highly sensitive pyrogen detection on medical devices by the monocyte activation test
TRANSCRIPT
Highly sensitive pyrogen detection on medical devicesby the monocyte activation test
Katharina Stang • Stefan Fennrich • Stefanie Krajewski • Sandra Stoppelkamp •
Iwan Anton Burgener • Hans-Peter Wendel • Marcell Post
Received: 30 October 2013 / Accepted: 23 December 2013
� Springer Science+Business Media New York 2014
Abstract Pyrogens are components of microorganisms,
like bacteria, viruses or fungi, which can induce a complex
inflammatory response in the human body. Pyrogen con-
tamination on medical devices prior operation is still critical
and associated with severe complications for the patients.
The aim of our study was to develop a reliable test, which
allows detection of pyrogen contamination on the surface of
medical devices. After in vitro pyrogen contamination of
different medical devices and incubation in a rotation model,
the human whole blood monocyte activation test (MAT),
which is based on an IL-1b-specific ELISA, was employed.
Our results show that when combining a modified MAT
protocol and a dynamic incubation system, even smallest
amounts of pyrogens can be directly detected on the surface
of medical devices. Therefore, screening of medical devices
prior clinical application using our novel assay, has the
potential to significantly reduce complications associated
with pyrogen-contaminated medical devices.
1 Introduction
In clinical routine, application of blood contacting medi-
cal devices, like stents, vascular grafts, components of
extracorporeal circulation systems or surgical instruments,
is associated with potential risks for the patients due to
different causes. Firstly, the material used for the medical
device itself can lead to various activation mechanisms
upon contact with blood or tissues, like thrombosis or
inflammation. On the other hand, contamination of med-
ical devices with pyrogens prior to the implantation can
lead to adverse reactions, such as alterations in hemos-
tasis, release of pro-inflammatory cytokines and induction
of fever [1].
Most pyrogens are of microbial origin, like bacteria [2],
viruses [3], yeasts and fungi [4, 5], but also environmental
particles can be considered as pyrogens [6]. The best
investigated pyrogens are components of bacterial cell
walls, such as lipopolysaccharides (LPS) of Gram-negative
bacteria or lipoteichoic acid (LTA) of Gram-positive bac-
teria, which lead to an activation of the immune system
upon contact [7]. The resulting symptoms are diverse,
including fever [8], systemic inflammatory response syn-
drome (SIRS) or even sepsis [9–12]. Also hypotension,
nausea, shivering, shock and even disseminated intravas-
cular coagulation (DIC) can occur [13].
Pyrogenic contamination of medical devices mostly
occurs during the production process through air contam-
ination, use of different detergents like metal working
fluids, contaminated packaging materials or microbial
remainings after sterilization [14]. To inactivate pyrogens a
dry heat treatment of the medical devices at [180 �C for
several hours is necessary [15]. So, pyrogenicity tests
should be performed as part of the individual company‘s
quality management.
Therefore, the U.S. Food and Drug Administration
(FDA) specifies an endotoxin limit of 0.5 EU/ml for
medical devices, 0.06 EU/ml for devices in contact with
liquor and 5 EU/kg/h for parenteral drugs [16, 17].
K. Stang � S. Fennrich � S. Krajewski � S. Stoppelkamp �H.-P. Wendel (&) � M. Post
Clinical Research Laboratory, Department of Thoracic, Cardiac
and Vascular Surgery, University Hospital Tuebingen,
Tuebingen University, Calwerstr. 7/1, 72076 Tuebingen,
Germany
e-mail: [email protected]
I. A. Burgener
Department of Small Animal Medicine, Faculty of Veterinary
Medicine, University of Leipzig, An den Tierkliniken 23,
04103 Leipzig, Germany
123
J Mater Sci: Mater Med
DOI 10.1007/s10856-013-5136-6
In order to prevent the potential use of contaminated
materials in clinical routine, two different tests for the
detection of pyrogens are currently available: the limulus
amoebocyte lysate assay (LAL), a bacterial endotoxin test
(BET) described in the U.S. Pharmacopeial convention
chapter 85 and the rabbit pyrogen test (RPT) [16].
