C1-esterase inhibitor attenuates the inflammatory response during human endotoxemia.

Abstract

Besides its role in regulation of the complement and contact system, C1-esterase inhibitor has other immunomodulating effects that could prove beneficial in patients with acute inflammation such as during sepsis or after trauma. We examined the immunomodulating properties of C1-esterase inhibitor during human experimental endotoxemia, in which the innate immune system is activated in the absence of activation of the classic complement pathway.
Double-blind placebo-controlled study.
Research intensive care unit of the Radboud University Nijmegen Medical Centre.
Twenty healthy volunteers.
Intravenous injection of 2 ng/kg Escherichia coli lipopolysaccharide. Thirty minutes thereafter (to prevent binding of lipopolysaccharide), C1-esterase inhibitor concentrate (100 U/kg, n = 10) or placebo (n = 10) was infused.
Pro- and anti-inflammatory mediators, markers of endothelial and complement activation, hemodynamics, body temperature, and symptoms were measured. C1-esterase inhibitor reduced the release of proinflammatory cytokines as well as C-reactive protein (peak levels of: interleukin-6 1521 ± 209 vs. 932 ± 174 pg/mL [p = .04], tumor necrosis factor-α 1213 ± 187 vs. 827 ± 167 pg/mL [p = .10], monocyte chemotactic protein-1 6161 ± 1302 vs. 3373 ± 228 pg/mL [p = .03], interleukin-1β 34 ± 5 vs. 23 ± 2 pg/mL [p < .01], C-reactive protein 39 ± 4 vs. 29 ± 2 mg/L [p = .02]). In contrast, release of the anti-inflammatory cytokine interleukin-10 was increased by C1-esterase inhibitor (peak level 73 ± 11 vs. 121 ± 18 pg/mL, p = .02). The increase in interleukin-1 receptor antagonist tended to be smaller in the C1-esterase inhibitor group, but this effect did not reach statistical significance (p = .07). Markers for endothelial activation were increased after lipopolysaccharide infusion, but no significant differences between groups were observed. The lipopolysaccharide-induced changes in heart rate, blood pressure, body temperature, and symptoms (all p < .001 over time) were not influenced by C1-esterase inhibitor. Complement fragment C4 was not increased after lipopolysaccharide challenge.
This study is the first to demonstrate that C1-esterase inhibitor exerts anti-inflammatory effects in the absence of classic complement activation in humans.

Full text

Page 1

C1-esterase inhibitor attenuates the inflammatory response during
human endotoxemia
Mirrin J. Dorresteijn, MD; Tjaakje Visser, MD; Laura A. E. Cox, BSc; Martijn P. Bouw, CCRN;
Janesh Pillay, MD; Anky H. L. Koenderman, PhD; Paul F. W. Strengers, MD; Luke P. H. Leenen, MD, PhD;
Johannes G. van der Hoeven, MD, PhD; Leo Koenderman, PhD; Peter Pickkers, MD, PhD
A
(1–4). The parenchymal damage and sub-
sequentorgandysfunctionarecausedbyan
cute respiratory distress syn-
drome and multiple organ dys-
function syndrome are the lead-
ing causes of death in medical
and surgical intensive care unit patients
overactivated inflammatory response (5).
The systemic release of several humoral
inflammatory mediators such as tumor ne-
crosis factor-?, interleukin (IL)-1?, and
IL-6 activate the vascular endothelium and
modulate activation and tissue infiltration
of circulating leukocytes (5–7).
A promising intervention to modu-
late the innate inflammatory response
is treatment with a high concentration
of C1-esterase inhibitor (C1INH).
C1INH is an acute-phase protein pro-
duced by the liver and important in
regulating the activation of the comple-
ment and contact system, which play a
role in opsonization and the regulation
of coagulation (8–10). Currently,
C1INH administration is applied in pa-
tients with a deficiency of the protein
causing hereditary angioedema (8).
Interestingly, several animal studies
have demonstrated that supraphysi-
ological levels of C1INH during models
of acute inflammation (sepsis and [ther-
mal] trauma) preserve endothelial func-
tion, prevent the occurrence of capil-
lary leak, and improve survival (11–14).
Furthermore, C1INH can inhibit adhe-
sion of leukocytes to the endothelium
and reduce tissue infiltration (11, 13, 15–
Objective: Besides its role in regulation of the complement and
contact system, C1-esterase inhibitor has other immunomodulating
effects that could prove beneficial in patients with acute inflamma-
tion such as during sepsis or after trauma. We examined the immu-
nomodulating properties of C1-esterase inhibitor during human ex-
perimental endotoxemia, in which the innate immune system is
activated in the absence of activation of the classic complement
pathway.
Design: Double-blind placebo-controlled study.
Setting: Research intensive care unit of the Radboud University
Nijmegen Medical Centre.
Subjects: Twenty healthy volunteers.
Interventions: Intravenous injection of 2 ng/kg Escherichia coli
lipopolysaccharide. Thirty minutes thereafter (to prevent binding
of lipopolysaccharide), C1-esterase inhibitor concentrate (100
U/kg, n ? 10) or placebo (n ? 10) was infused.
Measurements and Main Results: Pro- and anti-inflammatory
mediators, markers of endothelial and complement activation, he-
modynamics, body temperature, and symptoms were measured.
C1-esterase inhibitor reduced the release of proinflammatory cyto-
kines as well as C-reactive protein (peak levels of: interleukin-6
1521 ? 209 vs. 932 ? 174 pg/mL [p ? .04], tumor necrosis factor-?
1213 ? 187 vs. 827 ? 167 pg/mL [p ? .10], monocyte chemotactic
protein-1 6161 ? 1302 vs. 3373 ? 228 pg/mL [p ? .03], interleu-
kin-1? 34 ? 5 vs. 23 ? 2 pg/mL [p < .01], C-reactive protein 39 ?
4 vs. 29 ? 2 mg/L [p ? .02]). In contrast, release of the anti-
inflammatory cytokine interleukin-10 was increased by C1-esterase
inhibitor (peak level 73 ? 11 vs. 121 ? 18 pg/mL, p ? .02). The
increase in interleukin-1 receptor antagonist tended to be smaller in
the C1-esterase inhibitor group, but this effect did not reach statis-
tical significance (p ? .07). Markers for endothelial activation were
increased after lipopolysaccharide infusion, but no significant differ-
ences between groups were observed. The lipopolysaccharide-in-
duced changes in heart rate, blood pressure, body temperature, and
symptoms (all p < .001 over time) were not influenced by C1-
esterase inhibitor. Complement fragment C4 was not increased after
lipopolysaccharide challenge.
Conclusions: This study is the first to demonstrate that C1-
esterase inhibitor exerts anti-inflammatory effects in the absence
of classic complement activation in humans. (Crit Care Med 2010;
38:2139–2145)
KEY WORDS: C1-esterase inhibitor; sepsis; inflammation;
wounds and injuries; cytokines; lipopolysaccharide
From the Department of Intensive Care Medicine
(MJD, LAEC, MPB, JGVDH, PP), Radboud University
Nijmegen Medical Centre, The Netherlands; the De-
partments of Surgery (TV, LPHL) and Respiratory Med-
icine (JP, LK), University Medical Centre Utrecht, The
Netherlands; Sanquin Plasma Products (AHLK, PFWS),
Amsterdam, The Netherlands; and Nijmegen University
Centre for Infectious Diseases (JGVDH, PP), Radboud
University Nijmegen Medical Centre, The Netherlands.
Drs. Dorresteijn and Visser equally contributed to
this study. Dr. Leenen and Pickkers equally contributed
to this study.
Sanquin Plasma Products supplied the C1-
esterase inhibitor used to conduct the experiments.
Dr. Leo Koenderman has received a grant for a
public–private partnership between the University of
Utrecht, Dutch Government and GlaxoSmithKline,
Nycomed, Astra-Zeneca, and Danone (2008–2012).
Drs. Koenderman and Strengers both work for San-
quin, sponsors of this investigation. The remaining
authors have not disclosed any potential conflicts of
interest.
Supplemental digital content is available for this ar-
ticle. Direct URL citations appear in the printed text and
are provided in the HTML and PDF versions of this article
on the journal’s Web site (www.ccmjournal.com).
For information regarding this article, E-mail:
p.pickkers@ic.umcn.nl
Copyright © 2010 by the Society of Critical Care
Medicine and Lippincott Williams & Wilkins
DOI: 10.1097/CCM.0b013e3181f17be4
2139Crit Care Med 2010 Vol. 38, No. 11

