ARTICLE IN PRESS

Sleep Medicine ■■ (2015) ■■–■■

Contents lists available at ScienceDirect

Sleep Medicine
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / s l e e p

Original Article

Rapid maxillary expansion outcomes in treatment of obstructive

sleep apnea in children
Maria Pia Villa a,*, Alessandra Rizzoli a, Jole Rabasco a, Ottavio Vitelli a,
Nicoletta Pietropaoli a, Manuela Cecili a, Alessandra Marino b, Caterina Malagola b
a Neuroscience, Mental Health and Sense Organs Department, Paediatric Sleep Disorder Centre, Sant’Andrea Hospital, Faculty of Medicine and Psychology,

“Sapienza” University, Rome, Italy

b
Orthodontic Clinic Sant’Andrea Hospital, Faculty of Medicine and Psychology, “Sapienza” University, Rome, Italy

A R T I C L E I N F O A B S T R A C T

Article history: Objectives: The objectives of this study were to confirm the efficacy of rapid maxillary expansion in children
Received 1 August 2014 with moderate adenotonsillar hypertrophy in a larger sample and to evaluate retrospectively its long-term
Received in revised form 19 November benefits in a group of children who underwent orthodontic treatment 10 years ago.
2014
Methods: After general clinical examination and overnight polysomnography, all eligible children
Accepted 23 November 2014
Available online
underwent cephalometric evaluation and started 12 months of therapy with rapid maxillary expansion.
A new polysomnography was performed at the end of treatment (T1). Fourteen children underwent clinical
evaluation and Brouilette questionnaire, 10 years after the end of treatment (T2).
Keywords:
Rapid maxillary expansion Results: Forty patients were eligible for recruitment. At T1, 34/40 (85%) patients showed a decrease of
Pediatric OSA apnea–hypopnea index (AHI) greater than 20% (ΔAHI 67.45% ± 25.73%) and were defined responders. Only
Orthodontic treatment 6/40 (15%) showed a decrease <20% of AHI at T1 and were defined as non-responders (ΔAHI
Residual OSA −53.47% ± 61.57%). Moreover, 57.5% of patients presented residual OSA (AHI > 1 ev/h) after treatment. Disease
duration was significantly lower (2.5 ± 1.4 years vs 4.8 ± 1.9 years, p < 0.005) and age at disease onset was
higher in responder patients compared to non-responders (3.8 ± 1.5 years vs 2.3 ± 1.9 years, p < 0.05).
Cephalometric variables showed an increase of cranial base angle in non-responder patients (p < 0.05).
Fourteen children (mean age 17.0 ± 1.9 years) who ended orthodontic treatment 10 years previously
showed improvement of Brouilette score.
Conclusion: Starting an orthodontic treatment as early as symptoms appear is important in order to increase
the efficacy of treatment. An integrated therapy is needed.
© 2014 Elsevier B.V. All rights reserved.

1. Introduction Among all the causative factors, adenotonsillar hypertrophy is

the most common cause of childhood OSA [6,7]. Since it is a surgical
Obstructive sleep apnea (OSA) is a sleep disordered breathing char- therapy, adenotonsillectomy (AT) is limited by surgical risks and,
acterized by prolonged partial and/or intermittent collapse of the although it leads to significant improvements in respiratory indices,
airway during sleep, that interrupts normal ventilation and normal residual disease is present in a large proportion of children,
sleep patterns, with a prevalence of 1% to 5.7% in children [1–3] . especially if aged >7 years, where obesity, severe OSA before surgery
OSA is a multifactorial disease where different risk factors such and asthma are present [8]. Moreover, Guilleminault et al. reported
as craniofacial anomalies, adenotonsillar hypertrophy, obesity, al- in a recent study the recurrence of sleep respiratory symptoms on
terations in upper airway neuromotor tone and airway inflammation a cohort of OSA patients during adolescence, not depending on the
can co-exist. These lead to a decrease in nasopharyngeal airway di- standard therapies [9].
mensions that promotes a wide spectrum of symptoms ranging from In addition to large tonsils and adenoids, children with OSA may
primary snoring, to upper airway resistance syndrome, to frank OSA present narrow upper airways deriving from narrow and long faces,
[4,5]. maxillary constriction and/or high arched palates and/or some degree
of mandibular retrusion [10–12]. However these orthodontic and cra-
niofacial abnormalities in children with OSA have been widely
ignored even if, in the last decades, correction of mandibular or
* Corresponding author. Sleep Disorder Centre, Sant’Andrea Hospital, Via di
Grottarossa 1035-1039, 00189 Rome, Italy. Tel.: +39 0633775855; fax +39
maxillomandibular anomalies has been shown to improve OSA [13–18].
0633775941. Rapid maxillary expansion (RME) is a dentofacial orthopedic
E-mail address: mariapia.villa@uniroma1.it (M.P. Villa). treatment procedure commonly adopted in young patients for the

http://dx.doi.org/10.1016/j.sleep.2014.11.019
1389-9457/© 2014 Elsevier B.V. All rights reserved.

Please cite this article in press as: Maria Pia Villa, et al., Rapid maxillary expansion outcomes in treatment of obstructive sleep apnea in children, Sleep Medicine (2015), doi: 10.1016/
j.sleep.2014.11.019
 ARTICLE IN PRESS
2 M.P. Villa et al./Sleep Medicine ■■ (2015) ■■–■■