For the RPT, an eluate of the test material is injected
intravenously to rabbits or alternatively the medical device
itself is implanted in rabbits. Afterwards, the rectal tem-
perature of the animals is monitored for several hours. The
presence of pyrogens is associated with a significant rise in
body temperature [18]. However, besides being an animal
experiment, the implantation of a medical device itself can
cause inflammatory reactions resulting in increased body
temperature unrelated to any potential pyrogenic contam-
ination [15]. Moreover, sensitivity to pyrogens varies
between each animal depending on sex and age and in
some cases animal studies are not reflecting the reactions
taking place in the human body [19–21].
An alternative in vitro quality control was established
with the LAL test, which is based on the coagulation
reaction of horseshoe crabs hemolymph after contact with
LPS [22]. Unfortunately, this test only detects endotoxins
in aqueous eluates of medical devices, but not on surfaces.
Furthermore, other pyrogens, like LTA, viruses or yeasts
are not detectable using the LAL test [23].
Recently, the monocyte activation test (MAT) proce-
dure, with its five variants, has been implemented in the
European Pharmacopoeia [17]. This test is based on the
activation of human monocytes by pyrogens and the
measurement of the resulting cytokine release by ELISA
allowing for the first time pyrogen detection within the
human species [24, 25]. However, the MAT is only vali-
dated for pharmaceutical solutions so far and not for
medical devices [26], despite having great potential for
further applications [27].
The aim of our study was to develop a sensitive and
reliable assay for the detection of pyrogens on the surface
of medical devices. Previous studies already introduced the
MAT for the testing of artificial endotoxin contamination
on different metals used for medical devices [23]. Also on
aneurysm clips (implanted for the treatment of cerebral
aneurysms) and intraocular lenses pyrogen contaminations
were detected [15, 28]. Another field of application was
created by the evaluation of cleaning methods of medical
devices after the production process [15, 23]. In our study,
however, detection of pyrogen contamination on different
implantable devices was not reliable using the common
MAT protocol. Hence, we develop a new method com-
bining a modified MAT protocol and a dynamic incubation
system (Fig. 1). This new method allows highly sensitive
detection of LPS and LTA on the surface of medical
devices consisting of different materials.
2 Materials and methods
2.1 Blood collection
All blood sampling procedures and the use of blood in the
described experimental settings were specifically approved
by the Research and Ethics Unit of the University of Tueb-
ingen, Germany (Project approval number 270/2010BO1).
Before blood sampling written informed consent was
obtained from all voluntary blood donors. Fresh human
whole blood was obtained by venipuncture with a 20Gx3/
4TW needle (Sarstedt, Nuembrecht, Germany) from healthy
adult blood donors into heparinised monovettes (19 I.E.;
Sarstedt). For cell count analysis, the blood was taken using
an EDTA monovette (Sarstedt) and measured with the cell
counter (ABX, Micros 60, HORIBA medical, USA).
2.2 Sample materials
As pyrogenic stimuli, LPS (WHO International Standard
Endotoxin E. coli O113:H10:K, 2nd International Stan-
dard, NIBSC code: 94/580, UK) and LTA (purified LTA
SA, InvivoGen, code: tlrl-pslta, Germany) as a non-endo-
toxin were used. They were reconstituted and stored as
recommended in the manufacturer’s instructions. For the
standard curves LPS and LTA were dissolved either in
physiological saline solution or pyrogen-free water
(Charles River). One EU (Endotoxin Unit) of the WHO
International Standard Endotoxin E. coli O113:H10:K
corresponds to 100 pg of endotoxin.
Stainless steel plates made of 1.4301 steel with dimen-
sions of 8 9 8 9 0.5 mm (Rocholl GmbH, Aglasterhausen,
Germany) were used for pretesting. The electropolishing was
performed by ?ELYPO- Metallveredelung, Germany.
Standard endoluminal stents made of cobalt chromium
were obtained from Qualimed (Qualimed, Winsen, Ger-
many) and ePTFE vascular grafts from Jotec (Jotec GmbH
Hechingen, Germany).
2.3 Sample preparation
The electropolished stainless steel plates were washed in
acetone and sterile water before heat treatment at 250 �C
for at least 8 h. The stents and ePTFE vascular grafts were
only heat treated at 250 �C for at least 8 h.
2.4 Evaluation of interference of test materials
with the monocyte activation test
As test for interference for all experiments a liquid LPS-
and LTA- standard curve with or without the different test
materials in a range of 0–20 pg for LPS or 0–25 lg for
J Mater Sci: Mater Med
123
LTA was performed. The different LPS or LTA amounts
were each incubated with 1.2 ml diluted whole blood (1:12
physiological saline) overnight at 37 �C, before centrifu-
gation at 300 g for 5 min. The supernatants were then
stored at -20 �C until further ELISA analysis, if not used
instantly.
