Laser-assisted uvulopalatoplasty (LAUP) is an outpatient staged surgical
procedure for the treatment of snoring. Each patient undergoes a series of procedures with
the end point being patient and bed partner satisfaction in most cases.
The purpose of this study was to objectively evaluate the frequency,
pattern, and volume of snoring in patients prior to and following each LAUP procedure. A
sonographic device, SNAPÔ , which records oronasal
respiration, was used to assess patients at home. A digital analysis of the frequency,
pattern, and volume was then performed.
Twenty-seven patients have been completely evaluated with this
recording device. The findings demonstrate that the LAUP procedure alters the snoring
sound. The maximum, average, and velum-like respiratory noise loudness all showed a
statistically significant decrease when comparing the preoperative snoring to the final
recording after treatment was completed. In addition, the fundamental frequency of the
snoring increased significantly after each LAUP procedure. No change was seen in the
snoring index following treatment. These objective results correlated well with the
subjective responses of the patients and their bed partners. These findings are consistent
with the hypothesis that the LAUP procedure alters snoring in a favorable manner by both
objective data and subjective accounts.
LARYNGOSCOPE, 106:1372-1377,1996
INTRODUCTION
Self-reported data on snoring
is questionable at best. The patient's description is dependent on many factors such as
the perceived social stigma, health consciousness, the bed partners' complaints, and their
ability to objectively assess the snoring themselves.1 In a large study of 613
patients undergoing polysomnography, 471 patients who snored were unaware of their
snoring. In addition, when trained sleep technologists were asked to rate snoring during
these polysomnograms and objective testing was done simultaneously, the technologists'
subjective perceptions did not correlate well with the objective measure of the snoring
loudness.2
Numerous investigators have
studied snoring with various types of objective testing.2-11 Yet there is no
standardization of what spectral characteristics make up the snoring sound or what form of
objective testing is most accurate or clinically useful.1,7 Studies
have been done in which objective testing of snoring was compared prior to and after
uvulopalatopharyngoplasty (UPPP).12,13 There is no difference in the snoring in
the preoperative versus the postoperative groups with regard to the snoring index or
loudness. There did appear, however, to be a difference in the fundamental frequency of
the snoring.12-14
Despite these less than
impressive objective results, a significant number of patients perceived dramatic
improvements in their snoring. Thus it appears that to date we are not able to identify
which components of the snoring noise are most annoying to the human ear.
Laser-assisted
uvulopalatoplasty (LAUP) is a surgical treatment for snoring that has been used in the
United States since 1993. The procedure can be performed in the office using local
anesthesia and a carbon dioxide (C02) laser to reduce and reshape the uvula and
soft palate. Each patient undergoes a series of procedures until the snoring has
diminished to a level that is satisfactory to the patient and bed partner. Although
several reports have been published on the subjective outcome of this procedure, no
objective testing has been reported.15-17 The purpose of this study
was to objectively and subjectively evaluate snoring preoperatively and after each LAUP
procedure. The objective testing examined the volume, frequency, and pattern of snoring.
MATERIALS AND METHODS
Patient Selection
Between December 1994 and December 1995, 27 consecutive patients with
snoring were evaluated for LAUP treatment by two physicians (R.P.W.
and W.M.G.) Patients were advised to bring their bed partners to
the initial evaluation. All patients and their partners viewed an educational video and
answered a detailed sleep questionnaire prior to their evaluation.
During the initial visit, a complete history was obtained, and thorough
physical and otolaryngologic examinations were performed. The nasopharynx, hypopharynx,
and larynx were visualized using a flexible scope, and Müller's maneuver9 was performed at the levels of the nasopharynx and the
base of tongue. Initial body weight and height were recorded, and the body mass
index (BMI) was calculated.18
LAUP was recommended for patients with loud, disruptive snoring and no
medical contraindications to surgery. The procedure was not recommended if patients had a
hyperactive gag reflex, mandibular retrognathia, mandibular micrognathia, a bleeding
disorder, or if they were receiving anticoagulant therapy.
When appropriate, patients were encouraged to avoid sleeping supine, to
lose weight, to lengthen their bedtime, and to avoid alcohol and tobacco. All patients
were informed that the benefits, risks, and complications of LAUP in the treatment of
snoring were not fully known.
SNAP® Methods
A sonographic system, SNAPÒ (SNAP
Laboratories, Glenview, Illinois), was
utilized to record all patients prior to LAUP and following each procedure. Each patient
had between one and three LAUP procedures. The recordings were done before the patient
returned for their next procedure; thus, most recordings were done in 4 to 6 weeks after a
LAUP procedure.
