Abstract

Spontaneous liver hemorrhage of unknown origin is an extremely rare entity. In previous literature, the word “spontaneous” is used for liver hemorrhage for reasons other than trauma or iatrogenic injuries. The reported etiologies of spontaneous liver hemorrhage mainly include pregnancy-related condition, liver neoplasm, and medications. Very rarely is the identification of the cause of spontaneous liver hemorrhage difficult, and there have been 2 cases of such “idiopathic” spontaneous liver hemorrhage in previous reports.1,2
In this article, we present a case of spontaneous liver hemorrhage that was initially diagnosed as liver neoplasm based on imaging findings and was therefore treated with surgical resection. The surgical specimen revealed no tumor, and the liver hemorrhage was truly idiopathic. To the best of our knowledge, this is the third case of idiopathic spontaneous liver hemorrhage.
A 51-year-old man developed acute right shoulder pain when he was sitting relaxed on a sofa and thus visited a local hospital. Abdominal contrast-enhanced computed tomography (CT) revealed a large heterogeneous low-density area in the right hepatic lobe. Furthermore, a high-density area was located in the arterial phase, indicating hepatic subcapsular hematoma with extravasation (Figure 1A, above). There was no apparent tumor, abscess, or aneurysm in the liver. He had no relevant past medical history and was not taking antiplatelet or anticoagulant medication. Therefore, he was diagnosed with hepatic subcapsular hematoma of unknown origin. He was referred to our hospital for further examination and treatment. On admission, the patient’s vital signs were stable, and physical examination was negative for abdominal distension or tenderness. Blood test revealed slight anemia (hemoglobin 13.4 g/dL), increased hepatic enzyme levels (aspartate aminotransferase 83 U/L and alanine aminotransferase 89 U/L), inflammatory response (white blood cell count 15330/μL and C-reactive protein 1.99 mg/dL), and normal platelet count or absence of coagulation. Levels of tumor markers, including carcinoembryonic antigen, alpha-fetoprotein, carbohydrate antigen 19-9, and protein induced by vitamin K absence or antagonist 2, were not increased. Follow-up CT revealed an 18-mm-diameter intrahepatic mass in the segment 8, and extravasation was no longer identified in the hematoma (Figure 1A, below). Because there was no sign of active bleeding and his condition was already stable, invasive treatment including interventional radiology or surgery was not indicated, and conservative treatment with fasting, intravenous transfusion, and antibiotic therapy was performed. Abdominal contrast-enhanced magnetic resonance imaging (MRI) was performed 3 days after admission, and the intrahepatic mass was detected on CT. Although the intensity of the mass was not homogenous, the major part of the mass showed low intensity on T1-weighted image (T1WI) and high intensity on T2-weighted image (T2WI) and diffusion-weighted image (DWI) (Figure 1B), which was consistent with the pattern of most liver malignant neoplasms. Dynamic study revealed low attenuation in the arterial, portal, venous, and hepatic phases (Figure 1B). The differential diagnoses of the mass included a tumor that could either be poorly differentiated hepatocellular carcinoma (HCC), hepatic adenoma, hepatic angioma, or hepatic angiolipoma, and in any case, the origin of hemorrhage was suspected to be the tumor. Therefore, surgery to resect the tumor and remove the hematoma was scheduled, and 3 days before the surgery, transcatheter arterial embolization (TAE) was performed for the hepatic artery in the segment 8 to prevent intraoperative bleeding from the tumor. Seven days after admission, partial hepatectomy of the segment 8 following removal of the hematoma was performed. On laparotomy, there was a large subcapsular hematoma covering almost all of the right liver (Figure 2B). Intraoperative ultrasound revealed no mass. However, contrast-enhanced ultrasonography using Sonazoid revealed a hypoechoic lesion, 2 cm × 1 cm in size, and a wedge-shaped area adjacent to the lesion. The wedge-shaped area showed a slightly hypoechoic Kupffer phase (Figure 2B). Liver transection was performed so that the wedge-shaped area was included in the specimen. The operative time was 4 h 30 min, and the total blood loss was 600 mL. Preoperative imaging findings on CT and MRI. (A) Axial view of the liver on CT. A large subcapsular hematoma on the right liver with extravasation (arrowhead) was observed on the first CT (above), and there was an 18-mm-diameter intrahepatic mass (arrow) in the segment 8 on follow-up CT (below). (B) Axial view of the liver on MRI. The major part of the mass (arrowhead) showed low intensity on T1WI and high intensity on T2WI and DWI. A dynamic study revealed low attenuation in HBP. Abbreviations: CT, computed tomography; DWI, diffusion-weighted image; HBP, hepatobiliary phase; MRI, magnetic resonance imaging; T1WI, T1-weighted image; T2WI, T2-weighted image. (A) Intraoperative photographs during partial hepatectomy of the segment 8 following the removal of the hematoma. A large subcapsular hematoma covered almost all of the right liver (left). The resection surface of the liver after partial hepatectomy is shown (right). (B) Intraoperative contrast-enhanced ultrasonography using Sonazoid. Hypoechoic lesion (arrow), 2 cm × 1 cm in size, and wedge-shaped area (arrowhead) adjacent to the lesion are shown. (C) Macroscopic view of the surgical specimen. An intrahepatic hematoma (arrow) and a laceration from the hematoma to the subcapsular hematoma (arrowhead) are observed. (D) Hematoxylin and eosin staining of the surgical specimen. An intrahepatic hematoma (arrow) and a focal necrosis due to transcatheter arterial embolization (arrowhead) are observed.

