- Email: [email protected]
Release of anti-inflammatory mediators after major torso trauma correlates with the development of postinjury multiple organ failure
Release of Anti-Inflammatory Mediators after Major Torso Trauma Correlates with the Development of Postinjury Multiple Organ Failure David A. Partrick, MD, Ernest E. Moore, MD, Denver, Colorado, Frederick A. Moore, MD, Houston, Texas Walter L. Biffl, MD, Carlton C. Barnett, Jr., MD, Denver, Colorado
BACKGROUND: Soluble tumor necrosis factor receptor (sTNFr) and interleukin-1 receptor antagonist (IL-1ra) have been identified as endogenous inhibitors of TNF-␣ and IL-1. While TNF-␣ and IL-1 levels are not systematically elevated in postinjury patients who developed multiorgan failure (MOF), their involvement at the tissue level has been suggested. Our study hypothesis was that levels of sTNFr-I and IL-1ra would discriminate patients at risk for postinjury MOF. METHODS: Serial plasma levels of sTNFr and IL1ra were measured in 29 trauma patients at high risk for postinjury MOF. RESULTS: sTNFr-I levels were higher in MOF compared with non-MOF patients at 12, 84, and 132 hours postinjury. MOF patients also had higher IL-1ra values 36, 60, 84, and 132 hours postinjury. CONCLUSIONS: Anti-inflammatory mechanisms are activated after trauma. Since increased levels of sTNFr and IL-1ra correlate with postinjury MOF, they may contribute to our understanding of the pathogenesis as well as prediction of outcome. High levels of antagonists to TNF-␣ and IL-1 suggest tissue level involvement of these cytokines in postinjury hyperinflammation. Am J Surg. 1999;178:564 –569. © 1999 by Excerpta Medica, Inc.
M
ultiple organ failure (MOF) remains the most common cause of delayed mortality in surgical intensive care units. Once established, MOF defies our standard critical care supportive measures and mortality exceeds 50%. Despite intensive investigation, the pathogenesis of postinjury MOF remains unclear.1,2
From the Department of Surgery, (DAP, EEM, WLB, CCB), Denver Health Medical Center, University of Colorado Health Sciences Center, Denver, Colorado; and the Department of Surgery (FAM), UT-Houston Medical School, Houston, Texas. This work was supported in part by National Institutes of Health Grants P50GM49222-01 and T32GM08315-03. Requests for reprints should be addressed to Ernest E. Moore, MD, Chief, Department of Surgery, Denver Health Medical Center, 777 Bannock Street, Denver, Colorado 80204. Presented at the 51st Annual Meeting of the Southwestern Surgical Congress, Coronado, California, April 18 –21, 1999.
564
© 1999 by Excerpta Medica, Inc. All rights reserved.
Our global hypothesis is that postinjury MOF occurs as the result of a dysfunctional inflammatory response.1,3 Following major tissue injury and hemorrhagic shock, patients develop an early physiologic state of systemic hyperinflammation, referred to as the systemic inflammatory response syndrome (SIRS). Certain patients appear to be vulnerable (ie, primed) such that early secondary insults amplify SIRS, which produces organ dysfunction ultimately culminating in MOF. We have previously demonstrated that systemic levels of the cytokines interleukin-6 (IL-6) and interleukin-8 (IL-8) correlate with MOF following severe injury.4 The role of the other proinflammatory cytokines tumor necrosis factoralpha (TNF-␣) and interleukin-1 beta (IL-1) in the systemic inflammatory response syndrome leading to postinjury MOF, however, remains unclear. TNF-␣, produced from the monocyte/macrophage lineage, is well established as a pivotal cytokine responsible for the clinical manifestations of shock induced by endotoxin or bacteremia leading to SIRS following sepsis.5,6 Activation of the TNF-␣ regulatory protein NFB has been documented in alveolar macrophages from patients with adult respiratory distress syndrome (ARDS).7 However, TNF-␣ levels in trauma patients have not been found to be elevated, even in patients who eventually developed MOF.8 –10 Similarly, IL-1 is a proinflammatory cytokine that shares many of the proinflammatory properties of TNF-␣5,11 Indeed, TNF and IL-1 not only stimulate their own release, but also the release of each other (as well as IL-6 and IL-8), thus amplifying the cascade of inflammatory mediators.12 IL-1 has two bioactive forms: IL-1␣ and IL-1. IL-1 is the predominant IL-1 and a major product of activated human monocytes, tissue macrophages, and neutrophils. Local production of IL-1 is significantly increased following trauma as evidenced by increased levels in bronchoalveolar lavage fluid, but it has not been found elevated in the circulation.10 This raises the possibility that the local release of these proinflammatory cytokines may significantly differ from their blood levels. Naturally occurring inhibitors of these two proinflammatory cytokines, produced by macrophages and neutrophils, have been well characterized. Soluble TNF receptors (sTNFr) compete with membrane receptors for binding of free TNF-␣, thus eliminating TNF bioactivity that would otherwise occur with TNF receptor ligation.13 IL-1 receptor antagonist (IL-1ra) is structurally related to IL-1, binds to the same receptor, but does not induce a discernible biological response.14 In vitro experiments have demonstrated 0002-9610/99/$–see front matter PII S0002-9610(99)00240-8
RELEASE OF ANTI-INFLAMMATORY MEDIATORS AFTER TORSO TRAUMA/PARTRICK ET AL
that sTNFr protects against excessive TNF activity13 and that IL-1ra prevents endotoxin-induced lung injury in rats15 and improves survival after endotoxemia in mice.16 sTNFr and IL-1ra have been detected in increased serum concentrations after severe injury,8,10,17 but it has not been established if systemic levels of these anti-inflammatory mediators correlate with postinjury organ failure in humans. Our study hypothesis was circulating levels of sTNFr-I and IL-1ra would discriminate patients at risk for the development of postinjury MOF.
