Abstract
Endotoxin-stimulated blood cytokine responses have been widely used to describe compromised host defense mechanisms after trauma. We investigated whether blood cytokine production after endotoxin stimulation is able to define distinct trauma-induced alteration patterns and whether alteration patterns are associated with tumor necrosis factor (TNF) gene polymorphisms. In 48 patients undergoing joint replacement, the levels of TNF alpha (TNF-α), interleukin 6 (IL-6), and IL-8 production in blood after endotoxin stimulation were measured preoperatively on the day of surgery and 24 h thereafter. Patients were genotyped for the TNF-α position −308 G/A polymorphism and the TNF-β NcoI polymorphism. Postoperative alterations, i.e., increases or decreases of cytokine levels (TNF-α versus IL-6, P = 0.013; TNF-α versus IL-8, P = 0.001; IL-6 versus IL-8, P = 0.007), and relative postoperative changes, i.e., percentages of preoperative cytokine levels (TNF-α versus IL-6, rs = 0.491, P < 0.001; TNF-α versus IL-8, rs = 0.591, P < 0.001; IL-6 versus IL-8, rs = 0.474, P < 0.001 [where rs is the Spearman rank correlation coefficient]), had significant positive correlations among the cytokines. Overall enhanced postoperative alteration patterns were found in 10 patients, attenuated patterns were found in 18 patients, and mixed patterns were found in 20 patients. Preoperative cytokine production levels differed significantly between these groups (those of the overall enhanced pattern group were less than those of the mixed pattern group, which were less than those of the overall attenuated pattern group). TNF polymorphisms were not associated with overall alteration patterns, but the A*TNFB1 haplotype was associated with a postoperative increase in TNF-α production (P = 0.042). Whole-blood cytokine responses to endotoxin define the following preexisting patterns in leukocyte function: low baseline production and overall enhanced alteration patterns after trauma (type 1), intermediate baseline production and mixed alteration patterns (type 2), and high baseline production and overall attenuated alteration patterns (type 3). TNF gene polymorphisms were associated with changes in TNF-α production but do not explain the overall reaction patterns of cytokine production after trauma. The clinical correlate of these newly defined reaction types remains to be determined.
Infection, severe injuries, and major surgical trauma are known to affect cellular immune functions, which are regarded to be responsible for increased susceptibility to secondary complications, e.g., sepsis and multiple-system organ failure (8, 14, 19). Similar to measurements of monocytic HLA-DR expression, detection of ex vivo lipopolysaccharide (LPS)-stimulated blood cytokine responses has been widely used as an indicator of compromised host defense mechanisms following accidental or surgical trauma (1, 2, 3, 23).
Furthermore, inherited variabilities of cytokine production and genetic predisposition for fatal infectious diseases have been suggested (26). In sepsis patients and in severely injured patients, two tightly associated tumor necrosis factor (TNF) gene polymorphisms, the TNF alpha (TNF-α) position −308 G/A polymorphism, a guanine-to-adenine transition at position −308 in the TNF promoter, and the TNF-β NcoI polymorphism, a biallelic restriction fragment length polymorphism in the first intron of the TNF-β gene, have been described as being significantly associated with mortality and susceptibility to severe sepsis and septic shock (11, 12, 17, 20, 21, 26). However, functional consequences associated with these polymorphisms are controversial (5, 10, 22, 26), and data on corresponding blood cytokine-producing capacities in trauma and sepsis patients are rare.
It is well established that LPS-stimulated blood cytokine production is generally down-regulated after trauma (2, 8, 13, 14, 23). However, in previous studies which presented individual alterations after surgical trauma, a remarkable proportion of patients showed enhanced cytokine responses upon LPS stimulation (13, 23). Moreover, it was suggested that down-regulation of LPS-stimulated blood cytokine production in the very early posttraumatic period is less pronounced or not detectable in patients who develop severe sepsis after accidental trauma (4, 6, 15).
Based on the foregoing findings, we hypothesized (i) the existence of distinct trauma-induced alteration patterns in leukocyte function as reflected by either increased or decreased blood cytokine production after LPS stimulation and (ii) that these alteration patterns in leukocyte function might be associated with the TNF-β NcoI and TNF-α −308 G/A gene polymorphisms. Since the detection of distinct trauma-induced alteration patterns requires a comparison with individual leukocyte function under baseline conditions, we used blood samples from otherwise healthy patients undergoing joint replacement because of degenerative arthrosis as a clinical trauma model (3, 14, 23).
MATERIALS AND METHODS
Patient characteristics and study protocol.
The protocol used was approved by the ethics committee of the Faculty of Clinical Medicine Mannheim of Ruprecht-Karls University Heidelberg. Forty-eight patients who underwent elective bone and joint surgery in the Center for Orthopaedics and Trauma Surgery of the University Hospital Mannheim were included in the study. All patients fulfilled the following criteria: age over 18 years, absence of malignant or infectious disease, absence of rheumatoid arthritis, no steroid medication, and no known liver, kidney, or pancreas disease requiring treatment. Informed consent was obtained from all patients. The patient characteristics are shown in Table 1. All patients were Caucasians. Forty-six patients underwent hip replacements, one patient underwent humerus head replacement, and one patient underwent knee replacement because of degenerative arthrosis. None of the patients had activated osteoarthritis at the time of the operation. All patients used nonsteroidal antiphlogistics for pain management preoperatively. All patients received antibiotics for perioperative infection prophylaxis with cefuroxime and postoperative pain management with opioids during the study period. The clinical course was uneventful for all patients. None of the patients developed infectious complications. Blood samples were drawn on the day of surgery (preoperative sample) between 6:00 and 8:00 a.m. and 24 h thereafter (postoperative sample). Blood (9 ml) was collected in plastic tubes (NH4-heparin tubes; Sarsted, Nümbrecht, Germany) and immediately used for stimulation ex vivo.
