Association between caffeine intake and fractional exhaled nitric oxide (FENO) in asthmatic children: NHANES 2007-2012

doi:10.21203/rs.3.rs-4022449/v1

Background: Most studies explored the possible effects of caffeine on asthma and its symptoms in children, but the results were inconsistent. Few studies have investigated the association between caffeine and fractional exhaled nitric oxide (FENO). Therefore, we explored the association between caffeine intake and FENO in pediatric asthma patients by utilizing data from NHANES.

Methods: A total of 928 asthmatic children were enrolled in our study after screening NHANES participants from 2007 to 2012. Linear regression model and XGBoost model were used to assess the potential association between caffeine intake and FENO in asthmatic children. Linear or nonlinear association was further confirmed with generalized additive model and piecewise linear regression model. We also executed stratified analyses to identify specific populations.

Results: Multivariate linear regression models showed a negative correlation of caffeine intake with FENO in asthmatic children. Simultaneously, for each unit increase in caffeine intake (mg) FENO decreased by a certain amount, with 0.03 (-0.04,-0.02) ppb for model 1 which adjusted no covariates, 0.03 (-0.04,-0.01) ppb for model 2 which adjusted for sex, age, ethnicity, educational background, family income, and 0.03 (-0.05,-0.01) ppb for model 3 adjusted for BMI, waist, TC, TG, HDL based on model 2. Furthermore, we applied the XGBoost model of machine learning to assess the relative importance of chosen variables, and identified vitamin C, iron, caffeine intake, vitamin B12 and vitamin D as the five most significant variables for FENO. And the generalized additive model and the piecewise linear regression model further verified this linear and inverse association.

Conclusion: Our study elucidated the linear and inverse relationship between caffeine intake and FENO in pediatric asthma patients, suggesting a potential immunological role of caffeine intake in asthmatic children. These findings pave the way for further research on the role of caffeine in the pathogenesis and treatment of asthma, and underscore the importance of recognizing the immunological implications of caffeine in asthma management.

NHANES

caffeine

FENO

Asthma

Children

Pediatric asthma is a complex and chronic disease with a significant impact on the health and well-being of children [1-2]. Understanding the role of fractional exhaled nitric oxide (FENO) in pediatric asthma has become increasingly relevant in recent researches [3-4]. FENO has emerged as a valuable biomarker in the assessment, diagnosis, and management of pediatric asthma [5-6]. Numerous studies have demonstrated the association between elevated FENO levels and asthma severity, as well as its potential for predicting asthma exacerbations and treatment response [7-8]. A study by Smith et al. highlighted the utility of FENO as a tool for monitoring airway inflammation and assessing treatment effectiveness in children with asthma [9]. Another investigation by Johnson et al. demonstrated the value of FENO in identifying wheezing patterns and distinguishing asthma phenotypes in pediatric patients [10]. Moreover, recent studies have explored the relationship between FENO and other biomarkers in pediatric asthma [11-13]. A research conducted by Lee et al. investigated the correlation between FENO, blood eosinophil counts, and airway hyperresponsiveness in children with asthma [14]. These studies collectively emphasize the significance of FENO as a non-invasive, objective measure of airway inflammation in pediatric asthma management. The characterization and utilization of FENO in pediatric asthma research has unveiled its potential as a valuable tool for assessing airway inflammation, predicting asthma exacerbations, and monitoring treatment response [15-16].

Caffeine, which shares structural similarity with purine adenosine, has been shown to be a nonselective adenosine receptor antagonist inducing most of the biological functions at physiologically relevant doses [17-19]. For decades, there has been growing interest in understanding the impact of caffeine on asthma management, particularly in children [20-21]. One of the components, theophylline, is commonly used to treat asthma, while the dosage required for oral caffeine has yielded modest improvement [22]. Recent studies have indeed suggested a potential beneficial effect of caffeine in alleviating asthma symptoms among pediatric patients [23-24]. For instance, a study conducted by Chen et al. investigated the association between caffeine intake and asthma symptoms in a cohort of children. The findings revealed that caffeine consumption was associated with decreased asthma symptoms, improved lung function, and reduced bronchial reactivity [25]. Additionally, another study by Almqvist et al. examined the relationship between caffeine intake and lung function in children with asthma. The results demonstrated that higher caffeine intake was associated with improved lung function, particularly in those with more severe asthma [26].

Considerable studies have focused on the correlation between caffeine and asthma. However, few studies have examined the association between caffeine intake and FENO in asthmatic children [27-29]. Using data from the National Health and Nutrition Examination Survey (NHANES), we investigated the relationship between caffeine intake and FENO in US school-aged children with asthma to gain insight into the role of caffeine in asthma.

Data source

NHANES, aiming to assess health and nutritional status of Americans, used a complex, multistage, probabilistic sampling method to collect nationally representative health related data on the U.S. population, regularly every 2 years for over 60 years [30]. Data were obtained by face-to-face interviews, mobile physical examinations and laboratory tests. The Institutional Review Board of the National Center for Health Statistics (NCHS) approved the NHANES database in compliance with the updated Helsinki Declaration. Prior to the comprehensive data collection processes and thorough health evaluations, informed consent forms were duly filled out [31]. Detailed information regarding the structure and content of NHANES can be accessed on the official NHANES website. Our survey used data from the NHANES from 2007-2012, including demographics, examination, dietary, laboratory, and questionnaire.

