Introduction
Obstructive sleep apnea (OSA) is a breathing disorder that occurs during sleep, is
characterized by frequent attacks of upper airway collapse, with a subsequent reduction
(hypopnea) or absence (apnea) in airflow lasting for at least 10 seconds and accompanied
by either cortical arousal or a fall in blood SPO2.[1]
The pathophysiology of this syndrome is complex and multifactorial. Several factors
influence the stability of the pharynx and might contribute to its closure during
sleep when the activity of the pharyngeal dilating muscles is considerably reduced.[2]
Anatomically, the dorsum of the tongue is anterior to the soft palate. Based on the
observation of posterior displacement of the tongue during obstructive apnea, Isono
et al. hypothesized that the dorsum of the tongue pushes the anterior wall of the
soft palate posteriorly during inspiratory efforts, maintaining closure of the retropalatal
airway. On the other hand, they evaluated the dynamic interaction between the tongue
and the soft palate during OSA by manometric studies on 14 patients with sleep-disordered
breathing. They found close apposition between the tongue and soft palate in all patients,
and a progressive increase in contact pressure during OSA.[3]
Drug-induced sleep endoscopy (DISE) is widely performed, and several studies have
demonstrated its validity; in fact, it provides clinical information not available
by routine clinical inspection alone.[4] Another diagnostic tool that is useful in conjunction with the standard DISE is
the transoral fiber-optic endoscopy, which could give additional information in selected
patients if the mouth is open.[5]
Our study aims to assess the role of transoral drug-induced sleep endoscopy in the
evaluation of tongue-palate (TP) interaction (the absence of space between tongue
and palate with the visual impression that the tongue is pushing the soft palate),
and its impact on surgical outcomes in such patients.
Methods
This study included 42 patients, 24 males (57.1%) and 18 females (42.9%), and their
ages ranged between 25 and 54-years-old. This research was performed in the university
hospitals from August 2019 to December 2020 and approved by the university's ethical
committee with the code MS/19.07.719. Informed written consent was taken from all
patients after explanation.
Inclusion Criteria
We included adults over 18-years-old, with OSA syndrome, fit for general anesthesia,
having failed or refused CPAP therapy, with negative tongue collapse according to
the transnasal DISE.
Exclusion Criteria
The exclusion criteria were patients unfit for general anesthesia, those with previous
sleep surgeries, cases with tongue base and laryngeal collapse according to the transnasal
DISE, and cases with craniofacial anomalies.
Evaluation
A thorough medical history was obtained using a visual analogue scale (VAS) for snoring[6] and the Arabic version of the Epworth Sleepiness Scale (ESS).[7] A general examination was performed concerning body mass index (BMI) and neck circumference.
Additionally, an ENT examination was performed for a deviated septum, hypertrophied
turbinates, the Friedman tongue position (FTP), as well as the oral cavity for the
tongue, uvulae, tonsils (Brodsky scale, in five grades), and the relationship between
the tongue and soft palate to determine whether an enlarged tongue obscures the vision
of the palate.
Polysomnography
The polysomnography examination was performed preoperatively and 6 months postoperatively,
utilizing the SOMNOscreen plus (SOMNOmedics AG, Randersacker, Germany) device to diagnose
and evaluate the degree of OSAS by a chest consultant (the last author). Apnea and
hypopnea were the main obstructive events in OSA, with apnea being described as the
stoppage of airflow at the nostrils and mouth for at least 10 seconds, and hypopnea
as a decrease of more than 50% in oronasal airflow for at least 10 seconds accompanied
by more than 4% decrease in oxyhemoglobin saturation from baseline level and/or arousal.[8]
Drug-Induced Sleep Endoscopy (DISE)
Patients were placed in a supine position with basic cardiorespiratory monitoring.
The required depth of sedation was to reach the unconscious state, in which the subject
started to snore.
