First Episode Psychosis: Psychiatric or Neurologic?

Even when cerebrospinal fluid (CSF) was tested, patients with first episode psychosis did not have a high frequency of N-methyl-d-aspartate receptor (NMDAR) antibodies, a prospective study found.

The frequency of NMDAR antibodies in CSF was about the same as it was when measured in serum, reported Josep Dalmau, MD, of University of Pennsylvania in Philadelphia, and colleagues.

“The expectation of a high frequency of NMDAR-antibodies in psychiatric disorders has not been met by most first episode psychosis studies using serum, which usually show frequencies of seropositivity ≤ 3%, or not different from healthy controls,” Dalmau and co-authors wrote in Neurology.

“Because previous studies on first episode psychosis did not examine CSF antibodies, it has been postulated that if CSF was tested the number of autoimmune cases would be higher,” they added. “Our findings here show that even when CSF is tested, the frequency of autoimmune cases does not seem to be higher.”

People with a proposed type of autoimmune encephalitis (autoimmune psychosis) have first episode psychosis caused by NMDAR antibodies. These patients make up a small minority of all first psychosis cases and have a treatable neurologic disorder, though diagnosis and treatment may be delayed if primary psychiatric disease is presumed.

In their study, Dalmau and co-authors followed 105 patients who had been admitted to one of two Spanish hospitals between June 2018 and March 2020 with first episode psychosis, defined as new-onset disorganized behavior with delusions or hallucinations of less than 6 months duration.

Patients had a median age of 30 and 42% were female. Median hospital admission was 21 days and median duration of symptoms related to first episode psychosis was 79 days.

No patient had NMDAR antibodies in serum or CSF and none developed autoimmune psychosis. About one-third fulfilled two sets of warning signs for autoimmune psychosis and another 20% met criteria for possible or probable autoimmune psychosis.

Over a median followup of 18 months, the cause of first episode psychosis ultimately was determined to be psychiatric in 101 patients (96%) and non-psychiatric in four people (4%).

In an accompanying editorial, Maarten Titulaer, MD, PhD, of Erasmus MC University in Rotterdam, the Netherlands, and Gregory Day, MD, MSc, of Mayo Clinic in Jacksonville, Florida, noted that the discovery of patients with prominent psychosis and subacute decline in neurological function established anti-NMDAR encephalitis as a novel cause of new-onset psychoses.

While these patients “may be severely affected at presentation (earning the moniker ’brain on fire’), remarkable improvement is noted following early-induction of appropriate immunotherapies,” they added.

This study shows that, among patients with first episode psychosis, “a small percentage (most likely below 1%) have CSF NMDAR-antibodies,” Dalmau and co-authors observed.

“Patients with first episode psychosis and NMDAR-antibodies have anti-NMDAR encephalitis; if the CSF is antibody-negative, alternative etiologies must be considered,” they added. “All patients with first episode psychosis of unclear etiology should be carefully considered for the diagnosis of anti-NMDAR encephalitis.”

Important clues may differentiate anti-NMDAR encephalitis from non-autoimmune first episode psychosis and prompt a confirmatory CSF evaluation, the researchers said. Patients with anti-NMDAR encephalitis psychosis often:

• Develop neurologic symptoms within days or a few weeks of disease onset.
• May have an associated tumor (teratoma).
• Are prone to intolerance to antipsychotics.
• Have one or more abnormal paraclinical tests (e.g., >95% EEG or ~40% MRI).

These criteria depend on neurological features and abnormal CSF, MRI, or EEG tests, despite attempts to recognize autoimmune psychosis in patients with predominantly psychiatric symptoms, Dalmau and colleagues noted.

“Some patients with anti-NMDAR encephalitis initially present with psychotic behavior indistinguishable from primary psychiatric disorders without any of the neurological features listed in criteria for possible autoimmune psychosis,” they wrote. “Thus, the diagnosis will likely be missed.”

These findings are of great interest to practicing neurologists who often are asked to determine whether emergent psychiatric symptoms are a manifestation of autoimmune encephalitis, the editorialists noted. “The results of this paper call into question the ability to identify patients with first episode psychosis due to autoimmune psychosis by clinical criteria alone,” they wrote.

The study also raises the bar for studies considering autoimmune contributions to psychiatric disease, Titulaer and Day added.

“The high importance of early treatment of autoimmune encephalitis and autoimmune psychosis justifies efforts to improve early identification of the comparatively low proportion of patients (likely <2%) with first episode psychosis attributed to disease-associated auto-antibodies,” they wrote. “Larger studies or meta-analyses are required to define the prevalence of anti-NMDAR encephalitis in first episode psychosis, and to optimize the selection of patients with first episode psychosis who require CSF sampling.”

Limitations include potentially limited generalizability as the study cohort was from one European health care system. In addition to the study findings, Dalmau and co-authors included a diagnostic algorithm for autoimmune first episode psychosis in their paper.

1. Even when cerebrospinal fluid was tested, patients with first episode psychosis did not have a high frequency of N-methyl-d-aspartate receptor (NMDAR) antibodies, a prospective study found.

2. People with a proposed type of autoimmune encephalitis (known as autoimmune psychosis) have first episode psychosis caused by NMDAR antibodies; they make up a small percentage of cases and have a treatable neurologic disorder.

Paul Smyth, MD, Contributing Writer, BreakingMED™

Cat ID: 130

Topic ID: 82,130,730,130,192,925