Obesity and rhinitis in a nationwide study of children and adults in the United States

Abstract

Background

Obesity has been associated with higher risk of asthma and asthma severity, both in children and adults. However, studies evaluating the relation between obesity and rhinitis have yielded conflicting results.

Methods

We performed a cross-sectional study of obesity indicators and rhinitis using data from 8,165 participants in the 2005–2006 National Health and Nutrition Examination Survey. Allergic rhinitis was defined as physician-diagnosed hay fever or allergy, presence of symptoms in the past 12 months, and at least one positive allergen-specific IgE. Non-allergic rhinitis was defined as physician diagnosis and symptoms but no positive allergen-specific IgE. Multivariate regression was used to assess the relationship between obesity and rhinitis in children and adults.

Results

In adults, being overweight or obese was associated with increased odds of non-allergic rhinitis (adjusted odds ratio [aOR]=1.43, 95% confidence interval [CI]=1.06–1.93, P=0.02). Similarly, central obesity was associated with increased odds of non-allergic rhinitis in adults (aOR=1.61, 95% CI=1.20–2.16, P<0.01). In an analysis stratified by gender, the observed associations were attenuated and became non-statistically significant in female adults, but remained significant in male adults. Overweight, obesity or central obesity were not associated with allergic rhinitis in adults. In children, central obesity was associated with reduced odds of allergic rhinitis (aOR=0.35, 95% CI=0.19–0.64, P<0.01). After stratification by gender, this association was similar in female and male children.

Conclusions

In adults, obesity is associated with increased odds of non-allergic rhinitis, particularly in males. In children, central obesity is associated with reduced odds of allergic rhinitis, regardless of gender.

INTRODUCTION

Obesity and atopic diseases, such as asthma and allergic rhinitis, are major public health problems worldwide^{1, 2}. Multiple epidemiological studies have shown an association between obesity and asthma in children^3–5 and adults^6–9. While the mechanisms underlying the link between obesity and asthma are still unclear, there is growing evidence that obesity may be associated with the development of non-atopic asthma^{7, 10, 11} but may also worsen pre-existing atopic asthma^{3, 12}.

Even though a high proportion of subjects with asthma also suffer from allergic rhinitis¹³, the relation between obesity and allergic rhinitis (with or without co-existing asthma) is poorly understood. Studies of overweight or obesity and allergic rhinitis have predominantly included children and have yielded conflicting results^14–22. Whereas some studies reported that neither overweight nor obesity was associated with symptoms suggestive of allergic rhinitis in children¹⁹, others have shown a positive association between obesity and allergic rhinitis in children and young adults^20–22, with yet a few others showing an negative (inverse) association in school-aged children¹⁴ and adult men¹⁷. Of interest, some studies have shown that obesity is only associated with an increased risk of allergic rhinitis in female subjects, raising the question of whether hormonal surges around puberty predispose adolescent girls to allergic sensitization and allergic rhinitis^{20, 23}.

As symptoms of non-allergic rhinitis are sometimes similar to those of allergic rhinitis²⁴, misclassification may occur in studies in which a diagnosis of allergic rhinitis is based solely on a physician’s diagnosis without testing for allergic sensitization. Non-allergic rhinitis usually affects adults and presents with nasal symptoms. In children, the most common form of non-allergic rhinitis is infectious rhinitis². Non-allergic rhinitis differs from allergic rhinitis in that atopy is not involved, and thus affected patients have negative results for allergen-specific IgE, skin test reactivity (STR) to allergens, or nasal allergen challenges². Although both allergic and non-allergic rhinitis are risk factors for asthma²⁵, little is known about the relation between obesity and non-allergic rhinitis.

Since different types of obesity may have diverging effects on health, we examined if obesity (defined by body mass index, BMI) and central obesity (defined by waist circumference, WC) are differentially associated with rhinitis in a cross-sectional study of children and adults who participated in the National Health and Nutrition Examination Survey (NHANES) in the U.S.. Moreover, we examined whether the estimated effect of obesity on rhinitis differs according to atopic status (e.g. allergic vs. non-allergic rhinitis).

