Dobutamine Improves Liver Function after
Hemorrhagic Shock through Induction of Heme
Alexander Raddatz, Darius Kubulus, Johannes Winning, Inge Bauer, Sascha Pradarutti, Beate Wolf, Sascha
Kreuer, and Hauke Rensing
Department of Anesthesiology and Critical Care Medicine, University of the Saarland, Homburg, Germany
Rationale: Induction of heme oxygenase-1 (HO-1) protects the liver Although biliverdin is a potent antioxidant, CO plays a pivotal
against reperfusion injury after hemorrhagic shock. Previous data role as a putative vasodilator for regulation of hepatic vascular
suggest that the 1-adrenoceptor agonist dobutamine induces HO-1 resistance (6, 7), most notably under stressful conditions (8, 9).
in hepatocytes. Hemorrhagic shock leads to a profound hepatocellular induction
Objectives: To investigate the functional significance of dobutamine of HO-1 (5, 10). Blockade of the HO pathway after hemorrhagic
pretreatment for liver function after hemorrhagic shock in vivo. shock increases hepatocellular injury (5), indicating a protective
Methods: Anesthetized rats received either Ringer’s (Vehicle/ role of HO-1 under these conditions.
Shock), 10 g/kg/min of the 1-adrenoceptor agonist dobutamine Further pathways of HO-1 induction under stress conditions
(Dob/Shock), or 10 g/kg/min dobutamine and 500 g/kg/min of are described. In addition to oxidative stress (5, 11), a protein
the 1-adrenoceptor antagonist esmolol (Dob/Esmolol/Shock) for kinase A–dependent induction of HO-1 was observed (12). He-
6 h. Hemorrhagic shock was induced thereafter (mean arterial pres- patocytes exhibit both - and -adrenoceptors (13). Immunohis-
sure, 35 mm Hg for 90 min). Animals were resuscitated with shed tochemistry reveals a zonal expression of 1-adrenoceptors in
blood and Ringer’s. In addition, the HO pathway was blocked after
the pericentral region. Stimulation of 1-adrenoceptors leads to
dobutamine pretreatment with 10 mol/kg tin-mesoporphyrin-IX
a time- and dose-dependent induction of HO-1 primarily in
(Dob/SnMP/Shock) or animals received 100 mg/kg of the carbon
pericentral hepatocytes (14).
monoxide donor dichloromethane (DCM/Shock).
The functional role of HO-1 induction after dobutamine treat-
Measurements: Hepatocellular metabolism and liver blood flow were
ment has not yet been clariﬁed. The aim of this study is to assess
measured by plasma disappearance rate of indocyanine green
the functional signiﬁcance of dobutamine pretreatment for hepatic
(PDRICG) as a sensitive marker of liver function.
Main Results: Pretreatment with dobutamine induced HO-1 in peri-
function and perfusion after hemorrhagic shock in vivo.
central hepatocytes and improved PDRICG (Vehicle/Shock: 11.7
8.12%/min vs. Dob/Shock: 19.7 2.46%/min, p 0.006). Blockade METHODS
of the HO pathway after preconditioning and the combined pre-
treatment with dobutamine and esmolol decreased PDRICG (Dob/
SnMP/Shock: 12.6 4.24%/min, p 0.011; Dob/Esmolol/Shock: All experiments were performed in accordance with the German legisla-
10.2 4.34%/min, p 0.008). Pretreatment with a carbon monox- tion on protection of animals and the National Institutes of Health
guidelines for animal care. Male Sprague-Dawley rats (200–250 g body
ide donor improved PDRICG (DCM/Shock: 18 3.19%/min, p
weight) were obtained from Charles River (Sulzfeld, Germany).
0.022) compared with Vehicle/Shock.
Conclusions: These results suggest a 1-adrenoceptor–dependent Surgical Procedures
hepatic up-regulation of HO-1 and a better maintained hepatocellu-
Animals were anesthetized and prepared as described previously (5).
lar function after hemorrhagic shock in animals pretreated with
In addition, the right femoral vein was cannulated for continuous admin-
dobutamine. The improved hepatocellular function may be in part
istration of indocyanine green (ICG). Cardiac output was measured
mediated by carbon monoxide because of up-regulation of HO-1. by transpulmonary thermodilution technique (Cardiotherm 500; Co-
Pretreatment with dobutamine might be a potential means of phar- lumbus Instruments, Columbus, OH).
macologic preconditioning before ischemia-reperfusion of the liver. Hemorrhagic shock was induced by rapid arterial blood withdrawal
by way of the left femoral artery (mean arterial pressure [MAP]: 35
Keywords: dobutamine; heme oxygenase-1; hemorrhagic shock; liver; 5 mm Hg for 90 min). Animals were resuscitated with 60% of the shed
preconditioning blood withdrawn infused during the ﬁrst 10 min of resuscitation and
twice the shed blood volume as Ringer’s solution during the ﬁrst hour
Heme oxygenase (HO)-1 catalyzes the rate-limiting step in the of resuscitation. The infusion rate of Ringer’s solution was lowered to
degradation of heme to biliverdin, iron, and carbon monoxide a volume equaling the maximal bleed-out volume for the second hour
(CO). Two different isoforms of HO have been characterized. of resuscitation. At the end of the experiment, livers were harvested
Whereas the isoform HO-2 is constitutively expressed (1–3), the and stored at 70 C until further processing.
isoform HO-1 is highly inducible by a variety of stress conditions
To study the dose-dependent induction of HO-1, animals (n 6/group)
were treated with 10, 20, or 50 g/kg/min dobutamine for 6 h. At the
end of the experiment, livers were harvested and stored at 70 C until
further processing. In further control experiments, animals were treated
(Received in original form August 8, 2005; accepted in final form April 13, 2006 ) with 10 g/kg/min dobutamine for 6 h to assess the organ-speciﬁc
Correspondence and requests for reprints should be addressed to Hauke Rensing, expression of HO-1. Liver, heart, aorta, lung, kidney, and jejunum were
M.D., Klinik fur Anaesthesiologie und Intensivmedizin, Universitat des Saarlandes,
¨ ¨ harvested at the end of the experiment and stored at 70 C until further
D-66421 Homburg/Saar, Germany. E-mail: firstname.lastname@example.org processing.
