Endogenous prostaglandin F2 alpha in the hyperdynamic state of severe sepsis in man

Prostaglandin F2α receptor antagonist attenuates LPS-induced systemic inflammatory response in mice

Although it is known that prostaglandin (PG) F_2α level is elevated in the plasma of patients with sepsis, the roles of PGF_2α is still unknown. We aimed to clarify the roles of PGF_2α in the regulation of lipopolysaccharide (LPS)-induced systemic inflammation. At 24 hours after LPS administration, neutrophil infiltration in peritoneal cavity, the mRNA expression of pro-inflammatory cytokines such as tumor necrosis factor-α, interleukin (IL)-1β, IL-6, and macrophage inflammatory protein-2, and tissue damages in lung, liver, and kidney were all increased. Inhibition of FP receptors significantly decreased LPS-induced neutrophil infiltration and lowered the mRNA expression of the pro-inflammatory cytokines. At 6 hour after LPS administration, the level of anti-inflammatory cytokine, IL-10 in peritoneal lavage fluid was higher than that in naïve mice. Inhibition of FP receptors in these mice increased IL-10 level further. Stimulation of isolated peritoneal neutrophils by LPS increased the gene expression of IL-10, which was further increased by AL8810 treatment. Administration of an anti-IL-10 antibody antagonized the AL8810-decreased mRNA expression of pro-inflammatory cytokines and tissue damages. These results indicate that inhibition of FP receptors by AL8810 attenuated LPS-induced systemic inflammation in mice via enhanced IL-10 production.

Influence of heat shock protein 70 and metallothionein induction by zinc-bis-(DL-hydrogenaspartate) on the release of inflammatory mediators in a porcine model of recurrent endotoxemia

The manipulation of stress gene expression by heavy metals provides protection against the lethal effects of endotoxemia in murine models of septic shock. Recent in vitro studies with alveolar macrophages or monocytes show that induction of the stress response in these cells is followed by a decreased liberation of major cytokines [tumor necrosis factor-alpha (TNF alpha) and interleukin-1 (IL-1)] after endotoxin challenge. These findings suggest that the increased resistance to endotoxin in vivo after stress protein induction could be explained by an altered pattern of inflammatory mediator release. Therefore, we measured the time course of thromboxane-B2 (TxB2), 6-keto-PGF1 alpha, platelet activating factor (PAF), TNF alpha, interleukin-1 beta (IL-1 beta), and interleukin-6 (IL-6) formation with and without induction of the stress response in an established porcine model of recurrent endotoxemia (Klosterhalfen et al., Biochem Pharmacol 43: 2103-2109, 1992). Induction of the stress response was done by a pretreatment with Zn2+ (25 mg/kg zinc-bis-(DL-hydrogenasparate = 5 mg/kg Zn2+). Pretreatment with Zn2+ prior to lipopolysaccharide (LPS) infusion induced an increased heat shock protein 70 and metallothionein expression in the lungs, liver, and kidneys and increased plasma levels of TNF alpha, IL-1 beta, IL-6, and TxB2 as opposed to untreated controls. After LPS infusion, however, pretreated animals showed significantly decreased peak plasma levels of all mediators as opposed to the untreated group. The time course of mediator release was identical with the decreasing and increasing three peak profiles described previously. Hemodynamic data presented significantly decreased peak pulmonary artery pressures and significantly altered hypodynamic/hyperdynamic cardiac output levels in the pretreated group. In conclusion, the data show that the induction of stress proteins by Zn2+ could be a practicable strategy to prevent sepsis.

Concentrations of prolactin and prostaglandins during and after cardiopulmonary resuscitation

OBJECTIVES

To assess differences in plasma prolactin and prostaglandin concentrations in resuscitated and nonresuscitated patients during cardiopulmonary resuscitation (CPR), and to compare changes of prostaglandin and prolactin concentrations with hemodynamic variables in the immediate postresuscitation phase.

DESIGN

Prospective, descriptive study.

SETTING

Emergency medical service at a university hospital.

PATIENTS

Twenty-nine patients ranging in age from 39 to 87 yrs with out-of-hospital cardiac arrest.

INTERVENTIONS

Venous blood samples were taken during CPR and at 5, 15, 30, and 60 mins after restoration of spontaneous circulation in order to measure plasma concentrations of prolactin, prostaglandin F2 alpha, 15-keto-13,14-dihydro-prostaglandin F2 alpha, 6-keto-prostaglandin F1 alpha, and thromboxane B2 by immunoassay. Heart rate and blood pressure were measured at 5, 15, 30, and 60 mins after restoration of spontaneous circulation.

