Leiomyosarcomas best removed en bloc
“When you are creating public health care policy, decision analysis must begin with scientifically valid evidence,” asserted panelist Dr. Elizabeth Pritts, medical director of the Wisconsin Fertility Institute, Middleton.
Dr. Elizabeth Pritts
She and her colleagues undertook a comprehensive new meta-analysis assessing the prevalence of occult leiomyosarcomas at hysterectomy or myomectomy for presumed uterine fibroids, including 133 original articles describing 30,193 women having explicit pathology.
Analysis of all prospective data showed that the predicted prevalence rate of occult leiomyosarcoma was 0.12 per 1,000 operations for presumed benign fibroids.
The corresponding 1 in 8,300 operations needed to find a leiomyosarcoma in this new meta-analysis differs greatly from the 1 in 498 found in an FDA meta-analysis, mainly because of the differing evidence base, Dr. Pritts maintained. “It really has to do with initial search criteria,” she said, noting, for example, that the FDA’s search strategy missed studies in which no cancer was found and studies in languages other than English.
Dr. Pritts and her colleagues also conducted a new systematic review looking at outcomes after morcellation of an unsuspected leiomyosarcoma, which was recently published (J. Minim. Invasive Gynecol. 2014 Sept. 2 [doi: 10.1016/j.jmig.2014.08.781]).
Main analyses here were based on six papers that compared morcellation with en bloc removal of leiomyosarcomas, most of which found worse survival for women whose tumors were morcellated.
“Now this is not great evidence, but remember, in evidence-based medicine, you’ve got to look at the very best available evidence. This is it,” Dr. Pritts maintained. “En bloc removal confers benefit—don’t cut into these.”
On closer inspection, only 3 of the 81 cases of morcellation reported were confirmed to be power morcellation. Comparisons of outcome with power versus hand morcellation, albeit limited by small numbers, suggested no difference in survival or upstaging.
“There are no data to suggest that any type of morcellation is better or worse than another type, even when including simple tumor biopsies,” Dr. Pritts concluded.
Dr. Brown, Dr. Lieng, and Dr. Pritts disclosed that they had no relevant conflicts of interest.