help button home button
AJRCCM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Online Supplement
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dugernier, T. L.
Right arrow Articles by Pugin, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dugernier, T. L.
Right arrow Articles by Pugin, J.
American Journal of Respiratory and Critical Care Medicine Vol 168. pp. 148-157, (2003)
© 2003 American Thoracic Society

Original Article

Compartmentalization of the Inflammatory Response during Acute Pancreatitis

Correlation with Local and Systemic Complications

Thierry L. Dugernier, Pierre-François Laterre, Xavier Wittebole, Jean Roeseler, Dominique Latinne, Marc S. Reynaert and Jérôme Pugin

Department of Intensive Care and Emergency Medicine, and Experimental Immunology Unit, Department of Surgery, St. Luc University Hospital, Brussels, Belgium; and Division of Medical Intensive Care, University Hospital of Geneva, Geneva, Switzerland

Correspondence and requests for reprints should be addressed to Thierry Dugernier, M.D., Department of Intensive Care and Emergency Medicine, St. Luc University Hospital, Hippocrate Avenue, 10 B-1200 Brussels, Belgium. E-mail: thierry.dugernier{at}skynet.be


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Local and systemic inflammation has been implicated in the pathogenesis of acute pancreatitis and secondary multisystem organ failure. To assess the pro- and antiinflammatory response, the site of mediator production, and their route of diffusion, we sampled simultaneously ascites, thoracic lymph, and blood at the onset of end-organ dysfunction and for the following 6 days in 60 patients with acute pancreatitis. We used immunoassays to measure pro- and antiinflammatory cytokines and cell-based bioassays to assess the net pro- and antiinflammatory activity elicited by the biological fluids. Tumor necrosis factor-{alpha} and interleukin-1ß were detected in less than 15% of blood and lymph samples. Secondary pro- and antiinflammatory cytokines were found to be elevated early and throughout the sampling period in all compartments. Cytokine levels decreased from ascites to lymph to blood, suggesting a splanchnic origin. Prolonged diversion of ascites and lymph did not alter cytokine gradients, suggesting mediator transfer via the splanchnic blood circulation. Although a net proinflammatory activity ascribed to interleukin-1ß was detected in ascites, a net antiinflammatory activity was measured in virtually all lymph and blood samples, suggesting that the pancreas and the splanchnic area are sites of a proinflammatory response and that an early, dominant, and sustained antiinflammatory activity takes place in circulating compartments.

Key Words: acute necrotizing pancreatitis • cytokines • inflammation mediators • multiple organ failure • systemic inflammatory response syndrome

The overall mortality associated with severe acute pancreatitis still ranges from 15 to 30% (1, 2). The two major determinants of outcome are the severity of multisystem organ failure (MSOF) and the extent of (peri)pancreatic necrosis that creates the culture medium for bacterial proliferation (3–8). MSOF is observed early after the onset of severe acute pancreatitis and, in some cases, follows pancreatic infection (3).

Experimental studies on severe pancreatitis have demonstrated that inflammatory mediators play an important role in the local tissue injury and the development of MSOF (9–13). The administration of inhibitors of proinflammatory mediators prevents local and remote tissue damage and improves survival in several animal models (14–20). A close correlation between the severity of MSOF and the serum levels of mediators of the inflammatory cascade has been reported in a number of human studies (21–26). However, little is known about the contribution of these mediators to the extent of pancreatic necrosis. Furthermore, data on the site and the kinetics of cytokine production, the routes of cytokine diffusion, and the balance between the pro- and antiinflammatory response are scanty or nonexistent in humans (21, 27, 28).

In this study, we prospectively studied a large cohort of patients with severe acute pancreatitis in whom peritoneal lavage and thoracic duct drainage were performed at the onset of MSOF. This gave us the unique opportunity to address the compartmentalization of the immunoinflammatory response. The levels of pro- and antiinflammatory mediators and their bioactivities were determined in ascites, lymph, and blood. We found that both pancreatic damage and end-organ dysfunctions correlated with the magnitude of the inflammatory response. Inflammatory mediators were primarily released from the splanchnic area, and gained access to the systemic compartment mainly by the portal and suprahepatic circulation. The peritoneal compartment was the site of a net proinflammatory reaction to the pancreatic necrosis, whereas a net antiinflammatory response dominated in the lymph collected from the thoracic duct, as well as in the systemic circulation during the first week after the onset of MSOF.


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Patients
Sixty-three patients admitted to the medicosurgical intensive care unit (ICU) early after the onset of severe acute pancreatitis were consecutively enrolled in a conservative therapeutic strategy that included peritoneal lavage and thoracic duct drainage (29). The study protocol including drainage of the thoracic duct and sampling of lymph, ascites, and blood received institutional approval from the local Ethics Committee. Written informed consent was obtained from all patients or from the next of kin before enrollment.

Samples of at least two of three compartments were obtained in 60 of 63 patients (95%) who were selected for this study. All patients had at least one end-organ failure at the onset of thoracic duct drainage. The severity of acute pancreatitis and the magnitude of MSOF were assessed by modified Acute Physiology and Chronic Health Evaluation (APACHE II), Sequential Organ Failure Assessment (SOFA) scores, the number and type of organ system failure, Ranson prognostic signs, and the peak serum level of C-reactive protein (30, 31). Acute respiratory distress syndrome (ARDS) was defined according to criteria from the North American–European Consensus Conference (32).

The extent of local tissue injury was assessed by the 10-point computed tomography (CT) Severity Index (33), and by the volume of necrosis (<30%, 30–50%, >50%). Early infected pancreatic necrosis was defined as pancreatic infection occurring 14 days or less after the onset of pancreatitis.

Sampling Protocol and Cytokine Assays
Peritoneal lavage and thoracic duct drainage were performed as described (34, 35). Blood and lymph were sampled once daily and concurrently collected at the same time of day during the course of thoracic duct drainage. A complete sequence of lymph and blood samples was obtained in 59 (98%) and 36 (60%) patients, respectively. Ascites was collected once at the time of insertion of the peritoneal lavage catheter in 15 (25%) patients from whom pure undiluted fluid could be obtained.

