- Email: [email protected]
Effect modification by race–ethnicity of environmental tobacco smoke exposure on airway inflammation in US children
Ann Allergy Asthma Immunol 115 (2015) 477e480
Contents lists available at ScienceDirect
Effect modification by raceeethnicity of environmental tobacco smoke exposure on airway inflammation in US children Alem Mehari, MD *, y; Ngozi Enwerem, MD *; Yewande Odeyemi, MD *, y; and Richard. F. Gillum, MD * * Department y
of Internal Medicine, Howard University College of Medicine, Washington, DC Division of Pulmonary Diseases, Howard University College of Medicine, Washington, DC
A R T I C L E
I N F O
Article history: Received for publication July 24, 2015. Received in revised form September 5, 2015. Accepted for publication September 14, 2015.
A B S T R A C T
Background: Asthma mortality and morbidity are higher in black than in white children. Fractional exhaled nitric oxide (FeNO) is a noninvasive biomarker of eosinophilic airway inflammation. Identification of differences in the effect of environmental tobacco smoke (ETS) on airway inflammation by race and ethnicity from a large sample is needed. Objective: To estimate a racial difference in association with ETS and FeNO. Methods: Data from the 2007 to 2012 National Health and Nutrition Examination Survey were analyzed to compare associations of ETS and FeNO levels in US black and other children. No ETS exposure was defined as a serum cotinine level lower than 0.05 ng/mL and ETS exposure was defined as a serum cotinine level of at least 0.05 ng/mL. FeNO was measured using a device that relies on an electrochemical sensor. Analyses took the complex survey design into account. Results: The analytic sample was formed by 5,473 participants (6e11 years old, n ¼ 2,385; 12e19 years old, n ¼ 3,088) with complete data on demographics, serum cotinine levels, and 2 reproducible FeNO measurements. In weighted linear regression analyses at 6 to 11 years, the interaction term for ETS and black race was not significant (P ¼ .15). At 12 to 19 years, the interaction term was significant (P ¼ .03) in an analysis of all racial groups. In race-specific models, the coefficient for ETS exposure in blacks was 0.033 and that in others was 0.175, ie, ETS exposure was associated with a greater decrease in FeNO in non-blacks than in blacks. Conclusion: There was no evidence at 6 to 11 years of age for an effect modification by race of the association between ETS and FeNO. At 12 to 19 years, the data suggested an effect modification. Ó 2015 American College of Allergy, Asthma & Immunology. Published by Elsevier Inc. All rights reserved.
Introduction Asthma morbidity and mortality are higher in black than in white children.1 Asthma diagnosis and treatment are challenging in children because symptoms are often reported by a parent and are often minimized. Inhaled corticosteroids, which are the mainstay of therapy in asthma, are underused because of concerns about longterm side effects.2 Measurement of fractional exhaled nitric oxide (FeNO) recently has gained favor in identifying and monitoring eosinophilic airway inflammation in asthma.3 FeNO testing also is useful for monitoring adherence to asthma medications, identifying patients with persistent allergen exposure, and guiding the use and dosing of inhaled corticosteroid therapy.4 Thus, FeNO
Reprints: Alem Mehari, MD, Pulmonary and Critical Care, Howard University, 2041 Georgia Avenue NW, Washington, DC 20060; E-mail: [email protected]. Disclosures: Authors have nothing to disclose. Funding Sources: None.
provides health care providers with valuable information for assessment of disease severity and monitoring control. In atopic asthma, inducible NO synthetase (iNOS) is upregulated in respiratory epithelium and produces increased NO concentrations in exhaled air.5 Many factors can affect FeNO. FeNO increases with age in children and can be higher in the elderly than in younger adults.3 Levels can be higher in men than in women and in taller than in shorter adults.6 Smoking decreases FeNO7 and consumption of NO-rich foods increases levels.8 In addition, ozone air pollution, upper respiratory infections,9 and allergen exposure increase FeNO, and exercise decreases FeNO.3 However, major effects have been shown only for smoking, viral infections, and recent intake of NO-rich foods. Although young children do not use tobacco, they are frequently exposed to environmental tobacco smoke (ETS), most often in the home.10 In 2007 through 2008, an estimated 54% of children 3 to 11 years old and 47% of nonsmoking adolescents 12 to 19 years old were exposed to ETS.11 In 2007 through 2010, 54% of US children with asthma had ETS exposure.12 Studies have associated exposure
http://dx.doi.org/10.1016/j.anai.2015.09.010 1081-1206/Ó 2015 American College of Allergy, Asthma & Immunology. Published by Elsevier Inc. All rights reserved.
