Search

Background Currently, supplementary serological testing for β-D glucan (BDG) is often selected to diagnose deep mycosis in care covered by the health insurance in Japan. The Wako method used by our center has low sensitivity, and different studies have used different cut-off values due to factors that cause false positives and false negatives. One possible cause of false negatives is the use of platelet-rich plasma (PRP) as the sample material. Because phagocytic white blood cells (WBC) are precipitated by centrifugation and only plasma is measured, it seems unlikely that the actual amount of BDG is being measured when using PRP. Further, a frequent cause of false positives is contamination from blood products and gauze containing BDG. To resolve these issues, the blood cell separator, hydroxyethyl starch, is used to precipitate only the red blood cells to obtain leukocyte-rich plasma (LRP). We hypothesized that it might be possible to improve the diagnostic rate of deep mycosis by measuring the BDG content of plasma containing WBC and fungal components and by comparing the BDG content of PRP and LRP measured simultaneously. Materials and methods Healthy human blood, albumin-added blood, wrung-out gauze fluid-added blood, and fungal solution-added blood were prepared, and PRP and LRP were prepared using hydroxyethyl starch. The BDG content of each sample was measured using the Wako method and compared. In addition, PRP and LRP of fungal-added blood were Gram-stained and examined under a microscope, and the number of WBCs and phagocytosed fungi was counted visually and compared. Results Measuring the BDG content of LRP confirmed that there were no false positives with LRP, and in vitro experiments comparing albumin-added false-positive blood to fungal-added blood showed significant differences between PRP and LRP only in the fungal-added blood. Conclusion Calculating the BDG-ratio (LRP/PRP) by measuring both LRP and PRP may eliminate false positives and false negatives of true deep mycosis and improve the diagnostic rate.

We focused our attention on the endotoxin present within and on the surface of white blood cells and attempted to establish a new sample preparation method for endotoxin assays in leukocyte-rich plasma (LRP), taking advantage of the erythrocyte-aggregating property of hydroxyethyl starch. We used an endotoxin-specific turbidimetric kinetic assay, which is the conventional method used to assay endotoxin levels in platelet-rich plasma (PRP). Then, we comparatively assessed the assay results obtained with the endotoxin assay using PRP and LRP. It was found that the sensitivity of endotoxin assay in LRP was 88.5 %, which was superior to 73.1 % of the sensitivity in PRP in the diagnosis of infections caused by gram-negative bacteria. These results suggest that our newly developed LRP endotoxin assay may contribute to an improvement in the rate of sepsis diagnosis.

PURPOSE Laser scattering photometry (ESP) is a newly developed plasma endotoxin assay method using horseshoe crab amebocyte lysate (AL) that recognizes small particles produced by polymerization of coagulin under the stirring conditions at 1000rpm. We elucidated the effect of human serum album (HSA) in the ESP method. METHODS AL was dissolved with 630μL of the specimen and a 200-μL aliquot was used for ESP; this conventional protocol was regarded as the ESP630 method. The ESP210 method was also used, i. e. AL was dissolved with 210μL of the specimen and a 200-μL aliquot was used for ESP. RESULTS Water induced the agglutination, and HSA prolonged the agglutination time depending on its concentration especially in the ESP630 method. The water-induced agglutination was not inhibited by the addition of anti-factor C monoclonal antibody, and amidinophenyl benzoate hydrochloride, used as a clotting enzyme inhibitor, intensively inhibited the water-induced agglutination. Therefore, the water-induced agglutination was suggested to be a false-positive reaction to non-specific activation of the clotting enzyme. The HSA-induced prolongation of the reaction in the national health insurance-covered turbidimetric kinetic assay was not observed. CONCLUSION HSA or plasma protein seemed to affect the result, especially in the ESP630 method, and a non-specific reaction was found to occur in the ESP methods.

The goal of the study was to examine the effects of sivelestat sodium hydrate (sivelestat), a neutrophil elastase inhibitor, on production of cytokines in granulocytes and monocytes, using flow cytometry after cytokine staining in whole blood culture. Blood samples were collected from healthy volunteers. Vehicle (control group), lipopolysaccharide (LPS) (LPS group), or LPS + sivelestat (sivelestat group) were added to the whole blood, followed by addition of a protein transport inhibitor in each group. After incubation, staining for cytokines retained in the cells was performed by addition of an anti-interleukin 8 (IL-8) or anti-tumor necrosis factor-α (TNF-α) antibody. The cells were then analyzed using flow cytometry. Granulocytic production of IL-8 induced by 1 ng/ml LPS was significantly (P

The effects of sevoflurane on proinflammatory cytokines related to ischemic-reperfusion injury are not clear. The hypothesis was tested that sevoflurane decreases myocardial ischemic-reperfusion injury by suppressing proinflammatory cytokines.Prospective, randomized study.A medical university heart center.Twenty-three patients undergoing coronary artery bypass surgery allocated randomly into 2 groups.Anesthesia for 23 patients undergoing coronary artery bypass surgery was maintained using either fentanyl (30 microg/kg) with propofol (2-8 mg/kg/h) in the control group (n = 10) or fentanyl (30 microg/kg) with 0.5% to 1.0% sevoflurane in the sevoflurane group (n = 13).Interleukin (IL)-6, IL-8, IL-10, and IL-1 receptor antagonist (IL-1ra) were measured by enzyme-linked immunosorbent assay. Troponin-T and creatine kinase-MB isoenzyme (CK-MB) were measured by enzyme immunoassay and ultraviolet absorption spectrophotometry, respectively. Serum IL-6 and IL-8 concentrations in both groups increased significantly over baseline from 60 minutes after declamping the aorta (p0.001). The increases were greater in the control group than in the sevoflurane group (p0.05). Serum IL-10 and IL-1ra concentrations in both groups increased significantly over baseline from 60 minutes after declamping the aorta (p0.001). There were no differences between the two groups. Serum troponin-T and CK-MB concentrations increased significantly in both groups from 60 minutes after declamping the aorta (p0.001); the increases were greater in the control group (p0.05).Sevoflurane suppressed the production of IL-6 and IL-8, but not IL-10 and IL-1ra. Changes in the balance between pro- and anti-inflammatory cytokines may be one of the most important mechanisms of myocardial protection caused by sevoflurane.

