Abstract
Objective
Encephalitis is a serious and potentially life‐threatening condition of infectious or autoimmune cause. We aim to characterize the frequency and clinical spectrum of presenting psychiatric symptoms in encephalitis in order to inform earlier recognition and initiation of treatment.
Methods
This was a retrospective study of adult patients who met the 2013 International Encephalitis Consortium (IEC) and/or 2016 Graus criteria between February 2005 and February 2023. The study included two hospital systems in Houston, Texas, and Baltimore, Maryland and included a total of 642 patients. Psychiatric manifestations were grouped into five high‐level categories: behavior, psychosis, mood, sleep disturbances, and catatonia.
Results
In our cohort of 642 patients, 318 (49.6%) had psychiatric symptoms at the time of initial presentation, including 78.2% with autoimmune etiologies and 35.2% with viral etiologies (P < 0.001). Those with psychiatric symptoms were younger (median age 47.5 vs. 51.5; P < 0.001), and more likely to have a history of documented psychiatric disorders, as well as longer lengths of hospital stay, and poorer discharge outcomes. Of patients initially admitted to a psychiatric service (n = 28), most had autoimmune causes, although 3 out of 28 (10.7%) had herpes viral infections; admission to a psychiatric service was associated with substantially longer interval to initiation of antivirals and immunotherapy. Autoimmune and infectious etiologies differed in the spectrum and frequency of psychiatric manifestations.
Interpretation
Psychiatric symptoms are common across etiologies of encephalitis and are associated with longer lengths of hospital stay and worse clinical outcomes. Specific patterns and dimensionality of psychiatric symptoms distinguish autoimmune from infectious causes.
Introduction
Encephalitis is characterized by inflammation of the brain parenchyma and results from infectious or autoimmune causes. 1 , 2 Prompt recognition of the syndrome, identification of the specific etiology, and timely initiation of treatment are critical for optimal outcomes. 3 , 4 However, the manifold and heterogeneous clinical presentations of encephalitis can pose substantial challenges to its recognition. As a result, clinical case definitions have been developed, in part to assist clinicians in recognizing signs, symptoms, and laboratory studies consistent with encephalitis. Such case definitions have typically included an absolute requirement for sustained alteration of mental status, along with evidence of inflammation and focal neurologic involvement. 5 Psychiatric symptoms were largely unaddressed until more recently, when it has been recognized that autoimmune encephalitis, in particular, can present with psychiatric changes as the primary alteration in mentation. 6 , 7 This has resulted in the explicit incorporation of psychiatric symptoms in newer case definitions of autoimmune encephalitis. 8 When psychiatric symptoms are the leading presentation, autoimmune encephalitis has been referred to as psychiatric autoimmune encephalitis. 9
While this has resulted in improved identification of some patients with autoimmune encephalitis, the potential for misdiagnosis and exposure to unneeded immune therapies with concomitant adverse effects has arisen. Indeed, in two recent studies, each comprised of over 100 patients misdiagnosed with autoimmune encephalitis, approximately one‐fifth had a primary psychiatric disorder. 10 , 11 In addition, the emphasis of acute psychiatric manifestations in autoimmune encephalitis may bias clinicians from considering infectious causes. This may lead to delays in diagnosis and treatment of infectious encephalitis if psychiatric symptoms, which have been described in this group, 12 are common.
Thus, increased granularity regarding the psychiatric manifestations of encephalitis etiologies is needed. The purpose of this study was to characterize the frequency and spectrum of psychiatric symptoms in patients with all cause encephalitis.
Materials and Methods
Ethical considerations
The study was approved by the University of Texas Health Committee for the Protection of Human Subjects, the Harris Health System Research and Sponsor Programs, the Memorial Hermann Research Review Committee, and the John Hopkins Internal Review Board committee.
Methods
We followed the guidelines set forth by the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement 13 and checklist as best as possible considering our research question and data resources. Objective guidelines for data were followed in order to reduce potential sources of bias.
Study population
Patients aged 18 years or older with an encephalitis‐related discharge diagnosis identified with International Classification of Disease (ICD‐9) discharge codes corresponding to encephalitis 14 were identified from two hospital systems in Houston, Texas, between February 2005 and February 2023. Similarly, in Baltimore, Maryland, adult patients were identified from the Johns Hopkins Hospital and Johns Hopkins Bayview Medical Center between June 1, 2006 and March 15, 2016, with additional patients prospectively enrolled from the Johns Hopkins Hospital and outpatient center between January 2016 and December 2022. A total of 367 adults from the Houston cohort and 275 from the Baltimore cohort were included, all of whom met the International Encephalitis Consortium (IEC) and/or Graus criteria for encephalitis (Table S2). 5 , 8 Patients were included who met either or both of the IEC or Graus criteria.
Each case was analyzed for demographics, admission presentation, initial labs, diagnostic tests, and reports of imaging during hospitalization. 15 Glasgow Outcome Scales (GOS) was used to measure neurological outcomes. 16 Patients were designated as having psychiatric symptoms based on documentation from the clinical chart of any of the following symptoms within the five subtypes of symptoms: behavior (agitation, personality changes, combativeness), psychosis (delusions, paranoid ideas, hallucinations), mood disturbances (irritability, depression, anxiety), sleep disturbances (insomnia or hypersomnia), and catatonia. 17 We used a complete case analysis to address missing data.
