Published ahead of print on June 19, 2008, doi:10.1164/rccm.200802-301OC
© 2008 American Thoracic Society doi: 10.1164/rccm.200802-301OC
The Association between Obesity and AsthmaInteractions between Systemic and Airway Inflammation1 Respiratory Research Unit, Department of Medical and Surgical Sciences, Dunedin School of Medicine; 2 Department of Medical Microbiology, Division of Health Sciences; 3 Department of Medical and Surgical Sciences, Dunedin School of Medicine; and 4 Department of Preventive and Social Medicine, Dunedin School of Medicine, University of Otago, Dunedin, New Zealand Correspondence and requests for reprints should be addressed to Professor D. Robin Taylor, M.D., F.R.C.P.C., Dunedin School of Medicine, University of Otago, P.O. Box 913, Dunedin, New Zealand. E-mail: robin.taylor{at}stonebow.otago.ac.nz
Rationale: Both obesity and asthma are common conditions, and both are characterized by the presence of inflammation. Animal studies suggest that the development of airway hyperresponsiveness with antigen challenge is exaggerated with obesity. However, clear evidence for either an additive or a synergistic pathologic interaction between obesity and asthma is lacking in humans. Objectives: To identify whether an interaction between systemic and local inflammation occurs in obese subjects with asthma in a controlled observational study. Methods: We studied 79 women: obese patients with asthma (n = 20), normal-weight patients with asthma (n = 19), obese patients without asthma (n = 20), and normal-weight patients without asthma (n = 20). After corticosteroid withdrawal, between-group differences in spirometric values, lung volumes, exhaled nitric oxide, induced sputum cell counts, and biomarkers of inflammation in sputum supernatant and blood were measured, and interactions explored. Measurements and Main Results: Markers of systemic inflammation were increased with obesity, and Th2 cytokines were increased with asthma, but no important interactions were identified. Obesity adversely affected lung function with increases in functional residual capacity and residual volume in obese but not normal-weight patients with asthma, with a significant obesity by asthma interaction. Conclusions: The link between obesity and asthma is unlikely to be explained by enhancement of the "classical" forms of airway inflammation resulting from the systemic inflammatory effects of obesity itself. Other mechanisms, possibly related to innate immunity, may explain the relationship between obesity and asthma.
Key Words: asthma • obesity • inflammation • interaction
Both asthma and obesity are common clinical problems, often coexisting in the same patient, and epidemiologic studies have confirmed an association between the two conditions (1). Obese individuals have an increased prevalence of asthma and its persistence, as well as the need for increased treatment, also appears to be greater with obesity (2, 3). A number of hypotheses have been proposed to explain these relationships. First, the dietary lifestyle that predisposes to obesity may be important in the etiology of asthma (4). Second, the mechanical effect of obesity on the chest and abdominal wall affects respiratory function, and may increase the work of breathing, particularly with exercise. Alternatively, smaller tidal volumes and increased breathing frequency may cause an increase in airway smooth muscle shortening with enhanced airway hyperresponsiveness (AHR) (5). There is some evidence that obesity predisposes to AHR (6), although this is not a consistent finding (7). Recently, Sutherland and colleagues have reported that dynamic hyperinflation after bronchoprovocation is greater in obese patients with asthma, suggesting that the effects of bronchospasm differ between obese and normal-weight individuals (8). Thus, respiratory symptoms attributable to obesity may be interpreted as "asthma" by a patient or his or her clinician.
