Abstract

Objectives:

Varicella-zoster virus (VZV) is a significant pathogen of viral central nervous system (CNS) infections. There are many studies on VZV-associated CNS infections, but there is currently a lack of large-sample comparative studies on VZV-associated CNS infections in immunocompetent and immunocompromised herpes zoster (HZ) patients. In this study, we compared the clinical characteristics and prognosis of VZV-associated CNS infections in immunocompetent and immunocompromised HZ patients.

Methods:

This retrospective study, conducted at a tertiary hospital specializing in dermatology in East China, compares clinical characteristics, treatment, and prognosis at discharge in 117 immunocompetent and 49 immunocompromised HZ patients with VZV-associated CNS infection.

Results:

Compared with the immunocompetent group, the incidence of encephalitis in immunocompromised HZ patients with VZV-associated CNS infections was significantly higher (P<0.05), and the prognosis at discharge was worse (P<0.01). Blood white blood cell (WBC), red blood cell (RBC), blood platelet counts (BPC), hemoglobin (Hb), total protein, prealbumin, and albumin contents decreased (all P<0.05), while adenosine deaminase (ADA) levels in cerebrospinal fluid (CSF) and blood C-reactive protein-to-albumin ratio (CAR), systemic immune-inflammation index (SII), C-reactive protein (CRP), and β2 microglobulin levels were higher (all P<0.05). Further analysis revealed that the time between neurological symptoms and antiviral therapy and the occurrence of encephalitis were independent risk factors for poor prognosis at discharge in immunocompetent HZ patients with CNS infections, while the independent risk factors for poor prognosis in the immunocompromised group were age and the time between neurological symptoms and antiviral therapy.

Conclusions:

Immunocompromised HZ patients have a higher incidence of encephalitis and worse prognosis. Regardless of immunocompetent or immunocompromised HZ patients with concurrent CNS infections, the time between neurological symptoms and antiviral therapy is an independent risk factor for prognosis at discharge, highlighting the importance of early intervention and treatment.

Introduction

Herpes zoster(HZ) is a skin infectious disease caused by reactivation of the varicella-zoster virus (VZV), which is latent in the host (1). VZV, also known as human herpes virus 3, has a high affinity for skin and nerve tissue (2). After the first infection, the virus can remain dormant in the ganglia for a long time. Dormant VZV can reactivate and replicate in large quantities. The reactivated VZV can reach the nerve innervation area along the nerve axon, causing shingles symptoms such as erythema, blisters, and neuralgia on one side of the dermatome. Immune dysfunction is a significant risk factor for the occurrence of HZ in the host (3–6).

The majority of HZ patients can be completely cured after standardized treatment, but some patients still develop peripheral nervous system involvement, such as postherpetic neuralgia (PHN), herpes zoster ophthalmopathy, and cranial nerve paralysis. In some patients, VZV can involve the central nervous system (CNS), resulting in complications such as meningitis, encephalitis, myelitis, and cerebrovascular disease (7–9). Among them, meningitis and encephalitis are the main complications of VZV-associated CNS infections (10, 11). Tobias Tyrberg et al. (10) conducted a nationwide Swedish retrospective case-control study, which included 1488 patients with VZV-associated CNS infections. Among them, meningitis was the most frequent (45%), followed by encephalitis (38%). Elderly individuals have an increased risk of developing CNS infections and are more susceptible to encephalitis. HIV, hematological cancer, and treatment with specific immunosuppressants or high-dose glucocorticoids were significant risk factors for VZV-associated CNS infection. Laura Krogh Herlin et al. (12) conducted a retrospective analysis of clinical data from 92 VZV encephalitis patients nationwide in Denmark, which revealed that age and immunosuppression are high-risk factors for VZV encephalitis. In 2025, Ci XJ et al. (13) retrospectively analyzed the clinical characteristics of 108 immunocompetent patients diagnosed with VZV CNS infection and found that both VZV meningitis and encephalitis in immunocompetent individuals typically present with mild clinical symptoms and have a good prognosis.

Although there are increasing studies of CNS infections caused by VZV reactivation in clinical practice, few studies have examined the differences between immunocompromised and immunocompetent HZ patients with CNS infection, although CNS infections are more frequently reported in immunocompromised patients (14–16). Therefore, this study systematically compared the clinical characteristics, treatment, and prognosis of VZV-associated CNS infections in immunocompetent and immunocompromised patients hospitalized in our center from January 2020 to December 2025, in order to provide a reference for clinicians in the diagnosis and treatment of VZV-associated CNS infection.