Sleep was induced using propofol in a 1.5 mg/kg dose as a bolus and then maintained
with the simple manual controlled infusion. This drug allowed higher sedation throughout
the process of endoscopy.[9] Greater degrees of sedation were accompanied by a marked decrease in upper airway
dilator muscle tone and neuromuscular reflex stimulation, raising airway collapsibility
and avoiding oversedation during DISE. We did not record or analyze the collapse that
happened while oxygen saturation was lower than the saturation nadir in PSG.
A transnasal DISE was performed, with a flexible endoscope lubricated with lidocaine
gel being introduced to the nasal cavity. The observation was conducted according
to the lateral pharyngeal wall, palate, tongue, and larynx (LwPTL) classification,
where the sites generating snoring and obstruction were assessed. Then, a transoral
DISE was performed as the endoscopy was smoothly pushed among both incisors without
the need to force the mouth open, to evade the effect of mouth breathing on the upper
airway structure. Positive TP interaction was defined as the absence of space between
tongue and palate with the visual impression that the tongue pushed the soft palate,[10] as shown in [Figure 1].
Fig. 1 Endoscopic imaging showing (A) positive and (B) negative TP interactions.
Surgical Technique
The updated Cahali lateral pharyngoplasty[11] was performed in all cases, relying mainly on a palatopharyngeus muscle flap by
the senior author.[12]
Follow-up
All patients were postoperatively checked at 1-week, 2-weeks, 1-month, and 6-months
as outpatient visits to assess any complications and manifestation relief in terms
of snoring according to the visual analogue scale and sleepiness according to the
ESS. Additionally, polysomnography was performed 6 months postoperatively to calculate
the apnea–hypopnea index (AHI) and other respiratory parameters to compare it to the
preoperative data. Finally, the Sher success criteria were used to define the surgical
success of the operation; 50% reduction of the preoperative AHI and of the < 20 events/hour.[13]
Statistical Analysis
Data were fed to the computer and analyzed using the Statistical Package Social Sciences
(SPSS, IBM Corp., Armonk, NY, USA), version 22.0 for Windows. Qualitative data were
described using numbers and percentages. Quantitative data were described using the
mean and standard deviation (SD) for parametric data after testing normality with
the Kolmogorov–Smirnov test. The significance of the obtained results was judged at
the 0.05 level, and a highly significant difference was present if p ≤ 0.001. The qualitative data analysis was performed using the Chi-squared test to
compare two or more groups, and the Monte Carlo test as correction when more than
25% of cells have to count less than 5 in tables (>2*2). Data analysis of quantitative
data was performed using a Student t-test to compare two independent groups, while the paired t-test was to compare two studied periods. We used the Z test for comparing proportions.
Binary stepwise logistic regression analysis was used for the prediction of independent
variables with a binary outcome. Significant predictors in the univariate analysis
were entered into the regression model using the forward Wald method. Adjusted odds
ratio (OR) and their 95% confidence interval (CI) were calculated.
Results
This study included 42 patients, 24 males (57.1%) and 18 females (42.9%), and the
mean age of the studied cases was 38.95 ± 6.99, with a mean BMI of 32.13, ranging
from 26 to 39.1 kg/m2. There is no statistically significant age difference between the –ve and +ve TP
interaction groups). However, there is a statistically significant higher TP interaction
among females and those with a higher BMI ([Table 1]).
Table 1
Sociodemographic characteristics of the two studied groups
|
-ve TP interaction
n = 36
|
+ve TP interaction
n = 6
|
Test of significance
|
|
Age, years (mean ± SD)
|
38.58 ± 7.14
|
41.17 ± 6.11
|
t = 0.835
p = 0.409
|
|
Sex, n (%)
|
|
Female
|
13 (36.1)
|
5 (83.3)
|
χ2 = 4.7
|
|
Male
|
23 (63.9)
|
1 (16.7)
|
p = 0.03*
|
|
BMI (kg/m2), mean ± SD
|
30.94 ± 3.87
|
35.25 ± 2.69
|
t = 2.61
p = 0.012*
|
Abbreviations: BMI, body mass index; SD, standard deviation; TP, tongue-palate; t, Student t-test. Notes: *statistically significant if p < 0.05.
