The study was conducted from January 1, 2018, to June 30, 2024. This retrospective descriptive study was based on the records of patients hospitalized for herpes zoster in the Infectious and Tropical Diseases Department of Fann Hospital. The study population included all patients diagnosed with herpes zoster. The diagnosis was clinically confirmed by the presence of the characteristic vesicular rash in a unilateral metameric band, painful or not, with or without associated complications. Patients of all ages and both sexes were enrolled. For HIV-positive patients, confirmation of HIV status was made through serology documented in the medical file. Patients with HZ who did not consent to participation in the study were excluded from enrollment, as were individuals for whom hospital records could not be located. Cases with an uncertain diagnosis of HZ or possible confusion with other vesicular dermatoses were excluded.

The data were retrieved from patients’ medical records and hospital admission records. The survey form was completed in reliance on the information obtained from the patients’ records. Additionally, telephone calls were initiated to patients to complete specific data elements. Accordingly, each patient included in the study was administered an individual survey form. The following items were included in the form:

The data were initially recorded in an Excel spreadsheet and subsequently analyzed using IBM SPSS version 26 software. Quantitative variables were described in terms of their position (mean, median) and dispersion (standard deviation, interquartile range, or extremes) according to their applicability criterion. The qualitative variables were depicted in absolute terms (frequency) and relative terms (percentage). Microsoft Office Excel Professional Plus 2021 for macOS was employed to create graphs and tables.

The study was conducted with the approval of the department head. The participation of patients in this study was entirely voluntary. The data collected in the study were de-identified to maintain the confidentiality of the patient information. No financial compensation was provided for the data.

During the seven-year study period, 18 cases confirmed of herpes zoster were identified among 4471 hospital admissions, corresponding to a frequency of 0.4%. The median age of patients was 58 (IQR: 33.5 - 62.5; range: 18 - 75 years). The most represented age group was 50 - 64 years (n = 7; 38.89%). Women predominated, accounting for 55.56% of cases, yielding a male-to-female ratio of 0.8:1. Married individuals represented two-thirds of the cohort (n = 12; 66.67%). Geographic distribution revealed a strong urban concentration, with 88.89% of patients residing in Dakar region (n = 16), predominantly in suburban areas (n = 11). The socio-demographic characteristics of the patients are summarized in Table 1 .

At admission, one patient was known to be HIV-positive. This individual reported poor adherence to antiretroviral therapy, including an interruption exceeding six months. Two patients (11.11%) had a history of diabetes. The mean delay between symptom onset and medical consultation was 4.11 ± 1.97 days, with 44.45% of patients presenting within three days. The mean interval between the onset of pain and the appearance of rash was 1.83 ± 1.04 days. In most cases (82.35%), the rash developed within 48 hours of pain onset. Pain and rash were reported in all patients (100%). Headache was noted in nine patients (50.0%), while fever was documented in eight (44.44%).

Dermatological examination revealed isolated vesicles as the predominant lesion type (n = 7; 38.89%). Hemifacial distribution was noted in half the patients (9 cases), while hemithorax involvement was observed in four cases (22.22%).

Ophthalmic herpes zoster was the most common form (n = 8; 44.45%). Among these, one patient presented with a marked decrease in visual acuity. Another developed keratitis with severe conjunctival and palpebral oedema, leading to complete eye occlusion and requiring surgical intervention. Table 2 summarizes the clinical profile of the patients in the study. The particularities of HIV-infected patients are shown in Table 3 .

*PHN: Postherpetic Neuralgia.

*PHN: Postherpetic Neuralgia.

All patients were tested for HIV, with 10 positive cases (55.55%). Hematological assessment was available for 14 patients. White blood cell counts had values within normal limits in 85.71%. Elevated C-reactive protein (>6 mg/l) was identified in three-quarters of evaluated patients. Elevated blood glucose, urea, and creatinine levels were documented in three patients (23.08% of those with available data). Complete laboratory findings are summarized in Table 4 .

The mean interval from rash onset to antiviral therapy initiation was 3.67 ± 2.2

days. More than half of patients (55.55%) received treatment within 72 hours. Acyclovir was prescribed to all patients, and one additionally received topical ganciclovir ophthalmic ointment (5.56%). Step-2 analgesics were administered to 12 patients (66.67%), neuroleptics in seven cases (38.89%), and pregabalin in two patients (11.11%). Standardized local wound care with aqueous eosin solution was provided to all patients.

Hospital length of stay averaged 7.22 ± 3.19 days (range: 2-14 days). Post-hospital discharge follow-up revealed complications such as pruritus (44.44%), scarring with keloids (16.67%), and postherpetic neuralgia (PHN) in 3 patients (16.67%). Table 5 detailed the characteristics of patients with PHN. Mortality occurred in two cases (11.11%). Both deceased patients were HIV-positive women with ophthalmic herpes zoster. One case involved suspected varicella-zoster virus (VZV) encephalitis, but the patient died prior to confirmation of the diagnosis.

