Joel B. Cochran

Pneumococcus-induced T-antigen activation in hemolytic uremic syndrome and anemia

The hemolytic uremic syndrome (HUS) is most commonly associated with Escherichia coli, but has been associated with other infections such as Streptococcus pneumoniae. Pneumococcus-induced HUS carries an increased risk of mortality and renal morbidity compared with E. coli-induced HUS. The pneumococcal organism produces an enzyme, which can expose an antigen (T-antigen) present on erythrocytes, platelets, and glomeruli. Antibodies to the T-antigen, normally found in human serum, bind the exposed T-antigen, and the resultant antigen-antibody reaction (T-activation) can lead to HUS and anemia. Clinicians need to be aware to request specific testing when pneumococcus-induced HUS/anemia is suspected, as current blood banking techniques do not routinely test for the presence of the T-antigen. Once this association is documented, washing all blood products and avoiding plasma products, if possible, is recommended. Plasmapheresis can be considered for the more critically ill patient. The incidence of pneumococcus-induced HUS may be increasing. We report six cases of pneumococcus-induced HUS/anemia presenting at our hospital.

Acute liver failure in children.

Objective: To review the incidence, etiologies, pathophysiology, and treatment of acute liver failure (ALF) in children. Emphasis will be placed on the initial management of the multiple organ system involvement of ALF. Method: MEDLINE search from 1970 to March 2005 was performed. Search headings were as follows: acute liver failure, fulminant liver failure, pediatric liver failure, hepatic encephalopathy, and liver transplantation. Studies written in English were selected. Pediatric studies were emphasized. Adult studies were referenced if there were no pediatric studies available in regard to a specific aspect of liver failure. Conclusions: Pediatric acute liver failure is a rare but life-threatening disease. The common etiologies differ for given age groups. Management includes treating specific causes and supporting multiple organ system failure. Commonly associated disorders that require initial recognition and treatment include energy production deficiencies (hypoglycemia), coagulation abnormalities, immune system dysfunctions, encephalopathy, and cerebral edema. Criteria used to determine the need for liver transplant are reviewed as well as the difficulties associated with predicting which patients will meet these criteria and how rapidly liver transplant will become the only option. Finally, experimental procedures that may provide additional time for the liver to recover are briefly reported.

Neonatal Hypersusceptibility to Endotoxin Correlates with Increased Tumor Necrosis Factor Production in Mice

Septic shock is a major cause of mortality in neonates. The hypothesis was tested that neonatal age is associated with altered sensitivity to shock-inducing bacterial products or proinflammatory cytokines (or both). Mice of different ages were inoculated with various doses of lipopolysaccharide (LPS), superantigenic staphylococcal enterotoxin B (SEB), or recombinant tumor necrosis factor-alpha (rTNF-alpha), alone or in combination with the sensitizing agent D-galactosamine. Neonatal mice were markedly more susceptible to LPS-induced lethality but more resistant to SEB than were adults (P < .05). Mice of different ages did not differ, however, in their sensitivity to lethal activities of rTNF-alpha. Neonatal susceptibility to LPS and SEB correlated directly with plasma TNF-alpha but not IFN-gamma levels, which was confirmed by TNF-alpha and IFN-gamma blockade experiments. These data document marked age-related differences in the pathophysiology of septic shock and suggest that IFN-gamma is not an obligatory mediator of either LPS- or SEB-induced lethality in neonates.

Emergency cardiopulmonary bypass for cardiac arrest refractory to pediatric advanced life support

We report the application of emergent cardiopulmonary bypass (CPB) for three pediatric patients in the cardiac catheterization laboratory with cardiac arrest who did not respond to conventional resuscitation efforts. All three patients had return of baseline prearrest rhythms within minutes of the initiation of artificial cardiopulmonary support and the return of spontaneous circulation upon weaning CPB. Two patients had normal neurologic outcomes despite an interval of over 30 minutes from arrest to CPB. The continued judicious application and study of this technology in a small subpopulation of pediatric cardiac arrest patients is warranted.

Protective effect of tyrphostin AG-556 on shock induced by endotoxin or gram positive bacteria.

