3.1.1 Diagnostic Considerations
In our case, the patient’s diagnosis of GSD was made based on a combination of symptoms and imaging findings. Chronicity of his symptoms over several years and absence of a primary tumor made a malignant process highly unlikely. While osteolytic lesions do occur in the skull base in Langerhans cell histiocytosis (LCH), LCH bone lesions more commonly affect the calvarium and the temporal bone within the skull base. CSF leaks may also occur with LCH but are more frequently encountered in GSD. Furthermore, the presence of a mediastinal LM is suggestive of CLA.
Historically, lymphangiomatosis was an umbrella “diagnosis” that encompassed all CLA. The term, lymphangiomatosis, should be avoided since it is non-specific and could delay appropriate care or inadequately address potential complications. While our patient was diagnosed with “lymphangiomatosis” early in life, the exact type of CLA was not delineated because of diagnostic challenges. Sometimes the initial clinical presentation does not offer all the clues for an accurate diagnosis, especially in young children. Maintaining a high index of suspicion and re-analyzing the diagnosis when new complications occur is important.
Identification of the area of bone destruction and dura matter defect can be difficult. The exact mechanism in which CSF leaks develop in GSD is unclear but is hypothesized to occur when the lymphatic malformation destroys the bone cortex of vertebrae or skull base and then infiltrates into dura mater. Spinal instability secondary to significant vertebrae or skull base destruction may also result in CSF leak.
Our patient’s case was complicated by recurrent meningitis infections, a known complication of patients with GSD and skull base defects [17-19]. Although the etiology of recurrent meningitis isn’t completely understood, the highest risk of infection appears to be with osteolytic nasotemporal lesions [17]. In our patient, oropharyngeal bacteria presumably translocated from the sphenoid sinus through porous eroded bone and into the CSF space through the dural mater defect. Our patient was also asplenic, predisposing him to infection with encapsulated bacteria. Comprehensive immunologic testing did not identify a primary immunodeficiency.
Although our patient had thrombocytopenia, his platelet count normalized after splenectomy, consistent with ITP. Furthermore, he had normal fibrinogen levels and no hemorrhagic effusions, arguing against a diagnosis of KLA.