures in order to reduce the volume of rejectbeds are easily regenerated but deionize water less effi- water will lead to reduced permeate purity. If increasedciently than mixed beds, which have a considerably more pressures are needed over time to achieve the same perme-complex regeneration process. Rechargeable resin canisters ate flow, this is an indication of partial membrane blockagecan also be used for this purpose. that needs to be corrected before it becomes irreversiblyThe CEDI system uses a combination of mixed resin, se- fouled, and expensive membrane replacement is the onlylectively permeable membranes, and an electric charge, pro- option.viding continuous flow (product and waste concentrate) Other concerns associated with the design and operationand continuous regeneration. Water enters both the resin of RO units include membrane materials that are extremelysection and the waste (concentrate) section. As it passes sensitive to sanitizing agents and to particulate, chemical,through the resin, it is deionized to become product water. and microbial membrane fouling; membrane and seal integ-The resin acts as a conductor enabling the electrical poten- rity; the passage of dissolved gases, such as carbon dioxidetial to drive the captured cations and anions through the and ammonia; and the volume of wastewater, particularlyresin and appropriate membranes for concentration and re- where water discharge is tightly regulated by local authori-moval in the waste water stream. The electrical potential ties. Failure of membrane or seal integrity will result in prod-also separates the water in the resin (product) section into




uct water contamination. Methods of control involve suita- be demonstrated and validated as short-term, single-use fil-ble pretreatment of the influent water stream, appropriate ters at points of use in water systems that are not designedmembrane material selection, integrity challenges, mem- for endotoxin control or where only an endotoxin “polish-brane design and heat tolerance, periodic sanitization, and ing” (removal of only slight or occasional endotoxin levels)monitoring of differential pressures, conductivity, microbial is needed. Control and validation concerns include volumelevels, and TOC. and duration of use, flow rate, water conductivity and pu-

The development of RO units that can tolerate sanitizing rity, and constancy and concentration of endotoxin levelswater temperatures as well as operate efficiently and contin- being removed. All of these factors may have to be evalu-uously at elevated temperatures has added greatly to their ated and challenged prior to using this approach, makingmicrobial control and to the avoidance of biofouling. RO this a difficult-to-validate application. Even so, there may stillunits can be used alone or in combination with DI and CEDI be a possible need for additional backup endotoxin testingunits as well as ultrafiltration for operational and quality both upstream and downstream of the filter.enhancements.

Microbial-Retentive FiltrationUltrafiltration

Microbial-retentive membrane filters have experienced anUltrafiltration is a technology most often employed in evolution of understanding in the past decade that has

pharmaceutical water systems for removing endotoxins from caused previously held theoretical retention mechanisms toa water stream. It can also use semipermeable membranes, be reconsidered. These filters have a larger effective “porebut unlike RO, these typically use polysulfone membranes size” than ultrafilters and are intended to prevent the pas-whose intersegmental “pores” have been purposefully exag- sage of microorganisms and similarly sized particles withoutgerated during their manufacture by preventing the poly- unduly restricting flow. This type of filtration is widely em-mer molecules from reaching their smaller equilibrium prox- ployed within water systems for filtering the bacteria out ofimities to each other. Depending on the level of equilibrium both water and compressed gases as well as for vent filterscontrol during their fabrication, membranes with differing on tanks and stills and other unit operations. However, themolecular weight “cutoffs” can be created such that mole- properties of the water system microorganisms seem tocules with molecular weights above these cutoff ratings are challenge a filter’s microbial retention from water with phe-rejected and cannot penetrate the filtration matrix. nomena absent from other aseptic filtration applications,

Ceramic ultrafilters are another molecular sieving technol- such as filter sterilizing of pharmaceutical formulations priorogy. Ceramic ultrafilters are self supporting and extremely to packaging. In the latter application, sterilizing grade fil-durable, backwashable, chemically cleanable, and steam ters are generally considered to have an assigned rating ofsterilizable. However, they may require higher operating 0.2 or 0.22 µm. This rather arbitrary rating is associatedpressures than membrane type ultrafilters. with filters that have the ability to retain a high level chal-

All ultrafiltration devices work primarily by a molecular lenge of a specially prepared inoculum of Brevundimonassieving principle. Ultrafilters with molecular weight cutoff (formerly Pseudomonas) diminuta. This is a small microorgan-ratings in the range of 10,000–20,000 Da are typically used ism originally isolated decades ago from a product that hadin water systems for removing endotoxins. This technology been “filter sterilized” using a 0.45-µm rated filter. Furthermay be appropriate as an intermediate or final purification study revealed that a percentage of cells of this microorgan-step. Similar to RO, successful performance is dependent ism could reproducibly penetrate the 0.45-µm sterilizing fil-upon pretreatment of the water by upstream unit ters. Through historic correlation of B. diminuta-retainingoperations. tighter filters, thought to be twice as good as a 0.45-µm

Issues of concern for ultrafilters include compatibility of filter, assigned ratings of 0.2 or 0.22 µm with their success-membrane material with heat and sanitizing agents, mem- ful use in product solution filter sterilization, both this filterbrane integrity, fouling by particles and microorganisms, rating and the associated high level B. diminuta challengeand seal integrity. Control measures involve filtration me- have become the current benchmarks for sterilizing filtra-dium selection, sanitization, flow design (dead end vs. tan- tion. New evidence now suggests that for microbial-reten-gential), integrity challenges, regular cartridge changes, ele- tive filters used for pharmaceutical water, B. diminuta mayvated feed water temperature, and monitoring TOC and not be the best model microorganism.differential pressure. Additional flexibility in operation is pos- An archaic understanding of microbial-retentive filtrationsible based on the way ultrafiltration units are arranged such would lead one to equate a filter’s rating with the false im-as in a parallel or series configurations. Care should be taken pression of a simple sieve or screen that absolutely retainsto avoid stagnant water conditions that could promote mi- particles sized at or above the filter’s rating. A current un-croorganism growth in back-up or standby units. derstanding of the mechanisms involved in microbial reten-

tion and the variables that can affect those mechanisms hasyielded a far more complex interaction of phenomena thanCharge-Modified Filtration previously understood. A combination of simple sieve reten-tion and surface adsorption are now known to contribute toCharge-modified filters are usually microbially retentive fil- microbial retention.ters that are treated during their manufacture to have a pos- The following all interact to create some unusual and sur-itive charge on their surfaces. Microbial-retentive filtration prising retention phenomena for water system microorgan-will be described in a subsequent section, but the significant isms: the variability in the range and average pore sizes cre-feature of these membranes is their electrostatic surface ated by the various membrane fabrication processes, thecharge. Such charged filters can reduce endotoxin levels in variability of the surface chemistry and three-dimensionalthe fluids passing through them by their adsorption (owing structures related to the different polymers used in theseto the endotoxin’s negative charge) onto the membrane filter matrices, and the size and surface properties of thesurfaces. Although ultrafilters are more often employed as a microorganism intended to be retained by the filters. B.unit operation for endotoxin removal in water systems, diminuta may not be the best challenge microorganisms forcharge-modified filters may also have a place in endotoxin demonstrating bacterial retention for 0.2- to 0.22-µm ratedremoval particularly where available upstream pressures are filters for use in water systems because it appears to benot sufficient for ultrafiltration and for a single, relatively more easily retained by these filters than some water systemshort-term use. Charge-modified filters may be difficult to flora. The well-documented appearance of water system mi-validate for long-term or large-volume endotoxin retention. croorganisms on the downstream sides of some 0.2- toEven though their purified standard endotoxin retention can 0.22-µm rated filters after a relatively short period of usebe well characterized, their retention capacity for “natural” seems to support that some penetration phenomena are atendotoxins is difficult to gauge. Nevertheless, utility could




work. Unknown for certain is if this downstream appearance the destruction of ozone. With intense emissions at wave-is caused by a “blow-through” or some other pass-through lengths around 185 nm (as well as at 254 nm), mediumphenomenon as a result of tiny cells or less cell “stickiness”, pressure UV lights have demonstrated utility in the destruc-or by a “growth through” phenomenon as a result of cells tion of the chlorine-containing disinfectants used in sourcehypothetically replicating their way through the pores to the water as well as for interim stages of water pretreatment.downstream side. Whatever is the penetration mechanism, High intensities of this wavelength alone or in combination0.2- to 0.22-µm rated membranes may not be the best with other oxidizing sanitants, such as hydrogen peroxide,choice for some water system uses. have been used to lower TOC levels in recirculating distribu-