The postoperative course was uneventful except for persistent elevation of inflammatory marker levels of unknown origin, which were improved with antibiotic therapy. The patient was discharged on postoperative day 33. Pathological examination revealed no tumor in the specimen. There was laceration from the intrahepatic hematoma to the subcapsular hematoma (Figure 2C). The mass detected on CT, MRI, and ultrasonography turned out to be an intrahepatic hematoma. The wedge-shaped area identified on contrast-enhanced ultrasonography was considered to be focal necrosis due to TAE (Figure 2D). The cause of bleeding was unclear with no evidence of neoplasm, aneurysm, or other vascular and systemic diseases, including polyarteritis nodosa or amyloidosis. Four months after discharge, a follow-up CT was performed, and there was no mass in the liver or other organs.
A search for reported articles on spontaneous liver hemorrhage was conducted in the PubMed database on May 10, 2020. The search words were (“spontaneous” OR “idiopathic”) AND (“hemorrhage” OR “bleeding” OR “hematoma”) AND (“liver” OR “hepatic”) AND (“subcapsular” OR “intrahepatic”) in title and/or abstract. In total, 217 articles were found, and there were 2 cases of spontaneous liver hemorrhage of unknown origin, namely, “idiopathic” spontaneous liver hemorrhage,1,2 and the present report is the third such case (readers can contact us for a complete reference list of our search for spontaneous liver hemorrhage). Briefly, the previously reported 2 cases were of 40-year-old man and 43-year-old woman. They had neither underlying diseases nor any medications that could cause liver hemorrhage. The male patient visited the hospital with right upper quadrant pain and developed 16 × 10 cm hematoma in the right liver. The female patient complained right lower thoracic pain, and hematoma was detected in the right liver although the size was not descripted. Both hemorrhages were treated with conservative therapy; the male patient was successfully relieved, but the female patient died 2 days after admission. In the present case, surgical resection was performed for right liver subcapsular hematoma because liver neoplasm was initially suspected as the cause of hemorrhage. In consequence, pathological examination revealed no particular cause of liver hemorrhage; therefore, the hemorrhage was considered to be idiopathic. Considering the possible fatal course in the reported case, surgical treatment might be a choice for idiopathic spontaneous liver hemorrhage.
Differential diagnosis between intrahepatic hematomas and liver neoplasms is sometimes difficult. Currently, MRI is widely used to differentiate intrahepatic masses with high diagnostic accuracy. Magnetic resonance imaging is also used for diagnosing hematoma, and it is well-known that hematoma shows sequential changes in MRI signal intensity reflecting the degeneration of hemoglobin.3,4 Additionally, the signal intensity in hematoma is frequently heterogenic because hematoma usually comprises fresh and old hemorrhages. 4 The sequential change and heterogeneity of hematoma in MRI signal intensity would make the differential diagnosis difficult. In the present case, the origin of the hemorrhage was initially considered to be a certain liver neoplasm since MRI detected a well-circumscribed intrahepatic mass just close to extravasation, and the intensity of the mass was low on T1WI and high on T2WI and DWI without contrast enhancement, which was consistent with both hyperacute hematoma and some liver malignant neoplasms including HCC and metastatic liver cancer. Although intrahepatic hematoma can be treated with TAE or conservative therapy, liver neoplasm is the most frequent cause of spontaneous liver hemorrhage requiring surgical resection. Therefore, careful evaluation is required when encountering spontaneous liver hemorrhage with an intrahepatic mass on MRI.
Footnotes
Author Contributions
Fumiya Sato, Kyoji Ito, and Nobuyuki Takemura contributed to the conception and design of the study. All authors contributed to the acquisition and analysis of data. Fumiya Sato, Kyoji Ito, and Nobuyuki Takemura were major contributors in writing the manuscript. All authors have read and approved the final manuscript.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Grants-in-Aid for Research from the National Center for Global Health and Medicine (30-1021 to N.T.).
Data Accessibility Statement
The data sets analyzed during the current study are not publicly available because they contain information that could compromise the privacy of research participants but are available from the corresponding author upon reasonable request.
Ethical Statement
This article satisfied the consensus of the National Center for Global Health and Medicine Research Ethics Committee/Institutional Review Board.
