|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Divisions of Pulmonary Medicine (Dr. Amin and Ms. Jeffries), Biostatistics (Dr. Bean), and Psychology (Dr. Burklow), Cincinnati Childrens Hospital Medical Center, Cincinnati, OH.
Correspondence to: Raouf Amin, MD, Cincinnati Childrens Hospital Medical Center, Division of Pulmonary Medicine, 3333 Burnet Ave, Cincinnati, OH 45229; e-mail: aminr0{at}chmcc.org
| Abstract |
|---|
|
|
|---|
Objectives: To determine whether clinically stable pediatric patients with CF have lower sleep efficiency than healthy control subjects, and to examine the relationship between sleep efficiency and pulmonary function.
Methods: Forty-four CF patients and 40 control subjects completed 5 days of actigraphy recordings. Additionally, sleep questionnaires were independently completed by all study participants and their parents. Pulmonary function testing was performed in all patients with CF. Multiple regression analysis was used to measure the independent correlation between sleep variables and pulmonary function.
Results: CF patients had significantly lower sleep efficiency than control subjects. The FEV1 of these patients correlated positively with sleep duration and efficiency, and negatively with the number and duration of nocturnal awakenings, age, and body mass index (BMI). The independent effect of FEV1 on sleep was first examined. Age and FEV1 were the only variables that predicted sleep duration (R2 = 0.3; p = 0.0007), while FEV1 was the only variable predicting sleep efficiency (R2 = 0.28; p = 0.0002). When the independent effect of sleep on FEV1 was analyzed, sleep efficiency, BMI percentile, and gender predicted FEV1 (R2 = 0.46; p
0.0001). The frequency of nocturnal cough reported by patients and their parents was an independent predictor of FEV1.
Conclusions: Pediatric patients with CF and stable pulmonary function have lower sleep efficiency and more frequent nocturnal awakenings than do healthy control subjects. After adjustment for demographic characteristics, there was an independent and significant correlation between sleep parameters and FEV1, when either sleep variables or FEV1 were used as dependent variables. These findings suggest a bidirectional relationship between sleep disturbance and CF lung disease.
Key Words: cystic fibrosis pulmonary function sleep disturbance
| Introduction |
|---|
|
|
|---|
Over the past several decades, there has been a steady increase in the life expectancy of CF patients.2 This trend can be attributed to a better understanding of the factors that contribute to the rate of pulmonary function decline. Such factors include the roles of adequate nutrition,345 early diagnosis and management of glucose intolerance,6789101112 frequency of endobronchial infections,13 and implementation of airway clearance.14 Sleep loss is known to have negative effects on mental and neurocognitive functions,151617 and on multiple metabolic parameters, including glucose metabolism.1819202122 It has also been associated with alterations in the normal circadian cytokine network,23 the expression of adhesion molecules,24 and the balance between cellular and humoral immune responses.2526 Previous studies2728293031 have shown an association between sleep disturbance and CF lung disease in patients with severe airway obstruction and nocturnal hypoxemia. However, whether sleep disturbance exists in CF patients with mild-to-moderate airway obstruction, and whether it has a negative impact on the clinical course of CF lung disease are not known. Understanding the interaction between sleep disturbance and CF lung disease could have important therapeutic implications. The objectives of the present study were thus twofold, as follows: to determine whether patients with clinically stable CF and mild-to-moderate pulmonary disease had lower sleep efficiency than healthy control subjects, and to examine the relationship between sleep efficiency and pulmonary function. To accomplish these objectives, we tested the hypothesis that nocturnal pulmonary and GI symptoms contribute to sleep disturbance, which in turn may adversely affect pulmonary function. To this end, we objectively measured sleep quality in the home environment, assessed childrens perceptions of their own sleep, assessed parents perceptions of their childs sleep, and explored the independent effect of sleep disturbance on pulmonary function. This study was approved by our institutional review board; an assent form was obtained from participants who were > 11 years of age.
| Materials and Methods |
|---|
|
|
|---|
Age-matched, gender-matched, and race-matched control subjects who were free from any acute or chronic pulmonary disease or other medical conditions were also recruited.
Actigraphy
The actigraph (Basic Motionlogger; Ambulatory Monitoring; Ardsley, NY) is a miniature, wristwatch-like device that monitors activity levels for extended continuous periods. Previous work32333435 has demonstrated its validity in monitoring sleep-wake patterns among infants, children, and adults. Actigraphic measures were averaged across 5 weekdays of monitoring. Actigraphic sleep measures included the following: sleep duration; sleep efficiency (ie, the percentage of true sleep time in the total sleep period), the number and duration of night awakenings lasting
5 min, and the total duration of awakenings after sleep onset.
Daily Sleep Logs: Daily sleep logs were completed by all participants and their parents. The measures were averaged across the monitoring period. The derived measures included the following: sleep onset; sleep duration; and awakening time.
