Encephalatrophy and anti-N-methyl-D-aspartate receptor encephalitis

Objective: This study aims to find whether there are differences of clinical and laboratory features between anti-NMDAR encephalitis patients with and without brain atrophy. We also compare brain atrophy scale in anti-NMDAR positive patients with and without teratomas. Methods: Brain atrophy scales of 82 patients with anti-NMDAR encephalitis were measured with the median temporal lobe atrophy (MTA) scale and the global cortical atrophy (GCA) scale. They were divided into two groups by brain atrophy status. In addition, 48 female patients with GCA(+) than thoes without ovarian teratomas (P=0.006). Conclusions: Brain atrophy are not uncommon in patients with anti-NMDAR encephalitis. MTA and GCA may be associated with severity and prognosis of anti-NMDAR encephalitis.


Introduction
Anti-N-methyl-D -aspartate receptor (anti-NMDAR) encephalitis is a form of inflammatory encephalopathies in association with antibodies (immunoglobulin G) against the GluN1 subunit of the NMDAR. It predominantly affects young women with or without ovarian teratomas [1][2][3].
A variety of magnetic resonance imaging (MRI) findings have been reported in patients with anti-NMDAR encephalitis [2,[4][5][6]. However, few stuies focus on the brain atrophy in anti-N-methyl-D-aspartate receptor encephalitis. Iizuka et al found that cerebellar atrophy was irreversible and associated with poor outcome in patients with anti-N-methyl-D-aspartate receptor encephalitis [7]. However, This study based on only 15 patients. Further studies are needed to confirm the clinical association between brain atrophy and anti-N-methyl-D-aspartate receptor encephalitis.
This study aims to find whether there are differences of clinical and laboratory features between anti-NMDAR encephalitis patients with and without brain atrophy.

1. Study population
Our database comprised 82 patients with anti-NMDAR encephalitis who were admitted from March 2014 to December 2018 in the department of neurology of the Third Affiliated Hospital of Sun Yat-sen University. Anti-NMDAR encephalitis was diagnosed in the presence of a core syndrome of limbic encephalitis ( psychiatric symptoms, seizures, conscious disturbance and memory deficits) and detection of IgG antibodies against NMDAR [8]. And it is important to carefully exclude relevant differential diagnoses. Gender, age, modified Rankin Scale (mRS) and clinical manifestations were recorded. All patients underwent routine abdominal and pelvic ultrasound. And a pelvic computed tomography (CT) scan or magnetic resonance imaging (MRI) scan was performed when an ovarian teratoma was suspected. Treatments included first-line immunotherapy, second-line immunotherapy and tumour removal. First-line immunotherapy included the use of steroids, intravenous immunoglobulins or plasma exchange alone or combined.
Second-line immunotherapy included rituximab, azathioprine or cyclophosphamide treatment alone or combined. The initial treatment was recorded as a failure if no sustained improvement occurred within 1 month of initiation of immunotherapy or tumour removal, and if the mRS score remained at 4 or higher.  score means more atrophy in the region (table1) [9]. We evaluated severity of GCA according to Pasquier and co-workers on a 4-point rating scale ( (0 = no atrophy; 1=mild GCA, sulcal opening peripherally; 2=moderate GCA, widening along the length of the sulci; 3=severe GCA, gyral thinning [10]. A neurologist and a neuroradiologist independently read all MRIs, blinded to clinical data.

Statistics analyses
All statistical analyses were conducted by using the SPSS 13.0 package for Windows (SPSS Inc, Chicago, IL, USA). Continuous data were presented as the mean ± standard deviation or median (range).
Continuous data were compared by Student's t-test or the Mann Whitney U test. Nominal variables were analyzed using the chi-square test or Fisher's exact test.

Clinical correlation between anti-NMDAR encephalitis patients with and without brain atrophy
As shown in Table 3, patients with anti-NMDAR encephalitis were divided into two groups based on MTA and GCA scale. The GCA (+) group had higher percentage of female patients than that in GCA(-) group (p=0.029). No significant difference was found for median age and percentage of patients older than 18 years. As for the clinical presentation at onset, the GCA (+) group had higher percentage of patients with status epilepticus (p<0.001) than GCA (-) group and the MTA (+) group had higher percentage of patients with memory deficits (p < 0.001) than MTA (-) group. No significant difference was found for psychiatric symptoms and conscious disturbance. Higher percentages of severe infections were found in GCA(+) and MTA (+) groups than those in GCA(-) and MTA (-) groups (p=0.01 and p = 0.002, respectively). The percentage of patients needing ventilatory support was higher in MTA (+) group than that in MTA (-) group (p=0.015). The modified Rankin Scale (mRS) scores of GCA(+) and MTA (+) groups were significantly higher when compared with GCA(-) and MTA (-) groups both at day one (p<0.001 and p=0.001, respectively) and at discharge (p<0.001 and p=0.02, respectively).
Patients in GCA (+) group had higher median length of hospital stay than those in GCA (-) group (p<0.001). The median onset-to-MR time of GCA(+) group was higher than that of GCA(-) group (p=0.034).
Percentage of patients with limited response to treatment was higher in GCA (+) group than that in GCA(-) group (p<0.001). As for laboratory test results, the median 25 hydroxyvitamin D3 level was lower in the GCA (+) group than that in GCA(-) group (p=0.001). There was no difference in percentage of patients receiving first line combined with second line treatment, CSF abnormalities, median antibody titers, median vitamin B1 level, median vitamin B2 level, median vitamin B6 level, median vitamin B12 level and mean folic acid level between patients with and without brain atrophy.

