Postoperative infections are common and are a major cause of morbidity and mortality after different types of surgery (1). Incidence of gram-negative infections in surgical patients is about 30% and is associated with a substantial mortality2,3. The immune response against infections is, in the early phase, mainly determined by the innate immune system. One of its essential parts is the pattern recognition receptors that sense conserved microbial molecules, the so called "pathogen-associated molecular patterns" (4). These molecules originate in pathogenic microorganisms such as bacteria, yeasts, and viruses. Toll-like receptor (TLR) 4 is the main sensor for lipopolysaccharide (LPS) (5), which originates from gram-negative bacteria. Signal transduction via intracellular adaptor proteins and kinases leads to activation of nuclear factors that induce cytokine genes (6). These cytokines are released into the tissue and the bloodstream, exerting their defensive and inflammatory activities. It is widely accepted that LPS sensing and cytokine release is a major contributor to the inflammatory response to gram-negative bacterial infection.
A single nucleotide polymorphism (SNP) on the locus Asp299Gly/Thr399Ile of the TLR4 gene was associated with reduced airway response to inhaled endotoxin in humans, indicating an impaired function (7). Results regarding the effects of this SNP for septic complications have been conflicting. The incidence of gram-negative infections was higher in carriers of this SNP after elective surgery (8), but incidence of sepsis or mortality was not affected (9). In medical intensive care unit (ICU) patients, there was a predisposition to develop septic shock with gram-negative microorganisms (10), but in contrast, the same SNP has a protective effect against Legionella infections (11). In recent studies, there were conflicting data concerning the effects of the TLR4 polymorphisms on cytokine release in different populations upon stimulation with LPS in vitro. In healthy volunteers, as well as in surgical patients, the response to LPS was not affected (12,13).
Generally, in surgical patients, the LPS-induced release of proinflammatory cytokines such as TNF-α or IL-6 is decreased over several days after surgery (14). Similar results could be seen with ICU patients with sepsis where the diminished cytokine release was a risk factor for mortality, suggesting that the ability to respond to infectious agents is impaired (15).
To test the hypothesis that the common TLR4 SNP Asp299Gly/Thr399Ile leads to higher mortality and a difference in cytokine response, we analyzed 62 patients in a observational cohort study. All patients were scheduled to have major gastrointestinal surgical procedures, predominantly tumor resections. Secondarily, we performed cytokine analysis after LPS in vitro stimulation and we assessed incidence of infections and inflammatory response and severity of the postoperative course. Complete sets of laboratory tests were obtained in 36 individuals. Their clinical characteristics were comparable with the studied cohort of 62 patients.
MATERIALS AND METHODS
Patients enrolled in this prospective blinded cohort study were all unrelated European Caucasians scheduled to undergo major surgical procedures for cancer therapy. This study was approved by the local ethics committee. Inclusion criteria were informed consent of the patients and an age over 18 years. Exclusion criteria were disseminated tumor disease, acute or chronic hematological malignancies, chronic steroid medication (>0.5 mg/kg prednisolone or any equivalent glucocorticoid for more than 3 days preoperatively), or other immunomodulatory medication. Age, sex, site of infection, relevant microorganisms detected, type of surgery, development of systemic inflammatory response syndrome (SIRS), and infectious complications were recorded. Definition of SIRS and septic complications (sepsis, severe sepsis, and septic shock) were based on published criteria (16). Sepsis was defined as the presence of criteria for SIRS in response to a documented or clinically suspected acute infection. Severe sepsis was defined as sepsis associated with evidence of hypoperfusion with organ dysfunction or sepsis-induced hypotension. Septic shock was defined as sepsis with sepsis-induced hypotension requiring vasopressor therapy despite adequate fluid challenge along with the presence of hypoperfusion and organ dysfunction. Severity of disease was assessed on a daily basis by the Sequential Organ Failure Assessment (SOFA) Score and Simplified Acute Physiology Score (SAPS II) on admission (17, 18). Patients were followed until discharge from the hospital. All patients received standard supportive care. Antimicrobial therapy was carried out according to published standards and was adjusted to the microbiological diagnosis. Investigators blinded to the patients' genotype and clinical course performed data collection and sequence analysis. Patient characteristics are shown in Table 1).
