in Septic Patients After Trauma - NCBI

ANNALS OF SURGERY Vol. 229, No. 2, 246-254 C 1999 Lippinc Williams & Wilins, Inc.

HLA-DR Expression and Soluble HLA-DR Levels in Septic Patients After Trauma Markus Ditschkowski,* Ernst Kreuzfelder,* Vera Rebmann,* Stanislav Ferencik, PhD,* Mathias Majetschak, MD,t Ernst N. Schmid, PhD,t Udo Obertacke, MD,t Herbert Hirche,§ Ulrich F. Schade,ll and Hans Grosse-Wilde*

From the *Institute of Immunology, tDepartment of Trauma Surgery, tinstitute of Medical Microbiology, §Institute of Medical Informatics, Biometry and Epidemiology, and the IlClinical Research Group on Shock and Multiorgan Failure, University Hospital of Essen, Essen, Germany

Objective

Results

To determine if cellular and soluble HLA-DR molecules may be relevant in severely injured patients for the development of gram-positive or gram-negative sepsis.

HLA-DR expression on circulating T cells, calculated as mean fluorescence intensity in channels, was reduced at day 1 after admission in 20 patients with subsequent severe sepsis compared with 46 patients without sepsis. The septic patients immediately after trauma had significantly lower soluble HLA-DR plasma levels than the nonseptic patients. At day 2 after admission, HLA-DR expression on monocytes was significantly lower in the severe sepsis group than in the patients without sepsis, and lasted until day 14 after injury.

Summary Background Data HLA-DR molecules play a central role in the specific immune response to infection. The reduced HLA-DR expression on monocytes is considered to correlate with infectious complications and the development of sepsis. Data on the role of HLA-DR expression on T cells and soluble HLA-DR molecules are rare.

Methods HLA-DR expression on monocytes and T cells was measured by flow cytometry. Plasma levels of soluble HLA-DR were studied by enzyme-linked immunosorbent assay.

Despite new generations of antibiotics and great improvement in the fields of rescue and modem intensive care medicine, sepsis remains one of the most frequent causes of complications and death in severely injured patients.1-3 Besides shock, initial bacterial colonization, and application of invasive measures for diagnostic and therapeutic purposes, immune mechanisms are reported to be responsible Supported by the Deutsche Forschungsgemeinschaft (Gr6O8n-l, Schm 74/13-1) and Fonds der Chemischen Industrie (FUS). Part of this study is the fulfillment of M. Ditschkowski to receive a doctorate degree through the Medical Faculty, University of Essen. Correspondence: Prof. Dr. H. Grosse-Wilde, Institut fur Immunologie, Universitatsklinikum Essen, Virchowstr.171, D-45122 Essen, Germany.

Accepted for publication August 31, 1998.

246

Conclusions In severely injured patients, decreased levels of cellular and soluble HLA-DR appear as early indicators of an immune deviation associated with the development of severe sepsis. Moreover, immune alterations of different cell types may promote distinct kinds of septicemia.

for the increased susceptibility of patients to sepsis after multiple trauma.47 Variations have been described in the function of polymorphonuclear leukocytes,8 lymphocytes,9'10 macrophages/monocytes, 11,12 and humoral immune parameters.'3 To determine the impact of trauma and predict clinical outcome, different scoring systems have been developed,14-16 but they failed to predict the development of sepsis. Hershman et al17 formulated a predictive outcome score for sepsis and death after injury based on Injury Severity Score (ISS) expressed as the percentage of LD50, degree of bacterial colonization, and HLA-DR expression on monocytes. This classification appeared to discriminate patients surviving major injury without sepsis from patients who become septic after trauma.

HLA-DR molecules are required for antigen presentation and activation of helper T lymphocytes and, therefore, play a central role in the specific immune response to infection. 18 These molecules are expressed on the surface of professional antigen-presenting cells such as macrophages, dendritic cells, and B cells. Several studies have shown that HLA-DR expression on monocytes decreases after major injury,'9 surgery,2021 and organ transplantation.22'23 In most cases, the reduced HLA-DR expression on monocytes was considered to correlate with infectious complications and the development of sepsis. Besides membrane-bound HLA-DR, soluble forms exist in the peripheral blood.24 Recent studies have revealed that increased levels of soluble HLA-DR are correlated with transplantation-related complications and autoimmune diseases.24'25 For the latter, bacterial and in part virus-derived superantigens have been proposed as causative agents by inducing polyclonal T-cell activation26'27; the same holds true for the pathogenic mechanism of sepsis.28 In this context, it is important that soluble HLA-DR molecules can function as ligands for superantigens29 and, thus, may play a role in the "detoxification" of superantigens. To assess whether HLA-DR molecules on circulating lymphocytes and in plasma predict the development of sepsis after injury, a prospective study was performed in 77 consecutive trauma patients. We monitored HLA-DR surface expression on monocytes and T cells as well as soluble HLA-DR plasma levels in the patients immediately after trauma and serially during the subsequent 14 days.