Page 2

19). In septic patients, administration of
C1INH reduced leukocyte activation and
the release of cytotoxic mediators by de-
granulation (20). The mechanisms by
which C1INH exerts its actions are poorly
understood but appear to be (partly) in-
dependent of its effects on the comple-
ment and contact system, because its
beneficial effects in in vitro and animal
studies remain intact after cleavage of the
reactive center (12, 21).
Up to now, only a few reports exist of
the administration of C1INH to patients
undergoing major surgery or with severe
sepsis or septic shock. Although these
studies were only performed in very small
patient groups and not always in a place-
bo-controlled fashion, their results were
encouraging demonstrating a significant
reduction of renal impairment and small
case series suggest that the administra-
tion of C1INH is associated with less need
for vasopressor therapy and a reduced
hospital stay (8, 22–25).
However, to elucidate the mechanism
by which C1INH exerts its anti-inflamma-
tory effects, patient studies remain diffi-
cult to interpret as a result of heteroge-
neity of the underlying diseases. In
contrast, human experimental endotox-
emia provides an in vivo model with re-
producible systemic inflammation (26).
During human endotoxemia, the release
of several humoral mediators, activation
of leukocytes, and vascular endothelium
occur within a few hours after infusion.
Interestingly, this cascade is independent
of complement or contact system activa-
tion (27–29). We hypothesized that
C1INH can modulate the inflammatory
response during human endotoxemia in
the absence of classic complement sys-
tem activation.
MATERIALS AND METHODS
Subjects. This study was registered at
ClinicalTrials.gov as NCT00785018. After ap-
proval from the medical ethical committee, 20
healthy male subjects gave written informed
consent to participate in the experiments in
accordance with the Declaration of Helsinki.
Subjects using any drugs were excluded.
Screening of the subjects within 14 days be-
fore the test revealed no abnormalities in med-
ical history or physical examination. Labora-
tory tests (including serology on HIV and
hepatitis B) and electrocardiogram were nor-
mal. Ten hours before the experiment, sub-
jects refrained from the intake of caffeine,
alcohol, and food.
Study Design. After admission to the re-
search intensive care unit of the Radboud Uni-
versity Nijmegen Medical Centre, purified li-
popolysaccharide
Reference Endotoxin Escherichia coli O:113)
obtained from the Pharmaceutical Develop-
ment Section of the National Institutes of
Health (Bethesda, MD) was administered at a
dose of 2 ng/kg body weight at t ? 0 hr.
Thereafter, subjects were randomized by an
independent research nurse to receive C1INH
concentrate (n ? 10; Cetor; Sanquin, Amster-
dam, The Netherlands; 100 U/kg body weight,
infused in 30 mins) or an equivalent volume of
placebo (n ? 10; 0.9% saline; Baxter, Utrecht,
The Netherlands) using the sealed envelope
method. The dose of 100 U/kg C1INH was
chosen based on a study performed in humans
with acute myocardial infarction (30). C1INH
solution or placebo was prepared by the inde-
pendent research nurse and given to the in-
vestigator in identical containers ensuring the
double-blind fashion of the study. Intravenous
infusion of C1INH or placebo was started by
the investigators at 30 mins after LPS to pre-
vent binding to LPS (13, 28, 31, 32) (Fig. 1).
Hemodynamic and Clinical Response. For
continuous monitoring of blood pressure and
for blood sampling, the radial artery was can-
nulated with a 20-gauge arterial catheter.
Heart rate monitoring was performed using a
five-lead electrocardiogram. A cannula was
placed in an antecubital vein to permit infu-
sion of prehydration fluid, endotoxin, C1INH,
or placebo and the continuous infusion of 150
mL/hr 2.5% glucose/0.45% saline (detailed in-
formation is provided in the supplemental
data [see Supplemental Digital Content 1,
http://links.lww.com/CCM/A165]). Body tem-
perature (FirstTemp Genius; Tyco Healthcare,
Hampshire, UK) and symptoms were scored
every 30 mins. Subjective symptoms were
scored by the subjects using grades varying
(LPS) (USStandard
from 0 (symptoms: absent) to 5 (symptoms:
worst ever experienced) to define the severity
of nausea, headache, shivering, muscle, and/or
back pain. Thereafter, scores were added, lead-
ing to an arbitrary “total symptom score” with
a maximum value of 25 points.
Assays. Measurements of C1INH antigen
and activity, complement levels C4, various
cytokines, C-reactive protein, and soluble ad-
hesion molecules were performed before LPS
and serially thereafter. Additional details on
the methods of the performed assays are pro-
vided in the supplemental data (see Supple-
mental Digital Content 1, http://links.lww.
com/CCM/A165).
Objectives and Hypothesis. The primary
objective of the present study was to deter-
mine whether C1INH can modulate cytokine
release during human endotoxemia. Second-
ary objectives include the effects of C1INH on
C-reactive protein, hemodynamic and clinical
response, and the release of soluble adhesion
molecules after LPS challenge.
Data Analysis and Statistics. Values are
expressed as mean ? SEM unless described
otherwise. Kolmogorov-Smirnov tests indi-
cated a normal distribution of most data (a few
exceptions of less relevant time points).
Hence, a two-way repeated-measures analysis
of variance was used to test variation over
time, the variation between interventions, and
the interaction between time and intervention
(SPSS 16.0 software; SPSS, Chicago, IL).
Changes over time alone were analyzed by
one-way analysis of variance (Graphpad Prism
5; Graphpad Software, LA Jolla, CA). To com-
pare differences of baseline parameters be-
tween groups we used Student’s t test (SPSS
16.0 software; SPSS). Because the symptom
score is a discontinuous variable, we used
nonparametric Friedman analysis of variance
for changes over time followed by Mann-
Whitney U tests for differences between
groups (SPSS 16.0 software; SPSS). These
data are expressed as median and ranges. A p
value ?.05 was considered to indicate signifi-
cance. Given the explorative “proof-of-con-
cept” nature of this study, no formal sample
size calculation was performed. Furthermore,
no subgroup analyses were made.
RESULTS
Baseline Characteristics. After screen-
ing 23 healthy volunteers, 20 subjects
were enrolled in the study protocol and
randomized to receive C1INH or placebo
(Fig. 1). Besides a difference in body
weight (Student’s t test p ? .03), there
were no significant differences in baseline
characteristics between both groups. De-
mographic data are shown in Table 1.
Safety. No serious adverse events oc-
curred during the experiments. The
symptoms observed during the experi-
ments could be related to the administra-
Figure 1. Flow diagram of subjects included in
the study. After screening of 23 subjects, 20
subjects were included in the experiments. All
received 2 ng/kg Escherichia coli lipopolysac-
charide. Thereafter, subjects were randomized
to receive C1-esterase inhibitor (100 U/kg) or
placebo. ECG, electrocardiogram.
2140 Crit Care Med 2010 Vol. 38, No. 11

