[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 54.211.82.105. Please contact the publisher to request reinstatement.
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Case Report/Case Series
January 2014

Limbic Encephalitis Associated With Anti–Voltage-Gated Potassium Channel Complex Antibodies Mimicking Creutzfeldt-Jakob Disease

Author Affiliations
  • 1Yale Comprehensive Epilepsy Center, New Haven, Connecticut
  • 2Department of Neurology, Mount Sinai Hospital, New York, New York
JAMA Neurol. 2014;71(1):79-82. doi:10.1001/jamaneurol.2013.5179
Abstract

Importance  Limbic encephalitis that is associated with anti–voltage-gated potassium channel complex (VGKCC) antibodies (VGKCC syndrome) is an autoimmune, usually nonparaneoplastic form of encephalitis that is responsive to immunotherapy. Differentiating this treatable disease from others that have a similar presentation is thus important.

Observations  We present the case of a 58-year-old man who had a rapid onset of progressive confusion, twitching of the face and hand, and abnormal basal ganglia detected by magnetic resonance imaging. His conditions were initially diagnosed as Creutzfeldt-Jakob disease (CJD). Faciobrachial dystonic seizures, possibly pathognomonic for the VGKCC syndrome, had been misdiagnosed as myoclonus. Treatment led to a complete resolution of his symptoms.

Conclusions and Relevance  Given the similarities of the clinical features and, at times, the neuroimaging findings of VGKCC syndrome to CJD, recognizing VGKCC syndrome and the highly associated and distinctive faciobrachial dystonic seizures is very important. Because this syndrome is the most common treatable condition that mimics CJD, we believe that it is crucial to screen all patients with presumed CJD for this reversible condition.

Limbic encephalitis that is associated with anti–voltage-gated potassium channel complex (VGKCC) antibodies (VGKCC syndrome) is an autoimmune, usually nonparaneoplastic form of encephalitis, and most patients with this condition improve after early initiation of steroids or other immune-modulatory therapies.14 Thus, it is important to identify this treatable disease and to differentiate it from other diseases that have a similar presentation.

Herein, we present the case of a 58-year-old man who had a rapid onset of progressive but fluctuating dementia, twitching of the face and hand, and abnormal basal ganglia detected by magnetic resonance imaging (MRI). At another academic medical center, he had received a diagnosis of Creutzfeldt-Jakob disease (CJD). Faciobrachial dystonic (FBD) seizures, possibly pathognomonic for this syndrome (and more specifically for the nonparaneoplastic anti-LGI1 version of the syndrome),5 had been misdiagnosed as myoclonus. Treatment led to a complete resolution of his symptoms.

Report of a Case

A 58-year-old right-handed man who recently received a diagnosis of CJD at an outside hospital presented with a new-onset generalized seizure that occurred while he was sleeping. Two months before this presentation, he was admitted to the outside hospital with a 2- to 3-month history of intermittent confusion and twitching of his face and hand. At that time, MRI scans showed T2-weighted and fluid-attenuated inversion recovery hyperintensities in the right caudate, the right putamen, and the left greater than right medial temporal lobe, with borderline diffusion-weighted imaging hyperintensity in the right caudate and putamen and a normal attenuation diffusion coefficient (Figure 1). The results of blood work were notable for a low sodium level of 131 mEq/L (normal range,135-145 mEq/L [to convert to millimoles per liter, multiply by 1.0]). The results of cerebrospinal fluid (CSF) studies were unremarkable. Some results, including those for the CSF 14-3-3 protein level, the CSF and serum paraneoplastic panel, and the VGKCC antibody, were pending at the time of hospital discharge. The results of a routine electroencephalogram (EEG) were unremarkable. With a follow-up telephone call after additional results were received, he was told that he had CJD. Since that time, his family noted that his face and hand twitching worsened over 1 to 2 months, with the facial contortions and simultaneous hand dystonic movements lasting under 2 seconds and happening many times a day. His confusion and amnesia had gotten worse, and he was no longer able to work. The nocturnal convulsion led to his hospitalization.

Figure 1.
Initial Magnetic Resonance Imaging (MRI) Findings Showing Abnormal Basal Ganglia
Initial Magnetic Resonance Imaging (MRI) Findings Showing Abnormal Basal Ganglia

The MRI scans show T2-weighted and fluid-attenuated inversion recovery hyperintense signals in the right caudate and in the right putamen (A [arrowheads]) and in the left greater than right medial temporal lobe (B [arrowheads]). There is borderline diffusion-weighted imaging hyperintensity in the right caudate and putamen (C [arrowheads]), without a corresponding attenuation diffusion coefficient hypointensity (D).

During a neurologic examination, the patient was alert and oriented, with no language deficits; however, it was noted that he was intermittently confused and slow to respond. There was no myoclonus, fasciculations, or exaggerated startle. Frequent though very brief (0.5-2 seconds) facial grimaces were noted, on the right side of his face more than the left, sometimes with subtle right-hand movement (mostly wrist flexion). He was unaware of these movements unless it involved his chin and jaw. The results of cranial nerve, motor, sensory, coordination, and gait examinations were normal.