TABLE 1.
Epidemiological and clinical characteristics
| Patient characteristica | Value for all groups (n = 48) | Value for patients in indicated postoperative alteration pattern groupb
|
P | ||
|---|---|---|---|---|---|
| Overall enhanced (n = 10) | Mixed (n = 20) | Overall attenuated (n = 18) | |||
| Agee (yr) | 66 ± 1.3 | 68 ± 3.5 | 64 ± 2.4 | 68 ± 1.5 | 0.509c |
| No. of patients | |||||
| Female/male | 12/36 | 2/8 | 6/14 | 4/14 | 0.789d |
| Receiving anesthesia | |||||
| General | 14 | 4 | 5 | 5 | |
| Locoregional | 34 | 6 | 15 | 13 | 0.686d |
| Duration of surgerye (min) | 97 ± 5 | 102 ± 10 | 100 ± 9 | 90 ± 9 | 0.399c |
| Infusion during surgerye (liters) | |||||
| Crystalloids | 1.7 ± 0.1 | 1.8 ± 0.4 | 1.8 ± 0.2 | 1.4 ± 0.4 | 0.139c |
| Colloids | 0.8 ± 0.06 | 0.9 ± 0.2 | 0.9 ± 0.1 | 0.8 ± 0.1 | 0.669c |
| Hemoglobin levele (mg/dl) | |||||
| Pre | 12.7 ± 0.3 | 12.3 ± 0.5 | 13.3 ± 0.5 | 12.3 ± 0.5 | 0.295c |
| Post | 10.6 ± 0.2 | 11.1 ± 0.5 | 10.5 ± 0.4 | 10.4 ± 0.3 | 0.247c |
| No. of patients receiving autologous blood transfusion/no. receiving heterologous transfusion | |||||
| 0 U | 25/41 | 4/6 | 13/18 | 8/17 | |
| 1 U | 10/1 | 2/1 | 3/0 | 5/0 | |
| 2 U | 10/1 | 2/1 | 4/0 | 4/0 | 0.403d,f |
| 3 U | 3/5 | 2/2 | 0/2 | 1/1 | 0.130d,g |
| No. of WBC/nle | |||||
| Pre | 6.5 ± 0.4 | 7.0 ± 1.3 | 6.9 ± 0.4 | 5.7 ± 0.4 | 0.237c |
| Post | 7.5 ± 0.4 | 6.6 ± 0.6 | 7.7 ± 0.7 | 7.8 ± 0.5 | 0.369c |
| Tempe (°C) | |||||
| Pre | 36.8 ± 0.1 | 36.8 ± 0.1 | 36.8 ± 0.1 | 36.9 ± 0.1 | 0.669c |
| Post | 37.5 ± 0.1 | 37.4 ± 0.1 | 37.6 ± 0.1 | 37.5 ± 0.1 | 0.610c |
Pre, preoperative; Post, postoperative; Temp, rectal body temperature.
Data are for patients with different alteration patterns of cytokine (TNF-α, IL-6, and IL-8) production.
Values determined by MWU.
Values determined by χ2.
Data in each row are given as means ± standard deviations except where otherwise indicated.
P values represent the levels of statistical significance for differences in autologous blood transfusion between patients with different alteration patterns.
P values represent the levels of statistical significance for differences in heterologous blood transfusion between patients with different alteration patterns.
Ex vivo endotoxin stimulation of whole blood.
A human whole-blood assay was used as previously described (13). Briefly, whole blood mixed 1:1 (vol/vol) with cell culture medium (RPMI 1640 [Gibco BRL, Karlsruhe, Germany], 64 IU of penicillin/ml, 64 μg of streptomycin/ml) was transferred to microtiter plates (Falcon 3072 Microtest III tissue culture tubes; Becton Dickinson, Paramus, N.J.). Samples were prepared in duplicate. The mixtures were incubated at 37°C in 5% CO2 with endotoxin (LPS, 100 ng/ml; from a stock solution of Salmonella enterica serovar Friedenau [5 mg/ml; Sigma, Munich, Germany]) for 4 h. After incubation, the supernatants were separated and stored at −20°C until they were assayed for cytokine concentrations. All samples were assayed within 2 weeks and were not previously thawed.
Immunologic assays of cytokines.
Commercially available enzyme-linked immunosorbent assay kits were used for determinations of TNF-α, interleukin 6 (IL-6), and IL-8 (obtained from Milenia Biotech, Bad Nauheim, Germany) (lower detection limit for all assays, 30 pg/ml) in supernatants of whole-blood cultures according to the manufacturers' instructions.