Page 3

tion of LPS and are discussed subse-
quently.
Clinical and Hematologic Response.
As summarized in Table 2, endotoxin in-
fusion resulted in the expected changes
in clinical and hemodynamic parameters
in both groups. All endotoxin-induced
changes were statistically significant
(over time analysis using one-way analy-
sis of variance).
Endotoxin-induced symptoms typi-
cally started with headache approxi-
mately 1 hr after LPS administration.
Symptoms, as scored by the subjects,
peaked at t ? 1.5 hrs at a median value
of 5.5 (range, 3–10; out of a maximum
of 25) for subjects receiving C1INH vs.
6.5 (range, 1–11) in subjects receiving
placebo (over time p ? .001 using
Friedman, no significant difference be-
tween groups using Mann-Whitney U
test [p ? .542]). Blood pressure
dropped by 19 ? 3% in the placebo
group compared with 18 ? 2% in the
C1INH group (no significant difference
between groups [p ? .27]). A compara-
ble compensatory rise in heart fre-
quency was observed in both groups
(p ? .001). Also, the maximum inc-
rease in temperature of 1.7 ? 0.2°C
(C1INH, p ? .001) and 1.9 ? 0.2°C
(placebo p ? .001) was similar in both
groups (Table 2).
C1INH Antigen, Activity, and Levels of
Complement Factor 4. After LPS infu-
sion, levels of C1INH antigen and activity
demonstrated a modest increase in the
placebo group (Fig. 2, over time: p ?
.01). After administration of C1INH at 30
mins after endotoxin infusion, levels of
C1INH antigen increased from 0.20 ?
0.01 g/L to 0.54 ? 0.03 g/L at t ? 1 hr
after LPS administration (over time: p ?
.001, between groups: p ? .001) and re-
mained elevated throughout the experi-
ment. Also, levels of C1INH activity in-
creased from 0.94 ? 0.03 U/mL to 2.42 ?
0.13 U/mL (over time: p ? .001, between
groups: p ? .001, Fig. 2). The levels of C4
remained low in both groups (not signif-
icantly different over time in both groups
[for the group receiving placebo p ? .24,
for the group receiving C1INH p ? .97]
or between groups [p ? .30]).
Inflammatory Markers. After the ad-
ministration of endotoxin, all measured
cytokines showed a marked increase as
illustrated in Figure 3 (over time: all p ?
.001). The administration of C1INH at-
tenuated the release of all proinflamma-
tory cytokines. Compared with subjects
receiving placebo, peak levels of IL-6
were reduced by 39% and tumor necrosis
factor-? peak levels were abrogated by
32%. Concentrations of monocyte che-
motactic protein-1 and IL-1? decreased
by 45% and 32% compared with placebo,
respectively. Conversely, the release of
the anti-inflammatory IL-10 was in-
creased in subjects receiving C1INH by
66%. The increase in IL-1 receptor antag-
onist tended to be less in the C1INH
group, but this effect did not reach sta-
tistical significance (p ? .07). C-reactive
protein was significantly reduced in the
C1INH group compared with placebo
(p ? .02, Table 2).
Markers for Endothelial Activation.
Endotoxin administration is known to
cause a release of soluble adhesion mol-
ecules, suggesting activation of endothe-
lial cells (33, 34). As demonstrated in
Figure 4, all measured markers for endo-
thelial activation were significantly in-
duced by LPS infusion except for von
Willebrand factor. Von Willebrand factor
was not significantly induced after LPS
infusion in subjects receiving placebo
(over time p ? .09). No significant differ-
ence in concentrations of circulating en-
dothelial markers was observed between
both groups.
DISCUSSION
The present study is the first to dem-
onstrate that administration of a high
dose of C1INH can modulate a controlled
inflammatory response in humans as
elicited by in vivo infusion of endotoxin.
During experimental endotoxemia, the
release of proinflammatory cytokines was
attenuated by C1INH, whereas the release
of the anti-inflammatory cytokine IL-10
was potentiated. This C1INH-mediated
shift in the pattern of the inflammatory
response occurred in the absence of acti-
vation of the classic complement system
and could not be explained by binding to
endotoxin. After administration of
C1INH, concentrations of antigen and ac-
tivity were increased throughout the en-
tire experiment. An increase of C1INH
antigen and activity of 340 mg/L and 1.5
U/mL, respectively, is expected after an
average dose of 7500 U. This higher
plasma concentration is similar to that
found during infectious diseases (8), sug-
gesting that during pathophysiological
conditions, C1INH in acute-phase protein
concentration can exert immunoregulat-
ing effects.
The observed early increase of plasma
IL-10 in the C1INH group compared with
the placebo group is most remarkable.
Although the IL-10 levels were signifi-
cantly increased by C1INH at 1.5 hrs after
LPS infusion, the levels of proinflamma-
tory cytokines tumor necrosis factor-?,
IL-1?, IL-6, and monocyte chemotactic
protein-1 did not differ between groups
until 2 hrs after LPS infusion. Our find-
ings are in agreement with data pub-
lished by Storini et al (35), who showed a
more pronounced increase of IL-10
mRNA expression after ischemia–reperfu-
sion brain injury in mice treated with
C1INH. They also found a concurrent
smaller increase of mRNA levels of tumor
necrosis factor-? and IL-6 in the C1INH-
treated group. Interestingly, in our study,
the production of the anti-inflammatory
cytokine IL-1 receptor antagonist was not
potentiated but even moderately blunted
by C1INH. This indicates that C1INH
does not induce a general anti-inflamma-
tory response. It is, therefore, tempting
to speculate that C1INH acts as an anti-
inflammatory mediator in humans by en-
hancing IL-10 production. To our knowl-
edge, the direct effects of C1INH on IL-10
production have not been studied previ-
ously. IL-10 can block the release of the
proinflammatory cytokines tumor necro-
sis factor-?, IL-1?, IL-6, and IL-8 in vitro
Table 1. Demographic characteristics
PlaceboC1INH Total Group
p Value Between
Groups
Age, yrs
Height, cm
Weight, kg
BMI, kg/m2
HR, beats/min
MAP, mm Hg
21.6 ? 2.6
187 ? 7
80 ? 10
23 ? 2
66 ? 4
103 ? 10
22.6 ? 3.6
181 ? 7
71 ? 6
22 ? 1
70 ? 12
98 ? 8
22.1 ? 3.1
184 ? 8
75 ? 9
22 ? 2
68 ? 9
101 ? 9
.49
.07
.03
.21
.35
.41
C1INH, C1-esterase inhibitor; BMI, body mass index; HR, heart rate; MAP, mean arterial blood
pressure.
Data are presented as mean ? SD.
2141Crit Care Med 2010 Vol. 38, No. 11

