Clinical summary
A 63-year-old Japanese male was admitted to our hospital for the detailed examination of a hepatic cystic lesion, which was originally detected 14 years ago and had slowly enlarged (initial detected size was 16 mm in diameter). He had a clinical history of diabetes mellitus, chronic pancreatitis, and constitutional jaundice. Laboratory tests after the patient’s admission to hospital showed no remarkable abnormality but slight elevations of unconjugated bilirubin (1.2 mg/dL, normal range: < 0.5), glucose (181 mg/dL, normal range: 73–109), hemoglobin A1c (6.1%, normal range: 4.9–6.0), and amylase (188 U/l, normal range: 44–132). Serological tests for hepatitis B virus surface antigen and hepatitis C virus antibody were both negative. All evaluated serum tumor marker (carcinoembryonic antigen, carbohydrate antigen 19 − 9, alpha-fetoprotein, protein induced by vitamin K absence or antagonists-Ⅱ) were within the normal range. Abdominal computed tomography (CT) revealed a well-demarcated multicystic lesion with a thick septum that was 4.2 × 4.1 cm in diameter across segments 8 of the liver. Dynamic CT revealed delayed enhancement of the septum in the mass (Fig. 1a, 1b). T2-weighted magnetic resonance imaging clearly revealed a multicystic lesion with fluid content (Fig. 1c). Positron emission tomography showed no specific uptake. Endoscopic retrograde cholangiography (ERC) showed that there was no detectable communication between the multicystic lesion to the biliary tracts, and this finding was also confirmed by no remnant contrast medium being present in the multicystic lesion in abdominal CT scans after ERC.
From these radiological findings, MCN or MCBH were considered as potential clinical diagnoses. Considering the malignant potential, its isolation from the biliary system and anatomical location and the hepatic functional reserve of the patient, an optimal procedure for surgical resection was discussed. Right hepatectomy was too risky considering the liver function (the indocyanine green retention rate at 15 min was 27.3%) even if percutaneous transhepatic portal vein embolization could be performed.
We concluded that right anterior sectionectomy was optimal after simulation using three-dimensional reconstruction of abdominal CT scans (total liver volume: 1,513 mL, resection volume including tumor volume: 462 mL, resection rate: 30.5%) (Fig. 1d). Furthermore, we selected pure laparoscopic surgery which is less invasive than conventional open liver resection. The patient successfully underwent purely laparoscopic right anterior sectionectomy (operation time: 435 min, blood loss: 212 g). The lesion was completely resected, and no recurrence was observed at the time of writing this case report (7 months after surgery).
Pathological Findings Of Resected Specimens
The cut surfaces of resected specimens grossly exhibited a well-circumscribed multicystic lesion with a thick septum (Fig. 2). The content of the cysts was serous fluid.
Histologically, the multicystic lesion was clearly bordered by surrounding liver parenchyma and had a thick fibrotic septum (Fig. 3). At the fibrotic septum, many small ducts were observed among the fibrous tissue. Although invasive growth of these small ducts was suspected at low-magnification (Fig. 4a), the ducts lacked nuclear atypia and the mitotic figure could not be determined by observation at high-magnification (Fig. 4b). Thus, the possibility of invasive adenocarcinoma (intrahepatic cholangiocarcinoma) could be denied. The cyst wall was mainly covered by cuboidal epithelial cells resembling epithelium of the bile duct (Fig. 1d). Many small ducts morphologically resembled peribiliary glands (Fig. 5a). Possible differential diagnoses for this cystic lesion were intraductal papillary neoplasm of the bile duct (IPNB), MCN, biliary adenofibroma (BAF) or MCBH; however, IPNB was first ruled out because of isolation of the biliary tract.
Ovarian-like stroma, which are necessary for the pathological diagnosis of MCN [13], were not observed even after immunohistochemistry (IHC) to detect estrogen and progesterone receptors and alpha-inhibin. Thus, this lesion could not be categorized as MCN. The nuclei of both bile duct- and peribiliary gland-like epithelium showed no significant atypia and the mitotic figure could not be determined. In IHC analyses, both the bile duct-like and peribiliary gland-like epithelium were positive for cytokeratin (CK) 7 (Fig. 5b) and CK 19, but negative for CK 20, MUC 2 and p53. The Ki-67 labeling index was 2% at the hot spot (Fig. 5c). The bile duct-like epithelia were focally positive for MUC 5AC and MUC 6. Notably, peribiliary gland-like epithelia were diffusely positive for MUC 6 (Fig. 5d) and negative for MUC 5AC. This phenotype is consistent with peribiliary glands.
BAF usually is shown as a solid and/or sponge-like microcystic gross appearance and glandular structure of biliary-type epithelium [14] and therefore BAF was not suit for the diagnosis of the present case. From these findings and clinical considerations, we finally diagnosed the case as MBCH with a marked peribiliary gland component.