Incorrect Wording and Omissions

In the Caring for the Critically Ill Patient article titled "High-Dose Antithrombin III in Severe Sepsis: A Randomized Controlled Trial" published in the October 17, 2001, issue of THE JOURNAL (2001;286:1869-1878), several errors were printed. On page 1874 in the first column, second paragraph, the sentence should read "Antithrombin III resulted in a 15% relative improvement in 90-day mortality," not "absolute improvement." On page 1875, the title of Table 3 should read "Incidence of New Organ Dysfunction According to Logistic Organ Dysfunction Score. . . . " Also on page 1875, the legend for Figure 5 should indicate that the patients represented in the graph are the percentage discharged alive from ICU in relation to all patients (surviving and dead). On page 1876, the footnote to Table 4 was inadvertently omitted. It should have read "Number of patients actually treated with antithrombin III or placebo." This footnote clarifies the different patient numbers for the primary efficacy analysis (n = 1157 per group) vs the safety analysis (placebo: n = 1155; antithrombin III: n = 1161). On page 1876, the top paragraph in the third column should end with this sentence: "Up to and beyond the second interim analysis, an overall nominal survival benefit for antithrombin III prevailed, despite the recognized increased bleeding risk in that group."