Stability of tumor necrosis factor [alpha], interleukin 6, and interleukin 8 in blood samples of patients with systemic immune activation.
Abstract: * Context.--Tumor necrosis factor a, interleukin 6, and interleukin 8 serum/plasma levels are frequently used for the monitoring of patients with systemic immune activation/ sepsis. This requires comparability of test results over time. However, cytokines are usually not considered to be very stable after blood collection, which might artificially interfere with test results.

Objective.--To obtain better knowledge about stability of these cytokines in blood samples for interpretation of test results.

Design.--Blood of patients with systemic immune activation was collected in EDTA, lithium heparin, ammonium heparin, and serum tubes. Aliquots were analyzed after storage at room temperature for 2 to 8 hours. Additionally, storage conditions for separated serum/plasma for 24 hours and the reproducibility of repeated cytokine measurements by an automated DPC Immulite analyzer were tested.

Results.--Tumor necrosis factor a level was stable in EDTA plasma for 8 hours, while slightly increasing in heparin plasma and serum. Interleukin 6 concentrations were stable for 8 hours in all blood types, whereas interleukin 8 concentrations were stable only in EDTA plasma and were strongly increasing in heparin plasma and serum. Cytokine concentrations in separated serum/plasma were stable during 24 hours if stored at 4[degrees]C or frozen at -20 or -70[degrees]C. Reproducibility of repeated cytokine measurements revealed no significant differences for all blood types.

Conclusions.--Cytokine levels were most critically influenced by the period between blood collection and plasma separation, but its impact was strongly dependent on cytokine and anticoagulant. However, under appropriate conditions cytokine levels were surprisingly stable for up to 8 hours.
Article Type: Report
Subject: Tumor necrosis factor (Properties)
Interleukin-6 (Properties)
Interleukin-8 (Properties)
Immunopathology (Research)
Blood (Analysis and chemistry)
Blood (Methods)
Authors: Friebe, Astrid
Volk, Hans-Dieter
Pub Date: 11/01/2008
Publication: Name: Archives of Pathology & Laboratory Medicine Publisher: College of American Pathologists Audience: Academic; Professional Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2008 College of American Pathologists ISSN: 1543-2165
Issue: Date: Nov, 2008 Source Volume: 132 Source Issue: 11
Topic: Event Code: 310 Science & research Canadian Subject Form: Tumour necrosis factor
Geographic: Geographic Scope: Germany Geographic Code: 4EUGE Germany
Accession Number: 230246870
Full Text: Diagnostic analysis of cytokine levels in serum or plasma has become an important issue for several disease conditions. Especially in cases of systemic inflammation/sepsis, several cytokines, such as tumor necrosis factor (TNF) [alpha], interleukin (IL) 8, and IL-6, proved to be of diagnostic and prognostic value and are frequently used for the monitoring of septic patients. (1-7) Serum/plasma concentrations are usually measured by enzymelinked immunosorbent assay and automated analyzers are best suited for routine diagnostics. Nevertheless, cytokine measurement can be associated with some problems. Generally, results are only comparable if obtained by the same enzyme-linked immunosorbent assay test resulting from different epitope specificities of utilized antibodies. However, even when the same test is used, cytokine levels are usually regarded as being very susceptible and results might easily be influenced by the period between blood collection and arrival in the laboratory. (8, 9) Thus, separation of cells and plasma as quickly as possible is usually recommended, which is often difficult in clinical practice. Therefore, further knowledge about cytokine stability in blood samples is important for interpretation of test results, especially as a close cytokine monitoring requires comparability of values obtained at different time points.

The aim of this study was to investigate the stability of TNF-[alpha], IL-8, and IL-6 in blood samples. We investigated the influence of several periods ranging from 2 to 8 hours between blood collection and sample analysis. Additionally, we tested different storage conditions for serum/ plasma for 24 hours, in case samples could not be analyzed the same day. Finally, we investigated the reproducibility of repeated cytokine measurements by an automated DPC Immulite analyzer. All tests were performed with blood anticoagulated by lithium heparin, ammonium heparin, EDTA, and serum.

