Comparison of immunotherapy and prognosis of seizures caused by viral encephalitis and autoimmune encephalitis

Background: The immunotherapy that is more effective for seizures caused by viral encephalitis and autoimmune encephalitis and the long-term use of anti-epileptic drugs is not clear. We aimed to compare the immunotherapy and prognosis of seizures caused by viral encephalitis and autoimmune encephalitis. Methods: Clinical data of 121 patients with seizures caused by viral encephalitis and autoimmune encephalitis diagnosed and treated in the two largest tertiary general hospitals in the Yunnan Province were retrospectively collected to compare the immunotherapy used. Dynamic follow-up was performed to observe seizures and the use of antiepileptic drugs. Results: The seizure-free rates at 6 months and 12 months after the onset of viral encephalitis were 77.8% and 80.8%, respectively. In total, 79.1% of autoimmune encephalitis cases were seizure-free at 6 months after onset, and the seizure-free rate at 12 months was 91.9%. A total of 75.0% of viral encephalitis and 67.7% of autoimmune encephalitis patients discontinued antiepileptic drugs and were seizure-free at 12 months after onset. Patients with viral encephalitis treated with glucocorticoids alone had a lower risk of seizures after the acute phase than those treated with glucocorticoids combined with immunoglobulin (P < 0.05). The risk of seizures in patients with autoimmune encephalitis treated with glucocorticoids combined with immunoglobulin was lower than that in patients treated with glucocorticoids and immunoglobulin alone (P < 0.05). Conclusions: Immunotherapy may improve the seizure prognosis of patients with acute encephalitis. The prognosis of seizures due to viral encephalitis may be determined as early as 6 months after onset, while the seizure outcome of autoimmune encephalitis is further improved 12 months after onset. lobe in the hospital medical record The included patients met the following criteria: (1) clinical manifestations, inammatory cerebrospinal uid(CSF), Magnetic resonance imaging(MRI) characteristics suggesting inammation, or serum and/or CSF autoimmune encephalitis antibody test results meeting the diagnosis of autoimmune encephalitis or viral encephalitis [3,13]; (2) seizures; and (3) use of immunotherapy. The exclusion criteria included a history of epileptic seizures or the presence of brain trauma, tumours, vascular diseases and other diseases that may cause seizures. Demographic information (sex, age), main clinical manifestations in the acute phase (rst symptom, fever, disturbance of consciousness, abnormal mental behaviour, type of seizures, status epilepticus), brain MRI results, intensive care unit (ICU) admission, AEDs and the immunotherapy programme of each patient were collected. The types of seizures were divided into tonic-clonic seizures and focal seizures [6]. In this study, the patients were divided into viral encephalitis and autoimmune encephalitis. Informed consent was obtained from the research subjects. We compared the outcomes of symptomatic seizures with different immunotherapies. We found that patients with viral encephalitis treated with glucocorticoids alone had a better prognosis for seizures at 12 months after onset than those treated with glucocorticoids combined with immunoglobulin. Patients with symptomatic seizures caused by autoimmune encephalitis often use glucocorticoids combined with immunoglobulin in the acute phase, and glucocorticoids are used to suppress immune inammation at high doses. The statistical results showed that the prognosis of symptomatic seizures treated with glucocorticoids combined with immunoglobulin was better than that treated with glucocorticoids or immunoglobulin alone. In our study, patients with viral encephalitis had similar seizure rates at 6 months and 12 months after onset. However, patients with autoimmune encephalitis had signicantly lower seizure rates at 12 months after onset than at 6 months. Therefore, we hypothesize that the duration of the acute phase of viral encephalitis is shorter than that of autoimmune encephalitis and that the rate of symptomatic seizures largely does not change after the acute phase. We speculate that most patients without risk factors for post-encephalitis epilepsy may not need long-term AED after 6 months of viral encephalitis onset.


