INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Adrenomedullin (AM) is a newly identified endogenous hypotensive peptide that was isolated from human pheochromocytoma tissue by monitoring the increase in cyclic adenosine monophosphate (cAMP) in rat platelets exposed to extracts of this tissue (1). AM consists of 52 amino acids, and shares slight homology with calcitonin gene-related peptide (CGRP) (2). Several studies have shown that the vasodilatory effect of AM is potent and long-lasting, and comparable to that of CGRP (2). Ichiki and colleagues reported that AM was widely distributed in human tissues including the adrenal medulla, atrium, lung, pancreas, and small intestine (2). Recent studies have shown that AM is produced and secreted abundantly from cultured vascular smooth-muscle cells and vascular endothelial cells, and that its production is most potently augmented by lipopolysaccharide (LPS), tumor necrosis factor (TNF), and interleukin (IL)-1s (2). Furthermore, these vascular cells express AM-specific receptors linked to the adenylate cyclase-cAMP system. The presence of these receptors on vascular cells suggests that AM may be one of the major local factors regulating vascular tone. AM also circulates in plasma, and plasma levels of AM are reported to be increased in a variety of diseases such as essential hypertension, chronic heart failure, and chronic renal failure, in proportion to the severity of disease (2). These findings have suggested that AM functions not only as a local vasorelaxant but as a circulating systemic vasodilator.

Sepsis has been recognized as the systemic inflammatory response to an active infectious process in the host. A similar or identical systemic inflammatory response can be induced even in the absence of infection. This inflammatory response has been termed systemic inflammatory response syndrome (SIRS) (3). SIRS can be induced by a wide variety of insults, including burns, trauma, and pancreatitis in addition to infection. These insults stimulate the release of various mediators from inflammatory cells or vascular cells, subsequently leading to SIRS. In this context, proinflammatory cytokines such as TNF- and IL-1 are known to be the key mediators responsible for SIRS. Considering that these proinflammatory cytokines are the most potent stimuli of AM production in vascular cells, it is possible that AM also plays a pathophysiologic role in SIRS. On the basis of this possibility, we conducted a study of whether plasma levels of AM were increased in patients with various forms of SIRS, and whether their plasma AM levels were associated with the severity of their SIRS. Focusing on patients with traumatic shock and septic shock, we also measured plasma levels of several humoral mediators in addition to AM, including proinflammatory cytokines, plasminogen activator inhibitor (PAI)-1, and thrombomodulin (TM), each of which was previously reported as a marker of severity in trauma and sepsis. We also compared the levels of AM and of these mediators with severity of illness to identify the right marker for evaluating the severity of traumatic and septic shock.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Study Population

We studied 89 patients who were admitted to the emergency room and the intensive care unit of the Nara Medical University Hospital, after obtaining the approval of our institutional review board for the study. At the time of admission, all patients fulfilled the clinical criteria for SIRS as recently defined by the Members of the American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference Committee (3). Eighty-nine patients were classified into six groups, as follows: (1) 10 patients with major burns (age: 58.5 ± 7.3 yr [mean ± SEM]) (burns group); (2) six patients with acute pancreatitis (age: 69 ± 2.9 yr) (pancreatitis group); (3) 11 trauma patients without shock (age: 34.2 ± 6.8 yr) (trauma group); (4) 16 patients with traumatic shock (age: 45.5 ± 5.3 yr) (traumatic shock group); (5) 16 patients with severe sepsis (age: 54.1 ± 4.7 yr) (severe sepsis group); and (6) 30 patients with septic shock (age: 63.2 ± 2.6 yr) (septic shock group). The diagnostic criteria used on admission were: (1) extensive thermal injury with a total burn surface area of over 30% (burns); (2) characteristic abdominal pain in association with increases in serum amylase or lipase levels, and contrast enhanced computed tomography (CT) showing necrotizing pancreas (pancreatitis); (3) major trauma (Injury Severity Scores > 9) (4) and systolic pressure > 90 mm Hg (trauma); (4) major trauma with hypotension (systolic pressure < 90 mm Hg or decrease > 40 mm Hg from the baseline) (traumatic shock); (5) SIRS with a suspected or proven site of infection and hypotension (severe sepsis); and (6) sepsis with hypotension despite adequate fluid resuscitation (septic shock). All patients with traumatic shock or septic shock received increasing doses of dopamine as necessary to maintain systolic blood pressure above 90 mm Hg.

