High Prevalence of Herpes Zoster in
Patients With Depression
Chun-Hui Liao, MD, MSPHa,b; Chen-Shu Chang, MDe,f; Chih-Hsin Muo, MScc; and Chia-Hung Kao, MDb,d,*
D
epression, characterized by pervasive and persistent depressive mood, decreased confidence, and loss of interest in generallyagreeable activities, can lead to withdrawal from social interaction and influence the subsequent development of physical illnesses; it is a biological, psychological, and social illness. Depression has also been observed to increase patients’ risk of physical illness.1 The World Health Organization (WHO) indicated that depressive disorders were the third leading cause of disease in 2004, and they are expected to rank the highest by 2030.2
Many hypotheses support the interactions between the central nervous system and the immune system in chronic stress response and in depression. In these interactions, consistent activation of the hypothalamic-pituitary-adrenal (HPA) axis potentially impairs immune response and contributes to the development and progression of numerous medical disorders.3 Depression and anxiety have been observed to affect immune function negatively and lead to increased susceptibility to infection.4,5 Furthermore, Irwin et al6,7 demonstrated that patients with major depression have lower levels of varicella-zoster virus (VZV)–specific cell-mediated immunity, resulting in impaired cellular immunity and increased susceptibility to herpes zoster. Herpes zoster, a painful neurocutaneous syndrome, results from the reactivation and replication of latent VZV in the dorsal root and cranial nerve ganglia.7 A higher risk of occurrence was reported
among patients with depressive disorders.8 Despite providing evidence supporting depression as a risk factor for herpes zoster, the study had a cross-sectional design and the sample size was limited to 250 patients with herpes zoster, which precludes causal inference.
the association between depression and herpes zoster. One epidemiologic study9 revealed that psychiatric diseases are associated with herpes zoster. It revealed that patients aged 60 years or younger with affective psychosis and either neurotic illness or personality disorders are at a higher risk of developing herpes zoster. Hata et al10 revealed a slight association between herpes zoster and depression in a hospital-based study, but the association was nonsignificant. On the basis of these findings, the long-term presence of depressive disorders seems to be associated with herpes zoster, but the relationship between depression and herpes zoster remains unclear. Therefore, we conducted a 12-year follow-up, population-based cohort study to evaluate the risk factors for developing herpes zoster in patients with depression in comparison with a control group of patients without depression. The results of this study can provide an evidenced-based plan for a more comprehensive approach to the long-term care of patients diagnosed with depression, which incorporates the prevention of herpes zoster.
METHOD
Study Population
In this retrospective cohort study, we used the Longitudinal Health
Insurance Database (LHID), which is released by the Bureau of National Health Insurance (BNHI) of Taiwan. The BNHI
instituted a National Health Insurance (NHI) program on March 1, 1995, and this program covers nearly 99% of the population of Taiwan (http://www.nhi.gov.tw). The LHID covers 1 million insurants randomly selected from the 2000 registry and contains all medical records of each insurant from 1996 to 2011. The LHID identifies diseases based on the International Classification of Diseases, Ninth
Revision (ICD-9).11 The insurant’s personal information is encrypted to protect his or her identity before being sent to researchers. This study was also approved by the institutional review board of China Medical University Hospital.
Study Participants
Participants in the study were patients diagnosed with depressive disorders (ICD-9: 296.2, 296.3, 300.4, and 311) from 2000 to 2005. Those with a history of herpes zoster (ICD-9: 053) after the date of depressive disorder diagnosis
were excluded. The entry date was defined as the date of depression diagnosis. The 22,886 patients with depression were assigned to a depression cohort. For each patient in the depression cohort, we selected approximately 4 control patients from those without a history of depression or herpes zoster before the entry date. They were frequency matched with patients in the depression cohort based on age (in 5-year bands) and sex. Control patients were assigned the same index date as depression cohort patients. All patients were followed from the index date until herpes diagnosis, withdrawal from the program, or the end of 2011, whichever occurred first.
