Scand J Infect Dis 24: 361-368, 1992

Endotoxin, TNF-alpha, Interleukin-6 and Parameters of the Cellular Immune System in Patients with lntraabdominal Sepsis G E R H A R D HAMILTON', SUSANNE HOFBAUER' and BARBARA HAMILTON'

Scand J Infect Dis Downloaded from informahealthcare.com by University of Auckland on 12/04/14 For personal use only.

From the 'Experimental Reseurch of the First Departnient of Surgery, University School of Medicine, and 'Institute of General Biochernistrv, Uriiversity of Vienna, Vienna, Austria

The correlation of endotoxin (ET), tumor necrosis factor alpha (TNF-alpha), interlenkin-6 (IL-61, and cellular immune parameters with multiple organ failure and lethal outcome in intraabdominal infections was studied in a group of 18 patients with peritonitis, abscess or pancreatitis. Of these patients, 7 developed respiratory failure and 5 died due to multiple septic organ failure. The peak levels of ET (2.7 f 1.3 ng/ml) in the course of the disease were followed by moderate increases of TNF-alpha (mean 147 k 41 pg/ml) and I G 6 (170 2 61 pg/ml) within 2 days. Analysis of the parameters for the last 12 days prior to death or discharge showed, that the patient group with lethal outcome was characterized by significant lower mean plasma levels of TNF-alpha (< 75 pg/ml versus > 160 pg/ml) and IL-6 (< 130 pgiml versus > 270 pg/ml), as well as high rates of unstimulated thymidine uptake into peripheral mononuclear blood cells (> 44000 cpml8 X lo6 PMBCl18 h versus < 24000 cmp), T-lymphocyte depression (CD3; approximately > 40% reduction) with lower T-helperlinducer subset cell numbers (mean CD4:CDS ratio 1.0 f 0.55 versus 1.8 f 0.2) and lower lectin (PHA) stimulation values (1.9 f 1.4 versus 4.1 f 1.0). These data demonstrate an anergic immune status with low mediator levels and depressed T-lymphocyte function in patients with poor prognosis. Therefore the therapeutic modalities should consist of immunorestorative measures and all attempts to neutralize mediators as TNF-alpha and IG6 are likely to have further adverse effects on the immune system in the late stages of intraabdominal infections.

G . Hamilton, MD, Experimental Research of the First Department of Surgery, University School of Medicine, Alserstrasse 4 , A-1090 Vienna, Austria

INTRODUCTION Multiple septic organ failure is the most frequent cause of death in severe intraabdominal infections, as diffuse peritonitis, intraabdominal abscess or bacterially contaminated necrotizing pancreatitis (1, 2). Infection may induce a septic syndrome, characterized by leukocytosis, pyrexia, altered mentation and a hypermetabolic state resulting ultimately in multiple organ failure (3). The response of the host immune system to bacterial products contributes to the septic organ failure by elaborating excessive levels of mediators, as interleukin-1 (IL-l), tumor necrosis factor alpha (TNF-alpha). IL-6 and others, which are functional in the enhancement of the host defenses in low concentrations, but deleterious in high amounts or prolonged production (4, 5 ) . Endotoxins (ET), TNF-alpha and IL-6 have been correlated with severity and prognosis in bacteremia, meningitis and severe sepsis (6-and 8)polyclonal and monoclonal antibodies directed to any of these mediators have been demonstrated to prevent organ failure from septic syndrome in experimental animal models and subgroups of patients (9-12). In addition to altered cytokine patterns, the cellular immune system is affected and its anergy, as measured by in-vivo and in-vitro tests, as well as shifts in subpopulations of effector cells, has been correlated with outcome in septic patients (13-15). I n a previous study we found no correlation of ET plasma concentrations with prognosis in

362 G. Hamilton et al.

Scand J Infect Dis 24

i n t r a a b d o m i n a l sepsis, indicating possible differences in t h e c o u r s e of i n t r a a b d o m i n a l infections versus systemic vascular infections (16). T h e r e f o r e in this study an a t t e m p t w a s m a d e t o m e a s u r e t h e s t a t u s of t h e i m m u n e s y s t e m , as characterized by cellular i m m u n e p a r a m e t e r s a n d plasma c o n c e n t r a t i o n s o f t h e m e d i a t o r s T N F - a l p h a and IL-6 in ET-positive l a t e a b d o m i n a l infections and t o study t h e correlation of t h e s e f a c t o r s with organ failure a n d survival

Scand J Infect Dis Downloaded from informahealthcare.com by University of Auckland on 12/04/14 For personal use only.

