Search

To elucidate the time course of endotoxaemia and antiendotoxin antibodies in patients with acute pancreatitis.Prospective clinical study.University hospital, Germany.25 patients with oedematous (n = 9) or necrotising (n = 16) pancreatitis, and 20 healthy controls.Concentrations of endotoxin and immunoglobulins (classes G, M, and A) directed at two lipid A molecules, four lipopolysaccharides, and alpha-haemolysin of Staphylococcus aureus measurements in plasma during a 12 day period.There were no differences in the degree of endotoxaemia between patients with oedematous and necrotising pancreatitis on admission. However, from the day after admission and throughout the observation period patients with necrotising pancreatitis had significantly higher concentrations of endotoxin than those with oedematous pancreatitis. Concentrations of IgM specific for endotoxin peaked at day 4, and then decreased in patients with oedematous pancreatitis while remaining high for those with necrotising pancreatitis. There was only a slight increase in IgA specific for endotoxin, and IgG and immunoglobulins to gamma-haemolysin remained steady throughout the observation period. There was strong cross-reactivity (r0.7) between IgM specific for endotoxin (70%), but this was less with IgA (52%), and IgG (20%).Necrotising pancreatitis is accompanied by persistent endotoxaemia with an extended rise in antiendotoxin antibodies. Patients with oedematous pancreatitis have a transient endotoxaemia with a temporary increase of Ig specific for endotoxin. Endotoxin stimulates the synthesis of specific antibodies (IgM) despite general immunosuppression.

Surgical procedures usually result in elevation of body temperature and changes of blood cell count or electrolyte levels, commonly known as the postoperative acute phase reaction. The present prospective randomised study compares the intensity of the postoperative acute phase reaction and postoperative morbidity in groups of patients who underwent either laparoscopic or conventional cholecystectomy. Conventionally-operated patients had significantly higher endotoxin plasma levels and a lower endotoxin-neutralisation capacity in the plasma 20 min after skin incision. This difference persisted until 6 h postoperatively. The plasma levels of interleukin-6 (IL-6) increased postoperatively, but did not differ between the two groups. C-reactive protein (CRP) increased postoperatively, with a maximum at 48 h after surgery. When comparing both groups, laparoscopically-operated patients had lower CRP-levels. The postoperatively elevated body temperature decreased significantly earlier in laparoscopically-operated...

Plasma levels of endotoxin, TNFα, interleukin-1β (IL-1β), and interleukin-6 (IL-6) as well as the endotoxin-induced release of these cytokines from whole blood were determined in 20 patients after elective aseptic surgery and eight patients suffering from septic disease. Blood samples were drawn before surgery as well as once daily over 7 days postoperatively. Endotoxin plasma levels increased on the day of the operation from 0.025 ± 0.005 EU/ml to 0.099 ± 0.022 EU/ml and remained elevated for 1 day. No changes were noted in the plasma levels of TNFα and IL-1. Levels of IL-6 increased to 198 ± 52 pg/ml on the day of the operation, while the release of TNFα and IL-6 markedly declined after whole blood was stimulated with endotoxin. The difference between basal and stimulated TNF values amounted to 505 ± 129 pg/ml preoperatively and 132 ± 41 pg/ml after the operation and 297 ± 28 pg/ml and 137 ± 43 pg/ml for IL-6. From the second day of the postoperative period onwards, the endotoxin-induced release of IL-6 was higher than the preoperative values. The results of whole blood stimulation in patients with sepsis were similar to those achieved directly after elective surgery. The cytokine release of TNFα, IL-1β, and IL-6 was reduced. We conclude that, during the early aseptic acute phase reaction, the changes in the functioning of monocytes resemble - qualitatively - those occurring during septic disease, and that this may be related to endotoxemia originating in the gastrointestinal tract.