2.5 Sample contamination with LPS and LTA
In order to contaminate different sample surfaces, two
methods were used.
2.5.1 Pyrogen contamination by liquid incubation
The stents were incubated overnight or at least 10 h in
physiological saline solution (Fresenius Kabi) containing
200, 400 pg LPS or 25, 50 lg LTA. Afterwards, stents
were air-dried at room temperature before use.
2.5.2 Surface drying
For the second contamination method, 10 ll of a LPS-
solution containing 5, 10 or 20 pg or a LTA-solution
containing 2.5, 5 or 10 lg was evenly spread over each test
sample surface (i.e. (a) stainless steel plates; (b) stents or
(c) ePTFE vascular grafts) and dried overnight at room
temperature.
2.6 Detection of pyrogen contamination on various
surfaces
For all experiments a liquid standard curve with LPS
(0–20 pg) or LTA (0–10 lg) was performed for reference
with the different test material.
2.6.1 Pyrogen detection using the MAT under static
conditions
After contamination by liquid incubation, stents were
incubated in 6-well plates under static conditions with 6 ml
of diluted whole blood (1:12 in physiological saline)
overnight at 37 �C. The plates were placed in a 35� angle,
in order to completely cover the stents with blood.
After contamination using the surface drying method,
stainless steel plates and stents were transferred to 24-
or 48-well plates (Greiner Bio One) or reaction tubes
(Eppendorf, DNA LoBind) and incubated with 1.2 ml of
diluted whole blood at 37 �C overnight.
2.6.2 Pyrogen detection under dynamic conditions
by a novel modified MAT
All samples contaminated by surface drying were incubated
under dynamic conditions at 10 rpm (rotator, neoLab, Hei-
delberg Germany) with 1.2 ml of physiological saline-diluted
whole blood (1:12 dilution) or 1.2 ml of diluted whole blood in
pyrogen-free water including 100 ll saline concentrate solu-
tion (containing 9.9 mg saline) to reach physiological saline
conditions (1:12 dilution, modified MAT) overnight at 37 �C.
In another set of experiments, contaminated stainless
steel plates and stents were pre-incubated in reaction tubes
for 1, 2 or 24 h in a thermomixer at 350 rpm and tem-
peratures of 26, 37 and 72 �C to solve surface bound
endotoxins. After cooling to room temperature the samples
and eluates were incubated with diluted whole blood.
The ePTFE vascular grafts were cut into pieces of
10 mm in diameter and put in caps of reaction tubes before
incubation with diluted whole blood.
The following day all samples were resuspended and cen-
trifuged at 300 g for 5 min and the supernatant was transferred
Fig. 1 Schematic overview showing the static incubation method as well as the newly developed dynamic incubation method of the MAT for the
detection of pyrogenic contaminated medical implants
J Mater Sci: Mater Med
123
to new reaction tubes. The supernatants were stored at -20 �C
until further ELISA analysis, if not used instantly.
2.7 Detection of endotoxin contamination by LAL
After contamination with a defined amount of LPS-solu-
tion, the stainless steel plates and stents were put in reac-
tion tubes before pre-incubation with 1,000 ll pyrogen-
free water for 1, 2 or 24 h in a thermomixer at 26, 37 or
72 �C. After pre-incubation, the supernatants were mea-
sured by the Endosafe�-PTSTM (Charles river) with a
detection limit of 0.005–0.5 EU/ml endotoxin as described
in the manufacturer’s instructions. The stainless steel plates
were transferred to new reaction tubes and incubated with
diluted whole blood under dynamic conditions for further
analysis in the modified MAT.
2.8 Enzyme-linked immunosorbent assay (ELISA)
In order to analyze the supernatants, a validated in-house
sandwich ELISA was performed measuring the IL-1b cyto-
kine expression based on matching antibodies against human
IL-1b (R&D Systems, Germany) diluted with PBS contain-
ing 3 % BSA (R&D Systems). Coating of a U-bottom
96-well-microtiter plate (Nunc, Roskilde, Denmark) was
performed at 8 �C using 50 ll coating monoclonal anti-IL-1bantibody (concentration 4 lg/ml¸ R&D Systems) per well.
