The four cases of N-methyl-D-aspartate (NMDA) receptor antibody encephalitis with associated psychosis reported in December ^{Reference Barry, Hardiman, Healy, Keogan, Moroney and Molnar1} raise an important and emerging issue and highlight that psychiatrists should include the condition in the differential diagnosis for patients presenting with acute psychosis. But there are some aspects that need clarification. The authors state that ‘this case series demonstrates a new and treatable cause of psychosis’, inferring that the association of psychosis with these antibodies was previously unknown. However, since the first 100 patients with NMDA receptor antibody encephalitis were reported in 2008, ^{Reference Dalmau, Gleichman, Hughes, Rossi, Peng and Lai2} this association has been well documented; psychosis is typically the first presentation and many cases were seen by psychiatrists before neurologists become involved. ^{Reference Dalmau, Gleichman, Hughes, Rossi, Peng and Lai2,Reference Lennox, Coles and Vincent3}

The association of these antibodies with psychosis is highly relevant because they bind to key neuronal surface proteins and are therefore likely to be pathogenic. Indeed, NMDA receptor antibody encephalitis is a condition that responds to immunotherapy and, importantly, there is thought to be an initial ‘treatment window’ for optimal immunomodulation. ^{Reference Irani, Bera, Waters, Zuliani, Maxwell and Zandi4,Reference Vincent, Bie, Irani and Waters5} The authors ^{Reference Barry, Hardiman, Healy, Keogan, Moroney and Molnar1} speculate that ‘there may be a pure psychiatric presentation associated with lower antibody titres’. Indeed, a recent study found that 3 out of 46 patients with first-episode psychosis (with no neurological or other clinically distinguishing features) had NMDA receptor antibodies. ^{Reference Zandi, Irani, Lang, Waters, Jones and McKenna6} One patient made a significant clinical improvement with plasmapheresis and steroid treatment. An additional patient had voltage-gated potassium channel antibodies, which can also be found in patients with other psychiatric presentations. ^{Reference Vincent, Bie, Irani and Waters5,Reference Spinazzi, Argentiero, Zuliani, Palmieri, Tavolato and Vincent7} It now appears increasingly likely that other neuropsychiatric (e.g. catatonia) and psychiatric (e.g. obsessive-compulsive) symptoms may be associated with cell-surface neuronal antibodies. ^{Reference Kayser, Kohler and Dalmau8}

As Barry et al ^{Reference Barry, Hardiman, Healy, Keogan, Moroney and Molnar1} point out, the condition does indeed provide some support for the NMDA receptor hypofunction hypothesis for psychosis. Some proponents of this theory have linked NMDA receptor hypofunction to first-rank psychotic symptoms in particular. ^{Reference Stephan, Friston and Frith9} It is important that future studies of auto-antibody-associated psychosis characterise symptomatology in full, as this could allow for a level of clinical-pathological correlation rarely attained in psychiatry.