Microbial retention success in water systems has been re- tion systems. The organics are typically converted to carbonported with the use of some manufacturers’ filters arbitrarily dioxide, which equilibrates to bicarbonate, and incompletelyrated as 0.1 µm. There is general agreement that for a given oxidized carboxylic acids, both of which can easily be re-manufacturer, their 0.1-µm rated filters are tighter than their moved by polishing ion-exchange resins. Areas of concern0.2- to 0.22-µm rated filters. However, comparably rated include adequate UV intensity and residence time, gradualfilters from different manufacturers in water filtration appli- loss of UV emissivity with bulb age, gradual formation ofcations may not perform equivalently owing to the different UV-absorbing film at the water contact surface, incompletefilter fabrication processes and the nonstandardized micro- photodegradation during unforeseen source water hyper-bial retention challenge processes currently used for defining chlorination, release of ammonia from chloraminethe 0.1-µm filter rating. It should be noted that use of 0.1- photodegradation, unapparent UV bulb failure, and conduc-µm rated membranes generally results in a sacrifice in flow tivity degradation in distribution systems using 185-nm UVrate compared to 0.2- to 0.22-µm membranes, so whatever lights. Control measures include regular inspection or emis-membranes are chosen for a water system application, the sivity alarms to detect bulb failures or film occlusions, regu-user must verify that the membranes are suitable for their lar UV bulb sleeve cleaning and wiping, downstream chlo-intended application, use period, and use process, including rine detectors, downstream polishing deionizers, and regularflow rate. (approximately yearly) bulb replacement.

For microbial retentive gas filtrations, the same sievingand adsorptive retention phenomena are at work as in liq- Distillationuid filtration, but the adsorptive phenomenon is enhancedby additional electrostatic interactions between particles and Distillation units provide chemical and microbial purifica-filter matrix. These electrostatic interactions are so strong tion via thermal vaporization, mist elimination, and waterthat particle retention for a given filter rating is significantly vapor condensation. A variety of designs is available includ-more efficient in gas filtration than in water or product solu- ing single effect, multiple effect, and vapor compression.tion filtrations. These additional adsorptive interactions The latter two configurations are normally used in largerrender filters rated at 0.2–0.22 µm unquestionably suitable systems because of their generating capacity and efficiency.for microbial retentive gas filtrations. When microbially re- Distilled water systems require different feed water controlstentive filters are used in these applications, the membrane than required by membrane systems. For distillation, duesurface is typically hydrophobic (non-wettable by water). A consideration must be given to prior removal of hardnesssignificant area of concern for gas filtration is blockage of and silica impurities that may foul or corrode the heat trans-tank vents by condensed water vapor, which can cause me- fer surfaces as well as prior removal of those impurities thatchanical damage to the tank. Control measures include elec- could volatize and condense along with the water vapor. Intrical or steam tracing and a self-draining orientation of vent spite of general perceptions, even the best distillation pro-filter housings to prevent accumulation of vapor conden- cess cannot afford absolute removal of contaminating ionssate. However, a continuously high filter temperature will and endotoxin. Most stills are recognized as being able totake an oxidative toll on polypropylene components of the accomplish at least a 3–4 log reduction in these impurityfilter, so sterilization of the unit prior to initial use, and peri- concentrations. Areas of concern include carry-over of vola-odically thereafter, as well as regular visual inspections, in- tile organic impurities such as trihalomethanes (see Source ortegrity tests, and changes are recommended control Feed Water Considerations) and gaseous impurities such asmethods. ammonia and carbon dioxide, faulty mist elimination, evap-In water applications, microbial-retentive filters may be orator flooding, inadequate blowdown, stagnant water inused downstream of unit operations that tend to release condensers and evaporators, pump and compressor seal de-microorganisms or upstream of unit operations that are sen- sign, pinhole evaporator and condenser leaks, and conduc-sitive to microorganisms. Microbial-retentive filters may also tivity (quality) variations during start-up and operation.be used to filter water feeding the distribution system. It Methods of control may involve preliminary decarbona-should be noted that regulatory authorities allow the use of tion steps to remove both dissolved carbon dioxide andmicrobial-retentive filters within distribution systems or even other volatile or noncondensable impurities; reliable mistat use points if they have been properly validated and are elimination to minimize feedwater droplet entrainment; vis-appropriately maintained. A point-of-use filter should only ual or automated high water level indication to detect boilerbe intended to “polish” the microbial quality of an other- flooding and boil over; use of sanitary pumps and compres-wise well-maintained system and not to serve as the primary sors to minimize microbial and lubricant contamination ofmicrobial control device. The efficacy of system microbial feedwater and condensate; proper drainage during inactivecontrol measures can only be assessed by sampling the periods to minimize microbial growth and accumulation ofwater upstream of the filters. As an added measure of pro- associated endotoxin in boiler water; blowdown control totection, in-line UV lamps, appropriately sized for the flow limit the impurity concentration effect in the boiler to man-rate (see Sanitization), may be used just upstream of micro- ageable levels; on-line conductivity sensing with automatedbial-retentive filters to inactivate microorganisms prior to diversion to waste to prevent unacceptable water upon stilltheir capture by the filter. This tandem approach tends to startup or still malfunction from getting into the finishedgreatly delay potential microbial penetration phenomena water distribute system; and periodic integrity testing forand can substantially extend filter service life. pinhole leaks to routinely assure condensate is not compro-

mised by nonvolatized source water contaminants.Ultraviolet Light

Storage TanksThe use of low-pressure UV lights that emit a 254-nmwavelength for microbial control is discussed under Sanitiza- Storage tanks are included in water distribution systemstion, but the application of UV light in chemical purification to optimize processing equipment capacity. Storage also al-is also emerging. This 254-nm wavelength is also useful in lows for routine maintenance within the pretreatment train




while maintaining continuous supply to meet manufacturing for microorganism control. The system may be continuouslyneeds. Design and operation considerations are needed to operated at sanitizing conditions or sanitized periodically.prevent or minimize the development of biofilm, to mini-mize corrosion, to aid in the use of chemical sanitization of

INSTALLATION, MATERIALS OFthe tanks, and to safeguard mechanical integrity. These con-siderations may include using closed tanks with smooth in- CONSTRUCTION, AND COMPONENTteriors, the ability to spray the tank headspace using SELECTIONsprayballs on recirculating loop returns, and the use ofheated, jacketed/insulated tanks. This minimizes corrosion Installation techniques are important because they can af-and biofilm development and aids in thermal and chemical fect the mechanical, corrosive, and sanitary integrity of thesanitization. Storage tanks require venting to compensate system. Valve installation attitude should promote gravityfor the dynamics of changing water levels. This can be ac- drainage. Pipe supports should provide appropriate slopescomplished with a properly oriented and heat-traced filter for drainage and should be designed to support the pipinghousing fitted with a hydrophobic microbial retentive mem- adequately under worst-case thermal and flow conditions.brane filter affixed to an atmospheric vent. Alternatively, an The methods of connecting system components includingautomatic membrane-filtered compressed gas blanketing units of operation, tanks, and distribution piping requiresystem may be used. In both cases, rupture disks equipped careful attention to preclude potential problems. Stainlesswith a rupture alarm device should be used as a further steel welds should provide reliable joints that are internallysafeguard for the mechanical integrity of the tank. Areas of smooth and corrosion-free. Low-carbon stainless steel, com-concern include microbial growth or corrosion due to irreg- patible wire filler, where necessary, inert gas, automaticular or incomplete sanitization and microbial contamination welding machines, and regular inspection and documenta-from unalarmed rupture disk failures caused by condensate- tion help to ensure acceptable weld quality. Follow-upoccluded vent filters. cleaning and passivation are important for removing con-

tamination and corrosion products and to re-establish thepassive corrosion-resistant surface. Plastic materials can beDistribution Systemsfused (welded) in some cases and also require smooth, uni-form internal surfaces. Adhesive glues and solvents shouldDistribution system configuration should allow for thebe avoided due to the potential for voids and extractables.continuous flow of water in the piping by means of recircu-Mechanical methods of joining, such as flange fittings, re-lation. Use of nonrecirculating, dead-end, or one-way sys-quire care to avoid the creation of offsets, gaps, penetra-tems or system segments should be avoided whenever pos-tions, and voids. Control measures include good alignment,sible. If not possible, these systems should be periodicallyproperly sized gaskets, appropriate spacing, uniform sealingflushed and more closely monitored. Experience has shownforce, and the avoidance of threaded fittings.that continuously recirculated systems are easier to main-

