Reviews

Liver dysfunction in sepsis

Ewa A. Woźnica1,A–D, Małgorzata Inglot2,C,E, Ryszard K. Woźnica3,C,E, Lidia Łysenko1,C,E,F
1
Department of Anaesthesiology and Intensive Therapy, Wroclaw Medical University, Poland
2
Department of Infectious Diseases, Liver Diseases and Acquired Immune Deficiences, Wroclaw Medical University, Poland
3
Altnagelvin Area Hospital, Londonderry, UK

A – research concept and design; B – collection and/or assembly of data; C – data analysis and interpretation;
D – writing the article; E – critical revision of the article; F – final approval of the article

Advances in Clinical and Experimental Medicine, ISSN 1899-5276 (print), ISSN 2451-2680 (online) Adv Clin Exp Med. 2018;27(4):547–551

Address for correspondence
Ewa Woźnica
Abstract
E-mail: [email protected] Despite continuous progress in medicine, sepsis remains the main cause of deaths in the intensive care unit.
Liver failure complicating sepsis/septic shock has a significant impact on mortality in this group of patients.
Funding sources
None declared
The pathophysiology of sepsis-associated liver dysfunction is very complicated and still not well understood.
According to the Surviving Sepsis Campaign (SSC) Guidelines, the diagnosis of liver dysfunction during sepsis
Conflict of interest is based on the increase in bilirubin concentration >2 mg/dL and the occurrence of coagulation disorders with
None declared INR > 1.5. The lack of specificity and ability to distinguish acute liver failure from previous liver dysfunction
disqualifies bilirubin as a single parameter reflecting the complex liver function. Clinical manifestations
of sepsis-associated liver dysfunction include hypoxic hepatitis, sepsis-induced cholestasis and dysfunction
Received on October 12, 2016 of protein synthesis manifesting with, e.g., coagulopathies. Detoxifying liver dysfunction, which is associated
Reviewed on December 29, 2016
Accepted on January 9, 2017 with an increase in serum ammonia concentration, manifesting with e.g., confusion, loss of consciousness and
hepatic encephalopathy, may be disguised by analgosedation used in the intensive care unit. To determine
a liver dysfunction in a critically ill patient, the concept of shock liver may be used. It is a complex syndrome
of hemodynamic, cellular, molecular and immunologic changes leading to severe liver hypoxia. In clinical
practice, there is no standardized diagnostic panel that would allow for an early, clear diagnosis of acute
liver dysfunction, and there is no therapeutic panel enabling the full restoration of damaged liver function.
The aim of the article is to present the pathophysiology and clinical manifestations of sepsis-associated liver
dysfunction.
Key words: sepsis, MODS, liver dysfunction, shock liver

DOI
10.17219/acem/68363

Copyright
© 2018 by Wroclaw Medical University
This is an article distributed under the terms of the
Creative Commons Attribution Non-Commercial License
(http://creativecommons.org/licenses/by-nc-nd/4.0/)
548 E. Woźnica, et al. Liver dysfunction in sepsis