There is no statistically significant difference between studied groups regarding
minimal SpO2 preoperative, while there is a statistically significant difference between both
groups postoperative with a higher percentage of improvement among –ve than the +ve
TP interaction groups (11.65 vs. 9.7%, respectively). Also, there is no statistically
significant difference between studied groups regarding snoring index preoperative,
while there is a statistically significant difference between studied groups postoperative
with a higher percentage of improvement among –ve than the +ve TP interaction groups
(48.8 vs. 25.2%, respectively), as shown in [Table 2].
Table 2
Sleep parameters before and after surgery among the studied groups
|
-ve TP interaction
n = 36
|
+ve TP interaction
n = 6
|
Test of significance
|
|
Minimal preoperative SpO2 (mean ± SD)
|
78.69 ± 5.59
|
75.50 ± 4.76
|
t = 1.32
p = 0.195
|
|
Minimal postoperative SpO2
(mean ± SD)
|
87.86 ± 3.76
|
82.83 ± 2.48
|
t = 3.15
p = 0.003*
|
|
Paired t-test
|
t = 14.17
p < 0.001*
|
t = 4.03
p = 0.01*
|
|
|
Percent of change
|
11.65%
|
9.7%
|
|
|
Preoperative
snoring index (mean ± SD)
|
157.0 ± 35.36
|
184.67 ± 18.78
|
t = 1.86
p = 0.07
|
|
Postoperative snoring index
(mean ± SD)
|
80.33 ± 32.14
|
138.17 ± 18.71
|
t = 4.26
p < 0.001*
|
|
Paired t-test
|
t = 19.43
p < 0.001*
|
t = 4.98
p = 0.004*
|
|
|
Percent of change
|
48.8%
|
25.2%
|
|
|
Preoperative AHI
(mean ± SD)
|
33.75 ± 11.74
|
46.68 ± 8.72
|
t = 2.57
p = 0.014*
|
|
Postoperative
AHI (mean ± SD)
|
15.94 ± 8.40
|
33.0 ± 4.60
|
t = 4.82
p < 0.001*
|
|
Paired t-test
|
t = 14.0
p < 0.001*
|
t = 7.31
p = 0.001*
|
|
|
Percent of change
|
52.8%
|
29.3%
|
|
|
Preoperative
ESS
(mean ± SD)
|
18.08 ± 3.90
|
20.33 ± 3.56
|
t = 1.32
p = 0.194
|
|
Postoperative ESS
(mean ± SD)
|
12.17 ± 3.49
|
17.0 ± 4.05
|
t = 3.07
p = 0.004*
|
|
Paired t-test
|
t = 11.76
p < 0.001*
|
t = 3.78
p = 0.013*
|
|
|
Percent of change
|
32.7%
|
16.4%
|
|
|
Preoperative
VAS (mean ± SD)
|
7.5 ± 2.1
|
8.2 ± 1.9
|
t = 0.765
p = 0.449
|
|
Postoperative VAS
(mean ± SD)
|
3.24 ± 0.58
|
4.5 ± 2.1
|
t = 3.11
p = 0.003*
|
|
Paired t-test
|
t = 2.5
p < 0.001*
|
t = 3.1
p < 0.001*
|
|
|
Percent of change
|
56.8
|
45.1
|
|
Abbeviations: AHI, apnea-hypopnea index; ESS, Epworth Sleepiness Scale; SD, standard deviation;
SpO2, oxygen saturation; t, Student t-test; VAS, visual analogue scale.
Note: *statistically significant if p < 0.05.