PATIENTS AND METHODS Patierits 18 patients with intraabdominal infections. treated between November 19x9 and June 1990 at our institution, were included in this prospective study. This group consisted of 7 females and 11 males with a median age of 57 years (t 15 years, range 26-76). Abdominal septic state was defined as surgical intraabdominal infection, proven by bacteriological cultures. The infections were caused by diffuse pcritonitis ( 1 1 = 9), intraabdominal abscess ( n = 2) and infected pancreas necrosis (acute necrotizing pancreatitis with bacterial contamination of the necrotic tissue. ref 17; n = 7). The severity of the disease was classified by the APACHE I1 -score (18) and determined to be 14.5 (median, range 7-29) at the study entrance. Surgical treatment aimed to remove the source of infection and included the pancreatic necrosis blunt necrosectomy, application of laparostomies, followed by a regimen of scheduled and on demand revisions. All patients received their treatment at the intensive care medical unit and were supported with mechanical ventilation, continuous hemofiltration, fluid therapy with cristalloids and vasoactive drugs, if necessary. They obtained total parenteral nutrition and antibiotics directed to the microorganisms found in bacterial cultures of peritoneal exsudate, necrotic tissue, blood and bronchial secretion. Seven patients (39%) developed respiratory failure (defined according to Knaus et al. ref. 19) and required mechanical ventilation, and 5 patients (28%) died due to multiple septic organ failure, proven by autopsy. The study covered the period from admittance to the intensivc care unit until discharge (no late non-survivors) or death. Blood samplirig For the isolation of peripheral blood mononuclear cells and for the determination of ET, TNF-alpha and 1L-6. 10 ml heparinizcd blood samples werc collected from an arterial line 3 times every 2nd day (8 a.m., noon, 4 p.m.) during the study period. 5 ml were used for the isolation of peripheral mononuclear blood cells (PMBC) and the remaining blood was centrifuged (10 min, 500 g) and the plasma stored frozen at - 20°C. Determination of ET, TNF-alpha arid IL-6 Didoioxin. ET was mcasured as previously described (16). In brief, 0.5 ml plasma (stored frozen at -20°C) was centrifuged in molecular weight filtration tubes (prewashed by centrifugation with pyrogenfree water. cut off 20 KD; Centrisart I, Sartorius. Gottingen. FRG). The fraction > 20 KD was brought to 1 ml with pyrogcn-free distilled water and the proteins precipitated by addition of 1 ml watersaturated phenol (10 min, 68°C). After cooling to 4°C (30 min) and centrifugation (15 min, 1000g). the aqueous phase was collected and the phenolic phase once again extracted with 2 ml of water. The aqueous phases were pooled and aliquots were diluted 1:400 in pyrogen-free water and tested directly in an Limulus amoebocyte lysate quantitative assay system (Coatest Endotoxin, Kabi Vitrum, Nykoping. Sweden) without further purification. TNF-alpha. TNF-alpha was measured with the immunoradiometric assay (IRMA) from Medgenix diagnostics (Brussels, Belgium) according to the manufacturers instruction. IL-6. IL-6 was determined with an ELISA assay from Research and Diagnostics Systems (Minneapolis. MN, USA). Samples from healthy controls contained no detectable levels of IL-6. Isolation of PMBC 5 ml heparinized blood were diluted with 3.5 ml phosphate buffered saline (PBS) and layered carefully on the top of 4.5 ml Ficoll-Paque (Pharmacia, Uppsala, Sweden). Samples were centrifuged for 30 min at 260 g and the PMBC collected from the interphase and washed twice with medium (RPMI-1640/ Hepes, supplemented with 10% fetal bovine serum, 4mM glutamine and 75 pg/ml gentaniicin; Flow, Irvine, Scotland, UK). The cells were counted with an Microcellcounter (Sysmex, Tokyo, Japan) and

C‘ytokities wid T-lymphocytes iri intraubdotninol sepsis 363

S c a n d J Infect Dis 21

T h l e I. Within-day variability of tiw medicitor i ~ l r i s r n nh e l s Blood sampling Mediator

X a.m.

