A 69-year old previously healthy man presented in January 2011 with a two-week history of behavioural disturbances including anxiety, hypochondriac complaints and motor hyperactivity. Since the patient never had had any psychiatric symptoms his general practitioner chose to refer him to a neurologist rather than to a psychiatrist.
Neurological examination was unremarkable except for slight motor agitation. During the next days the patient had several episodes of sudden onset of unresponsiveness and orofacial automatisms. The episodes lasted for 30 to 60 seconds.
Magnetic resonance imaging (MRI) of the brain including T2-weighted and fluid attenuated inversion recovery (FLAIR) sequences was unremarkable (fig 1,2). However, EEG revealed epileptic activity in the left frontotemporal area (fig. 3). Fluorodeoxyglucose positron emission tomography (FDG PET) showed an area of increased metabolism in the left mesial temporal lobe (fig 4). Analysis of cerebrospinal fluid was notable for slightly increased protein (0.66 gram/litre) but was otherwise normal. A whole body PET computed tomography was without signs of systemic malignancy. An extensive search for infectious causes and anti-neuronal antibodies, including LGI1, CASPR2, NMDAR and anti-GAD65, was negative. We made a diagnosis of non-paraneoplastic limbic encephalitis without detectable anti-neuronal antibodies.
Treatment with steroids (1000mg prednisolone intravenously once daily during 3 days, followed by 1mg/kg oral prednisolone once daily), intravenous immunoglobulins (IVIG; 0.4g/kg once daily for 5 days) and levetiracetam (500mg twice daily) resulted in a return to baseline within the following months. Repeated neuroimaging in April and May 2011 revealed normal T2-weighted imaging (not shown), hyperintense signal change on MR diffusion weighted imaging (DWI; fig 5a,b) and FLAIR sequences (fig 6) but decreasing signal change on FDG PET (fig 7). The patient discontinued steroid treatment, which resulted a few weeks later in an episode of maniac mood disturbance and motor restlessness. IVIG and steroids were prescribed once again. Levetiracetam was replaced by lamotrigine (initially 25mg once daily, tapered gradually to 100mg once daily). Lamotrigine was chosen because of its mood-stabilizing effects. The patient made a near-complete recovery within the following two weeks. In August 2011 FDG PET showed further decrease of temporal lobe metabolism (fig 8). MR FLAIR and DWI in September revealed normalized signal in the left temporal lobe (fig. 9, 10). The patient is currently on low-dose prednisolone (12.5mg once daily) and lamotrigine. The last follow-up has been unremarkable except for minor irritability. A new MRI is scheduled for January 2012 in order to assess for possible temporal lobe atrophy.
*Legend figure 3: Electroencephalogram revealed epileptic spike and wave activity (red squares) bilaterally with predominance in the left prefrontal lead. This figure shows a 12 second long recording during which the patient was having a seizure with orofacial automatisms and unresponsiveness.
Case presented by:
Daniel Kondziella MD, PhD
Fellow of the European Board of Neurology
Staff neurologist
Department of Neurology
Rigshospitalet, Copenhagen University Hospital
DK-2100 Copenhagen
Comment by Massimo Filippi
Previous reports show that encephalitis frequently manifests as FDG-PET hypermetabolism (see below). Together with previous studies, this case suggests that hypermetabolism may be a sign of epileptic activity.
I have some questions regarding this interesting case:
1. Did the patient undergo a formal neuropsychological battery? It would be useful to see specific cognitive test performances (e.g., memory).
2. The authors should mention whether a Voltage-Gated Potassium Channel Antibodies titre has been performed.
3. Has hyponatraemia in the serum been excluded?
4. Thyroid function and autoantibodies tests should be reported.
5. How was the peripheral nervous system?
Fakhoury T, Abou-Khalil B, Kessler RM (1999) Limbic encephalitis and hyperactive foci on PET scan. Seizure 8(7):427-431.
Kassubek J, Juengling FD, Nitzsche EU, Lucking CH (2001). Limbic encephalitis investigated by 18FDG-PET and 3D MRI. J Neuroimaging 11:55-59.
Lee BY, Newberg AB, Liebeskind DS, Kung J, Alavi A (2004). FDG-PET findings in patients with suspected encephalitis. Clin Nucl Med 29(10):620-625.
Chatzikonstantinou A, Szabo K, Ottomeyer C, Kern R, Hennerici MG (2009). Successive affection of bilateral temporomesial structures in a case of non-paraneoplastic limbic encephalitis demonstrated by serial MRI and FDG-PET. J Neurol 256:1753-1755.
Massimo Filippi is Chair of the EFNS Scientist Panel on Neuroimaging and Director of the Neuroimaging Research Unit, INSPE, Division of Neuroscience, San Raffaele Scientific Institute and Viat-Salute San Raffaele University, Via Olgettina, 60; 20132 Milan, Italy
Comment by Angela Vincent and Christian Bien
This male patient presented with frequent focal seizures plus neuropsychiatric disturbance at the age of 69. Although there was no overt amnesia, and the MRI was negative initially, in the absence of any other identifiable cause (eg infection, tumours), these features in an otherwise healthy person of this age should raise suspicion of VGKC-Ab associated limbic encephalitis.
The lack of clear CSF abnormalities, merely a slightly raised protein, is typical of this syndrome. The authors did search for LGI1 and CASPR2 antibodies but they were negative. LGI1 and CASPR2 are antigens that are part of the VGKC-complex and major targets for the antibodies that immunoprecipitate VGKC-complexes from brain tissue extracts (as used in the radioactive “VGKC” antibody assay since the 1990s). However, these antigens alone do not cover the full spectrum of VGKC-complex Ab positivity. In our experience around 30% of patients with limbic encephalitis and VGKC-complex antibodies are not positive for LGI1 or CASPR2 (Irani et al Brain 2010), and it would have been better to request the VGKC-complex Abs first, and only do the LGI1 and CASPR2 antibodies if they were positive.
They also tested for NMDAR-Abs. These antibodies are often associated with seizures and behavioural disturbance initially but in many cases the patients progress to a much more complex encephalopathy which was not the case here. Moreover, they appear to be uncommon in older patients, particularly males.
The authors remark on the lack of a tumour; it is now clear that the majority of patients with limbic encephalitis, particularly presenting in this age range, are not tumour-associated.
Angela Vincent is Professor of Neurology at the Nuffield Department of Neurosciences, University of Oxford, John Radcliffe Hospital, Oxford OX3 9TH, United Kingdom
Christian Bien, PD Dr., Krankenhaus Mara, Epilepsie-Zentrum Bethel, Maraweg 21, 33617 Bielefeld, Germany
1 comment
I thank Prof. Filipo, Prof. Vincent and Prof. Bien for their important comments.
A formal neuropsychology battery has been scheduled; indeed, on the last follow-up the patient’s wife has been complaining about her husband’s forgetfulness. Hyponatremia in the serum has been excluded. The peripheral nervous system and thyroid function were normal, thyroid autiantibodies were negative.
In the meantime we have ordered a second antineuronal antibody analysis from a different laboratory. As a matter of fact, whereas the first analysis for LGI1 antibodies was negative, for unknown reasons, the second analysis has been reported as positive. Thus, we now believe that the patient has LGI1 related limbic encephalitis. I apologize for the confusion, but at least we now have a final diagnosis.
Kind regards from Copenhagen,
Daniel Kondziella