Materials of construction should be selected to be com-tain. Pumps should be designed to deliver fully turbulentpatible with control measures such as sanitizing, cleaning,flow conditions to facilitate thorough heat distribution (forand passivating. Temperature rating is a critical factor inhot water sanitized systems) as well as thorough chemicalchoosing appropriate materials because surfaces may be re-sanitant distribution. Turbulent flow also appear to eitherquired to handle elevated operating and sanitization tem-retard the development of biofilms or reduce the tendencyperatures. Should chemicals or additives be used to clean,of those biofilms to shed bacteria into the water. If redun-control, or sanitize the system, materials resistant to thesedant pumps are used, they should be configured and usedchemicals or additives must be utilized. Materials should beto avoid microbial contamination of the system.capable of handling turbulent flow and elevated velocitiesComponents and distribution lines should be sloped andwithout wear of the corrosion-resistant film such as the pas-fitted with drain points so that the system can be com-sive chromium oxide surface of stainless steel. The finish onpletely drained. In stainless steel distribution systems wheremetallic materials such as stainless steel, whether it is a re-the water is circulated at a high temperature, dead legs andfined mill finish, polished to a specific grit, or an electropol-low-flow conditions should be avoided, and valved tie-inished treatment, should complement system design andpoints should have length-to-diameter ratios of six or less. Ifprovide satisfactory corrosion and microbial activity resis-constructed of heat-tolerant plastic, this ratio should betance as well as chemical sanitizability. Auxiliary equipmenteven less to avoid cool points where biofilm developmentand fittings that require seals, gaskets, diaphragms, filtercould occur. In ambient temperature distribution systems,media, and membranes should exclude materials that per-particular care should be exercised to avoid or minimizemit the possibility of extractables, shedding, and microbialdead leg ratios of any size and provide for complete drain-activity. Insulating materials exposed to stainless steel sur-age. If the system is intended to be steam sanitized, carefulfaces should be free of chlorides to avoid the phenomenonsloping and low-point drainage is crucial to condensate re-of stress corrosion cracking that can lead to system contami-moval and sanitization success. If drainage of componentsnation and the destruction of tanks and critical systemor distribution lines is intended as a microbial control strat-components.egy, they should also be configured to be completely dried

Specifications are important to ensure proper selection ofusing dry compressed air (or nitrogen if appropriate em-materials and to serve as a reference for system qualificationployee safety measures are used). Drained but still moist sur-and maintenance. Information such as mill reports for stain-faces will still support microbial proliferation. Water exitingless steel and reports of composition, ratings, and materialfrom the distribution system should not be returned to thehandling capabilities for nonmetallic substances should besystem without first passing through all or a portion of thereviewed for suitability and retained for reference. Compo-purification train.nent (auxiliary equipment) selection should be made withThe distribution design should include the placement ofassurance that it does not create a source of contaminationsampling valves in the storage tank and at other locations,intrusion. Heat exchangers should be constructed to preventsuch as in the return line of the recirculating water system.leakage of heat transfer medium to the pharmaceuticalWhere feasible, the primary sampling sites for water shouldwater and, for heat exchanger designs where preventionbe the valves that deliver water to the points of use. Directmay fail, there should be a means to detect leakage. Pumpsconnections to processes or auxiliary equipment should beshould be of sanitary design with seals that prevent contam-designed to prevent reverse flow into the controlled waterination of the water. Valves should have smooth internal sur-system. Hoses and heat exchangers that are attached tofaces with the seat and closing device exposed to the flush-points of use in order to deliver water for a particular useing action of water, such as occurs in diaphragm valves.must not chemically or microbiologically degrade the waterValves with pocket areas or closing devices (e.g., ball, plug,quality. The distribution system should permit sanitization




gate, globe) that move into and out of the flow area should and quantification of residues of the sanitant or its objec-be avoided. tionable degradants is an essential part of the validation

program. The frequency of sanitization should be supportedby, if not triggered by, the results of system microbial moni-

SANITIZATION toring. Conclusions derived from trend analysis of the mi-crobiological data should be used as the alert mechanism

Microbial control in water systems is achieved primarily for maintenance. The frequency of sanitization should bethrough sanitization practices. Systems can be sanitized us- established in such a way that the system operates in a stateing either thermal or chemical means. Thermal approaches of microbiological control and does not routinely exceedto system sanitization include periodic or continuously circu- alert levels (see Alert and Action Levels and Specifications).lating hot water and the use of steam. Temperatures of atleast 80° are most commonly used for this purpose, butcontinuously recirculating water of at least 65° has also OPERATION, MAINTENANCE, AND CONTROLbeen used effectively in insulated stainless steel distributionsystems when attention is paid to uniformity and distribu- A preventive maintenance program should be establishedtion of such self-sanitizing temperatures. These techniques to ensure that the water system remains in a state of con-are limited to systems that are compatible with the higher trol. The program should include (1) procedures for operat-temperatures needed to achieve sanitization. Although ther- ing the system, (2) monitoring programs for critical qualitymal methods control biofilm development by either contin- attributes and operating conditions including calibration ofuously inhibiting their growth or, in intermittent applica- critical instruments, (3) schedule for periodic sanitization, (4)tions, by killing the microorganisms within biofilms, they are preventive maintenance of components, and (5) control ofnot effective in removing established biofilms. Killed but in- changes to the mechanical system and to operatingtact biofilms can become a nutrient source for rapid biofilm conditions.regrowth after the sanitizing conditions are removed or Operating Procedures—Procedures for operating thehalted. In such cases, a combination of routine thermal and water system and performing routine maintenance and cor-periodic supplementation with chemical sanitization might rective action should be written, and they should also definebe more effective. The more frequent the thermal sanitiza- the point when action is required. The procedures shouldtion, the more likely biofilm development and regrowth can be well documented, detail the function of each job, assignbe eliminated. Chemical methods, where compatible, can who is responsible for performing the work, and describebe used on a wider variety of construction materials. These how the job is to be conducted. The effectiveness of thesemethods typically employ oxidizing agents such as haloge- procedures should be assessed during water system valida-nated compounds, hydrogen peroxide, ozone, peracetic tion.acid, or combinations thereof. Halogenated compounds are Monitoring Program—Critical quality attributes and op-effective sanitizers but are difficult to flush from the system erating parameters should be documented and monitored.and may leave biofilms intact. Compounds such as hydro- The program may include a combination of in-line sensorsgen peroxide, ozone, and peracetic acid oxidize bacteria or automated instruments (e.g., for TOC, conductivity,and biofilms by forming reactive peroxides and free radicals hardness, and chlorine), automated or manual documenta-(notably hydroxyl radicals). The short half-life of ozone in tion of operational parameters (such as flow rates or pres-particular, and its limitation on achievable concentrations re- sure drop across a carbon bed, filter, or RO unit), and labo-quire that it be added continuously during the sanitization ratory tests (e.g., total microbial counts). The frequency ofprocess. Hydrogen peroxide and ozone rapidly degrade to sampling, the requirement for evaluating test results, andwater and oxygen; peracetic acid degrades to acetic acid in the necessity for initiating corrective action should be in-the presence of UV light. In fact, ozone’s ease of degrada- cluded.tion to oxygen using 254-nm UV light at use points allow it

Sanitization—Depending on system design and the se-to be most effectively used on a continuous basis to providelected units of operation, routine periodic sanitization maycontinuously sanitizing conditions.be necessary to maintain the system in a state of microbialIn-line UV light at a wavelength of 254 nm can also becontrol. Technologies for sanitization are described above.used to continuously “sanitize” water circulating in the sys-

Preventive Maintenance—A preventive maintenancetem, but these devices must be properly sized for the waterprogram should be in effect. The program should establishflow. Such devices inactivate a high percentage (but notwhat preventive maintenance is to be performed, the fre-100%) of microorganisms that flow through the device butquency of maintenance work, and how the work should becannot be used to directly control existing biofilm upstreamdocumented.or downstream of the device. However, when coupled with

conventional thermal or chemical sanitization technologies Change Control—The mechanical configuration and op-or located immediately upstream of a microbially retentive erating conditions must be controlled. Proposed changesfilter, it is most effective and can prolong the interval beshould be evaluated for their impact on the whole system.tween system sanitizations. The need to requalify the system after changes are made

It is important to note that microorganisms in a well-de- should be determined. Following a decision to modify aveloped biofilm can be extremely difficult to kill, even by water system, the affected drawings, manuals, and proce-aggressive oxidizing biocides. The less developed and there- dures should be revised.fore thinner the biofilm, the more effective the biocidal ac-tion. Therefore, optimal biocide control is achieved by fre-

SAMPLING CONSIDERATIONSquent biocide use that does not allow significant biofilmdevelopment between treatments.