Introduction The aim of this article is to explain the pathophysiology

and review the clinical manifestations of sepsis-associated
Despite the continuous progress in medicine, sepsis re- liver dysfunction.
mains the leading cause of deaths in the intensive care
units (ICUs). American data estimate the incidence of sep-
sis to be about 300 cases/100,000 people, and the mortality The liver in sepsis
rate ranges between 30% and 50%.1–3 That is more than
the number of deaths caused by prostate cancer, breast The liver is a parenchymatous organ composed of 3 types
cancer and AIDS all together.4 of cells: hepatocytes (HCs), Kupffer cells (KCs), and liver
The prevalence of sepsis in Poland according to recently sinusoidal endothelial cells (LSECs).
published data is estimated to be about 25% of patients hos- In  the  course of  sepsis/septic shock, the  metabolism
pitalized in the ICUs. Among patients diagnosed with severe of HCs is modified towards the inflammatory response.
sepsis, almost half of them (44%) developed septic shock.5 The main cytokine of the liver inflammatory response is in-
The development of multiple organ dysfunction syn- terleukin-6 (IL-6), which is responsible for synthesizing
drome (MODS) is  one of  the  complications of  sep- acute phase proteins such as C-reactive protein (CRP), α-1
sis. A study conducted by Kubler et al, which analyzed antitrypsin, fibrinogen, prothrombin, and haptoglobin.9
the course and outcome of severe sepsis in Poland, revealed The increase in acute phase protein concentrations leads
that patients admitted to ICUs were severely ill. Dysfunc- to inhibition of the protein C pathway, and thus it is respon-
tion of 1 or 2 organs was diagnosed in 9–12% of patients sible for the increase of coagulation factor activity. Secretion
at the time of admission, whereas most of the patients of IL-6 is induced by endotoxin (lipopolysaccharide – LPS)
(89%) developed dysfunction of 3 or more organs during and tumor necrosis factor-α (TNF-α).
their stay in ICUs.6 Lipopolysaccharide also stimulates secretion of TNF-α,
The level of organ dysfunction, including liver failure, interleukin-1β(IL-1β), interleukin-12 (IL-12) and interleu-
may vary from a mild organ dysfunction to life-threatening kin-18 (IL-18) by KCs.10 Il-18 is the main factor responsible
fulminant organ failure. for LPS-induced liver damage. IL-18 leads to interferon-γ
During sepsis, not only infection itself, but also hyper- (IFNγ) secretion, which results in hepatocyte apoptosis,
activity of the inflammatory response, microcirculatory an  increase in  TNF-α concentration and upregulation
failure, and side effects of  the  therapy are responsible of  CD14 expression. CD14 is  a  monocyte/macrophage
for liver injury. surface receptor responsible for binding the lipopolysac-
The liver plays a pivotal role in maintaining homeostasis. charide binding protein (LPS/LBP) complex.
Its functions include: metabolism of carbohydrates, lipids, Kupffer cells are macrophages of the liver, which are re-
proteins and hormones; biosynthesis of blood components, sponsible for scavenging portal vein blood from bacteria and
enzymes and clotting factors; production of bile; detoxi- endotoxins. As a response to LPS stimulation, KCs release
fication; metabolism of nitrogen compounds – synthesis TNF-α, IL-1β, IL-6, IL-12 and IL-18, reactive oxygen spe-
of urea; storage of glycogen, cholesterol, vitamins (A, D, cies (ROS), and nitric oxide (NO), which induce endothelial
B12), iron, and many more. cell and hepatocyte injury. During the early stages of sep-
The  incidence of  sepsis-associated liver dysfunction sis, as a response to KCs’ release of TNF-α and leukotriene
(SALD) is hard to establish due to the lack of a homogeneous B4, neutrophils are recruited to the liver.11 Cytokines pro-
definition, and consequently lack of a thorough registry. duced by neutrophils are responsible for further damage
What is more, the incidence of SALD varies, as currently of hepatocytes.
there are no specific diagnostic tools available, especially Endothelial cell (EC) dysfunction contributes to the de-
ones that could detect liver injury in the early stages. velopment of MODS.12 Liver sinusoidal endotheliac cells
In a study performed by Birrer R. et al., the presence are also involved in cytokine release in response to LPS
of hepatic injury was identified in 1.1% of admissions.7 stimulation, and they are the main hepatic source of en-
The causes of hepatic injury in the group of patients were: dothelin-1 (ET-1), which is a strong vasoconstrictor.13,14
hypotension, congestive heart failure (secondary liver hy- During sepsis, ET-1 is secreted in response to NO release
poperfusion), sepsis, respiratory failure resulting in hy- from inducible nitric oxide synthase (iNOS).15
poxia, and other causes resulting in hypoxemia. Sepsis Endothelin-1 has also been found to have a strong corre-
was diagnosed in 16.1% of patients who developed hypoxic lation with the inflammatory response. It involves the ex-
hepatitis. pression of cytokines such as TNF-α, IL-1 and IL-6, as well
Another study performed by Kobashi H. et al. observed as the activation of transcription factors such as nuclear
SALD in 34.7% of the patients.8 Among those who devel- factor kappa B (NF-κB).16 Endothelin-1 also increases syn-
oped SALD, the authors distinguished 3 groups: “hepa- thesis of TNF-α in monocytes and macrophages.17 In their
tocellular” (21.8%), “cholestatic” (48.1%) and “shock liver” study, Brauner et  al. found ET-1 concentrations to  be
(30.1%). In each group, jaundice as a complication was an early and sensitive predictor of mortality in patients
observed in 17.6%, 33% and 8.5% of cases, respectively. with septic shock.18
Adv Clin Exp Med. 2018;27(4):547–551 549

