Sleep, 17(6):516-521
© 1994 American Sleep Disorders Association and Sleep Research Society
Effect of Nasal Spray, Positional Therapy, and the
Combination Thereof in the Asymptomatic Snorer
*Howard M. Braver and
tAo Jay Block
* University of Florida College of Medicine,
Gainesville, Florida, U.S.A.
tDepartment of Veterans Affairs Medical Center,
Gainesville, Florida, U.S.A.
Summary: The benefits of using a nasal decongestant, sleeping on one's side and the combination thereof were
studied in 20 asymptomatic male snorers. Both the apnea-hypopnea index (AHI) and snoring were evaluated. Four
consecutive nocturnal polysomnographic studies were done. Night 1 was a control; the other 3 nights were randomly
assigned to nasal decongestant, best sleeping position and a combination of the two. Results were calculated based
on sleep period time. The mean control AHI ± the standard error of the mean (SEM) was 17.5 ± 6.5. AHI improved
to 14.1 ± 6.3 with sleep in the best position (p = 0.03). The AHI also improved to 13.2 ± 6.04 with both nasal
decongestant and position (p = 0.0012). Using nasal decongestant alone, the mean AHI was 18.1 ± 6.3 (p = 0.765).
During the control night, the mean number of snores/hour ± SEM was 356 ± 46.0. Using nasal decongestant alone,
the mean number of snores was 381 ± 50.4 (p = 0.50). With position alone, the mean number of snores was 356
± 46.0 (p = 0.8). Using the combination of nasal decongestant and position, mean snores were 352 ± 48.9 (p =
0.91). In conclusion, a statistically significant improvement in AHI was produced using the general measures of
altering the position of the body during sleep and by the combination of nasal decongestant and positional change.
There was no significant change in snoring using any of these general measures. Key Words: Snoring-TreatmentSleep apnea.
Large numbers of men who snore heavily exist in
the general population. Although many such men who
are loud snorers claim that they are without symptoms
of disease, a remarkable degree of apneas has been
found in these patients when they have been studied
by polysomnography (l). Studies have also shown that
snorers have an increased risk of developing hypertension, ischemic heart disease and strokes (2-4). The
proper management of such asymptomatic patients is
not clear from the published literature. Several general
measures have been recommended as treatments for
sleep apnea. These general measures include alteration
of sleeping body position (5) and improving nasal patency (6). Neither of these measures should have unpleasant side effects and might be acceptable to the
asymptomatic snorer. Both of these measures have been
studied alone and each has been shown to have some
efficacy in reducing the frequency of obstructive sleep
Accepted for publication May 1994.
Address correspondence and reprint requests to A. Jay Block,
M.D., Pulmonary Division, P.O. Box 100225 Health Center, University of Florida, Gainesville, FL 32610-0225, U.S.A.
apnea. No studies are available wherein the combination of these general measures of treatment have
been used in the same patient.
The purpose of this study was to determine if sleep
apnea and snoring in asymptomatic patients could be
eliminated or reduced in frequency by altering the position of the body during sleep, improving nasal patency by using a nasal decongestant and by the combined effect of body position and using a nasal
decongestant. The long-term goal of this project would
be to identify a successful, easy, noninvasive method
of therapy for the patient who is a loud snorer.
METHODS
Patients. The study group consisted of 20 male subjects, 27-63 years of age, who were all asymptomatic
snorers. Recruitment of the subjects was quite easy
through response to newspaper advertisements and
through the use of lists of subjects from earlier studies
(1,7). The polysomnography laboratory at the Gainesville YA Medical Center has files containing names
and addresses of more than 50 asymptomatic men who
516
TREATMENT OF MEN WHO SNORE
FIG.t. A reproduction of an actual polysomnographic tracing. The
first tracing is the EEG lead. The second tracing is the electromyogram lead. The third tracing is the EOG lead. The fourth tracing is
the microphone with the inflections representing snores. The fifth
tracing is the EKG lead.
snore heavily and who have participated previously in
studies of sleep apnea and snoring. Subjects were interviewed for their responses to a standardized series
of questions to assure that they were healthy, not addicted to drugs or alcohol, and had no predisposing
factors towards sleep apnea other than obesity. Spouses
were also interviewed similarly concerning symptoms
of which the subject was unaware or which he denied.
