Keywords
continuous positive airway pressure - obstructive sleep apnea - surveys and questionnaires
Introduction
Obstructive sleep apnea (OSA) is a common sleep disorder that affects more than 900
million adults aged between 30 and 69 years worldwide.[1] It is characterized by pathophysiological features, such as repetitive episodes
of breathing cessation during sleep, resulting in intermittent and/or sustained hypoxemia
and sleep disruption.[2] These pathophysiological features are closely associated with several comorbidities
of OSA, including hypertension, cardiovascular diseases, arrhythmias, stroke, and
neurocognitive disorders.[3]
[4] Continuous positive airway pressure (CPAP) therapy, which is the gold standard treatment
for OSA, has favorable outcomes, such as short and long-term decreased blood pressure[5]
[6]
[7] and reduced stroke risk,[8] depending on good adherence to this therapy. Meanwhile, the effect of CPAP therapy
on cardiovascular events is controversial.[9]
[10]
The pathophysiological features of OSA also cause sleep-related symptoms, including
snoring, insomnia, sleep restlessness, excessive daytime sleepiness, falling asleep,
and headache. Excessive daytime sleepiness adversely affects everyday activities and
is associated with an increased risk of motor vehicle and occupational accidents.[4] Moreover, it increases the risk of mortality in elderly patients with OSA.[11] Excessive daytime sleepiness can be assessed by objective measures, such as the
Maintenance of Wakefulness Test (MWT) and the Multiple Sleep Latency Test (MSLT),
and self-reported (subjective) measures, such as the Epworth Sleepiness Scale (ESS),
the Basic Nordic Sleepiness Questionnaire, and the Stanford Sleepiness Scale.[12]
[13] The ESS is the most widely used tool for assessing subjective sleepiness. The ESS
scores do not correlate well with OSA severity,[14]
[15]
[16] but correlate closely with the MWT and MSLT.[16] The poor correlation of the ESS with OSA may be partially due to the heterogeneity
of OSA.
Regarding the heterogeneity of OSA, patients with OSA experience combined sleep-related
symptoms and comorbidities, changes in apnea–hypopnea index (AHI), and symptoms in
response to CPAP therapy. Furthermore, they present with different outcomes after
CPAP therapy.[17]
[18] Thus, in recent years, cluster analysis based on sleep-related symptoms is performed
to characterize the disease subtypes of OSA, such as disturbed sleep, minimally symptom,
and excessive daytime sleepiness.[18]
[19]
[20]
[21]
[22]
[23] Attention has been paid to the subtype of excessive daytime sleepiness because this
subtype is associated with a higher risk of cardiovascular events.[19]
[20]
[21]
[22] However, it was not simply classified using the ESS.[20]
[24] In addition, the sleep-related symptoms that should be included in the cluster analysis
of OSA are unknown. Previous studies showed that sleep restfulness was associated
with a risk of cardiovascular diseases,[25] and it influenced mortality according to age and sleep duration.[26] Thus, sleep restfulness is an important sleep-related symptom in the general population.
However, it is rarely included in sleep-related questions in cluster analysis of OSA.[23] Thus, the role of sleep restfulness in patient with OSA is unknown. We hypothesized
that sleep restfulness closely correlates with adherence to CPAP therapy in patients
with OSA. Thus, the current study examined the role of sleep restfulness in patients
with OSA undergoing CPAP therapy.
Materials and Methods
Participants
We administered sleep-related questionnaires to 775 patients with OSA who received
CPAP therapy from April 2022 to July 2022 and who were followed-up at Kyoto University
Hospital. The present study was approved by the Ethics Committee of Kyoto University
Graduate School and Faculty of Medicine (approval no. R3618, approval date: July 22,
2022). The need to obtain a written informed consent for all participants was waived
off.
Data on clinical characteristics, including age, body mass index (BMI), smoking status,
comorbidities (hypertension, dyslipidemia, diabetes mellitus, cardiovascular diseases,
arrhythmia, stroke, and epilepsy), and AHI before CPAP therapy initiation, and duration
after CPAP therapy initiation, were obtained from the medical records.