MATERIALS AND METHODS

Patients

Blood was taken from 24 patients with mean age of 64.79 [+ or -] 14.45 years (range, 19-80 years) consisting of 15 men and 9 women after getting informed consent. Patients were recruited from an intensive care unit and all patients suffered from systemic immune activation, either sepsis or noninfective postsurgical systemic immune response syndrome. The study was approved by the local ethics committee.

[FIGURE 1 OMITTED]

Sample Preparation

Blood was collected in serum tubes and vacutainers containing EDTA, lithium heparin, and ammonium heparin as anticoagulants (Sarstedt, Niimbrecht, Germany).

Testing of Cytokine Stability in Whole Blood Samples.--Whole blood was transferred to Eppendorf tubes and 1 aliquot was immediately analyzed to obtain baseline values. Three aliquots remained at room temperature for 2, 4, and 8 hours. Cytokines were measured in serum/plasma obtained after centrifugation at 250g- for 10 minutes at room temperature.

Testing of Cytokine Stability in Separated Serum/Plasma.--Serum/ plasma was obtained immediately after blood collection. Samples were kept for 1 hour at room temperature, 24 hours at 4[degrees]C, 24 hours at--20[degrees]C, or 24 hours at--70[degrees]C, and cytokine concentrations were compared with baseline levels.

Reproducibility of Repeated Cytokine Measurements by an Automated DPC Immulite Analyzer.--Blood was collected from 5 patients in serum, lithium heparin, ammonium heparin, and EDTA tubes. Cytokine levels were determined in serum/ plasma 10 times for each blood type and for each patient.

[FIGURE 2 OMITTED]

Assays for Cytokine Detection

Immulite TNF-[alpha] assay (2-1000 pg/mL; DPC Biermann, Bad Nauheim, Germany), Immulite IL-6 assay (2-2000 pg/mL; DPC Biermann), and Immulite IL-8 assay (6-7500 pg/ml; DPC Biermann) were used. Tests were used with the IMMULITE Automated Analyzer (DPC Biermann).

Statistical Methods

All statistical analyses were performed by SPSS 12.0 for Windows (SPSS Inc, Chicago, Ill). The limit of significance was set at .05. Cytokine levels were normally distributed (KolmogorovSmirnov test). The effect of time for cytokine stability and the effect of storage conditions on cytokine levels were tested for significance by repeated-measures analysis of variance (ANOVA).

For analysis of test reproducibility the coefficient of variation (CV) was calculated for each test series: CV (%) = (SD X 100)/ mean. Coefficients of variation were tested for significant differences between blood types by ANOVA.

RESULTS

Stability of TNF-[alpha], IL-6, and IL-8 in Whole Blood

Baseline cytokine levels were determined in serum/ plasma immediately after blood collection. The remaining blood was kept unchanged at room temperature for 2, 4, and 8 hours, respectively, before cytokine levels were measured.

[FIGURE 3 OMITTED]

Results for TNF-[alpha] showed a stable concentration if measured in EDTA plasma for up to 8 hours. In heparin plasma and especially in serum, values were slowly but continually increasing, although not reaching statistical significance, and exceeded the initial concentration after 8 hours by approximately 25% (Figure 1, A and B).

Interleukin 6 concentrations turned out to be stable for all tested blood types and corresponded even after 8 hours with the initial values (Figure 1, C and D). In EDTA plasma, extreme outliers were repeatedly measured due to a disturbance of the test system caused by the presence of EDTA. Therefore, we excluded these results from further analysis. Meanwhile, this problem has been solved by the manufacturer and the IL-6 Immulite assay is now suitable for EDTA too.

Interleukin 8 concentrations were stable only in EDTA plasma. In heparin plasma and even more in serum, values were continually and strongly increasing over time exceeding initial values by approximately 5 times for heparin plasma and 20 times for serum after 8 hours (Figure 1, E and F). This increase reached levels of significance as tested by ANOVA (lithium heparin: n = 8, [F.sub.3] = 4.727, P = .01; ammonium heparin: n = 8, [F.sub.3] = 4.775, P = .009; serum: n = 8, [F.sub.3] = 4.727, P = .01).