All patients who entered this study, which was IRB approved, signed a
separate consent form for the recordings prior to entering the study. Using this system,
oronasal respiration was recorded continuously during 4 hours of sleep via a cannula
positioned on the patient's upper lip, with extensions positioned over the oral and nasal
apertures. The cannula was connected to an electret microphone and digital audio tape
recorder (DAT) (Sony TCD-D7). This recorded data are then digitized at 2.3 kHz to a
computer hard disk and analyzed using temporal, fast Fourier transform (FFT), pattern
recognition, and statistical analysis. Although the process is automated, all results are
verified by the system operator listening to and confirming statistically selected samples
of analyzed data.
The SNAPÔ system conforms to standards
recommended for the recording and analysis of respiratory sounds. The electret condenser
microphone and DAT recorder have a flat frequency response of 50 to 14,500 Hz, a
signal-to-noise ratio of 87 dB, dynamic range of 87 dB, and wow and flutter below
measurable limits. The FFT algorithm utilizes a Hanning window and amplitude spectrum. It
defaults to a 1024-point resolution, which can be reduced to 256 points, allowing the
accurate processing of nonstationary signals when necessary. While the recording system is
initially calibrated to an absolute sound-intensity level, all parameters are determined
relative to a baseline of quiet respiration rather than of an absolute magnitude of sound.
The SNAPÔ system identifies, counts, and characterizes all
respiratory events: quiet respiration, apnea, hypopnea, and snoring. Quiet respiration or
breathing is characterized by a diffuse spectral pattern occurring periodically.
Snoring is categorized into five subgroups that are spectrally defined
and confirmed by the system operator listening to the analyzed data:
A snoring event that is dominated by a harmonic pattern or single
tone with a fundamental frequency lower than 180 Hz;
A snoring event that is spectrally more diffuse but still dominated
by frequencies lower than 180 Hz;
A snoring event with both low- and high-frequency components;
A snoring event that is more diffuse (than 5) but is still dominated
by frequency components greater than 180 Hz; or
WL (wheeze-like)—A snoring event that is dominated by a harmonic
pattern or single tone with a fundamental frequency greater than 180 Hz.
Using the above classification the following parameters are calculated
by the SNAPÔ system: snoring index (SI), maximum relative
respiratory noise loudness (MRL), average relative snoring loudness (ARL), percentage of
low frequency (velum-like) snores (VL%), low frequency (velum-like) snoring relative
loudness (VRL), and the average fundamental frequency of snoring (FREQ) (Table I).
LAUP Surgical Technique
The same physicians, R.P.W. and W.M.G., performed
all of the LAUP procedures, which were done in a clinic using standard C02
laser precautions. Anesthesia began with 10% lidocaine spray (Xylocaine) applied to the
palate and the base of the tongue. This was followed by infiltration of either side of the
base of the uvula and, occasionally, the base of the uvula with a total of 2 mL of a
mixture of lidocaine, epinephrine, and triamcinolone acetonide (Kenalog). The procedure
was performed with a CO2 laser with handpieces designed to protect the
posterior pharyngeal wall. The laser power was set at 15 to 20 W, and vertical
transpalatal incisions, approximately 1 cm long,
were made bilaterally through the soft palate just lateral to the base of the uvula. This
was followed by partial vaporization of the uvula.
After the procedure, patients were observed for 10 to 20 minutes and
then sent home. Postoperative medications typically included ampicillin suspension for 7
days and acetaminophen with codeine elixer and/or anesthetic lozenges as needed for pain.
In addition, many of the patients received perioperative oral steroids.
Postoperative Evaluation
Patients were seen approximately every 4 to 6 weeks. All patients
completed an interview by an independent observer preoperatively and after each LAUP
procedure. Subjective reports were obtained from the patients and bed partners when
available. A scale of -2 to +2 was used to subjectively report on the snoring (Table II).
Statistical Analysis
All computations were performed on a personal computer using KWIK-STAT4
system software. Newman-Keuls multiple comparisons was used to determine significance in
all comparisons unless otherwise noted with statistical significance of P<.05. This
test is an all pairwise comparison of every combination of recording pairs. We can
conclude from large values of q that the difference of the two recordings being compared
is statistically significant. The data are expressed as the mean±standard error of the
mean.
RESULTS
Of the 27 habitual snorers tested, 6 (22% were women and 21
(78%) were men. Patient age ranged from 32 to 80 years, with a mean of 50.8 years. The
body mass index (BMI) ranged from 17.5 to 35.2 with a mean of 27.5. All patients were
recorded preoperatively, 16 were recorded after the first LAUP procedure, 24 after the
second LAUP procedure, and 20 after the third procedure. All patients had at least two
recordings, one preoperatively and
one after a LAUP procedure. The recordings were performed in the patients' homes. The
parameters which were determined for each recording session included the SI, MRL, ARL,
VRL, VL%, and FREQ. Each patient was tested prior to the next LAUP session with recording
times ranging from 3 weeks post-LAUP to 4 months following their last session. Most
patients were recorded 1 month following the LAUP procedure. None of the patients had
other surgical procedures performed in conjunction with the LAUP procedures.
