To the Editor.
—We are indebted to Dr Barquet and colleagues for the recent report of their vast experience of meningococcal infections and for their efforts to create a validated bedside scoring system.1 However, without further clarification it would be treacherous to interpret their data as proving that preadmission antimicrobials significantly decrease the risk of a fatal outcome.Based on the rapidity and severity of disease, the authors categorized the 624 cases of infection into fulminant, acute, and subacute subgroups. Yet they did not stratify by subgroup the relationship between preadmission antimicrobials and outcome. Although only 5% of the total number of cases of meningococcal infection were classified as "fulminant," 41% of the deaths occurred in this group (fatality rate of 44%). Contrast this to the "acute" group (71% of cases, 48% of deaths, fatality rate of 3.6%) and the "subacute" group (23% of cases, 8% of deaths, fatality