Sleep disturbance and daytime symptoms in wheezing school-aged children
Kristine Desager, Department of Pediatrics, University Hospital Antwerp, Wilrijkstraat 10, 2650 Antwerp, Belgium. Tel.: +32 3 821 34 48; fax: +32 3 829 11 94; e-mail: firstname.lastname@example.org
The aim of the study was to investigate whether wheezing is associated with disturbed sleep and increased daytime symptoms in school-aged children. A random sample of 1234 children, aged 6–14 years, participated in a respiratory health study in the region of Antwerp. The International Study of Asthma and Allergies in Childhood questionnaire and a separate sleep questionnaire were completed. In the children who wheezed in the last 12 months, sleep quality was more frequently disturbed due to nocturnal awakenings and restless sleep compared with children who did not wheeze. Daytime sleepiness and tiredness were more common in wheezing than in non-wheezing children. After adjusting for possible confounders a positive association was found between wheeze and: difficulties falling asleep [odds ratio (OR) = 2.0], restless sleep (OR = 5.0), daytime sleepiness (OR = 3.8) and daytime tiredness (OR = 5.1). Chronic cough (OR = 2.4), snoring (OR = 2.0), chronic rhinitis (OR = 2.6) and eczema (OR = 3.3) were associated with disturbed sleep. Chronic cough (OR = 2.5) and rhinitis (OR = 4.1) were related to daytime tiredness. Chronic rhinitis was an important risk factor for snoring (OR = 1.9). In wheezing school-aged children, decreased quality of sleep and increased daytime tiredness and sleepiness were more often reported. Upper airway symptoms were related to the sleep disturbances.
Asthma in childhood is not well controlled (Rabe et al., 2000). Recently, the extensive problem of sleep disruption in asthmatic adults has been highlighted (Fitzpatrick et al., 1991; Janson et al., 1990, 1996). In childhood, sleep disturbance as a consequence of asthma has been a neglected area: there are no large cross-sectional studies in children designed to investigate the possible relationship between asthma or wheeze and sleep. In a study with more than 1000 6–14-year-old children investigating the association of asthma with extra-respiratory symptoms, no association between asthma and sleep disturbance was found (Ronchetti et al., 2002). Lack of awareness of asthma-sleep association and its clinical implications can lead to a vicious cycle of poor control of asthma and impaired daytime activity (Fitzpatrick et al., 1991). Although nocturnal awakening is beginning to be recognized as a common problem among asthmatic children (Sadeh et al., 1998; Stores et al., 1998), the existence of other sleep disturbances, the clinical presentation and consequences on daytime functioning remain largely unclear.
Therefore, the aim of this study was to investigate the association between wheezing and disturbed sleep and daytime symptoms in school-aged children using a cross-sectional study design.
A cross-sectional study on respiratory health was conducted on 1234 school-aged children between 6 and 14 years from eight randomly selected primary schools in Mechelen, a city situated South of Antwerp with 75 943 inhabitants and 28 primary schools. Asthma symptoms were investigated using the International Study of Asthma and Allergies in Childhood (ISAAC) questionnaire (Pearce et al., 1993). The child was classified as having wheezing if the parents gave a ‘yes’ response to a question asking if their chest ever sounded wheezy or whistling in the last 12 months. The child was classified as suffering from chronic rhinitis if the parents answered ‘yes’ to the question: ‘Did your child suffer from running nose or obstructed nose in the last 12 months even when the child had no common cold?’ Other questions were related to hayfever, eczema and chronic cough in the last 12 months.
Allergy was documented using specific IgE determination for house dust mite on a capillary blood sample using a micromethod.
Pulmonary function test included forced vital capacity, forced expiratory volume in 1 s (FEV1) and maximal expiratory flow at 50 and 25% of vital capacity. Changes were recorded in the form of a maximal expiratory flow curve according to the standards of the American Thoracic Society (1987). FEV1 was performed baseline and 5 and 10 min after a 6-min free-running protocol (Ponsonby et al., 1996; Powell et al., 1996). For data analysis, two categories of baseline FEV1 were defined: <85% and ≥85% of predicted value. A decrease of >15% in FEV1 was defined as exercise-induced bronchospasm.