Noon

4 p.m.

Endotoxin

34.3219.9.’ 50.0f2S.S” 40.5 23.6

36.6i17.9 29.1+20.6 2h.8i14.9

2 9 . 4 i 14.2 21.0i12.8 32.5220.2

1”F -a1p ha

*

1L-(1

Scand J Infect Dis Downloaded from informahealthcare.com by University of Auckland on 12/04/14 For personal use only.

I’

I’

Mean peicentiigc of the mediator concentration a t cach timc point ( 2 SD) with 100% representing the win of all 3 detcrniinations for each day. Statistically significant different from the mean vnlues for noon and 4 p . m . according t o the rank correlation test (p < 0.05; Spearman).

brought to a concentration of 8 pla 1es .

X

10hcells/ml medium and distributed to the wells of 96 wells microtiter

m t l phenorye oj ilir PMNC For the rncasurement of spontaneous prolifer-ation two 100 pl aliquots were labeled with 37 kBq ’H-thymidine for 18 h under tissue culture conditions. For lectin stimulation the cultures were supplemented with 10 ngiml phytohcmagglutinin (Pharniacia. Uppsala, Sweden) and incubated for 3 days followed by labeling with thymidine f o r the last 18 h. The proliferative response was expressed as stiniulation index (SI) relatively to control culttires without lectin. At the end of the incubation time the PMRC were collected to filter disks using a cell harvester (Skatron, Lier. Norway) lysed with distilled watcr. and the incorporated radioactivity was measured by beta-counting. Lvmphocytic subpopulations were quantitated by tlow cytometry (FACS Analyzer cquipped with an FITCiPE filter set, FACSLite laser source and Consort 30 data handling system; Becton-Diekinson. Mountainview. CA, USA) after staining with appropriate direct-labeled antibodies (Leu3a-FITC. L e d a - P E . L e d - F I T C and anti-HLA-DR-PE; Bccton-Dickinson). Data were collected from 5000 lymphocytes as separated by a volume-side scatter live gate and the positive cell populations are given as perccntage of this gated cell population. Dcrrttiiiiuiioirof / h e uctiL’ity

Siuiisiicul unulysis Differences in the within-day variations of thc mediators were analyzed with the rank-correlation according to Spearman. Values for the mediators and cellular immune parameters were compared for the survivors-nonsurvivors and patients with and without respiratory failure and evaluated for statistically significant differences with the Wilcoxon test.

RESULTS Clinical coiirse of [he patients Seven patients (39%) developed respiratory failure and 5 patients (28%) died due to multiple septic organ failure as proven by autopsy. The surviving patients and the patients without respiratory failure had no significant age difference in comparison t o the other patients. Bacterial contamination was detected in peritoneal exsudate or pancreatic necrosis o f all patients and positive cultures were also obtained for the bronchial secretions from 11/18 patients, in 1/18 cases additionally in blood cultures. A total of 43 revisions were carried out for this group of 18 patients. Mcusurement o,f the mediators arid within-day vciriations Blood samples were collected at 8 a.m., noon and 4 p.m. every 2nd day and used for the isolation of PMBC or stored frozen for the analysis of ET, TNF-alpha and IL-6. For the

Scand J Infect Dis Downloaded from informahealthcare.com by University of Auckland on 12/04/14 For personal use only.

364

G. Hamilton et al.

o b s e r v a t i o n nr.