Statistical analysis
Data were analyzed by the Jamovi project open statistical software and IBM SPSS Statistics. Descriptive analysis based on clinical features such as demographics, presentation, and clinical management was performed. Individual bivariate analyses were conducted to examine the relationship between clinically cogent variables and the presence of psychiatric symptoms. This included the use of chi‐square or Fischer's exact tests for categorical variables, t‐tests for normally distributed continuous variables, and Mann–Whitney U‐test for non‐normally distributed continuous variables. In this study, post‐hoc pairwise analyses were conducted when comparing multiple groups, to identify which specific groups differed. Subsequently, Bonferroni correction was applied to the resulting P‐values to address multiple simultaneous comparisons. InteractiVenn software was used for Venn diagrams. 18
Results
Of the 642 patients included, 318 (49.5%) presented with psychiatric symptoms (Table 1). Patients with psychiatric symptoms were more likely to have a history of psychiatric disorders (19.2% vs. 10.8%, P = 0.003), with the most common being depression (9.75%) followed by anxiety (6.92%). There was no difference in the number of patients with documented schizophrenia or other psychotic disorder between those presenting with and without psychiatric symptoms (2.22% vs. 0.61%. P = 0.088). Of patients with psychiatric symptoms, 50.9% had autoimmune etiology and 25.2% had viral etiology. Among those with autoimmune etiologies, 78.2% presented with psychiatric symptoms, while of those with viral etiologies, 35.2% presented with psychiatric symptoms. As expected, all patients initially admitted to a psychiatry service had psychiatric symptoms. Of those initially admitted to a neurology service, 54.3% presented with psychiatric symptoms, while of those admitted to a medicine service, 43.3% presented with psychiatric symptoms.
Table 1.
Baseline characteristics, nature of presentation, and etiologies of 642 adults with encephalitis with and without psychiatric symptoms.
| Variable (n/N (%)) | Psychiatric symptoms (N = 318 (49.6%)) | No psychiatric symptoms (N = 324 (50.5%)) | P‐value a |
|---|---|---|---|
| Median age, years (range) | 47.5 (5–90) | 51.5 (17–92) | <0.001 |
| Male sex | 173/318 (54.4) | 152/324 (46.9) | 0.058 |
| Psychiatric history | 60/313 (19.2) | 34/316 (10.8) | 0.003 |
| Schizophrenia | 7/318 (2.22) | 2/324 (0.61) | 0.088 |
| Bipolar disorder | 6/318 (1.89) | 5/324 (1.54) | 0.737 |
| Depression | 31/318 (9.75) | 16/324 (4.94) | 0.019 |
| Anxiety | 22/318 (6.92) | 9/324 (2.77) | 0.014 |
| Presenting service b | |||
| Medicine | 164/318 (51.6) | 215/324 (66.4) | <0.001 |
| Neurology | 126/318 (39.6) | 106/324 (32.7) | |
| Psychiatry | 28/318 (8.81) | 0/324 (0.00) | |
| Other | 0/318 (0.00) | 1/324 (0.31) | |
| Etiologies | |||
| Viral | 80/318 (25.2) | 147/324 (45.4) | <0.001 |
| Autoimmune | 162/318 (50.9) | 45/324 (13.8) | <0.001 |
| Unknown | 52/318 (16.4) | 78/324 (24.1) | 0.015 |
| Other infectious | 25/318 (7.86) | 53/324 (16.4) | <0.001 |
P‐value for difference between patients with and without psychiatric symptoms;
Medical service which patient is initially admitted to P‐values bolded if significant. Other bolded text indicates new category of data.
Patients with psychiatric symptoms were less likely to have fever, headache, and nausea, but more likely to have seizures (P < 0.05; Table 2) and normal CSF WBC (29.8% vs. 18.9%, P = 0.002), CSF glucose (15.3% vs. 24.5%, P = 0.011), and CSF protein (42.6% vs. 19.3%, P = 0.001). There were no differences in the prevalence of abnormal EEG or MRI between patients with and without psychiatric symptoms.
Table 2.
Clinical characteristics, management, and outcomes in 642 adults with encephalitis with and without psychiatric symptoms.