MEASUREMENTS AND MAIN RESULTS

In 15 patients, restoration of spontaneous circulation was achieved; in the remaining 14 patients, successful resuscitation was not possible. During CPR, the mean plasma prolactin, prostaglandin F2 alpha, 15-keto-13,14-dihydro-prostaglandin F2 alpha, 6-keto-prostaglandin F1 alpha, and thromboxane B2 concentrations were 95.9 +/- 13.6 micrograms/L, 357 +/- 61 ng/L, 228 +/- 28 ng/L, 277 +/- 66 ng/L, and 375 +/- 78 ng/L, respectively, in resuscitated patients, and 23.9 +/- 5.6 micrograms/L (p = .0001), 192 +/- 22 ng/L (p = .005), 202 +/- 31 ng/L (p = .528), 221 +/- 40 ng/L (p = .713), and 344 +/- 77 ng/L (p = .780), respectively, in nonresuscitated patients. At 60 mins after restoration of spontaneous circulation, the mean plasma prolactin, prostaglandin F2 alpha, 15-keto-13,14-dihydro-prostaglandin F2 alpha, 6-keto-prostaglandin F1 alpha, and thromboxane B2 concentrations were 50.1 +/- 9.5 micrograms/L, 306 +/- 42 ng/L, 503 +/- 87 ng/L, 278 +/- 55 ng/L, and 355 +/- 30 ng/L, respectively. Mean values of systolic arterial blood pressure were 114 +/- 12 mm Hg at 30 mins and 123 +/- 18 mm Hg at 60 mins. No significant correlations were found between hemodynamic values and plasma concentrations of prolactin or prostaglandins.

CONCLUSIONS

Prolactin and prostaglandin concentrations were increased during cardiac arrest and CPR. Successful initial resuscitation was associated with increased prolactin and prostaglandin F2 alpha concentrations during CPR. Decreased concentrations in non-resuscitated patients may have been a result of exhaustion of the neuroendocrine and eicosanoid systems, or may be due to differences in bioavailability at the site of blood sampling based upon differences in hemodynamics.

Urine hydrogen peroxide during adult respiratory distress syndrome in patients with and without sepsis

BACKGROUND

The lung injury in adult respiratory distress syndrome (ARDS) has been associated with increased expiratory hydrogen peroxide (H2O2) concentrations. Furthermore, patients with sepsis and ARDS are reported to have greater serum scavenging of H2O2 than patients with ARDS only. We hypothesized that the systemic presence of H2O2 would be detectable in the urine of these two groups of patients and that, in the case of ARDS sepsis, the relative contribution of each disease to the production this analyte would be discernible. Accordingly, we used an in vitro radioisotope assay to follow the weekly course of urine H2O2 levels in ARDS patients with and without sepsis, and in samples from control non-ARDS patients with sepsis with indwelling urinary catheters and in samples provided by healthy volunteers.

METHODS

Thirty patients with ARDS were included in the study: 23 had sepsis and 7 were sepsis free. An indwelling catheter was used to collect urine from each patient over a 24-h period, first within 48 h of ICU admission and then every seventh day over the course of their illness. Urine H2O2 was measured by competitive decarboxylation of 1-14C-alpha-ketoglutaric acid by H2O2. Urine samples were provided by 20 healthy volunteers while, in 10 non-ARDS patients with sepsis, urine was collected over one 24-h period following a 5-day minimum with an indwelling urinary catheter.

RESULTS

Urine H2O2 concentration in healthy control subjects (88 +/- 4 mumol/L) and non-ARDS patients with urinary catheters (96 +/- 5 mumol/L) was not significantly different. During the first 48 h in the ICU, urine H2O2 in patients with ARDS only (295 +/- 29 mumol/L) was significantly lower (p < 0.05) than patients with ARDS and sepsis (380 +/- 13 mumol/L); however, the lung injury scores of these two groups did not differ. Furthermore, within the first 48 h, the urine H2O2 of the patients with ARDS and sepsis who did not survive (427 +/- 19 mumol/L; n = 7) was significantly higher than that in patients who survived sepsis (352 +/- 14 mumol/L; n = 15). Thereafter, the lung injury scores and urine H2O2 levels of the nonsurvivor ARDS-sepsis group remained significantly higher compared with the other two groups. At lung injury scores of 3 and 2, regardless of days in ICU, the patients with ARDS only had significantly lower urine H2O2 (266 +/- 30 mumol/L and 167 +/- 24 mumol/L, respectively) compared with the survivor ARDS-sepsis group (376 +/- 19 mumol/L and 250 +/- mumol/L). When the patients with ARDS (both ARDS only and with sepsis) recovered, their urine H2O2 concentration did not differ from the control groups (healthy donors and patients without ARDS).

