PET-CT detection of abdominal recurrence of ovarian cancer: Radiologic-surgical correlation

H. K. Pannu, C. Cohade, R. E. Bristow, E. K. Fishman, R. L. Wahl

Research output: Contribution to journalArticlepeer-review

71 Scopus citations

Abstract

The diagnosis of recurrent ovarian cancer can be difficult on cross-sectional imaging, and variable sensitivities and specificities have been reported for positron emission tomography (PET). Combined functional and anatomic imaging with PET plus computed tomography (CT) potentially allows for improved detection of tumor masses. We investigated the sensitivity, specificity, and accuracy of PET-CT for the diagnosis of recurrent ovarian cancer. Sixteen women with previously treated ovarian cancer underwent imaging on a combined PET-CT scanner followed by surgery to assess for possible recurrent disease. The fused PET-CT images were retrospectively reviewed for recurrent disease, and the results of PET-CT were compared with the operative notes. Eleven of the 16 patients had recurrent disease at surgery. The sensitivity, specificity, and accuracy of PET-CT for disease detection on a per-patient basis were 72.7%, 40%, and 62.5%, respectively. For cases of malignant adenopathy (n = 7), 100% were detected on PET-CT. For peritoneal lesions no larger than 1 cm (n = 23), 13% were detected on PET-CT. For peritoneal lesions larger than 1 cm (n = 8), 50% were detected on PET-CT. The sensitivity of PET-CT for recurrent ovarian cancer is moderate in patients with low volume disease. A trial involving a larger number of patients with a spectrum of disease volumes is necessary to determine the impact of PET-CT in clinical practice.

Original languageEnglish
Pages (from-to)398-403
Number of pages6
JournalAbdominal Imaging
Volume29
Issue number3
DOIs
StatePublished - Jan 1 2004

Keywords

  • Cancer
  • Computed tomography
  • Ovary
  • Positron emission toography
  • Recurrence

Fingerprint

Dive into the research topics of 'PET-CT detection of abdominal recurrence of ovarian cancer: Radiologic-surgical correlation'. Together they form a unique fingerprint.

Cite this