Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Mar;6(3):235-246.
doi: 10.1016/S2215-0366(19)30001-X. Epub 2019 Feb 11.

The Psychopathology of NMDAR-antibody Encephalitis in Adults: A Systematic Review and Phenotypic Analysis of Individual Patient Data

Affiliations
Free PMC article

The Psychopathology of NMDAR-antibody Encephalitis in Adults: A Systematic Review and Phenotypic Analysis of Individual Patient Data

Adam Al-Diwani et al. Lancet Psychiatry. .
Free PMC article

Abstract

Background: Early immunotherapy administration improves outcomes in patients with N-methyl-D-aspartate receptor (NMDAR)-antibody encephalitis. As most patients with NMDAR-antibody encephalitis present to psychiatrists, the psychopathology of NMDAR-antibody encephalitis needs to be clearly defined to encourage accurate clinical identification and prompt treatment.

Methods: For this systematic review, we searched PubMed for all studies published in English between Jan 1, 2005, and Oct 7, 2017, to identify individually reported adult patients (≥18 years) who satisfied consensus criteria for definite NMDAR-antibody encephalitis. After generating a list of 50 fine-grained, lower-level features, we extracted psychopathological data in addition to demographic and aetiological data. The lower-level features were later ordered within higher-level categories. As a means of quality control, we filtered the data according to proxy markers of psychiatric involvement in their description. Subsequently, we compared lower-level features from individual patient data with operationalised psychiatric syndromes using a constrained combination approach and principal component analysis, and did a network analysis to explore the inter-relationships between multiple lower-level features. The review protocol was prospectively registered with PROSPERO, number CRD42017068981.

Findings: Of 1096 records identified in PubMed, 333 satisfied inclusion criteria and described 1100 patients in total with NMDAR-antibody encephalitis. The psychopathology of 505 (46%) patients with reported psychiatric symptoms was described in more detailed terms than only psychiatric or behavioural. 464 (91%) of the 505 patients were from papers in which patient data were reported individually. The remainder of the analyses focused exclusively on these 464 patients. Median age was 27 years (IQR 22-34), 368 (79%) of 464 patients were female and in 147 (32%), NMDAR-antibody encephalitis was associated with ovarian teratoma. The five higher-level categories into which the 464 patients most frequently grouped were behaviour (316 [68%]), psychosis (310 [67%]), mood (219 [47%]), catatonia (137 [30%]), and sleep disturbance (97 [21%]). The overall pattern of lower-level features was statistically stable across subgroups classified by age, sex, pregnancy association, presence of ovarian teratoma, prior herpes simplex virus encephalitis, and isolated psychiatric presentations (two-way ANOVA p=0·6-0·9). Constrained combination and principal component analyses found that mixtures of mood and psychosis syndromes fit each patient better than any single diagnosis alone, particularly for the patients in the psychiatric-described subgroup (mean ΔAkaike information criterion -0·04 in non-psychiatric-described subgroup vs 0·61 in psychiatric-described subgroup). The overlapping nature of the higher-level features was also enriched upon analysis of the psychiatric-described data (221 [67%] of 329 overlaps in non-psychiatric-described subgroup vs 96 [81%] of 118 overlaps in psychiatric-described subgroup, p=0·0052). Network analysis confirmed that the features were closely related and consistent between individual patients; the psychiatric-described subgroup had a markedly high and narrow range of closeness centralities (92% above 0·93 in psychiatric-described subgroup vs 51% above 0·93 in the non-psychiatric group).

Interpretation: The distinctive aspect of NMDAR-antibody encephalitis psychopathology is complexity; core aspects of mood and psychotic disorders consistently coexist within individual patients. Alongside the predominant young female demographic, these psychopathological features could help psychiatrists identify patients who would benefit from cerebrospinal fluid testing and immunotherapies. Well-controlled prospective studies with bespoke inventories are needed to advance this clinically grounded approach.

Funding: Wellcome Trust, NIHR Oxford Biomedical Research Centre, NIHR Oxford Health Biomedical Research Centre, British Medical Association Foundation for Medical Research.

