Objective:

Characterize neurologic complications of varicella zoster virus (VZV) and identify clinical factors associated with VZV vasculopathy.

Background:

VZV is a neurotropic virus with diverse neurologic presentations and challenging diagnosis. Why some patients develop VZV vasculopathy while others develop non-vascular neurologic complications remains unclear. 

Design/Methods:

Cases with ICD-10 codes for VZV meningitis and/or encephalitis (B01.0, B01.1), and/or positive CSF VZV PCR, IgG, or IgM between 2010 and 2024 at an NYC hospital system were identified. Clinical presentation, comorbidities, and CSF parameters are being collected. VZV vasculopathy was defined as TIA, ischemic stroke, or hemorrhagic stroke, plus positive CSF VZV PCR or antibody testing or recent zoster rash. Non-vascular neuro-VZV was defined as other neurologic presentations attributed to VZV (meningitis, encephalitis, Ramsay-Hunt, myelitis, or other) as determined by the treating physician. Multivariate analysis compared parameters between the two groups. 

Results:

292 cases were identified. Of 159 reviewed to date, 54 were excluded because final diagnosis was unrelated to VZV, leaving 105 confirmed cases (43% female, median age 67 [IQR 45-78]). Comorbidities included hypertension (42%), hyperlipidemia (27%), diabetes (21%), atrial fibrillation (11%), HIV (11%), and other immunosuppressive conditions (35%). Sixty-one percent had a zoster rash (12% ophthalmic, 20% single dermatome, 21% multi-dermatome). Treatments received included acyclovir (86%), valacyclovir (26%), and steroids (12%). Twenty-nine (28%) had VZV vasculopathy and 76 (72%) non-vascular neuro-VZV. In the vasculopathy group, 79% had ischemic stroke, 27% hemorrhagic stroke, 7% TIA. Non-vascular neuro-VZV included meningitis (50%), encephalitis (32%), and Ramsay-Hunt (11%). Diabetes was more prevalent in the vasculopathy group (41% vs 13%, p=0.004), as was steroid treatment (41% vs 1%, p<0.001). There were no other significant differences in demographics, comorbidities, rash, or CSF parameters. One year mortality was 13% (21% vasculopathy, 11% non-vascular neuro-VZV). 

Conclusions:

In analysis to date, diabetes was significantly associated with VZV vasculopathy; other clinical features were similar between groups.