478
A. Mehari et al. / Ann Allergy Asthma Immunol 115 (2015) 477e480
Table 1 Coefficients from multiple regression models of natural logarithmic fractional excretion of exhaled nitric oxide for 2 age groups, National Health and Nutrition Examination Survey, 2007 through 2012 Variable
Multiple linear regression: 6e11 y Regression coefficienta (95% CI)
Age per yr Sex Female Male Standing height, cm Race or ethnicity Non-Hispanic black Non-Hispanic white Hispanic Hay fever in past 12 mo Present Absent Wheeze in past 12 mo Present Absent Self-reported asthma Present Absent ETS exposure Yes No ETS black race
0.0884 (0.0656e0.1112) 0.03373 (0.0996 to 0.0322) Ref 0.0153 (0.0066e0.0204) 0.1873 (0.0694e0.3054) Ref 0.1935 (0.1183e0.2686)
Multiple linear regression: 12e19 y P value
Regression coefficienta (95% CI)
P value
<.001
0.0104 (0.0034 to 0.0242)
.14
.31
0.2346 (0.3199 to 0.1492) Ref 0.0099 (0.0047e0.0151)
<.001
<.001 <.001 <.001
0.3119 (0.1887e0.4351) Ref 0.1701 (0.0654e0.2750)
<.001 <.001 .002
0.2417 (0.1058e0.377) Ref
.001
0.2289 (0.1029e0.35510) Ref
.001
0.1499 (0.0001 to 0.2998) Ref
.05
0.0569 (0.0930 to 0.2069) Ref
.45
0.1627 (0.2424 to 0.0829) Ref
<.001
0.1849 (0.2333 to 0.1365) Ref
<.001
0.0930 (0.1726 to 0.0132) Ref 0.1314 (0.0484 to 0.3110)
.02
0.1915 (0.2845 to 0.0985) Ref 0.1344 (0.0131e0.2556)
<.001
.15
.03
Abbreviations: CI, confidence interval; ETS, environmental tobacco smoke; Ref, reference group. a Coefficients are expressed as change in natural logarithmic fractional excretion of exhaled nitric oxide per 1-U change in the independent variable; means are expressed as parts per billion.
to ETS with asthma occurrence and asthma exacerbations in children.13 Previous studies in mostly small, selected volunteer or clinical samples have associated acute and chronic exposures to ETS with lower FeNO.14,15 Whether ETS exposure lowers FeNO values to the same degree in children, a degree that impairs its utility in asthma management overall or in high-risk subgroups, remains controversial. Whether any effect of ETS differs in black vs white children is unreported. In a large US national sample, this study tested these hypotheses: (1) ETS is associated with lower FeNO in children and youth and (2) this association varies by race or ethnicity. Methods The National Health and Nutrition Examination Survey (NHANES) is designed to assess the health and nutritional status of adults and children in the United States.16,17 Response rates for 2year study cycles exceed 75%. The National Center for Health Statistics Research ethics review board approved all protocols. All participants and their agents provided written informed consent. Trained interviewers administered questionnaires in the home on family and sample person characteristics using a ComputerAssisted Personal Interviewing (CAPI) system. The household smokers section of the family questionnaire provides information on exposure to secondhand smoke in the home setting. The medical conditions section provides self-reported personal interview data on diagnosed asthma. The respiratory disease section provides interview data for symptoms of wheezing. Reported hay fever episodes in the past 12 months were considered atopy. Participants who were nonsmokers were classified as having exposure to ETS if serum cotinine levels were at least 0.05 ng/L, the limit of detection. Serum cotinine, a metabolite of nicotine, was analyzed using an isotope dilution liquid chromatographic tandem mass spectrometry method. Participants with evidence of active smoking, as indicated by levels of at least 10 ng/L, were excluded.11 Children 6 to 11 years old were assumed to be nonsmokers if their serum cotinine concentration was no higher than 10 ng/mL. Youths 12 to 19 years old were considered nonsmokers if their
serum cotinine concentration was no higher than 10 ng/mL and they did not report smoking within the preceding 30 days or use of any nicotine-containing product within the preceding 5 days at their physical examination. The primary end point of the study was FeNO. FeNO was measured using the Aerocrine NIOX MINO, a portable hand-held NO analyzer (Aerocrine AB, Solna, Sweden). This device relies on an electrochemical sensor to detect exhaled NO levels and provides measurements from 5 to 300 ppb in whole numbers. The Respiratory Health ENO Procedures Manual18 provides a detailed description of the examination procedure. The NHANES protocol required 2 FeNO measurements that were reproducible within the 10 attempts allowed. This analysis uses the average of 2 reproducible measurements. The correlation between the first 2 tests was good (R2 ¼ 0.75, P < .001) and the geometric mean difference between the first and second tests was 1.3 ppb. FeNO distributions, normal values, and basic findings using this method have been reported.16,17 Statistical Analysis Distributions of FeNO were approximately logarithmically normal. This analysis compared the means of logarithmically transformed values; exponentially, the means of logarithmically transformed values yielded geometric mean values with 95% confidence intervals. Multivariable linear regression modeling using STATA 11 (STATA Corp, College Station, Texas) assessed ETS exposure by history or by cotinine level as a predictor of logarithmically transformed FeNO and assessed effect modification by race. Models controlled for age, sex, Hispanic ethnicity, asthma diagnosis, histories of wheezing and hay fever and smoking (12e19 years old only), height, and body mass index. Data analyses using STATA 11 accounted for the complex survey design, differential probability of sample selection, nonresponse, and sample non-coverage. Results Of 30,442 participants interviewed, 29,353 underwent examination; of those 7,459 were 6 to 19 years old; of these, 6,318 had 2