Water systems should be monitored at a frequency that isSanitization steps require validation to demonstrate thesufficient to ensure that the system is in control and contin-capability of reducing and holding microbial contaminationues to produce water of acceptable quality. Samples shouldat acceptable levels. Validation of thermal methods shouldbe taken from representative locations within the processinginclude a heat distribution study to demonstrate that sani-and distribution system. Established sampling frequenciestization temperatures are achieved throughout the system,should be based on system validation data and should coverincluding the body of use point valves. Validation of chemi-critical areas including unit operation sites. The samplingcal methods requires demonstrating adequate chemical con-plan should take into consideration the desired attributes ofcentrations throughout the system, exposure to all wettedthe water being sampled. For example, systems for Water forsurfaces, including the body of use-point valves, and com-Injection, because of their more critical microbiological re-plete removal of the sanitant from the system at the com-

pletion of treatment. Methods validation for the detection




quirements, may require a more rigorous sampling Change to read:frequency.

Analyses of water samples often serve two purposes: in-process control assessments and final quality control assess-

CHEMICAL CONSIDERATIONSments. In-process control analyses are usually focused onthe attributes of the water within the system. Quality con-

The chemical attributes of Purified Water and Water fortrol is primarily concerned with the attributes of the waterInjection in effect prior to USP 23 were specified by a seriesdelivered by the system to its various uses. The latter usuallyof chemistry tests for various specific and nonspecific attrib-employs some sort of transfer device, often a flexible hose,utes with the intent of detecting chemical species indicativeto bridge the gap between the distribution system use-pointof incomplete or inadequate purification. While these meth-valve and the actual location of water use. The issue of sam-ods could have been considered barely adequate to controlple collection location and sampling procedure is oftenthe quality of these waters, they nevertheless stood the testhotly debated because of the typically mixed use of the dataof time. This was partly because the operation of water sys-generated from the samples, for both in-process control andtems was, and still is, based on on-line conductivity meas-quality control. In these single-sample and mixed-data useurements and specifications generally thought to precludesituations, the worst-case scenario should be utilized. Inthe failure of these archaic chemistry attribute tests.other words, samples should be collected from use points

USP moved away from these chemical attribute tests tousing the same delivery devices, such as hoses, and proce-contemporary analytical technologies for the bulk waters Pu-dures, such as preliminary hose or outlet flushing, as arerified Water and Water for Injection. The intent was to up-employed by production from those use points. Where usegrade the analytical technologies without tightening thepoints per se cannot be sampled, such as hard-piped con-quality requirements. The two contemporary analytical tech-nections to equipment, special sampling ports may be used.nologies employed were TOC and conductivity. The TOCIn all cases, the sample must represent as closely as possibletest replaced the test for Oxidizable Substances that primarilythe quality of the water used in production. If a point-of-usetargeted organic contaminants. A multistaged Conductivityfilter is employed, sampling of the water prior to and aftertest which detects ionic (mostly inorganic) contaminants re-the filter is needed, because the filter will mask the micro-placed, with the exception of the test for Heavy metals, allbial control achieved by the normal operating procedures ofof the inorganic chemical tests (i.e., Ammonia, Calcium, Car-the system.bon dioxide, Chloride, Sulfate).Samples containing chemical sanitizing agents require

Replacing the heavy metals attribute was considered un-neutralization prior to microbiological analysis. Samples fornecessary because (a) the source water specifications (foundmicrobiological analysis should be tested immediately, orin the NPDWR) for individual Heavy metals were tighter thansuitably refrigerated to preserve the original microbial attrib-the approximate limit of detection of the Heavy metals testutes until analysis can begin. Samples of flowing water arefor USP XXII Water for Injection and Purified Water (approxi-only indicative of the concentration of planktonic (free float-mately 0.1 ppm), (b) contemporary water system construc-ing) microorganisms present in the system. Biofilm microor-tion materials do not leach heavy metal contaminants, andganisms (those attached to water system surfaces) are usu-(c) test results for this attribute have uniformly been nega-ally present in greater numbers and are the source of thetive—there has not been a confirmed occurrence of a singu-planktonic population recovered from grab samples. Micro-lar test failure (failure of only the Heavy metals test with allorganisms in biofilms represent a continuous source of con-other attributes passing) since the current heavy metaltamination and are difficult to directly sample and quantify.drinking water standards have been in place. Nevertheless,Consequently, the planktonic population is usually used asbecause the presence of heavy metals in Purified Water oran indicator of system contamination levels and is the basisWater for Injection could have dire consequences, its absencefor system Alert and Action Levels and Specifications. The con-should at least be documented during new water systemsistent appearance of elevated planktonic levels is usually ancommissioning and validation or through prior test resultsindication of advanced biofilm development in need of re-records.medial control. System control and sanitization are key in

Total solids and pH were the only tests not covered bycontrolling biofilm formation and the consequent planktonicconductivity testing. The test for Total solids was consideredpopulation.redundant because the nonselective tests of conductivitySampling for chemical analyses is also done for in-processand TOC could detect most chemical species other thancontrol and for quality control purposes. However, unlikesilica, which could remain undetected in its colloidal form.microbial analyses, chemical analyses can be and often areColloidal silica in Purified Water and Water for Injection isperformed using on-line instrumentation. Such on-line test-easily removed by most water pretreatment steps and evening has unequivocal in-process control purposes because itif present in the water, constitutes no medical or functionalis not performed on the water delivered from the system.hazard except under extreme and rare situations. In suchHowever, unlike microbial attributes, chemical attributes areextreme situations, other attribute extremes are also likely tousually not significantly degraded by hoses. Therefore,be detected. It is, however, the user’s responsibility to en-through verification testing, it may be possible to show thatsure fitness for use. If silica is a significant component in thethe chemical attributes detected by the on-line instrumenta-source water, and the purification unit operations could betion (in-process testing) are equivalent to those detected atoperated or fail and selectively allow silica to be releasedthe ends of the use-point hoses (quality control testing).into the finished water (in the absence of co-contaminantsThis again creates a single-sample and mixed-data use sce-detectable by conductivity), then either silica-specific or anario. It is far better to operate the instrumentation in atotal solids type testing should be utilized to monitor andcontinuous mode, generating large volumes of in-processcontrol this rare problem.data, but only using a defined small sampling of that data

The pH attribute was eventually recognized to be redun-for QC purposes. Examples of acceptable approaches in-dant to the conductivity test (which included pH as an as-clude using highest values for a given period, highest time-pect of the test and specification); therefore, pH wasweighted average for a given period (from fixed or rollingdropped as a separate attribute test.sub-periods), or values at a fixed daily time. Each approach

The rationale used by USP to establish its Purified Waterhas advantages and disadvantages relative to calculationand Water for Injection conductivity specifications took intocomplexity and reflection of continuous quality, so the userconsideration the conductivity contributed by the two leastmust decide which approach is most suitable or justifiable.conductive former attributes of Chloride and Ammonia,thereby precluding their failure had those wet chemistrytests been performed. In essence, the Stage 3 conductivityspecifications (see Water Conductivity ⟨645⟩, Bulk Water) were




established from the sum of the conductivities of the limit itself is the source of chemicals (inorganics and organics)concentrations of chloride ions (from pH 5.0–6.2) and am- that leach over time into the packaged water and can easilymonia ions (from pH 6.3–7.0), plus the unavoidable contri- be detected by the conductivity and TOC tests. The irony ofbution of other conductivity-contributing ions from water organic leaching from plastic packaging is that before the(H+ and OH–), dissolved atmospheric CO2 (as HCO3–), and advent of bulk water TOC testing when the Oxidizable Sub-an electro-balancing quantity of either Na+ or Cl–, depend- stances test was the only “organic purity” test for both bulking on the pH-induced ionic imbalance (see Table 1). The and packaged/sterile water monographs in USP, that test’sStage 2 conductivity specification is the lowest value on this insensitivity to many of the organic leachables from plastictable, 2.1 µS/cm. The Stage 1 specifications, designed pri- and elastomeric packaging materials was largely unrealized,marily for on-line measurements, were derived essentially by allowing organic levels in packaged/sterile water to be quitesumming the lowest values in the contributing ion columns high (possibly many times the TOC specification for bulkfor each of a series of tables similar to Table 1, created for water). Similarly, glass containers can also leach inorganics,each 5° increment between 0° and 100°. For example pur- such as sodium, which are easily detected by conductivity,poses, the italicized values in Table 1, the conductivity data but poorly detected by the former wet chemistry attributetable for 25°, were summed to yield a conservative value of tests. Most of these leachables are considered harmless by1.3 µS/cm, the Stage 1 specification for a nontemperature current perceptions and standards at the rather significantcompensated, nonatmosphere equilibrated water sample concentrations present. Nevertheless, they effectively de-that actually had a measured temperature of 25°–29°. Each grade the quality of the high-purity waters placed into these5° increment in the table was similarly treated to yield the packaging systems. Some packaging materials contain moreindividual values listed in the table of Stage 1 specifications leachables than others and may not be as suitable for hold-(see Water Conductivity ⟨645⟩, Bulk Water). ing water and maintaining its purity.