After 18–24 h of incubation, unbound antibody was removed
and 200 ll/well blocking buffer solution (PBS containing
3 % BSA) was added and incubated for 2 h at 25 �C. After-
wards plates were stored at -20 �C until use. Before use, the
plate was washed three times with 250 ll/well PBS con-
taining 0.05 % (v/v) Tween 20 (MerckMillipore, Darmstadt,
Germany). Afterwards, 50 ll of sample supernatant as well
as 50 ll of a biotinylated antibody (concentration: 0.2 lg/ml;
R&D Systems) were added per well and incubated for 2 h at
room temperature. After three washing steps, 100 ll of
streptavidin–peroxidase (R&D Systems) was added to each
well and incubated at room temperature for 30 min. Again
washing was performed four times followed by addition of
100 ll/well of 3,30,5,50tetramethylbenzidine/H2O2 (1:2)
(R&D Systems). The reaction was stopped by adding 50 ll/
well 1 M H2SO4 after 30 min. The absorbance was measured
at 450 nm in a microplate reader (Mithras, Bertold Tech-
nologie, Bad Wildbad, Germany).
3 Results
3.1 Evaluation of the interference of test materials
with the MAT
To exclude possible influences like activation or inhibition
of the different test materials on the MAT, all test materials
were incubated with increasing concentrations of liquid
preparations of LPS or LTA and diluted whole blood.
All sample materials incubated with various LPS or LTA
amounts were in between the valid specification limits of the
test (50–200 % recovery) and comparable to the liquid ref-
erence standards. Therefore, no interference effects with the
test materials and the MAT could be detected. Results of
interference tests for all materials displaying LPS- or LTA-
recovery are summarized in Table 1a, b, respectively.
3.2 Detection of pyrogen contamination on various
surfaces
3.2.1 Pyrogen detection using the MAT under static
conditions
For pyrogen detection on material surfaces, contamination of
testing devices was realized by incubation of stents in liquid
preparations with different LPS or LTA concentrations as
described. The stents used for contamination in LPS-prepa-
rations (400 or 200 pg dissolved in 1 ml) exhibited a low
signal in the MAT assay when compared to the liquid LPS
Table 1 Test for interference
Sample 20 pg 10 pg 5 pg 2.5 pg 1.25 pg
(a)
Stents 90.1 ± 8.3 86.0 ± 8.5 77.4 ± 8.1 52.5 ± 42.9 52.4 ± 8.5
Recovery LPS (%) ± SDStainless steel 94.7 ± 0.1 99.2 ± 3.8 88.8 ± 0.7 82.4 ± 3.6 100.9 ± 3.2
ePTFE grafts 86.3 ± 1.0 79.5 ± 2.9 113.7 ± 0.4 70.0 ± 0.1 103.2 ± 4.8
Sample 10 lg 5 lg 2.5 lg 1.3 lg 0.3 lg
(b)
Stents 86.9 ± 13.6 89.8 ± 3.4 74.3 ± 3.4 91.3 ± 33.6 74.3 ± 15.8
Recovery LTA (%) ± SDStainless steel 98.3 ± 4.6 98.9 ± 1.5 104.5 ± 3.4 93.6 ± 1.2 66.0 ± 15.3
ePTFE grafts 90.6 ± 0.3 101.9 ± 2.2 92.2 ± 2.0 98.8 ± 2.5 91.0 ± 4.1
Test for interference of liquid standard preparations of LPS (a) and LTA (b) with testing devices (n = 3)
J Mater Sci: Mater Med
123
standard curve as shown in Fig. 2a. The same effect was seen
in LTA-contaminated test samples (Fig. 2b).
In order to develop a reliable test for pyrogen detection
and since exact pyrogen amounts could not be verified on
the material surfaces with the ‘‘pyrogen contamination by
liquid incubation’’ procedure, contamination with a known
amount of LPS or LTA, which was applied directly on the
surfaces, was performed. The MAT procedure was effec-
tive in detecting surface dried amounts of LPS (Fig. 3a)
and LTA (data not shown) on contaminated stents.
However, contaminated stainless steel samples with a size
of 8 9 8 mm and ePTFE grafts induced only a low signal in
the MAT assay when compared to the liquid LPS standard
curve. These effects could be seen in all used incubation
containers and are exemplarily shown for polystyrene well
plates and stainless steel samples in Fig. 3b.