The effects of tyrphostin AG-556 (TYR), a tyrosine kinase inhibitor, were evaluated on shock induced by lipopolysaccharide (LPS) or group B streptococcus (GBS) in rats. Mortality and mean survival time were monitored. Plasma 6-keto prostaglandin F1alpha (6-keto PGF1alpha) was also measured at four hours after LPS injection. The effects of TYR on the production of 6-keto PGF1alpha thromboxane B2(TXB2) and nitrite (NO) from LPS or GBS stimulated in vitro peritoneal rat macrophage were also examined. Salmonella enteritidis LPS (12 mg/kg, i.v. ) (n=6) produced severe shock (100% mortality). Simultaneous treatment with TYR (n=6) significantly (p < 0.01) extended mean survival time and 33% of rats survived. Plasma 6-keto PGF1alpha concentrations were increased in LPS controls, whereas TYR (5 mg/kg) significantly (p < 0.05) decreased the production. Animals treated with GBS/D-galactosamine (n=9) also exhibited shock with 100% lethality and TYR again prolonged survival time (p < 0.05) with 55% of the animals surviving. To evaluate direct effects of TYR on mediator production induced by LPS or GBS, rat macrophages were stimulated with heat-killed GBS or LPS with or without TYR. Supernatants were collected at 24 h for determination of TXB2, 6-keto PGF1alpha and NO. All mediators measured were significantly increased (p < 0.05) with LPS or GBS. TYR inhibited (p < 0.05) the production of all mediators from macrophages induced by LPS or GBS. The decrease in eicosanoids was associated with a reduction of the content of cyclooxygenase-2 (COX-2) as determined by western blotting. Collectively, these results suggest that TYR ameliorates toxic shock induced by LPS or gram positive bacteria. This protection is associated with suppression of macrophage mediator production.

Central Venous Access Via External Jugular Vein in Children

Objective: To determine the success rate and complications of using the external jugular (EJ) vein for central venous access in pediatric patients. Methods: Prospective cohort study of children who underwent attempts at EJ vein central venous access while receiving care in an 11-bed pediatric intensive care unit at an urban childrens hospital. Results: Over a period of 15 months, 50 patients had EJ venous cannulation attempts. Central venous access was achieved in 45 patients (90%). Successful central venous access was performed in 4 children (50%) younger than 1 year and in 36 older children (98%). Catheter-tip malposition on chest radiograph required subsequent line manipulation in 2 patients. No complications of pneumothorax or carotid artery puncture occurred during line insertion. The catheters were used for an average of 7.5 days (range, 1-28 days). Catheter malfunction occurred in 4 (1.21/100 catheter-days), and catheter-related bloodstream infections occurred in 2 patients (6.04/1000 catheter-days). No thrombotic complications were clinically detected. Conclusions: The EJ vein is a viable site for central venous access with a low complication rate in pediatric patients.

Chloramphenicol toxicity revisited: a 12-year-old patient with a brain abscess.

Chloramphenicol, a broad-spectrum antibiotic, is rarely used in the United States due to its well-described adverse effects. Because of its limited use, many clinicians are unfamiliar with its indications, spectrum of activity, and potential adverse drug effects. We describe a 12-year-old patient who presented after two craniotomies for a persistent brain abscess complicated by long-term chloramphenicol administration. Findings for this patient were consistent with many of the adverse drug effects associated with chloramphenicol, including elevated chloramphenicol serum concentrations, anemia, thrombocytopenia, reticulocytopenia, and severe metabolic acidosis. Rare manifestations of chloramphenicol toxicity that developed in this patient included neutropenia, visual field changes, and peripheral neuropathy. Chloramphenicol administration was discontinued, and hemodialysis was initiated for severe metabolic acidosis. The patient recovered with severe visual field deficits. Although chloramphenicol is rarely indicated, it remains an effective antibiotic. Healthcare providers should become familiar with the pharmacology, toxicology, and monitoring parameters for appropriate use of this antibiotic.

Effect of nitric oxide donors and nitric oxide synthase inhibitors in neonatal rat endotoxic shock

Previous studies have shown an increased mortality in response to endotoxin in 24-hr-old neonatal rats compared with older neonates and adults. This increased susceptibility may be related to increased nitric oxide (NO) and thromboxane (TxB2) production. Twenty-four-hour-old neonatal rat pups were given either N(G)-nitro-L-arginine methyl ester (L-NAME; a nonspecific NO synthase inhibitor), S-methylthioisourea (SMT; a specific NO synthase inhibitor), or molsidomine (a NO donor) subcutaneously prior to or after an LD50 of intracardiac endotoxin. Mortality was followed for 72 hr. There was no statistically significant difference in mortality between control animals and those pretreated with L-NAME, SMT, or molsidomine. A trend toward increased mortality with nonspecific NO synthase inhibition and decreased mortality with the NO donor was noted. Splenic cells were obtained for in vitro cytokine stimulation studies. In vitro adherent splenic cell stimulation studies confirmed an increase in NO production with NO donor pretreatment and decreased production of NO with NO synthase inhibition pretreatment. There was no difference in TxB2 production with either the NO synthase inhibitor or the NO donor. In conclusion, at the several doses employed, neither nonselective or selective NO synthase inhibitors nor NO donors prevented endotoxin-induced mortality in rat neonatal shock. Although these findings do not preclude possible involvement of NO in neonatal pathophysiology, increased NO production thus does not appear to be the primary determinant of the increased susceptibility of the neonatal rat to endotoxic shock.

The effect of a tyrosine kinase inhibitor on endotoxin mortality and splenocyte mediator production in the neonatal rat.