CPAP Therapy and Devices
Data on CPAP therapy, such as CPAP devices (Philips Respironics or ResMed device),
pressure settings (autoadjusting or fixed), and interfaces, were obtained from the
medical records. Information on AHI, CPAP usage time, and mask leak was obtained using
Encore Pro 2 or Care Orchestrator (Philips Respironics), ResScan (ResMed), NemLink
(Teijin Pharma Ltd), or f'Rens (Fukuda Densi Ltd). Data were downloaded for 1 month
before the sleep-related questionnaires were administered. Adherence to CPAP therapy
(good adherence) was defined as the use of CPAP therapy for at least 4 h a night in
70% of nights for 1 month.[27] Residual AHI was defined as AHI ≥ 10 events/hour.[27] Large leaks were defined as large leaks lasting > 1 hour if Philips Respironics
devices were used;[27] and a 95th percentile leak > 24 L/minute with the nasal or pillow interface or > 36
L/minute with the oronasal interface if ResMed devices were used.[27]
Sleep-related Questionnaire
Questions of the Japanese version of the ESS (JESS), subjective sleep duration, sleep
restfulness, and use of sleep medications within the last month were included in the
sleep-related questionnaire.
The JESS is used to assess daytime sleepiness and is a self-administered questionnaire
with eight questions.[28] Patients were instructed to rate their usual chances of dozing off or falling asleep
while engaged in eight different activities using a 4-point scale (from 0 to 3). The
JESS score ranged from 0 to 24. If the score was higher, the risk of falling asleep
during daytime was greater. A JESS score ≥ 11 indicated subjective excessive daytime
sleepiness.
Sleep restfulness was rated using a 5-point Likert-type scale, with the score of 5
indicating restfulness (1 = restless; 2 = somewhat restless; 3 = neither; 4 = somewhat
restful; 5 = restful).[26]
Statistical Analysis
Data were expressed as median with an interquartile range for continuous variables
or frequencies for categorical variables. All statistical analyses were performed
using a statistical software package (JMP Pro 14 software; SAS Institute, Cary, NC,
United States). First, clinical characteristics, sleep-related questionnaire scores,
CPAP therapy, and CPAP device parameters were analyzed according to sleep restfulness
score via univariate analysis. The Kruskal–Wallis test and the chi-squared test were utilized to compare continuous variables
and categorical variables, respectively. Second, multivariate logistic regression
analysis was performed to identify which variables could best determine sleep restfulness.
In addition to gender, variables with a p-value < 0.05 in the univariate analysis were used in the multivariate analysis.
Continuous variables were categorized into quartiles, and the upper quartile was entered
in the multivariate analysis. A p-value < 0.05 was considered statistically significant.
Results
Participants
[Figure 1] presents the study flowchart. Out of the 775 patients who received the sleep-related
questionnaire, 695 (89.7%) completed it. In addition, 25 patients were excluded due
to lack of data on AHI before CPAP therapy initiation (n = 9), and CPAP device parameters (n = 16). Finally, 670 patients were examined.
Fig. 1 Study flowchart. Abbreviations: AHI. apnea–hypopnea index; CPAP, continuous positive
airway pressure; OSA, obstructive sleep apnea.
[Table 1] shows data on clinical characteristics, sleep-related questionnaire scores, CPAP
therapy, and CPAP device parameters. The cohort included 525 (78.4%) men and 145 (21.6%)
women. The average BMI of the patients was of 26.8 (24.2–29.9) kg/m2, and they presented comorbidities, such as hypertension (66.4%), dyslipidemia (51.2%),
diabetes mellitus (27.3%), cardiovascular diseases (22.1%), arrhythmia (18.4%), stroke
(9.9%), and epilepsy (2.7%). In total, 151 (22.5%) patients used sleep medications.
Furthermore, 582 (86.9%) patients had a JESS score < 11 (no subjective excessive daytime
sleepiness). The Philips Respironics device was used in 362 patients (54.0%) and the
ResMed device in 308 (46.0%). Regarding CPAP device parameters, the AHI was 2.1 (1.1–3.6)
events/hour, and 15 (2.2%) patients had residual AHI. A total of 467 patients (69.7%)
had good adherence to CPAP therapy. In terms of restfulness, 29 (4.3%), 124 (18.5%),
139 (20.8%), 235 (35.1%), and 143 (14.3%) patients answered restless, somewhat restless,
neither, somewhat restful, and restful, respectively.