Scand J lnfect Dis 24

Fig. I . Courses of the plasma concentrations (mean values _+ SEM) of ET (A----A),TNFalpha (W-W) and IL-6 (0-0) in patients with intraabdominal sepsis. The measurements (observations every second day) were aligned according to the maximal plasma concentration of endotoxin (n = 15).

determination of ET, low molecular weight plasma constituents and proteins interfering with the Limulus amoebocyte lysate (LAL) assay were removed by ultrafiltration and phenol denaturation. The method differs from that described by Berger et al. (20) in omitting the highly complicated purification procedures after: phenol denaturation, which proved to be of no advantage for the testing of diluted phenol-treated samples in the LAL test system. The distributions of the maximal plasma concentrations of the respective mediators for each day are shown in Table I as mean percentages of ET, TNF-alpha and IL-6 measured at the different time points for each day (sum of the results of all 3 determinations for each day = 100°/o). In contrast to E T , which is distributed randomly for the 3 time points, most of the TNF-alpha maximal plasma concentrations are found in the blood samples collected at 8 a.m. (p < 0.05). A similiar distribution was found for IL-6 but the differences do not reach statistical significance.

Production of TNF-alpha and IL-6 in response to increased ET concentrations When the serial measurements of the ET plasma concentrations were matched according to the maximal value for individual patients, the subsequent elevations in TNF-alpha and IL-6 became detectable at the next observation point (Fig. 1). The mean peak plasma concentrations of approximately 150 pg/ml for both TNF-alpha and IL-6 are low in comparison to measurements of other septic patients or transplant recipients. IL-6 was elevated for the whole observation period, most likely due to the surgical trauma associated with the revisions. Comparison of the concentrations of the mediators prior to and after operation revealed a significant rise in IL-6 (125% rfr 110% increase relatively to the value prior to operation, p < 0.01), a significant decrease in E T (52.2 k 38.5, p < 0.05), but no significant changes in unstimulated thymidine uptake (+ 43% ? 130%) o r TNF-alpha plasma concentrations (+ 23.6% ? 66%).

Correlation of survival with plasma concentrations of the mediators and cellular immune parameters The plasma concentrations of the mediators TNF-alpha and IL-6 and the cellular immune parameters were analyzed for significant differences between the surviving (n = 13) and non-surviving patients ( n = 5 ; Fig. 2). The mean values for each group were calculated for

Scand J Infcct Dis 24

Cytokines and T-lymphocytes in intraahdominal sepsis

Scand J Infect Dis Downloaded from informahealthcare.com by University of Auckland on 12/04/14 For personal use only.

1

Fig. 2. Comparison of the courses of TNF-alpha, IL-6 and cellular immune parameters in surviving (O---O)and nonsurviving (O-) patients with intraabdominal sepsis for the last 12 days before death or discharge from the intensive care unit (measurements every 2nd day). Mean values (2 SEM) from thc 5 non-survivors and 13 survivors are shown for TNF-alpha (A), IL-6 (B), spontaneous blastogenesis (C; SB, or unstimulated thymidine uptake into PMBC) and the percentage of T-cells (CD3) among peripheral lymphocytes. Statistically significant diffcrcnces are marked by an asterisk 0, < 0.05; Wilcoxon test).

..

I-- ---I-----!

'1

'2

'4

'5

observation

nr.

'3

'6

+

the different parameters using the daily maximal values and shown for the last h observations (12 days) prior t o discharge or death. The measurements at the beginning of this observation period showed no significant differences for survivors and non-survivors (Fig. 2). For ET no significant differences (p < 0.05) could be detected in the plasma concentrations for surviving and non-surviving patients (3.3 k 2.1 n g h l versus 4.4 k 3.4 ng/ml for the whole observation period; mean k SD; data not shown; normal controls: < 0.02 ng/rnl). TNF-alpha plasma concentrations were lower in non-survivors, with statistically significant differences for the observations 2 , 3 and 6 (Fig. 2 A; normal controls: < 5 pg/ml). A similar result was found for IL-6 with significantly lower mean values for the non-survivors for the observations 2, 5 and 6 (Fig. 2 B). For the cellular immune parameters the spontaneous blastogenesis was elevated significantly in non-survivors, when compared with the surviving patients (p < 0.05; with exception of observation 2; Fig. 2 C) and the T-lymphocytes (CD3 cells) were significantly lower in non-survivors for the whole investigation period (p < 0.01; Fig. 2 D; normal controls; 72% -t 8%). Quantitation of the T-helperhnducer lymphocytes and T-suppressorkytotoxic lymphocytes showed, that the low T-cell percentage is due to the loss of T-helperhnducer lymphocytes, resulting in a mean T-helperhnducer : T-suppressor