treatment of constricted maxillary arches. Several studies have shown We excluded patients with a history of previous treatment for
the short-term efficacy of orthodontic treatment with rapid max- OSA (including tonsillectomy, adenoidectomy and AT), severe ton-
illary expander with evidence of a significant improvement of OSA sillar hypertrophy (grade IV), obesity (BMI value ≥95th centile [20]),
even in children with adenotonsillar hypertrophy [14,15,18]. genetic disorders, cerebral palsy, neuromuscular diseases, cardiac
Pirelli et al. [16] demonstrated that all 31 children studied, with disease, renal disease any systemic diseases or chronic cardiore-
upper jaw contraction, oral breathing, nocturnal snoring, and OSA, spiratory or neuromuscular diseases, dysmorphism, major
achieved a normal anterior rhinometry and an apnea–hypopnea craniofacial abnormalities or associated chromosomic syndrome.
index (AHI) < 1 event per hour after four months of treatment with
RME. Our group has previously demonstrated in 14 children with 2.1. Study design
dental malocclusion, a body mass index (BMI) <85 percentile, and
OSA confirmed by polysomnography (PSG) a significant decrease The study design is shown in Fig. 1. After recruitment, all par-
in the AHI, hypopnea obstructive index and arousal index after 12 ticipants underwent a detailed personal and family history and
months of RME therapy [17]. Moreover questionnaires on daytime general clinical examination and had an ear, nose and throat (ENT)
and night-time, fulfilled before and after treatment, showed sig- and orthodontic assessment before overnight PSG (T0). Parents were
nificant decreases in the severity of symptoms. asked when their child started to present daytime and night-time
Only few studies have investigated the long-term effects of orth- symptoms.
odontic treatment in OSA by considering the growing and the skeletal All children who met inclusion criteria underwent cephalomet-
changes occurring through the years [18]. Ten of the 14 children ric evaluation and started 12 months of therapy with RME and
who completed our 12-month therapeutic trial using RME (see performed a new polysomnographic assessment (T1). Disease du-
above) performed 24 months follow-up after the end of the RME ration was defined as the time between onset of symptoms and the
orthodontic treatment. No significant changes in the AHI or in other beginning of the treatment. Parents fulfilled a questionnaire at T0
variables were observed. and T1.
Previous papers regarding orthodontic treatment, associated or Presence of daytime and night-time symptoms in children who
not with AT, studied small-size samples. For this reason, the primary had completed the 12-month therapeutic trial with RME [17] was
aim of this prospective study was to confirm our previous find- investigated through a questionnaire and clinical evaluation, 10 years
ings [17] on the efficacy of RME in children with moderate after the end of treatment (T2).
adenotonsillar hypertrophy, with a larger sample. The second aim
was to retrospectively evaluate any long-term benefit after onset 2.2. Questionnaire data
of puberty in a group of children who underwent orthodontic treat-
ment with RME 10 years ago. The participants’ parents completed the previously validated
Brouilette questionnaire [21], at T0, T1 and T2. The questionnaire
2. Methods elicited information on daytime symptoms of OSA (including sleep-
iness, irritability, headache, school problems, tiredness and oral
Children between 4 and 10 years of age who had been referred breathing) and night-time symptoms (including habitual snoring,
to our Paediatric Sleep Center (Sant’Andrea Hospital, Rome, Italy) apneas, restless sleep, and nightmares).
and satisfied the following inclusion criteria were included: clini-
cal signs of malocclusion (high, narrow palate associated with deep 2.3. Polysomnography
bite, retrusive bite or crossbite); tonsillar grading I–III [19], signs
and symptoms of OSA (including habitual snoring, apnea and rest- Standard overnight PSG recordings were obtained at baseline,
less sleep as witnessed by parents), AHI > 1 as defined by a laboratory before starting orthodontic treatment (T0) and after 12 months of
PSG recording. All the participants’ parents provided written in- treatment (T1) using a Grass heritage polygraph. The variables re-
formed consent to the study. The study procedures were approved corded included an electroencephalogram (EEG) with at least six
by the hospital ethics committee. channels (bilateral frontal, central temporal, and occipital monopolar

Fig. 1. Study design graph. , pilot study (N = 14); →, new sample enrolment (N = 26). AHI, apnea–hypopnea index; ENT, ear, nose and throat; OSA, obstructive sleep
apnea syndrome; PSG, polisomnography; RME, rapid maxillary expansion.

Please cite this article in press as: Maria Pia Villa, et al., Rapid maxillary expansion outcomes in treatment of obstructive sleep apnea in children, Sleep Medicine (2015), doi: 10.1016/
j.sleep.2014.11.019
 ARTICLE IN PRESS
M.P. Villa et al./Sleep Medicine ■■ (2015) ■■–■■ 3

montages referred to the controlateral mastoid), an electro- 2.4. Ear, nose and throat assessment
oculogram (electrodes placed 1 cm above the right outer cantus and
1 cm below the left outer cantus and referred to A1), a submental Before orthodontic assessment children underwent an ENT ex-
electromyogram, and an electrocardiogram (ECG) (1 derivation). Sleep amination to grade tonsillar hypertrophy according to a standardized
was subdivided into 30-s epochs, and sleep stages were scored ac- scale ranging from 1 to 4 [20].
cording to the standard criteria of the American Academy of Sleep
Medicine (AASM) [22]. The following conventional sleep param- 2.5. Orthodontic assessment and orthopedic therapy
eters were measured: total sleep time, defined as the time from sleep
onset to the end of the final sleep stage; sleep efficiency, defined The orthodontic evaluation detected the presence of malocclu-
as the percentage ratio between total sleep time and total record- sion. The malocclusion was classified according to Angle’s criteria
ing time (from lights-out clock time to lights-on clock time). The [24]. Lateral cephalometric films were obtained before the orth-
percentage of total sleep time in each stage was measured as follows: odontic treatment (T0). All the radiographs were taken using a
percentage of stage N1, stage N2, stage N3, and stage R (REM sleep). standardized technique, with teeth in occlusion and lip relaxed. The
Arousals were detected visually according to the criteria reported head was adjusted so that the Frankfurt horizontal plane was par-
in the recent annual for the scoring of sleep and associated events allel to the floor. All of the lateral cephalograms were traced and
by the AASM [22]. the measurements recorded by the same operator (M.C.). The method
Central, obstructive, and mixed apnea events were counted ac- error was determined by repeating the measurement process for
cording to the criteria established by the AASM [23]. Chest and 15 randomly selected radiographs again. Mean values from the first
abdomen movements were measured by strain gauges. Oronasal and second tracings were used to determine the method error
airflow was recorded by means of an oronasal thermocouple and through Dahlberg’s formula. The paired samples t-test showed no
nasal pressure. Arterial oxygen saturation (SpO2) was monitored with significant mean differences between the two series of records [25].
a pulse oximeter. The AHI was defined as the average number of Cephalometric measurements are illustrated in Fig. 2.
apneas and hypopneas per hour of sleep. All recordings started at At the end of evaluation a Rapid Maxillary Expander (RME) was
the patients’ usual bedtime and continued until spontaneous awak- applied. RME is a fixed device with an expansion screw and two
ening. All recordings were scored visually by one of the investigators, bands (Leone Sesto Fiorentino, Florence) cemented to the second
who was blinded to the subjects’ group, age, and sex. Residual OSA deciduous molars of the upper jaw because the first permanent
after treatment was defined as the presence of AHI >1/h [8]. molar in many subjects was not completely erupted. The screw was

