What's New in Shock, August 2011?

The article by Schuetz et al. (1) describes hypotensive patients in the emergency department with sepsis or nonsepsis conditions. It appears in the septic patients that their condition was associated with the appearance of endothelial cell biomarkers, s(soluble) E-selectin and sFLT-1 (the soluble receptor for vascular endothelial growth factor) and that these markers were fairly reliable indicators of sepsis, whereas other endothelial markers (such as intercellular adhesion molecule 1, vascular cell adhesion molecule 1, vascular endothelial growth factor, or plasminogen activator inhibitor 1) were less reliable markers. To what extent the presence of sE-selectin and sFLT-1 can be used as prognostic indicators of sepsis during treatment remains to be determined. The report by Park et al. (2) has a focus on plasma levels of the N-terminal pro-brain natriuretic peptide (NT-proBNP) in patients with acute lung injury or acute respiratory distress syndrome, the clinical presentations being septic shock and sequential organ failure associated with assessment (SOFA) scores. In nonsurvivors, NT-proBNP scores were significantly elevated on days 1 and 2 when compared with levels in survivors. There were parallel changes in 28-day mortality. It is suggested that the combination of NT-proBNP levels and SOFA scores may be of greater prognostic value as compared with SOFA scores alone. It has been known for some time that heart rate variability (HRV) occurs in humans infused with LPS and that this may correlate with the ensuing inflammatory response. In the study by Kox et al. (3), 40 healthy adult humans were infused with LPS. Of this group, 12 subjects were reinfused with LPS 2 weeks later. Somewhat unexpectedly, the HRV response did not correlate with the magnitude of the inflammatory response (plasma cytokines), despite large changes in HRV values. Interestingly, in subjects exposed to a second infusion of LPS, HRV changes were robustly expressed again, but plasma levels of cytokines were reduced. Accordingly, HRV responses do not correlate with the inflammatory responses (plasma cytokine levels) after infusion of LPS into humans. The role of gut-derived, biologically active products and the ability of these factors to trigger a systemic inflammatory response have been a topic of considerable debate for some time. The report by Willoughby et al. (4) deals with studies of 8 patients undergoing elective abdominal aortic resection, during which time there was aortic cross clamping, resulting in bowel ischemia followed by reperfusion. Numerous cytokines were measured in central venous and mesenteric blood plasma samples, before and after aortic cross-clamping and after reperfusion. Neither central nor mesenteric venous blood contained detectable bacterial DNA, although it is not clear if the same would be found in sequential samples of plasma. Neither the central nor the mesenteric plasma samples caused apoptosis or necrosis of human umbilical vein endothelial cell monolayers exposed to these samples, nor was there evidence of upregulation of intercellular adhesion molecule 1 on human umbilical vein endothelial cells, although the plasma samples caused increased expression of RAGE (receptor for advanced glycation products) on these cells. The authors concluded that abdominal aortic resection raised plasma cytokine levels in both mesenteric and central venous blood, but the same samples lacked evidence for the presence of either "toxic endothelial factors" or "gut-derived microorganisms." Obviously, this report needs to be considered in terms relevant to other human conditions. Whether a similar pattern would be found in patients undergoing emergency abdominal aortic resection would be an important subsequent study, but under such conditions, the frequent need for massive blood replacement greatly complicates interpretation of data. On the other hand, the current study adds to increasing doubts about the role of gut bacterial translocations in the setting of shock and sepsis. Smoke inhalation is known to cause fibrin deposition in the lung. In the report by Midde et al. (5), pigs were ventilated and exposed to wood bark smoke, which induced acute lung injury. As expected, extensive fibrin deposition occurred in the alveolar compartment, in small and large airways, and in the subpleural space. When pleural mesothelial cells or alveolar type II cells were exposed to an extract of wood bark smoke, mRNA for plasminogen activator inhibitor 1 was stabilized with attendant enhancement protein