PATIENTS AND METHODS Patients The study group consisted of 77 consecutive blunt trauma patients admitted immediately after injury to the Department of Trauma Surgery, University Hospital of Essen. Criteria for the exclusion of patients were age (younger than 18 or older than 65 years), admission >8 hour after injury, penetrating injuries, premedication with immunosuppressive agents, and underlying cardiac, pulmonary, hematologic, or immunologic diseases. With regard to the severity of injury and clinical outcome, the patients were categorized into three groups: minor injury (ISS < 16) and uneventful recovery (group 1, n = 11), severely injured patients (ISS > 16) without sepsis (group 2, n = 46), and severely injured patients in whom severe sepsis developed (group 3, n = 20). For the diagnosis of severe sepsis, the criteria of the American College of Chest Physicians consensus were used30: severe sepsis was defined as sepsis associated with organ dysfunction, hypoperfusion, or hypotension. Analysis of peripheral blood mononuclear cells for all patients was performed without knowledge of the clinical status. The clinical and demographic profile of the patients is shown in Table 1. There were no statistical differences

247

HLA Molecules and Sepsis After Injury

Vol. 229 * No. 2

Table 1. CLINICAL AND DEMOGRAPHIC PROFILE OF TRAUMA PATIENTS Group 1 (ISS* < 16)

Group 2 (ISS > 16)

Group 3 (ISS > 16)

11 9 4-15 38 23-52 2.7 0

46 25 17-41 36 18-60 2.3 2

20 33t 17-57 44 18-65 2.3 6t

Number of patients ISS (mean) ISS (range) Age in years (mean) Age in years (range) Male:female ratio Deceased patients *

ISS = Injury Severty Score

t p < 0.05 Group 2 vs. Group 3.

concerning age and sex ratio between groups 2 and 3. However, mean ISS and the number of deceased patients were found to be significantly higher in group 3 than in group 2. All patients received cefazoline (2 g intravenously every 8 hours) at the time of surgery. Additional antibiotics were given no earlier than day 4. In cases of severe sepsis, treatment was initiated with piperacillin and tazobactam or ciprofloxacin and imipenem, and changed to appropriate antibiotics after identification of bacteria species. In all patients, a central venous catheter was placed in the internal jugular or subclavian vein. In group 2 and 3 patients, an arterial catheter was also introduced. Routine replacement of catheters was performed on day 10 after trauma. If signs of local infection or a fever of unknown origin occurred, catheters were replaced earlier. In patients with hemodynamic instability despite adequate fluid infusion and blood volume replacement, a triple-lumen, thermistor-tipped 7.5F pulmonary artery catheter (Ohmeda, Murray Hill, NJ) was inserted into the internal jugular or subclavian vein.

Blood Specimens Venous EDTA blood samples were collected immediately after admission and serially at 8 a.m. on days 1, 2, 4, 6, 8, and 14 after injury.

Cell Counting and Flow Cytometry Leukocyte counts were assessed using an electronic cell counter (Sysmex, Hamburg, Germany). Three-part differentiation of whole-blood peripheral blood mononuclear cells was performed by flow cytometry (FACScan, Becton Dickinson, Heidelberg, Germany) using monoclonal antibodies specific for CD3, CD14, CD25, CD45, and HLA-DR (clone L243). With the exception of anti-HLA-DR (Becton Dickinson), all monoclonal antibodies were purchased from

248

Ditschkowski and Others

Ann. Surg. * February 1999 HLA-DR positive

HLA-DR negative

could be defined. An example for the quantification of HLA-DR+ T cells is given in Figure 2.

Quantification of Soluble HLA-DR Plasma levels of soluble HLA-DR were studied using a modified enzyme-linked immunosorbent assay format, as described previously.31 In short, for the capture of HLA-DR molecules, microtiter plates (Costar GmbH, Bodenheim, Germany) were coated overnight at 4°C with monoclonal antibodies from clone L243 (Becton Dickinson) at a final concentration of 0.04 ,jg/ml. After blocking free binding sites, undiluted EDTA plasma samples were incubated for 1.5 hours at 37°C and bound soluble HLA-DR molecules were detected by monoclonal antibody CR3/43 (IgGI) (Dakopatts, Hamburg, Germany) recognizing a monomorphic determinant on HLA class 11 3 chains. Bound monoclonal antibody CR3/43 was detected by an alkaline phosphatase conjugated antimurine IgGI (Serva, Heidelberg, Germany). Affinity purified HLA-DR1 (kindly provided by Dr. S. Jurcevic, Harefield Hospital, UK) was used as a reference to calculate protein concentrations of soluble HLA-DR.

