 Well, now we undertake our next teaching review of brain tumors. And we'll start out with a 45-year-old with cognitive decline and turn right to the images which are axial T2 sequences. And I'll let you gaze at them for a moment. I'm sure you've noticed that there is mass effect on the midline and that there is amorphous signal alteration and relaxation prolongation through the left cerebral hemisphere. And perhaps the right cerebral hemisphere, or not, you be the judge. The next sequence, taken a little bit more craniad, shows the same phenomenon in the left cerebral hemisphere, and perhaps the right cerebral hemisphere, or not, you decide. A temporal lobe-level axial image also shows signal alteration. This time, clearly on both sides. Brain looks perhaps a little bit expanded, but you be the judge. Some proton density imaging reaffirms what we saw on our axial T2 imaging, giving you a moment to reflect on the findings, but not too long as you might be taking an exam with cross-sectional imaging of this nature. Even in the thalamus, there's some signal alteration present on the accompanying PD imaging. And here is an axial PD or flare with the same findings seen previously, a little bit more clearly, and perhaps you might have reaffirmed in your mind that there is involvement of the contralateral hemisphere. As important as the contrast, T1 weighted image, and the amount of enhancement is something you should decide on. Is it dynamically enhancing intensely? Is it enhancing moderately, or not at all? We have one more image to show you, a coronal contrast enhanced image, and it's important for you to decide on the degree, the distribution, and the intensity of enhancement. So now that you've had a chance to review this case, which is somewhat complex, but a case you might see on an exam, what is the most likely diagnosis? Is it oligodendroglioma? Is it gliomatosis cerebri? Is it metastatic disease? Is it encephalitis, or is it progressive multifocal leukoencephalopathy, A, B, C, D, and E? Let's go on to question number two. You can pause if you want to reflect on your answer momentarily, because it will have bearing on what you answer for the subsequent questions. Which is not true regarding gliomatosis cerebri? A, diffuse, B involves two or more lobes, C involves the meninges in calvarium, D frequently bilateral, E, the clinical is outweighed or overshadowed by the imaging. Question number three. Which of the following statements regarding gliomatosis cerebri is true on imaging? A, diffusion restriction present, B, preserves underlying architecture with elevated MI or myo inositol on spec MRI, C, enhances intensely, D, high cerebral blood flow due to vascular hyperplasia, E. Pat FDG or fluoro deoxyglucose shows marked hypermetabolism. Let's go back to question number one. First choice, A, oligodendroglioma. This is a spreading infiltrative disorder, but while it does involve this distribution of the anterior centrum semi ovale, it doesn't cross the midline as often as some other diagnoses, it calcifies and it bleeds. This lesion neither calcified nor bled, oligodoso 90 plus percent of the time. Gliomatosis cerebri, that will be a point of discussion since that's the correct answer. It involves multiple distributions, lobes and hemispheres of the brain. More to come, but it is hypovascular which helps differentiate it from many of the other proposed diagnoses like metastatic disease which enhances since it disrupts the blood-brain barrier. There's usually not a broad infiltrative process, but several different focal masses that enhance. That is not the diagnosis. Encephalitis, it's not an unreasonable choice. Brain swelling, diffuseness, multi-focality, however these patients are acutely ill. They're not reported as having cognitive decline, but rapid descent into cognitive decline and coma. But more importantly, encephalitis is not going to be present without some enhancement even in a cerebritis and without some either reactive or infiltrative leptomeningial enhancement neither of which are present here. Encephalitis is not the choice. Progressive, multifocal, leukoencephalopathy caused by the JC virus, a slow-acting viral infection in an immunosuppressed patient. We didn't say the patient was immunosuppressed. It is a white matter abnormality with relatively large lesions that progress from peripheral to central around the ventricles, not present here. This is a diffuse process without respect for white matter or gray matter. It is clearly not isolated to the white matter. So PML is not a reasonable choice. The answer here is B, which is not true regarding gliomatosis cerebri. A, diffuse. It is diffuse. It's diffusely infiltrative, although sometimes it can even somewhat preserve the brain architecture. So that is true, therefore it's not the correct answer. B involves two or more lobes. Our lesion involves four or five lobes. Both sides of the brain, temporal, parietal, frontal. So that one is