Cystic Fibrosis Patients Have Poor Sleep Quality Despite Normal Sleep Latency and Efficiency

Cystic Fibrosis Patients Have Poor Sleep Quality Despite Normal Sleep Latency and Efficiency* Larry Jankelowitz, MD; Kathryn J. Reid, MD; Lisa Wolfe, MD; Joanne Cullina, MD; Phyllis C. Zee, MD; and Manu Jain, MD, FCCP

Study objectives: Cystic fibrosis (CF) patients may be predisposed to poor sleep quality due to upper and lower airway abnormalities and impaired gas exchange. Previous sleep investigations of CF patients using single-night polysomnography have reported conflicting results. We hypothesized that sampling sleep for a prolonged period in a patient’s normal environment may give a more representative assessment of sleep quality than a single-night polysomnogram, and that impaired sleep quality would correlate with pulmonary disease severity and self-assessed sleep quality. Design: Using wrist actigraphy, we measured sleep quality in clinically stable CF patients and age-matched control subjects. In addition, each CF patient and control subject completed the following three questionnaires: the Epworth sleepiness scale; the Pittsburgh sleep quality index (PSQI); and the Medical Outcomes Study 36-item short form. Results: Twenty CF patients and control subjects were enrolled in the study, and were well-matched for age, sex, and body mass index. The mean (ⴞ SD) FEV1 for CF patients was 61.0 ⴞ 20.1% predicted. CF patients and control subjects had similar sleep duration, sleep latency, and sleep efficiency. However, CF patients had higher PSQI scores (6.45 vs 4.55, respectively; p ⴝ .04), a higher fragmentation index (FI) [31.72 vs 18.02, respectively; p < 0.001], and less immobile time (88.87 vs 91.89, respectively; p ⴝ 0.02). There was a significant correlation of FI with FEV1 and PSQI scores. Conclusions: Stable CF patients have disrupted sleep, and sleep disruption may in part be related to the severity of pulmonary disease. In addition, the PSQI may be useful in detecting CF patients with poor sleep quality. (CHEST 2005; 127:1593–1599) Key words: actigraphy; cystic fibrosis; fragmentation; Pittsburgh sleep quality index; sleep Abbreviations: AI ⫽ arousal index; BMI ⫽ body mass index; CF ⫽ cystic fibrosis; CI ⫽ confidence interval; ESS ⫽ Epworth sleepiness scale; FI ⫽ fragmentation index; IT ⫽ immobile time; MAS ⫽ mean activity score; PSQI ⫽ Pittsburgh sleep quality index; SF-36 ⫽ Medical Outcomes Study 36-item short form; WA ⫽ wrist actigraphy

fibrosis (CF) is the most common, lifeC ystic shortening, genetic disorder in white individuals, with a median survival time of 33 years.1,2 Most CF patients develop bronchiectasis with airflow obstruction and air-trapping, which can lead to hypoxemia, *From the Departments of Medicine (Drs. Jankelowitz, Wolfe, Cullina, and Jain) and Neurology (Drs. Reid and Zee), Division of Pulmonary and Critical Care Medicine, Northwestern University Medical School, Chicago, IL. Grant Support: Pulmonary Critical Care Divisional Funds. Manuscript received June 22, 2004; revision accepted November 22, 2004. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Manu Jain, MD, Assistant Professor of Medicine, Division of Pulmonary and Critical Care Medicine, Northwestern University, 240 East Huron St, McGaw Pavilion second floor, Chicago, IL 60611; e-mail: m-jain@northwestern. edu www.chestjournal.org

hypercapnia, and cor pulmonale. Upper airway manifestations of CF include chronic sinusitis and nasal polyposis, which can increase upper airway resistance. Abnormalities of lung mechanics, gas exchange, and cardiac function may lead to sleep abnormalities that are similar to those seen in patients with asthma and COPD.3–9 Other contributors to CF sleep abnormalities may include the use of certain medications,10,11 chronic anxiety, depression,12 and vitamin deficiencies.13 Sleep disruption may significantly impair daytime functioning and health-related quality of life.14 Health consequences may include impaired immune responses and metabolic abnormalities, as well as endocrine and cardiovascular dysfunction15–19. Thus, sleep disruption could potentially exacerbate symptoms and worsen prognosis in an already comproCHEST / 127 / 5 / MAY, 2005