0

0)

0

a

Molecular HLA-DR Typing Typing was done using the polymerase chain reaction technique with sequence-specific primers, as described pre135

Channel intervals

Figure 1. Distribution of HLA-DR- and HLA-DR+ lymphocytes in 19 healthy persons (0), 11 nonseptic trauma patients (0), and 11 septic trauma patients (V) matched in age and sex to the septic patients at day 1 after trauma. The HLA-DR fluorescence intensities of CD3- cells (see lower part of the right dot plot in Fig. 2) were divided into intervals of 10 channels, and the sums of events were plotted against the intervals. Analysis of HLA-DR expression was performed by gating on the lymphocyte subpopulation of interest by virtue of its forward and sideward scattering (FSC, SSC) properties.

Coulter-Immunotech (Hamburg, Germany). Flow cytometric analysis was conducted using a linear format to measure channel fluorescence intensities as numerals and to calculate mean fluorescence intensity values. To define HLA-DR+ cells, lymphocytes were studied from 19 healthy donors, 11 septic patients, and 11 patients in group 2 matched for age, sex, and ISS. The fluorescence intensities obtained were divided into intervals of 10 channels, and the sums of events were plotted against the intervals. As Figure 1 shows, there was a distinct separation between HLA-DR+ and HLADR- cells of all individuals at channel 135 based on a

curve-fitting program (Jandel Scientific TableCurve, Krefeld, Germany). Using this threshold value, the fraction of HLA-DR+ cells and, therefore, HLA-DR+/CD3+ cells

viously.32'33

Bacteremia The Bactec system (Becton Dickinson) was used for blood cultures. For the identification of bacteria, the API system (Biomerieux, Niirtingen, Germany) was used.

Statistics The course of HLA-DR levels in the different patient groups was presented as means ± SEM. Confirmatory statistical analysis was performed by a multifactorial analysis of variances with the factors "patient group" and "time," based on repeated measures within a single patient. Logarithmic transformation was used to attain sufficient normality and homogeneous variances. Significance of single group differences was assessed by multiple testing according to Scheffe, whereas parallelism of the different curves was tested by the interaction term of "group X time" in the model. Comparisons at different times were assessed using exploratory interpretation of Kruskal-Wallis, Wilcoxon's unpaired test, and Fisher's test. Resulting probability values were given without further adjustment for multiple testing. Results were considered significant at p < 0.05 (two-tailed tests).


Vol. 229 * No. 2

12039717.04

0

249

12039717.04

0

U)

CM'

0 o

,

Lo 0 c-m

.-

I-

6 U)

'.

jA,

il

.i ..:

I 1.

-

)cO 0

i....

_

.1.

I

0

U, -

Rl o_

0-.. 0

so

ibo .i0

200

250

FSC-H

12039717.04

C)

II

N .

0

1I 0

lm

1Z .-

50

I

..m

~

ibo

150

&A4

I

1

200

s 250

HLA-DR

Histogram Statistics Tube: HLA-DR/CD3 Gated Events: 130 X Parameter: FL1-H HLA-DR (Log)

Gate: G4 Total Events: 20000

CV Marker Left, Right Events % Gated % Total Mean SD Median All 130 100.00 0.65 167.78 18.34 10.93 168.00 0, 255 130 100.00 0.65 167.78 M1 135, 255 18.34 10.93 168.00 Figure 2. Enumeration of HLA-DR+ T cells. Lymphocytes were gated in the left forward/side scatter (FSC/SSC) dot plot showing leukocytes. The right dot plot shows the fluorescence properties of phycoerythrin (PE)-coupled CD3+ cells (T cells) and of fluorescein isothiocyanate-coupled HLA-DR+ cells within the lymphocyte gate. T cells with a fluorescence intensity > 135 channels were considered HLA-DR + and used for the calculation of the mean fluorescence intensity (MFI), shown in the histogram below.

RESULTS With regard to peripheral blood mononuclear cell counts, significant differences between the three groups were seen mainly at day 8 after trauma (Table 2). Leukocyte levels were significantly elevated in group 3 patients from day 4 until day 14, whereas CD3 + T cell levels were significantly reduced in group 3 versus group 2 at day 14. No significant differences were found for CD2 + or CD25 + cell counts (data not shown). The lowest levels of HLA-DR on T cells were recorded in group 3 patients during the whole observation period (Fig. 3). Here, HLA-DR expression decreased as early as day 1 after injury and remained lower than in the other patient groups. Analysis of variance showed parallel courses of HLA-DR expression on T cells in patients of groups 1 and

2 (not significant), whereas group 3 patients had significantly lower levels. Based on these data, on day 1 a cutoff value of mean fluorescence intensity at channel 161 could be defined to be significantly predictive for the later development of sepsis (sensitivity 67%, specificity 72%). Similarly, the HLA-DR expression on monocytes was studied. As Figure 4 demonstrates, the HLA-DR expression was strongly reduced in the severely traumatized groups. Analysis of variance revealed significant differences between all groups. At day 2 there was a significant difference between groups 2 and 3, lasting until day 14. Using a cutoff value of mean fluorescence intensity at 90 channels, HLA-DR expression on monocytes became informative at day 2 after admission for patients with an ISS > 16 and at risk for sepsis (sensitivity 53%, specificity 76%).

