Clinical, Neuroimmunologic, and CSF Investigations in First Episode Psychosis

Abstract

Objectives:To report the neuropsychiatric features and frequency of N-methyl-d-aspartate receptor (NMDAR) and other neuronal IgG antibodies in patients with first episode psychosis (FEP), and assess the performance of reported warning signs and criteria for autoimmune psychosis (AP).Methods:Prospective observational study of patients with FEP assessed for neuropsychiatric symptoms, serum and CSF neuronal antibodies (brain immunohistochemistry, cell-based assays, live neurons), and warning signs and criteria of AP. Previous autoimmune FEP series were reviewed.Findings:105 patients were included; median age 30 years (range 14-75), 44 (42%) female. None had neuronal antibodies. Two/105 (2%) had CSF pleocytosis, 4/100 (4%) brain MRI abnormalities, and 3/73 (4%) EEG alterations. Thirty-four (32%) and 39 (37%) patients fulfilled two sets of warning signs of AP, and 21 (20%) fulfilled criteria of possible or probable AP, yet none developed AP. The cause of FEP was psychiatric in 101 (96%) and non-psychiatric in 4 (4%). During this study, 3 patients with psychosis caused by anti-NMDAR encephalitis were transferred to our center; 2 did not meet criteria for possible AP. Of 1159 reported FEP patients, only 7 (1%) had CSF studies; 36 (3%) had serum NMDAR-antibodies (without definite diagnosis of AP), and 4 had CSF NMDAR-antibodies (3 classic anti-NMDAR encephalitis and 1 with isolated psychiatric features).Conclusions:NMDAR-antibodies were not found in patients with FEP unless they had anti-NMDAR encephalitis. Warning signs and criteria for AP have limited utility when neurological symptoms are absent or paraclinical tests are normal. A diagnostic algorithm for autoimmune FEP is provided.

Neurofilament Light Chain Levels in Anti-NMDAR Encephalitis and Primary Psychiatric Psychosis

“…These findings indicate that, despite an overlap of sNfL levels between pFEP and NMDARe, patients with new-onset psychosis of unclear etiology and serum NfL ≥15 pg/mL had a 120 times higher chance of having NMDARe than pFEP. Considering the clinical setting in which the a priori chance of NMDARe is considerably lower than that of psychosis of psychiatric etiology (estimated prevalence of NMDARe in FEP 3% 2 ) and using the indicated sNfL sensitivity and specificity, the estimated positive predictive value of a patient with a first episode of psychosis and sNfL >15 pg/mL having NMDARe would be 40%, and the negative predictive value would be 99.5%.…”

“…These findings indicate that serum NfL testing can be an important diagnostic aid in these clinical settings, but it does not override the need of comprehensive neurologic evaluation along with brain MRI, EEG, and serum NMDAR antibodies, as recently proposed. 2 If serum NfL levels are ≥15 pg/mL or any of the other tests are abnormal, patients with new-onset psychosis without a clear etiology should undergo a neurologic examination and workup, including CSF analysis for NMDAR antibodies.…”

“…1 This is particularly important when psychosis is the initial or predominant manifestation because current criteria of NMDARe and autoimmune psychosis depend on a combination of neurologic and paraclinical features that are inadequate in this clinical setting. 2,3 Most patients with first episode of psychosis (FEP) need antipsychotic drugs, which in turn may trigger neurologic adverse events undistinguishable from NMDARe. 4 If this occurs, the differential diagnosis becomes even more difficult, in a way that new deficits such as dyskinesias, hyperthermia, catatonia, or coma may be attributed to a neuroleptic-induced complication in a psychiatric patient who, in fact, has an evolving NMDARe.…”

Autoimmune Encephalitis With Psychotic Manifestations and Cognitive Impairment Presenting as Schizophrenia: Case Report and Literature Review

“…Pollak et al proposed a novel and conservative approach to recognize autoimmune psychosis ( 16 ). However, it was found that the criteria for autoimmune psychosis have limited utility in the absence of neurological symptoms or when paraclinical test results are normal ( 17 ). Another study had revealed that anti-NMDA-receptor encephalitis as defined by consensus criteria rarely occurred in psychiatric patients ( 18 ).…”

Autoimmune Encephalitis in Long-Standing Schizophrenia: A Case Report

“…In the CNS, IgG NR1 antigen-antibody binding results in reversible titratable internalisation of the NMDAR that reduces NMDA signalling and results in the symptom development. Antibody presence in the serum alone is not sufficient to cause symptoms (3,4,7). However, processes disrupting the blood-brain barrier can allow antibody circulation within the CSF, leading to antigen-antibody binding (4).…”