A. Mehari et al. / Ann Allergy Asthma Immunol 115 (2015) 477e480
reproducible FeNO measurements. Those 5,473 who also had complete data on demographics, household ETS, and serum cotinine formed the analytic sample. Of these, 2,385 were 6 to 11 and 3,088 were 12 to 19 years old. Examination of mean natural logarithm (ln) FeNO by exposure to ETS, age group, sex, and race or ethnicity suggested that ETS was associated with lower FeNO in non-Hispanic (NH) whites but higher FeNO in blacks. The pattern in Mexican Americans was similar to that in NH whites. Patterns in other groups were unstable, likely owing to smaller numbers. In multiple linear regression analysis, there was a positive correlation of ln FeNO with age, nonwhite ethnicity, standing height, and wheeze or hay fever within the past 12 months. FeNO levels increased with increasing age significantly (P < .001) in children 6 to 11 years old and with increasing height (P < .001) in children 6 to 11 and youth 12 to 19 years old. After controlling for all other factors, NH blacks consistently showed higher mean FeNO levels compared with the others (P < .001) in the 2 age groups. Male sex was associated with higher FeNO in adolescents (P < .001) but not in children 6 to 11 years old. In participants 6 to 11 and 12 to 19 years of age, there was a negative correlation of FeNO with ETS exposure, female sex, body mass index, and self-reported history of asthma. In weighted linear regression analyses at 6 to 11 years old, the interaction term for ETS and race was not significant (P ¼ .15; Table 1). At 12 to 19 years old, the interaction term was significant (P ¼ .03) in an analysis of all racial groups (Table 1). In race-specific models, the coefficient for ETS exposure in blacks was 0.033 and that in others was 0.175, ie, ETS exposure was associated with a greater decrease in ln FeNO in non-blacks than in blacks. However, in sensitivity analyses, when other races were excluded, the interaction term no longer attained significance. In expanded regression models to control for additional confounders (asthma, hay fever, self-reported wheezing or undiagnosed asthma, height, body mass index, and steroid use), the interaction term for black and ETS also was not significant.
479
ng/mL in others. Therefore, other unmeasured or unknown factors must account for this finding. Most previous studies in children with asthma or healthy children have associated recent exposure to ETS with lower FeNO levels,17,19,20 but there are some exceptions.22 In multivariate models in the 2007 to 2010 NHANES, ETS exposure was independently and negatively associated with FeNO at 6 to 11 and 12 to 80 years of age.17 A follow-up study found that ETS exposure at 4 years was associated with higher FeNO at 7 years, suggesting ETS might have a chronic effect as a risk factor for future airway inflammation apart from its acute effect to lower FeNO.23 In nonsmokers at least 3 years old, prevalence of exposure to ETS was consistently higher in NH blacks than in NH whites.11 FeNO levels also were higher in NH blacks than in NH whites.16 Examination of effect modification by race was lacking. Of note is that multivariate models with individual characteristics could account for only a small percentage of variation in FeNO (10% at 6e11 years and 16% at 12e80 years).17 As presented in Table 1, a history of hay fever was significantly associated with FeNO in children and youth. The major strength of this study is the availability of data from the NHANES, which is a nationally representative and comprehensive survey drawing from a large and diverse sample of participants. The survey design avoided volunteer bias and enhanced the generalizability of the present results. The authors acknowledge that sampling and non-sampling errors, including measurement error, might occur. In some subgroup analyses, the confidence interval of regression coefficients could not be generated, likely owing to strong geographic clustering of ethnic groups. The present analysis controlled for possible factors known to be related to FeNO; however, other factors not included in this analysis might differentially affect children by race or ethnicity. In conclusion, there was no evidence at 6 to 11 years of age for an effect modification by race of the association between ETS and FeNO. At 12 to 19 years, data suggested an effect modification with greater effect of ETS in non-blacks than in blacks. Replication in longitudinal studies is needed.
