Epstein-Barr Virus-Associated Encephalopathy Presenting with Nonconvulsive Status Epilepticus in an Immunosuppressive State

Epstein-Barr virus (EBV) infects over 90% of the adult population worldwide. EBV infection usually occurs early in childhood and produces no symptoms. By contrast, delayed primary infection during adolescence can lead to infectious mononucleosis [1]. EBV primarily targets and remains latent in memory B cells. Reactivation of latent EBV is often found in immunosuppressive states, and is clinically important because it causes lymphoproliferative disease and central nervous system (CNS) complications [2]. CNS symptoms, including encephalitis, meningitis, cerebellitis, polyradiculomyelitis, transverse myelitis, cranial and peripheral neuropathies, and psychiatric abnormalities, can occur in 1–18% of patients with EBV infection [2]. EBV-associated encephalopathy/encephalitis often presents with nonspecific symptoms and its diagnosis is based on the detection of EBV antibodies and/or DNA in the cerebrospinal fluid (CSF). The clinical features and pathophysiology of EBV infection in the CNS in adulthood remain unknown.

Herein, we describe animmunosuppressed adult male with EBV infection in the CNS, presenting with nonconvulsive status epilepticus (NCSE), revealed by arterial spin labeling magnetic resonance imaging (ASL-MRI) and single-photon emission computed tomography (SPECT).

A 63-year-old man was admitted to our hospital due to gradually impaired consciousness. He had been diagnosed with acute myeloid leukemia at 59 years of age. Remission of leukemia was achieved after allogeneic hematopoietic stem cell transplantation, and consequently hewastreated with prednisolone (2 mg/day) and tacrolimus (0.2 mg/day). About 1 week before admission to our hospital, he presented with a slight fever and malaise. He became progressively unable to walk and talk. Subsequently, his consciousness level was impaired to 11 points (E3, V3, M5) on the Glasgow Coma Scale (GCS), and he was admitted to our hospital. One day after admission, his consciousness level further decreased to 9 points (E2, V2, M5) on the GCS. His vital signs (i.e., blood pressure, pulse, and body temperature) were normal. There were no physical findings of neck stiffness, lymphadenopathy, rash, hepatosplenomegaly, or jaundice. The impaired consciousness was not accompanied by paralysis or convulsions. Blood tests for factors that might cause impaired consciousness, such as inflammation, electrolyte imbalance, hypo/hyperglycemia, and uremia, were normal (Table 1). Fluid-attenuated inversion recovery (FLAIR) and diffusion-weighted imaging (DWI) of brain MRI revealed no abnormalities (Fig. 1a–d). Magnetic resonance angiography also revealed no abnormalities during hospitalization (Fig. 1e, f). However, ASL-MRI and brain perfusion SPECT showed cerebral hyperperfusion in the right temporal region (Fig. 2a, b). Apart from generalized slower theta or delta frequencies, wakeful electroencephalography revealed no abnormal findings that would indicate epilepsy. Examination of CSF showed slight increases in white blood cell count (14 cells/µL), protein (73.6 mg/dL), and glucose (62.0 mg/dL). CSF cytology found no malignant cells. EBV virus capsid antigen (VCA) immunoglobulin (Ig)M was negative in CSF, yet EBV VCA IgG was positive. Quantitative PCR revealed that EBV DNA in CSF was increased to 720 copies/mL, suggesting EBV reactivation in the CNS. In contrast, DNA of herpes virus type 1 or cytomegalovirus was not detected by PCR.

Based on the findings of MRI and EBV DNA in CSF, we diagnosed him with EBV-associated encephalopathy with mild inflammation. Administration of an antiepileptic, levetiracetam (1,000 mg/day) promptly improved his consciousness to 15 points (E4, V5, M6) on the GCS. He regained the ability to stand and walk without any support on the 14th hospital day. SPECT showed disappearance of hyperperfusion in the temporal region on the 17th day (Fig. 2c). After discharge, the dosage of tacrolimus was decreased from 0.2–0.1 mg/day.

