Nursery outbreak of Pseudomonas aeruginosa: epidemiological conclusions from five different typing methods.

In April 1971, nine cases of Pseudomonas aeruginosa septicemia occurred in a high-risk nursery. The epidemiology of the outbreak was studied by pyocin production, pyocin sensitivity, serological typing, antibiotic susceptibility, and phenotypic properties such as colonial morphology, pigment, and hemolysis. Ten isolates of P. aeruginosa were recovered from 9 newborn infants and from 13 environmental sources. Twenty-one of the 23 isolates had identical pyocin production patterns against 60 different indicator strains and were of the same serotype. These 21 isolates were designated as the "outbreak strain"; the other 2 isolates had no epidemiological significance. The results of pyocin sensitivity, antibiotic susceptibility tests, and phenotypic properties were dissimilar. They would yield incorrect epidemiological conclusions if used alone. The outbreak strain dissociated in vitro and these phenotypic changes accounted for the variable results by the latter three typing methods. Although the precise mode of introduction of the organism into the nursery could not be determined in retrospect, the epidemiological data strongly suggested that one infant contracted a P. aeruginosa infection, and this strain spread throughout the nursery by means of contaminated resuscitation equipment.