There is a statistically significant difference between the studied groups regarding
the AHI preoperative. While there is a statistically significant difference between
both groups postoperative with a higher percentage of improvement among the –ve compared
to the +ve TP interaction group (52.8 vs. 29.3%, respectively). There is a statistically
significant difference in the ESS between studied groups after surgery without significant
difference preoperative. A higher percentage of improvement among the –ve (32.7%)
compared with the +ve (16.4%) TP interaction groups was detected ([Table 2]).
There is a statistically significant difference in VAS score between studied groups
postoperative without significant difference preoperative with a higher percentage
of change among the group with –ve TP interaction (56.8 vs. 45.1%, respectively).
([Table 2])
Binary stepwise logistic regression analysis illustrates that preoperative AHI, BMI,
sex, and preoperative snoring index have no statistically significant effect on TP
interaction ([Table 3]).
Table 3
Multivariate analysis of confounding factors of TP interaction
|
B
|
SE
|
Wald
|
DF
|
p-value
|
OR
|
95% CI for OR
|
|
|
|
|
|
|
|
Lower
|
Upper
|
|
Preoperative AHI
|
0.114
|
0.081
|
1.993
|
1
|
0.158
|
1.121
|
0.957
|
1.314
|
|
BMI
|
−0.017
|
0.138
|
0.014
|
1
|
0.905
|
0.984
|
0.751
|
1.288
|
|
Sex
|
0.225
|
1.078
|
0.044
|
1
|
0.835
|
1.252
|
0.151
|
10.354
|
|
Preoperative
snoring index
|
0.001
|
0.031
|
0.000
|
1
|
0.987
|
1.001
|
0.942
|
1.063
|
|
Constant
|
−6.272
|
5.791
|
1.173
|
1
|
0.279
|
0.002
|
|
|
Abbreviations: AHI, apnea-hypopnea index; B, unstandardized regression weight; BMI, body mass index;
CI, confidence interval; DF, degree of freedom; OR, odds ratio; SE, standard error;
TP, tongue-palate.
There is a statistically significant higher percentage of overall success among the
group with negative TP interaction than the one with positive interaction (72.2% vs.
zero). However, a nonstatistically significant difference is detected between negative
and positive TP interaction as regards snoring index, minimal SpO2, AHI, and ESS percentage of improvement ([Table 4]).
Table 4
Percentage of improvement in snoring index, minimal SpO2, AHI, ESS among studied groups
|
Percentage of improvement (%)
|
-ve TP interaction
n = 36
|
+ve TP interaction
n = 6
|
Test of significance
(z test)
|
|
Snoring index
|
48.8
|
25.2
|
z = 1.07
p = 0.282
|
|
Minimal SpO2
|
11.65
|
9.7
|
z = 0.14
p = 0.889
|
|
AHI
|
52.8
|
29.3
|
z = 1.07
p = 0.287
|
|
ESS
|
32.7
|
16.4
|
z = 0.80
p = 0.422
|
|
Overall success
|
72.2
|
0.0
|
z = 3.36
p = 0.001*
|
Abbreviations: AHI, apnea-hypopnea index; ESS, Epworth Sleepiness Scale; SpO2, oxygen saturation; TP, tongue-palate.Note: *statistically significant if p < 0.05
[Table 5] showed that there was a statistically significant difference between studied groups
regarding DISE palate classification with 66.7% of the cases with +ve TP interaction
having low palatal collapse, compared with 58.3% of the cases with –ve.
