Authors: Armangue T, Spatola M, Vlagea A, et al. Frequency, symptoms, risk factors, and outcomes of autoimmune encephalitis after herpes simplex encephalitis: a prospective observational study and retrospective analysis. Lancet Neurol
Reviewed by: Kyle Hobbs, MD, Director, Neurocritical Care Fellowship, Assistant Professor of Neurology and Neurocritical Care, Wake Forest School of MedicineRead the article
*You will need to log in to US National Library of Medicine National Institutes of Health to read this article.
Neurological worsening occurs in some patients with HSV encephalitis and has been hypothesized to be related to autoimmune encephalitis, but the nature of this relationship remains poor elucidated. To investigate this further, these study authors designed a two-part study: Cohort A was a prospective observation study of patients with HSV encephalitis, while Cohort B was a retrospective study of patients who developed neurological worsening after confirmed HSV encephalitis. In Cohort A, IgG antibodies against NMDAR and other neuronal surface proteins were analyzed in serum and CSF at diagnosis, at completion of acyclovir, and at two, six and 12 months and if new symptoms occurred. Clinical information including modified Rankin Scale (mRS) also obtained at these intervals. In Cohort B, patients were evaluated by the study team while they had new symptoms after resolution of their HSV encephalitis and the same autoantibodies were evaluated as in Cohort A. Autoimmune encephalitis was defined in both cohorts as new onset of neurological symptoms or worsening of pre-existing deficits lasting more than 24 hours with negative CSF PCR for HSV and no alternate explanation for the worsening. Univariate logistic regression was used to identify factors associated with risk of developing autoimmune encephalitis and with one-year outcome; those with a p-value of 0.1 and those with biological relevance (i.e., age, CSF pleocytosis, protein concentration) were then included in a final multivariate binary logistic regression model.
Fifty-one patients (50 HSV1, 1 HSV2) were included in Cohort A. Of these, 14 patients (27 percent) developed new neurological symptoms or worsening of previous deficits indicating probable autoimmune encephalitis. Median time from HSV encephalitis to autoimmune encephalitis was 32 days (IQR 22-43; range 7-61). No differences in CSF WBC, total protein or MRI lesion volume were seen at symptom onset between those who did and did not develop autoimmune encephalitis, but cystic lesions were seen more commonly in follow-up scans (after four months). At onset of HSV encephalitis, none of the 51 patients in Cohort A had antibodies to neuronal surface antigens, but all patients who developed autoimmune encephalitis were antibody positive at onset of new CNS symptoms (nine patients with NMDAR antibodies, one with NMDAR and GABAA receptor antibodies, and five with antibodies to unknown antigens). In patients who did not develop autoimmune encephalitis, 11 (30 percent) were IgG antibody positive during follow-up (three with NMDAR antibodies, eight with antibodies to unknown antigens). Multivariate logistic regression identified the presence of antibodies at three-week follow-up as a risk factor for autoimmune encephalitis (OR 11.5, 95 percent CI 2.7-48.8; p<0.001). At one-year follow-up, patients who had autoimmune encephalitis had higher mRS scores (P<0.001), more neurological deficits (P<0.001) and were more frequently treated with antiepileptic medication (p=0.046). Of the 48 patients in Cohort B, 92 percent had neuronal surface antibodies (77 percent NMDAR, 23 percent against unknown antigens), and antibodies were more frequently detected in serum than CSF. Median time from HSV encephalitis to symptoms of autoimmune encephalitis was 31 days (IQR 25-49, range 11-306). Age-dependent symptoms (< four years old) were choreoathetosis, behavioral change, decreased level of consciousness, truncal hypotonia, dysphagia and frequent refractory seizures. These younger patients had shorter intervals between HSV and autoimmune encephalitis, more seizures, NMDAR antibodies and worse outcome at one year. Psychosis was more common in older patients, and they were more likely to respond to immunotherapy.
This study indicates that neurological worsening after HSV encephalitis is common and often due to the development of autoimmune encephalitis; in both cohorts, over 25 percent of patients developed worsening neurological symptoms with detectable antibodies to neuronal surface antigens. The lack of antibodies at diagnosis of HSV encephalitis suggests a causative role in the development of later autoimmune encephalitis, and that the HSV infection was the trigger for development of these antibodies, although a few patients without neurological worsening also had these antibodies. A strength of this study was the regular follow-up and antibody testing, as well as the prospective nature of Cohort A. This study is limited by small sample size as well as a predominance of serum testing on follow-up so that the presence of antibodies in CSF remains unknown. Future work involving multiple centers would add to the picture and help elucidate how treatment of this autoimmune phenomena can improve long-term outcomes. This study should result in a heightened awareness of the possibility of autoimmune encephalitis in HSV encephalitis patients, with rapid workup and the possible role of immune-suppressing therapy in the event of neurologic decline after acyclovir therapy. #LiteratureWatch #NEWSReview #KyleHobbs