MATERIALS AND METHODS Study Population During a 16-month period, patients admitted to Denver Health Medical Center between 18 and 65 years of age who had an Injury Severity Score (ISS) ⱖ25 or the combination of an ISS ⬎15 and a transfusion requirement of more than 6 units of blood within the first 12 hours were considered eligible for this study. Previous epidemiological studies at our institution had identified these patients as being high risk for the development of MOF.18 Patients with a major head injury (Glasgow Coma Scale score ⬍8) or significant preexisting disease were excluded in an attempt to accrue a homogeneous study population. Additionally, patients transferred from other hospitals or with time from injury to our emergency department admission exceeding 3 hours were excluded. Informed consent was obtained via a protocol approved by the Colorado Multiple Institutional Review Board, University of Colorado Health Sciences Center, Denver, Colorado. A pool of 6 healthy adult volunteers served as controls. Demographic, risk factor, and outcome data for these patients were collected as part of an ongoing, prospective study funded by the National Institutes of Health to investigate the pathogenesis of MOF after injury. The diagnosis of MOF was based on our standard score described previously.18 Briefly, four organ functions (pulmonary, hepatic, renal, and cardiac) were graded daily on a scale of zero to three. MOF was defined as a sum of the individual organ grades obtained simultaneously that was greater than or equal to four. Organ dysfunction scores obtained within the first 48 hours were not used to define organ failure because they may reflect reversible derangement induced by the inciting event or incomplete resuscitation. Sample Collection All patients were resuscitated using a previously described protocol employing defined end points.19 Patients requiring surgical intervention underwent standard surgical and postoperative intensive care unit treatment provided by the same surgical team. The first blood sample was collected from each patient within 12 hours after the estimated time of injury. Three subsequent blood samples were obtained at 24 hours intervals (36, 60, and 84 hours) postinjury and the final sample was collected at 132 hours postinjury (48 hours after previous sample). Eight milliliters of venous blood were collected at each time point in a sterile, pyrogen-free glass tube containing disodium ethylenediamine-tetraacetic acid (EDTA). These samples were first centrifuged at 300g for 10 minutes and then spun in an ultracentrifuge for 5 additional minutes to obtain platelet-
poor plasma. The plasma was removed and stored at ⫺70°C in a sterile, pyrogen-free tube. Immunologic Assays sTNFr: There are two different soluble forms of the TNFr: a 55kD receptor (sTNFr-I) and a 75kD receptor (sTNFr-II). In vitro data support a greater inhibitory role for sTNFr-I in neutralizing the cytotoxic effects of TNF␣,13 and in vivo data has demonstrated sTNFr-I to be more predictive of mortality following sepsis.20 Therefore, a commercially available enzyme-linked immunosorbent assay (ELISA) kit (Bender Med System, Vienna, Austria) for the quantitative measurement of human sTNFr-I (p55) was used. Ninety-six well polystyrene microtiter plates precoated with a monoclonal antibody directed against human recombinant sTNFr were employed. A series of twofold dilutions of recombinant human sTNF ranging from 100 ng/mL to 1 ng/mL was performed to generate a standard curve for each assay. sTNFr in the sample or standard was quantified by an enzymatic reaction, resulting in a detectable color change. The absorbance was measured using a microtiter plate-reading spectrophotometer (THERMOmax; Molecular Devices Corporation, Menlo Park, California) set at 450 and 550 nm. All samples and standards were run in duplicate and mean values were used in calculating sample concentrations using a linear regression analysis. The lower limit of detection for sTNFr was 0.4 ng/mL. IL-1ra: IL-1ra was measured because it is a recognized anti-inflammatory cytokine that prevents organ injury in vitro15 and may improve survival in the setting of sepsis.16 A commercially available ELISA kit (R&D Systems, Minneapolis, Minnesota) for the quantitative measurement of human IL-1ra was used. The assay was similar to that process described for sTNFr except wells were precoated with a monoclonal antibody directed against human IL-1ra and the standard series ranged from 600 ng/mL to 0.5 ng/mL using recombinant human IL-1ra. The lower limit of detection of IL-1ra was 48 pg/mL. Statistical Analysis Nominal data (eg, gender, mechanism of injury, and MOF outcome) were compared by Fisher’s exact test, injury severity score by Student’s t test, and continuous data (eg, length of intensive care unit stay, and plasma levels) were compared by an analysis of variance (ANOVA) using Scheffe’s F procedure for post-hoc comparisons. A P value of ⬍0.05 was considered statistically significant.
RESULTS Patient Population Twenty-nine patients from those enrolled in the NIH study had plasma samples available over the entire 5-day postinjury time period and comprise the study population. The mean age of the study patients was 33.6 ⫾ 2.0 years; 21 (72%) were male, the injury mechanism was blunt in 22 (76%), and mean ISS was 30.5 ⫾ 1.5. Nine (31%) of the 29 study patients developed MOF. The Table summarizes the demographics, MOF risk factors, and clinical outcome for MOF versus non-MOF study subjects. There were no significant differences in age, gender, mechanism of injury,
THE AMERICAN JOURNAL OF SURGERY® VOLUME 178 DECEMBER 1999
565
RELEASE OF ANTI-INFLAMMATORY MEDIATORS AFTER TORSO TRAUMA/PARTRICK ET AL
TABLE Demographics, Multiorgan Failure (MOF) Risk Factors, and Clinical Outcomes of Study Patients Who Did or Did Not Develop MOF Patient Variables Age (years) Male Blunt mechanism ISS ED SBP (mm Hg) URBC (first 12 h) Highest BD (0–12 h) Highest lactate (mmol/L) (12–24 h) Minor septic complications Major septic complications Nonseptic complications Length ICU stay (days) Death
MOF (n ⴝ 9)
Non-MOF (n ⴝ 20)
P Value
36.3 ⫾ 4.1 33.0 ⫾ 2.5 NS 6 (67%) 15 (75%) NS 8 (89%) 14 (70%) NS 33.8 ⫾ 3.6 28.9 ⫾ 1.5 NS 111.2 ⫾ 6.7 107.8 ⫾ 6.6 NS 17.7 ⫾ 4.7 8.1 ⫾ 1.8 ⬍0.05 18.2 ⫾ 4.4 12.2 ⫾ 1.6 NS (P ⫽ 0.15) 5.8 ⫾ 1.1 3.3 ⫾ 0.7 NS (P ⫽ 0.06) 7 (78%) 4 (20%) ⬍0.05 9 (100%) 3 (15%) ⬍0.0001 9 (100%) 8 (40%) ⬍0.05 32.2 ⫾ 5.5 6.7 ⫾ 1.7 ⬍0.0001 2 (22%) 0 NS (P ⫽ 0.09)
ED SBP ⫽ systolic blood pressure in emergency department; URBC ⫽ units of red blood cells transfused; BD ⫽ base deficit (mEq/L); ICU ⫽ intensive care unit; NS, not significant (ie, P ⬎0.05).
