Does early treatment improve outcomes in N-methyl-d-aspartate receptor encephalitis?
Abstract
N-methyl-d-aspartate (NMDA) receptor encephalitis is a treatable cause of autoimmune encephalitis in both children and adults. It is still unclear if the natural history of the condition in children is altered by early treatment with immunosuppressive therapy. We looked at the outcomes of five children (two males, three females; mean age 6y 9mo, range 4–8y) who were treated empirically for autoimmune encephalitis within a brief period of presentation. Features that led clinicians to suspect autoimmune encephalitis included prominent neuropsychiatric features, movement disorder, seizures, and dysautonomic features. Immunosuppressive therapy was carried out in all cases. In this series of children, in whom the median time from symptom onset to treatment was 5 days and median length of time for follow-up was 24 months, four out of the five (80%) recovered to their baseline. Early initiation of immunosuppressive therapy may result in improved clinical outcomes.
What this paper adds
- Early treatment of N-methyl-d-aspartate (NMDA) receptor encephalitis in children may improve outcome.
Abbreviation
-
- NMDA
-
- N-methyl-d-aspartate
N-methyl-d-aspartate (NMDA) receptor encephalitis is a cause of autoimmune encephalitis in both children and adults. It is still unclear if the natural history of the condition in children is altered by early treatment with immunosuppressive therapy. There is speculation, however, that earlier treatment results in improved outcomes with fewer residual deficits. Finke et al.1 demonstrated an improved cognitive outcome in a small cohort of adult patients with anti-NMDA encephalitis who were treated with immunomodulatory therapy within 3 months of disease onset compared with those who were treated at a later stage or not at all. The authors proposed that a delay in treatment led to permanent hippocampal damage.1 In their study of autoimmune encephalitis in children, Hacohen et al.2 report that 29% of children with anti-NMDA and anti-voltage-gated potassium channel autoimmune encephalitis made a full recovery; however, their report does not state the time interval between presentation and treatment. A study by Titulaer et al.3 examining a large cohort with anti-NMDA receptor encephalitis reported that the average time from symptom onset to treatment in children was 21 days, and that 60% of children had a modified Rankin Scale score4 of 0, at a median follow-up time of 24 months. No study to date has shown definitively that earlier immunological treatment in children alters outcome.
We studied the outcomes of five children who were treated empirically for autoimmune encephalitis within 1 week of symptom onset. The children presented at two teaching hospitals in Dublin between 2007 and 2012. Review of anonymized data was permitted by the hospital audit board.
Before the onset of symptoms, all five children had attained typical developmental milestones and attended mainstream school. Details of individual cases are presented in Table 1.
| Case no. | Age at onset (y) | Sex | Length of in-patient stay (wks) | Prodrome (duration) | Reason for presentation to medical services | Other features of NMDA encephalitis | Investigations | Intensive care unit | NMDA antibody | Time from admission to treatment | Outcome |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 6 | F | 6 | No infection | Change in behaviour for 3d then seizure with residual left hemiparesis | Dysphasia, hallucinations, orofacial dyskinesia, chorea, self-mutilation, sleep cycle disturbance, fever, dysautonomia | CSF white cells 26/μL, abnormal MRI, abnormal EEG, no tumour | No | +ve serum, CSF not tested | Day 2 i.v. methylprednisolone and oral taper | Complete recovery, mainstream school |
| 2 | 6 | M | 4 | Throat infection (10d) | Admitted through psychiatry service with 2d history of confusion, bizarre behaviour and hallucinations | Seizure, dysphasia, orofacial dyskinesia, tremor, sleep cycle disturbance, fever | CSF white cells 1/μL, abnormal MRI imaging, abnormal EEG, no tumour | No | +ve serum, +ve CSF | Day 2 i.v. methylprednisolone and oral taper, day 2 i.v. immunoglobulin for 5d | Complete recovery, mainstream school |