As stated above, this rather radical change to utilizing a The attributes of conductivity and TOC tend to revealconductivity attribute as well as the inclusion of a TOC attri- more about the packaging leachables than they do aboutbute allowed for on-line measurements. This was a major the water’s original purity. These currently “allowed” leach-philosophical change and allowed major savings to be real- ables could render the sterile packaged versions of originallyized by industry. The TOC and conductivity tests can also equivalent bulk water essentially unsuitable for many usesbe performed “off-line” in the laboratories using collected where the bulk waters are perfectly adequate.samples, although sample collection tends to introduce op- Therefore, to better control the ionic packaging leach-portunities for adventitious contamination that can cause ables, Water Conductivity ⟨645⟩ is divided into two sections.false high readings. The collection of on-line data is not, The first is titled Bulk Water, which applies to Purified Water,however, without challenges. The continuous readings tend Water for Injection, Water for Hemodialysis, and Pure Steam,to create voluminous amounts of data where before only a and includes the three-stage conductivity testing instructionssingle data point was available. As stated under Sampling and specifications. The second is titled Sterile Water, whichConsiderations, continuous in-process data sampling is excel- applies to Sterile Purified Water, Sterile Water for Injection,lent for understanding how a water system performs during Sterile Water for Inhalation, and Sterile Water for Irrigation.all of its various usage and maintenance events in real time, The Sterile Water section includes conductivity specificationsbut is too much data for QC purposes. Therefore, a justifia- similar to the Stage 2 testing approach because it is in-ble fraction or averaging of the data can be used that is still tended as a laboratory test, and these sterile waters wererepresentative of the overall water quality being used. made from bulk water that already complied with the three-

Packaged/sterile waters present a particular dilemma rela- stage conductivity test. In essence, packaging leachables aretive to the attributes of conductivity and TOC. The package the primary target “analytes” of the conductivity specifica-

Table 1. Contributing Ion Conductivities of the Chloride–Ammonia Model as a Function of pH(in atmosphere-equilibrated water at 25°)

Conductivity(µS/cm)

Combined Stage 3pH H+ OH– HCO3– Cl– Na+ NH4+ Conductivities Limit5.0 3.49 0 0.02 1.01 0.19 0 4.71 4.75.1 2.77 0 0.02 1.01 0.29 0 4.09 4.15.2 2.20 0 0.03 1.01 0.38 0 3.62 3.65.3 1.75 0 0.04 1.01 0.46 0 3.26 3.35.4 1.39 0 0.05 1.01 0.52 0 2.97 3.05.5 1.10 0 0.06 1.01 0.58 0 2.75 2.85.6 0.88 0 0.08 1.01 0.63 0 2.60 2.65.7 0.70 0 0.10 1.01 0.68 0 2.49 2.55.8 0.55 0 0.12 1.01 0.73 0 2.41 2.45.9 0.44 0 0.16 1.01 0.78 0 2.39 2.46.0 0.35 0 0.20 1.01 0.84 0 2.40 2.46.1 0.28 0 0.25 1.01 0.90 0 2.44 2.46.2 0.22 0 0.31 1.01 0.99 0 2.53 2.56.3 0.18 0 0.39 0.63 0 1.22 2.42 2.46.4 0.14 0.01 0.49 0.45 0 1.22 2.31 2.36.5 0.11 0.01 0.62 0.22 0 1.22 2.18 2.26.6 0.09 0.01 0.78 0 0.04 1.22 2.14 2.16.7 0.07 0.01 0.99 0 0.27 1.22 2.56 2.66.8 0.06 0.01 1.24 0 0.56 1.22 3.09 3.16.9 0.04 0.02 1.56 0 0.93 1.22 3.77 3.87.0 0.03 0.02 1.97 0 1.39 1.22 4.63 4.6




tions in the Sterile Water section of Water Conductivity ⟨645⟩. biofilm is an adaptive response by certain microorganisms toThe effect on potential leachables from different container survive in this low nutrient environment. Downstream colo-sizes is the rationale for having two different specifications, nization can occur when microorganisms are shed from ex-one for small packages containing nominal volumes of isting biofilm-colonized surfaces and carried to other areas10 mL or less and another for larger packages. These con- of the water system. Microorganisms may also attach to sus-ductivity specifications are harmonized with the European pended particles such as carbon bed fines or fractured resinPharmacopoeia conductivity specifications for Sterile Water particles. When the microorganisms become planktonic,for Injection. All monographed waters, except Bacteriostatic they serve as a source of contamination to subsequent puri-Water for Injection, have a conductivity specification that di- fication equipment (compromising its functionality) and torects the user to either the Bulk Water or the Sterile Water distribution systems.section of Water Conductivity ⟨645⟩. For the sterile packaged Another source of endogenous microbial contamination iswater monographs, this water conductivity specification re- the distribution system itself. Microorganisms can colonizeplaces the redundant wet chemistry limit tests intended for pipe surfaces, rough welds, badly aligned flanges, valves,inorganic contaminants that had previously been specified and unidentified dead legs, where they proliferate, formingin these monographs. a biofilm. The smoothness and composition of the surface

Controlling the organic purity of these sterile packaged may affect the rate of initial microbial adsorption, but oncewaters, particularly those in plastic packaging, is more chal- adsorbed, biofilm development, unless otherwise inhibitedlenging. Although the TOC test can better detect and thereby sanitizing conditions, will occur regardless of the surface.fore be better used to monitor and control these impurities Once formed, the biofilm becomes a continuous source ofthan the current Oxidizable Substances test, the latter has microbial contamination.many decades-old precedents and flexibility with the varietyof packaging types and volumes applicable to these sterile

ENDOTOXIN CONSIDERATIONSpackaged waters. Nevertheless, TOC testing of these cur-rently allowed sterile, plastic-packaged waters reveals sub-

Endotoxins are lipopolysaccharides found in and shedstantial levels of plastic-derived organic leachables thatfrom the cell envelope that is external to the cell wall ofrender the water perhaps orders of magnitude less organi-Gram-negative bacteria. Gram-negative bacteria that formcally pure than typically achieved with bulk waters. There-biofilms can become a source of endotoxins in pharmaceuti-fore, usage of these packaged waters for analytical, manu-cal waters. Endotoxins may occur as clusters of lipopolysac-facturing, and cleaning applications should only becharide molecules associated with living microorganisms,exercised after suitability of the waters’ purity for the appli-fragments of dead microorganisms or the polysaccharidecation has been assured.slime surrounding biofilm bacteria, or as free molecules. The■There is an analogous partitioning of Total Organic Car-free form of endotoxins may be released from cell surfacesbon ⟨643⟩ to better control the organic packaging leach-of the bacteria that colonize the water system, or from theables. The first part is titled Bulk Water, which applies to thefeed water that may enter the water system. Because of theTOC method for Purified Water, Water for Injection, Water formultiplicity of endotoxin sources in a water system, endo-Hemodialysis, and Pure Steam. The second part is titled Ster-toxin quantitation in a water system is not a good indicatorile Water, which applies to the TOC method for Sterile Puri-of the level of biofilm abundance within a water system.fied Water, Sterile Water for Injection, Sterile Water for Inhala-

Endotoxin levels may be minimized by controlling the in-tion, and Sterile Water for Irrigation. For these sterile waters,troduction of free endotoxins and microorganisms in thethe TOC method is provided as an alternative test to thefeed water and minimizing microbial proliferation in the sys-Oxidizable Substances test. The TOC limits for the sterile wa-tem. This may be accomplished through the normal exclu-ters are set to higher values than the TOC limits for bulksion or removal action afforded by various unit operationswaters.■1S (USP36)within the treatment system as well as through system sani-tization. Other control methods include the use of ultrafilters