injury severity score, or systolic blood pressure in the emergency department. However, compared with the non-MOF patients, MOF patients required more early blood transfusions, developed more septic and nonseptic complications, and stayed longer in the intensive care unit. Furthermore, there was a trend toward increased base deficit and lactate as well as increased mortality in MOF patients. sTNFr Response to Injury Analyzing all 29 study patients (Figure 1A), sTNFr levels were significantly elevated at all time points compared with the normal controls (1.8 ⫾ 0.3 ng/mL). However, there was no significant change in sTNFr levels over the study time course (12 hours, 8.1 ⫾ 0.8 ng/mL; 36 hours, 8.2 ⫾ 1.0 ng/mL; 60 hours, 7.6 ⫾ 1.2 ng/mL; 84 hours, 8.6 ⫾ 1.4 ng/mL; and 132 hours postinjury, 11.1 ⫾ 1.7 ng/mL). In comparison, there was a significant difference between sTNFr levels for the MOF and non-MOF patients (Figure 1B). sTNFr levels were significantly higher (approximately 1.5 times greater) in the MOF patients compared with the non-MOF patients at 12 hours postinjury (11.3 ⫾ 2.2 ng/mL versus 6.8 ⫾ 0.6 ng/mL). At 36 and 60 hours postinjury there were no significant differences between MOF and non-MOF patients (10.1 ⫾ 2.0 ng/mL versus 7.5 ⫾ 1.1 ng/mL and 10.2 ⫾ 4.0 ng/mL versus 6.7 ⫾ 0.7 ng/mL, respectively). However, at 84 and 132 hours postinjury, sTNFr levels in the MOF patients increased again and were significantly elevated (approximately twofold higher at 84 hours and threefold higher 5 days postinjury) compared with non-MOF patients (13.8 ⫾ 4.2 ng/mL versus 6.5 ⫾ 0.8 ng/mL at 84 hours, 19.9 ⫾ 3.2 ng/mL versus 6.8 ⫾ 0.8 ng/mL at 132 hours). IL-1ra Response to Injury Considering all 29 study patients (Figure 2A), IL-1ra levels were also significantly higher at all time points compared with normal controls (0.12 ⫾ 0.04 ng/mL). Furthermore, IL-1ra was elevated at the initial 12-hour time point (22.3 ⫾ 5.1 ng/mL) compared with 36 (4.2 ⫾ 566
1.1 ng/mL), 60 (1.8 ⫾ 0.4 ng/ml), 84 (2.2 ⫾ 0.4 ng/mL), and 132 hours (4.1 ⫾ 0.8 ng/mL) postinjury. Figure 2B depicts the differential between IL-1ra levels for the MOF and non-MOF patients. IL-1ra levels were significantly higher (approximately 3 times greater) in the MOF patients compared with the non-MOF patients at 36 (8.2 ⫾ 3.4 ng/mL versus 2.6 ⫾ 0.8 ng/mL), 60 hours (3.2 ⫾ 1.1 ng/mL versus 1.3 ⫾ 0.3 ng/mL), 84 hours (3.8 ⫾ 0.9 ng/mL versus 1.6 ⫾ 0.4 ng/mL), and 132 hours postinjury (7.7 ⫾ 1.3 ng/mL versus 2.4 ⫾ 0.8 ng/mL).
COMMENTS Our trauma center has been interested in the pathogenesis of postinjury MOF in order to develop an early predictive model to identify the cohort of patients at high risk of MOF. We have identified several independent clinical predictors of postinjury MOF including age ⬎55 years, ISS ⱖ25, blood transfusion ⱖ6 units within 12 hours, base deficit from 0 to 12 hours ⬎8 mEq/L, and lactate ⬎2.5 mmol/L from 12 to 24 hours.18 Further work in our laboratory has demonstrated that neutrophil (PMN)-mediated tissue injury is also implicated in the pathogenesis of postinjury MOF.3 The proinflammatory cytokines IL-6 and IL-8 contribute to this PMN-mediated tissue injury in vitro.21 In order to clinically validate these basic studies, we previously measured IL-6 and IL-8 in severely injury patients and found that elevated levels of these proinflammatory cytokines discriminate patients who ultimately develop postinjury MOF.4 Both IL-6 and IL-8 were found to be significantly elevated in MOF compared with non-MOF patients early after injury (within 12 hours) and thus may act as markers as well as potential mediators of the postinjury SIRS response. Although TNF-␣ and/or IL-1 are well established as proinflammatory mediators in the sepsis syndrome5,6,11 and infusion of recombinant TNF-␣ and/or IL-1 leads to similar clinical symptoms and organ lesions as seen during MOF,22 their role in SIRS leading to MOF after traumatic injury has been less evident. Elevated levels of IL-1 and
THE AMERICAN JOURNAL OF SURGERY® VOLUME 178 DECEMBER 1999
RELEASE OF ANTI-INFLAMMATORY MEDIATORS AFTER TORSO TRAUMA/PARTRICK ET AL
Figure 1. Plasma concentrations of soluble tumor necrosis factor receptor (sTNFr) in all 29 study patients over the postinjury time period. Results are shown as mean ⫾ SEM. *P ⬍0.05 compared with normal controls. B. Plasma concentrations of sTNFr in multiorgan failure (MOF) patients (—■—) versus non-MOF patients (—䊐—). Results are shown as mean ⫾ SEM. *P ⬍0.05 compared with non-MOF patients.
TNF have not been demonstrated following injury, but local production in tissue beds (ie, lung) has been documented.7,9,10 The current study was undertaken to determine whether injury results in a rise of their naturally occurring anti-inflammatory inhibitors, sTNFr and IL-1ra, thus indirectly implicating TNF-␣ and/or IL-1 in the postinjury hyperinflammatory process. In addition, we wanted to determine if the systemic levels of these antiinflammatory mediators discriminated patients at risk for postinjury MOF, further contributing to our understanding of its pathogenesis as well as prediction of outcome. In the present study, we observed that sTNFr and IL-1ra are elevated early (within 12 hours) following injury compared with controls. These results are in concordance with other published data concerning early elevation of antiinflammatory mediators.8,10,17,23,24 In addition, we have demonstrated that elevated levels of sTNFr and IL-1ra correlate with the development of postinjury MOF early
Figure 2. Plasma concentrations of interleukin-1 receptor antagonist (IL-1ra) in all 29 study patients over the postinjury time period. Results are shown as mean ⫾ SEM. *P ⬍0.05 compared with normal controls. †P ⬍0.05 compared with 36, 60, 84, and 132 hours postinjury. B. Plasma concentrations of IL-1ra in multiorgan failure (MOF) patients (—F—) versus non-MOF patients (—E—). Results are shown as mean ⫾ SEM. *P ⬍0.05 compared with non-MOF patients.