The snoring index was 353.6±36.3 preoperatively and decreased to
332.0±22.8 post-LAUP #1, to 295.6±33.0 post- LAUP #2, and to 245.5±35.9 post-LAUP #3
(Fig. 1).
The maximum relative respiratory noise loudness (MRL) was 18.8 dB±1.0
preoperatively and decreased to 17.3 dB±1.3 post-LAUP #1, to 13.7 dB±1.3 post-LAUP #2,
and to 14.1 dB±1.1 post-LAUP #3. The comparison of preoperative levels to post-LAUP #2
and post-LAUP #3 were statistically significant (P<.05) (Fig. 2).
The average relative respiratory noise loudness (ARL) was 12.7 dB±0.8
preoperatively and decreased to 12.4 dB±1.4 post-LAUP #1, to 9.7 dB±1.2 post-LAUP #2,
and to 8.7 dB±1.0 post-LAUP #3. The comparison of the preoperative level to post-LAUP #3
was statistically significant (P<.05) (Fig. 2).
The low frequency (velum-like) snoring relative loudness (VRL) was 13.9
dB±1.5 preoperatively and decreased to 10.1 dB±1.7 post-LAUP #1, to 5.5 dB±1.2
post-LAUP #2, and to 4.5 dB±1.4 post-LAUP #3. The comparison of the preoperative level to
post-LAUP #2 and post-LAUP #3 as statistically significant (P<.05) (Fig. 2).
The percentage of low frequency (velum-like) snores (VL%) was 87.2±1.9
preoperatively and decreased to 70.8±6.8 post-LAUP #1, to 64.0±4.9 post-LAUP #2, and to
59.6±6.8 post-LAUP #3. The comparison of the preoperative level to post-LAUP #1,
post-LAUP #2, and post-LAUP #3 was statistically significant (P<.05) (Fig. 3).
The average fundamental frequency of snoring (FREQ) was 106.4 Hz±10.0
preoperatively and increased to 176.7 Hz±34.0 post-LAUP #1, to 156.4 Hz±15.2 post-LAUP
#2, and to 178.6 Hz±17.1 post-LAUP #3. The comparison of the preoperative level to
post-LAUP #1, post-LAUP #2, and post-LAUP #3 was statistically significant (P<.05)
(Fig. 4).
The subjective results were also obtained and compared to the objective
results for each patient at each recording session. The subjective scale of -2 to
+2 (Table II) was, utilized, with -2 being a significant worsening in the snoring and +2
being a significant improvement. The low frequency (velum-like) snoring relative loudness
difference (VRL diff) is defined as the postoperative VRL minus the preoperative VRL.
Thus, a negative VRL difference is an improvement. At each point on the subjective scale,
the average VRL diff was plotted for all of the procedures. The comparison of the patients
who reported significant worsening to the patients who reported significant improvement
was statistically significant (P<.05). Thus, the comparison between the VRL diff and
subjective responses demonstrated a good correlation (Fig. 5).
DISCUSSION
Snoring is a common disorder which affects at least 30% of the adult
population. Lugaresi et al report that 60% of men and 40% of women between the ages of 40
to 65 snore nightly.19 Young et al. reported in 1993 a large series focused on
sleep-disordered breathing among middle-aged adults. In the initial stage of this study, a
questionnaire was sent to subjects, and self-reported results of 3513 respondents showed
that 28% of women and 44% of men between the ages of 30 and 60 were habitual snorers.20
Most epidemiologic data on snoring comes from self-reporting, which is
potentially limited. Wiggins et al. compared self-reporting to spouse reporting in 360
couples and found that in men, spouse reports yielded a higher prevalence rate of snoring,
33% on self-report vs. 43% on spouse report. In women, self-report yielded similar rates
as the spouse report of snoring 18% vs. 17%. This author recommends that both the spouse
or bed partner and patient should be questioned about sleep-related Symptoms.21
Hoffstein et al. compared the subjective perception of a sleep technologist to objective
measurement of snoring during 613 polysomnograms. They found that the trained
technologist's perception was qualitative at best. In addition, 471 patients in this study
were unaware that they snored.2 Thus the need
for accurate objective testing is
essential. Especially when attempting to document the efficacy of a new procedure for the
treatment of snoring, subjective data is not adequate.
