The Effect of Sleep Loss on Breathing in Chronic Obstructive Pulmonary Disease

The Effect of Sleep Loss on Breathing in Chronic Obstructive Pulmonary Disease* Barbara A PhiUips, M.D., F:C.C.~;t Kevin R. Cooper, M.D., and Thomas v Burke, M.D.§

F:C.C.~;*

We have previously shown that one night of sleep deprivation results in signi8cant deterioration of spirometric performance and ventilatory responsiveness to inhaled carbon dioxide in normal people. Since even a small decrease in pulmonary function may be clinically important in patients with chronic limitation of airftow, we undertook the present study to assess the effects of sleep loss on breathing in patients with chronic obstructive pulmonary disease (COPD ~ Criteria for inclusion in thisstudy were (1) a ratio of the forced expiratory volume in one second over the forced vital capacity (FEV/FVC) of less than 60 percent, (I) no hospital admission for pulmonary disease within two weeks of testing, (3) stable «30 percent variation) in tests of pulmonary function on two occasions within three months of testing, and (4) no history of asthma. We studied 15 men (mean age, 57 ± 3 years) on two consecutive mornings. Patients were studied with and without sleep deprivation in a randomized fashion. Patients were hospitalized for the study so that sleep deprivation, medications, smoking, and

diet could be monitored and enforced. We found small but statistically significant falls in FEV. (1.06 ± 0.11 to 1.00±0.09 L; p<0.05) and in FVC (2.56±O.IO to 2.43 ± 0.17 L; p
~culty in

sleeping is a common and bothersome problem, particularly in the elderly. In normal people, 24 hours of total deprivation of sleep results in a measurable decline in performance on spirometric tests' and ventilatory response to inhaled carbon dioxide'" but does not result in clinically significant impairment of respiration. Little is known about the effects of sleep loss on breathing in patients with pulmonary disease, but these individuals might show clinically important disturbances in breathing following sleep loss because their ventilatory performance is already impaired. Patients with pulmonary disease lose sleep for a variety of reasons. Coughing, which may be most severe at night due to pooling of secretions, disturbs sleep. Medications commonly prescribed for such patients, including p-adrenergic drugs, methylxanthines, and steroids, promote insomnia.t" In addition, the methylxanthines may cause esophageal reflux and dyspepsla.Y Moreover, because of concern about

the ventilatory depressant properties of sedative-hypnotic drugs, physicians may withhold these agents from their patients with pulmonary disease. Morning bronchospasm may waken such patients early. Hypoxemia and carboxyhemoglobinemia may disturb normal sleeping patterns. 9 The inevitable anxiety and depression associated with chronic pulmonary disease also results in sleepless nights. Exercise, which improves sleep, 10 may be impossible for these patients because of respiratory impairment. Finally, when patients require admission to an intensive care unit, sleep is severely deranged, usually resulting in less than two hours of sleep per night, and nearly total suppression of slow-wave and rapid-eye-movement (REM) sleep. 11 Thus, disturbance in sleep is common in patients with chronic obstructive pulmonary disease (COPD). We undertook the present study to determine the effects of a single night's loss of sleep on breathing in patients with stable, severe chronic obstruction of

*From the Pulmonary Division, Department of Medicine, University of Kentucky College of Medicine, Lexington, and Medical College of V~ia, Richmond. t Assistant Professor of Medicine, University of Kentucky. *Associate Professo~ Medical College of Virginia. IPulmonary Fellow. University of Kentucky College of Medicine. Supported by a grant from the Kentucky Lung Association. Reprint requests: Dr. Philliv" Pulmonary DWisionIMedicine, University ofKentucky Medical Center; Lexington 40536-0084

MATERIALS AND METHODS

airflow.