3.2.2 Pyrogen detection under dynamic conditions
by a novel modified MAT
For this experiment blood contact time on the contaminated
material surfaces was increased by a dynamic rotation
model, which keeps the blood cells during the incubation in
suspense preventing sedimentation of the blood cells.
Fig. 2 Pyrogen contamination
by liquid incubation and
pyrogen detection using the
MAT under static conditions:
comparison of optical density
(mean ± SD) of IL-1b specific
ELISA from residues on stents
after incubation in 400 or
200 pg/ml LPS preparations (a),
and 50 or 25 lg/ml LTA
preparations (b), using the
contamination in liquid
procedure (grey bars). Black
bars represent the liquid
standard preparation for
reference
J Mater Sci: Mater Med
123
Additionally, the modified MAT protocol with pyrogen-
free water and saline concentrate solution instead of the
standard variant with 0.9 % saline was performed. Both
modifications together contribute to significantly increased
pyrogen recovery rates as shown for LPS on stainless steel
samples (Fig. 4a), ePTFE grafts (Fig. 4b) and even on
stents (Fig. 4c) the recovery rate could be enhanced.
With the modified MAT assay, the pyrogen recovery
could also be improved greatly for LTA contaminations on
stainless steel samples (Fig. 5a), ePTFE grafts (Fig. 5b)
and on stents (Fig. 5c) with the dynamic rotation model.
3.3 Detection of endotoxin contamination by LAL
and modified MAT
After the elution process of surface dried endotoxins from
stainless steel samples or stents under different temperature
and incubation time conditions, eluates were tested with the
LAL assay. For all samples cartridges with a detection limit
of 0.005 EU/ml were used. The results of eluates obtained
after incubation of contaminated stainless steel samples or
stents with pyrogen free water tested with the LAL assay are
shown in Table 2.
In the eluates extremely low endotoxin levels were mea-
sured, which was also the case when other elution tempera-
tures or times were employed (data not shown). These results
were confirmed when eluates were analyzed using the mod-
ified MAT assay for stainless steel samples or stents (data
exemplarily shown for steel samples in Fig. 6). After elution,
the steel samples and stents were further incubated in the
modified MAT test and the pyrogen recovery was compared
to contaminated samples without prior pyrogen elution.
Recovery rates of samples after the elution process were
located between 71 and 85 %, whereas samples incubated
directly in the modified MAT showed recovery rates among
76 and 96 %.
Fig. 3 Pyrogen contamination
by surface drying followed by
pyrogen detection using the
MAT under static conditions:
IL-1b ELISA signal induced by
LPS residues on stents (a), and
8 9 8 mm stainless steel
samples in 24-well polystyrene
plates (b), contaminated with
well-defined amounts of LPS
(grey bars) in comparison to the
respective liquid standard
preparations (black bars). Mean
OD values ± SD are shown
J Mater Sci: Mater Med
123
Fig. 4 Pyrogen detection under
dynamic conditions by a novel
modified MAT on stainless
steel, stents and ePTFE grafts:
comparison of optical density
(mean ± SD) of IL-1b specific
ELISA from supernatant after
whole blood incubation.
Different amounts of standard
preparation for reference are
shown as black bars. Dark grey
bars represent measurements of
surface dried LPS on stainless
steel (a), ePTFE grafts (b), or
stents (c), with modified MAT
procedure. Light grey bars
represent OD signals after
whole blood incubation with the
standard MAT procedure
J Mater Sci: Mater Med
123
Fig. 5 Pyrogen detection under
dynamic conditions by a novel
modified MAT on stainless
steel, stents and ePTFE grafts:
comparison of optical density
(mean ± SD) of IL-1b specific
ELISA from supernatant after
whole blood incubation.
Different amounts of standard
preparation for reference are
shown as black bars. Dark grey
bars represent measurements of
surface dried LTA on stainless
steel (a), ePTFE grafts (b), or
stents (c), with modified MAT
procedure. Light grey bars
represent OD signals after
whole blood incubation with the
standard MAT procedure
J Mater Sci: Mater Med
123
4 Discussion
It is widely known that surface components of microor-
ganisms or virus particles can cause fever-inducing reac-
tions or under critical circumstances SIRS or even sepsis in
humans or animals [9–12]. For these biological particles
autoclave procedures are not sufficient to inactive them.
Only dry heat treatment at 180 �C or higher for several
hours or aggressive chemical solutions are able to fully
extinguish the biological activity of these molecules [15].