Table 1
Differences in terms of clinical characteristics and data on sleep-related questionnaire,
CPAP therapy, and CPAP device parameters according to sleep restfulness score
|
Total
|
Sleep restfulness score[a]
|
p-value[b]
|
|
1
|
2
|
3
|
4
|
5
|
|
Number
|
670
|
29
|
124
|
139
|
235
|
143
|
|
|
Gender – male/female: n (%)
|
525/145 (78.4/21.6)
|
23/6
(79.3/20.7)
|
96/28 (77.4/22.6)
|
111/28 (79.9/20.1)
|
178/57 (75.7/24.3)
|
117/26 (81.8/18.2)
|
0.7
|
|
Age in years
|
69 (59–75)
|
57 (51–68)
|
64 (57–73)
|
68 (59–75)
|
70 (62–75)
|
71 (63–77)
|
< 0.0001
|
|
BMI in kg/m2
|
26.8 (24.2– 29.9)
|
28.2 (26–30.6)
|
27.0 (24.4– 30.2)
|
26.7 (24.2– 30)
|
26.8 (24.1–30)
|
26.5 (24.4– 29.5)
|
0.61
|
|
Smoking history – never/past/current: n (%)
|
275/352/43 (41.1/52.5/6.4)
|
15/9/5 (51.7/31.0/17.3)
|
56/61/7 (45.2/49.2/5.6)
|
49/84/6 (35.3/60.4/4.3)
|
98/121/16 (41.7/51.5/6.8)
|
57/77/9 (39.8/53.9/6.3)
|
0.14
|
|
History
|
|
|
|
|
|
|
|
|
Hypertension, n (%)
|
445 (66.4)
|
18 (62.1)
|
71 (57.3)
|
92 (66.2)
|
164 (69.8)
|
100 (69.9)
|
0.15
|
|
Dyslipidemia: n (%)
|
343 (51.2)
|
15 (51.7)
|
65 (52.4)
|
77 (55.4)
|
119 (50.6)
|
67 (46.9)
|
0.7
|
|
Diabetes mellitus: n (%)
|
183 (27.3)
|
11 (37.9)
|
34 (27.4)
|
39 (28.1)
|
59 (25.1)
|
40 (28.0)
|
0.7
|
|
Cardiovascular diseases: n (%)
|
148 (22.1)
|
8 (27.6)
|
26 (21.0)
|
26 (18.7)
|
48 (20.4)
|
40 (28.0)
|
0.33
|
|
Arrhythmia: n (%)
|
123 (18.4)
|
3 (10.3)
|
14 (11.3)
|
15 (10.8)
|
60 (25.5)
|
31 (21.7)
|
0.0004
|
|
Stroke: n (%)
|
66 (9.9)
|
2 (6.9)
|
12 (9.7)
|
17 (12.2)
|
24 (10.2)
|
11 (7.7)
|
0.74
|
|
Epilepsy: n (%)
|
18 (2.7)
|
0 (0)
|
2 (1.6)
|
5 (3.6)
|
6 (2.6)
|
5 (3.5)
|
0.56
|
|
Sleep-related questionnaire
|
|
|
|
|
|
|
|
|
Sleep medication use: n (%)
|
151 (22.5)
|
14 (48.3)
|
35 (28.2)
|
34 (24.5)
|
45 (19.2)
|
23 (16.1)
|
0.002
|
|
Subjective sleep duration in hours
|
6 (5.5–7)
|
5 (4–6)
|
6 (5–6.5)
|
6 (5–6.5)
|
6.5 (5.7–7)
|
7 (6–8)
|
<0.0001
|
|
JESS score
|
5 (3–8)
|
9 (5–15)
|
7 (4–10)
|
5 (3–9)
|
5 (3–7)
|
3 (1–6)
|
<0.0001
|
|
JESS score < 11: n (%)
|
582 (86.9)
|
17 (58.6)
|
96 (77.4)
|
117 (84.2)
|
214 (91.1)
|
138 (96.5)
|
<0.0001
|
|
Data on CPAP therapy
|
|
|
|
|
|
|
|
|
AHI before CPAP therapy initiation–[c] events/hour
|
38.8
(27.3–51.6)
|
38.4
(31.9–62.3)
|
35.4
(25.7–47.8)
|
40.5
(27.6–56.1)
|
36.6
(26.2–48.8)
|
41
(29.7–53.1)
|
0.072
|
|
Duration after CPAP therapy initiation in years
|
6.5
(3.0–11.1)
|
6.5