He underwent continuous video-EEG monitoring. Several brief left temporal lobe seizures, during which time he was unable to name or answer questions, were captured; he was unaware of these spells. The facial grimaces and hand movements showed no EEG correlates. An interictal EEG showed diffuse slowing, a posterior-dominant rhythm of 7.5 Hz, and occasional left hemispheric slowing.

At this time, MRI, which was performed 2 months after the initial MRI (Figure 1), showed T2-weighted hyperintensity in the bilateral medial temporal lobes, with no abnormalities in the basal ganglia (Figure 2). Positron emission tomography of the brain revealed prominent hypermetabolism in the left caudate and putamen and less prominent hypermetabolism in the right putamen (Figure 3). There was decreased metabolism in the right temporal lobe. The results of blood work were again notable for a low sodium level of 130 mEq/L.

Figure 2.
Additional Magnetic Resonance Imaging (MRI) Findings Showing Improvement in the Basal Ganglia
Additional Magnetic Resonance Imaging (MRI) Findings Showing Improvement in the Basal Ganglia

The MRI fluid-attenuated inversion recovery axial scans reveal a marked improvement in the hyperintensity in the basal ganglia (A) and in the remaining hyperintensity in the bilateral medial temporal lobes (B [arrowheads]).

Figure 3.
Positron Emission Tomographic (PET) Findings Showing Hypermetabolism in the Basal Ganglia
Positron Emission Tomographic (PET) Findings Showing Hypermetabolism in the Basal Ganglia

The PET scans of the patient’s brain show hypermetabolism in the left caudate and putamen (A [arrowheads]) and less prominent hypermetabolism in the right putamen (B [arrowheads]).

He received a diagnosis of FBD seizures, likely related to anti-VGKCC syndrome (anti-LGI1), which was later confirmed. He was given intravenous methylprednisolone succinate (1 g/d for 5 days). An EEG showed no further electrographic seizures, and his cognition improved. However, he continued to have frequent facial and hand dystonic movements (FBD seizures) after 2 days of treatment with steroids. At that time, he had been treated with phenytoin sodium, and the FBD seizures resolved by the next day. He was discharged from the hospital and prescribed phenytoin and oral prednisone (60 mg/d), which was continued for a month, followed by a slow tapering of the dosage.

A serum paraneoplastic panel was sent to the Mayo Laboratories in Rochester, Minnesota, and was positive for the anti-VGKCC antibody at 1.13 nmol/L (reference range, ≤0.02), and this confirmed a diagnosis of VGKCC syndrome. In a review of the patient’s prior workup at the outside hospital, a serum paraneoplastic panel was also sent to the same laboratories and was also positive for the anti-VGKCC antibody, at 2.0 nmol/L. The CSF sample was also positive for 14-3-3 protein (National Prion Disease Pathology Surveillance Unit, Cleveland, Ohio).

A follow-up EEG a month after hospital discharge was normal, with an improved posterior-dominant rhythm of 10 Hz. Follow-up anti-VGKCC antibody testing was “indeterminate” 6 months later, and anti-VGKCC antibodies became undetectable in another 2 months despite the gradual lowering of the doses of steroids and with no other immunosuppressants being used. Initial neuropsychological testing, which was performed a month after treatment, showed a superior range of intellectual abilities (full-scale IQ of 121) with relative weakness in delayed recall of verbal and nonverbal information (low average), as well as slowed mental processing speed (average). When the testing was repeated 6 months later, substantial improvement was seen in delayed recall (now average) and mental processing speed (now superior). The patient returned to his previous baseline function and resumed working full time. Steroids were tapered down to 30 mg every other day without recurrence of symptoms as of more than 1 year after diagnosis and 1.5 years after initial presentation.

Discussion

Creutzfeldt-Jakob disease is suspected when a patient presents with a rapidly progressive dementia accompanied by neuropsychiatric features, myoclonus, extrapyramidal or cerebellar signs, or visual disturbances.6 The initial presentation of our patient, especially in combination with the abnormalities in the basal ganglia detected by MRI, was mimicking CJD. However, the absence of definite restricted diffusion should have raised a red flag against the diagnosis of CJD. The diffusion restriction is shown to be the most sensitive and specific feature of sporadic CJD.7,8 The European MRI-CJD Consortium criteria allow for the use of fluid-attenuated inversion recovery without diffusion-weighted imaging abnormalities for the diagnosis of CJD,9 and with the use of their criteria, this patient’s condition would have been misdiagnosed. In addition, although the most common MRI findings of VGKCC syndrome are bilateral or unilateral medial temporal lobe high T2-weighted signals, abnormalities in the basal ganglia and the cortical ribbon, which are characteristic of CJD, have also been reported for this condition.5,10 Moreover, what was considered to be myoclonus was actually FBD seizures. The key feature that distinguished the movement from myoclonus was the highly stereotyped, simultaneous face and hand movement, which was slower than myoclonus. This movement has been described and named as FBD seizures by Irani et al.5 They consist of very brief and very frequent unilateral dystonic movements, which always involve the arm and commonly also the ipsilateral face (76%) and leg (34%).5 The epileptic nature of these movements was suggested by the associated brief loss of awareness, the highly stereotyped semiology, and the infrequent ictal scalp EEG changes.5 Our patient’s response to phenytoin suggests this as well. It has been suggested that ictal dystonia reflects basal ganglia involvement, which is supported by fludeoxyglucose F18 positron emission tomography findings demonstrating an abnormal basal ganglia metabolism in some of these patients5,11 (including our patient). Electroencephalograms can be helpful because generalized periodic discharges would suggest CJD.6 Hyponatremia is another helpful clue favoring VGKCC syndrome.1

