Sandle, T.

(2015) Microbiological Assessment of Compressed Gases in Pharmaceutical

Facilities, Journal of Validation Technology, Vol. 21, Issue 2, August 2015, pp1-8

Microbiological assessment of compressed gases in pharmaceutical facilities

Tim Sandle

Introduction

Compressed gasses are used at different stages of the pharmaceutical manufacturing process.
Applications include weighing of vials on process line or the removal of fallen vials.
Furthermore, compressed gases such as air, nitrogen, and carbon dioxide are deployed in
operations involving purging or overlaying.

Compressed gas sampling for microorganisms is an important part of contamination control

assessment (1). While sampling is important, the method of sampling can hindered by the
design of the gas system, where sampling is not easily conducted in an aseptic manner, or by
the design of the air-sampling instrument. This paper reviews the important aspects of
compressed air sampling for microbiological assessment and looks at possible sources of
contamination, should microorganisms be recovered.

Compressed gas

Compressed gas is a general term for gas stored or held under pressure that is greater than
atmosphere. Gas is compressed via a compressor; here, the greater the quantity of the gas
then the higher the pressure is (degree of compression). The compressor takes gas or air,
which occupies a given space, and reduces it to a smaller space. Here the greater mass of air
or gas produces a greater pressure. For instance, compressed air operating at 100 pound-force
per square inch will have been compressed down to 1/8th of its original volume. 100 pounds
per square inch equals 7 bar; or, to use SI units, one pound per square inch equals
approximately 6894 Pascals).

Types of compressed gas include air, which is colorless and odorless, where the composition
is 78% nitrogen, 21% oxygen, and the remainder composed of trace elements. Other common
gases are oxygen, carbon dioxide and nitrogen. Aside from air, nitrogen is the most
commonly used gas in the pharmaceutical sector. Nitrogen has inert characteristics that lead
to its ideal use as a pressurizing agent. With this, tanks, pipelines, hoses, vessels, and other
process equipment can be tested for leaks with nitrogen gas. Nitrogen gas can also be used to
dispense or transfer most fluids from storage tanks or reservoirs. Nitrogen gas should be
composed of a minimum of 99% nitrogen gas, with trace impurities, like carbon monoxide,
carbon dioxide and oxygen permitted at <0.001% (2).

Purity is a factor that needs to be maintained with compressed gas; hence the gas should be
supplied oil free. Purity overall is achieved through a combination of filtration, purification
and separation. The process of creating the compressed gas can additionally introduce water
vapor; thus, a process must be in place to remove water vapor before the gas is expelled into
a critical zone like a cleanroom. Compressed gas is typically discharged from the compressor
hot and it will contain water vapor. Temperature is reduced by using a post-compressor
cooler and, as the gas condenses, the water vapor and other impurities can be removed. The
risk of water vapor is particularly high with compressed air, which is drawn into a
compressor via the atmosphere. Atmospheric air contains a high proportion of water vapor
(that is water in a gaseous form). Water removal is achieved through a combination of
filtration and dehumidification.

Where is drawn in from the outside, the process of drawing in also introduces
microorganisms, which require filtering out. The level of filtering depends upon whether
'sterile' air is required (absence of viable microorganisms) or air with a low bioburden.

Compressed gas can be supplied at source either sterile or non-sterile. Sterility is achieved
through the use of a bacterial retentive membrane filter (0.2 µm pore size). A sterile-filtered
gas is required where a gas contacts a sterilized component or material. It is important that
the sterilizing grade filter is maintained dry for condensate in a gas filter will most probably
cause blockage or lead to microbial contamination. Risks of condensate are controlled by
heating and use of hydrophobic filters (to prevent moisture residues in a gas supply system).
Filters should also be changed periodically. As part of on-going quality control, filters must
be integrity tested at installation and at end of use.

