Asthma as a risk factor for zoster in adults: A population-based case-control study

Abstract

BACKGROUND

We recently reported an increased risk of herpes zoster (shingles or zoster) in children with asthma, but little is known about whether it is true for adults with asthma. We determined whether asthma is associated with an increased risk of zoster in adults.

METHODS

This study was designed as a population-based case-control study. Zoster cases during the study period were identified among adults (aged ≥ 50 years) who resided in Olmsted County, Minnesota. We compared the frequency of asthma between zoster cases and birthday- and gender-matched controls (1:2 matching) without a history of zoster. Asthma status was ascertained by predetermined criteria. Conditional logistic regression model was used to assess the association of asthma with risk of zoster.

RESULTS

A total of 371 zoster cases and their 742 matched controls were enrolled. Of the 371 cases, 246 (66%) were females, 348 cases (94%) were Caucasians, and the mean (± standard deviation) age was 66.8±10.7 years. Twenty-three percent (N=87) of zoster cases had a history of asthma, compared to 15% (N=114) in controls. Controlling for pertinent covariates and confounders, there was a significant association between a history of asthma and risk of zoster (adjusted odds ratio: 1.70, 95% CI: 1.20–2.42, p=0.003). Population-attributable risk percent for asthma was about 10%.

CONCLUSIONS

Asthma is an unrecognized risk factor for zoster in adults. Consideration for immunizing adults with asthma aged over 50 years as a target group should be given.

Introduction

Herpes zoster (zoster or shingles) is estimated to occur in up to 30% of all adults by age 80 with nearly 1 million cases a year in the United States. ^1,2 The decade of life with the largest number of cases is 50 to 59 years, causing significant loss of work and productivity.¹ Yawn et al reported a rising trend of the incidence of zoster in Olmsted County, Minnesota, over time (3.2 to 4.1 per 1,000 person-years between 1996 and 2001).^1,3 A prior study completed in Olmsted County using comparable data sources and definitions reported the incidence rate of 1.3 cases per 1,000 person-years between 1945 and 1959, suggesting an increase greater than threefold over the 56 years studied.⁴ While the reasons for this increase in zoster rates are unknown, the introduction of the childhood varicella vaccination (1995) is unlikely to account for the rising trends of zoster in our study setting.

Despite the presumed immunosenescence with aging,^1,5,6 it is still unclear why some people develop zoster while others do not. Almost all (99.2–99.6%) of the United States population aged ≥ 40 years have serologic evidence for previous varicella infection and therefore are at risk for zoster,^5,7 but two-thirds of people will never develop zoster.¹ Only 8 to 10% of zoster cases have known significant immune suppression potentially suggesting unrecognized risk factors.¹ In this respect, little is known about the impact of conditions with immune dysfunction such as asthma on the risk of developing zoster.

Asthma represents one of the five most burdensome chronic diseases in the US ⁸ affecting 7–17% of the US population.^8–12 It increases the risk of serious and common microbial infections^13–19, which might be, in part, accounted for by suboptimal innate and adaptive immune functions.^20–24 Thus, we recently examined the relationship between asthma and the risk of zoster and found a significantly increased risk of zoster among children with asthma (adjusted odds ratio: 2.07, 95% CI: 1.24–3.52).²⁵

This population-based study expands our exploration further to assess whether asthma status is associated with the risk of zoster in adults. This study provides an important insight into the nature of the impact of asthma on the risk of infections because zoster is a unique non-airway-related latent viral activation in dorsal root ganglia, which is primarily controlled by cell-mediated immunity.^13,14,16,17

METHODS

Study setting

Olmsted County, Minnesota, provides unique advantages for population-based epidemiologic studies because medical care is virtually self-contained within the community. In addition, when patients register with any health care providers in the community, they or their parents/legal guardian are asked to grant or refuse authorization of use of their medical records for research. Authorization is granted by over 95% of all individuals.²⁶ Medical records research using the geographically defined population of Olmsted County was possible through the Rochester Epidemiology Project (REP) which has been continuously funded by NIH and maintained since 1960.^27–30 Within the REP, ³¹ each patient seen at any health care facility within the county has all diagnosis codes collected with health care services and site of service. This study protocol was approved by the IRBs at both Mayo Clinic and Olmsted County Medical Center.

