Postinfectious Myeloradiculoneuropathy With Cranial Nerve Involvements Associated With Human Herpesvirus 7 Infection FREE

Guillain-Barré syndrome (GBS) has been recognized as a postinfectious autoimmune disorder against the peripheral nervous system, characterized by acute muscle weakness and areflexia.¹ Many GBS cases have antiglycosphingolipid antibodies such as GM1 ganglioside in patients with Campylobacter jejuni infection² and GM2 ganglioside, which shares common epitopes between the infectious agents and peripheral nerves, in patients with cytomegalovirus (CMV) infection.³

Previous studies have shown that one of the most common classes of viral infection that precedes GBS is the family of herpesviruses. Of GBS cases with respiratory insufficiency and cranial nerve involvement, roughly 10% to 13% and 8% to 10% demonstrate serological evidence of recent exposure to CMV and Epstein-Barr virus, respectively.⁴ Another group of the herpesvirus family includes human herpesvirus (HHV) 6 and HHV-7. Primary infections with either HHV-6 or HHV-7 generally occur in children and are characterized by exanthema subitum and febrile illness.^{5 - 6} Human herpesvirus 6 is recognized as an opportunistic pathogen that causes limbic encephalitis in persons infected with human immunodeficiency virus.⁷ Human herpesvirus 7 has recently been described as a cause of encephalitis and myelitis in immunologically competent adults.^{8 - 9}

We report a case of acute myeloradiculoneuropathy mimicking GBS, with genetic evidence documenting the presence of HHV-7 in the cerebrospinal fluid (CSF).

A 26-year-old man was admitted to the hospital with a 2-day history of progressive motor weakness, tingling in the extremities, dysphasia, and nasal voice. He had preceding flu-like symptoms 2 weeks before admission. Initial neurological examination revealed moderate motor weakness in the extremities (score of 3 to 4 of 5 on the Medical Research Council scale), with mild hyperreflexia except for the absence of an Achilles tendon reflex. The plantar response was initially flexor and then temporarily extensor. There was evidence of cranial nerve involvement including the facial, glossopharyngeal, and hypoglossal nerves, and autonomic dysfunctions were manifested as a heart conduction block. Examination results of CSF samples taken at admission were normal, but successive examinations demonstrated an increase in protein (89 mg/dL [normal level <40 mg/ dL]) and IgG (23 mg/dL [normal level <4 mg/dL]) levels that was accompanied by a modest pleocytosis (8 cells/μL [normal level <5 cells/μL]) by day 20. Laboratory evaluation results for evidence of immunological compromise were negative. A nerve conduction study performed on day 2 and day 25 documented a decrease in compound muscle action potential amplitudes with a reduction of the F-wave frequency. Motor nerve conduction velocities and distal latencies were preserved. There were no temporal dispersions or conduction blocks. Sensory nerve conduction study results were normal. Auditory brainstem response as well as magnetic resonance imaging results of the brain and the spinal cord with gadolinium enhancement appeared normal and, therefore, did not support a diagnosis of brainstem encephalitis. We tentatively diagnosed the patient as having acute myeloradiculoneuropathy, and we treated him with a high dosage of intravenous immunoglobulin (400 mg/kg per day) for 5 days. After treatment, complete recovery of cranial nerve dysfunction was noted within a week, and motor weakness recovered gradually, with pronounced hyperreflexia in the extremities without pathological reflexes. Eight months after the onset of neurological symptoms, his muscle strength returned to subnormal levels (score of 4 to 5 of 5 on the Medical Research Council scale).

We performed serological testing and found a significant change in the serum titers of antibody against CMV and Epstein-Barr virus (Table) during the 2-week interval without evidence of a recent C jejuni infection, although DNA of neither virus was detected in the CSF by polymerase chain reaction, suggesting cross-reacting (heterologous) antibody responses to CMV and Epstein-Barr virus. We further investigated HHV-6 and HHV-7 DNA in the CSF on day 1 and day 20 (after treatment) by real-time polymerase chain reaction,⁵ and we found a significant decrease in the amount of HHV-7 DNA (2800 copies/mL to 0 copies/mL), although no HHV-6 or HHV-7 genomes were detected in the serum sample. Fluorescent antibody testing of serum samples on day 1 and day 20 demonstrated an increase in anti–HHV-7 titers from 1:16 to 1:64 (Table).