Compressed gas standards

Although national standards bodies have guidance documents for compressed air sampling,
and reference is made within FDA and EU GMPs, the general approach and requirements for
compressed gasses are set out in a multi-part ISO standard: ISO 8573. This standard consists
of the following parts (3):

Part 1: Contaminants and purity classes,

Part 2: Test methods for aerosol oil content,
Part 3: Test methods for measurement of humidity,
Part 4: Test methods for solid particle content,
Part 5: Test methods for oil vapor and organic solvent content,
Part 6: Test methods for gaseous contaminant content,
Part 7: Test method for viable microbiological contaminant content,
Part 8: Test methods for solid particle content by mass concentration,
Part 9: Test methods for liquid water content.

Part 1 outlines the required purity classes based on the concentration of particles and level of
impurities. The potential ‘impure’ contaminants for compressed air, that can affect whether a
required purity class is met, include:

Particles (such as dirt, rust, pipe scale), with particles assessed by size. For example,
as result of the mechanical compression process, additional impurities may be
introduced into the air system. Generated contaminants include compressor lubricant,
wear particles and vaporized lubricant. Furthermore, fittings and accessories can
contribute to particles.
Water (in both vapor and liquid forms). Water is typically assessed by vapor pressure
dew-point. This is the temperature at which the air can no longer "hold" all of the
water vapor which is mixed with).
Oil (including aerosol, vapor, and liquid forms).

Microbiological contamination sources are discussed below.

Microbial content itself does not influence the purity class assigned, although the standard
recommends that microbial levels are assessed. Acceptable microbial numbers are subject to
a separate assessment; with such an assessment is based on an interpretation of GMP. For
example, the 2004 FDA Aseptic Filling Guidance document states (4):

“A compressed gas should be of appropriate purity (e.g., free from oil) and its
microbiological and particle quality after filtration should be equal to or better than that of the
air in the environment into which the gas is introduced.”

With purity, many parts of the pharmaceutical industry will use class 1 compressed gas based
on the maximum number of permitted particulates. The particle limits are:

ISO 8573 class Particle size limits per m3

0.1 - 0.5 µm 0.5 - 1.0 µm 1.0 - 5.0 µm
1 ≤ 20,000 ≤ 400 ≤ 10

Outside of the ISO 8573 standard, supporting information is contained within the ISPE Good
Practice Guide - Process Gases (2011) (5). Here Table 7.1 of the guide indicates that particle
counts (both viable and inert) should "Typically equal to the at rest condition of the area
served." That is, for EU GMP Grade A / ISO 14644 class 5 areas, the microbial count should
be <1 CFU/m3 and the particle levels conform to the area at rest ≤ 3,520 particles per m3).

A separate standard exists for the production of compressed air. This is ISO 12500, a four
part standard:

ISO 12500-1:2007 - Filters for compressed air -- Test methods -- Part 1: Oil aerosols
ISO 12500-2:2007 - Filters for compressed air -- Test methods -- Part 2: Oil vapors
ISO 12500-3:2009 - Filters for compressed air -- Test methods -- Part 3: Particulates
ISO 12500-4:2009 - Filters for compressed air -- Methods of test -- Part 4: Water

With ISO 12500 there are no specific microbial testing requirements.

Microbial survival

Based on the above, the maximum level of microorganisms will be applicable to the
cleanroom class. This places a tight limit on ISO 14644 class 5 / EU GMP Grade A areas.
Although compressed gas and air systems are relatively harsh environments, they can aid
microbial survival if there are available nutrients. The availability of nutrients is dependent
upon the purity of the gas and airline. Nutrients suitable for metabolizing by microorganisms
include water and oil droplets. Another factor that can affect survival is temperature,
especially where temperatures are warmer (6).

