New-onset refractory status epilepticus (NORSE) secondary to Bartonella henselae infection in a pediatric patient

Abstract

New-onset refractory status epilepticus (NORSE) is a rare and devastating clinical entity, often without a clearly identified etiology despite extensive testing or known predisposing neurological disorders.¹ We describe a child with NORSE in the setting of active/recent Bartonella henselae infection who responded well to multiple anticonvulsants, immunomodulatory therapy, and antimicrobial therapy.

A normally developing five-year-old boy with attention deficit/hyperactivity disorder and no history of seizures was found unresponsive with generalized body stiffening and facial twitching. The patient had several days of cough and congestion with no fevers prior to his presentation. Seizures persisted despite repeated doses of benzodiazepines and therapeutic doses of levetiracetam, phenobarbital, and fosphenytoin. Electroencephalography (EEG) showed ictal and peri-ictal discharges in the right posterior and left posterior/lateral head regions consistent with status epilepticus, requiring midazolam infusion. Due to continued status despite up-titration of midazolam, pentobarbital was added to achieve burst suppression.

The patient underwent a broad toxic, metabolic, genetic, infectious disease, and autoimmune evaluation. Rhinovirus/enterovirus RNA was detected on the nasopharyngeal swab, and his cerebrospinal fluid (CSF) sample showed 0/cmm (cubic millimeter) white blood cells, 15/cmm and 38/cmm red blood cells in tubes 1 and 4, respectively, glucose 64 mg/dL, protein 41 mg/dL, negative cultures, and negative herpes simplex virus and enterovirus polymerase chain reaction. In addition to levetiracetam, lacosamide was added to his scheduled antiseizure regimen when pentobarbital was being weaned, and high-dose methylprednisolone was tried on hospital day five for a total of five days. Head computed tomography and magnetic resonance imaging with and without contrast were normal. Two doses of intravenous immunoglobulin (IVIG) were administered on hospital days six and seven. CSF cytokine profile demonstrated elevated interleukin-4 (IL-4) (35 pg/mL), IL-6 (1775 pg/mL), IL-8 (10 816 pg/mL), IL-10 (10 pg/mL), and granulocyte–macrophage colony-stimulating factor (GM-CSF) (4 pg/mL). Clobazam was added when midazolam was being weaned, and anakinra, an IL-1 receptor antagonist, was started at 4 mg/kg/day on hospital day seven. EEG at that point showed focal epileptiform discharges in the left occipital region with continuous background and evidence of state change and reactivity.

Due to cat exposure, serum B. henselae titers were checked, using a pre-IVIG serum sample, and revealed a recent infection (IgM < 1:20; IgG 1:8192) with no signs of lymphadenopathy or neuroretinitis. As a result, antimicrobial therapy with doxycycline and rifampin commenced on hospital day eight for a total of 14 days. Computerized tomography scans of chest, abdomen, and pelvis were unrevealing, and his serum/CSF autoimmune encephalitis panels were negative. The patient was extubated, and midazolam was discontinued on hospital day 11. While his neurological exam continued to improve with inpatient therapies, he did have one clinical event concerning for a seizure requiring an increase in maintenance clobazam and a low-dose lorazepam bridge. The CSF metagenomics panel came back positive for human herpesvirus 6 and CSF neopterin level was 60 nmol/L. Anakinra was discontinued and another dose of IVIG was given on hospital day 23. Per parents, the patient was close to baseline on discharge (hospital day 25) with no focal deficits except for decreased spontaneous speech output and volume. Whole-genome sequencing showed that he has a compound heterozygous variant of uncertain significance in MLC1 (modulator of VRAC current 1).

NORSE is associated with a high mortality and can lead to persistent disabilities in survivors, such as poorly controlled epilepsy and cognitive-behavioral disability.² The etiology remains unexplained in about half of the cases, representing the so-called cryptogenic NORSE with Bartonella being a rare cause of non-cryptogenic NORSE.³ B. henselae IgG titers > 1:256 suggest active or recent infection despite low IgM titers.⁴ As demonstrated in prior reports^{5, 6} and based on recent consensus recommendations for the management of NORSE,⁷ our patient highlights the importance of early detailed exposure history and testing to identify an etiology, prompt initiation of immunomodulatory therapies and antiseizure medications, and prompt initiation of targeted treatment when an etiology is identified.

Siefaddeen Sharayah: Conceptualization; data curation; investigation; writing—original draft; writing—review and editing. Maria M. Galardi: Conceptualization; writing—review and editing. Soe Mar: Writing—review and editing.

Soe Mar is a member of the ACNS editorial board. The remaining authors declare no conflicts of interest. [Correction added 22 September 2023, after first online publication: The Conflicts of Interest were revised to include ACNS editorial board membership.]