Patient Population We studied 15 patients, aged 52 to 65 years (mean, 57±3 years), with the clinicaldiagnosis ofCOPD (mean forced expiratory volume in one second [FEV1] , l.06±O.41 L). Requirements for the study were (1)ratio of FEV1 over forced vital capacity (FVC) ofless than 60 percent, (2) no episode of decompensated pulmonary disease within CHEST I 91 I 1 I JANUARY, 1987

29

two weeks of testing, (3) stable «30 percent variation) tests of pulmonary function on two separate occasions within three months

)f testing, and (4) no history of asthma or marked variability in symptoms. The testing protocol was approved by the institutional committees for conducting human research, and all patients gave written informed consent. Testing Protocol

Patients were admitted to the hospital and randomly assigned to be deprived of sleep or allowed to sleep on the first night of the study. Sleep deprivation was enforced by nurses who kept the patients out of bed , generally in front of the nurses' station, and woke them ifthey attempted to nap. Patients passed the time by talking, reading, and watching television. Although sleep could not be enforced or pharmacologically induced, nurses required the patients to stay in bed from 11 PM to 7 AM on the night that sleep was allowed. Each patient assigned to sleep deprivation on the first night was allowed to sleep on the second night and vice versa. Patients refrained from eaffeine-containing beverages, cigarettes, and ~-adrenergic-ago­ nistic inhalers for four hours prior to testing each day. Those taking theophylline preparations (n = 7) or steroids (n =2) were instructed to use them in the usual doses at the usual times each day. Serum levels of theophylline were measured on both days by the clinical laboratory in the seven patients receiving theophylline preparations. Patients received a standard hospital diet (2,500 kcal, 90 g of protein, 100 g of fat, and 300 g of carbohydrate). Tests were performed on two successive days; one days testing followed a night of sleep, and the other followed a sleepless night. (Previous.work shows that ventilatory disturbances following sleep loss are totally reversed by a single night of sleep. 1) Each day, subjects arrived in the laboratory within one hour of 9 AM. Technologists in the pulmonary function laboratory knew the hypothesis being tested, but not when the subjects hadbeen recently deprived of sleep. The subjects rested for at least ten minutes; oral temperatures were recorded during that time. A sample of arterial blood was drawn for blood gas analysis (Micro 13, Instrumentation Laboratories). The FVC, FEV 1, and maximal voluntary ventilation (MVV) were measured with a dry rolling-seal spirometer (Ohio Medical Products 840), or an automated pulmonary function analyzer (Maxi-Mod, Warren E. Collins). The same piece of equip-

ment was alwaysused in a given subject. Functional residual capacity (FRC) was measured by the plethysmographic method. II Ventilatory responsiveness to inhaled carbon dioxide was measured as previously described by ReadJ3 and modified by Adler," The slope of minute ventilation evE) over carbon dioxide pressure (Pco, (HCVR) was calculated using a standard linear regression equation. Rebreathing was performed in duplicate with a ten-minute to 15minute resting period between tests. The values of HCVR reported are the means of two runs on each day. Maximal inspiratory and expiratory pressures at the mouth (MIP and ME~ respectively) were measured as previously described," using a calibrated aneroid manometer. Patients were instructed to empty or fill up their lungs completely before inhaling or exhaling maximally into the mouthpiece. All patients made at least three attempts for each maneuver but were encouraged to try until they thought they had given their "best effort." (All instruments were calibrated daily.) Mean values of FVC, FEV 1, M~ arterial oxygen pressure (PaOJ, arterial carbon dioxide tension (PaC0sJ, pH, MI~ ME~ and HCVR for the sleep-deprived and rested states were compared by the one-tailed t-test because of previous work which showed that sleep loss results in a decline in pulmonary function. 1-3 RESULTS

Pulmonary Function Testing

Spirometric results are reported in Table 1. Sleep loss resulted in a significant fall in mean FVC and FEV b with changes of similar magnitude in mean MV\Z Seven of the 15 patients studied had a baseline

FEV. of 1.00 or less. Four of these experienced no change or an increase in FEV. with sleep loss; in the remaining three patients the range of fall in FEV. was 30 to 100 ml. The FRC was not changed in the five patients who had this test (7.23±1.40 L on both days).