Genotyping for the TNF-α −308 G/A and TNF-β NcoI polymorphisms.
Each patient's genomic DNA was extracted from whole blood by use of a commercially available DNA isolation kit (QIAmp blood kit; QIAGEN, Krefeld, Germany) according to the manufacturers' instructions. One microliter (20 to 80 ng) of genomic DNA was used. Genotyping for the TNF −308 G/A genetic polymorphism was done with a real-time PCR assay with specific fluorescence-labeled hybridization probes, as reported recently (sense probe, 5′-AAGGAAACAGACCACAGACCTG; antisense probe, 5′-GGTCTTCTGGGCCACTGAC; detection probe specific for the G allele [TNF1], 5′-AACCCCGTCCCCATGCC; anchor probe, 5′-CAAAACCTATTGCCTCCATTTCTTTTGGGGAC) (12, 18). A sample was classified as TNF −308 genotype A/A (TNF2/TNF2), A/G (TNF1/TNF2), or G/G (TNF1/TNF1) according to the derivative melting curve.
The genotype of the NcoI restriction fragment length polymorphism of the TNF-β gene was determined by PCR amplification and enzymatic digestion of the products with NcoI as described previously (11), with the following nucleotide sequences for PCR amplification: 5′-CCGTGCTTCGTGCTTTGGACTA-3′ and 5′-AGAGGGGTGGATGCTTGGGTTC-3′. The TNFB1 allele includes a restriction site for NcoI which results in 196- and 586-bp fragments after the digestion of the amplified 782-bp product. A sample was classified as having TNF-β NcoI genotype TNFB1/TNFB1 (196- and 586-bp fragments), TNFB1/TNFB2 (196-, 586-, and 782-bp fragments), or TNFB2/TNFB2 (a 782-bp fragment) according to the fragments resulting after NcoI digestion.
Statistical analysis.
Epidemiological and clinical parameters are described as means ± standard errors of the means, and cytokine concentrations are described as the medians with the 25th to 75th percentiles indicated. To test for normal distribution, the Kolmogorov-Smirnov test was used. Since the data were not normally distributed for all parameters, nonparametric statistics were used for comparisons. The Spearman rank correlation coefficient (rs) test, the chi-square test (χ2), Fisher's exact test (Fisher), the Wilcoxon matched-pairs signed-rank test (Wilcoxon), the Mann-Whitney U test (MWU), and the Kruskal-Wallis H test for multiple comparisons (KWH) were used to test for significant differences between the groups and were calculated with the SPSS for Windows program, release 10.0.7 (SPSS, Inc., Chicago, Ill.). A two-tailed P value of <0.05 was considered significant.
RESULTS
Perioperative changes in cytokine-producing capacities.
LPS-stimulated cytokine production levels of preoperatively drawn whole blood were as follows (median [25th and 75th percentiles, respectively]): 393 (210 and 608) pg of TNF-α/106 white blood cells (WBC), 2,774 (1,869 and 3,967) pg of IL-6/106 WBC, and 535 (294 and 1,034) pg of IL-8/106 WBC. As anticipated (13, 23), the postoperative cytokine production levels were slightly lower than the preoperative levels (288 [194 and 631] pg of TNF-α/106 WBC [P > 0.05, Wilcoxon], 2,152 [1577 and 2953] pg of IL-6/106 WBC [P = 0.033, Wilcoxon], and 465 [257 and 951] pg of IL-8/106 WBC [P > 0.05, Wilcoxon]).
The distribution of the individual postoperative alteration patterns in blood cytokine production after LPS stimulation is shown in Table 2. TNF-α production decreased postoperatively in 29 patients and increased in 19 patients, IL-6 production decreased in 31 patients and increased in 17 patients, and IL-8 production decreased in 27 patients and increased in 21 patients.
TABLE 2.
Perioperative alteration patterns of blood TNF-α, IL-6, and IL-8 production after LPS stimulationa
| Cytokine responses | No. of patients with:
|
|
|---|---|---|
| Increase in TNF-α production | Decrease in TNF-α production | |
| Increases in IL-6 and IL-8 production | 10 | 2 |
| Increase in IL-6 production and decrease in IL-8 production | 1 | 4 |
| Decrease in IL-6 production and increase in IL-8 production | 4 | 5 |
| Decreases in IL-6 and IL-8 production | 4 | 18 |
A postoperative increase or decrease in LPS-stimulated production of one of the three cytokines was associated with a corresponding uniform alteration for the other two cytokines (TNF-α versus IL-6, P = 0.013, Fisher; TNF-α versus IL-8, P = 0.001, Fisher; IL-6 versus IL-8, P = 0.007, Fisher). A total of 48 patients were studied.