CONCLUSION

Lung injury scores did not differentiate patients with ARDS and sepsis from patients with ARDS only during the first 10 days in the ICU; however, urine H2O2 levels were significantly greater in the patients with ARDS and sepsis. Moreover, despite no initial difference in lung injury, patients who did not survive ARDS and sepsis had consistently greater urine H2O2 concentration than patients who survived sepsis. The urine H2O2 level in the ARDS-only group was about 70 percent of the level in the survivor ARDS and sepsis group, suggesting that ARDS alone is the major contributor to the H2O2 oxidant processes during combined ARDS and sepsis. Furthermore, these studies demonstrate that urine H2O2 may be a useful analyte to differentiate the severity of oxidant processes in patients with ARDS and sepsis albeit the prognosis appears to be survival or nonsurvival.

Ibuprofen intervention in canine septic shock: reduction of pathophysiology without decreased cytokines

This study was undertaken to evaluate the effect of a cyclooxygenase inhibitor, ibuprofen, at various time intervals in a live Escherichia coli model of canine septic shock. Group I (control) animals (n = 5) received a LD100 dose of 10(9) live E. coli per kilogram were given no further treatment. Group II animals (n = 5) received a 10 mg/kg bolus of ibuprofen 10 min prior to bacterial infusion. Group III animals (n = 5) received ibuprofen 15 min after the bacterial infusion. Statistical analysis revealed the following: Group II animals had significantly higher MABP and significantly lower levels of serum fluorescent products (superoxide radical activity), plasma thromboxane B2, prostaglandin E2, and endotoxin levels compared to Group I animals (P less than 0.05). Plasma levels of tumor necrosis factor (TNF) and interleukin-6 (IL-6) were significantly elevated (P less than 0.05) from baseline in all animals (Groups I, II, and III), but ibuprofen treatment failed to either increase or decrease these levels. This study demonstrates that ibuprofen treatment can significantly reverse the deleterious hemodynamic and metabolic effects commonly seen in live E. coli septic shock without depressing the endogenous production of TNF or IL-6. These data support the hypothesis that sepsis initiates a cascade of mediators with the cytokines TNF and IL-6 being proximal events which in turn stimulate the next level, with ibuprofen probably exerting its inhibitory effect distal to this point in the cascade.

Time course of various inflammatory mediators during recurrent endotoxemia

The time course of thromboxane B2 (TxB2), 6-keto-PGF1 alpha (stable metabolite of prostacyclin), tumor necrosis factor-alpha (TNF alpha), platelet activating factor (PAF), and interleukin-6 (IL-6) formation after three lipopolysaccharide (LPS) infusions was studied in pigs over an 18-hr, period. The Escherichia coli endotoxin W0111:B4 was injected i.v. into 10 of the test group pigs at a dose of 0.5 micrograms/kg over 30 min at 0, 5 and 10 hr of the experiment. Three pigs injected with physiological saline served as controls. At defined time points before and after each LPS administration venous blood was withdrawn (0, 15, 30, 45, 60, 120, 180 min) and plasma levels of TxB2, 6-keto-PGF 1 alpha, PAF, TNF alpha and IL-6 were determined. Pulmonary artery pressure (PAP) and cardiac output (CO) were measured every 15 min. TxB2 and PAF peaked significantly between 30 and 45 min, TNF alpha and 6-keto-PGF 1 alpha between 30 and 60 min, and IL-6 between 120 and 180 min after each LPS injection. The mediators PAF, TNF alpha and TxB2 showed a decreasing three-peak profile whereas 6-keto-PGF1 alpha exhibited an increasing one. IL-6 plasma concentrations increased after each LPS injection. The peak after the third LPS administration, however, was surprisingly low compared to the previous two. The first LPS infusion in our test group led to a significant, sustained rise in mean PAP. After recurrent LPS injections the peak in PAP was not as marked as after the first infusion, indicating the development of a tolerance towards LPS. Initially, CO showed hypodynamic values, whereas the end stage of the experiment was characterized by hyperdynamic CO levels. In conclusion, we believe this porcine model of septic shock to be one of the first large animal models to describe in detail the time-course of various important inflammatory mediators.

Multiple Organ Failure Syndrome—Part I: Epidemiology, Prognosis, and Pathophysiology

The multiple organ failure syndrome (MOFS) is the leading cause of death in intensive care units. Although sepsis is an important cause of MOFS, it is clear that MOFS can occur in the absence of infection. The pathophysiology of MOFS is complex and multifactorial and includes derangements in oxygen delivery and consumption, the release of inflammatory and vasoactive mediators capable of inflicting tissue damage, and alterations in the barrier function of the intestinal mucosa. Although advances have been made in our understanding of MOFS, treatment remains nonspecific and largely supportive. Early and aggressive restoration of tissue perfusion, adequate treatment of infection, timely nutritional support, and support of individual failed organs remain the mainstay of therapy. Therapeutic agents directed against the various mediators associated with the pathophysiology of MOFS may prove useful in the future.