Other Results There were no Significant changes in ventilatory

Table I-Spirometric Data FEV 1, L

FVC,L Subject 1 2

3 4 5 6 7 S 9 10 11 12 13 14 15 Mean SE

Rested 4.26 3.50 3.20

3.06 3.0S 2.65 2.58 2.45 2.30 2.30 2.20 2.00 2.00 1.50 1.34 2.56* 0.20

SleepDeprived

Percent Change

3.50 3.85 3.05 2.74 2.60 2.55 2.45 2.50 2.10 2.06 2.00 2.10 1.85 1.55 1.55 2.43* 0.17

-IS +9 -5 -10 -15 -4 -5 +2 -9 -10 -9 +S -S +3 +14 -5

M~L

Rested

SleepDeprived

Percent Change

2.08 1.25 1.20 1.60 1.10 1.15 0.96 0.60 LIS 1.28 0.88 0.72 0.85 0.65 0.46 LOOt 0.11

1.98 1.25 1.15 1.26 1.00 1.05 0.99 0.65 0.98 1.10 0.85 0.75 0.75 0.65 0.S7 LOOt 0.09

-5 0 -4 -21

-9

-9 +3 +S -15 -14 -3 +4 -12 +19 -6

FEV 1/FVC%

Rested

SleepDeprived

Rested

SleepDeprived

62.0 63.0 33.0 66.0 48.3 35.0 40.0 21.5 44.0 48.0 51.0 30.0 30.0 27.5 20.0 41.3 3.9

61.0 59.0 32.0 53.0 48.6 33.0 41.0 19.3 45.S 33.0 48.0 32.0 27.0 38.0 20.0 39.4 3.4

0.49 0.36 0.38 0.52 0.36 0.43 0.37 0.24 0.50 0.56 0.40 0.36 0.42 0.43 0.34 0.41 0.02

0.57 0.32 0.38 0.46 0.38 0.41 0.40 0.26 0.47 0.53 0.42 0.36 0.41 0.42 0.37 0.41 0.02

*p
Sleep Loss and cOPO (PhIlips, Cooper, Burlce)

Table !-ArteritJl BloodGG8 Lewls PaO., mm Hg

DISCUSSION

pH

PaCO., mm Hg

.

SleepSleepSleepSubject Rested Deprived Rested Deprived Rested Deprived 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Mean SE

75.3 88.0 73.8t 70.1 69.6t so. 0 63. It 56.6t

65.9t 64.9t BO.

58.ot 77.0 68.7t 90.1 72.1 2.6

75.0 95.0 69.6 79.5 68.5 69.0 62.7 61.1 62.5 62.0 82.0 56.0 76.0 78.5 so. 0 71.8 2.7

52.8· 37.0 44.5· 39.9 39.5 41.0 36.9 56.6· 39.1 39.9 37.0 40.0 41.0 49.0· 46.6· 42.8 1.6

49.4 38.0 47.9 42.1 42.0 42.0 43.4 61.6 35.0 39.9 39.0 39.0 42.0 44.6 41.9 43.2 1.6

7.43 7.48 7.38 7.44 7.38 7.35 7.43 7.40 7.42 7.43 7.34 7.37 7.49 7.40 7.37 7.41 0.01

7.40 7.48 7.39 7.41 7.34 7.36 7.39 7.44 7.39 7.42 7.35 7.40 7.47 7.44 7.34 7.40 0.01

·PaC022:44 mm Hg. tPaO. <70 mm Hg.