DNA extraction and genotypic analysis
Buffy coats prepared by Ficoll density gradient centrifugation were used for DNA preparation carried out by Agova (Berlin, Germany) from blood cells. For TLR4 genotyping, we performed a PCR reaction using the Lightcycler/FRET method described by us (19). In brief, DNA was extracted from whole blood and PCR was performed in a volume of 20 μL containing 5 μL of DNA (2-4 ng/μL). For TLR4 genotyping, primers were used at 0.25 μmol/L, and PCR was performed in the presence of 4 mM MgCl2, 2 μL of 10× Lightcycler DNA master hybridization probes (Roche Diagnostics, Mannheim, Germany), and 0.2 μmol/L of each fluorescence probe. The sensor probe covering the Asp299Gly polymorphism was labeled with fluorescein 3′. The corresponding anchor probe was labeled with Lightcycler Red 640 5′. The sensor probe covering the Thr399Ile polymorphism was labeled with Lightcycler Red 705 5′. The corresponding anchor probe was labeled with fluorescein 3′.
For the whole blood LPS stimulation assay and for baseline cytokine measurements, 5 mL of venous blood was obtained and collected in endotoxin-free tubes containing EDTA as anticoagulant (Chromogenix, Molndal, Sweden). After separation for baseline cytokine measurement, LPS-induced TNF-α and IL-6 release was assessed according to the manufacturer's procedures using an ex vivo TNF-α release test (Milenia, Bad Nauheim, Germany). The supernatants were removed and assayed for TNF-α and IL-6 content by an enzyme-linked immunoabsorbant assay-based automated method with a detection limit of <4.8 pg/mL for TNF-α and <4 pg/mL for IL-6 (IMMULITE system; DPC Biermann, Bad Nauheim, Germany) according to the manufacturer's instructions. Briefly, 500 μL of blood was incubated with 50 pg/mL LPS (Salmonella abortus equii, strain ATCC 9842, rough LPS) in RPMI 1640 medium for 4 h, followed by 5 min of centrifugation at 1000g. Supernatants were frozen and stored at -70°C for later examination.
Categorical data (TLR4 genotype, sex ratio, type of surgery, and coexisting diseases) were analyzed for differences between the patient groups by means of chi square exact test. For expected frequencies of less than 5, the Fisher exact test two-tailed test was applied. For numerical data (cytokine levels, score data [SAPS II and American Society of Anesthesiologists], and age), we used the Mann-Whitney U test to determine differences between the study groups. For differences in bound variables, the Wilcoxon rank sum test was applied. For statistical calculation, the SPSS for Windows software, release 12.0 (Chicago, IL), was used. A two-tailed P < 0.05 was considered significant.
Incidence of the TLR4 polymorphism
Of 62 patients examined, nine were heterozygous for the Asp299Gly/Thr399Ile TLR4 polymorphism (14.5%; Table 1). None of the patients was homozygous. These frequencies were according to other previous studies and also to our own previous observations (20).
Morbidity and mortality
As shown in Table 1), overall mortality, rate of infectious complications, and sepsis in the two groups were comparable. Patient mortality was 5.7% (3/53) in the wild-type (WT) group and 11.1% (1/9) in the TLR4 SNP group (P = 0.475). The rate of infectious complications was similar in both groups, with 4/9 in the TLR4 SNP group (44.4%) and 16/53 (30.2%) in the WT group (P = 0.453). SIRS developed in 77.8% (7/9) patients in the TLR4 SNP group and 71.7% (38/53)patients in the WT group (P = 1.000). No effect of the TLR4 polymorphism on the rate of gram-negative infections was detectable (3/9) in the TLR4 SNP group (33.3%) and 12/53 (18.9%) in the WT group (P = 0.409). The overall length of ICU stay in the polymorphism group was equal to the WT group (8 days). Severity of disease was measured with the maximum SOFA score and the SAPS Score on admission and was similar for the respective values (SOFA [max] 5.5 in the TLR4 SNP group and 4.1 in the WT group with P = 0.608 and SAPS score on admission 21.3 in the TLR4 SNP group and 20.0 in the WT group with P = 0.531).
