OBJECTIVE: Both the intraperitoneal seeding and the uterine-vesical extension theory have been proposed to explain the pathogenesis of bladder endometriosis. The aim of this study was to describe MR imaging findings of bladder endometriosis and involvement of the anterior uterine wall in a tertiary referral centre for endometriosis in a effort to improve diagnosis and help clarify the pathogenesis. METHODS: In a single-centre, retrospective study (2004-2009), 463 consecutive patients analysed for deep infiltrating endometriosis (DIE) were studied independently by two experienced readers for the presence of bladder endometriosis. MR studies revealing bladder endometriosis were then analysed in consensus for: location, size, signal intensity characteristics, uterine involvement, continuity with adenomyosis and presence of cysts. There was histopathologic correlation in 9 patients who had undergone partial bladder resection. RESULTS: Bladder endometriosis was diagnosed in 32 patients on MR imaging (k=0.85). Most lesions showed heterogeneous isointensity compared to that of muscle on T2-weighed imaging, containing foci of high signal intensity, suggesting cystic ectopic endometrial glands. On T1-weighted imaging lesions showed heterogeneous isointensity with foci or small cysts, demonstrating high signal intensity, indicating hemorrhage, was observed. Uterine involvement was found in 94% of the lesions, with either "continuous" or "hourglass" configurations. Presence of contiguous adenomyosis was found in only 4 lesions. CONCLUSIONS: With MR imaging, uterine involvement in bladder endometriosis is frequently found and in most cases located subserosally, suggesting extensive DIE, favouring the intraperitoneal seeding theory.
OBJECTIVE: Both the intraperitoneal seeding and the uterine-vesical extension theory have been proposed to explain the pathogenesis of bladder endometriosis. The aim of this study was to describe MR imaging findings of bladder endometriosis and involvement of the anterior uterine wall in a tertiary referral centre for endometriosis in a effort to improve diagnosis and help clarify the pathogenesis. METHODS: In a single-centre, retrospective study (2004-2009), 463 consecutive patients analysed for deep infiltrating endometriosis (DIE) were studied independently by two experienced readers for the presence of bladder endometriosis. MR studies revealing bladder endometriosis were then analysed in consensus for: location, size, signal intensity characteristics, uterine involvement, continuity with adenomyosis and presence of cysts. There was histopathologic correlation in 9 patients who had undergone partial bladder resection. RESULTS:Bladder endometriosis was diagnosed in 32 patients on MR imaging (k=0.85). Most lesions showed heterogeneous isointensity compared to that of muscle on T2-weighed imaging, containing foci of high signal intensity, suggesting cystic ectopic endometrial glands. On T1-weighted imaging lesions showed heterogeneous isointensity with foci or small cysts, demonstrating high signal intensity, indicating hemorrhage, was observed. Uterine involvement was found in 94% of the lesions, with either "continuous" or "hourglass" configurations. Presence of contiguous adenomyosis was found in only 4 lesions. CONCLUSIONS: With MR imaging, uterine involvement in bladder endometriosis is frequently found and in most cases located subserosally, suggesting extensive DIE, favouring the intraperitoneal seeding theory.