Nicorandil (NCR), a KATP channel opener, has been reported to preserve microvascular integrity in patients with reperfused myocardial infarction. We tested the hypothesis that NCR suppresses myocardial ischemia and reperfusion injury via the attenuation of cytokine production. Forty patients who underwent coronary artery bypass graft surgery were studied. The patients were randomly divided into two groups, i.e., the patients with NCR (4-6 mg/h; N group, n = 20) or without NCR (C group, n = 20). Cardiac surgery was performed under anesthesia using fentanyl and propofol. Blood were sampled at the time of induction of anesthesia, pre-cardiopulmonary bypass, 60 min after aortic occlusion, and 60, 120, and 180 min after declamping the aorta. The activation of NF-kappaB, expression of adhesion molecules, and cytokine production were evaluated in blood samples from the control volunteers by flow cytometric analysis with or without lipopolysaccharide (LPS) stimulation in vitro. Serum IL-6 and IL-8 levels in both groups increased 60 min after declamping the aorta compared with the preoperative value (P < 0.001); the increases of these parameters in N group were lower than those in C group (P < 0.05). Serum creatine kinase with muscle and brain subunits and troponin-T levels increased 60 min after declamping the aorta in two groups (P < 0,001), but the increases of both parameters in N group were lower than those in C group (P < 0.05). NF-kappaB activation, CD11b/CD18 expression, and the production of TNF-alpha, IL-8, and IL-6 in monocytes and granulocytes were inhibited by NCR in vitro. NCR suppressed the increase of inflammatory cytokines such as IL-6 and IL-8 levels, and reduced myocardial reperfusion injury. The inhibition on NF-kappaB activation, adhesion molecule expression, and cytokine production may be one of the important mechanisms of myocardial protection of NCR.

The FasL-Fas system is one of the recognized apoptosis-inducing systems, and has been determined to have important functions in relation to homeostasis and biological defense mechanisms. In this study, we investigated the serum levels of soluble Fas (sFas), soluble FasL (sFasL) and tumor necrosis factor alpha (TNF-alpha) in patients with burns. The sFas levels were found to be significantly higher in the patients who eventually died as compared to those in the patients who survived (3.9+/-1.8ng/ml versus 2.6+/-1.0ng/ml). On the other hand, the sFasL levels were significantly higher in the patients who survived (61.5+/-29.9ng/ml versus 37.2+/-14.4ng/ml) than in those who eventually died. A positive correlation was noted between the TNF-alpha level and the sFas level, and a negative correlation was observed between the TNF-alpha level and the sFasL level. These findings suggest that worsening of the condition of a burns patient may be related to changes in the Fas-FasL system.

The silkworm larvae plasma (SLP) test has been established based on a cascade reaction triggered by either peptidoglycan or (1, 3)-beta-D-glucan to produce melanin. We applied this test to the diagnosis of bacterial meningitis. Cerebrospinal fluid (CSF) obtained from patients with bacterial meningitis due to gram-positive bacteria, gram-negative bacteria, or fungi, showed positive reactions to the test. In contrast, CSF from patients with viral meningitis or noninfectious illnesses gave negative reactions. Therefore, this test seems to be useful for diagnosis of bacterial and fungal meningitis. When this test was used together with two types of limulus tests, an endotoxin-specific test, and a conventional test, meningitis was further characterized as gram-positive, gram-negative or fungal meningitis. The SLP test requires a computerized instrument for quantitative colorimetric measurement. A qualitative alternative of this test also can be accomplished by visually observing the darkening color. Thus, this method can be applied for simple and rapid diagnosis of meningitis.

To examine the influence of interleukin (IL)-18 in acute pancreatitis from the points of view of the severity of the disease, presence of the complication of MODF and the prognosis, 17 patients with acute pancreatitis were enlisted. The IL-18 and TNF-α levels in the serum were determined by enzyme-linked immunosorbent assay. The levels of IL-18 were 4.3±5.3 pg/ml in the mild group, 18.7±67.1 pg/ml in the moderate group, and 269.6±2482.1 pg/ml in the severe group. The IL-18 levels increased significantly, corresponding to the severity of the disease. The group in which the complication of multiple organ dysfunction syndrome (MODS) developed showed significantly higher IL-18 levels than that in which this complication did not develop. The IL-18 levels in the survivor group were 132.0±130.7 pg/ml, and those in the nonsurvivor group were 3028.6±480.8 pg/ml, the levels being significantly higher in the nonsurvivor group. Significant correlations were observed between the maximum serum IL-18 and TNF-α levels. Significant correlations were observed between the total bilirubin and IL-18 levels. The results suggest that IL-18 may be used as a good indicator of severity of acute pancreatitis and that involved in the development of liver dysfunction in acute pancreatitis.

The serum levels of surfactant protein (SP)-A and SP-D in patients with septic acute respiratory distress syndrome (ARDS) were compared with those in septic patients without ARDS. SP-A levels in groups with and without ARDS were 37.6±16.2 and 31.8±9.6 ng/ml, respectively. There was no significant difference between the two groups. SP-D levels in groups with and without ARDS were 83.0±33.9 and 476.3±391.2 ng/ml, respectively. The level in the ARDS group was significantly higher than that in the group without ARDS. No significant correlation was observed between SP-A and SP-D levels. In the presence of ARDS, SP-D served as a good indicator of severity.

Hypercytokinemia is known to occur in severe acute pancreatitis, suggesting that the production, deposition, and degradation of the extracellular matrix (ECM) occur actively as a result of the actions of the complicated cytokine network.To examine the influence of cytokines and growth factors on the ECM in acute pancreatitis from the points of view of the severity of the disease, the complication of multiple organ dysfunction syndrome (MODS), and the prognosis, 25 patients with acute pancreatitis were divided into three groups according to the severity of the condition as assessed by the Ranson score. The serum levels of matrix metalloproteinase (MMP)-1, tissue inhibitor of metalloproteinases (TIMP)-1, the MMP-1.TIMP-1 complex, tumor necrosis factor (TNF)-alpha, and transfer growth factor (TGF)-beta1 were determined by enzyme-linked immunosorbent assay.Comparison of the three groups divided according to the severity of the disease revealed significant differences in the levels of MMP-1 and TNF-alpha among the three groups, with the levels being higher in patients with more severe disease. The TIMP-1/MMP-1 ratio and the TGF-beta1 levels were found to be significantly lower in patients with more severe disease. Comparison between the group with the complication of MODS and the group without the complication showed that the levels of MMP-1, the MMP-1.TIMP-1 complex, and TNF-alpha were significantly higher in the former group than in the latter group. The TIMP-1/MMP-1 ratio and the TGF-beta1 levels were significantly lower in the group with the complication of MODS than in the group without the complication. Comparison between the nonsurvivor group and the survivor group revealed significantly higher levels of MMP-1, TIMP-1, and TNF-alpha in the nonsurvivors than in the survivors. The TIMP-1/MMP-1 ratio and the TGF-beta1 levels were significantly higher in the survivors than in the nonsurvivors. A significant correlation was observed between MMP-1 levels and TNF-alpha levels. On the other hand, a significant negative correlation was noted between MMP-1 levels and TGF-beta1 levels.The results of our study suggest that the activity of the ECM catabolic enzyme MMP-1 and cytokines are related to the development of acute pancreatitis.