Fig. 2. Cephalometric measurements. SNA, maxillary prognathism; SNB, mandibular prognathism; ANB, difference between SNA and SNB; NSBa, cranial base angle; ML-
NSL, mandibular inclination in relation to the anterior cranial base; NL-NSL, maxillary inclination in relation to the anterior cranial base; ML-NL, mandibular inclination in
relation to the maxillary inclination; NSAr, sella angle; SarGo, articular angle; ArGoMe, gonial angle; S Go/NMe, total facial index; NANS/ANSMe, anterior facial index; PNSAd1,
linear distance between PNS and the nearest point of adenoids along the PNSBa line; PNSBa, distance between PNS and Ba; PNSAd2, linear distance between PNS and the
nearest point of adenoids along the line passing in PNS and perpendicular to the SBa line; PNSSo, perpendicular from PNS to the SBa line in point So (point of intersection
of line from PNS perpendicular to SBa line); Ad2So, linear distance between So and the outmost point of adenoids along the PNSSo line; Ad1Ba, linear distance between Ba
and the outmost point of adenoids along the PNSBa line; PNSAd1/PNSBa, superior nasopharyngeal adenoids gradient; Ad2So/PNSSo, superior nasopharyngeal airway gra-
dient; PNSAd2/PNSSo, inferior nasopharyngeal adenoids gradient; Ad1Ba/PNSBa, inferior nasopharyngeal airway gradient.

Please cite this article in press as: Maria Pia Villa, et al., Rapid maxillary expansion outcomes in treatment of obstructive sleep apnea in children, Sleep Medicine (2015), doi: 10.1016/
j.sleep.2014.11.019
 ARTICLE IN PRESS
4 M.P. Villa et al./Sleep Medicine ■■ (2015) ■■–■■

turned two consecutive turns once a day for several days (10.9 ± 1.2 Table 1
days) until the palatal cusp of the upper molars came into contact Clinical, anthropometric and polysomnographic parameters in children at baseline
(T0) and 1 year after treatment (T1).
with the buccal cusp of the lower molars. Finally the screw was fixed
with a steel ligature wire and acrylic in order to keep the achieved Baseline (T0) One year after p
N = 40 treatment (T1)N = 40
maxillary expansion. The device was removed after 12 months.
Mean ± SD Mean ± SD
2.6. Statistical analysis Anthropometric parameters
Age (years) 6.3 ± 1.6 7.6 ± 1.5 <0.001
BMI (kg/m2) 17.7 ± 3.3 18.2 ± 3.7 <0.05
Paired t-test or Wilcoxon test and the non-parametric test ANOVA BMI centile 66.2 ± 26.3 71 ± 15 NS
(Friedman test) were used to compare two or more repeated mea- Questionnaire
surements. Unpaired t-test or Mann–Whitney and non-parametric Brouillette score −0.03 ± 1.8 −3.0 ± 1.1 <0.001
N (%) N (%)
ANOVA test (Kruskal–Wallis) with post hoc Bonferroni were used
Daytime symptoms
to compare two or more groups. Contingency tables (χ2 test) with Oral breathing 25.8% 25% NS
Fisher’s correction were used for comparison of proportions. Pearson Bruxism 27.6% 17.9% NS
correlations and multiple regression analysis were used to assess Daily sleepiness 48.3% 14.8% 0.004
the relation between the dependent variables and potentially ex- Halitosis 65.5% 39.3% 0.04
Night-time symptoms
planatory variables. Snoring 96.8% 17.9% 0.000
The variance of AHI (ΔAHI) before and after the orthodontic Apneas 80.6% 10.7% 0.000
therapy (AHI T1 − AHI T0/AHI T0 × 100) was used to define the impact Respiratory effort 74.2% 11% 0.000
of treatment. Differences and correlations were considered to achieve Night-time sweating 37.9% 32.1% NS
Restless sleep 63.3% 25% 0.002
statistical significance when the p-value was <0.05. A statistical soft-
Mean ± SD Mean ± SD
ware package (SPSS 13, Chicago, Ill) was used for calculations. Polysomnographic parameters
AHI (events/h) 4.7 ± 4.4 1.6 ± 1.4 <0.001
3. Results SpO24 96.8 ± 1.5 97.5 ± 1.8 <0.05
TST (min) 402.1 ± 50.3 433.4 ± 67.2 <0.05
SE (TST/time in bed) (%) 86.1 ± 9.7 86.5 ± 10.1 NS
The previous 12 months’ follow-up study protocol [17] was con- Arousal index (number of 16.3 ± 7.9 13. 2 ± 6.7 <0.05
tinued for the following four years. Thirty out of 64 eligible patients events/hour of sleep)
for the orthodontic treatment agreed to start the therapy in the orth- S1 (%) 5.9 ± 4.6 8.3 ± 14.1 NS
odontic department of our hospital. Four of them refused to repeat S2 (%) 43.4 ± 10.1 44.3 ± 11.9 NS
SWS (%) 32.3 ± 10 29.3 ± 9.1 NS
PSG. Twenty-six patients completed the follow up and this allowed
REM (%) 17.8 ± 6.2 20.2 ± 7.2 NS
us to increase the sample size up to 40 patients (mean age 6.3 ± 1.6
years, range 4.3–10.5; 23 boys). At baseline orthodontic evaluation AHI apnea–hypopnea index; BMI, body mass index; S1, sleep stage 1; S2, sleep stage
2; SE, sleep efficiency; SWS, slow-wave sleep; SpO2%, average overnight arterial oxygen
all children presented with narrow palate, associated in 29 cases (72.5%) saturation; TST, total sleep time.
to malocclusion: 8/29 subjects (20%) had crossbite, 13/29 subjects
(32.5%) deep, and 5/29 subjects (12.5%) open bite. A retrusive bite
was present in 10/29 subjects (25%). Moreover 10 patients (25%) had to note that the parameter NSBa (cranial base angle) is significantly
moderate tonsillar hypertrophy (grade III) (Table 1). A questionnaire higher in the non-responder group (p = 0.03). Despite a tendency for
showed that snoring and apneas were the most common symptoms difference (p = 0.06) the anterior total facial ratio SGo/NMe was not
reported. However most of the daytime and night-time symptoms smaller in the responder group. This result is confirmed by the higher
improved significantly after the treatment (Table 1). angle, in the vertical plane, of the parameter ML-NSL (mandibular
The AHI decreased significantly from T0 to T1 (4.7 ± 4.4 ev/h vs inclination in relation to theanteriorcranial base) even though not
1.6 ± 1.4 ev/h, p < 0.001) as well as the arousal index (16.3 ± 7.9 n/h
vs 13.2 ± 6.7 n/h, p < 0.05), whereas total sleep time (402.1 ± 50.3 min
vs 433.4 ± 67.2 min, p < 0.05) and mean overnight oxygen satura-
tion (96.8 ± 1.5% vs 97.5 ± 1.8%, p < 0.05) increased significantly. BMI
centile did not increase significantly from T0 to T1 (Table 1).
ΔAHI (mean 49.31% ± 54.39%) was significantly related to disease
duration (p < 0.05) (Fig. 3), and there was no difference according
to type of malocclusion and presence of severe tonsillar hypertro-
phy. Stepwise linear multiple-regression analysis identified disease
duration as the only variable that was significantly correlated with
ΔAHI (p = 0.006; r2 = 0.18; standardized β coefficient = −0.431). We
defined responders patients as children who showed a decrease of
AHI greater than 20% at T1 and as non-responders children who
showed a decreasing <20% of AHI at T1.
The present data showed that in 6/40 patients (15%) AHI did not
respond to the treatment (2.1 ± 1.3 ev/h vs 2.9 ± 1.3 ev/h, ΔAHI
−53.47% ± 61.57%). In 34/40 patients (85%) AHI decrease was greater
than 20% from T0 to T1 (5.2 ± 4.7 ev/h vs 1.4 ± 1.3 ev/h, ΔAHI
67.45% ± 25.73%) (Fig. 4).
No differences for gender, malocclusion, tonsillar hypertrophy, BMI
centile, and atopy emerged between the two groups. Disease duration
was significantly lower (2.5 ± 1.4 years vs 4.8 ± 1.9 years, p < 0.005)
and age at disease onset was higher in responder patients compared
to non-responders (3.8 ± 1.5 years vs 2.3 ± 1.9 years, p < 0.05) (Table 2).
By examining baseline values of cephalometric variables it is possible Fig. 3. Correlation between ΔAHI and disease duration. AHI, apnea–hypopnea index.