| Variable (n/N (%)) | Psychiatric symptoms (N = 318 (49.6%)) | No psychiatric symptoms (N = 324 (50.5%)) | P‐value a |
|---|---|---|---|
| Headache | 132/272 (48.5) | 167/286 (58.4) | 0.020 |
| Nausea | 31/120 (25.8) | 96/207 (46.4) | <0.001 |
| Acute presentation b | 125/311 (40.2) | 186/312 (59.6) | <0.001 |
| Presence of fever | 138/296 (46.6) | 194/313 (62.0) | <0.001 |
| Seizures | 160/307 (52.1) | 95/313 (30.4) | <0.001 |
| CSF c glucose <45 mg/dL, (range) | 49/302 (16.3) | 78/319 (24.5) | 0.011 |
| CSF WBC </=5 cells/mm3 (range) | 90/302 (29.8) | 60/317 (18.9) | 0.002 |
| CSF protein <50 mg/dL (range) | 127/298 (42.6) | 61/316 (19.3) | <0.001 |
| Abnormal MRI d | 156/284 (54.9) | 152/241 (63.1) | 0.059 |
| Abnormal EEG e | 199/261 (76.2) | 139/174 (79.9) | 0.37 |
| ICU f admission | 134/310 (43.2) | 119/322 (37.0) | 0.108 |
| Mechanical ventilation | 106/313 (33.9) | 93/321 (29.0) | 0.184 |
| Antipsychotics | 147/318 (46.2) | 68/322 (21.1) | <0.001 |
| Benzodiazepines | 255/317 (80.4) | 172/316 (54.4) | <0.001 |
| Steroids | 196/311 (63.0) | 177/320 (55.3) | 0.049 |
| IVIg g | 90/312 (28.8) | 24/322 (7.45) | <0.001 |
| Acyclovir use | 186/315 (59.0) | 213/317 (67.2) | 0.034 |
| In‐hospital mortality | 13/318 (4.09) | 38/324 (11.7) | <0.001 |
| Median length of stay days, (range) | 13 (1–209) | 10 (0–207) | <0.001 |
| Glasgow Outcomes Scale h <5 | 199/263 (75.7) | 204/306 (66.7) | 0.019 |
P‐values bolded if significant. Other bolded text indicates new category of data.
P‐value for difference between patients with and without psychiatric symptoms.
Presentation within 5 days of symptom onset.
Cerebral spinal fluid.
Magnetic Resonance Imaging.
Electroencephalogram.
Intensive Care Unit.
Intravenous immunoglobulin.
The Glasgow Outcome Scale (GOS) is a global scale for functional outcome that rates patient status into one of five categories: Dead (1), Vegetative State (2), Severe Disability (3), Moderate Disability (4) or Good Recovery (5). A score of <4 indicates severe disability or worse.
In terms of management of patients with and without psychiatric symptoms, no differences were found in admission to the ICU or needing mechanical ventilation (P > 0.05; Table 2). Patients with psychiatric symptoms were more often treated with antipsychotics (46.2% vs. 21.1%, P < 0.001), benzodiazepines (80.4% vs. 54.4%, P < 0.001), corticosteroids (63.0% vs. 55.3%, P < 0.001), and IVIg (28.8% vs. 7.45%, P < 0.001), and less likely to receive acyclovir (59.0% vs. 67.2%, P = 0.034). Outcomes differed between the two groups. Although patients with psychiatric symptoms had lower in‐hospital mortality rates (4.1% vs. 11.7%, P < 0.001), discharge outcome as assessed by GOS <5 was poorer in those presenting with psychiatric symptoms (75.7% vs. 66.7%, P = 0.019). In addition, the median length of stay for patients with psychiatric symptoms was longer than those without psychiatric symptoms [13 days (1–209), vs. 10 days (0–207) (P < 0.001)], respectively.
Table 3 describes differences in those presenting initially to a psychiatric service in comparison with a medicine or neurology service. Patients who presented to a psychiatric service were more likely to be younger (33 years vs. 50 years, P = 0.003) and to have prior psychiatric history (35.7% vs. 14.1%, P = 0.002). Most of the 28 patients initially admitted to a psychiatric service had an autoimmune etiology – including 12 out of 28 (42.9%) patients with anti‐NMDA receptor encephalitis – while three had herpesviral infections (2 HSV and 1 VZV). Patients admitted to a psychiatric service had substantially longer intervals between onset of symptoms and admission for encephalitis (28.5 days vs. 5 days, P < 0.001), and longer intervals to initiation of autoimmune (40.5 days vs. 14 days, P = 0.005), and antiviral treatment (29 days vs. 7 days, P = 0.01). Patients initially admitted to a psychiatric service were also more likely to have a poorer neurological outcome on discharge as measured by GOS <5 (94.7% vs. 69.7%, P = 0.019).
Table 3.
Comparison of those initially presenting to psychiatric service versus a medicine or neurology service.
| Variable (n/N (%)) | Psychiatric Service (n = 28) | Medicine/Neurology Service (n = 617) | P‐value |
|---|---|---|---|
| Median age | 33 (12–77) | 50 (5–92) | 0.003 |
| Male sex | 12/28 (42.9) | 307/617 (49.8) | 0.475 |
| Prior psychiatric history | 10/28 (35.7) | 85/604 (14.1) | 0.002 |
| Depression | 5/28 (17.9) | 42/617 (6.81) | 0.028 |
| Anxiety | 3/28 (10.7) | 28/617 (4.54) | 0.135 |
| Bipolar Disorder | 2/28 (7.14) | 10/617 (1.62) | 0.034 |
| Schizophrenia | 2/28 (7.14) | 7/617 (1.13) | 0.008 |
| Etiology | |||
| Viral | 3/28 (10.7) | 225/617 (36.5) | 0.005 |
| Autoimmune | 23/28 (82.1) | 185/617 (30.0) | <0.001 |
| Unknown | 2/28 (7.14) | 128/617 (20.7) | 0.197 |
| Other infectious | 0/28 (0.00) | 79/617 (12.8) | 0.043 |
| Median days from symptoms onset to admission | 28.5 days (IQR: 7–79.5) | 5 days (IQR: 2–18) | <0.001 * |
| Median days of symptoms prior to autoimmune therapy | 40.5 (IQR: 23–88.3) | 14 (IQR: 4–33.5) | 0.005 |
| Median days from symptoms onset to acyclovir initiation | 29 days (IQR:4–76) | 7 days (IQR:3–19) | 0.01 |
| Median days from admission to acyclovir initiation | 0 days (IQR:0–0) | 0 days (IQR: 0–1) | 0.25 |
| Glasgow outcome scale <5 | 18/19 (94.7%) | 387/555 (69.7%) | 0.019 |
| Median length of stay | 19.5 (IQR: 7–45) | 11 (IQR: 6–22) | 0.051 |
P‐values bolded if significant. Other bolded text indicates new category of data.