365

366 G. Hamilton et al.

Scand J Infect Dis 24

cytotoxic T cell ratio of 1.0 k 0.55 for the non-survivors and a normal ratio of 1.8 k 0.2 for the survivors. B-lymphocytes were low and not significantly different between survivors (S.5"/0 k 1.4%) and non-survivors (4.6% k 2.1%). The mean stimulation index (thymidine incorporation into lectin-stimulated cultures (PHA)/thymidine incorporation into control cultures) was significantly lower for the non-survivors (1.9 f 1.4 versus 4.1 ? 1.0; p < 0.05).

Scand J Infect Dis Downloaded from informahealthcare.com by University of Auckland on 12/04/14 For personal use only.

Correlation of respiratory failure with plasma concentrations of the mediators and cellular immune parumeters The differences in the mean values for the mediator concentrations and cellular immune parameters were analyzed for the patients with respiratory failure ( n = 7) and for the patients without need of mechanical ventilation ( n = 11). No significant differences were found for TNF-alpha, IL-6, ET or spontaneous blastogenesis and T-lymphocyte subpopulations in late intraabdominal infection (data not shown).

DISCUSSION Multiple septic organ failure is the most frequent cause of death following intraabdominal sepsis. The bacterial cell wall endotoxins stimulate monocyteshacrophages to secrete TNF-alpha, which is an endogenous pyrogen and causes the abnormalities observed during sepsis including organ failure, when injected into experimental animals (6). This exaggerated or prolonged stimulation of the immune system is associated with the release of further mediators. as IL-I, IL-6 and IL-8, platelet activating factors, complement components and others, which interact and produce the detrimental effects of the septic syndrome (21). Monoclonal antibodies to ET/LPS, TNF-alpha and IL-6 were found to have therapeutic efficacy in experimental animal models as well as in first human clinical studies (9-12). Several findings indicate important differences in the significance of cytokines in intraabdominal sepsis in comparison to systemic intravascular sepsis (22). Antibodies to TNF-alpha showed a protective effect in systemic infection in rats but failed to change the course of experimental Escherichia coli peritonitis (10). In fact it was demonstrated, that TNF-alpha is necessary in an animal model of peritonitis for the successful response to an E. coli challenge: neutralizing anti-TNF-antibodies induced high lethality in mice with sublethal cecal ligation and puncture and this effect could be reversed by supplementation of recombinant TNF (23). In a previous study we have found that patients with abdominal sepsis had elevated plasma concentrations of ET, but that E T does not predict survival or death in this group (16). Therefore the measurement of TNF-alpha, IL-6 and cellular immune parameters was included in this investigation to search for differential effects of ET in survivors and non-survivors. All mediators were determined 3 times a day on every second day, to detect the transient and highly variable peak plasma concentrations of ET and TNF-alpha with higher probability. Operative revisions during the course of abdominal sepsis, as well as operations performed previous to sepsis, and the other therapeutic modalities further contribute to the high variability in cytokine production in these patients. TNF-alpha values were significantly higher in the blood samples collected at 8 a.m. and IL-6 showed a similar distribution, whereas the peak levels of ET were found with equal probability at any of the 3 daily time points. The daily maximal values of the mediators were used for further analysis. Alignment of the cytokine measurements according to the ET maximal values showed that these patients with advanced abdominal infections exhibit increases in TNF-alpha and IL-6 in response to rising ET levels. These mean maximal TNF-alpha plasma concentrations

Scand J Infect Dis Downloaded from informahealthcare.com by University of Auckland on 12/04/14 For personal use only.