Baseline Comorbidities
Comorbidities were identified before the index
date. We considered the following comorbidities in this study: anxiety (ICD-9: 300.00), sleep disorders (ICD-9: 307.4 and 780.5), diabetes (ICD-9: 250), hyperlipidemia (ICD-9: 272), hypertension (ICD-9: 401–405), renal diseases 9: 580–589), malignant conditions
(ICD-9: 140–208), rheumatic diseases (ICD-(ICD-9: 710, 714, and
725), autoimmune diseases (ICD-9: 245.2, 250.01, 340, 358, 555.9, 556.9, 579, 696.0, 696.1, 710.1, and 710.2), and organ transplants (ICD-9: 11.6, 33.5, 33.6, 37.5, 50.5, 52.8, and 55.6). All comorbidities were identified based on outpatient or inpatient claims data, except malignant conditions. Malignant conditions were identified using the Catastrophic Illness Patient Registry (http://www.nhi. gov.tw/English/webdata/webdata.aspx?menu=11&menu_ id=596&WD_ID=596&webdata_id=3180), which contains information on patients who have received pathological treatment, or computed tomography or magnetic resonance imaging scans. Antidepressant use was a confounder
for developing herpes zoster. A patient who received an antidepressant for more than 90 days within 180 days before the end point was defined as a user. Antidepressants were tricyclics (including amitriptyline, clomipramine, doxepin, and imipramine), selective serotonin reuptake inhibitors (including fluoxetine, paroxetine, sertraline,
citalopram, escitalopram, and etoperidone), monoamine oxidase inhibitors (including selegiline and moclobemide), heterocyclic antidepressants (including trazodone), and others (including bupropion, venlafaxine, duloxetine, and mirtazapine).
Statistical Analysis
The difference in distribution between the 2 cohorts was analyzed using a χ2 test and a t test for categorical and continuous variables. We calculated the incidence of herpes zoster (per 1,000 person-years) for the 2 cohorts,
and also assessed the risk for herpes zoster development and herpes zoster– associated risk factors by using a Cox
proportional hazards regression model, controlling for age, sex, and comorbidities. We used this multivariable model to estimate age-, sex-, and comorbidity-specific incidence and risk of herpes zoster. We examined the joint effect of herpes zoster and herpes zoster–associated top 3 risk factors on depression, hyperlipidemia, rheumatic diseases, and malignant conditions. The association between depression and antidepressant in herpes zoster was also assessed. The cumulative incidences of herpes zoster in the 2 cohorts were plotted using Kaplan-Meier analysis, and the difference was calculated using the log-rank test. We performed all statistical analyses using SAS 9.3 (SAS Institute Inc; Cary, North Carolina). A significance level of P < .05 was determined using a 2-tailed test.
RESULTS
Baseline Characteristics of the 2 Cohorts We included 114,428 patients in this
study, who were divided into 2 cohorts. The depression cohort contained 22,886 patients, and the control cohort contained 91,542 patients. The majority of the depression cohort were women (61.7% vs 38.3%), and the mean age was 46.3 years (SD = 17.9). Patients in the depression cohort had more comorbidities than did patients in the control cohort, particularly anxiety (17.2% vs 3.09%)
and sleep disorders (38.7% vs 10.2%; Table 1). About 20.5% of patients with depression and 1.58% of controls had received an antidepressant for more than 90 days within 180 days before the end point.
Incidence Rate and Hazard Ratios for Herpes Zoster and Herpes Zoster– Associated Risk Factors
During the study period, 774 patients with depression and 2,358 control patients developed herpes zoster. As compared to control patients, depressed patients had 1.30 times higher incidence of herpes zoster (4.58 in patients with depression vs 3.54 in control patients per 1,000 person-years) and, in the multivariable model, had 1.11 times higher incidence (95% CI, 1.01–1.21) (Table 2). At the 11-year follow-up, the cumulative incidence of herpes zoster in the depression cohort was approximately 1% higher than in the control cohort (log-rank P < .0001; Figure 1). Incidence of herpes zoster increased with age, from 1.00 to 8.39 per 1,000 person-years, and, in the multivariable model, the risk
increased from 1.03 in patients aged 25–34 years to 6.49 in patients aged 65 years and older. In the multivariable model, patients with malignant conditions had the highest risk
of developing herpes zoster (hazard ratio [HR] = 1.41; 95% CI, 1.15–1.72), followed by patients with rheumatic diseases (HR = 1.28; 95% CI, 1.14–1.44), hyperlipidemia (HR = 1.24; 95% CI, 1.14–1.36), renal diseases (HR = 1.21; 95% CI, 1.08–1.36), anxiety (HR = 1.21; 95% CI, 1.07–1.38), sleep disorder (HR = 1.20; 95% CI, 1.09–1.31), and hypertension (HR = 1.11; 95% CI, 1.02–1.21).