1593

mised population such as CF patients. In order to optimize the complex medical management of CF patients, it is important to understand the magnitude of sleep disruption in this population and to identify the predictors of at-risk patients in whom further evaluation may be indicated. Previous reports on sleep disturbances in CF patients have been conflicting. Some studies20 –22 have found CF patients to have reduced total sleep time, whereas others23 found no differences in sleep time when compared with healthy control subjects. Moreover, some reports found CF patients to have poor sleep efficiency22,23 and increased sleep latency,22 while others found no differences20,21 compared to control subjects. Thus, the presence and/or extent of sleep disruption in CF patients seems unclear. Conflicting results in the past could be in part due to the use of a single overnight polysomnogram, which is considered to be the “gold standard” for assessing sleep quality. Pulmonary symptoms and exacerbations are episodic in the CF population. Polysomnography may thus be limited in its usefulness because a single-night study in a laboratory may not accurately reflect sleep quality in a home environment over a period of time. A commonly used alternative to polysomnography for recording sleep and wakefulness is activity monitoring. Sleep-wake activity determined by activity monitoring correlates relatively highly with polysomnography in many healthy and patient populations.24 –26 An advantage of activity monitoring is that it allows the continuous assessment of both wake and sleep behavior in the patient’s own environment over extended periods of time. Using an activity monitor, we sought to determine whether sleep quality (eg, sleep duration, sleep efficiency, and fragmentation index [FI]) in CF patients with varying severities of lung disease is diminished compared to age-matched control subjects. We also wished to ascertain whether sleep parameters correlated with pulmonary disease severity, and validated questionnaires assessing quality of life and sleep. Materials and Methods Subjects CF patients who were ⱖ 18 years of age were recruited from the adult CF clinic at Northwestern University from January 2002 to April 2003. CF diagnoses had been confirmed by an abnormal sweat chloride test result and/or the presence of two disease-causing mutations. Patients were ineligible if they had received oral or IV antibiotics for a pulmonary exacerbation in the 8 weeks prior to enrollment or had primary cardiac disease, psychiatric disorders, or a known sleep disorder. If a patient 1594

experienced a pulmonary exacerbation during the study, that patient was excluded from the study. Healthy, age-matched control subjects were enrolled in the same study protocol. Control subjects were excluded if they had any history of medical or sleep disorders, or consumed excessive quantities of alcohol or caffeine. The study was approved by Northwestern University Institutional Review Board, and informed consent was obtained for all participants. Study Design Age, sex, height, and weight were obtained on the day of study enrollment. Each CF patient and control subject completed spirometry (V˙max 20 C-series; SensorMedics; Yorba Linda, CA; and Morris-Polgar prediction equations) and three questionnaires, the Pittsburgh sleep quality index (PSQI), Epworth sleepiness scale (ESS), and the Medical Outcomes Study 36-item short form (SF-36), on the day of study enrollment. Patients and subjects were given a form for keeping a sleep log for 14 days. The PSQI is a self-administered questionnaire that measures sleep quality during the previous month, and contains 19 selfrated questions yielding both a global score and seven “component” subscores. A global PSQI score of ⬎ 5 indicates that a person is a “poor sleeper.”27 A subjective assessment of sleepiness was performed using the ESS questionnaire, which is a reliable and validated, self-administered questionnaire for measuring daytime sleepiness in adults. ESS scores significantly correlate with sleep latency (an indication of sleepiness) measured during multiple sleep latency testing and overnight polysomnography.28,29 Patients are considered to be sleepy if they have an ESS score of ⬎ 10. The SF-36 is a quality-of-life questionnaire, yielding scores for eight domains, as well as two summary scores, a mental component summary score and a physical component summary score.30 This measure is constructed to satisfy the minimum psychometric standards that are necessary for group comparisons involving generic health concepts that are not specific to any age, disease, or treatment group. The eight domains consist of physical functioning, role limitations-physical, bodily pain, general health perceptions, vitality, social functioning, role limitations-emotional, and mental health. Each of the eight domains is scored on a scale of 100 points, with higher scores indicating better functioning. The scores are standardized to a mean (SD) of 50 in the general US adult population and can therefore be used to determine whether a group or individual in question scores above or below the US average. Scores above and below 50 indicate above and below average functioning, respectively.30 Wrist Actigraphy Wrist actigraphy (WA) detects physical motion, is a validated measure of sleep, and compares favorably to polysomnography.24 –26 CF patients and control subjects wore a wrist actigraph (Mini Motionloggers or Actillume; Ambulatory Monitoring Co; Ardsley, NY) for 14 days. Data stored in memory was downloaded and analyzed using a commercially available program (Action 3; Ambulatory Monitoring Co) with the actigraph worn on the nondominant arm. A sample actogram is shown in Figure 1. The program can score activity/inactivity and derive from it periods of sleep or wakefulness using a computer-generated algorithm. Sleep log data are used in the computer algorithm to calculate sleep latency. It is also used in conjunction with activity monitoring to confirm bedtimes.25 WA data were analyzed cumulatively for the 2-week period during which subjects wore the actigraph and then separately for weekdays (ie, Sunday to Thursday) and weekends (ie, Friday and Saturday). There are Clinical Investigations