after injury (12 to 36 hours). sTNFr and IL-1ra continue to remain elevated in MOF patients compared with nonMOF patients up to 5 days postinjury. This delayed elevation most likely represents persistent activation of the SIRS response in these critically ill patients. High levels of these endogenous antagonists to TNF-␣ and IL-1 suggest that these proinflammatory cytokines are likely involved in the postinjury hyperinflammatory response at the tissue level. Furthermore, these cytokine elevations correspond temporally to our previous observations concerning early PMN priming and sequestration in the same high-risk patient group.3 It remains unclear if therapeutic interventions to block this hyperinflammatory response will be beneficial in decreasing the morbidity and mortality associated with postinjury MOF. With such high levels of sTNFr and IL-1ra already circulating in these patients, it is unknown how much more would need to be administered to effect a positive clinical response. Animal experiments demon-
THE AMERICAN JOURNAL OF SURGERY® VOLUME 178 DECEMBER 1999
567
RELEASE OF ANTI-INFLAMMATORY MEDIATORS AFTER TORSO TRAUMA/PARTRICK ET AL
strate that increasing the ratio of IL-1ra to IL-1 to 1000 or even 10,000 results in improved survival rates.16 In human studies, increasing the plasma concentration of IL-1ra to 20 g/mL ensures nearly complete blockade of endothelial cell IL-1 receptors. Clinical trials using recombinant human IL-1ra have already been undertaken in patients with sepsis syndrome.25 Unfortunately, they were unable to demonstrate a statistically significant increase in survival time for recombinant IL-1ra treatment compared with placebo. However, there was a suggestion of a dose-related increase in survival time among patients with sepsis who have organ dysfunction and/or a predicted risk of mortality of 24% or greater.25 As in the case of IL-1ra, it is currently unclear whether the amount of soluble TNF receptor produced in disease is sufficient to reduce endogenous TNF activity. The difficulty in using an infectious model such as septic patients is the unknown timing of the initial infectious insult. Our results raise the question if recombinant human IL-1ra or sTNFr would be of benefit in trauma patients at high risk for developing postinjury MOF. The advantage of studying injured patients is the timing of the inflammatory insult (ie, traumatic event) is known at presentation. Injured patients at high risk for developing MOF can then be identified by clinical predictors as well as proinflammatory and anti-inflammatory cytokine profiles early after injury. Further study may determine if these high-risk patients with an easily identifiable time of insult may benefit from additional sTNFr or IL-1ra administration.
REFERENCES 1. Moore FA, Moore EE. Evolving concepts in the pathogenesis of postinjury multiple organ failure. Surg Clin North Am. 1995;75: 257–277. 2. Baue AE, Durham R, Faist E. Systemic inflammatory response syndrome (SIRS), multiple organ dysfunction syndrome (MODS), multiple organ failure (MOF): are we winning the battle? Shock. 1998;10:79 – 89. 3. Partrick DA, Moore FA, Moore EE, et al. Neutrophil priming and activation in the pathogenesis of postinjury multiple organ failure. New Horiz. 1996;4:194 –210. 4. Partrick DA, Moore FA, Moore EE, et al. The inflammatory profile of interleukin-6, interleukin-8, and soluble intercellular adhesion molecule-1 in postinjury multiple organ failure. Am J Surg. 1996;172:425– 431. 5. Damas P, Reuter A, Gysen P, et al. Tumor necrosis factor and interleukin-1 serum levels during severe sepsis in humans. Crit Care Med. 1989;17:975–978. 6. Strieter RM, Kunkel SL, Bone RC. Role of tumor necrosis factor-␣ in disease states and inflammation. Crit Care Med. 1993; 21:S447–S463. 7. Schwartz MD, Moore EE, Moore FA, et al. Nuclear factor-kappa B is activated in alveolar macrophages from patients with acute respiratory distress syndrome. Crit Care Med. 1996;24:1285–1292.
DISCUSSION David B. Hoyt, MD (San Diego, California): The antiinflammatory mechanisms are complex and include cytokines with opposing function and naturally occurring inhibitors such as soluble TNG receptor and IL-1 receptor antagonists. These inhibitors appear within hours of injury or sepsis and at least in animal models they have been very 568
8. Tan LR, Waxman K, Scannell G, et al. Trauma causes early release of soluble receptors for tumor necrosis factor. J Trauma. 1993;34:634 – 638. 9. Rabinovici R, John R, Esser KM, et al. Serum tumor necrosis factor-alpha profile in trauma patients. J Trauma. 1993;35:698 –702. 10. Keel M, Ecknauer E, Stocker R, et al. Different pattern of local and systemic release of proinflammatory and anti-inflammatory mediators in severely injured patients with chest trauma. J Trauma. 1996;40:907–914. 11. Dinarello CA. Biologic basis for interleukin-1 in disease. Blood. 1996;87:2095–2147. 12. Le J, Vilcek J. Tumor necrosis factor and interleukin-1: cytokines with multiple overlapping biological activities. Lab Invest. 1987;56:234 –248. 13. Zee KJV, Kohno T, Fischer E, et al. Tumor necrosis factor soluble receptors circulate during experimental and clinical inflammation and can protect against excessive tumor necrosis factor-␣ in vitro and in vivo. Proc Natl Acad Sci USA. 1992;89: 4845– 4849. 14. Carter DB, Deibel MR, Dunn CJ, et al. Purification, cloning, expression and biological characterization of an interleukin-1 receptor antagonist protein. Nature. 1990;344:633– 638. 15. Rose CE, Juliano CA, Tracey DE, et al. Role of interleukin-1 in endotoxin-induced lung injury in the rat. Am J Respir Cell Mol Biol. 1994;10:214 –221. 16. Alexander HR, Doherty GM, Buresh CM, et al. A recombinant human receptor antagonist to interleukin-1 improves survival after lethal endotoxemia in mice. J Exp Med. 1991;173:1029 –1032. 17. Cinat ME, Waxman K, Granger GA, et al. Trauma causes sustained elevation of soluble tumor necrosis factor receptors. J Am Coll Surg. 1994;179:529 –537. 18. Sauaia A, Moore FA, Moore EE, et al. Early predictors of postinjury multiple organ failure. Arch Surg. 1994;129:39 – 45. 19. Moore FA, Haenel JB, Moore EE, Whitehill TA. Incommensurate oxygen consumption in response to maximal oxygen availability predicts postinjury multiple organ failure. J Trauma. 1992; 33:58 – 67. 20. Goldie AS, Fearon KCH, Ross JA, et al. Natural cytokine antagonists and endogenous antiendotoxin core antibodies in sepsis syndrome. JAMA. 1995;274:172–177. 21. Biffl WL, Moore EE, Moore FA, et al. Interleukin-6 potentiates neutrophil priming with platelet-activating factor. Arch Surg. 1994; 129:1131–1136. 22. Okusawa S, Gelfand JA, Ikejima T, et al. Interleukin 1 induces a shock-like state in rabbits: synergism with tumor necrosis factor and the effect of cyclooxygenase inhibition. J Clin Invest. 1988;81: 1162–1172. 23. Ertel W, Keel M, Bonaccio M, et al. Release of anti-inflammatory mediators after mechanical trauma correlates with severity of injury and clinical outcome. J Trauma. 1995;39:879 – 887. 24. Seekamp A, Jochum M, Ziegler M, et al. Cytokines and adhesion molecules in elective and accidental trauma-related ischemia/ reperfusion. J Trauma. 1998;44:874 – 882. 25. Fisher CJ, Dhainaut JA, Opal SM, et al. Recombinant human interleukin 1 receptor antagonist in the treatment of patients with sepsis syndrome: results from a randomized double-blind, placebocontrolled trial. JAMA. 1994;271:1836 –1843.
