Peer Reviewed

A Practical Approach to Depyrogenation

Studies Using Bacterial Endotoxin
Tim Sandle

sodilatation, diarrhea, and fetal shock syndrome. Due

Depyrogenation devices, such as tunnels, are used in the to the level of risk, pharmaceutical water systems and
pharmaceutical industry to prepare components for asep- parenteral products are tested for pyrogens including,
tic filling. To qualify such devices, various pharmacopeias or exclusively, endotoxins (3).
require depyrogenation devices to be periodically chal- Despite the relative comprehensiveness of the phar-
lenged with high levels of bacterial endotoxin. Although macopeial monographs for the LAL test, one key ap-
the pharmacopeias state the acceptance criteria, little plication of endotoxin testing is not described in great
consideration is given to the practical approach. This detail: conducting depyrogenation studies. A depyro-
article discusses the theoretical concept of depyrogena- genation study is the key biological test, in addition to
tion. A case study of a depyrogenation tunnel is used to thermometric tests, for the qualification of depyroge-
define some of the practical aspects of a depyrogenation nation devices. Depyrogenation can be defined as the
study that need to be considered. elimination of all pyrogenic substances, including bac-
terial endotoxin, and is generally achieved by removal
or inactivation (4). Depyrogenation, like sterilization, is
INTRODUCTION an absolute term that can only be theoretically demon-
The bacterial endotoxin test (BET) is a relatively straight- strated because of test insensitivity.
forward test and has been a pharmacopeial method Some scientists regard depyrogenation purely as
since 1980, when it first appeared in the United States endotoxin destruction or inactivation, and endotoxin
Pharmacopeia (USP). The test, using Limulus amebocyte removal as a distinct and unrelated process. Here the
lysate (LAL) methodology, is described in detail in the former refers to inactivating or destroying any endotox-
harmonized chapters in both the European Pharmaco- in present on a component, the latter to the removal of
poeia (1) and the United States Pharmacopeia (2). The any endotoxin present (5). With depyrogenation inacti-
test describes the detection of the most common and vation, the total destruction of the “pyroburden” is as-
significant pyrogenic material found in pharmaceutical sumed; with endotoxin removal it is assumed that a sig-
production: gram-negative bacterial endotoxin. LAL is nificant portion of the pyroburden has been removed.
an extract from the lysed blood cells (amebocytes) of the Other scientists consider both processes to be part of
horseshoe crab Limulus polyphemus, or related species. depyrogenation.
The need to perform the LAL test for endotoxins This article examines the mechanism of endotoxin
is well established. Endotoxins can cause, to varying inactivation by dry heat and the practical steps to be
degrees depending upon potency and target site, en- taken for conducting a depyrogenation study. Depyro-
dotoxemia (i.e., the presence of bacterial toxins in the genation of glassware is important in the production of
blood) and septic shock (i.e., the prolonged presence of parenteral pharmaceuticals as residual pyrogens could
bacteria and bacterial toxins in the body). The effects of ultimately be injected into a patient resulting in an ad-
endotoxin in the human body include high fever, va- verse reaction. This is especially important as endotox-

