Viral load and pathophysiology-based framework in varicella-zoster virus meningitis: Focus on altered consciousness and outcomes.
Source: PubMed, NCBI / U.S. National Library of Medicine
Although varicella-zoster virus (VZV) meningitis is generally benign, some patients experience severe disability. Previous studies have lacked integration of host, viral, and clinical factors. The present study aimed to identify factors associated with poor outcomes, explore a descriptive, pathophysiology-informed framework and assess the feasibility of a simple, illustrative prognostic approach. We retrospectively analyzed 41 adults with polymerase chain reaction (PCR)-confirmed VZV meningitis from two Japanese centers (2013-2025), including 30 with quantitative PCR data. Baseline variables included age, immunosuppression, altered level of consciousness (LOC), rash, cerebrospinal fluid (CSF) findings, and timing of acyclovir initiation. Outcomes were discharge modified Rankin Scale (mRS) scores. Logistic regression and serial mediation analyses were performed. Predictive performance was evaluated using the area under the curve and internal-external cross-validation (IECV). The median age was 63 years; 22.0% of patients were immunosuppressed, and 29.3% presented with LOC. LOC, age, and CSF viral load were associated with poor outcomes (mRS ≥ 3). Mediation analysis suggested an indirect pattern of association linking immunosuppression, higher viral load, LOC, and disability. A sequential rule-Rules 1 (LOC → mRS ≥ 3), 2 (age ≥ 69 years→mRS ≥ 2), and 3 (VZV ≥ 70
Abstract
Although varicella-zoster virus (VZV) meningitis is generally benign, some patients experience severe disability. Previous studies have lacked integration of host, viral, and clinical factors. The present study aimed to identify factors associated with poor outcomes, explore a descriptive, pathophysiology-informed framework and assess the feasibility of a simple, illustrative prognostic approach. We retrospectively analyzed 41 adults with polymerase chain reaction (PCR)-confirmed VZV meningitis from two Japanese centers (2013-2025), including 30 with quantitative PCR data. Baseline variables included age, immunosuppression, altered level of consciousness (LOC), rash, cerebrospinal fluid (CSF) findings, and timing of acyclovir initiation. Outcomes were discharge modified Rankin Scale (mRS) scores. Logistic regression and serial mediation analyses were performed. Predictive performance was evaluated using the area under the curve and internal-external cross-validation (IECV). The median age was 63 years; 22.0% of patients were immunosuppressed, and 29.3% presented with LOC. LOC, age, and CSF viral load were associated with poor outcomes (mRS ≥ 3). Mediation analysis suggested an indirect pattern of association linking immunosuppression, higher viral load, LOC, and disability. A sequential rule-Rules 1 (LOC → mRS ≥ 3), 2 (age ≥ 69 years→mRS ≥ 2), and 3 (VZV ≥ 7000 copies/mL → mRS ≥ 2)- demonstrated high apparent discrimination and acceptable consistency in IECV. Early acyclovir initiation was associated with better outcomes only in patients without LOC at presentation. Immunosuppression may be associated with viral replication, preceding LOC, strongly associated with short-term functional outcome. This descriptive, hypothesis-generating framework helps contextualize previously inconsistent findings and supports a clinically accessible, pathophysiology-informed approach to summarizing short-term risk in VZV meningitis.