Proinflammatory cytokines (tumor necrosis factor [TNF]-{alpha}, interleukin [IL]-1ß, IL-6, IL-8), their inhibitors (soluble TNF receptors [sTNF-RI and sTNF-RII], IL-1 receptor antagonist [IL-1ra]), and the antiinflammatory cytokine IL-10 were measured in plasma, lymph, and ascites. Cytokine concentrations were determined in specific solid-phase sandwich enzyme-linked immunosorbent assays (TNF-{alpha}, IL-1ß, IL-6, IL-8, and IL-10 ELISAs [BD Biosciences Pharmingen, San Diego, CA]); and in enzyme-amplified sensitivity immunoassays (IL-1ra, sTNF-RI, and sTNF-RII [EASIAs; Biosource, Nivelles, Belgium]) according to the manufacturer's instructions.

Pro- and Antiinflammatory Activities
Proinflammatory activity.
Proinflammatory activity in blood, lymph, and ascites was defined as the capacity of the sample to induce the activation of cultured human pulmonary type II-like A549 cells (American Type Culture Collection, Manassas, VA) as measured by the surface upregulation of intracellular adhesion molecule (ICAM)-1 by these cells as described (36). Values above 1 U/ml were considered significant, indicating the presence of proinflammatory activity (37).

Antiinflammatory activity.
Antiinflammatory bioassays were developed to quantify the excess antiinflammatory activity contained in human body fluids. "Anti–IL-1 activity" and "anti-TNF activity" were defined as the capacity of the body fluid to inhibit A549 cell activation (ICAM-1 upregulation) induced by a known and fixed amount of recombinant human IL-1ß and TNF-{alpha}, respectively. For anti–IL-1 activity, a standard curve of recombinant IL-1ra inhibition of IL-1ß–induced ICAM-1 upregulation was used to quantify the level of inhibition induced by the body fluid (see Figure E1 in the online supplement). For anti-TNF activity, a standard curve of recombinant sTNF-RI inhibition of TNF-{alpha}–induced ICAM-1 upregulation was used to quantify the level of inhibition induced by the body fluid (see Figure E2 in the online supplement). Results were expressed as units of anti–IL-1 activity or anti-TNF activity per milliliter. Values above 1 U/ml were considered significant, indicating the presence of antiinflammatory activity. See the online supplement for details on bioassay methods.

Statistical Analysis
The Mann–Whitney U test and Wilcoxon test were used as appropriate to compare continuous data between two groups. The Kruskal–Wallis and Friedman tests were used for analysis of variance. Fisher's exact test or {chi}2 plus Yates correction were used as appropriate to compare categorical data. The Spearman rank sum test with rank correlation coefficients was calculated to assess correlations between cytokine concentrations in blood and lymph and between cytokine levels and scoring systems. These correlations were reported using nominal p values that were uncorrected for multiple comparisons. A p value <= 0.05 was considered significant. Statistical calculations were performed with the MedCalc software package (38) and SPSS 9.0 for Windows (SPSS, Chicago, IL).


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Patients
The characteristics of the patient population are outlined in Table 1 . Thirty-four (57%) patients were admitted to the ICU within 48 hours of the onset of symptoms and 47 (78%) patients had MSOF within 96 hours of the onset of disease. Peritoneal lavage and thoracic duct drainage were performed in 50 (85%) and 59 (98%) patients, respectively. Peritoneal lavage was initiated in all patients within 24 hours of the onset of MSOF whereas 55 (93%) patients underwent drainage of the thoracic lymph within this time interval. The median duration (range) of peritoneal lavage and thoracic duct drainage was 5 (1–17) and 6 (2–13) days, respectively. As patients were admitted and sampled at various intervals from the onset of disease, results were normalized and analyzed according to the onset of organ failure(s) for sake of comparison between patients. Reference values of cytokines and bioactivity in the lymph were obtained for comparison in a group of eight patients who underwent either an elective transthoracic resection for carcinoma of the esophagus (n = 6) or organ procurement after severe traumatic brain injury and secondary brain death (n = 2). Reference values in ascites were obtained in a group of five noninfected patients with either right heart failure (n = 3) or liver cirrhosis (n = 2).


View this table:
[in this window]
[in a new window]
  		TABLE 1. Characteristics of the patient population*

 
The early staging of severity and the assessment of local and distant tissue injury are summarized in Table 2 . Infected pancreatic necrosis was documented during the hospital stay in 28 (47%) patients. Ten of these patients had early infected pancreatic necrosis that was diagnosed during the course of peritoneal lavage and thoracic duct drainage and 6 (60%) had concurrent bacteremia. Infection was documented outside the pancreatic area during the period of lymph sampling in 15 (25%) patients: 14 had pulmonary infections, 2 with bacteremia, and 1 had bacteremia of pharyngeal origin. Nineteen (32%) patients died during the hospital stay, most of them (95%) in the ICU, 49 ± 39 days after ICU admission. No patient died during the period of lymph sampling. In 17 (89%) patients death was ascribed to MSOF secondary to infected pancreatic necrosis.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Early assessment of severity, extent of remote organ dysfunction and retroperitoneal injury, and outcome*

 
Intracompartmental Mediators and Bioactivities at the Onset of MSOF
Cytokine concentrations as well as pro- and antiinflammatory activities in the three compartments at the initiation of peritoneal lavage and thoracic duct drainage are outlined in Table 3 .


View this table:
[in this window]
[in a new window]
  		TABLE 3. Intracompartmental concentrations of cytokines and bioactivities at the onset of msof||

 
"Proximal" proinflammatory cytokines (TNF-{alpha} and IL-1ß) and corresponding proinflammatory activity were detected in a minority of patients, particularly in lymph and plasma. The IL-1ß level in ascites differed significantly from the other compartments. Similarly, proinflammatory activity was primarily detected in the ascites compartment and was correlated with IL-1ß (r = 0.73, p = 0.029) whereas no correlation could be demonstrated with TNF-{alpha} (r = 0.46, p = 0.14). Patients with detectable IL-1ß in ascites were sampled earlier after the onset of disease than those in whom IL-1ß was not detected (mean ± SD, 1 ± 0.8 versus 2.3 ± 0.8 days; p = 0.03).

Antiinflammatory activity ascribed either to inhibitors of TNF-{alpha} (anti-TNF activity) or IL-1ß (anti–IL-1 activity) predominated largely over proinflammatory activity in plasma and lymph, but not in ascites. There was no significant difference in antiinflammatory activity between the three compartments with the exception of anti–IL-1 activity in ascites, which was found to be significantly higher than in lymph. In all compartments anti-TNF activity was higher than anti–IL-1 activity.