250


Ann. Surg. * February 1999

Table 2. PERIPHERAL BLOOD MONONUCLEAR CELL COUNTS AND MEAN Day 2

Day 1*

Group I Leukocytes (cells/nl) Lymphocytes (%) Lymphocytes (cells/,ul) CD3+ Tcells(%) CD3+ T cells (cells/,Il) CD3+ /HLA-DR+ (%) CD3+ /HLA-DR+ (cells/,l) CD3+ /HLA-DR+ (MFI) Monocytes (%) Monocytes (cells/,l)

8.9 ± 11.2 ± 914 ± 64 ± 590 ± 2.8 ± 13 ± 163 ± 8.2 ± 750 ±

0.9 2 134 3.6 80 0.6 1 1.9 0.8 120

Group 2 8.1 ± 0.4 10.6 ± 0.7 860 ± 70 62 ± 2.3 540 ± 50 2.7 ± 0.3 14 ± 2 163 ± 0.7 7 ± 0.4 600 ± 53

Group 3 8.4 ± 11.4 ± 910 ± 67 ± 630 ± 2.8 ± 15 ± 160t ± 6.9 ± 910 ±

0.9 1.1 100 3.7 80 0.5 2. 1 0.5 100

Group I 9.2 ± 11.8 ± 1063 ± 66 ± 765 ± 2.4 ± 16 ± 162 ± 8.7 ± 790 ±

1 1.5 168 3.8 130 0.3 1.7 1.1 0.9 104

Day 4

Group 2

Group 3

Group I

9.7 ± 0.4 10.5 ± 0.7 990 ± 63 62 ± 2 630 ± 50 2.1 ± 0.3 13 ± 2 160 ± 0.8 6.8 ± 0.4 640 ± 41

10 ± 0.8 9.7 ± 0.8 910 ± 68 66 ± 2.9 600 ± 60 2.2 ± 0.3 13 ± 2 160 ± 1.4 6.9 ± 0.5 710 ± 90

7.4 ± 12 ± 890 ± 67 ± 594 ± 2.1 ± 12 ± 163 ± 8.9 ± 645 ±

0.7 1

98 3.6 70 0.2 1.5 1.8 0.8 66

Group 2 8.8 ± 11.3 ± 960 ± 64 ± 650 ± 2.2 ± 16 ± 161 ± 8.2 ± 710 ±

0.4 0.8 72 1.4 55 0.3 4.5 1 0.5 50

Group 3

10.4t4 ± 790 8.5t4 ± 0.9 910 ± 164 68 ± 2.7 646 ± 130 2.2 ± 0.2 16 ± 6 160 ± 1.6 7.2 ± 1 760 ± 115

Day after admission. t p < 0.05 group 3 vs. group 2. t p < 0.05 group 3 vs. group 1.

*

After trauma, CD14 expression on HLA-DR+ monocytes of group 3 patients was not found to be significantly reduced compared with group 2 (data not shown). Monitoring of soluble HLA-DR concentrations revealed that the severely traumatized patients with sepsis were characterized by low plasma levels on days 0, 1, 2, 4, and 6 (Fig. 5). Analysis of variance showed a parallel course of soluble HLA-DR levels for group 1 and 2 patients without significant differences, whereas group 3 patients had significantly lower levels. Using a threshold level of soluble HLA-DR of 16 and no sepsis, 13 (29.5%) carried the HLA-DRBI *01 allele, whereas this allele was found only once (5%) in group 3 (p < 0.05). In the 20 septic patients, positive blood cultures were obtained for the first time between days 4 and 13 after injury; 14 were gram-positive. Table 3 lists the bacteria, the day of detection, and the indicators under study. Using the above-mentioned mean fluorescence intensity cutoff values, the comparison between patients with gram-positive sepsis and group 2 patients revealed at day 1 significantly reduced HLA-DR expression on T cells in contrast to monocytes at day 2. In contrast, HLA-DR expression on monocytes of gram-negative septic patients at day 2 was significantly reduced compared with group 2 patients. For soluble HLA-DR plasma concentrations, there was at day 0 no significant difference between samples from gram-negative and gram-positive septic patients.