Please cite this article in press as: Maria Pia Villa, et al., Rapid maxillary expansion outcomes in treatment of obstructive sleep apnea in children, Sleep Medicine (2015), doi: 10.1016/
j.sleep.2014.11.019
 ARTICLE IN PRESS
M.P. Villa et al./Sleep Medicine ■■ (2015) ■■–■■ 5

Fig. 4. AHI changes from baseline (T0) to 1 year after treatment (T1) in group 1 and group 2. AHI, apnea–hypopnea index.

statistically significant in the same group. No other variables differed

between the two groups of patients (Table 3).
Table 2 Looking at the complete efficacy of treatment, defined as AHI < 1,
Clinical and polysomnographic parameters in responder and non-responder patients. 23/40 patients (57.5%) presented residual OSA since they had AHI > 1.
Responders Non-responders No significantly clinical anthropometric and cephalometric differ-
N = 34 N=6 ences were noted in children with residual OSA and children without
Mean ± SD Mean ± SD p (Table 4).
Age (years) 6.3 ± 1.4 7.1 ± 2.7 NS No patient was treated again with RME after the initial treatment.
Age at onset (years) 3.8 ± 1.5 2.3 ± 1.9 <0.05 Patients were followed-up by the orthodontists every year for five
Disease duration (years) 2.5 ± 1.4 4.8 ± 1.9 <0.005
years (they are still ongoing follow-up). Considering the second aim
N (%) N (%) p
Gender (males) 15 (44.1%) 2 (33.3%) NS of our study, Brouilette score was administered to 14 children (mean
Malocclusion 25 (73.5%) 4 (66.7%) NS age of these children was 17.0 ± 1.9 years, 10 males) who ended the
Tonsillar hypertrophy (grade III) 9 (26.4%) 1 (16.6%) NS orthodontic treatment 10 years previously, to evaluate the
Mean ± SD Mean ± SD p
recurrence of night-time and daytime symptoms, through the years.
AHI T0 5.2 ± 4.7 2.1 ± 1.3 <0.005
SpO2% T0 96.7 ± 1.6 97.3 ± 1.4 NS
The mean age of these children was 17.0 ± 1.9 years, 10 males
AHI T1 1.4 ± 1.3 2.9 ± 1.3 <0.05 (62.5%). Night-time and daytime symptoms were decreased
SpO2% T1 97.4 ± 1.9 98.0 ± 1.7 NS compared to T0 but they were not modified by the end of the
AHI, apnea–hypopnea index (events/h); BMI, body mass index; NS, not significant; treatment through the following 10 years. After 10 years (T2), the
SpO2%, average overnight arterial oxygen saturation. Brouilette score improved compared to baseline (Table 5).