Discussion The present study found a pattern indicating that exposure to ETS lowered FeNO in NH whites but had little effect on FeNO in NH blacks. However, this trend for effect modification by race did not attain statistical significance at 6 to 11 years of age and was inconsistently significant at 12 to 19 years when tested in linear regression models controlling for multiple confounders. The possible mechanism by which tobacco smoke lowers FeNO could be that it increases arginase activity and therefore decreases arginine availability for NO production in the airways.19 NO is formed in airway epithelium by the conversion of arginine to citrulline by NOS, principally iNOS. Arginine also is converted by arginase to ornithine, which is a precursor of prolines and polyamines, which leads to collagen and mucus formation and cell proliferation, respectively. Epithelial and smooth muscle cells from smokers have been found to have higher levels of arginase and ornithine decarboxylase (a downstream enzyme of the arginase pathway) than those of nonsmokers.19 Others have suggested negative feedback of NO in smoke on iNOS.20 A polymorphism in the NOS and related genes could modify the effect of smoking.21 It could be asked why in race-specific models for 12- to 19-yearolds the coefficient for ETS exposure in blacks was only 0.033 and that in others was 0.175 (ie, ETS exposure was associated with a greater decrease in ln FeNO in non-blacks than in blacks). One explanation could be that ETS exposure was less in blacks than in others. However, that is not supported by the finding that in 12- to 19-year-old nonsmokers classified as having ETS exposure, the median serum cotinine level was 0.30 ng/mL in blacks and 0.26
References [1] Akinbami LJ, Moorman JE, Bailey C, et al. Trends in asthma prevalence, health care use, and mortality in the United States, 2001e2010. NCHS Data Brief. 2012;94:1e8. [2] Boulet LP, FitzGerald JM, Reddel HK. The revised 2014 GINA strategy report: opportunities for change. Curr Opin Pulm Med. 2015;21:1e7. [3] Bjermer L, Alving K, Diamant Z, et al. Current evidence and future research needs for FeNO measurement in respiratory diseases. Respir Med. 2014;108: 830e841. [4] Dweik RA, Boggs PB, Erzurum SC, et al. An official ATS clinical practice guideline: interpretation of exhaled nitric oxide levels (FENO) for clinical applications. Am J Respir Crit Care Med. 2011;184:602e615. [5] Scott M, Raza A, Karmaus W, et al. Influence of atopy and asthma on exhaled nitric oxide in an unselected birth cohort study. Thorax. 2010;65:258e262. [6] Olin AC, Rosengren A, Thelle DS, Lissner L, Bake B, Torén K. Height, age, and atopy are associated with fraction of exhaled nitric oxide in a large adult general population sample. Chest. 2006;130:1319e1325. [7] Persson MG, Gustafsson LE, Zetterström O, Agrenius V, Ihre E. Single-breath nitric oxide measurements in asthmatic patients and smokers. Lancet. 1994; 343:146e147. [8] Zetterquist W, Pedroletti C, Lundberg JON, Alving K. Salivary contribution to exhaled nitric oxide. Eur Respir J. 1999;13:327e333. [9] Sanders SP, Proud D, Permutt S, Siekierski ES, Yachechko R, Liu MC. Role of nasal nitric oxide in the resolution of experimental rhinovirus infection. J Allergy Clin Immunol. 2004;113:697e702. [10] Marano C, Schober SE, Brody DJ, Zhang C. Secondhand tobacco smoke exposure among children and adolescents: United States, 2003e2006. Pediatrics. 2009;124:1299e1305. [11] Vital signs: nonsmokers’ exposure to secondhand smokedUnited States, 1999e2008. MMWR Morb Mortal Wkly Rep. 2010;59:1141e1146. [12] Quinto KB, Kit BK, Lukacs SL, Akinbami LJ. Environmental tobacco smoke exposure in children aged 3e19 years with and without asthma in the United States, 1999e2010. NCHS Data Brief. 2013;126:1e8. [13] Howrylak JA, Spanier AJ, Huang B, et al. Cotinine in children admitted for asthma and readmission. Pediatrics. 2014;133:e355ee362.
480
A. Mehari et al. / Ann Allergy Asthma Immunol 115 (2015) 477e480
[14] Yates DH, Breen H, Thomas PS. Passive smoke inhalation decreases exhaled nitric oxide in normal subjects. Am J Respir Crit Care Med. 2001;164: 1043e1046. [15] Hoyt JC, Robbins RA, Habib M, et al. Cigarette smoke decreases inducible nitric oxide synthase in lung epithelial cells. Exp Lung Res. 2003;29:17e28. [16] Brody DJ, Zhang X, Kit BK, Dillon CF. Reference values and factors associated with exhaled nitric oxide: U.S. youth and adults. Respir Med. 2013;107: 1682e1691. [17] See KC, Christiani DC. Normal values and thresholds for the clinical interpretation of exhaled nitric oxide levels in the US general population: results from the National Health and Nutrition Examination Survey 2007e2010. Chest. 2013;143:107e116. [18] Respiratory Health ENO Procedures Manual. http://www.cdc.gov/nchs/data/ nhanes/nhanes_07_08/ENO.pdf Accessed June 31, 2015.
[19] Bergeron C, Boulet L-P, Page N, et al. Influence of cigarette smoke on the arginine pathway in asthmatic airways: Increased expression of arginase I. J Allergy Clin Immunol. 2007;119:391e397. [20] La Grutta S, Ferrante G, Malizia V, Cibella F, Viegi G. Environmental effects on fractional exhaled nitric oxide in allergic children. J Allergy. 2012;2012: 916e926. [21] Spanier AJ, Kahn RS, Hornung RW, et al. Environmental exposures, nitric oxide synthase genes, and exhaled nitric oxide in asthmatic children. Pediatr Pulmonol. 2009;44:812e819. [22] Kovesi TA, Dales RE. Effects of the indoor environment on the fraction of exhaled nitric oxide in school-aged children. Can Respir J. 2009;16: e18ee23. [23] Perzanowski MS, Divjan A, Mellins RB, et al. Exhaled NO among inner-city children in New York City. J Asthma. 2010;47:1015e1021.
