Reappraisal of the role of endotoxin in the sepsis syndrome

Towards clinical applications of anti-endotoxin antibodies; a re-appraisal of the disconnect

Endotoxin is a potent mediator of a broad range of patho-physiological effects in humans. It is present in all Gram negative (GN) bacteria. It would be expected that anti-endotoxin therapies, whether antibody based or not, would have an important adjuvant therapeutic role along with antibiotics and other supportive therapies for GN infections. Indeed there is an extensive literature relating to both pre-clinical and clinical studies of anti-endotoxin antibodies. However, the extent of disconnect between the generally successful pre-clinical studies versus the failures of the numerous large clinical trials of antibody based and other anti-endotoxin therapies is under-appreciated and unexplained. Seeking a reconciliation of this disconnect is not an abstract academic question as clinical trials of interventions to reduce levels of endotoxemia levels are ongoing. The aim of this review is to examine new insights into the complex relationship between endotoxemia and sepsis in an attempt to bridge this disconnect. Several new factors to consider in this reappraisal include the frequency and types of GN bacteremia and the underlying mortality risk in the various study populations. For a range of reasons, endotoxemia can no longer be considered as a single entity. There are old clinical trials which warrant a re-appraisal in light of these recent advances in the understanding of the structure-function relationship of endotoxin. Fundamentally however, the disconnect not only remains, it has enlarged.

Prognostic value of endotoxemia in patients with Gram-negative bacteremia is bacterial species dependent

The prognostic impact of endotoxemia detection in sepsis is unclear. Endotoxemia is detectable in <70% of patients with Gram-negative (GN) bacteremias. Mortality proportion data were available from 27 published studies of patients with GN bacteremia in various settings. Among ten studies restricted to specific types of GN bacteremia, endotoxemia was associated with significantly increased mortality risk for Neisseria meningitidis (4 studies; 138 bacteremias; OR 26.0; 95% CI, 1.6-321) but not for Salmonella enterica serovar Typhi (3 studies; 36 bacteremias; OR 0.89; 95% CI, 0.01-74.1). For 17 unrestricted studies (319 GN bacteremic patients), endotoxemia was associated with an increased mortality risk with non-Escherichia coli Enterobacteriaceae such as Klebsiella and Enterobacter species (97 bacteremias; OR 3.7; 95% CI, 1.3-10.3). By contrast, E. coli (144 bacteremias; OR 0.78; 95% CI, 0.36-1.7), and non-Enterobacteriaceae species such as Pseudomonas species (78 bacteremias; OR 1.7; 95% CI, 0.7-4.6) had no increased mortality risk. That endotoxemia detection is predictive of mortality among patients bacteremic with non-E. coli Enterobacteriaceae but not E. coli is surprising given the presumed commonality of the hexa-acyl lipid A structure among Enterobacteriaceae species.

Endotoxemia and mortality prediction in ICU and other settings: underlying risk and co-detection of gram negative bacteremia are confounders

INTRODUCTION

The interdependence between endotoxemia, gram negative (GN) bacteremia and mortality has been extensively studied. Underlying patient risk and GN bacteremia types are possible confounders of the relationship.

METHODS

Published studies with ≥ 10 patients in either ICU or non-ICU settings, endotoxemia detection by limulus assay, reporting mortality proportions and ≥ 1 GN bacteremia were included. Summary odds ratios (OR) for mortality were derived across all studies by meta-analysis for the following contrasts: sub-groups with either endotoxemia (group three), GN bacteremia (group two) or both (group one) each versus the group with neither detected (group four; reference group). The mortality proportion for group four is the proxy measure of study level risk within L'Abbé plots.

RESULTS

Thirty-five studies were found. Among nine studies in an ICU setting, the OR for mortality was borderline (OR <2) or non-significantly increased for groups two (GN bacteremia alone) and three (endotoxemia alone) and patient group one (GN bacteremia and endotoxemia co-detected) each versus patient group four (neither endotoxemia nor GN bacteremia detected). The ORs were markedly higher for group one versus group four (OR 6.9; 95% confidence interval (CI), 4.4 -to 11.0 when derived from non-ICU studies. The distributions of Pseudomonas aeruginosa and Escherichia coli bacteremias among groups one versus two are significantly unequal.