Work history was verified to assure that excessive
sleepiness had not interfered with the subject's occupation. Subjects were included in the study only if they
had a h,<;tory of asymptomatic heavy snoring, were
healthy, were not addicted to drugs or alcohol, had no
predisposing factors towards sleep apnea other than
obesity and had normal pulmonary function tests.
All patients gave informed consent before entry into
the study as approved by the Institutional Review Board
of the University of Florida (IRB #45-90).
Study design. At the initiation of the study, each
subject underwent a brief examination that was performed by the principal investigator. Thereafter, baseline pulmonary function tests were performed. Following pulmonary function testing, a baseline night of sleep
was monitored. On the 1st night, each subject was
allowed to sleep in any position he chose. Position
during sleep was recorded on an 8-hour, Fuji 2-160
videotape using an RCA infrared closed circuit TV
camera and a VCR-monitor system. Time was constantly displayed on the videotape using a date/time
display generator (Vicon Industries, Melville, NY,
U.S.A.). This display allowed later review of the tape
517
to correlate numbers of apneas and hypopneas with
position. The monitor was constantly viewed by an
experienced sleep technologist to ensure that accurate
records were kept. Polysomnography was performed
as in previous studies (1,7). Continuous oxygen saturation was measured with an ear oximeter (Biox 3700,
Ohmeda Corp., Louisville, CO, U.S.A.). Oral and nasal
airflows were monitored separately with Grass thermistors (Quincy, MA, U.S.A.). Abdominal and chest
movements were sensed with a respiratory inductance
plethysmograph (Ambulatory Monitoring Inc., Ardslet, NY, U.S.A.), and a sum tracing was also recorded.
The inductance tracing was not used quantitatively,
but rather the deflections defined breathing rate and
rhythm. Electroencephalograms (EEG) and electrooculograms (EOG) were recorded simultaneously from
occipital, frontal and ocular leads. Electrocardiograms
(EKG) were recorded from a standard V2 lead. All
leads were recorded simultaneously on a 15-channel
polygraph with optical disc storage capability.
In addition to the usual montage of polysomnographic monitoring, we also used a method to record
and quantitate the frequency of snoring. This method
was used in a study of snorers by Abbey et al. (8). We
used a Grass model TR-21 miniature microphone
(Quincy, MA, U.S.A.) that monitored sounds generated by the passage of air through the upper airway
during all-night sleep. The small size of the microphone allowed placement on the substernal notch. This
location allowed easy identification of snores (Fig. 1)
and discrimination of the deflections made by snoring
from extraneous noise. The microphone generates a
relatively large signal response to airflow; approximately 1-5 microvolts. Sounds made by the sleeper
caused a deflection of the polygraph to which the microphone was cabled, giving a real-time record of each
noise. All sleepers were monitored by television, and
noise deflections from extraneous sounds were marked
on a continuous paper tracing of a multichannel recorder by the sleep technologist. For both clinical and
research purposes, such quantitation of the frequency
of snores per night has been achieved easily using the
microphone. We do not currently have equipment in
our sleep laboratories to establish snoring intensity.
The next morning, the sleep technologist reviewed
the videotape of the 1st night of sleep and assessed the
severity of sleep apnea and snoring, as well as the position(s) associated with breathing irregularities. A best
position for sleeping for each patient became obvious
on review of each videotape during the control night.
This position for every subject was with the subject
sleeping on his side.
On the following 3 nights the factors sleeping position (best position vs. free position), nasal spray (yes
as opposed to no) and the combination of best position
Sleep, Vol. 17, No.6, 1994
518
H. M. BRA VER AND A. J. BLOCK
TABLE 1. Demographic characteristics of the 20 subjects
Patient no.
Age
(years)
1
43
2
37
36
3
4
44
59
5
40
6
27
7
40
8
64
9
28
10
40
11
12
37
13
63
48
14
15
41
16
38
17
30
18
63
19
35
20
29
Mean ± SEM 42 ± 2.6
Weight
(kg)
Height
(m)
BMI"
(kglm2)
105
141
106
94
105
114
135
118
104
131
147
105
104
127
98
110
94
86
90
113
111 ± 3.8
1.78
1.78
1.68
1.80
1.70
1.75
1.85
1.73
1.70
1.83
1.85
1.75
1.78
1.70
1.80
1.75
1.70
1.80
1.78
1.80
1.77±O
33
45
38
29
36
37
39
39
36
39
43
34
33
44
30
36
33
27
28
35
36 ± 1.1
greater from the preceding baseline saturation occurred. Sleep period time (SPT) was defined as the total
time between falling asleep (onset of EEG Stage I) and
awakening in the morning. Total sleep time (TST) was
defined as SPT less any time the subject was awake
after falling asleep (TST = SPT - Stage 0). Apneahypopnea index (AHI) is the total apneas plus hypopneas per hour of SPT.
