Printer friendly version
Ultrasound Outperforms Symptom Analysis In Detecting Ovarian Cancer
08 July 2009
Doctors at the University of Kentucky Chandler Medical Center-Markey Cancer Center compared symptom analysis to ultrasound in predicting ovarian cancer.
They selected 272 women participating in annual trans-vaginal screening (TVS) from 31,748 women enrolled in a free screening project at the university, comparing symptom results to ultrasound and surgical pathology findings.
They found TVS performed better than symptoms analysis for detecting malignancies (73.3% versus 20% sensitivity).
While symptoms analysis performed better for distinguishing benign tumors (91.3% versus 74.4% specificity), adding symptom analysis to TVS actually resulted in poorer identification of malignancy (sensitivity = 16.7%), even as it improved the ability to distinguish benign tumors (specificity = 97.9%).
The authors say the data indicates that while symptoms do identify ovarian malignancies, they are not as accurate as TVS. They add that informative symptoms can be expected to be absent in 80 percent of ovarian malignancies.