DISCUSSION The most prominent observations of this study are the decrease of HLA-DR expression on circulating T cells and

the initially low soluble HLA-DR plasma levels in severely traumatized patients with subsequent severe sepsis. The reduced HLA-DR expression was obviously not influenced by the severity of trauma, because analysis of variance showed no difference between group 1 (ISS < 16) and group 2 patients (ISS > 16). With regard to expression analysis, our data contrast with those of Wakefield et al,2' who found an increase of HLA-DR on T cells in patients with infectious complications after surgical trauma. This difference may be explained by divergent evaluation procedures of HLA-DR+ T cells. In patients after accidental or surgical trauma followed by infection, sepsis, or death, an established finding is the decrease of HLA-DR expression on monocytes,18-21 which has led to the therapeutic use of IFN-,y.34 Based on our prospective study, we could also demonstrate that this phenomenon typically starts at day 2 after injury. Moreover, HLA-DR expression on monocytes is, in contrast to T cells, influenced by the severity of trauma: analysis of variance revealed a significant inverse relation. Because patients with minor injuries developed no sepsis, this group might be excluded from HLA-DR expression analysis in further studies. After definition of a cutoff value for HLA-DR expression on monocytes, we could identify on day 2 after injury the patients at risk for the development of severe sepsis. Using the same criteria for T cells, we recognized patients at risk even on day 1. However, the earliest predictive marker for severe sepsis-prone patients appears to be the low level of soluble HLA-DR immediately after admission. If this finding is substantiated by further prospective studies, soluble HLA-DR quantification by enzyme-linked immunosorbent assay may allow a fast and simple identification of trauma patients at risk for sepsis. Nevertheless, it remains to be clarified whether soluble HLA-DR levels have the same prognostic power for sepsis as the soluble CD14

Molecules and Sepsis After Injury ~~~~~~~~~~~~HLA

Vol. 229 * No. 2

Vol. 229 No. 2 -

251

FLUORESCENCE INTENSITY VALUES IN THE THREE STUDY GROUPS Day 6 Group I 7.7

t

0.6

15 ± 1.4 1165±t156 67 ±3.3 800 ±120 2.5 ±0.5 19 ±6 162 ±1.2 9.1 ± 0.7 726 ±110

Day 8

Group 2

Group 3

9.9 ± 0.6 12.1 ± 0.7 1180 ±90 63±t1.5 780 ±66 2.6 ±0.3 22 ±5 161 ±0.7 9.8 ± 0.5 990 ±75

11.7t4: ± 0.8 8.7t4: ± 1 990±t134 61 ±3.7 595 ±76 3.3t ±0.5 23 ±7 160 ±1.2 8.5 ± 0.7 990 ±115

Group I 8.9

t

1.2

16 t 2 1377 ±199 66±t3 930 ±150 2.7 ±0.4 23 ±3.6 164 ±1.6 8.1 ± 0.7 698 ±72

Day 14

Group 2 11.9 ± 0.5 12.3 ± 0.8 1390±t102 62±t1.7 875 ±70 3 ±0.3 25 ±3 162 ±1 8.9 ± 0.5 1010 ±56

Group 3

15.2t4: 6.7t4:

Group I

± 1.4 t 0.8

920t44±110 534:t±4.5

500t4:± 70 4.5 ±1 18 ±4 1614± 1 6.2t4: ± 1 842t ±121

Group 3

Group 2

9 ± 0.9

0.5

15.74:

12.2 ± 0.8 1360 ±76 67 ±1.4 910 ±60 3 ±0.3 26 ±3 163 ±0.8 7.1 ± 0.3 830 ±53

7.7t4:

11.5

17.3 t 2.3 1480±t175 71 ±3 1070 ±150 2.4 ±0.5 24 ±7.5 166 ±4.5 6.4 ± 0.6 590 ±90

t

t t

2.7

0.8 1100±t150 59t ±3.8 670tt ±100 3.3 ±0.5 22t ±6.4 162 ±1.7 St ± 0.6 770 ±145

quantification reported by Burgmann et aP35 and Landmann et al.6 Furthier, our data may allow an accurate prediction of

exotoxins, of which a maj'or group are the superantigens, known to be potent T-cell mitogens.262 This polyclonal

sepsis, resulting in a sensitivity of 88% and a specificity of

T-cell activation may finally result in T-cell anergy and shock via T-cell or monocyte-derived cytokines.2 Our striking finding that initially low levels of soluble HLA-DR are strongly associated with a risk for sepsis could hint at the underlying mechanism. Because soluble HLA-DR molecules can bind superantigen by complex formation, these exotoxins are consequently withdrawn for their action on T cells. Soluble HLA-DR molecules may therefore be considered physiologic "detoxifiers" of superantigens. This function may explaln why the administration of high-dose intravenous imnmunoglobulin preparations known to contain substantial amounts of soluble HLA-DR38-40 improved sur-