Statistical analysis. All data are expressed as the
mean ± SEM with associated 95% confidence intervals
(CIs). The statistical significance of differences among
the control and the three different treatments was evaluated using repeated analysis of variance (ANOV A)
measurements, with p levels <0.05 considered to indicate statistical significance.
RESULTS
A total of 20 subjects entered this study. All 20 fulfilled the enrollment criteria, and all 20 subjects com" BMI, body mass index.
pleted the study. Demographic characteristics of the
study group are shown in Table 1. The degree of obesity
of each patient was expressed in terms of body mass
(on side) and nasal spray were studied in a randomized index (BMI), defined as weight/height where weight is
complete-block design. The best sleeping position was expressed in kilograms and height in m 2 (10).
The mean ± SEM percent of time the 20 subjects
secured with the use of foam rubber wedges both behind and in front of the subject. Television monitoring spent sleeping supine versus in the lateral position durwas used to assure that the position was maintained ing the control night was 68 ± 5.3% and 3.2 ± 5.2%,
during sleep. Oxymetazoline (Afrin) 12-hour nasal spray respectively.
was the nasal spray used in this study. Two sprays of
Table 2 reveals the sleep structure including the mean
±
SEM for SPT and TST and the percent ofSPT spent
this medication were delivered to each nostril just prior
to bedtime on the appropriate nights. Complete poly- in wakefulness, Stages I and II, Stages III and IV, and
somnographic monitoring plus television monitoring rapid eye movement (REM) sleep during the contrOl,
was repeated on every night of the 4-night study.
nasal spray and combination nasal spray and positional
Definitions. Apnea was noted when flow ceased in treatment nights for the 20 subjects.
Tables 3 and 4 reveal the actual AHIIhour SPT and
the nose and mouth for 10 seconds or longer. If chest
and abdominal movement ceased during this period, numbers of snores/hour SPT, respectively, for each
the apnea was recorded as a central apnea. If chest and patient.
As shown in Table 3, during the control night, the
abdominal movement continued in opposite directions, the apnea was recorded as obstructive. An epi- mean AHI ± SEM was 17.5 ± 6.5. With sleep in the
sode of hypopnea was noted if flows in the nose and best position, the mean AHI was 14.1 ± 6.3 (95% CI,
mouth decreased, chest and abdominal movements 0.36-6.4; p = 0.03). Using nasal spray alone, the mean
decreased and desaturation occurred (9). An episode AHI was 18.1 ± 6.3 (95% CI -4.74-3.54; p = 0.765).
of desaturation was recorded when a fall of 4% or The mean AHI ± SEM with the combination of nasal
TABLE 2. Sleep structure
Control
SPT (minutes)
TST (minutes)
Awake (% SPT)
I-II (% SPT)
III-IV (% SPT)
REM (% SPT)
385 ±
346 ±
10 ±
74 ±
5±
12 ±
6.4
8.7
1.8
2.6
1.5
1.5
Spray
411 ± 2.9
383 ± 6.2
7 ± 1.3
72 ± 2.5
5 ± 1.6
16 ± 1.8
Position
412
376
9
73
4
15
±
±
±
±
±
±
3.6
7.7
1.7
2.5
1.0
1.5
Combination
402 ±
371 ±
8±
70 ±
6±
17 ±
5.2
6.7
1.2
1.9
1.4
1.3
SPT, sleep period time in minutes; TST, total sleep time in minutes; Awake (Stage 0), I-II, III-IV, REM, percent of sleep period time
..,
spent in wakefulness, Stage I and II, Stage III and IV, and REM sleep, respectively.
Sleep. Vol. 17. No.6. 1994
TREATMENT OF MEN WHO SNORE
519
TABLE 3. Apnea-hypopnea indexa/hour sleep period time
Patient no.