90 Journal of GXP Compliance

 Tim Sandle

ins are heat stable, making them resistant to most con- affected by the pH range, temperature, flow rate,
ventional sterilization processes and thus necessitating and amount of electrolytes in the solution.
separate tests for viable cells and endotoxin (6). t
Dilution or rinsing—the amount of endotoxin is
The assessment of depyrogenation involves the in- washed away or reduced using WFI.
troduction of purified endotoxin of a high potency and t
Distillation—functions by turning water from
post-process testing to assess if a minimum of a three- a liquid to a vapor and then from vapor back to
log reduction has been achieved. The concept is simi- liquid. Endotoxin is removed by the rapid boiling
lar to that of steam sterilization studies, where a much that causes the water molecules to evaporate and
higher level of microorganisms than would ever be the relatively larger lipopolysaccharide (LPS) mol-
found in the normal environment is introduced into a ecules to remain behind.
device as a worst-case challenge. In the case of steam t
Adsorption (e.g., activated carbon beds, where
sterilization, this is Geobacillus stearothermophilus. Depy- endotoxin is adsorbed into charcoal, or depth
rogenation follows a pattern similar to steam steriliza- filters)—this functions by attracting negatively-
tion, in that the log-reduction will follow a linear plot charged endotoxin molecules to the carbon bed.
(i.e., as time and temperature increase, the endotoxin This mechanism is only efficient to a small degree
level decreases). After this destruction continues to oc- and is affected by a range of environmental factors.
cur, it does not necessarily follow linear regression (i.e., t
Hydrophobic attachment—certain materials, like
it does not produce the same semi-log reduction seen polyethylene, can bind endotoxin.
for bioburden reduction in sterilization devices and be- t
Acid or base hydrolysis—this destroys the eight-
comes more noticeable at lower temperatures) and in- carbon sugar: two-keto-three-deoxyoctonic acid
stead follows a biphasic reduction (7). that links Lipid-A to the core polysaccharide and
causes physiochemical changes that decrease
DEPYROGENATION pyrogenicity. The separated Lipid-A loses its pyro-
Before focusing on the practical aspects of depyrogena- genic activity). An example is 0.05M HCl for 30
tion studies, it is appropriate to consider the process of min. at 100 ∘C or 0.5 M NaOH at 50 ∘C for 30
depyrogenation. The following are various mechanisms min.
by which depyrogenation is achieved (8-11): It is possible, with alkaline hydrolysis, that the
t
Ultrafiltration—the process works by excluding level of endotoxin may initially rise as part of the
endotoxin by molecular weight using an ultra-fine separation process.
filter that blocks molecules of 10,000 Daltons or The hydrolysis methods are frequently used for
greater. This is often coupled with 0.1μm filter. depyrogenating glassware. The efficiency of this
t
Reverse osmosis—primarily functions as a process is often connected to the cleanliness of the
size-excluding filter operating under a highly- glassware prior to treatment.
pressurized condition. It will block 99.5% of endo- t
Oxidation—works by peroxidation of the fatty
toxin and ions or salts, but allow water molecules acid in the Lipid-A region (e.g., using hydrogen
through. USP reverse osmosis (RO) can be used to peroxide).
make water for injection (WFI) (whereas to meet t
Ionizing radiation—a very slow and inconsistent
the European Pharmacopoeia requirement it can process.
only be produced by distillation); within Europe it t
Ethylene oxide—functions by nucleophilic substi-
is used to produce highly-purified water. tution in the glucosomne of Lipid-A. It is not the
t
Affinity chromatography (e.g., Diethylaminoethyl most efficient depyrogenation process, and where
cellulose [DEAE] sepharose or polymyxin-B, endotoxin inactivation occurs this is normally a
which binds endotoxin by using a positive charge side effect of sterilization.
to attract the negatively-charged endotoxin and t
Moist heat—conventional autoclaving will not
then allowing its elution)—such processes are destroy endotoxin. However, the combination of

