Hauke Rensing

Recovery of hepatocellular ATP and “pericentral apoptosis” after hemorrhage and resuscitation

Progressive liver dysfunction contributes significantly to the development of multiple organ failure after trauma/hemorrhage. This study tested the relative impact of necrotic and apoptotic cell death in a graded model of hemorrhagic shock (mean arterial blood pressure=35±5 mmHg for 1, 2, or 3 h, followed by 2 h, 1 h, or no resuscitation, respectively) in rats. Prolonged periods of hemorrhagic hypotension (3 h) were paralleled by a profound decrease of hepatic ATP levels and occurrence of pericentral necrosis. Resuscitation after shorter periods of hemorrhagic hypotension resulted in restoration of tissue ATP whereas hepatocellular function as assessed by indocyanine green clearance remained depressed (49.9±1.6 mL/(min·kg) at baseline, 28.8± 1.2 mL/(min·kg) after 2 h of resuscitation; P<0.05). Under these conditions, induction of caspase activity and DNA fragmentation were observed in pericentral hepatocytes that could be prevented by the radical scavenger tempol. Pretreatment with z‐Val‐Ala‐Asp(O‐methyl)‐flouromethylketone prevented de novo expression of caspase‐generated cytokeratin 18, DNA fragmentation, and depression of hepatocellular indocyanine green clearance. These data suggest that prolonged low flow/hypoxia induces ATP depletion and pericentral necrosis and restoration of oxygen supply and ATP levels after shorter periods of low flow ischemia propagate programmed cell death or “pericentral apoptosis.”—Paxian, M., Bauer, I., Rensing, H., Jaeschke, H., Mautes, A. E. M., Kolb, S. A., Wolf, B., Stockhausen, A., Jeblick, S., Bauer, M. Recovery of hepatocellular ATP and “pericentral apoptosis” after hemorrhage and resuscitation. FASEB J. 17, 993–1002 (2003)

Argatroban for anticoagulation in continuous renal replacement therapy

Objective:Argatroban, a direct thrombin inhibitor, was evaluated for anticoagulation in continuous renal replacement therapy (CRRT) in critically ill patients with heparin-induced thrombocytopenia type II and acute renal failure. The investigation focused on predictors for the maintenance doses of argatroban with efficacy and safety of argatroban being secondary outcomes. Design:Prospective, dose finding study. Setting:Two intensive care units (medical and surgical) of a university hospital. Patients:Medical and surgical patients (n = 30) with acute or histories of heparin-induced thrombocytopenia type II and acute renal failure with necessity for CRRT. Intervention:CRRT with argatroban for anticoagulation. Measurements and Main Results:Critical illness severity scores Acute Physiology and Chronic Health Evaluation (APACHE)-II, Simplified Acute Physiology Score (SAPS) II, and the indocyanine green plasma disappearance rate (ICG-PDR) were correlated to the argatroban maintenance doses. These diagnostic tools can help to identify patients with the necessity for decreased argatroban doses. The following recommendations for argatroban dosing during CRRT could be determined: a loading dose of 100 &mgr;g/kg followed by a maintenance infusion rate (&mgr;g/kg/min), which can be calculated from the scores as follows: for APACHE II: 2.15–0.06 × APACHE II (r = −.81, p < 0.001); for SAPS II: 2.06–0.03 × SAPS II (r = −.8, p < 0.001); and for ICG-PDR: −0.35 + 0.08 × ICG-PDR (r = .89, p < 0.001). The efficacy and safety of anticoagulation during CRRT were determined by the steady state of blood urea nitrogen (32.16 ± 18.02 mg/dL), mean filter patency at 24 hrs (98%), and the rate of bleeding episodes. Only two patients developed minor bleeding; no patient developed severe bleeding episodes. Conclusion:In critically ill patients with heparin-induced thrombocytopenia type II and necessity for CRRT critical illness scores (APACHE II, SAPS II) or ICG-PDR can help to predict the required argatroban maintenance dose for anticoagulation. These predictors identify decreased argatroban dosing requirements resulting in effective and safe CRRT.

