The role of the lumbar puncture in the admission sepsis evaluation of the premature infant.

Premature infants with respiratory distress commonly receive evaluation for sepsis, including a lumbar puncture, within a short time after admission to a neonatal intensive care unit. We questioned the use of the lumbar puncture during the early sepsis evaluation, and since 1979, have omitted this procedure as part of the initial evaluation for sepsis (within 6 hours of birth) of premature infants. We monitored this policy to detect any change in the incidence of meningitis, and now report results accumulated over a 7-year period. From 1979 to 1986, 1390 inborn premature infants of 34 weeks' gestation or less were evaluated for early sepsis within 6 hours of birth. Thirty-two infants (2.3%) were diagnosed with sepsis. Fifteen of these infants died in the first 24 hours of life. Meningitis was not demonstrated by autopsy evaluation. The surviving 17 infants diagnosed with sepsis did not have meningitis. One hundred twenty-three infants whose initial blood cultures were negative developed infection later in their hospital course. Eleven of these 123 infants had infections with perinatally acquired organisms; two had group B streptococcus (GBS) meningitis. Their cases were not compatible with missed meningitis. The remaining 112 infants developed nosocomial infections of which 38.3% developed meningitis without associated bacteremia. These results suggest that the omission of the lumbar puncture in the early sepsis evaluation of the premature infant did not result in any missed meningitis and spared many infants the procedure shortly after birth. The lumbar puncture, however, continues to be vital in the assessment of late infections of the neonate.