Disseminated HSV-2 presenting with relapsing encephalomyelitis

Practical Implications

Consider disseminated herpesvirus in the differential diagnosis of MS. Imaging with peripheral nerve and cauda equina involvement may be helpful in distinguishing infectious etiologies from a primary autoimmune process such as MS, which can further be confirmed with dedicated CSF analysis.

A previously healthy 39-year-old woman presented with 3 weeks of progressive leg numbness. Examination showed mild bilateral iliopsoas weakness and patchy leg numbness. She was admitted, and spinal MRI revealed T2 enhancing thoracic lesions, nonenhancing cervical cord lesions (figure, A), and subtle pial and cauda equina enhancement (figure, B). Brain MRI revealed multiple periventricular T2 hyperintensities (figure, C). Since the diagnosis of multiple sclerosis (MS) was considered likely, lumbar puncture (LP) was deferred and she received 5 days of IV methylprednisolone. One week later, she developed increasing right-sided hearing loss and vertigo with right cochlear enhancement (figure, D). Two weeks later, a morbilliform and vesicular flank and trunk rash appeared (figure, E and F). She was readmitted and treated with IV acyclovir for presumed disseminated varicella zoster virus (VZV). LP revealed a lymphocytic pleocytosis (90 leukocytes, 100% lymphocytes) and elevated protein (150 mg/dL). Oligoclonal bands were positive, and neuromyelitis optica and human T-cell lymphotropic virus–1 antibodies were negative. A PCR encephalitis panel was positive for herpes simplex virus (HSV)–2 and negative for VZV and HSV-1. Skin biopsy viral culture and PCR were positive for HSV. Three weeks after rash onset, the patient developed worsening leg numbness, received 5 more days of IV methylprednisolone, and soon after developed severe ataxia and weakness. Repeat MRI revealed new pontine lesions atypical for MS, punctate lesions following a vascular distribution, and new enhancing spinal cord lesions (figure, G and H). Repeat LP revealed a decreasing leukocyte count (17 leukocytes) and rising protein (298 mg/dL). CSF HSV-2 PCR was now negative, yet quantitative ELISA revealed positive HSV-2 immunoglobulin G (IgG) (8.83 antibody index [AI]) and negative HSV-1 IgG (0.3 [reference range 0.9 AI]). CSF HSV-1 and HSV-2 immunoglobulin M (IgM) (1.15) and IgG titers were elevated (27.93 [reference range 0.9 AI]). A unifying diagnosis of HSV-2 encephalomyelitis was made. The patient was treated with plasmapheresis, followed by IV immunoglobulin (IVIG), concurrently with 6 weeks of IV acyclovir. A follow-up LP 3 weeks after treatment initiation demonstrated decreasing HSV IgM (0.22 AI) and IgG (7.96 AI) titers, pleocytosis (6 leukocytes), and protein (111 mg/dL), consistent with declining inflammation. Monthly follow-up imaging showed interval resolution of spinal enhancement and no new lesions, commensurate with resolving sensory symptoms and ataxia.

Figure. Radiologic and dermatologic findings.

(A) Sagittal short T1 inversion recovery sequence with several T2-hyperintense signal lesions throughout the cervical and thoracic spinal cord. (B) Fluid-attenuated inversion recovery (FLAIR) sequence demonstrates several scattered, subcortical white matter lesions. (C) Axial image of the lumbar spinal cord with gadolinium-enhancing (Gd1) T1-weighted imaging demonstrates enhancement of the cauda equina. (D) Imaging of the internal auditory canal with Gd+ shows enhancement of the cochlea. (E) Photograph of the morbilliform rash over the right flank and (F) a vesicular rash on the left flank. (G) Follow-up FLAIR sequence images after steroid treatment demonstrate atypical T2 lesions following a venous distribution in the subcortical white matter (H) as well as in the cerebellum and pons.

DISCUSSION

Pial and cauda equina involvement, 8th cranial nerve involvement, rash, and worsening after steroids are atypical for MS. HSV-2 typically causes meningitis, but rarely encephalitis or myelitis in immunocompetent adults.¹ MRI can be nonspecific, there may be a predilection for the brainstem, and nerve root enhancement can occur.² This patient had no prior diagnosis of genital herpes. In searching for an underlying immunodeficiency, HIV antibodies and PCR were negative, and CD4 count was normal, but we uncovered low natural killer (NK) cell levels (47 cells/μL, reference range 59–401 cells/μL) at the time of her acute illness that later became normal. NK cells are known for their innate defense against herpesviruses³ and can be suppressed in the setting of steroids.⁴ Transient NK cell deficiency may have enabled dissemination of the virus throughout the nervous system.

We also uncovered an IgG deficiency (351 mg/dL [reference range 768–1632 mg/dL]) that transiently increased 1 month after IVIG (898 mg/dL), and later remained deficient at 2-, 5-, and 6-month follow-up (802 mg/dL, 651 mg/dL, and 702 mg/dL, respectively). IgG-mediated antibody-dependent cellular cytotoxicity and the classical complement pathway are important in the defense against genital herpes, and lower levels of IgG subclasses have been found in particularly severe and recurrent genital HSV infections.⁵ We also uncovered a persistent immunoglobulin A deficiency (26–51 mg/dL [reference range 68–378 mg/dL]) before and after IVIG, which may have created a further vulnerability. IVIG therapy has been associated with a striking reduction in the frequency of recurrences, duration, and severity of genital HSV infections as compared to acyclovir alone, thought to be mediated by expansion of NK cell populations.⁶

Once disseminated into the nervous system, HSV can have a secondary immune-mediated, relapsing course, which has been shown in a recent case series with patients additionally demonstrating NMDA receptor (NMDAR) and other synaptic protein antibodies.⁷ Relapses can occur even after the CSF PCR seroconverts to negative, and they respond well to immunotherapy, as did our case.⁷ While her serum NMDAR antibody was negative, not all patients had positive antibodies.⁷ We therefore hypothesize that the underlying immunodeficiency enabled dissemination of HSV-2 into the nervous system, while the relapsing nature of the disease was perhaps due to a secondary immune-mediated process, mimicking demyelinating disease. This is supported by continued relapses despite PCR seroconversion and immunotherapy treatment coinciding with halting clinical-radiographic disease activity and declining CSF markers of disease activity.

AUTHOR CONTRIBUTIONS

Clotilde Hainline: primary author, writing/editing for content. Dominique Rosales: writing/editing for content. Purvi Parikh: writing/editing for content. Eddie Louie: writing/editing for content. Jonathan Howard: writing/editing for content. Nina Kim: writing/editing for content. Steven L. Galetta: writing/editing for content.

STUDY FUNDING

No targeted study funding reported.

DISCLOSURES

C. Hainline and D. Rosales report no disclosures. P. Parikh serves on speakers' bureaus for and has received funding for travel or speaker honoraria from Meda Pharmaceuticals, Allergy and Asthma Network, Boehringer Ingelheim, and Baxalta; serves as a consultant for Allergy and Asthma Network; and receives research support from Genentech and Astra Zeneca. E. Louie reports no disclosures. J. Howard receives publishing royalties from Neurology Video Textbook DVD (Demos Publishing, 2013). N. Kim holds stock/stock options in Natus. S.L. Galetta has received funding for travel or speaker honoraria from Biogen and Genzyme; serves on the editorial boards of Neurology® and Journal of Neuro-ophthalmology; and serves as a consultant for Genzyme and Biogen. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/cp.