To assess the clinical relevance of a positive voltage-gated potassium channel (VGKC) test in patients lacking antibodies to LGI1 and Caspr2.

VGKC-positive patients were tested for LGI1 and Caspr2 antibodies. Patients lacking both antibodies were matched (1:2) to VGKC-negative patients. Clinical and paraclinical criteria were used to blindly determine evidence for autoimmune inflammation in both groups. Patients with an inconclusive VGKC titer were analyzed in the same way.

A total of 1,455 patients were tested by VGKC radioimmunoassay. Fifty-six patients tested positive, 50 of whom were available to be included. Twenty-five patients had antibodies to LGI1 (n = 19) or Caspr2 (n = 6) and 25 patients lacked both antibodies. Evidence for autoimmune inflammation was present in 7 (28%) of the VGKC-positive patients lacking LGI1 and Caspr2, compared to 9 (18%) of the VGKC-negative controls (p = 0.38). Evidence for autoimmune inflammation was mainly found in patients with limbic encephalitis/encephalomyelitis (57%), but not in other clinical phenotypes (5%, p < 0.01). VGKC titers were significantly higher in patients with antibodies to LGI1 or Caspr2 (p < 0.001). However, antibodies to Caspr2 could also be detected in patients with inconclusive low VGKC titer, while many VGKC-positive patients had no evidence for autoimmune inflammation.

VGKC positivity in the absence of antibodies to LGI1 and Caspr2 is not a clear marker for autoimmune inflammation and seems not to contribute in clinical practice. No cutoff value for the VGKC titer was appropriate to discriminate between patients with and without autoimmune inflammation.