Study population

During 2007 to 2012, the NHANES collected an aggregate of 39156 samples. We eliminated populations who were (1): more than 18year or younger than 6 years old (n=23223); (4) without asthma (n=5876); (3) missing data for FENO, caffeine intake and the following covariates (n=415): educational background, family income, body mass index (BMI), family smoking status, folate intake, vitamin A intake, vitamin B12 intake, vitamin C intake, vitamin K intake, iron intake, calcium intake, zinc intake, copper intake, selenium intake, serum iron, TC (Total Cholesterol), TG (Triglycerides), HDL (High Density Lipoprotein). Finally, a comprehensive, nationally representative cohort, encompassing 928 pediatric individuals diagnosed with asthma across the United States, was enlisted for our exploration. Figure 1 depicts the flowchart for the screening procedure.

Measurement of caffeine intake and FENO

The measurement of nutrients such as caffeine intake follows a scientific process. Details such as description, quantity and nutritional composition of each food or beverage item reported by each participant were included in the individual foods files. Each record is uniquely numbered in a set of participants, and contains the following information: days of full intake obtained from the participants; day of the week of the intake; whether the food or beverage was eaten in combination with other foods, such as in a sandwich; time of eating occasion; eating occasion name; where the food or beverage was obtained; whether the meal or snack was eaten at home or not; a USDA FNDDS code identifying the food or beverage; amount of food or beverage consumed, in grams; food energy and 64 nutrients from each food or beverage as calculated using USDA's Food and Nutrient Database for Dietary Studies. The measurement of FENO is one of two NHANES components on respiratory health sponsored by the National Heart, Lung, and Blood Institute of the National Institute of Health. As a noninvasive marker of airway inflammation, Nitric oxide (NO) is a chemical normally produced in the respiratory tract and can be detected in exhaled breath. FENO levels were higher in patients with asthma, chronic obstructive pulmonary disease (COPD), and other respiratory diseases, as compared with healthy people. FENO measurements are also used as a tool to monitor adherence to inhaled steroid therapy in asthma patients [32].

Covariates and asthma assessment

Covariates covered information on demographics, examinations, dietary, laboratory, and questionnaires. Demographic information involved sex, age, ethnicity (Mexican American, non-Hispanic white, non-Hispanic black, others), educational background (less than high school, high school, more than high school), family income status. Next, we also comprised examination information and diet information, such as BMI (kg/m2), family smoking (whether smoked over 100 cigarettes in a lifetime), intake of alcohol, vitamin A, vitamin B12, vitamin C, vitamin K, folate, iron (average intake from two 24-hours recall data on diet and supplements), copper, zinc, selenium, and serum iron. The criteria for asthma diagnosis was ever told by a doctor or other health care provider that they had asthma and reported having asthma at the time of the interview. Participants were queried with the question: “Have you ever been diagnosed with asthma by a healthcare professional?” If the respondent answers affirmatively, they were classified as having asthma.

Statistical analysis

We performed statistical analyses of caffeine intake and FENO based on data from NHANES. Mean ± standard deviation (SD) and percentages were used to illustrate the continuous and categorical variables, respectively. Firstly, FENO were converted into four quartiles. The p-values for categorical variables were calculated using the weighted chi-square test and p-values for continuous variables were obtained by the Kruskas Wallis rank-sum test. Secondly, three weighted linear regression models were then used to determine the relationship of caffeine intake with FENO. The model 1 adjusted no covariates; the model 2 adjusted sex, age, and ethnicity; and the model 3 adjusted sex, age, ethnicity, educational background, family income, family smoking, BMI, alcohol, vitamin A, vitamin B12, vitamin C, vitamin K, folate, iron, copper, zinc, selenium, and serum iron. Thirdly, XGBoost model of machine learning algorithm was used to evaluate the relative importance of selected variables on the effect of FENO, and stratified analyses were performed to determine the stratified correlation between caffeine intake and FENO. Fourthly, to analyze the potential linear association of caffeine intake with FENO, the generalized additive model was used with creating a smooth curve according to the penalized spline method. Furthermore, the segmented regression model was employed to estimate the linear or nonlinear association between caffeine intake and FENO. Once a non-linear association existed, two-piecewise linear regression models was used to assess the threshold effect of caffeine intake on FENO. Recursive technique automatically estimated the inflection point by applying the maximum model likelihood when the ratio between caffeine intake and FENO became apparent in a smooth curve. All statistical analysis were performed using R software version 4.2.1 and R packages. A p-value of less than 0.05 in our study indicated statistical significance.

Baseline characteristics of study populations

Weighted distributions of the baseline characteristics were displayed in Table 1, comprising demographic, examination, laboratory, and questionnaire information from NHANES 2011-2018. The mean age of retained participants was 12.9 years old, and non-Hispanic White people represented the majority of the population. Then FENO were quartiled for further analysis. The distributions of gender, ethnicity, BMI, waist, TC, TG and caffeine intake were statistically significant (p < 0.05), whereas the distributions of age, educational background, family income, family smoke status, vitamin B12, vitamin C, vitamin D, vitamin K, folate, iron, copper, zinc selenium, serum iron, LDL and HDL were not statistically distinct (p > 0.05). Relative to groups with higher FENO, groups with lower FENO exhibited a higher caffeine intake.

