Comparing and Contrasting

Barrier Isolator
Decontamination Systems
Jim Fisher and Ross A. Caputo*

I
n recent years, the use of barrier isolators in a variety of ap-
plications has increased. In 1998, R. Friedman called isola-
tor technology a “promising technology [that] may repre-
sent a significant stride forward in aseptic processing” (1).
PHARMACEUTICAL SYSTEMS, INC.

Isolators have become the approach of choice for sterility test-

ing (2, 3) and vapor-phase hydrogen peroxide (VHP) has be-
come the sterilant of choice. According to a survey conducted
by Lysfjord and Porter, 75% of surveyed participants indicated
that they used VHP for their decontamination processes (4). In
this context, a comparison was made of the validation, opera-
tion, and performance of two VHP generators, the “STERIS
VHP 1000ED” biodecontamination system and the BioQuell’s
“Clarus ‘C’” H2O2 gas generator shown in Figure 1.
A discussion of the validation and
operation of two commercially available Isolator equipment and decontamination systems
vapor-phase hydrogen peroxide Isolators. An isolator uses a biological barrier concept in which
decontamination systems is presented, the analyst or operator is physically isolated from the environ-
based on a hands-on examination of both ment. The isolator is maintained under positive or negative
systems. pressure (depending on the application) by HEPA-filtered air.
Using a glove port or half-suit, the analyst performs manipu-
lations in the isolator but is physically separated from the prod-
uct environment, so the risk of contamination during the pro-
cedure is minimized.
The choice of positive or negative pressure depends on the
application. If the contents of the isolator are carcinogenic or
radioactive, negative pressure is maintained to protect the user.
However, in most aseptic processing situations, positive pres-
sure is maintained to protect the contents from contamination
by the user. The US Food and Drug Administration (FDA)
aseptic processing guidance offers some basic guidelines for
validating isolators. According to these guidelines, the isolator
interior must maintain a Class 100 environment or better, and
the room surrounding an aseptic processing isolator should be
Ross A. Caputo, PhD, is the CEO of classified to ensure a consistent bioburden for the isolator’s
Pharmaceutical Systems, Inc. (PSI), 909
decontamination (5).
Orchard St., Mundelein, IL, 60060,
tel. 847.566.9229, fax 847.566.4960, The isolator’s control system maintains process set points
rcaputo@pharmsustems,com. Jim Fisher is such as the working pressure of the unit. Older isolators typi-
the vice-president of engineering at PSI. cally had dials that adjusted the fan speed and the pressure dif-
*To whom all correspondence should be addressed. ferential. Modern isolators generally have a programmable logic
controller (PLC) or a computer, and generally need less user
supervision because the controllers can maintain various set
68 Pharmaceutical Technology NOVEMBER 2004 www.phar mtech.com
points such as pressure and temperature. The controller also STERIS VHP1000ED biodecontamination system
can alert the user of any alarm conditions through audio and • Phase 1: Dehumidification. The water concentration in the
visual alarms, and can communicate with a VHP generator dur- isolator is reduced to the specified set point.
ing the decontamination cycle. • Phase 2: Conditioning. A hydrogen solution is injected
VHP generators. A VHP generator has its own PLC or com- into the air stream to rapidly increase the VHP concen-
puter controller that manages the various cycle set points and tration.
phase parameters, and communicates with the isolator. The • Phase 3: Decontamination. Hydrogen is injected into the
generators generally have four distinct phases of operation. Al- system to maintain the concentration necessary for decon-
though the nomenclature of the four phases differs from one tamination.
manufacturer to the next, they function in essentially the same • Phase 4: Aeration. All the VHP and water are removed.
way. These four phases are conditioning, ramping, decontam-
ination, and aeration. BioQuell Clarus “C” H2O2 gas generator
The conditioning phase of the generator cycle prepares the • Phase 1: Conditioning. The generator adjusts the relative
isolator environment for biodecontamination. Conditioning humidity and increases the temperature.
the isolator environment consists of drawing air from the iso- • Phase 2: Ramp gassing. Hydrogen peroxide solution is in-
lator through the generator to increase the temperature and jected into the air stream to increase the VHP concentra-
adjust the relative humidity (RH) to required levels. The tar- tion.
get conditions of temperature and RH differ in the two gener- • Phase 3: Dwell gassing. Hydrogen peroxide is injected into
