Pulmonary dysfunction in overweight and obese children with obstructive sleep apnoea

Abstract Introduction Overweight and obese children are at risk of obstructive sleep apnoea (OSA) and abnormal pulmonary function (PF). Aim Investigate the relationship between body mass index (BMI), OSA on PF in children. Materials & Method Seventy‐four children were recruited. Mixed obstructive apnoea‐hypopnea index (MOAHI), BMI, oxygen saturation (SpO2), forced expiratory volume one second (FEV1), forced vital capacity (FVC) and fractionated exhaled nitric oxide (FeNO) were measured. Results Twenty‐four and thirty children had mild OSA and moderate‐to‐severe OSA respectively. BMI correlated negatively with SpO2 nadir (r = −.363, p = .001). FVC, FEV1 and nadir SpO2 values decreased with OSA severity (p < .001). The odds of a child with OSA having an abnormal spirometry was 3.16 (95% CI: 1.08, 9.22). There was significant association between FeNO and AHI (r = .497, <.001). Discussion Overweight and obese children with OSA have significant abnormalities in pulmonary function independent of BMI. OSA severity and elevated FeNO also correlated with diminishing lung function.


| INTRODUC TI ON
With the emerging obesity epidemic, up to 60% of overweight and obese (OO) children are diagnosed with obstructive sleep apnoea (OSA) 1,2 and both conditions may independently or synergistically influence system inflammation. 3 It has been proposed that the adult sleep apnoea syndrome is related to sleep apnoea in children, and that the differences of OSA in patient populations of different age groups may represent different stages in the development of the adult form of OSA. 4 Several studies have reported pulmonary dysfunction (PD) in adults with OSA 5,6 ; however, limited data is available regarding the association among OO, OSA and PD in children. Van Eyck et al. found a correlation between OSA severity and diminished lung function, reporting that obesity being an important confounding factor in both OSA severity and pulmonary dysfunction. 7 Obesity has been recognized as a risk factor for the development of asthma with altered pulmonary function, poor treatment response and high morbidity. 8,9 The incidence of asthma is 1.47 times higher in obese individuals than in non-obese individuals, and a three-unit increase in BMI is associated with a 35% increase in the risk of asthma. 10 In a cohort study of more than 25,000 children and adults with asthma, Schatz et al. 11 showed that a higher BMI was associated with worsened asthma control and an increased risk of asthma exacerbations.
Pathophysiologically, OSA may be on the causal pathway between asthma and obesity. Plausible mechanisms by which sleep apnoea may exacerbate asthma are through enhanced vagal cholinergic tone induced by apnoea, upper airway repetitive obstructioninduced local airway inflammation, systemic inflammation and apnoea-related alterations in intrathoracic pressures, the latter resulting in bronchoconstriction. It is also plausible that low-grade inflammation and oxidative stress generated during apnoea-related hypoxaemia episodes could attribute to airway inflammation and bronchial hyperactivity. 3  Although the relationship between obesity and pulmonary dysfunction is becoming increasingly clear, there is still much controversy regarding whether it occurs in OO children with OSA. Therefore, the primary objective of this study was to investigate the relationship between OSA and PFT in OO children. We also sought to investigate if the metabolic inflammation would drive airway inflammation leading to airway hyperactivity in obese children with and without asthma.

| MATERIAL S AND ME THODS
In this study, consecutive OO children (aged 6-13 years) with suspected OSA were recruited from the paediatric clinic and sleep laboratory at University Malaya Medical Centre (UMMC) between March 2016 and May 2017. In this study, children with BMI 25 kg/ m 2 were classified as overweight and 30 kg/m 2 as obese as defined by the International Obesity Task Force criteria. 13 Children who were unable to perform spirometry; on supplemental oxygen or non-invasive ventilation; with chronic lung disease; neuromuscular or craniofacial syndromes were excluded. None of the patients had adenotonsillectomy at the time of this study. Informed consent was obtained from the parents/legal guardians and the ethics committee of UMMC approved this study (MECID 20162-2139). All patients recruited were given a validated questionnaire to assess their symptoms of sleep-disordered breathing (SDB). 14 All measured variables including body measurement, spirometry, polysomnography, fractionated exhaled nitric oxide, and bloods for cytokines, eosinophils and neutrophils were done within 3 months of each other with no significant changes in body measurement and SDB symptoms during this interval.

