A 70-year-old man presented with a complete right facial nerve palsy. Past medical history was only remarkable for hypertension and gastroesophageal reflux disease. He was diagnosed with an idiopathic Bell’s palsy, for which he was treated with a course of prednisone and valacyclovir. He had complete resolution of facial weakness within 3 weeks of symptom onset. Four weeks after initial symptom onset, however, he became agitated and began exhibiting bizarre behavior, including inappropriately raising his voice at people and urinating on the side of the road. He then reportedly acutely developed incoherence and bilateral arm weakness, leading to local hospital admission. Head computed tomography (CT) was unremarkable, and his acute symptomology subsided so he was discharged back home. However, his family noted that he was repetitive and continued to have bizarre behavior. Over the next 3 weeks, he had four visits to the emergency room because of aggressive outbursts, such as yelling, pushing people, and hitting his own head. He was described as having pressured speech and was pacing the room. In addition, he also complained of headache. He had remained afebrile throughout these presentations. Unenhanced head magnetic resonance imaging (MRI) and routine electroencephalography (EEG) were unremarkable. For his agitation, he was initially prescribed lorazepam and sertraline and then switched to quetiapine, without significant response.

On his fourth admission, he had severe agitation requiring four-point restraints and treatment with haloperidol. He was assessed by the Neurology service, and aside from his behavioral change his examination including vital signs, orientation, cranial nerves, motor function, reflexes, sensation, and coordination were grossly intact. Formal bedside cognitive testing was not performed. Metabolic workup, including thyroid-stimulating hormone, vitamin B₁₂, liver function, and electrolytes, were normal, except for mildly elevated creatinine at 117 micromoles per liter. Rheumatologic workup only revealed a positive antinuclear antibody (ANA) at a titer of 1:80 with speckled pattern but was otherwise negative for extractable nuclear antigen, antimicrosomal antibodies, and anti-neutrophil cytoplasmic antibody. An enhanced thoracic, abdominal, and pelvic CT scan followed by whole-body positron emission tomography (PET)/CT found no evidence of malignancy. Enhanced head MRI did not reveal any acute abnormality. Repeat routine EEG was again unremarkable, without any epileptiform activity or focal slowing. Lumbar puncture for cerebrospinal fluid (CSF) evaluation showed normal glucose and protein, three white blood cells per microlitre, and negative oligoclonal bands. Extensive infectious workup, including testing for herpes simplex virus (HSV), varicella zoster virus (VZV), West Nile virus, fungi, acid-fast bacilli, and syphilis, was unremarkable except for serum detection of IgM and IgG antibodies against Borrelia burgdorferi by enzyme immunoassay and western blot. No CSF testing for Borrelia burgdorferi infection was performed. In retrospect, he recalled a tick bite 1 to 2 months preceding his facial nerve palsy. The Infectious Disease service was consulted and recommended daily intravenous ceftriaxone two grams for 4 weeks for possible nervous system Lyme disease, although his neuropsychiatric symptoms were considered to be atypical. No dramatic improvement was initially observed following administration of ceftriaxone. Comprehensive neural antibody testing for autoimmune encephalitis panel was performed and was positive for N-methyl-D-aspartate receptor antibody (anti-NMDAR) by fixed cell-based assay in CSF, confirming a diagnosis of anti-NMDAR encephalitis. ^{Reference Graus, Titulaer and Balu1} He was treated with IVIg 2 g/kg over 2 days and then a 5-day course of IV solumedrol 1000 mg. His neuropsychiatric symptoms began to improve within a week of treatment initiation, facilitating discharge from hospital. At 4-month follow-up, he was doing well according to family. Eight months after his original admission to the hospital, he developed agitation and pressured speech prompting clinical re-evaluation. However, brain MRI was again unremarkable, and repeat CSF evaluation including re-testing for anti-NMDAR was negative. On history, significant psychosocial stressors were identified that were felt to be the cause of his worsening. He improved with only low-dose clonazepam without immunotherapy and returned to his baseline within 2 weeks. At 1-year follow-up, the patient continues to do well, with no neuropsychiatric symptoms identified.

We herein present a patient who initially presented with facial palsy and serologic testing indicative of Lyme disease, followed by neuropsychiatric decline and CSF testing that confirmed a diagnosis of anti-NMDAR encephalitis. A number of pathologies have been reported as possible triggers for anti-NMDAR encephalitis, the most established of which are ovarian teratoma and HSV encephalitis. ^{Reference Titulaer, McCracken and Gabilondo2,Reference Armangue, Spatola and Vlagea3} In contrast, anti-NMDAR encephalitis following Lyme disease is exceptionally rare, with only two previous cases reported in the literature. ^{Reference Predkele and Mednieks4,Reference Martinez, Olguin-Ramirez and Camara-Lemarroy5} These two cases, along with our own, are summarized in Table 1. All patients had clinical symptoms indicative of Lyme disease within 4 weeks preceding or at the time of assessment for anti-NMDAR encephalitis. Interestingly, the two previously reported cases had an elevated Borrelia burgdorferi CSF antibody index suggestive of central nervous system Lyme infection. However, all patients had a neuropsychiatric decline that was more typical of anti-NMDAR encephalitis than neuroborreliosis and all lacked initial response to antimicrobial therapy but improved following immunotherapy, supportive of an autoimmune rather than infectious encephalitic process. ^{Reference Graus, Titulaer and Balu1,Reference Titulaer, McCracken and Gabilondo2,Reference Halperin6} It is possible that Borrelia burgdorferi neuro-invasion may trigger anti-NMDAR encephalitis, like has been well described following HSV encephalitis. ^{Reference Armangue, Spatola and Vlagea3} Our findings suggest that CSF anti-NMDAR testing is warranted in patients diagnosed with Lyme disease followed shortly thereafter by subacute neuropsychiatric decline typical of anti-NMDAR encephalitis. In patients with anti-NMDAR encephalitis following Lyme disease, addition of immunotherapy should be considered to treat neuroinflammation possibly triggered by infection.

Table 1: Patients with anti-NMDAR encephalitis triggered by Lyme disease reported in the literature

CBA = cell-based assay; CSF = cerebrospinal fluid; EEG = electroencephalogram; IVIG = intravenous immunoglobulin; IVMP = intravenous methylprednisolone; MRI = magnetic resonance imaging; NMDAR = N-methyl-D-aspartate receptor; WBC = white blood cell.