A postoperative increase or decrease in LPS-stimulated production of one of the three cytokines was significantly associated with a corresponding uniform alteration for the two other cytokines (TNF-α versus IL-6, P = 0.013, Fisher; TNF-α versus IL-8, P = 0.001, Fisher; IL-6 versus IL-8, P = 0.007, Fisher). An analysis of the relationship of the relative postoperative changes in the levels of LPS-stimulated TNF-α, IL-6, and IL-8 production, as percentages of preoperative cytokine levels, showed highly significant positive correlations (TNF-α versus IL-6, rs = 0.491, P < 0.001; TNF-α versus IL-8, rs = 0.591, P < 0.001; IL-6 versus IL-8, rs = 0.474, P < 0.001 [where rs is the Spearman rank correlation coefficient]). Since these correlations suggest a common pattern of alterations in cytokine production, we used the simplest approach to further group the patients according to their overall reactions, which is based on three dichotomous parameters (increased or decreased responses for TNF-α, IL-6, and IL-8). Eighteen patients showed overall attenuated alteration patterns after surgery (decreased responses in all of the measured cytokines), 10 patients showed overall enhanced patterns (increased responses for all measured cytokines), and 20 patients showed mixed alteration patterns (both increased and decreased responses). The relative postoperative changes in blood cytokine production levels were significantly different between those three groups (Fig. 1) (P value of <0.01 for all cytokines, KWH).
FIG. 1.
Preoperative (pre) and postoperative (post) cytokine production in whole blood after LPS stimulation. Whole-blood TNF-α, IL-6, and IL-8 production levels after LPS stimulation in patients with overall attenuated (n = 18) (A), mixed (n = 20) (B), and overall enhanced (n = 10) (C) alteration patterns of cytokine responses after trauma are shown. Individual lines represent individual preoperative and postoperative values. % pre, percentage of the preoperatively determined cytokine production after LPS stimulation. (D) Postoperative cytokine production levels as percentages of preoperative levels from patients with overall enhanced (white bars), mixed (grey bars), and overall attenuated (striped bars) alteration patterns in blood cytokine production after LPS stimulation. Data represent median cytokine production levels (error bars represent the range between the 25th and 75th percentiles). A comparison of cytokine production levels between groups showed significant differences for all three cytokines (P < 0.01, KWH). Differences were localized by using MWU. *, P < 0.05; **, P < 0.01.
The absolute preoperative blood cytokine production level for each individual cytokine after LPS stimulation was significantly lower in patients with an increased postoperative response than in patients with a decreased postoperative response (Fig. 2). Conversely, postoperative levels of TNF-α and IL-6 production were significantly higher in patients with an increased postoperative response than in patients with a decreased postoperative response for each cytokine. No statistically significant difference between patients with postoperatively increased or decreased blood IL-8 production levels was found.
FIG. 2.
Whole-blood cytokine production after LPS stimulation in patients grouped according to postoperative alterations for each individual cytokine and overall alteration patterns. Cytokine production is given in picograms per 106 WBC. Each symbol represents one patient. Horizontal lines represent median values. *, P < 0.05, MWU; **, P < 0.01, MWU. Production levels of TNF-α (A), IL-6 (C), and IL-8 (E) were measured preoperatively in patients with an increased postoperative response (□), preoperatively in patients with a decreased postoperative response (○), postoperatively in patients with an increased postoperative response (▪), and postoperatively in patients with a decreased postoperative response (•). Preoperative production levels of TNF-α (B), IL-6 (D), and IL-8 (F) were measured preoperatively in patients with overall enhanced postoperative alteration patterns (▪), in patients with mixed postoperative alteration patterns (▴), and in patients with overall attenuated postoperative alteration patterns (•). A comparison of cytokine production levels between groups showed significant differences for all three cytokines (P values of <0.01 for TNF-α, 0.013 for IL-6, and 0.012 for IL-8; KWH). Differences were localized by using MWU.
A comparison of patients with overall enhanced, mixed, and overall attenuated alteration patterns after trauma showed significant differences preoperatively (P values of <0.01 for TNF-α, 0.013 for IL-6, and 0.012 for IL-8, KWH) (Fig. 2). Preoperative TNF-α, IL-6, and IL-8 production levels in LPS-stimulated blood were lowest in patients with overall enhanced postoperative cytokine responses, intermediate in patients with mixed alteration patterns after surgery, and highest in patients with overall attenuated postoperative alteration patterns.
A comparison of patients with increased and decreased blood TNF-α, IL-6, and IL-8 production after surgery (data not shown), as well as of patients with overall enhanced, mixed, and attenuated alteration patterns of cytokine production after surgery, with regard to epidemiological and clinical characteristics, showed no significant differences between any of the measured variables (Table 1).
Allele frequencies and genotype distribution.
The overall allele frequencies were 19% A and 81% G for the TNF-α −308 G/A polymorphism and 35% TNFB1 and 65% TNFB2 for the TNF-β NcoI polymorphism. The genotype distribution for the TNF-α −308 G/A polymorphism was 66.7% (n = 32) G/G, 29.2% (n = 14) A/G, and 4.2% (n = 2) A/A. For the TNF-β NcoI polymorphism, the genotype distribution was 41.7% (n = 20) TNFB1/TNFB2, 43.8% (n = 21) TNFB2/TNFB2, and 14.6% (n = 7) TNFB1/TNFB1.
As expected (12, 25), the TNF-α −308 G/A polymorphism was significantly (P < 0.001, χ2) associated with the TNF-β NcoI polymorphism (data not shown). Epidemiological and clinical parameters were equally distributed among patients with the different genotypes (P > 0.05, KWH; data not shown).
Association of cytokine-producing capacities with TNF gene polymorphisms.