Please cite this article in press as: Maria Pia Villa, et al., Rapid maxillary expansion outcomes in treatment of obstructive sleep apnea in children, Sleep Medicine (2015), doi: 10.1016/
j.sleep.2014.11.019
 ARTICLE IN PRESS
6 M.P. Villa et al./Sleep Medicine ■■ (2015) ■■–■■

Table 3 Table 4
Cephalometric variables in responder and non-responder patients. Cephalometric variables in children with residual OSA (AHI >1 ev/h) and in chil-
dren with complete resolution of OSA (AHI <1 ev/h) after treatment.
Responders Non-responders
N = 34 N=6 Residual OSAS No residual OSAS
N = 23 N = 17
Mean ± SD Mean ± SD p
Skeletal Mean ± SD Mean ± SD p
SNA (°) 79.61 ± 3.32 77.30 ± 2.90 NS Skeletal
SNB (°) 75.10 ± 3.40 73.76 ± 1.73 NS SNA (°) 79.31 ± 2.52 79.19 ± 4.06 NS
ANB (°) 4.33 ± 1.80 3.30 ± 1.62 NS SNB (°) 74.33 ± 2.84 75.46 ± 3.56 NS
NSBa (°) 132.05 ± 4.85 138.01 ± 6.49 0.03 ANB (°) 4.80 ± 2.01 3.41 ± 1.09 NS
ML-NSL (°) 37.96 ± 5.14 39.66 ± 3.93 NS NSBa (°) 132.42 ± 4.39 133.81 ± 6.58 NS
NL-NSL (°) 6.57 ± 2.87 9.05 ± 4.16 NS ML-NSL (°) 38.65 ± 5.33 37.95 ± 4.56 NS
ML-NL (°) 30.69 ± 5.42 31.11 ± 3.25 NS NL-NSL (°) 6.36 ± 2.67 7.65 ± 3.63 NS
NSAr (°) 123.36 ± 4.98 127.08 ± 4.60 NS ML-NL (°) 31.33 ± 5.35 30.18 ± 4.71 NS
SarGo (°) 147.44 ± 6.75 144.98 ± 4.70 NS NSAr (°) 122.44 ± 4.39 125.93 ± 5.27 NS
ArGoMe (°) 129.52 ± 6.67 129.73 ± 3.38 NS SArGo (°) 148.60 ± 7.13 145.15 ± 5.15 NS
SOMMAT (°) 400.30 ± 5.16 401.80 ± 4.34 NS ArGoMe (°) 129.26 ± 7.02 129.88 ± 5.14 NS
S Go/Nme ratio (%) 61.63 ± 3.49 59.33 ± 1.96 0.06 SOMMAT (°) 400.38 ± 5.33 400.86 ± 4.73 NS
NANS/ANSMe ratio (%) 76.31 ± 4.89 80.33 ± 5.24 NS S Go/NMe (ratio%) 61.16 ± 3.78 61.14 ± 2.93 NS
Nasopharynx NANS/ANSMe (ratio%) 76.26 ± 5.24 78.23 ± 5.03 NS
PNSAd1/PNSBa (mm) 31.42 ± 22.36 32.80 ± 9.73 NS Nasopharynx
Ad2So/PNSSo (mm) 73.47 ± 7.13 70.80 ± 8.28 NS PNSAd1/PNSBa (mm) 32.76 ± 28.23 30.61 ± 8.15 NS
Ad1Ba/PNSBa (mm) 25.57 ± 7.19 28.60 ± 8.26 NS Ad2So/PNSSo (mm) 75.15 ± 7.31 70.76 ± 6.79 NS
PNSAd2/PNSSo (mm) 71.61 ± 11.05 69.75 ± 7.54 NS Ad1Ba/PNSBa (mm) 23.92 ± 7.43 28.38 ± 6.78 NS
PNSAd2/PNSSo (mm) 72.76 ± 12.96 69.75 ± 7.07 NS
SNA, maxillary prognathism; SNB, mandibular prognathism; ANB, difference between
SNA and SNB; NSBa, cranial base angle; ML-NSL, mandibular inclination in relation SNA, maxillary prognathism; SNB, mandibular prognathism; ANB, difference between
to the anterior cranial base; NL-NSL, maxillary inclination in relation to the anterior SNA and SNB; NSBa, cranial base angle; ML-NSL, mandibular inclination in relation to
cranial base; ML-NL, mandibular inclination in relation to the maxillary inclination; the anterior cranial base; NL-NSL, maxillary inclination in relation to the anterior cranial
NSAr, sella angle; SarGo, articular angle; ArGoMe, gonial angle; S Go/NMe, total facial base; ML-NL, mandibular inclination in relation to the maxillary inclination; NSAr, sella
index; NANS/ANSMe, anterior facial index; PNSAd1, linear distance between PNS and angle; SarGo, articular angle; ArGoMe, gonial angle; S Go/NMe, total facial index; NANS/
the nearest point of adenoids along the PNSBa line; PNSBa, distance between PNS ANSMe, anterior facial index; PNSAd1, linear distance between PNS and the nearest
and Ba; PNSAd2, linear distance between PNS and the nearest point of adenoids along point of adenoids along the PNSBa line; PNSBa, distance between PNS and Ba; PNSAd2,
the line passing in PNS and perpendicular to the SBa line; PNSSo, perpendicular from linear distance between PNS and the nearest point of adenoids along the line passing
PNS to the SBa line in point So (point of intersection of line from PNS perpendicular in PNS and perpendicular to the SBa line; PNSSo, perpendicular from PNS to the SBa
to SBa line); Ad2So, linear distance between So and the outmost point of adenoids line in point So (point of intersection of line from PNS perpendicular to SBa line); Ad2So,
along the PNSSo line; Ad1Ba, linear distance between Ba and the outmost point of linear distance between So and the outmost point of adenoids along the PNSSo line;
adenoids along the PNSBa line; PNSAd1/PNSBa, superior nasopharyngeal adenoids Ad1Ba, linear distance between Ba and the outmost point of adenoids along the PNSBa
gradient; Ad2So/PNSSo, superior nasopharyngeal airway gradient; PNSAd2/PNSSo, line; PNSAd1/PNSBa, superior nasopharyngeal adenoids gradient; Ad2So/PNSSo, su-
inferior nasopharyngeal adenoids gradient; Ad1Ba/PNSBa, inferior nasopharyngeal perior nasopharyngeal airway gradient; PNSAd2/PNSSo, inferior nasopharyngeal adenoids
airway gradient. gradient; Ad1Ba/PNSBa, inferior nasopharyngeal airway gradient.

