Introduction

According to the World Health Organization, obesity is defined as having a body mass index (BMI) of more than 30 kg/m2. In Taiwan, the Ministry of Health and Welfare has established the following guidelines for obesity based on body mass index (BMI). A BMI of 27 kg/m2 or higher is considered the threshold for obesity, while a BMI exceeding 40 kg/m2 is classified as morbid obesity. These criteria help assess individuals' weight status and provide a reference for healthcare professionals in determining appropriate interventions or treatments.

In 1975, the global prevalence of obesity was less than 1%; however, it increased to 6%–8% in 20161. Between 2017 and 2018, the prevalence of obesity and severe obesity in adults was 42.4% and 9.2%, respectively, in the United States. In addition, adults aged 40–59 years had the highest prevalence of severe obesity2. In southern China, the prevalence of overweight and obesity was 25.8% and 7.9%, respectively3. A meta-analysis of 288 studies covering 13.2 million individuals found that the global prevalence of central obesity was 41.5%. The prevalence of central obesity was higher among older adults, women, urban dwellers, Caucasians, and individuals from high-income countries. Moreover, the highest prevalence of central obesity was 55.1% and 52.9% in Sothern America and Central American, respectively4. Because obesity is a risk factor for chronic diseases, Keramat et al. suggested that strategies for weight management should be incorporated into health promotion programs to prevent chronic diseases5.

Herpes zoster (HZ) is caused by the reactivation of varicella-zoster virus (VZV) and presents as painful vesicles with a dermatomal distribution. A systematic review of 69 studies found that the incidence of HZ was between 5.23 and 10.9 per 1000 person-years worldwide. In addition, the authors reported that women (6.05–12.8 per 1000 person-years) had a higher incidence than men (4.3–8.5 per 1000 person-years)6. Immunocompromised patients have a higher risk of HZ. Schröder et al. found that the incidence of HZ in immunocompromised and immunocompetent patients was 11.5 and 5.9 per 1000 person-years, respectively, indicating that HZ incidence is nearly double in immunocompromised patients compared with immunocompetent patients7.

Mehta et al. reported that only 1 of 112 saliva samples from eight astronauts was positive for VZV DNA before space flight. However, 61 of 200 samples were positive for VZV DNA during and after spaceflight. These findings indicate that VZV can be reactivated by stress8. In our previous study, we found that stress from disease was associated with HZ development. Musculoskeletal chronic pain9,10,11,12,13,14, which causes stress to affected individuals, has been associated with HZ development. In addition, endocrine disease, namely polycystic ovary syndrome (PCOS)15, also increases risk of HZ occurrence. To date, whether obesity is a risk factor for HZ reactivation remains unknown. This study investigated the association between obesity and HZ occurrence.

Methods

Data source

The National Health Insurance Research Database (NHIRD) contains health insurance claims data for 99% of Taiwanese residents. This study used a sample data from the NHIRD covering a population of 2 million in 2000. Disease diagnoses were coded according to the International Classification of Diseases, 9th Revision and 10th Revision, Clinical Modification (ICD-9-CM and ICD-10-CM).

Study population, outcome and comorbidities

This cohort study observed patients aged 20–100 years with obesity from 2000 to 2017 (tracked to 2018). Obesity (ICD-9-CM: 278.0; ICD-10-CM: E66) was indicated by the presence of two or more outpatient diagnoses or at least one admission record. To assess the impact of obesity severity, we categorized obesity into two groups: non-morbid obesity (ICD-9-CM: 278.00; ICD-10-CM: E66.09, E66.1, E66.8, E66.9) and morbid obesity (ICD-9-CM: 278.01; ICD-10-CM: E66.01, E66.2). The index date was defined as the date when obesity was diagnosed, and patients with HZ (ICD-9-CM: 053; ICD-10-CM: B02) before the index date were excluded. The obesity and control cohorts underwent 1:1 propensity score matching based on age, sex, comorbidities, and index year. The covariates were comorbidities, such as diabetes mellitus (DM) (ICD-9-CM: 250; ICD-10-CM: E08–E13), chronic kidney disease (CKD) (ICD-9-CM: 585; ICD-10-CM: N18.4–N18.9), coronary artery disease (CAD) (ICD-9-CM: 410–414; ICD-10-CM: I20–I25) and cancer (ICD-9-CM: 140–208; ICD-10-CM: C).