expression, presumably due to dissociation from plasminogen activator inhibitor 1 mRNA of 6-phospho-d-gluconate NADP oxidoreductase. These events may relate to extensive intrapulmonary deposition of fibrin occurring after smoke inhalation. In the report of Thakkar et al. (6), acute lung injury developed in mice following hemorrhagic shock and subsequent cecal ligation and puncture (CLP). End points were levels in lung homogenates of TNF-α, monocyte chemotactic protein 1 (CCL2), IL-6, and IL-10. Interventions included none or use of siRNA for FasL, or siRNA for green fluorescent protein. Only the siRNA to FasL was protective (reduction in mediator levels as well as reduced lung levels of myeloperoxidase), provided it was administered intratracheally. Intravenous administration had no measurable effects. These studies suggest that local (intrapulmonary) production of FasL triggers caspase 3 activation, followed by apoptosis of pulmonary cells. The combination of local (flank) thermal injury and smoke inhalation in sheep is well known to induce a variety of complications, especially in lung. The Galveston group of Bartha et al. (7) now describes an increase in blood leukocyte levels of poly(ADP-ribose) polymerase, which is a DNA repair enzyme activated after oxidative stress that causes DNA strand breaks. In this report, it was also shown that the kinetics of poly(ADP-ribose) polymerase activation in leukocytes paralleled the time course for formation of reactive oxygen species. It remains to be seen if similar events occur in humans after thermal injury and smoke inhalation. It is known that acute thermal burn to skin in rodents causes a variety of changes in circulating leukocytes, including the appearance of immature neutrophils (PMNs) and monocytes (iMos). In the current report by Noel et al. (8), immature blood PMNs and monocytes were isolated and assessed by flow cytometry. In the study protocols, cells were either unstimulated or stimulated in vitro with LPS or peptidoglycan, and production of IL-10 was studied. Strikingly, peptidoglycan-stimulated PMNs from burned animals produced nearly eight times more IL-10 than LPS-treated sham PMNs. There were also increases in TNF-α and IL-6 production in burn PMNs activated by peptidoglycan when compared with sham PMN, but the increment was much less than that for IL-10. It is suggested that PMNs might be a significant source of IL-10 after burn injury, whereas the predominant product of iMos appears to be IL-6. Finally, LPS-induced mortality of burn mice was intensified by antibody-induced depletion of PMN and iMos. These data suggest that PMNs might be a major source of IL-10 under the experimental conditions used. The study by Nin et al. (9) used CLP in Sprague-Dawley male rats to assess determinants of vascular dysfunction occurring 48 h after CLP. Traditional pharmacological approaches were used, including contraction of vessels exposed to norepinephrine and endothelial relaxation after exposure to acetylcholine (using aortic wall and aortic rings as well as microvessels). Several interventions were used, including superoxide dismutase, tempol (a superoxide dismutase mimetic and peroxynitrite scavenger), the NO synthase inhibitor (l-NAME), and MNTMPYP (which induces decomposition of peroxynitrite). The data indicate a range of defects in vasculature function following CLP, including impairment of vascular contractility and relaxation, increased presence of 3-nitrotyrosine in aortic and microvascular tissues, oxidized DHE, NOS2, and increased expression of NOS2 and Nox-1. The data also suggested that the defective contractility response of aortic and microvascular rings may be linked to the presence of peroxynitrite. Collectively, such data suggest that the vasculature is profoundly functionally defective following onset of CLP. The report by O'Leary et al. (10) follows accumulating evidence that burn injury to the dorsal skin of mice resulted in enhanced responsiveness of macrophages to in vitro exposure to LPS, accompanied by enhanced mediator production. There was increased production of TNF-α (assessed by flow cytometry) in GR-1+, F4/80+, and CD11b+ splenic macrophages stimulated with LPS after burn injury but not in GR-1− macrophages. When CD4−/− mice were used, expansion of GR-1+ splenic macrophages together with increased TNF-α production after exposure to LPS produced results similar to cells from CD4+/+ mice. However, other studies hinted that CD4+ cells might influence TNF-α production by macrophages. It was suggested that the systemic