Background and Study Aims : Previous clinical and experimental studies have shown evidence of a leakage of whole bacteria and bacterial products after major trauma through the gut barrier. By determining plasma endotoxin levels, products of the arachidonic pathway, interleukin-6, and the endotoxin-neutralizing capacity (ENC) of plasma during colonoscopy, we studied the gut barrier function and the pathogenetic sequelae of mediator release during a minimally invasive procedure. Patients and Methods : Thirty-two patients were enrolled in a controlled prospective study. Endotoxin and ENC were determined by a chromogenic modification of the limulus amebocyte lysate test. Prostanoids and interleukin-6 were measured using commercially available ELISA tests. C-reactive protein levels were checked by nephelometry. Results : Twenty-one of the 32 patients had elevated endotoxin plasma levels during colonoscopy. In one patient, gut-derived bacteria were detected in plasma. ENC decreased after 5 min, and thromboxane B 2 levels also started to increase at that time. No acute-phase response took place after 24 h. Conclusion : During colonoscopy, endotoxin can be detected in blood. ENC measurement was shown to be even more sensitive. The pathogenetic sequelae leading to gut barrier failure remain unclear, because mediator release and endotoxemia, as checked by ENC, took place simultaneously.

During a cardio-pulmonary bypass, as well as post-operatively, high levels of endotoxin, interleukin-6 (Il-6) and C-reactive protein (CRP) were measured in 30 patients. A significant increase in endotoxin plasma level occurred during surgery, culminating in a peak during reperfusion. Plasma levels of endotoxin continued to be slightly raised until the fifth day after surgery, whereas those of Il-6 rose at the time the operation came to an end and were at their highest 6 h later. CRP levels were also high, post-operatively, and were markedly raised on day 2. A definite, statistically significant correlation between the plasma levels of endotoxin and Il-6 during the operation was established, leading us to conclude that the endotoxin liberated during cardiac surgery acts as the main trigger in the release of Il-6 and thus induces the post-operative acute phase reaction. There was no evidence of a correlation between CRP and endotoxin or Il-6 plasma levels.

Patients who underwent surgical procedures usually develop elevated body temperature, changes of plasma levels of some proteins, and leucocytosis. These alterations are summarized as the postoperative acute-phase reaction. Also endotoxin can induce the described phenomena suggesting that endotoxin may play a role concerning the induction of the acute phase reaction. In order to test that hypothesis we determined endotoxin plasma levels preoperatively and daily postoperatively in patients who were operated on because of goiter (n = 20), colonic, pancreatic and gastric diseases (n = 58). A significant increase of endotoxin plasma levels was found at the first and third day after abdominal surgery whereas after goiter surgery the increase revealed to be only very slight. However, the decrease between the first and second postoperative day in the latter group was again statistically significant suggesting postoperative endotoxemia even after minor operations. Furthermore a correlation between the amount of circulating endotoxin and pulmonary or infectious complications could be established in patients after major operations even at the first postoperative day suggesting a pathogenetic relevance of postoperative endotoxemia.

The aim of this study was to establish the incidence of endotoxemia and the influence of endotoxin on specific antibody response after multiple injury. Blood samples were collected from 39 patients (median Injury Severity Score: 20.5) at 0-3 and 6-12 hours, and 1, 3, 5, and 10 days after admission. The endotoxin plasma levels were high at the first time point (mean = 0.421 endotoxin units/mL) and decreased in the later course. Total immunoglobulin levels of IgM, IgG, or IgA were low and increased throughout the observation period. Specific antibodies of the IgM class against two lipid A and four lipopolysaccharide preparations increased transiently but significantly on day 3 and/or day 5. No changes of specific antibody content against endotoxin or lipid A was seen in the IgG or IgA class. The specific antibody content of the different classes against alpha-hemolysin of Staphylococcus aureus did not differ during 10 days after trauma. The specific antibodies of the IgM class reacted with all lipid A and LPS lipopolysaccharide preparations demonstrating cross-reactivity. These results suggest that endotoxin may be a specific stimulator of IgM antiendotoxin antibody secretion following trauma. Language: en

