To the Editor .— In their article about the timing of lumbar puncture (LP) in bacterial meningitis, Kanegaye et al1 recommend delaying administration of antibiotics until cultures have been taken. We would like to point out that where meningococcal meningitis is suspected, this potentially life-threatening delay may be neither appropriate nor necessary. In the United Kingdom, patients suspected of having invasive meningococcal disease are investigated using polymerase chain reaction (PCR) assay of plasma and, where available, cerebrospinal fluid (CSF). This service is provided nationally by the Public Health Laboratory Service (PHLS) Meningococcal Reference Unit (MRU) in Manchester.2 The use of PCR by the PHLS has markedly increased the rate of microbiological confirmation in suspected cases of meningococcal disease. In a review of 103 such cases, 89% were found to be positive by PCR assay, compared with only 44% by more traditional tests such as culture and microscopy.3 It is important to note that PCR detected meningococci even after antibiotics had been administered. Of samples tested >12 hours after initiation of antibiotics, 10 (59%) of 17 were still positive by PCR, whereas only 3 (14%) of 21 samples were positive by culture or microscopy. For obvious methodologic and ethical reasons, it is difficult to demonstrate rigorously that early treatment improves the outcome in meningococcal disease. However, common sense dictates that treatment should not be delayed unnecessarily. Where postantibiotic diagnosis is available, treatment can be commenced even before hospital admission. This is reflected in practice in the United Kingdom, where family doctors administer intramuscular benzylpenicillin immediately upon suspicion of meningococcal disease and before sending the patient to hospital. For the hospital doctor, there is less pressure to perform an early LP, with the possible complications that might ensue.4 The current view of many pediatricians in the United Kingdom …