| Demographic data
Sociodemographic information was collected from both parents and patients. The diagnosis of asthma was based on history, clinical examination and investigations.

| Quantitative measurements of inflammatory cytokines; blood neutrophils and eosinophils
Blood was drawn for full blood count to look at percentage of neutrophil and eosinophil counts, and the expression level of inflammatory cytokines in plasma using a human Inflammation immune-assays kit (Bio-Rad Laboratories) as per manufacturer's instructions. The cytokine analysis was performed by MAGPIX, a Luminex® Instrument (R&D Systems). A standard curve was derived using the different concentrations of the assay standard. Intra-assay variability was represented as the coefficient of variation as per manufacturer's instructions. Internal control was used for data normalization.

| Statistical analysis
Statistical analysis was performed using IBM SPSS statistics version 23.0 (IBM Corp). Categorical variables summarized as frequencies and percentages. Normally distributed continuous variables were presented as means and standard deviations while skewed variables were described as medians and interquartile ranges.
For comparing group differences, a multivariate analysis was performed and the ANOVA procedure was used for normally distributed data and the nonparametric Kruskal-Wallis test was used for non-normal data. When significant differences were observed, post hoc tests were used to test pairwise difference. In testing the associations between continuously variables, either Pearson's or Spearman's correlations were used. For all tests, the significant values were set at p < .05. (2) = 6.89, p = .032). The odds of a child with OSA having an abnormal spirometry was 3.16 (95% CI: 1.08, 9.22 between those with asthma and without asthma (Table 4). There was also no significant correlation between measures of adiposity including BMI z-score and percentage of body fat mass with any spirometry parameters (Table 5).
FeNO levels measured were also not statistically significant between asthmatics and non-asthmatics in this study (z-value = −.033, p = .974).

| Inflammatory cytokines and blood neutrophils and eosinophils measurements
Plasma samples from 30 (40.5%) children (9 children without OSA, 12 with mild OSA and 9 with moderate-severe OSA) were assessed for the level of inflammatory cytokines. There were no significant associations between, both BMI and OSA, and cytokines levels (

| DISCUSS ION
This study investigated the relationship between lung function and OSA severity in children with high BMI. The major finding from this study is that OO children with OSA have significant abnormalities in pulmonary function compared to those without OSA especially The frequency of OSA in our study population was higher than previously proposed. 2 The possible explanation for the higher prevalence of OSA in our patient cohort could be due to known distinguished changes to the upper airway structure and facial geometry in Asian children. 18 With the emerging obesity epidemic, along with an increased risk of OSA due to the craniofacial structure in children of Asian descent, local paediatricians should be vigilant in screening for OSA in children.
Almost 56% of the children in this study had an abnormal spirometry with restrictive pattern being twice as much as obstructive pattern. 19 Of these group of children, 80% had OSA.
Pulmonary function correlated with OSA severity; increased OSA severity correlated with diminished lung function, especially FEV 1 and FVC. Desaturation during sleep was associated with worse FEV 1 . Obesity and sleep apnoea both may cause changes in thoracic mechanics resulting in dysfunction of lung compliance and reduced lung capacity. 19,20 The FEV 1 /FVC ratio is generally well preserved or elevated even in morbidly obese individuals, indicating that FEV 1 and FVC are affected at the same rate. 21 By reducing functional lung volume, obesity can change airway diameter due to the interdependence of the airway and the adjacent pulmonary parenchyma which may explain the changes seen in our patients' spirometry results. Because of the ineffectiveness of the respiratory muscles, there is closure of dependent airways with the formation of small areas of atelectasis and increased alveolar surface tension due to a reduction in FRC. 22 All of these factors lead to inspiratory overload which increases respiratory effort, heterogeneity of ventilation distribution and respiratory energy expenditure and therefore, increases neural respiratory drive, in addition to causing respiratory sleep disorders and eventually hypercapnic respiratory failure. 23

| CON CLUS ION
Abnormalities between awake pulmonary function and sleep respiratory parameters may be observed in Malaysian population of children with OSA independent of a higher BMI and/or asthma diagnosis. Diminished spirometry is associated with OSA severity. The correlation between FEV 1 , FeNO and OSA severity implies a possible role for local inflammation in the pathology of OSA.

AUTH O R CO NTR I B UTI O N S
Aina Salwa Kasim: Data curation (equal); formal analysis (equal).

CO N FLI C T O F I NTE R E S T S TATE M E NT
The author has no conflict of interest to declare.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.