Differences between the preoperative and postoperative cytokine-producing capacities of patients with genotypes of the TNF-α −308 G/A and TNF-β NcoI polymorphisms were not detectable (P > 0.05, KWH; data not shown).
Comparisons of the genotype distributions for the TNF-α −308 G/A polymorphism, the TNF-β NcoI polymorphism, and the TNF-α −308 G/A-TNF-β NcoI genotype combinations between patients with overall enhanced, mixed, and overall attenuated alteration patterns in postoperative blood cytokine production after LPS stimulation showed no significant differences (by χ2, P values of 0.704 for the TNF-β NcoI polymorphism, 0.068 for the TNF-α −308 G/A polymorphism, and 0.315 for the TNF-α −308 G/A-TNF-β NcoI genotype combinations; data not shown).
The genotype distribution in patients with increased or decreased postoperative levels of production for each individual cytokine showed a significant association of the TNF-α −308 G/A polymorphism and the TNF-α −308 G/A-TNF-β NcoI genotype combinations with alterations in LPS-stimulated TNF-α production (Table 3). In 75% (24 of 32) of the patients with genotypes homozygous for the G allele of the TNF-α −308 G/A polymorphism, TNF-α production decreased postoperatively. In contrast, TNF-α production decreased postoperatively in only about 30% (5 of 16) of the patients with the A/A and G/A genotypes of the TNF-α −308 G/A polymorphism.
TABLE 3.
Distribution of TNF genotypes in patients with a perioperative increase or decrease in TNF-α production after LPS stimulation
| Genotype | No. of patients with:
|
Pa | |
|---|---|---|---|
| Increase in TNF-α production | Decrease in TNF-α production | ||
| TNF-α −308 | |||
| A/A | 2 | 0 | |
| A/G | 9 | 5 | |
| G/G | 8 | 24 | 0.009 |
| TNF-β NcoI | |||
| TNFB1/TNFB1 | 5 | 2 | |
| TNFB1/TNFB2 | 6 | 14 | |
| TNFB2/TNFB2 | 8 | 13 | 0.153 |
| TNF-α −308-TNF-β NcoI | |||
| A/A-TNFB1/TNFB1 | 2 | 0 | |
| A/G-TNFB1/TNFB1 | 3 | 1 | |
| A/G-TNFB1/TNFB2 | 4 | 4 | |
| A/G-TNFB2/TNFB2 | 2 | 0 | |
| G/G-TNFB1/TNFB1 | 0 | 1 | |
| G/G-TNFB1/TNFB2 | 1 | 10 | |
| G/G-TNFB2/TNFB2 | 7 | 13 | 0.034 |
The P value determined by χ2 is the level of statistical significance for differences in genotype distribution between patients with increases and those with decreases in TNF-α production.
The postoperative levels of TNF-α production were 157 and 189% of the preoperative levels for two patients with genotypes homozygous for the A allele. The median (25th and 75th percentile) TNF-α production levels were 170% (55 and 320%) of preoperative production levels for patients with genotypes heterozygous for the G allele and 63% (39 and 102%) for patients with genotypes homozygous for the G allele (P = 0.011, KWH).
Haplotype analysis revealed that the −308 A*TNFB1 haplotype had a significant positive correlation with a perioperative increase in TNF-α production. The correlation of the haplotypes carrying the −308 G allele with a perioperative decrease in the TNF-α production level did not reach a level of statistical significance (Table 4).
TABLE 4.
TNF-α −308-TNF-β NcoI haplotypes and perioperative changes in LPS-stimulated blood TNF-α production
| Haplotype | No. of patients with increase/ no. of patients with decrease, in TNF-α-producing capacitya
|
rs (P)c | Pd | |||
|---|---|---|---|---|---|---|
| 2b | 1 | 0.5 | 0 | |||
| A*TNFB1 | 2/0 | 3/1 | 5/4 | 9/24 | 0.399 (0.005) | 0.042 |
| A*TNFB2 | 0/0 | 1/0 | 5/4 | 25/13 | 0.222 (0.129) | 0.229 |
| G*TNFB1 | 0/1 | 4/11 | 5/4 | 10/13 | −0.151 (0.304) | 0.422 |
| G*TNFB2 | 5/2 | 5/4 | 2/10 | 7/13 | −0.230 (0.116) | 0.081 |
Haplotypes of heterozygotes were scored as 0.5 for each haplotype.
Number of haplotypes.
rs, correlation between the numbers of the patients with an increase or decrease in TNF-α secretion.
Determined by χ2.
Although the genotype distribution of the TNF-α −308 G/A polymorphism showed a significant association with postoperative alterations in IL-8 production (P = 0.04, χ2; data not shown), alterations in IL-8 production were not significantly associated with the TNF-β NcoI or TNF-α −308 G/A-TNF-β NcoI genotype combination (P > 0.05, χ2). Significant associations between the TNF gene polymorphisms and postoperative alterations in IL-6 production were not detectable (P > 0.05, χ2).
DISCUSSION
In this study, the patients who underwent elective joint replacement because of degenerative arthrosis were otherwise healthy. We assumed that preoperative cytokine production most likely reflects normal leukocyte function. Although the possibility that degenerative arthrosis influenced LPS-stimulated cytokine production per se cannot be excluded, at least the preoperative data were not confounded by any influence of trauma or injury.