Modification by steroids of pulmonary oedema and prostaglandin E2 pharmacokinetics induced by endotoxin in rats

1. A single i.p. injection of bacterial endotoxin in rats (3.5 mg kg-1) caused lung injury assessed as changes in lung dry:wet weight ratio and leukopaenia over the subsequent 28 h. 2. This treatment also slowed the efflux of 14C from [14C]-prostaglandin E2 (PGE2), i.e., increased t1/2 and increased the survival of PGE2 in isolated perfused lungs over the same period. 3. These effects of endotoxin were reversed by methylprednisolone (30 mg kg-1), given 30 min after the endotoxin. 4. Another synthetic corticosteroid, budesonide (1.2 mg kg-1) given 1 h before endotoxin partially prevented the lung injury and leukopaenia but did not affect the increased t1/2 for PGE2 nor its survival. 5. The reversal by methylprednisolone of both the physical signs of lung injury and the changes in PGE2 pharmacokinetics caused by endotoxin suggests that changes in PGE2 pharmacokinetics could serve as an index of acute lung injury following sepsis.

[Endotoxin and eicosanoids--trigger function, endogenous and therapeutic elimination in 120 surgical patients]

In patients with diffuse bacterial peritonitis and septic shock a correlation is found between endotoxin and certain eicosanoids and of both factors to the clinical degree of severity of the disease. Respiratory dysfunction secondary to septic shock is associated with an impairment of endogenous prostaglandin elimination in favour of pathological pulmonary de novo synthesis. Therapeutical elimination of endotoxin and eicosanoids can be effectively achieved by continuous intraperitoneal lavage during and after surgery for the septic focus.

The clinical significance of prostaglandins and thromboxane as mediators of septic shock

An evaluation was made of 106 surgical patients with Gram-negative septic shock, both for clinical criteria as well as the biochemical mediators endotoxin, prostaglandin F2 alpha, prostaglandin I2 (prostacyclin), and thromboxane. These data were correlated to various defined shock phases, functional data of vital organs, and clinical outcome. Patients underwent invasive organ function monitoring and the usual laboratory tests of intensive care. Prostaglandins and thromboxane were measured radioimmunologically, endotoxin by the limulus amebocyte lysate test. Endotoxin proved to be a more accurate predictor of severe sepsis than did positive blood cultures. Endotoxin as well as prostaglandins and thromboxane are predominantly released in early shock phases, appearing in plasma concentrations, which correlate with the severity of organ failure. Sepsis-induced respiratory failure coincides with a deterioration of pulmonary prostaglandin inactivation, which contributes to the release mechanism. High systemic prostacyclin activity benefits the patients' organ functions and clinical outcomes, while a predominance of thromboxane seems to effect the opposite. Transpulmonary-thromboxane gradients correlate significantly with pulmonary hypertension in the early phases of septic shock.

Role of thromboxane, prostaglandins and leukotrienes in endotoxic and septic shock

Intravenous bolus endotoxin elicits a marked but transient increase in plasma TxB2 and 6-keto-PGF1 alpha in a large number of species. A smaller, delayed and more prolonged increase in TxB2 and 6-keto-PGF1 alpha are reported in animals with septic shock, i.e., those with fecal peritonitis or cecal ligation. Thromboxane synthetase inhibitors or antagonists attenuate endotoxin-induced acute cardiopulmonary changes, the delayed increase in serum lysosomal enzymes, fibrin/fibrinogen degradation products and the thrombocytopenia in a number of species. While these drugs increase survival of rats or mice following endotoxin they do not alter survival of rats in septic shock. These results support the hypothesis that TxA2 exerts a pathophysiologic effect in shock following bolus endotoxin. In contrast, nonsteroidal antiinflammatory drugs (NSAID) and dietary essential fatty acid deficiency increase survival of rats subjected to endotoxin shock, and survival time in models of septic shock. These results also suggest that some other cyclooxygenase product(s) is involved in septic shock due to fecal peritonitis or cecal ligation. Preliminary experimental studies indicate salutary effects of leukotriene inhibitors and antagonists in endotoxin shock and in models of acute pulmonary injury. Clinical studies have demonstrated elevated plasma TxB2 and 6-keo-PGF1 alpha concentrations in patients with septic shock, and elevated LTD4 in pulmonary edema fluid of patients with the adult respiratory distress syndrome. In view of these clinical and experimental results, clinical trials of NSAID and/or leukotriene inhibitors/antagonists should be considered.