Transferrin is reported to be a major lipopolysaccharide binding protein of human plasma, at least in vitro. By use of the limulus-amebocyte-lysate test the influence of transferrin on endotoxicity was studied. In the absence of any other protein human iron-free transferrin was able to strongly enhance endotoxicity in a concentration-dependent manner. Similar results were obtained when transferrin was added to primarily heat-inactivated plasma. Even in this assay the endotoxin recovery increased when transferrin was exogenously added. On the other hand, transferrin inhibited endotoxicity when inactivation of the plasma samples was performed after the addition of endotoxin and transferrin. These results lead to the conclusion that transferrin in fact interacts with lipopolysaccharide in a biologically important manner. In order to achieve neutralization of endotoxin, however, other plasma constituents are needed. The hypothetical function of transferrin is possibly a disaggregation of lipopolysaccharide micelles, following the interaction between the two molecules. The present data should justify further studies in order to clarify a possible benefit of the substitution of transferrin during gram-negative sepsis.

Translocation of endotoxin is a controversial issue. The ability of plasma to inactivate endotoxin is an indirect measure of endotoxemia. Endotoxin is a potent stimulator of the inflammatory response and affects the innate immune system.To elucidate the kinetics of endotoxemia and the ability of plasma to inactivate endotoxin in patients with major abdominal operations. To demonstrate the early time course of the acute-phase proteins C-reactive protein (CRP), serum amyloid A (SAA), alpha(1)-antitrypsin, alpha(2)-macroglobulin, transferrin, and interleukin 6 (IL-6), and to correlate them with the amount of endotoxemia.Twenty patients with elective major abdominal operation and 10 healthy controls were investigated. Blood was collected preoperatively, during the operation and regularly up to 12 days after surgery. Endotoxin was measured by Limulus amebocyte lysate test (LAL), the ability of plasma to inactivate endotoxin by modified LAL, the acute-phase proteins nephelometrically, and IL-6 by enzyme-linked immunosorbent assay (ELISA).Preoperative endotoxin plasma level (0.026 +/- 0.004 EU/mL) did not differ from healthy volunteers but increased during operation (0.09 +/- 0.02 EU/mL, P = 0.02). Endotoxemia peaked 1 hour after the surgical procedure (0.16 +/- 0.03 EU/mL; P0.0001 versus preoperative) and decreased to almost normal values after 48 hours. The capability of plasma to inactivate endotoxin was significantly reduced during (recovery, 0.16 +/- 0.03 EU/mL), 1 hour (0.25 +/- 0.04 EU/mL) and 24 hours (0.16 +/- 0.02 EU/mL) after the operation compared with preoperative (0.068 +/- 0.01 EU/mL) values. Plasma IL-6 was significantly increased for 48 hours with a peak 1 hour after surgery (470 +/- 108 pg/mL). CRP peaked at 210 +/- 19 mg/L (P0.0001 versus preoperative) 48 hours after operation and was significantly elevated for the rest of the observation period. SAA was significantly increased 24 hours after surgery (249 +/- 45 mg/L) and peaked additional 48 hours later (456 +/- 86 mg/L). alpha(1)-Antitrypsin, although a positive acute-phase protein, decreased initially to 1.38 +/- 0.1 g/L (preoperative, 2.33 +/- 0.18 g/L; P0.0001) and increased thereafter until day 12 (3.05 +/- 0.35 g/L, P = 0.11 versus preoperative). The same was true for alpha(2)-macroglobulin (preoperative, 2.2 +/- 0.16 g/L; intraoperative, 1.36 +/- 0.13 g/L; day 5, 2.8 +/- 0.4 g/L). Transferrin decreased already during surgery (1.6 +/- 0.1 g/L versus preoperative 2.8 +/- 0.17 g/L, P0.0001) and remained on this level for 5 days. Correlation analysis revealed a relationship between endotoxemia and the ability of plasma to inactivate endotoxin (r = 0.67, P0.0001) and also a relation between intraoperative endotoxemia on one hand and alpha(2)-macroglobulin (-0.53r-0.6, P0.05) as well as alpha(1)-antitrypsin (0.64r0.55, P0.05) on the other.Major abdominal surgery is associated with transient endotoxemia and a transient reduced endotoxin inactivation capacity of the plasma. Endotoxemia correlates with the endotoxin inactivation capacity. The surgical procedure causes substantial changes in plasma concentrations of acute-phase proteins. alpha(2)-Macroglobulin and alpha(1)-antitrypsin correlate moderately with endotoxemia.

