To the Editor.
—The suggestion that the E5 monoclonal antibody will be useful in reducing fatalities in gram-negative sepsis based on the article by Greenman et al1 and the accompanying editorial2 is less well supported by the data than is implied.The only significant reduction in fatality rates occurred in a single subset (those with gram-negative infections without shock) out of the several subsets examined. Of the subsets examined, this subset is least likely to have serious gram-negative bacillary infections. If clinical effectiveness of antibodies to endotoxin is to be critically assessed, those patients with known endotoxemia (ie, proven gram-negative bacteremia) provide the most valid indicator of effectiveness. Those patients with gram-negative infections severe enough to produce shock provide the second most valid indicator. No effect was demonstrated in either of these subsets.Past experience indicates that the enrollment criteria used in the study by Greenman et al
McCabe WR. Antibody to Endotoxin in the Treatment of Gram-negative Sepsis. JAMA. 1992;267(17):2325. doi:10.1001/jama.1992.03480170051015