Clinical features and prognostic factors in adults with viral meningitis

Abstract

Clinical features applicable to the entire spectrum of viral meningitis are limited, and prognostic factors for adverse outcomes are undetermined. This nationwide population-based prospective cohort study included all adults with presumed and microbiologically confirmed viral meningitis in Denmark from 2015 until 2020. Prognostic factors for an unfavourable outcome (Glasgow Outcome Scale score of 1-4) 30 days after discharge were examined by modified Poisson regression. In total, 1066 episodes of viral meningitis were included, yielding a mean annual incidence of 4.7 episodes per 100 000 persons. Pathogens were enteroviruses in 419/1066 (39%), herpes simplex virus type 2 in 171/1066 (16%), varicella-zoster virus in 162/1066 (15%), miscellaneous viruses in 31/1066 (3%) and remained unidentified in 283/1066 (27%). The median age was 33 years (IQR 27-44), and 576/1066 (54%) were females. In herpes simplex virus type 2 meningitis, 131/171 (77%) were females. Immunosuppression [32/162 (20%)] and shingles [90/149 (60%)] were frequent in varicella-zoster virus meningitis. The triad of headache, neck stiffness and hyperacusis or photophobia was present in 264/960 (28%). The median time until lumbar puncture was 3.0 h (IQR 1.3-7.1), and the median CSF leucocyte count was 160 cells/µl (IQR 60-358). The outcome was unfavourable in 216/1055 (20%) 30 days after discharge. Using unidentified pathogen as the reference, the adjusted relative risk of an unfavourable outcome was 1.34 (95% CI 0.95-1.88) for enteroviruses, 1.55 (95% CI 1.00-2.41) for herpes simplex virus type 2, 1.51 (95% CI 0.98-2.33) for varicella-zoster virus and 1.37 (95% CI 0.61-3.05) for miscellaneous viruses. The adjusted relative risk of an unfavourable outcome was 1.34 (95% CI 1.03-1.75) for females. Timing of acyclovir or valacyclovir was not associated with the outcome in meningitis caused by herpes simplex virus type 2 or varicella-zoster virus. In summary, the outcome of viral meningitis was similar among patients with different aetiologies, including those with presumed viral meningitis but without an identified pathogen. Females had an increased risk of an unfavourable outcome. Early antiviral treatment was not associated with an improved outcome in meningitis caused by herpes simplex virus type 2 or varicella-zoster virus.

Keywords: aseptic meningitis; herpesviridae; tick-borne encephalitis; toscana virus; virology.

Figure 1

Aetiologies of viral meningitis in adults.

Figure 2

Annual incidences of viral meningitis in adults.

Figure 3

Stratified analyses of sex-based differences in unfavourable outcome (GOS score 1–4) 30 days after discharge in adults with viral meningitis. ^aStratified analyses were adjusted for age (18–30, 31–50, ≥51 years), sex (male, female), immunosuppression (yes, no), triad of headache, neck stiffness and hyperacusis or photophobia (yes, no), CSF leucocyte count (0–100, 101–500, ≥501 cells/µl), CSF protein (0.0–0.5, 0.6–1.0, ≥1.1 g/l), and treatment with dexamethasone for bacterial meningitis (yes, no). The number of episodes used in adjusted analyses were 251 in unidentified pathogen, 369 in EVs, 140 in HSV-2, 140 in VZV and 922 in all viral meningitis.