active in reversing shock and multiple organ failure. Each has been a strategy to decrease mortality from sepsis in clinical trials during the last 5 years. These have been pursued, however, with very disappointing results. The present study measures the appearance of these two factors in response to injury and shows that elevated levels correspond to patients developing MOF. For both, this
THE AMERICAN JOURNAL OF SURGERY® VOLUME 178 DECEMBER 1999
BACKGROUND: Soluble tumor necrosis factor receptor (sTNFr) and interleukin-1 receptor antagonist (IL-1ra) have been identified as endogenous inhibitors of TNF-␣ and IL-1. While TNF-␣ and IL-1 levels are not systematically elevated in postinjury patients who developed multiorgan failure (MOF), their involvement at the tissue level has been suggested. Our study hypothesis was that levels of sTNFr-I and IL-1ra would discriminate patients at risk for postinjury MOF. METHODS: Serial plasma levels of sTNFr and IL1ra were measured in 29 trauma patients at high risk for postinjury MOF. RESULTS: sTNFr-I levels were higher in MOF compared with non-MOF patients at 12, 84, and 132 hours postinjury. MOF patients also had higher IL-1ra values 36, 60, 84, and 132 hours postinjury. CONCLUSIONS: Anti-inflammatory mechanisms are activated after trauma. Since increased levels of sTNFr and IL-1ra correlate with postinjury MOF, they may contribute to our understanding of the pathogenesis as well as prediction of outcome. High levels of antagonists to TNF-␣ and IL-1 suggest tissue level involvement of these cytokines in postinjury hyperinflammation. Am J Surg. 1999;178:564 –569. © 1999 by Excerpta Medica, Inc.
M
ultiple organ failure (MOF) remains the most common cause of delayed mortality in surgical intensive care units. Once established, MOF defies our standard critical care supportive measures and mortality exceeds 50%. Despite intensive investigation, the pathogenesis of postinjury MOF remains unclear.1,2
From the Department of Surgery, (DAP, EEM, WLB, CCB), Denver Health Medical Center, University of Colorado Health Sciences Center, Denver, Colorado; and the Department of Surgery (FAM), UT-Houston Medical School, Houston, Texas. This work was supported in part by National Institutes of Health Grants P50GM49222-01 and T32GM08315-03. Requests for reprints should be addressed to Ernest E. Moore, MD, Chief, Department of Surgery, Denver Health Medical Center, 777 Bannock Street, Denver, Colorado 80204. Presented at the 51st Annual Meeting of the Southwestern Surgical Congress, Coronado, California, April 18 –21, 1999.
564
© 1999 by Excerpta Medica, Inc. All rights reserved.
Our global hypothesis is that postinjury MOF occurs as the result of a dysfunctional inflammatory response.1,3 Following major tissue injury and hemorrhagic shock, patients develop an early physiologic state of systemic hyperinflammation, referred to as the systemic inflammatory response syndrome (SIRS). Certain patients appear to be vulnerable (ie, primed) such that early secondary insults amplify SIRS, which produces organ dysfunction ultimately culminating in MOF. We have previously demonstrated that systemic levels of the cytokines interleukin-6 (IL-6) and interleukin-8 (IL-8) correlate with MOF following severe injury.4 The role of the other proinflammatory cytokines tumor necrosis factoralpha (TNF-␣) and interleukin-1 beta (IL-1) in the systemic inflammatory response syndrome leading to postinjury MOF, however, remains unclear. TNF-␣, produced from the monocyte/macrophage lineage, is well established as a pivotal cytokine responsible for the clinical manifestations of shock induced by endotoxin or bacteremia leading to SIRS following sepsis.5,6 Activation of the TNF-␣ regulatory protein NFB has been documented in alveolar macrophages from patients with adult respiratory distress syndrome (ARDS).7 However, TNF-␣ levels in trauma patients have not been found to be elevated, even in patients who eventually developed MOF.8 –10 Similarly, IL-1 is a proinflammatory cytokine that shares many of the proinflammatory properties of TNF-␣5,11 Indeed, TNF and IL-1 not only stimulate their own release, but also the release of each other (as well as IL-6 and IL-8), thus amplifying the cascade of inflammatory mediators.12 IL-1 has two bioactive forms: IL-1␣ and IL-1. IL-1 is the predominant IL-1 and a major product of activated human monocytes, tissue macrophages, and neutrophils. Local production of IL-1 is significantly increased following trauma as evidenced by increased levels in bronchoalveolar lavage fluid, but it has not been found elevated in the circulation.10 This raises the possibility that the local release of these proinflammatory cytokines may significantly differ from their blood levels. Naturally occurring inhibitors of these two proinflammatory cytokines, produced by macrophages and neutrophils, have been well characterized. Soluble TNF receptors (sTNFr) compete with membrane receptors for binding of free TNF-␣, thus eliminating TNF bioactivity that would otherwise occur with TNF receptor ligation.13 IL-1 receptor antagonist (IL-1ra) is structurally related to IL-1, binds to the same receptor, but does not induce a discernible biological response.14 In vitro experiments have demonstrated 0002-9610/99/$–see front matter PII S0002-9610(99)00240-8
RELEASE OF ANTI-INFLAMMATORY MEDIATORS AFTER TORSO TRAUMA/PARTRICK ET AL
that sTNFr protects against excessive TNF activity13 and that IL-1ra prevents endotoxin-induced lung injury in rats15 and improves survival after endotoxemia in mice.16 sTNFr and IL-1ra have been detected in increased serum concentrations after severe injury,8,10,17 but it has not been established if systemic levels of these anti-inflammatory mediators correlate with postinjury organ failure in humans. Our study hypothesis was circulating levels of sTNFr-I and IL-1ra would discriminate patients at risk for the development of postinjury MOF.