Page 4

(36, 37). In addition, studies in vivo have
demonstrated that IL-10 can protect mice
from lethal endotoxemia (38, 39). Apart
from increased levels of inflammatory
mediators, IL-10 knockout mice were
also characterized by an enhanced cellu-
lar inflammatory response in tissue (40).
These results imply a protective role for
IL-10 in leukocyte-driven inflammation,
but the exact mechanism by which
C1INH increases IL-10 production re-
mains to be elucidated.
Earlier studies have suggested that the
LPS-induced inflammatory reaction can
be diminished as a result of scavenging of
LPS by C1INH. This hypothesis was based
on in vitro studies demonstrating the
ability of C1INH to bind LPS and to re-
duce LPS binding to endothelial cells (13,
31). However, it seems unlikely that the
binding of LPS to C1INH explains the
anti-inflammatory effects found after ad-
ministration of C1INH concentrate in our
study. Plasma concentrations of LPS are
known to decrease rapidly to undetect-
able levels within 15 to 20 mins after
administration (28, 32). In our study,
C1INH infusion was started 30 mins after
LPS infusion, well after complete plasma
clearance of LPS. Furthermore, treat-
ment with C1INH caused enhanced levels
of LPS-induced IL-10. Scavenging of LPS
resulting from an interaction with C1INH
would have resulted in decreased levels of
all cytokines, including IL-10 (41). There-
fore, it can be concluded that C1INH has
direct immune-modulating effects irre-
spective of a scavenging effect on LPS.
The observation that C1INH has a vast
anti-inflammatory effect, even when it is
administered after the induction of in-
flammation, is in agreement with several
animal studies demonstrating beneficial
effects of C1INH administration well after
the onset of inflammation (12, 42). In a
murine cecal ligation and puncture
model, C1INH even increased survival
when administered up to 6 hrs after cecal
ligation and puncture, but not as much
as when administered directly or at 3 hrs
after cecal ligation and puncture (12).
The anti-inflammatory effects of C1INH
seem, at least in part, independent of its
function as a serpin, because in our
study, no signs of activation of the classic
pathway of the complement system were
found. We only measured serum concen-
trations of C4 and therefore cannot rule
out possible activation of the classic path-
way through complement components
bound to microparticles and a possible
effect here by C1INH (43). Although
some studies with higher doses of LPS
have demonstrated complement activa-
tion (44), no studies with a similar low
dose of LPS have shown activation of the
final common pathway of the comple-
ment system so far (28, 29). In addition,
in a seminal study performed by Liu et al
(12), it was demonstrated that serpin-
inactive C1INH was at least as effective as
active C1INH to prevent mortality in a
murine cecal ligation and puncture
model for sepsis.
Some animal studies indicate that
C1INH can inhibit margination of tissue
leukocytes (8, 14, 18). However, these
studies did not evaluate simultaneous in-
creases in leukocyte counts in the periph-
eral blood. In our study, we did not find
any effect of C1INH on peripheral leuko-
cyte counts (Table 2), which is in agree-
ment with data from an endotoxemia
model in rats in which similar results
were observed (16). However, C-reactive
protein was significantly lower at t ? 24
hrs in the C1INH group, indicating that
these subjects have apparently endured
less severe inflammation.
As demonstrated in our study, the lev-
els of soluble adhesion molecules inter-
cellular adhesion molecule, vascular cell
adhesion molecule, and P-selectin as well
as von Willebrand factor increased after
induction of systemic inflammation by
LPS. Increases in these soluble adhesion
molecules are thought to reflect activa-
tion of endothelial cells (33).
However, no differences were found
between subjects receiving placebo or
C1INH. This may point at the inability of
C1INH to antagonize activation of endo-
thelial cells after LPS challenge. Appar-
ently, C1INH had no effect on the shed-
dases, which cause the quick release of
these molecules from the endothelial sur-
face. This is in contrast to animal studies
in which an attenuation of inflammation-
Table 2. Hemodynamic parameters, symptoms, hematologic, and biochemical laboratory data during human endotoxemia in the absence and presence of
C1INH
T ? 0T ? 1T ? 2T ? 3T ? 4T ? 8T ? 24
p Value Between
Groups
MAP, mm Hg Placebo
C1INH
Placebo
C1INH
Placebo
C1INH
Placebo
C1INH
Placebo
C1INH
Placebo
C1INH
Placebo
C1INH
Placebo
C1INH
103 ? 4
98 ? 3
66 ? 2
70 ? 5
0 ?0–1?a
0 ?0–3?a
NA
NA
8.4 ? 0.1
8.4 ? 0.1
5.3 ? 0.6
4.6 ? 0.3
187 ? 10
167 ? 10
?5
?5
104 ? 2
97 ? 3
73 ? 3
74 ? 5
0 ?0–2?a
0 ?0–3?a
0.6 ? 0.1
0.5 ? 0.2
8.5 ? 0.2
8.6 ? 0.1
3.1 ? 0.5
2.2 ? 0.4
181 ? 9
159 ? 11
ND
ND
93 ? 2
86 ? 3
81 ? 3
87 ? 4
4 ?0–6?a
3 ?1–6?a
1.3 ? 0.2
1.2 ? 0.2
ND
ND
ND
ND
ND
ND
ND
ND
85 ? 4
83 ? 3
90 ? 3
91 ? 3
3 ?0–9?a
2 ?0–6?a
1.9 ? 0.2
1.7 ? 0.2
8.3 ? 0.1
8.5 ? 01
6.3 ? 0.7
7.0 ? 0.5
189 ? 13
179 ? 8
ND
ND
86 ? 3
83 ? 3
91 ? 2
93 ? 3
1 ?0–8?a
1 ?0–4?a
1.8 ? 0.3
1.3 ? 0.2
ND
ND
ND
ND
ND
ND
ND
ND
83 ? 6
83 ? 3
81 ? 4
77 ? 4
0 ?0–1?a
0 ?0–3?a
1.0 ? 0.3
0.8 ? 0.2
8.3 ? 0.1
8.4 ? 0.1
11.4 ? 0.7
11.0 ? 0.7
190 ? 10
181 ? 12
9 ? 1
7 ? 1
ND
ND
ND
ND
ND
ND
ND
ND
.27
HR, beats/min.08
Total symptoms.54b
?Temperature, °C.35
Hb, mmol/L 8.4 ? 0.2
8.5 ? 0.1
5.5 ? 0.5
5.4 ? 0.4
196 ? 12
184 ? 12
39 ? 4
29 ? 2
.97
Leukocytes, ?109/L
.54
Thrombocytes, ?109/L
.24
CRP, mg/L.02
T, time expressed in hours after lipopolysaccharide administration; MAP, mean arterial blood pressure; HR, heart rate; Hb, hemoglobin, CRP, C-reactive
protein; ND, not determined; NA, not applicable.
aTotal symptoms are expressed as median ?range?;bp value signifies difference between groups at t ? 1.5 hrs determined by Mann-Whitney U test,
median at T ? 1.5: placebo 7 ?1–11? and C1INH 6 ?3–10?. All other data are expressed as mean ? SEM. p values are comparisons between groups over time
and were determined by two-way repeated-measures analysis of variance.
2142Crit Care Med 2010 Vol. 38, No. 11