ators studied; both generators, however, condition the isolator the system to maintain the concentration necessary for de-
environment to a desired temperature and relative humidity. contamination.
The second phase, ramping, uses elevated hydrogen perox- • Phase 4: Aeration. All the VHP and water are removed.
ide (H2O2) injection rates to rapidly raise the concentration of
VHP inside the isolator to the desired limit. This phase of VHP IQ and OQ validation tests
decontamination is analogous to the “come-up” time in a steam As with all processing equipment, installation qualification (IQ)
sterilization application. must be performed on any new VHP gas generator. During the
When the second phase is completed, the decontamination IQ, all of the purchase orders, mechanical specifications, and
phase begins. During the decontamination phase, the genera- drawings for the equipment should be reviewed and all the in-
tor maintains a specified VHP concentration in the isolator by struments must be calibrated. If the equipment has a PLC or
reducing the H2O2 injection rate. It is this phase of the genera- computer control, the version of the control software should
tor cycle that organisms are inactivated. During validation, this be verified. Users should conduct a software validation assess-
phase is often cut short to demonstrate overkill of the decont- ment to determine the extent of software validations necessary.
amination cycle. The final phase of the VHP generator cycle is All of the isolator’s materials of construction should be verified
the aeration phase, during which the VHP is removed from the for safe use with hydrogen peroxide. Finally, all of the supplied
isolator after decontamination. utilities (i.e., supplied electricity, compressed air, and exhaust
systems) should be checked to ensure that they are within the
Materials manufacturer’s specifications.
In this comparison study, a la Calhène three-glove transfer iso- In addition to the standard operational qualification (OQ)
lator was used. The isolator is a rigid-walled (glass and stain- tests to ensure the equipment’s operational functionality (i.e.,
less steel) component with PLC control (7) and was located in alarm and interlock testing, functionality testing, etc.), the OQ
a Class 10,000 environment. must include specific testing of the generator and isolator. Such
The isolator contents used for this comparison are typical testing includes leak-rate determination, pressure-differential
sterility-testing contents, including a “Steritest” device. The load testing, and ammonium hydroxide testing. Finally, the inter-
contents were placed on stainless steel wire racks to reduce the relationship between the generator and isolator must be tested.
amount of surface-to-surface contact within the isolator. If the generator and isolator control systems communicate with
The two VHP generators studied use different methods to each other, this communication must be challenged. Further
achieve biological inactivation. Although they both supply VHP more, the OQ must include a test that will determine if the gen-
to the interior of the isolator, the generators have different VHP erator progresses through all four phases of operation without
and water concentration goals. The manufacturer of the Claris any alarms.
“C,” BioQuell Pharma, states that the formation of micro- Both VHP generator suppliers offer validation services. In
condensation is the primary method of causing the deconta- both instances, it was found that additional supplemental test-
mination (8–10), which requires a high level of air saturation. ing was necessary to meet corporate guidelines for equipment
In contrast, the STERIS system targets higher VHP concentra- validation. The purchased vendor validation package, however,
tions while avoiding condensation of any kind in the isolator did cover many aspects of the equipment validation that would
(11,12). not need to be repeated in the supplemental validation pack-
To achieve these goals, the two generators use different cycle pa- age. It is recommended that supplier validation be included
rameters, which are based on the four basic generator phases. The with a new purchase of a gas generator, especially if it is the first
specific phases, as described by the manufacturers, are as follows: system of its kind being used by the owner.
70 Pharmaceutical Technology NOVEMBER 2004 www.phar mtech.com
 Operational tips and observations
In an effort to aid the industry with further validations of both generator systems cycle.The volume of the load containers, however, does affect the
and to improve their use,a list of tips and observations is included below. temperature response of the containers.For example, if a container holds 100
• The pressure fluctuations of the STERIS VHP 1000ED can be large enough to mL of liquid, the surface of the container will heat more slowly than if the