Therefore, it is essential for the health of patients to screen
for such components in medical preparations or on medical
devices. For the detection of endotoxins, especially LPS,
different testing systems are available. In contrast to the
common limulus amoebocyte lysate test or the recombinant
Factor C Test (an alternative endotoxin test based on the
recombinant Factor C protein from the LAL assay) [29–
31], the MAT and the RPT assays sensitively detect par-
ticles of Gram-positive bacteria, yeast, fungi or viruses [32,
33]. So far, the field of application for these endotoxin tests
is primarily focused on parenteralia or pure water testings.
Hence, the main focus is on testing of liquid preparations.
Detection of pyrogenic components, especially LPS, on
medical surfaces requires time intensive elution steps,
because most of the pyrogenic substances have a high
affinity to adhere or remain on surfaces. So it is doubtful
whether the use of an elution procedure for a quantitative
detection of surface bound endotoxins is appropriate [34].
This was also confirmed by our results indicating poor
elution of pyrogens after employing different elution times,
temperatures and pH values.
The aim of our study was to detect low amounts of
endotoxins or non-endotoxin pyrogens on different mate-
rials with a MAT-based assay. The use of the standard
MAT allowing detection of endotoxins on a titanium-based
medical device [15], on different intraocular lenses mate-
rials or on gelatin polymers was already shown before [28,
35]. However, in our study detection of pyrogens on
stainless steel and ePTFE was not satisfactory with the
standard MAT protocol. Therefore, a modified MAT pro-
tocol was developed. Additionally, we compared the
detection of low amounts of endotoxins in our modified
whole blood MAT with the LAL or standard MAT assays.
It can be stated that pyrogen detection in eluates com-
pletely underestimated the contamination level on the
medical devices. However, direct incubation of these con-
taminated samples in pyrogen-free water with saline con-
centrate solution in the dynamic rotation model yielded
more than 90 % pyrogen recovery rates indicated by IL-1brelease from whole blood cells. Therefore, we could dem-
onstrate with the dynamic model that the pyrogens still
remain in an active form on the medical surfaces after the
elution procedure.
Table 2 Detection of endotoxin contamination by LAL
Sample
name
(pg)
Elution
procedure
Sample
material
Liquid LPS
preparation
(EU/ml)
Eluate
(EU/
ml)
Recovery
rate (%)
40 37 �C 1 h stents 0.267 \0.007 \3
40 72 �C
24 h
stents 0.269 \0.008 \3
40 37 �C 1 h stainless
steel
0.288 under detection
limit
40 72 �C
24 h
stainless
steel
0.283 under detection
limit
LAL measurement of the eluate of the 40 pg stainless steel samples
and stents after 24 h at 72 �C and 1 h at 37 �C
Fig. 6 Detection of endotoxin
contamination on stainless steel
as well as in eluates by modified
MAT: comparison of surface
dried LPS on stainless steel
samples directly incubated in
the modified MAT (dark grey
bars) or after 24 h at 72 �C
elution process (light grey).
White bars demonstrate
measurements of eluates (which
were also used for the LAL
assay) and black bars represent
the liquid LPS preparation for
reference. Shown are mean OD
values ± SD
J Mater Sci: Mater Med
123
In order to avoid interference effects of the medical
devices or the incubation containers on the MAT, it is
pivotal that the described interference test has to be indi-
vidually performed for each material. None of the experi-
mental procedures used to elute surface-bound pyrogenic
contaminations from diverse materials could achieve
results above the detection limit in both, the MAT and the
LAL test, independent of 1, 2 or even 24 h pre-incubation
time and different treatment conditions (e.g. various tem-
peratures, intensive mixing).
Our data further indicate that two modifications of the
standard test greatly improved recovery rates. Namely,
rotation during the incubation time to keep blood cells in
suspension (especially on ePTFE grafts and stainless steel
plates) and the use of pyrogen-free water and additional
saline concentrate solution to realize isotonic conditions.
Combining these modifications better pyrogen recoveries
were achieved than with the standard incubation protocol
using 0.9% saline. Until now it is not completely known
why the switch to pyrogen-free water and additional saline
concentrate solution achieves better recovery rates. Inter-
ferences by contaminations could be excluded, because no
activation in the blank samples was measured. Unsteadi-
ness in the saline dilution could be excluded by conduc-
tivity measurements of the sample solutions (data not
shown).