MICROBIAL CONSIDERATIONS or charge-modified filters, either in-line or at the point ofuse. The presence of endotoxins may be monitored as de-

The major exogenous source of microbial contamination scribed in the chapter Bacterial Endotoxins Test ⟨85⟩.of bulk pharmaceutical water is source or feed water. Feedwater quality must, at a minimum, meet the quality attrib-

MICROBIAL ENUMERATIONutes of Drinking Water for which the level of coliforms isregulated. A wide variety of other microorganisms, chiefly CONSIDERATIONSGram-negative bacteria, may be present in the incomingwater. These microorganisms may compromise subsequent The objective of a water system microbiological monitor-purification steps. Examples of other potential exogenous ing program is to provide sufficient information to controlsources of microbial contamination include unprotected and assess the microbiological quality of the water pro-vents, faulty air filters, ruptured rupture disks, backflow from duced. Product quality requirements should dictate watercontaminated outlets, unsanitized distribution system “open- quality specifications. An appropriate level of control may beings” including routine component replacements, inspec- maintained by using data trending techniques and, if neces-tions, repairs, and expansions, inadequate drain and air- sary, limiting specific contraindicated microorganisms. Con-breaks, and replacement activated carbon, deionizer resins, sequently, it may not be necessary to detect all of the mi-and regenerant chemicals. In these situations, the exoge- croorganisms species present in a given sample. Thenous contaminants may not be normal aquatic bacteria but monitoring program and methodology should indicate ad-rather microorganisms of soil or even of human origin. The verse trends and detect microorganisms that are potentiallydetection of nonaquatic microorganisms may be an indica- harmful to the finished product, process, or consumer. Finaltion of a system component failure, which should trigger selection of method variables should be based on the indi-investigations that will remediate their source. Sufficient care vidual requirements of the system being monitored.should be given to system design and maintenance in order It should be recognized that there is no single methodto minimize microbial contamination from these exogenous that is capable of detecting all of the potential microbialsources. contaminants of a water system. The methods used for mi-

Unit operations can be a major source of endogenous microbial monitoring should be capable of isolating the num-crobial contamination. Microorganisms present in feed bers and types of organisms that have been deemed signifi-water may adsorb to carbon bed, deionizer resins, filter cant relative to in-process system control and productmembranes, and other unit operation surfaces and initiate impact for each individual system. Several criteria should bethe formation of a biofilm. In a high-purity water system, considered when selecting a method to monitor the micro-




bial content of a pharmaceutical water system. These in- especially during the validation of a water system, as well asclude method sensitivity, range of organisms types or spe- periodically thereafter. This concurrent testing could deter-cies recovered, sample processing throughput, incubation mine if any additional numbers or types of bacteria can beperiod, cost, and methodological complexity. An alternative preferentially recovered by one of the approaches. If so, theconsideration to the use of the classical “culture” ap- impact of these additional isolates on system control andproaches is a sophisticated instrumental or rapid test the end uses of the water could be assessed. Also, the effi-method that may yield more timely results. However, care cacy of system controls and sanitization on these additionalmust be exercised in selecting such an alternative approach isolates could be assessed.to ensure that it has both sensitivity and correlation to class- Duration and temperature of incubation are also criticalical culture approaches, which are generally considered the aspects of a microbiological test method. Classical method-accepted standards for microbial enumeration. ologies using high-nutrient media are typically incubated at

Consideration should also be given to the timeliness of 30°–35° for 48–72 h. Because of the flora in certain watermicrobial enumeration testing after sample collection. The systems, incubation at lower temperatures (e.g., 20°–25°)number of detectable planktonic bacteria in a sample col- for longer periods (e.g., 5–7 days) can recover higher micro-lected in a scrupulously clean sample container will usually bial counts when compared to classical methods. Low-nutri-drop as time passes. The planktonic bacteria within the sam- ent media are designed for these lower temperature andple will tend to either die or to irretrievably adsorb to the longer incubation conditions (sometimes as long as 14 dayscontainer walls, reducing the number of viable planktonic to maximize recovery of very slow-growing oligotrophs orbacteria that can be withdrawn from the sample for testing. sanitant-injured microorganisms), but even high-nutrientThe opposite effect can also occur if the sample container is media can sometimes increase their recovery with thesenot scrupulously clean and contains a low concentration of longer and cooler incubation conditions. Whether or not asome microbial nutrient that could promote microbial particular system needs to be monitored using high- or low-growth within the sample container. Because the number of nutrient media with higher or lower incubation tempera-recoverable bacteria in a sample can change positively or tures or shorter or longer incubation times should be deter-negatively over time after sample collection, it is best to test mined during or prior to system validation and periodicallythe samples as soon as possible after being collected. If it is reassessed as the microbial flora of a new water systemnot possible to test the sample within about 2 h of collec- gradually establish a steady state relative to its routine main-tion, the sample should be held at refrigerated temperatures tenance and sanitization procedures. The establishment of a(2°–8°) for a maximum of about 12 h to maintain the mi- “steady state” can take months or even years and can becrobial attributes until analysis. In situations where even this perturbed by a change in use patterns, a change in routineis not possible (such as when using off-site contract labora- and preventative maintenance or sanitization procedures,tories), testing of these refrigerated samples should be per- and frequencies, or any type of system intrusion, such as forformed within 48 h after sample collection. In the delayed component replacement, removal, or addition. The decisiontesting scenario, the recovered microbial levels may not be to use longer incubation periods should be made after bal-the same as would have been recovered had the testing ancing the need for timely information and the type of cor-been performed shortly after sample collection. Therefore, rective actions required when an alert or action level is ex-studies should be performed to determine the existence and ceeded with the ability to recover the microorganisms ofacceptability of potential microbial enumeration aberrations interest.caused by protracted testing delays. The advantages gained by incubating for longer times,

namely recovery of injured microorganisms, slow growers,or more fastidious microorganisms, should be balancedThe Classical Culture Approach against the need to have a timely investigation and to takecorrective action, as well as the ability of these microorgan-Classical culture approaches for microbial testing of water isms to detrimentally affect products or processes. In noinclude but are not limited to pour plates, spread plates, case, however, should incubation at 30°–35° be less than 48membrane filtration, and most probable number (MPN) h or less than 96 h at 20°–25°.tests. These methods are generally easy to perform, are less Normally, the microorganisms that can thrive in extremeexpensive, and provide excellent sample processing environments are best cultivated in the laboratory usingthroughput. Method sensitivity can be increased via the use conditions simulating the extreme environments from whichof larger sample sizes. This strategy is used in the mem- they were taken. Therefore, thermophilic bacteria might bebrane filtration method. Culture approaches are further de- able to exist in the extreme environment of hot pharmaceu-fined by the type of medium used in combination with the tical water systems, and if so, could only be recovered andincubation temperature and duration. This combination cultivated in the laboratory if similar thermal conditionsshould be selected according to the monitoring needs were provided. Thermophilic aquatic microorganisms do ex-presented by a specific water system as well as its ability to ist in nature, but they typically derive their energy forrecover the microorganisms of interest: those that could growth from harnessing the energy from sunlight, from oxi-have a detrimental effect on the product or process uses as dation/reduction reactions of elements such as sulfur orwell as those that reflect the microbial control status of the iron, or indirectly from other microorganisms that do derivesystem. their energy from these processes. Such chemical/nutritionalThere are two basic forms of media available for tradi- conditions do not exist in high-purity water systems,tional microbiological analysis: “high nutrient” and “low nu- whether ambient or hot. Therefore, it is generally consid-trient”. High-nutrient media, such as plate count agar ered pointless to search for thermophiles from hot pharma-(TGYA) and m-HPC agar (formerly m-SPC agar), are in- ceutical water systems owing to their inability to growtended as general media for the isolation and enumeration there.of heterotrophic or “copiotrophic” bacteria. Low-nutrient The microorganisms that inhabit hot systems tend to bemedia, such as R2A agar and NWRI agar (HPCA), may be found in much cooler locations within these systems, forbeneficial for isolating slow-growing “oligotrophic” bacteria example, within use-point heat exchangers or transfer hoses.and bacteria that require lower levels of nutrients to grow If this occurs, the kinds of microorganisms recovered areoptimally. Often some facultative oligotrophic bacteria are usually of the same types that might be expected from am-able to grow on high-nutrient media and some facultative bient water systems. Therefore, the mesophilic microbial cul-copiotrophic bacteria are able to grow on low-nutrient me- tivation conditions described later in this chapter are usuallydia, but this overlap is not complete. Low-nutrient and adequate for their recovery.high-nutrient cultural approaches may be concurrently used,