Transcriptional activation of heme oxygenase-1 and its functional significance in acetaminophen-induced hepatitis and hepatocellular injury in the rat.

BACKGROUND/AIM Glutathione depletion contributes to acetaminophen hepatotoxicity and is known to induce the oxidative stress reactant heme oxygenase-1. The metabolites of the heme oxygenase pathway, biliverdin, carbon monoxide, and iron may modulate acetaminophen toxicity. The aim of this study was to assess cell-type specific expression of heme oxygenase-1 and its impact on liver injury and microcirculatory disturbances in a model of acetaminophen-induced hepatitis. METHODS Gene expression of heme oxygenase-1 was studied by Northern- and Western analysis as well as immunohistochemistry. The time course of heme oxygenase-1 and -2, cytokine-induced neutrophil chemoattractant-1, and intercellular adhesion molecule-1 was studied by Northern analysis. DNA-binding activity of nuclear factor-kappaB was determined by electrophoretic mobility shift assay. Sinusoidal perfusion and leukocyte-endothelial interactions were assessed by intravital microscopy. RESULTS Acetaminophen caused a moderate sinusoidal perfusion failure (-15%) and infiltration of neutrophils along with activation of nuclear factor-kappaB and intercellular adhesion molecule-1 and cytokine-induced neutrophil chemoattractant-1 mRNAs. Induction of heme oxygenase-1 mRNA and protein (approximately 30-fold) in hepatocytes and non-parenchymal cells paralleled the inflammatory response. Blockade of heme oxygenase activity with tin-protoporphyrin-IX abrogated acetaminophen-induced hepatic neutrophil accumulation and nuclear factor-kappaB activation, but failed to affect sinusoidal perfusion and liver injury. CONCLUSIONS The inflammatory response associated with acetaminophen hepatotoxicity is modulated by the parallel induction of the heme oxygenase-1 gene. However, heme oxygenase-1 has no permissive effect on sinusoidal perfusion and does not affect liver injury in this model. These data argue against a central role of nuclear factor-kappaB activation and neutrophil infiltration as perpetuating factors of liver injury in acetaminophen toxicity.

Does the timing of tracheal intubation based on neuromuscular monitoring decrease laryngeal injury? A randomized, prospective, controlled trial

Vocal cord injuries (VCI) and postoperative hoarseness (PH) are common complications after general anesthesia. Poor muscle relaxation at the moment of tracheal intubation may result in VCI. There is a large interindividual variation in neuromuscular depression after administration of neuromuscular blocking drugs. Therefore, the optimal individual timing of tracheal intubation based on neuromuscular monitoring (monitoring) may decrease VCI. In this prospective trial, 60 patients were randomized into 2 groups: Monitoring group: tracheal intubation at maximum block based on monitoring after atracurium 0.5 mg/kg and 2-min group: tracheal intubation 2 min after injection of atracurium 0.5 mg/kg. Intubating conditions were evaluated with the Copenhagen score. VCI were examined by stroboscopy before and 24 and 72 h after surgery. PH was assessed at 24, 48, and 72 h after surgery by a standardized interview. Excellent intubating conditions were significantly increased in the monitoring group compared with the 2-min group: 8 versus 2 patients, respectively (P = 0.036). The incidence of PH between the study groups was comparable: 7 (monitoring) versus 8 patients (2-min) (P = 0.860). Similar findings were observed for VCI: 9 versus 5 patients; respectively (P = 0.268); type of VCI: thickening of the vocal cords: 8 (monitoring) versus 5 (2-min) patients (P = 0.423), hematomas: 2 patients in each group (not significant). The present study demonstrated that neuromuscular monitoring improved endotracheal intubating conditions. However, tracheal intubation at maximum intensity of neuromuscular block was not associated with a decrease in vocal cord injuries.

Impact of alloantigens and storage-associated factors on stimulated cytokine response in an in vitro model of blood transfusion.