Figure 1. Sample actogram for a 20-day period demonstrating the change in activity from awake to sleep.

several measures that can be derived from these algorithms, such as the FI, mean activity score (MAS), and immobile time (IT). The FI measures restlessness and is calculated by summing the percentage of minutes spent moving with the percentage time spent in the immobility phase per minute. MAS is the magnitude of activity on a per-epoch basis during sleep, and it is calculated by dividing the total activity score by the number of epochs during the assumed sleep period. IT is derived by dividing the number of minutes spent immobile by the assumed sleep time and multiplying by 100 (Action 3; Ambulatory Monitoring Co). Statistical Analysis Sample means and SDs were calculated for all variables. The independent sample Student t test for the comparison of two groups for all continuous variables was performed. (SPSS for Windows, version 11.5; SPSS; Chicago, IL). There are accepted cut points for the ESS and PSQI, indicating abnormal sleep parameters. For this reason, we performed ␹2 analysis to compare groups for ESS and PSQI. A p value of ⬍ 0.05 was considered to be significant. A Pearson correlation coefficient was calculated to determine the relationship of continuous variables. We performed jackknife estimates of correlation coefficients (obtained by eliminating one x-y pair at a time), which provide a robust estimate of the underlying linear relationship (correlation) between variables of interest.31 A 95% confidence interval (CI) was calculated, which provides the most plausible range for jackknife correlation coefficient under assumption of repeated sampling. www.chestjournal.org

Results Twenty-one CF patients and control subjects were initially enrolled into the study, of whom 20 individuals in each group completed the study. The CF and control groups were well-matched for age, sex, and body mass index (BMI) [Table 1]. The airflow of CF patients ranged from normal to severe obstruction, with a mean (⫾ SD) FEV1 of 61.0 ⫾ 20.1% predicted (95% CI, 31 to 91% predicted). None of the CF patients were receiving home oxygen therapy, and their mean oxygen saturation at the time of study enrollment was 96 ⫾ 1% predicted. Questionnaire Data The mean PSQI for CF patients was 6.45 ⫾ 3.31, which was significantly higher than that for control subjects (4.55 ⫾ 2.21; p ⫽ .04). Furthermore, in the CF group there were 11 patients with a PSQI score of ⬎ 5, compared to only 5 subjects in the control group (p ⫽ .04). There was also a trend toward higher sleep disturbance (p ⫽ 0.07) and habitual sleep efficiency (p ⫽ 0.06) PSQI subscores in the CF group, although this did not reach statistical signifiCHEST / 127 / 5 / MAY, 2005

1595

Table 1—Demographics and Questionnaires* Variables Age, yr Sex Male Female BMI FEV1, % predicted ESS PSQI SF-36 PCS MCS

CF Patients (n ⫽ 20)

Control Subjects (n ⫽ 20)

25.85 (6.00)

25.35 (3.72)

8 12 23.2 (4.44) 61 (20.1) 6.75 (3.32) 6.45 (3.31)

8 12 24.27 (3.17) NA 5.80 (3.52) 4.55 (2.21)

0.39 NA 0.39 0.04

44.90 (10.12) 45.20 (9.78)

56.95 (3.30) 51.95 (7.83)

⬍ 0.001 0.02

p Value 0.75 1.0

*Values given as mean (SD), unless otherwise stated. PCS ⫽ physical component score; MCS ⫽ mental component score; NA ⫽ not applicable.

