ObjectiveTo determine whether a continuous intravenous infusion of pentoxifylline, a methylxanthine derivative, alters the serum cytokine concentrations and/or hemodynamic measurements in patients with septic shock.DesignA prospective, randomized, double-blind, placebocontrolled study.SettingMedical intensive care unit in a university hospital.PatientsSixteen patients with septic shock.InterventionsPatients were randomly assigned to receive either pentoxifylline (1 mg/kg) followed by an infusion of 1.5 mg/kg/hr for 24 hrs (n equals 8), or placebo (n equals 8).Measurements and Main ResultsTumor necrosis factor (TNF) and interleukin (IL)-6 concentrations were measured by radioimmunoassays; IL-8 concentrations by an enzyme-linked immunosorbent assay (ELISA) and pentoxifylline concentrations by high-performance liquid chromatography at 0, 3, 6, 12, 18, 24 and 48 hrs after study entry. Pulmonary artery catheter-derived hemodynamics were measured at 0, 0.75, 3, 6, 12, 18, and 24 hrs. In pentoxifylline-treated patients, at 24 hrs, serum concentrations of TNF were significantly lower compared with controls (12 plus minus 2 vs. 42 plus minus 12 pg/mL, respectively, p equals .04). Serum concentrations of IL-6 and IL-8 did not differ between the two treatment groups. There were also no significant differences in any hemodynamic and oxygenation measurements comparing the two treatment groups. Pentoxifylline concentrations were 1544 plus minus 241 ng/mL after the initial dose, and 5776 plus minus 1781 ng/mL at the end of the 24-hr infusion. Five patients in the pentoxifylline group and four patients in the placebo group died.ConclusionsPentoxifylline is able to decrease serum TNF but not IL-6 or IL-8 serum concentrations during septic shock. Pentoxifylline was well tolerated by all eight patients with no adverse effect. Further studies are needed to determine if pentoxifylline's ability to lower circulating TNF concentration without altering hemodynamics will improve outcome in septic shock.(Crit Care Med 1996; 24:207-214)