A questionnaire based on the Pediatric Sleep Questionnaire (PSQ) and on an adult sleep questionnaire was used to evaluate sleep problems (Chervin et al., 2000; Janson et al., 1995). Parents were asked if their child was troubled by any of the following problems in the last 6 months: (1) difficulties falling asleep; (2) awakening at night (nocturnal awakening); (3) pauses in the breathing pattern at night (apnoeas); (4) snoring; (5) restless sleep; (6) falling asleep during the day (daytime sleepiness); (7) tiredness in the absence of any exercise (daytime tiredness) (see Appendix). The child was classified as having difficulties falling asleep if it took 1 h or more to fall asleep. From the questions related to the time to go to bed and get up during the week, the total sleep time was calculated.
Table 1 describes the participation rates for each part of the study protocol. The questionnaires were distributed and collected via the schools.
Table 1. Participation rates for the different parts of the study protocol
|Number of children invited||1234|
|Lung function determination||599|
|Specific IgE house dust mite||428|
The study was approved by the Ethical Committee of the University of Antwerp. Informed consent was obtained from the parents.
The statistical analysis was performed with Statistical Package for Social Sciences software (release 10.0.5; SPSS, Chicago, IL, USA). Data are summarized by their mean and standard deviation when symmetrically distributed and by their median and range when the distribution was not symmetrical. Distributions of continuous data were tested for normality by Shapiro–Wilk's W-test. Differences between groups were tested by t-test for normally distributed data and by Mann–Whitney U-test when the distribution was not normal. Differences in categorical data were tested by chi-square or by Fisher exact test when appropriate.
First, the influence of wheezing on sleep disturbance and daytime symptoms was assessed for the whole group of 959 children. Six different outcome variables were identified: difficulties falling asleep, nocturnal awakenings, apnoeas, restless sleep, daytime sleepiness and daytime tiredness. For each of these variables a multiple logistic regression analysis was performed, adjusting for confounding variables: snoring, passive smoking, BMI and gender. Only variables influencing the logistic regression coefficient (confounding variables) of wheezing were entered in the model.
Finally, the role of asthma-related risk factors on sleep disturbance and daytime symptoms, adjusted for the confounding variables was assessed by multiple logistic regression analysis. A P-value <0.05 was regarded as statistically significant.
The demographic characteristics of the group of wheezing and non-wheezing children are presented in Table 2. In this random sample of school-aged children wheezing was present in 83/943 (8.8%) children. Wheezing children were more frequently males, had a higher prevalence of exercise-induced bronchospasm, house dust mite allergy and chronic rhinitis in the last 12 months than the non-wheezing children. They were treated more often with nasal or asthma medication, underwent more often adenoidectomy and/or tonsillectomy in the past and their baseline FEV1 was more often lower than 85%. Snoring was present in 28% of the children. All children were Caucasian except four Asian and four black, South African children.
Table 2. Characteristics of children with and without wheezing
|Age (years)||9.6 (7–14)||9.6 (7–13)||0.76|
|Passive smoking (%)||28||23||0.34|
|Exercise-induced bronchospasm (%)||4||20||<0.001|
|House dust mite sensitization (%)||28||49||0.004|
|Hayfever last 12 months (%)||6||25||<0.001|
|Treatment nose last 12 months (%)||18||46||<0.001|
|Treatment wheezing last 12 months (%)||4||59||<0.001|
|Adenoidectomy and/ or tonsillectomy (%)||15||28||0.004|
Wheezing children had a significantly higher prevalence of snoring, restless sleep, nocturnal awakenings, daytime sleepiness and tiredness than the non-wheezing children (Table 3).
Table 3. Occurrence of sleep disturbance and daytime symptoms in wheezing and non-wheezing children
|Restless sleep (%)||10||26||<0.001|
|Difficulties falling asleep (%)||16||19||0.410|
|Nocturnal awakenings (%)||20||30||0.022|
|Apnoea (%)|| 5|| 6||0.740|
|Total sleep time (h) (SD)||10:21 (0:42)||10:25 (0:42)||0.659|
|Daytime sleepiness (%)||13||25||0.004|
|Daytime tiredness (%)|| 7||17||<0.001|
Multiple logistic regression analysis with adjustment for possible confounders showed associations between wheeze and: difficulties falling asleep, restless sleep, daytime sleepiness and tiredness (Table 4).
