The diagnosis of anti-NMDAR encephalitis is very difficult to make at the point of first medical contact in the ED. To confirm the diagnosis, a positive anti-NMDAR test result from a patient’s CSF is needed, however this may not be available in many medical institutions. Fortunately, anti-NMDAR encephalitis presents with a constellation of characteristic symptoms. In this study, we found fever, abnormal (psychiatric) behaviors, a decreased level of consciousness, and seizures were the most common symptoms, all with an incidence above 50%. Unfortunately, the psychiatric symptoms associated with anti-NMDAR encephalitis can be so significant that many family or emergency physicians refer these patients to psychiatrists for consultation7,8. Our study found rates of misdiagnosis were relatively high with 47 (54.0%) patients experiencing a misdiagnosis. A key hallmark of anti-NMDAR encephalitis is that, unlike in cases of ‘pure’ psychosis, it is often associated with other symptoms such as fever, decreased level of consciousness, or seizures.
An important first step in the ED is to search for causes of fever, identify possible drug overdoses or other metabolic factors which may lead to decreased levels of consciousness and seizures, but anti-NMDA receptor encephalitis should be on the differential diagnosis for patients with a fever and psychiatric symptoms. Since many patients presenting to the ED with such symptoms will require a lumbar puncture for CSF testing as part of routine ED evaluation, we recommend considering testing for anti-NMDAR antibodies if a clinician suspects the diagnosis and if the medical institution can perform the necessary tests. Meanwhile, more ancillary tests are needed to support the diagnosis of anti-NMDAR encephalitis if antibody testing is unavailable.
In our study, we found the sensitivity of CSF oligoclonal band testing (OB) was 78.9%, and an abnormal EEG 71.2%, but serum antibody against NMDAR had a sensitivity of only 66.3%. Compared with EEG and serum antibody against NMDAR, CSF OB testing is more sensitive and more widely available in many EDs. Graus et al. reported CSF OB sensitivity to be > 60% and could be a useful ancillary test for the clinical diagnosis of anti-NMDAR encephalitis9.
Patients who presented to the ED were quite ill. According to the m-RS scores, more than half were ≥ 4, and, of these, > 80% received monitoring. Overall, in our study, more than half the patients needed to be monitored after being assessed by an ED physician, and half of those monitored patients were subsequently intubated and admitted to an ICU. We analyzed the reasons why patients were monitored, and we found that seizures, autonomic dysfunction, central hypoventilation, or a decreased level of consciousness were the top three reasons. As is common for many other serious ED conditions, airway and ventilation abnormalities or risks were of most concern in these patients. On a more positive note, only one out of 87 patients with anti-NMDAR encephalitis died.
Female patients suspected of anti-NMDAR encephalitis need tumor screening and teratoma removal operations. A majority of anti-NMDAR encephalitis patients in our study were female, and 14 had teratomas. In these 14 patients, 10 had an m-RS of 5, and one had a m-RS of 4, so 78.6% had an m-RS score ≥ 4. Patients with teratomas in our study had higher m-RS scores compared to those without teratomas (p = 0.002). Given this association, we reviewed the source database (ED and non-ED patients) and found that, out of 201 female patients, 39 had a teratoma (19.4%). Of those patients with a teratoma, 27 had an m-RS of 5, and patients with a mRS ≥ 4 take up 74.4% of all patients, like the results in this ED study. We strongly suggest that female patients diagnosed with anti-NMDAR encephalitis should receive tumor screening, especially for ovarian tumors such as teratomas. The most common method for this would be via abdominal or pelvic ultrasound, with computer tomography or MRI of the pelvis as possible options as well.
Immunotherapy is the first line treatment for the anti-NMDAR encephalitis: every patient received a combination of steroids and IVIG to treat anti-NMDAR encephalitis. 39 patients received steroid pulse therapy (methylprednisolone 500 ~ 1000mg daily). All steroid pulse therapy plans were decided after consultation with a neurologist. 84 (96.6%) patients received IVIG, which was sometimes ordered by ED physicians before consultation a neurologist.