Aunt Mickey (They Look the Same until You Undress Them). Meningioma or Something Else?

73 year old lady presented with six weeks history of odd behavior, increasing apathy, expressive aphasia, and mild headache. An MRI including diffusion and perfusion imaging was obtained. MRI revealed a T1 hypointense (to cortex) and slightly T2 hyperintense (to cortex) extra-axial mass in the left frontal region. There were multiple central T2 hyperintense areas. On post contrast T1 weighted sequence, there was intense enhancement of the mass except the central T2 hyperintense areas.There were prominent diffusion restriction and high rCBV in most parts of the tumor in diffusion and perfusion imaging respectively.

At this point, what is your diagnosis? Meningioma…… right?

Our pre-operative diagnosis was also meningioma. Only concern was that the diffusion restriction was little too much for a meningioma, even for a densely cellular anaplastic meningioma.

There were two surprises for us both from surgeons as well as from pathologists. When the surgeons opened the dura, the mass was intra-axial! When the pathologists saw the tumor under microscope, they found diffuse large B-cell lymphoma with very high proliferative index (90-95%) and officially they called it “Large B-cell lymphoma, diffuse, with high proliferative rate and intermediate features between Burkitt lymphoma and large cell lymphoma (WHO classification 2008)”.

This is a very interesting case because the tumor grew without following the ‘basic rules of neuroradiology’. Though the tumor arose from intra-axial compartment the way it enlarged and its internal morphology gave it a look of an extra-axial mass. This is not an uncommon dilemma of day to day neuroradiology practice. The best way to differentiate between pathologies at a given anatomic location is to execute a mental workout (I call it ‘Curé’s algorithm’) which is consisting of a) correct identification of the compartment from where the lesion arises, b) recall of normal structures of that compartment, c) recall of possible pathologies that can arise from the normal structures of that compartment, d) careful evaluation of the imaging appearances of the lesions in different imaging sequences/modalities and finally e) to perform an intensive mental database search for ‘curve fitting’ of the imaging appearances with the possible pathologies considered before. As the first step of the ‘Curé’s algorithm’ went wrong in this case, we ended up a wrong diagnosis even though there were subtle clues to the right diagnosis.

Though the patient presented with relatively short duration of frontal lobe symptoms and the tumor had prominent diffusion restriction we did not consider primary CNS lymphoma (PCNSL) as our provisional diagnosis because a) we thought the tumor was extra-axial, b) the incidence of primary leptomeningeal lymphoma (LL) is exceedingly rare and typical LL does not look like this, c) the tumor had high rCBV, a perfusion characteristic perfusion characteristic typically not seen in lymphoma and finally d) all imaging appearances other than diffusion restriction was ‘best-fit’ for meningioma.

PCNSL is an extranodal manifestation of non-Hodgkin’s lymphoma limited in central neuraxis including orbit, leptomeninges, spinal cord and cranial/spinal nerves without any other systemic manifestation. PCNSL constitutes 6-15.4% of all brain tumor, more common in men and usually occurs in patients >60 years of age. PCNSL is more commonly supratentorial (87%) and solitary (66%) in immunocompetent patient. Primary leptomeningeal lymphoma is a rare subtype of PCNSL limited to the meninges. Primary ocular lymphoma is another rare subset of PCNSL that involves vitreous, sub-retinal space or anterior chamber without systemic or CNS involvement. Up to 90% of these patients subsequently develop CNS lymphoma. Some authors also consider neurolymphomatosis as a variant of PCNSL. Neurolymphomatosis is very difficult to diagnose on MRI, particularly if the patient has no known lymphoma, either systemic or PCNSL.

The imaging appearance of the tumor depends upon the immune status of the patient. In immune competent patients, PCNSL is solid, T1 hypointense and iso- to slightly hyperintense on T2 weighted sequence. It can involve any area of brain but frontal lobe and basal ganglia are most commonly involved. It enhances brightly and homogenously with contrast. On diffusion weighted imaging, there is prominent diffusion restriction due to intense ‘packing’ of the tumor. Lymphoma is usually cold on perfusion imaging. PCNSLs in immunocompromised patients typically appear as peripherally located ring-enhancing lesions with central necrosis. They are often multiple. There may be diffusion restriction only at the periphery.

Iwamoto FM, DeAngelis LM. An update on primary central nervous system lymphoma. Hematol Oncol Clin North Am. 2006;20:1267-85

Shenkier TN. Unusual variants of primary central nervous system lymphoma. Hematol Oncol Clin North Am. 2005;19:651-64

This entry was posted in Aunt Mickeys, Brain and tagged , , , . Bookmark the permalink.

2 Responses to Aunt Mickey (They Look the Same until You Undress Them). Meningioma or Something Else?

  1. Asim Asim says:

    That was an excellent discussion. I appreciate this in-depth analysis of the case.

    These two images were essential for the post. Sorry for not including them in the original post.

  2. Girish Fatterpekar says:

    I do like the diagnosis of meningioma. However, let us revisit this case and evaluate it once again. Will we still entertain the diagnosis of meningioma? Here we go: On all the submitted images, the lesion demonstrates an acute (and not obtuse) angle at the dura-parenchymal interface (analogy of pleura-parenchymal interface when evaluating a lung versus pleural based lesion), the cortex is swollen, and no distinct CSF cleft is seen. These findings favor an intra-axial pathology. Now, let us look at the DWI – restricted diffusion! There are several lesions that can cause restricted diffusion. However, when it is seen so homogenously in a lesion of this size, lymphoma (high nuclear to cytoplasmic ratio) strongly merits consideration. I bet if you have a non-contrast CT study, you would see this lesion as being hyperdense. Now, let us move to the contrast-enhanced portion of the study, there is no any evidence for adjacent dural enhancement i.e. dural tail. Again, it is a non-specific sign of meningioma but, should be seen more often than not in dural-based lesions. Further, I always look for secondary signs such as hyperostosis and pneumosinus dilatans, difficult to appreciate on MR, but can be seen. These too were not noted in this lesion. Hence, though I would have considered the diagnosis of meningioma, I would have thought of lymphoma as a more likely possibility. The high rCBV as reported is not typically seen in primary CNS lymphoma. However, I would have loved to see the images. Again, this is all with a retrospectoscope and therefore the case still remains an aunt Mickey. As diagnostic sleuths we need to look out for those subtle foot-prints that can aid us in identifying the correct culprit. The goal is to convert aunt Mickey to aunt Minnie. Hope the discussion was helpful.

Leave a Reply