Am J Respir Crit Care Med Vol 174. pp 198–207, 2006
Sham-operated animals (n 8) received a constant infusion of
Originally Published in Press as DOI: 10.1164/rccm.200508-1221OC on April 20, 2006 10 ml/kg/h Ringer’s solution during the entire period of the experiment
Internet address: www.atsjournals.org but did not undergo hemorrhage.
Raddatz, Kubulus, Winning, et al.: Pharmacologic Preconditioning with Dobutamine 199
Animals received either 10 ml/kg/h Ringer’s (Vehicle/Shock; n 9) or
10 g/kg/min dobutamine (Dob/Shock; n 9) or a combination of
10 g/kg/min dobutamine and 500 g/kg/min of the 1-adrenoceptor
antagonist esmolol (Dob/Esmolol/Shock; n 5) for 6 h before induction
of hemorrhagic shock. Drug infusion was stopped 30 min before the
onset of hemorrhagic shock.
In control experiments, the HO pathway was blocked after dobutam-
ine pretreatment with the false substrate tin mesoporphyrin-IX (SnMP-
IX; Porphyrin Products, Logan, UT) 15 min before induction of hemor-
rhagic shock (Dob/SnMP/Shock; n 9). SnMP-IX was prepared and
administered as described previously (15).
To assess the role of CO, animals received 100 mg/kg of the CO
donor dichloromethane (DCM) 6 h before induction of shock (DCM/
Shock; n 9). DCM was administered by a stomach tube (16). CO
blood content was measured spectrophotometrically at baseline and
before induction of hemorrhagic shock.
Assessment of ICG Plasma Disappearance Rate
After 1 h of reperfusion, ICG (Pulsion, Munich, Germany) was continu-
ously infused (5 mg/h) by the right femoral vein for 1 h to achieve a
steady state. All syringes and tubes were covered with tinfoil to avoid
phototoxicity. Animals were heparinized with 300 IE/kg body weight
ﬁfteen min before taking blood samples. Blood samples (0.3 ml) were
taken at 0, 2, 4, 6, 8, 10, 15, and 20 min after switching off the perfusor
and immediately centrifuged at 10,000 g for 5 min. ICG absorbances
were determined spectrophotometrically at a wavelength of 800 nm.
Measured ICG absorbances were converted into the corresponding
plasma concentrations using a dose–response relationship. ICG plasma
disappearance rate (PDRICG) was deﬁned as the percentage decrease Figure 1. Dose-dependent induction of heme oxygenase (HO)-1. West-
in ICG-plasma concentration per minute (%/min). ern blot analysis was performed as described. Aliquots of liver protein
were fractionated by gel electrophoresis and electroblotted to polyvinyl-
Quantitative Determination of Serum Enzyme Levels idene membranes. Membranes were incubated with an HO-1 primary
Alanine aminotransferase (ALAT), glutamate dehydrogenase (GLDH), antibody and the antigen antibody conjugate detection was achieved
troponin T, lipase, and creatinine were analyzed with commercially by an enhanced chemiluminescent reaction. Signal detection quantifica-
available kits (Roche Diagnostics, Mannheim, Germany). tion was reached by a short exposure to film and subsequent densito-
metric analysis. Two protein samples of control animals, sham-operated
Western Blot Analysis and Immunohistochemistry animals and animals treated with different dosages of dobutamine for
Western blot analysis and immunohistochemistry were performed as 6 h, are shown in a representative Western blot in A. The densitometric
described previously (1). data are shown in B. HO-1 immunoreactive protein was below the
detection limit in control animals. Only a slight HO-1 expression was
Statistical Analysis observed in sham-operated animals. A dose-dependent increase in de
Data are presented as means SD. Differences were evaluated using novo HO-1 protein synthesis was observed after dobutamine treatment.
analysis of variance followed by Student-Newman-Keuls test; p 0.05 Arbitrary densitometric units are given as mean SD. *p 0.05 com-
was considered signiﬁcant. pared with sham.
Dose-dependent Hepatocellular Induction of HO-1 after cells, primarily in periportal ﬁelds. HO-1 immunoreactive pro-
Dobutamine Treatment tein was barely detectable in hepatocytes of sham-operated con-
In dose–response experiments, animals were treated with 10, 20, trol animals (Figures 2A and 2B). Treatment with dobutamine
or 50 g/kg/min dobutamine for 6 h or received 10 ml/kg/h led to a dose-dependent and substantial de novo expression
Ringer’s solution (sham). Untreated animals served as controls. of HO-1 immunoreactive protein in parenchymal cells in the
HO-1 protein was barely detectable in livers from unmanipu- midzonal and pericentral region and tended to increase HO-1 im-
lated controls. A slight induction of HO-1 was observed in sham- munoreactive protein in nonparenchymal cells (Figures 2C–2F).
operated animals, whereas a signiﬁcant and dose-dependent
increase of HO-1 immunoreactive protein after treatment with Organ-specific Induction of HO-1 after Dobutamine
dobutamine was observed in vivo (Figure 1). Treatment
In control experiments, animals were treated with 10 g/kg/
Immunohistochemical Detection of HO-1 Expression min dobutamine for 6 h. Compared with sham-operated control
Cell-type–speciﬁc and spatial expression pattern of HO-1 on the animals, a profound de novo synthesis of HO-1 immunoreactive
protein level was studied by immunohistochemistry. Representa- protein was observed in liver, heart, aorta, lung, kidney, and
tive liver sections were obtained from time-matched sham- jejunum after dobutamine treatment (Figure 3).