Control
Spray
Position
I
0
12
23
0
0
I
2
6
8
0
0
2
0
0
2
0
0
0
83
71
63
7
6
3
83
101
100
II
67
67
64
12
0
0
0
13
10
0
0
14
3
0
0
15
2
I
0
16
2
10
3
17
41
27
28
18
14
46
0
19
4
2
5
20
4
0
0
- Mean + SEM
17.5+6.5
14.1 ± 6.3
18.1 ± 6.3
a Apnea-hypopnea index, number of apneas and hypopneas per hour sleep period time.
0
20
I
2
3
4
5
6
7
8
9
10
spray and positional treatment was 13.2 ± 6.04 (95%
CI 1.93-6.66; p = 0.0012). AHI improved significantly
with both positional treatment and with the combination of positional treatment and nasal spray.
As shown in Table 4, during the control night, the
mean snores ± SEM were 356 ± 46.0. Using nasal
spray, mean snores were 381 ± 50.4 (95%CI, -100.9350.83; p = 0.50). With positional treatment alone, mean
snores were 356 ± 46.0 (95% CI, 259-452; p = 0.8).
With the combination of nasal spray and position, mean
snores were 352 ± 48.9 (95% CI, -66-73.9; p = 0.91).
TABLE 4. Number of snores/hour sleep period time
Patient no.
I
2
3
4
5
6
7
8
9
10
II
12
13
14
15
16
17
18
19
20
Mean ± SEM
Control
184
432
94
48
408
50
693
595
250
446
326
508
581
151
439
762
245
227
363
316
356 ± 46
Spray
242
431
83
247
89
66
938
366
420
550
611
517
534
139
379
666
544
259
198
71
381 ± 50
CombinaPosition
tion
363
435
40
29
347
33
505
426
250
565
553
603
723
231
294
375
525
14
426
381
356 ± 46
328
588
37
85
4
47
712
414
327
525
548
514
495
189
178
693
362
166
376
451
352 ± 49
Combination
I
4
0
0
0
0
0
70
I
90
59
0
0
I
0
4
28
5
I
0
13.2 ± 6.0
There was no statistically significant change in snoring
with the use of nasal spray, best position or the combination of nasal spray and positional treatment in our
subjects.
As shown in Table 3, three subjects (subject nos. 8,
10 and 11) had AHI > 65. These subjects were included in the study because they fulfilled all of our
inclusion criteria, including being asymptomatic,
healthy, not addicted to drugs or alcohol, having no
predisposing factors towards sleep apnea other than
obesity and having normal pulmonary function tests.
DISCUSSION
The results of our study were both disappointing and
surprising. In accordance with the literature, our study
showed an improvement in the AHI with positional
treatment during sleep. In 1984, Cartwright reported
that the apnea index was twice as high in 24 patients
being evaluated for the sleep apnea syndrome if they
slept supine rather than on their sides (11). In 1986,
Phillips et al. verified that sleeping in the lateral decubitus position was associated with improvement in
obstructive apneas and hypopneas (12). Physiological
support for the finding of increased numbers of apneas
while sleeping in the supine position has also been
reported (13). Although the current study did demonstrate a statistically significant reduction in AHI while
patients slept on their sides, the clinical significance of
this small change is doubtful. The lack of any beneficial
effect on snoring is distressing.
The most surprising part of our study was that increasing nasal patency did not improve the AHI or
Sleep, Vol. 17, No.6, 1994
520
H. M. BRA VER AND A. J. BLOCK
snoring. In the early 1980s, several papers demonstrated that obstructive sleep apnea could be worsened by
nasal obstruction (6,14,15). Posterior nasal packing,
which caused complete nasal obstruction after septoplasty, produced obstructive sleep apnea even in subjects who were completely normal preoperatively (16).
Polysomnography done on patients with seasonal allergic rhinitis showed twice as many apneas during the
allergic season when the nares were obstructed (17).
Why nasal obstruction causes sleep-disordered breathing is unclear.