Page 5

induced mRNA synthesis of intercellular
adhesion molecule-1, vascular cell adhe-
sion molecule-1, and P-selectin by C1INH
was demonstrated (15, 39). However,
mRNA synthesis does not always reflect
the levels of the circulating adhesion
molecules (45).
In our study, no difference in clin-
ical response such as hemodynamic
changes or symptoms could be demon-
strated. Obviously, human endotox-
emia, although a very useful tool to
study pharmacologic modulation of the
innate immune response, can by no
means mimic the complete cascade of
events occurring in intensive care unit
patients with inflammatory disorders
such as septic shock or severe trauma.
Because an enhanced cytokine release
during severe inflammation is associ-
ated with development of acute respira-
tory distress syndrome and multiple or-
gan dysfunction syndrome and eventually
death (3, 46, 47), it is tempting to spec-
ulate that the ability of C1INH to shift the
inflammatory response to a more anti-
inflammatory pattern, especially early in
the disease process, could be beneficial in
patients with these acute inflammatory
syndromes.
A limitation of our study is the fact
that only healthy young male subjects
were included. Because this study
should be considered a “proof-of-concept
study,” to demonstrate the effects of C1INH
on the innate immune response, we aimed
to create a homogeneous study population.
Therefore,theextrapolationoftheeffectsof
C1INH on LPS challenge to putative clini-
cal effects should be taken with caution.
Although a few studies have been reported
applying C1INH in sepsis or trauma pa-
tients, future randomized controlled pa-
tient studies are of pivotal importance to
determine the effects of this protein in the
clinical setting.
CONCLUSIONS
In the present study, C1INH has po-
tentiated the release of the anti-
Figure 2. Levels of C1-esterase inhibitor activity
(A), antigen (B), and complement factor 4 (C4)
(C) after administration of 2 ng/kg Escherichia
coli lipopolysaccharide at t ? 0 hr. At t ? 0.5 hr,
C1-esterase inhibitor activity at a dose of 100
U/kg intravenously over 30 mins (?) or placebo
(●) was infused. Data are expressed as mean ?
SEM. p values in the figure express differences
between groups obtained by repeated-measures
analysis of variance over the complete curve. Be-
cause SEM values for C1-esterase inhibitor and
antigen in the placebo group were so small, they
all are within the reach of the mark (●) and are
therefore not clearly visible.
Figure 3. Cytokine concentration in the absence
(●) and presence (?) of C1-esterase inhibitor
after administration of 2 ng/kg Escherichia coli
lipopolysaccharide at t ? 0 hr. IL-6, interleukin-6
(A); TNF-?, tumor necrosis factor-? (B); MCP-1,
monocyte chemotactic protein-1 (C); IL-1?, in-
terleukin-1? (D); IL-1RA, interleuking-1 receptor
antagonist (E); IL-10, interleukin-10 (F). Data are
expressed as mean ? SEM. p values in the figure
express differences between groups obtained by
repeated-measures analysis of variance over the
complete curve.
Figure 4. Concentrations of soluble adhesion
molecules in the absence (●) and presence (?) of
C1-esterase inhibitor after administration of 2
ng/kg Escherichia coli lipopolysaccharide at t ?
0 hr. VCAM, vascular cell adhesion molecule (A);
ICAM, intercellular adhesion molecule (B); Von
Willebrand factor (C); Pselectin, platelet selectin
(D). Data are expressed as mean ? SEM p values in
the figure express differences between groups
obtained by repeated-measures analysis of vari-
ance over the complete curve.
2143 Crit Care Med 2010 Vol. 38, No. 11

