Etiology of Fever in Young Infants: A Reassessment

Abstract

Source: Greenhow TL, Hung Y-Y, Herz AM. Changing epidemiology of bacteremia in infants aged 1 week to 3 months. Pediatrics. 2012; 129(3): e590– e596; doi: 10.1542/peds.2011-1546Investigators from Kaiser Permanente Northern California characterized the etiology of bacteremia in infants aged 1 week to 3 months by retrospectively reviewing the results of blood cultures (BC) obtained from a cohort of 160,818 full-term, previously healthy infants during a 5-year period (2005–2009). Cultures drawn from outpatients, in the emergency department, and within the first 24 hours of admission to the hospital were included. Excluded were infants with underlying medical problems and those born at <37 weeks estimated gestational age. Efforts were made to identify repeatedly positive cultures, which were counted as a single episode of bacteremia. Isolates such as coagulase-negative staphylococci, Micrococcus, and diphtheroids were defined as contaminants (3 children with viridans streptococci and 1 of enterococcal bacteremia were managed as infections).During the 5-year period 340 of 4,255 (8%) BCs were positive. Of these, 93 (2%) were clinically significant and 247 (6%) were considered contaminated. The incidence of true bacteremia was 0.57 per 1,000 live births in this cohort. Escherichia coli was the most common pathogen (56% of cases of bacteremia), followed by group B Streptococcus (GBS) (21%), Staphylococcus aureus (8%), and Streptococcus pneumoniae (3%). There were 3 cases of viridans streptococcal bacteremia, 2 cases each of Klebsiella and Salmonella bacteremia, and single cases of bacteremia due to Citrobacter, Enterococcus faecalis, Moraxella, and Streptococcus pyogenes. Of the 19 episodes of GBS bacteremia, 12 occurred between days 7 and 28, 6 in the second month of life, and 1 in the third month of life. For all other pathogens, there was no age-related pattern.A significantly higher proportion of contaminated BCs were drawn in the emergency department, compared to the outpatient or inpatient setting. Of the 48 infants with E coli bacteremia who had a urine culture (UC) performed, 47 (98%) had a concomitant urinary tract infection (UTI). Of the 38 such infants for whom a cerebrospinal fluid (CSF) culture was performed, 4 (11%) had concomitant meningitis. Of 17 infants with GBS bacteremia who had CSF culture performed, 5 (29%) had concomitant meningitis. There was 1 case of pneumococcal meningitis. Compared to nonbacteremic controls, bacteremic infants had a significantly higher average white blood cell count (13,820/μL vs 10,770/μL). Three of 52 (6%) of the E coli isolates were resistant to ampicillin and gentamicin.The authors conclude that empiric ampicillin treatment in the first 3 months of life may no longer be required, given the absence of Listeria infection and the observed antibiotic resistance in their cohort.Dr Tolan has disclosed no financial relationship relevant to this commentary. Dr. Tolan is on the Speaker’s Bureau for Novartis. This commentary does not contain a discussion of an unapproved/investigative use of a commercial product/device.This is the first study to address the incidence of bacteremia in infants 1 to 3 months of age since the advent of widespread pneumococcal conjugate vaccination. The most important finding is the prominence of UTI among serious bacterial infections (SBI) in this age group.1,2 Despite the striking occurrence of concomitant E coli UTI and bacteremia (98% when the urine was sampled), 4 infants with E coli bacteremia did not have UC performed. In fact, 11 (12%) of all the bacteremic infants did not have UC performed. Similarly, 32 (34.8%) of infants sick enough to have BC performed (and subsequently found to be bacteremic) did not have CSF cultures performed.The authors note the absence of Listeria infections in the study cohort, and suggest that empiric therapy for this organism may no longer be necessary. However, the rate of Listeria infection in the United States in 2009 was 0.05 cases per 100,000 persons aged ≤1 year.3 Thus, one would expect only 0.1 cases in their cohort over a 5-year span. Since the investigators excluded premature infants and those under 1 week of age, newborns with Listeria infections may have been missed.4 The results of this study, therefore, do not support a recommendation to discontinue empiric ampicillin therapy in the first months of life. However, assuring adequate treatment of E coli (including antibiotic-resistant strains) is of paramount importance in the management of SBI in this age group.The results described are likely generalizable to similar infants, but caution should be exercised when confronted with a febrile or ill-appearing infant born prematurely or having underlying medical conditions, for whom these data may be less applicable. In any event, the importance of UTI among infants with SBI in the first months of life is undeniable.