The patients were considered "survivors" when they were alive 28 d after inclusion in the study, even if they had already been discharged from the hospital; the others were considered "nonsurvivors."

Samples Collection

All blood samples were obtained within 6 h after admission, and were collected with disodium ethylenediamine tetraacetate (Na²⁺ EDTA) (1 mg/ml) and aprotinin (Trasylol; Bayer, Leverkusen, Germany; 500 kIU/ml) for the AM assay, and with sterile heparin for other assays, and were cooled on ice. Plasma was separated by centrifugation at 3000 rpm for 10 min at 4° C and stored at 80° C until assayed. Thirteen blood samples were obtained from healthy volunteers (seven men, six women; age 23 to 44 yr, mean: 32 yr) as controls.

Radioimmunoassay for AM

Plasma AM was measured by radioimmunoassay (RIA) after extraction and purification, as previously described (5).

Assays for Cytokines, PAI-1, and TM

Plasma levels of cytokines (TNF-, IL-6, IL-8) and PAI-1 were quantitated with commercially available enzyme-linked immunosorbent assay (ELISA) kits (IMMUNOTECH International, Tassigny, France, for cytokines; Momozyme, Hoersholm, Denmark, for PAI-1). Plasma levels of TM were also measured with a commercially available enzyme immunoasssay (Fuji Revio, Tokyo, Japan).

Additional Clinical Data

In the case of patients with traumatic or septic shock, the clinical severity of these conditions was evaluated at the time of blood sampling, using the Acute Physiology and Chronic Health Evaluation (APACHE) II system (6). Furthermore, the multiple organ failure (MOF) score, as previously reported by Goris (7), was determined on each study day. The peak score for MOF during the first month after admission was used to grade organ failure in each patient.

Presentation of Data and Statistical Analysis

Values were expressed as the mean ± SEM. Student's t test and the Mann-Whitney U test were used to examine differences between patient groups when indicated. Spearman's and Pearson's correlation coefficients were used for correlation analysis when indicated. We considered values of p < 0.05 to be statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Plasma Levels of AM

In the healthy volunteers, plasma levels of AM averaged 5.1 ± 0.2 fmol/ml, ranged from 4.1 to 6.6 fmol/ml, and were extremely constant. By comparison, plasma levels of AM in patients with SIRS ranged from 4.6 to 786.4 fmol/ml, and the mean (88.4 ± 13.2 fmol/ml) was significantly higher than that of controls (p < 0.01) (Figure 1, Table 1). Mean plasma levels of AM in each group of SIRS patients were as follows: burns group, 20.5 ± 3.2 fmol/ml, range: 5.3 to 40.4 fmol/ml; pancreatitis group, 13.8 ± 3.8 fmol/ml, range: 4.9 to 31.2 fmol/ml; trauma group, 14.9 ± 2.5 fmol/ml, range: 4.6 to 30.8 fmol/ml; traumatic shock group, 41.1 ± 7.8 fmol/ml, range: 5.6 to 127.7 fmol/ml; severe sepsis group, 59.9 ± 11.2 fmol/ml, range: 10.0 to 126.9 fmol/ml; septic shock group, 193.5 ± 30.1 fmol/ml, range: 13.6 to 786.4 fmol/ml. The average for each group was significantly higher than that for the control group. The highest level and the highest mean value were observed in the septic shock group. Plasma AM level overlapped to some extent between any two groups of patients examined, but an extremely high level of plasma AM (> 150 fmol/ml) was observed only in the septic shock group. The mean level of plasma AM in the traumatic shock group was significantly higher than that in the trauma group. Similarly, the mean plasma AM level in the septic shock group was significantly higher than that in the severe sepsis group.

View larger version (13K): [in this window] [in a new window]   Figure 1.   Plasma levels of AM in patients with SIRS. Open circles and bars indicate mean value and SEM, respectively.

Influence of Renal Function on Plasma AM Levels

To clarify the influence of renal clearance on plasma AM, we compared plasma AM levels with serum creatinine levels in each study group. Table 1 shows serum creatinine levels in each group. Serum creatinine levels correlated positively with plasma AM levels in the trauma (r = 0.701, p = 0.016) and severe sepsis (r = 0.618, p = 0.011) groups, whereas such correlation was not observed in the burns group, pancreatitis group, traumatic shock group, or septic shock group. The presence of renal dysfunction seemed to contribute to the increase in plasma AM levels when patients were not in the shock state.

