PEH, a well recognised but infrequently reported histological diagnosis, is a common vascular tumour of soft tissues and was first described by Pierre Masson in 1923.1,2
The lesion is thought to be the result of reactive endothelial proliferation rather than neoplasia. It is characterised by papillary lobules of proliferating endothelial cells with an underlying fibrous stroma and often resembles angiosarcoma.2,5
It may occur in any blood vessel in the body but is commonly found on the fingers, head, neck, and trunk. PEH may occur either in a pure primary form, as a focal change in a pre-existing vascular lesion (haemangioma, pyogenic granuloma, or vascular malformation), and rarely in an extravascular location following organisation of a haematoma.5,6
It is generally accepted that stasis and thrombosis are prerequisites for the development of PEH.2
A cardiac haemangioma with PEH was diagnosed in the present patient. Histologically, both the cardiac and hepatic lesions were associated with blood and thrombus, and contained multiple papillae with fibrous cores covered by a single layer of bland endothelial cells. The lesions were distinguishable from angiosarcoma by virtue of the more regular papillae and minimal endothelial cell mitosis. Cardiac haemangiomas constitute less than 2.8% of all cardiac masses. Case reports of solitary cardiac haemangiomas with PEH are rare4
; haemangiomas with PEH affecting both the heart and liver have not been previously described in the English literature. Because of the highly vascular nature of both the cardiac and hepatic lesions on magnetic resonance imaging, cardiac malignancy with liver metastases became the presumptive, albeit puzzling, initial diagnosis. Despite the apparently well progressed nature of the disease, complete surgical resection still seemed indicated, owing to the potential for the floating mass in the left ventricular cavity to induce ventricular arrhythmias, sudden death, or systemic dissemination of the thrombus. Ultimately this led to the correct diagnosis. This case serves to highlight how hypervascular cardiac or hepatic tumours may in fact be of vascular origin, rather than metabolically active malignant or metastatic tumours with correspondingly increased vascularity.