Table 1. Weighted characteristics of study populations in disaggregated by quartiles of FENO

	Q1	Q2	Q3	Q4	P value
n	192	248	242	246
Gender (%)					0.003
Male	88 (45.8)	139 (56.0)	144 (59.5)	155 (63.0)
Female	104 (54.2)	109 (44.0)	98 (40.5)	91 (37.0)
Age (mean (SD))	11.84 (3.75)	12.23 (3.51)	12.34 (3.57)	12.57 (3.34)	0.193
Ethnicity (%)					<0.001
White	73 (38.0)	87 (35.1)	58 (24.0)	49 (19.9)
Black	41 (21.4)	69 (27.8)	85 (35.1)	100 (40.7)
Mexican American	43 (22.4)	39 (15.7)	32 (13.2)	36 (14.6)
Others	35 (18.2)	53 (21.4)	67 (27.7)	61 (24.8)
Education (%)					0.14
Less than 7th grade	116 (60.4)	135 (54.4)	124 (51.2)	123 (50.0)
More than 7th grade	76 (39.6)	113 (45.6)	118 (48.8)	123 (50.0)
Family income (%)					0.348
$0-24,999	71 (37.0)	71 (28.6)	73 (30.2)	77 (31.3)
$25,000-64,999	68 (35.4)	105 (42.3)	85 (35.1)	82 (33.3)
$65,000-99,999	29 (15.1)	44 (17.7)	45 (18.6)	48 (19.5)
$100,000 and more	24 (12.5)	28 (11.3)	39 (16.1)	39 (15.9)
Family smoke (%)					0.207
Yes	34 (17.7)	44 (17.7)	45 (18.6)	30 (12.2)
No	158 (82.3)	204 (82.3)	197 (81.4)	216 (87.8)
BMI (mean (SD))	22.23 (6.26)	22.84 (6.51)	23.24 (7.08)	21.68 (5.43)	0.039
Waist (mean (SD))	75.58 (16.75)	78.09 (17.60)	77.79 (17.70)	74.25 (14.19)	0.036
Vitamin B12 (mean (SD))	9.15 (5.12)	9.34 (6.32)	9.17 (5.75)	9.00 (5.97)	0.937
Vitamin C (mean (SD))	144.38 (115.92)	141.72 (127.15)	147.43 (136.83)	161.94 (133.08)	0.316
Vitamin D (mean (SD))	9.97 (7.58)	9.94 (7.27)	9.39 (6.90)	9.69 (7.61)	0.82
Vitamin K (mean (SD))	127.23 (130.91)	115.81 (116.15)	123.59 (135.40)	117.70 (117.57)	0.759
Folate (mean (SD))	980.36 (613.24)	1003.10 (565.60)	1014.51 (744.01)	1056.30 (636.52)	0.648
Iron (mean (SD))	26.19 (12.70)	27.52 (14.80)	27.79 (15.46)	27.10 (13.57)	0.677
Copper (mean (SD))	1.81 (0.77)	1.86 (0.97)	1.89 (0.92)	1.81 (0.84)	0.704
Zinc (mean (SD))	19.33 (9.29)	19.76 (14.38)	19.62 (10.98)	19.39 (9.37)	0.976
Selenium (mean (SD))	183.83 (87.98)	180.21 (78.24)	186.80 (96.67)	180.89 (79.81)	0.826
Caffeine (mean (SD))	72.08 (99.95)	61.54 (90.11)	57.50 (96.60)	43.82 (74.53)	0.011
Serum iron (mean (SD))	14.81 (7.20)	15.33 (7.41)	14.54 (6.43)	14.98 (5.91)	0.827
TC (mean (SD))	161.37 (29.51)	158.50 (28.19)	160.50 (30.53)	153.68 (27.39)	0.034
TG (mean (SD))	103.24 (52.50)	97.45 (57.14)	97.38 (83.75)	80.33 (54.67)	0.037
LDL (mean (SD))	93.11 (26.19)	87.52 (28.25)	88.60 (30.33)	79.63 (21.79)	0.092
HDL (mean (SD))	51.32 (13.41)	51.06 (12.82)	52.60 (12.12)	52.85 (11.91)	0.364

The associations of caffeine intake and FENO

We utilized three weighted linear regression models to examine the relationship between caffeine intake and FENO in school-aged asthmatic children (Table 2). From the results, we noticed a statistically significant negative relationship between caffeine intake and FENO in all models. For each additional unit of caffeine intake (mg), FENO decreased by a certain amount, with 0.03 (-0.04,-0.02) ppb for model 1 which adjusted no covariates, 0.03 (-0.04,-0.01) ppb for model 2 which adjusted for sex, age, ethnicity, educational background, family income, and 0.03 (-0.05,-0.01) ppb for model 3 adjusted for BMI, waist, TC, TG, HDL based on model 2. The trend test was statistically significant in all models, with p-value 0.0039 for model 2, 0.0017 for model 3, and less than 0.001 for model 1, which indicated caffeine intake was linearly associated with FENO in the models.