Materials and methodsMethods

A retrospective comparative study was conducted at the Department of Neurology, Hangzhou Third People’s Hospital, Zhejiang, China, a tertiary hospital specializing in dermatology in East China. Clinical data were collected from the electronic medical records system, including the diagnosis, treatment, and follow-up of patients with VZV-associated meningitis/encephalitis hospitalized in our center from January 2020 to December 2025. The collected data included sex, age, location of shingles, course of disease, comorbidities, clinical symptoms, CSF, cranial CT/MRI, electroencephalogram, the time between shingles and neurological symptoms, the time between neurological symptoms and antiviral therapy, hospitalization days, and prognosis at discharge. This study was approved by the Ethics Committee of Hangzhou Third People’s Hospital (No. 2021KA013).

Patients with VZV-associated meningitis/encephalitis were included according to our previous criteria (17, 18). (The diagnosis of HZ met the criteria of the European consensus-based guideline (19), and the diagnosis of VZV encephalitis met the consensus statement of the International Encephalitis Consortium (20)). In brief, for shingles within 4 weeks, patients with symptoms/signs of CNS infection, CSF white blood cell counts >5×106/L or protein levels > 45mg/dL or VZV DNA (+), and metagenomic next-generation sequencing (mNGS) detection of pathogenic microorganisms in CSF excluding bacterial, fungal, and other microbial infections were included. All included patients had no history of VZV vaccination and did not receive antiviral treatment before admission.

Immunocompromised individuals were defined by previous study (21) as those with impaired immune function. These included solid organ malignancies, hematological malignancies, solid organ transplantation, hematopoietic stem cell transplantation, human immunodeficiency virus (HIV) infection (CD4+< 400×106/L), end-stage renal disease, congenital immune deficiencies, or autoimmune diseases undergoing chemotherapy or glucocorticoid treatment (taking prednisone ≥20mg/day or an equivalent dose of corticosteroids for at least one month). Individuals with shingles were defined as immunocompromised HZ patients, whereas patients without these conditions were defined as immunocompetent HZ patients.

Prognosis was assessed at discharge and categorized according to the Glasgow Outcome Scale (GOS), which ranges from 1 to 5 points. Prognosis at discharge was divided into two categories: Good: GOS ≥ 4 points and Poor: GOS ≤ 3 points. Neurologists in our research team assessed the prognosis of patients.

Statistical analysis

Data processing and statistical analysis were performed using SPSS 26.0 software. The Kolmogorov-Smirnov test was used to assess the normality of each group. Measurement data conforming to a normal distribution were expressed as x ± sd. The t-test was used for comparing the means of two samples. For data with non-normal distribution, the median (interquartile range) (M [P25, P75]) was used for description, and the Mann-Whitney U test was used for comparing between two groups. Counting data were expressed as frequency and percentage, and the chi-square test or Fisher’s exact test was used for group comparisons. Factors with statistical significance in univariate analysis were selected for multivariate binary logistic regression analysis. P<0.05 was considered statistically significant.

ResultsBaseline characteristics of VZV-associated CNS infection in immunocompetent and immunocompromised HZ patients

According to the criteria for distinguishing immune function and the diagnostic criteria of VZV-associated CNS infection, 117 patients were included in the immunocompetent VZV-associated CNS infection group, including 72 males and 45 females, with an average age of 51.7 ± 15.8 years. There were 49 patients in the immunocompromised VZV-associated CNS infection group, including 27 males and 22 females, with an average age of 65.4 ± 11.3 years. The baseline characteristics of both groups are shown in Table 1. Patients tended to be younger (51.7 ± 15.8 vs 65.4 ± 11.3 years; standardized mean differences (SMD)=0.97) in the immunocompetent HZ group, which also had a higher proportion of males (61.5% vs 55.1%; SMD = 0.13) and a lower proportion of hypertension comorbidity (24.8% vs 38.8%; SMD = 0.21). The head was the leading location of shingles of VZV-associated CNS infection in both immunocompetent and immunocompromised HZ patients (72.6% vs 63.3%, respectively; SMD = 0.11). All included patients underwent mNGS testing for pathogenic microorganisms in CSF. The results showed that 69 out of 117 immunocompetent patients (69/117, 59.0%) and 34 out of 49 immunocompromised patients (34/49, 69.4%) had VZV DNA (+) in CSF.