Given the similarities in the clinical and neuroimaging features between VGKCC syndrome and CJD, the recognition of VGKCC syndrome and its highly associated and distinctive seizure type (FBD seizures) is very important, especially since this syndrome is treatable. In fact, our patient, whose condition was previously misdiagnosed as CJD, became symptom free with treatment. Because this syndrome is the most common treatable condition that mimics CJD, we believe that it is important to screen all patients with presumed CJD for this reversible condition.

Back to top
Article Information

Accepted for Publication: September 17, 2013.

Corresponding Author: Ji Yeoun Yoo, MD, Department of Neurology, Mount Sinai Hospital, 1 Gustave L. Levy Pl, 2nd Floor, Box 1052, New York, NY 10029 (jiyeoun.yoo@mssm.edu).

Published Online: November 18, 2013. doi:10.1001/jamaneurol.2013.5179.

Author Contributions: Drs Yoo and Hirsch had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Hirsch.

Analysis and interpretation of data: Yoo.

Drafting of the manuscript: Yoo.

Critical revision of the manuscript for important intellectual content: Hirsch.

Conflict of Interest Disclosures: None reported.

Additional Contributions: We thank the neurology residents, neuropsychologists, and nurses for their care of our patient.

References
1.
Vincent  A, Buckley  C, Schott  JM,  et al.  Potassium channel antibody-associated encephalopathy: a potentially immunotherapy-responsive form of limbic encephalitis. Brain. 2004;127(pt 3):701-712.
PubMed
2.
Thieben  MJ, Lennon  VA, Boeve  BF, Aksamit  AJ, Keegan  M, Vernino  S.  Potentially reversible autoimmune limbic encephalitis with neuronal potassium channel antibody. Neurology. 2004;62(7):1177-1182.
PubMedArticle
3.
Lai  M, Huijbers  MG, Lancaster  E,  et al.  Investigation of LGI1 as the antigen in limbic encephalitis previously attributed to potassium channels: a case series. Lancet Neurol. 2010;9(8):776-785.
PubMedArticle
4.
Irani  SR, Alexander  S, Waters  P,  et al.  Antibodies to Kv1 potassium channel-complex proteins leucine-rich, glioma inactivated 1 protein and contactin-associated protein-2 in limbic encephalitis, Morvan’s syndrome and acquired neuromyotonia. Brain. 2010;133(9):2734-2748.
PubMedArticle
5.
Irani  SR, Michell  AW, Lang  B,  et al.  Faciobrachial dystonic seizures precede Lgi1 antibody limbic encephalitis. Ann Neurol. 2011;69(5):892-900.
PubMedArticle
6.
Brown  P, Cathala  F, Castaigne  P, Gajdusek  DC.  Creutzfeldt-Jakob disease: clinical analysis of a consecutive series of 230 neuropathologically verified cases. Ann Neurol. 1986;20(5):597-602.
PubMedArticle
7.
Shiga  Y, Miyazawa  K, Sato  S,  et al.  Diffusion-weighted MRI abnormalities as an early diagnostic marker for Creutzfeldt-Jakob disease. Neurology. 2004;63(3):443-449.
PubMedArticle
8.
Vitali  P, Maccagnano  E, Caverzasi  E,  et al.  Diffusion-weighted MRI hyperintensity patterns differentiate CJD from other rapid dementias. Neurology. 2011;76(20):1711-1719.
PubMedArticle
9.
Zerr  I, Kallenberg  K, Summers  DM,  et al.  Updated clinical diagnostic criteria for sporadic Creutzfeldt-Jakob disease. Brain. 2009;132(pt 10):2659-2668.
PubMedArticle
10.
Geschwind  MD, Tan  KM, Lennon  VA,  et al.  Voltage-gated potassium channel autoimmunity mimicking Creutzfeldt-Jakob disease. Arch Neurol. 2008;65(10):1341-1346.
PubMedArticle
11.
Kamaleshwaran  KK, Iyer  RS, Antony  J, Radhakrishnan  EK, Shinto  A.  18F-FDG PET/CT findings in voltage-gated potassium channel limbic encephalitis. Clin Nucl Med. 2013;38(5):392-394.
PubMedArticle
×