Herpes zoster is a relatively uncommon condition among hospitalized patients. The frequency observed in our cohort was notably low at approximately 0.4%. These results align with findings from two major dermatology departments in Dakar in 2019 [16] . In sub-Saharan Africa, the lack of an effective disease surveillance system represents a major challenge to accurately estimating herpes zoster incidence. As a result, limited epidemiological data are available. Higher frequencies have been documented in Mali and Cameroon, where prevalence exceeded 3% [17] [18] . These disparities can be attributed to methodological differences. In Cameroon, the series included HIV-positive patients, a population known to have higher herpes zoster prevalence. In Mali, data were collected through dermatology consultations. In contrast, enhanced monitoring protocols in Europe have revealed a more widespread prevalence of the disease burden. For example, Germany reports an annual incidence of over 400,000 cases [19] .

Herpes zoster is not considered a sex-linked disease [20] . The female predominance noted in our series corroborates the findings of several studies [2] [3] [8] [13] . The age-related epidemiological pattern observed in our study strongly supports established literature demonstrating increased herpes zoster incidence with advancing age, particularly among individuals aged 50 years and older [4] [8] [13] [20] . Also, hospitalization rates increase sharply in this age group [2] [3] [21] . Our cohort’s median patient age of 58 is consistent with this trend. The increased susceptibility to herpes zoster with aging is attributed to immunosenescence [5] . Beyond acute disease manifestations, herpes zoster in elderly populations frequently results in a loss of autonomy and reduced social engagement, impacting their quality of life [6] [14] .

Patients with herpes zoster typically seek medical care early, often within the first three days of symptom onset. This behavior results in relatively brief consultation delays. This pattern of early referral has been consistently reported in population-based and clinical studies, particularly in primary care settings, where most patients report promptly after rash or pain onset [22] . In our study, nearly one-third of patients consulted within 48 hours of symptom onset, likely due to the severity of acute pain. Pain constitutes the cardinal clinical manifestation of herpes zoster and was universally reported. However, atypical or painless presentations, although less common, can hinder timely diagnosis and increase the risk of inadequate management [23] [24] . Other clinical features such as fever and headache were present in less than half of the cases, aligning with previous reports from the region [17] [18] . The rash typically appeared rapidly, on average two days after the pain onset, and was predominantly vesicular, in agreement with prior literature [4] [18] .

A striking finding was the predominance of ophthalmic herpes zoster, which accounted for half of our cases, a proportion substantially higher than the 10 to 20% risk commonly reported [2] [15] . This risk increases with advanced age [25] . Ophthalmic involvement presents headache as the predominant symptom, with associated complications including keratitis, uveitis, conjunctivitis, and blindness [2] [15] [25] , as observed in two of our patients. The risk of severe sequelae, including permanent visual impairment, underscores the need for rapid diagnosis, timely referral to an ophthalmologist, and early antiviral therapy. In our setting, systematic ophthalmological evaluation was often insufficient, highlighting a clear need to implement routine specialist assessment for ocular herpes zoster. Additionally, thoracic herpes zoster ranked second in lesion distribution in this series, differing from most studies, where it represents the primary site [16] - [18] .

Investigations frequently aim to detect comorbidities HIV infection, malignancies, or other infectious complications. HIV infection is a well-established risk factor for herpes zoster [8] [9] [20] , and a primary comorbidity particularly prominent in African series [3] [18] . The risk of herpes zoster increases with declining CD4 count [6] [7] [26] . Among our HIV-positive patients, CD4 counts were available for only three individuals (30%), all of whom exhibited levels below 200 cells/mm³. The limited availability of CD4 count data represents a significant limitation in this study. This missing data restricts our ability to accurately correlate the severity of herpes zoster with the degree of immunosuppression, which is a critical factor in understanding the disease’s pathophysiology. In the absence of comprehensive immunological profiling, the association between immune status and clinical outcomes remains insufficiently defined. Future studies should systematically incorporate CD4 monitoring to address this gap. Although highly active antiretroviral therapy (HAART) decreases this risk, it remains elevated compared to the general population [7] . In our series, over half of the patients tested were HIV-positive, consistent with a Nigerian study reporting a prevalence of nearly 70% [3] . These findings confirm that herpes zoster frequently serves as a marker for undiagnosed HIV infection, with an especially high positive predictive value in cases of ophthalmic involvement [3] . However, within our series, ophthalmic herpes zoster was equally distributed between HIV-negative and HIV-positive patients, with four patients in each group. These data indicate that HZ ophthalmicus is not restricted to HIV-infected individuals. Nonetheless, HIV infection predisposes to more severe clinical manifestations, including necrotic and hemorrhagic forms, as well as complications like postherpetic neuralgia [3] [21] [25] . In this study, three cases of postherpetic neuralgia (PHN) were identified, two of them occurring in HIV-negative patients, with no apparent correlation to lesion anatomical site.