Statistical analysis

In this analysis, the frequencies and percentages of age, sex, and comorbidities as variables are presented. To compare the differences between categorical and continuous variables in the two groups, a chi-square test and t-test were employed, respectively. The variables of age and follow-up period are presented as mean and ± standard deviation (SD). To determine the incidence rate (IR) and hazard ratios (HR) of HZ, we utilized multivariable Cox proportional hazards models that were adjusted for age, sex, and comorbidities. Furthermore, we applied the Schoenfeld Individual Test to assess the proportional hazards assumption, similar to how the Shapiro–Wilk test evaluates normality. If the p-value is less than 0.05, it indicates a violation of the assumption. In our analysis, the Schoenfeld test yielded a p-value of 0.07, which is greater than 0.05, confirming that there is no violation of the proportional hazards assumption. We used the Kaplan–Meier method to derive the cumulative incidence curve of obesity and fitted the curve using the log-rank test. A two-sided p-value less than 0.05 was considered significant. All statistical analyses were performed using SAS, version 9.4 (SAS Institute Inc., Cary, NC). A graph of the cumulative incidence of obesity was obtained using R Studio.

Ethics approval and consent to participate

The study was conducted according to the Declaration of Helsinki and it was approved by the institutional review board of Ditmanson Medical Foundation, Chia-Yi Christian Hospital, Chia-Yi, Taiwan (approval number: IRB2022115, date of approval: December 19, 2022). Because the NHIRD was used for this study and the patients’ information has been de-identified, the informed consent was waived by the Institutional Review Board of Ditmanson Medical Foundation, Chia-Yi Christian Hospital, Chia-Yi, Taiwan.

Results

A flow chart of the patient’s selection from the NHIRD shown in Fig. 1. After propensity score matching, there were 18,855 patients in both the obesity and control cohorts. No significant differences in sex, age, and comorbidities were found between the two cohorts. Among the patients, the obesity and control cohorts had more than half women (58.35% and 58.37%, respectively). The mean age in the obesity and control cohorts were 43.31 (± 14.01) and 43.35 (± 14.06) years, respectively (Table 1). The median follow-up times in years for our cohorts were 7.47 for the control group and 7.76 for the obesity group. In the groups classified as non-morbid obesity and morbid obesity, the median follow-up times were 8.19 and 5.85 years, respectively.

Figure 1
figure 1

The flow chart of study sample selection from NHIRD Taiwan.

Full size image
Table 1 Baseline characteristics for individuals with and without obesity.
Full size table

Table 2 presents events, person-year, IR, and HR for obesity, sex, age, and comorbidities in patients with HZ. The obesity cohort [adjusted HR (aHR) 1.09, 95% CI 1.00–1.19] had a significantly higher risk of HZ than the control cohort. Further analysis, the morbid obesity group (aHR 1.47, 95% CI 1.21–.77) had a significantly higher risk of HZ than the non-morbid obesity group. The age groups of 30–39 (aHR 1.35, 95% CI 1.12–1.62), 40–49 (aHR 1.87, 95% CI 1.57–2.23) and ≥ 50 (aHR 3.43, 95% CI 2.91–4.05) had a significantly higher risk of HZ than the age group of 20–29. The risk of HZ in patients with comorbidities was significantly higher than in those without comorbidities. The comorbidities included DM (aHR 1.18; 95% CI 1.07–1.30), CAD (aHR 1.31; 95% CI 1.17–1.46) and cancer (aHR 1.62; 95% CI 1.26–2.08). Figure 2 presents the cumulative incidence of HZ, it shows that the cumulative incidence of HZ in the obesity cohort was higher than that in the control cohort (p = 0.04).

Table 2 Incidences and hazard ratios of herpes zoster for individuals with and without obesity by age, gender and comorbidities.
Full size table
Figure 2
figure 2

Kaplan–Meier curves of the cumulative incidence rate of herpes zoster during the follow-up period between obesity and control groups.