Table 4. Multiple logistic regression analysis of risk factors for sleep disturbance and daytime symptoms. Each column represents a separate multiple logistic regression analysis with the outcome variable in the heading of the column. Odds ratios (95% confidence interval) are adjusted for the independent variables, indicated in the rows of the table
|Wheeze||2.0* (1.0–4.0)||1.5 (0.7–3.4)||0.7 (0.1–3.1)||5.0*** (2.2–11.1)||3.8*** (1.8–8.0)||5.1*** (2.2–12.1)|
|Snoring||1.0 (0.6–1.7)||1.6 (0.9–2.5)||2.7** (1.2–5.9)||1.4 (0.8–2.6)||1.4 (0.8–2.4)||1.7 (0.9–3.4)|
|Passive smoking||0.7 (0.4–1.2)||0.6 (0.3–1.1)||0.6 (0.2–1.6)||1.1 (0.6–2.2)||0.8 (0.4–1.4)||0.9 (0.4–2.0)|
|BMI||1.0 (0.2–4.8)||1.6 (0.4–6.5)||1.7 (0.2–13.8)||0.9 (0.1–7.4)||1.3 (0.3–6.4)||5.8*** (1.3–25.9)|
|Gender||1.0 (0.7–1.6)||1.3 (0.8–2.0)||0.7 (0.3–1.4)||1.1 (0.6–2.0)||1.3 (0.8–2.2)||1.8 (0.9–3.7)|
No association was found between exercise-induced bronchospasm or baseline lung function and sleep disturbance or daytime symptoms after adjusting for possible confounders.
Table 5 shows the odds ratio (OR) adjusted for possible confounders with the following associations for sleep problems: chronic cough and difficulties falling asleep, snoring and nocturnal awakenings, chronic rhinitis and nocturnal awakenings, snoring and apnoeas. For the daytime symptoms an association was found between chronic cough and daytime tiredness, chronic rhinitis and daytime tiredness.
Table 5. Asthma-related risk factors for sleep disturbance and daytime symptoms. Each column represents a separate multiple logistic regression analysis with the outcome variable in the heading of the column. Odds ratios (95% confidence interval) adjusted for gender, passive smoke exposure, body mass index, adenoidectomy/tonsillectomy, baseline FEV1, exercise-induced bronchospasm, house dust mite sensitization and variables in the table
|Snoring||0.7 (0.4–1.4)||2.0* (1.1–3.8)||2.4* (1.0–6.2)||1.8 (0.8–3.9)||1.2 (0.5–3.1)|
|Chronic cough||2.4* (1.0–5.6)||2.2 (0.9–5.3)||1.9 (0.5–7.3)||1.4 (0.4–4.5)||2.5** (0.9–7.5)|
|Chronic rhinitis||1.0 (0.5–2.0)||2.6** (1.4–4.9)||0.8 (0.3–2.3)||1.3 (0.5–3.1)||4.1*** (1.7–9.5)|
|Hayfever||1.7 (0.6–4.4)||0.5 (0.2–1.6)||0.7 (0.1–3.5)||0.7 (0.1–3.6)||1.1 (0.3–4.2)|
|Eczema||1.0 (0.4–2.3)||1.3 (0.6–3.0)||1.3 (0.4–4.4)||3.3** (1.3–8.2)||1.1 (0.3–3.5)|
Wheeze was related to an increased risk for snoring (OR = 1.9, 95% CI: 1.0–3.9). Table 6 shows that chronic rhinitis is associated with snoring. If the child underwent surgical removal of adenoids and tonsils in the past, this was also associated with snoring.
Table 6. Asthma-related risk factors for snoring. Odds ratios (95% confidence interval) adjusted for gender, passive smoke exposure, body mass index (BMI), adenotomy/tonsillectomy, baseline FEV1 and exercise-induced bronchospasm (EIB)
|Chronic rhinitis||1.9 (1.2–3.1)**|
|Passive smoke exposure||1.2 (0.7–1.9)|
|BMI ≥ 25 kg m−2||0.3 (0.08–1.4)|
|FEV1 (<85%pred)||0.7 (0.4–1.2)|
This large cross-sectional study of nearly 1000 children demonstrates that wheezing in school-aged children is associated with decreased quality of sleep: more difficulty falling asleep and restless sleep were reported in the wheezing children compared with the non-wheezing children. This resulted in an increased prevalence of reported sleepiness and tiredness during the day. Chronic cough and rhinitis were associated with disturbed sleep and with daytime tiredness, whereas chronic rhinitis was a risk factor for snoring.
In this random sample of school-aged children the prevalence of wheezing was nearly 9%, which is in agreement with previously reported prevalence in Belgium (Wieringa et al., 2001).
In our analyses, respiratory symptoms (wheeze, cough, rhinitis) were more predictive of sleep disturbance than specific diagnoses of respiratory disease (asthma, hayfever). Therefore, analyses in this study were performed using symptoms.