operated controls, animals treated with 10 g/kg/min dobutam-
ine, or animals treated with 50 g/kg/min dobutamine for 6 h. Macrohemodynamic Parameters
Regarding the cell-type–speciﬁc and spatial expression pattern Baseline values of MAP, heart rate, and cardiac output were
of HO-1, sham-operated control animals expressed only small comparable in all groups. Sham-operated animals exhibited
amounts of HO-1 immunoreactive protein in nonparenchymal normal and stable hemodynamic conditions throughout the
200 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 174 2006
Figure 2. Cell type and spatial expression
pattern of HO-1 in vehicle- and dobutam-
ine-treated animals. Immunohistochemi-
cal detection of HO-1 immunoreactive
protein was performed in paraffin-
embedded dewaxed liver sections using
a polyclonal HO-1 primary antibody. Rep-
resentative liver sections were obtained
from time-matched sham-operated con-
trol animals (A, B ), animals treated with
10 g/kg/min dobutamine (C, D ), or ani-
mals treated with 50 g/kg/min dobu-
tamine (E, F ) for 6 h. Regarding the cell-
type–specific and spatial expression
pattern of HO-1, sham-operated control
animals expressed only small amounts of
HO-1 immunoreactive protein in nonpa-
renchymal cells, primarily in periportal
fields. HO-1 immunoreactive protein was
barely detectable in hepatocytes of sham-
operated control animals (A, B ). Treat-
ment with dobutamine led to a dose-
dependent and substantial de novo
expression of HO-1 immunoreactive pro-
tein in parenchymal cells in the midzonal
and pericentral region and tended to in-
crease HO-1 immunoreactive protein in
nonparenchymal cells (C–F). (A, C, E ):
original magnification, 10; (B, D, F ):
original magnification, 40. Asterisks
mark central veins. Arrowheads mark peri-
portal fields. White arrows mark hepato-
cytes. Black arrows mark nonparenchymal
experiment. Treatment with 10 g/kg/min dobutamine led to a out blockade (data not shown). Administration of DCM (DCM/
signiﬁcant increase in heart rate (Figure 4A), whereas no signiﬁ- Shock) did not lead to signiﬁcant changes in MAP, heart rate, and
cant differences in MAP were observed (Figure 4B). Cardiac cardiac output compared with the Vehicle/Shock group (data not
output was increased during preconditioning with dobutamine shown).
(Figure 4C). A decrease in heart rate and cardiac output toward
baseline was observed within 15 min after termination of dobu- Blood Gas Analysis
tamine pretreatment. Dob/Esmolol/Shock animals showed com- Blood gas analysis was performed immediately after blood sam-
parable hemodynamic parameters with the Vehicle/Shock group. pling. Baseline values of respiratory parameters (Po2, Pco2), acid-
Before onset of hemorrhagic shock, all groups exhibited compa- base metabolism (pH, base excess), and hemoglobin content
rable macrohemodynamic conditions. Induction of hemorrhagic were comparable in all groups. A signiﬁcant reduction of base
shock led to an initial profound decrease in heart rate in Vehicle/
excess was observed in all shock groups indicating a comparable
Shock animals, whereas it decreased only slightly in the Dob/
severity of hemorrhagic shock. Base excess recovered during
Shock group (Figure 4A). Hemorrhagic shock was reversible in
resuscitation in all shock groups (Table 2).
all animals as reﬂected by recovery of MAP, heart rate, and
cardiac output after retransfusion of shed blood. Shed blood PDRICG
volumes (milliliter per kilogram) were comparable in all groups
subjected to hemorrhagic shock (Table 1), suggesting a compara- Hemorrhagic shock and resuscitation signiﬁcantly impaired
ble insult. During reperfusion, macrohemodynamic parameters PDRICG compared with sham-operated animals. Pretreatment
were stable in all groups (Figure 4). with dobutamine before induction of hemorrhagic shock im-
Blockade of the HO pathway with SnMP in dobutamine- proved PDRICG. A combined pretreatment with dobutamine and
treated animals (Dob/SnMP/Shock) had no signiﬁcant inﬂuence esmolol impaired PDRICG after hemorrhagic shock compared
on macrohemodynamic parameters like MAP, heart rate, and with dobutamine alone. Blockade of the HO pathway with
cardiac output compared with dobutamine-treated animals with- SnMP-IX after dobutamine treatment abolished the observed
Raddatz, Kubulus, Winning, et al.: Pharmacologic Preconditioning with Dobutamine 201
protection. Application of DCM before shock led to a signiﬁcant
increase in CO hemoglobin (from 3.5 0.56% to 15.1 3.51%,
p 0.001) and improved PDRICG (Figure 5).
1 -Adrenoceptor–dependent Induction of HO-1
Samples of sham-operated, vehicle-treated, and dobutamine-
treated (10 g/kg/min) animals and animals treated simultaneously
with dobutamine (10 g/kg/min) and esmolol (500 g/kg/min) were
analyzed. Dobutamine treatment for 6 h led to a signiﬁcant in-
crease of HO-1 expression compared with sham-operated ani-
mals. Blockade of 1-adrenoceptors with the 1-adrenoceptor
antagonist esmolol prevented this increase (Figure 6).
Influence of DCM Pretreatment on HO-1 Induction
In further control experiments, the inﬂuence of DCM pretreat-
ment on HO-1 induction in the liver was investigated to assess
possible effects of the reagent on hemoglobin inhibiting oxygen
delivery and enhancing hypoxic preconditioning. No signiﬁcant
induction of HO-1 was observed after DCM treatment, making
a hypoxic preconditioning through DCM pretreatment unlikely
Determination of Serum Enzyme Levels as Markers of Injury
Hepatocellular injury was assessed by measurement of serum
enzyme levels of ALAT and GLDH.
Troponin T was measured as a marker of cardiac injury.
Lipase serum enzyme levels were assessed as a marker of pancre-
atic injury. Creatinine levels were used as a marker of renal
injury. Hemorrhage and subsequent resuscitation led to a sig-
niﬁcant increase in ALAT and GLDH activity in Vehicle/Shock
and Dob/Shock animals. No signiﬁcant differences were ob-
served between both groups (Figure 8).
Pretreatment with dobutamine decreased serum enzyme lev-
els of troponin T compared with vehicle, indicating less cardiac
injury after dobutamine pretreatment (Figure 9). No signiﬁcant
changes were observed for creatinine or lipase serum enzyme
The present study investigated the functional signiﬁcance of
pretreatment with the 1-adrenoceptor agonist dobutamine be-
fore hemorrhage and resuscitation on liver function and perfu-
sion. Preconditioning with dobutamine led to a hepatocellular
HO-1 induction and signiﬁcantly increased PDRICG after hemor-
rhagic shock. This protective effect was abrogated by blocking
the HO pathway with SnMP-IX. Blockade of 1-adrenoceptors
with esmolol during dobutamine preconditioning attenuated in-
duction of HO-1 and PDRICG. Administration of a CO donor
before induction of hemorrhagic shock improved PDRICG. These
results suggest a 1-adrenoceptor–dependent up-regulation of
HO-1 in the liver through pretreatment with dobutamine. This
up-regulation of HO-1 after preconditioning contributes to an
improved hepatic function and perfusion after hemorrhagic
shock, which might be in part mediated through the increased
Figure 3. Organ-specific induction of HO-1 after dobutamine pretreat- HO-dependent production of CO. Furthermore, induction of
ment. Western blot analysis was performed as described previously. For
HO-1 is observed in several other organs, such as heart, aorta,
each tissue, one representative sample after 6 h dobutamine pretreat-
lung, jejunum, and kidneys, after 6 h of dobutamine pretreat-
ment and one sample after 6 h vehicle treatment is shown in this
Western blot. Spleen tissue served as positive control for HO-1 protein.
