Introduction

Adenomatoid tumor (AT) is a rare benign genital tract neoplasm of mesothelial origin, and commonly affects fallopian tubes and the uterus. AT arising from uterus typically occurs in women of reproductive age, and is found in 1-5% of hysterectomy specimens. AT is usually located subserosal or in the outer myometrium, and typically a small solid mass but occasionally larger, or may be cystic^1-4. Most of uterine AT may be misdiagnosed as leiomyoma because it shared the same imaging features of leiomyoma^5-7. Because leiomyoma is sensitive for hormone therapy, unnecessary hormone therapy may be administered for AT which is misdiagnosed as leiomyoma. In addition, because AT is densely adherent to the surrounding myometrium without separating layer or capsule, it may be difficult to enucleate and surgeons should modify the procedure during laparoscopic “myomectomy”^{8, 9}. Therefore, it is important to differentiate AT from leiomyoma for the adequate treatment. The purpose of this study is to evaluate the MRI findings of AT including DWI for the differentiation from leiomyoma.

Methods

Pathologically proven 10 ATs in 9 women (29-54 years of age, mean 42 years) who had undergone MRI examinations with 3T (2 lesions in 2 cases) or 1.5T (8 lesions in 7 cases) superconducting MRI systems before surgery were retrospectively evaluated (Fig. 1). Fast spin-echo T2WI and spin-echo or gradient-echo T1WI with/without fat-saturation were obtained in all 9 patients. Gadolinium-enhanced T1WI were obtained in 7 lesions in 6 cases. DWI (b=800s/mm²) was obtained in 9 lesions. Susceptibility-weighted sequence (SWAN) was obtained in 6 lesions. Two gynecologic radiologists both with more than 20-years experiences qualitatively evaluated the images: location, shape, size, margin, signal intensity on T1WI (plain/CE), T2WI, DWI (b=800), and computed DWI (b=1500); the presence of hemorrhagic foci (high intensity on T1WI and signal voids on SWAN). The reviewers examined all images of the cases independently and then resolved discrepancies by consensus. The mean ADC values were measured in 3 circular ROIs within a mass (entire mass, high intensity area and low intensity area on DWI, respectively) from ADC maps generated by using b-values of 0 and 800 on the workstation (AW4.2).

Results and Discussions

All 10 tumors exhibited as subserosal or outer-myometrial round solid masses, and no predominantly cystic mass was observed. The maximum diameter of the tumor was 2.2 to 5.0 cm (mean, 3.3 cm). Nine of 10 lesions showed heterogeneous signal intensity with admixture of partially ill-defined high signal intensity areas and well-defined myoma-like low signal intensity areas on T2WI. The other one lesion showed slight low signal intensity with totally ill-defined margin on T2WI. Ring-like high intensity area was observed on T2WI in 4 ATs. Because AT is composed of mesothelial origin tubules of varying size with smooth muscle hypertrophy (Fig. 2), areas containing small cystic dilatations of abundant tubules may increase the signal intensity on T2WI, whereas areas of smooth muscle hypertrophy may show low signal intensity like leiomyoma^{3, 4}. In our study, the admixture of areas of smooth muscle hypertrophy showing low signal intensity with well-defined margins, and areas of abundant tubules showing high signal intensity with ill-defined margins on T2WI may be characteristic for AT reflecting its pathological features (Fig. 3, 5). Areas of abundant tubules showed high signal intensity on DWI (b=800) and signal decrease on computed DWI (b=1500) with relatively high ADC (high intensity areas: 1.4 to 1.73, mean 1.54 x 10^-3 mm²/s; low intensity areas: 1.14 to 1.43, mean 1.33 x 10^-3 mm²/s) in 6 of 9 lesions suggesting T2 shine-through effect (Fig. 4, 5). Ring-like high intensity area was observed on DWI in 3 ATs, which may be rarely observed in leiomyomas. In the other 3 lesions, 2 lesions did not contain high signal intensity areas on DWI, and one lesion showed high signal intensity both on DWI and computed DWI (b=1500) with relatively low ADC (1.14) suggesting water diffusion restriction. However uterine myometrial mass with ill-defined margins and/or with high intensity areas on T2WI and DWI may have a probability of malignancy such as sarcomas, however, relatively high ADC of most ATs may be suggestive for its benignity. High intensity hemorrhagic foci on T1WI were not observed in all 10 lesions. SWAN showed small signal voids reflecting hemorrhage in one lesion with hyalinized necrosis possibly due to partial ischemic changes. Intra-tumoral hemorrhagic necrosis is suggestive finding for uterine sarcomas¹⁰. In this study, most ATs showed no intra-tumoral hemorrhage, which may be compatible with benign tumor. Co-existing leiomyomas were demonstrated in 6 of 9 patients. Degree of enhancement was either lower than (n=6) or equal to that of the myometrium (n=1). Relatively weak contrast-enhancement of ATs may be due to unenhanced small cystic spaces formed by the dilatation of abundant tubules (Fig. 5).

Conclusion

We conclude that typical AT may appear as a heterogeneous signal intensity mass containing myoma-like low signal intensity areas with well-defined margins and high intensity areas with ill-defined margins on T2WI. AT may contain high signal intensity areas on DWI, occasionally appearing as ring-like high intensity area, with relatively high ADC due to T2 shine-through effect reflecting abundant tubules. Intra-tumoral hemorrhage may also be rare on MRI.

Acknowledgements

No acknowledgement found.