CharacteristicsImmunocompetent (n=117)Immunocompromised (n=49)Absolute SMDAge(x ± sd)51.7 ± 15.865.4 ± 11.30.97Sex.(n,%) Male72(61.5%)27(55.1%)0.13 Female45(38.5%)22(44.9%)Comorbidities(n,%) Hypertension29(24.8%)19(38.8%)0.21 Type 2 diabetes14(12.0%)9(18.4%)0.19 Coronary heart disease4(3.4%)5(10.2%)0.22 Stroke4(3.4%)5(10.2%)0.22Location of shingles(n,%) Head85(72.6%)31(63.3%)0.11 Neck15(12.8%)9(18.4%)0.06 Chest and Back14(12.0%)10(20.4%)0.14 Waist and Abdomen1(0.9%)1(2.0%)0.11 Upper limb2(1.7%)00.16 Lower limbs2(1.7%)1(2.0%)0.03

Baseline characteristics of immunocompetent and immunocompromised HZ patients with VZV-associated CNS infection.

SMD, standardized mean differences.

Clinical manifestations

The most common clinical manifestations among the 117 immunocompetent patients were fever (80.3%), headache (70.9%), meningeal irritation (29.1%), cranial nerve paralysis (13.7%), and dizziness (12.8%). In contrast, the common clinical manifestations in 49 immunocompromised patients were headache, fever, meningeal irritation, cranial nerve paralysis, and consciousness/mental disorders, accounting for 75.5%, 63.3%, 22.4%, 18.4%, and 14.3%, respectively. The proportion of fever in immunocompromised patients was lower than that in immunocompetent patients (P<0.05), while the proportion of consciousness/mental disorders was significantly higher (P<0.01) (Figure 1).

Clinical manifestations of patients in the two groups (%).

Blood and CSF testing in two groups

Compared with the immunocompetent group, immunocompromised patients had decreased levels of white blood cell (WBC), red blood cell (RBC), hemoglobin (Hb), blood platelet count (BPC), total protein, prealbumin, and albumin (all P<0.05), but increased levels of C-reactive protein-to-albumin ratio (CAR), systemic immune-inflammation index (SII), C-reactive protein (CRP), and β2-microglobulin (all P<0.05). CSF adenosine deaminase (ADA) levels were significantly elevated in immunocompromised patients (p<0.05) (Table 2).

CharacteristicsImmunocompetent (n=117)Immunocompromised (n=49)t/zpBlood WBC(×109/L)7.2 ± 2.86.2 ± 2.32.127a0.0350 RBC(×1012/L)4.6 ± 0.64.2 ± 0.63.409a0.0008 Hb (g/L)137.6 ± 22.1128.0 ± 21.22.515a0.0129 BPC(×109/L)210.7 ± 60.6183.0 ± 63.12.577a0.0109 PLR148.3 ± 69.0171.5 ± 101.61.418a0.1581 NLR3.6 ± 2.54.1 ± 2.61.146a0.2535 CAR0.09(0.01, 0.32)0.21(0.03, 0.45)2.829b0.0053 SII648.6 ± 443.4852.2 ± 568.32.175a0.0312 CRP (mg/L)3.0(0.5, 12.1)7.9(1.6, 16.3)2.388b0.0181 Total Protein(g/L)63.5 ± 5.061.4 ± 5.82.293a0.0232 Prealbumin(mg/L)246.5 ± 66.2218.7 ± 62.62.436a0.016 Albumin(g/L)37.0 ± 3.234.7 ± 3.73.844a0.0002 Globulin(g/L)26.7 ± 3.526.7 ± 4.30.0945a0.9249 β2-microglobulin(mg/L)1.7 ± 0.82.7 ± 2.03.614a0.0004 CD4+(×106/L)561.2 ± 334.9554.6 ± 364.90.0941a0.9250 CD8+(×106/L)348.5 ± 207.7350.3 ± 165.30.0449a0.9643 CD4+/CD8+1.8 ± 0.81.8 ± 1.10.2450a0.8086CSF Leukocyte count(×106/L)25.0(4.3, 81.5)10.0(3.0, 65.5)0.1768b0.8599 Protein (mg/dl)73.1 ± 38.868.4 ± 28.20.7738a0.4402 Chlorine (mmol/L)123.6 ± 4.6124.1 ± 5.70.6760a0.5000 LDH (U/L)29.6 ± 14.029.2 ± 12.20.1985a0.8429 Glucose (mmol/L)3.9 ± 1.03.8 ± 1.00.6164a0.5385 ADA (U/L)1.2(0.6,2.0)1.6(1.0,3.5)2.6840b0.0080

