| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Studies |
Departments of Internal Medicine (A.B.) and Clinical Chemistry (I.V., C.H.), Medical Spectrum Twente, Hospital Group, 7500 KA Enschede, The Netherlands; and Medical Intensive Care Unit, Free University Hospital (A.B., L.G.T.), 1081 HV Amsterdam, The Netherlands
Address all correspondence and requests for reprints to: I. Vermes, M.D., Ph.D., Department of Clinical Chemistry, Medical Spectrum Twente, Hospital Group, P.O. Box 50.000, 7500 KA Enschede, The Netherlands. E-mail: i.vermes{at}wxs.nl
Abstract
In patients with septic shock (n = 32), multitrauma (n = 8),
and hospitalized matched controls (n = 41), we serially measured
serum macrophage inhibitory factor (MIF), cortisol, plasma ACTH, tumor
necrosis factor-
, and interleukin-6 (IL-6) immunoreactivity during
14 days or until discharge/death. MIF levels were significantly
elevated on day 1 in septic shock (14.3 ± 4.5 µg/L), as opposed
to trauma (3.1 ± 1.7 µg/L) and control patients (2.5 ±
2.1 µg/L). The time course of MIF, parallel to cortisol, but in
contrast to ACTH, showed persistently elevated levels in septic
patients. On admission, nonsurvivors of septic shock (n = 11)
showed significantly higher MIF levels than survivors (18.4 ± 4.8
and 10.2 ± 4.2 µg/L, respectively). Patients with septic adult
respiratory distress syndrome (ARDS; n = 8) showed higher MIF
levels than those who did not develop ARDS (19.4 ± 4.7
vs. 9.2 ± 4.3 µg/L, respectively). Multiple
logistic regression analysis demonstrated that both MIF and ARDS were
independent predictors of adverse outcome. On admission, tumor necrosis
factor-, IL-6, procalcitonin, and lipopolysaccharide-binding protein
levels were higher in patients with septic shock than in patients with
multitrauma. In septic patients, regression analysis showed significant
correlations between MIF and cortisol as well as between MIF and IL-6
levels and disease severity scores. No relation was found between MIF
and markers of the acute phase response (procalcitonin, C- reactive
protein, and lipopolysaccharide-binding protein). In multitrauma
patients, MIF levels were not elevated at any time point and were not
related to other variables.
Our data suggest that during immune-mediated inflammation (such as septic shock) MIF is an important neuroendocrine mediator: a contraregulator of the immunosuppressive effects of glucocorticoids.
MACROPHAGE MIGRATION inhibitory factor
(MIF) is a multifunctional mediator with pituitary hormone,
macrophage-derived cytokine, and catalytic enzyme activities, which
appear to be of pivotal importance to the regulation of the host immune
and inflammatory response (1, 2). Based on animal studies,
it has been shown that MIF displays a proinflammatory spectrum of
actions that is most strikingly represented by its capacity to override
the antiinflammatory and immunosuppressive actions of glucocorticoids
(GC) (3, 4, 5, 6). MIF expression is biphasically regulated by
GC, following a bell-shaped dose-response curve. At low physiological
GC concentrations, MIF synthesis and release are induced, in contrast
to any other cytokine, whereas at high GC concentrations its overriding
capacity is diminished (5, 7). There appears to be a dual
role of MIF in optimizing inflammatory activity: being directly
proinflammatory and indirectly inhibiting maximal antiinflammatory GC
activity (1, 8). MIF release, the main source of which is
the pituitary or monocytes/macrophages, is stimulated by inflammatory
stimuli (microbial products and toxins), cytokines [tumor necrosis
factor- (TNF) and interferon-
], and stress-induced
activation of the hypothalamus-pituitary-adrenal (HPA) axis (4, 9, 10, 11, 12, 13).
During severe stress conditions, such as sepsis or trauma, activation of the HPA axis and hypercortisolism are present, whereas a discordant low ACTH level occurs in the chronic phase of critical illness (14). The antiinflammatory and immunosuppressive properties of GC are often beneficial in protecting the organism to mute its own inflammatory cascade (15), but to ensure host homeostasis the immunosuppressive effects of the HPA axis need to be counterregulated. Surprisingly, until recently no such system or no antagonizing mediator for GC was known. The rediscovery of MIF as a pituitary hormone/cytokine led to the awareness about MIF being a contraregulator to endogenous steroids within the host defense system (5, 16, 17).