Page 6

inflammatory cytokine IL-10 and si-
multaneously reduced the release of
proinflammatory cytokines during hu-
man experimental endotoxemia. This
shift in the pattern of the inflammatory
response occurred in the absence of ac-
tivation of the classic complement sys-
tem and could not be explained by bind-
ing of C1INH to LPS.
ACKNOWLEDGMENTS
We thank J. Bloem for her effort in the
measurements of C4 and C1INH antigen
and activity and Dr J. Over for his help in
the design of the research and his critical
appraisal of the manuscript (both San-
quin Plasma Products, Amsterdam, The
Netherlands). Furthermore, we thank
Trees Jansen for performing all cytokine
and adhesion molecule measurements.
REFERENCES
1. Barie PS, Hydo LJ: Epidemiology of multiple
organ dysfunction syndrome in critical sur-
gical illness. Surg Infect (Larchmt) 2000;
1:173–185; discussion 185–186
2. Deitch EA: Multiple organ failure: Patho-
physiology and potential future therapy. Ann
Surg 1992; 216:117–134
3. Roumen RM, Hendriks T, van der Ven-
Jongekrijg J, et al: Cytokine patterns in pa-
tients after major vascular surgery, hemor-
rhagic shock, and severe blunt trauma
Relation with subsequent adult respiratory
distress syndrome and multiple organ fail-
ure. Ann Surg 1993; 218:769–776
4. Schoenberg MH, Weiss M, Radermacher P:
Outcome of patients with sepsis and septic
shock after ICU treatment. Langenbecks
Arch Surg 1998; 383:44–48
5. Nuytinck HK, Offermans XJ, Kubat K, et al:
Whole body inflammation in trauma pa-
tients; an autopsy study. Prog Clin Biol Res
1987; 236A:55–61
6. Abraham E, Singer M: Mechanisms of sepsis-
induced organ dysfunction. Crit Care Med
2007; 35:2408–2416
7. Cohen J: The immunopathogenesis of sepsis.
Nature 2002; 420:885–891
8. Caliezi C, Wuillemin WA, Zeerleder S, et al:
C1-esterase inhibitor: An anti-inflammatory
agent and its potential use in the treatment
of diseases other than hereditary angio-
edema. Pharmacol Rev 2000; 52:91–112
9. Davis AE 3rd, Mejia P, Lu F: Biological ac-
tivities of C1 inhibitor. Mol Immunol 2008;
45:4057–4063
10. Sim RB, Arlaud GJ, Colomb MG: Kinetics of
reaction of human C1-inhibitor with the hu-
man complement system proteases C1r and
C1s. Biochim Biophys Acta 1980; 612:
433–449
11. Kochilas L, Campbell B, Scalia R, et al: Ben-
eficial effects of C1 esterase inhibitor in mu-
rinetraumaticshock.
8:165–169
12. Liu D, Lu F, Qin G, et al: C1 inhibitor-
mediated protection from sepsis. J Immunol
2007; 179:3966–3972
13. Liu D, Zhang D, Scafidi J, et al: C1 inhibitor
prevents Gram-negative bacterial lipopo-
lysaccharide-induced vascular permeability.
Blood 2005; 105:2350–2355
14. Radke A, Mottaghy K, Goldmann C, et al: C1
inhibitor prevents capillary leakage after
thermal trauma. Crit Care Med 2000; 28:
3224–3232
15. Cai S, Dole VS, Bergmeier W, et al: A direct
role for C1 inhibitor in regulation of leuko-
cyte adhesion. J Immunol 2005; 174:
6462–6466
16. Croner RS, Lehmann TG, Fallsehr C, et al: C1-
inhibitor reduces hepatic leukocyte–endothelial
interaction and the expression of VCAM-1 in
LPS-induced sepsis in the rat. Microvasc Res
2004; 67:182–191
17. Schmidt W, Stenzel K, Gebhard MM, et al:
C1-esterase inhibitor and its effects on endo-
toxin-induced leukocyte adherence and
plasma extravasation in postcapillary
venules. Surgery 1999; 125:280–287
18. Vangerow B, Hafner D, Rueckoldt H, et al:
Effects of C1 inhibitor and r-SP-C surfactant
on oxygenation and histology in rats with
lavage-induced acute lung injury. Intensive
Care Med 2001; 27:1526–1531
19. Cai S, Davis AE 3rd: Complement regulatory
protein C1 inhibitor binds to selectins and
interferes with endothelial–leukocyte adhe-
sion. J Immunol 2003; 171:4786–4791
20. Zeerleder S, Caliezi C, van Mierlo G, et al:
Administration of C1 inhibitor reduces neu-