Pour Plate Method or“Instrumental” ApproachesDrinking Water Membrane Filtration Methoda

Examples of instrumental approaches include microscopic Sample volume—1.0 mL minimumb

visual counting techniques (e.g., epifluorescence and immu- Growth medium—Plate Count Agarc

nofluorescence) and similar automated laser scanning ap- Incubation time—48–72 h minimumproaches and radiometric, impedometric, and biochemically Incubation temperature—30°–35°based methodologies. These methods all possess a variety of Pour Plate Method oradvantages and disadvantages. Advantages could be their Purified Water Membrane Filtration Methoda

precision and accuracy or their speed of test result availabil- Sample volume—1.0 mL minimumb

ity as compared to the classical cultural approach. In gen- Growth medium—Plate Count Agarc

eral, instrument approaches often have a shorter lead time Incubation time—48–72 h minimumfor obtaining results, which could facilitate timely system Incubation temperature—30°–35°control. This advantage, however, is often counterbalanced

Water for Injection Membrane Filtration Methodaby limited sample processing throughput due to extended

Sample volume—100 mL minimumbsample collection time, costly and/or labor-intensive sampleGrowth medium—Plate Count Agarcprocessing, or other instrument and sensitivity limitations.Incubation time—48–72 h minimumFurthermore, instrumental approaches are typically de-Incubation temperature—30°–35°structive, precluding subsequent isolate manipulation for

a A membrane filter with a rating of 0.45 µm is generally consideredcharacterization purposes. Generally, some form of microbialpreferable, although the cellular width of some of the bacteria in theisolate characterization, if not full identification, may be asample may be narrower than this. The efficiency of the filtrationrequired element of water system monitoring. Conse-process still allows the retention of a very high percentage of thesequently, culturing approaches have traditionally been pre-smaller cells and is adequate for this application. Filters with smallerferred over instrumental approaches because they offer aratings may be used if desired, but for a variety of reasons the abilitybalance of desirable test attributes and post-test capabilities.of the retained cells to develop into visible colonies may be compro-mised, so count accuracy must be verified by a reference approach.Suggested Methodologies b When colony counts are low to undetectable using the indicatedminimum sample volume, it is generally recognized that a larger

The following general methods were originally derived sample volume should be tested in order to gain better assurancefrom Standard Methods for the Examination of Water and that the resulting colony count is more statistically representative.Wastewater, 17th Edition, American Public Health Association, The sample volume to consider testing is dependent on the user’sWashington, DC 20005. Although this publication has un- need to know (which is related to the established alert and actiondergone several revisions since its first citation in this chap- levels and the water system’s microbial control capabilities) and theter, the methods are still considered appropriate for estab- statistical reliability of the resulting colony count. In order to test alishing trends in the number of colony-forming units larger sample volume, it may be necessary to change testing tech-observed in the routine microbiological monitoring of phar- niques, e.g., changing from a pour plate to a membrane filtrationmaceutical waters. It is recognized, however, that other approach. Nevertheless, in a very low to nil count scenario, a maxi-combinations of media and incubation time and tempera- mum sample volume of around 250–300 mL is usually considered ature may occasionally or even consistently result in higher reasonable balance of sample collecting and processing ease and in-numbers of colony-forming units being observed and/or dif- creased statistical reliability. However, when sample volumes largerferent species being recovered. than about 2 mL are needed, they can only be processed using the

The extended incubation periods that are usually required membrane filtration method.by some of the alternative methods available offer disadvan- c Also known as Standard Methods Agar, Standard Methods Platetages that may outweigh the advantages of the higher Count Agar, or TGYA, this medium contains tryptone (pancreatic di-counts that may be obtained. The somewhat higher base- gest of casein), glucose, and yeast extract.line counts that might be observed using alternate culturalconditions would not necessarily have greater utility in de- IDENTIFICATION OF MICROORGANISMStecting an excursion or a trend. In addition, some alternatecultural conditions using low-nutrient media tend to lead to Identifying the isolates recovered from water monitoringthe development of microbial colonies that are much less methods may be important in instances where specificdifferentiated in colonial appearance, an attribute that mi- waterborne microorganisms may be detrimental to thecrobiologists rely on when selecting representative microbial products or processes in which the water is used. Microor-types for further characterization. It is also ironical that the ganism information such as this may also be useful whennature of some of the slow growers and the extended incu- identifying the source of microbial contamination in a prod-bation times needed for their development into visible colo- uct or process. Often a limited group of microorganisms isnies may also lead to those colonies being largely nonviable, routinely recovered from a water system. After repeated re-which limits their further characterization and precludes covery and characterization, an experienced microbiologisttheir subculture and identification. may become proficient at their identification based on only

Methodologies that can be suggested as generally satis- a few recognizable traits such as colonial morphology andfactory for monitoring pharmaceutical water systems are as staining characteristics. This may allow for a reduction in thefollows. However, it must be noted that these are not refe- number of identifications to representative colony types, or,ree methods nor are they necessarily optimal for recovering with proper analyst qualification, may even allow testingmicroorganisms from all water systems. The users should shortcuts to be taken for these microbial identifications.determine through experimentation with various approacheswhich methodologies are best for monitoring their water

ALERT AND ACTION LEVELS ANDsystems for in-process control and quality control purposesas well as for recovering any contraindicated species they SPECIFICATIONSmay have specified.

Although the use of alert and action levels is most oftenassociated with microbial data, they can be associated withany attribute. In pharmaceutical water systems, almost every




quality attribute, other than microbial quality, can be very corrective actions can immediately be taken to bring therapidly determined with near-real time results. These short- process back into its normal operating range. Such remedialdelay data can give immediate system performance feed- actions should also include efforts to understand and elimi-back, serving as ongoing process control indicators. How- nate or at least reduce the incidence of a future occurrence.ever, because some attributes may not continuously be A root cause investigation may be necessary to devise anmonitored or have a long delay in data availability (like mi- effective preventative action strategy. Depending on the na-crobial monitoring data), properly established Alert and Ac- ture of the action level excursion, it may also be necessarytion Levels and Specifications can serve as an early warning or to evaluate its impact on the water uses during that time.indication of a potentially approaching quality shift occur- Impact evaluations may include delineation of affectedring between or at the next periodic monitoring. In a vali- batches and additional or more extensive product testing. Itdated water system, process controls should yield relatively may also involve experimental product challenges.constant and more than adequate values for these moni- Alert and action levels should be derived from an evalua-tored attributes such that their Alert and Action Levels and tion of historic monitoring data called a trend analysis.Specifications are infrequently broached. Other guidelines on approaches that may be used, ranging

As process control indicators, alert and action levels are from “inspectional” to statistical evaluation of the historicaldesigned to allow remedial action to occur that will prevent data have been published. The ultimate goal is to under-a system from deviating completely out of control and pro- stand the normal variability of the data during what is con-ducing water unfit for its intended use. This “intended use” sidered a typical operational period. Then, trigger points orminimum quality is sometimes referred to as a “specifica- levels can be established that will signal when future datation” or “limit”. In the opening paragraphs of this chapter, may be approaching (alert level) or exceeding (action level)rationale was presented for no microbial specifications being the boundaries of that “normal variability”. Such alert andincluded within the body of the bulk water (Purified Water action levels are based on the control capability of the sys-and Water for Injection) monographs. This does not mean tem as it was being maintained and controlled during thatthat the user should not have microbial specifications for historic period of typical control.these waters. To the contrary, in most situations such speci- In new water systems where there is very limited or nofications should be established by the user. The microbial historic data from which to derive data trends, it is commonspecification should reflect the maximum microbial level at to simply establish initial alert and action levels based on awhich the water is still fit for use without compromising the combination of equipment design capabilities but below thequality needs of the process or product where the water is process and product specifications where water is used. It isused. Because water from a given system may have many also common, especially for ambient water systems, to mi-uses, the most stringent of these uses should be used to crobiologically “mature” over the first year of use. By theestablish this specification. end of this period, a relatively steady state microbial popula-