Our previous study demonstrated that soluble CD14 (sCD14) modulates the biologic activity of circulating endotoxin, which appears after surgery. In this study, we examined the behavior of endotoxin-binding proteins, such as sCD14, lipopolysaccharide-binding protein (LBP), and bactericidal/permeability-increasing protein (BPI), in patients' plasma after major abdominal surgery and the phagocytic secretion of sCD14 from peripheral blood mononuclear cells (PBMCs) throughout the observation period. In a prospective study, 15 patients undergoing major abdominal surgery (gastrectomy, n = 3; pancreatectomy, n = 10; colectomy, n = 2) were involved in this study. The endotoxin-binding proteins were perioperatively (preoperatively; postoperative hour 6; days 1, 2, 3, 4, 5, 7, and 10) measured by an enzyme-linked immunosorbent assay (ELISA). To exclude the hemodilution effect of samples, each parameter was corrected by dividing the respective value by the albumin concentration. The phagocytic activity at each time point was tested as an ex vivo sCD14 secretion from PBMCs in the presence and absence of exogenously added endotoxin, Escherichia coli 055B5 (1 ng/ml). Significant endotoxemia (0.35 +/- 0.13 EU/ml; p

Antibiotics are known to induce the release of bioactive endotoxin (LPS) from gram-negative bacterial cells. Because varying data have been published on the influence of quinolone antibiotics on LPS liberation, we studied the effect of ciprofloxacin on a culture of Escherichia coli by determining bacterial killing and free LPS concentrations in comparison with imipenem and ceftazidime. LPS levels were measured by three different methods, namely (1) the Limulus amebocyte lysate test, (2) an ELISA method based on capture of LPS by monoclonal antibodies, and (3) indirect determination by measuring the ability of antibiotic-induced LPS to trigger TNFalpha release from a monocytic cell line. With both the Limulus and ELISA tests, a low endotoxin-releasing activity of ciprofloxacin was confirmed. In contrast to previous studies, this LPS also had low bioactivity in terms of TNFalpha induction. Limulus LPS determinations correlated more precisely with LPS bioactivity than did ELISA values, an observation which underlines the crucial role of LPS determination methods in studies of antibiotic-induced LPS release.

The in vitro effects of meropenem on Pseudomonas aeruginosa were examined by studying (i) the inhibitory and bactericidal concentrations of meropenem versus those of imipenem for clinical isolates; (ii) changes in bacterial morphology during in vitro culture; and (iii) release of endotoxin induced by meropenem compared with that induced by other antipseudomonal compounds. Meropenem MIC90 and MBC90 values for 108 clinical isolates were 2 and 4.8 mg/l compared to 4.5 and 9.6 mg/l for imipenem. Morphological studies using phase-contrast and scanning electron microscopy showed that meropenem induced the formation of indeterminate bacterial cell forms at drug concentrations of 1-2.5 mg/l (0.5- to 1.25-fold the MIC), while spheroplasts predominated at drug levels exceeding 5 mg/l (2.5-fold the MIC). Determination of free and EDTA-releasable endotoxin activity by means of the Limulus lysate test showed that both meropenem and imipenem liberated significantly less endotoxin than did ceftazidime. Therefore, although meropenem binds to penicillin-binding proteins (PBPs) 2 and 3 (in contrast to imipenem, which binds to PBP2 only), endotoxin release should not be a cause of concern when treating systemic gram-negative infections with this drug.

