Neurology India, Vol. 58, No. 3, May-June, 2010, pp. 468-470
Changing signal intensity of a craniopharyngioma
Somenath Chatterjee1, Chandrasekharan Kesavadas1, Girish Menon2, Suresh Nair2, Vishnupuri Venkatraman Radhakrishnan3
1 Department of Imaging Sciences and Interventional Radiology, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Trivandrum - 695 0ni11, India
Date of Acceptance: 01-Feb-2010
Code Number: ni10119
Craniopharyngiomas can present as suprasellar cystic lesion with varied imaging appearance. In magnetic resonance imaging using T1-weighted sequence, the cyst can show hypointense, isointense or hyper intense signals depending on the cyst content. We report a case where the T1 signal intensity of a craniopharyngioma changed over time. The hypointense lesion had become hyper intense in the follow-up scan after six months. Such change in signal intensity is described with Rathke's cleft cyst but has not been reported with craniopharyngioma. The possible reason for this change in signal intensity is discussed.
Keywords: Alteration of signal, craniopharyngioma, MRI, T1WI
Craniopharyngiomas are epithelial tumors arising in the path of craniopharyngeal duct. The majority (94-95%) has a suprasellar component: purely suprasellar, 20-41%; both supra and intrasellar, 53-75%, , whereas the purely intrasellar ones represent the least common variety (5-6%).  Histologically, two primary subtypes have been recognized, the adamantinomatous and the papillary, but transitional or mixed forms have also been described. , Macroscopically, they show cystic and/or solid components. ,, Signal intensity of the cystic component varies according to cyst content. 
A 51-year-old-lady presented to another facility in November 2008 with decreased vision in the both eyes of one month duration. The extra ocular movements were normal. She had early right optic atrophy on ocular fundal examination. Magnetic resonance imaging (MRI) revealed cystic lesion in suprasellar cistern, more towards the right side compressing the optic chiasm. It could be identified separately from the pituitary gland with hypointense diaphragma sella in between. It showed hypo intensity in T1-weighted image (T1-WI) and hyper intensity in T2-WI and showed only a thin rim of enhancing soft tissue in the superior and superomedial aspect [Figure - 1].
She was operated at the other facility in December 2008 by transnasal transsphenoidal approach. Surgical note revealed distorted nasal anatomy and severe bleeding on aspiration after opening the sellar floor. Hemostasis was achieved and the procedure was abandoned. The bleeding probably was related to carotid injury. Postoperatively, she had total left ophthalmoplegia and worsening of left sided visual disturbance. She was started on intravenous methylprednisolone and the left third cranial nerve palsy improved gradually. Unfortunately, her vision did not improve and there was no light perception on left eye.
MRI done postoperatively did not reveal any significant change in signal intensity of the mass lesion compared to the initial scan. It showed mild increase in size of the mass with appearance of subtle lobulations at the margin [Figure - 2]. MRA excluded any major arterial injury. In the next five months she did not have any new complaints. Again, in June 2009 she noticed progressively increasing vision loss in the right eye mainly for distant objects. On examination there was right temporal field cut which was not present in previous visual field charting.
Repeat MRI revealed increase in size and also change in the signal intensity. It had lobulated margin with retrosellar as well as presellar extension. It also revealed hyper intense signal in both T1-WI and T2-WI [Figure - 3]. The tumor was approached through a right pterional craniotomy and transylvian route. The lesion was seen in the suprasellar cistern occupying the optico carotid, interoptic and lateral carotid spaces. The lesion was cystic with machine oil like contents characteristic of a craniopharyngioma. The cyst wall was densely adherent to the right optic nerve, carotid artery and its perforators. A near total excision of the cyst wall was done. A small part adherent to the carotid artery was not removed. Histopathology was suggestive of craniopharyngioma [Figure 4].
Craniopharyngioma is purely or predominantly cystic in 46-64% of the cases, , purely or predominantly solid in 18-39% ,, and mixed in 8-36%. , Cyst of the craniopharyngioma may show variable signal intensity according to the cyst content. The cystic component is usually hypointense on T1-W and hyperintense on T2-W sequences.  Protein, cholesterol, and methemoglobin may cause high signal on T1-W images,  whereas very concentrated protein, calcification, and various blood products may be associated with low T2-W signal.  Postcontrast T1-weighted images may demonstrate thin peripheral contrast-enhancing rim of the cyst.  Interestingly, edema in the adjacent brain parenchyma (representing a reaction to the craniopharyngioma itself or a focal disturbance in the cerebrospinal fluid flow) spreading along the visual pathways may be present.
Increased signal intensity of cystic fluid in craniopharyngiomas on T1-weighted MR images can be caused by a protein concentration greater than or equal to 9,000 mg/dL (90.00 g/L), the presence of free methemoglobin, or both.  The initial and second MRI in our patient showed T1 hypo intensity with T2 hyper intensity suggestive of a fluid low in protein or methemoglobin content. However, in the next MRI it showed change of T1 signal from hypointense to hyper intense with T2 signal maintained as hyper intense. On a literature search we could not find any case with the description of a black cystic craniopharyngioma on T1WI turning white on serial imaging.
There were two isolated case reports of neuroenteric cysts with change of signal on follow up imaging along with an increased size. [ni11], In the more recent report,  the cyst signal in T1-WI was initially hyper intense and on follow up imaging became isointense. The authors postulated that the signal changes and lesion expansion were due to the breakdown of balance between epithelial secretion and absorption. Alternate possible explanation suggested was dilution of the cyst contents by cerebrospinal fluid (CSF). Thinning of the cyst wall and an osmotic response to cyst content might have caused inward permeation of CSF, resulting in dilution of cyst content. There was, however, no evidence to support these explanations from the surgical findings or histological examination of the cyst wall. Recent hemorrhage also might have caused this expansion and signal change on the MR images, but there was no evidence of previous hemorrhage at surgery.  Our case, along with increased size showed a marked change in T1-WI signal. It is almost opposite to the previously described signal change for neuroenteric cyst. Per-operative machinery oil like fluid was aspirated from the cyst. There was no evidence of frank hemorrhage within.
The wall of craniopharyngioma is layered with the innermost lining facing the cyst lumen being squamous epithelium.  The flat squames are desquamated singly or in distinctive stacked clusters forming nodules of "wet" keratin. In our case it seems likely that there has been more desquamation in the cyst lumen in the interval period between the second and third scan and it has resulted in cyst growth as well as signal change. Vascular endothelial growth factor has been detected in the epithelial cells of these tumors,  and the degree of its expression is probably related to the development of macroscopic cysts.  Notably, recurrent craniopharyngiomas, when compared with nonrecurrent ones, show higher micro vessel density values,  so there seems to be a relation between growth potential of a cyst with VEGF expression and resultant micro vessel density. Hence it seems likely that cyst growth may also be related to more VEGF expression.
In our case, along with cyst growth there must have been increased desquamation of the lining epithelium contributing to the change in signal. We are not able to comment definitely on any probable trigger which might have led to these events. Whether the bleeding due to initial surgery changed the local milieu and played a precipitating role for the chain of events, seems only a postulate. Otherwise it can be purely a part of natural course of the cyst.
To conclude, we report a rare case of craniopharyngioma with altered signal intensity in follow-up MR scan. Alteration of signal intensity had made us think of the possibility of Rathke's cleft cyst in the pre-operative diagnosis. This case report illustrates that a craniophryngioma can change signal intensity similar to a Rathke's cleft cyst.
Copyright 2010 - Neurology India
The following images related to this document are available:
Photo images[ni10119f3.jpg] [ni10119f2.jpg] [ni10119f1.jpg]