ment. Looking at markers of injury, a decrease of troponin
The mean of the densitometric data of four animals are shown in the serum enzyme levels after dobutamine pretreatment indicates
bar graphs. Dobutamine treatment led to a significant HO-1 induction a protective role for cardiac injury after shock.
in all investigated organs. Densitometric units are given as mean SD; The phenomenon that pre-exposure of the liver to transient
*p 0.05 compared with control. sublethal stress increases the tolerance to reperfusion injury is
known as “hepatic preconditioning.” Organ preconditioning is
an emerging means of protecting the body against ischemia-
induced injury as a consequence of surgical procedures. Several
202 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 174 2006
Figure 4. Hemodynamic effects of dobutamine preconditioning. Heart
rate (A ), mean arterial pressure (MAP) (B ), and relative changes in
cardiac output (C ). Dobutamine pretreatment led to a significant in-
crease in heart rate and cardiac output, whereas no significant changes
in MAP were observed. Vehicle-treated animals and Dob/Esmolol/Shock
animals exhibited stable hemodynamic parameters during precondi-
tioning. After termination of dobutamine administration, heart rate and
cardiac output in the Dob/Shock group returned to baseline. At the
onset of hemorrhagic shock, all groups had comparable hemodynamics.
Induction of hemorrhagic shock led to a significant decrease in MAP
and cardiac output in all groups. A significant decrease in heart rate
was observed in Vehicle/Shock animals, whereas it remained stable in
dobutamine-pretreated animals. After retransfusion of shed blood heart
rate, MAP and cardiac output returned to baseline values in all groups.
No significant differences in MAP between the groups were observed
during the entire experiment. Cardiac output was calculated as relative
changes of baseline values. Data are shown as mean SD of nine
(Vehicle/Shock, Dob/Shock) or five (Dob/Esmolol/Shock) independent
experiments. #p 0.05 compared with Vehicle/Shock; $p 0.05 com-
pared with baseline.
Substantial disadvantages of the described ways of precondi-
tioning include serious side effects (protoporphyrins) or technically
complex protocols (adenoviral HO-1 gene transfer). Pretreat-
ment with dobutamine offers the opportunity of a receptor-
dependent pharmacologic preconditioning before ischemia-
reperfusion. Dobutamine is a preferential 1-agonist commonly
used for cardiac support in the case of myocardial dysfunction.
Recent evidence suggests that treatment of brain-dead donors
with catecholamines reduces the rejection risk and improves
long-term transplantation outcome (27, 28). The mechanisms
are not fully understood but may include up-regulation of HO-1.
In several transplant models, preinduction of HO-1 by metallo-
porphyrins before cold ischemia has been shown to limit injury
in the post-transplant period (29). In the present study, pretreat-
ment with dobutamine was well tolerated and did not increase
parameters of liver injury. The observed hemodynamic changes
include tachycardia and an increase in cardiac output. These
changes were reversible after stopping the dobutamine infusion.
Dobutamine pretreatment dose-dependently induced HO-1 in
pericentral hepatocytes. The observed zonal expression pattern
of HO-1 correlates with the previously described zonal expres-
sion pattern of 1-adrenoceptors in the liver (30). To avoid the
possibility that dobutamine might directly inﬂuence hepatocellu-
lar function and sinusoidal perfusion by raising cardiac output
and liver blood ﬂow, administration of the drug was terminated
30 min before the onset of hemorrhagic shock. Considering the
short half-life of dobutamine (31), a signiﬁcant inﬂuence on
TABLE 1. SHED BLOOD VOLUME
Shed Blood Volume (ml/kg)
Vehicle/Shock 33.5 1.53
Dob/Shock 35.3 3.24
Dob/SnMP/Shock 35.7 6.40
preconditioning protocols are reported, including brief ischemia DCM/Shock 40.9 7.47
followed by reperfusion (17, 18), whole-body hyperthermia (19– Dob/Esmolol/Shock 41.7 6.75
21), and chemical induction of heat shock proteins with geranylg-
Definition of abbreviations: DCM dichloromethane; Dob dobutamine;
eranylacetone (22–24) and cobalt-protoporphyrin (25). Hepatic SnMP substrate tin mesoporphyrin.
preinduction of HO-1 with cobalt-protoporphyrin or with adeno- No significant differences in shed blood volumes were observed to achieve a
viral HO-1 gene transfer before ischemia-reperfusion improved mean arterial blood pressure of 35 5 mm Hg in the shock groups. Data are
hepatocellular function in vivo (26). means SD. Power of the performed test, 0.8.