METHODS

Subject recruitment

NHANES is a cross-sectional nationwide survey designed to assess the health and nutritional status of the non-institutionalized population of the U.S. NHANES combines interviews and physical examinations of participants by highly trained personnel, which ensures standardization. Study participants were selected using a stratified multistage probability design. By design, persons 60 years and older and ethnic minorities (African Americans and Hispanics) were oversampled to increase the statistical power for data analysis and to represent U.S. population after accounting for sampling weight. Both children (ages 6 to 17 years) and adults (18 years or older) who participated in the 2005–2006 NHANES are included in this analysis.

NHANES was approved by the Institutional Review Board of the National Center for Health Statistics of the U.S. Centers for Disease Control and Prevention (CDC). Informed consent was obtained from all participants. A proxy provided information for survey participants who were under 16 years of age and for individuals who could not answer the questions by themselves. Details of the methods, protocols and definitions used in NHANES can be found at http://www.cdc.gov/nchs/nhanes.htm.

Study procedures

BMI and WC were measured by trained health technicians, according to recommendations from the Anthropometric Standardization Reference Manual of NHANES. BMI was calculated dividing the weight in kg by the height squared in meter (m²). Overweight was defined as a BMI ≥25 kg/m² in adults, and as a BMI ≥85^th percentile in children; obesity was defined as a BMI ≥30 kg/m² in adults, and as a BMI ≥95^th percentile in children. Central obesity was defined as a WC ≥ 95^th percentile for both adults and children. For the analysis of BMI as a continuous variable, raw BMI (kg/m²) was used in adults, and BMI z-scores were used in children. In children, BMI z-scores were calculated in SAS using equations based on the 2000 U.S. CDC growth charts²⁶. Similarly, raw WC (in centimeters) and sample-specific WC z-scores was used for data analysis in adults and children, respectively.

Serum IgE to each of fifteen common aeroallergens (dust mite [Der p 1 and Der f 1], dog dander, cat dander, German cockroach, Alternaria, Aspergillus, mouse urinary protein, rat urinary protein, ragweed pollen, rye grass pollen, Bermuda grass pollen, oak pollen, birch pollen, and thistle) was measured using the ImmunoCAP 1000 System (Pharmacia Diagnostics, Kalamazoo, MI). For each allergen tested, a serum IgE equal or greater than 0.35 KU/L (the detection limit of the ImmunoCAP system) was considered positive. Serum cotinine was measured by an isotope dilution-high performance liquid chromatography/atmospheric pressure chemical ionization tandem mass spectrometry (ID HPLC-APCI MS/MS). Details of the laboratory methods used in NHANES can be found under http://www.cdc.gov/nchs/nhanes/nhanes2005-2006/lab_methods_05_06.htm.

Allergic rhinitis was defined as physician-diagnosed hay fever or allergy with symptoms of hay fever or allergy in the previous 12 months, and at least one positive IgE to the allergens included in NHANES. Non-allergic rhinitis was defined as physician diagnosis of hay fever and allergy with symptoms in the past year but without any positive IgE to the allergens tested.

Statistical analysis

Primary sampling units and strata for the complex design of NHANES were taken into account for data analysis. Sampling weights, stratification, and clusters provided in the NHANES data set were incorporated into the analysis to obtain proper estimates and standard errors (SEs). All multivariate analyses were performed using logistic regression within the SURVEY procedure in SAS (SAS Institute, Cary, NC). All multivariate models were adjusted for age, gender, race/ethnicity (classified as non-Hispanic white, non-Hispanic Black, Hispanic, or other), annual household income (<$20,000/year vs. ≥$20,000/year), current asthma (having had asthma diagnosed by a doctor or health care professional and still have asthma), serum cotinine, and C-reactive protein (CRP) level. In adults, models were additionally adjusted for apnea symptoms (as such data were only available for participants older than 15 years), defined as a participant’s report of ever having been told by a doctor or health professional that they have sleep apnea and/or that they ever had snorted, gasped, or stopped breathing while asleep in the past 12 months. All statistical analyses were conducted with SAS 9.3 software.

