Authors’ reply

We thank Dr Moran for highlighting the importance of immunologically mediated encephalitis when considering differential diagnoses for atypical psychosis. Dr Moran suggests that the aetiological association of anti-NMDA receptor encephalitis with ovarian neoplasms, in particular teratomas, was perhaps understated in our case series of four patients, where we reported ovarian pathology (a dermoid cyst) in one patient. ¹ By contrast, Dalmau et al's original series of 100 cases identified ovarian teratomas in 54 of the 58 cases with ovarian pathology and early removal of such tumours was associated with better outcomes. ² However, more recent studies have not observed such high rates of ovarian pathology. ^3,4 In a series of 44 patients with anti-NMDA receptor encephalitis, Irani and colleagues found tumours in 9 patients of which 8 cases were ovarian teratomas. Furthermore, 25% of cases overall were male. In keeping with Dalmau and colleagues, the identification and removal of an ovarian tumour was associated with a better outcome, ² although the best outcome was predicted by adequate immunotherapy during initial illness. ⁴

As noted previously, it is still unclear whether there is a purely psychiatric presentation to this disorder. However, the constellation of symptoms including some or all of catatonia, dyskinesias and seizures with psychosis certainly warrants anti-NMDA receptor antibody testing. We agree with Dr Moran that future studies are required to determine whether routine screening for NMDA receptor antibodies is indicated for atypical presentations, treatment-resistant cases and first-onset psychosis.