70%, by combining the following two parameters: initially low soluble HLA-DR level and decreased HLA-DR expression on T cells at day 1. Moreover, our findings point to a very early disturbance of host defense mechanisms regarding T cells and monocytes. Indeed, dysfunctions of T cells, rendering trauma patients more susceptible to microbial attack and prone to posttraumatic sepsis, were observed by Chaudry and

Ayala.3 The molecular events leading to gram-positive sepsis are still unclear. Nevertheless, gram-positive bacteria secrete

170,

160.

T cell HLA-DR

Monxocyte HLA-DR

140. 0

165. 0

U

120.

U

z 100.~

160.

80* *

155.

*

*

*

*

60J

6

i ~

4

6 i

6 6 t 6 6 10 1'1 1'2 1'3 1'4 1'5

Days after trauma T in patients with cells HLA-DR on 3. of expression Figure Monitoring minor injuries (0) and severe injuries without (40) and with (V) severe sepsis. MFI, mean fluorescence intensity. MFI values are given as mean ± SEM. *p < 0.05 group 3 vs. group 2; t p < 0.05 group 3 vs. group 1.-

~

6 6

6 6 lb 11l 1'2

13

1'4 15

Days after trauma

Figure 4. Monitoring of HLA-DR expression on monocytes in patients with minor injuries (0) and severe injuries without (0) and with (V) severe sepsis. MFI, mean fluorescence intensity. MFI values are given as mean ± SEM. *p < 0.05 group 3 vs. group 2; t p < 0.05 group 3 vs. group 1.


252 3.0-


Soluble HLA-DR

2.5E

a .a

2.0. 1.5.

re

1.0.

0.5. *

*

*

*

*

0.0.

0(

5

i

i

5

10 11

7'

12

1'3

1'4

1'5

Days after trauma

Figure 5. Monitoring of soluble HLA-DR plasma levels in patients with minor injuries (0) and severe injuries without (0) and with (V) severe sepsis. Mean soluble HLA-DR plasma levels ± SEM. * p < 0.05 group 3 vs. group 2; t p < 0.05 group 3 vs. group 1.

vival of septic patients. 42 Alternatively, persons with high levels of soluble HLA-DR are to a greater extent "protected" from superantigen-induced disturbances of the immune system. In this context, the increased frequency of HLA-

DRBI *01 alleles in the nonseptic patients and the presence of this allele in only one patient in the sepsis group point to a possible genetic background. With regard to HLA-DR expression being under genetic control by respective promoter regions, Singal et a143 have recently reported that the HLA-DRB1 *01-linked promoters are characterized by a four- to sixfold higher activity. This increased promoter activity may eventually result in an elevated plasma level of soluble HLA-DR. The pathogenic mechanisms leading to gram-negative sepsis are rather well known. Endotoxins released from the cell wall of gram-negative bacteria can bind to the lipopolysaccharide-binding protein. This complex finds its receptor (CD14) on macrophages, followed by their activation and cytokine secretion.4445 It has been reported that the CD14 expression on monocytes is significantly decreased in nonsurvivors versus survivors after trauma,46 but our data could not corroborate these findings: we did not observe significantly different CD14 expression on monocytes in severely injured patients with sepsis. This could be explained by the low incidence of gram-negative sepsis in our septic collective. Our prospective study indicates that different immune cells appear to be involved in the development of grampositive and gram-negative sepsis after severe trauma.

Table 3. POSITIVE BLOOD CULTURES AND HLA-DR LEVELS IN SEPTIC PATIENTS

Patient A.B. B.C. C.G. D.D. E.K. W.K. T.J. K.K. G.M. R.K. H.Z. F.B. G.F. S.T. S.B. N.J. M.K. W.S. U.V. A.K. *

Bacteria Species K. pneumoniae K oxytoca K pneumoniae P. aeruginosa Enterobacter Enterobacter S. epidermidis S. epidermidis S. aureus S. aureus S. aureus S. epidermidis Enterococcus S. epidermidis S. epidermidis S. aureus S. aureus S. aureus S. epidermidis S. epidermidis

MFI: mean fluorescence intensity

t Day after admission 4 n.d., not done , below threshold value above threshold value

