A phase II randomized, controlled trial of continuous hemofiltration in sepsis

To study the effect of early and continuous venovenous hemofiltration (CVVH) on the plasma concentrations of several humoral mediators of inflammation and subsequent organ dysfunction in septic patients.

Randomized, controlled trial.

Intensive care unit of a tertiary hospital.

Twenty-four patients with early septic shock or septic organ dysfunction.

Random allocation to receive 48 hrs of isovolemic CVVH at 2 L/hr of fluid exchange or no hemofiltration.

We measured the plasma concentrations of complement fractions C3a and C5a, interleukins 6, 8, and 10, and tumor necrosis factor alpha at baseline and 2, 24, 26, 48, and 72 hrs. A multiple organ dysfunction score (MODS) was calculated daily for each patient until death or discharge from the intensive care unit. The concentrations of most mediators decreased between baseline and 72 hrs. Some significant falls in concentration could be identified between specific time points, but CVVH was not associated with an overall reduction in any plasma cytokine concentrations. There was also no difference between the mean cumulative MODS for control survivors (43.3 +/- 19.7) and CVVH survivors (33.2 +/- 19.0; p = .30), and no difference between the average MODS calculated for all controls (4.1 +/- 1.9) and all CVVH subjects (3.3 +/- 1.7; p = .26). CVVH did not improve oxygenation, lower the platelet count, or reduce the duration of vasopressor support and mechanical ventilation.

Early use of CVVH at 2 L/hr did not reduce the circulating concentrations of several cytokines and anaphylatoxins associated with septic shock, or the organ dysfunction that followed severe sepsis. CVVH using current technology cannot be recommended as an adjunct to the treatment of septic shock unless severe acute renal failure is present.