The relationship between airway burn and nitric oxide (NO) levels in the early burn stage was investigated by quantifying nitrite/nitrate (NOx), which are the final metabolites of NO, in 22 burn patients. Although total body surface area (TBSA) and burn index (BI) were significantly higher in patients with airway-burn than in patients without airway burn (P=0.0347 and 0.0422, respectively), no significant difference in NOx levels was observed between the two groups (P=0.6196). The NOx levels were found to correlate significantly with TBSA (r=0.4775, P=0.0246). A significant correlation was also noted between the NOx levels and BI (r=0.4391, P=0.0409). These results suggest that NO reflects the intensity of inflammation in the early burn stage, but that excessive NO formation is unlikely to be induced by stress, such as that caused by airway burn.

In this study, we investigated multiple organ dysfunction syndrome (MODS) and nuclear matrix protein (NMP), which is an indicator for apoptosis, in burn patients. The subjects were 17 patients with a total burn surface area (TBSA) of 26% or more. Eight of them had MODS, and nine did not. Seven patients died, and 10 survived. The NMP value in the group with MODS (610 +/- 318 units/ml) was significantly higher than in the group without MODS (82 +/- 31 units/ml). The NMP values of the patients who died (645 +/- 326 units/ml) were significantly higher than in the group who survived (111 +/- 95 units/ml), and they rose significantly as the number of impaired organs increased. Our findings suggested that the NMP level in burn cases associated with MODS was higher when apoptosis developed and larger numbers of organs were impaired.

Magnesium (Mg) inhibits the influx of calcium in vascular smooth muscle cells. The purposes of this study were to test the hypothesis that an intravenous administration of magnesium might effect the complement response and to determine the effects of a magnesium pretreatment of patients with acute myocardial infarction (AMI) on the incidence of reperfusion injuries. Thirty-eight AMI patients were treated with coronary reperfusion therapy within 6 hours of onset. They were randomly divided into two groups: group pretreated with intravenous magnesium sulfate (0.27 mmol/kg) (magnesium group, n=19), and nonpretreated controls (placebo group). The reperfu sion injuries observed within 1 hour after the coronary reperfusion included arrhyth mias, aggravated chest pain, and ST segment elevation in 12-lead electrocardiograms. Coronary recanalization was performed in 36 patients. The incidence of reperfusion arrhythmia was significantly lower in the magnesium group than in the placebo group (17% vs 78%, p2+ may inhibit arrhythmic recurrence and the production of IL-6 and MMP-1 after reperfusion and prevent the increase of myocardial lesions caused by calcium overload on myocytes. The increased IL-6 production may induce MMP-1, leading to tissue organ injury. Pretreatment with magnesium sulfate may protect the myocardium of AMI patients from reperfusion injuries.

To understand the presence or absence of bacterial infection in patients with systemic inflammatory response syndrome (SIRS), the level of procalcitonin (PCT), a precursor of calcitonin, was determined. Subjects consisted of 14 SIRS patients without complication by bacterial infection, 14 SIRS patients complicated by sepsis, and 14 SIRS patients complicated by severe sepsis and septic shock. PCT levels in SIRS patients with sepsis (2.9 +/- 2.3 ng/ml) were significantly higher than those in SIRS patients without complication by infection (0.7 +/- 1.1 ng/ml). However, there were no significant differences in the levels of C-reactive protein (CRP), interleukin 6 (I-6) or tumor necrosis factor-alpha (TNF-alpha) between the two groups. PCT levels in SIRS patients with severe sepsis and septic shock (172.2 +/- 276.3 ng/ml) were significantly higher than those in SIRS patients with sepsis. Levels of CRP, IL-6 and TNF-alpha were also significantly higher in the patients with sepsis compared to those in patients with local infection. Significant correlations were observed between the levels of PCT and those of CRP, IL-6 and TNF-alpha in SIRS patients. It was suggested that to measure the levels of procalcitonin in patients with SIRS is useful to diagnose the infection and severity of illness.

炭酸脱水酵素阻害薬(アセタゾルアミド;AZ)の血流増加作用について,AZ投与時とCO2負荷時の臓器血流量と血管作動因子の関係を検討した。方法:2群の家兎の各々にAZ 4～12mg/kgの投与後,また5%と10% CO2負荷時の肝臓,腎臓,胃壁,腹筋の血流量をレーザー血流計で測定した。さらに心拍出量と血中PGI2(プロスタグランデイン),エンドセリン(ET-1),NOxを測定した。結果:AZ投与時には肝臓と腎臓の血流量は増したが,胃壁と腹筋の血流量および心拍出量には増加が認められなかった。血中のPGI2, ET-1, NOxにも変化は認められなかった。一方,CO2負荷時に心拍出量は増加し4臓器すべての血流量と血中PGI2とET-1は増加したが,NOxだけに変化は認められなかった。結論:AZ投与による臓器の血流増加は炭酸脱水酵素の豊富な臓器で認められ,その機序としてはCO2負荷時と同様な血管作動因子の作用ないしまったく異なった機序が関与して臓器血流を増すものと推察された。

エンドトキシン(LPS)に関する最新の報告を簡単にレビューし,さらにLPSに特異的なリムルステストとわれわれの開発した血漿前処理法(New PCA法)を駆使し,ここ十数年行ってきた研究の成果に基づいてLPSの病態における役割を考えた。1)LPS・LBP複合体は膜上のCD14と最近発見されたTLRとの複合体に結合し,シグナルが伝達される。2)感染を伴わない全身性炎症反応症候群(SIRS)ではLPSはほとんど基準値以下であった。3)重症の腹膜炎でもLPSの陽性率は決して高くなく,また血中濃度も高値でなかったことから,感染局所からのLPSの血中への侵入は無秩序には起こらない。4)敗血症や敗血症性ショックで病態と相関するのはLPSではなく炎症性サイトカインであった。5)抗LPSモノクローナル抗体(E5)の治験の成績のうち,血中LPS量と血液培養の結果から,菌血症でもLPS陰性(bacteremia but non-endotoxemia)の場合があり,リムルステストの検出感度が低いことが考えられた。6)LPSはむしろ感染局所で全身性の炎症反応に働いている。7)血漿蛋白に結合したLPSも多くの場合リムルステストで測定されている。8)保険適用測定法の一部にはまだ問題点がある。