Parisiet al. showed in 1989 that pharmacologic vasoconstriction of both the nasopharynx and oropharynx using phenylephrine intranasally and by gargling,
respectively, reduces tissue blood volume, thereby increasing cross-sectional area and decreasing collapsibility ofthe upper airway (18). Wasicko et al. showed
in 1990 that pharmacologic systemic vasodilation increases tissue blood volume, thereby reducing the pharyngeal cross-sectional area, which likely makes the
airway more susceptible to collapse (19), and in 1991
Wasicko et al. showed that the oropharyngeal and nasopharyngeal topical application of phenylephrine decreased both nasal and pharyngeal resistance and that
the effect of phenylephrine on nasal resistance was independent of the effect on pharyngeal resistance (20).
In 1990, Hutt et al. showed that oropharyngeal and
nasopharyngeal topical application of oxymetazoline,
a vasoconstrictor, caused significant reductions in sleepdisordered breathing in patients with obstructive sleep
apnea (21).
The reason why oxymetazoline did not cause a significant reduction in sleep-disordered breathing in our
study is not clear. The most likely explanation lies in
the fact that oxymetazoline has been shown to cause
increased nasal discharge in patients (22). Table 3 shows
that 13120 subjects actually had some decrease in AHI
with the use of oxymetazoline as compared to the control night. Table 4 shows that 9/20 subjects had some
decrease in the amount of snoring with the use of oxymetazoline as compared to the control night. It is very
possible that some of our subjects could have experienced an increase in nasal discharge secondary to the
oxymetazoline that caused increased nasal obstruction.
The increased nasal obstruction may have secondarily
caused an increase in the AHI and an increase in snoring. One should note that the subjects in our study
were only treated with intranasal instillation of oxymetazoline. They were not treated by both nasopharyngeal and oropharyngeal topical vasoconstriction as
was done in the previous studies mentioned (18,20,21).
It is possible that our patients had less oropharyngeal
vasoconstriction than those in the studies done by Parisi et al. (18), Wasicko et al. (20) and Hutt et al. (21)
and therefore had increased collapsibility ofthe upper
Sleep. Vol. 17. No.6. 1994
airway as compared to the subjects in these other studies. Alternatively, our study only had 20 subjects. Another study using a much larger number of subjects
could possibly show a statistically significant improvement in sleep-disordered breathing in patients treated
with topical vasoconstriction.
Snoring was not improved by any intervention or
combination of interventions. Because the noise of
snoring and its consequent spousal irritation are common reasons such patients seek medical attention, these
results are very disappointing. Hoffstein et al. demonstrated in 1988 that nasal airflow resistance correlates with snoring (23). More recent studies by Miljeteig
et al. (24) and Hoffstein et al. (25) suggest that reduction
in nasal resistance during sleep may not correlate with
snoring. We believed that reducing nasal resistance
during sleep with oxymetazoline would lead to a reduction in snoring. We did not measure resistance in
our study. It is possible that the nasal resistance in our
subjects was reduced by the oxymetazoline, but the
recent studies by Miljeteig et al. (24) and Hoffstein et
al. (25) were correct in suggesting that the reduction in
nasal resistance during sleep may not correlate with
snoring. It is also possible that the snoring in our patients stayed the same after using the nasal decongestant, because nasal resistance did not actually change.
Furthermore, it is possible that the reduction in the
AHI from position alteration in our subjects allowed
more inhalations to occur that were partially obstructed rather than totally obstructed, increasing rather than
decreasing the frequency of snoring.
On the basis of our results, it is obvious that the
institution of only positional treatment and nasal spray
cannot cure sleep-disordered breathing in the asymptomatic snorer. The combination of positional treatment and nasal spray only minimally improved the
sleep-disordered breathing in our subjects. It is possible that the addition of weight loss to the small benefits seen in this study could be additive and yield more
clinically impressive results. We will add the weight
loss intervention to our future studies to assess this
possibility .
Acknowledgements: This work was supported by a merit
review grant from the Veterans Administration. Special thanks
are due to Mr. Nick D' Amato and Mr. Don Hellard who
performed the sleep studies and stared endlessly at videotapes of men sleeping.
REFERENCES
1. Block AJ, Hellard DW, Cicale MJ. Snoring, nocturnal hypoxemia, and the effect of oxygen inhalation. Chest 1987;92:411-7.
2. Lugaresi E, Cirignotta F, Coccagna G, Piana C. Some epidemiological data on snoring and cardiocirculatory disturbances.
Sleep 1980;3:221-4.
TREATMENT OF MEN WHO SNORE
3. Koskenvuo M, Partinen M, Sarna S, Kaprio J, Langinvainio H,
Heikkila K. Snoring as a risk factor for hypertension and angina
pectoris. Lancet 1985; 1:893-5.
