A meta-analysis of FDG-PET for the evaluation of breast cancer recurrence and metastases

Abstract

To evaluate the diagnostic performance of 18F-2-deoxy-2-fluoro-D-glucose-positron emission tomography (FDG-PET) in the evaluation of breast cancer recurrence and metastases. Articles published in medical journals during January 1995-June 2004 were identified by a systematic Medline search, supplemented by a manual search of the references listed in original and review articles. Studies that evaluated FDG-PET with a dedicated camera for the diagnosis of breast cancer recurrence or metastases, and reporting sufficient data to permit calculation of sensitivity and specificity, were included in the analysis. Two reviewers independently reviewed the eligibility and abstracted data regarding the sample population, technical imaging characteristics of FDG-PET, and the number of true positives, true negatives, false positives and false negatives. Differences between readers were resolved by consensus. We used meta-analytic methods to estimate the pooled sensitivity, false positive rate, and the maximum joint sensitivity and specificity. Eighteen studies met the inclusion criteria. Sixteen studies included patient-based data, comprising a sample size of 808 subjects, and eight studies included lesion-based data, totaling 1013 lesions. Among the studies with patient-based data, the median sensitivity was 92.7%, and the median specificity was 81.6%. The pooled sensitivity was 90% [95% confidence interval (86.8-93.2)], and the pooled false positive rate was 11% [95% confidence interval (7.8-14.6)], after the exclusion of outliers. The maximum joint sensitivity and specificity, was 88% [95% confidence interval (86.0-90.6)]. These results indicate that FDG-PET is a valuable tool for detecting breast cancer recurrence and metastases.