The prevalence of occult leiomyosarcoma at surgery for presumed uterine fibroids: a meta-analysis

Elizabeth A. Pritts, David J. Vanness, Jonathan S. Berek, William Parker, Ronald Feinberg, Jacqueline Feinberg, David L. Olive

There is a concern regarding the risk of occult leiomyosarcomas found at surgery for presumed benign fibroids. We sought to produce a comprehensive review of published data addressing this issue and provide high-quality prevalence estimates for clinical practice and future research. A comprehensive literature search using the PubMed/MEDLINE database and the Cochrane Library was performed. Inclusion criteria were human studies, peer-reviewed, with original data, involving cases for surgery in which fibroid-related indications were the primary reason for surgery, and histopathology was provided. Candidate studies (4864) were found; 3844 were excluded after review of the abstract. The remaining 1020 manuscripts were reviewed in their entirety, and 133 were included in the Bayesian binomial random effect meta-analysis. The estimated rate of leiomyosarcoma was 0.51 per 1000 procedures (95 % credible interval (CrI) 0.16-0.98) or approximately 1 in 2000. Restricting the meta-analysis to the 64 prospective studies resulted in a substantially lower estimate of 0.12 leiomyosarcomas per 1000 procedures (95 % CrI <0.01-0.75) or approximately 1 leiomyosarcoma per 8300 surgeries. Results suggest that the prevalence of occult leiomyosarcomas at surgery for presumed uterine fibroids is much less frequent than previously estimated. This rate should be incorporated into both clinical practice and future research.