Few data about the pathophysiological role of MIF in humans exist (17), although in animal models an important role is suggested in a variety of pathological conditions (18, 19, 20, 21, 22, 23, 24, 25), such as endotoxic shock and the adult respiratory distress syndrome (ARDS) (4, 7, 9, 13, 26, 27). When coinjected into mice with lipopolysaccharide (LPS), MIF potentiates LPS lethality (9), whereas anti-MIF antibody offers full protection against lethal endotoxemia (4, 7, 9) and lethal bacterial peritonitis (27).
We investigated the intriguing role of MIF in the complex network of neuroendocrine adaptation during severe critical illness, such as severe trauma and septic shock, by measuring the time course of serum MIF levels in its relation to the HPA axis and to other cytokines, the severity of the disease, the occurrence of ARDS, and the final clinical outcome [intensive care unit (ICU) mortality].
Subjects and Methods
Study population
The study was performed in a total of 81 subjects. Approval for the study was obtained from our institutional human subjects research committee, and written informed consent from first degree relatives was mandatory. Forty consecutive patients admitted to the intensive care unit with septic shock or severe multiple trauma were included in the study within 6 h of admission.
Thirty-two patients had septic shock (12 females; mean age, 64 yr) as defined by clinical evidence of infection, temperature above 38.5 C or below 35.6 C, tachycardia (>90/min), tachypnea (>20/min), or necessity of mechanical ventilation. Shock was defined as a fall in systolic arterial blood pressure below 90 mm Hg, the need for vasopressors, together with signs of inadequate tissue perfusion (oliguria, mental alterations, lactic acidosis, coagulation abnormalities) (28). Eight patients (one female; mean age, 51 yr) were victims of multiple trauma of mechanical origin (injury severity score, >20) (29).
Exclusion criteria were age below 18 yr, use of corticosteroids or other drugs affecting the HPA axis, unexplained hypo- or hyperkalemia, preexisting adrenal insufficiency or known abnormalities of the HPA axis, and multitrauma with head injury.
In both patient groups the severity of disease was scored according to APACHE II scores (30), multiorgan failure [sepsis-related organ failure assessment (SOFA) scores] (31), injury severity scores (29), the presence of ARDS, bacteriological findings, and clinical outcome (ICU mortality).
ARDS was defined as a condition involving impaired oxygenation
(PaO2/FiO2, 
200)
regardless of the PEEP level, the detection of bilateral pulmonary
infiltrates on the frontal chest radiograph, and a pulmonary capillary
wedge pressure of 18 mm Hg or less or no clinical evidence of elevated
left atrial pressure (32).
Control patients (18 females; mean age, 62 yr) were patients without acute medical illness admitted to the medical department for routine diagnosis and treatment. Written informed consent was obtained from all control patients. Patients with infectious diseases, autoimmune diseases, cancer, renal disease, and HPA abnormalities were excluded from this group.
Blood samples
Blood samples were drawn daily from arterial lines at 0600 h. Follow-up was performed for 14 days or until discharge from ICU or death. Blood was collected into plain tubes or into prechilled tubes containing ethylenediamine tetraacetate. Specimens were immediately centrifuged at 4 C and stored at -25 C. Blood samples of the control subjects were collected by venipuncture between 0800–1000 h in the supine position after 30 min of rest. All hormone and cytokine measurements were performed in duplicate. Simultaneously, routine biochemistry (Roche, Basel, Switzerland) and hematology (HST 430, Sysmex Corp., Kobe, Japan) parameters were assessed.
Assays
Immunoreactive cortisol, ACTH, tumor necrosis factor-
(TNF), and interleukin-6 (IL-6) concentrations were measured with
commercially available chemiluminescent enzyme immunoassays using the
Immulite Automated Immunoassay System (Diagnostic Products, Los Angeles, CA). The cortisol assay is a solid phase
chemiluminescent enzyme immunoassay that has an analytical sensitivity
of 5 nmol/L and shows intra- and interassay coefficients of variation
(CVs) below 7% measured at three levels.