Autumn 2011 Volume 15 Number 4 91

Peer Reviewed

a chemical additive (e.g., hydrogen peroxide) and Various items can be depyrogenated by dry heat. The
physical variations (e.g., five hours at 121∘C with a main example referred within this paper is glass vials
pressure of 20 PSI and a pH of 3.8) are sometimes used for the filling of parenteral product, notwithstand-
effective. Alternatively, some studies have shown ing that many of the principles might apply to other
that autoclaving for prolonged periods of time or primary packaging materials providing that they can
at higher temperatures can be effective at reducing withstand high temperatures.
endotoxin and other pyrogenic substances (12, 13). Depyrogenation by dry heat for glass in the pharma-
t
Endotoxin can be inactivated by wet heat, ceutical industry is the primary endotoxin destruction
although this is only effective with far lower con- method used (17). This process both sterilizes and de-
centrations of endotoxin and is applied to non-heat pyrogenates and is mainly used for glass components.
stable materials. Destruction of endotoxin is far Dry heat involves subjecting the components to a high
more difficult to achieve, and lower log reductions level of heat (normally between 180 and 250∘C) for a
when compared with dry heat are achieved (14). defined time (the higher the temperature, the shorter
t
Dry heat (physical destruction), such as convection the time required). The typical cycle is 250∘C for not
(transfer of heat by movement of fluid or air), con- less than 30 minutes. For example, the European Phar-
duction (transfer of heat from adjacent molecules), macopoeia in chapter 2.6.8 states two possible time-
or irradiation (emission of heat by electromagnetic temperature combinations for depyrogenation: 60 min-
radiation). utes at 200∘C or 30 minutes at 250∘C. A quantity of
endotoxin destroyed at 250∘C for 60 minutes would not
This paper primarily focuses on one method of in- necessarily be totally destroyed at 200∘C at 60 minutes,
activation: by dry heat. The qualification of a depyro- based on the non-linearity of the thermal destruction
genation device operating by dry heat involves chal- curve. Endotoxin destruction at low temperature is of
lenging the device, such as a depyrogenation tunnel, the second-order (18).
with a known level of purified endotoxin (LPS). This is Depyrogenation dry heat devices include ovens
sometimes described as an “endotoxin indicator,” which and tunnel sterilizers. To operate, depyrogenation de-
is analogous to the “biological indicator” used to test a vices require a series of parameters to be controlled.
sterilization device like an autoclave (15). These parameters include laminar airflow controlled
by high-efficiency particulate air (HEPA) filters, with
Depyrogenation by Dry Heat—Case Study a specification for air velocity and particulates. Where
The following is a case study of depyrogenation of glass the device is a depyrogenation tunnel, the rate of speed
vials in a hot air tunnel sterilizer. (e.g., minimum, maximum, and nominal) must be
Depyrogenation is an important part of the manu- measured and verified. The key function for depyro-
facture of pharmaceutical products and is distinct genation is temperature control. Such depyrogenation
from sterilization. Sterilization refers to the destruction devices require qualifying as part of validation. This is
of living cells. However, the process does not neces- performed along the familiar lines of design qualifica-
sarily destroy microbial by-products and toxins. En- tion, installation qualification, operational qualifica-
dotoxin is one toxin that is extremely heat stable and tion, and performance qualification, as well as annual
is not destroyed by standard sterilization cycles (e.g., re-qualifications. A depyrogenation study is a test of the
autoclaving). If only sterilization is required to be dem- physical capabilities of a device to depyrogenate an ar-
onstrated, this can be achieved using biological indica- ticle or device. It is demonstrated by physical measure-
tors impregnated with endospores from a heat resistant ments (including temperature) and biological (using
bacteria (e.g., Bacillus subtilis var. niger [often used for bacterial endotoxin). The biological test is the concern
dry heat] or Geobacillus stearothermophilus [often used of this paper.
for moist heat, although the microorganism also has a As part of the validation, normally at the performance
high resistance to dry heat] [16]). qualification stage, depyrogenation devices are biologi-