In order to maintain a proper hepatic perfusion, a bal- proteins. 30 Energy shortage caused by hypoxemia/liver
ance of vasoactive effects of ET-1 and NO is needed. hypoperfusion may impair most bile synthesis steps.31
Nitric oxide is responsible for the relaxation of vascular Liver histological examinations in patients with jaundice
smooth muscle cells, the regulation of hepatic blood flow occurring during bacterial infection revealed the pres-
and the inhibition of platelet aggregation and the adhesion ence of intrahepatic cholestasis. The impairment of bile
of leukocytes to endothelium.19 The impact of NO on liver transport is a result of alterations in genes activating tran-
depends on its source. The endogenous NO released from scription and modifying posttranslational treatment of bile
endothelial nitric oxide synthase (eNOS) helps protect acids transporting proteins caused by LPS and proinflam-
the liver cells from damage caused by vasoconstriction matory cytokines.32
induced by  endothelin-1 (ET-1) release, whereas iNOS Laboratory test abnormalities include an increase of to-
promotes microvascular dysfunction and thereby SALD.15 tal bilirubin (>2 mg/dL), alkaline phosphatase, ALT, and
The other components that regulate the vasculature tone AST.33
are carbon monoxide (CO) and hydrogen sulfide (H2S).
One of the products of cysteine metabolism is H2S, which Hypoxic hepatitis
is synthesized in the brain, in the liver, and in vessels.20
H2S may also by synthesized by microflora of the gastro- Hypoxic hepatitis (HH) may be the cause of fulminant
intestinal tract and transferred to the liver via the portal hepatitis. In septic shock, an increase in blood flow and
circulation.21 Synthesis of H2S is increased during sepsis.22 cardiac output is not enough to compensate increased
Hydrogen sulfide relaxes vascular smooth muscle cells hepatic oxygen demand.34 Decreased hepatic blood flow
and inhibits their proliferation and platelet aggregation.23 in shock does not always cause HH; it may occur in patients
Finally, H 2S oxidation may contribute to  exacerbation with normal blood pressure.
of sepsis-associated tissue hypoxia.24 Other risk factors causing HH are LPS and inflammatory
Carbon monoxide (CO) is one of the products of heme cytokines. Hypoxic hepatitis may also be a result of the re-
degradation by  heme oxygenases (HO) (Fig. 1). Carbon oxygenation phase in the course of the ischemia/reperfu-
monoxide is responsible for maintaining the liver’s regional sion phenomenon.
perfusion, resulting in the activation of leucocytes.25,26 Fur- In the course of HH, except for a rapid (24 h from the
thermore, HO-1/CO prevents EC apoptosis via suppressing onset of shock) substantial increase in aminotransferases
inflammatory reactions contributing to EC apoptosis. CO and lactate dehydrogenase activity, an early decrease in
generated through heme catabolism by HO has an anti- serum prothrombin concentration is observed.35
apoptotic effect on ECs through activation of mitogen-ac-
tivated protein kinases (MAPK).27 It is still not clear which Coagulopathy
of the above-mentioned properties of CO has a hepato-
protective influence in sepsis. A wide range of coagulopathy may be observed in sep-
In a study on rodent model ischemia-reperfusion in- sis, starting with mild deviation in  laboratory results
duced systemic inflammation, exogenous CO was shown (prolonged clotting time, decreased number of platelets)
to have a hepatoprotective effect via improving liver cell to severe coagulopathy and/or disseminated intravascular
integrity and the redox state as well as protecting the liver coagulation (DIC).
microcirculation.28 The main cause of coagulopathy in sepsis is microvas-
cular endothelial injury resulting in an imbalance between
fibrinolysis and coagulation. 36 The changes seen in en-
Clinical manifestations of SALD dothelial injury include loss of vascular tone, capillary
obstruction by platelet or fibrin clots, as well as the degra-
In septic patients, the spectrum of liver dysfunction may dation of heparan sulfate leading to a pro-coagulant state.37
vary from subclinical to symptomatic liver failure. In criti- Coagulopathy may be another symptom of liver disease.
cally ill patients the concept of “shock liver” may be used. Several factors may contribute to  hemostatic changes
“Shock liver” is a syndrome of hemodynamic, cellular, im- in liver disease (Table 1).38
munologic and molecular disorders.29 SALD can manifest
in 2 clinical forms – jaundice/sepsis-induced cholestasis,
and hypoxic hepatitis (HH).19 Coagulopathy may be another Assessment of liver function
symptom of SALD. These processes are still not well un-
derstood due to the complexity of their pathomechanisms. Diagnosis of  liver dysfunction in  sepsis, according
to the Surviving Sepsis Campaign (SSC) Guidelines, is based
Jaundice/sepsis-induced cholestasis on an increase in serum bilirubin concentration >2 mg/dL
(34.2 μmol/L) and occurrence of coagulopathy (INR > 1.5).39
The synthesis of bile is a complex process, requiring prop- Currently, there are no specific biomarkers available
er energy input and normal function of transmembrane that  would allow for  an  early diagnosis of  acute liver
550 E. Woźnica, et al. Liver dysfunction in sepsis

