Bedside ultrasonography for acute appendicitis: An updated diagnostic meta-analysis

Int J Surg. 2019 Oct;70:1-9. doi: 10.1016/j.ijsu.2019.08.009. Epub 2019 Aug 9.


Background: Bedside ultrasonography is a promising tool for identification of acute appendicitis. We assessed the accuracy and clinical value of bedside ultrasonography for diagnosis of acute appendicitis in the emergency department.

Methods: Pubmed, Embase and Cochrane Library were searched from inception to November 2018. The diagnostic accuracy of bedside ultrasonography was compared with that of surgery and/or CT scan, which was used as reference standard. Pooled summary estimates of sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), and diagnostic odds ratio (DOR) from each included study were estimated using bivariate logistic regression model. Inter-study heterogeneity was examined using I2 statistic. Meta-regression was performed to further investigate the source of heterogeneity. Deeks's funnel plot was used to test publication bias.

Results: Our search yielded 5394 citations, of which 27 satisfied the inclusion criteria. Bivariate analysis yielded a mean sensitivity of 90% (95% CI 82%-0.95%) and specificity of 95% (95% CI 89%-98%). The area under the receiver operating characteristic curve was 0.97 (95% CI 0.95-0.98). There was significant inter-study heterogeneity (I2 = 96%, 95% CI 94%-99%). Meta-regression analysis suggested that study region and patient sample size could attribute to the heterogeneity. Deeks's funnel plot did not indicate the existence of publication bias (P = 0.15).

Conclusion: Bedside ultrasonography, a radiation-free and noninvasive modality, provides superior diagnostic performance in the diagnosis of acute appendicitis, but its value in different abdominal emergencies warrants further development and research.

Keywords: Appendicitis; Diagnosis; Ultrasonography; meta-Analysis.

Publication types

  • Meta-Analysis

MeSH terms

  • Acute Disease
  • Appendicitis / diagnostic imaging*
  • Humans
  • Logistic Models
  • Ultrasonography / methods*