Contents lists available at ScienceDirect
Effect modification by raceeethnicity of environmental tobacco smoke exposure on airway inflammation in US children Alem Mehari, MD *, y; Ngozi Enwerem, MD *; Yewande Odeyemi, MD *, y; and Richard. F. Gillum, MD * * Department y
of Internal Medicine, Howard University College of Medicine, Washington, DC Division of Pulmonary Diseases, Howard University College of Medicine, Washington, DC
A R T I C L E
I N F O
Article history: Received for publication July 24, 2015. Received in revised form September 5, 2015. Accepted for publication September 14, 2015.
A B S T R A C T
Background: Asthma mortality and morbidity are higher in black than in white children. Fractional exhaled nitric oxide (FeNO) is a noninvasive biomarker of eosinophilic airway inflammation. Identification of differences in the effect of environmental tobacco smoke (ETS) on airway inflammation by race and ethnicity from a large sample is needed. Objective: To estimate a racial difference in association with ETS and FeNO. Methods: Data from the 2007 to 2012 National Health and Nutrition Examination Survey were analyzed to compare associations of ETS and FeNO levels in US black and other children. No ETS exposure was defined as a serum cotinine level lower than 0.05 ng/mL and ETS exposure was defined as a serum cotinine level of at least 0.05 ng/mL. FeNO was measured using a device that relies on an electrochemical sensor. Analyses took the complex survey design into account. Results: The analytic sample was formed by 5,473 participants (6e11 years old, n ¼ 2,385; 12e19 years old, n ¼ 3,088) with complete data on demographics, serum cotinine levels, and 2 reproducible FeNO measurements. In weighted linear regression analyses at 6 to 11 years, the interaction term for ETS and black race was not significant (P ¼ .15). At 12 to 19 years, the interaction term was significant (P ¼ .03) in an analysis of all racial groups. In race-specific models, the coefficient for ETS exposure in blacks was 0.033 and that in others was 0.175, ie, ETS exposure was associated with a greater decrease in FeNO in non-blacks than in blacks. Conclusion: There was no evidence at 6 to 11 years of age for an effect modification by race of the association between ETS and FeNO. At 12 to 19 years, the data suggested an effect modification. Ó 2015 American College of Allergy, Asthma & Immunology. Published by Elsevier Inc. All rights reserved.
Introduction Asthma morbidity and mortality are higher in black than in white children.1 Asthma diagnosis and treatment are challenging in children because symptoms are often reported by a parent and are often minimized. Inhaled corticosteroids, which are the mainstay of therapy in asthma, are underused because of concerns about longterm side effects.2 Measurement of fractional exhaled nitric oxide (FeNO) recently has gained favor in identifying and monitoring eosinophilic airway inflammation in asthma.3 FeNO testing also is useful for monitoring adherence to asthma medications, identifying patients with persistent allergen exposure, and guiding the use and dosing of inhaled corticosteroid therapy.4 Thus, FeNO
Reprints: Alem Mehari, MD, Pulmonary and Critical Care, Howard University, 2041 Georgia Avenue NW, Washington, DC 20060; E-mail: [email protected]. Disclosures: Authors have nothing to disclose. Funding Sources: None.
provides health care providers with valuable information for assessment of disease severity and monitoring control. In atopic asthma, inducible NO synthetase (iNOS) is upregulated in respiratory epithelium and produces increased NO concentrations in exhaled air.5 Many factors can affect FeNO. FeNO increases with age in children and can be higher in the elderly than in younger adults.3 Levels can be higher in men than in women and in taller than in shorter adults.6 Smoking decreases FeNO7 and consumption of NO-rich foods increases levels.8 In addition, ozone air pollution, upper respiratory infections,9 and allergen exposure increase FeNO, and exercise decreases FeNO.3 However, major effects have been shown only for smoking, viral infections, and recent intake of NO-rich foods. Although young children do not use tobacco, they are frequently exposed to environmental tobacco smoke (ETS), most often in the home.10 In 2007 through 2008, an estimated 54% of children 3 to 11 years old and 47% of nonsmoking adolescents 12 to 19 years old were exposed to ETS.11 In 2007 through 2010, 54% of US children with asthma had ETS exposure.12 Studies have associated exposure
http://dx.doi.org/10.1016/j.anai.2015.09.010 1081-1206/Ó 2015 American College of Allergy, Asthma & Immunology. Published by Elsevier Inc. All rights reserved.