One purpose of objective testing is to identify the components of the
snoring noise that are annoying to the human ear. Another purpose is to measure outcomes
after a surgical intervention. Although there remains no standardization of what spectral
characteristics make up the snoring sound, it appears that the low frequency velum-like
respiratory noise correlates best with the patients' subjective perception of snoring in
this series.22,23 One advantage of the SNAPÔ system
is the placement of the recording cannula at the oral and nasal orifices, which allows for
an extremely sensitive detection of respiratory sounds. This system is not
position-dependent because the recording cannula is securely attached to the patient's
head throughout the recording. In addition, the respiratory sounds were not selectively
sampled, which ensures a more accurate assessment of the respiratory sounds.
No objective measurements of the snoring sound changes following the
LAUP procedure have been published. However, the subjective reported results of LAUP for
snoring are very encouraging.15-17 In Walker's initial study of 105 snorers who
completed LAUP treatment, 60% of patients and bed partners reported complete or near
complete elimination in the snoring, 29% noted a partial improvement, and 10% noted no
improvement.15 Kamami's most recent report on the effects of LAUP on snoring
showed that of 517 patients who completed treatment 70% had complete or near complete
elimination in the snoring, 25% noted a partial improvement, and 5% showed no improvement.24
These LAUP results are comparable to the UPPP subjective results.25-32
The UPPP literature shows that short-term improvement in snoring occurs in 76% to 95% of
treated patients. There is also objective data in the UPPP literature comparing
preoperative and postoperative snoring. Miljeteig et al. evaluated 69 patients who
underwent UPPP for snoring and obstructive sleep apnea. Snoring was measured during a
polysomnogram with a sound meter and a microphone at the nasion. The parameters they
examined were snoring index, the mean and the maximum sound intensity, as well as the
subjective responses of the patients. The results demonstrated no difference between the
preoperative and postoperative snoring index or the mean or maximum sound intensity. The
follow-up polysomnogram. was 13±15 months postoperatively. However, the patients reported
a 78% reduction or disappearance in snoring. This author concluded that snoring is
difficult to quantitate objectively.13 Schafer recorded 20 patients prior to
and following UPPP. The snoring sounds above a preset sound pressure level were
recorded and frequency spectrum analysis was performed. After UPPP, sound pressure level
was reduced and the low-frequency content of the spectrum below
400 Hz was also reduced.12 Weingarten objectively
assessed 8 patients prior to and following snare uvulopalatoplasty with the SNAPÔ device. He found that there was a significant decrease in the ARL
and VRL and there was an elevation in the fundamental frequency. The snoring index did not
change significantly in this study.33
The findings of the present study suggest that LAUP alters the snoring
sound. The maximum, average, and velum- like respiratory noise loudness (MRL, ARL, and
VRL) all showed a statistically significant decrease when the preoperative snoring was
compared to the final recording after treatment was completed.' The low frequency
(velum-like) snores decreased the most. In addition, the fundamental frequency of the
snoring increased after each LAUP procedure. Although the snoring index did not change
significantly, there was a downward trend from the preoperative to the final postoperative
recording.
Our objective results compared favorably with the subjective responses
of the patients. An independent person interviewed the patient and bed partner by
telephone or in person each time a recording was submitted for analysis. An independent
person was chosen to perform the interview to help avoid the bias of a patient reporting
directly to the treating physician. We found that the velum- like relative loudness
correlated best with the subjective perception of the snoring. It is of note that although
all patients ultimately were found to have both a subjective and objective improvement in
their snoring, several patients noted a temporary worsening in their snoring after a
specific LAUP session. This subjective worsening or lack of improvement was detected with
the recording. Thus it is not uncommon for a patient's snoring to fluctuate during the
series of procedures, and patients need to be aware of these changes.
LAUP is a relatively new procedure for the treatment of snoring. With
the development of less invasive, cost- effective procedures that can be performed in the
outpatient setting, there has been a renewed interest in the surgical treatment of
snoring. According to the American Sleep Disorders Association Report on LAUP, more
objective data are needed to document the efficacy of this procedure for the treatment of
snoring.34 The findings in the present study help to address such concerns.
LAUP significantly alters the snoring noise in a favorable manner by both objective and
subjective standards. The SNAPÔ recording device allows for an
accurate assessment of oronasal respiration and correlates well with the subjective data.
Future studies will need to accumulate further objective data on LAUP patients. In
addition, follow-up studies are needed to determine the long-term effectiveness of LAUP in
mitigating or arresting snoring.
ACKNOWLEDGMENTS
The authors recognize the invaluable assistance of Dr. Gil Raviv
and Mr. Ron Elesh (SNAP Laboratories,
Glenview, Ill.), Ms. Sandra Argento, and the generous contribution of the Loyola
University Department of Otolaryngology Research Fund. |