The study was undertaken to measure plasma interleukin-1 (IL-1) receptor antagonist and IL-10 concentrations in patients with acute myocardial infarction and to analyze their relationship to the hemodynamics, severity, and prognosis of myocardial infarction in its acute stages. We attempted to define the kinetics of IL-1 receptor antagonist and IL-10 in patients with acute myocardial infarction (n = 34, age 42-91 years, mean 68 years). Plasma IL-1 receptor antagonist and IL-10 levels were measured by enzyme-linked immunosorbent assay. Patients in group A (n = 17) had uncomplicated acute myocardial infarction (Killip class I). Patients in group B (n = 17) had severe acute myocardial infarction (Killip class II, III, or IV). Peak Il-1 receptor antagonist and IL-10 levels in group B were significantly higher (p0.05) than those of group A. In group B, the peak IL-1 receptor antagonist levels were significantly correlated with white blood cell counts (r = 0.63, p = 0.006), pulmonary capillary wedge pressure (r = 0.78, p = 0.0002), and cardiac index (r = -0.51, p = 0.04). Peak IL-10 levels were significantly correlated with white blood cell counts (r = 0.60, p = 0.01), the pulmonary wedge pressure (r = 0.73, p = 0.0008), and cardiac index (r = -0.50, p = 0.04). Moreover, a significant correlation was found between the peak IL-1 receptor antagonist and IL-10 levels (r = 0.91, p0.0001). The peak IL-1 receptor antagonist levels in nonsurvivors (n = 13) were significantly higher (p0.01) than those in survivors (n = 21). The plasma IL-1 receptor antagonist and IL-10 levels were closely correlated with the severity of hemodynamics in acute myocardial infarction and with the clinical status of patients with severe acute myocardial infarction. Results suggest that plasma IL-1 receptor antagonist and IL-10 can serve as prognostic indicators in cases of sever acute myocardial infarction.

Plasma concentrations of endothelin-1 (ET-1) and thrombomodulin (TM) were determined in patients with burns to examine their relation to the severity of illness. Tumor necrosis factor-alpha (TNF-alpha) was also measured, and its relationship to ET-1 and TM determined. Twenty-three burn patients were evaluated, who had a total burn surface area (TBSA) of at least 20 per cent. ET-1 was measured by radioimmunoassay (RIA). TM and TNF-alpha were measured by enzyme-linked immunosorbent assay (ELISA). Both the ET-1 and TM concentrations were significantly higher in the patients who developed sepsis than in those who did not and in the patients who eventually died than in those who survived. Maximum plasma concentrations of ET-1 and TM were significantly correlated with the acute physiological and chronic health evaluation II score. There was also a significant correlation between the plasma levels of TNF-alpha and both ET-1 and TM. ET-1 and TM closely reflect the severity of illness in patients with burns in the infectious stage; TNF-alpha may be involved in the production of ET-1 and TM.

To investigate whether ulinastatin pretreatment (6000 U.kg-1 before CPB and before declamping of aorta) influenced the production of cytokines and adhesion molecules in the peripheral circulation.This prospective randomized study was performed in 22 patients undergoing cardiac surgery. They were divided into two groups. Patients in Group I were untreated and in Group II treated with ulinastatin. The soluble intercellular adhesion molecule-1 (S-ICAM-1), soluble endothelial leukocyte adhesion molecule-1 (S-ELAM-1), interleukin 8 and 6 (IL- 8, 6) were measured using ELISA kits.Serum S-ICAM-1 concentration in Group I increased from the preoperative value of 297 +/- 27 ng.kg-1 to 418 +/- 106 ng.kg-1 at 60 min after declamping of the aorta (P0.01) but did not change in Group II. Serum S-ELAM-1 concentration did not change in either group. Serum concentration of IL-8 and IL-6 in Group I (37 +/- 44 pg.kg-1, and 59 +/- 59 pg.kg-1, preoperatively) increased to 169 +/- 86 pg.kg-1 and 436 +/- 143 pg.kg-1 at 60 min after declamping of the aorta (P0.001, P0.001). The increases were greater than those from 25 +/- 6 pg.kg-1 and 30 +/- 26 pg.kg-1 to 56 +/- 36 pg.kg-1 and 132 +/- 78 pg.kg-1 in Group II (P0.001, P0.001). The levels of S-ICAM-1 correlated with those of IL-8 (r = 0.5, P0.001).These results suggest that ulinastatin may suppress the increase in IL-8 production and the expression of ICAM-1 during cardiac surgery.

The time courses of serum complement levels and the severity of sepsis were compared in two groups of septic patients, one in which the patients survived (surviving group) and one in which they did not (nonsurviving group). The components of the complement system, namely, C3a, C4a, C5a, CH50, C3, C4, and C5, were measured at several points in time after the diagnosis of sepsis had been established. A 2-antibody radioimmunoassay was used to measure C3a, C4a, and C5a; the latex agglutination test was used to measure C3 and C4; nephelometry was used to measure C5; and Meyer's 50% hemolysis method was used to measure CH50. Following the diagnosis of sepsis, the levels of CH50, C3, and C4 were significantly lower in the nonsurviving than the surviving group, while the levels of C3a and C4a were significantly higher in the nonsurviving than the surviving group. The C5a levels were significantly higher in the nonsurviving than the surviving group, although no significant intergroup differences were subsequently noted. These results suggest that the serum levels of C3a, C4a, C5a, CH50, C3, and C4 could serve as indices of the severity of sepsis. Thus, monitoring the complement system may be useful for predicting the outcome of patients with sepsis.

