A wide variety of clinical syndromes has been associated with antibodies to voltage-gated potassium channels (VGKCs). Six years ago, it was discovered that patients do not truly have antibodies to potassium channels, but to associated proteins. This enabled the distinction of three VGKC-positive subgroups: anti-LGI1 patients, anti-Caspr2 patients and VGKC-positive patients lacking both antibodies. Patients with LGI1-antibodies have a limbic encephalitis, often with hyponatremia, and about half of the patients have typical faciobrachial dystonic seizures. Caspr2-antibodies cause a more variable syndrome of peripheral or central nervous system symptoms, almost exclusively affecting older males. Immunotherapy seems to be beneficial in patients with antibodies to LGI1 or Caspr2, stressing the need for early diagnosis. Half of the VGKC-positive patients lack antibodies to both LGI1 and Caspr2. This is a heterogeneous group of patients with a wide variety of clinical syndromes, raising the question whether VGKC-positivity is truly a marker of disease in these patients. Data regarding this issue are limited, but a recent study did not show any clinical relevance of VGKC-positivity in the absence of antibodies to LGI1 and Caspr2. The three VGKC-positive subgroups are essentially different, therefore, the lumping term ‘VGKC-complex antibodies’ should be abolished.