Raddatz, Kubulus, Winning, et al.: Pharmacologic Preconditioning with Dobutamine 203
TABLE 2. BLOOD GAS ANALYSIS
End of End of
Baseline Preconditioning End of Shock Experiment
Sham 7.36 0.019 — — 7.42 0.049
Vehicle/Shock 7.41 0.024 7.48 0.030 7.36 0.127 7.49 0.049
Dob/Shock 7.42 0.045 7.49 0.035 7.45 0.067 7.52 0.037
Dob/SnMP/Shock 7.38 0.040 7.49 0.034 7.47 0.088 7.50 0.042
DCM/Shock 7.41 0.016 7.48 0.031 7.49 0.063 7.48 0.032
Dob/Esm/Shock 7.46 0.041 7.52 0.017 7.42 0.126 7.50 0.049
Sham 77.0 11.91 — — 85.0 6.47
Vehicle/Shock 83.0 7.73 86.0 11.05 109.6 14.04 82.5 12.17
Dob/Shock 82.4 8.73 85.4 11.11 107.1 15.22 76.6 7.86
Dob/SnMP/Shock 78.7 8.75 77.8 10.35 107.1 14.10 79.8 11.14
DCM/Shock 89.6 5.80 81.3 11.69 105.8 12.70 87.4 5.54
Dob/Esm/Shock 75.6 8.37 80.1 4.54 103.4 17.64 79.9 11.99
Sham 32.2 14.50 — — 33.7 5.22
Vehicle/Shock 38.1 3.09 28.8 3.30 17.4 8.05 27.2 3.19
Dob/Shock 35.3 4.20 30.4 10.43 14.9 4.92 25.4 3.62
Dob/SnMP/Shock 42.5 4.57 26.0 2.15 19.1 6.80 27.1 5.12
DCM/Shock 39.4 3.02 31.7 6.83 19.7 6.45 33.1 1.81
Dob/Esm/Shock 32.2 5.98 31.0 6.66 18.3 6.09 25.3 4.04
Sham 12.7 2.14 — — 12.8 0.89
Vehicle/Shock 11.4 1.86 10.2 1.47 6.9 1.73 8.7 1.66
Dob/Shock 11.4 1.54 10.7 1.61 7.2 1.38 8.7 1.58
Dob/SnMP/Shock 13.3 1.06 11.9 0.72 8.7 1.08 9.1 2.12
DCM/Shock 13.3 0.91 11.9 1.34 8.3 1.30 9.2 0.87
Dob/Esm /Shock 13.0 1.24 10.7 2.00 6.5 2.10 6.8 0.93
Sham 2.1 1.14 — — 0.5 1.24
Shock 0.5 1.64 0.4 1.59 13.8 2.75 # 1.3 3.84
Dob/Shock 0.5 1.72 1.4 2.46 12.1 3.50 # 0.8 2.44
Dob/SnMP/Shock 0.1 2.22 2.0 1.39 9.3 6.61 # 0.6 4.11
DCM/Shock 0.9 1.38 1.3 3.23 8.5 5.84 # 2.7 2.71
Dob/Esm/Shock 0.3 3.48 1.7 1.60 11.1 4.02 # 3.0 3.87
Definition of abbreviations: BE base excess; DCM dichloromethane; Dob dobutamine; Esm esmolol; Hb hemoglobin;
SnMP substrate tin mesoporphyrin.
Blood gas analysis revealed normal baseline values of PO2 (mm Hg), PCO2 (mm Hg), pH, BE, and Hb concentrations (g/dl). Hb
concentrations and BE values were significantly lower in animals subjected to hemorrhagic shock compared with sham-operated
animals. At the end of the experiment, PCO2 values of all groups were below the normal range, indicating a mild hyperventilation
at this time point. Data are means SD; #p 0.05 compared to baseline.
systemic hemodynamics is unlikely. Furthermore, cardiac output The organ-speciﬁc expression of HO-1 after dobutamine pre-
at the beginning and at the end of shock was comparable between treatment showed an induction in the liver, heart, aorta, lung,
dobutamine-pretreated animals and those without pretreatment, kidney, and jejunum compared with sham-operated control ani-
indicating a comparable macrohemodynamic situation. A direct mals. Serum enzyme levels after hemorrhagic shock with or
adrenergic effect of dobutamine on hepatic function and perfu- without dobutamine pretreatment, including the serum enzyme
sion reﬂected by PDRICG is unlikely. levels of troponin T as a marker of cardiac injury, the enzyme
The question whether a post-shock treatment with dobutam- levels of lipase as a marker of pancreatic injury, and the serum
ine may lead to protection, as observed with the pretreatment, enzyme levels of creatinine as a marker of kidney function,
cannot be answered with the current data. Hemorrhagic shock showed unchanged creatinine and lipase levels, whereas troponin
T was lower after dobutamine pretreatment. These data indicate
and resuscitation lead to a profound induction of HO-1. Induc-
a protective role of dobutamine pretreatment for cardiac injury
tion of HO-1 under these conditions contributes to maintenance
after shock apart from the protective effects observed on liver
of liver blood ﬂow and hepatocellular integrity. The time course
of HO-1 induction reaches a maximum 6 h after shock (32). Different pathways of HO-1 induction under stress conditions
Treatment with dobutamine after shock might further increase have been described. Generation of reactive oxygen species is
the shock-dependent induction of HO-1 and might possibly in- one possible mechanism to induce HO-1 (32). Immenschuh and
crease the protection already observed. An essential experimen- coworkers (12) described a protein kinase A–dependent induc-
tal problem to assess the protective effect of post-treatment is the tion of HO-1 in vitro. In addition, other receptor-dependent
increase in cardiac output through dobutamine. To differentiate pathways affect intracellular cAMP levels. In a previous study,
between the effects of dobutamine on HO-1 induction and the the authors found a cAMP-dependent hepatocellular HO-1 in-
positive effects of dobutamine on hemodynamics might be quite duction after application of the selective 1-adrenoceptor ago-
difﬁcult under these conditions. nists dobutamine and xamoterol in vitro and in vivo (14). These
204 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 174 2006
Figure 5. Plasma disappearance rate of indocyanine green (PDRICG) after
2 h of resuscitation from hemorrhagic shock. ICG absorbances were
determined spectrophotometrically at a wavelength of 800 nm in
plasma samples. Measured ICG absorbances were converted into the
corresponding plasma concentrations using a dose–response relation-
ship. PDRICG was defined as the percentage decrease in ICG plasma
concentration per minute (%/min). Hemorrhagic shock (Vehicle/Shock;
n 9) led to a significant impairment of PDRICG compared with sham-
operated animals. Pretreatment with the 1-adrenoceptor agonist dobu-
tamine (Dob/Shock; n 9) before hemorrhagic shock significantly
improved PDRICG after reperfusion. Blockade of 1-adrenoceptors with
esmolol (Dob/Esmolol/Shock; n 5) during dobutamine-pretreatment
or blockade of the HO-pathway with SnMP-IX (Dob/SnMP/Shock; n 9)
prevented the observed protection through dobutamine. Pretreatment
with the carbon monoxide donor dichloromethane (DCM; DCM/Shock;
n 9) significantly increased PDRICG compared with Vehicle/Shock ani-
mals. Data are shown as mean SD. *p 0.05 compared with sham; Figure 6. 1-Adrenoceptor–dependent induction of HO-1. (A ) A repre-
p 0.05 compared with Vehicle/Shock; §p 0.05 compared with sentative Western blot. Liver protein samples of sham-, vehicle-,
Dob/Shock. dobutamine- (10 g/kg/min), or dobutamine- (10 g/kg/min) and
esmolol- (500 g/kg/min) treated animals were used. (B ) The densito-
metric data. Dobutamine treatment led to a significant increase in HO-1
expression compared with sham-operated animals. This increase was
prevented by simultaneous blockade of 1-adrenoceptors during dobu-
data are in line with the observed induction of HO-1 after dobu-
tamine-treatment. Data are presented as mean SD. *p 0.05
tamine pretreatment in the present study. Dobutamine pretreat-
compared with sham.
ment before onset of hemorrhagic shock increased PDRICG after
hemorrhage, indicating a protective effect.