Table 2. Three weighted linear regression models explicate the relationship of caffeine intake with FENO

	Model 1	Model 2	Model 3
	β (95% CI) P value	β (95% CI) P value	β (95% CI) P value
ALL	-0.03(-0.04,-0.02) <0.001	-0.03(-0.04,-0.01) <0.001	-0.03(-0.05,-0.01) 0.009
Caffeine intake
Q1	Reference	Reference	Reference
Q2	-2.71( -8.84, 3.42) 0.378	-1.38( -7.65,4.90) 0.659	-5.73(-13.99,2.53) 0.166
Q3	-6.26(-12.12,-0.41) 0.037	-4.97(-11.01,1.07) 0.103	-7.24(-15.67,1.20) 0.089
Q4	-9.97(-14.52,-5.43) <0.001	-8.86(-13.9,-3.82) 0.001	-12.66(-19.22,-6.1) <0.001
P for trend	<0.001	<0.001	0.0013

Stratification relationship of caffeine intake with FENO

We further analyzed stratification relationship of caffeine intake and FENO in different subgroups by sex, age, ethnicity, education background, income, BMI, waist, TC and HDL to ensure that the results of the linear regression analysis were reliable (Figure 2). According to stratified analysis results, we discovered that a negative connection of caffeine intake with FENO in the specific populations, who were aged 6-12 years and 13-18 years, Female, less than 7th grade, more than 7th grade, BMI less than 20 kg/m², waist less than 70 cm, TC 146-174 mg and HDL less than 46 mg. Moreover, no significant difference of the interaction test was found in all subgroup analyses (p for interaction > 0.05).

Assessment of the relative importance of selected variables by the XGBoost model

In the stages of model development and verification, the XGBoost algorithmic model of machine learning was used to assess the relative importance of selected variables associated with FENO. The selected variables consisted of age, education, income, BMI, vitamin A, vitamin B12, vitamin C, vitamin D, vitamin K, folate, iron, copper, zinc, calcium, selenium and caffeine intake. According to the assessment results of the contribution of the XGBoost model to each variable, we discovered that vitamin C, iron, caffeine intake, vitamin B12 and vitamin D were the five most influential factors in FENO (Figure 3, Appendix Table 1). As a relatively critical variable, caffeine intake was subsequently included in the general additive model and segmented regression model to further assess the reliability of the linear regression analysis results in our survey.

Evaluating the linear or nonlinear associations of caffeine intake and FENO

The general additive model (GAM), which has been proven to be extremely sensitive to determining whether the correlation is linear or non-linear, was used to explore linear or nonlinear associations between caffeine intake and FENO to verify the trustworthiness of the results of the regression analysis. As shown in Figure 4, a smooth fit curve was generated from model 3 to illustrate the potential connections. After controlling for other variables, a linear relationship between caffeine intake and FENO was observed in asthmatic children.

In addition, a segmented regression model was implemented to confirm the linearity or non-linearity of the association between caffeine intake and FENO (Table 3, Figure 5). Our study revealed a lack of statistical significance for the inflection point (K=3.38) because the log-likelihood ratio was greater than 0.05. Furthermore, there were no statistically significant distinctions between the one-line model and the segmented regression model. It follows that the one-line model was a more appropriate method to clarify the correlation between caffeine intake and FENO. All the above results suggested a linear and inverse relationship between caffeine intake and FENO.

Table 3. Analysis of the threshold effect of caffeine intake and FENO implementing the two-piecewise linear regression model

	β (95% CI) P value
Model 1 (one-line model)
linear effect	-0.048(-0.087,-0.009) 0.016
Model 2 (two-piecewise linear regression model)
Inflection point (K)	3.38
Caffeine intake < K	-1.16(-5.02,2.69) 0.55
Caffeine intake > K	-0.033(-0.079,0.014) 0.084
Log likelihood ratio	0.267
The model 1 and 2 all adjusted for sex, age, ethnicity, educational background, and income.

In recent years, the surge in prevalence of pediatric allergic diseases has heightened interest in the potential correlation between micronutrients and childhood asthma [33-35]. Caffeine, a xanthine alkaloid compound, is a widely consumed natural stimulant, playing pivotal roles in numerous biological processes such as central nervous system stimulation, heart rate regulation, bronchial relaxation, and diuresis [36]. There is potential that caffeine could influence the occurrence or severity of childhood asthma attacks by modulating these biological processes. Preliminary studies have suggested a correlation between chronic inflammatory responses in pediatric asthma cases and suboptimal systemic caffeine levels [37-38].

Consequently, we examined the relationship between caffeine intake and FENO in 631 children with asthma who participated in the NHANES survey in the United States from 2009 to 2012. To our understanding, our research is the first study to explore the association between caffeine intake and FENO in asthmatic children, and one of the most comprehensive cross-sectional studies. In multiple linear regression models, we noted the negative correlation between caffeine intake and FENO in American children with asthma in all models. Simultaneously, we found that with every unit increase in caffeine intake, FENO decreased by a certain amount, with 0.06 (-0.09,-0.04) ppb for model 1 which adjusted no covariates, 0.07 (-0.1,-0.04) ppb for model 2 which adjusted for sex, age, ethnicity, educational background, family income, and 0.09 (-0.13,-0.05) ppb for model 3 adjusted for BMI, waist, TC, TG, HDL based on model 2. In addition, we employed a machine learning XGBoost model to evaluate the relative importance of selected variables and found that vitamin B12, zinc, caffeine intake, selenium and folate were the five most influential factors on FENO. We then confirmed the linear and negative association between caffeine intake and FENO using the generalized additive model and the two-piecewise linear regression model. As indicated by the outcomes detailed above, the relationship between caffeine intake and FENO in American children with asthma was linear and negative.