Bacterial translocation following gut barrier dysfunction is a well-documented phenomenon after major trauma and burn injury [1-5]. The clinical relevance, however, has yet to be elucidated. The same question applies to the recently observed bacterial translocation during abdominal surgery [6]. Similarly, circulating endotoxin has been described after trauma, burn and major surgical procedures, again indicating gut barrier dysfunction [1, 7-12]. Therefore, the severity of perioperative endotoxemia could be diminished by selective decontamination of the gastrointestinal tract before cardiac surgery, supporting the view of endotoxemia originating from the gastrointestinal tract [13]. Although some pathogenetic sequelae leading to the translocation of whole bacteria, such as the role of radical formation and preceding endotoxemia, are understood, no definite explanation of the mechanisms of bacterial and endotoxin translocation exists [14].

The pathogenetic role of endotoxin in gram-negative sepsis is generally accepted [1, 2]. In septic diseases the role of intestinal endotoxin reaching the bloodstream was experimentally demonstrated by Cuevas and Fine more than 20 years ago [3]. In recent studies endotoxemia has been observed during and after major surgical procedures such as vascular, abdominal and cardiac operations [4-9]. In the case of abdominal surgery there is clear evidence of a pathogenetic role of endotoxin in terms of an association of endotoxemia and the occurrence of infectious complications [4]. In patients who had had cardiac surgery the endotoxin plasma levels correlated with the catecholamines needed to obtain adequate hemodynamic conditions at the end of cardiopulmonary bypass [9]. Martinez et al. demonstrated in cardiacpatients that the severity of endotoxemia could be reduced by preoperative selective decontamination of the gastrointestinal tract, supporting the relevance of endotoxin translocation from the gastrointestinal tract [10].

Recent attempts to protect hematopoietic progenitor cells from cytarabine (ara-C)-induced toxicity by transfer of the cytidine deaminase (CDD) gene resulted in efficient in vitro inducibility of ara-C resistance. Another enzyme involved in intracellular ara-CTP inactivation is the deoxycytidylate deaminase (dCMPD). We therefore transfected the human dCMPD cDNA gene into murine fibroblasts and investigated the relationship of forced dCMPD expression and resistance induction to ara-C. Several cell lines were established which demonstrated a 1.7–3.5-fold increase in cellular dCMPD activity and an up to 2-fold increase in the IC 50 value of ara-C. However, increases in dCMPD activities did not show a positive linear correlation with the induction of ara-C resistance. In addition, CD34+ hematopoietic progenitor cells revealed the highest endogenous dCMPD enzyme levels among different human hematopoietic cells. Thus, despite the documented role for dCMPD in ara-CTP inactivation of certain cell types, these results suggest that the dCMPD gene may prove less useful than the CDD gene as a therapeutic target in attempts to attenuate ara-C-induced bone marrow toxicity.

Little is known about the role of peripheral blood mononuclear cells (PBMCs) in lipopolysaccharide (LPS) elimination. We studied the endotoxin elimination capacities (EEC) of PBMCs of 15 healthy volunteers, 13 patients with sepsis, and 1 patient suffering from paroxysmal nocturnal hemoglobinuria (PNH). Although expression of CD14, the best-characterized receptor for LPS to date, was reduced from 93.6% ± 0.8% in healthy subjects to 50.5% ± 6.5% in patients with sepsis and was 0.3% in a patient with septic PNH, EEC were found to be unchanged. There was no difference in the amount of tumor necrosis factor alpha (TNF-α) released by PBMCs of healthy donors and patients with sepsis. Anti-CD14 antibodies (MEM-18) completely suppressed EEC, binding of fluorescein isothiocyanate-labeled LPS to monocytes as determined by FACScan analysis, and TNF-α release in all three groups studied. The concentrations of soluble CD14 (sCD14) secreted by endotoxin-stimulated PBMCs from healthy donors and patients with sepsis amounted to 4.5 ± 0.4 and 20.1 ± 1.8 ng/ml, respectively. Based on our results, we suggest that PBMCs eliminate LPS by at least two different mechanisms; in healthy subjects, the membrane CD14 (mCD14) receptor is the most important factor for LPS elimination, while in patients with sepsis (including the septic state of PNH), increased sCD14 participates in LPS elimination. Secretion of sCD14 is strongly enhanced under conditions of low expression of mCD14 in order to counteract the reduction of mCD14 and maintain the function of monocytes. This sCD14 may substitute the role of mCD14 in LPS elimination during sepsis. The elimination of LPS by PBMCs correlates with the binding reaction and the secretion of TNF-α.