Tyrosine kinases mediate cellular signal transduction to endotoxin. A class of tyrosine kinase inhibitors, the tyrphostins, have been shown to protect mice from endotoxin-induced lethality. Neonatal rats and mice have been shown to be uniquely susceptible to lethal endotoxic shock. In our study, the effect of a lipophilic tyrphostin, AG 556, on endotoxin-induced neonatal and adult mortality and in vitro neonatal splenic cell thromboxane (TxB2), tumor necrosis factor-alpha (TNF-alpha), and nitric oxide (NO) production were examined. Neonatal rats (<24 h old) were administered tyrphostin (100 microg subcutaneous) 2 h before an approximate LD50 dose of Salmonella enteritidis endotoxin (.024 mg/kg/intracardiac). There was a significant decrease in mortality in the animals pretreated with 100 microg of tyrphostin (29% mortality in the treated group, n = 41 versus 53% in the vehicle control group, n = 40; p < .05). Also in adult rats tyrphostin (5 mg/kg intraperitoneal) 2 h before endotoxin (10 mg/kg intravenous) significantly improved survival (50% drug treated versus 84% in control, n = 12/group; p < .05). Adherent neonatal splenic cell mediator production of TxB2, TNF-alpha, and NO (measured by nitrite) in tyrphostin pretreated splenic cells were compared with endotoxin-stimulated splenic cells in vitro. The studies (n = 4) demonstrate an increase (p < .05) in the production of TxB2, TNF-alpha, and NO in the endotoxin- (10 microg/mL) stimulated adherent splenic cells compared with basal. Tyrphostin pretreatment (10, 20, 50 microM) produced a dose-dependent decrease (p < .05) in endotoxin-stimulated TxB2 and TNF-alpha production. NO production was not significantly reduced. In conclusion, tryphostin appears to have a protective effect on mortality in both adult and neonatal rat endotoxic shock. Tyrphostin decreased specific mediator production in stimulated neonatal cells. Thus, inhibition of signal transduction pathways of endotoxin activation by tyrosine kinase inhibition may provide an effective approach to treat endotoxic shock in the neonate.

Endotoxin-induced cross-tolerance to Gram-positive sepsis

The manifestations of Gram-positive sepsis and Gram-negative sepsis share some common clinical features suggesting common pathways of activation. The goal of this study was to assess whether lipopolysaccharide (LPS) can produce cross-tolerance to Gram-positive sepsis induced by group B streptococcus (GBS). Thromboxane (TxB2), tumor necrosis factor (TNFα), and nitric oxide (NO) production by in vitro LPS- and heat killed GBS-stimulated rat peritoneal macrophages were measured. Since our previous studies have demonstrated altered macrophage activation of extracellular signal-regulated kinases 1 and 2 (ERK 1/2) in tolerance, we also examined the effect of LPS and killed GBS on ERK 1/2 activation in normal and LPS tolerant macrophages. Tolerance was induced in rats by intraperitoneal injection of Salmonella enteritidis LPS or vehicle for two consecutive days at doses of 0.1 and 0.5 mg/kg body weight. Three days after the second LPS dose, rats were injected intravenously with viable GBS (5 x l09 cfu/kg) and D-galactosamine (1 g/kg). LPS tolerance significantly prolonged (P <0.05) mean survival time to severe GBS sepsis in D-galactosamine sensitized rats from 12.9 ± 1.7 h in control rats to 44.0 ± 8.9 h in tolerant rats. Peritoneal macrophages from LPS tolerant rats exhibited suppressed LPS induced in vitro TxB2 and TNFα production (P <0.05). Tolerance also decreased in vitro heat killed GBS-induced TNFα production, but did not significantly affect TxB2 production. NO production stimulated by LPS (10 µg/ml was not impaired in LPS tolerance; however at lower doses (0.02—1.25 µg/ml), NO production was significantly decreased (P <0.05). NO production was augmented (P <0.05) in response to stimulation with GBS (10 µg/ml) and unaltered at lower doses (0.02—1.25 µg/ml) in tolerant cells. LPS activated ERK 1/2 in control macrophages, but activation of ERK 1/2 was suppressed in LPS tolerance. GBS did not significantly affect ERK 1/2 activity in control or tolerant macrophages. Nevertheless, the selective mitogen-activated kinase (MAPK)/ERK kinase (MEK) inhibitor, PD 98059 blocked (P <0.05) both GBS- and LPS-induced TNFα and TxB2 production, but not NO production. Thus, some level of ERK 1/2 activity appears essential for GBS- and LPS-induced macrophage activation. In conclusion, LPS tolerance induces partial cross-tolerance to Gram-positive sepsis induced lethality, and alters LPS- and GBS-induced in vitro peritoneal macrophage mediator production. This suggests common pathways of cellular activation for GBS and LPS that are altered by LPS tolerance.