set off alarms when using small volume isolators (,30 ft3). same container holds 50 mL of liquid.The load containers should hold the
• A proporational integral derivative (PID) control algorithm in the VHP 1000ED maximum amount of liquid during the validation cycles to present the worst-
controls pressure fluctuations in the isolator.The factory PID settings for this case thermal load for the isolator.
control algorithm yielded pressure fluctuations in the test isolator from • The material properties of the load contents should be checked to ensure
20.05 to 1.5 in.of water control.PID settings are adjustable by the user and compatibility with VHP.Cellulosic and paper materials and some flexible
will yield a stable pressure fluctuation when properly adjusted.In several plastics absorb VHP.This will lower the VHP concentration in the isolator and
instances,however,the PID control settings were not stored in the internal cause the aeration phase to be extended as a result of off-gassing of the VHP
memory of the VHP 1000ED after user adjustment. by the material.VHP has a minimal effect on materials such as glass, stainless
• A software bug was found on the VHP 1000ED during the validation process steel, aluminum, and many hard plastics such as nonwoven high-density
that causes a valve to remain open during the dehumidification phase, polyethylene and polyvinyl chloride.
allowing the machine to occasionally inject VHP into the isolator, causing • Sufficient water content is required if Dräger tubes are used to measure
condensation in the interior of the isolator.Under these conditions, the cycle residual VHP concentration after aeration is completed to ensure that the
should be aborted.STERIS was contacted and made aware of the issue the VHP concentration is ,1 ppm.The Dräger tubes require 3–10 mg/L of water
company has made changes to the software to correct this bug.It is vapor in the air to give an accurate reading.If the air in the isolator is too dry,
suggested that the most current version of the software be installed on the a small petri dish of water can be placed into the isolator such that the
VHP 1000ED to eliminate this problem. readings can be taken directly above the water.
• The temperature profiling of the isolator interior is critical to a successful • A glove-holding device and half-suit hangers should be used to keep gloves
decontamination cycle.Successful temperature mapping will identify isolator and half-suits from contacting any surfaces during decontamination.Gloves
chill plates,which are areas of the isolators interior or walls that do not and half-suits are the primary method of transport of samples and test
increase in temperature as fast as the rest of the unit.For example,the surface instruments within the isolator.Therefore, great care should be taken to
of the isolator where the legs of the isolator connect to the body could ensure that every surface is decontaminated thoroughly.
become a chill plate:because the floor area above the leg contains a large • The room surrounding the isolator should be temperature controlled.
mass of steel that needs to be heated, this area may be cooler than the rest of Fluctuations in room temperature will cause fluctuations in the temperature
the isolator, causing localized condensation. of the isolator’s exterior surface, leading to condensation on the isolator’s
• During the validation of an isolator system,biological indicators (BIs), interior surfaces.Fluctuations in the isolator’s exterior surface can be caused
chemical indicators (CIs),and thermocouples should be hung throughout the by localized flow from the HVAC system or general fluctuations in room
isolator.Tape or adhesive must be placed carefully.Masking,autoclave,and temperature.Therefore, it is important to consider both the temperature
electrical tape will leave a sticky residue on the surface of the isolator,and can control of the room and the location of the HVAC vents.
affect the integrity of flexible PVC if it is left in the isolator for several cycles. • The STERIS VHP1000ED has a regeneration-scheduling feature programmed
Experience has shown that “3M Command”adhesive strips and hooks work into the controller.The user can schedule the VHP 1000ED to automatically
well for hanging thermocouples,BIs,and CIs. run a regeneration cycle at a specified time.This feature allows the unit to be
• The surface area of the load and the material is more important than the regenerated during off-peak hours.It is suggested that users carefully
volume and contents of the load.Because VHP only interacts with the outside consider the timing of the regeneration cycle to reduce the possibility of
surfaces of the load content, volume has little effect on the decontamination requiring the use of the machine during the regeneration cycle.

