Systemic Infection Risk in Infants With Focal Bacterial Infections

Abstract

Source: Vidwan G, Geis GL. Evaluation, management, and outcome of focal bacterial infections (FBIs) in nontoxic infants under two months of age. J Hosp Med. 2010; 5(2): 76– 82; doi: 10.1002/jhm.583To determine the risk of concomitant systemic infection (CSI) in infants who present with a focal bacterial infection (FBI), investigators from Joe DiMaggio Children’s Hospital in Hollywood, FL, and Cincinnati Children’s Hospital Medical Center (CCHMC) retrospectively reviewed charts from a consecutive series of infants under two months of age who presented to the CCHMC emergency department between January 2000 and December 2005. Participants were included if they were well-appearing on examination; had normal heart rate, respiratory rate, blood pressure, and oxygen saturation for age; and received a discharge diagnosis of FBI. FBIs were pre-defined as soft tissue infection, cellulitis, mastoiditis, abscess, otitis media (OM), omphalitis, mastitis, mammitis, paronychia, balanitis, posthitis, impetigo, or lymphadenitis. CSIs were defined as bacteremia, urinary tract infection (UTI), meningitis, septic arthritis, osteomyelitis, or pneumonia. Patients with immunodeficiency, a central venous catheter, tracheostomy, gastrostomy tube, chronic lung disease, previous admission for bacterial infection, or who were taking systemic antibiotics were excluded. The authors also sought to determine the risk of CSI in relation to the presence of fever, recorded at home or in the emergency room.Two hundred forty-six patients under 60 days of age were diagnosed with a FBI. Thirty-seven of these met exclusion criteria and charts from 12 others could not be found. Of the 197 remaining participants, 158 were afebrile and 39 had fever. The mean age was 29.6 days (40.7 days for the febrile and 26.8 days for the afebrile). Four FBI patients (3 febrile, 1 afebrile) had a documented CSI for an overall risk of 2.0%. Of the three febrile patients, two had Escherichia coli UTIs and one had Streptococcus pneumoniae bacteremia accompanying periorbital cellulitis. The only afebrile infant with a CSI was an 11-day-old male who presented with OM and had an E coli UTI. An afebrile 11-day-old female infant with mastitis and a normal urinalysis had both E coli and normal flora on urine culture that was considered contaminated by the primary team. If this were considered a UTI, it would increase the overall risk of CSI to 2.5% among infants presenting with FBI and 1.3% among those who were afebrile. Febrile infants had a statistically higher risk of CSI than afebrile infants (OR=13.08; 95% CI, 1.32–129.46). The authors conclude that the risk of CSI in well-appearing afebrile infants with a FBI is small, but that febrile infants with a FBI have a risk of CSI similar to young infants presenting with fever without a source.There are several guidelines for the management of a well-appearing, febrile infant younger than 2 months of age who is without a source of fever.1 More recently the literature has assessed the risk of CSI in young infants who have a clinically apparent or laboratory-confirmed viral infection, where the risk of a CSI was found to be much lower than in those without a source of fever.2,3 Several studies have evaluated how the presence of OM affects the risk of CSI, but no studies have looked at the general category of FBI and the risk of CSI in infants with or without fever. Primary care and emergency room doctors sometimes admit afebrile well-appearing young infants with an FBI to the hospital out of concern for a CSI. Although still infrequent, such hospitalizations may be more common with the increase of community-acquired methicillin-resistant Staphylococcus aureus skin and soft tissue infections. Management strategies span the gamut from outpatient oral antibiotics to a full sepsis work-up with admission and intravenous antibiotics.This study, a retrospective analysis, is limited by the accuracy of the physicians’ diagnostic coding. A physician may neglect to code for OM if a more serious CSI is diagnosed during the visit. The authors also had to assume that infants who developed CSIs that were not tested for at the index visit would be identified when they returned to the CCHMC emergency department. This assumption may not be completely valid, despite the reassuring fact that it is the sole admitting pediatric center in the area. Finally, although only 12 charts (6%) could not be found, if one or two of those infants had a CSI it could greatly affect the results. These are important limitations; nonetheless this retrospective study from a single institution supports managing well-appearing, afebrile infants with FBI with oral antibiotic therapy and careful follow-up but not a sepsis work-up or hospitalization.