Study design

The study was designed as a population-based case-control study, which compared the frequency of asthma prior to index date among incident cases of zoster with that among birth date (±1 year)-, gender-, initial clinic or hospital registration date, and index-date (±1 year)-matched corresponding controls (1:2 individual matching) without a history of zoster. Index (incident) date for cases was the date of the first zoster diagnosis and for controls was the closest clinic visit date (within 1 year) to the index date of corresponding cases.

Case ascertainment

All potential zoster cases (≥ 50 years of age) occurring between January 1, 2010, and October 31, 2011were identified from medical records using case ascertainment algorithm used for our previous study.^32,33 All potential zoster incident cases were identified from a broad search of diagnostic codes (International Classification of Diseases, Ninth Revision (ICD-9) codes 053.xx) for zoster and zoster complications. Medical records regarding each potential incident case were reviewed to verify that the case was indeed a new case of zoster based on criteria for zoster. Confirmation required either 1) positive laboratory data supported with a Tzanck smear, viral culture, or a positive polymerase chain reaction test for zoster virus or 2) clinical criteria used in our previous study: a) a statement of characteristic rash (vesicular rash on dermatome) in medical records, b) signs or a statement of pain or itching at rash site during interview, and c) a physician diagnosis of zoster in medical record. Exclusion criteria included: 1) individuals without research authorization, 2) non-Olmsted County, Minnesota, residents, 3) another diagnosis explaining the rash, such as culture positive for herpes simplex, and 4) clinical conditions making ascertainment of asthma difficult listed in Table 1.

Table 1.

Definition of asthma

Patients were considered to have definite asthma if a physician had made a diagnosis of asthma and/or if each of the following three conditions were present, and they were considered to have probable asthma if only the first two conditions were present: History of cough with wheezing, and/or dyspnea, OR history of cough and/or dyspnea plus wheezing on examination, Substantial variability in symptoms from time to time or periods of weeks or more when symptoms were absent, and Two or more of the following: Sleep disturbance by nocturnal cough and wheeze Nonsmoker Nasal polyps Blood eosinophilia higher than 300/uL Positive wheal and flare skin tests OR elevated serum IgE History of hay fever or infantile eczema OR cough, dyspnea, and wheezing regularly on exposure to an antigen Pulmonary function tests showing one FEV1 or FVC less than 70% predicted and another with at least 20% improvement to an FEV1 of higher 70% predicted OR methacholine challenge test showing 20% or greater decrease in FEV1 Favorable clinical response to bronchodilator

Patients were excluded from the study if any of these conditions were present: Pulmonary function tests that showed FEV1 to be consistently below 50% predicted or diminished diffusion capacity Tracheobronchial foreign body at or about the incidence date Hypogammaglobulinemia (IgG less than 2.0 mg/mL) or other immunodeficiency disorder Wheezing occurring only in response to anesthesia or medications Bullous emphysema or pulmonary fibrosis on chest radiograph PiZZ alpha1-antitrypsin Cystic fibrosis Other major chest disease such as juvenile kyphoscoliosis or bronchiectasis FVC forced vital capacity; FEV1, forced expiratory volume in 1 sec.

Selection of controls

Two matched controls were randomly selected from the Rochester Epidemiology Project (i.e., population-based sampling frame for Olmsted County, MN, residents). The literature showed that 99.2–99.6% of the US population aged ≥40 years had serologic evidence for previous varicella infection which were quite steady over time (1988–2004)^5,7. We did not assess serologic evidence for a history of varicella infection in the past for controls.

