Positron emission tomography in Rasmussen's encephalitis

18 F-FDG PET/CT Imaging Findings in Adult-Onset Rasmussen Encephalitis

ABSTRACT

Rasmussen encephalitis is a rare condition characterized by chronic, progressive encephalitis. Typically, it involves one cerebral hemisphere and manifests through intractable epileptic seizures. Its occurrence in adults is notably infrequent. In this case, we presented the observations from an 18 F-FDG PET/CT scan of a 70-year-old man diagnosed with Rasmussen encephalitis, where MRI displayed nonspecific findings.

Crossed cerebellar diaschisis on 18F-FDG PET: Frequency across neurodegenerative syndromes and association with 11C-PIB and 18F-Flortaucipir

We used 18F-FDG-PET to investigate the frequency of crossed cerebellar diaschisis (CCD) in 197 patients with various syndromes associated with neurodegenerative diseases. In a subset of 117 patients, we studied relationships between CCD and cortical asymmetry of Alzheimer's pathology (β-amyloid (11C-PIB) and tau (18F-Flortaucipir)). PET images were processed using MRIs to derive parametric SUVR images and define regions of interest. Indices of asymmetry were calculated in the cerebral cortex, basal ganglia and cerebellar cortex. Across all patients, cerebellar 18F-FDG asymmetry was associated with reverse asymmetry of 18F-FDG in the cerebral cortex (especially frontal and parietal areas) and basal ganglia. Based on our operational definition (cerebellar asymmetry >3% with contralateral supratentorial hypometabolism), significant CCD was present in 47/197 (24%) patients and was most frequent in corticobasal syndrome and semantic and logopenic variants of primary progressive aphasia. In β-amyloid-positive patients, mediation analyses showed that 18F-Flortaucipir cortical asymmetry was associated with cerebellar 18F-FDG asymmetry, but that cortical 18F-FDG asymmetry mediated this relationship. Analysis of 18F-FDG-SUVR values suggested that CCD might also occur in the absence of frank cerebellar 18F-FDG asymmetry due to symmetrical supratentorial degeneration resulting in a bilateral diaschisis process.

Functional neuroimaging in Rasmussen syndrome

PURPOSE

For a diagnosis of Rasmussen syndrome (RS), clinical course together with electroencephalography (EEG) and magnetic resonance imaging (MRI) findings are considered important, but there are few reports on functional neuroimaging. This study investigated cerebral blood flow (CBF)-single photon emission computed tomography (SPECT), central benzodiazepine receptor (BZR)-SPECT, and fluorine-18 fluorodeoxy glucose-positron emission tomography (FDG-PET) in RS patients, and correlated neuroimaging results with MRI and pathological findings.

METHODS

Twenty-three patients diagnosed with RS according to Bien's (2005) diagnostic criteria (including 12 patients with a histological diagnosis) were studied. CBF-SPECT, BZR-SPECT and FDG-PET images were visually evaluated, and the findings correlated with MRI and histological findings.

RESULTS

Hypoperfusion areas were observed in 16 of 22 patients by interictal CBF-SPECT. Hyperperfusion areas were observed in 10 of 12 patients by ictal CBF-SPECT, which correlated with ictal onset area by ictal EEG (IOAE). In the limited data of BZR-SPECT in nine patients, lowered uptake was detected in all nine patients, including two with no MRI abnormalities. Lowered glucose metabolism was observed in affected areas in all five patients by FDG-PET. Histological examination revealed findings of chronic encephalitis in all 12 patients examined, concomitant with focal cortical dysplasia in five patients.

CONCLUSION

In RS patients, functional neuroimaging reveals clear abnormal findings, even before the appearance of MRI abnormalities. BZR-SPECT and FDG-PET could detect the IOAE efficiently even in the absence of MRI abnormalities, while interictal CBF-SPECT occasionally failed to detect IOAE if MRI was normal. Based on BZR-SPECT, refractory epileptic seizures in RS may suggest possible impairment of inhibitory neurons.

Arterial spin labeling hyperperfusion in Rasmussen's encephalitis: Is it due to focal brain inflammation or a postictal phenomenon?

BACKGROUND AND PURPOSE

The study evaluated the utility of arterial spin labeling (ASL) perfusion imaging in Rasmussen's encephalitis (RE).

MATERIAL AND METHODS

The hospital electronic database was searched using the search words "encephalitis," "autoimmune encephalitis" and "Rasmussen's encephalitis" for the period of 1 Jan 2015 to 31 Jan 2017. Clinically diagnosed cases of RE for which epilepsy protocol magnetic resonance imaging (MRI) with perfusion imaging (ASL) performed on a 3T scanner were retrieved. The diagnosis of RE was based on Bien's criteria (Bien et al., 2005). We obtained patient's demographic details, clinical features, electrophysiological studies, and follow-up data from electronic hospital records.

RESULTS

We included nine patients with RE of whom seven patients showed increased perfusion, and two patients decreased perfusion. Among these patients, MRI changes of gyral hyperintensity without volume loss corresponded to regional ASL hyperperfusion in six patients and ASL hypoperfusion in one patient. Two patients who showed ASL hypoperfusion had corresponding atrophy on MRI. Eight patients of RE had epilepsia partialis continua (EPC) or daily seizures, and one patient was seizure-free post-surgery. Five patients showed a concordance of ASL hyperperfusion with clinical ictal onset zone. Among the seven patients with ASL hyperperfusion, the finding was concordant (complete or partial) with the electroencephalogram (EEG) ictal onset zone in six patients and with interictal epileptiform discharges (IED) in seven patients.

CONCLUSION

Increased perfusion in ASL of the involved brain parenchyma in RE is a common MRI finding and may be due to either active inflammation of the brain involved or a seizure-related finding.

The Janus-faced nature of Rasmussen's encephalitis

Rasmussen encephalitis (RE) is an inflammatory unilateral progressive medically refractory epilepsy associated with hemiparesis, cognitive dysfunction, and hemispheric atrophy. Here, we present 2 cases from our institution that demonstrate the dual nature of RE in 2 similarly aged children. Overall, 2 types of RE have been described: type 1 has a short prodromal phase and more explosive onset and type 2 has a longer prodromal of partial seizures followed by hemiparesis and atrophy. Younger patients are more likely to fit into the type 1 presentation and have been described as more likely to have dual pathology. Perhaps the patients with a more acute onset are more likely to have a dual pathology as is found in our 2 cases. We review the typical findings in RE and discuss current treatment options, highlighting new experimental treatments.

Correlation between magnetic resonance imaging and histopathologic grades in Rasmussen syndrome

The aim of this study was to investigate the correlation between magnetic resonance imaging (MRI) and histopathologic findings in Rasmussen syndrome. Serial MRIs were obtained for five patients who had histologically proven Rasmussen syndrome. The histopathologic grades of the lesions were subdivided into phases: active 1-3, resolving 1-3, and chronic inflammatory. The images were then correlated with histopathologic findings. Neuropathologic findings in the central areas on MRI demonstrated only the chronic and resolving grades, but active inflammatory abnormalities were present not only at the margins of the lesions, but also in areas of subtle signal abnormality on MRI. Atrophic areas on MRI exhibited all grades of histopathologic abnormalities, but chronic and resolving grades were predominant. Seizure duration of less than 6 months was associated with very active grades, duration of 1-2 years with variable grades, and duration greater than 6 years with chronic and resolving grades only. The MRI images correlated highly with histopathologic analysis. These findings suggest that the lesions initially arise from one site in the brain, and so support the centrifugal spreading theory of this disease. Findings also suggest that the margin rather than the center of the MRI abnormality may be the most ideal site for biopsy.