Article Text

Download PDFPDF

Neurological manifestations related to level of voltage-gated potassium channel antibodies
  1. Henning Olberg1,
  2. Mette Haugen1,
  3. Anette Storstein1,
  4. Christian A Vedeler1,2
  1. 1 Department of Neurology, Haukeland University Hospital, University of Bergen, Bergen, Norway
  2. 2 Department of Clinical Medicine, University of Bergen, Bergen, Norway
  1. Correspondence to Professor Christian A Vedeler, Department of Neurology, Haukeland University Hospital, University of Bergen, Bergen N-5021, Norway; christian.vedeler{at}

Statistics from

Voltage-gated potassium channels (VGKC) are essential for the cellular action potential. Antibodies against the VGKC-complex are associated with a wide spectrum of diseases involving both the peripheral and central nervous system, including neuromyotonia, Morvan disease, limbic encephalitis, faciobrachial dystonic seizures and cerebellar ataxia.1 Leucine-rich, glioma-inactivated 1 antigen (LGI1) and contactin-associated protein-2 (Caspr2) have been found to be part of the VGKC-complex and targets for auto-antibodies.2 ,3 However, for the majority of the VGKC-complex antibodies, the specific antibody targets remain unknown. Moreover, apart from a recent study,4 little information exists on the relevance of VGKC-complex antibody level and neurological diseases, or the subgroups of VGKC antibodies and the presence of other encephalitic or paraneoplastic antibodies. We decided therefore to characterise our patients according to these parameters. The study was approved by the regional committee for medical and health research ethics in Western-Norway.

We identified 36 patients tested positive for VGKC-complex antibodies (tested by radioimmunoprecipitation in Oxford, >100 pM) at Haukeland University Hospital, Bergen, Norway, between 2001 and 2012. The sera were sent by neurologists who decided the indications for the test. The highest VGKC-complex antibody level was chosen for each patient (table 1). Two patients were of foreign origin and excluded due to lack of data accessibility. Two patients refused to participate. The medical records of the 32 participating patients were analysed in 2012. Six of the patients were dead of non-neurological disorders at the time of analysis. Sera from 29 of the participating patients were available and reanalysed at the Neurology research laboratory, Haukeland University Hospital for the following antibodies: LgI1, Caspr2, NMDAR, AMPA1/2R and GABA1/2R (using transfected cells;; and Hu, Yo, Ri, amphiphysin, CRMP5, Ma1/2, Sox1 and GAD65 (using line blots;

Table 1

Neurological manifestations related to VGKC levels

There were 12 patients with VGKC-complex antibody levels >500 pM of which 10 had presented with limbic encephalitis (levels 576–5143 pM; table 1) comprising both cognitive and psychiatric manifestations. MRI showed temporal lobe hypersignal intensity in eight of the 10 patients while one of two patients with normal MRI had a pathological PET. The other patient with VGKC-complex levels of 1043 pM was not investigated with PET and had a faciobrachial dystonic seizure variant of limbic encephalitis. All 10 patients had pathological EEG, five had intrathecal immunoglobulin (Ig)G synthesis and five hyponatremia (Na <130 mM). LGI1 antibodies were present in eight of the 10 patients with limbic encephalitis. In addition, one patient with AChR antibody-positive, but thymoma-negative myasthenia gravis had VGKC-complex antibody level of 827 pM. This patient also had sensory-motor polyneuropathy. Subtyping the VGKC-complex antibodies was not performed in this case due to lack of serum. Another patient with neuromyotonia had VGKC antibody level of 555 pM and LGI1 antibodies. This patient had cramps and myokymia with continuous motor unit activity by neurophysiological examination, but no CNS symptoms. All 12 patients with VGKC-complex antibody levels >500 pM responded to immunotherapy which included corticosteroids, intravenous IgG and/or plasma exchange.

The 20 patients with lower antibody levels were more diverse, and there was no clear association with autoimmune neurological disease. One patient with a VGKC-complex antibody level of 209 pM had cramp fasciculation syndrome and improved without treatment after 3 months with no evidence for neuromyotonia. One patient with a level of 155 pM had vasculitis. One patient with a level of 116 pM had AChR/MuSK antibody-negative myasthenia gravis and pulmonary leiomyosarcoma, but no thymoma. This patient was also VGCC antibody negative. Another sero-negative and thymoma-negative myasthenia gravis patient had VGKC-complex antibody level of 109 pM.

LGI1 antibodies were found in one patient with a VGKC-complex antibody level of 140 pM. This patient had a postinfectious, acute inflammatory demyelinating polyneuropathy, based on clinical and neurophysiological tests, and no CNS symptoms. The patient with acute inflammatory demyelinating polyneuropathy received intravenous IgG treatment and the patients with vasculitis or myasthenia gravis were treated with corticosteroids. The other patients in this group of low-level VGKC-complex antibodies did not receive immunotherapy.

Malignant disease was found in seven of the 32 patients. Medullary thyroid cancer and pulmonary leiomyosarcoma developed within 1 year of the neurological symptoms, which could therefore be paraneoplastic. The five other malignancies presented more than 5 years before the neurological disease and were therefore unlikely to be paraneoplastic. VGKC-complex antibodies have been associated with tumour in 33% of patients with neuromyotonia and 41% of patients with Morvan syndrome, usually with thymoma.5 None of our 29 patients had Hu, Yo, Ri, amphiphysin, CRMP5, Ma1/2, Sox1, GAD65, Caspr2, NMDAR, AMPA1/2R or GABA1/2R antibodies.

We found that levels of VGKC-complex antibody >500 pM were most likely associated with autoimmune neurological disease, usually limbic encephalitis. One patient had neuromyotonia which also has been associated with VGKC-complex antibodies. In addition, one patient with myasthenia gravis without thymoma had high VGKC-complex antibody level which has not previously been described. There was a significant association between high levels of VGKC-complex antibodies and the presence of LGI1 antibodies, p value=0.007, using multinomial logistic regression (Wald's test) which supports the recent results of a cohort of 316 patients with VGKC antibodies.4 Similar to previous studies, we found that LGI1 antibodies were primarily associated with CNS disease.2–4 No patients with Caspr2 antibodies, which have been associated with primarily PNS disease,3 ,4 were found in our patient cohort. Further, we found no patients with VGKC-complex antibodies in the range of 210–554 pM and cannot comment on the clinical relevance of these antibody levels. In our material, VGKC-complex antibodies of <210 pM were most likely not associated with autoimmune disease, but two patients with myasthenia gravis were also present in this group. LGI1 antibodies were found in one patient with low VGKC-complex levels, but with no clinical relevance.

We conclude from this small cohort of patients that VGKC-complex antibody levels >500 pM are most likely associated with autoimmune neurological disease and these patients also respond to immunotherapy.


We thank Angela Vincent (Neuroimmunology Group, Nuffield Department of Clinical Neurosciences, University of Oxford, John Radcliffe Hospital, Oxford, UK) for testing VGKC-complex antibodies and for proof reading.



  • Competing interests None.

  • Ethics approval Regional Committee for Medical and Health Research Ethics (REK Vest).

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.