4. Discussion needed in addition to standard therapy as reported by several authors

[32–35]. Several considerations could have been studied in the non-
Our data support the usefulness and the efficacy of orthodontic responder patients. Our interest was to evaluate whether patients
treatment in children with OSA and support previous findings [13–18]. who do not benefit at all from treatment had specific craniofacial
A larger sample size highlights that the best results are achieved characteristics. Looking at baseline cephalometric evaluation, it is
when an early orthodontic/orthopedic treatment is undertaken in possible to note that the parameter NSBa (cranial base angle) is sig-
late deciduous dentition and early mixed dentition. Moreover it is nificantly higher in the non-responder group (p = 0.03), indicating
known that early recognition and early treatment of OSA could that non-responder patients showed a more retrognathic facial type.
prevent its complications such as neurocognitive and behavioral dis- Moreover the vertical parameter, SGo/NMe, was lower in the non-
turbances, systemic inflammation, and cardiovascular and metabolic responder group indicating an increase in the anterior total face
dysfunction [26–31]. height, even if not statistically significant (p = 0.06) (Table 3).
Although sleep respiratory parameters improved after 12 months According to Jarabat classification [36], facial height ratio (FHR)
of treatment, our results highlighted that oral breathing is still re- or Jarabat quotient is 59–63% in the neutral growth pattern. If we
ported suggesting that a precocious re-education intervention is compare the non-responder group with the responder group they

Table 5
Anthropometric parameters and questionnaire results of treated subjects at baseline (T0), 1 year after treatment (T1) and 10 years after the end of treatment (T2).

Baseline (T0) 1 year after treatment (T1) 10 years after the end of 1 vs 2 vs 3 p
N = 14 N = 14 treatment (T2)N = 14
1 vs 2 1 vs 3 2 vs 3

Mean ± SD Mean ± SD Mean ± SD

Clinical and anthropometric parameters
Age (years) 5.9 ± 1.7 7.3 ± 1.7 17.0 ± 1.9
Percentile of BMI 77.4 ± 34 82.5 ± 11.9 69.4 ± 22.6 NS
Brouillette score 0.8 ± 1.4 −2.9 ± 1.2 −2.4 ± 1.1 <0.005 <0.005 <0.005 NS
Clinical symptoms N (%) N (%) N (%)
Snoring 14 (100%) 6 (42.8%) 10 (71.4%) <0.005 <0.005 NS NS
Oral breathing 13 (92.8%) 10 (71.4%) 8 (57.1%) NS NS NS NS
Bruxism 5 (35.7%) 4 (28.5%) 3 (21.4%) NS NS NS NS
Daily sleepiness 10 (71.4%) 3 (21.4%) 1 (7.1%) <0.005 <0.005 <0.005 NS
Halitosis 9 (64.3%) 7 (50%) 11 (78.6%) NS NS NS NS

Please cite this article in press as: Maria Pia Villa, et al., Rapid maxillary expansion outcomes in treatment of obstructive sleep apnea in children, Sleep Medicine (2015), doi: 10.1016/
j.sleep.2014.11.019
 ARTICLE IN PRESS
M.P. Villa et al./Sleep Medicine ■■ (2015) ■■–■■ 7