Equipment operational differences to five hours to complete and must be done after approximately
The Clarus “C” and VHP 1000ED generators have many fea- 1000 g of hydrogen peroxide have been injected. The Clarus “C”
tures that are similar although not identical. Both store multi- unit uses refrigeration principles to withdraw the water from
ple cycles, provide printouts, have alarms and safeties, and can the isolator. This system does not require regular regeneration.
send and receive remote input and output. The most impor- Single versus dual air flow loops. The Clarus “C” has two airflow
tant differences in the operation of the two machines include loops. One loop includes the catalyst and refrigeration unit for
the methods for VHP and water removal; the use of single or removing moisture and VHP from the air, and is used for the
dual airflow loops; the availability of a parametric gassing op- conditioning and aeration phases. The second loop is for the
tion; and the use of a wet or dry cycle. ramping and decontamination phases. The second loop does
VHP and water removal. Both isolators use a heavy-metal cata- not include the catalyst or refrigeration units, so the VHP and
lytic converter to break the VHP down into water and oxygen. water content are not removed from the isolator. The genera-
The VHP 1000ED uses a desiccant dryer unit that absorbs and tor vaporizes additional VHP and injects it into the stream of
holds the water removed from the isolator until the unit is re- air that already has VHP entrained.
generated. When the unit goes through its regeneration cycle, The VHP 1000ED uses a single-loop system. As a result, the
it heats up the dryer and passes heated air through it to remove generator removes all VHP and a majority of the water content
all moisture from the desiccant material. This cycle takes four from the air stream that the generator draws from the isolator.
72 Pharmaceutical Technology NOVEMBER 2004 www.phar mtech.com
 Wet versus dry decontamination cycle.
The major difference between the two
systems is that one uses a wet decon-
tamination cycle and the other uses a
dry cycle. The STERIS VHP 1000ED
operates on the principles that dry air
holds more VHP, and that the higher
the VHP concentration, the higher
the kill rate. Because the air inside the
isolator can only hold a finite amount
of VHP and water before condensa-
(a) (b) (c) tion begins, the VHP 1000ED dries
Figure 1: (a) The BioQuell Clarus “C” H2O2 gas generator, (b) the STERIS VHP 1000ED the air before gassing. This deconta-
biodecontamination system, and (c) the la Calhène isolator used in the study. mination method, known as the dry
cycle, allows the unit to produce a
higher VHP concentration without condensation.
The biodecontamination method of the BioQuell Clarus “C”
uses a wet-cycle with microcondensation. When hydrogen per-
oxide and water condense out of air they do not condense at
equal rates. In fact, the hydrogen peroxide condenses at a faster
rate than the water, creating a high hydrogen peroxide concen-
tration condensate. This condensate is believed to kill organ-
isms more quickly than VHP alone.
The microcondensation is purported to occur in extremely
small droplets that are invisible to the human eye, and is re-
quired over every surface of the isolator. A challenge in validat-
ing a wet cycle is showing that every surface has conditions suf-
ficient to yield microcondensation. Because of the high
saturation level of the air, and the fine line between micro- and
macro-condensation and temperature control of the room is
critical if a wet cycle is used.
Figure 2: Transfer isolator load.
Performance qualification (PQ)
The objective of PQ testing in an isolator decontamination ap-
plication is to demonstrate that the generator consistently biode-
contaminates the isolator using a specified cycle. During PQ,
decontamination is tested at the maximum and minimum iso-
lator load conditions.
The current aseptic processing guidance have left room for
interpretation in reference to the biological challenge during
validation, stating, “[Decontamination] cycles should be de-
veloped with an appropriate margin of extra kill to provide
confidence in robustness of the decontamination processes.
Normally, a four- to six-log reduction can be justified depend-
Figure 3: Flow trajectory plot. ing on the application” (5). Even though the type of biologic
challenge has not been addressed in the guidance, Geobacillus
During the ramping and decontamination phases, the unit in- stearothermophilus spores have become the industry-accepted
jects VHP into this stream of dry air to maintain the VHP con- organism for the decontamination challenge (6). A rather con-
centration within the isolator. servative approach was used in this biodecontamination vali-
Parametric gassing. The Clarus “C” has a parametric gassing dation. The cycles used in the validation are intended to pro-
option that allows the generator to monitor the decontamina- vide a minimum six-log reduction of a resistant biological
tion cycle and adjust parameters to meet decontamination ob- indicator (BI) when exposed to an overkill decontamination
jectives. BioQuell offers a condensation monitor that can au- cycle. G. stearothermophilus spores were used with a minimum
tomatically adjust the ramp gassing time based on the conditions initial concentration of 106 spores per indicator. The inactiva-