First Positive Blood Culture

T Cell HLA-DR < 161

Day

Day It

8 12 12 8 8 8 6 4 6 4 5 4 6 10 6 9 9 10 13 7

1 T

MFI* Channels

4

n.d.t 4 4 4 4 4 4 4 4 4 4 4 4 4 4 n.d. 4

Monocyte HLA-DR < 90 MFI Channels Day 2

Plasma Soluble HLA-DR < 0.5 ,lg/ml Day 0

4 4 4 4 4 4 1 4

4 4 4 4 4 4 4 4 4

t t t n.d. 4 4 n.d. 4 4 4 4

T t n.d. 4 4 4 4 4 n.d. 4 4

Vol. 229 * No. 2

Whereas the reduction of HLA-DR on T cells at day 1 was associated with gram-positive sepsis, decreased HLA-DR levels on monocytes at day 2 were followed by gramnegative sepsis. These observations, however, are still based on a limited number of patients. Therefore, specific studies on this topic are urgently needed, because their results may have important consequences for the specific application of antibiotics. In summary, monitoring of HLA-DR molecules on two cell subsets and soluble HLA-DR are promising in vitro parameters to predict within 2 days after admission severely injured patients at risk for the development of sepsis. This diagnostic improvement may enable earlier therapeutic intervention.

Acknowledgments The authors thank Mrs. S. Plewa, Mrs. B. Nyadu, and Mrs. B. Thiam for their excellent technical assistance.

References 1. Baker CC, Oppenheimer L, Stevens B, et al. Epidemiology of trauma deaths. Am J Surg 1980; 140:144-150. 2. Goris RJA, Draaisma J. Causes of death after blunt trauma. J Trauma 1982; 22:141-146. 3. Centers for Disease Control. Increase in national hospital discharge survey rates for septicemia-United States, 1979-1987. JAMA 1990; 263:937-938. 4. O'Mahony JB, Palder SB, Wood JJ, et al. Depression of cellular immunity after multiple trauma in the absence of sepsis. J Trauma 1984; 24:869-874. 5. Polk HC, George CD, Wellhausen SR, et al. A systematic study of host defense processes in badly injured patients. Ann Surg 1986; 204:282297. 6. Faist E, Kupper TS, Baker CC, et al. Depression of cellular immunity after major injury. Arch Surg 1986; 121:1000-1005. 7. Nishijima MK, Takezawa J, Hosotsubo KK, et al. Serial changes in cellular immunity of septic patients with multiple organ-system failure. Crit Care Med 1986;14:87-91. 8. Christou NV, McLean APH, Meakins JL. Host defense in blunt trauma: interrelationships of kinetics of anergy and depressed neutrophil function, nutritional status, and sepsis. J Trauma 1980; 20:833839. 9. Keane RM, Birmingham W, Shatney CM, et al. Prediction of sepsis in the multitraumatic patient by assays of lymphocyte responsiveness. Surg Gynecol Obstet 1983; 156:163-167. 10. O'Mahony JB, Wood JB, Rodrick ML, Mannick JA. Changes in T-lymphocyte subsets following injury. Ann Surg 1985; 202:580-586. 11. Faist E, Mewes A, Strasser T, et al. Alteration of monocyte function following major injury. Arch Surg 1988; 123:287-292. 12. Livingston DH, Appel SH, Wellhausen SR, Sonnenfeld G. Depressed interferon-gamma production and monocyte HLA-DR expression after severe injury. Arch Surg 1988; 123:1309-1312. 13. Faist E, Ertel W, Baker CC, Heberer G. Terminal B-cell maturation and immunoglobulin (Ig) synthesis in vitro in patients with major injury. J Trauma 1989; 29:2-9. 14. Baker SP, O'Neill B. The injury severity score: an update. J Trauma 1976; 16:882-886. 15. Greenspan L, McLellan BA, Greig H. Abbreviated injury scale and injury severity score: a scoring chart. J Trauma 1985; 25:60-64.