We used a 24-h interval after surgical trauma to study changes in LPS-evoked cytokine responses, because we showed previously that a depression of this leukocyte function occurs in patients with isolated fractures after 24 h and that operative trauma significantly reduced this leukocyte function, according to a comparison of preoperative and 24-h-postoperative data (13). As expected and in line with previous studies (13, 23), the overall reduction of LPS-stimulated blood cytokine production 24 h after the operation was small. This outcome suggests that the consequences of controlled orthopedic trauma on LPS-stimulated cytokine production are comparable to limited blunt accidental trauma, i.e., isolated fractures (13).
Analysis of the individual data showed that 40% of patients reacted with increased cytokine responses after surgical trauma and 60% of patients reacted with decreased cytokine responses. Postoperative alterations in ex vivo TNF-α, IL-6, and IL-8 release were significantly associated with each other. Surprisingly, we found that these subsequent responses after surgical trauma were associated with significant differences preoperatively. This result strongly indicates preexisting differences in leukocyte function which influence specific alteration patterns after trauma.
Based on these findings, distinct types of leukocyte function in patients who are indistinguishable by epidemiological and clinical characteristics can be defined for patients with overall enhanced, mixed, or overall attenuated blood TNF-α, IL-6, and IL-8 responses to LPS after trauma. These types are as follows: low cytokine (TNF-α, IL-6, and IL-8) responses to LPS under normal conditions and enhanced responses after trauma (type 1), intermediate cytokine responses under normal conditions and mixed alterations in cytokine production after trauma (type 2), and high cytokine responses under normal conditions and attenuated responses after trauma (type 3).
This assumption is strengthened by findings from others, which show that LPS-induced IL-6 production in whole blood is regulated independently from TNF-α production (24) and that ex vivo production of TNF is influenced by genetic factors (26), with a variation of 60% for TNF on the basis of heritability alone.
It certainly is a limitation of the present study that postoperative blood cytokine production after LPS stimulation was measured only 24 h after surgery. Measurements at several postoperative time points, as well as measurements in patients undergoing more than one operative procedure consecutively, are clearly required to establish the above-described assumption with confidence. Nevertheless, trauma-induced alterations of leukocyte function are known to be time dependent, but as yet, there is no evidence from the literature that individual alterations of leukocyte function change direction in principle (3, 4, 6, 8, 13, 14, 23). Therefore, the 24-h measurement rather underestimates the differences between the defined reaction types if the peak of alteration was missed (23).
In the present study, patients with increased blood TNF-α and IL-6 responses to LPS after trauma were found to have significantly increased postoperative levels of these cytokines compared to patients with decreased responses after trauma. Although we were not able to detect this relationship in patients grouped according to their overall alteration patterns, only 21% (10 of 48) of the patients showed overall enhanced reaction patterns, and therefore our sample size might have been too small to detect significant differences.
The reaction patterns used certainly oversimplify the complex regulation of whole-blood LPS responsiveness. However, it was the aim of the present study to use whole-blood cytokine responses to LPS stimulation as a global parameter of the patients' actual leukocyte function after trauma. Therefore, these categories indicate differences in actual leukocyte function and mechanistically describe a general predisposition for distinct alterations in inflammatory conditions independent from the regulatory mechanism.
The allele frequencies, genotype distributions, and association of the TNF-α −308 G/A polymorphism with the TNF-β NcoI polymorphism in the present study are in agreement with previously published data (11, 17, 25). In line with previous studies by others (7, 22), we were not able to detect a direct association of patients with these genotypes and protein production levels in preoperatively or postoperatively obtained blood.
Methodological differences may explain these controversial findings (5). Such differences may also suggest that, if it exists, the association of the TNF-α −308 G/A and TNF-β NcoI polymorphisms with the cytokine-producing capacities of normal stimulated peripheral blood mononuclear cells or whole blood is weak and probably caused by an association with currently unknown genetic elements responsible for the regulation of those leukocyte functions.
The presence of the A allele of the TNF-α −308 G/A polymorphism, as well as TNFB2 homozygosity, has been shown to correlate with sepsis susceptibility and mortality in critically ill patients (11, 12, 17, 20, 21). The significant association of trauma-induced alterations in LPS-stimulated TNF-α production with the TNF-α −308 G/A polymorphism and the A*TNFB1 haplotype suggests an influence of these TNF polymorphisms on leukocyte reactivity to inflammatory stimuli. However, this association cannot explain the significant differences in the preoperative TNF-α production levels of patients with the defined reaction types. Besides the cytokine network, several other molecules have been described to regulate leukocyte function after trauma (9, 16). Since the TNF-α −308 G/A and TNF-β NcoI polymorphisms did not correlate with protein production, differences in release of regulatory molecules seem more likely to account for the observed differences.
The findings that postoperative alterations in blood TNF-α, IL-6, and IL-8 responses to LPS are significantly associated with each other and that preoperative cytokine production was found to correlate significantly with the defined reaction types indicate that genetic elements other than these TNF gene polymorphisms regulate the uniform alterations in cytokine production. Nevertheless, the association of the A*TNFB1 haplotype with increased blood TNF-α responses to LPS after trauma and the finding that increased TNF-α responses after trauma are accompanied by higher-level TNF-α production postoperatively may correspond to previous findings involving severely injured patients. Those findings showed that those who develop severe posttraumatic sepsis present with increased cytokine production after LPS stimulation (4, 6, 15) and that the A allele of the −308 G/A polymorphism was found to be associated with increased susceptibility to posttraumatic sepsis (20).