Comparison of blood and CSF tests of immunocompromised and immunocompetent HZ patients with VZV-associated CNS infection.

WBC, white blood cell; RBC, red blood cell; Hb, hemoglobin; BPC, blood platelet count; PLR, Platelet-to-Lymphocyte Ratio; NLR, Neutrophil-to-Lymphocyte Ratio; CAR, C-Reactive Protein-to-Albumin Ratio; SII, Systemic Immune-Inflammation Index (SII =(Platelets x Neutrophils)/Lymphocytes). LDH, lactate dehydrogenase; ADA, adenosine deaminase. Note: the column “t/z” refers to the test statistic from the t-test and the z-test, used for group comparison. a: t-value; b: z-value.

Antiviral therapy regimens and prognosis at discharge in two groups

All included patients received intravenous acyclovir (10mg/kg, q8h). Among them, two cases in the immunocompetent group and one case in the immunocompromised group had renal dysfunction; these three patients had prolonged medication intervals (10mg/kg, q12h). The duration of intravenous acyclovir treatment in the immunocompetent group was 7–22 days (14.9 ± 3.6), while in the immunocompromised group it was 3–23 days (14.9 ± 4.6). There were 50 patients in the immunocompetent group who simultaneously received intravenous methylprednisolone (40-80mg/d) for 3–7 days, while 20 patients in the immunocompromised group received methylprednisolone (40-80mg/d). There were no significant differences between the two groups in the duration of antiviral treatment, use of glucocorticoids, hospitalization days, the time between shingles and neurological symptoms, or the time between neurological symptoms and antiviral therapy (all P>0.05) (Table 3).

CharacteristicsImmunocompetent (n=117)Immunocompromised (n=49)t/z/x2pTime between shingles and neurological symptom (days), (M[P25, P75])4.0(1.0, 7.0)4.0(1.0, 7.0)0.1251b0.9006Time between neurological symptoms and antiviral therapy(days), (M[P25, P75])3.0(1.0, 5.0)2.0(1.0, 5.0)0.7030b0.4831Hospitalization days(x ± sd)16.7 ± 4.517.8 ± 6.41.302a0.1946Treatment Days of antiviral treatment(x ± sd)14.9 ± 3.614.9 ± 4.60.0121a0.9904 Using glucocorticoids(n)50200.0521c0.8194VZV-associated CNS infection(n) Meningitis111399.2581c0.0023 Encephalitis610Prognosis at discharged(n) Good104374.8323c0.0279 Poor1312

Antiviral therapy regimens and prognosis at discharge in immunocompromised and immunocompetent HZ patients.

the column “t/z/χ²” refers to the test statistic from the t-test, z-test, or chi-square test, respectively, used for group comparison. a: t-value; b: z-value; c:χ²-value.

Among the 117 immunocompetent patients with VZV-associated CNS infection, there were 111 cases of VZV meningitis and 6 cases of VZV encephalitis. A total of 104 cases had a good prognosis and 13 cases had a poor prognosis at discharge. Among the 49 immunocompromised patients, there were 39 cases of VZV meningitis and 10 cases of VZV encephalitis. A total of 37 cases had a good prognosis and 12 cases had a poor prognosis at discharge. The proportion of VZV encephalitis in immunocompromised patients was significantly higher than in immunocompetent patients (P<0.01), and the prognosis was worse at discharge (P<0.05, Table 3).

