Limbic encephalitis (LE) is characterized by a subacute onset of episodic memory impairment, disorientation, agitation, seizures and histological evidence of mesial temporal lobe inflammation. It is usually considered to have a paraneoplastic origin and is associated with specific autoantibodies in various cancers [1]. The non-paraneoplastic form has similar clinical and neuroradiological presentation. In these cases antibodies usually target voltage-gated potassium channels (VGKC) or glutamic acid decarboxylase and are, therefore, accepted as immunotherapyresponsive syndromes. Potassium channel antibodies were reported in two patients with reversible LE for the first time in 2001 [2]. Very recently, however, Dalmau et al. showed that the autoantigen associated with LE previously attributed to VGKC is, in fact, leucine-rich, glioma inactivated 1 (LGI1), a neuronal secreted protein [3]. Since this is a relatively rare condition, there have been very few publications on the clinical findings, treatment and follow-up results in this group of patients [3–6] and the optimum treatment regimen is still unknown. In this case report we present a young male patient who did not respond to plasma exchange therapy, but improved rapidly and significantly after IV pulse steroid infusion. A 31-year-old man was admitted for confusional state that had started 1 month ago. He had fever, vomiting and headache at the onset. His past medical history was unremarkable. Physical and neurologic examinations were normal apart from confusion and short-term memory deficits. Cranial MRI revealed hyperintense signals in bilateral hippocampi and the amygdaloid nuclei with little contrast enhancement (Fig. 1). At the onset of his complaints he was admitted to another hospital where he was diagnosed with herpes encephalitis and was given acyclovir without any benefit. In our clinic, blood studies including complete blood count, liver and renal function tests, tumor markers and paraneoplastic antibody analysis (anti-Hu, -Yo, -Ri, -Ma, -CV2) were all normal. Serum electrolyte analysis revealed mild hyponatremia (132 mEq/L). Cerebrospinal fluid (CSF) protein and glucose levels were normal. Analysis of CSF cytology showed increased mononuclear cells. Herpes PCR was negative. Thorax– abdomen CT scans, scrotal-pelvic ultrasonography and whole body fluorodeoxyglucose-positron emission tomography were negative for malignancy. He had two temporal lobe seizures during his follow up. Electroencephalography revealed bilateral generalized mild slow wave paroxysms, without epileptiform discharges. The patient was put on oxcarbazepine (OXC), and did not have any further seizures. In order to investigate the presence of nonparaneoplastic LE, serum and CSF samples were sent to Dr. J. Dalmau’s laboratory where antibodies for NMDA receptors and VGKC antibodies were studied. Meanwhile, the patient had six exchanges of 1.5 plasma volumes on alternate days. We could not give intravenous immunoglobulin (IVIg) because his IgA levels were very low. No clinical response was obtained in the first week after treatment with plasmapheresis and the patient was discharged from hospital. One month later we were informed that he had antibodies against LGI1. The patient was hospitalized again for IV pulse steroid treatment. At that time, his orientation in time and space had partially improved; however, he still had severe memory deficits. He had difficulty in remembering some recent events. After the B. Kaymakamzade T. Kansu E. Tan N. Dericioglu (&) Department of Neurology, Faculty of Medicine, Hacettepe University, Ankara, Turkey e-mail: noroloji2011@gmail.com
Read full abstract