MR Imaging Appearance of Ruptured Rathke Cleft Cyst and Associated Bone Marrow Enhancement

DISCUSSION

Our study is the first to describe enhancing basisphenoid bone marrow below the sella as an imaging finding of rRCC. We present 7 cases of rRCC, all of which showed bone marrow enhancement. This finding is potentially important because the leading differential for an rRCC is cystic pituitary adenoma, which, in our limited experience reported here, does not have basisphenoid bone marrow enhancement.

RCCs arising from failed regression of the cleft between the adeno- and neurohypophysis are frequently incidental and asymptomatic findings.¹⁰ RCCs can become symptomatic and present with headaches, visual deficits, or hormonal abnormalities and have an overlapping imaging appearance with cystic adenoma.¹¹ Park et al¹² created a diagnostic model using MR imaging to differentiate cystic pituitary adenoma and RCC, but the model does not specifically assess rRCC. Prior case reports of rRCCs have not described specific findings to differentiate them from a cystic adenoma.

Intrasellar rupture of an RCC was described in 1988,¹⁰ with associated marked inflammation of the pituitary gland thought to be secondary to the spillage of RCC contents. Numerous additional case reports have since been presented in the literature of symptomatic patients with xanthogranulomatous changes of the pituitary gland.^{3⇓⇓⇓⇓-8,13} It is thought that mucin leaking out of the rRCC can trigger a cascade of surrounding inflammation.¹⁴ The abnormal bone marrow enhancement in our cases, therefore, likely represents an imaging manifestation of the adjacent inflammatory changes. This is most evident in the patient presented in Fig 2, in which the native pituitary was markedly enlarged with suprasellar extension, preoperatively thought to be an adenoma. In this case, the RCC was found to be ruptured, and the pituitary biopsy findings were thought to represent reactive hypophysitis. Further testing was not compatible with lymphocytic hypophysitis or immunoglobulin G4–related disease. Given that the bone marrow enhancement is likely secondary to the inflammatory cascade set off by the RCC rupture, we would not expect to see abnormal enhancement with nonruptured RCCs (Fig 4).

• Download figure
• Open in new tab
• Download powerpoint

FIG 4.

Example of a pathology-proved nonruptured RCC. The precontrast T1-weighted (left) images show normal signal of the basisphenoid bone marrow edema below the T1-hyperintense RCC. The fat-saturated postcontrasted image (right) shows normal basisphenoid bone marrow without abnormal enhancement.

Similar bone marrow imaging findings have been recently described with granulomatous hypophysitis,¹⁵ which is a noncystic inflammatory process of the pituitary gland. In that article, 100 pituitary adenomas were evaluated, none of which had basisphenoid bone marrow enhancement. Therefore, bone marrow enhancement is reflective of an inflammatory process, regardless of the cystic or solid nature of the sellar process. The current study is the first to describe bone marrow enhancement associated with an rRCC.

Differentiating between a RCC and a cystic pituitary adenoma can be challenging because they often have an overlapping imaging appearance. Pituitary adenomas have been described as off-midline in position and as having fluid levels, a hypointense T2 rim, and septations,¹² while RCCs are associated with T2-hypointense nodules and are typically midline in position.¹¹ Our study is limited by the small number of patients; however, none of our cystic adenomas had basisphenoid bone marrow enhancement, nor has this previously been described in the literature. The differential for cystic lesions of the pituitary gland extends beyond RCC and cystic adenoma, but such lesions are expected to demonstrate different imaging findings and clinical presentations. Adamantinomatous craniopharyngiomas occur more frequently in children and are typically suprasellar and calcified.¹⁶ Pituitary abscess could also appear as a cystic lesion and, due to marked surrounding inflammatory changes, could also have bone marrow signal changes; however, an abscess should also present with restricted diffusion,¹⁷ whereas an RCC would not.¹⁸ Our patients did not have DWI for us to analyze. The broad category of hypophysitis has many primary and secondary causes, ranging from immunoglobulin G4–related disease to infectious causes such as tuberculosis.¹⁹ While these entities, outside of granulomatous hypophysitis,¹⁵ have not been previously described as having bone marrow enhancement, it is certainly possible, given the associated inflammation. These should present as solidly enhancing enlargement of the pituitary gland, however, rather than the cystic appearance expected for an RCC.

Diagnosing basisphenoid bone marrow inflammatory changes on MR imaging is predicated on having bone marrow below the sella (non-type 4 sphenoid pneumatization) and the correct MR imaging protocol, and this requirement limits generalization of this finding to all pituitary cases. The findings are most easily seen when the T2-weighted and postcontrast T1-weighted images are fat-saturated. Even in the absence of fat saturation, however, precontrast T1-weighted images can show dark bone marrow signal (Fig 3). We believe that the presence of MR signal abnormalities consistent with basisphenoid inflammation is an important finding that can potentially help to identify an rRCC in the appropriate clinical setting. Although the small number of patients included in this report limits its generalizability, it can serve as a starting point for future studies that include more patients.