trophil activation in patients with sepsis.
Clin Diagn Lab Immunol 2003; 10:529–535
21. Davis AE 3rd, Cai S, Liu D: C1 inhibitor:
Biologic activities that are independent of
protease inhibition. Immunobiology 2007;
212:313–323
22. Caliezi C, Zeerleder S, Redondo M, et al:
C1-inhibitor in patients with severe sepsis
and septic shock: Beneficial effect on renal
dysfunction. Crit Care Med 2002; 30:
1722–1728
23. Fattouch K, Bianco G, Speziale G, et al: Ben-
eficial effects of C1 esterase inhibitor in ST-
elevation myocardial infarction in patients
who underwent surgical reperfusion: A ran-
domised double-blind study. Eur J Cardio-
thorac Surg 2007; 32:326–332
24. Fronhoffs S, Luyken J, Steuer K, et al: The
effect of C1-esterase inhibitor in definite and
suspected streptococcal toxic shock syn-
drome: Report of seven patients. Intensive
Care Med 2000; 26:1566–1570
25. Hack CE, Voerman HJ, Eisele B, et al: C1-
esterase inhibitor substitution in sepsis. Lan-
cet 1992; 339:378
26. Bahador M, Cross AS: From therapy to ex-
perimental model: A hundred years of endo-
toxin administration to human subjects.
J Endotoxin Res 2007; 13:251–279
Shock
1997;
27. Minnema MC, Pajkrt D, Wuillemin WA, et al:
Activation of clotting factor XI without de-
tectable contact activation in experimental
human endotoxemia. Blood 1998; 92:
3294–3301
28. van Deventer SJ, Buller HR, ten Cate JW, et
al: Experimental endotoxemia in humans:
Analysis of cytokine release and coagulation,
fibrinolytic, and complement pathways.
Blood 1990; 76:2520–2526
29. Moore FD Jr, Moss NA, Revhaug A, et al: A
single dose of endotoxin activates neutro-
phils without activating complement. Sur-
gery 1987; 102:200–205
30. de Zwaan C, Kleine AH, Diris JH, et al: Con-
tinuous 48-h C1-inhibitor treatment, follow-
ing reperfusion therapy, in patients with
acute myocardial infarction. Eur Heart J
2002; 23:1670–1677
31. Liu D, Cai S, Gu X, et al: C1 inhibitor pre-
vents endotoxin shock via a direct interac-
tion with lipopolysaccharide. J Immunol
2003; 171:2594–2601
32. Jellema WT, Veerman DP, De Winter RJ, et
al: In vivo interaction of endotoxin and re-
combinant bactericidal/permeability-increas-
ing protein (rBPI23): Hemodynamic effects
in a human endotoxemia model. J Lab Clin
Med 2002; 140:228–235
33. Leeuwenberg JF, Smeets EF, Neefjes JJ, et al:
E-selectin and intercellular adhesion mole-
cule-1 are released by activated human en-
dothelial cells in vitro. Immunology 1992;
77:543–549
34. Wilson M, Blum R, Dandona P, et al: Effects
in humans of intravenously administered en-
dotoxin on soluble cell-adhesion molecule
and inflammatory markers: A model of hu-
man diseases. Clin Exp Pharmacol Physiol
2001; 28:376–380
35. Storini C, Rossi E, Marrella V, et al: C1-inhibitor
protects against brain ischemia–reperfusion in-
jury via inhibition of cell recruitment and in-
flammation. Neurobiol Dis 2005; 19:10–17
36. Leon LR, Kozak W, Kluger MJ: Role of IL-10
in inflammation: Studies using cytokine
knockout mice. Ann N Y Acad Sci 1998;
856:69–75
37. Leon LR, Kozak W, Rudolph K, et al: An
antipyretic role for interleukin-10 in LPS
fever in mice. Am J Physiol 1999; 276:
R81–R89
38. Berg DJ, Kuhn R, Rajewsky K, et al: Inter-
leukin-10 is a central regulator of the re-
sponse to LPS in murine models of endotoxic
shock and the Shwartzman reaction but not
endotoxin tolerance. J Clin Invest 1995; 96:
2339–2347
39. Howard M, Muchamuel T, Andrade S, et al:
Interleukin 10 protects mice from lethal
endotoxemia. J Exp Med 1993; 177:
1205–1208
40. Hickey MJ, Issekutz AC, Reinhardt PH, et al:
Endogenous interleukin-10 regulates hemo-
dynamic parameters, leukocyte–endothelial
cell interactions, and microvascular perme-
ability during endotoxemia. Circ Res 1998;
83:1124–1131
2144Crit Care Med 2010 Vol. 38, No. 11