To assess the role of 1-adrenoceptors for HO-1 induction, 1-
adrenoceptors were blocked with the antagonist esmolol during mechanism and seems to be protective regarding liver function
dobutamine preconditioning. Esmolol attenuated induction of after hemorrhagic shock.
HO-1 through dobutamine and decreased PDRICG after hemor- Microcirculatory disturbances as consequences of ischemia-
rhagic shock. Esmolol is a 1-selective adrenergic receptor reperfusion have a serious impact on reperfusion injury (33,
blocking agent with a very short duration of action, and no 34). Although the underlying mechanisms are not completely
signiﬁcant intrinsic sympathomimetic activity at therapeutic dos- understood, recent evidence suggests a dysregulation of vaso-
ages. It is rapidly metabolized by hydrolysis of the ester linkage active mediators during resuscitation. A complex interaction of
by the esterases in the cytosol of red blood cells; the metabolism endothelins, catecholamines, and gaseous oxides regulates liver
of esmolol is not limited by the rate of blood ﬂow to metabolizing blood ﬂow under these conditions. The imbalance between vaso-
tissues. At the onset of hemorrhagic shock, 30 min after stopping dilators and vasoconstrictors has been identiﬁed as a potential
the infusion of esmolol and dobutamine, a signiﬁcant activity target for therapeutic interventions (35). HO-1 is induced after
of these agents at 1-adrenoceptors is unlikely. The observed hemorrhage and resuscitation (5, 10) and plays an essential role
attenuated HO-1 induction after preconditioning with dobutam- in protecting the liver against ischemia-reperfusion injury (5, 8).
ine and esmolol and the decreased PDRICG after hemorrhagic Mechanisms contributing to the observed protection through
shock under these conditions are likely to be dependent on an HO-1 may include the vasodilatory potential of CO (9, 10).
attenuated 1-adrenoceptor activation in the presence of esmolol Under stress conditions, HO-1 becomes the major site of CO
compared with dobutamine alone. Pretreatment with dobutam- generation (36). Consistent with this concept, the use of SnMP-
ine is likely to induce HO-1 by a 1-adrenoceptor–dependent IX as a speciﬁc blocker of the HO-CO pathway impaired portal
Raddatz, Kubulus, Winning, et al.: Pharmacologic Preconditioning with Dobutamine 205
Figure 7. Influence of DCM on hepatocellular HO-1 expression. (A ) A
representative Western blot. Two representative liver protein samples
of sham-, vehicle-, and DCM-treated animals were used. (B ) The densito-
metric data. No significant differences in HO-1 induction were observed
between sham-, vehicle-, and DCM-treated animals after 6 h.
(10) and sinusoidal blood ﬂow (8) and led to an increased hepato-
cellular injury (5).
In the present study, the authors observed an induction of
HO-1 after pretreatment with a 1-adrenoceptor agonist accom-
panied by an improved liver function and perfusion after shock.
Blockade of the HO pathway under these conditions abolished
the observed protection after pretreatment. In control experi-
ments animals were fed with the CO donor DCM to estimate
the role of CO as a mediator of protection. DCM increased
signiﬁcantly CO hemoglobin levels and improved PDRICG com-
pared with vehicle. Up-regulation of HO-1 after stimulation of Figure 8. Hepatocellular injury was assessed by plasma concentrations
1-adrenoceptors contributes to an improved hepatic perfusion of alanine aminotransferase (ALAT; A ) and glutamate dehydrogenase
and function after hemorrhagic shock, which might be mediated (GLDH; B ). Pretreatment with dobutamine did not increase plasma
in part through the increased HO-dependent production of CO. levels of ALAT or GLDH. Hemorrhagic shock and reperfusion led to a
A decrease in ALAT and GLDH serum levels after precondi- significant increase in GLDH and ALAT activity. No significant differences
tioning could not be observed but might be explained by an between vehicle- and dobutamine-treated animals were observed. Data
improved sinusoidal blood ﬂow and subsequent ﬂush out of liver are shown as mean SD. $p 0.05 compared with baseline.
enzymes into the systemic circulation in dobutamine-treated ani-
mals. In low-ﬂow situations like hemorrhagic shock, PDRICG
reﬂects hepatocellular function and liver perfusion more reliably sal perfusion after dobutamine administration in patients with
than serum liver enzyme levels (37, 38). Systemic hemodynamics septic shock is reported in several studies (43–47). This improve-
provides a poor estimate of hepatic perfusion. Measurement of ment might be the result of a receptor-mediated redistribution
PDRICG delivers information about hepatic perfusion and metab- of blood ﬂow from the muscularis to the mucosa (48). Data
olism (39, 40). ICG is a nontoxic dye, which is eliminated exclu- suggest that dobutamine by itself has no relevant effect on hepa-
sively by the liver in unaltered form without enterohepatic circu- tocytic clearance of ICG (48, 49).
lation (41, 42). PDRICG is a well-established quantitative marker In summary, the results indicate a 1-adrenoceptor–dependent
of liver function in intensive care units. After intravenous adminis- induction of HO-1, resulting in a protection of hepatic function
tration, ICG is immediately bound to plasma proteins and is con- and perfusion during hemorrhagic shock. In line with these results,
ﬁned to the vascular compartment. PDRICG is directly inﬂuenced dobutamine pretreatment might be a potential means to induce
by hepatic blood ﬂow and hepatocellular metabolism. In case of HO-1 in expected settings of ischemia-reperfusion events after
a reduced metabolic capacity or a dose-dependent saturation of major liver surgery or transplantation. Furthermore, the ability
hepatic dye elimination capability, the ﬂow-dependency of ICG of 1-agonists to stimulate the HO-catalyzed production of
elimination is no longer crucial and PDRICG becomes dependent the vasodilator CO may play a role in liver blood ﬂow regulation
on hepatocellular metabolism. An improvement in gastric muco- (e.g., during dobutamine therapy in critically ill patients).
206 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 174 2006
Figure 9. Cardiac injury was estimated by measurement of troponin
serum enzyme levels (A ), pancreatic injury was estimated by serum
enzyme levels of lipase (B ), and renal injury was estimated by measure-
ment of serum enzyme levels of creatinine (C ). Serum levels of troponin
at the end of the experiment were lower in dobutamine-treated animals
compared with the vehicle group (A ). There was no difference in serum
creatinine and lipase levels between dobutamine- and vehicle-treated
animals at the end of the experiment (C ). $p 0.05 compared with
baseline; #p 0.05 compared with vehicle.