Determination of exposure status (asthma status)

We applied predetermined criteria for asthma delineated in Table 1, which have been extensively used in research for asthma epidemiology and have shown excellent construct validity and reliability (0.72–92).^{13,14,17,18,34,35} As most subjects with probable asthma (85%) became definite asthma over time, we combined both probable and definite asthma.³⁶

Other variables

We collected pertinent variables for this study: asthma medication and control status, comorbid conditions at the time of index date, other atopic conditions prior to the index date, zoster vaccination prior to index date, and proxy measures for health care access (influenza vaccination in a preceding year and pneumococcal polysaccharide vaccination). The comorbid conditions examined in this study are listed in Table 2. Other atopic conditions were ascertained by a diagnosis of atopic dermatitis, allergic rhinitis, and food allergy documented in medical records prior to the index date of zoster. We collected data on corticosteroid therapies and asthma control status within six months of index date of zoster. We defined poorly controlled asthma as any asthma symptoms documented in medical records, systemic corticosteroid therapy for asthma symptoms, unscheduled visits for asthma, visit to the emergency department for asthma, or hospitalization for asthma within six months prior to index date.^37,38

Table 2.

Sociodemographic and clinical characteristics of subjects and their associations with risks of herpes zoster

Variable	Zoster Cases (N=371)	Controls (n=742)
Female gender	246 (66%)	492 (66%)
Age at index date	66.8±10.8	66.8±10.7
Age of asthma onset	50.4±16.8	48.2±18.0
Years since registration	39.3±17.3	41.5±16.8
Ethnicity:
Caucasian/non-Hispanic	348 (94%)	683 (92%)
Non-Caucasian	18 (5%)	51 (7%)
Unknown	5 (1%)	8 (1%)
Education:
< High school	20 (5%)	31 (4%)
High school graduate	113 (30%)	240 (32%)
College	120 (32%)	225 (30%)
Post graduate	107 (29%)	212 (29%)
Unknown	11 (3%)	34 (5%)
Asthma	87 (23%)	114 (15%)
Allergic rhinitis	129 (35%)	216 (29%)
Atopic dermatitis	45 (12%)	58 (8%)
Food allergy status	9 (2%)	10 (1%)
Comorbid Conditions*
Alcohol addiction	5 (1%)	24 (3%)
Autoimmune disease**	35 (9%)	44 (6%)
COPD	17 (5%)	24 (3%)
Chronic renal insufficiency	26 (7%)	41 (6%)
Cirrhosis	5 (1%)	10 (1%)
Congestive heart failure	12 (3%)	18 (2%)
Coronary artery disease	63 (17%)	106 (14%)
Diabetes mellitus	58 (16%)	133 (18%)
History of stroke	10 (3%)	38 (5%)
Hematologic cancer	28 (8%)	4 (1%)
Transplant recipients	2 (1%)	1 (0%)
Skin disease†	18 (5%)	27 (4%)
Psychologic conditions††	84 (23%)	196 (26%)
Any comorbidity	208 (56%)	409 (55%)
Physical trauma	14 (4%)	9 (1%)
Smoking status:
Never smoked	194 (52%)	379 (51%)
Active	17 (5%)	63 (8%)
Passive	2 (1%)	4 (1%)
Ex-smoker	156 (42%)	272 (37%)
Unknown	2 (1%)	24 (3%)
Influenza vaccination	253 (68%)	425 (57%)
Pneumococcal vaccination	193 (52%)	357 (48%)
History of zoster vaccination	83 (22%)	196 (26%)

^*Comorbid conditions are not mutually exclusive since some subject had multiple comorbid conditions and thus the sum of percentages is greater than the percentage correspondence to “any comorbidity”;

^**Autoimmune disease include SLE, Rheumatoid arthritis, and inflammatory bowel disease;

^†Skin disease includes any skin conditions other than atopic dermatitis such as psoriasis;

^††Psychologic conditions include a physician diagnosis of psychologic/psychiatric conditions such as mood disorders, anxiety disorders, conduct or behavioral disorder, and adult attention deficit disorder

Data analysis

Matched analysis via conditional logistic regression was used, with zoster case status as the target of endpoint in the model and asthma status as the primary explanatory variable. In each model the matching variables (age, gender, and years since registration) were included, as adjusting covariates to control for any residual confounding not prevented by the matching. Asthma was carried forward into a multivariable model along with other factors screened from univariable modeling using an alpha level of 0.2.³⁹ Atopic dermatitis, allergic rhinitis, and food allergy which share similar immunologic characteristics as asthma were left out of this model to avoid collinearity, though each was included in separate multivariable models (removing asthma) to test the association with zoster. We calculated the population attributable risk percent (PAR%) of asthma for zoster.⁴⁰ We examined a differential effect of asthma on the risk of zoster among different age groups by assessing statistical significance of interaction term between asthma and age groups. Incidence of zoster was estimated among adults in the county during the two-year study time frame, with rates adjusted for age and gender by direct standardization against the 2010 US population. All analyses were performed with SAS, version 9.3 (SAS Institute, Cary, NC, USA).