Introduction
Encephalitis has a high incidence and may lead to death and disability. Epileptic seizures are a common symptom of acute encephalitis and an important factor affecting its prognosis. The incidence of symptomatic seizures may be up to 70% or 50% in the acute phase of autoimmune encephalitis or viral encephalitis, and status epilepticus occurs in some patients [1][2][3]. Autoimmune encephalitis is gradually being recognized and valued. Immunotherapy is the main treatment for autoimmune encephalitis [4]. The application of immunotherapy and anti-epileptic drugs (AEDs) substantially improves the prognosis of autoimmune encephalitis, including symptomatic seizures [5][6][7]. In fact, immunotherapy, especially glucocorticoids, is sometimes used in acute viral encephalitis [8]. It was previously reported that viral encephalitis is mainly treated with antiviral therapy and can be supplemented with corticosteroid therapy [9,10,11]. However, whether immunotherapy has a positive effect on the prognosis of acute viral encephalitis and the conclusions of various studies are still inconsistent [12]. In addition, current research reports are mostly related to autoimmune encephalitis and childhood viral encephalitis. Our study observed and compared the clinical application of immunotherapy and the e cacy of epileptic seizure control in patients with new-onset epileptic seizures caused by viral encephalitis and autoimmune encephalitis. Prognostic research on symptomatic seizures can help identify good prognosis, avoid unnecessary use of anti-epileptic drugs for too long, guide early intervention measures, and improve prognosis.

Patient selection
Using "autoimmune encephalitis", "viral encephalitis", "Anti-N-methyl-D-aspartate Receptor (anti-NMDAR) encephalitis", "Anti-Leucine-rich Glioma-inactivated Protein1(anti-LGI1) encephalitis", and "limbic lobe encephalitis" as the search terms, we queried the two largest tertiary general hospitals in the Yunnan Province in Southwest China (The First A liated Hospital of Kunming Medical University and The First People's Hospital of Yunnan Province) from June 2010 to June 2019 in the hospital medical record systems. The included patients met the following criteria: (1) clinical manifestations, in ammatory cerebrospinal uid(CSF), Magnetic resonance imaging(MRI) characteristics suggesting in ammation, or serum and/or CSF autoimmune encephalitis antibody test results meeting the diagnosis of autoimmune encephalitis or viral encephalitis [3,13]; (2) seizures; and (3) use of immunotherapy. The exclusion criteria included a history of epileptic seizures or the presence of brain trauma, tumours, vascular diseases and other diseases that may cause seizures. Demographic information (sex, age), main clinical manifestations in the acute phase ( rst symptom, fever, disturbance of consciousness, abnormal mental behaviour, type of seizures, status epilepticus), brain MRI results, intensive care unit (ICU) admission, AEDs and the immunotherapy programme of each patient were collected. The types of seizures were divided into tonic-clonic seizures and focal seizures [6]. In this study, the patients were divided into viral encephalitis and autoimmune encephalitis. Informed consent was obtained from the research subjects.

Immunotherapy
The immunotherapy drugs used were glucocorticoids (methylprednisolone, dexamethasone, prednisone) and intravenous immunoglobulin (5 days, 0.4 g/kg/day). There were no second-line immunotherapy drugs.
Immunotherapy was divided into only glucocorticoids and only immunoglobulin and glucocorticoids combined with immunoglobulin. According to dexamethasone 0.75 mg = prednisone 5 mg = methylprednisolone 4 mg, all glucocorticoid doses were converted into equivalent methylprednisolone. The maximum dose of methylprednisolone ≥500 mg/day was de ned as high-dose methylprednisolone, and <500 mg/day was de ned as non-high-dose methylprednisolone.

Follow up
Epileptic seizures and AED administration at 6 and 12 months after onset were obtained by telephone or outpatient follow-up. During the follow-up period, if the patient and their relatives or specialists did not nd any clinical symptoms (including aura) of seizures (focal or generalized seizures), it was de ned as seizure-free [6].
Data analysis SPSS version 23.0 (SPSS Inc., Chicago, IL, USA) and GraphPad Prism 5 (GraphPad Software, San Diego, California USA) were applied for the statistical analyses. The continuous variables did not follow a normal distribution and were described as the medians (P25, P75). The Mann-Whitney U rank sum test was used for comparisons between two groups. The Pearson chi-square or Fisher's exact test was used to evaluate differences in the categorical variables, described by n (%). Binary logistic regression was used to analyse the risk factors for seizures at 6 and 12 months after onset. The test level was 0.05, and P <0.05 indicated a statistically signi cant difference.