1. Bauer I, Wanner GA, Rensing H, Alte C, Miescher EA, Wolf B, Pannen
BH, Clemens MG, Bauer M. Expression pattern of heme oxygenase
isoenzymes 1 and 2 in normal and stress-exposed rat liver. Hepatology
2. McCoubrey WK Jr, Huang TJ, Maines MD. Isolation and characteriza-
tion of a cDNA from the rat brain that encodes hemoprotein heme
oxygenase-3. Eur J Biochem 1997;247:725–732.
3. Goda N, Suzuki K, Naito M, Takeoka S, Tsuchida E, Ishimura Y,
Tamatani T, Suematsu M. Distribution of heme oxygenase isoforms
in rat liver: topographic basis for carbon monoxide-mediated microvas-
cular relaxation. J Clin Invest 1998;101:604–612.
4. Choi AM, Alam J. Heme oxygenase-1: function, regulation, and implica-
tion of a novel stress-inducible protein in oxidant-induced lung injury.
Am J Respir Cell Mol Biol 1996;15:9–19.
5. Rensing H, Bauer I, Datene V, Patau C, Pannen BH, Bauer M. Differen-
tial expression pattern of heme oxygenase-1/heat shock protein 32 and
nitric oxide synthase-II and their impact on liver injury in a rat model
of hemorrhage and resuscitation. Crit Care Med 1999;27:2766–2775.
6. Suematsu M, Goda N, Sano T, Kashiwagi S, Egawa T, Shinoda Y,
Ishimura Y. Carbon monoxide: an endogenous modulator of sinusoidal
tone in the perfused rat liver. J Clin Invest 1995;96:2431–2437.
7. Kyokane T, Norimizu S, Taniai H, Yamaguchi T, Takeoka S, Tsuchida
E, Naito M, Nimura Y, Ishimura Y, Suematsu M. Carbon monoxide
from heme catabolism protects against hepatobiliary dysfunction in
endotoxin-treated rat liver. Gastroenterology 2001;120:1288–1291.
8. Pannen BH, Kohler N, Hole B, Bauer M, Clemens MG, Geiger KK.
Protective role of endogenous carbon monoxide in hepatic microcircu-
latory dysfunction after hemorrhagic shock in rats. J Clin Invest 1998;
9. Rensing H, Bauer I, Zhang JX, Paxian M, Pannen BH, Yokoyama Y,
Clemens MG, Bauer M. Endothelin-1 and heme oxygenase-1 as modu-
lators of sinusoidal tone in the stress-exposed rat liver. Hepatology
10. Bauer M, Pannen BH, Bauer I, Herzog C, Wanner GA, Hanselmann R,
Zhang JX, Clemens MG, Larsen R. Evidence for a functional link
between stress response and vascular control in hepatic portal circula-
tion. Am J Physiol 1996;271:929–935.
11. Bauer I, Rensing H, Florax A, Ulrich C, Pistorius G, Redl H, Bauer M.
Expression pattern and regulation of heme oxygenase-1/heat shock
protein 32 in human liver cells. Shock 2003;20:116–122.
12. Immenschuh S, Kietzmann T, Hinke V, Wiederhold M, Katz N, Muller-
Eberhard U. The rat heme oxygenase-1 gene is transcriptionally in-
duced via the protein kinase A signaling pathway in rat hepatocyte
cultures. Mol Pharmacol 1998;53:483–491.
13. Morgan NG, Waynick LE, Exton JH. Characterisation of the alpha 1-
adrenergic control of hepatic cAMP in male rats. Eur J Pharmacol
14. Rensing H, Bauer I, Kubulus D, Wolf B, Winning J, Ziegeler S, Bauer M.
Heme oxygenase-1 gene expression in pericentral hepatocytes through
beta1-adrenoceptor stimulation. Shock 2004;21:376–387.
15. Kubulus D, Rensing H, Paxian M, Thierbach JT, Meisel T, Redl H,
Bauer M, Bauer I. Inﬂuence of heme-based solutions on stress protein
expression and organ failure after hemorrhagic shock. Crit Care Med
1-Agonists may be a useful pharmacologic tool to modulate 16. Ozawa N, Goda N, Makino N, Yamaguchi T, Yoshimura Y, Suematsu M.
hepatocellular induction of HO-1 in vivo to provide pharmaco- Leydig cell-derived heme oxygenase-1 regulates apoptosis of premei-
logic preconditioning. otic germ cells in response to stress. J Clin Invest 2002;109:457–467.
17. Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a
Conflict of Interest Statement : None of the authors has a financial relationship delay of lethal cell injury in ischemic myocardium. Circulation 1986;
with a commercial entity that has an interest in the subject of this manuscript. 74:1124–1136.
Raddatz, Kubulus, Winning, et al.: Pharmacologic Preconditioning with Dobutamine 207
18. Raeburn CD, Cleveland JC Jr, Zimmerman MA, Harken AH. Organ reperfusion: evidence for a role for superoxide anion. Hepatology
preconditioning. Arch Surg 2001;136:1263–1266. 1992;15:507–514.
19. Yonezawa K, Yamamoto Y, Yamamoto H, Ishikawa Y, Uchinami H, 34. Vollmar B, Glasz J, Leiderer R, Post S, Menger MD. Hepatic microcircu-
Taura K, Nakajima A, Yamaoka Y. Suppression of tumor necrosis latory perfusion failure is a determinant of liver dysfunction in warm
factor-alpha production and neutrophil inﬁltration during ischemia- ischemia-reperfusion. Am J Pathol 1994;145:1421–1431.
reperfusion injury of the liver after heat shock preconditioning. 35. Clemens MG, Bauer M, Pannen BH, Bauer I, Zhang JX. Remodeling
J Hepatol 2001;35:619–627. of hepatic microvascular responsiveness after ischemia/reperfusion.