RESULTS

Study subjects

Of the total 475 potential zoster cases identified during the study period, 93 subjects were excluded from incidence analyses due to the following reasons: non-Olmsted County residents (n=62), no research authorization (n=19), and no physician diagnosis of zoster in medical records during the study period (n=12). For analyses pertaining to the relationship between asthma and zoster, an additional 10 subjects were excluded due to clinical conditions making asthma ascertainment difficult (2 for pulmonary fibrosis, 2 for bronchiectasis, 5 for significant decreased pulmonary function defined by FEV1<50% or diminished diffusion capacity, and 1 for significant scoliosis) and 1 due to insufficient medical record to determine asthma status.

Based on a total of 382 incident cases, the overall age- and gender-adjusted incidence rate of zoster during our study period was 4.3 (95% CI, 3.9–4.7) per 1000 person-years. The age-adjusted sex-specific incidence rates (per 1000 person-years) were 5.3 (95% CI: 4.7–6.0) in females and 3.2 (95% CI: 2.7–3.8) in males. The sex-adjusted age-specific incidence rates were 3.1 (95% CI: 2.6–3.7) in subjects aged 50–59 years, 4.5 (95% CI: 3.7–5.4) in those aged 60–69 years, and 5.8 (95% CI: 4.9–6.8) in those aged 70 years or older. The characteristics of the zoster cases and their matched controls are summarized in Table 2.

Association between asthma and risk of zoster

We found a significant association between asthma and risk of zoster in our study population. The results are summarized in Table 3. Of the 371 zoster cases, 87 (23%) were asthmatics, compared to 114 of 744 controls (15%) who had asthma (OR, 1.73; 95% CI, 1.26–2.39; p<0.001). The mean (±SD) age at the onset of asthma was 49.1 (±17.5) years; 50.4 (± 16.8) in zoster cases and 48.2 (±18.0) in controls. Adjusted for potential confounders identified from univariate models (Table 3), the association between asthma and zoster remained significant (adjusted OR, 1.70; 95% CI, 1.20–2.42; p=0.003; PAR% = 9.7%). Based on a test of interaction between asthma and age categories of <60 years (n=356), 60–69 years (n=332), and ≥70 years (n=425), there was no evidence of a differential asthma effect across age strata (p=0.978).

Table 3.

The association between asthma and risk of herpes zoster in univariable and multivariable analyses