Scand J lnfcct Dis 24

Cymkines and T-lymphocytes in intraabdominal sepsis 367

(approximately 150 ng/ml) are rather low in comparison to those reached in healthy volunteers after application of LPS (> 250 ng/ml; 4), transplant rejection (> 500 ng/ml; 24) or lethal TNF-alpha levels in experimental animals (up to 20000 nglml; 6). The levels of IL-6 were elevated for the whole observation period, a finding in agreement with studies demonstrating TNF-alpha-independent IL-6-production (25). Our data are consistent with the role of IL-6 as negative modulator of TNF-alpha production as shown by the decline in TNFalpha following IL-6 peak concentration (1 1). To investigate the parameters, which measure immunological functions possibly linked to survival in intraabdominal sepsis, we compared the values for the surviving and nonsurviving patients for the last 12 days prior to discharge or death. The two groups showed no significant differences at the beginning of this observation period. Our data show that, in the presence of similar plasma concentrations of ET, the non-survivors of intraabdominal infections exhibit lower plasma concentrations of TNF-alpha and IL-6 in comparison to surviving patients. T-lymphocytes, in particular T-helperhnducer T-subpopulation are below normal, as well as the functional activity of the T-cells as assessed by stimulation with lectin in vitro. This decrease in CD3 and CD4 cells is known to be an effect of corticosteroids produced in response to E T (26). Increased rates of unstimulated thymidine uptake have been described previously as predictive parameter of sepsis in burn patients and interpreted as overstimulated immune effector cells (27). In summary, the non-survivors are characterized by an anergic immune system, associated with depressed T-lymphocytes and with low levels of TNF-alpha and IL-6, which are necessary for the normal host defense system in low amounts. Therapeutic agents, aiming to neutralize the effects of TNF-alpha, o r inactivate inducers of TNF-alpha, may have adverse effects under these conditions and the therapeutic modalities should try to restore the functional capacity of the immune system in late intraabdominal sepsis. ACKNOWLEDGEMENTS We thank Ms Maria Prettenhofer and Ms H. Zommer for technical assistance

REFERENCES 1. Fry DE, Pearlstein L, Fulton RL, Polk HC. Multiple system organ failure. Arch Surg 1.55: 13G-140, 1980. 2. Polk HC, Shields CL. Remote organ failure: A valid sign of occult intraabdominal infection. Surgery 81: 31(L313, 1977. 3. Marshall J. Microbial infection and the septic response in critical surgical illness. Arch Surg 125: 17-23, 1990. 4. Michie HR, Manogue KR, Spriggs DR, Revhaug A , O’Dwyer S, Dinarello CA, Cerami A , Wolff SH. Willmore DW. Detection of circulating tumor necrosis factor after endotoxin administration. N Engl J Med 318: 1481-1486, 1988. 5 . Helfgott DC, Tatter SB, Santhanam U , Clarick RH, Bhardwaj N, May LT, Sehgal PB. Multiple forms of IFN-beta-2/IL-6 in serum and body fluids during acute bacterial infection. J lmmunol 142: 948-953, 1989. 6. Hesse DG, Tracey KJ, Fong Y , Manogue KR, Pallandino MA, Cerami A , Shires GT, Lowry SE Cytokine appearance in human endotoxemia and primate bacteremia. Surg Gynecol Obstet 166: 147-153, 1988. 7. Brandtzaeg P, Kierulf P, Ganstad P. Skulberg A , Bunner JN, Halvorsen S, Sorensen E. Plasma endotoxin as predictor of multiple organ failure and death in systemic meningococcal disease. J lnfect Dis 159: 195-204, 1989. 8. Girardin E, Grau GE, Dayer JM, Roux-Lombard P, Larnbert PH. Tumor necrosis factor and interleukin-1 in the serum of children with severe infectious purpura. N Engl J Med 319: 397400, 1988. 9. Silva AT, Appelmelk BJ. Buurman WA, Bayston KF, Cohen J. Monoclonal antibody to endotoxin 24 Scand J lnfect DIS

Scand J Infect Dis Downloaded from informahealthcare.com by University of Auckland on 12/04/14 For personal use only.