The definition of asthma remains a challenge for epidemiologists. Validated questionnaires have been developed to deal with this problem. However, lung function tests may help in formulating the case definition of asthma, as these tests are less influenced by recognition or awareness of symptoms. Lung function can be evaluated relatively easily in school-aged children by measuring FEV1. However, in asthmatic children it is highly variable as the disease is characterized by variable narrowing of the airways and the majority of asthmatic children have little or no abnormality of lung function most of the time (Le Souef, 2001). Bronchial hyperresponsiveness tests have been developed for epidemiological surveys. Tsanakas has even suggested the free running test as a screening tool for undiagnosed asthma (Tsanakas et al., 1988). A study of Remes evaluated how much bronchial hyperresponsiveness tests would increase the diagnostic accuracy of childhood asthma after the symptom history had been taken into account (Remes et al., 2002). Despite the high specificity of bronchial hyperresponsiveness tests, diagnostic accuracy was not significantly improved. Sensitivity was found to be low, whereas several symptoms had both high sensitivity and specificity in relation to clinical asthma. They conclude that a written symptom questionnaire is a valid method for studying the epidemiology of asthma and bronchial hyperresponsiveness is not required for defining asthma in epidemiological studies.
The association between asthma or wheeze and impaired sleep has been documented in adults. In large population-based studies, asthmatic patients had more difficulties of initiating (Fitzpatrick et al., 1991; Janson et al., 1990, 1996) and of maintaining sleep (Janson et al., 1990; Vir et al., 1997) compared with healthy subjects. Sleep was found to be less refreshing in asthmatics compared with non-asthmatics (Fitzpatrick et al., 1991) and Janson showed more early morning awakenings in asthmatics (Janson et al., 1996; Vir et al., 1997). Physiological sleep laboratory assessments of adult asthmatic patients have provided some objective evidence of poor quality of sleep. Compared with healthy controls, patients were found to spend a greater time awake at night with less than usual amounts of deep non-rapid eye movement sleep (Fitzpatrick et al., 1991).
Large cross-sectional studies, investigating the association between wheezing and sleep disturbance in childhood, are lacking. One study with more than 1000 6–14-year-old children could not show a relationship between sleep disturbance and asthma (Ronchetti et al., 2002). However, the aim of this study was to evaluate if and what extra-respiratory symptoms were associated with asthma. Extra-respiratory symptoms included headache, restlessness, sleep disturbances, etc. Therefore, it seems likely that the design of that study was not sensitive enough to detect any association between asthma and sleep difficulties. In our study including nearly 1000 school-aged children, the association between wheezing and impaired sleep was clearly demonstrated: wheezing children were two times more at risk of having difficulties falling asleep and five times more at risk of having a restless sleep than non-wheezing children. Two authors report on the association between asthma and sleep in children (Camhi et al., 2000; Redline et al., 1999). However, in these studies the major drawback is that children were from selected populations. As part of the Tucson Epidemiologic Study on Obstructive Airway Disease, Camhi administered health questionnaires to children aged 3–14 of adult cohort members (Camhi et al., 2000). He identified wheezing as a risk factor for disorders of initiating and maintaining sleep. In our study, wheezing children did not show an increased risk of nocturnal awakenings compared with the non-wheezers. In the study of Redline, children were recruited as members of families with a proband with known sleep apnoea (Redline et al., 1999). In these children, sinus problems and persistent wheeze each independently predicted sleep-disordered breathing.
In childhood, polysomnographic data are conflicting and based on limited numbers of patients. Kales demonstrated that asthmatic children had a shorter total sleep time and spent less time in stage 4 (Kales et al., 1970). However, bronchodilators were withheld for at least 6 h before bedtime, a group of retrospective control subjects was used and asthma was documented only by symptoms and not by objective measurements of airway obstruction. Stores showed that the sleep of 21 asthmatic children was disrupted by very much higher rates of both brief and longer awakenings, producing a much reduced sleep efficiency (Stores et al., 1998). Sadeh monitored 40 asthmatic children using wrist actigraphs and reported lower percentages of quiet sleep and increased activity level during sleep (Sadeh et al., 1998). Avital demonstrated that sleep architecture was normal in asthmatic adolescents. However, this could be attributed to the fact that their asthma was well-controlled (Avital et al., 1991).