Variable	Matched Analysis Result: OR (95% CI) [p-value]
	Univariable Models‡	Multivariable Model
Ethnicity:	p=0.313
Caucasian/non-Hispanic	1.0 (ref)
Non-Caucasian	0.65 (0.36, 1.15)
Unknown	1.18 (0.38, 3.63)
Education:	p=0.599
< High school	1.0 (ref)
High school graduate	0.81 (0.44, 1.48)
College	0.88 (0.48, 1.59)
Post graduate	0.78 (0.42, 1.45)
Unknown	0.51 (0.21, 1.25)
Asthma	1.73 (1.26, 2.39) [<.001]	1.70 (1.20, 2.42) [0.003]
Allergic rhinitis#	1.30 (1.00, 1.70) [0.054]
Atopic dermatitis#	1.71 (1.12, 2.61) [0.013]
Food allergy status#	1.77 (0.71, 4.38) [0.219]
Comorbid Conditions*
Alcohol addiction	0.42 (0.15, 1.13) [0.086]	0.40 (0.14, 1.17) [0.094]
Autoimmune disease**	1.65 (1.03, 2.64) [0.037]	1.41 (0.84, 2.38) [0.190]
COPD	1.48 (0.77, 2.84) [0.236]
Chronic renal insufficiency	1.35 (0.78, 2.34) [0.284]
Cirrhosis	1.17 (0.39, 3.48) [0.775]
Congestive heart failure	1.31 (0.59, 2.90) [0.511]
Coronary artery disease	1.24 (0.86, 1.79) [0.246]
Diabetes mellitus	0.83 (0.59, 1.16) [0.271]
History of stroke	0.50 (0.25, 1.03) [0.059]	0.60 (0.28, 1.27) [0.181]
Hematologic cancer	13.88 (4.86, 39.62) [<.001]	13.75 (4.66, 40.55) [<.001]
Transplant recipients	5.78 (0.45, 73.52) [0.176]	5.89 (0.12, 292.36) [0.373]
Skin disease†	1.36 (0.73, 2.55) [0.329]
Psychologic conditions††	0.80 (0.59, 1.08) [0.148]	0.79 (0.57, 1.10) [0.159]
Any comorbidity	1.04 (0.79, 1.35) [0.794]
Trauma	3.06 (1.27, 7.36) [0.013]	2.72 (1.07, 6.92) [0.036]
Smoking status:	p=0.025	p=0.018
Never smoked	1.0 (ref)	1.0 (ref)
Active	0.58 (0.33, 1.02)	0.60 (0.32, 1.12)
Passive	1.09 (0.19, 6.08)	1.15 (0.20, 6.59)
Ex-smoker	1.13 (0.86, 1.48)	1.27 (0.95, 1.69)
Unknown	0.17 (0.04, 0.75)	0.20 (0.05, 0.89)
Influenza vaccination	1.05 (0.93, 1.19) [0.402]
Pneumococcal vaccination	1.33 (0.95, 1.85) [0.092]	1.14 (0.78, 1.66) [0.497]
History of zoster vaccination	0.77 (0.55, 1.07) [0.124]	0.68 (0.47, 0.97) [0.035]

^‡Each model includes the matching variables – gender, age and years since registration – as adjusting covariates;

^#other atopic conditions (eg, atopic dermatitis) were not included in this model testing the association of asthma with risk of zoster, due to biological collinearity;

^*Comorbid conditions are not mutually exclusive since some subject had multiple comorbid conditions;

^**Autoimmune disease include SLE, Rheumatoid arthritis, and inflammatory bowel disease;

^†Skin disease includes any skin conditions other than atopic dermatitis such as psoriasis;

^††Psychologic conditions include a physician diagnosis of mood disorders, anxiety disorders, conduct or behavioral disorder, and adult attention deficit disorder

Associations of asthma control and inhaled corticosteroids (ICS) with risk of zoster

Use of ICS was not shown to be a risk factor for zoster in our study population. Among all asthmatics of cases and controls (n=201), only 11 (6%) were considered to be poorly controlled at the time of zoster assessment, and 62 (31%) had been treated with ICS and/or systemic corticosteroids (the dose of ICS was defined by the National Asthma Education and Practice Program guidelines): 49 on low-dose ICS (79%), 7 on medium-dose ICS (11%), 2 on high-dose ICS (3%), and 4 on high-dose ICS and oral corticosteroid therapy (6%). Twenty-five of 87 zoster cases (29%) were treated with ICS, compared to 37 of 114 controls (32%) (OR: 0.83, 95% CI: 0.45–1.54, p=0.547). For active (current) vs. inactive asthma status, both inactive asthma (OR: 1.90, 95%CI: 1.21–3.00) and active (current) asthma (OR: 1.61, 95%CI: 1.07–2.41) were associated with increased risks of zoster compared to those without asthma (p=0.003), but there was no difference in the risk of zoster between inactive and active asthma (p=0.61). For asthma control status, 6 of 75 zoster cases (7%) and 5 of 114 controls (4%) had poorly controlled asthma (OR: 1.53, 95% CI: 0.44–5.25, p=0.501).