show a lower value and they are closer to the hyperdivergent growth oral breathing, both of which are involved in airway muscle func-
pattern (FHR < 59%). This result is confirmed by the greater tion and upper airway patency in order to ameliorate symptoms after
amplitude in the vertical plane of the angle ML-NSL even though 10 years. As suggested by data from the literature, best results are
not statistically significant. It is well documented, indeed, that achieved with a combination of oropharyngeal exercises and other
mouth-breathing, presented in children with OSA, induced the therapeutic options [32–35,44]. It is therefore clear that rehabili-
development of longer faces [37–41]. The switch from a nasal to tation should follow or integrate treatments for sleep disorders to
an oronasal (mouth and nose combined) breathing pattern, in fact, prevent future complications.
induces functional adaptations that include an increase in total This study presents several limitations such as the lack of
anterior face height and vertical development of the lower anterior polysomonographic data in the 10 years follow-up group. Since they
face. Moreover, it is important to bear in mind that in this group, refused to undergo a new PSG, only the questionnaire could be ad-
disease duration was significantly higher and age at disease onset ministered. Moreover the age of onset of symptoms was parentally
was lower compared to responders, suggesting that the earlier the reported and this could have also influenced the assessment of
oral breathing occurs, the more serious skeletal alterations typical disease duration.
of snoring and OSA become, the more difficult it will be to achieve We administered Brouillette questionnaire to evaluate the re-
OSA resolution. This is the reason why the patients should be treated currence of symptoms in adolescent patients 10 years after the end
as early as symptoms appear in order to interrupt the vicious circle. of orthodontic treatment in order to compare it with the question-
Maybe a new treatment with RME or with a different orthodontic naire previously administered, although Brouillette questionnaire
device should be performed. was originally validated for children until the age of 10 years. On
Although it is well known that a successful outcome is achieved the other hand the questions of Brouilette questionnaire used for
in both AT and orthodontic treatment, our goal was to highlight the the score are about snoring, apneas and restless sleep and there should
role of RME in OSA children with malocclusion and without sig- be no difference in asking the questions of a child or of a teenager.
nificant tonsillar hypertrophy. Although only 42.5% of patients Further studies with larger samples are needed to assess the pre-
achieved an AHI < 1 as determined by the post-RME PSG, most chil- dictive role of cephalometric analysis in the choice of treatment and
dren had evidence of residual OSA after RME treatment. It is its efficacy.
important to underline that residual OSA was present also in re-
sponder children (AHI T1 2.44 ± 1.2 ev/h ΔAHI 48.59% ± 20.78%). The 5. Conclusions
occurrence of residual OSA is supported by different studies un-
derlying that a single treatment is often not enough to obtain a In conclusion, our results confirm our previous findings and data
complete resolution of the disease even if a significant ΔAHI after by other authors. Starting orthodontic treatment as early as symp-
treatment is present and that a multidisciplinary approach could toms appears to be, once again, an important message to transmit
be successful [8,14,42–46]. Moreover since the AHI < 1 may not be in order to increase the efficacy of treatment. Non-responder pa-
appropriate and an agreement of AHI cut-off value for residual OSAS tients, in fact, have a longer duration of disease and a lower age of
is currently unavailable, cut-off value could influence the evalua- onset. Moreover, it could be suggested that, in the presence of some
tion of treatment efficacy. It is easy to understand that considering structural characteristics, the orthodontic treatment should be per-
a higher cut-off to define residual OSAS, the percentage of pa- sonalized based on the patient’s phenotype. An early myofunctional
tients who showed a resolution of the disease consequently increases. rehabilitation may be useful.
There are very important open questions on the definition of re-
sidual OSA that will clearly be the focus of further studies in the
next years. Conflict of interest
The ideal treatment should be able to eradicate the disease, as
suggested by the normalization of AHI independently from the de- The authors confirm that there are no known conflicts of inter-
creasing rate of it, so we evaluated the cephalometric parameters est associated with this publication and there has been no significant
in order to understand whether there were structural issues that financial support that could have influenced its outcome.
can predict the treatment efficacy or characteristics that might The ICMJE Uniform Disclosure Form for Potential Conflicts of In-
suggest alternative or integrated orthodontic treatments. terest associated with this article can be viewed by clicking on the
No differences in cephalometric measurements were present following link: http://dx.doi.org/10.1016/j.sleep.2014.11.019.
comparing the group of patients with residual OSA with the pa-
tients who achieved complete resolution. However it is possible to References
note that the patients with residual OSA showed a higher value of
the cephalometric parameter ANB (difference between maxillary [1] Marcus CL, Brook LJ, Davidson Ward S, et al. Diagnosis and management of
prognathism and mandibular prognathism) angle, even if it was not childhood obstructive sleep apnea syndrome. Pediatrics 2012;130:e714.
[2] Katz ES, D’Ambrosio CM. Pathophysiology of pediatric obstructive sleep apnea.
statistically significant. Proc Am Thorac Soc 2008;5(2):253–62.
Guilleminault et al. showed that recurrence of symptoms during [3] Alabi BS, Abdulkarim AA, Musa IO, et al. Prevalence of snoring and symptoms
teenage years in a group of adolescents with OSA, previously treated of sleep disordered breathing among primary school pupils in Ilorin, Nigeria.
Int J Pediatr Otorhinolaryngol 2012;76(5):646–8.
with AT and orthodontia at a mean age of 7.5 years, may be due to [4] Guilleminault C, Li KK, Khramstov A, et al. Sleep disordered breathing: surgical
hormonal status changes [9]. outcomes in prepubertal children. Laryngoscope 2004;14(1):132–7.
Our data on follow-up evaluation of 14 patients after 10 years [5] Zonato AI, Martinho FL, Bittencourt LR, et al. Head and neck physical
examination: comparison between nonapneic and obstructive sleep apnea
showed that orthodontic treatment has long-lasting benefits on
patients. Laryngoscope 2005;115(6):1030–4.
symptoms and prevents, after the onset of puberty, the recur- [6] Marcus CL. Sleep-disordered breathing in children. Am J Respir Crit Care Med
rence of baseline conditions (Table 5). It is important to underline 2001;164:16–30.
[7] Arens R, McDonough JM, Costarino AT, et al. Magnetic resonance imaging of
that symptoms such as snoring and oral breathing persist after the
the upper airway structure of children with obstructive sleep apnea syndrome.
onset of puberty or are present again after 10 years from the end Am J Respir Crit Care Med 2001;164(4):698–703.
of treatment showing that RME is not the definitive treatment and [8] Bhattacharjee R, Kheirandish-Gozal L, Spruyt K, et al. Adenotonsillectomy
that a closer follow-up is needed. outcomes in treatment of obstructive sleep apnea in children: a multicenter
retrospective study. Am J Respir Crit Care Med 2010;182(5):676–8.
Once again we could suppose that myofunctional re-education [9] Guilleminault C, Huang YS, Quo S, et al. Teenage sleep-disordered breathing:
could help subjects to acquire physiological breathing and to correct recurrence of syndrome. Sleep Med 2013;14(9):927–8.

Please cite this article in press as: Maria Pia Villa, et al., Rapid maxillary expansion outcomes in treatment of obstructive sleep apnea in children, Sleep Medicine (2015), doi: 10.1016/
j.sleep.2014.11.019
 ARTICLE IN PRESS
8 M.P. Villa et al./Sleep Medicine ■■ (2015) ■■–■■

