Downloaded from on November 3, 2023

One-way Distribution System for Water for Injection: Process

Management, Microbiological Quality Control, and Meeting
Regulatory Requirements
Michael Jahnke

PDA J Pharm Sci and Tech 2001, 55 3-9

 Downloaded from on November 3, 2023

TECHNOLOGY/APPLICATION

One-way Distribution System for Water for Injection:

Process Management, Microbiological Quality Control,
and Meeting Regulatory Requirements

Michael Jahnke

Pharma Hameln GmbH

SHORT TITLE: Process Management and Control of One-way Distribution Systems for WFI

ABSTRACT: The specifications for pharmaceutical water, the qualification and validation of water preparation facili-
ties, strategies to prevent contamination by water-borne bacteria and lastly, the monitoring of microbiological purity
are the topics of frequent seminars on Pharma Water Total Quality Management. The same subdivisions are used in
the following paper on the process management and microbiological control of a one-way distribution system for
Water for Injection.

Since 1990, such a system has been in use in the production department of Pharma Hameln GmbH, a contract manu-
facturer of parenterals. Using this system as an example, the twin needs for flawless microbiological process control
and for suitable measures to monitor water quality are discussed, which, together with extensive documentation of the
qualification of the production and distribution system, ultimately led to acceptance of the system by regulatory authorities.

KEYWORDS: Pharmaceutical water, Water for injection, Quality assurance, Microbiological monitoring, Trend analy-
sis, Method validation

Requirements of a One-way Distribution System for The system for producing WFI in the production area
Water for Injection for parenteral dosage forms of Pharma Hameln GmbH
consists of a multiple-effect still, a storage tank, and a
The main requirements of a distribution network for cascade of heat exchangers and delivery pumps to pump
Water for Injection (WFI) in the production area for the WFI produced via three one-way systems to the use
parenteral pharmaceuticals generally include a perma- sites (Figure 1). The still is fed with water of pharma-
nent supply of hot and cooled WFI, storage of the wa- ceutical quality: Purified Water (Aqua Purificata),
ter at 85 +/- 5ºC, and supply of both cooled and hot which is obtained from the public water supply and is
WFI to a large number of points of use. In addition, it then subjected to softening and dechlorinating pro-
must be possible to completely empty the distribution cesses, reverse osmosis, and ion exchange chromatog-
system and carry out a safe and reliable sterilization in raphy. Cooled WFI and hot water of WFI quality are
the event of contamination [1, 2]. available at all legs for cleaning purposes, preparation
of solutions. and disinfection measures.

Correspondence to author should be addressed to: Sanitization Measures

Pharma Hameln GmbH, Quality Assurance - Microbi-
ology, Langes Feld 13, D-31789 Hameln. E-mail: In view of the need for consistent water quality, the deci-
m.jahnke@pharma-hameln.de sion to use a one-way system leads to far-reaching require-
ments regarding process management and control [3]. For