RESULTS

The main characteristics of the 8,165 participants (2,358 children and 4,906 adults) are shown in Table 1. Of the 2,358 participating children aged 6–17 years, 61.6% were non-Hispanic white, more than 86% had a household income ≥$20,000/year or were covered by health insurance, slightly over a third were overweight or obese, 8.2% had central obesity, 11.8% had current asthma, 12.5% had allergic rhinitis and 7% had non-allergic rhinitis. Of the 4,906 adult participants, 71.9% were non-Hispanic white, more than 81% had a household income ≥$20,000/year or were covered by health insurance, two thirds were overweight or obese, 6.5% had central obesity, 14.7% had allergic rhinitis and 11.5% had non-allergic rhinitis.

Table 1.

Characteristics of study participants in NHANES 2005–2006

	Children (6–17 years) n=2,358	Adults (≥ 18 years) n=4,906
Age (years)	11.8 (0.1)	45.7 (0.8)
Male gender	1173 (51.2)	2360 (48.1)
Race
Non-Hispanic White	626 (61.6)	2329 (71.9)
Non-Hispanic Black	722 (13.6)	1168 (11.3)
Hispanic	880 (17.5)	1212 (11.6)
Other	130 (7.3)	197 (5.2)
Household income ≥ $20,000/year	1815 (86.3)	3671 (85.4)
Covered by health insurance	1953 (88.4)	3767 (81.0)
Current asthma	265 (11.8)	381 (8.2)
Serum cotinine (ng/ml)	7.8 (1.2)	65.6 (3.3)
C-reactive protein (mg/dl)	0.19 (0.02)	0.42 (0.02)
Apnea and/or symptoms1	-	906 (20.1)
Obesity/adiposity measures
Body mass index (kg/m2)	20.9 (0.2)	28.5 (0.2)
Overweight	387 (16.1)	1632 (32.5)
Obese	562 (19.3)	1677 (33.9)
Overweight and obese	949 (35.4)	3309 (66.5)
Waist circumference (cm)	73.2 (0.5)	97.4 (0.7)
Central obesity	239 (8.2)	308 (6.5)
Allergic rhinitis	273 (12.5)	616 (14.7)
Non-allergic rhinitis	124 (7.0)	431 (11.5)

Data are shown as mean (SE) for continuous variables, and as number (%) for binary variables. Numbers may vary due to missingness.

¹Data only available among adults.

Table 2 shows the results of multivariate analysis of overweight or obesity and rhinitis (allergic and non-allergic). Among children, there was no significant association between BMI or overweight/obesity and rhinitis (allergic or non-allergic). Among adults, however, each 1 kg/m² increment in BMI was associated with ~2% increased odds of non-allergic rhinitis (Model 1). The observed associations remained unchanged after additional adjustment for CRP level and apnea symptoms (Model 2). Neither BMI nor overweight/obesity was significantly associated with allergic rhinitis in adults.

Table 2.

Multivariate analysis of body mass index (BMI), overweight or obesity, and allergic rhinitis in children and adults