92 Journal of GXP Compliance

 Tim Sandle

cally challenged using a known level of a high concen- The testing of a depyrogenation device, at perfor-
tration of Escherichia coli endotoxin. The preparation mance qualification, firstly involves running the de-
used is a freeze-dried extract from the Gram-negative vice with a full set of containers in normal operation.
bacterial cell wall lipopolysaccharide (LPS). The prepa- Simultaneously, the depyrogenation device is tempera-
ration is similar to the control standard endotoxin (CSE) ture mapped using thermocouples. The thermocouples
used for routine LAL testing, although the concentra- indicate where the cold spots (i.e., areas of the lowest
tion, once reconstituted, is far greater. temperatures) are within the device. The run is then re-
As indicated earlier, the mechanism for endotoxin peated using endotoxin challenges at these colder areas,
destruction by depyrogenation is not fully understood alongside thermocouples. One problem to consider, be-
(19). Depyrogenation by dry heat follows what Ludwig fore undertaking such a study, is how to identify the in-
and Avis (18) describe as a “biphasic destruction curve.” oculated vials once they enter the tunnel. Many tunnel
This is a logarithmic reduction until three logs of endo- sterilizers have a capacity for many thousands of bottles.
toxin have been eliminated and is a varied slow down A second consideration is the number of challenge vials
when the inactivation of endotoxin ceases to decrease at to use in the study. The number used will be based on
a log-step rate. This is partially the reason why the crite- the size of the depyrogenation device. Typically, five to
ria for a successful depyrogenation validation are set at a ten challenge vials are sufficient to assess the depyroge-
minimum of three-logs to provide an over-kill and com- nation capabilities of the device being tested.
panies go further and stipulate a minimum of four-logs The following are two approaches for adding the en-
(20). The time and temperature combination is of im- dotoxin challenges:
portance. Achieving a three-log reduction at 170∘C may t
Using a high potency endotoxin spike directly
take 100 min. under one set of conditions. Whereas by onto the surface of the container to be depyroge-
increasing the temperature to 250∘C, the time would nated; allowing this to dry or to freeze dry and
theoretically be reduced to approximately 30 min. then placing it into the depyrogenation device
A further variability, outside of the theoretical ap- t
Using vials of high concentration endotoxin and
proach of such studies, would be the different potencies substituting these for the containers.
and thermal death of environmental endotoxin from
different Gram-negative bacterial lipopolysaccharide Of the two methods, the former is the one that is clos-
(21). er to a representative challenge. The second approach,
Although the first ever depyrogenation studies were due to the endotoxin being contained in a vial of a dif-
conducted in the early 1940s using rabbits to detect re- ferent size (and invariably of a different type of glass), is
sidual pyrogens, it is not considered ethical, or scien- less of a representative challenge. The second method is
tifically valid, to use the rabbit test for depyrogenation not often accepted by regulatory authorities.
studies. Any of the three established LAL methods can In both cases the endotoxin challenge is typically
be used to conduct a depyrogenation study: gel-clot, 1000 Endotoxin Units (EU) or greater. The level of the
turbidimetric, and chromogenic. However, where pos- challenge is determined by using control vials that are
sible, the gel-clot method is not recommended because not subjected to the depyrogenation cycle. These are
of the large number of dilutions required. tested alongside the test vials on completion of the de-
The endotoxin used in the study should be licensed pyrogenation run. It is not necessary to recover all of the
by the US Food and Drug Administration and pur- endotoxin from the control vials, and it is acknowledged
chased with a certificate of analysis for the lot of lysate that recovering endotoxin is difficult. A study by Plant
used, with the potency determination having been gave a range of between 20 and 70% as the typical re-
undertaken by the supplier against reference standard covery range (22). It is only necessary to recover a level
endotoxin. This potency is used to convert the labeled of endotoxin to show that the necessary log reduction
weight of the endotoxin (expressed in nanograms) into has been achieved. Therefore, if a three-log reduction
Endotoxin Units (expressed as X EU/ng). was required and 1000 EU were the theoretical chal-