20. Mato JM, Corrales FJ, Avila MA. Heat shock: a barrier to hepatic Shock 1997;8:80–85.
ischemic injury. J Hepatol 2001;35:663–665. 36. Wakabayashi Y, Takamiya R, Mizuki A, Kyokane T, Goda N, Yamaguchi
21. Matsumoto K, Honda K, Kobayashi N. Protective effect of heat precondi- T, Takeoka S, Tsuchida E, Suematsu M, Ishimura Y. Carbon monoxide
overproduced by heme oxygenase-1 causes a reduction of vascular
tioning of rat liver graft resulting in improved transplant survival.
resistance in perfused rat liver. Am J Physiol 1999;277:1088–1096.
37. Jalan R, Plevris JN, Jalan AR, Finlayson ND, Hayes PC. A pilot study
22. Fudaba Y, Tashiro H, Miyata Y, Ohdan H, Yamamoto H, Shibata S,
of indocyanine green clearance as an early predictor of graft function.
Nishihara M, Asahara T, Fukuda Y, Goto A, et al. Oral administration Transplantation 1994;58:196–200.
of geranylgeranylacetone protects rat livers from warm ischemic injury. 38. Plevris JN, Jalan R, Bzeizi KI, Dollinger MM, Lee A, Garden OJ, Hayes
Transplant Proc 1999;31:2918–2919. PC. Indocyanine green clearance reﬂects reperfusion injury follow-
23. Fudaba Y, Ohdan H, Tashiro H, Ito H, Fukuda Y, Dohi K, Asahara T. ing liver transplantation and is an early predictor of graft function.
Geranylgeranylacetone, a heat shock protein inducer, prevents pri- J Hepatol 1999;30:142–148.
mary graft nonfunction in rat liver transplantation. Transplantation 39. Gottlieb ME, Sarfeh IJ, Stratton H, Goldman ML, Newell JC, Shah DM.
2001;72:184–189. Hepatic perfusion and splanchnic oxygen consumption in patients
24. Fudaba Y, Tashiro H, Ohdan H, Miyata Y, Shibata S, Shintaku S, postinjury. J Trauma 1983;23:836–843.
Nishihara M, Ito H, Fukuda Y, Asahara T, et al. Prevention of warm 40. Matuschak GM, Pinsky MR, Rogers RM. Effects of positive end-
ischemic injury in rat liver transplantation by geranylgeranylacetone. expiratory pressure on hepatic blood ﬂow and performance. J Appl
Transplant Proc 2000;32:1615–1616. Physiol 1987;62:1377–1383.
25. Redaelli CA, Tian YH, Schaffner T, Ledermann M, Baer HU, Dufour 41. Caesar J, Shaldon S, Chiandussi L, Guevara L, Sherlock S. The use of
JF. Extended preservation of rat liver graft by induction of heme indocyanine green in the measurement of hepatic blood ﬂow and as
oxygenase-1. Hepatology 2002;35:1082–1092. a test of hepatic function. Clin Sci 1961;21:43–57.
26. Amersi F, Buelow R, Kato H, Ke B, Coito AJ, Shen XD, Zhao D, Zaky 42. Wheeler HO, Cranston WI, Meltzer JI. Hepatic uptake and biliary excre-
J, Melinek J, Lassman CR, et al. Upregulation of heme oxygenase-1 tion of indocyanine green in the dog. Proc Soc Exp Biol Med 1958;
protects genetically fat Zucker rat livers from ischemia/reperfusion
43. Silverman HJ, Tuma P. Gastric tonometry in patients with sepsis: effects
injury. J Clin Invest 1999;104:1631–1639.
of dobutamine infusions and packed red blood cell transfusions. Chest
27. Schnuelle P, Lorenz D, Mueller A, Trede M, van der Woude FJ. Donor 1992;102:184–188.
catecholamine use reduces acute allograft rejection and improves graft 44. Gutierrez G, Clark C, Brown SD, Price K, Ortiz L, Nelson C. Effect of
survival after cadaveric renal transplantation. Kidney Int 1999;56:738– dobutamine on oxygen consumption and gastric mucosal pH in septic
746. patients. Am J Respir Crit Care Med 1994;150:324–329.
28. Schnuelle P, Berger S, de Boer J, Persijn G, van der Woude FJ. Effects 45. Neviere R, Mathieu D, Chagnon JL, Lebleu N, Wattel F. The contrasting
of catecholamine application to brain-dead donors on graft survival effects of dobutamine and dopamine on gastric mucosal perfusion in
in solid organ transplantation. Transplantation 2001;72:455–463. septic patients. Am J Respir Crit Care Med 1996;154:1684–1688.
29. Katori M, Busuttil RW, Kupiec-Weglinski JW. Heme oxygenase-1 system 46. Levy B, Bollaert PE, Charpentier C, Nace L, Audibert G, Bauer P,
in organ transplantation. Transplantation 2002;74:905–912. Nabet P, Larcan A. Comparison of norepinephrine and dobutamine
30. Cardani R, Zavanella T. Immunohistochemical localization of beta to epinephrine for hemodynamics, lactate metabolism, and gastric
1-adrenergic receptors in the liver of male and female F344 rat. tonometric variables in septic shock: a prospective, randomized study.
Histochem Cell Biol 2001;116:441–445. Intensive Care Med 1997;23:282–287.
47. Levy B, Bollaert PE, Lucchelli JP, Sadoune LO, Nace L, Larcan A.
31. Vallet B, Dupuis B, Chopin C. Dobutamine: mechanisms of action and
Dobutamine improves the adequacy of gastric mucosal perfusion in
use in acute cardiovascular pathology. Ann Cardiol Angeiol (Paris)
epinephrine-treated septic shock. Crit Care Med 1997;25:1649–1654.
1991;40:397–402. 48. Joly LM, Monchi M, Cariou A, Chiche JD, Bellenfant F, Brunet F,
32. Rensing H, Bauer I, Peters I, Wein T, Silomon M, Jaeschke H, Bauer Dhainaut JF. Effects of dobutamine on gastric mucosal perfusion and
M. Role of reactive oxygen species for hepatocellular injury and heme hepatic metabolism in patients with septic shock. Am J Respir Crit
oxygenase-1 gene expression after hemorrhage and resuscitation. Care Med 1999;160:1983–1986.
Shock 1999;12:300–308. 49. De Backer D, Creteur J, Noordally O, Smail N, Gulbis B, Vincent JL.
33. Koo A, Komatsu H, Tao G, Inoue M, Guth PH, Kaplowitz N. Con- Does hepato-splanchnic VO2/DO2 dependency exist in critically ill
tribution of no-reﬂow phenomenon to hepatic injury after ischemia- septic patients? Am J Respir Crit Care Med 1998;157:1219–1225.