われわれは,大量に産生されるサイトカインを制御し,病態改善を図るため抑制性biological response modifier (BRM)としてのmethylprednisolone (MPS)の比較的,短時間持続投与を敗血症患者4例に行なった.すなわちMPSの初回投与量は5mg/kgとし,その後0.65mg/kg・hを48時間持続投与した.また,MPSの血中濃度の推移を測定した.われわれの投与方法ではMPSの濃度は投与後30分で2μg/mlに達し,以後2～4μg/ml前後で推移した.MPS投与により血中のTNFα,IL-6,IL-8は一様に低下した.臨床症状と各種検査値は投与後24時間には有意に改善された.以上より,敗血症症例での高サイトカイン血症が原因である臓器機能不全に対して,MPSの比較的少量持続投与は有効な治療手段の一つとなりうることが示唆された.

The plasma concentrations of type II phospholipase A 2 (type II PLA 2 ) and eicosanoids, such as leukotriene B 4 (LTB 4 ), 6-keto-prostaglandin F 1α (6-keto-PGF 1α ), and thromboxane B 2 (TXB 2 ), were determined by radioimmunoassay in 23 patients with burns covering at least 20 per cent of their body surface. Cytokines such as tumour necrosis factor-α (TNF-α), interleukin 6 (IL-6) and interleukin 8 (IL-8) were determined by enzyme-immunosorbent assay. There was no increase in type II PLA 2 concentration in the early stage of burns, but an increase in type II PLA 2 concentration was triggered by infection ( P 2 was significantly higher in the non-surviving group than in the surviving group ( P = 0.0006), suggesting that it reflects the severity of the disease. There was a significant correlation between the maximum level of type II PLA 2 and TNF-α ( r = 0.6346, P = 0.0011). There was a significant correlation between the maximum level of type II PLA 2 and the accompanying plasma concentrations of LTB 4 , 6-keto-PGF 1α , and TXB 2 throughout the observation period ( r = 0.4814, P = 0.0200; r = 0.5943, P = 0.0028; r = 0.4368, P = 0.0372 respectively). Plasma levels of LTB 4 , and TXB 2 were significantly higher in the burn patients who died than in those who survived ( P = 0.0493; P = 0.0493 respectively).

It has been reported that interleukin 8 (IL-8) and interleukin 6 (IL-6) are two of the chemical mediators causing myocardial injury. It is not clear whether treatment with corticosteroids in vitro in these patients can prevent the production of interleukin 8 and 6. This prospective study was conducted to investigate whether methylprednisolone (MP) pretreatment (30 mg.kg-1 before CPB and before declamping of aorta) influenced the production of IL-8 and 6 in the peripheral circulation in 27 patients undergoing elective coronary artery bypass surgery. The IL-8 and IL-6 concentrations were measured by ELISA kit. We also studied the effect of MP pretreatment on postoperative cardiac function. Serum concentration of IL-8 in non-MP-treated patients (37 +/- 44 pg.ml-1 preoperatively) increased to 169 +/- 86 pg.ml-1 60 min after declamping of the aorta (P < 0.001). The increase was greater than the increase from 22 +/- 8.9 pg.ml-1 to 52 +/- 35 pg.ml-1 in the MP-treated patients (P < 0.01). Serum IL-6 concentration in non-MP-treated patients increased from the preoperative value of 59 +/- 30 pg.ml-1 to 436 +/- 143 pg.ml-1 60 min after declamping of the aorta (P < 0.001). The increase was greater than the increase from 36 +/- 15 pg.ml-1 to 135 +/- 85 pg.ml-1 in the MP-treated patients (P < 0.01). Furthermore, postoperative cardiac index in MP-treated patients (3.6 +/- 1.1 L.min-1.m-2) was higher than 2.3 +/- 0.8 L.min-1.m-2 of non MP-treated patients (P < 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)

The endotoxin activity assay (EAA) is a FDA-approved blood endotoxin assay that is reported as a useful tool for the diagnosis of gram-negative bacterial infection. However, discrepancies between the results of the EAA and those of the limulus amebocyte lysate (LAL) assay have been reported. Thus, we verified these methods. Blood was incubated with anti-endotoxin antibody, the resultant polymorphonuclear activation to produce oxidants was measured and the EAA level calculated. As a reference endotoxin assay, we used an endotoxin-specific LAL assay. Significant increases in plasma LAL assay levels were observed only in patients with sepsis caused by gram-negative bacterial infections, whereas higher EAA levels were observed in almost all the sepsis cases and the SIRS cases, especially those with acute pancreatitis. Graded amounts of LPS (1–10,000 pg/ml) were spiked into normal blood to obtain dose–response curves: a good dose–response curve, from 1 to 1,000 pg/ml, was obtained for the LAL assay. A good dose–response curve was barely obtained for the EAA; the lowest detection limit seemed to be 1,000 pg/ml. Addition of methylprednisolone decreased the EAA levels. Interleukin-8 (IL-8) induced elevation in EAA levels when IL-8 was added to volunteers’ blood samples. Overall, the EAA kit could not measure clinically relevant doses of endotoxin. Because IL-8 induced an increase in EAA level, it is suggested that the EAA level reflects the primed state of polymorphonuclear leukocytes.

The roles of cytokines and endotoxin in hemorrhagic shock, particularly the translocation of endotoxin and bacteria during hemorrhagic shock, were investigated.Prospective study.Critical care and emergency center of a university hospital.Twenty-nine patients with hemorrhagic shock and 20 healthy controls.Serial blood samples were collected from both study and control patients. Standard resuscitation techniques were used.Plasma levels of endotoxin and various cytokines were determined repeatedly during hemorrhagic shock. Endotoxin was measured using an endotoxin-specific assay in addition to a new perchloric acid method for pretreatment of plasma. Cytokines were measured by commercial enzyme-linked immunosorbent assays. Plasma endotoxin concentrations remained within the normal range for 7 days after admission. Although levels of tumor necrosis factor-alpha and several interleukins increased slightly in some patients, these cytokines did not reach the levels seen in septic shock.Translocation of bacteria or endotoxin from the gastrointestinal tract into the bloodstream has been noted in animal experiments; however, translocation was not detected in our patients with hemorrhagic shock.

In order to study the network between human Hepatocyte Growth Factor (hHGF) and cytokines, we measured the changes in serum hHGF, interleukin-6 (IL-6) and interleukin-8 (IL-8) levels by ELISA method after hepatectomy in 7 patients. And plasma endotoxin levels were also measured by means of endospecy method. Serum levels of IL-6 and IL-8 reached the maximum immediately after hepatectomy, and rapidly decreased. Meanwhile serum hHGF levels reached the peak on just 1st postoperative day, and gradually decreased. Plasma endotoxin levels were all under normal level. No significant differences in the changes of serum hHGF were observed among the cases with or without liver cirrhosis, or the grades of hepatectomy. No significant correlations were found between peak serum levels of hHGF and those of IL-6. The interval time between the peak serum level of IL-6 and that of hHGF was 15 hours on the average. These results suggest that IL-6 can be a factor influencing the production of hHGF.

