Abstract
Background: Superficial neck infections including lymphadenitis and abscesses are commonly encountered in pediatric emergency departments (PEDs). It is often unclear which patients are likely to develop an abscess that necessitates surgical drainage. In evaluating these patients, computed tomography (CT) and ultrasound are often used to identify/confirm abscess formation. The criteria for determining the need for imaging studies are not well defined.
Design/Methods: All visits to the study PED were examined in 2009 to 2010. Visits with the diagnosis of cervical lymphadenitis or abscess were identified. Records were retrospectively reviewed to determine the duration of symptoms, fever, previous antibiotic therapy, prior PED visit, size of neck swelling, fluctuance on physical examination, white blood cell count, and results of CT and/or ultrasound obtained in the PED. Data were analyzed to determine which of these characteristics were more likely to be associated with an abscess that was operatively drained.
Results: A total of 768 patients were evaluated for neck infections. One hundred twelve (14%) of these pediatric patients underwent abscess drainage in the operating room. Two hundred eighty-nine patients underwent a neck CT and/or ultrasound, of which 119 were positive for abscess. Factors associated with surgical drainage included fluctuance (odds ratio [OR], 18.92; 95% confidence interval [CI], 3.66-31.37), previous emergency department visit (OR, 2.79; 95% CI, 1.34-5.84), and age less than 4 years (OR, 3.01; 95% CI, 1.15-9.87). A recursive partitioning model stratified patients' risk for going to the operating room. Patients without fluctuance and with no prior emergency department visit, along with no prior antibiotic use, have less than 4% chance of having an abscess that necessitates surgical drainage.
Conclusions: Pediatric patients who are more likely to have a neck infection that necessitates surgical drainage can be stratified based on clinical characteristics. This knowledge may allow physicians to better predict the resource needs including hospital admission and emergent imaging for neck infection. (C) 2013 Elsevier Inc. All rights reserved.