IS Case 2: Rasmussen's encephalitis

Ben Wandtke, MD and Justin Brucker, CC4

Imaging Sciences URMC

Imaging Sciences URMC 2008
Publication Date: 2009-05-18


New onset of right-sided seizure disorder at 27 years of age. Negative neurovascular examination at that time with normal neuroimaging. Progressive decrease in cognitive function and persistent seizures over the subsequent seven year period. Brain biopsy demonstrated a T-cell mediated encephalitis consistent with clinical suspicion of Rasmussen's encephalitis.


Asymmetric atrophy of the left cerebral hemisphere, most notably within the occipital and parietal lobes. PET / CT scan demonstrates approximately 50% decreased FDG uptake in this atrophic region.


Rasmussen's encephalitis.


Rasmussen's encephalitis (RE) is a chronic, progressive inflammatory brain disorder, characterized by unilateral cerebral atrophy and seizures. Seizures are characteristically focal or multifocal, confined to the affected hemisphere, and unresponsive to antiepileptic medications. In addition, RE patients suffer progressive mental deterioration, hemaniopsia, hemiparesis, and dysphasia (if the dominant hemisphere is affected). [1-11]

Despite scarce epidemiologic data, a review of the literature confirms that this disease is universally rare. Though generally referred to as a pediatric illness, there is actually a small bimodal distribution (median ages 5.3 and 18.9 years), with as many as 10% of cases appearing in early adulthood [3]. This latter group has been described as having "adult variant" [11] or "Type 2" Rasmussen's encephalitis [3]; these patients typically exhibit a milder, slower progression of RE. Baseline developmental status prior to disease acquisition may account for the phenotypic age discrepancy, since a similar disease process is seen in both populations.

The etiology and pathogenesis of RE are unclear. One theory is that RE is virally-induced, which may explain its incidence in previously healthy individuals as well as the isolated reports of "viral" prodromes in some. A myriad of viruses (HSV, EBV, CMV, etc.) [2,11] has been identified in some brain specimens, and there are some case reports of improvement with antiviral medication [2], but a clear causal-link has not yet been demonstrated. Another widely shared idea is that RE is an autoimmune disease, either triggered by viral infection [9] or endogenous factors, such as proinflammatory cytokines [10] or glutamate-receptor autoantibodies (anti-GluR3) [2]. However, the role of a humoral response in RE is questionable for several reasons: 1) autoantibodies are not found in all RE patients, 2) anti-GluR3 is found in other epileptic syndromes [2,8], and 3) brain biopsy reveals a sparse number of B lymphocytes.

Histopathologic studies of RE usually demonstrate proliferation of CD8+ T lymphocytes, microglial nodules, astrocytosis, and neuronal loss as well as perivascular cuffing within the gray and white matter [4,6,11] Neuronal loss tends to occur late in disease, while inflammatory processes are best appreciated in the early stages of RE. The observed profile indicates a cell-mediated, cytotoxic T-cell autoimmune reaction, with components of neuroinflammation and autoantibody-mediated injury [2,4,9,10]. Histopathology might provide some insight, but it is not necessary for diagnosis [6].

Diagnosis and disease progression are determined clinically and radiologically [1,4]. According to the Montreal Neurological Institute (MNI) standards, RE can be divided into three stages [3]: 1) Prodromal -- initial seizure activity and rare mild hemiparesis, 2) Acute -- frequent seizures and hemiparetic progression, with lesional spreading within the ipsilateral cortex, and 3) Residual -- decreased persistent seizure activity and stable permanent hemiparesis. The acute phase of injury correlates with increased signal on T2-weighted MR and FLAIR. More specifically, MR imaging criteria can be broken up into 4 stages [2]: 1) increased unilateral brain volume with hyperintense T2W/FLAIR signal, corresponding to acute injury and parenchymal infiltration by inflammatory cells [1-4]; 2) normal brain volume with increased signal; 3) decreased volume unilaterally with hyperintense signal; and 4) progressive atrophy with normal signal, corresponding to the residual stage of disease, when the inflammatory process has more or less "burned out" [1-4]. The atrophy usually begins within the fronto-temporo-insular areas and ultimately spreads throughout the entire cerebral hemisphere [4,8,11]. In addition to white matter and cortical gray matter, the basal ganglia will often show disease on MR, especially the head of the caudate nucleus [4]. As a quantitative tool, one source reported that the "Hemispheric Ratio" (HR = affected hemisphere volume/normal hemisphere volume) correlated with aspects of clinical severity, suggesting another dimension to MR's utility in assessing RE [3].

Other imaging modalities are less commonly used. For instance, CT will demonstrate progressive volume loss eventually, but may be normal in the early course of disease [1]. PET and SPECT scans performed in the late stage RE have demonstrated hypometabolism and hypoperfusion in affected areas [4]. Depressed FDG-PET uptake has also been reported in early adult-variant RE [11]. Similarly, proton MR spectroscopy yields a decreased NAA signal and MRA reveals small vessels [4], all in keeping with an atrophic brain. EEG recordings show unilateral disorganized background activity, with a mixture of focal theta-delta slowing and interictal spiking [8,11].

