102 / 232
outcomes were stratified by preoperative OSA severity
level, which showed a mean improvement in oAHI of
0.93, 4.3, and 17.6 for mild, moderate, and severe dis-
ease, respectively (Table IV).
DISCUSSION
In children with DS, lingual tonsil hypertrophy and
persistent OSA following T&A, LT significantly improved
AHI, oAHI, and the oxygen saturation nadir. After sur-
gery, 19% had complete resolution of OSA, whereas an
additional 42.9% had only residual mild disease. At the
same time, the median oxygen saturation nadir
improved significantly. In this small sample, we did not
see any difference in PSG outcomes by age. Fourteen
percent of the children had severe OSA after their LT,
and all of these patients had moderate or severe disease
before surgery. Two patients had worsening of the oAHI
following the procedure (oAHI from 6.9 to 10.3 and 6.4
to 6.7), which were not clinically significant and could be
a result of test-retest issues with the PSG.
A recent review of non–continuous positive airway
pressure treatment options for children with persistent
OSA following T&A, found that LT was the most com-
monly reported surgical intervention.
10
An additional
study regarding children who underwent LT reported a
mean improvement of the respiratory disturbance index
from 14.7 to 8.1 events/hour in 26 patients.
13
This study
population was more heterogeneous than our population,
as 46% had no comorbidities, and the remaining patients
had a variety of syndromes (including DS), although the
percentage of patients with DS was not reported. Anoth-
er study by Abdel-Aziz et al. reviewed 16 children who
underwent LT and noted a mean improvement in AHI
from 10.5 events/hour before surgery to 3.2 events/hour
postoperatively; one patient in the study had DS.
14
Truong et al. reported on 27 children who underwent
LT, and demonstrated that this procedure decreased the
mean AHI from 18.3 to 9.7 events/hour (
P
<
.05).
15
Their
population was again heterogeneous, with 26% of
patients having comorbidities that included DS, cranio-
facial syndromes, and other neurologic diseases.
It has been shown that there is a high likelihood
that patients with DS have multiples sites of upper air-
way obstruction associated with their OSA.
9
In addition,
airway obstruction in patients with DS is further compli-
cated by the presence of muscular hypotonia with pha-
ryngeal hypotonia, midface hypoplasia, glossoptosis, and
relative macroglossia. In light of this, one might expect
a significant failure rate and resolution rates to be lower
in children with DS when compared to children without
comorbidities. When comparing our results to previously
published reports, we find similar rates of improvement
for our cohort with DS as were seen in mixed patient
populations (children with and without comorbidities).
One reason for this better than expected resolution rate
could be that the patients in our study had lingual tonsil
hypertrophy diagnosed with the assistance of a dynamic
cine MRI as opposed to drug-induced sleep endoscopy.
Although many surgeons use flexible endoscopy to deter-
mine the presence or absence of lingual tonsil
TABLE II.
Mean and Median Polysomnography Outcomes for Children With Down Syndrome Who Underwent Lingual Tonsillectomy for Obstructive Sleep Apnea After Adenotonsillectomy.
Preoperative PSG
Postoperative PSG
Overall Change
P
Value
AHI, events/hr, mean (SD), median [range]
14.1 (12.2), 9.1 [3.8 to 43.8]
5.9 (6.3), 3.7 [0.5 to 24.4]
5.1 [
2
2.9 to 41]
<
.0001
Obstructive AHI, events/hr, mean (SD), median [range]
13.0 (11.8), 7.6 [2.9 to 43.8]
4.9 (5.7), 2.8 [0 to 22.2]
5.3 [
2
2.9 to 41]
<
.0001
O
2
nadir, %, mean (SD), median [range]
0.84 (0.08), 0.86 [0.58 to 0.91]
0.89 (0.05), 0.91 [0.76 to 0.94]
2
0.05 [
2
0.31 to 0.08]
.004
Percent REM, %, mean (SD), median [range]
19.6 (8.0), 19 [0 to 35]
18.8 (5.3), 19 [8 to 30]
0.02 [
2
0.15 to 0.18]
.67
% time CO
2
>
50 mm Hg, mean (SD), median [range]
41.6 (37.2), 34.5 [0 to 99]
33.6 (42.4), 5 [0 to 100]
8 [
2
91 to 87.6]
.52
Apnea index, events/hr, mean (SD), median [range]
5.3 (5.9), 2.6 [0.12 to 17.1]
1.9 (2.1), 1.2 [0.16 to 6.6
g
1.3 [
2
1.71 to 14.82]
.013
Hypopnea index, events/hour, mean (SD), median [range]
7.6 (7.5), 5.7 [1.03 to 32.1]
4.1 (4.6), 2.3 [0.58 to 15.5]
3.8 [
2
8.01 to 29.32]
.012
Maximum ET CO
2
, mm Hg, events/hr, mean (SD), median [range]
53.8 (6.6), 53 [44 to 66]
52.3 (6.4), 52 [42.7 to 69]
1 [
2
12 to 16.3]
.89
Central index, events/hr, mean (SD), median [range]
1.14 (1.14), 0.8 [0 to 3.9]
0.95 (1.24), 0.3 [0 to 4.2]
0 [
2
2.3 to 3.9]
.57
Postoperative oAHI 1 event/hr, n (%)
0
4 (19.1%)
.07
Postoperative oAHI 5 events/hr, n (%)
4 (19.1%)
13 (61.9%)
.007
The number for each field is based upon all 21 patients’ results unless otherwise noted.
Pre- and postoperative changes in continuous variables were tested with the Wilcoxon signed rank test, whereas changes in categorical variables were tested with the McNemar test.
AHI
5
apnea-hypopnea index; ET
5
end tidal; MA
5
mixed apnea; OA
5
obstructive apnea; oAHI
5
obstructive apnea-hypopnea index; PSG
5
polysomnography; REM
5
rapid eye movement; SD
5
standard deviation.
Prosser et al.: PSG Outcomes of Lingual Tonsillectomy in DS
80