LAUSR

Objectives To assess computed tomography (CT) and magnetic resonance imaging (MRI) findings of intramuscular hemangiomas (IMHs) in oral and maxillofacial region and correlate them with the histopathological classifications for selecting optimum management. Methods The clinical data and pretreatment findings of 32 patients with pathologically proven IMHs on CT (n = 10), MRI (n = 27), or both (n = 5) were analyzed retrospectively. Correspondence of clinical and imaging characters with 3 different pathological classifications (cavernous, capillary, and mixed) of IMHs was studied. A number of pitfalls and overlap of imaging features can result in misdiagnosis of different IMHs lesions. Results Four patients had multi-muscular lesions, and 28 had single-muscular lesions. The predilection site were the tongue (11 cases, 34.4%) and the masseter muscle (10 cases, 31.2%). Cavernous type (17 cases, 53.1%) was the most common IMHs type. All patients showed slightly hypointense or isointense on T1-weighted imaging, 3 patients showed hyperintense on T2-weighted imaging and the others showed slightly hyperintense. The most common enhancement pattern was progressive (29 cases, 90.6%). The capillary type (9 cases, 28.1%) and mixed type (6 cases, 28.1%) of IMHs on imaging indicated characteristics of lesions with rich blood supply status, the cavernous type (17cases, 53.1%) of IMHs belonged to relatively poor blood supply lesions. A total of 5 patients (15.6%) were initially misdiagnosed, there were recurrences in 4 IMHs patients. Extra functional MRI (fMRI) was performed on these 5 misdiagnosed patients, the average ADC of the 5 patients was 1.50 × 10−3 mm2/s. The presence of vermicular vessels was different among these three types of IMHs. Conclusions The reason for the misdiagnosis in localized IMHs may be the obvious border of mass-like lesions and the lack of enlarged vessels. Combined evaluation of presence of vermicular vessels and fMRI might be more accurately for determining the IMHs and create a preoperative plan.