A VLBW Infant with Suspected Sepsis: To Tap or Not To Tap—That is the Question

Abstract

Source: Stoll BJ, Hansen N, Fanaroff AA, et al. To tap or not to tap: high likelihood of meningitis without sepsis among very low birth weight infants. Pediatrics. 2004;113:1181–1186.The National Institute of Child Health and Human Development (NICHD) Neonatal Research Network of major tertiary care academic centers collaborated to carry out a multicenter study of the epidemiology of lateonset meningitis (LOM) in very low birth weight (VLBW) infants and evaluate the concordance of cerebrospinal fluid (CSF) and blood culture (BC) results. LOM was defined as CSF culture positive for bacteria or fungi obtained after 72 hours of age. BCs taken within 7 days of each positive CSF culture were used to classify positive CSF cultures as indicative of: 1) meningitis without sepsis, 2) meningitis with sepsis, and 3) probable contaminant. Of 11,028 infants born in the participating centers between September 1, 1998 and December 31, 2001, 9,828 survived at least 72 hours. Infants with CSF shunts (187) were excluded, leaving 9,641 infants for analysis. Of the 6,056 infants who had at least 1 BC performed after 72 hours of life, only 2,877 had a lumbar puncture (LP). Of infants who had an LP, 98% had at least one BC. Of the 228 infants with a positive CSF culture, 41% (94) were considered contaminants by the study site and not treated. The remaining 134 infants with LOM constituted 5% of those who had a CSF culture (2,877) and 1.4% of all infants (9,641). One-third of the infants with LOM (45/134) did not have a positive BC.An increased risk of meningitis was significantly associated with lower gestational age and with a prior episode of sepsis, but not with prior intraventricular hemorrhage, periventricular leukomalacia, or with an infant’s birthweight, gender, or race. Infants with LOM were 15 times more likely to have seizures (OR: 14.8), spend more time on ventilators (28±1.9 versus 18±1 days), take longer to attain full feeds (31±1.5 versus 21±1.8 days), and have longer hospital stays (91±3.9 versus 79±2.1 days) than those without LOM (P<.001 P for each). There was no increased risk of death in infants who had an LP. The authors conclude that LOM may be underdiagnosed in VLBW infants.The necessity of performing an LP in the VLBW newborn has been debated because of the perceived risks of the procedure in very sick infants.1,2 At least for now, this study should settle the argument. It not only shows that an LP is safe, but that not doing one may be detrimental to the health of the newborn. The fact that in a large percentage of infants who had an initial positive CSF culture, a repeat LP still yielded the same organism as the first LP is important information. If the LP had not been done, the meningitis might have gone unrecognized. The diagnosis of meningitis may also be of prognostic significance affecting key events in the life of a sick neonate (eg, weaning from mechanical ventilation and hospital discharge). Merely obtaining a blood culture would miss meningitis in a number of infants. Conversely, treating infants for meningitis if they have a positive BC and not performing an LP would result in gross overuse of antimicrobials, a practice to be condemned in view of increasing recognition of antimicrobial resistance.This article suggests that there are few reasons (eg, severe thrombocytopenia) for not performing an LP in a newborn being evaluated for sepsis. We must return to one of the basic tenets of pediatric care—if you think of meningitis, perform an LP.