damage in the course of sepsis/septic shock and distin- However, these parameters cannot be used for contin-
guishing it from a pre-existing liver pathology. Liver func- uous and rapid monitoring of liver function in patients
tion can be assessed using static and dynamic parameters. treated in  ICU, nor are they  diagnostic or  prognostic
Static parameters include: in this group of patients.40,41 As mentioned above, accord-
–– secretory capacity – bilirubin; ing to the SSC Guidelines, serum bilirubin concentration
–– parameters of cholestasis – alkaline phosphatase, (>2 mg/dL or >34.2 μmol/L) is used as a single marker
γ-glutamyltransferase; guideline to diagnose liver dysfunction.39 Due to a number
–– intracellular enzymes activity – alanine amino- of drawbacks limiting its application, serum bilirubin is not
transferase, aspartate aminotransferase, glutamate an appropriate marker to reflect complex liver function.
dehydrogenase; Increase in serum bilirubin concentration is neither spe-
–– synthesizing capacity – albumin, clotting factors V cific nor does it allow acute liver dysfunction to be distin-
and VII.40 guished from a pre-existing liver pathology.41 Table 2 shows
causes of hyperbilirubinemia in sepsis.33
Table 1. Factors contributing to hemostatic changes in liver dysfunction Dynamic parameters assessing liver function include:
Hemostatic changes in liver dysfunction –– indocyanine green (ICG), caffeine and bromosul-
promoting hemostasis impairing hemostasis
fophthalein clearance;
–– liver detoxification capacity – measuring the concen-
• low plasminogen activity • reduced hematocrit
• decreased protein • thrombocytopenia tration of  14CO2 in exhaled air (measuring the con-
S, protein C and • production of nitric oxide and centration of  [14 C]aminopyrine, [14 C]methacetin,
antithrombin activity prostacyclin [14 C]erythromycin metabolites) and measuring
• increased VWF activity • low serum concentration of coagulation
• increased factor VIII serum factors II, V, VII, IX, X, and XI
the  concentration of  lidocaine/midazolam serum
concentration • dysfibrinogenemia metabolites;
• vitamin K deficiency –– ability to eliminate galactose.40
• elevated activity of tPA
• low serum concentrations of plasmin
inhibitor, factor XII and TAFI

