The aim of our study was to assess the effect of methylene blue infusion on plasma levels of cytokines in severe sepsis. In a prospective, randomized, double-blind, placebo-controlled study, patients received either methylene blue 0.5 mg.kg-1.h-1 (MB group, n = 15) or similar volume of isotonic saline (control group, n = 15) i.v. for 6 hours. Plasma concentrations of tumour necrosis factor-alpha, interleukin-1, interleukin-2 receptor, interleukin-6, interleukin-8 were measured by sensitive immunoassays at basal (15 min before start of the study), immediately after, and at 24 and 48 hours after methylene blue infusion. We evaluated haemodynamic parameters (mean arterial pressure, heart rate), blood gases, methaemoglobin levels, and biochemical parameters at the same time. Methylene blue administration had no significant effect on plasma cytokine levels, blood gases and biochemical parameters. When compared to placebo infusion in controls, methylene blue administration resulted in significantly higher mean arterial pressure (85 +/- 14 mmHg vs 74.1 +/- 10.3 mmHg; P < 0.01), and methaemoglobin levels (1.06 +/- 0.22% vs 0.9 +/- 0.05%; P < 0.05). Furthermore, comparison with baseline levels revealed significantly increased both mean arterial pressure (85 +/- 14 mmHg and 74.1 +/- 10.2 mmHg; P < 0.05) and methaemoglobin levels (1.06 +/- 0.22% and 0.88 +/- 0.06%; P < 0.05) in MB group. There was no difference in mortality rates between the groups. We found that methylene blue infusion did not change cytokine levels or outcome in severe sepsis. The administration of methylene blue, however, resulted in a transient increase in arterial pressure. Because of the limited size of the present study, and the short period of observation, our findings need to be confirmed by larger clinical trials of methylene blue infused in a dose-titrated manner.