Full size image

Table 3 shows the results of further stratification of obesity. In the obesity cohort, the risk of HZ in women (IR 8.10 vs. 7.04; aHR 1.12; 95% CI 1.00–1.24), those aged 30–39 years (IR 4.81 vs. 3.66; aHR 1.33; 95% CI 1.06–1.66), and without comorbidities (IR 6.09 vs. 5.13; aHR 1.18; 95% CI 1.05–1.33) was significantly higher than that of the control cohort. Table 4 presents the results from a more detailed stratification between non-morbid and morbid obesity cohorts. Within the morbid obesity group, the risk of HZ across different genders was significantly higher than in the control group. For females, the IR was 8.30 compared to 6.63 in the control group, with an aHR of 1.36 (95% CI 1.06–1.74). For males, the IR were 6.82 for the control group and 6.83 for the morbid obesity group, with an aHR of 1.66 (95% CI 1.24–2.23). Among the age groups 40–49 and ≥ 50, the risk of HZ in the morbid obesity cohort was also significantly higher compared to the non-morbid obesity cohort, with aHR of 1.59 (95% CI 1.10–2.30) and 1.58 (95% CI 1.18–2.12) respectively. Additionally, the analysis of comorbidities revealed that the morbid obesity cohort consistently displayed a higher risk of developing HZ compared to the non-morbid obesity cohort, irrespective of the presence of comorbid conditions. In Fig. 3, the Kaplan–Meier plots for non-morbid obesity and morbid obesity showed overlapping curves, suggesting a crossing lines phenomenon. To further investigate this, we subdivided the study population based on follow-up duration: less than 14 years and 14 years or more. This allowed us to separately evaluate the risk of developing HZ in the non-morbid and morbid obesity groups. We found a significant difference in the cumulative incidence of HZ between these groups during the shorter follow-up period (p < 0.001), with the morbid obesity group exhibiting a higher incidence compared to the non-morbid obesity group. However, no significant difference was observed in the groups with a follow-up of 14 years or more.

Table 3 Cox proportional hazards regression analysis for the risk of herpes zoster.
Full size table
Table 4 Cox proportional hazards regression analysis for the risk of herpes zoster in different type of obesity.
Full size table
Figure 3
figure 3

Kaplan–Meier curves of the cumulative incidence rate of herpes zoster during the follow-up period between non-morbid obesity and morbid obesity groups.

Full size image

Discussion

Extensive research has revealed a strong correlation between obesity and various chronic diseases, such as DM and CKD. Additionally, obesity has been found to be associated with musculoskeletal conditions like sciatica and endocrine-related disorders such as PCOS. It is noteworthy that these diseases have also been linked to the reactivation of HZ. This connection underscores the intricate interplay between obesity, its related health conditions, and their potential impact on the reactivation of HZ.

In a previous systematic review, Freemantle et al. reported a strong association between abdominal obesity and DM16. In addition, Lu et al. found that compared with the normal weight population, the odds ratios (OR) for DM were 1.55 and 1.85 for individuals with abdominal and compound obesity, respectively17. The authors suggested weight control to decrease the risk of DM17, particularly reducing waist circumference16. A population-based study, which used data from Taiwan’s NHI program, found that the incidence of HZ in patients with and without DM was 7.85 and 6.75 per 1000 person-years, respectively18. A recent meta-analysis found a similar HZ incidence among patients with DM, with an incidence of HZ of 7.22 and 4.12 per 1000 person-years in patients with and without DM, respectively19. Patients with DM had a higher risk of HZ than those without DM, and the relative risk (RR) was 1.38. Huang et al. suggested that HZ vaccination is necessary for patients with DM20.

Obesity increases the risk of CKD development. Chang et al.21 reported that compared with the normal weight population, the 5-year cumulative incidence of CKD in patients with overweight and obesity was 3.5 and 6.7 per 1000 person-years. A meta-analysis of 21 studies with an average follow-up period of 9.86 years revealed that patients with obesity had a higher risk of CKD than patients without obesity, and the RR was 1.81. The authors suggested that obesity can be used as a predictive factor for CKD development22. In patients with obesity, the mechanism of renal function impairment may be due to lipotoxicity and changes in adipose tissue secretions, resulting in renal inflammation, oxidative stress, and fibrosis23. A nationwide study found that the incidence of HZ in patients with predialysis CKD and patients without CKD was 8.76 and 6.27 per 1000 person-years, respectively. The authors reported that patients with predialysis CKD had a 1.4-fold increased risk of developing HZ compared with patients without CKD24. Dialysis was associated with a higher risk of HZ among patients with CKD25. Lin et al.26 presented the risk of HZ in patients with CKD treated with various therapies. They found that compared with the control group, the HR of HZ risk in patients with hemodialysis, peritoneal dialysis, and renal transplantation were 1.35, 3.61, and 8.46, respectively. Therefore, HZ vaccination is recommended for patients with CKD25.