responsiveness to inhaled carbon dioxide following sleep loss. Mean VE/Pco 2 was 0.78±0.10 Umin/mm Hg (range, 0.29 to 1.62) in the rested state and O.73±O.07 Umin/mmHg (range, 0.16 to 1.29) after sleep loss. Valuesfor VEIPC02are very low; as is typical of patients with airflow obstruction. Data on blood gas levels are reported in Table 2. Means values were remarkably consistent between the rested and sleep-deprived days. The greatest individual changes were a fall in Pa02from 90.1to BO. 0 mm Hg and a rise in PaC02 from 36.9 to 43.4 mm Hg. Five individuals were hypercapnic (PaC02~44.0 mm Hg) at baseline. The PaC02 rose slightly in two of these individuals (subjects 3 and 8) following sleep loss, but pH also rose. Seven patients had baseline values for Pa02below 70 mm Hg. The greatest fall in PaOI in this group of relatively hypoxemic patients was from 61.0 to 56.0mm Hg. There was a trend toward reduced inspiratory muscular strength due to sleep loss, but the difference was not statistically significant. Mean values for MIP were 81.5 ± 8.8 vs 75.9 ± 7.6 and for MEP were 76.1± 5.8 vs 79.3±5.6 for the rested and sleep-deprived days, respectively. There were no significant changes in baseline VE (14.04±0.96 Umin rested; 13.01±0.83 Umin sleepdeprived; N = 10), oral temperature (36.9°C [98.4°F±0.I°F] rested; 36.8°C [98.2°F±0.2°F] sleep-deprived), or serum theophylline levels (8.4~glml ± 1.4~g/ml rested; 7.4~/ml ± 1.4~g/ml sleep-deprived; N = 7). Patients generally reported feeling tired following the night of sleep loss but did not subjectively feel that their breathing was affected.

The declines in FVC and FEV 1 due to sleep loss could result from a decrease in respiratory muscular performance or from increased obstruction of the airways and air trapping. Since we found no change in FEVlFVC and no increase in FRC (in the five patients tested), we believe that obstruction and air trapping are unlikely. Moreover, the reductions in FVC and FEV 1 were accompanied by slight declines in mean MVVand MI~ which although not statistically significant, are consistent with reduced muscular performance. Such a reduction in muscular performance could be due to actual weakness or to a reduction in voluntary effort. There were no correlations between changes in FVC and changes in MIP or ME~ which suggests that muscular fatigue or weakness is unlikely to be the cause of changes in FVC. Thus, although we cannot rule out an actual decrease in muscular strength due to sleep loss, a reduction in effort (ie, the patients did not try as hard in spite of identical coaching and willingness to cooperate because they were sleepy) is more likely. Normal subjects experience a 20 percent decline in the slope of HCVR followingsleep loss.1.3 Our patients with COPD experienced only a 6 percent decline (not significant), but they all had very small baseline values. Subjects with small values of HCVR are less likely to experience a change due to training" and perhaps are less responsive to other stimuli as well. Alternatively, perhaps the reduced responsiveness to carbon dioxide due to abnormal pulmonary mechanics in patients with COPD would tend to mask the more subtle effects of sleep loss. Since there were no changes in arterial blood gas levels, VE, or ventilatory response to carbon dioxide, we conclude that sleep loss had no major effect on respiratory drive, although perhaps more sensitive indices of muscular function such as diaphragmatic electromyography could yield further information. The clinical importance of the fall in FVC and FEV 1 is uncertain. We generally regard these parameters as measures of breathing ability and interpret a decline as a loss of pulmonary reserve. Patients with COPD frequently experience episodes of worsening symptoms and respiratory failure and are prevented from adequate sleep by illness. Thus, a small decline in FEV 1 and FVC due to sleep loss itself could have important clinical consequences because it occurs exactly when the patient is least able to tolerate any further challenge to his ventilatory ability. This study evaluated the effect of a single night of total sleep loss on breathing in patients with COPD. In reality, of course, patients with COPD experience chronic sleep disturbances," so that the cumulative effects of many days of sleep deprivation in such patients may be greater. Since the FRC and vital capacity are both larger in the upright than in the CHEST I 91 I 1 I JANUARY, 1987

31

supine positions," one might question whether the upright position required during the night of wakefulness may have helped to preserve pulmonary function following sleep loss; however, the patients in this study were awake and upright for at least an hour prior to each day's testing, so that persistence of posture-related changes in pulmonary function is unlikely. Furthermore, FRC was not changed by sleep loss in the five patients in whom it was measured. Marini and others" have shown that patients with severe COPD experience negligible changes in FRC with varying postures. Participants in this study refrained from the use of metered-dose inhalers for at least four hours prior to testing each day. Metaproterenol, which has a duration of action of less than three hours, to was the only type of metered-dose inhaler used by these patients. Thus, although it is possible that patients used inhalers more frequently during the night they were awake, it is unlikely that self-medication significantly influenced results of pulmonary function testing. We conclude that in patients with COPD, loss of sleep results in a decline in spirometric performance which is small but statistically significant. This small decline is probably clinically unimportant in patients with stable pulmonary disease but could be important in the setting of acute-on-chronic respiratory failure. The importance of sleep as a restorative process in maintaining ventilatory ability should be appreciated. REFERENCES