4. Partinen M, Palomaki H. Snoring and cerebral infarction. Lancet 1985;2: 1325-6.
5. McEvoy RD, Sharp DJ, Thornton AT. The effects of posture
on obstructive sleep apnea. Am Rev Respir Dis 1986; 133:662-6.
6. Zwillich CW, Pickett C, Hanson FN, Weil JV. Disturbed sleep
and prolonged apnea during nasal obstruction in normal men.
Am Rev Respir Dis 1981;124:158-60.
7. Block AJ, Hellard DW, Switzer DA. Nocturnal oxygen therapy
does not improve snorers' intelligence. Chest 1989;95:274-8.
8. Abbey NC, Block AJ, Green D, Mancuso A, Hellard DW. Measurement of pharyngeal volume by digitized magnetic resonance
imaging. Effect of nasal continuous positive pressure. Am Rev
Respir Dis 1989; 140:717-23.
9. Block AJ, Boysen PG, Wynne JW, Hunt LA. Sleep apnea, hypopnea, and oxygen de saturation in normal subjects. A strong
male predominance. New Engl J Med 1979;300:513-7.
10. Thomas AE, McKay DA, Cutlip MB. A nomograph method for
assessing body weight. Am J Clin Nutr 1976;29:302-4.
II. Cartwright RD. Effect of sleep position on sleep apnea severity.
Sleep 1984;7: 110-4.
12. Phillips BA, Okeson J, Paesani D, Gilmore R. Effect of sleep
position on sleep apnea and parafunctional activity. Chest 1986;
90:424-9.
13. Brown IB, McClean PA, Boucher R, Zamel N, Hoffstein V.
Changes in pharyngeal cross-sectional area with posture and
application of continuous positive airway pressure in patients
with obstructive sleep apnea. Am Rev Respir Dis 1987; 136:62832.
14. Wynne JW. Obstruction of the nose and breathing during sleep.
Chest 1982;82:657-8.
15. Olsen KD, Kern EB, Westbrook PR. Sleep and breathing dis-
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
521
turbance secondary to nasal obstruction. Otolaryngol Head Neck
Surg 1981 ;89:804-10.
Taasan V, Wynne JW, Cassisi N, Block AJ. The effect of nasal
packing on sleep-disordered breathing and nocturnal oxygen
desaturation. Laryngoscope 1981; 7: 1163-72.
McNicholas WT, Tarlo S, Cole P, et al. Obstructive apneas
during sleep in patients with seasonal allergic rhinitis. Am Rev
Respir Dis 1982; 126:625-8.
Parisi RA, Wasicko MJ, Hutt DA, Mandel M, Santiago TV,
Edelman NH. Mucosal vasoconstriction increases pharyngeal
size in normal subjects. Am Rev Respir Dis 1989;139:A374 (abstract).
Wasicko MJ, Hutt DA, Parisi RA, Neubauer JA, Mezrich R,
Edelman NH. The role of vascular tone in the control of upper
airway collapsibility. Am Rev Respir Dis 1990;141:1569-77.
Wasicko MJ, Leiter JC, Erlichman JS, Strobel RJ, Bartlett D.
Nasal and pharyngeal resistance after topical mucosal vasoconstriction in normal humans. Am Rev Respir Dis 1991; 144:104852.
Hutt DA, Parisi RA, Santiago TV. Pharyngeal mucosal vasoconstriction reduces the frequency of disordered breathing events
in patients with obstructive sleep apnea. Am Rev Respir Dis
1990; 141 :A865 (abstract).
Huff BB, Dowd AL, eds. Physician's desk reference for nonprescription drugs. Oradell, NJ: Medical Economics Incorporated, 1989:682.
Hoffstein V, Chaban R, Cole P, Rubinstein I. Snoring and upper
airway properties. Chest 1988;94:87-9.
Miljeteig H, Savard P, Mateika S, Cole P, Haight JS, Hoffstein
V. Snoring and nasal resistance during sleep. Laryngoscope 1993;
103:918-23.
Hoffstein V, Mateika S, Metes A. Effect of nasal dilation on
snoring and apneas during different stages of sleep. Sleep 1993;
16:360-5.
Sleep, Vol. 17, No.6, 1994