Share this article on:

News from the AAN Annual Meeting: Aggressive Treatment, Team Approach and Patience Needed for Anti-NMDA Receptor Encephalitis

Goodman, Alice

doi: 10.1097/01.NT.0000417968.93019.70
Back to Top | Article Outline


An analysis of more than 500 cases of anti-NMDA receptor encephalitis, an acute form of encephalitis identified about five years ago, points to the need for earlier, more aggressive treatment.

NEW ORLEANS—It's been five years since anti-NMDA receptor encephalitis, an acute form of encephalitis caused by autoimmune reaction against NR1- and NR2-subunits of the glutamate NMDA (N-methyl-D-aspartate) receptor, was first identified by Josep Dalmau, MD, PhD, a neurologist then at the University of Pennsylvania who is now at the University of Barcelona in Spain. Since 2007, Dr. Dalmau and colleagues have identified and followed more than 500 cases of this potentially devastating autoimmune disorder associated with tumors, memory problems, and neuropsychiatric issues.

The disease is defined by eight major symptoms, which typically occur during the first month: behavioral or cognitive problems, memory deficit, speech disorder, seizures, compromised level of consciousness, movement disorders, central hyperventilation, and autonomic symptoms. More than 50 percent of patients have six to eight major symptoms within the first month.

The characterization of this disorder is evolving, and the findings to date have several important implications, said Maarten Titulaer, MD, PhD, a visiting fellow from the Netherlands now at University of Pennsylvania, who described their analysis of patients at a plenary session here during this year's AAN annual meeting.

“We now know that we have to treat anti-NMDA receptor encephalitis aggressively. If first-line immunotherapy doesn't work within one month, we need to initiate second-line therapy promptly with rituximab and cyclophosphamide. Patients can recover fully or almost to their full pre-illness capacity, but neurologists need to be patient, because this can take a very long time. In some patients, it can take several months before initial improvement is noted, and the process to full recovery can take as long as two years,” Dr. Titulaer said, in an interview with Neurology Today. He noted that a multidisciplinary approach is needed to optimize outcomes.

Back to Top | Article Outline


The analysis focused on 565 patients diagnosed with anti-NMDA receptor encephalitis to date; of these, comprehensive clinical data are available on 550 and treatment information is available on 475 patients, with a median follow-up of 18 months.

Peak age of onset is between the ages of 15 and 35 (median age, 19); 36 percent of the cases were children and 5 percent were age 45 or above. To date, the patients have been predominately females (81 percent) — with 88 percent between the age of 12 and 45 years old.

Thirty-nine percent of the patients had associated tumors and 96 percent of the tumors were teratomas. Tumors were rare in patients under the age of 12, but 54 percent of females over the age of 12 had ovarian teratomas.

Two out of three adults have psychiatric symptoms as the initial presentation, while this is true of only one in three children under the age of 12; 50 percent of these children present with seizure and movement disorders, he said.

“Not many new symptoms develop after the first month,” Dr. Titulaer commented. “Eighty-seven percent of patients were treated in the ICU, reflecting the severity of disease.”

Factors that influence outcome include disease severity and delay in starting treatment. Eighty-nine percent of patients with modified Rankin Score (mRS) 3–4 had a good outcome at two years, as did 73 percent of those treated in the ICU. Among 25 patients (5 percent) who were not treated for any reason, 38 percent died or were in a vegetative state at two years.

“The message is the quicker you treat, the better the outcome,” he emphasized.

After one month of first-line immunotherapy — steroids, plasma exchange, or IViG — 50 percent of patients responded and started improving, 45 percent did not respond (an additional 5 percent were not treated). Of those who improved, 98 percent were able to perform activities of daily living independently and 90 percent returned to school or work. Among non-responders, about half were treated with repeat first-line immunotherapy or no additional therapy and the other half received rituximab, cyclophosphamide, or both, depending on the treating physician's choice. At 24 months, patients who received second-line therapy had much better outcomes, as 75% improved, he said.

“It's clear that second-line treatment helps. At the moment, we advise using both rituximab and cyclophosphamide as second-line therapy, because we don't know which is better. They have different spectrums of activity,” he stated.

Overall, the risk of relapse at two years was 13 percent: 9 percent if the patient had teratoma and 16 percent if no tumor was found. “Most relapses are less severe. About 16 percent had more severe relapse, but these were mainly patients in whom the disease had not been recognized during the first episode,” Dr. Titulaer said.



Aggressiveness of treatment influenced risk of relapse. Patients who received second-line treatment had a 5 percent risk of relapse in the first 2 years, whereas those who only received first-line treatment had a 20 percent risk of relapse.

“We have an evolving picture of this disease. The majority of patients who are treated optimally will recover to a good level but it will take time. For some patients, the improvement can be dramatic and occur in a matter of days, while for others, it can take as long as 2.5 years to recover. We used to think that if a patient with encephalitis is in the ICU for a long time, the chance of recovery is small and he or she should go to a nursing home, but the lesson we've learned with anti-NMDA receptor encephalitis is that a patient can be severely ill in the ICU for many months and still recover,” he stated.

Dr. Titulaer noted that increased awareness of the disease and commercialized diagnostic tests have contributed to the increase in the number of cases identified. However, the incidence also depends on diagnostic accuracy, he said. “The diagnostic test used in all laboratories is highly specific, but if the criteria of Dalmau et al. are not followed, the interpretation of test results can be misleading,” Dr. Titulaer said.

The criteria defined by Dr. Dalmau and colleagues in 2008 in Lancet Neurology include the study of both CSF and serum, Dr. Titulaer added. “Sometimes an additional test is required if only serum is studied. We have identified this problem [of requiring an additional test] in referrals with schizophrenia, mitochondrial disease, or prion disease misdiagnosed with serum NMDA receptor antibodies. If strict criteria are not applied, the data on incidence, demographics, and spectrum of symptoms will appear altered.”

Back to Top | Article Outline


Angela Vincent, MD, an immunologist in the department of clinical neurosciences at Oxford University in the UK, discussed the paper at the plenary session. She said that antigens have been identified in autoimmune neurological disorders of the peripheral systems and more recently, in the CNS. In some of these diseases, the role of antigens is unclear, she noted. “They may be markers, and not necessarily pathogenic.”

“NMDA receptor is a novel and exciting finding associated with ovarian teratoma. This presentation was based on an impressive number of patients, had excellent clinical and outcome data, and treatment guidelines that are extremely useful,” Dr. Vincent said.

Because case ascertainment is easier in the UK, Dr. Vincent and colleagues have a more accurate idea of the incidence of anti-NMDA receptor encephalitis than researchers in the U.S. “We estimate that each year there are 1 to 2 cases of anti-NMDA receptor encephalitis per million per year in the UK. We see a hint of an older population [than Dr. Titulaer's series] and we find tumors less frequently, probably because of wider intake,” she told the audience.

“The potential for this field is expanding,” she continued. “We need to be aware that antibodies can be a red herring and may be secondary to pathology. A positive test does not always exclude a different diagnosis.”

Back to Top | Article Outline


• Goodman A. Clinical symptoms – and test – described for detecting anti-NMDA receptor encephalitis, Neurology Today, May 15, 2008:
    • Dalmau J, Gleichman AJ, et al. Anti-NMDA-receptor encephalitis: Case series and analysis of the effects of antibodies. Lancet Neurol 2008;7(12):1091–1098. E-pub 2008 Oct 11.
      ©2012 American Academy of Neurology