32例の食道静脈瘤直達手術術後に, 血中エンドトキシソ (Et) をNew perchloric acid (New PCA) 処理後エンドスペシー (生化学工業) を用いて測定し, polymyxin B sulfate (PL-B), tinidazol (TDZ) 投与の直腸内細菌, 血中Etに及ぼす効果を調べた.症例を非投与 (A) 群13例と, PL-B300万単位 (B) 群9例, PL-B600万単位 (C) 群5例, PL-B300万単位, TDZ 1,000mg (D) 群5例に分け, 術前に3日間投与した.投与前 (手術3日前), 術後1, 3, 5病日と, 投与群では術前日 (投与後) にも末梢血を採取しEtを測定した.A群は術前5.8 (平均, pg/ml) から第1病日に7.4に増加し, その後漸減した.B群では投与後に16%の減少がみられ, 術後も同じレベルで推移した.一方, C, D群は投与後に30%以上減少したが術後はB群に比し高値であった.PL-B投与は, 術後血中Etの増加を抑制したが, PL-B増量やTDZ併用は必ずしもEt濃度の減少を来さず, 細菌叢の変化あるいは死滅した細菌が放出したEtの増大が考えられた.

A clinical Phase I study of E5, edobacomab, an anti-endotoxin monoclonal IgM antibody was carried out to evaluate its efficacy, safety and pharmacokinetics and in parallel, explore an optimum dosage in the treatment of patients with gram-negative sepsis (incl. those suspected). A total of 10 patients were administered a dose of 1 mg/kg, 2 mg/kg or 4 mg/kg of E5 by infusion over one hour. 1. Peak blood levels after an intravenous dose of 1 mg/kg, 2 mg/kg and 4 mg/kg of E5 were 9.9, 11.5 and 35.6 micrograms/ml, respectively and the half-lives of elimination were 8.7, 11.7 and 10.8 hours, respectively, about the same at all dose levels. The area under the blood concentration-time curve (AUC) were 129.8, 200.8 and 574.4 micrograms.hr/ml, respectively, showing an increase in proportion with the dosage. 2. Urinary E5 concentration was not detectable in all the patients and renal excretion of unchanged E5 was not observed. 3. Out of the 8 patients with a positive pre-treatment endotoxin level with a value of above 9.8 pg/ml, those who showed its reduction to a level of below 9.8 pg/ml within one hour after dosing of E5 were 0 of 2 patients (0%) at a dose of 1 mg/kg, 3 of 3 (100%) at a dose of 2 mg/kg and 1 of 3 (33.3%) at a dose of 4 mg/kg. 4. The clinical efficacy following a single intravenous dose of E5 was rated as being "good" or "excellent" for 3 of 3 patients (100%) at a dose of 1 mg/kg, 4 of 4 (100%) at a dose of 2 mg/kg and 2 of 3 (66.7%) at a dose of 4 mg/kg. 5. As for antipyretic effect as assessed by the mean value of temperature, a fall in fever to the level of 36 degrees C was noted from 12 hours after administration of E5 in the patients treated with 2 mg/kg. The degree of improvement in WBC, platelet count and CRP was also more evident in this 2 mg/kg group. 6. The safety was rated as "no problem in safety" for all 10 patients. No adverse reaction and clinical laboratory abnormalities suspected of their relation to E5 was observed. 7. Positive antibody (HAMA) response to E5 was observed in 2 of the 10 patients. In one patient (4 mg/kg) of them, response because negative 9 weeks later and in the other patient (1 mg/kg) it tended to decrease.(ABSTRACT TRUNCATED AT 400 WORDS)

A study was carried out to investigate the efficacy of therapy with recombinant human granulocyte colony-stimulating factor (rhG-CSF) in 24 patients with granulocytopenia and sepsis who had failed to respond to antibiotics. The mean leukocyte count at the start of the study was 911 +/- 334/microliter. Patients were injected subcutaneously with 75 micrograms rhG-CSF once daily for a mean of 5.2 days. The plasma G-CSF concentration was measured by ELISA. The leukocyte count increased approximately 9-fold after 1 week in 19 patients and the percentage of granulocytes rose from 46.2% to 78.9%. These 19 patients survived, while the 5 patients with no leukocyte response to rhG-CSF died. High plasma G-CSF levels were found in patients with granulocytopenia. Plasma G-CSF levels decreased as levels of granulocyte increased in survivors. A high plasma G-CSF concentration persisted in the 5 non-responding patients resulting in a fatal outcome. This study suggests that rhG-CSF both increased the leukocyte count and was a useful therapeutic manoeuvre for sepsis.

食道静脈瘤術後にみられたトキシカラーとエンドスペシー (ともに生化学工業, 東京) のエンドトキシン (以下, Et) 値の相違に関して検討を加えた.症例 (13例) において手術前後にNew PCA (過塩素酸) 法前処理で測定したトキシカラーとエンドスペシーとの値の差 (G因子活性化因子値, 以下G値) は, 術後46.3 (平均, pg/ml) より術後第1病日140.6に有意に上昇した (p

The modernized limulus lysate assay on cerebrospinal fluid pretreated by the new perchloric acid (PCA) treatment was evaluated in one patient with Escherichia coli meningitis and four non-meningitis patients. The endotoxin (ET) concentration was much higher by the new PCA treatment than by the ordinary dilution method in the exacerbation phase of E. coli meningitis. The recovery rate of added ET also demonstrated the superiority of the new PCA treatment.

Levels of plasma tumour necrosis factor-alpha (TNF-alpha) were determined consecutively in 42 patients with burns > 20 per cent of the total body surface area using an enzyme-linked immunosorbent assay. In the early period after injury (including the period of burn shock), 24 patients had detectable TNF-alpha levels in their plasma. However, the plasma TNF-alpha levels at the time of admission were very low and did not correlate with the extent of the burn or the prognosis. In contrast, the maximum plasma TNF-alpha level over the whole clinical course was significantly correlated with the area of the burn and the prognosis. No correlation was found between the plasma TNF-alpha and plasma endotoxin levels. TNF-alpha may be produced locally in infected burns and monitoring of plasma TNF-alpha levels may be a useful prognostic indicator for burns patients.