A further possible mechanism concerns immunoregulation and inflammation. Adipose tissue, far from being biologically inert, secretes important regulatory adipokines (9). Some have proinflammatory effects, leading to an association between obesity and chronic mild systemic inflammation. For example, C-reactive protein (CRP) (10), IL-6 (11), and tumor necrosis factor (TNF)-
Proinflammatory cytokines are also central to the pathogenesis of asthma (15). Although they are believed to act locally in the airways, there is some evidence that systemic inflammation also occurs in asthma. For example, in the NHANES (National Health and Nutrition Examination Study), subjects with current asthma had elevated CRP levels (16). Of more importance to the present investigation, some cytokines associated with obesity may play a modifying role in asthma. For example, IL-6 has the potential to modulate the T-helper 2 (Th2) immunity (17); TNF- Taken together, these data raise the possibility that the relationship between obesity and asthma may be mediated, at least in part, by overlapping or interacting pathogenic mechanisms. We hypothesized that the low-grade systemic inflammation present in obesity would augment the inflammation of asthma (a synergistic effect) or, alternatively, that the inflammation of obesity might affect the airways independently (an additive effect), perhaps resulting in a separate or mixed inflammatory cell phenotype. To test these hypotheses, we conducted an observational controlled study to compare systemic and airway inflammatory profiles in obese and normal-weight individuals with and without asthma.
See the online supplement for additional information on methods. Study participants were female, aged between 18 and 50 years old, with approximately 20 subjects in each of the following four groups: obese asthmatic, obese nonasthmatic, normal-weight asthmatic, and normal-weight nonasthmatic. Obesity was defined as a body mass index (BMI) greater than 30 kg/m2 and normal weight as less than 25 kg/m2. Asthma was defined as typical symptoms with either an increase in FEV1 of 12% or more after bronchodilator or significant AHR, defined as a provocative dose of inhaled methacholine causing a 20% reduction in FEV1 (PD20) of less than 8 µmol. We sought to minimize any potential confounders. Only premenopausal women were recruited because the association between obesity and asthma appears to be stronger in women (22, 23) and may be different in postmenopausal women (24). Exclusion criteria included the following: current smoker or ex-smoker with a greater than 10 pack-year smoking history, comorbidity that could potentially increase systemic inflammatory markers, medication use (including recent oral steroids) that could alter systemic inflammation (e.g., statins or hormonal contraceptives), the presence of any respiratory symptoms, and atopy or AHR in the nonasthmatic subjects. After a screening visit, subjects with asthma taking inhaled corticosteroids had them withdrawn until loss of control or 4 weeks, whichever came sooner, at which time they undertook the study procedures. Loss of control was defined using a priori criteria. Those who did not lose control were only included if they demonstrated AHR to methacholine or bronchodilator reversibility. Those who did not lose control, together with the steroid-naive subjects with asthma and all nonasthmatics, were assessed during the first week of the next menstrual cycle.
Study Procedures
Induced sputum was processed using a standardized procedure (30). Sputum supernatant was frozen and stored at –80°C until analyzed. Two cytospins were performed and cell counts of 1% or greater for eosinophils and 61% or greater for neutrophils were considered to indicate eosinophilic and neutrophilic asthma, respectively (31).
CRP was measured in plasma using a highly sensitive assay (Roche Diagnostics, Mannheim, Germany). Multiplex bead array assays were used to analyze for the following cytokines in plasma and sputum supernatant (32): interleukins 1β, 2, 4, 5, 6, 8, 10, and 13; IFN- The study was approved by the Otago Regional Ethics Committee, and subjects gave written, informed consent.
Statistical Analysis
A total of 79 subjects (2 were Maori) completed the study procedures. Their demographic and anthropometric data are shown in Table 1. There were 39 subjects with asthma, of whom 6 had mild intermittent (3 obese, 3 normal weight), 23 had mild persistent (11 obese, 12 normal weight), 5 had moderate (2 obese, 3 normal weight), and 5 had severe (4 obese, 1 normal weight) asthma according to the GINA (Global Initiative for Asthma) classification. Thirty-five subjects (90%) were atopic. Of the 33 subjects with asthma taking regular inhaled steroids, 28 experienced loss of control after steroid withdrawal within 3 to 30 days. For safety reasons, these subjects were seen immediately rather than with regard to the phase of their menstrual cycle. Thus, 16 of 28 were seen in the proliferative phase and the remaining 12 of 28 were seen in the luteal phase or at midcycle. Of the steroid-naive subjects with asthma (n = 6), the subjects with asthma who did not lose control (n = 5), and the subjects without asthma (n = 40), 49 of 51 were seen in the proliferative phase, based on a combination of biochemical measurements and self-reporting of cycle day. Two women were diagnosed post hoc as being postmenopausal on biochemical measurements, but separate analyses excluding their data did not influence the overall results.