Immunoreactive ACTH, TNF, IL-6, and LPS-binding protein (LBP)
concentrations were measured with solid phase two-site chemiluminescent
immunometric assays. The ACTH assay showed less than 0.001%
cross-reactivity with ACTH-(1–18), ACTH-(1–24), ACTH-(22–39), and
MSH. The results were expressed as picomoles per L using
ACTH-(1–39) as a standard. The assay has an analytical sensitivity of
0.2 pmol/L. The intra- and interassay CVs measured at three levels
ranged from 3.1–8.2% and from 5.1–9.7%, respectively. The TNF
assay showed no significant cross-reactivity to IL-2, -6, -8, or -10.
The detection limit of the TNF assay was 2 ng/L, and the intra- and
interassay CVs were less than 3% and less than 6%, respectively. The
IL-6 assay showed nondetectable cross-reactivity to other ILs and
0.05% to TNF. The detection limit is 5 ng/L, and the intra- and
interassay CVs measured at three levels were less than 6% and less
than 10%, respectively. The LBP assay showed nondetectable
cross-reactivity to IL-6, IL-8, TNF, and C-reactive protein (CRP).
The analytical sensitivity was 0.2 mg/L, the interassay CVs measured at
three levels were less than 10%, and the calibration range was up to
200 mg/L. Procalcitonin (PCT) concentrations were measured by an
immunoluminometric assay using two antigen-specific monoclonal
antibodies (LUMItest PCT, Brahms Diagnostica, Berlin, Germany). The
minimal detected concentration with a maximal interassay CV was 0.3
µg/L. Immunoturbidimetric determination of CRP was made by an
automated analyzer (Roche).
MIF serum levels were measured using a novel time-resolved fluorometry-based detection method (Delfia, Perkin-Elmer-Wallac, Inc., Gaithersburg, MD) (33). The results are expressed as micrograms per L using recombinant MIF as a standard (R\|[amp ]\|D Systems, Inc., Minneapolis, MN). The detection limit of the assay was 30 ng/L. The inter- and intraassay CVs varied from 1.2–3.1% and from 0.9–6.4%, respectively.
Statistical analysis
Values are expressed as the mean ± SD. All
statistical analyses were performed using a statistical software
package (version 9.0.1, SPSS, Inc., Chicago, IL).
Qualitative values were analyzed using the 
2
test. For data that were not normally distributed, the Mann-Whitney U
test was used if only two groups were compared, and the Kruskal-Wallis
one-way ANOVA test was used if more than two groups were compared.
Serial data were analyzed using Friedman’s repeated measures ANOVA on
ranks, followed by Dunn’s test for specific comparisons. The Spearman
rank order correlation coefficient (rs) was used
to estimate the relation between MIF and the other variables. We used
two-factor ANOVA to determine whether survivors and nonsurvivors as
well as patients with ARDS and non-ARDS differed with respect to the
MIF level and to calculate the interference between the presence or
absence of survival and ARDS. The best cut-off for MIF was chosen using
Youden’s index, with calculation of sensitivity, specificity, positive
predictive value (PPV), and negative predictive value (NPV). Then
multiple logistic regression was used to test the predictive value of
altered MIF concentrations and ARDS for outcome in patients with septic
shock by calculating the odds ratios (OR) and its confidence limits
(CI). For all statistical analysis, P < 0.05 was
considered significant.
Results
Clinical and biochemical characteristics on admission of both
patient and control groups are summarized in Table 1
. Age, mortality, length of stay on ICU,
and SOFA and APACHE II scores were higher (P < 0.05)
in septic shock patients compared with multitrauma patients.
|
-3). ACTH levels
displayed the characteristic biphasic pattern in both patient groups,
with paradoxically low concentrations from day 5 (Figs. 2 and 3).