In addition to vegetative cells, bacterial spores are well equipped to survive the harsh
environmental conditions. Spores are resistant to the types of temperature ranges and
moisture levels found within compress gas lines. Another risk exists with biofilm, where
microbial communities can potentially form and develop through attachment to air lines and
tubing.
Although these risk factors exist, typically no microorganisms would be expected to be
recovered from compressed gas lines. Research has shown that many microorganisms can
survive and multiply in pressurized systems up to 10 bar and some are at least able to recover
after being pressurized. However, at 160 bar pressure upwards, survival rates are very low.
Where low level counts are recovered, these require investigation. More often the source is
adventitious contamination, although a fault with the compressed air line cannot be ruled out.

Although microbial contamination of compressed air or gas is a rare event, incidents can
occur. Sources of contamination include:

Source of the air of gas. With air, this is air intake air from surroundings (which can
contain oil, dirt/dust and moisture/ water vapor, microorganisms).
Piping distribution systems. Piping distribution and air storage tanks, more prevalent
in older systems, will have contaminant in the form of rust, pipe scale, mineral
deposits, in addition to bacteria.
Bacterial retentive filter: the filter may become blocked, lose its integrity, or become
wet.
Compressor failure: The compressor itself can create a contaminated environment.
For example; the compressor's pre-filters can become overloaded with dust and lint,
causing the filter to cease functioning properly.
Sample valve: the point-of-use sample valve may not be designed correctly or become
faulty.

Sampling and testing requirements

As indicated above, compressed gas requires assessment against a number of parameters,

including particles and viable microorganisms. The part of the standard used for making
assessments is ISO 8573 Compressed air —Part 7: “Test method for viable microbiological
contaminant content” (7).

When sampling compressed air for microorganisms, it is important that the air is
depressurized and that the flow rate is controlled. Control of the flow rate is important to
ensure that a cubic meter of air is sampled within the required sampling time (this time will
be instrument dependent). With this, compressed gas is typically at 160 pounds per square
inch or greater. An external regulator will be needed to bring this down to the sampling rate
of instrument. If the air sampler takes 36 minutes to capture a cubic meter of air, then it will
be sampling at one cubic foot per minute. The regulator will need to reduce the air down to
allow for this. This is assessed using a flow meter. Pressure reduction to atmospheric
conditions is of great importance and knowing the flow allows the agar exposure time to be
assessed, so that one cubic meter of air is sampled.

It is also important that isokinetic sampling of the air occurs and that air velocity is reduced
until it is within the range of the sampler as identified by the manufacturer. This is not only
necessary for obtaining the correct sample size; it also impacts on the possibility of microbial
survival. The level of impact stress has been shown to affect microbial recovery on agar and
be dependent upon the impaction velocity of the cells into the agar as well as the design and
operating parameters. Due to the fact that any microorganisms present are transported under
pressure and then suddenly released into atmospheric conditions, they may be damaged by
the immediate expansion of the gas and the resulting shearing forces.
The head of the instrument and any attachments must be sterile before use, to avoid
contamination. The culture medium used with the instrument should be sterile (normally by
irradiation) and a representative item should have been tested for growth promotion. With
most samplers the head will be autoclavable. Some users disinfect the tubes and hoses used to
connect the sampler with a disinfectant like 70% isopropyl alcohol. This is mentioned as an
option in the ISO standard, although this is erroneously described as "sterilization." Where a
disinfectant is used it is important to run the air through the sampler without any agar plate in
place; this is necessary to evaporate the disinfectant and to remove any residues. The
presence of disinfectant could potentially lead to a 'false negative'.

With sampling, the sample inlet is connected to the compressed gas line and air is directed
over an agar plate or strip. The method works by compressed gas, under reduced pressure,
called 'partial flow´, is forced over the surface of an agar plate. Any microorganisms are
impinged onto the surface of the agar.

The sampling time should be sufficient in order to sample one cubic meter of the agar. After
sampling, the agar plate or strip is removed and incubated within a microbiology laboratory.
At the end of incubation, the agar is examined for colony forming units.

If colony forming units are recovered, these should be assessed against the appropriate limit.
It is good practice to identify the contaminants recovered; the identification may provide
important information as to the origin of the bacteria.