VWF – von Willebrand factor; TAFI – thrombin activatable fibrinolysis

Maximal liver function capacity
inhibitor; tPA – tissue plasminogen activator.
Kaffarnik et  al., in  their study published in  Critical
Care in 2013, attempted to assess the maximal liver func-
Table 2. Causes of hyperbilirubinemia in sepsis
tion capacity (LiMax) test as a useful tool for early di-
1. Cholestasis agnosis of sepsis-associated liver dysfunction.42 LiMax
2. Hemolysis – drug-induced/infection-related is a non-invasive breath test using 13C-labeled methace-
• in normal RBCs tin, which is exclusively metabolized by cytochrome P450
• in RBCs with enzyme defects
(1A2) to  13CO2 and acetaminophen. The idea of the test
3. Hepatic dysfunction
• ischemia: hypotension, hypoxia
is to measure the amount of exhaled 13CO2 . The result
• hepatocellular damage is given as the LiMax value in μg per kg of body weight
• bilirubin transport dysfunction: decreased uptake, canalicular per hour. It  shows the  speed of  substrate metabolism,
transport, clearance
thus  allowing one to  evaluate the  metabolic capacity
RBCs – red blood cells. of the liver.

Table 3. Types of drug-induced liver injury. Examples of hepatotoxic drugs

Cellular Cholestatic
Mixed
(↑alanine aminotransferase) (↑total bilirubin, ↑alkaline phosphatase)
• Acetaminophen • Amoxicillin/clavulanate • Amitriptyline
• Allopurinol • Anabolic steroids • Azathioprine
• Amiodarone • Chlorpromazine • Captopril
• Baclofen • Clopidogrel • Carbamazepine
• Isoniazid • Oral contraceptives • Clindamycin
• Ketoconazole • Erythromycins • Enalapril
• Lisinopril • Irbesartan • Nitrofurantoin
• Losartan • Phenothiazines • Phenobarbital
• Methotrexate • Tricyclic antidepressants • Phenytoin
• NSAIDs • Sulfonamides
• Omeprazole • Trimethoprim/sulfamethoxazole
• Pyrazinamide • Verapamil
• Rifampin
• Statins
• Tetracyclines
• Valproate
Adv Clin Exp Med. 2018;27(4):547–551 551

In the study, a pathologic deterioration of LiMax values These include:

in patients with septic shock was observed within 2 days 1. avoiding potentially hepatotoxic drugs;
after the onset of sepsis. Among patients with LiMax, 2. early enteral feeding of hemodynamically stable patients;
<100 μg/kg/h the mortality rate was 55%, and with LiMax 3. glucose concentration monitoring and adequate glu-
>100 μg/kg/h the mortality rate was 0%. The authors’ cose supply if necessary;
conclusion was that LiMax values <100 μg/kg/h could be 4. extracorporeal liver support – Molecular Adsorbent
a good predictor of morbidity and mortality. Recirculating System (MARS) albumin dialysis, single-
pass albumin dialysis (SPAD).19,46,47
Drugs are an important cause of liver injury. Examples
Indocyanine green clearance of potentially hepatotoxic drugs are given in Table 3.48
La Mura proposed simvastatin administration for the pre-
Indocyanine green (ICG) is a non-toxic, water-soluble vention of LPS-induced intrahepatic endothelial dysfunc-
fluorescent dye. Its spectrophotometric evaluation does tion. He found that  prophylactic simvastatin prevents
not depend on  oxygen saturation and serum bilirubin endotoxemia-induced liver injury, reduces liver inflamma-
concentration. Indocyanine green clearance (ICG PDR) tion, and prevents microvascular dysfunction in rodents.49
can be used to reflect the liver function. Because it is not In other studies, prophylactic simvastatin has also been
metabolized, it is secreted almost exclusively by the liver reported as being able to correct endothelial dysfunction.50
and it is not subject to enterohepatic circulation.43
A significant limitation in using ICG PDR is the hemo-
dynamic condition of the patient, as ICG PDR depends Summary
on hepatic blood flow.44 Additional parameters limiting
the use of this test are the serum bilirubin concentration, The incidence of sepsis-associated liver failure is hard
serum albumin concentration, body weight, and patient’s to  estimate, but it  is  incontestable that  liver failure
age.45 Studies have shown that ICG PDR may be a diag- as a complication of sepsis dramatically worsens the out-
nostic and prognostic tool in monitoring acute liver fail- come of the patients. It is important to remember that dur-
ure in critically ill patients in the ICU, but there are still ing sepsis not only the infection itself is responsible for liver
no randomized control trials clearly confirming the utility dysfunction, but also hyperreactivity of the inflamma-
of ICG PDR in daily clinical practice.43 tory response, microcirculatory failure, and side effects
It is worth underlining that in clinical practice there are of the therapy. Only an early diagnosis of sepsis and its
no standardized diagnostic panels allowing for an early, complications as well as quick implementation of thera-
clear diagnosis of acute liver dysfunction. Until now only peutic bundles allow reducing the incidence of severe organ
a few studies have been published, and their results remain complications, to shorten the hospitalization time and
equivocal. improve patients’ quality of life.