The natural outcome for all RE patients is progressive atrophy of the affected hemisphere with residual seizure activity. Hemiparesis is inevitable, however hemishpherectomy has been shown to provide relief from the seizures [5,6]. Obviously, this is a daunting option for patients and their families, but early intervention (< 2years ) has been shown to minimize the extent of hemiparesis and overall loss of function [4-7]. These patients typically have favorable outcomes [5,6] without significant change in cognitive ability when compared to their preoperative status [7]. Many continue on with mainstream schooling, sometimes with supportive services [6]. Removal of the dominant hemisphere does impair language abilities, particularly word-finding [6,7]. Hemaniopsia and hemiparesis are unavoidable postoperative complications [6].

Less invasive therapies target the autoimmune features of RE. Steroids, intravenous immunoglobulin, and plasmapheresis have been shown to help slow progression of this disease for a short time [4], and may be helpful in patients who are poor surgical candidates or do not yet have a secure diagnosis [6,7]. Due in part to brain plasticity, very young patients demonstrate more favorable outcomes [5]. Therefore, older patients who possess a milder, stabler form of RE, may lean towards immunomodulatory therapies [5-7]. Conversely, delaying surgery with such treatments may reduce the brain-plasticity window in a young patient's recovery period [5] Immuno-ablation by cyclophosphamide has shown some promise in young patients, as well, but further studies are warranted [6].


  1. Atlas S. Magnetic Resonance Imaging of the Brain and Spine, 3rd ed. Lippincott Williams and Wilkins. 2002; Vol 1: 1122-1125.
  2. Bauer J, Bien CG, Lassmann H. Rasmussen's encephalitis: a role for autoimmune cytotoic T lymphocytes. Curr Opin Neurol. 2002 Apr;15(2):197-200. [PMID: 11923635]
  3. Bien CG, Widman G, Urbach H, Sassen R, Kuczaty S, Wiestler OD, Schramm J, Elger CE. The natural history of Rasmussen's encephalitis. Brain. 2002 Aug;125(Pt 8):1751-9. [PMID: 12135966]
  4. Chiapparini L, Granata T, Farina L, Ciceri E, Erbetta A, Ragona F, Freri E, Fusco L, Gobbi G, Capovilla G, Tassi L, Giordano L, Viri M, Dalla Bernardina B, Spreafico R, Savoiardo M. Diagnostic imaging in 13 cases of Rasmussen's encephalitis: can early MRI suggest the diagnosis? Neuroradiology. 2003 Mar;45(3):171-83. [PMID: 12684722]
  5. Daniel RT, Villemure JG. Experience with immunomodulatory treatments in Rasmussen's encephalitis. Neurology. 2004 Nov 9;63(9):1761-2; author reply 1761-2. [PMID: 15534287]
  6. Freeman JM. Rasmussen's syndrome: progressive autoimmune multi-focal encephalopathy. Pediatr Neurol. 2005 May;32(5):295-9. [PMID: 15866428]
  7. Granata T, Fusco L, Gobbi G, Freri E, Ragona F, Broggi G, Mantegazza R, Giordano L, Villani F, Capovilla G, Vigevano F, Bernardina BD, Spreafico R, Antozzi C. Experience with immunomodulatory treatments in Radmussen's encephalitis. Neurology. 2003 Dec 23;61(12):1807-10. [PMID: 14694056]
  8. Granata T, Gobbi G, Spreafico R, Vigevano F, Capovilla G, Ragona F, Freri E, Chiapparini L, Bernasconi P, Giordano L, Bertani G, Casazza M, Dalla Bernardina B, Fusco L. Rasmussen's encephalitis: early characteristics allow diagnosis. Neurology. 2003 Feb 11;60(3):422-5. [PMID: 12578922]
  9. Takahashi Y, Mori H, Mishina M, Watanabe M, Kondo N, Shimomura J, Kubota Y, Matsuda K, Fukushima K, Shiroma N, Akasaka N, Nishida H, Imamura A, Watanabe H, Sugiyama N, Ikezawa M, Fujiwara T. Autoantibodies and cell-mediated autoimmunity to NMDA-type GluRepsilon2 in patients with Rasmussen's encephalitis and chronic progressive epilepsia partialis continua. Epilepsia. 2005;46 Suppl 5:152-8. [PMID: 15987271]
  10. Tekgul H, Polat M, Kitis O, Serdaroglu G, Tosun A, Terlemez S, Kutukculer N, Ersahin Y, Gökben S. T-cell subsets and interleukin-6 response in Rasmussen's encephalitis. Pediatr Neurol. 2005 Jul;33(1):39-45. [PMID: 15876522]
  11. Villani F, Pincherle A, Antozzi C, Chiapparini L, Granata T, Michelucci R, Rubboli G, Simone I, Bellomo R, Spreafico R. Adult-onset Rasmussen's encephalitis: anatomical-electrographic-clinical features of 7 Italian cases. Epilepsia. 2006;47 Suppl 5:41-6. [PMID: 17239105]

3 images