Instrumentation

The type of instrument recommended in the ISO standard is a “slit-sampler, a type of

impaction air tester”, although alternative samplers can be used, if justified. With a standard
impactor sampler, air is drawn through a sampling head via a pump or fan and accelerated,
usually through a perforated plate (sieve samplers), or through a narrow slit (slit samplers).
This process creates a laminar flow through the sampler head. Hence, the air sampler should
be fitted with a diffuser capable of maintaining laminar flow conditions. This is necessary so
that particles pass through the sample head in a controlled flow.

The velocity of the air is determined by the diameter of the holes in sieve samplers and the
width of the slit in slit samplers. When the air strikes the collection surface on the agar plate,
it makes a tangential change of direction. This causes any suspended particles to be thrown
out by inertia, impacting onto the collection surface. When the correct volume of air has been
passed through the sampling head, the agar plate can be removed and incubated (8).

When selecting a suitable sampler, three parameters should be checked and evaluated. These
are (9):

The physical efficiency of the sampler. This is the relative efficiency of the sampler in
collecting particles over a range of sizes. Physical efficiency is measured against
membrane filtration sampling and the d50 value assessed. The d50 is the aerodynamic
diameter, above which the collection efficiency of the impactor approaches 100%.
Knowing the d50 value gives an indication of the sizes of particles likely to be
collected by the sampler for the d50 is equivalent to particle size at which 50% of the
particles are collected, and 50% pass through the sampler because that are too small to
impact (10).
 The biological efficiency. This is the relative efficiency of the sampler in collection of
microorganisms on a surface so that they remain viable and can be counted post-
incubation. Biological efficiency is compared with an established reference sampler
such as the Casella slit sampler. Assessment of air samplers involves the use of a
controlled microbial population passed into a nebulizing chamber.
The flow rate of the sampler. With all sample sizes, the flow rate of air through the
sampling head is critical to the accuracy of the result.

In terms of culture media, sampler models available either collect air samples onto contact
plates (55mm or 84mm diameter), standard petri dishes (90mm diameter), or onto agar strips.

Outside of these requirements, the ideal device should be portable to permit sampling
throughout a series of cleanrooms. Devices should also be cleanable and resistant to common
cleanroom disinfectants. The ideal material of construction is stainless steel.

Sampling concerns

Compressed air sampling should form part of an environmental monitoring program, along
with cleanroom assessments. The program should take into account air points to be tested.
This could be every point; points considered to be of greater risk (such as product contact); or
representative points along a loop. The frequency of testing must also be considered, and this
too would need to tie into risk.

An appropriate agar must be selected. An example is tryptone soya agar, which is a generally
nutritious medium designed to recover a range of bacteria and fungi. A key factor to take into
account is whether the process of sampling leads to undue desiccation of the agar, rendering
any recovered microorganisms unable to grow on the gar due to depletion of growth
nutrients. This will be affected by the flow rate, type of compressed gas, and the model of air-
sampler, together with the type of culture medium. A risk will remain that microbial cells will
become damaged by mechanical stress during the sampling process and lose viability. These
factors should be evaluated through a study (11).

In addition, the agar medium should be removed from the sampler as quickly as is practicable
and transferred to the required incubator. This is to avoid the culture medium from drying out
or deteriorating.

With the incubation conditions selected, the time and temperature should be suitable for the
recovery of the a general range of microorganisms, particularly Gram-positive organisms
given that such bacteria are better equipped to survive in dry environments (12). The typical
requirement is to look for mesophilic bacteria and fungi (those that would grow across the
temperature range 20-30oC). Here some users would elect to use one representative
temperature whereas others would elect to use a two-step incubation regime, such as (13):

20-25oC for 3-5 days, followed by

30-35oC for 3-5 days.