Vol. 55, No. 1, January/ February 2001 3

 Downloaded from on November 3, 2023

instance, regular flushing must be undertaken to decon- demonstrating a cycle of at least 121°C for at least 15
taminate the distribution system [4]. Distillation un- min. at all user points (valves). Sterilization demon-
doubtedly supplies sterile water; experience shows, strated lethal effects on endospores derived from Ba-
however, that this does not remain free of microorgan- cillus stearothermophilus at spore counts of 10 6 spores
isms even when stored at 80°C and with a complete with D-values of greater 1.5 min. at 121°C. For time-
reduction of nutrient supply [5]. Accordingly, suitable managed sterilization, the heating-up time of the net-
measures for routine sanitization of the distribution work has to be determined in order to establish the steam
network have to be introduced. inflow time required. This latter time should be checked
with regard to a successful sterilization using
In addition to steam sterilization, flushing with hot bioindicators with a population density of 10 6 spores
water at 80°C every working day is intended to prevent per indicator system. The heat resistance of these
contamination of the distribution system. A standard bioindicators should be reflected in killing kinetics that
operating procedure was drawn up which specifies the correspond to an effect of saturated steam at a tempera-
flushing times and the amount of water to be used for ture of 121ºC for at least 15 min. (F0121oC = 15 min.).
each section of the system. The aim of this flushing During steam sterilization of the distribution network,
procedure is to exchange the volume of water present the temperature is recorded by valve PT100 tempera-
in the pipe network at least three times. The flow engi- ture probes placed in the bleeder valves, condensate
neering calculations required for this process were veri- separators, and also at other suitable points (i.e., those
fied by the German Technical Testing Organization determined during qualification as the most sluggish).
(TÜV), on the basis of the blueprint (detailed drawing) Stainless steel adapters with a two-part inner chamber
of the distribution system, which enabled the length and separated by a perforated plate are used to hold the
diameter of the piping in the individual sections of the bioindicators. Bioindicators are placed in one half
three main sections and the various subsections to be whilst steam can pass freely through the other cham-
determined. ber, ensuring contact with the bioindicators. During
the steaming procedure, these adapters are arranged in
For each flushing phase, the volumes necessary for a series with condensate separators (Figure 2).
three-fold exchange of water were calculated from the
geometry of the individual piping sections. These vol- With regard to gaining acceptance by regulatory authori-
umes are checked by mobile, calibrated flow meters and ties, the qualification of steaming measures is of spe-
documented every working day. cial importance. Although one-way distribution sys-
tems are not ruled out per se [3], there is nonetheless
The initiation of the daily flushing procedure is con- some initial reservations concerning the reliability of
trolled automatically. The flushing time is shown, and the process management of such systems. This is also
an additional flushing procedure of the piping from the reflected, for instance, in the demand for suitable ster-
outlet of the WFI storage tank to the distribution points ilization measures if (one-way) pipe lengths exceed six
in the three main sections is undertaken via the opera- times the pipe diameter [3].
tion of various valves. If 48 h have elapsed since the
last flushing phase, the feed of WFI into the respective Microbiological Monitoring and Trend Analyses
main section is automatically stopped via the control
of this “source sanitization.” The 48-h time bar was Microbiological monitoring serves to check the quality
considered suitable on the basis of trend analyses of of the process water and provides early warning of de-
the microbiological monitoring results. terioration in quality. The sample volumes, warning,
and action limits and measures to be introduced if these
Qualification of Steam Sterilization Measures values are exceeded must be specified (Table 1) and
documented in standard operation procedures (SOP) and
Steam sterilization of the production, storage, and dis- validation protocol. As part of the routine testing car-
tribution system must be possible in case trend analy- ried out on water production and distribution systems,
ses and/or microbiological results indicate contamina- water samples are taken daily and tested for their mi-
tion, or following work to the piping system [6]. Quali- crobial content, for the absence of Escherichia coli as
fication of sterilization measures was thus performed faecal indicator, for the absence of Pseudomonas

4 PDA Journal of Pharmaceutical Science & Technology

 Figure 1: Diagram of the production and distribution system for Water for Injection (WFI) in the production area.

Distribution Panel
00001 00001 00001 00001 00001 00001
Heat Exchanger X0101 X0104 X0105 X0106 X0107 X0108
Storage Tank
Pumps
Distillation

A2 A3 A4 A6 A7 A8

Vol. 55, No. 1, January/ February 2001

A10 A11 A12 A14 A15 A16 A17 J11
00001
X0207

J15
A18
A19
K1 A1 00001 00001 00001 00001 00001 00001
X0201 X0202 X0204 X0205 X0206 X0208 A20
Downloaded from on November 3, 2023

00001
J1 J7 J8 X0313
J2
J5 J6
J4 J10
00001
J9 X0311
00001
J13
X0312

J14

00001 00001 00001 00001 00001 00001 00001 00001 00001 00001 00001
X0301 X0302 X0303 X0304 X0305 X0317 X0306 X0307 X0308 X0309 X0310

5
 Downloaded from on November 3, 2023

Figure 2: Plan for steam sterilization of a one-way distribution system for WFI.
Clean steam was supplied via valve no. 1 x 0032 and bled by subsequent opening of the valves equipped with terminal
condenser valves. Thus all user points were accessible by clean steam of at least 121° C for a time period of at least 15 min.
A temperature probe was installed at valve no. 1 x 0206, representing the coolest spot of this distribution line.

00001 00001 00001 00001 00001 00001

X0201 X0202 X0204 X0205 X0206 X0207

TIR

00001
X0032

Clean Steam

Stainless Steel Pipings

00001
X0019
00001
X0024 Condenser Valve
00001
X0029 TIR
Temperature Probe

Table 1: Quality control requirements on daily water testing, based on national and international pharmacopoeias.
In addition to the quantity of samples required for differing water qualities, warning and action limits for microbial counts
and endotoxin content of a water sample are specified. In addition, tests are undertaken for the absence of enterobacteria and
pseudomonads.