Etiologies of the 28 patients initially admitted to a psychiatric hospital are as follows: 12 anti‐NMDAR, 8 seronegative, 2 HSV, 1 VZV, 2 anti‐Caspr2, 1 anti‐GAD65, and 2 unknown.
We next examined the distribution of psychiatric categories among the 318 patients in our cohort presenting with psychiatric symptoms. Overall, behavioral disturbance was most common (230/318, 72.3%) while catatonia was least common (22/318, 6.9%), Figure 1A. When comparing the two largest subcategories (Fig. 1), viral versus autoimmune etiologies, differences were seen in the proportion of those with psychosis (38.8% vs. 57.8%, P = 0.005), catatonia (2.50% vs. 11.8%, P = 0.016), and sleep disturbances (17.5% vs. 47.8%, P < .001). With respect to the number of symptom categories, those with an autoimmune etiology were more likely to present with three or more categories compared to patients with other causes of encephalitis (45.1% vs. 19.1%, P < 0.001). In addition, while no patients with a viral etiology presented with all five psychiatric categories, five patients with autoimmune etiologies had psychiatric manifestations spanning all categories (Table S1).
Figure 1.
Venn diagrams showing the distribution of the five main symptom clusters (behavior, psychosis, sleep disturbances, mood symptoms, and catatonia) in all cause encephalitis (A) as well as divided by etiology into viral (B), autoimmune (C), unknown (D), and other (E). Venn diagrams were made by using InteractiVenn online software, inputting the patients that exhibited each symptom cluster and creating overlapping groups that show the number of patients who exhibit multiple symptoms. Behavior was the most common symptom in all cause encephalitis, and autoimmune etiology showed the most symptom overlaps.
We then examined the frequency and pattern of psychiatric categories in the two most common specific etiologies of encephalitis, HSV and anti‐NMDAR, in comparison with other confirmed autoimmune and viral etiologies (Table 4 and Fig. S1). Presenting psychiatric symptoms occurred in 41.1% of those with HSV, 30.0% of those with other viral encephalitides, 94.3% of those with anti‐NMDAR, and 69.1% of those with other autoimmune encephalitides (P < 0.001). More patients with anti‐NMDAR than HSV presented with three or more categories of psychiatric manifestations (34/50, 68.0% vs. 9/44, 20.5%; P < 0.001) and 4 out of 50 (8%) presented with all five psychiatric categories, as compared to none with HSV or other infectious causes. Catatonia was strongly associated with anti‐NMDAR encephalitis versus all other causes of encephalitis (14/50, 28.0% vs. 3/268, 1.1%; P < 0.001) (Table S1).
Table 4.
Comparison of psychiatric categories between HSV, anti‐NMDAR, and other encephalitides.