Survey Questionnaire
A modified validated questionnaire consisting of 53 questions was administered to children with CF and age-matched, gender-matched, and race-matched control subjects.3637 The questionnaire included demographic data, and questions pertaining to the childs and parents medical histories. Thirty-three questions pertained to symptoms of sleep-disordered breathing, including nocturnal cough, sleep habits (eg, nocturnal use of bathroom), and symptoms of daytime sleepiness. The responses were graded as follows: never; less than once per week; one to two times per week; three to four times per week; and almost daily. To measure the degree of agreement between the responses of children and their parents, identical questionnaires were administered to parents. Both parents and children were instructed to complete the questionnaires independently.
Pulmonary Function Testing
All patients with CF underwent pulmonary function testing. Spirometry was performed (SensorMedics; Yorba Linda, CA) according to American Thoracic Society standards.38 Values were expressed as percentages of the predicted normal values, with adjustments made for age, sex, and height.
Statistical Analysis
Data from actigraphy were analyzed by t test, and
2 analysis was performed to measure differences in questionnaire responses between CF and control groups. For each group,
-statistics were calculated to measure the degree of agreement between parents and childrens responses. In the CF group, multiple regression analysis was performed to measure the independent correlation between sleep variables and FEV1. In the regression analysis, male gender was coded "2" and female gender was coded "1."
| Results |
|---|
|
|
|---|
|
|
|
|
|
The independent effect of sleep parameters on FEV1 was examined in a regression model that included demographic variables, age, gender, BMI expressed as a percentile, and sleep efficiency and duration as measured by actigraphy. Sleep efficiency (ß = 1.38; p = 0.0002), BMI percentile (ß = 0.29; p = 0.0013), and gender (ß = 10; p = 0.04) predicted FEV1 (R2 = 0.46; p = < 0.0001). When sleep duration rather than sleep efficiency was entered as an independent variable, BMI percentile (ß = 0.24; p = 0.02) and sleep duration (ß = 0.09; p = 0.04) were the only variables predicting FEV1 (R2 = 0.28; p = 0.0012).
Survey Questionnaire
Childrens Responses:
Compared to control subjects, children with CF reported more frequent awakenings with cough (p < 0.0001), more frequent awakenings for bathroom use (p = 0.005), more frequent difficulty initiating sleep (p = 0.04), and more frequent watching of television at bedtime (p = 0.03). The reason for more frequent nocturnal bathroom use was defecation for children with CF vs micturition for control subjects (p = 0.007).
Parents Responses: Compared to the parents of control subjects, the parents of children with CF reported that their children demonstrated more frequent awakenings with cough (p < 0.0001), more frequent awakenings for bathroom use (p = 0.02), more nocturnal sweating (p = 0.014), loud snoring (p = 0.007), and morning sleepiness (p = 0.007), and more frequent difficulty initiating sleep (p = 0.04). The reason for nocturnal bathroom use was defecation for children with CF vs micturition for control subjects (p = 0.015).
Agreement Between Measured and Reported Sleep Periods:
For all participants, the duration of sleep measured by actigraphy correlated with the sleep period reported by children (r = 0.48; p
0.0001) and that reported by parents (r = 0.58; p
0.0001). A lower correlation between reported and measured sleep duration was observed in the CF group vs the control group. In the CF group, the duration of sleep measured by actigraphy showed no significant correlation with the duration of sleep reported by children (r = 0.18; p = 0.2) or with that reported by parents (r = 0.29; p = 0.06). However, in the control group, the duration of sleep measured by actigraphy showed a significant correlation both with that reported by children (r = 0.71; p = < 0.0001) and that reported by parents (r = 0.79; p = < 0.0001). For both the CF patients and control subjects, the measured sleep duration was shorter than that reported by either children or their parents. This difference was larger for the CF group (Table 4 ).
|
CF Group: In the CF group, children and parents differed in their responses to three questions. Compared to parents responses, childrens responses indicated more frequent positive responses to watching television at bedtime (p = 0.008), more frequent positive responses to eating and drinking during the night (p = 0.02), and less frequent nocturnal sweating (p = 0.025).
Control Group: Compared to parents responses, children reported a higher degree of sleepiness on awakening (p = 0.01) and during the day (p = 0.01).
Relationship Between Nocturnal Symptoms and FEV1
Childrens Responses:
In the multiple regression analysis, which included demographic variables, age, gender, BMI expressed as a percentile, and frequency of nocturnal cough, a model that consisted of BMI expressed as a percentile (ß = 0.38; p
0.001), nocturnal cough (ß = 4.7; p = 0.01), and gender (ß = 9; p = 0.04) predicted FEV1 (R2 = 0.44; p
0.0001).
Parents Responses: In the multiple regression, which included demographic variables, age, gender, BMI expressed as a percentile, and frequency of nocturnal cough, a model that consisted of BMI percentile (ß = 0.25; p = 0.007) and nocturnal cough (ß = 8.2; p = 0.001) predicted FEV1 (R2 = 0.41; p = < 0.0001).
| Discussion |
|---|
|
|
|---|
The objective measures of sleep quality provided by actigraphy and the subjective measures provided by questionnaires completed by children and their parents demonstrated that children with CF have lower sleep efficiency and more frequent nocturnal awakenings compared to age-matched and gender-matched control subjects. Sleep efficiency and duration were closely associated with FEV1. This independent and significant association between sleep variables and FEV1 was demonstrated when either sleep parameters or FEV1 was used as an outcome variable.