Interleukin-6, IL-8, soluble TNF receptors, and IL-1ra levels were well above the limit of detection in the three compartments and were detected in nearly all patients in spite of marked interpatient variability. Levels of cytokines were found to be highest in the ascites compartment. The balance between the pro- and antiinflammatory response as expressed by molar ratios of antiinflammatory to proinflammatory cytokines and bioactivities is presented in Table 4 .


View this table:
[in this window]
[in a new window]
  		TABLE 4. Molar ratios of antiinflammatory to proinflammatory cytokines

and bioactivities at the onset of msof*

 
With the exception of TNF-{alpha}, IL-1ß, and proinflammatory activity, the mediator concentrations and antiinflammatory activities in the three compartments were significantly higher than their respective control levels (p < 0.05 for all cytokines versus controls in lymph, plasma, and ascites). In ascites, levels of IL-1ß were significantly increased in patients when compared with controls (median, 5 pg/ml; range, 0–265 versus 0 pg/ml, respectively; p < 0.001) (see Table E1 in the online supplement).

Proinflammatory Mediators and Biological Activities during Drainage
The number of patients and samples in whom TNF-{alpha}, IL-ß, and pro- or antiinflammatory activity was detected in lymph and plasma during the period of observation is outlined in Table 5 . TNF-{alpha} and IL-1ß and the corresponding proinflammatory activity were found in few patients. In the vast majority of the patients a net antiinflammatory activity largely predominated in lymph and plasma, with similar levels in both compartments. In the few samples in which TNF-{alpha}, IL-1ß, or proinflammatory activity was detected they were not concentrated at the onset of organ failure but, rather, were randomly distributed in the 7-day period of observation. Anti–IL-1 and anti-TNF activity in lymph and plasma correlated significantly with IL-1ra concentrations (r = 0.60; p = 0.006) and sTNF-RII levels (r = 0.39; p = 0.017), respectively.


View this table:
[in this window]
[in a new window]
  		TABLE 5. Detection of tnf-{alpha} and IL-1ß and pro- and antiinflammatory activity during thoracic duct drainage and peritoneal lavage

 
Time Course of Intracompartmental Mediator Levels and Intercompartmental Gradients
All secondary mediators and antiinflammatory activity remained markedly elevated in lymph and plasma throughout the study period (see Figure E3 in the online supplement). Concentrations of mediators were significantly higher in lymph than in plasma, except for IL-1ra levels, which were higher in plasma, and antiinflammatory activities, which were not significantly different in the two compartments. Plasma cytokine levels were found to be even more elevated in patients with shock, ARDS, and/or renal failure. The difference in mediator concentrations between the lymph and the blood compartments was accordingly reduced or abolished in some cases (see Figures E4–E9 in the online supplement). Except for IL-6 in lymph and plasma and IL-10 in lymph, the concentrations of which fell significantly during the period of the study, bioactivity and all other mediator levels did not change whatever the end-organ failure(s) (see Figures E3 and E4–E9 in the online supplement). Accordingly, molar ratios of antiinflammatory to proinflammatory cytokines and bioactivities did not differ significantly in lymph and plasma from the onset of MSOF throughout the course of drainage. Renal replacement therapy was initiated during the sampling period in 10 patients. Mediator levels in lymph and plasma were not significantly altered by this procedure (see Figures E10 and E11 in the online supplement).

Intercompartmental mediator gradients were analyzed only in paired samples during the first week after the onset of MSOF. A significant gradient between the three compartments was observed for all mediators except for anti-TNF activity and IL-10 between ascites and lymph, and anti-IL-1 activity between ascites and plasma (Figure 1) . The highest mediator levels were measured in ascites; levels were intermediate in the lymph and the lowest concentrations were found in plasma with the exception of IL-1ra. Accordingly, gradients were highest between ascites and plasma. We observed a nonsignificant tendency to decreased gradients between lymph and plasma with time. At the onset of drainage, lymph mediator levels significantly correlated with their corresponding levels in plasma except for IL-6 and anti–IL-1 activity. After 6 days of thoracic lymph drainage, a significant correlation between lymph and corresponding plasma concentrations was still found for all mediators (Table 6) .



View larger version (25K):
[in this window]
[in a new window]
  		Figure 1. Time course of intercompartmental gradients of cytokines [(A–D), and (G) and (H)] and antiinflammatory activity (E and F) during thoracic duct drainage and peritoneal lavage. Intercompartmental gradients were expressed as ratios of cytokine concentrations or bioactivity in paired samples (solid diamonds, gradient between ascites and lymph; solid triangles, gradient between ascites and plasma; solid circles, gradient between lymph and plasma). The abscissa indicates the line of identity between two compartments. Note the logarithmic scale on the ordinate in some panels. Figures at the bottom of each panel denote the number of paired samples. Anti–IL-1 = antiinflammatory activity ascribed to IL-1 inhibitors; anti-TNF = antiinflammatory activity ascribed to TNF inhibitors. *p < 0.05, **p < 0.01; levels of cytokines within one compartment compared with the other compartment, Wilcoxon's rank sum test.

 

View this table:
[in this window]
[in a new window]
  		TABLE 6. Spearman's correlation coefficients between plasma and corresponding lymph concentrations of cytokines and bioactivity

 
Mediator Levels and Bioactivities and Local and Systemic Complications
As TNF-{alpha}, IL-1ß, and proinflammatory activity could be detected in only 15% of the samples, peak levels were used to analyze their relationship with distant organ failures, local tissue injury, and outcome. For antiinflammatory activity and the other mediators, mean levels were used.

Systemic complications were quantified by APACHE II and SOFA scores and by the occurrence of major end-organ failure. No correlation was found between peak levels of TNF-{alpha}, IL-1ß, and proinflammatory activity in lymph and plasma and APACHE II and SOFA scores as well as the presence of end-organ failure (see Tables E2 and E3 in the online supplement). Plasma levels of all other cytokines and anti–IL-1 activity were closely correlated with APACHE II and SOFA scores. The correlation between severity and organ dysfunction scores and lymph mediator levels were looser and not significant for IL-6. Patients with shock had significantly higher mean plasma levels of IL-6, IL-8, and both soluble TNF receptors. The presence of criteria for ARDS significantly altered neither plasma levels of mediators nor bioactivities. Patients with shock and/or ARDS had higher lymph levels of IL-1ra and anti–IL-1 activity. Patients with renal failure (n = 30) showed an elevation in all mediators in both compartments compared with patients with normal renal function (see Tables E2 and E3 in the online supplement). The correlation between APACHE II score and lymph and plasma mediator levels remained significant after excluding patients with renal failure (data not shown).