During goitre surgery (25 patients) and after major abdominal surgery (52 patients), we studied the plasma levels of endotoxin, interleukin-6 (IL-6), C reactive protein (CRP), and the so called myeloid-related proteins (MRP), MRP8, MRP14, and the heterocomplex of both single proteins, MRP8/MRP14 in three intervals: pre-, intra-, and postoperative. We observed that CRP levels began to increase on the first postoperative day, reaching a maximum on day 2 (median levels of 185 mg/L after major surgery and 77 mg/L after goitre surgery). IL-6 levels peaked at the end of the operation, remaining elevated for 6 h following abdominal surgery (299 pg/mL) and peaked on day 1 after goitre surgery (63 pg/mL). An increase in MRP8/MRP14 levels began toward the end of abdominal surgery, and maximum levels were recorded until 5 days after the operation (5,695 micrograms/L). Plasma levels were significantly elevated 2 and 6 h after minor surgery (3,619 micrograms/L), while no changes were observed in the plasma levels of MRP8 and MRP14. Evidence of significant endotoxemia was found after the induction of anesthesia in the abdominal surgery group (.13 endotoxin units (EU)/mL) and after skin incision (.07 EU/mL) in the thyroid surgery group. The observed time sequence, starting with the release of bacterial products at an early stage, followed by the secondary stimulation of factors inherent to the acute phase led us to conclude that certain bacterial compounds, probably deriving from the gastrointestinal tract, trigger the postoperative acute phase reaction and are responsible for the activation of monocytes/macrophages and granulocytes.

In search of potential therapeutic targets for enhancing the cellular detoxification of cytosine-arabinoside (ara-C), we transfected the human CDD cDNA gene into murine fibroblast cells and examined the effect of increased CDD expression on the cellular resistance to ara-C. Several cell lines were established which consistently demonstrated a 2.5-3.0-fold increase in cellular CDD activity and exhibited an up to 3.0-3.5-fold increase in the IC50 value of ara-C with regard to inhibition of cell proliferation or DNA synthesis, respectively. In the absence of ara-C, these cells maintained unaltered morphology and growth characteristics as compared with parental cells and mock-transfected controls. In addition, we examined the CDD and dCK activity in hematopoietic cells and found CD34+ hematopoietic progenitor cells to reveal a low ratio of CDD/dCK activity. Together, these data suggest that constitutive overexpression of the CDD gene correlates with increased ara-C resistance and thus may prove useful in considering attempts to attenuate ara-C-induced toxicity on hematopoietic progenitor cells.

To investigate the time course of endotoxaemia and its effects on the synthesis of antiendotoxin antibodies in patients with multiple injuries.Prospective clinical study.University hospital, Germany.40 Patients with multiple injuries and 20 healthy volunteers who served as controls.Blood samples were collected up to 10 days after injury and the concentrations of endotoxin, antiendotoxin antibodies to four endotoxins, and for anti-alpha-haemolysin of Staphylococcus aureus were measured. The kinetics of endotoxaemia and antiendotoxin antibodies were investigated.Endotoxaemia peaked 0-3 hours after injury at 0.425 EU/ml and decreased thereafter to 0.04 EU/ml within five days. Total concentrations of IgM, IgA, and IgG increased continuously after day 3 (p0.05), but the specific IgM response to all endotoxins was only temporary and the relative content of specific antibodies to all endotoxins peaked at day 3 (p0.05). Antiendotoxin antibodies of IgM class cross-reacted among themselves. There was no general increase in specific antiendotoxin antibodies of IgA and IgG class. The relative content of specific antibodies to alpha-haemolysin of S aureus of all classes of immunoglobulins (IgM, IgA, IgG) remained on the same level from day 0-10.Multiple injuries are associated with early and temporary endotoxaemia which causes specific increases in antiendotoxin antibodies of the IgM-class. IgM antibodies to endotoxins cross-react among themselves.