inflammatory (following burn injury, sepsis, or endotoxemia) may be manipulated to regulate GR-1+ macrophage expansion together with CD4+ T-cell responses, collectively containing an excessive inflammatory response. It is well known that acute renal injury is often accompanied by acute pulmonary dysfunction. In the study by Feltes et al. (11), acute renal injury was induced by ischemia/reperfusion (I/R), and a focus was directed on the responses of isolated pulmonary rat endothelial cells and serum from normal rats, I/R rats, and from rats that had been bilaterally nephrectomized. Serum from I/R or nephrectomized rats but not normal serum induced in pulmonary endothelial cells a variety of proinflammatory genes, such as those for ICAM-1 and E-selectin, tissue plasminogen activator, and plasminogen activator inhibitor 1, as well as RANTES, Fas ligand and IL-6, and others. These sera also induced apoptosis of pulmonary vascular endothelial cells. Such outcomes might provide insights into why acute renal injury is often associated with acute lung injury. The group in Newark [Deitch et al. (12)] has already shown in rats that trauma/hemorrhagic shock (T/HS) followed by reinfusion of heparin-anticoagulated shed blood induces increased lipase activity in mesenteric lymph fluids. This has been interpreted to be linked to in vitro induction of toxicity to endothelial cells. In the current study, similar studies were also done with shed blood using the anticoagulants heparin, fondaparinux sodium (Arixtra), or citrate. Use of Arixtra did not increase lymph lipase levels, and little endothelial cell toxicity occurred with such lymph fluids. On the other hand, when Arixtra-treated rats were subjected to T/HS, they still developed acute lung injury that was prevented by lymph direct ligation. When citrated T/HS rats were used, similar outcomes occurred. The data suggest that lymph lipase is not responsible for the effects of T/HS mesenteric lymph on endothelial cells. There have been numerous efforts to divert metabolism of arachidonic acid away from formation of proinflammatory arachidonate metabolites. In the study by Cao et al. (13), rats were subjected to CLP. Central venous lines were installed before CLP and maintained for the next 5 days. Infusions included normal saline, total parenteral nutrition containing soybean oil emulsion or fish oil as a supplement to parenteral nutrition. Treatment with the fish emulsion increased levels of eicosapentaenoic and docosahexaenoic acids. Rats subjected to CLP receiving the fish oil emulsion also had in spleen increased levels of CD3+ and CD3+CD4+ T cells and increased ratio of CD4+ to CD8+ cells. At the same time, regulatory T cells (CD4+CD25+, Foxp3+) were reduced in animals receiving the fish oil supplement. This therapeutic manipulation also reduced the levels of HMGB-1 and IL-10 in plasma after CLP. Colony-forming unit levels in the peritoneum and in the blood, lung, and spleen were also reduced as a result of the treatment. Survival curves appeared to be improved, although more studies are needed. These data suggest that dietary manipulation can move indicators of sepsis into more favorable directions. The studies of Smith Sonneborn and Rutten (14) focused on rats that were subjected to hemorrhagic shock in the absence of fluid resuscitation and the question of whether survival would be improved when the animals were treated (before and after induction of shock) with AICAR (5-aminoimidazale-4-carboxamide ribonucleoside), which induces adenosine monophosphate-activated protein. Treatment with AICAR extended survival time, reduced heart rate, and lowered lactate levels. Such therapy might be relevant to treatment of humans with hemorrhagic shock. The studies of Namas et al. (15) were focused on the extent to which hypoxia-induced upregulation of the cell death protein BNIP3 might be linked to upregulation of hypoxia-inducible factor 1α (HIF-1α), which is well known to be upregulated under conditions of hypoxia. Primary cultures of mouse hepatocytes were exposed for varying times (1-48 h) to hypoxia, whereas livers were exposed to segmental lobar ischemia followed by reperfusion. In hypoxic hepatocytes, there was induction of the BNIP3 homodimer. However, similar responses in mouse macrophages (RAW 264.7) were not found. With the use of siRNA technology to reduce HIF-1α levels, such treatment did not affect hypoxia-induced elevations of BNIP3 protein, suggesting that HIF-1α is not involved in hypoxia-induced elevations of BNIP3 in hepatocytes exposed to hypoxic conditions.