Local complications were evaluated by CT Severity Index, extent of pancreatic necrosis, and early pancreatic infection. In plasma no correlation was found between levels of mediators and bioactivities and the extent of local damage as assessed by CT Severity Index (see Tables E4 and E5 in the online supplement). Plasma levels of mediators and bioactivities did not differ significantly in patients with pancreatic necrosis (< 50% necrosis or >= 50% necrosis). In contrast, lymph TNF-{alpha}, IL-1ra, and proinflammatory activity levels correlated significantly with the CT Severity Index. Patients with >= 50% necrosis had significantly higher lymph IL-6, IL-10, IL-1ra, and anti–IL-1 activity levels than did those with < 50% necrosis (Figure 2 ; and see Tables E4 and E5 in the online supplement).



View larger version (11K):
[in this window]
[in a new window]
  		Figure 2. Box-and-whisker plots of lymph (A) and plasma (B) mean concentrations of IL-6 according to the volume of pancreatic necrosis as graded by contrast-enhanced computed tomography. The central box represents values from the lower to upper quartile. The middle line represents the median. The vertical line extends from the minimum to the maximum value, excluding outside values depicted as square markers. The dots show individual outliers. *p < 0.05, < 30% versus 30–50%; **p < 0.01, < 30% versus > 50% (analysis of variance performed by using log10-transformed data).

 
Early pancreatic infection was not associated with a significant change in the rate of detection of TNF-{alpha}, IL-1ß, and proinflammatory activity in both compartments. However, mean plasma IL-8, IL-1ra, and sTNF-RII levels and mean lymph IL-6, IL-8, sTNF-RI, and sTNF-RII increased significantly with infection (see Tables E4 and E5 in the online supplement). Globally, the cytokine profile showed a parallel time course in plasma and lymph (Figure 3) . In the noninfected group, mediator levels did not change significantly with time, except for IL-6. In the infected group, mediator levels increased significantly with time in plasma and lymph, except for IL-10 in both compartments, IL-1ra in plasma, and sTNF-RII in lymph. In patients who developed an extrapancreatic infection, no significant differences were noted either in plasma or in lymph concentrations of mediators (data not shown).



View larger version (36K):
[in this window]
[in a new window]
  		Figure 3. (A–F) Changes with time in cytokine concentrations in lymph (solid squares) and plasma (solid triangles) according to the presence (solid lines) or absence (dashed lines) of early infected pancreatic necrosis. Data are expressed as medians plus pooled interquartile range. *P < 0.05, **P < 0.01, ***P < 0.001, infection compared with no infection (Mann–Whitney U test); {dagger}P < 0.05, days 5–6 compared with days 1–2, {dagger}{dagger}P < 0.05, days 5–6 compared with days 3–4 (Friedman analysis of variance followed by Wilcoxon's rank sum test).

 
Concentrations of mediators were significantly higher in nonsurvivors than in survivors except for IL-1ra and anti–IL-1 activity in plasma, IL-6 and anti–IL-1 activity in lymph, and TNF-{alpha}, IL-1ß, and proinflammatory activity in both compartments (see Tables E4 and E5 in the online supplement).


   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
A specific treatment directed to the inflammatory and necrotizing process, as well as to early MSOF, would certainly decrease the considerable morbidity and mortality that still affect patients with severe acute pancreatitis. Therefore, improved knowledge of the pathophysiologic mechanisms underlying both local necrosis and distant organ damage is of utmost importance to impinge on these factors by more than supportive measures. Unlike previous related investigations (21–26), the main goal of the present study was to address the net immunologic and biological balance between the pro- and antiinflammatory mediators simultaneously in the systemic circulation and in two anatomic compartments that drain the primary inflammatory focus. The design of the study allowed us to assess the relative contribution of local versus distant mediator production as well as the routes of mediator transfer from one compartment to the other, giving insight as to how a local inflammatory process propagates into a systemic illness.

Balance between Pro- and Antiinflammatory Responses
Whereas a sustained and early release of several pro- and antiinflammatory cytokines could be demonstrated in the three compartments, the proximal mediators of the inflammatory cascade, TNF-{alpha} and IL-1ß, were detected in the thoracic lymph and plasma in only a minority of patients. In contrast, IL-1ß was detected in the peritoneal compartment in half the patients. Data on cytokine levels in ascites and lymph during human acute pancreatitis are scarce. High and matched concentrations of TNF-{alpha} were detected in blood and lymph from a small group of six patients with severe acute pancreatitis and ARDS, but four of these patients had either infected necrosis or bacterial pneumonia (28). In accordance with our results, high levels of IL-1ß and other cytokines were measured in ascites or lesser sac aspirate from 20 patients with severe acute pancreatitis (21). A 100-fold increase in IL-1ß was found locally compared with peak systemic concentrations (21). Data on the cytokine profile in peripheral blood of patients with severe acute pancreatitis are, by comparison, considerable. Two studies have shown an early surge of secondary pro- and compensatory antiinflammatory cytokines that persisted throughout an extended period of time (21, 24). In line with our results, TNF-{alpha} and IL-1ß were detected only in 6 to 36% of the patients (21, 23, 24). Several factors may account for the low detection rate of proinflammatory cytokines in the blood and lymph compartments. These cytokines have a short half-life; their peak concentrations could easily be missed by daily blood and lymph sampling whereas clearance might be slower in the peritoneal cavity (39). These cytokines are produced primarily within tissues, act in an autocrine or paracrine fashion, and may not always reach the circulation. Transient and repetitive bursts in the circulation or spillover from the peritoneal cavity may explain the sporadic detection of TNF-{alpha} and IL-1ß in some of our patients throughout the period of sampling. Finally, natural circulating inhibitors and activated proteases released into the circulation by the gland may either interfere with immunoassays or alter cytokine levels or function.