inside the isolator. This ensures the desired VHP and water con- tion of 106 spores at the three-quarter decontamination cycle
centration levels have been reached before the dwell-gassing time provides a theoretical total organism reduction of 108 or-
phase begins. ganisms for a full cycle.
74 Pharmaceutical Technology NOVEMBER 2004 www.phar mtech.com
 of the cycle. In VHP decontamination, the surface
conditions of the load are the most critical factors
to consider because VHP is a surface-decontam-
ination process. The first surface condition to con-
sider is the compatibility of the surface material
with VHP. Ideally the load contents should not
react with, absorb, or adsorb the VHP. Secondly,
the more surface area the VHP has to interact with,
the longer it will take to decontaminate all of the
Figure 4: Velocity cut plots. (Left) 10.5 in. from rear wall. (Right) 40 in. from left wall. surfaces. Another concern is overloading the iso-
lator. If the load is packed too tightly, there may
be areas of low VHP concentration within the load resulting
from to reduced airflow through the load.
The final consideration for loading an isolator is to avoid sur-
face-to-surface contact between items or parts of the load it-
self, and between the load and the isolator. Areas where surface-
to-surface contact occurs are less likely to receive sufficient VHP
concentration, and therefore can pose a potential for contam-
ination. To help avoid areas of surface-to-surface contact, every-
thing should be placed on stainless steel wire racks.
The load used to test the VHP 1000ED consisted of forty-six
250-mL bottles of rinse fluid and samples, eight 100-mL vials
of media, two “Steritest kits,” scissors, forceps, pens, and a Dräger
pump with tubes. A similar load was used to test with the Clarus
Figure 5: Location of thermocouples, biological indicators, and “C” generator. The majority of the 250-mL bottles were placed
chemical indicators. on the bottom shelf of the two-shelf rack. The rest of the items
were placed on the top shelf. The loading configuration is shown
Table I: Settings used for each cycle during performance in Figure 2.
qualification.
Conducting CFD and smoke studies
Steris VHP 1000ED biodecontamination system
CFD is a finite element analysis for determining air patterns
Airflow 20 SCFM
and temperature distributions around or inside a given model.
Phase I: Dehumidification 15 min and RH ,4.6 mg/L The most recognizable use for this software is in the automo-
Phase II: Conditioning 2 min at 5.6 g/min bile and aerospace industries, where air patterns around a car
Phase III: Decontamination 34 min at 3.5 g/min or over the wing of an airplane can be modeled and analyzed.
Phase IV: Aeration 2 to 5 hours This technology can be used to analyze air patterns inside an
isolator. Once the patterns are established, the areas of worst
Bioquell Clarus “C” H2O2 gas generator flow or temperature can be found.
Airflow 500 L/min (~18 SCFM) There are four basic steps in conducting a CFD analysis on
Phase I: Conditioning 10 min and 40% RH an isolator. The first step is to create a three-dimensional solid
Phase II: Ramp gassing 15 min at 1.5 g/min model of the isolator, its components, and the placement of
Phase III: Dwell gassing 12 min at 1.1 g/min load inside the isolator. Once the system’s physical shapes and
Phase IV: Aeration 2.5 to 3 h dimensions are defined, the initial fluid conditions must be es-
The cycle times listed above are the full cycle times. During the tablished. These initial fluid conditions include properties such
PQ, the VHP 1000ED's Phase III and the Clarus's Phases II and III as temperature and flow rates of the inlet, the exhaust, any dis-
were run at three quarters of the times listed above. tribution fans, the isolator walls, and the air itself. When all the
initial conditions have been defined, the analysis is performed.
Several tests must be completed in an isolator PQ. The first The flow trajectories, the flow velocities, and the tempera-
step is to define the isolator load conditions for both minimum tures throughout the isolator are the three most important re-
and maximum loading. Next, the user must conduct a compu- sults of a CFD in an isolator application. VHP is distributed
tational fluid dynamics analysis (CFD) and/or smoke studies through the isolator by two methods: it can be directly car-
to determine the “worst case” airflow locations in the isolator. ried on the air stream, and it also can be spread by diffusion
The last step of the PQ is the biological challenge at both the from areas of high concentration to areas of low concentra-
minimum and the maximum loads to prove that the stated goals tion. It is preferable for the VHP to be carried on the air stream
of 106 spore reduction in a cycle time are met. to the decontamination site because it is distributed faster this
Defining the load. As with most other decontamination or ster- way.
ilization processes, load determination is critical to the efficacy The airflow trajectory plot generated from the CFD analy-
76 Pharmaceutical Technology NOVEMBER 2004 www.phar mtech.com
 Total injected hydrogen peroxide vs time STERIS VHP1000ED and BioQuell Clarus "C" hydrogen
140 peroxide concentration vs time during the gassing phases
VHP 10000ED 2.0
Total injected peroxide (grams)

