Reply

Reply: We thank Dr. Schomburg for his interest in our study on sodium selenite (1) and fully agree with him regarding the importance of selenoprotein P (Sel-P) in the process of septic shock (2, 3). As mentioned by Dr. Schomburg, there are indeed two faces to selenium (4): on the one hand, selenoproteins represent the main antioxidant defense mechanism in mammals, but small selenocompounds, particularly selenite, are well known to be dangerous oxidant poisons (4). This toxicity of small selenocompounds may, however, be of interest to transiently reduce overactivated phagocytic cells in septic shock (3, 5). Clinical studies have shown that sodium selenite administration may indeed reduce morbidity and mortality in critically ill patients, although most of these investigations were small single-center trials, and results were not consistent (6, 7). Recently, the results of two multicenter clinical trials on selenium, one using a bolus dose followed by continuous infusion (8) and the other just a continuous infusion (9), were published. The two studies reached different conclusions. Several explanations for the discrepant results can be proposed, including the different dosing strategies, the actual dose administered (10), the timing of administration, and the smaller number of patients in the study by Forceville et al. (9). In an attempt to assess further the different dosing schedules, we conducted this experimental study (1) using a well-resuscitated clinically relevant large-animal model in which bolus and continuous selenite infusion were administered 1) at the same time, 9 h after the initiation of peritonitis, and 2) using a similar dose of 2 mg. Importantly, this is a relatively large dose when considered in relation to the total body selenium in a 30-kg sheep (∼4 - 8 mg), representing a therapeutic intervention and not just "supplementation." As mentioned by Dr. Schomburg, we reached oxidant selenite concentrations in plasma with the bolus administration but not with continuous administration. Although not explicitly excluded because we did not measure Sel-P concentrations, it seems unlikely that the differences in outcomes in our groups were related to Sel-P induction because 1) serum IL-6 concentrations were significantly lower in the bolus injection group than in the continuous infusion group at T10 (just 1 h after selenite administration)-such an immediate effect seems unlikely to be due to Sel-P given the complex mechanism by which selenium is incorporated into selenoproteins (11); 2) moreover, if the effect was related to Sel-P induction, it should have been seen in both groups because animals receiving the continuous infusion received a large-enough amount of selenium for Sel-P synthesis. In addition, Sel-P seems to be a negative acute-phase protein in sepsis, as has been nicely demonstrated in mice (12). Nevertheless, as we acknowledged (1), our study does not allow a thorough exploration of the underlying mechanisms, for which plasma Sel-P concentration measurements would be necessary. Recently, an alternative potential process has been proposed, which may support the oxidant mechanism. The ADMA/DDAH system is an important intracellular regulator of NOS activity in which ADMA is metabolized by DDAH enzyme into citrulline and dimethylamine (13). Inhibition of DDAH enzyme activity may be of therapeutic significance in the treatment of septic shock because excessive iNOS activity and NO overproduction play a central role in sepsis-induced vasodilatation and microcirculatory dysfunction. Intriguingly, oxidant compounds have been observed to abrogate DDAH activity because a crucial cysteine residue held in its active center is susceptible to oxidation (13). A further study is underway to investigate whether DDAH inhibition plays a part in the observed beneficial effects elicited by the oxidant doses of sodium selenite in our earlier study (1). As stated by Dr. Schomburg, selenocompound administration represents a promising adjuvant therapeutic option in septic shock, but clearly, further studies are necessary to fully elucidate the mechanisms of action of these agents in such patients and to determine optimal dosing strategies. In the case of selenocompounds such as sodium selenite, especially if administered as a bolus, the transient oxidative action may be rapidly followed by a beneficial induction of Sel-P and other antioxidant selenoenzymes, reducing reactive oxygen species damage, limiting the toxicity of selenocompounds such as selenite, and increasing immunity. Zhen Wang Jean-Louis Vincent Erasme Hospital, Belgium Xavier Forceville Hopital Saint Faron, France