Plasma Levels of Humoral Mediators

Plasma levels of proinflammatory cytokines (TNF-, IL-6, IL-8), PAI-1, and TM were measured in patients with traumatic shock and those with septic shock. Mean plasma levels of TNF-, IL-6, IL-8, and PAI-1 in these two groups exceeded the normal upper limits, although the mean plasma level of TM did not exceed the normal upper limit of normal in the traumatic shock group. The mean levels of TNF-, IL-6, IL-8, and TM as well as of AM in the septic shock group were significantly higher than those in the traumatic shock group, whereas no significant difference between these two groups was observed for PAI-1 (Table 2).

Correlation of Plasma AM Levels with Plasma Levels of Humoral Mediators

To obtain an insight into the factors influencing plasma AM concentration, we compared plasma AM levels with levels of other humoral mediators. In the traumatic shock group, plasma AM levels did not correlate with plasma levels of TNF-, IL-6, IL-8, PAI-1, or TM, whereas in the septic shock group, plasma levels of AM correlated with plasma TNF- levels (r = 0.510, p = 0.004), but not with plasma IL-6, IL-8, PAI-1, or TM levels.

Correlation of Plasma AM and Humoral Mediator Levels with APACHE II Scores and MOF Scores

In the traumatic shock group, only plasma IL-6 levels showed a significant correlation with APACHE II scores (p = 0.023) and peak MOF scores (p = 0.019); no other humoral mediators were correlated with either APACHE II scores or peak MOF scores. In the septic shock group, plasma levels of TNF- (p = 0.015), IL-6 (p = 0.009), IL-8 (p = 0.023), TM (p = 0.011), and AM (p = 0.0003) showed significant correlations with APACHE II scores (Figure 2A). Moreover, plasma levels of TNF- (p = 0.0007), IL-8 (p = 0.026), PAI-1 (p = 0.001), and AM (p = 0.0002) were also correlated with peak MOF scores (Figure 2B).

View larger version (12K): [in this window] [in a new window]   View larger version (12K): [in this window] [in a new window]   Figure 2.   (A) Correlation between APACHE II scores and plasma levels of AM in patients with septic shock. (B) Correlation between MOF scores and plasma levels of AM in patients with septic shock.

Comparison of Plasma Levels of AM and Humoral Mediators in Survivors and Nonsurvivors

In the traumatic shock group, only plasma IL-6 levels in the nonsurvivors were significantly higher than those in the survivors (p = 0.011). In the septic shock group, levels of IL-6 (p = 0.009), IL-8 (p = 0.006), and AM (p = 0.022) in the nonsurvivors were also significantly higher than those in the survivors. However, considerable variation in levels of these mediators were observed for individual patients. For example, in the septic shock group, the highest plasma AM level in the survivors was 786.4 fmol/ml, which was greater than the highest plasma AM levels observed in nonsurvivors.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study demonstrated that circulating AM levels were markedly increased in patients with various forms of SIRS, and that the increase was proportional to the severity of the patients' diseases.

Since the discovery of AM in 1993, it has attracted the interest of researchers in the cardiovascular field because of its powerful and characteristically long-lasting vasodepressor activity. The mean plasma AM level in patients with heart failure was found to be about four times higher than that of controls, being higher in patients in Class III or IV of the New York Heart Association functional classification than in those in Class I or II (2). However, little is known about the pathophysiologic role of AM in other diseases (e.g., inflammation).

We have recently demonstrated that remarkably increased plasma AM levels correlate with relaxation of vascular tone in patients with septic shock (5). These observations led us to speculate that AM may play a role in the pathophysiology of "inflammation," in addition to its role in the regulation of cardiovascular functions.

To estimate the contribution of AM to inflammation, we measured plasma AM levels in patients with various types of SIRS. In each group of SIRS patients, plasma AM levels were increased by from 2.8- to 39-fold as compared with the control value. Cockcroft and colleagues (8) reported that the lowest dose of AM inducing significant arteriolar dilatation was comparable with plasma levels found in patients with heart failure. The AM levels in SIRS are closely similar to or higher than those found in heart failure; furthermore, since AM derived from vascular cells acts in autocrine/paracrine manner, the AM level in local sites is thought to be much higher than the plasma AM level, suggesting that plasma AM in SIRS patients may affect vascular tone directly. In addition, plasma AM levels in patients with traumatic shock and septic shock were significantly increased over those in patients with trauma and severe sepsis, respectively, suggesting that plasma AM levels are increased in proportion to the severity of SIRS. These observations indicated that AM may play a role in the pathophysiology of SIRS. Moreover, we have recently demonstrated that AM is produced and secreted from peripheral blood granulocytes, lymphocytes, monocytes, monocyte-derived macrophages, and fibroblasts, all of which are involved in the inflammatory process (9, 10). These findings, taken together with our earlier study demonstrating that AM stimulates IL-6 production in Swiss mouse 3T3 cells (11), makes it possible to say that AM, like cytokines, participates actively in the pathophysiology of inflammation.