| HSV N = 104 | Anti‐NMDAR N = 53 | Viral, not HSV (C) N = 114 | Autoimmune, not anti‐NMDAR (D) N = 64 | P‐value | |
|---|---|---|---|---|---|
| Behavior (a) | 22/104 (21.2) | 47/53 (88.7) | 35/114 (30.7) | 34/64 (53.1) |
B vs. A < 0.001 B vs. C < 0.001 B vs. D < 0.001 |
|
D vs. A < 0.001 D vs. C = 0.03 | |||||
| Psychosis (b) | 17/104 (16.3) | 38/53 (71.7) | 17/114 (14.9) | 24/64 (37.5) |
B vs. A < 0.001 B vs. C < 0.001 B vs. D = 0.001 |
|
D vs. A = 0.02 D vs. C = 0.001 | |||||
| Mood (c) | 23/104 (22.1) | 30/53 (56.6) | 23/114 (20.2) | 30/64 (46.9) |
B vs. A < 0.001 B vs. C < 0.001 |
|
D vs. A = 0.01 D vs. C = 0.001 | |||||
| Catatonia (d) | 1/104 (1) | 14/53 (26.4) | 1/114 (0.8) | 1/64 (1.6) |
B vs. A < 0.001 B vs. C < 0.001 B vs. D < 0.001 |
| Sleep (e) | 9/104 (8.7) | 29/53 (54.7) | 10/114 (8.8) | 23/64 (35.9) |
B vs. A < 0.001 B vs. C < 0.001 |
|
D vs. A < 0.001 D vs. C < 0.001 | |||||
| P‐value |
d vs. a < 0.001 d vs. b < 0.001 d vs. c < 0.001 d vs. e = 0.01 |
a vs. b = 0.03 a vs. c < 0.001 a vs. d < 0.001 a vs. e < 0.001 |
a vs. b = 0.003 a vs. c = 0.03 a vs. d < 0.001 a vs. e < 0.001 |
d vs. a < 0.001 d vs. b < 0.001 d vs. c < 0.001 d vs. e < 0.001 |
|
|
e vs. a = 0.01 e vs. c = 0.002 |
b vs. d < 0.001 b vs. e = 0.01 |
d vs. b < 0.001 d vs. c < 0.001 d vs. e = 0.01 |
e vs. a = 0.02 | ||
|
d vs. c = 0.005 d vs. e = 0.01 |
e vs. c = 0.003 |
Discussion
In this study of over 600 patients with encephalitis, notable findings included the following: (1) psychiatric symptoms on presentation were common, occurring in half of patients; (2) although more common in those with autoimmune encephalitis, presenting psychiatric symptoms were present in over one‐third of patients with infectious causes of encephalitis, including 40% of those with HSV encephalitis; (3) the distribution and number of categories of psychiatric symptoms differed between autoimmune and infectious causes, with profiles that distinguished anti‐NMDA receptor encephalitis from other autoimmune encephalitis and from HSV encephalitis; (4) hospital length of stay was longer, and outcome by GOS poorer, in those with psychiatric manifestations; and (5) those initially admitted to a psychiatry, as opposed to medicine or neurology, service had a substantially longer duration of symptoms prior to initiation of treatment.
We found that psychiatric symptoms were common across etiologies of encephalitis, and most common in those with autoimmune encephalitis. Knowledge of the psychiatric manifestations of autoimmune encephalitis has increased markedly in recent years, 6 , 7 largely driven by the clinical and laboratory characterization of anti‐NMDAR encephalitis. 19 A decrease in glutamatergic signaling resulting from the binding of IgG antibodies with subsequent internalization of the NMDA receptor from the neuronal cell surface 20 has been proposed to underlie some of the psychiatric manifestations, which are similar to those seen in the setting of administration of the NMDA antagonists PCP or ketamine and reminiscent of the psychotic symptoms of schizophrenia. 21 , 22 However, several arguments suggest that the bases of psychiatric manifestations in encephalitis are likely to be far more extensive than can be explained by the glutamatergic dysfunction hypothesis: (1) psychiatric symptoms in anti‐NMDAR encephalitis span multiple domains, beyond psychosis 17 ; (2) other autoimmune encephalitides also present with an array of psychiatric manifestations despite the lack of evidence of direct glutamatergic disruption 23 ; and (3) in our study, over one‐third of infectious encephalitis patients also present with psychiatric manifestations, including half of those with HSV encephalitis and a third of those with VZV or WNV encephalitis. This raises the possibility that broad underlying mechanisms in infectious and autoimmune encephalitis, such as microglial activation or inflammation‐induced network disruption, with varying brain regional specificities depending upon etiology and host, may contribute to the array of psychiatric manifestations seen in both autoimmune and infectious encephalitis. 24 , 25 , 26 Taken together, while psychiatric symptoms have often been ascribed to autoimmune encephalitis, they are common in all‐cause encephalitis and their presence should not preclude consideration of an infectious cause.
Our findings of the psychopathology of anti‐NMDAR encephalitis bear similarities to a recent study by Al‐Diwani et al. consisting of a systemic review of the literature. 17 Of the five higher level psychiatric categories utilized in that study, behavioral and psychotic disorders were found in the majority of patients, followed in descending order by mood disorder, catatonia, and sleep disturbance. Similarly, in our cohort, patients with anti‐NMDAR encephalitis most commonly had behavioral and psychotic disorders, followed by mood disturbances. In addition, in approximately three‐quarters of patients in the Al‐Diwani study, the five features overlapped, most commonly in the form of mood, psychosis, and behavioral features, with or without catatonia. In that study, approximately 75% of patients exhibited more than a single psychiatric category, while in our study more than 90% exhibited more than a single category, supporting the multidimensional nature of the psychopathology of anti‐NMDAR encephalitis. Notably, in that study the proportion of anti‐NMDAR encephalitis patients with psychiatric manifestations was lower compared to ours, as was the proportion of each of the psychiatric categories, suggesting the possibility of under‐reporting of psychiatric symptoms when relying on the published literature.