Background Transfusion of blood may contribute to immunosuppression in major surgery. The authors assessed the impact of alloantigens and storage on function of peripheral blood mononuclear cells cultured in their physiologic environment. Methods Blood units (whole blood, packed erythrocytes) were prepared with or without prestorage leukodepletion and stored for 24–26 days. Blood samples were coincubated with allogeneic fresh blood, autologous, or allogeneic stored blood. Endotoxin-stimulated release of tumor necrosis factor-&agr; (TNF-&agr;) and interleukin 10 (IL-10) was measured after 24 h of culture by enzyme-linked immunosorbent assay. Results Coincubation with equal amounts of allogeneic fresh blood showed almost no influence on TNF-&agr; (−12%, not significant) and IL-10 (+11%, not significant) release. Stored allogeneic whole blood resulted in a significant TNF-&agr; depression (−61%) and IL-10 induction (+221%). These effects were diminished but not prevented by prestorage leukodepletion (TNF-&agr; −42%, IL-10 +110%) and required the presence of soluble factors (TNF-&agr; suppression) and cellular components (IL-10 induction). TNF-&agr; decrease and IL-10 increase were in the same order of magnitude (−40%, +134% with, −65%, +314% without leukodepletion) after coincubation with autologous blood. In contrast, allogeneic erythrocytes had only little effects (TNF-&agr; −6%, IL-10 +36%) even at this high transfusion equivalent. Conclusion These data suggest that banked whole blood has an immunosuppressive effect that is largely attributable to storage-dependent factors. These factors are partially removed by prestorage leukodepletion, while the contribution of alloantigens is of minor significance. Immunosuppressive effects are least apparent with leukodepleted erythrocytes, suggesting that the presence of plasma during storage is required for the immunosuppressive effect to develop.

Differential activation pattern of redox-sensitive transcription factors and stress-inducible dilator systems heme oxygenase-1 and inducible nitric oxide synthase in hemorrhagic and endotoxic shock

ObjectiveTo investigate the role of redox-sensitive transcription factors nuclear factor kappa-B (NF-&kgr;B) or activator protein-1 (AP-1) for hepatic gene expression of heme oxygenase (HO)-1 and inducible nitric oxide synthase (iNOS) in models of hemorrhagic or endotoxic shock. DesignProspective controlled laboratory study. SettingAnimal research laboratory at a university hospital. SubjectsMale Sprague-Dawley rats (250–350 g). InterventionsAfter anesthesia, animals were assigned to hemorrhagic shock (mean arterial pressure 35–40 mm Hg for 60 mins), sham operation, or endotoxemia (1 mg/kg intraperitoneally). To assess the role of reactive oxygen species for activation of NF-&kgr;B or AP-1, animals were treated with the antioxidant trolox (6 mg/kg body weight). In additional experiments, animals were pretreated with dexamethasone (10 mg/kg body weight), an inhibitor of the transactivating function of DNA-bound AP-1 or with actinomycin-D (2 mg/kg body weight), an inhibitor of DNA-directed RNA synthesis. Activation of NF-&kgr;B or AP-1 was assessed by electrophoretic mobility shift assay. HO-1 and iNOS gene expression were assessed by Northern and Western blot. Measurements and Main Results Hemorrhage and resuscitation induced hepatic HO-1 transcripts 12-fold. Induction was abolished by actinomycin-D and was attenuated by dexamethasone and the antioxidant trolox. Activation of AP-1 was observed after hemorrhagic but not after endotoxic shock. AP-1 activation was inhibitable by trolox and correlated with accumulation of HO-1 transcripts. In contrast, a weak activation of NF-&kgr;B was observed after hemorrhage that was not affected by trolox. A profound activation of NF-&kgr;B after endotoxic shock correlated with induction of iNOS but failed to induce HO-1 transcripts. ConclusionsThese data suggest that AP-1 but not NF-&kgr;B activation is dependent on reactive oxygen intermediates in vivo and contributes to HO-1 gene expression. Thus, AP-1-dependent HO-1 induction under oxidative stress conditions may subserve a similar function as a stress-inducible vasodilator system as does NF-&kgr;B-dependent iNOS expression in liver inflammation.