Page 7

41. Suffredini AF, Hochstein HD, McMahon
FG: Dose-related inflammatory effects of
intravenous endotoxin in humans: Evalua-
tion of a new clinical lot of Escherichia coli
O:113 endotoxin. J Infect Dis 1999; 179:
1278–1282
42. Longhi L, Perego C, Ortolano F, et al: C1-
inhibitor attenuates neurobehavioral deficits
and reduces contusion volume after con-
trolled cortical impact brain injury in mice.
Crit Care Med 2009; 37:659–665
43. Biro E, Nieuwland R, Tak PP, et al: Activated
complement components and complement
activator molecules on the surface of cell-
derived microparticles in patients with rheu-
matoid arthritis and healthy individuals. Ann
Rheum Dis 2007; 66:1085–1092
44. Soop A, Albert J, Weitzberg E, et al: Comple-
ment activation, endothelin-1 and neuropep-
tide Y in relation to the cardiovascular re-
sponse to endotoxin-induced systemic
inflammation in healthy volunteers. Acta An-
aesthesiol Scand 2004; 48:74–81
45. Videm V, Albrigtsen M: Soluble ICAM-1 and
VCAM-1 as markers of endothelial activation.
Scand J Immunol 2008; 67:523–531
46. Maier B, Lefering R, Lehnert M, et al: Early
versus late onset of multiple organ failure is
associated with differing patterns of plasma
cytokine biomarker expression and outcome
after severe trauma. Shock 2007; 28:
668–674
47. Takala A, Jousela I, Takkunen O, et al: A
prospective study of inflammation markers
in patients at risk of indirect acute lung
injury. Shock 2002; 17:252–257
2145 Crit Care Med 2010 Vol. 38, No. 11