In our overall cohort, behavioral symptoms were by far the most common of the high‐level psychiatric categories, while catatonia was infrequent. A similar pattern was seen among the viral subgroup, and differed from the autoimmune subgroup where psychosis, sleep disturbance, and catatonia were more prominent. The large number of patients in our study afforded us the opportunity to compare the psychopathology of several subgroups of patients, including those with HSV encephalitis, anti‐NMDAR encephalitis, and other autoimmune encephalitides, leading to some notable observations. First, in our cohort of patients with all cause encephalitis, the presence of catatonia was virtually diagnostic of anti‐NMDAR encephalitis. While well recognized in association with anti‐NMDAR encephalitis, 19 it is surprising that catatonia was not seen more commonly in other subgroups where psychosis was prevalent, and suggests that there may be specific pathophysiological mechanisms by which anti‐NMDAR antibodies drive catatonia. Second, while psychosis was a feature of all groups, it was far more prevalent in anti‐NMDAR encephalitis than viral causes of encephalitis including HSV. Also distinguishing anti‐NMDAR from HSV encephalitis was psychiatric multidimensionality, which was much more apparent in anti‐NMDAR encephalitis, where more patients had three or more psychiatric categories and some had all five, as opposed to those with HSV encephalitis. Thus, while both HSV encephalitis and anti‐NMDAR encephalitis may present with an infectious prodrome and limbic symptomatology, 27 the nature and extent of psychiatric manifestations on presentation may help to distinguish between the two. Additionally, among those with psychiatric manifestations was the prominence of sleep disturbances in autoimmune encephalitis, occurring in half of those with anti‐NMDAR encephalitis, a third with other autoimmune encephalitides, and only one tenth of those with viral causes including HSV encephalitis. While best recognized in the setting of anti‐NMDAR encephalitis and anti‐IgLON5 encephalitis, sleep disturbances have been described in a wide variety of other autoimmune encephalitides, and studies from animal models and post‐mortem tissue have suggested a number of distinct mechanisms linking the two. 28 Given the multitude of brain networks involved in initiation and regulation of sleep, however, it is surprising that viral encephalitides do not more commonly present with sleep disturbances. It is possible that sleep symptoms are overshadowed by other manifestations of viral disease – and thus under‐reported – a prospect that could be clarified by prospective studies.
In patients with new psychiatric symptoms, a history of prior psychiatric disease is often considered to be more supportive of a primary psychiatric disorder, rather than an organic disease such as autoimmune encephalitis. 29 However, we found that a substantial proportion of our patients who presented with psychiatric manifestations – one in five – had a history of prior psychiatric disease. Thus, a prior psychiatric history should not preclude consideration of the diagnosis of encephalitis in individuals who present with psychiatric symptoms. Notably, a prior psychiatric history was twice as common in those with psychiatric manifestations as compared to those without psychiatric manifestations. This observation can be hypothesized to be due to the association between neuroinflammation and psychiatric disorders. This has been considered in the literature in many different contexts, most notably in the setting of schizophrenia. Some neuroimaging studies have revealed elevated levels of inflammatory markers in the brains of individuals with schizophrenia. 30 There have also been differences seen in genes related to immune functioning in individuals with schizophrenia. Another notable relationship between inflammation and psychiatric disorders is autoimmune conditions concomitant with schizophrenia. 31 Studies that have looked specifically at autoimmune encephalitis, a small proportion of patients with schizophrenia that have been identified in one study as having anti‐NMDAr antibodies. 30 Mechanistic links between a prior psychiatric history and the presence of psychiatric symptoms in acute neuroinflammation are poorly understood and would benefit from further exploration in studies incorporating structural and functional imaging modalities.
Striking delays in admission and treatment of the underlying encephalitis occurred in those who presented to a psychiatry service first, rather than to a medicine or neurology service. Indeed, those who presented to a psychiatry service underwent evaluation for encephalitis 4 weeks after the onset of symptoms, a full 3 weeks later than those who presented to a medicine or neurology service. This, in turn, was associated with a substantially longer interval between symptom onset and initiation of treatment. Notably, while the majority of our patients first admitted to a psychiatry unit had autoimmune encephalitis, three patients had infectious causes, including two with HSV for whom early treatment with acyclovir is optimal. 29 Taken together, these findings underscore the importance of the characterization of the spectrum and pattern of psychiatric manifestations of autoimmune and infectious encephalitis, which could inform screening criteria for patients presenting to a psychiatric service, 32 , 33 thus potentially resulting in earlier recognition and management of encephalitis.
Our study had limitations. Because this was a retrospective study, it was limited to the information provided within the electronic medical record. Thus, we relied on clinicians' documentation of psychiatric symptoms rather than use of standardized rating scales, potentially leading to underreporting, as has been recognized for symptoms such as catatonia. 34 , 35 In addition, some patients who may have been diagnosed at a later time point with autoimmune or viral encephalitis may have instead been classified as unknown cause during the hospitalization. Only adult patients from the United States were included, which limits the generalizability. However, the study was strengthened by inclusion of two distinct geographic cohorts. Overall, our study is the first to document the pattern of psychiatric symptoms in a large cohort of adults presenting with multiple etiologies of encephalitis and to characterize its impact on management and outcomes.
Conclusions
Presenting psychiatric symptoms are common in encephalitis – regardless of etiology – and are associated with longer hospital stays and poorer outcomes, contributing to an increased health care burden. Misdiagnosis of patients resulting in initial admission to a psychiatry service results in substantially prolonged times to initiation of treatment for an underlying autoimmune or infectious cause. The type and number of psychiatric categories can distinguish between viral and autoimmune causes, and thus it is important to appreciate these distinctions in order to inform optimal triage and management of patients with encephalitis.