Factors of prognosis in 117 immunocompetent HZ patients with VZV-associated CNS infection

Based on the prognosis criteria for patients at discharge, there were 104 patients (104/117, 88.9%) in the good prognosis group, including 62 males and 42 females, with an average age of 51.2 ± 16.0 years. Among them, 101 cases (97.1%) were VZV meningitis and 3 cases (2.9%) were VZV encephalitis. There were 13 patients (13/117, 11.1%) in the poor prognosis group, including 10 males and 3 females, with an average age of 55.8 ± 13.8 years. Among them, 10 cases (76.9%) were VZV meningitis and 3 cases (23.1%) were VZV encephalitis. The incidence of encephalitis was significantly higher in the poor prognosis group (P<0.01) (Table 4).

CharacteristicsGood (n=104)Poor (n=13)t/z/x2pAge(x ± sd)51.2 ± 16.055.8 ± 13.81.007a0.3159Sex.(n) Male62101.4625c0.2265 Female423VZV-associated CNS infection(n) Meningitis101109.6841c0.0019 Encephalitis33 Time between shingles and neurological symptoms(days), (M[P25, P75])4.0(1.3,7.0)3.0(1.0,7.0)2.158b0.0330 Time between neurological symptoms and antiviral therapy(days), (M[P25, P75])2.0(1.0,4.8)5.0(3.0,19.0)3.526b0.0006 Hospitalization days(x ± sd)16.4 ± 4.318.9 ± 5.11.1945a0.0542Treatment Days of antiviral treatment(x ± sd)14.7 ± 3.516.2 ± 3.61.426a0.1567 Using glucocorticoids(n)4370.7378c0.3904Blood WBC(×109/L)7.2 ± 2.86.8 ± 1.90.5761a0.5660 RBC(×1012/L)4.6 ± 0.64.5 ± 0.70.5088a0.6119 Hb(g/L)137.9 ± 22.6135.4 ± 17.20.3628a0.7175 BPC(×109/L)213.3 ± 61.4188.7 ± 50.21.3360a0.1843 PLR169.6 ± 102.8187.5 ± 93.20.5767a0.5653 NLR4.0 ± 2.64.9 ± 2.61.082a0.2816 CAR0.09(0.01,0.32)0.03(0.01,0.64)0.6357b0.5264 SII846.3 ± 579.1902.0 ± 486.20.3195a0.7499 CRP (mg/L)3.0(0.5,11.3)1.4(0.5,26.4)2.358b0.0201 Total Protein(g/L)63.6 ± 4.862.3 ± 6.20.8915a0.3746 Prealbumin(mg/L)243.9 ± 64.6268.3 ± 77.81.207a0.2299 Albumin(g/L)37.1 ± 3.035.3 ± 4.41.6610a0.0996 Globulin(g/L)26.5 ± 3.128.0 ± 5.71.4290a0.1558 β2-microglobulin(mg/L)1.7 ± 0.52.6 ± 2.52.747a0.0074 CD4+(×106/L)587.5 ± 338.2301.3 ± 134.12.508a0.0138 CD8+(×106/L)357.8 ± 211.7256.3 ± 139.71.404a0.1635 CD4+/CD8+1.8 ± 0.91.4 ± 0.71.041a0.1266CSF Leukocyte counts(×106/L)30.0(4.0,85.0)8.0(4.0,26.5)0.9647b0.3367 Protein (mg/dl)74.7 ± 39.760.7 ± 29.61.229a0.2218 Chlorine (mmol/L)123.5 ± 4.7124.1 ± 4.50.4168a0.6776 LDH (U/L)29.9 ± 13.927.3 ± 15.90.6284a0.5310 Glucose (mmol/L)3.9 ± 1.04.1 ± 0.80.7622a0.4475 ADA (U/L)1.1(0.5,2.0)1.2(0.95,3.0)1.472b0.1440

Comparison of the characteristics of different prognosis at discharge in 117 immunocompetent patients with VZV-associated CNS infection.

the column “t/z/χ²” refers to the test statistic from the t-test, z-test, or chi-square test, respectively, used for group comparison. a: t-value; b: z-value; c:χ²-value.

Patients in the poor prognosis group had a shorter time between shingles and neurological symptoms (P<0.05) and a longer time between neurological symptoms and antiviral therapy (P<0.01). However, there were no significant differences in the duration of antiviral treatment, use of glucocorticoids, or hospitalization days between the two groups (P>0.05). In the poor prognosis group, CRP and β2-microglobulin levels were elevated (P<0.05, P<0.01), while blood CD4+ levels were decreased (P<0.05) (Table 4).

Independent risk factors for prognosis at discharge in immunocompeten