Meningococcal septic shock has a rapid onset and characteristic skin hemorrhages that allow bedside diagnosis. Initial plasma endotoxin levels are high and correlate closely with clinical outcome. In a double-blind, randomized, placebo-controlled trial (planned, n = 270; actual, n = 269), we compared the effectiveness of HA-1A (6 mg/kg of body weight iv; maximum, 100 mg), a human monoclonal antibody to endotoxin, and placebo in reducing the 28-day all-cause mortality rate among children with a presumptive clinical diagnosis of meningococcal septic shock. Treatment groups were well balanced for baseline characteristics and prespecified prognostic variables. In this trial no significant benefit of HA-1A could be demonstrated. The 28-day mortality rates in the intention-to-treat analysis were as follows: placebo, 28%; HA-1A, 18%; reduction in mortality, 33% (P = .11, per Fisher's exact test, two-tailed; odds ratio = 0.59; 95% confidence interval for the difference, 0.31-1.05). All patients tolerated HA-1A well, and no antibodies to HA-1A were detected.