Demographics and baseline characteristics
The baseline characteristics of all included patients and diagnostic subgroups are shown in Table 1 According to the diagnosis, there were 54 patients (44.6%) in the viral encephalitis group and 67 patients (55.4%) in the autoimmune encephalitis group. There were no signi cant differences between the two groups in terms of demographic characteristics, most clinical manifestations in the acute phase, abnormal MRI rate or ICU admission rate.
Compared with the autoimmune encephalitis group, the viral encephalitis group had relatively fewer mental and behavioural abnormalities (P = 0.021), manifested as focal seizures (P = 0.001), and they developed status epilepticus (SE) (P = 0.008) at a lower rate. Nine patients in the viral encephalitis group were not treated with AEDs to control seizures, and fewer patients were treated with AEDs in combination than those in the autoimmune encephalitis group (20.4% vs 43.3%, P = 0.003).

Immunotherapy
All patients received immunotherapy with rst-line drugs. Glucocorticoids were administered to 113 patients (93.3%), of whom 50 (41.3%) were treated alone and 63 (52.1%) with combination immunoglobulin. Only 8 patients (6.6%) received immunoglobulin alone. Compared with autoimmune encephalitis, there were signi cantly more patients treated with glucocorticoids alone in the viral encephalitis group (64.8% vs 22.4%) and signi cantly fewer patients treated with glucocorticoids in combination with immunoglobulin (29.6% vs 70.1%). There were statistically signi cant differences between the two groups in terms of immunotherapy for encephalitis (P < 0.001) ( Table 1). Patients in the viral encephalitis group received adjuvant corticosteroids, most of which were dexamethasone, prednisolone, or methylprednisolone <500 mg/day (42/51, 82.4%), and only nine patients received high-dose methylprednisolone. Patients in the autoimmune encephalitis group were mostly treated with intravenous pulses of high-dose methylprednisolone (44/62, 71.0%). The median dose of methylprednisolone was 106 mg/day in the viral encephalitis group and 500 mg/day in the autoimmune encephalitis group (P < 0.001) ( Figure 1).