We performed cytokine analysis in 30 WT patients and six patients with the TLR4 polymorphism preoperatively (T1) and after 60 to 102 h (T2) postoperatively. The study design is depicted in Figure 1). Pre- and postoperative cytokine values were as follows. TNF-α and IL-6 were low in both groups preoperatively, with almost 80% of all tests below the detection level. Postoperatively, only few patients showed elevated TNF-α (mean 5.8 pg/mL) when over the lower detection level, whereas IL-6 rose from 10.43 pg/mL (SD 9.45) to 69.43 pg/mL (SD 118.64) in the TLR-4 group (P = 0.313) and from 16.13 pg/mL (SD 53.88) to 66.05 pg/mL (SD 120.25) in the WT group (P < 0.001). There was no difference between the two groups at either time point.
Ex vivo cytokine release
In both groups, there was a substantial increase in TNF-α and IL-6 concentrations after stimulation with LPS at both time points. The release of TNF-α was comparable in the two groups at T1 422.97 pg/mL (SD: 662.57; range: 4.00-3816) WT group versus 310.83 pg/mL (SD: 117.53; range 112.00-453.00) TLR4 SNP group, P = 0.85. There was no difference in the release of IL-6 preoperatively with 298.20 pg/mL (SD: 202.75; range 22.60-659.00) WT group versus 266.90 pg/mL (SD: 139.63; range 91.40-425.00) TLR4 SNP group, P = 0.98. At T2, the LPS-induced release of TNF-α was markedly decreased in both groups: 191.68 pg/mL (SD: 147.26; range 21.40-667.00) WT group versus 134.08 pg/mL (SD: 91.49; range 41.50-298:00) TLR4 SNP group, with no difference when comparing the two groups (P = 0.42). The same reaction was detectable for IL-6: 135.08 pg/mL (SD: 118.07; range 5.40-525.00) WT group versus 125.32 pg/mL (SD: 91.81; range 23.50-271.00) TLR4 SNP group with P = 0.98 for the comparison. As shown in Figure 2), the release of TNF-α and IL-6 was significantly diminished at T2 for WT patients (TNF-α, P < 0.001 and IL-6, P < 0.001) and for TLR4 patients (TNF-α, P < 0.031 and IL-6, P < 0.031).
A comparison of these patients with respect to the development of inflammatory responses is shown in Figure 3). Cytokine release before and after surgery grouped by different disease entities shows the overall tendency of diminished cytokine releases. Preoperative TNF-α values in WT patients were 637.12 pg/mL (SD 1016.95) for SIRS, 238.00 pg/mL (SD 111.63) for sepsis, 133.33 pg/mL (SD 113.96) for severe sepsis, and 520.00 pg/mL for septic shock. Postoperative values for TNF-α in these patients were 209.48 pg/mL (SD: 191.86) for SIRS, 147.90 pg/mL (SD 85.37) for sepsis, 235.00 pg/mL (SD 98.79) for severe sepsis, and 421.00 pg/mL for septic shock. The trend over all WT patients shows preoperative TNF-α values at 297.11 pg/mL (SD 200.00) and postoperatively at 267.97 pg/mL (SD 139.50). Preoperative TNF-α values in TLR4 patients with SIRS were 393 pg/mL (SD 51.97), 112 pg/mL with sepsis, 324 pg/mL with severe sepsis, and 250 pg/mL with septic shock. Postoperative values for TNF-α in the TLR4 patients with SIRS were 136.67 pg/mL (SD 8.33), 41.5 pg/mL with sepsis, 298 pg/mL with severe sepsis, and 55 pg/mL with septic shock. As well, the trend over all TLR4 patients shows preoperative TNF-α at 228.67 pg/mL (SD 107.60) and postoperative TNF-α values at 131.50 pg/mL (SD 144.35).
Preoperative IL-6 values in WT patients were 320.76 pg/mL (SD 208.10) for SIRS, 207.95 pg/mL (SD 181.50) for sepsis, 117.93 pg/mL (SD 121.67) for severe sepsis, and 577 pg/mL for septic shock, and IL-6 postoperatively in WT patients, values were 133.94 pg/mL (SD 141.78) for SIRS, 127.23 pg/mL (SD 108.64) for sepsis, 121.67 pg/mL (SD 18.56) for severe sepsis, and 416 pg/mL for septic shock. The trend also decreased from 300.96 pg/mL (SD 243.26) to 221.63 pg/mL (SD 168.35). In TLR4 patients, preoperative IL-6 values for SIRS were 369.67 pg/mL (SD 50.85),91.40 pg/mL for sepsis, 303 pg/mL for severe sepsis, and 98 pg/mL for septic shock. The postoperative values for IL-6 in TLR4 patients were 139.13 pg/mL (SD 36.78) for SIRS, 23.50 pg/mL for sepsis, 271 pg/mL for severe sepsis, and 40.00 pg/mL for septic shock. As well as the WT patients, the trend over all TLR4 patients was, for IL-6, preoperative, 164.13 pg/ml (SD:120.31) and postoperative, 111.50 pg/mL (SD 138.38). Although the subgroups, particularly in the TLR4 SNP group, are small, this seems to be a persistent pattern. There is a trend toward an overall diminished IL-6 release in the TLR4 group, which is not statistically significant. Patients with no inflammatory response postoperatively were excluded. When adjusting the cytokine values for white blood cell count at the respective time points, there was also no difference between the two groups (data not shown).