References
Therapeutic considerations 1. Blanco J, Muriel-Bombín A, Sagredo V, et al. Incidence, organ dys-
function and mortality in severe sepsis: A Spanish multicentre study.
Crit Care. 2008;12:R158.
Currently, there is  no  specific therapeutic treatment 2. Martin GS, Mannino DM, Eaton S, Moss M. The  epidemiology
available for the full restoration of damaged liver func- of sepsis in the United States from 1979 through 2000. N Engl J Med.
tion. The therapy, according to the SSC Guidelines, should 2003;348:1546–1554.
focus on eradicating infection and treating sepsis and its 3. Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pin-
sky MR. Epidemiology of severe sepsis in the United States: Anal-
complications. 39 Furthermore, there are tools available ysis of incidence, outcome, and associated costs of care. Crit Care
that could reduce the risk of further damage to this organ. Med.2001;29:1303–1310.

O₂ CO

HEME BELIVERDIN
HO
NADPH NADP+ + H₂O Fe²+

Fig. 1. Heme degradation products

HO – heme oxygenase; CO – carbon monoxide; NADPH – r educed form of nicotinamide adenine dinucleotide phosphate; NADP+ – nicotinamide adenine
dinucleotide phosphate.
552 E. Woźnica, et al. Liver dysfunction in sepsis