The biological activity of a cytokine in a given biological fluid results from the balance between the cytokine and its antagonist(s) and, therefore, is not reflected by the simple measurement of the antigenic level in immunoassays. The net proinflammatory activity of the three biological fluids sampled in the present study was measured by quantifying the upregulation of ICAM-1 in target human alveolar type II-like A549 cells. This bioassay, sensitive to minute quantities of bioactive TNF-{alpha} and IL-1ß, has been previously validated for the measurement of proinflammatory activity in plasma, bronchoalveolar lavage, and edema fluids (36, 37, 40). The absence of net proinflammatory activity in a fluid may result from an excess of natural TNF-{alpha} and IL-1ß inhibitors, that is, soluble receptors and receptor antagonists, blocking the proinflammatory cytokines. To measure a possible net antiinflammatory activity, we developed antiinflammatory bioassays that allowed us to quantify net "anti–TNF-{alpha} activity" and "anti–IL-1ß activity." These bioassays reflected the amount of natural circulating inhibitors to either TNF-{alpha} or IL-1ß. The net biological activity in the vast majority of lymph and peripheral blood samples was strongly antiinflammatory from the onset of MSOF throughout the first week. This suggests that a rapid, robust, and dominant antiinflammatory response takes place in circulatory compartments, most probably in response to local proinflammatory stimuli. This response was consistent with the presence in virtually all lymph and plasma samples of a 100- to 1,000-fold excess molar ratio of soluble TNF receptors and IL-1 receptor antagonist (41, 42). A modest proinflammatory activity was sporadically detected throughout the course of the study in only about 10% of several hundred lymph and blood samples. It should be emphasized that these biological fluids could possibly have elicited a proinflammatory response by using a different target cell or by other proinflammatory mediators that are short-lived, or cannot be detected with the A549 cell bioassay. In addition, it is possible that we missed an early but short-lasting window of circulating proinflammatory activity because of the delay between the onset of the attack and sampling, inasmuch as most patients were sampled within 24 hours of the onset of MSOF. Nevertheless, our results concur with previous studies conducted in other "systemic inflammatory response syndrome" situations such as sepsis and ARDS that demonstrated the prominence of the counterregulatory, antiinflammatory response in the systemic circulation mirroring a net proinflammatory reaction within the alveolar compartment (42–44). Severe acute pancreatitis is therefore another severe condition with acute local tissue damage that is accompanied by a dominant systemic antiinflammatory response. It has been proposed that an excessive systemic antiinflammatory response during tissue injury may induce immune suppression, and be responsible for an increased risk of developing secondary, nosocomial infection (45).

Although the number of samples was limited, significant net proinflammatory activity was found in the peritoneal compartment and correlated with the antigenic level of IL-1ß in ascites. Despite increased concentrations of IL-1ra, it did not balance local IL-1ß bioactivity in these samples. A similar close relationship between the bioassay and the immunologic measurement of IL-1ß was found in the bronchoalveolar lavage fluid of patients with ARDS (36, 40). A lack of significant correlation between IL-1ra and the net biological activity of IL-1ß was also observed in these patients (40). The presence of net proinflammatory activity in the peritoneal compartment suggests that local cells, possibly peritoneal or tissue macrophages, preferentially produce proinflammatory cytokines. In contrast, circulating leukocytes dominantly express antiinflammatory mediators to prevent the deleterious and unwanted systemic spreading of bioactive proinflammatory mediators (45, 46). The findings of the present study are consistent with those of other studies done in patients with various inflammatory conditions (36, 37, 40, 47–49). In these studies, proinflammatory cytokines were confined within the primary inflammatory focus with no or minor spillover into the circulation.

Site of Mediator Production and Pathways of Transfer
Cytokine levels as measured by immunoassays were consistently highest in ascites, intermediate in lymph, and lowest in plasma except for IL-1ra. Similarly, biological activity, either proinflammatory (caused by IL-1ß) or antiinflammatory (ascribed to IL-1ra), was highest in the peritoneal compartment. Several mechanisms may account for the markedly elevated cytokine levels in peritoneal ascites: (1) a spillover of mediators from the primary retroperitoneal inflammatory process, (2) a local synthesis by activated peritoneal macrophages, and (3) a spillover of cytokines from the vascular compartment caused by increased capillary permeability with peritoneal trapping. However, analyses of mediator and biological activity gradients strongly suggest a splanchnic source of many of these mediators. Our data are also in agreement with clinical and animal studies that demonstrated gene induction and cytokine production early after the initiation of pancreatitis, first in the pancreas and after a significant delay in distant organs (9, 21, 28, 39).

Despite the continuous lavage of the peritoneal cavity and external drainage of the thoracic duct, the gradients in cytokine levels between the lymph and blood compartments remained mostly unaltered during the 7-day period of observation irrespective of the volume of diverted lymph. Thus, unless complete anatomic diversion of the thoracic lymph was not achieved and significant systemic synthesis or lung release of mediators compensated for mediator removal, neither peritoneal lavage nor thoracic duct drainage influenced plasma levels of cytokines. Because the primary source of cytokine production seems to be in the splanchnic area, portal and hepatic venous transfer to the systemic circulation most likely accounts for the drainage of inflammatory mediators into the systemic compartment. Altogether, these results suggest that both plasma and lymph compartments are "fueled" by the same inflammatory focus located in the splanchnic territory. The relative contribution of circulating leukocytes to systemic levels of cytokines was comparatively less significant, except perhaps for IL-1ra. Supporting this, results obtained in experimental acute pancreatitis demonstrated that tissue rather than circulating leukocytes are the major source of inflammatory mediators (9, 50).

Correlation with Local and Systemic Complications
Cytokine levels in blood and lymph were closely associated with the severity of illness on admission, the magnitude of MSOF as well as with outcome. The statistical association between severity scores and levels of mediators was stronger in plasma than in lymph. The low detection rate of TNF-{alpha}, IL-1ß, and proinflammatory activity accounts for the lack of significant relationship between these mediators and organ dysfunction or severity scores.

The pathogenic role of the inflammatory cascade in local tissue injury, although well established in experimental acute pancreatitis (9, 10), has not been properly assessed in human disease. A significant association between mediator levels and the extent of local tissue injury was found with lymph but not with plasma. The lymphatic compartment is therefore more representative than the systemic circulation to assess local tissue damage. Spillover of mediators from distant organs into the circulation may account for a looser association between mediator concentrations and the volume of pancreatic necrosis (9).