120 Clarus "C"

Peroxide concentration (mg/L)

Steris decon start 1.8
Clarus dwell start
100 1.6
1.4
80
1.2
60 1.0 NH H202 Clarus
0.8 FL H202 STERIS
40 FL H202 Clarus
0.6 Start of Decon
0.4 Start of Dwell
20 Clarus gassing end
0.2 NL H202 STERIS
0 0.0
0 4 8 12 16 20 24 28 32 36 0.0 3.0 6.0 9.0 12.0 15.0 10.0 21.0 24.0 27.0 30.0 33.0 36.0

Time (h) Time (min)

Figure 6: Total injected aqueous H2O2 versus time. Figure 7: VHP concentration versus time.

sis, showed the approximate paths of air through the isolator

and its velocity at each point (see Figure 3). The plot shows that STERIS VHP1000ED and BioQuell Clarus "C"
Water concentration vs time during the gassing phases
the airflow tends to travel around the outer perimeter of the 22.0
NH H202 Clarus
isolator, around the load, and does not flow much into the rapid 20.0

Water concentration (mg/L)

FL H202 STERIS
18.0 FL H202 Clarus
transfer port (RTP). 16.0 Start of Decon
Start of Dwell
To obtain more detailed information, air velocity cut plots 14.0 Clarus Gassing End
12.0 NL H202 STERIS
were then generated at points showing the low flow areas of the 10.0
isolator identified in the airflow trajectory plot. From velocity 8.0
6.0
cut plots (see Figure 4), it was determined that the locations of 4.0
worst flow were on the left side of the bottom shelf in between 2.0
0.0
the left and middle gloves, in the RTP, and inside the sleeves of 0.0 3.0 6.0 9.0 12.0 15.0 18.0 21.0 24.0 27.0 30.0 33.0 36.0
the gloves. In the velocity cut plot, higher velocity flows are
Time (min)
shown in red while darkening shades of blue show low flows.
Smoke studies were conducted to provide a physical confir- Figure 8: Water concentration versus time.
mation of the CFD analysis. During these studies, the loaded iso-
lator was mapped with visible smoke, and the established smoke
patterns were analyzed to determine the worst-case flow loca- firmation of the initial population was performed according to
tions. In both generators studied, the airflow entered and exited USP Chapter ^55&, and it was verified that the BIs were within
at the same places at a similar volumetric flow rate, 20 standard acceptable population limits at 1.1 3 106 organisms per carrier.
ft3/min. The smoke studies for generators yielded similar worst- The D-value of the BIs was also tested, using a VHP biolog-
case points for airflow as those seen in the CFD analysis. ical indicator evaluator resistometer (BIER) unit (VhyPer, PSI)
to expose the BIs to a controlled concentration of VHP and
Placement of thermocouples, biological indicators, and water vapor under a constant controlled temperature (13). This
chemical indicators test was conducted using a square wave shuttle over various
Twenty-two calibrated thermocouples, biological indicator (BIs), timed exposures at 32 8C with a VHP concentration of 2.0 mg/L.
and chemical indicators (CIs) were placed at standard locations The test identified the D-value of the BIs as 0.2 min. This is
throughout the isolator. Additionally, four thermocouples, BIs, lower than the manufacturer’s specifications, probably because
and CIs were placed in the worst-case locations determined in the VhyPer allows for more exact determination of the D-value.
the CFD and smoke studies. Worst-case points were defined as
the area of lowest flow and/or lowest temperature. The BIs and Results and equipment performance differences
CIs covered the entire interior of the isolator, with a concentra- Separate performance qualification tests were conducted for
tion of approximately one BI and CI per cubic feet. Figure 5 the VHP 1000ED and the Clarus “C”. Both generators’ cycles
shows the placement of the thermocouples, BIs, and CIs. passed the acceptance criteria of 100% inactivation of the bio-
G. stearothermophilus was used as the BI for the PQ of both logical challenge for both the minimum and maximum load
generators. The BIs were commercially prepared on a stainless conditions on three consecutive runs each. During PQ, several
steel carrier and placed into a nonwoven high-density poly- performance differences were noticed. The Clarus’s cycle is
ethylene pouch. Because the PQ of the two machines was exe- shorter and uses considerably less hydrogen peroxide solution.
cuted more than six months apart, two different lots of BIs, In addition, the VHP and water concentration profiles differ
from a single manufacturer, were used. The manufacturer’s cer- significantly because of the differences in their methods of op-
tified D-value for both lots of BIs was approximately two min- eration (wet versus dry).
utes with an initial population of 2.2 3 106. An in-house con-
78 Pharmaceutical Technology NOVEMBER 2004 www.phar mtech.com
 tion, absorption, and decomposition of VHP in an isolator will
Temperature mapping of the transfer isolator cause VHP and water concentrations to vary from the theoret-
during the STERIS VHP 1000ED Cycle
36
Start of cycle
ical concentrations as calculated by a mass balance (15). The
Start of
34
conditioning guided wave vapor monitor uses near infrared spectroscopy to
Start of
32
decon measure the actual VHP and water vapor concentrations in
mg/L (15). The guided wave’s probe was positioned in the ap-
Temperature (°C)