CONCLUSIONS

The co-detection of GN bacteremia and endotoxemia is predictive of increased mortality risk versus the detection of neither but only in studies undertaken in a non-ICU setting. Variation in GN bacteremia species types and underlying risk are likely unrecognized confounders in the individual studies.

Pyrogenic reactions and hemorrhage associated with intrinsic exposure to endotoxin-contaminated intravenous solutions

OBJECTIVE

An epidemiological investigation was conducted to determine risk factors for adverse reactions among patients in hospitals and the possibility of extrinsic or intrinsic contamination of intravenous solutions.

DESIGN

A retrospective cohort study was conducted to identify solutions associated with adverse reactions. Implicated lots were cultured for bacteria, and endotoxin concentrations were measured.

SETTING

Five hospitals in the state of Pernambuco, Brazil, were investigated from February through March 2002.

PATIENTS

Surgical inpatients or outpatients receiving intravenous solutions during the study period.

RESULTS

Of 355 surgical patients or outpatients treated at hospitals, 28 (8%) developed illness within a mean of 2.5 hours after exposure to intravenous solutions implicated in adverse reactions; 5 (17.9%) of the case patients died. Laboratory testing of bottles from the lots of Ringer's lactate solution implicated in deaths demonstrated a high mean endotoxin concentration of 88.3 endotoxin units (EU)/mL (range, 9.7-298.0 EU/mL), compared with the permitted limit in Brazil of <0.5 EU/mL. Testing of metronidazole implicated in adverse reactions at another hospital and produced by the same company that manufactured the lots Ringer's lactate solution also showed high endotoxin concentrations (mean level, 8.3 EU/mL [range, 5.0-58.3 EU/mL]). The outbreak was controlled after a national recall of the implicated brand of intravenous solutions.

CONCLUSIONS

Case patient status was associated with use of Ringer's lactate solution and metronidazole from large bottles, both of which were produced by the same company. High endotoxin concentrations were demonstrated in unopened bottles of implicated products, which is consistent with intrinsic contamination. The high mortality rate may have been compounded by the fact that clinicians administered additional volumes of contaminated 0.9% isotonic sodium chloride solution in response to hypotension or bleeding to some surgical patients. No additional case patients were identified after a national recall of products, inspection, closure of the implicated company's manufacturing facility, and establishment of random quality-control testing of intravenous solutions.

Preoperative oral application of immunoglobulin-enriched colostrum milk and mediator response during abdominal surgery

Our objective was to evaluate the influence of pre-operative oral application of an immunoglobulin-enriched milk preparation on endotoxin translocation and mediator release during and after abdominal surgery. Forty patients who had been treated by partial (n = 4) or total gastrectomy (n = 8) or pancreatic resection (n = 28) were enrolled in a placebo-controlled pilot study. Pre-operatively, patients were randomly treated for 3 days by oral application of a bovine milk preparation (lactobin 56g/day, n = 20) or placebo (n = 20). In both groups, endotoxin translocation and mediator release was studied pre- and intraoperatively by measuring endotoxin, endotoxin-neutralizing capacity (ENC), interleukin 6, C-reactive protein, transferrin, alpha-2-macroglobulin, albumin, apoliprotein-A1/-B, IgG, IgA, and IgM. The clinical course was followed up by daily evaluation of the Apache-II-score. Clinical data were comparable in both groups. The lactobin group showed significantly lower levels of endotoxin and ENC compared to the placebo group. Acute phase response, endotoxin-binding proteins, and clinical outcome did not differ between both groups. We conclude that prophylactic oral application of lactobin reduces perioperative endotoxemia and prevents reduction of ENC, suggesting a stabilization of gut barrier during abdominal surgery.