	Children (6–17 years)		Adults (≥ 18 years)
	Model 11	Model 22	Model 11	Model 23
	Odds Ratio (95% confidence interval)
Allergic rhinitis
BMI (continuous)4	0.94 (0.82–1.09)	0.96 (0.83–1.12)	1.00 (0.99, 1.02)	1.00 (0.99–1.02)
Normal weight	1.0	1.0	1.0	1.0
Overweight	1.27 (0.83–1.93)	1.31 (0.87–1.96)	1.07 (0.82, 1.41)	1.06 (0.80–1.40)
Obese	0.73 (0.41–1.27)	0.76 (0.42–1.38)	0.91 (0.75, 1.09)	0.88 (0.71–1.09)
Overweight or obese	0.98 (0.77–1.24)	1.03 (0.80–1.34)	1.01 (0.80, 1.28)	1.00 (0.77–1.28)
Non-allergic rhinitis
BMI (continuous)4	0.99 (0.81–1.21)	0.99 (0.80–1.23)	1.02 (1.01, 1.04)**	1.02 (1.01–1.03)**
Normal weight	1.0	1.0	1.0	1.0
Overweight	0.69 (0.37–1.26)	0.70 (0.38–1.31)	1.42 (1.04, 1.93)*	1.44 (1.05–1.97)*
Obese	1.16 (0.73–1.85)	1.18 (0.75–1.86)	1.27 (0.99, 1.63)	1.19 (0.95–1.50)
Overweight or obese	0.88 (0.63–1.22)	0.89 (0.63–1.24)	1.43 (1.06, 1.93)*	1.38 (1.05–1.82)*

¹Model 1 adjusted for age, gender, race, annual household income, serum cotinine and asthma status.

²Model 2 additionally adjusted for CRP in children, and for CRP and apnea symptoms in adults³.

⁴BMI in kg/m² for adults and z-score for children.

^*P<0.05;

^**P<0.01

Table 3 shows the results of multivariate analysis of central obesity and rhinitis. Among children, central obesity was significantly associated with 65% decreased odds of allergic rhinitis, and each 1 z-score increment in WC was significantly associated with 16% lower odds of allergic rhinitis (Model 1). Central obesity remained significantly associated with lower odds of allergic rhinitis after additional adjustment for CRP level (Model 2). Neither central obesity nor WC was significantly associated with non-allergic rhinitis in children. Among adults, central obesity was associated with 61% increased odds of non-allergic rhinitis, and each 1-centimeter increment in WC was significantly associated with 18% increased odds of non-allergic rhinitis (Model 1). Similar results were obtained after additional adjustment for CRP level and apnea symptoms (Model 2).

Table 3.

Multivariate analysis of waist circumference (WC) or central obesity and allergic rhinitis

	Children (6–<18 years)		Adults (≥ 18 years)
	Model 11	Model 22	Model 11	Model 23
	Odds Ratio (95% confidence interval)
Allergic rhinitis
WC (continuous)4	0.84 (0.73–0.98)*	0.86 (0.73, 1.00)	1.02 (0.92–1.12)	1.01 (0.90, 1.14)
Normal weight	1.0	1.0	1.0	1.0
Centrally obese	0.35 (0.19–0.64)**	0.36 (0.19, 0.68)**	1.07 (0.75–1.52)	1.03 (0.69, 1.53)
Non-allergic rhinitis
WC (continuous)4	1.06 (0.92–1.22)	1.07 (0.94, 1.23)	1.18 (1.06–1.31)**	1.15 (1.04, 1.26)**
Normal weight	1.0	1.0	1.0	1.0
Centrally obese	1.40 (0.59–3.34)	1.43 (0.60, 3.37)	1.61 (1.20–2.16)**	1.46 (1.14, 1.89)**

¹Model 1 adjusted for age, gender, race, annual household income, serum cotinine and asthma status.

²Model 2 in children additionally adjusted for CRP and ³in adults for CRP and apnea symptoms.

⁴WC in cm for adults and z-score for children.

^*P<0.05;

^**P<0.01

Figure shows estimates of the prevalence of allergic rhinitis and non-allergic rhinitis in children and adults, by overweight, obesity, and central obesity. Among adults, those with central obesity had a higher prevalence of non-allergic rhinitis than those without central obesity. There was no significant difference in allergic rhinitis or non-allergic rhinitis by overweight or obesity. Among children, the prevalence of allergic rhinitis was lower in those with central obesity than in those without central obesity.

Figure.