Recognition of pathogens by TLRs triggers signaling pathways mediated by the Toll/IL-1 receptor homologous region (TIR) domain-containing adaptors such as myeloid differentiation response factor 88 (MyD88), MyD88 adaptor-like/TIR-associated protein (MAL/TIRAP), Toll receptor-associated molecule (TRAM), and Toll receptor-associated activator of interferon (TRIF) (21). The activation of the TLR4 heterotrimeric complex with CD14 and MD-2 is caused by LPS and potentially endogenous mediators such as fibronectin, heat shock protein 60 (hsp60), and other ligands (22). After activating, intracellular signal transduction nuclear factor κB (NF-κB) is translocated into the nucleus and cytokine release follows. The consecutively released cytokines exert the acute-phase response to the inflammatory stimuli such as infection, surgery, or trauma.
Frequent polymorphisms in the TLR4 locus (Asp299Gly and Thr399Ile) were shown to influence the incidence of gram- negative infections in surgical patients (8). In septic shock caused by gram-negative microorganisms, the incidence of TLR4 SNPs was found to be higher (10). Therefore, the possible reduced function in the endotoxin receptor seems to potentially impair innate immune response to infections.
Certain types of surgical procedures, especially tumor resections, carry a high risk of developing SIRS (23). Surgical procedures, as well as other inflammatory triggers such as multiple trauma, lead to a diminished cytokine release after in vitro LPS stimulation in humans by causing SIRS (24, 25). This endotoxin hyporesponsiveness is detected up to 10 days after the insult (15). Thus, in the later postoperative course, cytokine response is influenced and immune response could be critically impaired, leading to higher susceptibility to infections after surgery. It has not yet been examined whether frequent SNPs of the TLR4 gene are associated with this endotoxin hyporesponsiveness after surgery. Therefore, we tested the cytokine release in normal WT patients versus carriers of the TLR4 polymorphism in a setting with a high probability of inflammatory response and infection.
Cytokine release after LPS stimulation in healthy volunteers was not affected when comparing carriers of the TLR4 polymorphism with WT patients12, 13, 26). The same results could be seen shortly after cardiac surgery (2 h) with similarly changed endotoxin-induced cytokine production comparing TLR4 SNP patients and WT patients (27). Our data show that in the later course of inflammatory responses after major gastrointestinal surgical procedures, there is also a significant decrease in the release of proinflammatory cytokines after LPS stimulation. These results may be different when looking at major surgery not crossing mucous membranes such as vascular surgery or some types of thoracic surgery.
Because many postoperative infections develop after surgical procedures, it is of interest whether changes in the ability to produce proinflammatory cytokines as a result of infection depend on the TLR4 SNP. We show that there is no difference in cytokine release postoperatively in surgical patients with the TLR4 polymorphism as compared with WT patients, indicating that the mechanisms responsible for higher susceptibility of TLR4 mutation carriers for gram-negative infections do not exclusively depend on the function of the TLR4. In general, patients with the TLR4 SNP show the same hyporesponsiveness to LPS after surgery or trauma as do WT patients.
The comparison of the different degrees of infection severity with a tendency toward higher mortality in the TLR4 SNP group (11%) than in the WT group (6%) as well as the higher frequency of septic complications (45% versus 30%) and the higher incidence of gram-negative infections (33% versus 19%), however, could potentially indicate a mild influence of this SNP. The relatively small sample of our observational study is certainly of concern for any statistical calculation. The power analysis revealed that to achieve statistical significance for 20% difference in mortality, at least 500 patients had to be included in a prospective study.
