The clinician is occasionally faced with the diagnostic dilemma of an indeterminate upper abdominal mass. Despite thorough evaluation with cross-sectional imaging, endoscopy, endoscopic ultrasound (EUS), and even biopsy, not all upper abdominal masses can be diagnosed preoperatively. As described in the case report by Kochar and colleagues, hepatic lesions—including hemangiomas—can infrequently appear as apparent extrahepatic tumors.1 In this case report, a symptomatic, exophytic hepatic hemangioma abutting the greater curve of the stomach was misdiagnosed as a gastric subepithelial tumor (SET) by both computed tomography (CT) scan and EUS. The correct diagnosis was subsequently made by core biopsy and laparoscopy.
In contrast to other liver tumors, hemangiomas can usually be diagnosed using cross-sectional imaging. On CT scan, hepatic hemangiomas are typically hypodense on unenhanced images.2 Following administration of intravenous contrast medium, hepatic hemangiomas have a characteristic pattern of enhancement, with early peripheral nodular enhancement, followed by centripetal filling in of the lesion on delayed phases. This delayed-phase CT imaging can be particularly helpful in the diagnosis of an hepatic hemangioma. In fact, a late-contrast phase should be included when evaluating all masses in proximity to the liver, and care must be taken not to interpret imaging of tumors adjacent to the liver solely based on noncontrast or early-phase CT scan.
Magnetic resonance imaging (MRI) can be particularly accurate for diagnosing a liver hemangioma. Even without contrast, hemangiomas have a pathognomonic appearance on MRI in most cases. On MRI, these tumors are characterized as homogeneous, well-demarcated lesions with very high signal intensity on T2-weighted images. Like CT, contrast-enhanced MRI of hepatic hemangiomas typically display early peripheral nodular enhancement with centripetal fill-in on delayed images. In one recent study by Szurowska and colleagues, the sensitivity of contrast-enhanced CT for a hemangioma was 76%, and the specificity was 90%.3 In comparison, the sensitivity of MRI (both noncontrast and with contrast) was 98%, and the specificity was 99%.
Unlike smaller hemangiomas, very large heman-giomas can have more heterogeneous features on CT or MRI.4 This is often secondary to hemorrhage, thrombosis, fibrosis, or extensive hyalinization. Although such tumors can be more difficult to diagnose on cross-sectional imaging, they still commonly display characteristic peripheral nodular enhancement with delayed centripetal filling after contrast administration on both CT and MRI. However, the centripetal filling of these lesions is often incomplete, representing the areas of hemorrhage, fibrosis, or hyalinization. This underscores the importance of the use of delayed-contrast CT and MRI in the evaluation of indeterminate masses in or adjacent to the liver.
Biopsy is rarely required for the diagnosis of a liver hemangioma. Early reports of fine-needle aspiration (FNA) of a liver hemangioma demonstrated significant risk of often fatal complications as a result of bleeding.5-7 In addition, the diagnostic yield from FNA is low. In one early report by Taavitsainen and colleagues, FNA was performed in 36 patients with suspected hepatic hemangiomas; however, diagnostic cellular material could only be obtained in 21 patients.7 Core biopsy, although perhaps riskier than FNA, may improve the accuracy of needle biopsy because a larger volume of tissue is sampled. However, biopsy of lesions suspected to be hepatic hem-angiomas is rarely advised, given the high sensitivity and specificity of contrast-enhanced CT and MRI, the risk of bleeding associated with percutaneous biopsy, and the low diagnostic yield of percutaneous biopsy.
The presence of symptoms requiring tumor resection for palliative indications can make the preoperative evaluation of tumors in this region less complicated in most cases. When symptomatic, surgical resection is generally recommended in patients with liver hemangiomas. Most commonly, patients with symptomatic hepatic hemangiomas present with pain or with symptoms of compression of extrinsic organs (nausea, vomiting, and early satiety), as did the patient in the case presented by Kochar and colleagues.1 Complications from hepatic hemangiomas, including spontaneous rupture, are exceedingly rare. In addition, when followed over time, the size of most hemangiomas remains stable. Thus, when asymptomatic, even giant hepatic hemangiomas can safely be observed.8,9 Prior to resection, it is imperative that other potential causes of a patient’s symptoms be fully investigated and ruled out. In a series published by Pietrabissa and colleagues, 50% of patients continued to be symptomatic following intervention for a symptomatic hepatic hemangioma,9 indicating the existence of an alternative cause for the patients’ symptoms.