Association of other factors with risk of zoster

From univariable analyses, the association of zoster was statistically significant for atopic dermatitis (OR: 1.71, 95% CI: 1.12–2.61, p=0.013). In separate models that controlled for the same adjusting covariates in Table 3, the odds of zoster were significantly higher in subjects with atopic dermatitis (adjusted OR, 1.72; 95% CI: 1.09–2.71; p=0.020) than those without. With asthma and other atopic conditions (allergic rhinitis and food allergy) additionally entered into the model, both asthma (adjusted OR: 1.54, 95%CI: 1.06–2.25; p=0.024) and atopic dermatitis (adjusted OR: 1.63, 95%CI: 1.03–2.59; p=0.038) were independently associated with risk of zoster. We also examined a potential interaction between asthma and atopic dermatitis in relation to the impact of asthma on the risk of zoster by stratifying the effect of asthma on the risk of zoster by atopic dermatitis status. While there was no significant interaction (p=0.287), the impact of asthma on the risk of zoster appeared to be greater among patients with atopic dermatitis (OR: 2.73, 95%CI: 1.02–7.29) than those without atopic dermatitis (OR: 1.55, 95%CI: 1.07–2.26). However, the associations of food allergy (p=0.145) and allergic rhinitis (p=0.070) with the risk of zoster only approached statistical significance in the multivariate analysis. A history of trauma and hematologic malignancy had increased risks of zoster compared to those without such conditions. As expected, neither influenza vaccine nor pneumococcal vaccine was associated with the risk of zoster, while zoster vaccine was associated with a lower risk of zoster.

DISCUSSION

This population-based study in adults upholds our previous finding on asthma as an independent risk factor for zoster in children with asthma. This may also be true for other atopic conditions such as atopic dermatitis. The impact of asthma on risk of infection may go beyond airways.

The overall age-and gender-adjusted incidence rate of zoster in our study (4.3, 95% CI, 3.9–4.7, per 1000 person-years) was similar to that in the United States (2.8–4.4 depending on study years).⁴¹ Our study results corroborated previously reported factors associated with the risk of zoster such as female gender, immunocompromised conditions, mechanical trauma, and zoster vaccination. ^25,42–44

Our study results on the association between asthma and risk of zoster (adjusted OR, 1.70; 95% CI, 1.20–2.42; p=0.003) are unlikely to be due to a detection bias because all cases were population-based incident zoster cases and no significant differences in influenza and pneumococcal vaccinations before the index date of zoster between cases and controls as a surrogate measure for access to health care were found. Regardless, the estimated effect of asthma was adjusted for these variables. No significant differences in access to health care services (e.g., general medical examination or evaluation for acute illnesses) between individuals with and without a physician diagnosis of asthma were found in our study setting.^17,45 Susceptibility bias might be an important concern given the relatively older population in our study subjects and higher comorbid conditions. To address this concern, we adjusted significant comorbid conditions individually, which were associated with the risk of zoster as shown in Table 3. Controlling comorbid conditions did not significantly change the effect size (1.73 to 1.70). Our study results suggest that there is no evidence of a differential impact of asthma across age strata. While both active and inactive asthma were associated with the increased risk of zoster, asthma medications and control status did not influence the association, and this is consistent with the literature.⁴⁶ Finally, other atopic condition such as atopic dermatitis (adjusted OR, 1.72; 95% CI: 1.09–2.71; p=0.020) was associated with risk of zoster, while both allergic rhinitis (adjusted OR, 1.30; 95% CI, 0.98–1.74; p=0.070) and food allergy (adjusted OR, 2.05; 95% CI, 0.78–5.38; p=0.145) approached statistical significance. When adjusted for atopic dermatitis, asthma was still significantly associated with the risk of zoster and there was no significant interaction between asthma and atopic dermatitis in relation to the effect of asthma on the risk of zoster. These results suggest the impact of asthma on the risk of zoster is independent of atopic dermatitis. These results potentially suggest biological coherence with our main study findings on the association between asthma and risk of zoster.

The results of this present work are consistent with our previous study results on the increased risk of zoster among children with asthma (adjusted OR: 2.09, 95% CI: 1.24–3.52, p=0.006). ²⁵ Recent studies conducted in Spain and England characterized adult patients with zoster without a priori hypothesis and showed that asthma was one of the conditions associated with an increased risk of zoster.^47,48 In addition, previous studies showed increased risks of herpes simplex ocular infections ^49,50 and eczema herpeticum⁵¹ among patients with atopic conditions.