MATERIALS AND METHODS Study Population During a 16-month period, patients admitted to Denver Health Medical Center between 18 and 65 years of age who had an Injury Severity Score (ISS) ⱖ25 or the combination of an ISS ⬎15 and a transfusion requirement of more than 6 units of blood within the first 12 hours were considered eligible for this study. Previous epidemiological studies at our institution had identified these patients as being high risk for the development of MOF.18 Patients with a major head injury (Glasgow Coma Scale score ⬍8) or significant preexisting disease were excluded in an attempt to accrue a homogeneous study population. Additionally, patients transferred from other hospitals or with time from injury to our emergency department admission exceeding 3 hours were excluded. Informed consent was obtained via a protocol approved by the Colorado Multiple Institutional Review Board, University of Colorado Health Sciences Center, Denver, Colorado. A pool of 6 healthy adult volunteers served as controls. Demographic, risk factor, and outcome data for these patients were collected as part of an ongoing, prospective study funded by the National Institutes of Health to investigate the pathogenesis of MOF after injury. The diagnosis of MOF was based on our standard score described previously.18 Briefly, four organ functions (pulmonary, hepatic, renal, and cardiac) were graded daily on a scale of zero to three. MOF was defined as a sum of the individual organ grades obtained simultaneously that was greater than or equal to four. Organ dysfunction scores obtained within the first 48 hours were not used to define organ failure because they may reflect reversible derangement induced by the inciting event or incomplete resuscitation. Sample Collection All patients were resuscitated using a previously described protocol employing defined end points.19 Patients requiring surgical intervention underwent standard surgical and postoperative intensive care unit treatment provided by the same surgical team. The first blood sample was collected from each patient within 12 hours after the estimated time of injury. Three subsequent blood samples were obtained at 24 hours intervals (36, 60, and 84 hours) postinjury and the final sample was collected at 132 hours postinjury (48 hours after previous sample). Eight milliliters of venous blood were collected at each time point in a sterile, pyrogen-free glass tube containing disodium ethylenediamine-tetraacetic acid (EDTA). These samples were first centrifuged at 300g for 10 minutes and then spun in an ultracentrifuge for 5 additional minutes to obtain platelet-
poor plasma. The plasma was removed and stored at ⫺70°C in a sterile, pyrogen-free tube. Immunologic Assays sTNFr: There are two different soluble forms of the TNFr: a 55kD receptor (sTNFr-I) and a 75kD receptor (sTNFr-II). In vitro data support a greater inhibitory role for sTNFr-I in neutralizing the cytotoxic effects of TNF␣,13 and in vivo data has demonstrated sTNFr-I to be more predictive of mortality following sepsis.20 Therefore, a commercially available enzyme-linked immunosorbent assay (ELISA) kit (Bender Med System, Vienna, Austria) for the quantitative measurement of human sTNFr-I (p55) was used. Ninety-six well polystyrene microtiter plates precoated with a monoclonal antibody directed against human recombinant sTNFr were employed. A series of twofold dilutions of recombinant human sTNF ranging from 100 ng/mL to 1 ng/mL was performed to generate a standard curve for each assay. sTNFr in the sample or standard was quantified by an enzymatic reaction, resulting in a detectable color change. The absorbance was measured using a microtiter plate-reading spectrophotometer (THERMOmax; Molecular Devices Corporation, Menlo Park, California) set at 450 and 550 nm. All samples and standards were run in duplicate and mean values were used in calculating sample concentrations using a linear regression analysis. The lower limit of detection for sTNFr was 0.4 ng/mL. IL-1ra: IL-1ra was measured because it is a recognized anti-inflammatory cytokine that prevents organ injury in vitro15 and may improve survival in the setting of sepsis.16 A commercially available ELISA kit (R&D Systems, Minneapolis, Minnesota) for the quantitative measurement of human IL-1ra was used. The assay was similar to that process described for sTNFr except wells were precoated with a monoclonal antibody directed against human IL-1ra and the standard series ranged from 600 ng/mL to 0.5 ng/mL using recombinant human IL-1ra. The lower limit of detection of IL-1ra was 48 pg/mL. Statistical Analysis Nominal data (eg, gender, mechanism of injury, and MOF outcome) were compared by Fisher’s exact test, injury severity score by Student’s t test, and continuous data (eg, length of intensive care unit stay, and plasma levels) were compared by an analysis of variance (ANOVA) using Scheffe’s F procedure for post-hoc comparisons. A P value of ⬍0.05 was considered statistically significant.
RESULTS Patient Population Twenty-nine patients from those enrolled in the NIH study had plasma samples available over the entire 5-day postinjury time period and comprise the study population. The mean age of the study patients was 33.6 ⫾ 2.0 years; 21 (72%) were male, the injury mechanism was blunt in 22 (76%), and mean ISS was 30.5 ⫾ 1.5. Nine (31%) of the 29 study patients developed MOF. The Table summarizes the demographics, MOF risk factors, and clinical outcome for MOF versus non-MOF study subjects. There were no significant differences in age, gender, mechanism of injury,
THE AMERICAN JOURNAL OF SURGERY® VOLUME 178 DECEMBER 1999
565
RELEASE OF ANTI-INFLAMMATORY MEDIATORS AFTER TORSO TRAUMA/PARTRICK ET AL
TABLE Demographics, Multiorgan Failure (MOF) Risk Factors, and Clinical Outcomes of Study Patients Who Did or Did Not Develop MOF Patient Variables Age (years) Male Blunt mechanism ISS ED SBP (mm Hg) URBC (first 12 h) Highest BD (0–12 h) Highest lactate (mmol/L) (12–24 h) Minor septic complications Major septic complications Nonseptic complications Length ICU stay (days) Death
MOF (n ⴝ 9)
Non-MOF (n ⴝ 20)
P Value
36.3 ⫾ 4.1 33.0 ⫾ 2.5 NS 6 (67%) 15 (75%) NS 8 (89%) 14 (70%) NS 33.8 ⫾ 3.6 28.9 ⫾ 1.5 NS 111.2 ⫾ 6.7 107.8 ⫾ 6.6 NS 17.7 ⫾ 4.7 8.1 ⫾ 1.8 ⬍0.05 18.2 ⫾ 4.4 12.2 ⫾ 1.6 NS (P ⫽ 0.15) 5.8 ⫾ 1.1 3.3 ⫾ 0.7 NS (P ⫽ 0.06) 7 (78%) 4 (20%) ⬍0.05 9 (100%) 3 (15%) ⬍0.0001 9 (100%) 8 (40%) ⬍0.05 32.2 ⫾ 5.5 6.7 ⫾ 1.7 ⬍0.0001 2 (22%) 0 NS (P ⫽ 0.09)
ED SBP ⫽ systolic blood pressure in emergency department; URBC ⫽ units of red blood cells transfused; BD ⫽ base deficit (mEq/L); ICU ⫽ intensive care unit; NS, not significant (ie, P ⬎0.05).