Adiposity and prevalence of allergic and non-allergic rhinitis in children (left panel) and adults (right panel)

Because gender has been shown to modify the estimated effect of obesity on asthma, we repeated the analysis in children and adults after stratification by gender (Table 4). In children, this stratified analysis yielded similar results to those of the non-stratified analysis, with an inverse association between central obesity and allergic rhinitis in both boys and girls. Although there was an inverse association between overweight and non-allergic rhinitis in boys, this was not conclusive, as we found no significant association between obesity or central obesity and allergic rhinitis in boys. In adults, overweight, overweight or obesity and central obesity were all significantly associated with non-allergic rhinitis in males but not in females.

Table 4.

Multivariate analysis of overweight, obesity, or central obesity and allergic rhinitis in children and adults, stratified by gender

	Children (6–17 years)		Adults (≥ 18 years)
	Male (n=1173)	Female (n=1185)	Male (n=2360)	Female (n=2546)
	Odds Ratio (95% confidence interval)
Allergic rhinitis
Overweight1	1.26 (0.79, 2.01)	1.56 (0.80, 3.06)	1.50 (0.99, 2.27)	0.82 (0.54, 1.22)
Obese1	0.70 (0.33, 1.49)	0.82 (0.42, 1.59)	0.81 (0.58, 1.14)	0.96 (0.68, 1.35)
Overweight or Obese1	0.92 (0.58, 1.47)	1.22 (0.78, 1.90)	1.29 (0.89, 1.86)	0.85 (0.58, 1.25)
Centrally obese2	0.39 (0.18, 0.86)*	0.32 (0.15, 0.69)**	0.99 (0.53, 1.86)	1.08 (0.58, 2.00)
Non-allergic rhinitis
Overweight1	0.11 (0.03, 0.44)**	1.16 (0.49, 2.73)	1.70 (1.13, 2.55)*	1.34 (0.88, 2.05)
Obese1	1.47 (0.66, 3.24)	1.06 (0.46, 2.46)	1.26 (0.88, 1.79)	1.18 (0.86, 1.61)
Overweight or Obese1	0.65 (0.34, 0.26)	1.10 (0.58, 2.09)	1.64 (1.06, 2.52)*	1.32 (0.95, 1.83)
Centrally obese2	0.93 (0.34, 2.56)	1.74 (0.64, 4.73)	1.98 (1.31, 2.98)**	1.20 (0.72, 2.00)

All models adjusted for age, race, annual household income, serum cotinine, asthma status, and CRP. Models in adults additionally adjusted for apnea symptoms.

¹Comparing to those with normal weight (defined by BMI in kg/m² for adults and BMI z-score for children)

²Comparing to those without central obesity (define by WC in cm for adults and WC z-score for children)

^*P<0.05;

^**P<0.01

DISCUSSION

To our knowledge, this is the first study to examine the relation between overweight or obesity (both “general obesity” defined by BMI, and central obesity, defined by WC) and rhinitis (both allergic and non-allergic) in a large sample of children and adults. Among adults, we report that overweight or obesity and central obesity are both associated with increased odds of non-allergic rhinitis in males. Among children, we found that central obesity (but not “general obesity”) is associated with reduced odds of allergic rhinitis in both males and females.

Most studies reporting a correlation between obesity and allergic rhinitis have been conducted in adolescents^{20, 22, 23, 27}, with some demonstrating increased prevalence of allergic rhinitis in obese boys^{16, 27} and others in obese girls^{20, 23, 28}. In two of these studies, a higher BMI was a significant predictor of allergic rhinitis²⁰ and atopy (assessed by STR to allergens) in girls but not in boys²³. On the other hand, two separate studies showed that cockroach sensitization was slightly more common in obese or overweight male adolescents¹⁶, and that atopy was more common among overweight boys²⁷. Several mechanisms have been proposed that could explain the effect of adipose tissue on the immune system, including chronic inflammatory changes found in obese individuals, with higher levels of interleukin (IL)-6, leptin, and TNF-α, which can downregulate regulatory T cells (T_Reg). Conversely, increased adipose tissue also leads to reduced adiponectin levels, which in turn downregulates the secretion of IL10 and decreases the regulatory effect of T_Regs²⁹. Several studies, particularly in adults, have reported that subjects with obesity-related asthma have a predominantly neutrophilic airway inflammation³⁰, which could explain the higher prevalence of non-allergic rhinitis in our analysis. Obese mice exposed to diesel exhaust particles show airway inflammation with increased neutrophilia and decreased eosinophilia³¹.

