Accuracy of ultrasonography performed by examiners with varied training and experience in predicting specific pathology of adnexal masses. (6th April 2015)

• Record Type: Journal Article
• Authors: Sayasneh, A.; Kaijser, J.; Preisler, J.; Smith, A. A.; Raslan, F.; Johnson, S.; Husicka, R.; Ferrara, L.; Stalder, C.; Ghaem‐Maghami, S.; Timmerman, D.; Bourne, T.
• Is Part Of: Ultrasound in obstetrics & gynecology. Volume 45:Number 5(2015:May)
• Page Start: 605
• Page End: 612
• Publication Date: 2015-04-06
• Subjects: ovarian cancer -- ovarian neoplasm -- ovary -- ultrasonography -- visual pattern recognition; Ultrasonics in obstetrics -- Periodicals; Generative organs, Female -- Diseases -- Diagnosis -- Periodicals; Diagnosis, Ultrasonic -- Periodicals; Genital Diseases, Female -- ultrasonography -- Periodicals; Ultrasonography, Prenatal -- Periodicals; 618.047543
• DOI: 10.1002/uog.14675
• Languages: English
• ISSNs: 0960-7692

Abstract

Abstract : Linked Comment: Ultrasound Obstet Gynecol 2015; 45: 513–515 ABSTRACT: Objectives: To assess the diagnostic performance of subjective assessment by Level II ultrasound examiners in predicting the specific histology of adnexal masses. Methods: The women included in this prospective multicenter cross‐sectional study were older than 16 years of age and had at least one adnexal mass. They underwent transvaginal sonography (TVS) performed by Level II examiners, all of whom were familiar with the International Ovarian Tumor Analysis (IOTA) group definitions of ultrasound features of ovarian masses. The final outcome was histology. Specific diagnoses were categorized into 16 groups. Agreement between subjective assessment and final histology was measured using unweighted kappa coefficients. Sensitivities and specificities were obtained for subjective assessment. Results: Of the 1279 women who underwent TVS, 313 were included in the final analysis. Overall agreement (16 × 16 table) between subjective assessment and histology was moderate, with a Cohen's kappa coefficient of 0.59 (95% CI, 0.53–0.65). The specificity of subjective assessment ranged between 91% and 100% for all histological subgroups. Highest sensitivities were achieved in the diagnosis of simple cysts (100% (95% CI, 61–100%)), hydrosalpinges (100% (95% CI, 34–100%)), mature teratomas (88% (95% CI, 74–96%)), endometriomas (75% (95% CI, 61–85%)), ovarian fibromas (88% (95% CI, 47–100%)), tubo‐ovarian abscesses (88% (95% CI, 47–100%)) and serous cystadenocarcinomas (82% (95% CI, 66–93%)). Serous cystadenomas were misdiagnosed most commonly (40.5%). The sensitivity of subjective assessment in diagnosing adnexal torsion was 54% (95% CI, 25–81%); the 17 confirmed and/or suspected cases of adnexal torsion were not included in the 313 cases examined and analyzed for diagnostic performance. Conclusion: Overall, subjective assessment by Level II examiners was good for the detection of simple cysts, endometriomas, mature teratomas, hydrosalpinges, fibroma, tubo‐ovarian abscess and serous cystadenocarcinomas. Copyright © 2014 ISUOG. Published by John Wiley & Sons Ltd.