Sir.—It is hard for me to discern exactly what hypothesis Fischer et al were testing in their recent article, "Lumbar Punctures and Meningitis" (129:590-592, 1975). In their cases 1 through 3 there is no evidence that the lumbar puncture itself contributed in any way to the subsequent manifestation of meningitis in these bacteremic children. It is reasonable to expect that certain microorganisms, after seeding the bloodstream, may infect the meninges. In newborns, essentially every organism— Escherichia coli with capsular antigen Kl, especially1—possesses this virulence. It is unreasonable to ascribe this to the lumbar puncture. Their fourth case, in which pus was found near the site of the puncture, is more suggestive of some iatrogenic involvement in the presence of viremia.
That a second lumbar puncture ought to be performed in the presence of a positive blood culture and deteriorating clinical condition is meritorious. That the lumbar