In this study, 10 patients (17%) had unequivocal pancreatic infection shortly after initiation of thoracic duct drainage. We considered that the other patients had sterile necrosis, an assumption that was substantiated by their clinical course and microbiological cultures although some patients may have had subclinical infectious processes. Interestingly, early infected pancreatic necrosis was not associated with a significantly higher detection rate of lymph and blood TNF-{alpha}, IL-1ß, and proinflammatory activity. This finding is consistent with the concept of compartmentalization of the initial proinflammatory response at the site of the inflammation/infection, with minor or absent mediator spillover into the circulatory compartments, even in the lymph that drains infected tissues (36, 37, 40, 47–49). In contrast to TNF-{alpha} and IL-1ß, "secondary" cytokines and chemokines, as well as antiinflammatory mediators, increased significantly and ran a parallel course in blood and lymphatic compartments during infection. This mediator surge most likely represented the systemic sequel of the local release of first-line proinflammatory cytokines after bacterial contamination of necrosis. Finally, a strong "antiinflammatory" systemic response may well represent a risk factor for the development of sepsis in this population (45).

Conclusion and Limitations
Several potential limitations of this study should be noted. The study population was relatively young, the initiating factor was predominantly alcohol abuse, and the majority of the patients were transferred, albeit with only a short delay, from referring hospitals. However, this population does provide an accurate representation of the severe end of the spectrum of necrotizing pancreatitis (51). Only a small subset of susceptible patients with acute pancreatitis progresses to extensive necrosis and distant organ failure, and until now a rapid and reliable prediction of these patients on an individual basis is not possible (24). Therefore, the onset of MSOF, and not the onset of acute pancreatitis, was chosen as the reference point in our study, which may have imposed an additional uncontrolled variable on data analysis. This was dictated by the design of the clinical protocol, in which patients had peritoneal and thoracic duct drainage only when extensive necrosis and organ dysfunction were present. We have focused our study on inflammatory mediators as potential pathogenic factors in severe acute pancreatitis. It should be clear, however, that other important local and systemic biological pathways such as coagulation, fibrinolysis, and protease and lipase enzymatic cascades may play important roles in the pathogenesis of the disease, possibly in close relationship with inflammation.

In summary, the prominent findings of our study are the following: (1) severe acute pancreatitis and early MSOF are associated with a sustained release of both pro- and antiinflammatory cytokines in the ascitic fluid, the thoracic lymph, and the systemic circulation; (2) a net proinflammatory activity imputable to bioactive IL-1ß was measured only in the peritoneal compartment, which is in close vicinity to the site of pancreatic inflammation and necrosis; (3) the net biological activity is antiinflammatory in circulating compartments as soon as the acute pancreatitis is accompanied by remote organ dysfunctions. Severe acute pancreatitis represents therefore another clinical situation, in addition to sepsis and ARDS, with a local proinflammatory response, and a systemic antiinflammatory response syndrome; (4) gradients of inflammatory mediators from ascites to lymph to blood persisted throughout the duration of the study, despite prolonged lavage of the peritoneal cavity and drainage of the thoracic duct; and (5) the extent of local necrosis correlates only with lymph inflammatory mediator levels, whereas plasma levels of mediators and biological activity are closely associated with the severity of MSOF.

Although experimental studies support a pathogenic role of inflammatory mediators in both the extent of local necrosis and the development of remote organ dysfunction, these associations have not been proven in humans. A definite proof of such an association should come from therapeutic trials aimed at modulating inflammatory pathways. In that case, investigators should definitely consider the importance of the compartmentalization of the inflammatory response, and the routes of drainage of toxic mediators in this disease.


   Acknowledgments
 
The authors thank Jean Paul Squifflet for thoracic duct cannulation, Geneviève Hanique for statistical advice, and Dominique Lisenko and Mbaye Cissé for excellent technical assistance. The authors also express their gratitude to Carline Vandenbroucke and Viviane Van Meensel for invaluable assistance in the preparation of the manuscript.


   FOOTNOTES
 
This article has an online supplement, which is accessible from this issue's table of contents online at www.atsjournals.org