4. National Institute of General Medical Sciences – Sepsis Fact Sheet. 29. Strassburg CP. Gastrointestinal disorders of the critically ill. Shock
http://www.nigms.nih.gov/Education/Pages/factsheet_sepsis.aspx Liver. Best Pract Res Clin Gastroenterol. 2003;17:369–381.
Accessed on December 28, 2016. 30. Trauner M, Meier PJ, Boyer JL. Molecular pathogenesis of cholesta-
5. Kübler A, Adamik B, Ciszewicz-Adamiczka B, Ostrowska E. Severe sep- sis. N Engl J Med. 1998;339:1217–1227.
sis in intensive care units in Poland – Point prevalence study in 2012 31. Fuchs M, Sanyal AJ. Sepsis and cholestasis. Clin Liver Dis. 2008;12:151–
and 2013. Anaesthesiol Intensive Ther. 2015;47:315–319. 172. ix.
6. Kübler A, Adamik B, Durek G, et al. Results of the severe sepsis reg- 32. Moseley RH. Sepsis and cholestasis. Clin Liver Dis. 2004;8:83–94.
istry in intensive care units in Poland from 2003–2009. Anaesthesiol 33. Chand N, Sanyal AJ. Sepsis-induced cholestasis. Hepatology.
Intensive Ther. 2015;47:7–13. 2007;45(1):230–241.
7. Birrer R, Takuda Y, Takara T. Hypoxic hepatopathy: Pathophysiology 34. Dahn MS, Lange P, Lobdell K, Hans B, Jacobs LA, Mitchell RA. Splanch-
and prognosis. Intern Med. 2007,46(14):1063–1070. nic and total body oxygen consumption differences in septic and
8. Kobashi H, Toshimori J, Yamamoto K. Sepsis-associated liver injury: injured patients. Surgery. 1987;101:69–80.
Incidence, classification and the clinical significance. Hepatol Res. 35. Henrion J. Hypoxic hepatitis. Liver Int. 2012;32(7):1039–1052.
2013;43(3):255–266. 36. Taylor FB Jr, Toh CH, Hoots WK, Wada H, Levi M. Scientific Subcom-
9. Aninat C, Seguin P, Descheemaeker P, Morel F, Malledant Y, Gullou- mittee on Disseminated Intravascular Coagulation (DIC) of the Inter-
zo A. Catecholamines induce an inflammatory response in human national Society on Thrombosis and Haemostasis (ISTH): Towards
hepatocytes. Crit Care Med. 2008,36:848–854. definition, clinical and laboratory criteria, and a  scoring system
10. Kolios G, Valatas V, Manousou P, Xidakis C, Notas G, Kouroumalis E. for  disseminated intravascular coagulation. Thromb Haemost.
Nitric oxide and MCP-1 regulation in LPS activated rat Kupffer cells. 2001;86:1327–1330.
Mol Cell Biochem. 2008;319:91–98. 37. Lipinska-Gediga M. Sepsis and septic shock-is a microcirculation
11. Doi F, Goya T, Torisu M. Potential role of  hepatic macrophages a main player? Anaesthesiol Intensive Ther. September 23, 2016. doi:
in neutrophil-mediated liver injury in rats with sepsis. Hepatology. 10.5603/AIT.a2016.0037. [Epub ahead of print]
1993;17:1086–1094. 38. Lisman T, Leebeek FWG. Hemostatic alterations in  liver disease:
12. Aird WC. The role of the endothelium in severe sepsis and multiple A review on pathophysiology, clinical consequences, and treatment.
organ dysfunction syndrome. Blood. 2003;101(10):3765–3777. Dig Surg. 2007;24:250–258.
13. Wang D, Yin Y, Yao Y. Advances in sepsis-associated liver dysfunc- 39. Dellinger RP, Levy MM, Rhodes A, et al. Surviving Sepsis Campaign:
tion. Burns Trauma. 2014;2:97–105. International Guidelines for Management of Severe Sepsis and Sep-
14. Kwok W, Lee SH, Culberson C, Korneszczuk K, Clemens M. Caveolin-1 tic Shock: 2012. Crit Care Med. 2013;41(2):580–637.
mediates endotoxin inhibition of endothelin-1-induced endothelial 40. Sakka S. Assessing liver function. Curr Opin Crit Care. 2007;13:207–214.
nitric oxide synthase activity in liver sinusoidal endothelial cells. Am 41. Marshall JC, Cook DJ, Christou NV, Bernard GR, Sprung CL, Sibbald
J Physiol Gastrointest Liver Physiol. 2009;297(5): G930–G939. WJ. Multiple organ dysfunction score: A reliable descriptor of a com-
15. Hyun-Ae Eum, Sang-Won Park, Sun-Mee Lee. Role of nitric oxide plex clinical outcome. Crit Care Med. 1995;23:1638–1652.
in the expression of hepatic vascular stress genes in response to sep- 42. Kaffarnik, Lock JF, Vetter H, et al. Early diagnosis of sepsis-related
sis. Nitric Oxide. 2007;17:126–133. hepatic dysfunction and its prognostic impact on survival: A pro-