In septic patients, we found on each individual day a strong positive
correlation between MIF and cortisol levels (day 1:
rs = 0.72; P < 0.01; pooled
data: rs = 0.77; P < 0.01; Fig. 4). Also, on admission, a significant
correlation was observed between MIF and IL-6 concentrations
(rs = 0.40; P = 0.042), APACHE
scores (rs = 0.70; P < 0.01),
and SOFA scores (rs = 0.57; P <
0.05). In both patient groups, no correlations in time were found
between MIF and other variables (ACTH, TNF, CRP, PCT, and LBP).
|
|
|
|
. Nonsurvivors showed higher MIF levels
than survivors of septic shock (18.4 ± 4.8 and 10.2 ± 4.2
µg/L, respectively; P < 0.01). Patients with septic
ARDS also had higher MIF levels than non-ARDS patients (19.4 ±
4.7 and 9.2 ± 4.2 µg/L, respectively). However, two-factor
ANOVA showed a significant difference between MIF levels in survivors
and nonsurvivors (F = 13.97; df = 1,28; P =
0.001), but no significant difference regarding ARDS (F = 2.67;
df = 1,28; P = 0.115), and no significant
interaction between survival or the presence of ARDS was observed
(F = 0.65, df = 1,28; P = 0.43). Furthermore,
we calculated the specificity, sensitivity, Youden’s index, and
predictive values for MIF levels on day 1 to determine a cut-off point
that allows discrimination of patients with septic shock at risk for
adverse outcome. A threshold of 14 µg/L showed a sensitivity of 82%,
a specificity of 86%, a positive predictive value of 75%, and a
negative predictive value of 90%. Multiple logistic regression
analysis comparing increased MIF levels and ARDS in terms of relative
risk for adverse outcome demonstrated that MIF was a better predictor
of ICU mortality (MIF, >14 µg/L: OR, 27; 95% CI, 3.8–191;
P = 0.001 vs. ARDS: OR, 11.4; 95% CI,
1.74–76.9; P = 0.01).
|
Discussion
HPA function plays a crucial role in metabolic and immunological homeostasis during critical illness (34). We (14) and others (35) showed that the HPA axis displays a biphasic pattern during the course of critical illness. The present data confirm high cortisol levels in these severely stressed patients together with the biphasic pattern of ACTH concentrations, showing paradoxically low concentrations in the chronic phase, indicating that cortisol release in this phase may be driven through an alternative (non-ACTH) pathway (14). Serum MIF, however, displayed significantly increased concentrations in critically ill septic patients. This release of MIF into the circulation indicates that the host also has the capacity to antagonize the systemic antiinflammatory properties of GC. However, it has been observed that MIF release follows a bell-shaped dose-response curve with respect to microbial toxins, and that its overriding capacity is diminished at high GC concentrations suggests the existence of counterregulatory mechanisms within the MIF/glucocorticoid system (1, 11).
We present the first data in humans directed upon the temporal course
of MIF concentrations in severely ill patients suffering from
multitrauma or septic shock. Patients with septic shock demonstrated
increased MIF levels during acute and prolonged critical illness
parallel to and strongly correlated with serum cortisol concentrations.
Nonsurvivors, patients with positive blood cultures and ARDS, and the
most severely ill patients showed the highest MIF (and cortisol)
levels, suggesting that MIF might be an indicator of the severity of
the septic shock. In animal experiments MIF acts to override
GC-mediated inhibition of cytokine production (7). On the
other hand, macrophage MIF is released upon stimulation with TNF,
and at high levels (>100 µg/L) MIF induces TNF secretion by
macrophages (27), augmenting each other in a
proinflammatory loop (7). Our results showed high MIF and
cytokine levels in the initial phase of septic shock, but a
dissociation in the more chronic phase, when MIF levels were
persistently high, whereas cytokine levels significantly decreased. We
only observed a weak relation between serum IL-6 and MIF
(P = 0.042) on admission, but no relation between
TNF and MIF. IL-6 is a known activator of the HPA axis
(36), but IL-6 is not known, in contrast to TNF or
interferon-, to induce MIF production (10).
