Use of Ketamine Needs Caution in NMDA-R Encephalitis Related Status Epilepticus

Abstract

Anti-N-methyl-D-aspartate receptor (NMDAR) encephalitis is characterised by the antibodies decreasing the NMDAR surface density and synaptic localization by selective antibody-mediated capping and internalization of surface NMDAR. Ketamine is a NMDAR antagonist which can produce dose-dependent, self-limiting side effects such as hypersalivation, hyperreflexia, transient clonus, dizziness, nausea, vomiting, tachycardia, hypertension, and detrusor muscle overactivity. A 22-year-old girl presented with recent onset behavioral change, progressive movement disorder and later lapsed into super refractory status epilepticus (SRSE). She was treated with at least four antiseizure medications, immunomodulation as well as anesthetic drugs with only partial relief in status. Ketamine was added for SRSE and led to a significant worsening of clinical symptoms with abatement after stopping. She finally responded to thiopentone infusion. Ketamine has been found to be beneficial in SRSE in encephalitis including anti-NMDAR encephalitis. This is seemingly counterintuitive given how the action of the drug mimics the pathophysiology of the disease. This report highlights the risk of paroxysmal sympathetic hyperactivity and aggravation of clinical and electrographic features of anti-NMDAR encephalitis with ketamine. Hence, ketamine and similar medications acting on the NMDAR should be used with caution in anti-NMDAR encephalitis.

Key words: Status epilepticus, Intravenous anaesthetics, Electroencephalography, Nonconvulsive, Seizures

Introduction

Autoimmune encephalitis (AIE) presents with highly diverse clinical features, depending on the specific neuronal antigens and targeted brain regions. These features include behavioural disturbances, cognitive dysfunction, involuntary movements, intractable seizures, sleep disturbances, autonomic instability, and a decreased level of consciousness. Anti-N-methyl-D-aspartate receptor (NMDAR) encephalitis, first reported by Dalmau et al.1 is an AIE in which seizures (both clinical and electrographic seizure) are common (76%). These seizures are either localization-related or secondarily generalized, resistant to anti-seizure medications (ASM), and rapidly progress to convulsive status epilepticus and non-convulsive status epilepticus in 25% of cases, refractory status epilepticus (RSE) in 13.8%, and super refractory status epilepticus (SRSE) in 10.2%. A case of SRSE with AIE is described here, focussing on some of the inherent challenges in managing this condition, with special reference to ketamine.

Case Report

A 22-year-old girl presented with recent disturbed sleep-wake cycle, disorganized thought and altered psychomotor activity 2 weeks after she noticed transient galactorrhea. She had become overtly anxious with increased talkativeness, agitation, grandiose delusions, inappropriate cry and laughter, and increased religiosity. She received antipsychotic and antidepressant medications (olanzapine 5 mg and escitalopram 10 mg per day), which she stopped abruptly after 4–5 days. A week later, she lapsed into altered sensorium with tonic posturing and immobility suggesting catatonia. She underwent treatment for catatonia with lorazepam (up to 16 mg per day) and electroconvulsive therapy with partial relief. Magnetic resonance imaging of brain was unremarkable (Fig. 1A). Two weeks later, she developed abnormal tonic posturing of limbs, dyskinetic tongue, and orofacial movements and was transferred to the Neurology Department. The patient was in altered sensorium and extremely restless. All extremities were hypotonic without any obvious paucity of power, normal deep tendon jerks and plantar response.

Hematological and biochemical investigations (hemogram, kidney, liver, and thyroid function tests, and electrolytes) were unremarkable except low vitamin D level (19.6 IU/mL). Cerebrospinal fluid (CSF) analysis showed normal protein (21 mg/dL), sugar (71 mg/dL), and eight lymphocytes per mL. Tests for connective tissue disorders (extractable nuclear antigen profile, angiotensin-converting enzyme levels, and rheumatoid factor), vasculitis (antineutrophil nuclear antibodies), and anti-thyroid peroxidase antibodies were normal. Serum prolactin was 1,741 mIU/L (normal range, 132–498 mIU/L), though no active galactorrhea or abnormality on pelvic ultrasound was observed. Computed tomography thorax and abdomen were unremarkable.

Electroencephalography (EEG) showed a background of anterior dominant delta activity with intermittent epileptiform discharges independently over the left anterior temporal and right mid temporal regions (Fig. 1B). Considering the possibility of encephalitis, antibiotics and ASMs (lacosamide 200 mg per day followed by oxcarbazepine 900 mg per day) were started. An indirect immunofluorescence assay on transfected cell lines for qualitative determination of human antibodies in CSF was strongly positive for anti-NMDA antibodies, and intravenous methylprednisolone (1 g) was given for 5 days. Fluorodeoxyglucose positron emission tomography-computed tomography of brain demonstrated hypermetabolism involving fronto-temporal regions and severe hypometabolism in parieto-occipital regions, depicting antero-posterior glucose metabolic gradient consistent with AIE (Fig. 1C, D).

