Cytokines clearance during venovenous hemofiltration in the trauma patient.

The objective of the study was to investigate whether continuous venovenous hemofiltration (CVVH) would facilitate removal of substantial amounts of tumor necrosis factor (TNF) and interleukin-6 (IL-6) from the circulation in traumatized critically ill patients with multiple organ dysfunction syndrome. The study design was a prospective, nonblind, randomized controlled trial that was set in the trauma intensive care unit of a tertiary university referral hospital. Thirty consecutive critically ill, mechanically ventilated trauma patients with multiple organ dysfunction syndrome (without renal failure) were included in the study. Patients were randomized to either CVVH or conventional treatment. Blood and ultrafiltrate samples were collected from each patient before the initiation of CVVH and after 24, 72, and 168 hours of therapy. In the control group, blood samples were collected during the same periods. In the 30 patients studied, 15 had hemofiltration and 15 did not. Both groups were similar with regard to age (36+/-18 years v 36+/-14 years) and severity scores (injury severity score, 32+/-16 v 30+/-11; APACHE II score, 22+/-7 v 21+/-6; Goris score, 5.2+/-1.7 v 5.2+/-1.8). Before CVVH, TNF and IL-6 could be detected in the serum of all patients. The mean concentration of TNF was 17+/-22 pg/mL in patients and 22+/-20 pg/mL in control subjects (P = NS). The mean concentration of IL-6 was 2,153+/-2,824 pg/mL in patients and 1,774+/-1,637 pg/mL in control subjects (P = NS). We found a TNF and IL-6 substantial elimination with CVVH (excretion of TNF [microg/d] at 24, 48, and 168 hours: 112.6+/-161.2, 105.2+/-149.4, and 143.1+/-170.0; excretion of IL-6 [microg/d]: 1,655+/-719, 3,091+/-489, and 2,420+/-366). However, no significant difference was found in serum cytokines concentration between groups during the study: mean serum TNF concentration decreased from the pretreatment level to a mean level of 12+/-9.6 pg/mL in patients and 21+/-27 pg/mL in control subjects. Similar results were found with IL-6 concentration that decreased from the pretreatment level to a mean of 554+/-731 pg/mL in patients and 382 +/-568 pg/mL in control subjects. In conclusion, CVVH is associated with removal of substantial amounts of TNF and IL-6 from the circulation in traumatized critically ill patients, but the profile of these mediators is similar to that of controls, suggesting a nonclinically relevant elimination. Further prospective, randomized, clinical trials are needed to support our results.