| 3 | 7 | F | 7 | Fever and arthralgia (2d) | Generalized seizure | Behavioural change, dysphasia, agitation, hallucinations, orofacial dyskinesia, chorea, myoclonus, fever, dysautonomia | CSF white cells 4/μL, abnormal MRI, abnormal EEG, no tumour | Yes, 16d | −ve serum, +ve CSF | Day 4 i.v. methylprednisolone and oral taper, day 11 i.v. immunoglobulin for 5d | Ongoing seizures and behavioural issues, special school |
| 4 | 11 | F | 12 | Abdominal pain (14d) | Admitted with 3d history of confusion, behavioural change, aggression, inappropriate speech and cursing | Seizure, dysphasia, agitation, hallucinations, orofacial dyskinesia, chorea, echolalia, sleep cycle disturbance, fever, dysautonomia | CSF white cells 107/μL, normal MRI, abnormal EEG, no tumour | Yes, 23d | +ve serum, +ve CSF | Day 3 i.v. methylprednisolone and oral taper, day 11 i.v. immunoglobulin for 5d, day 16 PLEX (eight cycles) | Complete recovery but with weight gain, mainstream school |
| 5 | 4 | M | 3 | Gum infection (2d) | Admitted with lethargy, confusion and coma for 2d (GCS score 11/15) | Dysphasia, agitation, low sodium, fever, dysautonomia | CSF white cells 1/μL, abnormal MRI, abnormal EEG, no tumour | Yes, 7d | +ve serum, CSF not tested | Day 3 i.v. methylprednisolone and oral taper | Complete recovery, mainstream school |
- Features that led clinicians to suspect NMDA receptor encephalitis are highlighted in bold. F, female; M, male; CSF, cerebrospinal fluid; MRI, magnetic resonance imaging; EEG, electroencephalography; GCS, Glasgow Coma Scale; PLEX, plasma exchange.
The median time from neurological or psychiatric symptom onset to empirical treatment with immunosuppression was 5 days (range 3–6d).
The median length of follow-up was 24 months (range 2–36mo). At follow-up, four children had a modified Rankin Scale score of 0 and were attending mainstream school (cases 1, 2, 4, and 5). Only one child (case 3) did not recover to the pre-illness baseline. This child continues to experience seizures, exhibits marked behavioural and cognitive difficulties, and attends a special school (modified Rankin Scale score 2).
Immunosuppressive therapy was instituted in all children before the results of the NMDA antibody test became available because they had features consistent with autoimmune encephalitis. Intravenous methylprednisolone was prescribed at a dose of 30mg/kg/day for 3 days, followed by oral prednisone for 2 to 4 weeks (2mg/kg/d) and, where prescribed, 0.4g/kg/day intravenous immunoglobulin for 5 days.
Features that led clinicians to suspect autoimmune encephalitis in cases 1 to 4 included prominent neuropsychiatric features (sudden and marked behavioural change and the onset of vivid visual hallucinations), prominent movement abnormalities (orofacial dyskinesia, choreoathetosis), and dysautonomic features (unstable pulse, blood pressure, and flushing). In case 5, although there was only one white cell in the cerebrospinal fluid, the diagnosis was considered as the child presented with coma, bitemporal magnetic resonance image changes, and hyponatraemia, a feature more often described with anti-voltage-gated potassium channel receptor encephalitis than with anti-NMDA receptor encephalitis.
All children continued on antimicrobial therapy and antiviral therapy until polymerase chain reaction or cultures were negative.
In this series of five children, in whom the median time from symptom onset to treatment was less than 1 week, four recovered completely, in comparison with 29% and 60% of children in previous studies.2, 3
This may suggest that earlier initiation of immunosuppressive therapy results in improved outcomes; however, this case series is limited by small numbers. In addition, follow-up neuropsychological assessment was not carried out in cases 2 and 5 because of rapid return to baseline and good outcome reported by parents and school.
A systematic review and meta-analysis of NMDA cases in the literature might determine whether early treatment of anti-NMDA receptor encephalitis results in an improved outcome.