Additional comorbid conditions that enhance herpes zoster susceptibility include diabetes mellitus [9] [27] , autoimmune diseases [8] [28] , and hematological malignancies [29] . Two cases of diabetes were identified in this study. Psychological stress, frequently overlooked, is an important trigger [30] [31] , particularly when severe and occurring within the preceding three months [32] . Our study did not assess stress levels. Future clinical evaluations should inquire systematically about recent stressful life events such as examinations, work overload, or bereavement. This would add a valuable dimension to future studies.

Herpes zoster treatment objectives encompass pain alleviation, acceleration of lesion healing, and prevention of postherpetic neuralgia [4] [33] . Thus, effective pain management is paramount, though its efficacy, both acute and chronic, tends to decline with patient age [33] . All our patients received analgesics, with opioids and tricyclic antidepressants reserved for severe and refractory pain. Antiviral therapy, mainly acyclovir, was systematically administered. Acyclovir remains the most commonly used antiviral in low- and middle-income countries due to accessibility [34] , as observed in our cohort, whereas valacyclovir with superior bioavailability is less frequently used because of cost constraints. Antiviral administration within the initial 72-hour window generally reduces acute pain duration and rash severity, accelerates healing, and mitigates complication risk [4] [35] . Furthermore, antiviral therapy reduces viral dissemination and prevents new lesion formation. More than half of our patients received antivirals within this critical time frame. Nevertheless, starting antiviral therapy beyond 72 hours may still be clinically justified, particularly with continued lesion development or complication occurrence [35] [36] .

The average hospitalization duration for uncomplicated HZ is approximately seven days for patients over 50, with longer duration in the presence of postherpetic neuralgia [20] . Our data showed similar hospitalization lengths.

Postherpetic neuralgia (PHN) remains the most common complication, with reported frequency ranging from 5% to over 30%, depending on study design, patient age, and definition criteria [2] . In our study, PHN occurred in three patients, consistent with this range. No correlation between advanced age and postherpetic neuralgia was observed in our series, as cases involved two younger adults aged 18 and 29, and one elderly patient aged 73. This contrasts with most studies reporting higher postherpetic neuralgia incidence in patients aged 50 years and older [4] [21] [36] . Additional risk factors for PHN include ophthalmic herpes zoster [4] , comorbidities such as diabetes or immunosuppression [37] . Postherpetic neuralgia may persist beyond one year in 30% - 50% of patients [2] , representing a significant therapeutic challenge [4] potentially causing physical disability, emotional distress, loss of autonomy, and insomnia [1] . Due to study limitations, the duration of postherpetic pain could not be formally assessed due to absent formalized follow-up protocols.

Herpes zoster–associated mortality remains generally low [2] [38] . In hospital settings, case-fatality rates can reach up to 15% [34] . Our study recorded two deaths among 18 patients (11.1%), both involving HIV-positive women over 60 years with ophthalmic herpes zoster. Advanced age [2] , female sex [34] [38] and comorbidity such as HIV infection [21] have been identified as mortality risk factors in herpes zoster. One patient was suspected of VZV meningoencephalitis but died rapidly before confirmation, while the second died two months after hospitalization; in this latter case, herpes zoster was unlikely to be the direct cause. The complex interplay of multiple risk factors likely contributed to these fatal outcomes.

Long-term follow-up is crucial for managing herpes zoster complications. Our cohort lacked structured follow-up protocols, including defined schedules and geriatric assessments, despite the risk of frailty and loss of autonomy in older patients. This gap underscores the urgent need for geriatric involvement in patient management.

Given herpes zoster risks, routine vaccination is recommended for individuals aged 50 years and older, as well as immunocompromised patients, including those with HIV [39] . Two vaccines have demonstrated strong efficacy in reducing herpes zoster incidence, hospitalizations, ophthalmic involvement, and postherpetic neuralgia [40] [41] , and are considered cost-effective [39] . Consequently, several high-income countries have integrated vaccination into immunization programs for older adults [42] [43] .

In Senegal, herpes zoster vaccination remains unavailable and is therefore not routinely offered to older adults. Further epidemiological studies assessing herpes zoster incidence locally are essential to support national vaccine policy decisions.

This study has several limitations. The relatively small sample size of 18 cases over a six-year period substantially limits statistical power and precludes meaningful subgroup analyses or the identification of significant associations. The retrospective design introduces inherent constraints, including incomplete laboratory data for several parameters, which further limit the robustness of the findings. The single-center nature of the study restricts external validity, as results may reflect institutional practices and patient demographics specific to SMIT Fann rather than broader regional patterns. Despite these limitations, the study provides valuable insights into the epidemiological and clinical profile of herpes zoster in a sub-Saharan context. Future research should focus on prospective cohort designs with larger sample sizes, standardized diagnostic criteria, systematic laboratory assessments, and structured follow-up protocols to strengthen the evidence base and better inform prevention and management strategies.