Donald Le Ly MD, BEng, Presenter: Nothing to Disclose

Seng Thipphavong MD, Abstract Co-Author: Nothing to Disclose

Sara H. Gray MD, FRCPC, Abstract Co-Author: Nothing to Disclose

Anthony Edward Hanbidge MBBCh, Abstract Co-Author: Nothing to Disclose

Mostafa Atri MD, Abstract Co-Author: Nothing to Disclose

Korosh Khalili MD, Abstract Co-Author: Nothing to Disclose

To determine the attitudes of emergency department (ED) physicians (MDs) towards non-visualization of the appendix (NVA) on ultrasound (US) scans for RLQ pain and to assess ultrasounds’ diagnostic performance.

A survey was administered to 166 ED MDs at several academic hospitals to determine their interpretation and practice after receiving an ultrasound report with NVA. Retrospective review of three large academic EDs revealed 1672 US scans performed for appendicitis in 2012. 291 (17.4%) explicitly indicated NVA and underwent a chart review of US findings, follow-up imaging, and surgical findings to determine the negative predictive value of NVA and utility of secondary findings. Univariate analysis was performed to determine which secondary signs of appendicitis were significant (P<0.05) predictors of a positive CT scan.

95/166 (57%) of ED MDs completed the survey; 53% had >10 years experience. After receiving an US report with NVA, 92% indicated re-examining the patient; 79% felt further imaging was required and 61% would proceed with CT. Only 18/95 (19%) of ED MD believed that follow up CT is positive for appendicitis in less than 10% of NVA; 71% agreed that a level of certainty scale would be helpful on radiology reports for appendicitis. Of 291 US scans with NVA, 246 (85%) were female (mean age 31y). In 229/291 (79%), no alternate diagnosis was found and a CT was done in 94/229 (41%). Appendicitis was found in 13/229 (5.7%) patients with NVA and no alternate diagnosis; in 9/13 secondary signs of appendicitis were noted. Therefore negative predictive value (NPV) for NVA was 216/229 (94.3%). Inflammatory changes in RLQ (p=0.01) and focal tenderness (p=0.02) noted on US were significant predictors of a positive CT scan.

Current perceptions and practice of some ED physicians equate NVA on US as an inadequate study to exclude appendicitis. However, NVA is itself a highly predictive sign (94.3%) of absence of appendicitis when an alternate cause of pain is not seen.

Dissemination of the high NPV of ultrasound in RLQ pain among ED MDs may lead to diminished resource utilization. Radiologists can add value to ED MDs by providing an evidence-based level of certainty scale for US findings when assessing for appendicitis. 

Ly, D, Thipphavong, S, Gray, S, Hanbidge, A, Atri, M, Khalili, K, When the Appendix is not Seen on Ultrasound for RLQ Pain: Does the Interpretation of Emergency Department Physicians Correlate with Diagnostic Performance?.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14045567.html