Sonohysterography: The utility of diagnostic criteria sets

Mousumi Bhaduri, Mahmoud Khalifa, George Tomlinson, Phyllis Glanc

Research output: Contribution to journalArticlepeer-review

4 Scopus citations

Abstract

OBJECTIVE. The objective of our study was to determine whether applying specific diagnostic criteria to the interpretation of sonohysterography would improve the diagnostic accuracy of the interpretation. MATERIALS AND METHODS. This retrospective study included 200 consecutive patients who underwent both sonohysterography and a procedure that resulted in a positive pathologic diagnosis. The initial interpretation (reader 1) was performed at the time of the examination. Subsequently, a reviewer with limited expertise (reader 2) interpreted the sonohysterograms masked to both the original medical imaging report and the final pathologic diagnosis. Reader 2 used a set of standardized diagnostic criteria to aid in arriving at one of the following diagnoses: normal, endometrial polyp, endometrial hyperplasia, endometrial carcinoma, submucosal fibroid, adenomyosis, adhesions, or clots. These results were compared with the initial diagnostic report (reader 1) and the final pathologic findings. RESULTS. Overall agreement with pathology findings was 76.7% for reader 1 and 84.9% for reader 2. Comparison of the readers' interpretations using the pathologic diagnosis as the reference standard showed that reader 2's interpretations, which were established using the diagnostic criteria set, were uniformly better than those of reader 2. CONCLUSION. The application of standardized diagnostic criteria may enhance the diagnostic accuracy of sonohysterography.

Original languageEnglish (US)
JournalAmerican Journal of Roentgenology
Volume198
Issue number1
DOIs
StatePublished - Jan 1 2012

Keywords

  • Diagnostic criteria
  • Endometrium
  • Resident education
  • Sonohysterography
  • Standardization

Fingerprint Dive into the research topics of 'Sonohysterography: The utility of diagnostic criteria sets'. Together they form a unique fingerprint.

Cite this