Selenium replacement in patients with severe systemic inflammatory response syndrome improves clinical outcome

OBJECTIVE: To determine the effect of selenium replacement on morbidity and mortality in patients with systemic inflammatory response syndrome (SIRS). DESIGN: Controlled, randomized prospective open-label pilot study comparing patients with and without selenium replacement. SETTING: Intensive care unit of a university hospital for internal medicine. PATIENTS: Forty-two patients with SIRS caused by infection and a minimal Acute Physiology and Chronic Health Evaluation (APACHE) II score of 15 points on the day of admission were included. The selenium replacement group of patients (Se+; n = 21) received sodium selenite for 9 days (535 microg 6.77 micromol for 3 days, 285 microg 3.61 micromol for 3 days, and 155 microg 1.96 micromol for 3 days) and thereafter, 35 microg (0.44 micromol) per day iv. The control group (Se-, n = 21) received 35 microg of sodium selenite throughout the total treatment period. INTERVENTIONS: Morbidity and clinical outcome was monitored by scoring using the APACHE III score, occurrence of acute renal failure, need and length of mechanical ventilation, and hospital mortality. Blood samples on days 0, 3, 7, and 14 were analyzed for serum selenium concentration and glutathione peroxidase (GSH-Px) activity. MEASUREMENTS AND MAIN RESULTS: The median APACHE II score at admission, age, gender, underlying diseases, serum selenium levels, and GSH-Px activities at admission were identical in both groups. In Se+ patients, serum selenium levels and GSH-Px activity normalized within 3 days, whereas in controls, both variables remained significantly low (p .05; day 7, p = .018; and day 14, p = .045 Se+ compared with Se-). Hemodialysis with continuous veno-venous hemodialysis because of acute renal failure was necessary in nine Se- compared with three Se+ patients (p = .035). Overall mortality in the Se- group was 52% vs. 33.5% in the Se+ group (p = .13). CONCLUSIONS: Selenium replacement in patients with SIRS seems to improve clinical outcome and to reduce the incidence of acute renal failure requiring hemodialysis.