In addition, we observed a potential influence of different inflammatory states such as SIRS, sepsis, severe sepsis, and septic shock on ex vivo cytokine release (Fig. 3). We observed a more pronounced postsurgical decrease in TNF-α and IL-6 release in noninfected versus infected patients. As presented in Figure 3), inflammatory response and infection lead to changed TNF-α and IL-6 release with a more pronounced decrease in noninfected versus infected patients. In patients with severe sepsis, even an increase of cytokine induction comparing pre- and postsurgical stimulation is detectable, but clearly, the small number of patients does not allow a meaningful statistical analysis.
Several reasons may exist to explain the hyporesponsiveness. One possible mechanism is an altered surface expression of TLR4. It has been shown, however, that surface expression of the TLR4 on monocytes as well as TLR4 mRNA is upregulated in patients with inflammatory states or infections compared with healthy controls28, 29). No data exist up to now concerning the effect of the TLR4 SNP at the TLR4 locus on its change of surface expression. Thus, potentially, other factors such as binding of endotoxin or changes of coproteins might counteract the upregulated expression.
Furthermore, differences in the concentrations of LPS released from bacteria depend on their species or subtypes (30). This is especially important because there is an. indication of dose dependence for endotoxin as shown in recently published stimulation experiments in patients after cardiac surgery (27). There is as well a possibility that whole blood is not the main compartment of action for LPS sensing because most infections are localized in certain tissues, such as pneumonia or peritonitis in patients after surgery (31).
Spreading of the microbes after initial sensing might already have influenced TLR4 function through the above-mentioned regulative mechanisms. It is of interest that already preoperatively there seems to be a blunted response to LPS in a whole blood assay in surgical patients who were developing postoperative infections, indicating an ongoing acute-phase response that might lead to immunocompromise in these patients (32).
In conclusion, patients with the TLR4 SNP have no change in postoperative cytokine production after in vitro LPS stimulation compared with WT patients. The TLR4 SNP does not influence the LPS hyporesponsiveness as a possible adaptive reaction detected in our patients postoperatively.
The authors thank Guntram Schulze for his assistance with laboratory tests. We also thank Christoph Ulmer and Tobias Schulze for support in preparing the stimulation assay. Furthermore, we thank Diane Wöllner and Fränzi Creutzburg for excellent technical assistance in DNA preparation and genotyping.
1. Pessaux P, Msika S, Atalla D, Hay JM, Flamant Y: Risk factors for postoperative infectious complications in noncolorectal abdominal surgery: a multivariate analysis based on a prospective multicenter study of 4718 patients. Arch Surg
2. Wallace WC, Cinat ME, Nastanski F, Gornick WB, Wilson SE: New epidemiology for postoperative nosocomial infections. Am Surg
3. Cohen J, Cristofaro P, Carlet J, Opal S: New method of classifying infections in critically ill patients. Crit Care Med
4. Beutler B: Innate immunity: an overview. Mol Immunol
5. Poltorak A, He X, Smirnova I, Liu MY, Van Huffel C, Du X, Birdwell D, Alejos E, Silva M, Galanos C, Freudenberg M, Ricciardi-Castagnoli P, Layton B, Beutler B: Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4
6. Takeda K, Akira S: TLR signaling pathways. Semin Immunol
7. Arbour NC, Lorenz E, Schutte BC, Zabner J, Kline JN, Jones M, Frees K, Watt JL, Schwartz DA: TLR4
mutations are associated with endotoxin
hyporesponsiveness in humans. Nat Genet
8. Agnese DM, Calvano JE, Hahm SJ, Coyle SM, Corbett SA, Calvano SE, Lowry SF: Human toll-like receptor 4 mutations but not CD14 polymorphisms are associated with an increased risk of gram-negative infections. J Infect Dis
9. Feterowski C, Emmanuilidis K, Miethke T, Gerauer K, Rump M, Ulm K, Holzmann B, Weighardt H: Effects of functional toll-like receptor-4 mutations on the immune response to human and experimental sepsis. Immunology