368 G. Hamilton et al.

Scand J Infect Dis 24

core protects mice from Escherichia coli sepsis by a mechanism independent of tumor necrosis factor and interleukin-6. J Infect Dis 162: 454459, 1990. 10. Bagby GJ, Plessala KJ, Wilson LA, Thompson JJ, Nelson S. Divergent efficacy of antibody to tumor necrosis factor-alpha in intravascular and peritonitis model of sepsis. J Infect Dis 163: 83-88, 1991. I t . Starnes HF Jr. Pearce MK, Tewari A, Yim JH, Zou JC, Abrams JS. Anti-IL-6 monoclonal antibodies protect against lethal Escherichia coli infection and lethal tumor necrosis factor-alpha in mice. J Imrnunol 145: 41854191, 1990. 12. Ziegler EJ, Fisher CJ Jr, Sprung CL, Straube RC, Sadoff JC, Foulke GE, Wortel CH, Fink MP, Dellinger RP, Teng NNH, et al. Treatment of gram-negative bacteremia and septic shock with HA-1A human monoclonal antibody against endotoxin. N Engl J Med 324: 429436, 1991. 13. Christou NV. Host defense mechanisms in surgical patients: a correlative study of the delayed hyperscnsitivity skin test response. granulocyte function and sepsis in 2202 patients. Can J Surg 28: 3 9 4 9 , 1985. 14 Deitch EA. Infection in the compromized host. Surg Clin North Am 68: 181-197, 1988. 15 Miller-Graziano CL. Szabo G , Kodys K, Griffey K . Aberrations in post-trauma monocyte (MO) subpopulation: role in septic shock syndrome. J Trauma 12: 8 6 9 6 , 1090. 16. Fiigger R, Hamilton G, Rogy M, Herbst F, Kwasny W, Schemper M, Schultz F. Prognostic significance of endotoxin determination in patients with severe intraabdominal infection. J Infect Dis 161: 1314-1315. 1990. 17. Frey CF. Bradley EL. Beger HG. Progress in acute pancreatitis. Surg Gynecol Obstet 167: 282-286, 1988. 18. Knaus WA. Draper EA, Wagner DP, Zimmermann JE. APACHE 11: A severity of disease classification system. Crit Care Med 13: 818-829, 1985. 19. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. Prognosis in acute organ-system failure. Ann Surg 202: 685493, 1985. 20. Berger D, Marzinzig E , Marzinzig M, Berger HG. Quantitative endotoxin determination in blood chromogenic modification of the Limulus amoebocyte lysate test. Eur Surg Res 20: 128-136, 1988. 21 van Deventer SJ. Buller HR, ten Cate JW, Aarden LA, Hack CE, Sturk A . Experimental endotoxemia in humans: analysis of cytokine release and coagulation, fibrinolytic, and complement pathways. Blood 76: 252G-2526, 1990. 22 Christou NV. Systemic and peritoneal host defense in peritonitis. World J Surg 14: 184-190, 1990. 23 Echtenacher B, Falk W, Mannel DN, Krammer PH. Requirement of endogenous tumor necrosis factorkachcctin for recovery from experimental peritonitis. J Immunol 145: 3762-3766. 1990. 24 Maury DPJ, Teppo AM. Raised serum levels of cachectinitumor necrosis factor in renal allograft rejection. J Exp Med 166: 1132-1136, 1987. 25 Havell EA, Sehgal PB. Tumor necrosis factor-independent I L 6 production during murine listeriosis. J Immunol 146: 756761, 1991. 26 Richardson RP. Rhyne CD, Fong Y, Hesse DG, Tracey KJ, Marano MA, Lowry SF, Antonacci AC, Calvano SE. Peripheral blood leucocyte kinetics following in vivo lipopolysaccharide (LPS) administration to normal human subjects. Influence of elicited hormones and cytokines. Ann Surg 210: 239-245, 1989. 27 Deitch EA, Landry KN, McDonald JC. Postburn impaired cell-mediated immunity may not be due to lazy lymphocytes but to overwork. Ann Surg 201: 793-802, 1985.

Endotoxin, TNF-alpha, interleukin-6 and parameters of the cellular immune system in patients with intraabdominal sepsis.

The correlation of endotoxin (ET), tumor necrosis factor alpha (TNF-alpha), interleukin-6 (IL-6), and cellular immune parameters with multiple organ f...
559KB Sizes 0 Downloads 0 Views