Table 5
The DISE distribution among the studied groups
|
-ve TP interaction
n = 36 (%)
|
+ve TP interaction
n = 6 (%)
|
Test of significance
|
|
DISE larynx
|
|
|
FET
p = 1.000
|
|
L0
|
36 (100)
|
6 (100)
|
|
L1
|
0 (0.0)
|
0 (0.0)
|
|
DISE tongue
|
|
|
MC
p = 1.000
|
|
T0
|
36 (100)
|
6 (100)
|
|
TH
|
0 (0.0)
|
0 (0.0)
|
|
TL
|
0 (0.0)
|
0 (0.0)
|
|
DISE palate
|
|
|
MC
p = 0.001*
|
|
P0
|
15 (41.7)
|
0 (0.0)
|
|
PHL
|
0 (0.0)
|
2 (33.3)
|
|
PL
|
21 (58.3)
|
4 (66.7)
|
|
DISE lateral wall
|
|
|
MC
p = 0.478
|
|
LH
|
4 (11.1)
|
0 (0.0)
|
|
LS
|
5 (13.9)
|
0 (0.0)
|
|
LV
|
25 (69.4)
|
6 (100)
|
|
LW0
|
2 (5.6)
|
0 (0.0)
|
Abbreviations: DISE, drug-induced sleep endoscopy; FET, Fischer exact test; L0, no laryngeal collapse;
L1, laryngeal collapse; LH, lateral wall at hypopharynx; LS, lateral wall at salpingopharyngeal
folds; LV, lateral wall at the velum; LW0, no collapse at the lateral wall; MC, Monte
Carlo test; P0, no palatal collapse; PHL, high palate collapse of the vertical palate;
PL, low palate collapse of the oblique palate; T0, no tongue collapse; TH, high tongue
base collapse; TL, low tongue base collapse; TP, tongue-palate.Note: *statistically significant.
In our study, the Friedman tongue position (FTP) in the +ve TP interaction group was
grade III in 1 patient, II in 3 patients, and I in 2.
There were 2 cases (4.76%) with secondary bleeding and 2 (4.76%) with velopharyngeal
insufficiency in terms of complications.
Discussion
The current was a prospective interventional study conducted on 42 cases who underwent
transoral sleep endoscopy, demonstrating that 6 cases had +ve and 36 had -ve TP interaction.
The cases were operated on by updated lateral palatopharyngoplasty at a tertiary university
hospital.
Cases with either high or low tongue base collapse during transnasal DISE were excluded
from the study because they needed tongue surgery. Still, those with TP interaction
without tongue collapse were the subject of study whether this interaction affected
the response of palatopharyngoplasty surgery or not, and if this necessitated additional
management. In our study, positive TP interaction meant a visual impression of the
oropharyngeal tongue pushing the soft palate, causing either its collapse or at least
limiting the oral airflow.
We performed the DISE to assess the level and the pattern of obstruction, as the studies
comparing it to awake evaluations showed alteration in the surgical management plan
in approximately 50% of the patients.[14] The drug propofol was used in our study, as most studies comparing sedation and
natural sleep used it or midazolam as a single agent. These drugs had an effect that
mimics the critical closing pressure during natural sleep without a significant difference
in the AHI.[15]
[16]
[17]
Other studies[18]
[19] used simulation models to detect the level of obstruction depending on computational
fluid dynamics (CFD). However, in the present study, these tools could not be used
their unavailability in our country and the dependence on DISE to detect the level
of collapse pattern.
We selected a new Cahali[11] lateral pharyngoplasty with the concept that it can correct all the retropalatal
collapse in all dimensions, and also it can stiffen the lateral pharyngeal wall. The
study by Cahali proposed that all OSA patients will benefit from the new lateral pharyngoplasty
procedure, regardless of the level and pattern of airway obstruction, as neither the
thickness of the posterior tonsillar pillar nor the soft TP position are selection
factors. His technique for treating sleep apnea has evolved due to factors such as
the retropalatal airway being the primary site of obstruction, TP coupling, lateral
wall extension, lack of anteroposterior enlargement of the retropalatal area in previous
LPs, and residual supine obstruction. His approach was modified to prevent stretching
of the pharyngeal mucosa, promote retropalatal expansion, and splint the upper lateral
pharyngeal wall with a myomucosal palatopharyngeus flap.[11] Therefore, if the new lateral pharyngoplasty was properly done, any residual collapse
should be either at the level of the tongue or larynx, which can be excluded by the
routine transnasal DISE, or the palate is being pushed by the tongue, which the transoral
DISE can demonstrate.