10. Lorenz E, Mira JP, Frees KL, Schwartz DA: Relevance of mutations in the TLR4
receptor in patients with gram-negative septic shock. Arch Intern Med
11. Hawn TR, Verbon A, Janer M, Zhao LP, Beutler B, Aderem A: Toll-like receptor 4 polymorphisms are associated with resistance to Legionnaires' disease. Proc Natl Acad Sci USA
12. Heesen M, Bloemeke B, Kunz D: The cytokine synthesis by heterozygous carriers of the toll-like receptor 4 Asp299Gly polymorphism does not differ from that of wild type homozygotes. Eur Cytokine Netw
13. von Aulock S, Schroder NW, Gueinzius K, Traub S, Hoffmann S, Graf K, Dimmeler S, Hartung T, Schumann RR, Hermann C: Heterozygous toll-like receptor 4 polymorphism does not influence lipopolysaccharide-induced cytokine release in human whole blood. J Infect Dis
14. Haupt W, Riese J, Mehler C, Weber K, Zowe M, Hohenberger W: Monocyte function before and after surgical trauma. Dig Surg
15. Ertel W, Kremer JP, Kenney J, Steckholzer U, Jarrar D, Trentz O, Schildberg FW: Down-regulation of proinflammatory cytokine release in whole blood from septic patients. Blood
16. Bone RC, Sibbald WJ, Sprung CL: The ACCP-SCCM consensus conference on sepsis and organ failure. Chest
17. Vincent JL, Moreno R, Takala J, Wilatts S, De Mendonca A, Bruining H, Reinhart CK, Suter PM, Thijs LG: The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-related Problems of the European Society of Intensive Care Medicine. Intensive Care Med
18. Le Gall JR, Lemeshow S, Saulnier F: A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study. JAMA
19. Hamann L, Hamprecht A, Gomma A, Schumann RR: Rapid and inexpensive real-time PCR for genotyping functional polymorphisms within the toll-like receptor -2, -4, and -9 genes. J Immunol Methods
20. Schroder NW, Schumann RR: Single nucleotide polymorphisms of toll-like receptors and susceptibility to infectious disease. Lancet Infect Dis
21. Beutler B: Inferences, questions and possibilities in toll-like receptor signalling. Nature
22. Takeda K, Akira S: Microbial recognition by toll-like receptors. J Dermatol Sci
23. Haga Y, Beppu T, Doi K, Nozawa F, Mugita N, Ikei S, Ogawa M: Systemic inflammatory response syndrome and organ dysfunction following gastrointestinal surgery. Crit Care Med
24. Ordemann J, Jacobi CA, Schwenk W, Stosslein R, Muller JM: Cellular and humoral inflammatory response after laparoscopic and conventional colorectal resections. Surg Endosc
25. Majetschak M, Flach R, Kreuzfelder E, Jennissen V, Heukamp T, Neudeck F, Schmit-Neuerburg KP, Obertacke U, Schade FU: The extent of traumatic damage determines a graded depression of the endotoxin
responsiveness of peripheral blood mononuclear cells from patients with blunt injuries. Crit Care Med
26. Erridge C, Stewart J, Poxton IR: Monocytes heterozygous for the Asp299Gly and Thr399Ile mutations in the toll-like receptor 4 gene show no deficit in lipopolysaccharide signalling. J Exp Med
27. Schippers EF, van't Veer C, van Voorden S, Martin CA, le Cessie S, van Dissel JT: TNF-α promoter, Nod2 and toll-like receptor-4 polymorphisms and the in vivo and ex vivo response to endotoxin
28. Armstrong L, Medford AR, Hunter KJ, Uppington KM, Millar AB: Differential expression of toll-like receptor (TLR)-2 and TLR-4 on monocytes in human sepsis. Clin Exp Immunol
29. Harter L, Mica L, Stocker R, Trentz O, Keel M: Increased expression of toll-like receptor-2 and -4 on leukocytes from patients with sepsis. Shock
22: 403-409, 2004.
30. Kido N, Nakashima I, Kato N: Correlation between strong adjuvanticity of klebsiella O3 lipopolysaccharide and its ability to induce interleukin-1 secretion. Cell Immunol
31. Kimura Y, Yaegashi Y, Sato N: Tumor necrosis factor-α production after esophageal cancer surgery: differences in the response to lipopolysaccharide stimulation among whole blood, pleural effusion cells, and bronchoalveolar lavage fluid cells. Surg Today
32. Haupt W, Zirngibl H, Klein P, Riese J, Hohenberger W: Reduced TNF-α and IL-6 production in patients who mount a preoperative acute phase response. Langenbecks Arch Surg
Postoperative sepsis; SNP; TLR4; cytokine induction; endotoxin