The mechanisms underlying the association between asthma and risk of zoster are unknown. Waning cell-mediated immunity has been suggested to increase the risk of zoster^1,5,6 and asthma might play a role in waning of adaptive immunity.^19,52 Asthma as a Th2-predominent condition, which results in reciprocal suppression of Th1 immunity (cell-mediated immunity), could potentially in part explain the association observed in our study.^53,54 As viral gene products are needed to establish and maintain latency and host responses (T-cell immunity) determine latency vs. reactivation,⁵⁵ it is important to determine how asthma as a host factor influences the interaction between viral gene expression (e.g, ORF61 gene for varicella zoster virus) and host-immune response (e.g., interferon α, β, and γ).^56,57 Impairment in innate immune functions in skin and airways is well documented in patients with asthma or atopic dermatitis.^15,18,19 Although there is limited information on adaptive immune functions in asthmatics, there is some evidence that patients with asthma or other atopic conditions may have impairment or incompetence in cell-mediated immunity.^19,21,58

Clinicians and patients need to be aware that a history of asthma and atopic dermatitis might be unrecognized risk factors for zoster in both adults and children, although the underlying mechanisms are unknown. Neither asthma-related medications such as corticosteroids nor asthma control status affected the risk of zoster, although statistical power might be limited to address these associations. Given the large proportion of adults affected by asthma and availability of effective vaccine preventing zoster, consideration for immunizing adults with asthma or atopic dermatitis over 50 years or older as a target group for zoster vaccine should be given.

The main strength of the study is a population-based study design, which is based on incident zoster cases, minimizing misclassification as reflected in the similar incidence rate of zoster at the national level. Another important strength is epidemiologic advantages of our study setting with self-contained health care environment and the availability of all medical records for study subjects under the REP. Also, we used predetermined criteria for asthma (instead of self-report or ICD-9 codes), which were extensively used for previous epidemiologic investigations for the association between asthma and risk of various microbial infections relating our current results to our previous research work ensuring consistency and coherence.

Our study has limitations. Even if our study setting is self-contained and case identification was based on incident cases, we might have missed some zoster cases, thereby the incidence of zoster could be underestimated. However, the incidence estimate in our study was similar to that at the national level. Secondly, the study subjects were predominantly a Caucasian population, which might limit generalizability of our results to other ethnic groups or study settings. In addition to the comparable incidence rates of zoster in our study setting to that at the national level, our study results corroborated previously reported risk factors for zoster, suggesting suitability of our study subjects and setting in addressing the study question. Thirdly, asthma control status (i.e., poorly controlled asthma) was ascertained based on medical chart review resulting in the possibility of missing some asthma symptoms and/or administration of asthma medications unless they visited the clinic, ER, or were hospitalized for asthma and the use of asthma medications was documented in the medical record. Finally, our study subjects were a relatively older population (mostly born in 1940s when asthma incidence was low^59,60) and thus, comorbid conditions might be a significant confounder and the relationship between asthma and risk of zoster might vary across age strata. The association between asthma and risk of zoster was adjusted for significant comorbid conditions, and there was not a differential impact of asthma across different age strata. Apart from this concern, a potential misclassification of COPD as asthma, we limited our analysis to subjects without a history of smoking. Asthma status was still significantly associated with the risk of zoster (adjusted OR: 1.61, 95%CI: 1.00–2.57, p=0.048).

In conclusion, asthma might be an unrecognized risk factor for zoster in adults and the impact of asthma on risk of microbial infections or immune dysfunction may go beyond airways. This may be true for atopic dermatitis. Consideration for immunizing adults with asthma or atopic dermatitis over 50 years or older as a target group for zoster vaccine should be given.

Clinical implications.

Asthma is an unrecognized risk factor for herpes zoster in adults. Consideration for immunizing adults with asthma aged over 50 years as a target group should be given.

Abbreviation

HZ

herpes zoster

OR

Odds ratio

95%CI

95% confidence interval

REP

Rochester Epidemiology Project