(2.8–10.6)
|
6.8
(3.5–10.8)
|
7.2
(2.8–11.8)
|
5.9
(2.8–10.8)
|
6.1
(3.1–11.3)
|
0.57
|
|
CPAP devices – Philips Respironics/ResMed: n (%)
|
362/308
(54/46)
|
18/11 (62.1/37.9)
|
77/47 (62.1/37.9)
|
71/68 (51.1/48.9)
|
127/108 (54.0/46.0)
|
69/74 (48.3/51.8)
|
0.17
|
|
Pressure settings – autoadjusting/fixed: n (%)
|
634/36
(94.6/5.4)
|
27/2
(93.1/6.9)
|
114/10 (91.9/8.1)
|
128/11 (92.1/7.9)
|
228/7 (97.0/3.0)
|
137/6
(95.8/4.2)
|
0.15
|
|
Interface – nasal/pillow/oronasal: n (%)
|
600/40/30
(89.5/6.0/4.6)
|
26/2/1
(89.7/6.9/3.4)
|
113/7/4 (91.1/5.7/3.2)
|
125/8/6
(89.9/5.8/4.3)
|
210/13/12
(89.4/5.5/5.1)
|
126/10/7
(88.1/7.0/4.9)
|
1.00
|
|
CPAP device parameters
|
|
|
|
|
|
|
|
|
AHI – events/hour
|
2.1 (1.1–3.6)
|
2.5 (1.4–4.1)
|
2.3 (1.2–4.2)
|
2.1 (1.1–3.3)
|
2.1 (1–3.6)
|
2 (1–3.4)
|
0.23
|
|
Mean duration of CPAP use on the days used: hours/day
|
5.7
(4.7–6.6)
|
4.6
(3.8–5.6)
|
5.3
(4.1–6.1)
|
5.5
(4.6–6.4)
|
5.9
(4.9–6.6)
|
6.3
(5.4–7.1)
|
<0.0001
|
|
Percentage of days of CPAP use: %
|
100 (86.7–100)
|
90.3 (73.4–100)
|
96.7 (76.9–100)
|
96.7 (73.3–100)
|
100 (90–100)
|
100 (96.7–100)
|
<0.0001
|
|
Large leaks:[d]
n (%)
|
187 (27.9)
|
8 (27.6)
|
25 (20.2)
|
39 (28.1)
|
71 (30.2)
|
44 (30.8)
|
0.27
|
|
Good adherence:[e]
n (%)
|
467 (69.7)
|
13 (44.8)
|
71 (57.3)
|
90 (64.8)
|
171 (72.8)
|
122 (85.3)
|
<0.0001
|
Abbreviations: AHI, apnea–hypopnea index; BMI, body mass index; CPAP, continuous positive
airway pressure; JESS, Japanese version of the Epworth Sleepiness Scale.
Notes: Data were expressed as median with an interquartile range for continuous variables
or frequencies for categorical variables.
a Sleep restfulness was rated using a 5-point Likert-type scale, with the score of
5 indicating restfulness (1 = restless; 2 = somewhat restless; 3 = neither; 4 = somewhat
restful; 5 = restful).
b Kruskal–Wallis test (continuous variables) or chi-squared test (categorical variables).
c Before CPAP therapy initiation, 577 (86.1%) patients underwent polysomnography, and
93 (13.9%) had a portable monitor.
d Large leaks were defined as large leaks lasting > 1 hour, if Philips Respironics
devices were used; and a 95th percentile leak > 24 L/minute with the nasal or pillow
interface or > 36 L/minute with the oronasal interface, if ResMed devices were used.
e Good adherence to CPAP therapy was defined as the use of CPAP therapy for at least
4 hours a night in 70% of nights for 1 month.