Autumn 2011 Volume 15 Number 4 93

Peer Reviewed

lenge per container, but only 500 EU per container were adding a known level of pyrogen free water to the con-
recovered, then provided the test vials showed <0.5 EU trol and test vials. The amount of water should be suf-
per device recovery, the three-log reduction would have ficient to cover the base of the device and allow rins-
been achieved. Nevertheless, some level of laboratory ing. The vials require different techniques in order to
competence should be demonstrated, in that the level remove the endotoxin from the glass surface. These are
of endotoxin recovered from control vials can be done typically variations of vortex mixing and ultrasonica-
so reproducibly. tion. The actual times required for optimal endotoxin
When carrying out a depyrogenation study, pyrogen- recovery will need to be assessed by the user. A dis-
free pipettes and glassware must be used throughout. persing agent or buffer may also be used in place of
Exposed vials should also be covered with a pyrogen- pyrogen-free water.
free covering, such as Parafilm-M (Pechiney Plastic An aliquot is then tested against an endotoxin stan-
Packaging). The vials that are challenged with endo- dard series (consisting of a minimum of three-log con-
toxin require preparing the day before the validation centrations of endotoxin). The standard series should be
run in order to allow for the endotoxin to dry onto the prepared using the same lot of endotoxin used to chal-
base of the vial. This takes, in this author’s experience, a lenge the vials.
surprisingly long time. The endotoxin should be applied The control vials require dilution prior to testing. The
to the base of the vial and the vials dried (so that there level of dilution will depend upon the expected level of
is no visible endotoxin) under a unidirectional airflow endotoxin to be recovered. Negative control vials are
cabinet to protect the exposed vials from contamina- also required. Normally, two types of negative controls
tion. There will be a degree of loss through evaporation, are used. The first set is uninoculated vials, which are
although the extent of this is assessed through the re- not put into the depyrogenation device. The result from
covery from the control vials. Typically, a 0.1 ml inocu- these will indicate if any residual endotoxin is pres-
lum of endotoxin solution can take eight hours to dry ent. The second set consists of vials that have passed
onto the surface of the glass. In order to increase the re- through the depyrogenation device. These are tested in
covery of endotoxin, Novitsky (23) recommends freeze- the event that the elevated temperature inside the de-
drying the challenge (rather than air-drying), although pyrogenation device has resulted in the leeching of any
this requires specialized equipment that is not available interfering substances that might inhibit the LAL assay.
to all laboratories. Other techniques include using an Both types of negative controls should show a low level
elevated temperature to increase the rate of drying. The of residual endotoxin of below the calculated three-logs.
method used to dry endotoxin should be validated, spe- The negative control vials are tested in the same way as
cifically to ensure that the method does not affect the the spiked test vials. In order to claim depyrogenation,
challenge level of endotoxin. the device must show a three-log reduction of endotoxin
Prior validation work should be conducted to deter- of a 1000 EU or more challenge as per USP<1211>.
mine the time needed between spiking the vials and The following factors introduce variability into de-
placing them into the depyrogenation device (and at pyrogenation studies and may affect the success of the
what temperature they are required to be stored at, with study:
2-8∘C being typical). Work must also be carried out t
The material being challenged. For example, a
to determine the expiry time of vials that have passed glass vial behaves differently to an aluminum cap.
through the depyrogenation device (i.e., how long can t
For glass, the type of glass the challenge vials are
the vials remain before they are required to be tested. made from (e.g., Type I or Type II glass. Type I
Typically, this time should not exceed 24 hours). glass is borosilicate characterized by a high degree
Once the depyrogenation cycle has been completed, of hydrolytic stability; Type II glass is soda-lime).
the spiked containers or endotoxin vials are removed t
The type of depyrogenation device and its effi-
and tested using the LAL assay, and the remaining ciency. One key difference between dry heat ovens
level of endotoxin is assessed. This is performed by is the HEPA filter type and housing. Some filters