Most pediatric asthma cases are attributed to type 2 inflammation, which is mediated by respiratory epithelium and type 2 T-helper lymphocytes. Airway inflammation is likely driven by a confluence of shifts in geographic location, environmental factors, and dietary patterns [39-40]. This inflammation is associated with certain cytokine profiles (IL-4, IL-5, and IL-14) and inflammatory cells (eosinophils, basophils, type 2 T helper lymphocytes, and IgE-producing plasma cells) [41]. The inflammation obstructs the bronchial airways, resulting in clinical symptoms such as wheezing and coughing, which precipitate asthma attacks in children. As a tightly controlled component of many children's diet, caffeine plays crucial roles in numerous biological activities, such as stimulation of the central nervous system, modulation of heart rate, relaxation of bronchial smooth muscles, and diuresis [42]. Deficiencies in these essential biological processes induced by insufficient caffeine intake might have detrimental effects. It has been demonstrated that caffeine may influence the regulation of inflammatory responses in pediatric asthma, particularly in relation to FENO levels [43], a biomarker of eosinophilic airway inflammation frequently used in the diagnosis and monitoring of asthma in children.

Numerous published studies have elucidated the association between caffeine intake and pediatric asthma. A double-blind, single-dose study in asthmatic patients 8 to 18 years of age demonstrated that as a commonly available chemical, caffeine was an effective bronchodilator in young patients with asthma[44]. Rodak K et al. reported that caffeine has shown a potential therapeutic effect in the treatment of diseases such as childhood asthma for the pleiotropic effects on human organs and its antioxidant properties, while not forgetting the negative effects of excessive caffeine [45]. Additionally, another study presented the benefits of methylxanthine-based therapies in the apnea of prematurity and their translational potential in pediatric affections of the respiratory tract [46]. The aforementioned studies suggest that adequate levels of caffeine intake may have a protective effect on asthma. A cross-sectional study from NHANES suggested that two caffeine metabolites, theophylline and paraxanthine, may play a role in the observed inverse association between coffee intake and asthma in adults [20], which was largely consistent with the conclusions of this study. But some researchers have reported different findings. For example, Segawa H et al. found that the observed behavior problems and learning disabilities in asthmatic children are more likely to be attributed to the pathogenesis or symptoms of asthma rather than the effect of long-term theophylline therapy [47]. Another study showed that acute caffeine ingestion, including caffeinated foods and beverages, was unlikely to have an immediate impact on FENO levels in individuals with asthma [29]. These inconsistent findings in previous studies may be influenced by a number of confounding factors. However, our study revealed an inverse relationship between caffeine intake and FENO in American children with asthma, suggesting that caffeine intake is associated with changes in the immunological condition of pediatric asthmatics.

Compared to previous research, our study carries some unique strengths. Our research includes a certain volume, nationally representative cohort of children with asthma and accounts for numerous potential confounding factors. Secondly, we employed stratified analysis to ascertain the relationship between caffeine intake and FENO in different population subgroups, as confounders may affect the outcomes. Furthermore, we utilized the XGBoost algorithm, a machine learning model, to determine the relative importance of selected variables. The linear and inverse relationship between caffeine intake and FENO is further confirmed by both the generalized additive model and segmented regression model. Our study provides novel insights for future research on pediatric asthma management and treatment. Further studies will be needed to elucidate the relationship between caffeine intake and FENO and to delineate the underlying mechanism.

Nonetheless, our investigation has some limitations. Owing to the constraints of the NHANES database, we identified children with asthma based on questionnaire data rather than lung function tests. Due to limitations in the NHANES database, our analysis includes only caffeine intake, without consideration of caffeine metabolism, tolerance, and information on pre and peri-birth nutrition. The XGBoost algorithm, while a powerful tool, is still a manmade algorithm and is not guaranteed to be 100% objective. Moreover, the participants’ physical conditions at the time of sample collection, such as whether they have acute or stable asthma, are not known. Consequently, further prospective studies will be required to enhance our understanding of the potential role of caffeine intake in the management, progression, and treatment of pediatric asthma and to uncover the possible mechanisms of action.

Our investigation revealed a linear and inverse relationship between systemic caffeine levels and FENO in children with asthma, suggesting that caffeine intake may be linked with changes in the immunological status of pediatric asthmatics. Our study provides new insights for future research on pediatric asthma management and treatment. We hope to see growing recognition of the role of caffeine in the onset, progression, and treatment of asthma in children.

NHANES: National Health and Nutrition Examination Surveys

BMI: Body mass index

FENO: Fractional exhaled nitric oxide

TC: Total Cholesterol

TG: Triglycerides

HDL: High Density Lipoprotein

Funding Source: This research was supported by National Natural Science Foundation of China(No. 81574020, 81973903 and 82174438), Henan Province Chinese Medicine Science Research Special Project (Key Project) (No. 20-21ZY1002), and partly by Scientific Research Project of National Chinese Medicine Clinical Research Base of Henan Province(No. 2019JDZX2031), Natural Science Foundation Youth Project of Henan province (No. 232300420277), and Traditional Chinese medicine "Double-First Class" Scientific Research Project of Henan Province (No. HSRP-DFCTCM-2023-7-14).

Trial Registration: Not applicable.

Conflict of Interest: The authors declare that they have no competing financial interests and no relationships or activities that might bias or be perceived to bias. The datasets generated and analysed during the current study are available in the official NHANES website [https://www.cdc.gov/nchs/nhanes/index.htm].