30

28 proximate center of the isolator to determine the concentra-

26 tions in and around the load. Because the guided wave meas-
Start of Aeration
24 ures VHP and water vapor continuously, these concentrations
22 can be measured over time during the generator decontamina-
20 tion cycle. Figures 6 and 7 show the concentration profiles of
18 VHP and water vapor during the gassing cycles. In both in-
0 15 30 45 60 75 90 105 120 135 150 165 180 195 210 225 240 255 270 stances the maximum and minimum load conditions are plot-
Time (min) ted. The readings gathered from the guided wave guaranteed
that the isolator and load had similar vapor concentrations pro-
files from run to run.
Figure 9: STERIS temperature profile.
Because the VHP 1000ED and the Clarus “C” use different
biodecontamination methods (wet versus dry) VHP and water
Temperature mapping of the transfer isolator levels were different in the two generators (see Figures 8 and 9).
during the clarus "C" cycle
Cycle started at 0 minutes The VHP 1000ED had significantly higher VHP concentrations
36
Start of ramp gassing throughout the cycle than the Clarus “C.” In contrast, the VHP
34
Start of dwell gassing
1000ED has significantly lower water concentrations through-
32
Start of aeration out the cycle. The similarity between the maximum- and
Temperature (°C)

30
minimum-loaded conditions demonstrates that the load con-
28
tents had little effect on the VHP concentrations during the gassing
26
cycle.
24
The temperature profiles for the two generator systems are
22
shown in Figures 9 and 10.The temperature profiles for the two
20
generators are very similar. In other generators, the tempera-
18
0 15 30 45 60 75 90 105 120 135 150 165 180 195 210
ture increased continuously during the decontamination phase
Time (min)
and did not reach a steady state temperature until the aeration
phase.
Figure 10: Clarus “C” temperature profile.
Conclusion
After completing an installation qualification, operational qual-
Compared biodecontamination cycles ification, and performance qualification, and comparing the
The cycles that were programmed into each generator are listed operation and performance of the STERIS VHP 1000ED Bio-
in Table I. Aeration was continued until the isolator environ- decontamination System and BioQuell Clarus “C” H2O2 gas
ment was ,1 ppm VHP, as read by a Dräger pump and tube, generator, it can be concluded that both units can be validated,
to meet OSHA permissible exposure limit guidelines for accept- and are capable of effectively decontaminating an isolator. There
able hydrogen peroxide contact levels (14). Once the VHP level are differences between the two systems; however, these differ-
was below this limit, the cycle was considered complete and the ences do not affect the effectiveness of either unit to decontam-
BIs and CIs were harvested. inate an isolator system.
The first three phases (conditioning, ramping, and decont-
amination) of the VHP 1000ED totaled 62 min with an addi- References
tional 3 to 5h of aeration, for a total cycle time of 4 to 6h. The 1. R. Friedman, “Design of Barrier Isolators for Aseptic Processing: A
first three phases of the Clarus “C” totaled 37 min with 2.5–3h GMP Perspective,” Pharm. Eng. 18 (2), 28–33 (1998).
2. J.C. Lyda, “Regulatory Aspects of Isolator/Barrier Technology,” PDA
of aeration, for a total cycle time of 3 to 3.5. Because the Clarus Pharm. Sci. Tech., 49 (6), 200–304 (1995).
“C” has a shorter cycle time and lower hydrogen peroxide in- 3. C.M. Wagner and J. Raynor, “Industry Survey on Sterility Testing Iso-
jection rates, it uses less hydrogen peroxide during a cycle. Dur- lators: Current Status and Trends,” Pharm. Eng. 21 (2), 124–140 (2001).
ing the two three-quarter cycles, the Clarus “C” used 22.5 g of 4. J. Lysfjord and M. Porter, “Barrier Isolation History and Trends, A Mil-
hydrogen peroxide whereas the VHP 1000ED used 130.2 g. lennium Update,” Pharm. Eng. 21 (2), 142–145 (2001).
5. Food and Drug Administration, Guidance for Industry: Sterile Drug
Products Produced by Aseptic Processing–Current Good Manufacturing
Water and hydrogen peroxide concentrations and Practice (FDA, Rockville, MD, 2001).
temperatures 6. “Hydrogen Peroxide Vapour Biological Efficacy,” V2.5 (BioQuell
To monitor the concentration of VHP and water vapor in the Pharma, Andover Hampshire, United Kingdom, February 2003).
isolator, a guided wave vapor monitor was used. The adsorp- 7. “La Calhène Three Glove Transfer Isolator User’s Manual,” (La Cal-