253

16. Chamion HR, Copes WS, Sacco WJ, et al. A new characterization of injury severity. J Trauma 1990; 30:539-549. 17. Hershman MJ, Cheadle WG, Kuftinec D, Polk Jr HC. An outcome predictive score for sepsis and death following injury. Injury 1988; 19:263-266. 18. Cheadle WG. The human leukocyte antigens and their relationship to infection. Am J Surg 1993; 165:75S-81S. 19. Hershman MJ, Cheadle WG, Wellhausen SR, et al. Monocyte HLA-DR expression characterizes clinical outcome in the trauma patient. Br J Surg 1990; 77:204-207. 20. Cheadle WG, Hershman MJ, Wellhausen SR, Polk HC. HLA-DR expression on peripheral blood monocytes correlates with surgical infection. Am J Surg 1991; 161:639-645. 21. Wakefield CH, Carey PD, Foulds S, et al. Changes in major histocompatibility complex II expression in monocytes and T cells of patients developing infection after surgery. Br J Surg 1993; 80:205-209. 22. Baehr RV, Volk HD, Reinke P, et al. An immune monitoring program for controlling immunosuppressive therapy. Transplant Proc 1989; 21:1189-1191. 23. Reinke P, Volk HD. Diagnostic and predictive value of an immune monitoring program for complications after kidney transplantation. Urol Int 1992; 49:69-75. 24. McDonald JC, Adamashvili I, Hayes JM, et al. Soluble HLA class II concentrations in normal individuals and transplant recipients. Comparison with soluble HLA class I concentrations. Transplantation 1994; 58:1268-1272. 25. Nicolle MW, Nag B, Sharma SD, Willcox N, et al. Specific tolerance to an acetylcholine receptor epitope induced in vitro in myasthenia gravis CD4+ lymphocytes by soluble major histocompatibility complex class 1I-peptide complexes. J Clin Invest 1994; 93:1361-1369. 26. Kotb M. Bacterial pyrogenic exotoxins as superantigens. Clin Microbiol Rev 1995; 8:411-426. 27. Huber BT, Hsu PN, Sutkowski N. Virus-encoded superantigens. Microbiol Rev 1996; 60:473-482. 28. Gutierrez-Ramos JC, Bluethmann H. Molecules and mechanisms operating in septic shock: lessons from knockout mice. Immunol Today 1997; 18:329-334. 29. Mottershead DG, Hsu PN, Urban RG, et al. Direct binding of the Mtv7 superantigen (Mls-1) to soluble MHC class II molecules. Immunity 1995; 2:149-154. 30. American College of Chest Physicians/Society of Critical Care Medicine consensus conference. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Crit Care Med 1992; 20:864-874. 31. Westhoff U, Grosse-Wilde H. Soluble HLA class I and class II concentrations in factor VIII and PCC preparations. Vox Sang 1995; 68:73-76. 32. Olerup 0, Zetterquist H. HLA-DR typing by PCR amplification with sequence-specific primers (PCR-SSP) in 2 hours: an alternative to serological DR typing in clinical practice including donor-recipient matching in cadaveric transplantation. Tissue Antigens 1992; 39:225235. 33. Olerup 0, Aldener A, Fogdell A. HLA-DQB 1 and -DQA1 typing by PCR amplification with sequence-specific primers (PCR-SSP) in 2 hours. Tissue Antigens 1993; 41:119-134. 34. Docke WD, Randow F, Syrbe U, et al. Monocyte deactivation in septic patients: restoration by IFN-y treatment. Nature Med 1997; 3:678681. 35. Burgmann H, Winkler S, Locker GJ, et al. Increased serum concentration of soluble CD14 is a prognostic marker in gram-positive sepsis. Clin Immunol Immunopathol 1996; 80:307-310. 36. Landmann R, Reber AM, Sansano S, Zimmerli W. Function of soluble CD14 in serum from patients with septic shock. J Infect Dis 1996; 173:661-668.

254


37. Chaudry IH, Ayala A. Effect of hemorrhage on specific immune responses. In Chaudry IH, Ayala A (eds). Immunological aspects of haemorrhage. Austin: RG Landes Co.; 1992:53-82. 38. Grosse-Wilde H, Blasczyk R, Westhoff U. Soluble HLA class I and class II concentrations in commercial immunoglobulin preparations. Tissue Antigens 1992; 39:74-77. 39. Blasczyk R, Westhoff U, Grosse-Wilde H. Soluble CD4, CD8, and HLA molecules in commercial immunoglobulin preparations. Lancet 1993; 341:789-790. 40. Lam L, Whitsett CF, McNicholl JM, et al. Immunologically active proteins in intravenous immunoglobulin. Lancet 1993; 342:678. 41. Dominioni L, Dionigi R, Zanello M, et al. Effects of high-dose IgG on survival of surgical patients with sepsis scores of 20 or greater. Arch Surg 1991; 126:236-240.


42. Schedel I, Dreikhausen U, Nentwig B, et al. Treatment of gramnegative septic shock with an immunoglobulin preparation: a prospective, randomized clinical trial. Crit Care Med 1991; 19:11041113. 43. Singal DP, Qiu X. Polymorphism in the upstream regulatory region and level of expression of HLA-DRB genes. Mol Immunol 1994; 31:1117-1120. 44. Wright SD, Ramos RA, Tobias PS, et al. CD14, a receptor for complexes of lipopolysaccharide (LPS) and LPS binding protein. Science 1990; 249:1431-1433. 45. Tobias PS, Ulevitch RJ. Lipopolysaccharide binding protein and CD14 in LPS dependent macrophage activation. Immunobiology 1993; 187: 227-232. 46. Heinzelmann M, Mercer-Jones M, Cheadle WG, Polk HC. CD14 expression in injured patients correlates with outcome. Ann Surg 1996; 224:91-96.