injury severity score, or systolic blood pressure in the emergency department. However, compared with the non-MOF patients, MOF patients required more early blood transfusions, developed more septic and nonseptic complications, and stayed longer in the intensive care unit. Furthermore, there was a trend toward increased base deficit and lactate as well as increased mortality in MOF patients. sTNFr Response to Injury Analyzing all 29 study patients (Figure 1A), sTNFr levels were significantly elevated at all time points compared with the normal controls (1.8 ⫾ 0.3 ng/mL). However, there was no significant change in sTNFr levels over the study time course (12 hours, 8.1 ⫾ 0.8 ng/mL; 36 hours, 8.2 ⫾ 1.0 ng/mL; 60 hours, 7.6 ⫾ 1.2 ng/mL; 84 hours, 8.6 ⫾ 1.4 ng/mL; and 132 hours postinjury, 11.1 ⫾ 1.7 ng/mL). In comparison, there was a significant difference between sTNFr levels for the MOF and non-MOF patients (Figure 1B). sTNFr levels were significantly higher (approximately 1.5 times greater) in the MOF patients compared with the non-MOF patients at 12 hours postinjury (11.3 ⫾ 2.2 ng/mL versus 6.8 ⫾ 0.6 ng/mL). At 36 and 60 hours postinjury there were no significant differences between MOF and non-MOF patients (10.1 ⫾ 2.0 ng/mL versus 7.5 ⫾ 1.1 ng/mL and 10.2 ⫾ 4.0 ng/mL versus 6.7 ⫾ 0.7 ng/mL, respectively). However, at 84 and 132 hours postinjury, sTNFr levels in the MOF patients increased again and were significantly elevated (approximately twofold higher at 84 hours and threefold higher 5 days postinjury) compared with non-MOF patients (13.8 ⫾ 4.2 ng/mL versus 6.5 ⫾ 0.8 ng/mL at 84 hours, 19.9 ⫾ 3.2 ng/mL versus 6.8 ⫾ 0.8 ng/mL at 132 hours). IL-1ra Response to Injury Considering all 29 study patients (Figure 2A), IL-1ra levels were also significantly higher at all time points compared with normal controls (0.12 ⫾ 0.04 ng/mL). Furthermore, IL-1ra was elevated at the initial 12-hour time point (22.3 ⫾ 5.1 ng/mL) compared with 36 (4.2 ⫾ 566
1.1 ng/mL), 60 (1.8 ⫾ 0.4 ng/ml), 84 (2.2 ⫾ 0.4 ng/mL), and 132 hours (4.1 ⫾ 0.8 ng/mL) postinjury. Figure 2B depicts the differential between IL-1ra levels for the MOF and non-MOF patients. IL-1ra levels were significantly higher (approximately 3 times greater) in the MOF patients compared with the non-MOF patients at 36 (8.2 ⫾ 3.4 ng/mL versus 2.6 ⫾ 0.8 ng/mL), 60 hours (3.2 ⫾ 1.1 ng/mL versus 1.3 ⫾ 0.3 ng/mL), 84 hours (3.8 ⫾ 0.9 ng/mL versus 1.6 ⫾ 0.4 ng/mL), and 132 hours postinjury (7.7 ⫾ 1.3 ng/mL versus 2.4 ⫾ 0.8 ng/mL).
COMMENTS Our trauma center has been interested in the pathogenesis of postinjury MOF in order to develop an early predictive model to identify the cohort of patients at high risk of MOF. We have identified several independent clinical predictors of postinjury MOF including age ⬎55 years, ISS ⱖ25, blood transfusion ⱖ6 units within 12 hours, base deficit from 0 to 12 hours ⬎8 mEq/L, and lactate ⬎2.5 mmol/L from 12 to 24 hours.18 Further work in our laboratory has demonstrated that neutrophil (PMN)-mediated tissue injury is also implicated in the pathogenesis of postinjury MOF.3 The proinflammatory cytokines IL-6 and IL-8 contribute to this PMN-mediated tissue injury in vitro.21 In order to clinically validate these basic studies, we previously measured IL-6 and IL-8 in severely injury patients and found that elevated levels of these proinflammatory cytokines discriminate patients who ultimately develop postinjury MOF.4 Both IL-6 and IL-8 were found to be significantly elevated in MOF compared with non-MOF patients early after injury (within 12 hours) and thus may act as markers as well as potential mediators of the postinjury SIRS response. Although TNF-␣ and/or IL-1 are well established as proinflammatory mediators in the sepsis syndrome5,6,11 and infusion of recombinant TNF-␣ and/or IL-1 leads to similar clinical symptoms and organ lesions as seen during MOF,22 their role in SIRS leading to MOF after traumatic injury has been less evident. Elevated levels of IL-1 and
THE AMERICAN JOURNAL OF SURGERY® VOLUME 178 DECEMBER 1999
RELEASE OF ANTI-INFLAMMATORY MEDIATORS AFTER TORSO TRAUMA/PARTRICK ET AL
Figure 1. Plasma concentrations of soluble tumor necrosis factor receptor (sTNFr) in all 29 study patients over the postinjury time period. Results are shown as mean ⫾ SEM. *P ⬍0.05 compared with normal controls. B. Plasma concentrations of sTNFr in multiorgan failure (MOF) patients (—■—) versus non-MOF patients (—䊐—). Results are shown as mean ⫾ SEM. *P ⬍0.05 compared with non-MOF patients.
TNF have not been demonstrated following injury, but local production in tissue beds (ie, lung) has been documented.7,9,10 The current study was undertaken to determine whether injury results in a rise of their naturally occurring anti-inflammatory inhibitors, sTNFr and IL-1ra, thus indirectly implicating TNF-␣ and/or IL-1 in the postinjury hyperinflammatory process. In addition, we wanted to determine if the systemic levels of these antiinflammatory mediators discriminated patients at risk for postinjury MOF, further contributing to our understanding of its pathogenesis as well as prediction of outcome. In the present study, we observed that sTNFr and IL-1ra are elevated early (within 12 hours) following injury compared with controls. These results are in concordance with other published data concerning early elevation of antiinflammatory mediators.8,10,17,23,24 In addition, we have demonstrated that elevated levels of sTNFr and IL-1ra correlate with the development of postinjury MOF early
Figure 2. Plasma concentrations of interleukin-1 receptor antagonist (IL-1ra) in all 29 study patients over the postinjury time period. Results are shown as mean ⫾ SEM. *P ⬍0.05 compared with normal controls. †P ⬍0.05 compared with 36, 60, 84, and 132 hours postinjury. B. Plasma concentrations of IL-1ra in multiorgan failure (MOF) patients (—F—) versus non-MOF patients (—E—). Results are shown as mean ⫾ SEM. *P ⬍0.05 compared with non-MOF patients.
after injury (12 to 36 hours). sTNFr and IL-1ra continue to remain elevated in MOF patients compared with nonMOF patients up to 5 days postinjury. This delayed elevation most likely represents persistent activation of the SIRS response in these critically ill patients. High levels of these endogenous antagonists to TNF-␣ and IL-1 suggest that these proinflammatory cytokines are likely involved in the postinjury hyperinflammatory response at the tissue level. Furthermore, these cytokine elevations correspond temporally to our previous observations concerning early PMN priming and sequestration in the same high-risk patient group.3 It remains unclear if therapeutic interventions to block this hyperinflammatory response will be beneficial in decreasing the morbidity and mortality associated with postinjury MOF. With such high levels of sTNFr and IL-1ra already circulating in these patients, it is unknown how much more would need to be administered to effect a positive clinical response. Animal experiments demon-
THE AMERICAN JOURNAL OF SURGERY® VOLUME 178 DECEMBER 1999
567
RELEASE OF ANTI-INFLAMMATORY MEDIATORS AFTER TORSO TRAUMA/PARTRICK ET AL
strate that increasing the ratio of IL-1ra to IL-1 to 1000 or even 10,000 results in improved survival rates.16 In human studies, increasing the plasma concentration of IL-1ra to 20 g/mL ensures nearly complete blockade of endothelial cell IL-1 receptors. Clinical trials using recombinant human IL-1ra have already been undertaken in patients with sepsis syndrome.25 Unfortunately, they were unable to demonstrate a statistically significant increase in survival time for recombinant IL-1ra treatment compared with placebo. However, there was a suggestion of a dose-related increase in survival time among patients with sepsis who have organ dysfunction and/or a predicted risk of mortality of 24% or greater.25 As in the case of IL-1ra, it is currently unclear whether the amount of soluble TNF receptor produced in disease is sufficient to reduce endogenous TNF activity. The difficulty in using an infectious model such as septic patients is the unknown timing of the initial infectious insult. Our results raise the question if recombinant human IL-1ra or sTNFr would be of benefit in trauma patients at high risk for developing postinjury MOF. The advantage of studying injured patients is the timing of the inflammatory insult (ie, traumatic event) is known at presentation. Injured patients at high risk for developing MOF can then be identified by clinical predictors as well as proinflammatory and anti-inflammatory cytokine profiles early after injury. Further study may determine if these high-risk patients with an easily identifiable time of insult may benefit from additional sTNFr or IL-1ra administration.