Follow-up results
At the follow-up 6 months after onset, 95 of the 121 patients (78.5%) with encephalitis were seizure-free, 53 (43.8%) continued to take AEDs, and 61 (50.4%) had no seizures after discharge and no seizures after discontinuation of AEDs. There were no statistically signi cant differences in the seizure-free rate (77.8% vs 79.1%) or AEDs rate (35.2% vs 50.7%) between the viral encephalitis group and the autoimmune encephalitis group (P > 0.05). A total of 61.1% of patients with viral encephalitis achieved discontinuation of AEDs without seizures, which was signi cantly higher than that of the autoimmune encephalitis group (41.8%) (P = 0.009) ( Table 2).
A total of 26 (21.4%) of these patients with encephalitis still had seizures at 6 months after onset ( Figure 2a); 19 were still taking AEDs, with a drug taking rate of 73.1%, which was signi cantly higher than the 35.8% taking rate of 95 patients without seizures (P = 0.001). The same trend was observed in the viral encephalitis and autoimmune encephalitis groups (Figure 2b).
A total of 56.2% of encephalitis patients (including 35 viral encephalitis and 33 autoimmune encephalitis) had withdrawn AEDs at 6 months after onset. The seizure-free rate among these patients was 89.7%, higher than the 64.2% rate among the 53 patients who took AEDs. Moreover, 33 (94.3%) patients with viral encephalitis and 28 (84.8%) patients with autoimmune encephalitis experienced remission of seizures after discharge, and there was no statistically signi cant difference between the two groups (P > 0.05) (Figure 2c). After 12 months, 4 patients were lost to follow-up, and 3 died of myocardial infarction or accident. In total, 114 patients, including 52 with viral encephalitis and 62 with autoimmune encephalitis, were followed up. Ninety-nine (86.8%) patients with encephalitis were seizure-free. The seizure-free rate of the autoimmune encephalitis group (57 patients, 91.9%) was higher than that of the viral encephalitis group (42 patients, 80.8%), but the difference between the two groups was not statistically signi cant (P > 0.05). The overall rate of taking AEDs was reduced to 28.1% (32 patients). The rate of taking in the viral encephalitis group (12 patients, 23.1%) was similar to that of the autoimmune encephalitis group (20 patients, 32.3%), and they all took only one AED. In this period, the rate of discontinuation of AEDs without seizures in the viral encephalitis group (75.0%) was higher than that in the autoimmune encephalitis group (67.7%), and the difference was statistically signi cant (P = 0.015) ( Table 2).
By 12 months after onset, the epileptic seizure rate of encephalitis had dropped to 13.2%, 19.2% for viral encephalitis and 8.1% for autoimmune encephalitis (Figure 2a). Of the 15 patients who still had seizures (10 with viral encephalitis and 5 with autoimmune encephalitis), only one (6.7%) patient did not continue to take AEDs.
Among those who were seizure-free, the rate of continued use of AEDs was reduced to 18.2% (18 patients), with only 3 (7.1%) patients with viral encephalitis and 15 (26.3%) patients with autoimmune encephalitis continuing to take AEDs (Figure 2b). The seizure-free rate of patients with autoimmune encephalitis and viral encephalitis who stopped taking AEDs was 100% and 97.5%, respectively, both of which were signi cantly higher than those in the continuing group (Figure 2c).
The seizure-free rates of viral encephalitis and autoimmune encephalitis at 12 months after onset were higher than those at 6 months after onset. The difference in seizures between the two follow-up periods was statistically signi cant (P < 0.001). Meanwhile, the rate of AED use at 12 months was lower than that at 6 months (P < 0.001). In addition, 6 new patients with viral encephalitis and 14 patients with autoimmune encephalitis achieved discontinuation of AEDs without seizures 12 months after the onset of disease, except for the 61 patients who had no seizures when the drug was stopped at the follow-up 6 months after the onset of the disease. Thus, 81 patients (71.1%) were seizure-free after the drug was stopped, and the drug was successfully withdrawn. There was a statistically signi cant difference in the seizure-free rate after AED withdrawal between the two follow-up periods (P < 0.001) ( Table 2). The prognosis of epileptic seizures at 12 months after onset was better than that at 6 months.
Demographic and acute phase characteristics and prognosis of epileptic seizures in viral encephalitis and autoimmune encephalitis The relationships between the prognosis of seizures with viral encephalitis and autoimmune encephalitis and demographic information, clinical symptoms in the acute phase, and treatment were assessed (Table 3). Univariate analysis showed that fever in the acute phase (P = 0.026, OR 11.7 95% CI 1.4-101.1) affected the prognosis of epileptic seizures 6 months after the onset of viral encephalitis, while admission to the ICU and glucocorticoid therapy alone were correlated with the prognosis of epileptic seizures at 6 and 12 months (P < 0.05). Immunotherapy was related to the prognosis of epileptic seizures at 6 and 12 months after the onset of autoimmune encephalitis (P < 0.05) ( Table 3).
The variables with P < 0.1 in the univariate analysis of the two groups of encephalitis as well as SE and ICU were included in the binary logistic regression analysis. The results showed that fever was the only risk factor for a poor prognosis of epileptic seizures after 6 months of viral encephalitis. Patients with fever were 9.5 times more likely to have seizures 6 months after onset than those without fever (P = 0.043, OR 9.5 95%CI 1.1-84.5). Patients with viral encephalitis treated with glucocorticoid alone had a lower risk of seizures after 12 months than those treated with glucocorticoids combined with immunoglobulin (P = 0.006, OR 0.95% CI 0.0-0.5) ( Table 4).
Multivariate regression analysis showed that immunotherapy could affect the outcome of autoimmune encephalitis epileptic seizures. The risk of epileptic seizures in patients treated with glucocorticoids alone at 6 months and 12 months was 5.6 times and 13.5 times higher than that in patients treated with glucocorticoids combined with immunoglobulin, respectively (P = 0.015, OR 5.6 95%CI 1.4-22.4; P = 0.031, OR 13.5 95%CI 1.3-143.6). Patients treated with immunoglobulin alone were 12.6 times more likely to have seizures at 6 months than those treated with glucocorticoid combined with immunoglobulin (P = 0.014, OR 12.6 95%CI 1.7-94.5) ( Table   4).