The selected incubation time should be based on growth promotion studies. If certain
microorganisms are considered a problem, alternate incubation times or culture media can be
considered.
Sampling frequencies

The user will need to determine whether each compressed gas line requires testing and the
frequency of testing. Certainly all product contact compressed gases should be assessed. A
sampling plan should also consider, and adapt to, the following:

Increased or reduced production schedules,

Seasonal changes,
Equipment changes and modifications,
Replacement of hardware or filters and dryers,
Inactivity of system.

Reporting requirements

When reporting the results from a compressed air sampling session, in relation to microbial
counts, the following information is advised in the ISO 8573 standard:

1. Whether the compressed airline was “sterile” or “non-sterile”,

2. The date of sampling,
3. The date of measurements, and
4. The location of the sample.

In addition to this, the result should be added, expressed as colony forming units per cubic
meter of air (CFU/m3).

Bacterial endotoxin

The ISO 8573 standard has the option of sampling compressed air for bacterial endotoxin.
Such testing remains relatively uncommon and it is only necessary should the compressed
gas have a direct product contact and where there is a concern with Gram-negative bacteria.
In most cases there should be no likelihood of endotoxin being present.

The sampling method for bacterial endotoxin is tricky and inexact. Either colonies are
examined for Gram-negative bacteria, and assessment is made about endotoxin risk; or the
compressed is passed through pyrogen-free water.

Summary

The assessment of compressed gas is an important part of quality control within

pharmaceutical organizations, whether that assessment is for chemical impurities, particulates
or microorganisms. This paper has assessed the appropriate standards for compressed air and
focused on microbiological sampling. In doing so, the important features of air sampling have
been raised together with the factors that can lead to microbial contamination occurring.
These aspects should be built into a biocontamination control program.
References

1. Sandle, T. (2013). Contamination Control Risk Assessment in Masden, R.E. and

Moldenhauer, J. (Eds.) Contamination Control in Healthcare Product Manufacturing,
Volume 1, DHI Publishing, River Grove: USA, pp423-474
2. Baseman, H. (2010) Sterile product manufacture using form fill seal technologies. In
Agalloco, J. and Akers, J. (Eds.) Advanced Aseptic Processing Technology, Informa
Healthcare, New York, USA, pp164-165
3. ISO 8573-1:2010 Compressed Air Contaminants and Purity Classes, International
Standards Organization, Geneva, Switzerland
4. FDA (2004) Guidance for Industry Sterile Drug Products Produced by Aseptic
Processing — Current Good Manufacturing Practice, U.S. Department of Health and
Human Services Food and Drug Administration, Rockville, MD, USA
5. IPSE Good Practice Guide: Process Gases, 2011, IPSE, USA, pp81-86
6. Stewart, S. L., S. A. Grinshpun, K. Willeke, S. Terzieva, V. Ulevicius, and J.
Donnelly (1995) Effect of Impact stress on microbial recovery on an agar surface.
Appl. Environ. Microbiol. 61:1232-1239.
7. ISO 8573-7:2003 Test method for viable microbiological contaminant content,
International Standards Organization, Geneva, Switzerland
8. Sandle, T. (2011): Key points for active air samplers, Clean Air and Containment
Review, Issue 5, pp8-10
9. Sandle, T. (2010) Selection of active air samplers, European Journal of Parenteral
and Pharmaceutical Sciences, 15 (4): 119-124
10. Hinds, W. C. (1982) Aerosol technology, John Wiley & Sons, New York, p. 104–126
11. Morring, K. L., W. G. Worenson, and M. D. Attafield (1983) Sampling for airborne
fungi: a statistical comparison of media. Am. Ind. Hyg. Assoc. J. 44: 662-664
12. Moissl-Eichinger C, Rettberg P, Pukall R. (2012) The first collection of spacecraft-
associated microorganisms: a public source for extremotolerant microorganisms from
spacecraft assembly clean rooms, Astrobiology 12(11):1024-34
13. Sandle, T. (2014) Examination of the Order of Incubation for the Recovery of
Bacteria and Fungi from Pharmaceutical Cleanrooms, International Journal of
Pharmaceutical Compounding, 18 (3): 242 – 247