16. Yeager ME, Belchenko DD, Nguyen CM, Colvin KL, Ivy DD, Stenmark spective study with the LiMAx test. Crit Care. 2013;17(5):R259.
KR. Endothelin-1, the unfolded protein response, and persistent 43. Vos JJ, Wietasch JKG, Absalom AR, Hendriks HGD, Scheeren TWL.
inflammation: Role of pulmonary artery smooth muscle cells. Am J Green light for  liver function monitoring using indocyanine
Respir Cell Mol Biol. 2012;46:14–22. green? An overview of current clinical applications. Anaesthesia.
17. Bellisai F, Morozzi G, Scaccia F, et al. Evaluation of the effect of bosen- 2014;69:1364–1376.
tan treatment on proinflammatory cytokine serum levels in patients 44. Janssen MW, Druckrey-Fiskaaen KT, Omidi L, et al. Indocyanine green
affected by  systemic sclerosis. Int J Immunopathol Pharmacol. R15 ratio depends directly on liver perfusion flow rate. J Hepatobili-
2011;24:261–264. ary Pancreat Sci. 2010;17:180–185.
18. Brauner JS, Rohde LE, Clausell N. Circulating endothelin-1 and tumor 45. Kim GY, Bae KS, Noh GJ, Min WK. Estimation of indocyanine green
necrosis factor-a: Early predictors of mortality in patients with sep- elimination rate constant k and retention rate at 15 min using patient
tic shock. Intensive Care Med. 2000;26:305–313. age, weight, bilirubin and albumin. J Hepatobiliary Pancreat Sci.
19. Nesseler N, Launey Y, Aninat C, Morel F, Mallédant Y, Seguin P. Clini- 2009;16:521–528.
cal review: The liver in sepsis. Critical Care. 2012;16:235. 46. Sauer IM, Goetz M, Steffen I, et al. In vitro comparison of the Molec-
20. Bukovska G, Kery V, Kraus JP. Expression of human cystathionine ular Adsorbent Recirculation System (MARS) and single-pass albu-
beta-synthase in Escherichia coli: Purification and characterization. min dialysis (SPAD). Hepatology. 2004;39(5):1408–1414.
Protein Expr Purif. 1994;5:442–448. 47. Woznica R. Single Pass Albumin Dialysis for treatment of the acute
21. Blachier F, Davila AM, Mimoun S, et al. Luminal sulfide and large intes- liver failure – A case report. Med Intens Ratunk. 2007;10(4):233–237.
tine mucosa: Friend or foe? Amino Acids. 2010;39:335–347. 48. Herrine SK. Liver injury caused by drugs. http://www.merckmanu-
22. Zhang H, Zhi L, Moore PK, Bhatia M. Role of hydrogen sulfide in cecal als.com/professional/hepatic-and-biliary-disorders/drugs-and-the-
ligation and puncture induced sepsis in the mouse. Am J Physiol Lung liver/liver-injury-caused-by-drugs Accessed on December 28,2016.
Cell Mol Physiol. 2006;290:L1193–1201. 49. La Mura V, Pasarín M, Meireles CZ, et al. Effects of simvastatin admin-
23. Altaany Z, Moccia F, Munaron L, Mancardi D, Wang R. Hydrogen sul- istration on rodents with lipopolysaccharide-induced liver microvas-
fide and endothelial dysfunction: Relationship with nitric oxide. Curr cular dysfunction. Hepatology. 2013;57:1172–1181.
Med Chem. 2014;21(32):3646–3661. 50. Abraldes JG, Albillos A, Banares R, et al. Simvastatin lowers portal
24. EJ Norris, CR Culberson, S Narasimhan, MG Clemens. The  liver pressure in patients with cirrhosis and portal hypertension: A ran-
as a central regulator of hydrogen sulfide. Shock. 2011;36(3):242–250. domized controlled trial. Gastroenterology. 2009;136:1651–1658.
25. Pannen BHJ, Köhler N, Hole B, Bauer M, Clemens MG, Geiger KK. Pro-
tective role of endogenous carbon monoxide in hepatic microcir-
culatory dysfunction after hemorrhagic shock in rats. J Clin Invest.
1998,102:1220–1228.
26. Hoetzel A, Dolinay T, Schmidt R, Choi AMK, Ryter SW. Carbon mon-
oxide in sepsis. Antioxid Redox Signal. 2007;11(9):2013–2026.
27. Brouarda S, Otterbeinb LE, Anratherc J, et al. Carbon monoxide gen-
erated by heme oxygenase 1 suppresses endothelial cell apoptosis.
JEM. 2000;192(7):1015–1026.
28. Wunder C, Brock RW, Frantz S, et  al. Carbon monoxide, but not
endothelin-1, plays a major role for the hepatic microcirculation
in a murine model of early systemic inflammation. Crit Care Med.
2005;33:2323–2331.