Barry et al reported four cases of anti-N-methyl-d-aspartate (NMDA) receptor antibody encephalitis that presented psychiatrically. ^{Reference Barry, Hardiman, Healy, Keogan, Moroney and Molnar1} This report was welcome in highlighting the importance of immunologically mediated encephalitides (or synaptopthies), both primary autoimmune and paraneoplastic, that has emerged over recent years. However, two points are worthy of emphasis.

First, the aetiological association of anti-NMDA receptor antibody encephalitis with ovarian neoplasms was perhaps understated in the paper. In a large study by Dalmau et al, around 50% of cases were associated with ovarian neoplasms and 80% of such patients improved following tumour removal and first-line immunotherapy, whereas only 48% of patients without an identified tumour responded as well to first-line immunotherapy. ^{Reference Dalmau, Lancaster, Martinez-Hernandez, Rosenfeld and Balice-Gordon2} Therefore, the identification and resection of ovarian tumours in patients with this syndrome is a primary concern.

Second, Barry et al conclude that it is important to consider anti-NMDA receptor antibody encephalitis in new-onset psychosis associated with catatonia, seizures and dyskinesia, and that it is unclear whether there is a pure psychiatric presentation. Zandi et al explored this question prospectively in 46 unselected patients with new-onset psychosis, finding anti-NMDA receptor antibodies in 2 patients. ^{Reference Zandi, Irani, Lang, Waters, Jones and McKenna3} It was also found that there were no clinical features that differentiated between antibody positive and negative patients. Also of note, this study identified one patient positive for anti-voltage-gated potassium channel antibodies (probably, in fact, anti-leucine-rich, glioma inactivated 1 (LGI1)). It is recognised that psychosis may be a feature of autoimmune encephalitides associated with serum antibodies against a number of proteins, including LGI1 and glutamic acid decarboxylase. Further psychiatric studies are required to determine whether a screen for antibodies associated with encephalitis should be routine in new-onset psychosis.