There were significant differences between the obese and normal-weight subjects for blood pressure, fasting glucose, and fasting high-density lipoprotein, low-density lipoprotein, and triglyceride, but no obesity–asthma interaction was observed for any of these variables (data not shown).
Lung Function and FENO
Sputum Cell Counts Satisfactory samples of induced sputum were obtained in 67 of the 79 subjects (85%). Satisfactory slides for sputum cell counts were made in 62 subjects, and the results are reported in Table 3. The asthmatic groups had significantly higher eosinophil counts than their nonasthmatic counterparts, but there was no significant difference between obese and normal-weight subjects with asthma. There were no significant differences in neutrophil counts between asthmatic and nonasthmatic groups. No obesity–asthma interactions were identified.
There were no significant differences between obese and normal-weight subjects with asthma in the proportions of patients with sputum eosinophils of 1.0% or more (86.7 and 93.8%, respectively) or sputum with mixed cellularity (6.7 and 6.3%, respectively). None of the subjects with asthma had a purely neutrophilic phenotype (neutrophil count  61%). The majority of subjects without asthma (29/31, 93.5%) were classified as paucigranulocytic.
Inflammatory Biomarkers
Sputum supernatant levels of IL-1β, IL-5, IL-6, and IL-8 were all significantly higher in subjects with asthma compared with nonasthmatic subjects. All of these relationships remained significant after adjusting for sputum total cell count. There were no obesity-by-asthma interactions for any of the sputum supernatant biomarkers. Post hoc analyses based on results for sputum supernatant cytokines were performed to assess the numbers needed to make the interaction between asthma and obesity statistically significant at the 5% level. The numbers ranged from 530 per group for IL-4 to 1.04 x 106 for IL-6.
As far as we are aware, this is the first controlled observational study designed to explore the possibility that airway inflammation in asthma might be altered or augmented in patients who are obese. We sought to obtain proof-of-concept evidence that might help to explain the relationship between obesity and asthma observed in both epidemiologic and clinical studies. We failed to identify a significant obesity–asthma interaction, both in regard to biomarkers of systemic and airway inflammation, and in the airway inflammatory cell phenotype. As expected, biomarkers of systemic inflammation were increased in obese versus normal-weight individuals (e.g., IL-6 and CRP). Although the highest levels occurred in the obese asthmatic group, the differences compared with nonobese subjects with asthma were not statistically significant. Likewise, as expected, biomarkers in sputum supernatant (IL-1β, IL-5, IL-6, IL-8) were significantly increased in subjects with asthma compared with subjects without asthma. Although the values, notably for IL-5 and IL-6, were highest in the obese asthmatic group, the differences compared with nonobese asthmatic group were not significant. For CRP, the effects of obesity and asthma were additive. Again, there was no clear evidence of an interaction between obesity-related adipokines and airway inflammation, at least for the range of predominantly Th2-related biomarkers which were measured in the present study. Higher levels of IL-6 have previously been reported in obese subjects with asthma in comparison with nonobese subjects with asthma (34). However, even with greater numbers, it seems unlikely that we would have reported a different overall outcome. Whether the effects of one cytokine in isolation might be important is difficult to judge, although it is theoretically possible that increases in IL-6 in particular might result in "priming" of susceptibility to proinflammatory triggers in obese subjects with asthma, leading to increased asthma symptoms. Similarly, the increase in plasma IL-4 and IL-6 in obese subjects might enhance the response to an allergic or other stimulus, even in the absence of an increase in sputum eosinophils, but this is speculative. Shore and colleagues have previously explored these mechanisms in animal models, providing evidence that, in obese mice, there is an increase in leptin-mediated AHR (35) as well as enhancement of the innate immune response to triggers such as ozone (14). More recently, Johnston and coworkers have confirmed that ovalbumin sensitization and subsequent challenge result in increased AHR in obese compared with nonobese mice, but independently of Th2-mediated pathways (36). Thus, if an inflammatory mechanism does indeed explain the obesity–asthma link, it may be operating beyond the scope of those pathways that were explored in the present study, and our results may therefore be falsely negative. Perhaps