Pediatric lateral neck infections - Computed tomography vs ultrasound on initial evaluation

Review the evaluation of children with a deep lateral neck infection and define the impact of initial imaging modality on outcomes and costs.

Case series, pediatric patients <18 years of age admitted to a tertiary care hospital with lateral neck infection between 01/01/14-05/31/16 as identified by ICD-9 and ICD-10 codes: 289.3 (lymphadenitis, unspecified), 682.1 (cellulitis and abscess of neck), 683 (acute lymphadenitis), I88.9 (nonspecific lymphadenitis, unspecified), L02.11 (cutaneous abscess of neck), L03.221 (cellulitis of neck), and L03.222 (acute lymphangitis of neck). Patients were divided into two groups based on initial imaging modality: primary ultrasound or primary computed tomography. Differences in length of stay, type and total number of imaging studies obtained, number of procedures, hospital readmission, and hospital cost were compared between cohorts.

There were 40 (31%) primary ultrasound and 88 (69%) primary computed tomography patients (128 total). Median length of stay was 46 (IQR: 25,90) hours (1.9 days) for primary ultrasound and 63 (IQR: 39,88) hours (2.6 days) for primary computed tomography patients (p = 0.33). Drainage was performed in 48% of both groups. Additional imaging occurred in 17 (43%) primary ultrasound and 18 (20%) primary computed tomography patients (p = 0.02). Readmission occurred in 8 patients (6.3%). Retropharyngeal infection was encountered in 13 patients (10%); this was only discovered in patients who had a computed tomography performed. Median cost per primary ultrasound patients was $5363 (IQR: 3011, 7920) and $5992 (IQR: 3450, 8060) for primary computed tomography patients.

The primary imaging modality (ultrasound or computed tomography) used to work-up children with a lateral neck infection did not impact length of stay or hospital cost. However, a significant subset had a coexisting retropharyngeal infection that was only identified on computed tomography. Future studies are needed to identify appropriate criteria for imaging in the work-up of lateral neck infections.