Requirements on Daily Water Testing

(in house specification on basis of pharmacopoeial requirements)

WFI DIW Pure steam Mains water

Sample Volume: [ml] 300 100 100 (condensate) 100

Warning level: 1 CFU / 100 ml 10 CFU/ ml Spore-formers 50 CFU/ml

Action limit: 10 CFU / 100 ml 50 CFU / ml Sterile 100 CFU / ml

Enterobacteria: E. coli negative E. coli negative — E. coli negative

Pseudomonads: P. aeruginosa negative P. aeruginosa negative — P. aeruginosa negative

Endotoxin [EU / ml ]

Warning level: > 0.125 < 0.25 > 0.25 < 0.5 > 0.125 < 0.25 record

Action limit: 0.25 — 0.25

6 PDA Journal of Pharmaceutical Science & Technology

 Downloaded from on November 3, 2023

aeruginosa, and for their endotoxin content. A sam- record of the water system quality is achieved. Under
pling plan ensures that all distribution legs are covered these conditions, a validation period of 10 days, based on
daily. A more intensive sampling is undertaken as part the minimum time for a process validation, is quite ac-
of process validation: over a period of at least 10 pro- ceptable.
duction days, all sites of use are sampled and tested for
chemical and physical properties and microbiological The data relating to the microbial load of the water sys-
contamination. Over a minimum of 10 production days, tem are tabulated and evaluated graphically (Figure 3).
a twice-yearly process validation of the reverse osmo- The results may, for example, show that the hygiene
sis plant (feed water for the distillation) also takes place. status at various sampling sites differs. Once the cause
If the process validations also take into account poten- of such variation has been identified, remedial action can
tial seasonal variations and a routine monitoring plan be taken. For instance, unusually high microbial counts
(endotoxins, total microbial count, and detection of or endotoxins exceeding warning levels at a particular site
problem organisms) is followed, then a comprehensive may be an early indication that valves are leaking.

Figure 3: Example of trend analysis.

As an example, the data are plotted as colony forming units (CFU) per 100 ml sample volume on the ordinate against
the working day (abscissae) for use sites 1.1 to 1.8 and for the storage tank (Tank) for WFI.
The figure provides data for a year’s working days, demonstrating daily monitoring of the storage tank and monitoring
of user points 1.1. to 1.8 according to a sampling plan. The maximum total count of microorganisms was 3 cfu/100 ml
in this example.
Within day 10 to 21 and day 143 to 158 validation of the system was performed resulting in condensed, e.g. daily
monitoring of all user points.

10

8
CFU / 100 ml

0 Tank
1 21 41 1,7
61 81 1,5
101 121
141 161 181 1,1
201 222
Working Day 242

Vol. 55, No. 1, January/ February 2001 7

 Downloaded from on November 3, 2023

Method Validation number and species range of bacteria and to trend ana-
lyze the results. The latent risk of development of
Even though the culture conditions for microorganisms biofilms is especially a focus of interest in distribution
potentially present in a water sample appear to have systems of process water. Reports on remarkable dif-
been validated, it is important to revalidate them, tak- ferences when analyzing species ranges of water
ing into account those organisms that are part of the samples taken from outlets of pipe wall taps and samples
normal environment of the production area [3]. Very taken from the center of water stream [7] support this
often, sublethally damaged populations characterized risk assessment.
by slow growth are present, or nonfastidious organisms,
which also grow slowly in a nutrient-rich medium. The In this respect, it appears to be a preconceived idea that
incubation conditions (temperature and time) should biofilms are more prominent in one-way/dead-leg sys-
not merely be suitable for the particular growth char- tems rather than circulating systems: “obviously, water
acteristics of an organism, but should also allow the in constant motion is less liable to have high levels of
possibility of subsequent identification. When check- contamination” [3].
ing the growth-promoting properties of the culture
media, it is important to use the in-house populations However, in view of microbiological status — and this
as well as the pharmacopoeial (especially the USP) ref- is true for all kinds of distribution systems —only the
erence organisms. following statements are valid:

• Bacterial contamination may occur sooner or later.