REFERENCES 1. Moore FA, Moore EE. Evolving concepts in the pathogenesis of postinjury multiple organ failure. Surg Clin North Am. 1995;75: 257–277. 2. Baue AE, Durham R, Faist E. Systemic inflammatory response syndrome (SIRS), multiple organ dysfunction syndrome (MODS), multiple organ failure (MOF): are we winning the battle? Shock. 1998;10:79 – 89. 3. Partrick DA, Moore FA, Moore EE, et al. Neutrophil priming and activation in the pathogenesis of postinjury multiple organ failure. New Horiz. 1996;4:194 –210. 4. Partrick DA, Moore FA, Moore EE, et al. The inflammatory profile of interleukin-6, interleukin-8, and soluble intercellular adhesion molecule-1 in postinjury multiple organ failure. Am J Surg. 1996;172:425– 431. 5. Damas P, Reuter A, Gysen P, et al. Tumor necrosis factor and interleukin-1 serum levels during severe sepsis in humans. Crit Care Med. 1989;17:975–978. 6. Strieter RM, Kunkel SL, Bone RC. Role of tumor necrosis factor-␣ in disease states and inflammation. Crit Care Med. 1993; 21:S447–S463. 7. Schwartz MD, Moore EE, Moore FA, et al. Nuclear factor-kappa B is activated in alveolar macrophages from patients with acute respiratory distress syndrome. Crit Care Med. 1996;24:1285–1292.
DISCUSSION David B. Hoyt, MD (San Diego, California): The antiinflammatory mechanisms are complex and include cytokines with opposing function and naturally occurring inhibitors such as soluble TNG receptor and IL-1 receptor antagonists. These inhibitors appear within hours of injury or sepsis and at least in animal models they have been very 568
8. Tan LR, Waxman K, Scannell G, et al. Trauma causes early release of soluble receptors for tumor necrosis factor. J Trauma. 1993;34:634 – 638. 9. Rabinovici R, John R, Esser KM, et al. Serum tumor necrosis factor-alpha profile in trauma patients. J Trauma. 1993;35:698 –702. 10. Keel M, Ecknauer E, Stocker R, et al. Different pattern of local and systemic release of proinflammatory and anti-inflammatory mediators in severely injured patients with chest trauma. J Trauma. 1996;40:907–914. 11. Dinarello CA. Biologic basis for interleukin-1 in disease. Blood. 1996;87:2095–2147. 12. Le J, Vilcek J. Tumor necrosis factor and interleukin-1: cytokines with multiple overlapping biological activities. Lab Invest. 1987;56:234 –248. 13. Zee KJV, Kohno T, Fischer E, et al. Tumor necrosis factor soluble receptors circulate during experimental and clinical inflammation and can protect against excessive tumor necrosis factor-␣ in vitro and in vivo. Proc Natl Acad Sci USA. 1992;89: 4845– 4849. 14. Carter DB, Deibel MR, Dunn CJ, et al. Purification, cloning, expression and biological characterization of an interleukin-1 receptor antagonist protein. Nature. 1990;344:633– 638. 15. Rose CE, Juliano CA, Tracey DE, et al. Role of interleukin-1 in endotoxin-induced lung injury in the rat. Am J Respir Cell Mol Biol. 1994;10:214 –221. 16. Alexander HR, Doherty GM, Buresh CM, et al. A recombinant human receptor antagonist to interleukin-1 improves survival after lethal endotoxemia in mice. J Exp Med. 1991;173:1029 –1032. 17. Cinat ME, Waxman K, Granger GA, et al. Trauma causes sustained elevation of soluble tumor necrosis factor receptors. J Am Coll Surg. 1994;179:529 –537. 18. Sauaia A, Moore FA, Moore EE, et al. Early predictors of postinjury multiple organ failure. Arch Surg. 1994;129:39 – 45. 19. Moore FA, Haenel JB, Moore EE, Whitehill TA. Incommensurate oxygen consumption in response to maximal oxygen availability predicts postinjury multiple organ failure. J Trauma. 1992; 33:58 – 67. 20. Goldie AS, Fearon KCH, Ross JA, et al. Natural cytokine antagonists and endogenous antiendotoxin core antibodies in sepsis syndrome. JAMA. 1995;274:172–177. 21. Biffl WL, Moore EE, Moore FA, et al. Interleukin-6 potentiates neutrophil priming with platelet-activating factor. Arch Surg. 1994; 129:1131–1136. 22. Okusawa S, Gelfand JA, Ikejima T, et al. Interleukin 1 induces a shock-like state in rabbits: synergism with tumor necrosis factor and the effect of cyclooxygenase inhibition. J Clin Invest. 1988;81: 1162–1172. 23. Ertel W, Keel M, Bonaccio M, et al. Release of anti-inflammatory mediators after mechanical trauma correlates with severity of injury and clinical outcome. J Trauma. 1995;39:879 – 887. 24. Seekamp A, Jochum M, Ziegler M, et al. Cytokines and adhesion molecules in elective and accidental trauma-related ischemia/ reperfusion. J Trauma. 1998;44:874 – 882. 25. Fisher CJ, Dhainaut JA, Opal SM, et al. Recombinant human interleukin 1 receptor antagonist in the treatment of patients with sepsis syndrome: results from a randomized double-blind, placebocontrolled trial. JAMA. 1994;271:1836 –1843.
active in reversing shock and multiple organ failure. Each has been a strategy to decrease mortality from sepsis in clinical trials during the last 5 years. These have been pursued, however, with very disappointing results. The present study measures the appearance of these two factors in response to injury and shows that elevated levels correspond to patients developing MOF. For both, this
THE AMERICAN JOURNAL OF SURGERY® VOLUME 178 DECEMBER 1999