Author Contributions
Concept and design: RH, RG, JP, AV, and PB. Acquisition, analysis, or interpretation of data: all authors. Drafting of the manuscript: AV and PB. Critical revision of the manuscript for important intellectual content: all authors. Statistical analysis: RH, PB, RH, AV. Supervision: RH, AV.
Funding Information
None.
Conflicts of Interest
Potential conflicts of interest include RH has research support and personal fees from Biofire®. J.P. is a site investigator for a study sponsored by Genentech.
Supporting information
Figure S1.
Table S1.
Acknowledgements
The authors would like to acknowledge all current and prior members of the encephalitis research collaboration between Johns Hopkins and McGovern Medical School which have made this work possible.
Data Availability Statement
The participants of this study did not give written consent for their data to be shared publicly, so supporting data are not available.
References
- 1. Venkatesan A, Michael BD, Probasco JC, Geocadin RG, Solomon T. Acute encephalitis in immunocompetent adults. Lancet. 2019;393(10172):702‐716. doi: 10.1016/S0140-6736(18)32526-1 [DOI] [PubMed] [Google Scholar]
- 2. Bloch KC, Glaser C, Gaston D, Venkatesan A. State of the art: acute encephalitis. Clin Infect Dis. 2023;77(5):e14‐e33. doi: 10.1093/cid/ciad306 [DOI] [PubMed] [Google Scholar]
- 3. Venkatesan A, Geocadin RG. Diagnosis and management of acute encephalitis: A practical approach. Neurol Clin Pract. 2014;4(3):206‐215. doi: 10.1212/CPJ.0000000000000036 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Lancaster E. The diagnosis and treatment of autoimmune encephalitis. J Clin Neurol. 2016;12(1):1‐13. doi: 10.3988/jcn.2016.12.1.1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Venkatesan A, Tunkel AR, Yoder J, et al. Case definitions, diagnostic algorithms, and priorities in encephalitis: consensus statement of the international encephalitis consortium. Clin Infect Dis. 2013;57:1114‐1128. http://hdl.handle.net/1959.13/1067273 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Dalmau J, Graus F. Antibody‐mediated encephalitis. N Engl J Med. 2018;378(9):840‐851. doi: 10.1056/NEJMra1708712 [DOI] [PubMed] [Google Scholar]
- 7. Bost C, Pascual O, Honnorat J. Autoimmune encephalitis in psychiatric institutions: current perspectives. Neuropsychiatr Dis Treat. 2016;12:2775‐2787. doi: 10.2147/NDT.S82380 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Graus F, Titulaer MJ, Balu R, et al. A clinical approach to diagnosis of autoimmune encephalitis. Lancet Neurol. 2016;15(4):391‐404. doi: 10.1016/S1474-4422(15)00401-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Hansen N, Timäus C. Autoimmune encephalitis with psychiatric features in adults: historical evolution and prospective challenge. J Neural Transm. 2021;128(1):1‐14. doi: 10.1007/s00702-020-02258-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Jacobs S. Autoimmune encephalitis misdiagnosis can lead to harm in adult patients. Neurol Advisor. 2022. [Google Scholar]
- 11. Flanagan EP, Geschwind MD, Lopez‐Chiriboga AS, et al. Autoimmun enecephalitis misdiagnosis in adults. JAMA Neurol. 2023;80(1):30‐39. doi: 10.1011/jamaneurol.2022.4251 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Caroff SN, Mann SC, Gliatto MF, Sullivan KA, Campbell EC. Psychiatric manifestations of acute viral encephalitis. Psych Annal. 2001;31(3):193‐204. doi: 10.3928/0048-5713-20010301-10 [DOI] [Google Scholar]
- 13. Ev E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. Strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med. 2007;335(7624):806‐808. doi: 10.1136/bmj.39335.541782.AD [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. George BP, Schneider EB, Venkatesan A. Encephalitis hospitlizations rates and inpatietns mortality in the United States, 2000‐2010. PLoS One. 2014;9(9):e104169. doi: 10.1371/journal.pone.0104169 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Sejvar J, Eastwood K, Stahl J‐P, et al. Case definitions, diagnostic algorithms, and priorities in encephalitis: consensus statement of the international encephalitis consortium. Clin Infect Dis. 2013;57(8):1114‐1128. doi: 10.1093/cid/cit458 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. McMillan T, Wilson L, Ponsford J, Levin H, Teasdale G, Bond M. The Glasgow outcome scale — 40 years of application and refinement. Nat Rev Neurol. 2016;12(8):477‐485. doi: 10.1038/nrneurol.2016.89 [DOI] [PubMed] [Google Scholar]