478
A. Mehari et al. / Ann Allergy Asthma Immunol 115 (2015) 477e480
Table 1 Coefficients from multiple regression models of natural logarithmic fractional excretion of exhaled nitric oxide for 2 age groups, National Health and Nutrition Examination Survey, 2007 through 2012 Variable
Multiple linear regression: 6e11 y Regression coefficienta (95% CI)
Age per yr Sex Female Male Standing height, cm Race or ethnicity Non-Hispanic black Non-Hispanic white Hispanic Hay fever in past 12 mo Present Absent Wheeze in past 12 mo Present Absent Self-reported asthma Present Absent ETS exposure Yes No ETS black race
0.0884 (0.0656e0.1112) 0.03373 (0.0996 to 0.0322) Ref 0.0153 (0.0066e0.0204) 0.1873 (0.0694e0.3054) Ref 0.1935 (0.1183e0.2686)
Multiple linear regression: 12e19 y P value
Regression coefficienta (95% CI)
P value
<.001
0.0104 (0.0034 to 0.0242)
.14
.31
0.2346 (0.3199 to 0.1492) Ref 0.0099 (0.0047e0.0151)
<.001
<.001 <.001 <.001
0.3119 (0.1887e0.4351) Ref 0.1701 (0.0654e0.2750)
<.001 <.001 .002
0.2417 (0.1058e0.377) Ref
.001
0.2289 (0.1029e0.35510) Ref
.001
0.1499 (0.0001 to 0.2998) Ref
.05
0.0569 (0.0930 to 0.2069) Ref
.45
0.1627 (0.2424 to 0.0829) Ref
<.001
0.1849 (0.2333 to 0.1365) Ref
<.001
0.0930 (0.1726 to 0.0132) Ref 0.1314 (0.0484 to 0.3110)
.02
0.1915 (0.2845 to 0.0985) Ref 0.1344 (0.0131e0.2556)
<.001
.15
.03
Abbreviations: CI, confidence interval; ETS, environmental tobacco smoke; Ref, reference group. a Coefficients are expressed as change in natural logarithmic fractional excretion of exhaled nitric oxide per 1-U change in the independent variable; means are expressed as parts per billion.
to ETS with asthma occurrence and asthma exacerbations in children.13 Previous studies in mostly small, selected volunteer or clinical samples have associated acute and chronic exposures to ETS with lower FeNO.14,15 Whether ETS exposure lowers FeNO values to the same degree in children, a degree that impairs its utility in asthma management overall or in high-risk subgroups, remains controversial. Whether any effect of ETS differs in black vs white children is unreported. In a large US national sample, this study tested these hypotheses: (1) ETS is associated with lower FeNO in children and youth and (2) this association varies by race or ethnicity. Methods The National Health and Nutrition Examination Survey (NHANES) is designed to assess the health and nutritional status of adults and children in the United States.16,17 Response rates for 2year study cycles exceed 75%. The National Center for Health Statistics Research ethics review board approved all protocols. All participants and their agents provided written informed consent. Trained interviewers administered questionnaires in the home on family and sample person characteristics using a ComputerAssisted Personal Interviewing (CAPI) system. The household smokers section of the family questionnaire provides information on exposure to secondhand smoke in the home setting. The medical conditions section provides self-reported personal interview data on diagnosed asthma. The respiratory disease section provides interview data for symptoms of wheezing. Reported hay fever episodes in the past 12 months were considered atopy. Participants who were nonsmokers were classified as having exposure to ETS if serum cotinine levels were at least 0.05 ng/L, the limit of detection. Serum cotinine, a metabolite of nicotine, was analyzed using an isotope dilution liquid chromatographic tandem mass spectrometry method. Participants with evidence of active smoking, as indicated by levels of at least 10 ng/L, were excluded.11 Children 6 to 11 years old were assumed to be nonsmokers if their serum cotinine concentration was no higher than 10 ng/mL. Youths 12 to 19 years old were considered nonsmokers if their
serum cotinine concentration was no higher than 10 ng/mL and they did not report smoking within the preceding 30 days or use of any nicotine-containing product within the preceding 5 days at their physical examination. The primary end point of the study was FeNO. FeNO was measured using the Aerocrine NIOX MINO, a portable hand-held NO analyzer (Aerocrine AB, Solna, Sweden). This device relies on an electrochemical sensor to detect exhaled NO levels and provides measurements from 5 to 300 ppb in whole numbers. The Respiratory Health ENO Procedures Manual18 provides a detailed description of the examination procedure. The NHANES protocol required 2 FeNO measurements that were reproducible within the 10 attempts allowed. This analysis uses the average of 2 reproducible measurements. The correlation between the first 2 tests was good (R2 ¼ 0.75, P < .001) and the geometric mean difference between the first and second tests was 1.3 ppb. FeNO distributions, normal values, and basic findings using this method have been reported.16,17 Statistical Analysis Distributions of FeNO were approximately logarithmically normal. This analysis compared the means of logarithmically transformed values; exponentially, the means of logarithmically transformed values yielded geometric mean values with 95% confidence intervals. Multivariable linear regression modeling using STATA 11 (STATA Corp, College Station, Texas) assessed ETS exposure by history or by cotinine level as a predictor of logarithmically transformed FeNO and assessed effect modification by race. Models controlled for age, sex, Hispanic ethnicity, asthma diagnosis, histories of wheezing and hay fever and smoking (12e19 years old only), height, and body mass index. Data analyses using STATA 11 accounted for the complex survey design, differential probability of sample selection, nonresponse, and sample non-coverage. Results Of 30,442 participants interviewed, 29,353 underwent examination; of those 7,459 were 6 to 19 years old; of these, 6,318 had 2