1 Cooper KR, Phillips BA. Effect of short-term sleep loss on breathing. J Appl Physioll982; 53:855-58 2 Schiffman PL, 'Irontell MC, Mazar MF, Edelman NH. Sleep deprivation decreases ventilatory response to CO 2 but not load compensation. Chest 1983; 84:695-98 3 White D~ Douglas NJ, Pickett CK, Zwillich C~ Weil Sleep deprivation and the control of ventilation. Am Rev Respir Dis 1983; 128:984-86 4 Innes 1ft, Nickerson M. Norepinephrine, epinephrine, and the sympathomimeticamines. In: Goodman LS, Gilman A, eds. The pharmacologic basis of therapeutics. 5th ed. New York: Mac-

rv

32

millan Publishing Co, Inc, 1975:477-511 5 Rhind GB, Connaughton JJ, McFie J, Douglas NJ, Flenley DC. Sustained release choline theophyllinate in nocturnal asthma. Br Med J 1985; 291:1605-07 6 Bukowskyj M, Nakatsu K, Munt PW Theophylline reassessed. Ann Intern Med 1984; 101:63-73 7 Haynes RC, Lamer J. Adrenocorticotropic hormone; adrenocortical steroids and their synthetic analogs; inhibitors of adrenocortical steroid biosynthesis. In: Goodman LS, Gilman A, eds. The pharmacologic basis of therapeutics. 5th 00. New York: Macmillan Publishing Co, Inc, 1975:1472-506 8 Abramowicz M, 00. Drugs for asthma. Moo Letter 1982; 24:83-6 9 Pappenheimer JR. Hypoxic insomnia: effects of carbon monoxide and acclimatization. J Appl Physiol: Respir Environ Exercise Physioll984; 57:1696-703 10 How to get a better night's sleep. In: Lamberg L. The American Medical Association guide to better sleep. New York: Random House, 1984, chap 15 11 Aurell J, Elmquist D. Sleep in the surgical intensive care unit: continuous polygraphic recording of sleep in nine patients receiving postoperative care. Br Moo J 1985; 290:1029-32 12 DuBois AB, Botelho SY, Bedell GN, Marshall R, Com roe J8. A rapid plethysmographic method for measuring thoracic gas volume: a comparison with a nitrogen washout method for measuring functional residual capacity in normal subjects. J Clin Invest 1956; 35:322-26 13 Read OJ. A clinical method for assessing the ventilatory response to carbon dioxide. Australas Ann Med 1966; 16:20-32 14 Adler JJ. A simplified method for evaluating the ventilatory response to carbon dioxide. Am Rev Respir Dis 1973; 108: 1449-50 15 Black LF, Hyatt RE. Maximal respiratory pressures: normal values and relationship to age and sex. Am Rev Respir Dis 1969; 99:696-702 16 Cooper KR, Phillips BA. Learning effect of repeated hypercapnic ventilatory response testing. Am J Med Sci 1986; 291:386-90 17 Klink ME, Quan SF: Prevalence of sleep disturbance and its relationship to chronic respiratory disease in a general population. Am Rev Respir Dis 1986; 133:AI49 18 Structure, expansion, and movement of the lung. In: Cotes JE. Lung function assessment and application in medicine. Oxford, England: Blackwell Scientific Publications, 1979 19 Marini JJ, Tyler ML, Hudson LD, Davis BS, Huseby JS. Influence of head-dependent positions on lung volume and oxygen saturation in chronic airHow obstruction. Am Rev Respir Dis 1984; 129:101-05 20 Passamonte PM, Martinez AJ. Effect of inhaled atropine or metaproterenol in patients with chronic airway obstruction and therapeutic serum theophylline levels. Chest 1984; 85:610-15

Sleep Loss and COPO(Phillips, Cooper, Burlce)