The incidence of infectious complications in patients undergoing elective joint replacement is very small, and none of our patients developed even minor wound infections. Therefore, it is unquestionable that the clinical trauma model of elective orthopedic procedures is incapable of studying an association between susceptibility to severe infections and the newly defined reaction types. In order to address the potential clinical significance of those leukocyte function types and to define preoperative threshold values which identify distinct alteration patterns, studies of larger patient populations with a significant risk for severe infectious complications are necessary.
Acknowledgments
We thank Anja Bistron for excellent technical assistance.
We thank Udo Obertacke and Hanns-Peter Scharf for supporting this study.
REFERENCES
- 1.Adrie, C., M. Adib-Conquy, I. Laurent, M. Monchi, C. Vinsonneau, C. Fitting, F. Fraisse, A. T. Dinh-Xuan, P. Carli, C. Spaulding, J. F. Dhainaut, and J. M. Cavaillon. 2002. Successful cardiopulmonary resuscitation after cardiac arrest as a “sepsis-like” syndrome. Circulation 106:562-568. [DOI] [PubMed] [Google Scholar]
- 2.Cabie, A., C. Fitting, J. C. Farkas, C. Laurian, J. M. Cormier, J. Carlet, and J. M. Cavaillon. 1992. Influence of surgery on in-vitro cytokine production by human monocytes. Cytokine 4:576-580. [DOI] [PubMed] [Google Scholar]
- 3.Ditschkowski, M., E. Kreuzfelder, V. Rebmann, S. Ferencik, M. Majetschak, E. N. Schmid, U. Obertacke, H. Hirche, F. U. Schade, and H. Grosse-Wilde. 1999. HLA-DR expression and soluble HLA-DR levels in septic patients after trauma. Ann. Surg. 229:246-254. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Flach, R., M. Majetschak, T. Heukamp, V. Jennissen, S. Flohè, J. Börgermann, U. Obertacke, and F. U. Schade. 1999. Relation of ex vivo stimulated blood cytokine synthesis to posttraumatic sepsis. Cytokine 11:173-178. [DOI] [PubMed] [Google Scholar]
- 5.Heesen, M., D. Kunz, B. Bachmann-Mennenga, H. F. Merk, and B. M. Bloemeke. 2003. Linkage disequilibrium between tumor necrosis factor (TNF)-alpha-308 G/A promoter and TNF-beta NcoI polymorphisms: association with TNF-alpha response of granulocytes to endotoxin stimulation. Crit. Care Med. 31:211-214. [DOI] [PubMed] [Google Scholar]
- 6.Heesen, M., U. Obertacke, F. U. Schade, B. Bloemeke, and M. Majetschak. 2002. The interleukin-6 (IL-6) G(−174)C polymorphism and the ex vivo IL-6 response to endotoxin in severely injured blunt trauma patients. Eur. Cytokine Netw. 13:72-77. [PubMed] [Google Scholar]
- 7.Hoffmann, S. C., E. M. Stanley, E. D. Cox, N. Craighead, B. S. DiMercurio, D. E. Koziol, D. M. Harlan, A. D. Kirk, and P. J. Blair. 2001. Association of cytokine polymorphic inheritance and in vitro cytokine production in anti-CD3/CD28-stimulated peripheral blood lymphocytes. Transplantation 72:1444-1450. [DOI] [PubMed] [Google Scholar]
- 8.Keel, M., N. Schregenberger, U. Steckholzer, U. Ungethum, J. Kenney, O. Trentz, and W. Ertel. 1996. Endotoxin tolerance after severe injury and its regulatory mechanisms. J. Trauma 41:430-437. [DOI] [PubMed] [Google Scholar]
- 9.Kovacs, E. J., K. A. Messingham, and M. S. Gregory. 2002. Estrogen regulation of immune responses after injury. Mol. Cell. Endocrinol. 193:129-135. [DOI] [PubMed] [Google Scholar]
- 10.Louis, E., D. Franchimont, A. Piron, Y. Gevaert, N. Schaaf-Lafontaine, S. Roland, P. Mahieu, M. Malaise, D. de Groote, R. Louis, and J. Belaiche. 1998. Tumour necrosis factor (TNF) gene polymorphism influences TNF-alpha production in lipopolysaccharide (LPS)-stimulated whole blood cell culture in healthy humans. Clin. Exp. Immunol. 113:401-406. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Majetschak, M., S. Flohé, U. Obertacke, J. Schröder, K. Staubach, D. Nast-Kolb, F. U. Schade, and F. Stüber. 1999. Relation of a TNF gene polymorphism to severe sepsis in trauma patients. Ann. Surg. 230:207-214. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Majetschak, M., U. Obertacke, F. U. Schade, M. Bardenheuer, G. Voggenreiter, B. Bloemeke, and M. Heesen. 2002. Tumor necrosis factor gene polymorphisms, leukocyte function, and sepsis susceptibility in blunt trauma patients. Clin. Diagn. Lab. Immunol. 9:1205-1211. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Majetschak, M., R. Flach, E. Kreuzfelder, V. Jennissen, T. Heukamp, F. Neudeck, U. Obertacke, K. P. Schmit-Neuerburg, and F. U. Schade. 1999. The extent of traumatic damage determines a graded depression of the endotoxin responsiveness of peripheral blood mononuclear cells from patients with blunt injuries. Crit. Care Med. 27:313-318. [DOI] [PubMed] [Google Scholar]