Acute immunotherapy programmes and prognosis of epileptic seizures in autoimmune encephalitis
Patients with autoimmune encephalitis received three immunotherapy programmes in the acute phase, and there were differences in the prognosis of epileptic seizures within 12 months of onset. The rate of achieving seizurefree AEDs in patients treated with glucocorticoids alone and immunoglobulin alone in the acute phase was 60.0%

Discussion
For the rst time, we compared the prognosis of symptomatic seizures in acute viral encephalitis and autoimmune encephalitis under different immunotherapies. Patients with viral encephalitis are mostly treated with non-high-dose glucocorticoid monotherapy, while patients with autoimmune encephalitis are mainly treated with high-dose methylprednisolone combined with immunoglobulin. After discharge, the epileptic seizures of patients with both viral encephalitis and autoimmune encephalitis were gradually controlled over time, and the demand for AEDs was gradually reduced, with most patients, particularly those with autoimmune encephalitis, successfully withdrawing AEDs within 12 months. Immunotherapy can improve the prognosis of symptomatic seizures caused by acute autoimmune encephalitis, as well as acute viral encephalitis. Patients with autoimmune encephalitis treated with combined glucocorticoids and immunoglobulin and patients with viral encephalitis treated with glucocorticoids alone have a better prognosis for symptomatic seizures. This study will be helpful for optimizing the treatment strategy of symptomatic epileptic seizures caused by acute autoimmune and viral encephalitis and shortening the course of unnecessary AEDs.
In our study, 13.2% of patients with acute viral and autoimmune encephalitis still had seizures, and 28.1% insisted on taking AEDs. Moreover, these patients only needed to receive a single AED treatment 12 months after onset. Compared with some previous studies on similar causes of encephalitis, our study showed a better prognosis of seizures and a lower dependence rate of AEDs [14,15]. Singh de ned post-encephalitis epilepsy patients as those who had taken AEDs for more than 12 months, with a median follow-up time of 43 months. Overall, 29.9% of patients with acute encephalitis developed post-encephalitis epilepsy, while 46.3% of patients with symptomatic seizures in the acute phase of encephalitis developed post-encephalitis epilepsy and required long-term medication [14]. In another study on the seizure prognosis of infective and autoimmune encephalitis in children, 21% of children with acute encephalitis developed post-encephalitis epilepsy, 10% had drug-resistant epilepsy, and 15% needed long-term antiepileptic drugs after a follow-up time ≥24 months [15]. All these studies were followed up for longer than 12 months, and the seizure rate during follow-up was higher than that in our study. In terms of the speci c causes of encephalitis, 20.9% of patients with autoimmune encephalitis had seizures 6 months after onset, while only 8.1% of patients had seizures at 12 months after onset, which was similar to that in the study of seizures caused by autoimmune encephalitis [6]. In a study of post-encephalitis epilepsy in children in Taiwan . In our study, only 19.2% of the patients with viral encephalitis still had seizures 12 months after onset. In our study, symptomatic seizures may have had a better prognosis because the subjects of our study were mainly adults. In contrast, the central nervous system of children has not yet been fully developed, and the outcome of epileptic seizures after encephalitis may be different from that of adults. In addition, in the Taiwan cohort, viral encephalitis with poor prognosis was the predominant aetiological encephalitis. Most importantly, our subjects were all treated with immunotherapy. Glucocorticoids contribute to the control of epileptic seizures [17,18], especially those associated with in ammation [19]. Children with herpes simplex viral encephalitis who are treated with glucocorticoids can be free of seizures after treatment [20]. After statistical analysis, our results showed that immunotherapy has a positive effect on the prognosis of epileptic seizures caused by viral and autoimmune encephalitis. Immunotherapy can therefore improve the prognosis of autoimmune-related seizures [21][22][23].
We compared the outcomes of symptomatic seizures with different immunotherapies. We found that patients with viral encephalitis treated with glucocorticoids alone had a better prognosis for seizures at 12 months after onset than those treated with glucocorticoids combined with immunoglobulin. Patients with symptomatic seizures caused by autoimmune encephalitis often use glucocorticoids combined with immunoglobulin in the acute phase, and glucocorticoids are used to suppress immune in ammation at high doses. The statistical results showed that the prognosis of symptomatic seizures treated with glucocorticoids combined with immunoglobulin was better than that treated with glucocorticoids or immunoglobulin alone.