We herein report a late middle-aged man who developed NCSE with focal hyperperfusion. Because he was in an immunosuppressive state and we detected significant amounts of EBV DNA and elevated EBV VCA IgG in CSF [2], we speculate that immunosuppression-associated reactivation of EBV may have caused encephalopathy leading to NCSE with mild inflammation. To the best of our knowledge, this is the first reported adult case of EBV-associated encephalopathy presenting with NCSE.

An obvious abnormal finding related to impaired consciousness in the present case was focal hyperperfusion detected by ASL-MRI and SPECT, and which is a characteristic of focal epilepsy [3-6]. In addition, the fact that the administration of levetiracetam promptly improved both consciousness disturbance and focal hyperperfusion supported our diagnosis of NCSE in this case. Even so, repeated electroencephalography did not identify any obvious epileptic spikes.

The symptoms and clinical features of EBV infection in the CNS are heterogeneous and may be dependent upon the degree of cellular dysfunction or destruction, i.e., from encephalopathy to encephalitis. There are two reported pediatric cases with NCSE due to EBV encephalitis [7, 8]. Both cases exhibited hyperintensity lesions on T2-weighted images, while EBV VCA IgM was positive, indicating primary infection of EBV. Abnormal brain MRI findings suggest that primary infection of EBV during pediatric periods may cause destructive neural damage accompanied by severe inflammation. In contrast, Park et al. [9] reported a case of a 30-year-old woman with EBV cerebellitis, who had no prior history of illness. This case showed hyperperfusion in both cerebellar hemispheres on SPECT without any abnormalities on MRI. Moreover, she recovered fully without any neurological sequelae, similar to our case. In our present case, reactivation of EBV may have caused neural dysfunction leading to NCSE with mild inflammation.

Usefulness of an antiviral or immunomodulatory treatment for EBV infection has not yet been established. The above two cases were treated with acyclovir [7, 8] and corticosteroid [8]. Meanwhile, both the case of Park et al. [9] and our case showed hyperperfusion without any morphological abnormalities and improved with only supportive therapies. The degree of inflammation may provide a good indication on how to treat these patients.

While EBV typically infects and remains in memory B cells in the latent phase [10] (in which EBV remains “hidden” in healthy hosts), EBV DNA is often detected in CSF in patients under immunosuppressive conditions [2]. In our present case, continuous treatment with prednisolone and tacrolimus after stem cell transplantation may have caused latent infection and reactivation of EBV in the CNS. EBV can enter B cells by binding to gp350 and complement receptor 2 (CR2), which are highly expressed on the plasma membranes of B cells [10]. Because CR2 is also highly expressed in astrocytes in the CNS [1, 11], EBV can also enter astrocytes with high affinity and remain latent [12], ultimately causing astrocytic dysfunction. It is also suggested that some astrocytoma cases are associated with EBV infection [13]. A possible pathophysiological mechanism of the present case may be as follows: initially, EBV was infected and reactivated in astrocytes in an immunosuppressive state. Next, the dysfunction of astrocytes may have caused neuronal hyperexcitability leading to epilepsy with cerebral hyperperfusion [6, 14, 15].

In conclusion, we report an interesting case with immunosuppression-associated reactivation of EBV in the CNS that may possibly cause NCSE accompanied with focal hyperperfusion and mild inflammation. This case suggests a new concept of EBV-associated encephalopathy leading to epilepsy, particularly in immunosuppressive patients and including aged people. Further investigations to uncover the pathophysiology of EBV-associated encephalopathy may be needed in future.

We thank Rachel James, PhD, from the Edanz Group (www. edanzediting.com) for English editing.

There are no ethical conflicts to declare. The patient was informed of the purpose of the case presentation, and provided written informed consent for the publication, including images, case history, and clinical data.

The authors have no conflicts of interest to declare.

The authors have no funding sources to declare.

All authors read and approved the manuscript before submission. Yuichiro Ohya and Kuniyuki Nakamura reviewed the clinical data, performed the literature search, and drafted the manuscript. Yoshinobu Wakisaka, Hiroaki Sato, Kayo Wakisaka, Masaya Kumamoto, Yohei Muraya, Junya Kuroda, Hiroshi Nakane, Goichi Yoshimoto, Takanari Kitazono, and Tetsuro Ago contributed to the diagnosis and revised the manuscript.