Impact of Transcervical Ultrasound for the Diagnosis of Pediatric Peritonsillar Abscesses on Emergency Department Performance Measures

J Ultrasound Med. 2020 Apr;39(4):715-720. doi: 10.1002/jum.15150. Epub 2019 Nov 11.


Objectives: To evaluate the impact of transcervical ultrasound (US) as the initial imaging study for suspected peritonsillar abscesses (PTAs) on pediatric emergency department (ED) throughput measures.

Methods: A retrospective cohort study of patients evaluated for suspected PTAs between January 2009 and April 2017 was conducted. We compared the ED length of stay (LOS) before and after implementation of transcervical US to diagnose a PTA. The balancing measure was the rate of return visits within 2 weeks.

Results: There were 387 eligible patients over the study period. A total of 101 patients were evaluated for PTAs with computed tomography and 286 with US. The mean LOS was significantly less for patients who had US (347 minutes; 95% confidence interval [CI], 330, 364 minutes) compared to computed tomography (426 minutes; 95% CI, 392, 459 minutes), with an absolute difference of 79 minutes (95% CI, 44, 113 minutes). Patients who were evaluated with US did not have an increased rate of return visits (5.9% versus 8.0%; P = .66).

Conclusions: The introduction of transcervical US was associated with a decrease of greater than 1 hour in the ED LOS for patients with suspected PTAs. Given the better radiation profile of US and no increase in the rate of return visits after its implementation, we propose the adoption of a transcervical US-first approach for the diagnosis of PTAs in pediatrics.

Keywords: emergency department; length of stay; pediatrics; peritonsillar abscess; ultrasound.

MeSH terms

  • Child
  • Cohort Studies
  • Emergency Service, Hospital*
  • Female
  • Humans
  • Male
  • Palatine Tonsil / diagnostic imaging*
  • Peritonsillar Abscess / diagnostic imaging*
  • Reproducibility of Results
  • Retrospective Studies
  • Tomography, X-Ray Computed / methods
  • Ultrasonography / methods*