The differential diagnoses of gastric SETs are extensive and include gastrointestinal stromal tumors (GISTs), leio-myomas, lipomas, gastric varices, tumors of neural origin, pancreatic rests, duplication cysts, carcinoid tumors, lym-phomas, and metastases.10 SETs outside the stomach can mimic gastric tumors, originating from adjacent structures (including the liver, spleen, biliary tract, and pancreas), and cause extrinsic compression of the stomach. For gastric SETs that are clearly resectable, biopsy is generally not needed because initial resection is indicated, even when the SET is not symptomatic. For borderline resectable and unresectable SETs—or if lymphoma is suspected based on imaging—biopsy can be useful in dictating management and is generally indicated. For example, preoperative ima-tinib therapy may be useful in some cases of a borderline or unresectable GIST to improve resectability. Similarly, gastric lymphoma is generally treated with either Helicobacter pylori eradication, systemic therapy, radiation, or a combination of the above therapies, and surgery is rarely indicated for gastric lymphoma. When the diagnosis is uncertain, the asymptomatic lesion will often undergo more thorough diagnostic evaluation to rule out tumors that do not require resection. Although cross-sectional imaging alone is often sufficient to diagnose hepatic hem-angioma, indeterminate lesions may necessitate further testing, including endoscopy, EUS, or biopsy.
In summary, this interesting case illustrates that the diagnosis of upper abdominal masses can be challenging. Although hepatic hemangiomas are typically less difficult to diagnose using triple-phase contrast imaging, the diagnosis of those that may not have classic imaging features may be problematic. In particular, atypical hemangiomas that are exophytic and abut other organs such as the stomach, as in the case reported by Kochar and colleagues,1 may masquerade as gastric SETs.
References
- 1.Kochar R, Atiq M, Lee JH, Landon G, Mansfield PF, Bhutani MS. Giant hepatic hemangioma masquerading as a gastric subepithelial tumor. Gastroenterol Hepatol (NY). 2013;9:396–398. [PMC free article] [PubMed] [Google Scholar]
- 2.Heiken JP. Distinguishing benign from malignant liver tumours. Cancer Imaging. 2007;7(spec no A):S1–S14. doi: 10.1102/1470-7330.2007.9084. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Szurowska E, Nowicki T, Izycka-Swieszewska E, et al. Is hepatotropic contrast enhanced MR a more effective method in differential diagnosis of hemangioma than multi-phase CT and unenhanced MR? BMC Gastroenterol. 2011;11:43. doi: 10.1186/1471-230X-11-43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Vilgrain V, Boulos L, Vullierme MP, Denys A, Terris B, Menu Y. Imaging of atypical hemangiomas of the liver with pathologic correlation. Radiographics. 2000;20:379–397. doi: 10.1148/radiographics.20.2.g00mc01379. [DOI] [PubMed] [Google Scholar]
- 5.Terriff BA, Gibney RG, Scudamore CH. Fatality from fine-needle aspiration biopsy of a hepatic hemangioma. AJR Am J Roentgenol. 1990;154:203–204. doi: 10.2214/ajr.154.1.2104717. [DOI] [PubMed] [Google Scholar]
- 6.Davies R. Haemorrhage after fine-needle aspiration biopsy of an hepatic hae-mangioma. Med J Aust. 1993;158:364. doi: 10.5694/j.1326-5377.1993.tb121823.x. [DOI] [PubMed] [Google Scholar]
- 7.Taavitsainen M, Airaksinen T, Kreula J, Paivansalo M. Fine-needle aspiration biopsy of liver hemangioma. Acta Radiol. 1990;31:69–71. [PubMed] [Google Scholar]
- 8.Gilon D, Slater PE, Benbassat J. Can decision analysis help in the management of giant hemangioma of the liver? J Clin Gastroenterol. 1991;13:255–258. [PubMed] [Google Scholar]
- 9.Pietrabissa A, Giulianotti P, Campatelli A, et al. Management and follow-up of 78 giant haemangiomas of the liver. Br J Surg. 1996;83:915–918. doi: 10.1002/bjs.1800830710. [DOI] [PubMed] [Google Scholar]
- 10.Humphris JL, Jones DB. Subepithelial mass lesions in the upper gastrointestinal tract. J Gastroenterol Hepatol. 2008;23:556–566. doi: 10.1111/j.1440-1746.2007.05232.x. [DOI] [PubMed] [Google Scholar]