A. Mehari et al. / Ann Allergy Asthma Immunol 115 (2015) 477e480
reproducible FeNO measurements. Those 5,473 who also had complete data on demographics, household ETS, and serum cotinine formed the analytic sample. Of these, 2,385 were 6 to 11 and 3,088 were 12 to 19 years old. Examination of mean natural logarithm (ln) FeNO by exposure to ETS, age group, sex, and race or ethnicity suggested that ETS was associated with lower FeNO in non-Hispanic (NH) whites but higher FeNO in blacks. The pattern in Mexican Americans was similar to that in NH whites. Patterns in other groups were unstable, likely owing to smaller numbers. In multiple linear regression analysis, there was a positive correlation of ln FeNO with age, nonwhite ethnicity, standing height, and wheeze or hay fever within the past 12 months. FeNO levels increased with increasing age significantly (P < .001) in children 6 to 11 years old and with increasing height (P < .001) in children 6 to 11 and youth 12 to 19 years old. After controlling for all other factors, NH blacks consistently showed higher mean FeNO levels compared with the others (P < .001) in the 2 age groups. Male sex was associated with higher FeNO in adolescents (P < .001) but not in children 6 to 11 years old. In participants 6 to 11 and 12 to 19 years of age, there was a negative correlation of FeNO with ETS exposure, female sex, body mass index, and self-reported history of asthma. In weighted linear regression analyses at 6 to 11 years old, the interaction term for ETS and race was not significant (P ¼ .15; Table 1). At 12 to 19 years old, the interaction term was significant (P ¼ .03) in an analysis of all racial groups (Table 1). In race-specific models, the coefficient for ETS exposure in blacks was 0.033 and that in others was 0.175, ie, ETS exposure was associated with a greater decrease in ln FeNO in non-blacks than in blacks. However, in sensitivity analyses, when other races were excluded, the interaction term no longer attained significance. In expanded regression models to control for additional confounders (asthma, hay fever, self-reported wheezing or undiagnosed asthma, height, body mass index, and steroid use), the interaction term for black and ETS also was not significant.
479
ng/mL in others. Therefore, other unmeasured or unknown factors must account for this finding. Most previous studies in children with asthma or healthy children have associated recent exposure to ETS with lower FeNO levels,17,19,20 but there are some exceptions.22 In multivariate models in the 2007 to 2010 NHANES, ETS exposure was independently and negatively associated with FeNO at 6 to 11 and 12 to 80 years of age.17 A follow-up study found that ETS exposure at 4 years was associated with higher FeNO at 7 years, suggesting ETS might have a chronic effect as a risk factor for future airway inflammation apart from its acute effect to lower FeNO.23 In nonsmokers at least 3 years old, prevalence of exposure to ETS was consistently higher in NH blacks than in NH whites.11 FeNO levels also were higher in NH blacks than in NH whites.16 Examination of effect modification by race was lacking. Of note is that multivariate models with individual characteristics could account for only a small percentage of variation in FeNO (10% at 6e11 years and 16% at 12e80 years).17 As presented in Table 1, a history of hay fever was significantly associated with FeNO in children and youth. The major strength of this study is the availability of data from the NHANES, which is a nationally representative and comprehensive survey drawing from a large and diverse sample of participants. The survey design avoided volunteer bias and enhanced the generalizability of the present results. The authors acknowledge that sampling and non-sampling errors, including measurement error, might occur. In some subgroup analyses, the confidence interval of regression coefficients could not be generated, likely owing to strong geographic clustering of ethnic groups. The present analysis controlled for possible factors known to be related to FeNO; however, other factors not included in this analysis might differentially affect children by race or ethnicity. In conclusion, there was no evidence at 6 to 11 years of age for an effect modification by race of the association between ETS and FeNO. At 12 to 19 years, data suggested an effect modification with greater effect of ETS in non-blacks than in blacks. Replication in longitudinal studies is needed.