80 Pharmaceutical Technology NOVEMBER 2004 www.phar mtech.com

 hène, Vendome, France).
8. R. Watling and C. Parks, “Theoretical Analysis of the Condensation FYI
of Hydrogen Peroxide Gas and Water Vapour as Used in Surface De-
AAPS announces 2004 fellows
contamination,” Pharm. Sci. Tech. 56 (6), 291–299 (2002).
9. Clarus “C” User Manual, STD2000-005, Rev. 4 (BioQuell Pharma, An- The American Association of Pharmaceutical Scientists (AAPS,Arlington,VA,
dover Hampshire, United Kingdom, 2002). www.aapspharmaceutica.com) has unveiled its annual list of AAPS Fellows.This
10. Cycle Development Guide, STD2000-006, Rev. 1, (BioQuell Pharma, year,24 people were honored for professional excellence in the pharmaceutical
Andover Hampshire, United Kingdom, 2002). sciences:
11. VHP 1000ED Biodecontamination System Operator Manual, (STERIS
Corporation, Erie, PA, 2002). Kevin D.Altria,PhD Peter Kleinebudde,PhD
12. VHP 1000ED-C Biodecontamination System Maintenance Manual, Jeffrey S.Barrett,PhD Ah-Ng Kong,PhD
(STERIS Corporation, Erie, PA, 2003).
Ronald R.Bowsher,PhD Jean W.Lee,PhD
13. D. Khorzad et al., “Design and Operational Qualification of a Vapor-
Phase Hydrogen Peroxide Biological Indicator Evaluator Resistome- Richard N.Dalby,PhD Hans Lennernäs,PhD
ter (BIER) Uni,” Pharm. Technol. 27 (11), 84–90 (2003). John W.A.Findlay,PhD Joyce J.Mordenti,PhD
14. M. Ebers, MSDS No. A121, “Hydrogen Peroxide Solution (31%–35%)”,
STERIS, 19 February 2002. Joseph C.Fleishaker,PhD Ram B.Murty,PhD
15. G.P. Brown, et al.,“Calibration of Near-Infrared (NIR) H2O2 Vapor Joseph A.Fix,PhD Fridrun Podczeck,Dr.sc.nat.habil
Monitor,” Pharm. Eng. 18, (6), 66–76, (July–August 1998).
Andrea Gazzaniga,PhD Joseph W.Polli,PhD
Bruno C.Hancock,PhD Jagdish Singh,PhD
Please rate this article.
On the Reader Service Card, circle a number: Guenther Hochhaus,PhD Tetsuya Terasaki,PhD
345 Very useful and informative Jin-Ding Huang,PhD Peter Veng-Pedersen,PhD
346 Somewhat useful and informative Peter F.Kador,PhD Lawrence X.Yu,PhD
347 Not useful or informative
Your feedback is important to us.

Circle/eINFO 61
82 Pharmaceutical Technology NOVEMBER 2004 www.phar mtech.com