Sleep Restfulness
[Table 1] shows the differences in terms of clinical characteristics, sleep-related questionnaire
scores, CPAP therapy, and CPAP device parameters according to sleep restfulness score.
The subjective sleep duration ([Figure 2A]) and mean CPAP duration ([Figure 2B]) significantly increased with greater sleep restfulness score (from 1 to 5) (5 [4–6]
hours; 6 [5–6.5] hours; 6 [5–6.5] hours; 6.5 [5.7–7] hours; and 7 [6–8] hours; p < 0.0001; and 4.6 [3.8–5.6] hours/days; 5.3 [4.1–6.1] hours/days; 5.5 [4.6–6.4] hours/days; 5.9 [4.9–6.6] hours/days; and 6.3 [5.4–7.1] hours/days; p < 0.0001 respectively). In addition, the percentage of patients using sleep medications
([Figure 2C]) significantly decreased with greater sleep restfulness score (from 1 to 5) (48.3%;
28.2%; 24.5%; 19.2%; and 16.1%; p = 0.002 respectively). Meanwhile, the percentage of patients with a JESS score < 11
([Figure 2D]) and good adherence to CPAP therapy ([Figure 2E]) significantly increased (58.6%; 77.4%; 84.2%; 91.1%; and 96.5%; p < 0.0001; and 44.8%; 57.3%; 64.8%; 72.8; and 85.3%; p < 0.0001 respectively).
Fig. 2 Differences in terms of subjective sleep time (A), mean duration of CPAP use on the days used (B), and the percentage of sleep medication use (C), JESS score < 11 (D), and good adherence to CPAP therapy (E) according to sleep restfulness score. Abbreviations: CPAP, continuous positive airway
pressure; JESS = Japanese version of the Epworth Sleepiness Scale.
[Table 2] shows data on sleep restfulness based on the multivariate logistic regression analysis.
Subjective sleep duration ≥ 7 hours, good adherence to CPAP therapy, mean CPAP duration
of 6.6 hours/day, and JESS score < 11 were independently and positively associated
with sleep restfulness. Meanwhile, the use of sleep medications was negatively associated
with sleep restfulness.
Table 2
Multivariate logistic regression analysis of sleep restfulness
|
Model 1
|
Model 2
|
Model 3
|
|
Odds ratio (95%CI)
|
p-value
|
Odds ratio (95%CI)
|
p-value
|
Odds ratio (95%CI)
|
p-value
|
|
Age ≥ 75 years
|
1.48 (0.96–2.28)
|
0.078
|
1.58 (1.03–2.42)
|
0.036
|
1.47 (0.95–2.27)
|
0.082
|
|
Gender (male)
|
1.18 (0.71–1.96)
|
0.52
|
1.15 (0.70–1.91)
|
0.58
|
1.15 (0.69–1.92)
|
0.59
|
|
Arrythmia
|
1.01 (0.61–1.65)
|
0.98
|
1.03 (0.63–1.68)
|
0.92
|
1.03 (0.62–1.68)
|
0.84
|
|
Sleep medication use
|
0.48 (0.28–0.82)
|
0.0068
|
0.46 (0.27–0.78)
|
0.0042
|
0.46 (0.27–0.79)
|
0.0044
|
|
JESS score < 11
|
3.99 (1.55–10.25)
|
0.0041
|
4.12 (1.61–10.57)
|
0.0032
|
3.99 (1.55–10.31)
|
0.0042
|
|
Good adherence
|
2.27 (1.35–3.81)
|
0.0021
|
2.07 (1.21–3.54)
|
0.0076
|
1.95 (1.14–3.35)
|
0.015
|
|
Subjective sleep time ≥ 7 hours
|
2.69 (1.80–4.02)
|
< 0.0001
|
−
|
−
|
2.26 (1.46–3.49)
|
0.0003
|
|
Mean duration of CPAP use on the days used ≥ 6.6 hours/day
|
−
|
−
|
2.33 (1.51–3.60)
|
0.0001
|
1.64 (1.02–2.63)
|
0.042
|
Abbreviations: 95%CI, 95% confidence interval; CPAP, continuous positive airway pressure;
JESS, Japanese version of the Epworth Sleepiness Scale.