The effect of the severity of pulmonary disease on sleep has been examined in previous studies2728293031 that have focused primarily on CF patients with moderate-to-severe lung disease. These studies proposed that sleep-related intermittent hypoxia and hypercapnia are likely mechanisms for sleep disruption. Although severe hypoxemia and hypercapnia could have contributed to increased frequency of arousals in these studies, it is unlikely that they played an important role in our study population. Ninety-three percent of our CF patient group had FEV1 in the normal to moderately abnormal range. In the absence of acute pulmonary infection and with only a mild abnormality of FEV1, it is less likely that severe blood gas abnormalities played an important role in causing sleep disturbance.
Our study explored the association between nocturnal pulmonary symptoms and sleep disturbance. The frequency of awakening with nocturnal cough was reported more frequently by patients with CF and their parents than by control subjects, and was independently associated with FEV1. This observation demonstrates that with the decline in lung function, the frequency of nocturnal cough increases. If nocturnal cough is a mechanism for sleep disturbance in clinically stable CF patients, then one would expect further sleep disruption when FEV1 approaches its nadir during pulmonary infection. However, a causal relationship between cough and sleep disturbance could not be substantiated. This indicates that cough can only occur during consciousness. Experiments performed on sleeping animals and humans have shown that cough is suppressed during sleep and occurs only following an arousal.394041 It is possible that by different mechanisms the decline in pulmonary function in CF patients lowers the threshold for awakening, thereby allowing nocturnal cough to occur.
Previous studies have shown that BMI345 and gender21342 are factors that have an important effect on the rate of pulmonary function decline in CF patients. These reports indicate that malnutrition and female gender are associated with a worse outcome. The present study indicates that in addition to these variables, sleep disturbance has a negative correlation with FEV1.
The negative association between sleep disturbance and FEV1 could be explained by two possible mechanisms. Some human studies have suggested a supportive influence of regular nocturnal sleep on the immune responses to experimental infection.43 It has been shown2526 that sleep shifts the T helper (Th) type 1/Th2 balance toward Th1 activity favoring the cellular response to infection. This activity is blocked by sleep deprivation. Some animal studies44 have shown that in mice infected with Pseudomonas aeruginosa a Th1 response was associated with faster bacterial clearance and milder lung inflammation than that in mice showing a predominant Th2 response. These findings indicate that a Th1 response might be beneficial in CF patients with chronic P aeruginosa lung infection.45 Future studies are needed to determine whether sleep disturbance in CF patients affects the frequency of pulmonary exacerbations.
The second plausible mechanism occurs through the effect of sleep disruption on glucose metabolism. Several studies have linked various sleep disturbances to glucose intolerance. Gonzalez-Ortiz et al21 showed that 1 night of sleep deprivation in healthy subjects is associated with decreased insulin sensitivity. In the context of obstructive sleep apnea, there is also evidence that the severity of the disorder is associated with insulin resistance.22 This observation raises the question of whether sleep disturbance could add to the already existing susceptibility to glucose intolerance in CF patients. Diabetes mellitus is becoming increasingly prevalent in CF patients, and is associated with the deterioration of pulmonary function and increased mortality.678910 Insulin deficiency is thought to be the primary cause for CF-related diabetes mellitus. However, there is evidence that decreased insulin sensitivity also contributes to abnormal glucose metabolism.46 The early diagnosis and treatment of CF-related diabetes mellitus is associated with improved nutrition, pulmonary function, and long-term outcomes.1112 Although our study excluded patients with diabetes mellitus, it did not include a glucose tolerance test to determine whether CF patients were in a prediabetic state.
The present study showed a very modest difference in sleep parameters between CF patients and control subjects. These results are not unexpected, given the mild degree of severity of the disease. It is, however, important to note that the tight link between sleep parameters and pulmonary function indicates that those differences are of clinical significance.
The following two parameters have been identified as occurring more frequently in CF patients than in control subjects: nocturnal use of bathroom for defecation; and watching television at bedtime. These observations indicate that additional factors such as CF-related GI disorders and inadequate sleep hygiene might also contribute to disturbed sleep in CF patients.
This study has several limitations. First, the cross-sectional design did not ascertain a causal relationship between sleep disturbance and decline in FEV1. A longitudinal design will be better suited to answering this question. Second, the exclusion of CF patients with symptoms of pulmonary infection and/or drop in FEV1 by > 10% from baseline led to the underrepresentation of patients with more severe lung disease.
In summary, the present study showed both objectively and subjectively that clinically stable pediatric patients with CF have a higher prevalence of sleep disturbance than do healthy children. Future research is needed both to examine the nature of the interaction between CF pulmonary disease and sleep, and to determine whether optimizing sleep quality in pediatric CF patients could modify the clinical course of the disease.
| Footnotes |
|---|
Received for publication January 12, 2005. Accepted for publication February 15, 2005.
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |