Improvement in diagnostic accuracy of ultrasound of the pediatric appendix through the use of equivocal interpretive categories.

OBJECTIVE

The purpose of this article is to evaluate the diagnostic performance of ultrasound of the pediatric appendix using standardized structured reports that incorporate equivocal interpretive categories.

MATERIALS AND

METHODS

Standardized structured appendix ultrasound reports using a five-category interpretive scheme were reviewed. Interpretive categories were positive, intermediate likelihood, or negative when the appendix was visualized, and secondary signs or no secondary signs when the appendix was not visualized. Interpretations were compared with clinical and pathologic follow-up. Diagnostic accuracy was compared with the accuracy of a simulated binary interpretive scheme based on the same data.

RESULTS

One thousand three hundred fifty-seven examinations were included, with appendicitis present in 16.9% (230/1357) of cases. The appendix was visualized in 47.2% (641/1357) of cases, with interpretations as follows: positive, 27.5% (176/641); intermediate likelihood, 9.7% (62/641); and normal, 62.9% (403/641). The appendicitis rate in each group was 92.6% (163/176), 25.8% (16/62), and 0.5% (2/403), respectively. The appendix was not visualized in 52.8% (716/1357) of cases, with secondary findings identified in 8.5% (61/716) and no secondary findings in 91.5% (655/716) of cases. The appendicitis rate was 39.3% (24/61) and 3.8% (25/655) in these groups, respectively. Appendicitis was present in 32.5% of equivocal (intermediate likelihood and not visualized, secondary findings) cases and 2.6% of negative (normal and not visualized, no secondary findings) cases. Diagnostic accuracy of a five-category scheme was 96.8% versus 94.1% for a binary scheme.

CONCLUSION

Appendix ultrasound examinations interpreted according to a scheme that incorporates equivocal categories better convey diagnostic certainty and increase diagnostic accuracy compared with a binary interpretive scheme.