To evaluate the patient’s serum for the presence of antiganglioside antibodies, mixtures of gangliosides (GM1, GM2, GM3, GD1a, GD1b, GT1b, GQ1b, and asialo GM1) processed by thin-layer chromatography (using a solvent of chloroform, methanol, and 0.02% calcium chloride in a 55:45:10 vol/vol/vol ratio) were blotted onto a polyvinylidene difluoride membrane by an electrothermal blotter (ATTO Co Ltd, Tokyo, Japan). This polyvinylidene difluoride membrane was probed using patient sera taken on day 1 and day 20 (×1000 dilution) in blocking buffer (2% nonfat milk in the wash buffer, which was phosphate-buffered saline containing 0.5% Nonidet P-40 [Nakarai Tesque Inc, Kyoto, Japan]). After treatment with the second antibody, a positive band was sought using an enhanced chemiluminescence reagent (New England Nuclear, Boston, Mass). A band was present in the positive control (anti-GM1–antibody positive), but no band was detected using the patient’s serum samples (Figure).

Human herpesvirus 6 can be silently harbored in the human brain following primary infection.¹¹ Detection of virus DNA in the CSF, however, is considered a reliable diagnostic tool for infections of the central nervous system,¹² although a DNA polymerase chain reaction can occasionally yield false-positive results.¹³ In our case, HHV-7 DNA was the only viral DNA detected in the CSF, and, more importantly, evidence of HHV-7 DNA disappeared following therapy. These changes were accompanied by concomitant changes in anti–HHV-7 serum antibody titers. Judging from this evidence, we speculate that the development of the clinical features of the patient is closely related to the reactivation of HHV-7 in the nervous system. At present, however, we do not know why our patient developed a reactivation of HHV-7 in the nervous system, even though an immunologically competent state was absent as described in other cases of HHV-6.¹⁴ Based on the time sequence of the onset of the symptoms, we also speculate that HHV-7 might result in the production of neuronal autoantibodies, as has been described with antiglycosphingolipid antibody following CMV infection,³ although there was no antiganglioside antibody in the serum of this patient. Primary HHV-7 infection of the spinal cord also remains a distinct possibility.

The clinical presentation of this case exhibited several characteristics of GBS. This was especially true for the nerve conduction study data. However, there were also signs of modest involvement of the spinal cord as evidenced by the transient presence of positive pathological reflexes. Previous analyses on 229 patients with GBS have suggested that all of the patients with CMV or Epstein-Barr virus infection showed demyelinating neuropathy¹⁵ whereas the present patient showed an axonal neuropathy.

This case supports the contention that HHV-7 may be a pathological factor in the development of acute myeloradiculoneuropathy.

Correspondence: Tatsuro Mutoh, MD, PhD, Department of Neurology, Fujita Health University School of Medicine, 1-98 Dengakugakubo, Kutsukake-cho, Toyoake, Aichi 470-1192, Japan (tmutoh@fujita-hu.ac.jp).

Accepted for Publication: February 22, 2005.

Author Contributions:Study concept and design: Mutoh. Acquisition of data: Mihara, Yoshikawa, and Yano. Analysis and interpretation of data: Asano and Yamamoto. Drafting of the manuscript: Mihara and Mutoh. Critical revision of the manuscript for important intellectual content: Yoshikawa, Yano, Asano, and Yamamoto. Obtained funding: Mutoh. Administrative, technical, and material support: Mutoh. Study supervision: Asano.

Funding/Support: This work was supported in part by the grant-in-aid for the Center of Excellence program, High-Tech Research Project, and Scientific Research of Priority Area (functional glycomics) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan, Tokyo (Dr Mutoh).