In 148 patients after major surgical procedures urinary endotoxin levels were determined and compared with bacteriological results. The study was designed as a screening study. Urine samples were collected once by suprapubic or transurethral catheters. In a first series of 49 patients urine bacteriology was positive (mainly, Gram-negative rods were found) in 3 cases. However, endotoxin determination was positive in these 3 patients and in a further 10 patients receiving antibiotic therapy for other reasons. Therefore, the following 99 patients were studied also by urinalysis by reagent strips for leukocytes and nitrite. In the second series, 12 urine cultures positive for bacteria were observed. Eleven samples were also endotoxin positive. Five more patients were endotoxin positive and had pathological but unspecific reagent strip results. These patients were treated with antibiotics for other reasons. Patients with candida found in the urine culture (n = 5) were endotoxin negative. Thus, endotoxin determination in urine obtained by suprapubic or transurethral catheters proved to be a very sensitive method for diagnosis of bacterial contamination, even during antibiotic treatment.

It has been suggested that tumor necrosis factor alpha (TNF alpha) acts not only by direct toxicity, but also as a proximal mediator which is able to induce the production of other cytokines, especially interleukin 6 (IL-6) and interleukin 1 beta (IL-1 beta). In order to test the dependence of the release of these two cytokines from leukocytes upon induction by TNF alpha, we stimulated whole blood in vitro with TNF alpha and compared the cytokine levels with those induced by endotoxin. The cytokine release was also determined after stimulation by endotoxin with added TNF alpha and by endotoxin with monoclonal antibodies against TNF alpha (anti-TNF alpha) added in order to reduce TNF alpha. Unstimulated blood samples were used as controls. The plasma levels of both IL-6 and IL-1 beta were significantly higher after stimulation with endotoxin than after stimulation with TNF alpha. TNF alpha did not induce cytokine levels significantly higher than controls. The cytokine levels were the same whether or not anti-TNF alpha was included together with the endotoxin. Plasma from samples with added anti-TNF alpha had no detectable TNF alpha. Our results indicate that the leukocyte-derived production of IL-6 and IL-1 beta in whole blood is stimulated directly by endotoxin and is not mediated by TNF alpha.

In 92 patients who underwent abdominal and goiter surgery endotoxin and endotoxin-neutralizing capacity (ENC) were determined in plasma preoperatively and daily postoperatively. Endotoxin plasma levels started to increase on day 1 in patients who were laparotomized. Correspondingly ENC was reduced during the 1st postoperative week. Even on the 1st postoperative day determination of ENC made the differentiation between patients possible who showed an uneventful course and patients who developed pneumonia, pulmonary failure, or mental disorders. Furthermore the need for diuretics in order to maintain sufficient renal function was associated with lower ENC during days 1-3. Endotoxin plasma levels were significantly elevated in patients who developed pneumonia during days 1-4, whereas the occurrence of pulmonary failure was only correlated with elevated endotoxin levels on day 3. Endotoxemia was pronounced in patients needing diuretics on days 1, 4, and 5. In patients who underwent goiter surgery ENC changed significantly as found for endotoxin plasma values during the postoperative course. Determination of endotoxin and especially ENC with the use of the limulus amebocyte lysate test turned out to be a reliable method correlating with impending complications at least in postsurgical patients.

Today endotoxins or lipopolysaccharides are accepted as representing the main pathogenetic mediators of gram-negative bacteria [1]. In a variety of diseases endotoxin was found circulating in the blood stream [2–6]. Ever since the mediator principle of septic disease has been generally accepted endotoxin has been seen as the primary trigger molecule of different mediator systems, at least in gram-negative sepsis [7, 8]. Recently, some excellent studies have been published demonstrating a correlation between endotoxin plasma levels and the course of sepsis that supports this interpretation [9, 10]. Thus, from a clinical point of view an adjuvant therapy seems desirable.