true, and therefore can't be the answer. We're looking for the untrue answer. It involves the meninges and calvarium, where all our lesion did not involve the meninges and calvarium. PXA, pleomorphic xanthoastrocytoma might do that, but not gliomatosis cerebri. So C is not true, therefore it's the correct answer. Frequently bilateral, our lesion is bilateral, and that is true of gliomatosis cerebri. And finally, the clinical is outweighed or overshadowed by the imaging. In other words, the imaging looks worse than the patient. And that is true, therefore it's not the correct answer to question number two. The correct answer is C. Which of the following statements regarding gliomatosis cerebri is true on imaging? A, diffusion restriction present, you would think with amylignancy, this infiltrative that you would have diffusion restriction, but unfortunately not. It has an interstitial growth pattern in between the white matter tracts and sparing nerve cells in certain loci. Therefore diffusion restriction is not present. B, preserves underlying architecture with elevated MI on spec MRI. That is true. We already implied that it preserves underlying architecture when we answered question number one and also in describing its interstitial growth pattern. Sometimes it can preserve the pattern of the gyri and the sulci so profoundly that it mimics an infarct and to the inexperienced practitioner is called an infarct inadvertently or incorrectly. Seeing this happen many times. C, enhances intensely. Well no, it doesn't enhance intensely, that's untrue. Even though it's amylignancy, even though it's huge, even though it's infiltrative, our lesion demonstrated the typical appearance. Virtually no enhancement and certainly no leptomeningial enhancement. D, high cerebral blood flow due to vascular hyperplasia. Well we know that one is most likely untrue because we hardly have any enhancement. Although enhancement on MRI, if it's not dynamic, really reflects breakdown of the blood-brain barrier. And on dynamic imaging, you're looking at vascular hyperplasia. But these lesions don't have high cerebral blood flow. They don't have vascular hyperplasia. E, pet FDG shows marked hypermetabolism. You would think that a large, huge, diffusely infiltrated malignancy would have hypermetabolism, but it doesn't. It's hypometabolic, another fooler. So the correct answer in this case is B, preserves underlying architecture with elevated MI on spec MRI. Remember it preserves the architecture so much in some cases that if you don't have history or you're not tapped into the clinical, you might accidentally or inadvertently and embarrassingly call this an infarct. Let's talk a little bit more about gliomatosis cerebrite. This is a World Health Organization grade three lesion, not as malignant as glioblastoma multiforme, which is grade four. They are supertentorial and diffuse, infiltrative, multi-lobar, multi-hemispheric occurring around age 45. They cross the midline via the corpus callosum or the mass of intermediate. There is white matter involvement with gray matter extension. It may balloon or expand the gray matter, but preserve the gyroform shape of the cortex. The expansion is rather diffuse. The lesions are ill-defined, it's hard to put your finger on that mass. But rather, it appears that the entire brain is enlarged or involved. It's ill-defined, hypovascular on C plus MR and CT because of the interstitial growth pattern and the lack of blood brain barrier breakdown. It is hypo metabolic on PET, even though it's a malignancy. And on T2 imaging, as well as flair, it's still very ill-defined. A unique feature of gliomatosis cerebri is when you do diffusion tensor imaging to actually image the white matter tracks or pathways like the callosum, like the centrum semi ovale, etc. They appear preserved. They may be moved from side to side. You may see them spread apart, but they are not interrupted or destroyed. We have a malignancy, but we do not have restricted diffusion in this entity. If you do dynamic contrast enhanced MRI, you're going to see decreased cerebral blood flow with first pass MRI imaging. It's kind of contrary to your intuition for a malignancy. And on SPECT MRI, there is marked elevation of the MI peak. MI is a transporter that takes ATP back and forth across membranes. So if it's sitting there doing nothing due to obstruction and encasement by gliomatosis cerebri, preventing the MI from carrying ATP back and forth, it will accumulate in the extracellular interstitial space. Let's talk a little bit about the differential diagnosis of gliomatosis cerebri. As stated on multiple occasions, one of the most common mistakes I see residents and fellows make is to call this entity a vascular injury, a vascular cerebritis, or an infarct. That is incredibly embarrassing when you take a huge malignancy and you call it a stroke. If somebody with a stroke of this size is going to be severely acutely impaired, so the clinical