Tumor necrosis factor measurement and use of different anticoagulants: possible interference in plasma samples and supernatants from endotoxin-stimulated monocytes.

OBJECTIVE

Unfractionated heparin is frequently used as an anticoagulant during blood sampling and in cell culture experiments. In the present study we investigated whether heparin and other anticoagulants (citrate and EDTA) interfered with measurements of plasma tumor necrosis factor alpha (TNF alpha) concentrations or with TNF alpha release from endotoxin-stimulated monocytes.

METHODS

TNF alpha was measured by a WEHI 164 bioassay in the plasma of 16 septic patients anticoagulated with heparin, citrate, or EDTA. Anticoagulants were incubated with the bioassay cell line and cell lysis was monitored. To exclude falsely low TNF alpha concentrations, anticoagulants were incubated in increasing amounts with human recombinant TNF alpha/saline solution, and rTNF alpha recovery was measured either with the WEHI 164 bioassay or an ELISA test. Further, anticoagulants were incubated with monocytes isolated from healthy volunteers and stimulated with endotoxin. Supernatants were analyzed for TNF alpha with both test systems.

RESULTS

No biologically active TNF alpha was detected in the plasma with heparin anticoagulation, whereas with citrate, reproducible, TNF alpha-induced cytotoxicity was detectable in blood samples of 13 of the 16 patients. Anticoagulation with EDTA resulted in fairly high, variable and poorly reproducible TNF alpha values. Only EDTA produced falsely high values by unspecific lysis of WEHI cells. Only heparin at a concentration of 20 I.U./ml or more was found to produce falsely low values by interaction with the TNF alpha bioassay, but also with the ELISA test. In monocyte culture experiments, heparin significantly attenuated the stimulatory effect of endotoxin on TNF alpha release already at the lowest concentration tested (25 I.U./ml).

CONCLUSIONS

Heparin and EDTA may have significant adverse effects on TNF alpha measurement when used for blood sampling. Citrate does not interfere with the TNF alpha bioassay or ELISA, and seems, therefore, to be the anticoagulant of choice. Due to intrinsic interactions with various cell systems (including the WEHI cell and monocytes), one should be careful in using heparin in cell culture studies in which effects of TNF alpha or of endotoxin are being studied.