Mucin-producing cystic neoplasms of the liver consist of MCN and cyst-forming IPNB, which are considered premalignant or malignant lesions with similar macroscopic features (
2). MCN includes biliary cystadenoma (BCA) and biliary cystadenocarcinoma (BCAC), and is defined as a cyst-forming epithelial neoplasm composed of mucin-producing epithelium and ovarian-type stroma (
1). Both lesions are usually multilocular with enhanced walls, multiple septa, and variable degrees of calcification (
4). Enhanced mural nodules are more commonly encountered in BCAC than BCA (
5). Cyst-forming IPNB is a subtype of IPNB, which was previously categorized as biliary papilloma and papillomatosis (
6). Cyst-forming IPNB lacks ovarian-type stroma and communicates with the bile duct, unlike MCN (
2). However, the imaging findings for cyst-forming IPNB are similar to those of MCN, except for the presence of a mural nodule, and upstream and downstream biliary dilatation. Communication between hepatic cystic lesions and the bile duct is more common in cyst-forming IPNB (
7). A recent study using gadoxetic acid-enhanced MRI demonstrated contrast uptake on 20-minute HBP imaging in cyst-forming IPNB, reflecting the presence of communication between the lesion and the bile duct (
8). Our case showed a mural nodule and contrast uptake on HBP imaging, which led to cyst-forming IPNB as our first differential diagnosis.
Although rare, differentiating biliary cysts and mucin-producing cystic neoplasms can pose diagnostic challenges, as in our case. Biliary cysts usually do not require treatment unless symptomatic, while mucin-producing cystic neoplasms are treated with surgical resection (
9). Thus, differentiation of these lesions is important. Kim et al. reported that the presence of septa, central septa, mural nodule, upstream bile duct dilatation, and downstream bile duct dilatation were significant CT findings differentiating mucin-producing cystic neoplasms from solitary bile duct cysts (P < 0.05 for each finding) (
7). Septa were present in both lesions but the incidence was significantly higher in mucin-producing cystic neoplasms (P = 0.001). Mural nodules (P < 0.001), mosaic pattern (P = 0.113), and communication with the bile duct (P = 0.053) were only present in mucin-producing cystic neoplasms. They concluded that when CT findings were used in combination, mucin-producing cystic neoplasms could be differentiated from biliary cysts with a high degree of diagnostic accuracy (
7). In our case, the mass showed septa and mural nodules (which later proved to be congeries of small capillaries due to secondary revascularization in the process of organization), without central septa or biliary dilatation. Since upstream and downstream biliary dilatation are highly specific (90% - 100% specificity) findings for mucin-producing cystic neoplasms, biliary cyst should have been considered during the differential diagnosis in the present case (
7).
Based on the clinical and imaging findings, the first differential diagnosis for our patient was cyst-forming IPNB with invasive carcinoma. Even though the patient was a 79-year-old male, the second differential diagnosis was MCN with invasive carcinoma due to the multilocular cystic appearance of the mass with enhanced walls. There were strong enhanced portions in the periphery of the cystic mass, and these lesions showed prolonged enhancement. This finding was regarded as indicating an invasive carcinoma arising from an underlying mucin-producing cystic neoplasm. However, upon radiologic-pathologic correlation, the organizing tissue component composed of large muscular vessels surrounded by congeries of small capillaries was compatible with the enhanced portions on imaging studies. In addition, the multilocular cystic appearance on CT and MRI proved to indicate small capillary linings around an organized hematoma, not true septa or multilocular cysts. On 20-minute HBP imaging, the mass exhibited higher SI than on dynamic phase imaging. This could be interpreted as indicating either bile duct communication within the mass or extracellular contrast agent pooling due to excessive fibrosis (
10). Gadoxetic acid is a widely-used hepatocyte-specific MR contrast agent, which has characteristics of both extracellular and liver-specific contrast agents with approximately 50% of the intravenously injected gadoxetic acid being excreted via the bile duct (
11). When a lesion shows contrast uptake on 20-minute HBP imaging, there are two possible mechanisms: 1) the presence of functioning hepatocytes with biliary excretion and/or biliary communication enabling contrast uptake, or 2) the presence of severe fibrosis with abundant extracellular space causing pooling of the contrast agent (
12,
13). However, the latter mechanism is usually seen in solid tumors such as intrahepatic mass-forming type cholangiocarcinomas (
14). On gross pathologic findings, our case revealed that the bile duct was inside the mass, compatible with bile duct communication, and this explained the lesion’s contrast uptake. Most recently, Ying et al. showed that gadoxetic acid revealed an intraductal mucin component and biliary communication in IPNB (
8). We hypothesize that gadoxetic acid-enhanced MRI may be a promising imaging modality for the diagnosis of biliary communication within cystic lesions, although additional study is warranted.
On diffusion weighted imaging, the mass revealed no areas of diffusion restriction. Several recent studies have shown that diffusion restriction could be useful for intraductal IPNB solid component detection and tumor invasiveness determination (
15-
17). Based on retrospective findings, mismatch between the prolonged enhanced portion and the absence of diffusion restriction could be an important clue for differential diagnosis of biliary cysts with hemorrhage and revascularization, rather than invasive carcinoma, in mucin-producing cystic neoplasms. However, the diagnostic value of diffusion weighted imaging for differentiating mucin-producing cystic neoplasms and biliary cysts requires validation by additional studies.
In conclusion, we presented a biliary cyst case complicated by hemorrhage with secondary revascularization in an elderly male patient. Gadoxetic acid-enhanced MRI could be a useful imaging modality for detecting bile duct communication within a cystic lesion. When a complex cystic mass with a persistent enhanced solid portion is detected, although rare, the possibility of complicated biliary cyst with secondary revascularization should be included in the differential diagnosis in addition to mucin-producing cystic neoplasms.