Discussion The present study found a pattern indicating that exposure to ETS lowered FeNO in NH whites but had little effect on FeNO in NH blacks. However, this trend for effect modification by race did not attain statistical significance at 6 to 11 years of age and was inconsistently significant at 12 to 19 years when tested in linear regression models controlling for multiple confounders. The possible mechanism by which tobacco smoke lowers FeNO could be that it increases arginase activity and therefore decreases arginine availability for NO production in the airways.19 NO is formed in airway epithelium by the conversion of arginine to citrulline by NOS, principally iNOS. Arginine also is converted by arginase to ornithine, which is a precursor of prolines and polyamines, which leads to collagen and mucus formation and cell proliferation, respectively. Epithelial and smooth muscle cells from smokers have been found to have higher levels of arginase and ornithine decarboxylase (a downstream enzyme of the arginase pathway) than those of nonsmokers.19 Others have suggested negative feedback of NO in smoke on iNOS.20 A polymorphism in the NOS and related genes could modify the effect of smoking.21 It could be asked why in race-specific models for 12- to 19-yearolds the coefficient for ETS exposure in blacks was only 0.033 and that in others was 0.175 (ie, ETS exposure was associated with a greater decrease in ln FeNO in non-blacks than in blacks). One explanation could be that ETS exposure was less in blacks than in others. However, that is not supported by the finding that in 12- to 19-year-old nonsmokers classified as having ETS exposure, the median serum cotinine level was 0.30 ng/mL in blacks and 0.26
References [1] Akinbami LJ, Moorman JE, Bailey C, et al. Trends in asthma prevalence, health care use, and mortality in the United States, 2001e2010. NCHS Data Brief. 2012;94:1e8. [2] Boulet LP, FitzGerald JM, Reddel HK. The revised 2014 GINA strategy report: opportunities for change. Curr Opin Pulm Med. 2015;21:1e7. [3] Bjermer L, Alving K, Diamant Z, et al. Current evidence and future research needs for FeNO measurement in respiratory diseases. Respir Med. 2014;108: 830e841. [4] Dweik RA, Boggs PB, Erzurum SC, et al. An official ATS clinical practice guideline: interpretation of exhaled nitric oxide levels (FENO) for clinical applications. Am J Respir Crit Care Med. 2011;184:602e615. [5] Scott M, Raza A, Karmaus W, et al. Influence of atopy and asthma on exhaled nitric oxide in an unselected birth cohort study. Thorax. 2010;65:258e262. [6] Olin AC, Rosengren A, Thelle DS, Lissner L, Bake B, Torén K. Height, age, and atopy are associated with fraction of exhaled nitric oxide in a large adult general population sample. Chest. 2006;130:1319e1325. [7] Persson MG, Gustafsson LE, Zetterström O, Agrenius V, Ihre E. Single-breath nitric oxide measurements in asthmatic patients and smokers. Lancet. 1994; 343:146e147. [8] Zetterquist W, Pedroletti C, Lundberg JON, Alving K. Salivary contribution to exhaled nitric oxide. Eur Respir J. 1999;13:327e333. [9] Sanders SP, Proud D, Permutt S, Siekierski ES, Yachechko R, Liu MC. Role of nasal nitric oxide in the resolution of experimental rhinovirus infection. J Allergy Clin Immunol. 2004;113:697e702. [10] Marano C, Schober SE, Brody DJ, Zhang C. Secondhand tobacco smoke exposure among children and adolescents: United States, 2003e2006. Pediatrics. 2009;124:1299e1305. [11] Vital signs: nonsmokers’ exposure to secondhand smokedUnited States, 1999e2008. MMWR Morb Mortal Wkly Rep. 2010;59:1141e1146. [12] Quinto KB, Kit BK, Lukacs SL, Akinbami LJ. Environmental tobacco smoke exposure in children aged 3e19 years with and without asthma in the United States, 1999e2010. NCHS Data Brief. 2013;126:1e8. [13] Howrylak JA, Spanier AJ, Huang B, et al. Cotinine in children admitted for asthma and readmission. Pediatrics. 2014;133:e355ee362.
480
A. Mehari et al. / Ann Allergy Asthma Immunol 115 (2015) 477e480
[14] Yates DH, Breen H, Thomas PS. Passive smoke inhalation decreases exhaled nitric oxide in normal subjects. Am J Respir Crit Care Med. 2001;164: 1043e1046. [15] Hoyt JC, Robbins RA, Habib M, et al. Cigarette smoke decreases inducible nitric oxide synthase in lung epithelial cells. Exp Lung Res. 2003;29:17e28. [16] Brody DJ, Zhang X, Kit BK, Dillon CF. Reference values and factors associated with exhaled nitric oxide: U.S. youth and adults. Respir Med. 2013;107: 1682e1691. [17] See KC, Christiani DC. Normal values and thresholds for the clinical interpretation of exhaled nitric oxide levels in the US general population: results from the National Health and Nutrition Examination Survey 2007e2010. Chest. 2013;143:107e116. [18] Respiratory Health ENO Procedures Manual. http://www.cdc.gov/nchs/data/ nhanes/nhanes_07_08/ENO.pdf Accessed June 31, 2015.
[19] Bergeron C, Boulet L-P, Page N, et al. Influence of cigarette smoke on the arginine pathway in asthmatic airways: Increased expression of arginase I. J Allergy Clin Immunol. 2007;119:391e397. [20] La Grutta S, Ferrante G, Malizia V, Cibella F, Viegi G. Environmental effects on fractional exhaled nitric oxide in allergic children. J Allergy. 2012;2012: 916e926. [21] Spanier AJ, Kahn RS, Hornung RW, et al. Environmental exposures, nitric oxide synthase genes, and exhaled nitric oxide in asthmatic children. Pediatr Pulmonol. 2009;44:812e819. [22] Kovesi TA, Dales RE. Effects of the indoor environment on the fraction of exhaled nitric oxide in school-aged children. Can Respir J. 2009;16: e18ee23. [23] Perzanowski MS, Divjan A, Mellins RB, et al. Exhaled NO among inner-city children in New York City. J Asthma. 2010;47:1015e1021.