Methicillin treatment of severe staphylococcal disease. Observations in 146 cases.

The cases were classified according to the degree to which the staphylococcal infection was authenticated and the extent to which the effects of methicillin could be evaluated. The most frequent staphylococcal infections treated were those of the lower respiratory tract. Among 63 such cases the mortality was 57%. However, when patients treated with methicillin for less than 24 hr. or in whom the staphylococcal aetiology was not established are omitted, the mortality in the remaining 40 cases was 40%; a majority of the deaths among the latter were due to super-infections with gram-negative bacilli (mostly Pseudomonas aeruginosa and Klebsiella aerobacter). There were 37 cases with staphylococcal bacteraemia, but only 6 of them were in cases of respiratory-tract infections. The fatality rate in the bacteraemic cases was 18% in the 11 in which adequate drainage or debridement of focal lesions was carried out, and 69% among the 26 in which such foci were not drained, remained unrecognized or were inaccessible. Superinfection with gram-negative bacteria was a major problem, occurring in nearly a third of the cases of staphylococcal infection of the respiratory tract that were treated with methicillin for more than 24 hr. Two thirds of the patients with such superinfections died, and in most of those patients the deaths appeared to be due to the gram-negative bacillary infection, which became manifest after a median of 5 days of methicillin treatment. Repopulation of staphylococcal wounds with similar gram-negative organisms did not generally result in serious disease. Although there appeared to be adequate evidence in these cases for the effectiveness of methicillin in the treatment of staphylococcal infections, particularly those due to penicillinase-producing staphylococci and in which other antimicrobial agents had failed, no valid conclusions could be drawn about the exact value and limitations of this agent because most of the cases treated were complex and no adequate control cases are available for comparison.