Figures

Figure 1
Figure 1
Study selection
Figure 2
Figure 2
Demographics of 464 patients with N-methyl-D-aspartate receptor-antibody encephalitis (A) Age distribution of patients at onset of N-methyl-D-aspartate receptor (NMDAR)-antibody encephalitis, including female patients and those with ovarian teratomas. Trend lines were smoothed with a second-order function with eight nearest neighbours. (B) Week of pregnancy and ages of patients at onset of pregnancy-associated NMDAR-antibody encephalitis. One case associated with recent miscarriage is not shown.
Figure 3
Figure 3
Frequencies and overlaps of higher-level and lower-level psychopathological features from individual patient data 464 patients were from papers in which patient data were reported individually. (A) Number of patients who manifest the eight higher-level and 50 lower-level features. (B) Venn diagram of overlaps between the five most common higher-level features shows frequent coexistence in individual patients. The numbers within the Venn diagram represent numbers of patients who had that pattern of overlaps. (C) Percentage frequencies of the 50 lower-level features in six subgroups of patients classified by age, sex, pregnancy, presence of ovarian teratoma, previous HSVE, and isolated psychiatric presentations. Two-way ANOVA with Bonferroni correction was used to compare the subgroups. None of the subgroups had significantly different frequencies of lower-level features. HSVE=herpes simplex virus encephalitis.
Figure 4
Figure 4
Initial psychiatric diagnoses and lower-level features of patients in the psychiatric-described and non-psychiatric-described subgroups (A) Initial psychiatric diagnoses reported in 44 patients before confirmation of N-methyl-D-aspartate receptor-antibody encephalitis. The comorbidity diagnosis consisted of somatoform disorder NOS, psychotic disorder NOS, and dissociative disorder NOS. (B) Proportion of patients in the psychiatric-described subgroup and non-psychiatric-described subgroup grouped according to the number of psychopathological features used to describe them. NOS=not otherwise specified.
Figure 5
Figure 5
Classification of individual patients using operationalised psychosis and mood spectrum diagnoses (A) Common primary psychiatric syndromes (n=14) were compared with 50 lower-level features to generate a signature for each syndrome. Subsequently, as illustrated by the example of comparison label, each syndrome was compared with each patient in the psychiatric-described subgroup (n=129) and the non-psychiatric-described subgroup (n=335; the heatmap for both subgroups is in the appendix). (B) Principal component analysis of lower-level features. Jaccard indices were generated to assess the overlaps between the 14 operationalised diagnostic categories and individual patient data. Analysis of the variables contributing to the first two principal components (PC1 and PC2) show a clear distribution of patients with N-methyl-D-aspartate receptor-antibody encephalitis defined by a combination of psychosis and mood dimensions. Overlay of the operationalised diagnoses showed orthogonal vectors from mood (blue) and psychotic (red) categories, with mainly mixed disorders (green) within an intermediate space. Histograms on each axis show the density of data mapped across PC1 and PC2. APPD=acute polymorphic psychotic disorder, Cat Sz=catatonic schizophrenia, D=depression, Heb Sz=hebephrenic schizophrenia, M=mania, NOS=not otherwise specified, PC=principal component, PPP=post-partum psychosis, P Sz=paranoid schizophrenia, Sz=schizophrenia, SzAD=schizoaffective disorder, + cat=with catatonia, + psy=with psychotic features.
Figure 6
Figure 6
Features of N-methyl-D-aspartate receptor-antibody encephalitis psychopathology in the psychiatric-described and non-psychiatric-described subgroups (A) Venn diagram of overlaps between the five most common higher-level features in psychiatric-described subgroups and non-psychiatric-described subgroups. (B) Estimation of fit using AIC and coherence estimation using network analysis in the non-psychiatricdescribed (n=335) and psychiatric-described (n=129) subgroups. The figure shows the number of patients modelled to each diagnosis (best single diagnosis histogram), and the number of patients with each pair of diagnoses (best pair of diagnoses heatmap). AIC=Akaike information criterion. APPD=acute polymorphic psychotic disorder. Cat Sz=catatonic schizophrenia. D=depression. Heb Sz=hebephrenic schizophrenia. M=mania. PPP=post-partum psychosis. P Sz=paranoid schizophrenia. Sz=schizophrenia. SzAD=schizoaffective disorder. + cat=with catatonia. + psy=with psychotic features. (C) Histograms showing the change in AIC in each group (left and right panels). The mean AIC by two diagnoses than by one. AIC=Akaike information criterion. (D) Network analysis of the 27 lower-level features found at greater than 10% relative frequencies of the most common feature (agitation). The lower-level features are represented as nodes; the size of the nodes is proportionate to the frequency of the feature and the edge thickness is proportionate to the frequency of co-occurrences. The nodes are colour-coded by closeness centrality, a measure of interconnectedness, where one is complete. Agit=agitation. Aggr=aggression. Ahal =auditory hallucinations. A-V hal=auditory and visual hallucinations. Anx=anxiety or fear. Del=delusion. Depr=depression. Disorg=disorganised or bizarre. Disinh=disinhibition. Gibb=talking gibberish. Hal=hallucinations. Insom=insomnia. Irrit=irritability or mood instability. Lau-cry=incongruent laughter-crying. Mani=mania. Mut=mutism. Para=paranoid theme. Phag=hypophagia. Post=posturing. Scr=screaming. Sex=sexual disinhibition. Som=hypersomnia. Stup=stupor. Suic=suicidal thoughts. V-hal=visual hallucinations. Viol=violence. Wax=waxy flexibility.
Figure 6
Figure 6
Features of N-methyl-D-aspartate receptor-antibody encephalitis psychopathology in the psychiatric-described and non-psychiatric-described subgroups (A) Venn diagram of overlaps between the five most common higher-level features in psychiatric-described subgroups and non-psychiatric-described subgroups. (B) Estimation of fit using AIC and coherence estimation using network analysis in the non-psychiatricdescribed (n=335) and psychiatric-described (n=129) subgroups. The figure shows the number of patients modelled to each diagnosis (best single diagnosis histogram), and the number of patients with each pair of diagnoses (best pair of diagnoses heatmap). AIC=Akaike information criterion. APPD=acute polymorphic psychotic disorder. Cat Sz=catatonic schizophrenia. D=depression. Heb Sz=hebephrenic schizophrenia. M=mania. PPP=post-partum psychosis. P Sz=paranoid schizophrenia. Sz=schizophrenia. SzAD=schizoaffective disorder. + cat=with catatonia. + psy=with psychotic features. (C) Histograms showing the change in AIC in each group (left and right panels). The mean AIC by two diagnoses than by one. AIC=Akaike information criterion. (D) Network analysis of the 27 lower-level features found at greater than 10% relative frequencies of the most common feature (agitation). The lower-level features are represented as nodes; the size of the nodes is proportionate to the frequency of the feature and the edge thickness is proportionate to the frequency of co-occurrences. The nodes are colour-coded by closeness centrality, a measure of interconnectedness, where one is complete. Agit=agitation. Aggr=aggression. Ahal =auditory hallucinations. A-V hal=auditory and visual hallucinations. Anx=anxiety or fear. Del=delusion. Depr=depression. Disorg=disorganised or bizarre. Disinh=disinhibition. Gibb=talking gibberish. Hal=hallucinations. Insom=insomnia. Irrit=irritability or mood instability. Lau-cry=incongruent laughter-crying. Mani=mania. Mut=mutism. Para=paranoid theme. Phag=hypophagia. Post=posturing. Scr=screaming. Sex=sexual disinhibition. Som=hypersomnia. Stup=stupor. Suic=suicidal thoughts. V-hal=visual hallucinations. Viol=violence. Wax=waxy flexibility.