Incidence and Risk of Herpes Zoster in the Depression Cohort Versus the Control Cohort Based on Age, Sex, and Comorbidity
In a sex-specific analysis, incidence of herpes zoster in the depression cohort was higher than in the control cohort, but risk was significantly different in men (HR = 1.26; 95% CI,
1.09–1.47; Table 3). Only patients aged 45 to 54 years showed significant changes in incidence and risk of herpes zoster
(HR = 1.44; 95% CI, 1.19–1.73). Patients with depression and without any comorbidity had a significantly higher risk than controls without any comorbidity (HR = 1.29; 95% CI, 1.07–1.56).
Joint Effect of Herpes Zoster and Herpes Zoster–Associated Risk Factors
Compared with study subjects without depression,
rheumatic diseases, and malignant conditions, a 1.27-fold risk for herpes zoster was found in subjects with only depression (95% CI, 1.14–1.42; Table 4) in the age- and sex-adjusted model. Patients with depression had an increased risk with hyperlipidemia or rheumatic disease compared with patients with only depression (data not shown). Compared with subjects not receiving antidepressant treatment, subjects who received antidepressant treatment had a decreased risk (HR = 0.66; 95% CI, 0.47–0.93), but patients with depression who were taking antidepressant treatment had an increased risk (HR = 1.28; 95% CI, 1.09–1.49). Furthermore, we
also found the patients with depression had higher risk of postherpetic neuralgia after herpes zoster infection. Depressive patients without antidepressants had more risk to suffer from the complaints (see Supplementary eTable 1 at Psychiatrist.com).
DISCUSSION
The results of this retrospective, population-based cohort study indicate that the incidence of herpes zoster in patients with depression is 1.30 times higher than in patients without depression. A significantly increased risk of herpes zoster in depressed patients, with an HR of 1.11 (95% CI, 1.01–1.21), was observed after adjusting for potential confounding factors, including demographic factors and potential risk factors for herpes zoster. Further analysis revealed that the increased risk of herpes zoster was significant among
patients aged 45 to 54 years.
Research has proven that reactivation of the latent VZV
causes herpes zoster.12,13 Researchers have also revealed that the decline of cellular immunity to VZV predisposes patients to developing herpes zoster.14–17 One study18 indicated that depression is associated with infection because it causes immunosuppression. Irwin et al6,7 suggested that patients with major depression had lower VZV-specific cell-mediated immunity. The results of our study provide epidemiologic evidence that depression is positively associated with herpes zoster. Furthermore, our analysis indicates that the prevalence of anxiety and sleep problems is greater among patients with depression. Previous research has revealed that patients with insomnia are more susceptible to infection.19 Anxiety can also cause vulnerability to infection.20 This could partially explain why patients with depression in our study had a higher risk of herpes zoster.
Moreover, research has linked depression to nutritional deficiencies.21–23 Nutritional deficiencies might decrease specific immune response to VZV.24–26
This research is supported by the results of our study, and is consistent with the hypothesis that patients with depression are prone to developing herpes zoster because of impaired cellular immunity or nutritional deficiencies. We also observed that the risk of developing herpes zoster increased significantly with age. This finding also correlates with previous studies27,28 that have indicated that VZV-specific cell-mediated immunity declines with age. Forbes et al29 found that patients with depression who were younger than 50 years had higher risk of herpes zoster than a control group in UK Clinical Practice Research Datalink primary care data. Our age-specific analysis determined that patients diagnosed with depression and aged 45 to 54 years had a significantly higher risk of developing herpes zoster. Frank et al30 revealed that early onset depression might be associated with a reduction in natural killer cell activity and
number, negatively affecting antiviral defense mechanisms.31 However, this hypothesis could not explain why patients with
depression aged 45 to 54 years were more sensitive to herpes zoster infection in our cohort study. Further investigation is required to answer this question. The result also indicates that patients aged 45 to 54 years with depression might need vaccination for herpes zoster.