Although short-time exposure of LPS is sufficient to
activate monocytes in liquid preparations [36], a short
contact time with contaminated medical devices did not
yield the expected recovery rates. Using our dynamic
model to ascertain pyrogens on medical surfaces, we
demonstrated a possibility to detect contaminations below
the current required limits for implants (0.5 EU/ml), which
cannot be detected with the used standard MAT protocol or
the commonly used LAL assay. Although the modified
MAT assay takes up to 20 h at the moment, which is
comparable with the time needed for the RPT, it has sev-
eral crucial advantages. In the modified MAT, the test
material is directly incubated, so no underestimation of
pyrogenic burdens by eluting the devices can occur. Fur-
thermore, no animal testing is performed. The most sig-
nificant aspect of our study is a quantification of a full
pyrogen spectrum with a test system based on the human
species, which is not achieved by the LAL and RPT test
systems.
5 Conclusion
Our proposed rotation model combined with the modified
MAT assay offers a novel reliable and sensitive test system
for pyrogen quantification of even smallest amounts of
pyrogens. Our work clearly demonstrates that the modified,
dynamic MAT greatly enhanced the detection of pyrogenic
burden on the surface of medical devices, whereas the
standard MAT assay and elution procedures might lead to
an underestimation of the actual contamination load.
Furthermore, the modified MAT assay allows scale-ups
to detect higher pyrogen burdens and even large medical
devices can be tested dynamically by increasing the liquid
components of the assay and using a 3D-agitator for blood
suspension.
Therefore, the modified MAT could be performed as
part of the manufacturers’ individual quality management
to provide product release decisions and in process control.
Applying this novel test method would greatly improve
the safety of patients, who require implantation of a med-
ical device. We therefore recommend that testing of med-
ical devices should be performed with prudence, always
taking into account the full range of pyrogenic contami-
nations and their characteristics (e.g. differences in their
soluble and surface bound forms).
Acknowledgments The authors would like to express their grati-
tude to the companies Qualimed and Jotec, who generously made the
stents and vascular grafts available.
References
1. Sefton MV, Gorbet MB. Biomaterial-associated thrombosis: roles
of coagulation factors, complement, platelets and leukocytes.
Biomaterials. 2004;25:5681–703.
2. Probey TF, Pittman M. The pyrogenicity of bacterial contami-
nants found in biologic products. J Bacteriol. 1945;50:397–411.
3. Atkins E, Huang WC. Studies on the pathogenesis of fever with
influenzal viruses. I. The appearance of an endogenous pyrogen
in the blood following intravenous injection of virus. J Exp Med.
1958;107:383–401.
4. Kobayashi GS, Friedman L. Characterization of the pyrogenicity
of Candida albicans, Saccharomyces cerevisiae, and Crypto-
coccus neoformans. J Bacteriol. 1964;88:660–6.
5. Braude AI, McConnell J, Douglas H. Fever from pathogenic
fungi. J Clin Invest. 1960;39:1266–76.
6. Monn C, Becker S. Cytotoxicity and induction of proinflamma-
tory cytokines from human monocytes exposed to fine (PM2.5)
and coarse particles (PM10–2.5) in outdoor and indoor air.
Toxicol Appl Pharmacol. 1999;155:245–52.
7. Ginsburg I. Role of lipoteichoic acid in infection and inflamma-
tion. Lancet Infect Dis. 2002;2:171–9.
8. Dinarello CA. Review: infection, fever, and exogenous and
endogenous pyrogens: some concepts have changed. J Endotoxin
Res. 2004;10:201–22.
9. Rotta J. Endotoxin-like properties of the peptidoglycan. Z Im-
munitatsforsch Exp Klin Immunol. 1975;149:230–44.
10. Atkins E, Morse SI. Studies in staphylococcal fever. VI.
Responses induced by cell walls and various fractions of staph-
ylococci and their products. Yale J Biol Med. 1967;39:297–311.
11. Morath S, Geyer A, Hartung T. Structure-function relationship of
cytokine induction by lipoteichoic acid from Staphylococcus
aureus. J Exp Med. 2001;193:393–7.
12. Watson DW. Host-parasite factors in group A streptococcal
infections. Pyrogenic and other effects of immunologic distinct
J Mater Sci: Mater Med
123
exotoxins related to scarlet fever toxins. J Exp Med.
1960;111:255–84.