In our study, patients with viral encephalitis had similar seizure rates at 6 months and 12 months after onset.
However, patients with autoimmune encephalitis had signi cantly lower seizure rates at 12 months after onset than at 6 months. Therefore, we hypothesize that the duration of the acute phase of viral encephalitis is shorter than that of autoimmune encephalitis and that the rate of symptomatic seizures largely does not change after the acute phase. We speculate that most patients without risk factors for post-encephalitis epilepsy may not need long-term AED after 6 months of viral encephalitis onset.
Patients with acute epileptic seizures, epileptic status, abnormal electroencephalography (EEG) or admission to the ICU are at increased risk of epilepsy after the acute phase [15,16,24,25]. Our study found that acute fever is a risk factor for seizures in patients with viral encephalitis 6 months after onset. Fever can induce seizures in the acute phase of encephalitis [26,27], and the proin ammatory response of the brain caused by seizures can also cause fever. This process forms a vicious cycle, resulting in the recurrence or aggravation of seizures. Previous studies have shown that epilepsy in some patients after the acute phase of encephalitis is resistant to drugs, and patients are prone to developing drug-refractory epilepsy [28,29]. In this study, fever in the acute phase was a risk factor for poor prognosis of epileptic seizures in patients with viral encephalitis; thus, it is necessary to be cautious in the discontinuation of AEDs in such patients. Rismanchi proposed that children who require AEDs when discharged from the hospital were more likely to develop chronic epilepsy [30]. Statistically, we found that choosing whether to take AEDs for a long time was not the main factor affecting the prognosis of epileptic seizures. Long-term use of AEDs in patients with encephalitis led to a higher seizure rate than in those who stopped taking AEDs. This nding does not prove that long-term AEDs are a risk factor for seizures. Instead, the risk may be due to poor seizure control in patients with encephalitis and the need for long-term use of AEDs.
In this study, the identi cation rate of viral pathogens of viral encephalitis was relatively low, and the antibody types of some patients with autoimmune encephalitis were not completely determined. Clinical diagnosis and treatment were mainly based on clinical symptoms, blood, cerebrospinal uid, and MRI. For patients with acute epileptic seizures who are clinically diagnosed with viral encephalitis or autoimmune encephalitis, the viral pathogens and antibody types are not yet clear. In addition to actively looking for the speci c cause of encephalitis, we suggest that immunotherapy may be given in the acute phase as appropriate. Viral encephalitis can be treated with anti-in ammatory non-high-dose glucocorticoids, but the prognosis of seizures in patients with acute fever is poor. Patients should be followed up to observe the epileptic seizures, and targeted and personalized AED guidance should be given. Patients with autoimmune encephalitis are given high-dose glucocorticoids combined with immunoglobulin therapy in the acute period to improve the prognosis of symptomatic epileptic seizures. During immunotherapy, attention should be paid to the side effects of drugs.
However, this study also has some de ciencies. First, it was a retrospective observational cohort study, which may have partial recall bias. In addition, due to the limitation of sample size, this study did not analyse the subtypes of autoimmune encephalitis antibody type or viral encephalitis virus type. A prospective study with a larger sample size will be conducted in the future to supplement and verify the results.

Conclusion
Immunotherapy can be used in acute autoimmune encephalitis, as well as acute viral encephalitis. Patients with autoimmune encephalitis are usually treated with combined glucocorticoids and immunoglobulin in the acute period. Viral encephalitis is mainly treated with non-high-dose glucocorticoid adjuvant therapy. After immunotherapy for acute encephalitis, the prognosis of epileptic seizures of viral encephalitis may be determined as early as 6 months after onset, while the seizure outcome of autoimmune encephalitis will be further improved 12 months after onset.

Abbreviations
AEDs:Anti-epileptic drugs; Anti-NMDAR: Anti-N-methyl-D-aspartate Receptor; anti-LGI1:Anti-Leucine-rich Glioma- Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Competing interests The authors declare that they have no competing interests.   Acute immunotherapy and prognosis of seizures in patients with autoimmune encephalitis (a); prognosis of seizures in patients with acute fever and viral encephalitis (b) M month.