Comment in

Similar articles

See all similar articles

Cited by 8 articles

See all "Cited by" articles

References

    1. Dalmau J, Tüzün E, Wu HY. Paraneoplastic anti-N-methyl-D-aspartate receptor encephalitis associated with ovarian teratoma. Ann Neurol. 2007;61:25–36. - PMC - PubMed
    1. Armangue T, Leypoldt F, Málaga I. Herpes simplex virus encephalitis is a trigger of brain autoimmunity. Ann Neurol. 2014;75:317–323. - PMC - PubMed
    1. Hacohen Y, Deiva K, Pettingill P. N-methyl-D-aspartate receptor antibodies in post-herpes simplex virus encephalitis neurological relapse. Mov Disord. 2013;29:90–96. - PubMed
    1. Dalmau J, Lancaster E, Martinez-Hernandez E, Rosenfeld MR, Balice-Gordon R. Clinical experience and laboratory investigations in patients with anti-NMDAR encephalitis. Lancet Neurol. 2011;10:63–74. - PMC - PubMed
    1. Titulaer MJ, McCracken L, Gabilondo I. Treatment and prognostic factors for long-term outcome in patients with anti-NMDA receptor encephalitis: an observational cohort study. Lancet Neurol. 2013;12:157–165. - PMC - PubMed

Publication types

MeSH terms

Feedback