Where appropriate, a microbial specification could be tion (microorganism types and levels) will have been al-qualitative as well as quantitative. In other words, the num- lowed or promoted to develop as a result of the collectiveber of total microorganisms may be as important as the effects of routine system maintenance and operation, includ-number of a specific microorganism or even the absence of ing the frequency of unit operation rebeddings, backwash-a specific microorganism. Microorganisms that are known to ings, regenerations, and sanitizations. This microbial popula-be problematic could include opportunistic or overt patho- tion will typically be higher than was seen when the watergens, nonpathogenic indicators of potentially undetected system was new, so it should be expected that the datapathogens, or microorganisms known to compromise a pro- trends (and the resulting alert and action levels) will increasecess or product, such as by being resistant to a preservative over this “maturation” period and eventually level off.or able to proliferate in or degrade a product. These micro- A water system should be designed so that performance-organisms comprise an often ill-defined group referred to as based alert and action levels are well below water specifica-“objectionable microorganisms”. Because objectionable is a tions. With poorly designed or maintained water systems,term relative to the water’s use, the list of microorganisms the system owner may find that initial new system microbialin such a group should be tailored to those species with the levels were acceptable for the water uses and specifications,potential to be present and problematic. Their negative im- but the mature levels are not. This is a serious situation,pact is most often demonstrated when they are present in which if not correctable with more frequent system mainte-high numbers, but depending on the species, an allowable nance and sanitization, may require expensive water systemlevel may exist, below which they may not be considered renovation or even replacement. Therefore, it cannot beobjectionable. overemphasized that water systems should be designed for

As stated above, alert and action levels for a given process ease of microbial control, so that when monitored againstcontrol attribute are used to help maintain system control alert and action levels, and maintained accordingly, theand avoid exceeding the pass/fail specification for that attri- water continuously meets all applicable specifications.bute. Alert and action levels may be both quantitative and An action level should not be established at a level equiv-qualitative. They may involve levels of total microbial counts alent to the specification. This leaves no room for remedialor recoveries of specific microorganisms. Alert levels are system maintenance that could avoid a specification excur-events or levels that, when they occur or are exceeded, indi- sion. Exceeding a specification is a far more serious eventcate that a process may have drifted from its normal operat- than an action level excursion. A specification excursion maying condition. Alert level excursions constitute a warning trigger an extensive finished product impact investigation,and do not necessarily require a corrective action. However, substantial remedial actions within the water system thatalert level excursions usually lead to the alerting of person- may include a complete shutdown, and possibly even prod-nel involved in water system operation as well as QA. Alert uct rejection.level excursions may also lead to additional monitoring with Another scenario to be avoided is the establishment of anmore intense scrutiny of resulting and neighboring data as arbitrarily high and usually nonperformance-based actionwell as other process indicators. Action levels are events or level. Such unrealistic action levels deprive users of mean-higher levels that, when they occur or are exceeded, indi- ingful indicator values that could trigger remedial systemcate that a process is probably drifting from its normal oper- maintenance. Unrealistically high action levels allow systemsating range. Examples of kinds of action level “events” in- to grow well out of control before action is taken, whenclude exceeding alert levels repeatedly; or in multiple their intent should be to catch a system imbalance before itsimultaneous locations, a single occurrence of exceeding a goes wildly out of control.higher microbial level; or the individual or repeated recovery Because alert and action levels should be based on actualof specific objectionable microorganisms. Exceeding an ac- system performance, and the system performance data aretion level should lead to immediate notification of both QA generated by a given test method, it follows that those alertand personnel involved in water system operations so that and action levels should be valid only for test results gener-




ated by the same test method. It is invalid to apply alert performance testing of semisolid drug products. Generaland action level criteria to test results generated by a differ- chapter Topical and Transdermal Drug Products—Productent test method. The two test methods may not equiva- Quality Tests ⟨3⟩ provides information related to productlently recover microorganisms from the same water samples. quality tests for topical and transdermal dosage forms, DrugSimilarly invalid is the use of trend data to derive alert and Release ⟨724⟩ provides procedures and details for testingaction levels for one water system, but applying those alert drug release from transdermal systems, and this chapterand action levels to a different water system. Alert and ac- ⟨1724⟩ provides procedures for determining drug releasetion levels are water system and test method specific. from semisolid dosage forms.

Nevertheless, there are certain maximum microbial levelsabove which action levels should never be established.

INTRODUCTIONWater systems with these levels should unarguably be con-sidered out of control. Using the microbial enumeration

This chapter provides general information for in vitro test-methodologies suggested above, generally considered maxi-ing of semisolid drug products. Semisolid dosage forms in-mum action levels are 100 cfu/mL for Purified Water and 10clude creams, ointments, gels, and lotions. Semisolid dosagecfu/100 mL for Water for Injection. However, if a given waterforms may be considered extended-release preparations,system controls microorganisms much more tightly thanand their drug release depends largely on the formulationthese levels, appropriate alert and action levels should beand manufacturing process. The release rate of a givenestablished from these tighter control levels so that they canproduct from different manufacturers is likely to be different.truly indicate when water systems may be starting to trend

out of control. These in-process microbial control parame-ters should be established well below the user-defined mi- DRUG PRODUCT QUALITY ANDcrobial specifications that delineate the water’s fitness for

PERFORMANCE TESTSuse.Special consideration is needed for establishing maximum

A USP drug product monograph contains tests, analyticalmicrobial action levels for Drinking Water because the waterprocedures, and acceptance criteria. Drug product tests areis often delivered to the facility in a condition over whichdivided into two categories: (1) those that assess generalthe user has little control. High microbial levels in Drinkingquality attributes, and (2) those that assess product perfor-Water may be indicative of a municipal water system upset,mance, e.g., in vitro release of the drug substance from thebroken water main, or inadequate disinfection, and there-drug product. Quality tests assess the integrity of the dos-fore, potential contamination with objectionable microor-age form, but performance tests, such as drug release, as-ganisms. Using the suggested microbial enumeration meth-sess attributes that relate to in vivo drug performance.odology, a reasonable maximum action level for DrinkingTaken together, quality and performance tests are intendedWater is 500 cfu/mL. Considering the potential concern forto ensure the identity, strength, quality, purity, comparabil-objectionable microorganisms raised by such high microbiality, and performance of semisolid drug products.levels in the feed water, informing the municipality of the

Details of drug product quality tests for semisolid drugproblem so they may begin corrective actions should be anproducts can be found in chapter ⟨3⟩. Product performanceimmediate first step. In-house remedial actions may or maytests for semisolid drug products are conducted to assessnot also be needed, but could include performing additionaldrug release from manufactured pharmaceutical dosagecoliform testing on the incoming water and pretreating theforms. In vitro performance tests for semisolid products dowater with either additional chlorination or UV light irradia-not, however, directly predict the in vivo performance oftion or filtration, or a combination of approaches.drugs, as the primary factor that impacts bioavailability andclinical performance are the barrier properties of the epithe-lia to which the product is applied (epidermal or mucosaltissues). Although product performance tests do not directlymeasure bioavailability and relative bioavailability (bioe-quivalence), they can detect in vitro changes that may cor-respond to altered in vivo performance of the dosage form.These changes may arise from changes in physicochemicalcharacteristics of the drug substance and/or excipients or toAdd the following:the formulation itself, changes in the manufacturing process,shipping and storage effects, aging effects, and other for-mulation and/or process factors.■⟨1724⟩ SEMISOLID DRUG

At present, a product performance test is available to eval-uate in vitro drug release for creams, ointments, lotions, andPRODUCTS—PERFORMANCEgels. Several available apparatus can be used for this evalua-tion, including the vertical diffusion cell, immersion cell, andTESTSa special cell used with USP Apparatus 4. Because of thesignificant impact of in vitro test parameters, such as releasemedia, porous membrane and dosing, and the interactionof these parameters with a given drug product, the primaryuse of in vitro drug release testing is comparison testing inSCOPEwhich any difference in delivery rate is undesirable. Drugrelease testing is most suitable for evaluation of small formu-The scope of this general chapter is to provide generallation and process changes, manufacturing site changes,information for performance testing of semisolid drug prod-and stability testing. The evaluation or comparison of largeucts, various types of equipment employed for such testing,formulation changes may provide unmeaningful results, un-and potential applications of the performance testing.less extensive validation is performed to select test parame-ters that ensure that the sensitivity of the test is meaning-

PURPOSE fully correlated with in vivo performance. The only requiredregulatory use of the in vitro release test is to determine the

This chapter provides general information about perfor- acceptability of minor process and/or formulation changesmance testing of semisolid drug products, the theory and in approved semisolid dosage forms (see FDA Guidance forapplications of such testing, information about the availabil- Industry—Nonsterile Semisolid Dosage Forms—Scale-Up andity of appropriate equipment, and likely developments in Postapproval Changes: Chemistry, Manufacturing, and Con-



12300

Documents

Transcript of 12300