were no differences between weekdays and weekends for these parameters in either group (data not shown). Compared to control subjects, CF patients also displayed greater night-to-night variability in FI (p ⬍ 0.003) and IT (p ⬍ 0.02), as evidenced by greater SDs when each night was assessed separately (Table 2). These findings are consistent with the overall more disrupted sleep for CF patients. CF patients’ lung function, as assessed by FEV1, had small but significant associations with both the FI and the IT (Fig 2). The confidence interval calculated by jackknife analysis31 suggests a robust relationship between decrement in lung function and the severity of sleep disturbance. In contrast, there were no significant associations between resting room air O2 saturation or BMI and any measured sleep parameter (data not shown).

cance. The mean ESS scores for the CF and control groups were 6.75 ⫾ 3.32 and 5.72 ⫾ 3.63, respectively (p ⫽ 0.39), with four CF patients and three control subjects having scores of ⬎ 10. The SF-36 scores for the CF patients were significantly lower for both the physical and mental components compared to our healthy control subjects as well as the population means.32 WA Data Control and CF patients had similar total sleep time, sleep latency, and sleep efficiency. Sleep onset and offset times were also comparable, meaning that CF patients and control subjects were going to bed and waking up at approximately the same times. Both CF patients and control subjects went to sleep an average of 1 h later on the weekends than during the week. The total activity score and MASs were also similar. In contrast, CF patients had a higher FI and less IT than control subjects (Table 2). There

Table 2—WA Data* Variables

CF (n ⫽ 20)

Sleep onset 12:52 am Awakening time 8:29 am Sleep duration, h 7.52 (.74) Sleep latency, min 12.75 (14.77) Sleep efficiency, % 85.2 (6.15) IT, % 88.87 (4.99) Total activity score 8,429 (3,330) Mean activity score 9.13 (3.78) FI 31.72 (12.40) Night-to-night IT 4.97 (1.03) variability Night-to-night FI 12.91 (1.64) variability

Controls (n ⫽ 20)

p Value

12:49 am 8:01 am 7.1 (.94) 13.85 (10.05) 85.4 (5.60) 91.89 (2.35) 6,908 (4,110) 7.68 (3.88) 18.04 (6.22) 2.17 (0.31)

0.91 0.36 0.12 0.79 0.91 0.02 0.21 0.24 ⬍ 0.001 ⬍ 0.02

6.81 (0.84)

*Values given as mean (SD), unless otherwise indicated. 1596

⬍ 0.003

Figure 2. Top, A: correlation coefficients and CIs for FI with FEV1. The Pearson correlation coefficient between the FI on WA and FEV1 was ⫺0.26 (95% CI, ⫺0.28 to ⫺0.25). Bottom, B: correlation coefficients and CIs for IT with FEV1. The Pearson correlation coefficient between the FI on WA and FEV1 was 0.27 (95% CI, 0.24 to 0.28). Clinical Investigations

Comparison of Activity and Questionnaire Data There were significant correlations for FI, IT, and MAS with the PSQI for CF subjects, but not for control subjects (Fig 3). In contrast, there were no significant correlations between the ESS and SF-36 instruments and any sleep parameter measured by WA (data not shown).

Figure 3. Top, A: correlation between MAS and PSQI. There was a statistically significant positive relationship between the MAS and PSQI. The Pearson correlation coefficient between the MAS on WA and the PSQI was 0.80. Middle, B: correlation between IT and PSQI. There was a statistically significant inverse relationship between the IT and PSQI. The Pearson correlation coefficient between the IT on WA and the PSQI was ⫺0.64. Bottom, C: correlation between FI and PSQI. There was a statistically significant inverse relationship between the FI and PSQI. The Pearson correlation coefficient between the FI on WA and the PSQI was ⫺0.48. www.chestjournal.org