Furthermore, the method for detecting low microbial
counts in WFI and of finding enterobacteria and • The extent of contamination is driven by the con-
pseudomonads must be validated. In practice, it has centration of nutrients, cell counts of feeding wa-
proved worthwhile to give staff engaged in microbial ter, and seasonal influences.
counts some training prior to a method validation. This
• The quality of the surface of piping determines the
can take the form of exercises in undertaking micro-
extent of potential biofilms.
bial counts of samples of known density and of finding
certain types of microorganisms deliberately planted • Microbiological control and sanitation measures in-
in a sample, or of identifying critical micro-organisms fluence the extent of potential biofilms.
in a mixed population.
When biofilms in high-quality water systems have been
Discussion demonstrated to exist, it turned out that control strate-
gies, e.g. irregular operation, lack of/inefficient sanita-
Since microbiological contamination of pharmaceuti- tion, and periodic sterilization, have failed [8]. As long
cal process water has been shown to be a major poten- as microbiological monitoring goes along with regular
tial risk of manufacturing problems and process fail- sanitation, one-way/dead-leg distribution systems can
ures, it is widely accepted to monitor and control the operate constantly with reliable, high quality water [9].

8 PDA Journal of Pharmaceutical Science & Technology

 Downloaded from on November 3, 2023

References

1. “Upgrading requirements for pharmaceutical 6. R. Noy, “Validation of pure water systems,”

grades of water: Proposed revision”, Pharmacopoe- Pharmac. Manufact. Rev., 6, 28-30 (1995).
ial Forum, 20, 7526-7545 (1994).

2. J. Cross, “Upgrading a pharmaceutical water puri- 7. M. P. E Patterson,. G. R. Husted, A. Rutkowski,

fication system to meet FDA requirements,” Mi- and D. C. Mayette “Bacteria - Isolation, identifi-
crobiology Europe, 4, 16-17 (1996). cation, and microscopic properties of biofilms in
high-purity water distribution systems,” Ultrapure
3. U.S. Food and Drug Administration (FDA), “Guide Water 8 (4), 18-24 (1997).
to Inspections of High Quality Water Systems,”
Rockville, Maryland (1993).

4. J. Cross, “Removing microbial contaminants from 8. W. P. Olson, “Biofilms in the pipeline and in the
process water.” Pharmac. Manufact. Rev., 9. 2-4 patient,” PDA J. Pharm. Sci. & Tech. 51 (6), 252
(1998). -260 (1997).

5. G. E. Martyak, J. C Carmody,. and G. R. Husted,

“Characterizing biofilm growth in deionized 9. M. Jahnke,. “Microbiological leak test of supply
ultrapure water piping systems, “ Microcontam- pipes for pharmaceutical process water,” European
ination, 11 No. 1, 39-44 (1993). J. Parenteral Sci., 5 (3), 75-78 (2000).

Vol. 55, No. 1, January/ February 2001 9

 Downloaded from on November 3, 2023

An Authorized User of the electronic PDA Journal of Pharmaceutical Science and

Technology (the PDA Journal) is a PDA Member in good standing. Authorized Users are
permitted to do the following:

·Search and view the content of the PDA Journal

·Download a single article for the individual use of an Authorized User
·Assemble and distribute links that point to the PDA Journal
·Print individual articles from the PDA Journal for the individual use of an Authorized User
·Make a reasonable number of photocopies of a printed article for the individual use of an
Authorized User or for the use by or distribution to other Authorized Users

Authorized Users are not permitted to do the following:

·Except as mentioned above, allow anyone other than an Authorized User to use or access the
PDA Journal
· Display or otherwise make any information from the PDA Journal available to anyone other than an
Authorized User
·Post articles from the PDA Journal on Web sites, either available on the Internet or an Intranet, or in
any form of online publications
·Transmit electronically, via e-mail or any other file transfer protocols, any portion of the PDA
Journal
·Create a searchable archive of any portion of the PDA Journal
·Use robots or intelligent agents to access, search and/or systematically download any portion of the
PDA Journal
·Sell, re-sell, rent, lease, license, sublicense, assign or otherwise transfer the use of the PDA
Journal or its content
·Use or copy the PDA Journal for document delivery, fee-for-service use, or bulk reproduction or
distribution of materials in any form, or any substantially similar commercial purpose
·Alter, modify, repackage or adapt any portion of the PDA Journal
·Make any edits or derivative works with respect to any portion of the PDA Journal including any text
or graphics
·Delete or remove in any form or format, including on a printed article or photocopy, any copyright
information or notice contained in the PDA Journal