Concerning the effects of asthma on daytime behaviour, more daytime sleepiness was reported in asthmatic adults (Gislason and Almqvist, 1987) (Janson et al., 1990, 1996; Klink et al., 1994; Vir et al., 1997). No such effect was found by Fitzpatrick et al. (1991). Janson reported more daytime tiredness in asthmatic subjects (Janson et al., 1996).
Our results show that wheezing children are nearly four times more at risk of having daytime sleepiness than non-wheezing children. This is in agreement with the results of Stores, reporting that asthmatic children were found to fall asleep more frequently during daytime compared with normal controls (Stores et al., 1998).
Wheezing children were found to be five times more at risk of having daytime tiredness compared with non-wheezing children. The tiredness could be related to their asthma. Sadeh found moderately significant Pearson correlations between sleep measures of sleep per cent and mean activity during sleep and evening and morning peakflow measures (Sadeh et al., 1998). Although in our study, wheezing children had significantly more exercise-induced bronchospasm and a higher prevalence of baseline FEV1 below 85% compared with the non-wheezing children, no association was found between exercise-induced bronchospasm or baseline lung function and sleep disturbance or daytime symptoms after adjusting for possible confounders. Therefore, it is not very likely that daytime tiredness in our population can only be attributed to poorly controlled asthma.
Apart from the association of wheeze with quality of sleep and daytime symptoms, a higher prevalence of snoring was found in the wheezing children: 46% of the wheezing children snored versus 26% of the non-wheezing children. In adults, Janson, Fitzpatrick and Vir also found this association (Fitzpatrick et al., 1991; Janson et al., 1996; Vir et al., 1997). In adult patients with obstructive sleep apnoea, Lin found an increased prevalence in bronchial hyperresponsiveness (Lin and Lin, 1995), whereas Janson did not find a relationship between bronchial hyperresponsiveness and snoring or apnoeas (Janson et al., 1996). In our study no association was found between exercise-induced bronchospasm and snoring.
In order to investigate the possible underlying mechanism causing the impaired quality of sleep in wheezing school-aged children, the association between asthma-related variables and sleep disturbance was analysed (Table 5). Chronic cough and chronic rhinitis were strongly related to impaired quality of sleep. On the contrary, chronic rhinitis was found to be independently associated with snoring (Table 6). Chronic rhinitis is a frequent problem in school-aged children. Postnasal drip with nasopharyngeal inflammation leads to lymphoid hypertrophy with prominence of adenoidal and tonsillar tissue, which may be associated with poor appetite, poor growth and obstructive sleep apnoea (Lack, 2001). It can be the presentation of an underlying atopic condition, but is usually not diagnosed as ‘hayfever’ in school-aged children. This is in agreement with the finding that no significant relationships were found for ‘hayfever’ in our study. Our results indicate that the upper airways are a major factor in causing sleep problems in wheezing children.
The gold-standard in the diagnosis of sleep problems is polysomnography, but the time, effort and expense of laboratory studies has limited relevant research and particularly epidemiological research that requires large samples. Research in adults has profited from the existence of several validated questionnaire instruments to assess sleep-related symptoms (Douglass et al., 1994; Johns, 1991). However, few published questionnaires have been designed for use in children. For assessing sleep-related breathing disorders, a three-question-item obstructive sleep apnoea score and a PSQ have been developed (Brouillette et al., 1984; Chervin et al., 2000). The last one has been validated by comparison with polysomnographic data. In our study, a modified PSQ was used: items concerning snoring and daytime sleepiness were derived from PSQ and questions concerning nocturnal awakening and difficulties falling asleep were added, based on a questionnaire used in adults in large epidemiological studies (Janson et al., 1995).
The sleep of wheezing school-aged children is impaired with difficulties falling asleep and restless sleep and as a consequence the children fall asleep during the day and are tired. Attention should be given to upper airway symptoms, such as chronic rhinitis, which are probably responsible for the sleep problems in these children.
In this questionnaire there are several different types of questions. The questions refer to how your child has slept in the past 6 months.
1. At what time goes your child usually to bed during the week? …….
2. At what time does your child usually get up during the week? …….
3. How long does it take for your child to fall asleep?
• 1/2 h or less
• 1 h
• 2 h
• more than 2 h
4. Does your child wake up in the night more than once a week?
• 2 times
• 3 times
• more than three times
If yes: how many times:
5. Does your child stop breathing during the night?
6. While sleeping, does your child always snore?
7. Does your child have restless sleep?
8. Does your child have a problem with sleepiness during the day?
9. Is your child tired during the day?