Stability of TNF-[alpha], IL-6, and IL-8 in Serum/Plasma Under Different Storage Conditions

We aimed to determine optimal storage conditions for serum/plasma during 24 hours in case an immediate cytokine analysis is not possible. Blood was centrifuged immediately after collection and cytokine baseline levels were determined. Remaining serum/plasma aliquots were stored for 1 hour at room temperature, 24 hours at 4[degrees]C, 24 hours at -20[degrees]C, and 24 hours at -70[degrees]C.

Results showed that cytokine concentrations corresponded with the initial values obtained immediately after blood collection for all storage conditions, which was independent if cytokines were measured in EDTA, lithium heparin, ammonium heparin plasma, or serum (Figure 2, A through F).

Reproducibility of Repeated TNF-[alpha], IL-6, and IL-8 Measurements

Cytokines were measured 10 times in EDTA, lithium heparin, ammonium heparin plasma, and serum from 5 patients. The coefficient of variation (CV = [SD X 100]/ mean) was calculated for each test series and finally, the mean resulting from 5 test series was determined for each blood type.

Repeated measurement of TNF-[alpha] resulted in a comparable amount of variation between test results for EDTA (CV = 5.7%), lithium heparin (CV = 6.3%), ammonium heparin plasma (CV = 6.2%), and serum (CV = 6.1%). There were also no statistically significant differences between coefficients of variations for IL-6 measured in lithium heparin (CV = 10%), ammonium heparin plasma (CV = 6.3%), and serum (CV = 7.1%) and no significant differences for IL-8 in EDTA (CV = 4.0%), lithium heparin (CV = 6.2%), ammonium heparin plasma (CV = 4.4%), and serum (CV = 5%) (Figure 3, A and B).

COMMENT

The aim of this study was to investigate the stability of TNF-[alpha], IL-6, and IL-8 in blood samples of patients with systemic immune activation. We investigated the influence of several time periods ranging from 2 to 8 hours between blood collection and sample analysis. TNF-[alpha] was most stable in EDTA plasma, while slightly increasing in heparin plasma and even more in serum, which is probably due to spontaneous TNF-[alpha] production by monocytes in response to syringe material. IL-6 concentrations were stable in all tested blood types. This is in line with the results of Kenis et al (8) who showed that IL-6 serum levels were stable for up to 4 days of clotting at 4[degrees]C and 20[degrees]C. However, Thavasu et al (9) reported that TNF-[alpha] as well as IL-6 concentrations were continually declining in EDTA, heparin plasma, and serum during 12 hours. These differences might result, in addition to a different enzyme-linked immunosorbent assay test that had been used, from the fact that tests were performed with blood from healthy volunteers supplemented with recombinant cytokines (as healthy subjects often do not have detectable levels of cytokines) and not from patients with an activated immune system, which might result in ongoing release of cytokines after blood collection.

Interleukin 8 concentrations were stable only in EDTA plasma and were strongly increasing in heparin plasma and serum probably due to a release of IL-8 by erythrocytes. This resulted in a 5-fold increase of IL-8 levels in heparin plasma and a nearly 20-fold increase in serum during 8 hours. Thus, when comparing cytokine levels of healthy controls and systemic immune response syndrome patients as shown for example by Rodrlguez-Gaspar et al, (10) IL-8 concentrations can be affected by specimen handling in a way that might lead to erroneous clinical impressions.

In summary, EDTA plasma seems to be most suitable for stability reasons, but unexpected problems can occur due to an interference of EDTA with the test system. Serum seems to be most critical regarding stability and should be avoided for routine parameter analysis. If heparin plasma is used, plasma should be separated as quickly as possible within 4 hours after blood collection.

Cytokine concentrations in separated serum/plasma seem to be very stable and can be stored for 24 hours at 4[degrees]C or can be frozen by--20[degrees]Cor--70[degrees]C. Storage conditions for IL-6 were also investigated in the study by Kenis et al (8) and IL-6 was stable in serum for 21 days at room temperature (20[degrees]Cand30[degrees]C) and probably for several years when stored at--20[degrees]C. Regarding the reproducibility of repeated cytokine measurements by the DPC Immulite analyzer, there were no significant differences between blood types for TNF-[alpha], IL-6, and IL-8.