Discussion Total sleep time, sleep efficiency, and sleep latency were similar between CF patients and agematched control subjects. However, sleep was significantly more fragmented, with more night-to-night variability. Self-perceived sleep quality was also poor in CF patients. In addition, there is evidence that the degree of sleep disruption may be related to pulmonary disease severity (as measured by FEV1). Previous investigations20 –23 in CF patients using polysomnography have reported conflicting results in sleep parameters including, sleep efficiency, sleep latency, and arousal index (AI). In the current study, using wrist activity monitoring, we found no differences in sleep efficiency and latency between control subjects and CF patients. In addition, sleep duration was 2 h longer in our CF patients than previously reported.20,22 In contrast to control subjects, CF patients had a higher FI and a lower IT (ie, they spent less time lying still), which is consistent with more disrupted sleep. The FI, unlike sleep efficiency, measures the number of times an individual moves rather than the percentage of time that they spend awake during the sleep period. The FI may be a more useful measure to examine sleep disturbance because it may better detect repetitive short events that disrupt sleep. The discrepancies between our results and previous findings may be due at least in part to methodological differences. WA is less cumbersome and can be performed in a subject’s natural sleep environment, as opposed to polysomnography, which is usually performed in a sleep laboratory. In addition, we sampled sleep for a 2-week period compared to a single night, which is typical for studies using polysomnography. By measuring sleep for 2 weeks in the home environment, we think that our study offers a more representative assessment of sleep-wake habits of CF patients and age-matched control subjects, especially in light of the greater night-to-night variability displayed by our CF subjects. While CF patients have more disrupted sleep and rate their sleep as being worse than control subjects on the PSQI, they do not report greater sleepiness on the ESS. Self-assessed ratings such as the ESS, however, can have limited correlation to objective measures of daytime sleepiness.33,34 Therefore, the increased sleep fragmentation observed in the CF patients may still have consequences for daytime function. Future studies of CF patients should evaluate the relationship between sleep fragmentation and objective measures of daytime sleepiness such as the multiple sleep latency test and/or neurocognitive performance. Compared to previous studies of sleep in CF CHEST / 127 / 5 / MAY, 2005

1597

patients, our study included patients with milder disease. Our patients had an average FEV1 of 61% predicted and a BMI of 23, both of which are higher than those values in CF patients, in whom sleep data has been previously reported.20 –23 Nearly two thirds of CF patients who are ⱖ 18 years of age have an FEV1 of ⬍ 70% predicted.2 Our results need to be validated in a larger study but suggest that a significant proportion of CF patients may experience sleep disruption, which would put them at risk for reduced neurocognitive function, and potentially poorer health and sense of well being.22 There is limited information about sleep fragmentation in CF patients and how this relates to measures of pulmonary disease severity.21,22 While daytime hypoxemia predicts nocturnal oxyhemoglobin desaturation in CF patients,21,35 and nocturnal hypoxemia correlated with sleep efficiency in one study,22 oxygen supplementation has not resulted in improved sleep quality.20,36,37 In another study, maximum expiratory strength correlated with rapid eye movement respiratory disturbance index scores, but spirometric function and arterial blood gas levels did not correlate with the respiratory disturbance index scores.21 This suggests that lung function and gas exchange are not good predictors of sleep fragmentation but respiratory muscle strength may be. However, in our study the FEV1 correlated with actigraphically recorded FI, a measure of disrupted sleep. One reason for this apparent discrepancy may be that prior studies did not assess the relationship between pulmonary function and the AI. The AI is the one polysomnography parameter that is most closely analogous to the FI in our study. Previous studies38,39 have shown inverse correlations between PSQI scores and polysomnographymeasured sleep efficiency. When combined with our results, the PSQI appears to be a useful screening tool to identify CF patients who are at risk for sleep disturbance, especially if they have a decreased FEV1. Compared to polysomnography, the PSQI is simple to perform, inexpensive, could be followed over time, and does not involve a hospital stay. If the PSQI suggests poor sleep quality, then WA and/or polysomnography studies may be obtained for further evaluation of the underlying sleep disorders. Given the night-to-night variability observed in CF patients and the relative cost of polysomnography, it may be more accurate and cost-effective to obtain wrist activity monitoring. That many CF patients experience poor sleep is clear from the present study, but the underlying etiology of the sleep disturbance is not. Possibilities include nocturnal coughing,40,41 the use of ␤2-agonists before sleep, increased work of breathing,42,43 psychiatric disorders, and/or periodic leg move1598