New criteria for selecting the proper antimicrobial chemotherapy for severe sepsis and septic shock

The mortality rate resulting from severe bacterial sepsis, particularly that associated with shock, still approaches 50% in spite of appropriate antimicrobial therapy and optimum supportive care. Bacterial endotoxins that are part of the cell wall are one of the cofactors in the pathogenesis of sepsis and septic shock and are often induced by antimicrobial chemotherapy even if it is administered rationally. Not all antimicrobial agents are equally capable of inducing septic shock; this is dependant on their mechanism of action rather than on the causative pathogen species. The quantity of endotoxin released depends on the drug dose and whether filaments or spheroplast formation predominates. Some antibiotics such as carbapenems, ceftriaxone, cefepime, glycopeptides, aminoglycosides and quinolones do not have the propensity to provoke septic shock because their rapid bactericidal activity induces mainly spheroplast or fragile spheroplast-like bacterial forms.

Antibiotic-induced release of bacterial cell wall components in the pathogenesis of sepsis and septic shock: a review

This article reviews the new criteria for selecting the proper antimicrobial agent and dosage regimen for standard treatment of severe sepsis, with the intention of preventing septic shock. After introducing new concepts on the pathogenesis of sepsis and septic shock, the authors analyze the parameters of betalactam antibacterial activity, the antibiotic-induced release of bacterial endotoxin and the interrelationships between pharmacokinetics and pharmacodynamics of antibiotics in the search for an optimum dosage regimen of antimicrobial mono- or polytherapy for severely ill septic patients admitted to the intensive care unit.

Continuous antibiotic treatment for experimental abdominal sepsis: effects on organ inflammatory cytokine expression and neutrophil sequestration

BACKGROUND

Tumour necrosis factor (TNF) alpha and interleukin (IL) 1 beta are produced in the lung after peritonitis and may contribute to neutrophil-mediated organ injury. It was hypothesized that, during experimental peritonitis, continuous rather than intermittent antibiotic therapy would reduce lung expression of TNF-alpha and IL-1 beta messenger RNA (mRNA) and neutrophil sequestration.

METHODS

After caecal ligation and puncture, mice received either intermittent or continuous cefoxitin, or continuous metronidazole or aztreonam. Cytokine mRNAs were determined by reverse transcription differential polymerase chain reaction and lung neutrophil content by myeloperoxidase (MPO) assay.

RESULTS

Continuous cefoxitin reduced median (interquartile range (i.q.r.)) lung IL-1 beta mRNA expression ((ratio to beta-actin): continuous 0.18 (0.14-0.34), intermittent 0.46 (0.44-0.49), saline 0.43 (0.38-0.53), P < 0.05) and median (i.q.r.) lung MPO content (continuous 22.5 (9.7-40), intermittent 65 (57.5-76), saline 47 (41-64), P < 0.05) compared with intermittent therapy and saline controls. Continuous infusion was also associated with reduced bacteraemia (P < 0.05) but not serum TNF-alpha or endotoxin levels. Both continuous metronidazole and aztreonam reduced lung MPO concentration (P < 0.05) and TNF-alpha and IL-1 beta mRNA expression (P < 0.05) compared with those in saline controls. These effects were dependent on a reduction in the number of susceptible bacteria rather than serum TNF-alpha or endotoxin levels.

CONCLUSION

The stimulus for organ inflammatory cytokine production and neutrophil sequestration during peritonitis is the level of bacteraemia present, which is more effectively controlled with continuous antibiotic therapy.

The pulmonary inflammatory response to experimental fecal peritonitis: relative roles of tumor necrosis factor-alpha and endotoxin

The roles of endotoxin (LPS) and tumor necrosis factor-alpha (TNF-alpha) in the causation of organ injury during sepsis are unclear. To study LPS and TNF-alpha in the genesis of lung inflammation after cecal ligation and puncture (CLP), we used endotoxin-resistant (C3H/HeJ) and endotoxin-sensitive mice (C3H/HeOuJ). We examined lung neutrophil sequestration, interleukin 1 (IL-1)beta mRNA expression, IL-1 beta protein expression, and injury. We also determined the expression of two C-X-C chemokine mRNAs, macrophage inflammatory protein-2 (MIP-2) and KC, in the lung to determine whether in vivo, endotoxin, or TNF-alpha are significant modulators of MIP-2 and KC mRNA expression. After CLP, increased neutrophils sequestrated in the lungs of both strains of mice and coincided with an increase in expression of IL-1 beta, MIP-2 and KC mRNAs, and IL-1 beta protein. Lung and serum TNF-alpha were significantly increased in the C3H/HeOuJ strain but not in the C3H/HeJ strain. Histologic studies of the lung revealed similar injury in both strains. Our results suggest that bacterial factors other than endotoxin cause lung neutrophil sequestration and injury after CLP and, further, that TNF-alpha production is not a prerequisite. Our findings also suggest a potential role for local pulmonary chemokine production in the control of neutrophil sequestration after CLP.

Effect of endotoxin on gentamicin pharmacokinetics in old and young adult rats

The effect of endotoxin administration on gentamicin pharmacokinetics in young adult (2-3 months) and old (22-24 months) rats was studied. Gentamicin (3 mg/kg, iv) was administered 24 hours after an endotoxin challenge (5 mg/kg, ip). Some blood biochemical parameters, viz. urea, AST, GGT activities in addition to PCV and Hb concentration and creatinine clearance were also measured. In young animals, endotoxin caused prolongation in gentamicin half life (t1/2), increased area under the plasma concentration-time curve (AUC) and reduced total body clearance (ClB) and volume of distribution (Vd). Endotoxin effects in the old rats were qualitatively similar to those induced in the young but were more pronounced. They included more than 10 fold increase in the t1/2 and AUC. In addition, a rising early phase in gentamicin plasma concentration was noticed in old rats treated with endotoxin which was, probably, due to an early redistribution process of gentamicin. The results indicate that aging and endotoxin, individually, can significantly alter gentamicin pharmacokinetics in the rat. These alterations were exacerbated when endotoxemia was induced in old rats.

Microbial/host interactions in health and disease: who controls the cytokine network?

The interacting cellular and molecular systems which we classify as immunity and inflammation evolved to protect the organism from exogenous parasites including viruses and bacteria. Cytokines play a pivotal, but paradoxical, role both in immunity and inflammation. These local peptide hormone-like molecules form a major arm of the organisms, defenses against infectious microorganisms but they are also implicated as potent mediators of the pathology of infectious diseases. The apparently lethal effects of interleukin-1 and tumor necrosis factor in experimental septic shock testify to the latter. In the current paradigm, cytokine induction, as a protective or pathological mechanism, is a direct response to the presence of infectious microorganisms. Evidence is now accumulating that cytokines play a much more complex role in the interplay between exogenous microorganisms and the host. For example, it has been established that viruses have evolved pro-active methods of subverting the cytokine network by producing: (i) soluble cytokine receptors which bind and inactivate cytokines, (ii) immunomodulatory cytokine homologues, and (iii) ICE inhibitors. The possibility exists that the major role of these 'viral cytokines' is to neutralize certain host responses. Recent cytokine transgenic knockouts demonstrate that the normal benign response to commensal gut microflora becomes a lethal inflammatory state in the absence of the cytokines interleukin 2 or interleukin 10. The human body contains an enormous number of microorganisms which constitute the normal microflora. It is estimated that the average human contains 10(13) eukaryotic cells but 10(14) bacteria. We propose that the ability of the multicellular organism to live harmoniously with its commensal microflora must depend on mutual signalling involving eukaryotic cytokines and prokaryotic cytokine-like molecules. Such interactive signalling sets up non-inflammatory cytokine networks in tissues which form the background on which responses to infectious microorganisms must be built and related. The capacity of bacteria to induce cytokine synthesis was believed to be due to a small number of components, such as lipopolysaccharide (LPS), which is only active as a complex with host factors (lipopolysaccharide binding protein and CD14). However, it is now clear that bacteria contain and produce a large number of diverse molecules which can selectively induce the synthesis of both pro-inflammatory and immunomodulatory/anti-inflammatory cytokines. Many toxins are potent inducers of cytokine release or synthesis and some can inhibit LPS-induced cell activation. We have introduced the term bacteriokine to describe these bacterial cytokine inducers. The question that has to be addressed therefore is - who controls the cytokine network (eukaryotic or prokaryotic cells) and how is it controlled? It is proposed that an understanding of this question will bring with it an understanding of how to control the pathological inflammatory response and may allow the development of truly effective anti-inflammatory agents.

Advanced and controlled drug delivery systems in clinical disease management

Advanced and controlled drug delivery systems are important for clinical disease management. In this review the most important new systems which have reached clinical application are highlighted. Microbiologically controlled drug delivery is important for gastrointestinal diseases like ulcerative colitis and distally localized Crohn's disease. In cardiology the more classic controlled release systems have improved patient compliance and decreased side effects. In the treatment of intractable pain the spinal and transdermal route is well documented. In neurology the flattened peak-through levels of antiepileptic drugs and anti Parkinson's drugs represents a more predictable kinetic profile. Tracheal delivery of corticosteroids and sympaticomimetics in asthma and Chronic Obstructive Pulmonary Disease is fully accepted in clinical practice: delivery by this route results in better efficacy and a better safety profile. In gynaecology the delivery of pulsatile hormones (LHRH) is used for pregnancy induction, while transdermal oestrogens are promising in the prevention of osteoporosis. In surgical practice the use of antibiotic impregnated bone cement and antibiotic impregnated biodegradable collagens is well established. To prevent infections intravascular catheters coated with heparin or antibiotics are used. In ophthalmology the Ocusert systems provide a controlled release of different drugs in the eye. Most spectacular is the clinical introduction of the first liposomal drugs; amfotericine B and daunorubicine. Liposomal formulations of these drugs have enhanced activity and decreased toxicity compared to conventional formulations.

Effects of UR-12633, a new antagonist of platelet-activating factor, in rodent models of endotoxic shock

1. The effects of the selective and potent novel platelet-activating factor (PAF) antagonist, UR-12633 (1-(3,3-diphenylpropionyl)-4-(3-pyridylcyanomethyl)piperidin e) on several markers of endotoxic shock syndrome were evaluated in rats and mice. 2. UR-12633, administered 60 min after E. coli lipopolysaccharide (LPS), reversed the LPS-induced sustained hypotension in rats at doses of 0.01 to 1 mg kg-1, i.v. The reference compound WEB-2086 (1 mg kg-1) also reversed the LPS-induced hypotension. UR-12633 (1 mg kg-1), administered 10 min before LPS, almost fully inhibited sustained hypotension. The immediate hypotension (within 1 min) caused by LPS was not prevented by either UR-12633 or WEB-2086. 3. Pretreatment with 10 mg kg-1, i.v. of either UR-12633 or WEB-2086 inhibited the increase in disseminated intravascular coagulation markers, such as activated partial thromboplastin time (55 and 74% inhibition, respectively), and prothrombin time (22 and 72% inhibition) and prevented the decrease in plasma fibrinogen content (100 and 29% inhibition). 4. Increases in acid phosphatase (ACP) plasma activity, a marker of lysosomal activation, and in lactate dehydrogenase (LDH), a marker of tissue damage, were inhibited by pretreatment with 10 mg kg-1, i.v. of either UR-12633 or WEB-2086 (100% and 69% inhibition, ACP; 62 and 48% inhibition, LDH). Hyperglycaemia (71 and 46%) and hyperlactacidaemia (92 and 56%) were also inhibited. 5. UR-12633, but not WEB-2086, inhibited the LPS-induced increase in vascular permeability in rats, as shown by prevention of haemoconcentration and, to a lesser degree, the increase in Evans blue dye extravasation. 6. In a series of nine reference compounds and UR-12633, we found a high correlation (P < 0.001) between PAF antagonist activity, measured as the inhibition of PAF-induced rabbit platelet aggregation or PAF-induced mortality in mice and the inhibition of LPS-induced mortality. 7. In spite of the multifactorial nature of endotoxic shock, in which many mediators may be involved, the new potent PAF antagonist, UR-12633, proved effective in protecting against changes in most shock markers. These data strongly suggest a key role for PAF in the pathogenesis of endotoxic shock in rodents.

Early detection and markers of sepsis

In vitro and animal models of sepsis have provided a template for studies of the pathogenesis of sepsis in patients at risk for and with the syndrome. Numerous potential markers have been identified in these models and then looked for in patients. No single marker or combination of markers convincingly identifies sepsis, predicts the development of sepsis, predicts the development of complications of sepsis, or predicts mortality. As discussed, the clinical studies have been complicated by many confounding variables, including the lack of adherence to rigorous definitions, differences in assay methods, differences in timing of the studies, and differences in outcome variables analyzed. In spite of the limitations, the studies have been critical in helping determine the pathogenesis of sepsis in humans. As new mediators and modulators of inflammation are identified, it will be important to study their role as markers, individually and in combination, in human disease.

The association among gastric mucosal pH, endotoxemia, and low systemic vascular resistance after cardiopulmonary bypass

OBJECTIVE

Previously, it was found that a number of patients suffer a "low systemic vascular resistance syndrome" after cardiopulmonary bypass, and this was hypothesized to be secondary to endotoxemia, resulting from intestinal ischemia during bypass.

DESIGN

Prospective cohort.

SETTING

University teaching hospital.

PARTICIPANTS

Thirty-two patients undergoing cardiac surgery.

INTERVENTIONS

A number of variables relating to adequacy of tissue perfusion were measured at seven time periods perioperatively: cardiac output, systemic vascular resistance, oxygen delivery and consumption, oxygen extraction ratio, gastric mucosal pH, serum lactate, and endotoxin levels.

MEASUREMENTS AND MAIN RESULTS

Investigators could not find any association between systemic vascular resistance and mucosal pH or endotoxin levels after bypass. There were significant changes in oxygen flux and extraction ratio (p < 0.001) as well as serum lactate (p < 0.001). There was no significant change in endotoxin levels or mucosal pH. The systemic vascular resistance at 6 hours postbypass could be predicted from the vascular resistance reading at 1 hour postbypass by a regression equation. A significant correlation between systemic vascular resistance and mixed venous oxygen was found at 4 and 6 hours postbypass (p < 0.01) as well as with oxygen extraction (p < 0.01). There was a negative correlation between mucosal pH and serum lactate, particularly at 6 hours postbypass (p < 0.01). There was no correlation between mucosal pH and endotoxin levels, oxygen flux, or cardiac output.

CONCLUSIONS

The investigators therefore could not find any evidence that intestinal ischemia during bypass, as measured by gastric mucosal pH, predisposes to endotoxemia, or low systemic vascular resistance after cardiac surgery.

Cellular functions during activation and damage by pathogens: immunogold studies of the interaction of bacterial endotoxins with target cells

Bacterial endotoxins (lipopolysaccharides or LPS) are active components of Gram-negative bacteria that act on numerous cellular functions through the processes of cell activation and damage. The molecular mechanisms involved in the "endotoxic phenomenon" are not defined yet, although extensive studies have been carried out. Immunogold and electron microscopy (EM) have contributed to identify the primary target cells of endotoxins and the subcellular systems that receive the direct action of these bacterial agents. Here, we review our studies on immunogold detection of endotoxins in cellular and subcellular systems. The analysis of the interaction between endotoxins and cells was focussed on the following aspects: (1) morphological characteristics of the LPS aqueous suspensions used in experimental work; (2) binding of endotoxins to the plasma membrane of type II pneumocytes and alveolar macrophages (two of their cellular targets), and influence of the state of aggregation of the LPS; (3) movement and distribution of endotoxins inside the cell, from the plasma membrane to the nucleoplasm; and (4) interaction of LPS with microtubules and its effects on the integrity of the microtubular network. These approaches provide information at the molecular level as well as data for the establishment of physiological models of endotoxicity.

Reappraisal with meta-analysis of bacteremia, endotoxemia, and mortality in gram-negative sepsis

Among patients with suspected sepsis, endotoxemia is variably present in association with gram-negative bacteremia. A total of 738 patients with suspected sepsis from 11 studies could be classified into four groups: 131 (18%) patients had both endotoxemia and gram-negative bacteremia (group 1), 87 (12%) had only gram-negative bacteremia (group 2), and 143 (19%) had only endotoxemia (group 3); in 377 (51%) patients neither could be detected (group 4). By the statistical techniques of meta-analysis, the fatality risk for patients with either endotoxemia or gram-negative bacteremia or both was estimated with group-specific case fatality rates from these studies and expressed as an odds ratio (OR; 95% confidence interval [95% CI]) versus patients with these factors absent. This risk was increased marginally when endotoxemia was detected without gram-negative bacteremia (OR, 2.3; 95% CI, 1.3 to 4.0) or the converse (OR, 2.0; 95% CI, 1.0 to 3.8), in contrast to the striking increase when both endotoxemia and gram-negative bacteremia were detected (OR, 6.3; 95% CI, 4.0 to 10.0). Alone, neither endotoxemia nor gram-negative bacteremia is a strong predictor of outcome in patients with suspected sepsis. However, in combination, the two identify a subpopulation with a substantially increased risk of mortality.