Zhou et al.27 reported that BMI was significantly associated with an increased risk of sciatica, and the OR was 1.33. The authors also found that waist and hip circumference, total body fat mass, and fat percentage were associated with an increased risk of sciatica. A meta-analysis of 26 studies found that both overweight and obesity increased the risk of sciatica, and the OR were 1.12 and 1.31. In addition, the two factors also increased the risk of sciatica-related hospitalization, and the OR was 1.16 for overweight and 1.38 for obesity28. To prevent sciatica, weight control is recommended for patients with obesity27,28. In our previous study, patients with sciatica were 1.19-fold more likely to develop HZ compared with those without sciatica14. This may be because of immune system dysfunction in patients with chronic pain29.

Another meta-analysis found that women with PCOS had a higher rate of overweight and obesity. Compared with women without PCOS, women with PCOS had a RR of 1.95 for overweight, 2.27 for obesity, and 1.73 for central obesity30. PCOS typically manifests clinically after weight gain, and the main characteristics are hyperandrogenism, reproductive and metabolic dysfunction31. This is because the obese women with PCOS have a lower level of sex hormone-binding globulin and higher levels of total testosterone and fasting insulin32. Glueck et al.33 found that the symptoms of PCOS can be improved after 5–10% weight loss. Thus, weight control is critical for patients with PCOS33. In our previous study, we found that patients with PCOS had a higher risk of HZ development. The incidence of HZ in patients with and without PCOS was 3.92 and 3.17 per 1000 person-years, respectively. Patients with PCOS had a 1.26-fold higher risk of developing HZ than patients without PCOS15.

Previous studies have established a connection between DM, CKD, sciatica, PCOS, and the occurrence of HZ. However, our present study yielded distinct findings. We observed that among patients without any comorbidity, those with obesity alone exhibited a significantly higher risk of HZ compared to non-obese patients. This suggests that obesity could potentially act as a stressor that triggers the reactivation of HZ. These novel results highlight the role of obesity as a potential contributing factor in HZ reactivation, even in the absence of other comorbidities.

As this is retrospective study in nature, it is important to acknowledge its limitations. Firstly, the influence of weight loss medication on the study results cannot be fully accounted for. Since weight loss pills are not covered by the NHI in Taiwan, information regarding medication for weight loss is not available in the NHIRD. However, we believe that the impact of weight loss pills on the findings is minimal, considering the limited number of individuals who self-pay for such medication, thus allowing us to reasonably ignore any potential bias arising from weight loss pills. Second, previous studies have indicated that exercise can have a positive impact on the immune function of the human body34 and it may potentially influence the HZ occurrence. However, lifestyle factors, including exercise, are not available in the NHIRD. Therefore, the study is unable to account for the potential impact of exercise on the results. Thirdly, the NHIRD does not provide access to laboratory data such as glucose or creatinine levels. This absence of laboratory data limits the assessment of disease severity, which may potentially influence the study outcomes. Fourthly, since the study utilizes data from the NHIRD, the study population is identified using ICD diagnostic codes. The database does not explicitly record height and weight, making it impossible to calculate the body mass index. Although diagnoses may be made by physicians from various specialties, Taiwan's National Health Insurance Administration enforces a stringent audit system that penalizes inappropriate diagnoses and treatments. Therefore, the diagnoses in the study are considered reliable. Despite these limitations, it is worth noting that this study is based on nationwide data with a large sample size. Thus, it can still provide substantial evidence and serve as a valuable reference for clinicians and researchers in the field.

Conclusion

Obese patients are at a higher risk of HZ development, especially in the patients with morbid obesity. Weight reduction is critical for preventing the onset of chronic diseases and decreasing the risk of HZ in patients with obesity.