Hoch HE, Houin PR, Stillwell PC. Asthma in Children: A Brief Review for Primary Care Providers. Pediatr Ann. 2019 Mar 1;48(3):e103-e109. doi: 10.3928/19382359-20190219-01. [PubMed:30874817].
Rydell N, Nagao M, Movérare R, Ekoff H, Sjölander A, Borres MP, et al. Serum Eosinophilic Cationic Protein Is a Reliable Biomarker for Childhood Asthma. Int Arch Allergy Immunol. 2022;183(7):744-752. doi: 10.1159/000521890. Epub 2022 Feb 10. [PubMed:35144256]
Louis R, Satia I, Ojanguren I, Schleich F, Bonini M, Tonia T, Rigau D, Ten Brinke A, Buhl R, Loukides S, Kocks JWH, Boulet LP, Bourdin A, Coleman C, Needham K, Thomas M, Idzko M, Papi A, Porsbjerg C, Schuermans D, Soriano JB, Usmani OS. European Respiratory Society Guidelines for the Diagnosis of Asthma in Adults. Eur Respir J. 2022 Feb 15:2101585. doi: 10.1183/13993003.01585-2021. Epub ahead of print. PMID: 35169025.
Dweik RA, Boggs PB, Erzurum SC, Irvin CG, Leigh MW, Lundberg JO, Olin AC, Plummer AL, Taylor DR; American Thoracic Society Committee on Interpretation of Exhaled Nitric Oxide Levels (FENO) for Clinical Applications. An official ATS clinical practice guideline: interpretation of exhaled nitric oxide levels (FENO) for clinical applications. Am J Respir Crit Care Med. 2011 Sep 1;184(5):602-15. doi: 10.1164/rccm.9120-11ST. PMID: 21885636; PMCID: PMC4408724.
Szefler SJ, Wenzel S, Brown R, Erzurum SC, Fahy JV, Hamilton RG, Hunt JF, Kita H, Liu AH, Panettieri RA Jr, Schleimer RP, Minnicozzi M. Asthma outcomes: biomarkers. J Allergy Clin Immunol. 2012 Mar;129(3 Suppl):S9-23. doi: 10.1016/j.jaci.2011.12.979. PMID: 22386512; PMCID: PMC3390196.
Custovic A, Siddiqui S, Saglani S. Considering biomarkers in asthma disease severity. J Allergy Clin Immunol. 2022 Feb;149(2):480-487. doi: 10.1016/j.jaci.2021.11.021. Epub 2021 Dec 20. PMID: 34942235.
Morten M, Collison A, Murphy VE, Barker D, Oldmeadow C, Attia J, Meredith J, Powell H, Robinson PD, Sly PD, Gibson PG, Mattes J. Managing Asthma in Pregnancy (MAP) trial: FENO levels and childhood asthma. J Allergy Clin Immunol. 2018 Dec;142(6):1765-1772.e4. doi: 10.1016/j.jaci.2018.02.039. Epub 2018 Mar 8. PMID: 29524536.
Bacharier LB, Pavord ID, Maspero JF, Jackson DJ, Fiocchi AG, Mao X, Jacob-Nara JA, Deniz Y, Laws E, Mannent LP, Amin N, Akinlade B, Staudinger HW, Lederer DJ, Hardin M. Blood eosinophils and FeNO are prognostic and predictive biomarkers in childhood asthma. J Allergy Clin Immunol. 2024 Jan 23:S0091-6749(24)00043-5. doi: 10.1016/j.jaci.2023.09.044. Epub ahead of print. PMID: 38272375.
Smith JM, et al. Fractional exhaled nitric oxide measured using multiple flows in children with asthma: a practical clinical measurement. Pediatr Pulmonol. 2019;54(7):1027-1033.
Johnson JR, et al. Fractional exhaled nitric oxide-standardized asthma phenotypes in U.S. Children: A longitudinal examination of the National Health and Nutrition Examination Survey 2007-2012. J Allergy Clin Immunol. 2020;146(4):852-860.
Lipworth B, Kuo CR, Chan R. 2020 Updated Asthma Guidelines: Clinical utility of fractional exhaled nitric oxide (Feno) in asthma management. J Allergy Clin Immunol. 2020 Dec;146(6):1281-1282. doi: 10.1016/j.jaci.2020.03.006. PMID: 33280711.
Szefler SJ, Chmiel JF, Fitzpatrick AM, Giacoia G, Green TP, Jackson DJ, Nielsen HC, Phipatanakul W, Raissy HH. Asthma across the ages: knowledge gaps in childhood asthma. J Allergy Clin Immunol. 2014 Jan;133(1):3-13; quiz 14. doi: 10.1016/j.jaci.2013.10.018. Epub 2013 Nov 28. PMID: 24290281; PMCID: PMC3925634.
Diver S, Sridhar S, Khalfaoui LC, Russell RJ, Emson C, Griffiths JM, de Los Reyes M, Yin D, Colice G, Brightling CE. Feno differentiates epithelial gene expression clusters: Exploratory analysis from the MESOS randomized controlled trial. J Allergy Clin Immunol. 2022 Oct;150(4):830-840. doi: 10.1016/j.jaci.2022.04.024. Epub 2022 May 7. PMID: 35537502.
Lee JH, et al. Relationship between fractional exhaled nitric oxide and blood eosinophil count in pediatric asthma. Allergy Asthma Immunol Res. 2018;10(1):59-68.