if the measured biomarkers had been more closely allied to innate immunity (e.g., soluble CD-14 and endotoxin, or the expression of innate immune receptors) a different picture might have emerged. The inflammatory phenotypes, as identified by sputum cell counts, were similar in obese and normal-weight asthmatic groups. Two alternative outcomes had been considered possible. First, we speculated that the intensity of airway inflammation might have been relatively greater in the obese asthmatic group. This would have suggested that obesity has a synergistic effect on existing airway inflammation. There was no evidence of this. Our results are consistent with those obtained from a larger population–based study, in which, although obesity was associated with asthma in women, there was no clear relationship to eosinophilic airway inflammation as judged by FENO measurements (37). Second, we considered whether the inflammatory phenotype in obese subjects with asthma might be characterized by a greater predominance of noneosinophilic inflammation and associated Th1 cytokines. If the number of subjects who demonstrated either a sputum neutrophil count of 61% or more or mixed cellularity had been greater in the obese groups, we would have concluded that obesity conferred an additive inflammatory effect. This was not found, although it was theoretically possible given that leptin may indirectly enhance neutrophilic airway inflammation (38). Neutrophilic inflammation is associated with resistance to inhaled corticosteroid therapy (39), and if present, this might potentially explain why obese patients with asthma require increased levels of antiinflammatory treatment (40). In our study, the obese asthmatic group had the highest leptin levels (see Table 4). This appeared to be associated with a higher total cell count in sputum compared with nonobese individuals, but these differences were not significant. Overall, our results are consistent with those reported in a much larger study comprising 727 patients with asthma, in which no relationship between body mass index and induced sputum differential cell counts was found, even after controlling for inhaled corticosteroid use (41). Our study design had both weaknesses and strengths. First, we were able to obtain adequate sputum specimens in only 78% of subjects. Second, we were aware that the study size might be too small to identify significant between-group differences. However, post hoc analyses of the sputum cytokine data showed that very large numbers would have been needed in each group to demonstrate a significant interaction between obesity and asthma (e.g., n = 530 for IL-4). In the absence of previous data upon which to base a study size estimate, we included two control groups, thus permitting obesity–asthma interactions to be more rigorously assessed. We also applied very strict inclusion criteria, notably that only female patients should be studied and that inhaled steroid therapy should be withdrawn. These criteria aimed to minimize any possible confounding effect by factors such as female sex hormones and antiinflammatory therapy. In addition, in the nonasthmatic groups, those with atopy, AHR, or increased FENO were excluded. This meant that over 200 individuals were screened to recruit the 79 participants. In conclusion, in this proof-of-concept study, we found no evidence that the relationship between obesity and asthma in humans is mediated via enhancement of "classical" forms of airway inflammation resulting from the systemic inflammatory effects of obesity itself. Even though they were shown to coexist, systemic and airway inflammation appear to operate independently of one another. Other mechanisms, possibly related to innate immunity, may explain the epidemiologic relationship between obesity and asthma, but in individual patients with asthma, it seems more likely that the impact of obesity is mediated via dynamic changes in lung function that are exaggerated (8). This might explain the apparent need for increased inhaled antiinflammatory therapy in obese patients with asthma and the difficulties that are encountered in achieving good asthma control in such individuals.
The authors thank all the participants who volunteered for this study.
Supported by the Otago Medical Research Foundation and the Dean of the Dunedin School of Medicine Bequest Fund. T.J.T.S. was the recipient of the Francis Cotter Research Fellowship of the Dunedin School of Medicine. This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org Originally Published in Press as DOI: 10.1164/rccm.200802-301OC on June 19, 2008 Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form February 20, 2008; accepted in final form June 16, 2008
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
AT A GLANCE COMMENTARY
(
; monocyte chemotactic protein-1; and TNF-