94 Journal of GXP Compliance

 Tim Sandle

shed a high number of particles during tempera- REFERENCES

ture transition (i.e., the rate of temperature 1. EU, Chapter 2.6.12, European Pharmacopoeia, edition 7.0,
change). Generally, the faster the temperature European Directorate for the Quality of Medicines and
increases the more particles that are generated. Healthcare, 2011.
This phenomenon is also affected by degree of 2. USP, Chapter <85>, United States Pharmacopeia, 34, USP
airflow uniformity and pressure balance. Some Convention, 2011.
of these particles may contain endotoxin or 3. Parillo, J. E., “Endotoxemia in Human Septic Shock,” Chest,
interfere with the LAL assay (24). January 1991.
t
The method used to dry the endotoxin to the 4. Department of Health, Education, and Welfare Public
container being tested. Service, Food and Drug Administration, ITG Subject: Bac-
t
The mechanism of the depyrogenation device. terial Endotoxins / Pyrogens, Date: 3/20/85 Number: 40 at:
Akers found that devices that dry heat depyro- http://www.fda.gov/ora/inspect_ref/itg/itg40.html (accessed
genate using infrared are more effective (25). 4th November 2010).
t
Different manufacturers endotoxin varies 5. Williams, K., “Differentiating Endotoxin Removal from In-
based on the extent of natural or artificial activation in Vial Component Depyrogenation Validations,”
contaminants. This is dependent upon how Pharmaceutical Technology, October 1995.
pure the endotoxin is (i.e., whether other cel- 6. Williams, K.L., Endotoxins: Pyrogens, LAL testing and
lular components are present) and whether the Depyrogenation, 2nd Edition. Drugs and the Pharmaceutical
endotoxin contains fillers, such as glycol. These Sciences Volume 111, Marcel Dekker Inc., New York, USA.
factors may increase or decrease the time taken Chapters 1, 2, 7 & 8, 2001.
to achieve heat inactivation. 7. Tsuji, K. et al., “Dry-Heat Destruction of Lipopolysaccha-
ride,” App. and Env. Microbial, 36:705-179, 1978.
CONCLUSION 8. Weary, M. and Pearon III, P., “A Manufacturer’s Guide to
This paper has examined a mechanism of depyro- Depyrogenation,” Biopharm, pp2- 7, 1988.
genation using dry heat inactivation. Depyrogena- 9. Guy, D., “Endotoxins and Depyrogenation” in Hodges, N.
tion forms part of a critical process in many phar- and Hanlon, G., Industrial Pharmaceutical Microbiology:
maceutical production facilities, particularly where Standards and Controls, Euromed, pp12.1 – 12.15, 2003.
glass vials and bottles are required for aseptic filling 10. Hou, K. C. and Zaniewski, R., “Depyrogenation by
operations. There remain some important questions Endotoxin Removal with Positively Charged Depth Filter
relating to depyrogenation studies, such as the num- Cartridge,” Journal of Parenteral Science and Technology, Vol.
ber of validation runs and frequency of re-validation. 44, No.4, pp204 – 209, 1990.
The number of validation runs is commonly set at 11. McCullough, K. Z. and Novitsky, T. J., “Detoxification of
three in order to demonstrate reproducibility, but Endotoxin by Acid and Base,” Ch.8, PDA Technical Report
this number is not fixed. The frequency of re-vali- No.7, Parenteral Drug Association, Inc., Philadelphia, 1985.
dation is to be determined by the user based on risk 12. Mosier, L. D., Bosworth, M.E., Jurgens Jr.,R.W., Mohl, M.H.
assessment, although annual re-qualification using and Rizzolo, R.R., “The Effects of pH. Time, Temperature,
endotoxin is not uncommon. and Autoclaving on E. coli Endotoxin in Selected Paren-
It should not be forgotten, even achieving success- teral Solutions,” PDA Journal of Pharmaceutical Science and
ful depyrogenation, that the pyroburden that pres- Technology, Vol. 41, No. 1, pp 21-25, 1987.
ents a risk to pharmaceuticals is derived from a com- 13. Moesby L, Timm M, Hansen E.W., “Effect of moist heat
bination of raw materials, water, active ingredients, sterilisation on the pyrogenicity of cell wall components
environment, and primary packaging materials. from Staphylococcus aureus,” Eur J Pharm Sci. Vol. 35, No.
Risks of endotoxin or other pyrogenic contamina- 5, pp442-6, 2008.
tion can arise from multiple sources and not simply 14. Fujii, S., Takai, M. and Maki, T., “Wet Heat Inactivation of
from the final containers alone. Lipopolysaccharide from E.coli Serotype 055:B5,” PDA Jour-