Received in original form April 10, 2002; accepted in final form March 19, 2003


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Rau B, Pralle U, Uhl W, Schoenberg MH, Beger HG. Management of sterile necrosis in instances of severe acute pancreatitis. J Am Coll Surg 1995;181:279–288.[Medline]
  2. Neoptolemos JP, Raraty M, Finch M, Sutton R. Acute pancreatitis: the substantial human and financial costs. Gut 1998;42:886–891.[Abstract/Free Full Text]
  3. Beger HG, Bittner R, Block S, Büchler M. Bacterial contamination of pancreatic necrosis: a prospective clinical study. Gastroenterology 1986;91:433–438.[Medline]
  4. Widdison AL, Karanjia ND. Pancreatic infection complicating acute pancreatitis. Br J Surg 1993;80:148–154.[Medline]
  5. deBeaux AC, Palmer KR, Carter DC. Factors influencing morbidity and mortality in acute pancreatitis: an analysis of 279 cases. Gut 1995;37:121–126.[Abstract/Free Full Text]
  6. Renner IG, Savage WT, Pantoja JL, Renner VJ. Death due to acute pancreatitis: a retrospective analysis of 405 autopsy cases. Dig Dis Sci 1985;30:1005–1018.[CrossRef][Medline]
  7. McKay CJ, Evans S, Sinclair M, Carter CR, Imrie CW. High early mortality rate from acute pancreatitis in Scotland, 1984–1995. Br J Surg 1999;86:1302–1305.[CrossRef][Medline]
  8. Johnson C, Kingsnorth A, Imrie C, McMahon M, Neoptolemos JP, McKay C, Toh S, Skaife P, Leeder P, Wilson P, et al. Double blind, randomised, placebo controlled study of a platelet activating factor antagonist, lexipafant, in the treatment and prevention of organ failure in predicted severe acute pancreatitis. Gut 2001;48:62–69.[Abstract/Free Full Text]
  9. Norman J, Fink G, Denham W, Yang J, Carter G, Sexton C, Falkner J, Gower W, Franz M. Tissue specific cytokine production during experimental acute pancreatitis: a probable mechanism for distant organ dysfunction. Dig Dis Sci 1997;42:1783–1788.[CrossRef][Medline]
  10. Rinderknecht H. Fatal pancreatitis, a consequence of excessive leukocyte stimulation? Int J Pancreatol 1988;3:105–112.[Medline]
  11. Lundberg A, Granger N, Russel J, Sabek O, Henry J, Gaber L, Kotd M, Gaber AO. Quantitative measurement of P- and E-selectin adhesion modecules in acute pancreatitis: correlation with distant organ injury. Ann Surg 2000;231:213–222.[CrossRef][Medline]
  12. Frossard JL, Saluja A, Bhagat L, Lee HS, Bhatia M, Hofbauer B, Steer ML. The role of intercellular adhesion molecule 1 and neutrophils in acute pancreatitis and pancreatitis-associated lung injury. Gastroenterology 1999;116:694–701.[CrossRef][Medline]
  13. Hofbauer B, Saluja AK, Bhatia M, Frossard JL, Lee HS. Bhagat L Steer ML. Effect of recombinant platelet-activating factor acetylhydrolase on two models of experimental acute pancreatitis. Gastroenterology 1998;115:1238–1247.[CrossRef][Medline]
  14. Yang J, Denham W, Tracey K, Wang H, Kramer A, Slahab K, Norman J. The physiologic consequences of macrophage pacification during severe acute pancreatitis. Shock 1998;10:169–175.[Medline]
  15. Osman M, Jacobsen N, Kristensen J, Deleuran B, Gesser B, Larsen C, Jensen S. IT 9302, a synthetic interleukin-10 agonist, diminishes acute lung injury in rabbits with acute necrotizing pancreatitis. Surgery 1998;124:584–592.[Medline]
  16. Lundberg A, Fukatsu K, Gaber L, Callicutt S, Kotb M, Wilcox H, Kudsk K, Gaber A. Blocking pulmonary ICAM-1 expression ameliorates lung injury in established diet-induced pancreatitis. Ann Surg 2001;233:213–220.[CrossRef][Medline]
  17. Imrie CW. The possible role of platelet-activating factor antagonist therapy in the management of severe acute pancreatitis. Baillieres Clin Gastroenterol 1999;13:357–364.[CrossRef]
  18. Osman MO, Kristensen JU, Jacobsen N, Lausten S, Deleuran B, Deleuran M, Gesser B, Matsushima K, Larsen C, Jensen S. A monoclonal anti-interleukin 8 antibody (WS-4) inhibits cytokine response and acute lung injury in experimental severe acute necrotising pancreatitis in rabbits. Gut 1998;43:232–239.[Abstract/Free Full Text]
  19. Hughes CB, Grewal HP, Gaber L, Kotb M, El-din AB, Mann L, Gaber A. Anti-TNF-{alpha} therapy improves survival and ameliorates the pathophysiologic sequelae in acute pancreatitis in the rat. Am J Surg 1996;171:274–280.[CrossRef][Medline]
  20. Tanaka N, Murata A, Uda K, Toda H, Kato T, Hayashida H, Matsuura N, Mori T. Interleukin-1 receptor antagonist modifies the changes in vital organs induced by acute necrotizing pancreatitis in a rat experimental model. Crit Care Med 1995;23:901–908.[CrossRef][Medline]
  21. Mayer J, Rau B, Gansauger F, Beger H. Inflammatory mediators in human acute pancreatitis: clinical and pathophysiological implications. Gut 2000;47:546–552.[Abstract/Free Full Text]
  22. McKay CJ, Gallagher G, Brooks B, Imrie CW, Baxter JN. Increased monocyte cytokine production in association with systemic complications in acute pancreatitis. Br J Surg 1996;83:919–923.[Medline]
  23. de Beaux AC, Goldie AS, Ross JA, Carter DC, Fearon KC. Serum concentrations of inflammatory mediators related to organ failure in patients with acute pancreatitis. Br J Surg 1996;83:349–353.[Medline]
  24. Brivet F, Emilie D, Galanaud P. Pro- and anti-inflammatory cytokines during acute severe pancreatitis: an early and sustained response, although unpredictable of death. Parisian Study Group on Acute Pancreatitis. Crit Care Med 1999;27:749–755.[CrossRef][Medline]
  25. Ikei S, Ogawa M, Yamaguchi Y. Severity of acute pancreatitis: blood concentrations of polymorphonuclear leucocyte elastase and interleukin-6 are indicators for the occurrence of multiple organ failures at the early stage of acute pancreatitis. J Gastroenterol Hepatol 1998;13:1274–1283.[CrossRef][Medline]
  26. Hirota M, Nozawa F, Okabe A, Shibata M, Beppu T, Shimada S, Egami H, Yamaguchi Y, Ikei S, Okajima K, et al. Relationship between plasma cytokine concentration and multiple organ failure in patients with acute pancreatitis. Pancreas 2000;21:141–146.[CrossRef][Medline]
  27. Denham W, Yang J, Fink G, Zervos E, Carter G, Norman J. Pancreatic ascites as a powerful inducer of inflammatory cytokines: role of known versus unknown factors. Arch Surg 1997;132:1231–1236.[Abstract]
  28. Montravers P, Chollet-Martin S, Marmuse JP, Gougerot-Picidalo MA, Desmonts JM. Lymphatic release of cytokines during acute lung injury complicating severe pancreatitis. Am J Respir Crit Care Med 1995;152:1527–1533.[Abstract]
  29. Reynaert MS, Dugernier T, Kestens PJ. Current therapeutic strategies in severe acute pancreatitis. Intensive Care Med 1990;16:352–362.[CrossRef][Medline]
  30. Frossard J-L, Hadengue A, Pastor C. New serum markers for the detection of severe acute pancreatitis in humans. Am J Respir Crit Care Med 2001;164:162–170.[Free Full Text]
  31. Moreno R, Vincent JL, Matos R, Mendoça A, Cantraine F, Thijs L, Takala J, Sprung C, Antonelli M, Bruining H, et al. The use of maximum SOFA score to quantify organ dysfunction/failure in intensive care: results of a prospective, multicentre study. Intensive Care Med 1999;25:686–696.[CrossRef][Medline]
  32. Bernard R, Artigas A, Brigham K. The American–European consensus conference on ARDS: definitions, mechanisms, relevant outcomes and clinical trial coordination. Am J Respir Crit Care Med 1994;149:818–824.[Abstract]
  33. Balthazar EJ, Robinson DL, Megibow AJ, Ranson JH. Acute pancreatitis: value of CT in establishing prognosis. Radiology 1990;174:331–336.[Abstract/Free Full Text]
  34. Reynaert MS, Bshouty ZH, Otte JB, Kestens PJ, Tremouroux J. Percutaneous peritoneal dialysis as an early treatment of acute necrotic hemorrhagic pancreatitis. Intensive Care Med 1985;11:123–128.[Medline]
  35. Dugernier T, Reynaert MS, Deby-Dupont G, Roeseler JJ, Carlier M, Squifflet JP, Deby D, Pincemail J, Lamy M, De Maeght S, et al. Prospective evaluation of thoracic-duct drainage in the treatment of respiratory failure complicating severe acute pancreatitis. Intensive Care Med 1989;15:372–378.[Medline]
  36. Pugin J, Ricou B, Steinberg P, Suter PM, Martin TR. Proinflammatory activity in bronchoalveolar lavage fluids from patients with ARDS, a prominent role for interleukin-1. Am J Respir Crit Care Med 1996;153:1850–1856.[Abstract]
  37. Pugin J, Verghese G, Widmer MC, Matthay MA. The alveolar space is the site of intense inflammatory and profibrotic reactions in the early phase of acute respiratory distress syndrome. Crit Care Med 1999;27:304–312.[CrossRef][Medline]
  38. Schoonjans F, Zalata A, Depuydt C, Comhaire FH. MedCalc: a new computer program for medical statistics. Comp Methods Prog Biomed 1995;48:257–262.[CrossRef][Medline]
  39. Grewal HP, Kotb M, El-din AB, Ohman M, Salem A, Gaber L, Gaber AO. Induction of tumor necrosis factor in severe acute pancreatitis and its subsequent reduction after hepatic passage. Surgery 1994;115:213–221.[Medline]
  40. Park W, Goodman R, Steinberg K, Ruzinski J, Radella II F, Park D, Pugin J, Skerrett S, Hudson L, Martin T. Cytokine balance in the lungs of patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 2001;164:1896–1903.[Abstract/Free Full Text]
  41. Heaney ML, Golde DW. Soluble receptors in human disease. J Leukoc Biol 1998;64:135–146.[Abstract]
  42. Arend WP, Joslin FG, Thompson RC, Hannum CH. An IL-1 inhibitor from human monocytes: production and characterization of biologic properties. J Immunol 1989;143:1851–1858.[Abstract]
  43. Brandtzaeg P, Osnes L, Ovstebo R, Joo GB, Westvik AB, Kierulf P. Net inflammatory capacity of human septic shock plasma evaluated by a monocyte-based target cell assay: identification of interleukin-10 as a major functional deactivator of human monocytes. J Exp Med 1996;184:51–60.[Abstract/Free Full Text]
  44. Goldie A, Fearon K, Ross J, Barclay G, Jackson R, Grant I, Ramsay G, Blyth A, Howie J. Natural cytokine antagonists and endogenous antiendotoxin core antibodies in sepsis syndrome. JAMA 1995;274:172–177.[Abstract]
  45. Munford R, Pugin J. Normal responses to injury prevent systemic inflammation and can be immunosuppressive. Am J Respir Crit Care Med 2001;163:316–321.[Free Full Text]
  46. Satoh A, Shimosegawa T, Masamune A, Fujita M, Koizumi M, Toyota T. Ascitic fluid of experimental severe acute pancreatitis modulates the function of peritoneal macrophages. Pancreas 1999;3:268–275.
  47. Dehoux MS, Boutten A, Ostinelli J, Seta N, Dombret MC, Crestani B, Desches M, Trouillet JL, Aubier M. Compartmentalized cytokine production within human lung in unilateral pneumonia. Am J Respir Crit Care Med 1994;150:710–716.[Abstract]
  48. Leblond-Tillie I, Pugin J, Marquette CH, Lamblin C, Saulnier F, Brichet A, Wallaert B, Tonnel AB, Gosset P. Balance between proinflammatory cytokines and their inhibitors in bronchial lavage from patients with status asthmaticus. Crit Care Med 1999;159:487–494.
  49. Holzheimer R, Schein M, Wittmann DH. Inflammatory response in peritoneal exudate and plasma of patients undergoing planned relaparotomy for severe secondary peritonitis. Arch Surg 1995;130:1314–1320.[Abstract]
  50. Fink GW, Norman J. Intrapancreatic interleukin 1 gene expression by specific leukocyte populations during acute pancreatitis. J Surg Res 1996;63:369–373.[CrossRef][Medline]
  51. Isenmann R, Rau B, Beger H. Early severe acute pancreatitis: characteristics of a new subgroup. Pancreas 2001;22:274–278.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 J. Leukoc. Biol.Home page
S. F. Lowry and S. E. Calvano
Challenges for modeling and interpreting the complex biology of severe injury and inflammation
J. Leukoc. Biol., March 1, 2008; 83(3): 553 - 557.
[Abstract] [Full Text] [PDF]