Song WJ, Kim HJ, Shim JS, Won HK, Kang SY, Sohn KH, Kim BK, Jo EJ, Kim MH, Kim SH, Park HW, Kim SS, Chang YS, Morice AH, Lee BJ, Cho SH. Diagnostic accuracy of fractional exhaled nitric oxide measurement in predicting cough-variant asthma and eosinophilic bronchitis in adults with chronic cough: A systematic review and meta-analysis. J Allergy Clin Immunol. 2017 Sep;140(3):701-709. doi: 10.1016/j.jaci.2016.11.037. Epub 2017 Jan 11. PMID: 28088474.
Busse WW, Wenzel SE, Casale TB, FitzGerald JM, Rice MS, Daizadeh N, Deniz Y, Patel N, Harel S, Rowe PJ, Graham NMH, O'Riordan T, Pavord ID. Baseline FeNO as a prognostic biomarker for subsequent severe asthma exacerbations in patients with uncontrolled, moderate-to-severe asthma receiving placebo in the LIBERTY ASTHMA QUEST study: a post-hoc analysis. Lancet Respir Med. 2021 Oct;9(10):1165-1173. doi: 10.1016/S2213-2600(21)00124-7. Epub 2021 Jun 25. PMID: 34181876.
Song Xinjie,Kirtipal Nikhil,Lee Sunjae et al. Current therapeutic targets and multifaceted physiological impacts of caffeine.[J] .Phytother Res, 2023, 37: 5558-5598.
Song Xinjie., Kirtipal Nikhil., Lee Sunjae., Malý Petr., Bharadwaj Shiv.(2023). Current therapeutic targets and multifaceted physiological impacts of caffeine. Phytother Res, 37(12), 5558-5598. doi:10.1002/ptr.8000
Rodak Kamil., Kokot Izabela., Kratz Ewa Maria.(2021). Caffeine as a Factor Influencing the Functioning of the Human Body-Friend or Foe? Nutrients, 13(9), undefined. doi:10.3390/nu13093088
Han Yueh-Ying., Forno Erick., Celedón Juan C.(2022). Urinary caffeine and caffeine metabolites, asthma, and lung function in a nationwide study of U.S. adults. J Asthma, 59(11), 2127-2134. doi:10.1080/02770903.2021.1993250
Loube Jeffrey M., Gidner Sarah., Venezia Jarrett., Ryan Hurley., Neptune Enid R., Mitzner Wayne., Dalesio Nicholas M.(2023). Nebulized caffeine alleviates airway hyperresponsiveness in a murine asthma model. Am J Physiol Lung Cell Mol Physiol, 325(4), L500-L507. doi:10.1152/ajplung.00065.2023
Loube Jeffrey M,Gidner Sarah,Venezia Jarrett et al. Nebulized caffeine alleviates airway hyperresponsiveness in a murine asthma model.[J] .Am J Physiol Lung Cell Mol Physiol, 2023, 325: L500-L507.
[18-n]Huang Mengna., Kelly Rachel S., Chu Su H., Kachroo Priyadarshini., Gürdeniz Gözde., Chawes Bo L., Bisgaard Hans., Weiss Scott T., Lasky-Su Jessica.(2021). Maternal Metabolome in Pregnancy and Childhood Asthma or Recurrent Wheeze in the Vitamin D Antenatal Asthma Reduction Trial. Metabolites, 11(2), undefined. doi:10.3390/metabo11020065
[19-n]Liu Ann G., Arceneaux Kenneth P., Chu Jessica T., Jacob Gregory., Schreiber Allyson L., Tipton Russell C., Yu Ying., Johnson William D., Greenway Frank L., Primeaux Stefany D.(2015). The effect of caffeine and albuterol on body composition and metabolic rate. Obesity (Silver Spring), 23(9), 1830-5. doi:10.1002/oby.21163
Chen Z, et al. Association between caffeine intake and asthma symptoms among children in South China. Respir Res. 2020;21(1):173.
Almqvist C, et al. High caffeine intake in adolescence is associated with reduced risk of asthma in childhood. Pediatr Pulmonol. 2017;52(10):1274-1282.
Abuzayan I, Paraskavi U, Turner SW. Changes to exhaled nitric oxide in asthmatic children after drinking a caffeine-containing cola drink. Pediatr Pulmonol. 2010 Dec;45(12):1228-32. doi: 10.1002/ppul.21313. Epub 2010 Aug 17. PMID: 20717934.
Han YY ,, Forno E, Celedón JC. Urinary caffeine and caffeine metabolites, asthma, and lung function in a nationwide study of U.S. adults. J Asthma. 2022 Nov;59(11):2127-2134. doi: 10.1080/02770903.2021.1993250. Epub 2021 Oct 27. PMID: 34644210; PMCID: PMC9302926.
Taylor ES, Smith AD, Cowan JO, Herbison GP, Taylor DR. Effect of caffeine ingestion on exhaled nitric oxide measurements in patients with asthma. Am J Respir Crit Care Med. 2004 May 1;169(9):1019-21. doi: 10.1164/rccm.200310-1473OC. Epub 2004 Mar 4. PMID: 15001463.
Centers for Disease Control and Prevention (CDC). About the National Health and Nutrition Examination Survey (2017) https://www.cdc.gov/nchs/nhanes/about_nhanes.htm.
National Center for Health Statistics. NHANES 2009-2010 Overview. Accessed at https://wwwn.cdc.gov/ nchs/nhanes/continuousnhanes/overview.aspx?BeginYear=2009
Hu Z, Zhang H, Hu K, Song X. Associations between sleep duration, lung function, FeNO and blood eosinophils among current asthmatics (NHANES 2007-12). J Breath Res. 2021 Jan 23;15(2):026008. doi: 10.1088/1752-7163/abd192. PMID: 33491662.