- 17. Al‐Diwani A, Handel A, Townsend L, et al. The psychopathology of NMDAR‐antibody encephalitis in adults: a systematic review and phenotypic analysis of individual patient data. Lancet Psychiatry. 2019;6(3):235‐246. doi: 10.1016/S2215-0366(19)30001-X [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Heberle H, Meirelles GV, da Silva FR, Telles GP, Minghim R. InteractiVenn: a web‐based tool for the analysis of sets through Venn diagrams. BMC Bioinformatics. 2015;16(1):169. doi: 10.1186/s12859-015-0611-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Dalmau J, Armangué T, Planagumà J, et al. An update on anti‐NMDA receptor encephalitis for neurologists and psychiatrists: mechanisms and models. Lancet Neurol. 2019;18(11):1045‐1057. doi: 10.1016/S1474-4422(19)30244-3 [DOI] [PubMed] [Google Scholar]
- 20. Hughes EG, Peng X, Gleichman AJ, et al. Cellular and synaptic mechanisms of anti‐NMDA receptor encephalitis. J Neurosci. 2010;30(17):5866‐5875. doi: 10.1523/jneurosci.0167-10.2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Kayser MS, Dalmau J. Anti‐NMDA receptor encephalitis, autoimmunity, and psychosis. Schizophr Res. 2016;176(1):36‐40. doi: 10.1016/j.schres.2014.10.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Javitt DC, Zukin SR, Heresco‐Levy U, Umbricht D. Has an angel shown the way? Etiological and therapeutic implications of the PCP/NMDA model of schizophrenia. Schizophr Bull. 2012;38(5):958‐966. doi: 10.1093/schbul/sbs069 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Mueller C, Elben S, Day GS, et al. Review and meta‐analysis of neuropsychological findings in autoimmune limbic encephalitis with autoantibodies against LGI1, CASPR2, and GAD65 and their response to immunotherapy. Clin Neurol Neurosurg. 2023;224:107559. doi: 10.1016/j.clineuro.2022.107559 [DOI] [PubMed] [Google Scholar]
- 24. Najjar S, Steiner J, Najjar A, Bechter K. A clinical approach to new‐onset psychosis associated with immune dysregulation: the concept of autoimmune psychosis. J Neuroinflammation. 2018;15(1):40. doi: 10.1186/s12974-018-1067-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Peer M, Prüss H, Ben‐Dayan I, Paul F, Arzy S, Finke C. Functional connectivity of large‐scale brain networks in patients with anti‐NMDA receptor encephalitis: an observational study. Lancet Psychiatry. 2017;4(10):768‐774. doi: 10.1016/S2215-0366(17)30330-9 [DOI] [PubMed] [Google Scholar]
- 26. Heine J, Prüss H, Kopp UA, et al. Beyond the limbic system: disruption and functional compensation of large‐scale brain networks in patients with anti‐LGI1 encephalitis. J Neurol Neurosurg Psychiatry. 2018;89(11):1191‐1199. doi: 10.1136/jnnp-2017-317780 [DOI] [PubMed] [Google Scholar]
- 27. Armangue T, Leypoldt F, Dalmau J. Autoimmune encephalitis as differential diagnosis of infectious encephalitis. Curr Opin Neurol. 2014;27(3):361‐368. doi: 10.1097/wco.0000000000000087 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Munoz‐Lopetegi A, Graus F, Dalmau J, Santamaria J. Sleep disorders in autoimmune encephalitis. Lancet Neurol. 2020;19(12):1010‐1022. doi: 10.1016/S1474-4422(20)30341-0 [DOI] [PubMed] [Google Scholar]
- 29. Funayama M, Koreki A, Takata T, et al. Differentiating autoimmune encephalitis from schizophrenia spectrum disorders among patients with first‐episode psychosis. J Psychiatr Res. 2022;151:419‐426. doi: 10.1016/j.jpsychires.2022.05.008 [DOI] [PubMed] [Google Scholar]
- 30. Hansen N, Luedecke D, Maier HB, et al. NMDAR1 autoantibodies as potential biomarkers for schizophrenia phenotyping. Lancet Psychiatry. 2024;11(10):780‐781. doi: 10.1016/S2215-0366(24)00282-7 [DOI] [PubMed] [Google Scholar]
- 31. Jeppesen R, Benros ME. Autoimmune diseases and psychotic disorders. Front Psychiatry. 2019;10:131. doi: 10.3389/fpsyt.2019.00131 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Herken J, Prüss H. Red flags: clinical signs for identifying autoimmune encephalitis in psychiatric patients. Front Psychiatry. 2017;8:25. doi: 10.3389/fpsyt.2017.00025 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Warren N, Flavell J, O'Gorman C, et al. Screening for anti‐NMDAR encephalitis in psychiatry. J Psychiatr Res. 2020;125:28‐32. doi: 10.1016/j.jpsychires.2020.03.007 [DOI] [PubMed] [Google Scholar]
- 34. Llesuy JR, Medina M, Jacobson KC, Cooper JJ. Catatonia under‐diagnosis in the general hospital. J Neuropsychiatry Clin Neurosci. 2018;30(2):145‐151. doi: 10.1176/appi.neuropsych.17060123 [DOI] [PubMed] [Google Scholar]
- 35. van der Heijden FMMA, Tuinier S, Arts NJM, Hoogendoorn MLC, Kahn RS, Verhoeven WMA. Catatonia: disappeared or under‐diagnosed? Psychopathology. 2005;38(1):3‐8. doi: 10.1159/000083964 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1.
Table S1.
Data Availability Statement
The participants of this study did not give written consent for their data to be shared publicly, so supporting data are not available.