ments. While our data do not allow us to distinguish between these possibilities, they do suggest avenues for future investigation. Future studies evaluating these potential mechanisms may suggest therapeutic interventions (eg, cough suppression, anxiolytics, bilevel positive airway pressure ventilation), which could be tested in clinical trials. There are potential limitations of this study. The sample size was relatively small, which may limit the generalizability of these findings and should be confirmed in a larger study. A limitation of WA is the inability to detect differences between quiet wakefulness and sleep, resulting in an underestimation of sleep disruption or duration. However, in the current study CF patients had greater sleep disruption than control subjects. There is also the potential to overestimate total sleep time if subjects are lying quietly awake, which would lead to an overestimation of sleep efficiency. Based on these considerations, CF patients may actually have a greater number of sleep abnormalities than those measured in our study. While we do not have concomitant polysomnography data to confirm the WA data, several previous studies25 in other patient populations have successfully used activity monitoring to characterize sleep-wake characteristics. Sleep fragmentation and decreased sleep quality are common in CF patients and likely to contribute to decreased daytime function and quality of life. It is not known whether improving sleep quality can reduce disease severity or associated health problems in this population. In light of the consequences of prolonged sleep loss/sleep disruption on health and well-being, the identification and treatment of sleep disorders in CF patients should be an integral part of disease management. ACKNOWLEDGMENT: This study is dedicated to the memory of Steven Baker, who originated the idea for this study and provided immeasurable support. We wish to thank Borko Jovanovic for statistical support.

References 1 FitzSimmons SC. The changing epidemiology of cystic fibrosis. J Pediatr 1993; 122:1–9 2 Foundation CF. Patient registry 2001 annual data report. Bethesda, MD: Cystic Fibrosis Foundation, 2001 3 Guilleminault C, Cummiskey J, Motta J. Chronic obstructive airflow disease and sleep studies. Am Rev Respir Dis 1980; 122:397– 406 4 Fleetham J, West P, Mezon B, et al. Sleep, arousals, and oxygen desaturation in chronic obstructive pulmonary disease: the effect of oxygen therapy. Am Rev Respir Dis 1982; 126:429 – 433 5 Wynne JW, Block AJ, Hemenway J, et al. Disordered breathing and oxygen desaturation during sleep in patients with chronic obstructive lung disease (COLD). Am J Med 1979; 66:573–579 6 Calverley PM, Brezinova V, Douglas NJ, et al. The effect of Clinical Investigations

7 8 9

10

11 12 13 14 15 16 17 18 19 20

21 22 23 24 25

oxygenation on sleep quality in chronic bronchitis and emphysema. Am Rev Respir Dis 1982; 126:206 –210 Cormick W, Olson LG, Hensley MJ, et al. Nocturnal hypoxaemia and quality of sleep in patients with chronic obstructive lung disease. Thorax 1986; 41:846 – 854 Fitzpatrick MF, Martin K, Fossey E, et al. Snoring, asthma and sleep disturbance in Britain: a community-based survey. Eur Respir J 1993; 6:531–535 Janson C, De Backer W, Gislason T, et al. Increased prevalence of sleep disturbances and daytime sleepiness in subjects with bronchial asthma: a population study of young adults in three European countries. Eur Respir J 1996; 9:2132–2138 Avital A, Sanchez I, Holbrow J, et al. Effect of theophylline on lung function tests, sleep quality, and nighttime Sao2 in children with cystic fibrosis. Am Rev Respir Dis 1991; 144:1245–1249 Lipsky BA, Baker CA. Fluoroquinolone toxicity profiles: a review focusing on newer agents. Clin Infect Dis 1999; 28:352–364 Thase ME. Depression, sleep, and antidepressants. J Clin Psychiatry 1998; 59(suppl):55– 65 Williams DC. Periodic limb movements of sleep and the restless legs syndrome. Va Med Q 1996; 123:260 –265 Moyer CA, Sonnad SS, Garetz SL, et al. Quality of life in obstructive sleep apnea: a systematic review of the literature. Sleep Med 2001; 2:477– 491 Spiegel K, Sheridan JF, Van Cauter E. Effect of sleep deprivation on response to immunization. JAMA 2002; 288: 1471–1472 Tasali E, Van Cauter E. Sleep-disordered breathing and the current epidemic of obesity: consequence or contributing factor? Am J Respir Crit Care Med 2002; 165:562–563 Van Cauter E, Copinschi G. Interrelationships between growth hormone and sleep. Growth Horm IGF Res 2000; 10(suppl):S57–S62 Spiegel K, Leproult R, Van Cauter E. Impact of sleep debt on metabolic and endocrine function. Lancet 1999; 354:1435– 1439 Lanfranchi PA, Somers VK. Sleep-disordered breathing in heart failure: characteristics and implications. Respir Physiol Neurobiol 2003; 136:153–165 Milross MA, Piper AJ, Norman M, et al. Low-flow oxygen and bilevel ventilatory support: effects on ventilation during sleep in cystic fibrosis. Am J Respir Crit Care Med 2001; 163:129 – 134 Milross MA, Piper AJ, Norman M, et al. Predicting sleepdisordered breathing in patients with cystic fibrosis. Chest 2001; 120:1239 –1245 Dancey DR, Tullis ED, Heslegrave R, et al. Sleep quality and daytime function in adults with cystic fibrosis and severe lung disease. Eur Respir J 2002; 19:504 –510 Bradley S, Solin P, Wilson J, et al. Hypoxemia and hypercapnia during exercise and sleep in patients with cystic fibrosis. Chest 1999; 116:647– 654 Sadeh A, Hauri PJ, Kripke DF, et al. The role of actigraphy in the evaluation of sleep disorders. Sleep 1995; 18:288 –302 Ancoli-Israel S, Cole R, Alessi C, et al. The role of actigraphy in the study of sleep and circadian rhythms. Sleep 2003; 26:342–392