Mouse anti-endotoxin monoclonal antibody (E5) is now under clinical trial (Phase II) in Japan by assessing clinical findings and plasma endotoxin levels. We attempted to evaluate an inhibitory effect of E5 on hypotension induced by endotoxin and TNF-alpha. Systolic blood pressure was measured with a programmable sphygmomanometer using the tail-cuff method. The mixture of endotoxin (E. coli O111:B4, Sigma, 2 micrograms/mouse) and recombinant mouse TNF-alpha (Genzyme, 8000 units) were injected into mice (ddY, 6-8-weeks-olds), and the change of blood pressure was observed for 3 hrs. The mixture significantly decreased the blood pressure to about 70% of the control. E5 (20 micrograms/mouse), which was injected 10 min before, significantly abrogated the effect of endotoxin and TNF-alpha. The hypotension induced by the mixture was definitely inhibited by the injection of E5 (50, 100 micrograms/mouse), injected 10 min later, or the injection of E5 (100 micrograms/mouse) 30 min later. We previously found that E5 prevents the lethality induced by a concomitant injection of endotoxin, TNF-alpha and IL-2. These results suggest that E5 effective for inhibition of endotoxin activity in vivo.

Plasma endotoxin concentrations were determined in 42 patients with burns covering more than 20 per cent of the body surface area, using the endotoxin-specific Endospecy assay and treatment of plasma by a new method developed by ourselves. The normal endotoxin level was 9.8 pg/ml or less. In the early period after injury when no infection was present, very few patients had an endotoxin level above 9.8 pg/ml and endotoxin levels did not correlate with the area of the burns or with prognosis. However, later in the clinical course, endotoxin levels were correlated significantly with the burned area and with the prognosis.

We established a new method of plasma treatment for the removal of interfering factors in the plasma to allow detection of endotoxin by limulus test. The limulus test used was an endotoxin-specific chromogenic test, the Endospecy test. Perchloric acid (PCA) treatment and centrifugation (PCA method) is usually used to remove interfering factors from plasma, with the precipitate being discarded and the supernatant used to detect endotoxin. As the solubilized precipitates of endotoxin-spiked plasma and some patient plasma were found to contain the Endospecy activity, we have devised a new method assaying endotoxin in both the supernatant and precipitate. This study confirmed that the solubilized precipitate of endotoxin-spiked plasma had Endospecy activity and found that the precipitate had other endotoxin activities, such as lethality in galactosamine-sensitized mice and pyrogenicity in rabbits. We also confirmed that interfering factors were completely removed from plasma samples by this new method. The endotoxin level after the new PCA method was found to be about 8 times higher than that determined after PCA treatment and the new PCA method surpasses the conventional PCA method with regard to the positive rate of endotoxin contents in clinical samples. These results indicate that the new PCA method is superior to the PCA method as a plasma pretreatment method for limulus test.

A case of congenital Listeria septicemia is reported. A 2256 g male infant suffering from respiratory and circulatory failure with shock-like symptoms and high levels of inflammatory cytokines (tumor necrosis factor-alpha, interleukin-1 beta, -6, and -8), was admitted to the Morioka Red Cross Hospital. Listeria monocytogenes was cultured from cord blood, contents from the external ear canal, rectum and stomach. The infant was treated with surfactant replacement as well as conventional therapy. The high levels of interleukin-1 beta decreased with the improvement of the circulatory function, which might have been the major cause of the poor clinical state.

Our purpose was to evaluate the effect of labor pain on the concentrations of cytokines and soluble adhesion molecules in cord blood.Umbilical cord interleukin-6 (IL-6), interleukin-8 (IL-8), tumor necrosis factor-alpha (TNF-alpha), granulocyte elastase (GEL), intercellular adhesion molecule-1 (ICAM-1), and endothelial lymphocyte adhesion molecule-1 (ELAM-1) were measured in 21 normal appropriately grown newborns delivered vaginally (VD group), and 20 normal appropriately grown newborns delivered by elective cesarean section (ECS group). Umbilical cord arterial blood pH and PO2 were also measured.Umbilical artery blood pH and PO2 in the VD group were not significantly different from those in the ECS group. There were no significant differences for concentrations of IL-6, IL-8, TNF-alpha, GEL, ICAM-1, and ELAM-1 in cord blood between VD and ECS groups.These results suggests that labor pains do not affect the concentrations of cytokines and soluble adhesion molecules in cord blood.

To evaluate the effect of labor pains or fetal distress on concentrations of interleukin-1 receptor antagonist (IL-1ra) in cord blood.Umbilical cord interleukin-1 beta (IL-1 beta) and IL-1ra were measured in 24 normal appropriately grown newborns delivered vaginally (VD group), 10 normal appropriately grown newborns delivered by elective cesarean section (ECS group), and 5 appropriately grown newborns with fetal distress (FD group). Umbilical cord arterial blood pH and PO2 were also measured.Umbilical artery blood pH and PO2 in the VD group were not significantly different from those in the ECS group. Umbilical artery blood pH and PO2 in the FD group were significantly lower than those in the VD group (p0.05). Umbilical artery blood pH in the FD group was significantly lower than that in the ECS group (p0.05), but no significant difference for umbilical artery blood PO2 was found between the FD and ECS groups. IL-1 beta levels were undetectable in the three groups of neonates. There were no significant differences for concentrations of IL-1ra in cord blood among the groups.These results suggest that the labor pain or fetal distress does not affect concentrations of IL-1ra in cord blood.

Interleukin (IL)-18 ( interferon-γ inducing factor) is one of cytokines, produced by macrophage, take part in differentiation T-helper (Th) to Th1 and interferon γ producing. T helper1 play role in cellular immunity especially in viral infection include dengue. A descriptive correlative study has done to know the correlation between IL-18 levels and disease severity in admitted dengue fever (DF) and dengue hemorrhagic fever (DHF) patients. In 42 subjects consist of 20 (47.6%) DF and 22 (53.3%) DHF (grade I to IV WHO criteria, 1999) showed that IL-18 levels significantly higher in DHF than DF patients. There are significant correlation between IL-18 levels and hematocrit and low platelet value. This study supports the possible role of IL-18 in pathogenesis DHF in adults. (Med J Indones 2004; 13: 86-9) Keywords: dengue, dengue hemorrhagic fever, IL-18, cytokine, pathogenesis

To evaluate antibody responses against lipopolysaccharide (LPS: O157, O26, and O111) in enterohemorrhagic Escherichia coli(EHEC) infection, sera of 24 schoolchildren associated with the Morioka outbreak in 1997 and of 74 sporadic patients suspected of having EHEC infection were examined. Using a positive standard serum, quantitative evaluation of LPS antibodies by an enzyme-linked immunosorbent assay (ELISA) was established. High levels of specific IgM and IgA antibodies against homologous E. coli LPS were present in the acute period and are characteristic of EHEC. This could be used for the serological diagnosis of EHEC infection, except for early infants and the elderly. In addition to the specific homologous response, multiple antibody responses against different serotypes other than those isolated were demonstrated in many cases by qualitative analysis using Western blotting.