Elzayat et al.[20] conducted their study on 40 cases undergoing the new Cahali pharyngoplasty operation.
They concluded that the technique could be used as a standalone procedure in all OSA
patients except those who had a lateral wall collapse at the level of the hypopharynx
(LH), high tongue base collapse (TH), laryngeal collapse (L1), and TP interaction.
Our definition of positive TP interaction as absent space between tongue and palate
is much different from Friedman or Mallampati scores because interaction usually happens
on a dynamic basis, and when revising the static examination data of the 6 positive
cases, we found they were not necessarily high grades Friedman or Mallampati.
In our study, the Friedman tongue position (FTP) in the +ve TP interaction group was
grade III in 1, II in 3, and I in two patients, meaning that a high Friedman tongue
position can't be used as a predictor for positive TP interaction.
According to the European position paper on DISE,[5] the transoral type can be used to assess the degree of tongue retraction and position,
as it could be evaluated from the oral cavity and nasopharynx, highlighting a secondary
anteroposterior soft palate collapse, due to the tongue position but they did not
study its effect on the management plan.
Additionally, Elsobki et al.[10] conducted a cross-sectional study on a total of 30 cases with OSA who underwent
DISE according to a new classification system called LwPTL. They demonstrated that
93.3% of the cases presented lateral pharyngeal wall collapse, usually at the level
of the velum (73.3%), and 80% presented multilevel collapse. They classified the findings
of transoral DISE into TP contact, which occurred between the oropharyngeal tongue
and soft palate laterally, and this normally happened in all patients and TP interaction
with a visual impression of the tongue pushing the palate back. They mentioned that
either positive or negative TP interaction did not affect treatment plans due to lack
of standardization.
There is no marked difference in age between both groups. However, +ve TP interaction
in females is marked higher than –ve, while BMI in +ve cases is higher than in –ve
ones.
In terms of outcomes, the success rate of the studied cases was 61.9% after lateral
pharyngoplasty. In addition, there was a marked difference among studied groups in
terms of success rate after surgery; with 100% of cases with +ve TP interaction failed
versus 27.8% of cases with –ve. The 100% failure does not mean that patients with
positive interaction did not gain any improvement but means that their improvement
did not meet Sher's success criteria.
Similarly, Elzayat et al.[20] conducted their study on 40 cases with known OSAS undergoing the new Cahali pharyngoplasty
operation. They revealed that there were 28 (70%) cases with successful surgery results.
Yi et al.[21] also reported a comparable success rate (64.7%) in their study on OSA cases that
underwent Z-palatopharyngoplasty (ZP3).
The current study demonstrated a marked difference in minimal oxygen saturation and
snoring index among studied groups postoperatively (p = 0.003* and < 0.001*, respectively). In harmony with the current study, Park et
al.[22] reported that various parameters such as respiratory disturbance index (RDI), lowest
O2 saturation, mean O2 saturation, oxygen desaturation index, supine AHI, and ESS markedly improved after
surgeries in both groups.
In the same line, El Sobki et al.[12] demonstrated that postsurgical STOP-BANG score, AHI, and snoring index were markedly
diminished compared with presurgical data. On the contrary, minimal and baseline SpO2 were markedly elevated after surgery.
Regarding AHI, the current study demonstrated a marked difference among studied groups
in the preoperative score. At the same time, there is a marked difference among studied
groups postoperatively, with a higher percentage of improvement among the –ve compared
with the +ve TP interaction group. In the same line, Dizdar et al.[23] displayed that the average presurgical AHI of lateral pharyngoplasty cases was 23.5,
and the average postsurgical was 11.5. Thus, the average minimal SpO2 markedly increased postsurgically in both groups; on the other hand, it was markedly
reduced in the lateral pharyngoplasty group compared with the uvulopalatopharyngoplasty
(UPPP) ones.
In terms of the ESS, the current study demonstrated a marked difference among studied
groups after surgery, without significant difference presurgery. A higher percent
of improvement was detected among the –ve TP compared with the +ve TP interaction
group.
