Selenium pharmaconutrition in sepsis: To give or not to give? Is this still the question?

Abstract

The Surviving Sepsis Campaign (SSC) International Guidelines for Management of Severe Sepsis and Septic Shock [1] were recently updated to assist critical care practitioners in managing the proliferation of emerging evidence for the treatment of sepsis syndromes and improving the prognosis of sepsis. Regarding pharmaconutrition with high-dose intravenous (IV) selenium, we find it surprising that these new SSC guidelines suggest, “not using intravenous selenium for the treatment of severe sepsis (grade 2C)” [1]. We respectfully disagree with this recommendation and question the rationale for this conclusion. In our opinion, the SSC may not have considered much of the published clinical evidence. In fairness, the author’s recommendation does not exclude the use of low-dose selenium in parenteral nutrition (PN) but no reference is made to the American Society for Parenteral and Enteral Nutrition (ASPEN) position paper on micronutrients [2] indicating that, although controversial, “patients who are deficient in selenium or who are critically ill, septic, or have severe burns may benefit from short termvery high daily doses [of selenium] administered separately from the PN” [2]. In fact, several observational studies have demonstrated that: low selenium status is associated with severity of illness; septic shock patients exhibit the lowest plasma selenium levels; and serum selenium shows a relatively good predictive value for mortality in the intensive care unit (ICU) [3–5]. Therefore, it seems logical that selenium supplementation is not only able to correct a nutritional deficiency but can help to improve outcome for septic patients. One of the most relevant selenium pharmaconutrition trials, quoted by the SSC was the SIC (Selenium in Intensive Care) study [6]. The SIC protocol incorporated an initial selenite loading dose by IV bolus to optimize selenium levels and extracellular glutathione peroxidase activity, followed by continuous infusion for 14 d. Although the reduction in 28-d mortality for supplemented patients failed to reach statistical significance in the intentionto-treat analysis, mortality was significantly lower in the perprotocol group (56.7% versus 42.6%; P 1⁄4 0.049). Additionally, mortality was significantly reduced in those septic patients with