alone is going to put you in the right direction. But you don't always have the clinical, but you should. Both are gyroform, but gliomatosis cerebri does not enhance, whereas infarcts in the late acute and subacute phase will give you gyroform enhancement. The degree of mass effect may be a little bit different. Infarcts may give you mass effect at around 3, 4, 5, even up to 7 days, whereas the mass effect of gliomatosis cerebri is concentric, expanding the brain in gradual fashion. Vocal symptoms of a stroke are present, such as hand weakness, whereas gliomatosis cerebri, the symptoms are rather subtle, progressive, and amorphous. Vascularitis is often on the differential diagnostic list for the same reasons. It may produce an infarct-like pattern, but often vascularitis has hemorrhage associated with it. You may see aneurysms or pseudoaneurysms on MRA. One feature of a vascularitis that is very consistent is headache, and the headache is even more severe when they bleed. Anaplastic astrocytoma, or grade 3 astrocytoma, not as large, not as multi-lobar, not as likely to cross the midline, not as diffusely expansile, and it may enhance. Viral encephalitis, this is an important one. These two patterns are very similar. It looks like somebody went in there and just stirred up the brain with an instrument in the kitchen. But in viral encephalitis, there's an acute phase. The patients have a stiff neck. There's meningial enhancement. They have meningismus. Remember, you're an imaging clinician. If it's herpes, it's going to be temporal limbic with hemorrhage in the region, but there will be enhancement of the peritemporal meninges. D or dysmyelination. Well, you'd have to have pretty massive D or dysmyelination, as in, say, severe metachromatic leukodystrophy, or if you had severe MS, you're still going to have skip areas, even in some of the severe forms of MS, like the shielders variant or the bellows variant. MS is going to give you much less mass effect. It's not going to appear to cross the midline. When it enhances, it's going to be ring enhancement that's open on the peel side. You may have optic nerve involvement and spine involvement, neither of which are present in gliomatosis cerebri. Progressive, multifocal, leukoencephalopathy. They're not going to have enhancement, just like gliomatosis cerebri, but they're going to be immunosuppressed. It likes the parieto-oxypital white matter closing down into the periventricular region. It may involve the corpus callosum. Patients with AIDS are prone to PML, but any immunosuppressive state can lead to it. Metabolic dysmyelination. Well, we already alluded to D and dysmyelination earlier, but the metabolic portion of this choice refers to mitochondrial cytopathies, perioxosomal diseases, Mela syndrome, Merf syndrome, Lay's disease, disorders of pyruvate carboxylase, other enzymes. This occurs in a younger age range. You don't see gliomatosis cerebri in 10-year-olds or even 15-year-olds. There's no mass effect. There are multiple infarcts. There may be multiple symmetric calcifications. There may be multiple symmetric areas of relaxation change in the thalamus or in the globus pallidus or putamen. They have an elevated lactate in the serum and on MR spect in these mitochondrial forms of metabolic disease. There are other metabolic diseases that we won't cover right now. Limfoma. This is a deep, irregular, strange-looking, periventricular infiltrative process, but it enhances. It enhances intensely. So, picking limfoma as a choice is 180 degrees out of phase with the appearance of gliomatosis cerebri. It's multifocal. It likes the deep brain. It likes the septum palusitum. It likes the corpus callosum. It's hyper-dense on CT. It's iso-intense on MRI because there's very little cytoplasm. Little cytoplasm, little T2 hyper-intensity. The brain architecture, pathologically, is surprisingly preserved in this nasty infiltrative disease. There is no necrosis, a helpful side. There is no neovascularity. There is blurring or wax-on, wax-off appearance of the gray white matter junction under the microscope. There are infiltrates between the myelinated nerves as opposed to direct myelin infiltration and destruction as might occur with a dismyelination or a demyelinating disease. There are varying degrees of mitosis. You can have a lower-grade form of leomatosis cerebri, World Health Organization Grade II, although that's less common. The majority of them, more than three-fourths, are World Health Organization Grade III, the classic form. So in closing, a 40-something-year-old patient with headaches, weakness, seizures, confusion, cognitive decline that is non-acute, non-enhancing, large, expansile, infiltrative, non-necrotic, non-hemorrhagic, multi-lobar, multi-hemispheric, that is, leomatosis cerebri. The mortality is 50% at one year, despite its who-grading of World Health Organization Class III. That concludes leomatosis cerebri. Let's move on to the next case.