The endotoxin in blood was measured to establish both the cut-off value and to detect Gram-negative septicemia. We employed a new perchloric acid treatment method using an endotoxin-specific chromogenic Limulus test (Endospecy test). The cut-off value of endotoxin in blood was 11.2 pg/mL. All cases of septicemia (n = 7) showed high values of endotoxin. Three cases were Group B streptococci, and two cases were Escherichia coli. The others were showed to be negative in blood cultures. The paired values of endotoxin titers during a 48 h interval were useful to evaluate the effectiveness of antibiotics.

Nitric oxide (NO) acts as a vasorelaxant. We investigated the relationship between nitrite/nitrate (NOx), which are the final metabolites of NO, and hemodynamics during septic shock. We also examined tumor necrosis factor alpha (TNF-alpha), interleukin-8 (IL-8), and endotoxin. A significant negative correlation was observed between NOx levels and pulmonary capillary wedge pressure (PCWP; r = -0.6075, P = 0.0028). A significant positive correlation was noted between NOx levels and the cardiac index (CI; r = 0.5934, P = 0.0038). A significant negative correlation was found between NOx levels and the systemic vascular resistance index (SVRI; r = -0.4354, P = 0.0485). A significant positive correlation was observed between NOx levels and the stroke volume index (SVI; r = 0.5040, P = 0.0186). A significantly close positive correlation was also observed between TNF-alpha levels and NOx levels (r = 0.7848, P0.0001). These findings suggest that NOx levels are closely associated with hemodynamics during septic shock, resulting in a vascular relaxing effect.

To determine the influence of prostaglandin E1 (PGE1) on the cytokine balance and myocardial protection during cardiac surgery.Prospective, randomized, nonblinded study.University hospital.A total of 19 patients on cardiopulmonary bypass undergoing cardiac surgery.According to randomized sequence, the patients received PGE1 (0.02 approximately 0.05 microg x kg(-1) x min(-1)) from the beginning of surgery to the end of study (PGE1 group, n = 11) or nothing (control group, n = 8).Interleukin (IL)-6, IL-8, IL-10, IL-1 receptor antagonist (IL-1ra), soluble tumor necrosis factor receptor I (sTNF RI), and soluble tumor necrosis factor receptor II (sTNF RII) were measured by enzyme-linked immunosorbent assays. Troponin-T and isoenzyme of creatine kinase with muscle and brain subunits (CK-MB) were measured by enzyme immunoassay and ultraviolet absorption spectrophotometry method, respectively. Serum IL-6 and IL-8 concentrations in both groups increased significantly from 60 mins after declamping the aorta compared with preoperative value (p.001), However, the increases were greater in the control group than in the PGE1 group (p.01). Serum IL-10, IL-1ra, sTNF RI, and sTNF RII concentrations increased significantly from 60 mins after declamping the aorta compared with preoperative values in two groups (p.001, respectively). There were no differences between the two groups. Serum troponin T and CK-MB concentrations increased significantly in the two groups from 60 mins after declamping the aorta (p.001), but these increases were greater in the control group than in the PGE1 group (p.01). IL-6 and IL-8 levels correlated with CK-MB concentration (r2 = 0.49, r2 = 0.36; p.001 respectively).PGE1 suppressed the production of IL-6 and IL-8 but not IL-10, IL-1ra, sTNF RI, or sTNF RII. The change in the balance between pro-and anti-inflammatory cytokines may be one of the most important cytoprotective mechanisms of PGE1.

Tumor necrosis factor alpha (TNF-alpha) and tumor necrosis factor receptor I and II (TNFRI and TNFRII) were studied in 24 burn patients who had a total burn surface area (TBSA) of 50.2 +/- 20.4%. Immediately after the injury, both the TNFRI and TNFRII levels correlated significantly with TBSA (r = 0.7344, P < 0.0001; r = 0.6074, P = 0.0012). The TNFRI and TNFRII levels immediately after the injury were significantly higher in the 11 patients who later died of their burns than in the 13 patients who survived (0.8 +/- 0.4 ng/ml vs. 1.8 +/- 0.7 ng/ml, P = 0.0002; 2.3 +/- 1.1 ng/ml vs. 4.5 +/- 1.6 ng/ml, P = 0.0009). The TNF-alpha levels immediately after the injury did not differ significantly between the group that survived and the group that died. The TNFRI and TNFRII values for the entire follow-up period also correlated significantly with TBSA. Peak TNFRI and TNFRII levels were significantly higher in the group that died than in the group that survived (6.0 +/- 4.7 ng/ml vs. 14.1 +/- 7.8 ng/ml, P = 0.0009; 7.0 +/- 5.1 ng/ml vs. 16.7 +/- 5.2 ng/ml, P = 0.0003). The TNF-alpha levels correlated significantly with both the TNFRI and the TNFRII levels. The TNFRI and TNFRII levels thus closely reflected the severity of the burns in both the acute postburn period and the subsequent follow-up period. In other words, these parameters well reflected the severity and outcome of the burns, irrespective of the presence or absence of accompanying infection.

We measured endotoxins, inflammatory cytokines and soluble adhesion molecules in the blood of 17 severe burn patients to determine the involvement of these factors in the pathophysiology of severe burns. All seventeen patients had burns with a total burn surface area of 20% or more and a burn index of 15% or more. Endotoxin was measured by an endotoxin-specific assay and tumor necrosis factor-alpha, interleukin 6 and interleukin 8 and soluble adhesion molecules were measured by enzyme-linked immunosorbent assay. CD11a, CD11b and CD18, measured by flow cytometry, were elevated in the non-surviving group, the septic shock group and the multiple organ dysfunction syndrome group, suggesting a close connection between these adhesion molecules and burns complicated by infection. Soluble adhesion molecules were found to indirectly reflect the level of endothelial cell adhesion molecules, suggesting that inflammatory cytokines may also be involved in their production.