Uterine artery embolization with trisacryl gelatin microspheres in women treated for leiomyomas: a clinicopathologic analysis of alterations in gynecologic surgical specimens.

To characterize the histologic range of alterations due to uterine artery embolization with trisacryl gelatin microspheres in gynecologic specimens containing leiomyomas in detail, we report our clinicopathologic experience with 26 cases (longest postuterine artery embolization interval, 1.9 yr). Microspheres were observed in 85% of cases and could be seen up to 1.9 years after embolization. They were mainly present in leiomyomas and nonneoplastic myometrium but could be found in other nontargeted sites, such as the cervix, endometrium, ovaries, and fallopian tubes; however, infarction (present in 96% of cases) was confined to leiomyomas and did not involve other nonneoplastic tissues. The appearance of the infarcts was correlated with time after embolization, and coagulative necrosis/necrosis of indeterminate type was restricted to the early period after uterine artery embolization (before 10 wk postuterine artery embolization) whereas hyaline necrosis was seen predominantly in the late period (mostly after 10 wk, up to 1.9 yr). Of the 14 hysterectomy specimens with microspheres in extravascular spaces (almost all of which were in close proximity to the arteries), pseudoaneurysms were also focally present in 8 (57%) specimens. Microspheres were usually associated with mild inflammatory reactions, which persisted >1 year after embolization but did not become more severe over time. Morphologic and histochemical features of trisacryl gelatin microspheres were compared with other embolization agents, which can also be encountered in surgical specimens [polyvinyl alcohol (PVA) particles and PVA microspheres]. Trisacryl gelatin microspheres were negative with periodic acid-Schiff and orange-pink with Movat stains whereas PVA was positive with periodic acid-Schiff and black with Movat. Our study, the largest histologic analysis to date, confirms and extends the observations of earlier studies of trisacryl gelatin microspheres. In addition, we conclude that, as expected, the histologic appearance of microsphere-induced infarcts is a function of time, similar to healing of infarcts in nongynecologic sites. Pseudoaneurysms are a likely mechanism for the production of microspheres in extravascular spaces. Inflammation associated with microspheres can persist in gynecologic tissues but does not seem to result in the destruction of nontargeted sites. Finally, trisacryl gelatin microspheres can be distinguished from PVA particles and PVA microspheres based on a combination of morphologic features and histochemical stains.