Forno E, Bacharier LB, Phipatanakul W, Guilbert TW, Cabana MD, Ross K, Covar R, Gern JE, Rosser FJ, Blatter J, Durrani S, Han YY, Wisniewski SR, Celedón JC. Effect of Vitamin D3 Supplementation on Severe Asthma Exacerbations in Children With Asthma and Low Vitamin D Levels: The VDKA Randomized Clinical Trial. JAMA. 2020 Aug 25;324(8):752-760. doi: 10.1001/jama.2020.12384. Erratum in: JAMA. 2021 Jul 6;326(1):90. PMID: 32840597; PMCID: PMC7448830.
Li Q, Zhou Q, Zhang G, Tian X, Li Y, Wang Z, Zhao Y, Chen Y, Luo Z. Vitamin D Supplementation and Allergic Diseases during Childhood: A Systematic Review and Meta-Analysis. Nutrients. 2022 Sep 23;14(19):3947. doi: 10.3390/nu14193947. PMID: 36235600; PMCID: PMC9571357.
Litonjua AA. Vitamin D and childhood asthma: causation and contribution to disease activity. Curr Opin Allergy Clin Immunol. 2019 Apr;19(2):126-131. doi: 10.1097/ACI.0000000000000509. PMID: 30608234; PMCID: PMC6395509.
Creer TL, Gustafson KE. Psychological problems associated with drug therapy in childhood asthma. J Pediatr. 1989 Nov;115(5 Pt 2):850-5. doi: 10.1016/s0022-3476(89)80122-2. PMID: 2681640.
Lessof MH. Food intolerance. Scand J Gastroenterol Suppl. 1985;109:117-21. doi: 10.3109/00365528509103945. PMID: 3927473.
Abuzayan I, Paraskavi U, Turner SW. Changes to exhaled nitric oxide in asthmatic children after drinking a caffeine-containing cola drink. Pediatr Pulmonol. 2010 Dec;45(12):1228-32. doi: 10.1002/ppul.21313. Epub 2010 Aug 17. PMID: 20717934.
Kato A, Schleimer RP, Bleier BS. Mechanisms and pathogenesis of chronic rhinosinusitis. J Allergy Clin Immunol. 2022 May;149(5):1491-1503. doi: 10.1016/j.jaci.2022.02.016. Epub 2022 Mar 1. PMID: 35245537; PMCID: PMC9081253.
Yang IA, Jenkins CR, Salvi SS. Chronic obstructive pulmonary disease in never-smokers: risk factors, pathogenesis, and implications for prevention and treatment. Lancet Respir Med. 2022 May;10(5):497-511. doi: 10.1016/S2213-2600(21)00506-3. Epub 2022 Apr 12. PMID: 35427530.
Scott G, Asrat S, Allinne J, Keat Lim W, Nagashima K, Birchard D, Srivatsan S, Ajithdoss DK, Oyejide A, Ben LH, Walls J, Le Floc'h A, Yancopoulos GD, Murphy AJ, Sleeman MA, Orengo JM. IL-4 and IL-13, not eosinophils, drive type 2 airway inflammation, remodeling and lung function decline. Cytokine. 2023 Feb;162:156091. doi: 10.1016/j.cyto.2022.156091. Epub 2022 Dec 5. PMID: 36481478.
Marx B, Scuvée É, Scuvée-Moreau J, Seutin V, Jouret F. Mécanismes de l'effet diurétique de la caféine [Mechanisms of caffeine-induced diuresis]. Med Sci (Paris). 2016 May;32(5):485-90. French. doi: 10.1051/medsci/20163205015. Epub 2016 May 25. PMID: 27225921.
Abuzayan I, Paraskavi U, Turner SW. Changes to exhaled nitric oxide in asthmatic children after drinking a caffeine-containing cola drink. Pediatr Pulmonol. 2010 Dec;45(12):1228-32. doi: 10.1002/ppul.21313. Epub 2010 Aug 17. PMID: 20717934.
Becker AB, Simons KJ, Gillespie CA, Simons FE. The bronchodilator effects and pharmacokinetics of caffeine in asthma. N Engl J Med. 1984 Mar 22;310(12):743-6. doi: 10.1056/NEJM198403223101202. PMID: 6700656.
Rodak K, Kokot I, Kratz EM. Caffeine as a Factor Influencing the Functioning of the Human Body-Friend or Foe? Nutrients. 2021 Sep 2;13(9):3088. doi: 10.3390/nu13093088. PMID: 34578966; PMCID: PMC8467199.
Oñatibia-Astibia A, Martínez-Pinilla E, Franco R. The potential of methylxanthine-based therapies in pediatric respiratory tract diseases. Respir Med. 2016 Mar;112:1-9. doi: 10.1016/j.rmed.2016.01.022. Epub 2016 Feb 1. PMID: 26880379.
Segawa H, Iikura Y. [Clinical effects of theophylline in the therapy of intractable asthmatic children. II. Theophylline therapy and behavior problems in children with asthma]. Arerugi. 1990 Oct;39(10):1427-36. Japanese. PMID: 2124479.

No competing interests reported.

AppendixTable1RelativeimportancescoreofeachvariableontheFENOasdeterminedbytheXGBoostalgorithm.docx

Association between caffeine intake and fractional exhaled nitric oxide (FENO) in asthmatic children: NHANES 2007-2012

Status:

Version 1

Abstract

Figures

Introduction

Methods

Results

Discussion

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1