Autumn 2011 Volume 15 Number 4 95

Peer Reviewed

nal of Pharmaceutical Science and Technology, Vol. 56, No.4, ARTICLE ACRONYM LISTING
pp220-227, 2002. BET Bacterial Endotoxin Test
15. Sandle, T., “Some Considerations on the Use of Endotoxin CSE Control Standard Endotoxin
Indicators in Depyrogenation Studies,” Pharmaceutical DEAE Diethylaminoethyl Cellulose
Microbiology Forum Newsletter, Vol. 13, No. 10, November EU Endotoxin Units
2007. HCI Hydrochloric Acid
16. Wood, J. P., Lemieux P., Betancourt D., Kariher P. and HEPA High-Efficiency Particulate Air
Gatchalian N.G., “Dry thermal resistance of Bacillus LAL Limulus Amebocyte Lysate
anthracis (Sterne) spores and spores of other Bacillus spe- LPS Lipopolysaccharide
cies: implications for biological agent destruction via waste NaOH Sodium Hydroxide
incineration,” J. Appl. Microbiol, Vol. 109, No. 1, pp99-106, RO Reverse Osmosis
2010. USP United States Pharmacopeia
17. Nakata, T., “Destruction of typical endotoxins by dry heat WFI Water for Injection
as determined using LAL assay and pyrogen assay,” J Par-
enter Sci Technol, Sep-Oct;47(5):258-64, 1993. ABOUT THE AUTHOR
18. Ludwig, J. D. and Avis, K. E., “Dry Heat Inactivation of En- Tim Sandle, Ph.D., is the head of the microbiology department at
Bio Products Laboratory Limited, a pharmaceutical organization
dotoxin on the surface of Glass,” Journal of Parenteral Science
owned by the UK Department of Health. Dr. Sandle is a chartered
and Technology, pp4-12, Vol. 44, No.1, January-February biologist (Society for Biology) and holds a first class honors degree
1990. in applied biology, a Masters degree in education, and obtained his
19. Agalloco, J.P. and Carleton, F.J., Validation of Pharmaceutical doctorate from Keele University. He may be reached by e-mail at
Processes, Edition: 3, Published by CRC Press, p233, 2007. tim.sandle@bpl.co.uk; website: http://www.pharmig.blogspot.com.
20. Hecker, D., Witthaeur, D. and Staerk, A., “Validation of dry
heat inactivation of bacterial endotoxins,” PDA J Pharm Sci
Technol, 48(4):197-204. 1994.
21. Tours, N. and Sandle, T., “Comparison of Dry-Heat Depy-
rogenation using Three Different Types of Gram-negative
Bacterial Endotoxin,” European Journal of Parenteral and
Pharmaceutical Sciences, Volume 13, No.1, pp17-20, 2008.
22. Plant, I., “Destruction of Typical Endotoxins by Dry Heat as
Determined using LAL Assay and Pyrogen Assay,” J Parenter
Sci Technol, Sep-Oct; 47 (5):258-64, 1993.
23. Novitsky, T.J., “Limulus Amebocyte Lysate (LAL) Assays”
in Olson, W.P. (Ed.) Automated Microbial Identification and
Quantitation, pp277-298, 1996.
24. Melgaard, H., “Filter Shedding and Automatic Pressure
Balance in Batch Depyrogenation Ovens,” Pharmaceutical
Engineering, Vol. 8, No.6, pp37-43, 1988.
25. Akers, M.J., Avis, K.E., and Thompson, B., “Validation Stud-
ies of the Fostoria Infrared Tunnel Sterilizer,” PDA Journal
of Pharmaceutical Science and Technology, 34 (5): 330-347,
1980. GXP

96 Journal of GXP Compliance