Home page
 Arch SurgHome page
D. Parekh
Laparoscopic-Assisted Pancreatic Necrosectomy: A New Surgical Option for Treatment of Severe Necrotizing Pancreatitis
Arch Surg, September 1, 2006; 141(9): 895 - 903.
[Abstract] [Full Text] [PDF]

Home page
 Innate ImmunityHome page
J.-M. Cavaillon and D. Annane
Invited review: Compartmentalization of the inflammatory response in sepsis and SIRS
Innate Immunity, June 1, 2006; 12(3): 151 - 170.
[Abstract] [PDF]

Home page
 Innate ImmunityHome page
R. S. Munford
Invited review: Detoxifying endotoxin: time, place and person
Innate Immunity, April 1, 2005; 11(2): 69 - 84.
[Abstract] [PDF]

Home page
 Am. J. Respir. Crit. Care Med.Home page
M. J. Tobin
Critical Care Medicine in AJRCCM 2003
Am. J. Respir. Crit. Care Med., January 15, 2004; 169(2): 239 - 253.
[Full Text] [PDF]

Home page
 Am. J. Respir. Crit. Care Med.Home page
J. P. Neoptolemos
Compartments That Cause the Real Damage in Severe Acute Pancreatitis
Am. J. Respir. Crit. Care Med., July 15, 2003; 168(2): 141 - 142.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Online Supplement
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dugernier, T. L.
Right arrow Articles by Pugin, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dugernier, T. L.
Right arrow Articles by Pugin, J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Proc. Am. Thorac. Soc. Am. J. Respir. Cell Mol. Biol.
Copyright © 2003 American Thoracic Society