Expression pattern and regulation of heme oxygenase-1/heat shock protein 32 in human liver cells.

Heme oxygenase-1 (HO-1) is a stress response protein that is highly inducible under various conditions, such as oxidative or heat stress. The present study investigated expression pattern and regulation of HO-1 in human liver. Expression pattern of HO-1 immunoreactive protein was studied in liver biopsies by immunohistochemistry, revealing constitutive expression in Kupffer cells but not in hepatocytes. HO-1 was, however, inducible in hepatocytes and vascular tissue under pathological conditions, e.g. associated with fatty degeneration or liver malignancies. Regulation of HO-1 gene expression was further studied by Northern blot analysis in HepG2 cells and freshly isolated peripheral blood mononuclear cells as model systems of parenchymal and nonparenchymal liver cell populations, respectively. HO-1 mRNA was inducible in HepG2 cells and mononuclear cells by various agents inducing oxidative stress. However, HO-1 gene expression was not inducible by heat shock. Pyrrolidine dithiocarbamate, an inhibitor of nuclear factor &kgr;B-dependent gene expression, dose dependently decreased HO-1 mRNA transcripts in human mononuclear cells subjected to oxidative stress while slightly increasing HO-1 gene expression in HepG2 cells. In contrast, HO-1 induction upon oxidative stress was attenuated in HepG2 cells by cycloheximide and dexamethasone. Although activator protein-1 has been reported as the predominant redox-sensitive transcription factor inducing HO-1 expression in murine macrophages, nuclear factor &kgr;B seems to play a significant role in human mononuclear cells. Our data are consistent with a role for activator protein-1in HO-1 induction in human HepG2 hepatoma cells. These data suggest a differential regulation of HO-1 gene expression in parenchymal and non-parenchymal human liver cells and may provide a topographic basis for the understanding of the role of the heme oxygenase/carbon monoxide pathway in human liver disease.

Endotoxin desensitization of human mononuclear cells after cardiopulmonary bypass: role of humoral factors.

Background The ability of leukocytes to release proinflammatory cytokines on lipopolysaccharide stimulation in vitro is impaired after cardiopulmonary bypass (CPB). This study tested contribution and interaction of humoral factors in altered leukocyte responsiveness to lipopolysaccharide. Methods Whole blood and isolated peripheral-blood mononuclear cells (PBMCs) from 10 patients obtained after induction of anesthesia (T1) and 20 min (T2) and 24 h (T3) after CPB were cultured in the absence or presence of lipopolysaccharide and assessed for release of tumor necrosis factor &agr; (TNF-&agr;) and interleukin (IL)-1&bgr; and their functional antagonists, IL-1 receptor antagonist (IL-1ra) and IL-10. In addition, dose–response characteristics and interaction of IL-10 and norepinephrine as modulators of TNF-&agr; release were studied. Results Cardiopulmonary bypass induced release of antiinflammatory (T2: IL-10: median 25 pg/ml, 25th–75th percentile 9–42; IL-1ra: median 1,528 pg/ml, 25th–75th percentile 1,075–17,047;P < 0.05 compared with T1) but failed to induce proinflammatory cytokines (T2: TNF-&agr;: median 0 pg/ml, 25th–75th percentile 0–6; IL-1&bgr;: median 1 pg/ml, 25th–75th percentile 0–81; nonsignificant). Removal of plasma at T2 increased TNF-&agr; response to lipopolysaccharide (+83.8%;P < 0.05), whereas it suppressed IL-10 (−36.8%;P < 0.05). Similarly, incubation of PBMCs (T1) with plasma obtained after CPB (T2) as well as addition of IL-10 or norepinephrine in concentrations present in plasma after CPB led to a reduced lipopolysaccharide-stimulated TNF-&agr; and an increased IL-10 response. Coadministration of norepinephrine and IL-10 had synergistic effects. Although pretreatment with an anti–IL-10 antibody and labetalol before addition of plasma obtained at T2 largely restored the TNF-&agr; response in vitro, their addition post-treatment failed to restore the monocytic TNF-&agr; response. Conclusions Plasma contains interacting factors that inhibit the release of TNF-&agr; and increase the release of IL-10, presumably attenuating the inflammatory response to CPB. Although norepinephrine fails to induce a cytokine response in the absence of other stimuli, its administration seems to augment the antiinflammatory IL-10 response while attenuating the TNF-&agr; response.

