The spectrum of imaging manifestations of Gorham–Stout disease: a novel dynamic contrast-enhanced MR lymphangiography

GSD, also known as vanishing bone disease or massive osteolysis, is a rare lymphatic malformation of unknown etiology [13]. Progressive osteolysis can affect single or multiple bones by progressive destruction of the osseous matrix due to an imbalance of osteoclasts and osteoblasts, as well as proliferation of endothelial cells in lymphatic structures [14]. Therefore, when considering the pathogenesis of GSD, imaging of the lymphatic vessels is essential to accurately assess the diagnosis, disease extent, and treatment response. This study summarizes the clinical history, osseous involvement, and extraosseous manifestations of GSD. In particular, the recently introduced imaging modality, DCMRL, was used to assess the disease based on its mechanism, despite the small number of patients.

The clinical manifestations of GSD vary widely depending on the site and extent of involvement [15]. GSD results in pain, swelling, weakness, and pathological fractures of the affected bones or adjacent areas. Therefore, conventional plain radiography focused on assessing osseous involvement of the spine, pelvis, lower extremities, and ribs. Osteolytic lesions in GSD are associated with local lymphatic vessel proliferation. In this study cohort, the most affected sites were the spine (73.3%), pelvic bone (60.0%), lower extremities (40.0%), and ribs (40.0%), as seen on conventional plain radiography. Despite the variance of bone lesions depending on the stage, plain radiographs can provide the most significant clues for diagnosing GSD and can be helpful for accurately assessing the range of bone destruction.

However, lymphatic vessel proliferation can occur anywhere; thus, extraosseous involvement can lead to chylous effusion in the pleural or abdominal cavity, resulting in poor prognosis and life-threatening symptoms such as dyspnea and infection [16‐18]. This study also revealed chylothorax in seven patients (46.7%) and chylous ascites in one patient. There was a relatively high incidence of edematous changes in the trunk and lower extremities (60.0%). These changes might be related to the high rate of sepsis history, such as pneumonia or cellulitis (80.0%). Conventional MRI can help assess the extent of non-osseous involvement, such as lung, pleura, spleen, and soft-tissue changes [19].

Considering the mechanisms that drive osteolysis and lymphangiogenesis in GSD, an accurate assessment of lymphatic vessels might be needed in patients with GSD. DCMRL is an emerging imaging tool for evaluating lymphatic anatomy and physiologically assessing the lymphatic flow. Compared to the previous lymphatic imaging modalities, such as lymphoscintigraphy or lipiodol lymphangiography, DCMRL has better spatial resolution and more physiologic lymphatic flow due to non-oily contrast agents. To our knowledge, there are few reports on DCMRL findings in patients with GSD. DCMRL, as a one-stop imaging modality, can be used to evaluate non-osseous involvement and lymphatic flow. DCMRL can reveal the phenomenon of abnormal propagation of contrast agents from the inguinal lymph nodes through the central conducting lymphatics and dermal backflow along the chest and abdominal walls. It can directly show the proliferation of lymphatic vessels within the bone in patients with GSD, which are useful imaging findings that cannot be seen on plain radiographs or other imaging modalities. All patients who underwent DCMRL in this study presented with altered anatomical lymphatics and functional flow with collateralization. In particular, the giant thoracic duct seen in two patients with GSD may be a tortuous dilated thoracic duct deformed due to weak central conducting lymph flow, and DCMRL data collection about the giant thoracic duct would be required to determine whether it is one of the characteristic findings in GSD patients. Furthermore, DCMRL could help determine further treatment so that interventional treatment such as embolization could proceed. Therefore, evaluation through DCMRL might be necessary and crucial to assess patients with GSD.

GSD may often be misdiagnosed as a neoplasm, tuberculosis, or chronic osteomyelitis owing to its rarity and unique clinical characteristics [20]. Therefore, a combination of clinical history, radiologic findings, and histopathological examination is the gold standard for the diagnosis of GSD. This study is useful in that it has shown in several cases that it is possible to directly view and evaluate lymphatic vessel proliferation using DCMRL in patients with GSD. However, because GSD is an extremely rare disease, this study has a limitation in that it investigated only a relatively small number of patients. Since all patients who underwent interventional embolization were also treated with sirolimus and pamidronate, it is difficult to know what the actual treatment effect was due to. There has yet to be a consensus on the most effective treatment for GSD. The treatment options for GSD patients consists of the mTOR inhibitor sirolimus and bisphosphonates such as pamidronate [21]. The effect of sirolimus blocks the expression of vascular endothelial growth factor as an mTOR inhibitor, which shows anti-angiogenic results in patients with GSD because of stopping the proliferation of lymphatic vessels [22]. Bisphosphonate agents are also reported as one of the medical treatment drugs because of their anti-osteoclastic and anti-angiogenic effects [21‐23]. Based on the pathogenesis and previous case reports, our medical treatment protocol for patients with GSD adopted sirolimus and pamidronate for anti-osteoclastic and anti-angiogenic effects under off-label use. However, a randomized controlled trial is still needed to select the optimal medical treatment method.