13. Lemke HD. Methods for the detection of endotoxins present
during extracorporeal circulation. Nephrol Dial Transplant.
1994;9(Suppl 2):90–5.
14. Luginbuehl Reto FA. Analysis of endotoxin residues on cleaned
implant materials. J ASTM Intern. 2008;5(2).
15. von Aulock S, Mazzotti F, Beuttler J, Zeller R, Fink U, et al.
In vitro pyrogen test–a new test method for solid medical devices.
J Biomed Mater Res A. 2007;80A:276–82.
16. FDA Guideline for industry. 2012:8.
17. European Pharmacopoeia (EP). 2010;6.7:5440–45.
18. ISO 10993 part 11 biological evaluation of medical devices test
for systemic toxicity. 2009:29.
19. Lipton JM, Ticknor CB. Influence of sex and age on febrile
responses to peripheral and central administration of pyrogens in
the rabbit. J Physiol. 1979;295:263–72.
20. Greisman SE, Hornick RB. Comparative pyrogenic reactivity of
rabbit and man to bacterial endotoxin. Proc Soc Exp Biol Med.
1969;131:1154–8.
21. Wilsnack RE. Quantitative cell culture biocompatibility testing of
medical devices and correlation to animal tests. Biomater Med
Devices Artif Organs. 1976;4:235–61.
22. Levin J, Bang FB. The role of endotoxin in the extracellular
coagulation of limulus blood. Bull Johns Hopkins Hosp.
1964;115:265–74.
23. Hasiwa M, Kullmann K, von Aulock S, Klein C, Hartung T. An
in vitro pyrogen safety test for immune-stimulating components
on surfaces. Biomaterials. 2007;28:1367–75.
24. Hartung T, Aaberge I, Berthold S, Carlin G, Charton E, et al.
Novel pyrogen tests based on the human fever reaction–the report
and recommendations of ECVAM workshop 43. Altern Lab
Anim. 2001;29:99–123.
25. Fennrich S, Fischer M, Hartung T, Lexa P, Montag-Lessing T,
et al. Detection of endotoxins and other pyrogens using human
whole blood. Dev Biol Stand. 1999;101:131–9.
26. Hartung T, Schindler S, von Aulock S, Daneshian M. Develop-
ment, validation and applications of the monocyte activation test
for pyrogens based on human whole blood. Altex. 2009;26:265–77.
27. Fennrich S, Wendel A, Hartung T. New applications of the
human whole blood pyrogen assay (PyroCheck). Altex.
1999;16:146–9.
28. Werner L, Tetz M, Mentak K, Aldred M, Zwisler W. Detection of
pyrogens adsorbed to intraocular lenses: evaluation of limulus
amoebocyte lysate and in vitro pyrogen tests. J Cataract Refract
Surg. 2009;35:1273–80.
29. Levin J, Bang FB. A description of cellular coagulation in the
Limulus. Bull Johns Hopkins Hosp. 1964;115:337–45.
30. Tokunaga F, Nakajima H, Iwanaga S. Further studies on lipo-
polysaccharide-sensitive serine protease zymogen (factor C): its
isolation from Limulus polyphemus hemocytes and identification
as an intracellular zymogen activated by alpha-chymotrypsin, not
by trypsin. J Biochem. 1991;109:150–7.
31. Ding JL, Ho B. A new era in pyrogen testing. Trends Biotechnol.
2001;19:277–81.
32. Hartung T, Fennrich S, Fischer M, Montag-Lessing T, Wendel A.
Development and evaluation of a pyrogen test based on human
whole blood. Altex. 1998;15:9–10.
33. van Dijck P, van de Voorde H. Factors affecting pyrogen testing
in rabbits. Dev Biol Stand. 1977;34:57–63.
34. Pfeiffer M. Testing medical disposables using the limulus amoe-
bocyte lysate (LAL) test. Med Device Technol. 1990;1:37–41.
35. Mohanan PV, Banerjee S, Geetha CS. Detection of pyrogenicity
on medical grade polymer materials using rabbit pyrogen, LAL
and ELISA method. J Pharm Biomed Anal. 2011;55:1170–4.
36. Gallay P, Jongeneel CV, Barras C, Burnier M, Baumgartner JD,
et al. Short time exposure to lipopolysaccharide is sufficient to
activate human monocytes. J Immunol. 1993;150:5086–93.
J Mater Sci: Mater Med
123