Regarding the VAS score of snoring, the present study revealed that changes in VAS
score after lateral pharyngoplasty were significant in both –ve and +ve TP interaction
groups (3.24,4.5) respectively, with p = 0.003*.
The study by Li-Ang Lee and Jen-Fang Yu[24] used two procedures to treat snoring, palatal implant and radiofrequency surgery,
showing changes in VAS scores following surgeries were significant in the two groups.
The current study displayed a markedly higher average improvement among cases with
–ve TP interaction than those with +ve in terms of snoring index, AHI, and ESS. In
the same line, Elzayat et al.[20] studied the relationship between the findings of DISE and the operation's success;
they reported that transoral DISE can predict operation outcomes while the level of
palatal collapse had no significant correlation with the outcome of the operation.
Pang et al. described the modified cautery-assisted palatal stiffening operation (CAPSO)
approach, which is performed under local anesthesia.[25] This approach has shown promising outcomes for patients with snoring and mild OSA.[25] In 2009, the updated CAPSO procedure was anterior palatoplasty, focusing on the
anterior surface of the soft palate.[26]
Barbed anterior palatoplasty is a modified technique, with an identical procedure
to the traditional one, with the addition of barbed thread to suspend the suture across
multiple mucosa and muscle planes without tying knots.[27] Pang et al.[28] performed a systematic review that has shown that anterior palatoplasty has comparably
favorable results to other methods of palatal surgery in adults. The procedure is
simple to perform, anatomically sound, and has minimal complications.
There were 2 cases (4.76%) with secondary bleeding and 2 (4.76%) with velopharyngeal
insufficiency in terms of complications. In harmony with the current study, Elsobki
et al.[12] demonstrated minimal postsurgical complications with no significant long-term morbidity
as; secondary bleeding in 2 (4.3%) cases and 1 patient (2.2%) had velopharyngeal insufficiency.
Additionally, Park et al.[22] reported that velopharyngeal insufficiency developed in 1 patient, and postsurgical
hemorrhage in 4 patients (10%), with stopped immediately in all cases without the
need for emergency surgeries.
Palatopharyngeal expansion is usually expected to cause temporary VPI because of the
extra effort needed by the velar structures to close the intentionally widened sphincter,
with compensation being the rule rather than the exception.[11] Our cases were treated by speech therapy.
According to the results of our study, patients with positive TP interaction do need
additional treatment rather than isolated palatopharyngeal expansion, even if their
tongues were not collapsing during the routine transnasal DISE. However, further research
is needed to evaluate different treatment plans.
The transoral DISE had disadvantages, such as difficulty of introduction, and the
need to keep the mouth slightly open which is not physiological. It also provides
no data about the nose condition or palatal collapse. The current lack of standardization
for its applicability should also be considered.
To the best of our knowledge, this is the first study designed to thoroughly evaluate
the effect of positive TP interaction on decision-making in sleep surgery and the
possible role of transoral DISE in its evaluation.
One major limitation of this study is its dependence on the only examiner's subjective
impression or visual inspection. Future standardization using computer-assisted evaluation
may overcome this limitation. Additionally, a larger number of cases and longer observation
periods are needed to assess any delayed complications or failures. Another limitation
was the absence of a bispectral (BIS) index monitor and target-controlled infusion
(TCI) in our institution; however, we always strive to make our DISE data reliable.
As such, this last limitation, as well as collapse happening while O2 saturation is lower than the nadir in polysomnography, was overlooked.
Transoral DISE is a complementary tool to the conventional transnasal exam, with the
objective being to exclude TP interaction to prevent the oral tongue from pushing
the soft palate, contributing to its collapse and thus making isolated palatopharyngeal
surgery less successful. This was corroborated by our results, despite its small sample
size. Other limitations of our work were cases of central retropalatal collapse (which
require modified anterior palatoplasty) and lack of positional endoscopy.