Discussion
The present study showed that sleep restfulness was independently and positively associated
with not only long subjective sleep duration and a JESS score < 11 (no subjective
excessive daytime sleepiness) but also good adherence to CPAP therapy. The favorable
outcomes of CPAP therapy, such as decreased blood pressure and risk of cardiovascular
diseases and stroke, significantly depend on good adherence to CPAP therapy.[5]
[6]
[7]
[8]
[9]
[10] Thus, sleep restfulness can be an indicator of a subtype that has favorable outcomes
after CPAP therapy. However, sleep restfulness is rarely included in sleep-related
questionnaires used in the cluster analysis of OSA.[23] In addition, sleep-related symptoms are largely influenced by sleep quantity (sleep
duration), as shown in our study results. However, sleep duration has not been included
in the cluster analysis of OSA. Therefore, sleep duration and restfulness should be
included in the new cluster analysis of OSA to differentiate the heterogeneity of
OSA.
Sleep restfulness was found to be independently associated with adherence to CPAP
therapy. To the best of our knowledge, only one report showed the association between
sleep restfulness and adherence to CPAP therapy in patients with OSA. Pien et al.[23] revealed that the longitudinal change in sleep restfulness (from restless to restful
response to CPAP therapy) was larger in patients with good adherence to CPAP therapy
than in those without in each subgroup such as disturbed sleep, minimally symptomatic,
and sleepy classified via cluster analysis. Their results were consistent with ours.
Sleep restfulness is closely associated with the risk of cardiovascular diseases and
mortality.[25]
[26] Thus, it is an important sleep-related symptom in the general population. The favorable
outcomes of CPAP therapy are significantly based on adherence to this therapy. Thus,
future prospective studies should evaluate the association between outcomes and adherence
to CPAP therapy in patients with or without sleep restful under receiving this therapy
are needed to further identify the role of sleep restfulness in patients with OSA.
Sleep restfulness was independently associated with subjective sleep duration. Generally,
sleep issue is assessed according to sleep quantity (sleep duration) and sleep quality
(sleep-related symptoms). Regarding sleep duration, several systematic reviews have
shown that short sleep duration is not only a predictor of mortality but also a possible
marker of increased risk of hypertension, diabetes mellitus, cardiovascular diseases,
and stroke.[29]
[30]
[31]
[32] Regarding sleep quality, several studies have shown that poor subjective sleep quality
is associated with an increased risk of all-cause mortality and cardiovascular diseases.[25]
[26]
[33]
[34]
[35] In addition, sleep duration and sleep quality may affect health outcomes.[36] However, sleep duration has not been included in the cluster analysis of OSA. Thus,
another cluster analysis of OSA including sleep duration must be developed.
Sleep restfulness was independently associated with nonusage of sleep medications.
Although sleep medications can improve subjective sleep quality,[37]
[38] they have several adverse effects, including daytime fatigue, cognitive impairment,
difficulty concentrating, oversleeping, and nightmares.[38]
[39] These adverse effects could have induced the lack of sleep restfulness in the present
report, even if patients consumed sleep medications. However, it is not recommended
to discontinue sleep medications to get sleep restfulness.
The present study had several major limitations. First, this was a cross-sectional
study. Thus, the present study did not validate the association between sleep restfulness
and outcome. Second, 86.9% of the patients had a JESS score < 11. Hence, most patients
did not complain of excessive daytime sleepiness. Third, the JESS questions, subjective
sleep duration, sleep restfulness, and use of sleep medications were only included
in the sleep-related questionnaire. Fourth, we did not evaluate types of sleep medications,
such as benzodiazepines, nonbenzodiazepines, melatonin receptor agonist, antidepressants,
antipsychotics, and anticonvulsants. Fifth, only subjective sleep duration was examined
in the present study. However, there is a difference between subjective and objective
sleep duration.[40] Future prospective studies, which include several questions on sleep-related symptoms
and examine kinds of sleep medications and objective sleep duration, must be conducted
to identify the association among sleep-related symptoms, including sleep restfulness,
sleep duration, and outcome.
Conclusion
Sleep restfulness was associated with subjective sleep duration and good adherence
to CPAP therapy in patients with OSA. Thus, sleep restfulness can be an indicator
of a subtype that has favorable outcomes after CPAP therapy.