Cytokine levels were most critically influenced by the period between blood collection and plasma separation, but its impact was strongly dependent on cytokine and anticoagulant. Send-out laboratories try to solve this problem by accepting only frozen serum/plasma samples and recommend in their user guidelines separation from cells within 30 minutes. However, because in daily routine this period might easily be exceeded, cytokine assay validation should particularly include careful attention to specimen type, either if samples are sent out or if readers wish to measure cytokines in their own laboratories. Nevertheless, if appropriate conditions are applied, cytokine levels were surprisingly stable for up to 8 hours.

Accepted for publication May 8, 2008.

References

(1.) Volk H, Reinke P, Doecke WD. Immunological monitoring of the inflammatory process: which variables? When to assess? Eur J Surg Suppl. 1999;584: 70-72.

(2.) Kox W, VolkT, Kox S, Volk H. Immunomodulatory therapy in sepsis. Intensive Care Med. 2000;2 6:124-12 8.

(3.) Adamik B, Zimecki M, Wlaszczyk A, Kubler A. Immunological status of septic and trauma patients: high tumor necrosis factor alpha serum levels in septic and trauma patients are not responsible for increased mortality; a prognostic value of serum interleukin 6. Arch Immunol Ther Exp. 1997;45(2-3):169-175.

(4.) Oda S, Hirasawa H, Shiga H, Nakanishi K, Matsuda K, Nakamua M. Sequential measurement of IL-6 blood levels in patients with systemic inflammatory response syndrome (SIRS)/sepsis. Cytokine. 2005;29(4):169-175.

(5.) Cojocaru IM, Musuroi C, Iacob S, Cojocaru M. Investigation of TNF-alpha, IL-6, IL-8 and of procalcitonin in patients with neurologic complications in sepsis. Rom J Intern Med. 2003;41(1):83-93.

(6.) Perenvi A, Johann-Liang R, Stavola JJ. Assessment of cord blood IL-6 levels as an indicator of neonatal sepsis. Am J Perinatol. 1999;16:525-530.

(7.) Horisberger T, Harbarth S, Nadal D, Baenziger O, Fischer JE. G-CSF and IL-8 for early diagnosis of sepsis in neonates and critically ill children--safety and cost effectiveness of a new laboratory prediction model: study protocol of a randomized controlled trial. Crit Care. 2004;8(6):R443-R450.

(8.) Kenis G, Teunissen C, De Jongh R, Bosmans E, Steinbusch H, Maes M. Stability of interleukin 6, soluble interleukin 6 receptor, interleukin 10 and CC16 in human serum. Cytokine. 2002;19(5):228-235.

(9.) Thavasu PW, Longhurst S, Joel SP, Slevin ML, Balkwill FR. Measuring cytokine levels in blood: importance of anticoagulants, processing, and storage conditions. J Immunol Methods. 1992;153(1-2):1 1 5-124.

(10.) Rodriguez-Gaspar M, Santolaria F, Jarque-Lopez A, et al. Prognostic value of cytokines in SIRS general medical patients. Cytokine. 2001;15(4):232-236.

Astrid Friebe, MD; Hans-Dieter Volk, MD

From the Institute of Medical Immunology, Charite-Universitatsmedizin Berlin, Berlin, Germany. Dr Friebe is now at the Center for Internal Medicine and Dermatology, Biomedizinisches Forschungshaus, Charite-Universitatsmedizin Berlin, Berlin, Germany.

The authors have no relevant financial interest in the products or companies described in this article.

Reprints: Astrid Friebe, MD, Center for Internal Medicine and Dermatology, Charite-Universitatsmedizin Berlin, Biomedizinisches Forschungshaus, Augustenburger Platz 1, 1 3353 Berlin, Germany (e-mail: astrid.friebe@charite.de).
Gale Copyright: Copyright 2008 Gale, Cengage Learning. All rights reserved.