The mechanism by which plasma AM levels were increased in patients with SIRS remains unknown. Since serum creatinine levels showed a positive correlation with plasma AM levels in the trauma and severe sepsis groups, decreased clearance of AM from the kidney may have contributed to the increased AM levels in these patient groups. However, no such correlations were observed in the traumatic shock or septic shock groups. Moreover, since the mean levels of plasma AM in the traumatic shock, severe sepsis, and septic shock groups were much higher than those in patients with end-stage renal failure (2), reduced renal clearance may not be necessarily responsible for the increased plasma AM levels in these patients, especially in the shock state. On the basis of these findings, taken together with the fact that AM is produced in and secreted from inflammatory cells (9) in addition to vascular cells (2), we may assume that production and secretion of AM were enhanced in SIRS.

In order to identify the factors regulating the plasma levels of AM in SIRS, we compared plasma levels of AM with those of proinflammatory cytokines in the traumatic shock and septic shock groups. We found a significant correlation between plasma levels of AM and TNF- in the septic shock group. It is possible that the increased levels of plasma TNF- enhanced AM production in vascular cells. Furthermore, to investigate the other factors influencing the plasma AM level, we compared plasma AM levels with plasma PAI-I and TM levels, which are used as markers of endothelial injury. We found no correlations between plasma AM level and degree of endothelial injury in the traumatic shock or septic shock groups. Further studies are required to elucidate the mechanisms by which plasma AM levels increase in SIRS.

The mean levels of AM in the septic shock group were significantly higher than those in the traumatic shock group. Martin and coworkers (12) showed that plasma levels of TNF- and IL-6 in patients with septic shock were increased to a higher level than those in traumatic shock, suggesting that the degree of inflammatory response in septic shock is greater than that in traumatic shock. The difference in the plasma AM level in infection and trauma may reflect the difference in the degree of inflammatory response.

Many studies have shown that plasma levels of humoral mediators increase according to the severity of illness in cases of trauma and sepsis, but most of those are mediators that are reported to take part in the inflammatory process. Such a potent vasoactive substance as AM is not reported to be as remarkably increased in disease (2) as the plasma AM levels in our study. We therefore investigated the relationship of plasma levels of AM and of other humoral mediators to two types of severity score, the APACHE II score and the MOF score, in the traumatic shock and septic shock groups in our study. In the traumatic shock group, only plasma levels of IL-6 correlated with APACHE II scores and MOF scores, suggesting that plasma levels of IL-6 may be a marker of severity and be related to mortality in traumatic shock. In the septic shock group, APACHE II scores and MOF scores correlated with plasma levels of AM, TNF-, and IL-8. These results suggest that plasma levels of AM, in addition to those of TNF- and IL-8, may be a good marker for evaluating severity, and may be an early predictor of subsequent MOF in septic shock. Furthermore, we investigated the differences between survivors and nonsurvivors in plasma levels of AM and humoral mediators. Plasma IL-6 levels differed significantly between survivors and nonsurvivors in the traumatic shock group, suggesting that IL-6 is a useful prognostic index in traumatic shock. Plasma levels of AM, IL-6, and IL-8 differentiated survivors from nonsurvivors in our septic shock group. These findings suggest that high plasma AM levels in septic shock patients may predict poor outcome. With an AM cutoff value of 150 fmol/ml for prognosis in patients with severe sepsis and septic shock, the sensitivity of the test was 65% and the specificity was 92%.

In conclusion, we found that plasma AM levels increased in patients with SIRS in proportion to its severity, suggesting that AM, an endothelium-derived, potent vasodilator, may play an important role in the pathophysiology of inflammation. These findings, taken together with our previous finding that plasma AM levels correlate with the relaxation of vascular tone in patients with septic shock (5), AM, especially in septic shock, appears to be not only a marker for evaluating disease severity, but also an early predictor correlating with subsequent organ dysfunction and outcome. Although further studies are needed to clarify the contribution of AM to inflammation, the regulation and modulation of function of AM may provide a new therapeutical strategy for patients with SIRS, including those with septic shock.