Discussion
This study focused on the characteristics of anti-NMDAR encephalitis patients with brain atrophy. Previous studies demonstrated grey matter atrophy in patients with multiple sclerosis and neuromyelitis optical [11].
In multiple sclerosis, deep grey matter has been recognised as a crucial component of the disease and has been associated with disability [12]. In neuromyelitis optical, hippocampal volume is the main MRI predictor of cognition [13]. However, little is known about brain atrophy in anti-NMDAR encephalitis. In this study, more than one third had brain atrophy on plain MRI films measured with MTA scale and GCA scale in total 82 patients with anti-NMDAR encephalitis. This result demonstrated that brain atrophy were not uncommon in patients with anti-NMDAR encephalitis. Medial temporal lobe atrophy as assessed by MTA was observed in 14.1% of patientson at least 1side and this finding is consistent with that in a previous study [14].
We found the GCA (+) group had higher percentage of female patients. This result may due to that all patients with teratoma were female in our study and we found that anti-NMDAR positive female patients with ovarian teratomas were more likely to have GCA than thoes without teratomas.
For clinical symptoms, we found that the MTA (+) group had higher percentage of patients with memory deficits than MTA (-) group.
MTA-scale was usually used to assess degree of hippocampus atrophy on plain MRI films. A previous study demonstrated that the degree of MTA was significantly correlated with scores on memory tests, which was consistent with our finding [9].
We found that the median onset-to-MR time of the GCA (+) group was 60 which was higher than that of the GCA (-) group. A previous study demonstrated that brain atrophy in anti-NMDAR encephalitis started 1 to 2 month after symptom onset, which was consistent with our finding [7].
We also found that the modified Rankin Scale (mRS) scores of MTA (+) groups were significantly higher when compared with MTA (-) groups. A previous literature indicated that left hippocampal volume could predict disease severity [15] which was consistent with that in our study. Also higher mRS scores were observed in patients with GCA.
Patients with GCA were more likely to have limited response to treatment, need ventilatory support and had higher median length of hospital stay.
The findings suggested that MTA and GCA may be associated with severity and prognosis of anti-NMDAR encephalitis. Lizuka et al reported several patients with anti-NMDAR encephalitis developed diffuse cerebral atrophy who had severe dyskinesias and they slowly recovered 3 to 4 years after symptom onset. Also, their brain atrophy were reversible [16]. So further studies are needed to confirm the association between long-term MRI changes and clinical outcome of patients with anti-NMDAR encephalitis.
The mechanisms of brain atrophy in anti-NMDAR encephalitis remain largely unknown. In this study we found that patients with brain atrophy had higher percentages of status epilepticus, severe infection and lower 25 hydroxyvitamin D3 level. According to previous studies, these factors may be associated with brain atrophy [17][18][19].
Previous studies indicated that an ovarian tumor could be found in approximately 26.9% to 44.2% of female patients with anti-NMDAR encephalitis [5,20,21] which was consistent with our finding (35.4%).
We found that anti-NMDAR positive patients with ovarian teratomas were more likely to have GCA than patients without ovarian teratomas.
We also found that patients with teratomas had higher mRS scores which was consistent with the result in a previous literature [21]. Patients with a teratoma develop more robust immune responses than those without a tumour [22] and tend to present more severe neurological conditions [21].
In additon, we found that patients with teratomas had lower level of folic acid. A previous study conducted by Snowdon et al showed that low serum folate was strongly associated with atrophy of the cerebral cortex [23], though the mechanism remains to be determined.

Limitations and conclusions
There are some limitations in this study: (a) because we only enrolled patients with anti-NMDAR encephalitis, our conclusions may not be applied to patients with other types of autoimmune encephalitis; (b) bias is inevitable in retrospective studies; (c) we did not perform follow up of MR changes in this study.
In summary, this study confirmed that brain atrophy was not uncommon in patients with anti-NMDAR encephalitis. MTA and GCA may be associated with severity and prognosis of anti-NMDAR encephalitis. Also we found that anti-NMDAR positive patients with ovarian teratomas were more likely to have GCA than patients without ovarian teratomas. The exact mechanism of brain atrophy in anti-NMDAR encephalitis need further study.

Ethics approval and consent to participate
This research was approved by the ethics committee of the Third Affiliated Hospital of Sun Yat-sen University.

Consent for publication
All participants involved in this study provided written informed consent.

Data availability statements
The datasets analysed during the current study available from the corresponding author on reasonable request.