We determined that patients diagnosed with depression are more likely to have comorbid chronic medical diseases than control patients32; after controlling for potential
confounding effects caused by comorbidities, the results still indicate a significantly increased risk of developing herpes zoster in patients with depression. Our data also reveal that patients with malignant conditions, rheumatic diseases, hyperlipidemia, renal diseases, anxiety, sleep disorder, and hypertension have a significantly higher risk of developing herpes zoster. In addition to renal diseases, malignant conditions, rheumatic diseases, and autoimmune diseases, other medical conditions can also induce the reactivation of VZV, causing herpes zoster to develop.10 We also observed that patients with hyperlipidemia have a higher incidence of herpes zoster. Our epidemiologic study links hyperlipidemia to herpes zoster; diminished cellular immunity caused by hyperlipidemia might have increased patients’ susceptibility to reactivation of cells causing VZV infection.33 Several studies have indicated that obesity causes a potential risk for infection.34–38 Obesity and hyperlipidemia have a close positive correlation.39 A more thorough investigation is required to evaluate the incidence of herpes zoster among the patients with hyperlipidemia.
The main strengths of our study are the large sample size, length of the longitudinal follow-up, and a large control
Table 4. Joint Effect Between Herpes Zoster and Herpes Zoster–Associated Risk Factor Model 1a Depression Hyperlipidemia Rheumatic Diseases Malignant Condition
Cases, n Population, n Hazard Ratio (95% CI) No No No No 1,573 75,771 1.00 No Yes No No 477 9,995 1.46 (1.31–1.62)*** No No Yes No 153 2,801 1.69 (1.43–2.00)*** No No No Yes 42 1,106 1.46 (1.07–1.98)* No Yes Yes No 82 1,449 1.56 (1.24–1.95)*** No Yes No Yes 18 288 2.10 (1.32–3.34)** No No Yes Yes 10 102 3.52 (1.89–6.57)*** No Yes Yes Yes 3 30 3.13 (1.01–9.73)* Yes No No No 434 16,842 1.27 (1.14–1.42)*** Yes Yes No No 206 3,627 1.76 (1.51–2.04)*** Yes No Yes No 68 1,101 1.98 (1.55–2.53)*** Yes No No Yes 12 373 1.41 (0.80–2.50) Yes Yes Yes No 41 748 1.41 (1.03–1.93)* Yes Yes No Yes 9 123 2.33 (1.21–4.50)* Yes No Yes Yes 2 50 1.46 (0.36–5.84) Yes Yes Yes Yes 2 22 2.84 (0.71–11.4) Model 2b Depression Antidepressant No No 2,325 90,092 1.00 No Yes 33 1,450 0.66 (0.47–0.93)* Yes No 586 18,203 1.06 (0.97–1.17) Yes Yes 188 4,683 1.28 (1.09–1.49)***
aAdjusted for age, gender, and antidepressant use.
bAdjusted for age, gender, and comorbidity. *P < .05. **P < .01. ***P < .001.
Abbreviations: HR = hazard ratio.
cohort with stratified age and sex. This cohort study allows for prospective reporting of depression and herpes zoster to validate cases, thus reducing recall bias caused by events that occurred prior to the time of enrollment.
This study also has several limitations. First, although we controlled for several potential confounding factors in
the statistical analysis, numerous possible confounding variables associated with herpes zoster, including smoking, alcohol use, diet, body weight, and presence of human immunodeficiency virus (HIV), were not included in our database. Information on these variables is not available for patients; therefore, it could not be controlled for directly in our analysis. The association among depression, HIV, and herpes zoster must be further evaluated. Second, we did not consider risk factors such as immunosuppressant drug use because of the complexity of this information. Additionally, the depression cohort exhibited a high risk of comorbidities causing immunodeficiency diseases, but the influence of immunodeficiency diseases (if any) most likely decreased after treatment. However, we might have underestimated the association between immunodeficiency diseases and the development of herpes zoster, because not all patients with such diseases received persistent drug treatment. Third, we could not evaluate the severity or status of depression because of the limited data provided by the NHI. Therefore, the correlation between severity of depression and herpes zoster was not within the scope of this study. Nonetheless, on the basis of statistical significance of our results, we do not believe that these limitations adversely affected our study. In conclusion, this national cohort study indicated that
depression is associated with an increased risk for developing herpes zoster, particularly in patients aged 45 to 54 years
and patients with comorbidities, including malignant conditions, rheumatic diseases, hyperlipidemia, renal diseases, anxiety, sleep disorder, and hypertension. Further investigation is required to identify the underlying causes of this association and determine whether appropriate
treatment of depression can decrease the risk for developing herpes zoster and decrease the risk of postherpetic neuralgia.
Indeed, a vaccination policy by the Centers for Disease Control and Prevention40 suggested that only adults 60 years and older receive the vaccination. This research raises the question of whether younger patients with depression, especially those aged 45 to 54 years, might benefit from