Apparently both severe trauma and septic shock are conditions
leading to activation of the HPA axis with a paradoxically low ACTH
level during prolonged illness. However, the changes in MIF
concentrations were distinctive; the persistently elevated levels of
MIF in septic shock contrasted with the normal levels in multiple
trauma. This fact suggests that the trigger of MIF release is not the
stress-induced HPA activation, and the source of the high MIF
concentration found in septic patients is not the pituitary. The
bidirectional communication between the immune and neuroendocrine
systems during the stress of inflammation could be the explanation for
this discrepancy (37, 38). Possibly, endo- or exotoxins
are more potent inducers of MIF release from the macrophages (4, 27) than the pituitary MIF release induced by hypothalamic
activation during trauma. In fact, on admission we found significantly
higher concentrations of IL-6, TNF, procalcitonin, and LBP in
patients with septic shock than in patients with multiple trauma. Only
IL-6 had a statistically significant relation, although weak, to serum
MIF. No relation was present between MIF and procalcitonin, a promising
marker and possible neuroendocrine mediator of sepsis (39, 40). We also examined the relation between MIF and LBP, a novel
acute phase protein with possible protective effects against endotoxins
(41, 42, 43). No relation was found between LPB and MIF levels
on admission. Interestingly, LBP was strongly related to PCT, also an
acute phase protein (data not shown). One can speculate that one of
these substances could be responsible for the inflammation-induced MIF
release in patients with septic shock.
Although MIF may function as a counterregulatory hormone for GC on
inflammation and immune function in conditions of severe illness, its
regulation must be different in view of the differences found in MIF
levels between patients with septic shock and multiple trauma. In
addition to its directly acting, proinflammatory functions, the
secretion of MIF during immune-mediated inflammation suggests that
there is a direct neurohumoral cytokine response to infection and
tissue invasion. Its localization in central (pituitary) (9, 12) and peripheral (immune cells) (10, 44) sites is
consistent with its pivotal physiological role within immune and
endocrine defense systems (1, 17). Pituitary-derived MIF
may serve to prime systemic immune responses once a localized
inflammatory site fails to contain an invasive agent. Another
explanation is that MIF acts as a central nervous system-derived stress
signal to activate the immune system in anticipation of an impending
invasive stimulus (9, 17). Accordingly, MIF may act in
concert with ACTH and the adrenocortical axis to modulate systemic
inflammatory responses (12). There is a sufficient
molecular biological basis to suggest such an interplay between MIF and
GC at the receptor/intracellular level (45, 46, 47, 48). The
antagonistic role of MIF regarding GC might be linked to its
interaction with Jab1, by affecting I
B or by modulation of
Jab1-steroid interactions (45).
In conclusion, we found markedly elevated serum MIF levels (5-fold) in septic shock compared with trauma and control patients. The prolonged stimulation of MIF release appears to be characteristic of septic shock, which supports the hypothesis that MIF might be a neuroendocrine modulator of systemic inflammation. In addition, nonsurvivors and more severely ill septic patients showed higher MIF levels at most time points. Also, we found markedly increased serum MIF levels in ARDS up to 14 days, which, in accordance with previously reported high local MIF levels in the lungs (49), suggests that the balance between MIF (proinflammatory) and cortisol (antiinflammatory) plays a role in the intensity of the inflammatory reaction. This might have important implications for GC treatment of sepsis or ARDS (8, 49, 50, 51). Both increased MIF levels and the presence of ARDS were independent predictors of adverse outcome.
GC are essential for survival during critical illness partly due to their potent antiinflammatory effects. However, GC also cause immunosuppression through GC receptor-mediated mechanisms. Prolonged, uncompensated elevation of GC receptor signaling would be maladaptive with regard to the immune system (46). GC might exert dual feedback effects at the level of the pituitary gland, inhibiting ACTH secretion while stimulating MIF secretion. High levels of MIF may be a crucial factor limiting the immunosuppressive effects of even high dose GC treatment for septic shock or late ARDS (52, 53). Anti-MIF therapy in combination with low dose GC may be considered an important treatment option for sepsis. In relative adrenal insufficiency, for which we are still searching for reliable diagnostic tests (54, 55), MIF might be the crucial factor in determining which patients will respond to GC therapy.
Received October 3, 2000.
Revised January 10, 2001.
Revised February 26, 2001.
Accepted March 2, 2001.
References
, and IFN- administered orally via attenuated
Salmonella typhimurium. J Immunol. 160:1285–1289.B. Am J Physiol Regul Integr Comp Physiol.