Despite plasma exchange, her sensorium progressively worsened, with EEG showing generalized rhythmic delta activity (RDA) predominating over the anterior head regions and possible extreme delta brush (EDB) pattern (Fig. 1E). She was treated for non-convulsive status epilepticus with a midazolam infusion, titrated up to 0.8 mg/kg/hour, and escalated to four ASMs (lacosamide 400 mg, oxcarbazepine 1,200 mg, phenobarbital 180 mg, and perampanel 8 mg per day) after elective intubation and ventilation under continuous electroencephalographic monitoring (CEEG). Although she transiently achieved burst suppression, electrographic seizures occurred subsequently with episodes of tachycardia and epileptic hippus suggesting SRSE. Cycling of anaesthetic drugs was done and ketamine was added with withdrawal of midazolam.

Within 24 hours of initiation and gradual escalation of ketamine from 5 to maximum 8 mg/kg/hour, she started having continuous choreiform and high amplitude dyskinetic limb movements with orofacial dyskinesia, tonic seizures, tachycardia and hyperhidrosis (Fig. 2, Supplementary Video 1). CEEG showed increased frequency of RDA and definite EDB with accentuated brush pattern compared to all previous EEGs (Fig. 1F). In view of exacerbation of clinical and electrographic status and ketamine induced hyperkinetic disorder, ketamine was tapered off, and thiopental sodium infusion started instead (Fig. 2). There was a clear temporal relation between stopping ketamine and reduction of the hyperkinetic movements and dysautonomia.

She subsequently responded to thiopental infusion with significant reduction of clinical and electrographic seizures, followed by tapering and cycling with midazolam. The EEG background activity improved from delta to theta range with occasional discharges and intermittent persistent orofacial and limb dyskinesias. After completing seven plasma exchanges (3 per week on alternate days) in the acute phase, intravenous immunoglobulin (0.4 mg/kg/day for 5 days) was administered as a monthly pulse and 3 months after symptom onset, rituximab was administered (induction and 6 monthly maintenance). The patient showed improvement in terms of longer seizure free intervals, though dyskinesias with perioral involvement persisted, albeit with decreased frequency.

Discussion

The highlight of our case is that it brings to the fore the putative role of ketamine in anti-NMDAR encephalitis. NMDAR, the most abundant excitatory receptor in central nervous system, is the target in anti-NMDAR encephalitis, where antibodies localise to its extracellular NR1 and NR2 subunits, leading to internalization and decrease in surface receptor density proportional to the antibody titre. Ketamine, or 2-(o-chlorophenyl)-2-methyl-amino cyclohexane hydrochloride, is primarily a non-competitive inhibitor of NMDAR, leading to decrease in duration and frequency of channel opening and preventing glutaminergic transmission. Additionally ketamine gets trapped within the closed channel and continues channel blockade.

The role of ketamine in RSE suggests that it antagonizes glutamate excitotoxicity, does not increase intracranial pressure, maintains cerebral perfusion pressure, and does not cause cardiac or respiratory suppression. In fact, the literature, including a meta-analysis of 238 RSE patients, found ketamine a safe and efficacious agent (seizure cessation up to 70%) in a dose of 0.07–15 mg/kg/hour.1 In anti-NMDAR encephalitis, ketamine was found beneficial for dyskinesias and RSE. MacMahon et al.2 described a dramatic improvement in severe orofacial and limb dyskinesia after ketamine infusion following initial refractoriness to multiple immunomodulation. Santoro et al.3 observed clinical and/or electrographic seizure cessation within 48 hours after ketamine administration in three patients with anti-NMDAR encephalitis.

Contrarily, the present case highlights the need for cautious use of ketamine in anti-NMDAR encephalitis, as it may worsen the condition. Ketamine may produce symptoms akin to anti-NMDAR encephalitis, such as psychosis, dystonia, autonomic instability, arrhythmias and hypertension, similar to the dose-related escalation of tachycardia, hypermotor activity and intensity of EDB observed by us during ketamine infusion (Fig. 2). Ketamine may induce a similar EDB pattern, as evidenced by Chauhan et al.4 in a case of illicit ketamine use, where the patient presented with stereotyped limb and oral-facial movements, with EEG showing global dysfunction with EDB. This suggests that EDB may be specific to NMDA receptor blockade rather than autoimmune encephalitis alone.

As more data accumulates, anti-NMDAR encephalitis continues to present with many diverse interesting challenges. The pharmacology of ketamine is more complex than the direct antagonism of NMDAR. Hence, drugs such as ketamine may behave unpredictably at the NMDAR, and its use in management of RSE and SRSE in anti-NMDAR encephalitis deems caution.