- 14.Majetschak, M., R. Flach, T. Heukamp, V. Jennissen, U. Obertacke, F. Neudeck, K. P. Schmit-Neuerburg, and F. U. Schade. 1997. Regulation of whole blood tumor necrosis factor production upon endotoxin stimulation after severe blunt trauma. J. Trauma 43:880-887. [DOI] [PubMed] [Google Scholar]
- 15.Majetschak, M., R. Flach, V. Jennissen, U. Obertacke, T. Heukamp, F. Neudeck, K. P. Schmit-Neuerburg, and F. U. Schade. 1997. Ex-vivo cytokine production after severe trauma: association with posttraumatic clinical course. Shock 7:102. [Google Scholar]
- 16.Majetschak, M., U. Krehmeier, M. Bardenheuer, C. Denz, M. Quintel, G. Voggenreiter, and U. Obertacke. 2003. Extracellular ubiquitin inhibits the TNF-alpha response to endotoxin in peripheral blood mononuclear cells and regulates endotoxin hyporesponsiveness in critical illness. Blood 101:1882-1890. [DOI] [PubMed] [Google Scholar]
- 17.Mira, J. P., A. Cariou, F. Grall, C. Delclaux, M. R. Losser, F. Heshmati, C. Cheval, M. Monchi, J. L. Teboul, F. Riche, G. Leleu, L. Arbibe, A. Mignon, M. Delpech, and F. Dhainaut. 1999. Association of TNF2, a TNF-alpha promoter polymorphism, with septic shock susceptibility and mortality: a multicenter study. JAMA 282:561-568. [DOI] [PubMed] [Google Scholar]
- 18.Moos, V., M. Rudwaleit, V. Herzog, K. Hohlig, J. Sieper, and B. Müller. 2000. Association of genotypes affecting the expression of interleukin-1beta or interleukin-1 receptor antagonist with osteoarthritis. Arthritis Rheum. 43:2417-2422. [DOI] [PubMed] [Google Scholar]
- 19.Munoz, C., J. Carlet, C. Fitting, B. Misset, J. P. Bleriot, and J. M. Cavaillon. 1991. Dysregulation of in vitro cytokine production by monocytes during sepsis. J. Clin. Investig. 88:1747-1754. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.O'Keefe, G. E., D. L. Hybki, and R. S. Munford. 2002. The G→A single nucleotide polymorphism at the −308 position in the tumor necrosis factor-α promoter increases the risk for severe sepsis after trauma. J. Trauma 52:817-825. [DOI] [PubMed] [Google Scholar]
- 21.Stüber, F., M. Petersen, F. Bokelmann, and F. U. Schade. 1996. A genomic polymorphism within the tumor necrosis factor locus influences plasma tumor necrosis factor-α concentrations and outcome of patients with severe sepsis. Crit. Care Med. 24:381-384. [DOI] [PubMed] [Google Scholar]
- 22.Stüber, F., I. A. Udalova, M. Book, L. N. Drutskaya, D. V. Kuprash, R. L. Turetskaya, F. U. Schade, and S. Nedospasov. 1996. −308 tumor necrosis factor (TNF) polymorphism is not associated with survival in severe sepsis and is unrelated to lipopolysaccharide inducibility of the human TNF promoter. J. Inflamm. 46:42-50. [PubMed] [Google Scholar]
- 23.Van Deuren, M., T. B. Twickler, M. C. de Waal Malefyt, H. Van Beem, J. van der Ven-Jongekrijk, C. M. M. Verschueren, and J. W. M. van der Meer. 1998. Elective orthopedic surgery, a model for the study of cytokine activation and regulation. Cytokine 10:897-903. [DOI] [PubMed] [Google Scholar]
- 24.Vasilescu, C., D. Berger, K. Buttenschon, M. Seidelmann, and H. G. Beger. 1996. Endotoxin-induced release of interleukin 6 and interleukin 1 beta in human blood is independent of tumor necrosis factor alpha. Eur. Surg. Res. 28:55-62. [DOI] [PubMed] [Google Scholar]
- 25.Waterer, G. W., M. W. Quasney, R. M. Cantor, and R. G. Wunderink. 2001. Septic shock and respiratory failure in community-acquired pneumonia have different TNF polymorphism associations. Am. J. Respir. Crit. Care Med. 163:1599-1604. [DOI] [PubMed] [Google Scholar]
- 26.Westendorp, R. G. J., J. A. M. Langermans, T. W. J. Huizinga, C. L. Verweij, and A. Sturk. 1997. Genetic influence on cytokine production and fatal meningococcal disease. Lancet 349:170-173. [DOI] [PubMed] [Google Scholar]