279:R1043–R1049.
This article has been cited by other articles:
 |
|
 |
|
  T. Deak Immune Cells and Cytokine Circuits: Toward a Working Model for Understanding Direct Immune-to-Adrenal Communication Pathways Endocrinology, April 1, 2008; 149(4): 1433 - 1435. [Full Text] [PDF]
|
|
|
|
|
|
L. Engstrom, K. Rosen, A. Angel, A. Fyrberg, L. Mackerlova, J. P. Konsman, D. Engblom, and A. Blomqvist Systemic Immune Challenge Activates an Intrinsically Regulated Local Inflammatory Circuit in the Adrenal Gland Endocrinology, April 1, 2008; 149(4): 1436 - 1450. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. V. Crichlow, K. F. Cheng, D. Dabideen, M. Ochani, B. Aljabari, V. A. Pavlov, E. J. Miller, E. Lolis, and Y. Al-Abed Alternative Chemical Modifications Reverse the Binding Orientation of a Pharmacophore Scaffold in the Active Site of Macrophage Migration Inhibitory Factor J. Biol. Chem., August 10, 2007; 282(32): 23089 - 23095. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Flaster, J. Bernhagen, T. Calandra, and R. Bucala The Macrophage Migration Inhibitory Factor-Glucocorticoid Dyad: Regulation of Inflammation and Immunity Mol. Endocrinol., June 1, 2007; 21(6): 1267 - 1280. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kwaijtaal, A. J. van der Ven, R. van Diest, C. A. Bruggeman, F. W. H. M. Bar, T. Calandra, A. Appels, and F. C. G. J. Sweep Exhaustion is Associated With Low Macrophage Migration Inhibitory Factor Expression in Patients With Coronary Artery Disease Psychosom Med, January 1, 2007; 69(1): 68 - 73. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. MASTORAKOS and I. ILIAS Interleukin-6: A Cytokine and/or a Major Modulator of the Response to Somatic Stress Ann. N.Y. Acad. Sci., November 1, 2006; 1088(1): 373 - 381. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. M. Arafah Hypothalamic Pituitary Adrenal Function during Critical Illness: Limitations of Current Assessment Methods J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 3725 - 3745. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. McDevitt, J. Xie, G. Shanmugasundaram, J. Griffith, A. Liu, C. McDonald, P. Thuma, V. R. Gordeuk, C. N. Metz, R. Mitchell, et al. A critical role for the host mediator macrophage migration inhibitory factor in the pathogenesis of malarial anemia J. Exp. Med., May 15, 2006; 203(5): 1185 - 1196. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. N. Silverman, A. H. Miller, C. A. Biron, and B. D. Pearce Characterization of an Interleukin-6- and Adrenocorticotropin-Dependent, Immune-to-Adrenal Pathway during Viral Infection Endocrinology, August 1, 2004; 145(8): 3580 - 3589. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. A. Clark, L. M. Alleva, A. C. Mills, and W. B. Cowden Pathogenesis of Malaria and Clinically Similar Conditions Clin. Microbiol. Rev., July 1, 2004; 17(3): 509 - 539. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Shimizu, J. Nishihira, H. Watanabe, R. Abe, A. Honda, T. Ishibashi, and H. Shimizu Macrophage Migration Inhibitory Factor Is Induced by Thrombin and Factor Xa in Endothelial Cells J. Biol. Chem., April 2, 2004; 279(14): 13729 - 13737. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Roger, C. Froidevaux, C. Martin, and T. Calandra Macrophage migration inhibitory factor (MIF) regulates host responses to endotoxin through modulation of Toll-like receptor 4 (TLR4) Innate Immunity, April 1, 2003; 9(2): 119 - 123. [Abstract] [PDF]
|
|
|
|
|
|
A. Beishuizen and L. G. Thijs Review: Endotoxin and the hypothalamo-pituitary-adrenal (HPA) axis Innate Immunity, February 1, 2003; 9(1): 3 - 24. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
), multitrauma (•),
and hospitalized matched controls (
).
) concentrations in patients with septic shock.
, Survivors (n = 21);