Central obesity may be superior to general obesity as an indicator of risk for obesity-related complications^{12, 15}. Similar to previous findings of an association between central obesity and non-atopic asthma¹⁰, we found that overweight or obesity, as well as central obesity, were positively associated with non-allergic rhinitis in adults. Moreover, we show that the magnitude of the association between central obesity and non-allergic rhinitis in adults is slightly greater than that for the association between overweight or obesity and non-allergic rhinitis in this age group. Distinct types of adiposity may be associated with different obesity complications. For example, subjects with “normal weight central obesity” (normal BMI but high waist-to-hip ratio) may have the highest risk for coronary artery disease³². Visceral adipose tissue may be more closely related to metabolic abnormalities and may exhibit a more pro-inflammatory phenotype (Th₁ and TH₁₇ cells) in obese individuals compared to individuals of normal weight³³. Of interest, our findings for non-allergic rhinitis in adults were stronger in males than in females, which could be due to hormonal effects on immune responses or rhinitis, or to misdiagnosis of symptoms of sleep apnea as rhinitis in male adults. Obstructive sleep apnea (OSA) is a common comorbidity of obesity³⁴ or allergic rhinitis^{35, 36}, and uncontrolled rhinitis can lead to OSA³⁷. Although we lacked data from sleep studies in study participants, adjustment for symptoms of OSA (not considered in previous studies) did not change the association between obesity and non-allergic rhinitis in adults.

In contrast to our findings for non-allergic rhinitis in adults, central obesity was associated with reduced odds of allergic rhinitis in children. Our results are consistent with those of two prior studies in the U.S.^{15, 18}, which demonstrated a negative association between obesity and allergic rhinitis in children. This is consistent with a Th₁-predominant phenotype in another disease related to obesity (“obese asthma”).

Our study has considerable strengths, including a large sample size (both children and adults) representative of the U.S. population, and objective measurements of atopic sensitization in both children and adults. We also accounted for several potential confounders, including cigarette smoking (measured by serum cotinine), low grade systematic inflammation (measured by CRP) and sleep apnea symptoms in our multivariate analysis. Moreover, we used not only BMI but also WC as adiposity measures. We also acknowledge several limitations of our findings. Firstly, we cannot determine a temporal relationship between obesity and rhinitis in this cross-sectional study. Secondly, we lack data on factors that may influence allergic sensitization or allergic rhinitis, such as hormonal levels at puberty. However, we found similar results in an analysis stratified by both age (6–11 years vs. 12–17 years) and gender in children (data not shown). Thirdly, while NHANES tested for the 15 most common allergens found in the general U.S. population, some participants may have been sensitized to other allergens that were not included. Finally, information on gastro-esophageal reflux, which could contribute to rhinitis symptoms, was not available in NHANES.

In conclusion, obesity was associated with non-allergic rhinitis in adults, particularly in males. In children, central obesity was negatively or inversely associated with allergic rhinitis, regardless of gender. Future longitudinal studies in obese individuals should aim to determine the temporal relationship between obesity and the development of both allergic and non-allergic rhinitis.

Clinical Implications.

Obesity is associated with increased odds of non-allergic rhinitis in adults, particularly in males. Allergy skin testing or measurement of IgE to common allergens should be considered in obese adults with rhinitis symptoms.

Abbreviations

STR

skin test reactivity

BMI

body mass index

WC

waist circumference

NHANES

National Health and Nutrition Examination Survey

CRP

C-reactive protein

OSA

obstructive sleep apnea