Kupffer cells and neutrophils as paracrine regulators of the heme oxygenase-1 gene in hepatocytes after hemorrhagic shock.

Heme oxygenase (HO) plays a pivotal role for the maintenance of liver blood flow and hepatocellular integrity after hemorrhagic shock. We investigated the role of Kupffer cells and neutrophils as paracrine modulators of hepatocellular HO-1 gene expression in a rat model of hemorrhage and resuscitation. Male Sprague-Dawley rats (n = 6-10/group) were anesthetized (pentobarbital, 50 mg/kg intraperitonal) and subjected to hemorrhagic shock (mean arterial blood pressure: 35 mmHg for 60 min) or a sham protocol. Based on the time course of HO-1 gene expression, the effect of various antioxidants, Kupffer cell blockade [gadolinium chloride (GdCl3); 10 mg/kg; 24 h prior to hemorrhage or dichloromethylene diphosphonate (Cl2MDP); 1 mg/kg; 2 days prior to hemorrhage], or neutrophil depletion (vinblastine, 0.5 mg/kg, 5 days prior to hemorrhage) on induction of the HO-1 gene was assessed at 5 h of resuscitation, i.e., the time point of maximal induction. Kupffer cell blockade and antioxidants abolished HO-1 mRNA and protein induction after hemorrhage, while neutrophil depletion failed to affect hepatocellular HO-1 gene expression. In addition, Kupffer cell blockade aggravated hepatocellular injury. N-formyl-methionine-leucyl-phenylalanin (fMLP) induced a substantial influx of neutrophils into the liver but failed to induce hepatocellular HO-1 mRNA expression. These data suggest that Kupffer cells but not neutrophils induce an adaptive hepatocellular stress response after hemorrhage and resuscitation. Oxygen-free radicals released by Kupffer cells may serve as paracrine regulators of a hepatocellular stress gene which is necessary to maintain liver blood flow and integrity under stress conditions.

Monocyte deactivation in severe human sepsis or following cardiopulmonary bypass.

We investigated the specificity for gram-negative stimuli as well as the contribution of signal transduction pathways for leukocyte hyporesponsiveness in sepsis or following cardiopulmonary bypass (CPB). Whole blood of nine patients undergoing CPB and 25 patients with severe sepsis was stimulated ex vivo with LPS (E. coli O111:B4) or with Staphylococcus aureus Cowan strain I (SAC-I) lysate in the absence or presence of inhibitors of protein kinase C (PKC), protein-tyrosine kinase (PTK), or protein-tyrosine phosphatase (PTP). Both toxins stimulated a TNF-&agr; response through PTK signaling. Although suppression of the cytokine response was similar for LPS and SAC-I after CPB, it was significantly more pronounced for SAC-I in sepsis. Inhibition of PTP failed to increase TNF-&agr; upon LPS, whereas a moderate increase was observed with SAC-I. Impaired TNF-&agr; responses occur in sepsis and after CPB. Although this has primarily been reported for gram-negative stimuli, our data suggest that this is even more pronounced for gram-positive stimuli in severe sepsis. Although PTK was the predominant signaling pathway, inhibition of PTP only partially restored the TNF-&agr; response to SAC-I. Our results suggest that cellular mechanisms underlying monocyte deactivation are different in sepsis or following CPB and are discriminate for gram-positive and gram-negative toxins.