Endotoxemia: methods of detection and clinical correlates

As an assay for endotoxin, the Limulus amebocyte lysate assay has several desirable properties: sensitivity, specificity, and potential for adaptation to a quantitative format. Several modifications have been developed to enhance its potential for clinical application. The modifications that allow quantitative measurement of endotoxin and also improve its application to blood samples are described in this review. In fluids other than blood, the detection of endotoxin with the Limulus amebocyte lysate assay can be used as an aid to identify the presence of gram-negative bacteria, and the assay has established utility. With blood, however, there are a range of factors that interfere with the detection of endotoxemia and there are disparate views with respect to the diagnostic and prognostic significance of the test results. In general, the clinical significance of the finding of endotoxemia broadly parallels the frequency and importance of gram-negative sepsis in the patient groups studied and a decline in endotoxin levels accompanies clinical improvement. However, with therapies designed to reduce levels of endotoxin, or to antagonize its effects, it is unclear whether clinical improvement occurs as a consequence of changes in the levels of endotoxemia.

Endotoxemia and bacteremia in patients with sepsis syndrome in the intensive care unit

STUDY OBJECTIVES

To study circulatory endotoxin (ET) in patients with sepsis syndrome (SS) in order to answer three questions: (a) How often and at which concentration is ET present in the plasma of patients with SS and is the presence of ET a prognostic marker in this situation? (b) Is detection of ET helpful in predicting Gram-negative bacterial infections with or without bacteremia? (c) What are the kinetics of clearance of ET concentrations in plasma?

DESIGN

Prospective study of consecutive patients fulfilling Bone's criteria for SS.

SETTING

Medical ICU in a teaching hospital.

PATIENTS

The study included 93 patients. The simplified acute physiologic score was 19 +/- 6, 49 percent were in shock, and 54 percent were mechanically ventilated. The mortality at day 28 was 53 percent.

MEASUREMENTS

Endotoxin determinations and blood cultures were performed simultaneously at the onset (day 1) of SS. Samples were collected on several days from 48 patients. Endotoxin concentration was determined using an end point chromogenic Limulus assay. For the first ET determination, the mean circulatory level (mean +/- SEM) was calculated among patients with detectable ET, thus excluding patients with a null value for ET.

RESULTS

On day 1, ET was detected in 44 patients (47 percent; 60.2 +/- 16.5 pg/ml) and was statistically more frequent in patients with shock, elevated plasma lactate, and organ failure. There was no statistical difference for age, gender, ratio of PaO2 to fraction of inspired oxygen. Among patients with proven Gram-negative bacterial infection (n = 46), ET was detected in 67 percent as compared with 28 percent without Gram-negative bacterial infection (p = 0.0001). On day 1, among 19 patients who had positive blood cultures with Gram-negative bacteria (GNB), 15 had detectable ET (79 percent, 61 +/- 22 pg/ml). In 14 other patients whose blood cultures were positive for GNB but became negative on day 1, 9 had detectable ET (64 percent; 36 +/- 6.5 pg/ml). Endotoxin declined linearly between days 1 and 4.

CONCLUSION

In our study, the plasma ET concentration predicts neither Gram-negative infection, with or without bacteremia, nor the outcome. However, when ET is present in the plasma of patients with SS it remains detectable for a long period of time as compared to its rapid disappearance from plasma of animals or healthy human volunteers receiving ET intravenously. This slow clearance of ET suggests either a continuous release or a defect in its clearance.

Rifampicin in pneumococcal meningoencephalitis

The cases are reported of two infants with pneumococcal meningitis in whom initial antibiotic treatment was ineffective despite the organisms being sensitive to the drugs used. A clinical and radiological diagnosis of meningoencephalitis was made. A rapid improvement followed the addition of rifampicin treatment.