www.chestjournal.org

26 Littner M, Kushida CA, Anderson WM, et al. Practice parameters for the role of actigraphy in the study of sleep and circadian rhythms: an update for 2002. Sleep 2003; 26:337– 341 27 Buysse DJ, Reynolds CF III, Monk TH, et al. The Pittsburgh sleep quality index: a new instrument for psychiatric practice and research. Psychiatry Res 1989; 28:193–213 28 Johns MW. Reliability and factor analysis of the Epworth sleepiness scale. Sleep 1992; 15:376 –381 29 Johns MW. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep 1991; 14:540 –545 30 Ware JE Jr, Gandek B. Overview of the SF-36 Health Survey and the International Quality of Life Assessment (IQOLA) Project. J Clin Epidemiol 1998; 51:903–912 31 Efron B. The jackknife, the bootstrap and other resampling plans. Philadelphia, PA: Society for Industrial and Applied Mathematics, 1982 32 Ware JE Jr, Kosinski M, Bayliss MS, et al. Comparison of methods for the scoring and statistical analysis of SF-36 health profile and summary measures: summary of results from the Medical Outcomes Study. Med Care 1995; 33: AS264 –A279 33 Sangal RB, Sangal JM. Rating scales for inattention and sleepiness are correlated in adults with symptoms of sleep disordered breathing syndrome, but not in adults with symptoms of attention-deficit/hyperactivity disorder. Sleep Med 2004; 5:133–135 34 Pelin Z, Karadeniz D, Ozturk L, et al. The role of mean inspiratory effort on daytime sleepiness. Eur Respir J 2003; 21:688 – 694 35 Versteegh FG, Bogaard JM, Raatgever JW, et al. Relationship between airway obstruction, desaturation during exercise and nocturnal hypoxaemia in cystic fibrosis patients. Eur Respir J 1990; 3:68 –73 36 Tepper RS, Skatrud JB, Dempsey JA. Ventilation and oxygenation changes during sleep in cystic fibrosis. Chest 1983; 84:388 –393 37 Spier S, Rivlin J, Hughes D, et al. The effect of oxygen on sleep, blood gases, and ventilation in cystic fibrosis. Am Rev Respir Dis 1984; 129:712–718 38 Backhaus J, Junghanns K, Broocks A, et al. Test-retest reliability and validity of the Pittsburgh Sleep Quality Index in primary insomnia. J Psychosom Res 2002; 53:737–740 39 Milross MA, Piper AJ, Norman M, et al. Subjective sleep quality in cystic fibrosis. Sleep Med 2002; 3:205–212 40 Stokes DC, McBride JT, Wall MA, et al. Sleep hypoxemia in young adults with cystic fibrosis. Am J Dis Child 1980; 134:741–743 41 Francis PW, Muller NL, Gurwitz D, et al. Hemoglobin desaturation: its occurrence during sleep in patients with cystic fibrosis. Am J Dis Child 1980; 134:734 –740 42 O’Neill S, Leahy F, Pasterkamp H, et al. The effects of chronic hyperinflation, nutritional status, and posture on respiratory muscle strength in cystic fibrosis. Am Rev Respir Dis 1983; 128:1051–1054 43 Marks J, Pasterkamp H, Tal A, et al. Relationship between respiratory muscle strength, nutritional status, and lung volume in cystic fibrosis and asthma. Am Rev Respir Dis 1986; 133:414 – 417

CHEST / 127 / 5 / MAY, 2005

1599