[10] Guilleminault C, Li KK, Khramtsov A, et al. Sleep disordered breathing: surgical [28] James AL, Runciman M, Burton MJ, et al. Investigation of cardiac function in chil-
outcomes in prepubertal children. Laryngoscope 2004;114:132–7. dren with suspected obstructive sleep apnea. J Otolaryngol 2003;32(3):151–4.
[11] Tasker C, Crosby JH, Stradling JR. Evidence for persistence of upper airway [29] Gozal D, Capdevila OS, Kheirandish-Gozal L. Metabolic alterations and systemic
narrowing during sleep, 12 years after adenotonsillectomy. Arch Dis Child inflammation in obstructive sleep apnea among nonobese and obese
2002;86:34–7. prepubertal children. Am J Respir Crit Care Med 2008;177(10):1142–9.
[12] Souki BQ, Pimenta GB, Souki MQ, et al. Prevalence of malocclusion among [30] Chervin RD, Dillon JE, Bassetti C, et al. Symptoms of sleep disorders, inattention,
mouth breathing children: do expectations meet reality? Int J Pediatr and hyperactivity in children. Sleep 1997;20(12):1185–92.
Otorhinolaryngol 2009;73:767–73. [31] Beebe DW, Gozal D. Obstructive sleep apnea and the prefrontal cortex: towards
[13] Villa MP, Bernkopf E, Pagani J, et al. Randomized controlled study of an oral a comprehensive model linking nocturnal upper airway obstruction to daytime
jaw positioning appliance for the treatment of obstructive sleep apnea in cognitive and behavioral deficits. J Sleep Res 2002;11(1):1–16.
children with malocclusion. Am J Respir Crit Care Med 2002;165:123–7. [32] Villa MP, Brasili L, Ferretti A, et al. Oropharyngeal exercises to reduce symptoms
[14] Villa MP, Castaldo R, Miano S, et al. Adenotonsillectomy and orthodontic therapy of OSA after AT. Sleep Breath 2015;19:281–9.
in pediatric obstructive sleep apnea. Sleep Breath 2014;18:533–9. [33] Guimarães KC, Drager LF, Genta PR, et al. Effects of oropharyngeal exercises on
[15] Villa MP, Miano S, Rizzoli A. Mandibular advancement devices are an alternative patients with moderate obstructive sleep apnea syndrome. Am J Respir Crit Care
and valid treatment for pediatric obstructive sleep apnea syndrome. Sleep Breath Med 2009;179(10):962–6.
2012;16(4):971–6. [34] Guilleminault C, Huang YS, Monteyrol PJ, et al. Critical role of myofascial re-
[16] Pirelli P, Saponara M, Guilleminault C. Rapid maxillary expansion in children education in pediatric sleep-disordered breathing. Sleep Med 2013;14(6):518–
with obstructive sleep apnea syndrome. Sleep 2004;27(4):761–6. 25.
[17] Villa MP, Malagola C, Pagani J, et al. Rapid maxillary expansion in children with [35] Moeller J, Paskay LC, Gelb ML. Myofunctional therapy: a novel treatment of
obstructive sleep apnea syndrome: 12-month follow-up. Sleep Med pediatric sleep-disordered breathing. Sleep Med Clin 2014;9(2):235–43.
2007;8:128–34. [36] Siriwat PP, Jarabat J. Malocclusion and facial morphology. Is there a relationship?
[18] Villa MP, Rizzoli A, Miano S, Malagola C. Efficacy of rapid maxillary expansion Angle Orthod 1985;55(2):127–36.
in children with obstructive sleep apnea syndrome: 36 months of follow-up. [37] Flores-Mir C, Korayem M, Heo G, et al. Craniofacial morphological characteristics
Sleep Breath 2011;15:179–84. in children with obstructive sleep apnea syndrome: a systematic review and
[19] Friedman M, Ibrahim H, Joseph NJ. Staging of obstructive sleep apnea/hypopnea meta-analysis. J Am Dent Assoc 2013;144(3):269–77, Review.
syndrome: a guide to appropriate treatment. Laryngoscope 2004;114(3):454–9. [38] Vieira BB, Itikawa CE, de Almeida LA, et al. Cephalometric evaluation of facial
[20] Cole TJ, Bellizzi MC, Flegal KM, et al. Establishing a standard definition for pattern and hyoid bone position in children with obstructive sleep apnea
child overweight and obesity worldwide: international survey. BMJ 2000;320 syndrome. Int J Pediatr Otorhinolaryngol 2011;75(3):383–6.
(7244):1240–3. [39] Arens R, Marcus CL. Pathophysiology of upper airway obstruction: a
[21] Brouillette R, Hanson D, David R, et al. A diagnostic approach to suspected developmental perspective. Sleep 2004;27(5):997–1019.
obstructive sleep apnea in children. J Pediatr 1984;105:10–14. [40] Marino A, Ranieri R, Chiarotti F, et al. Rapid maxillary expansion in children
[22] Iber C, Ancoli-Israel S, Chesson A, et al. for the American Academy of Sleep with Obstructive Sleep Apnoea Syndrome (OSA). Eur J Paediatr Dent
Medicine. The AASM manual for the scoring of sleep and associated events: 2012;13(1):57–63.
rules, terminology and technical specifications. 1st ed. Westchester, IL: American [41] Souki BQ, Lopes PB, Pereira TBJ, et al. Mouth breathing children and
Academy of Sleep Medicine; 2007. cephalometric pattern: does the stage of dental development matter? Int J
[23] Berry RB, Budhiraja R, Gottlieb DJ, et al. for the American Academy of Sleep Pediatr Otorhinolaryngol 2012;76:837–41.
Medicine. Rules for scoring respiratory events in sleep: update of the 2007 AASM [42] Tauman R, Gulliver TE, Krishna J, et al. Persistence of obstructive sleep apnea
Manual for the Scoring of Sleep and Associated Events. Deliberations of the syndrome in children after adenotonsillectomy. J Pediatr 2006;149(6):803–8.
Sleep Apnea Definitions Task Force of the American Academy of Sleep Medicine. [43] Guilleminault C, Monteyrol PJ, Huynh NT, et al. Adeno-tonsillectomy and rapid
J Clin Sleep Med 2012;8(5):597–619. maxillary distraction in pre-pubertal children, a pilot study. Sleep Breath
[24] Angle EH. Treatment of malocclusion of the teeth: Angle’s system. Philadelphia: 2011;15(2):173–7.
SS White Dental Manufacturing Co; 1907. [44] Huang YS, Guilleminault C. Pediatric obstructive sleep apnea and the critical
[25] Houston WJ. The analysis of errors in orthodontic measurements. Am J Orthod role of oral-facial growth: evidences. Front Neurol 2013;3:1–7.
1983;83(5):382–90. [45] Guilleminault C, Partinen M, Praud JP, et al. Morphometric facial changes and
[26] Bhattacharjee R, Kim J, Kheirandish-Gozal L, et al. Obesity and obstructive sleep obstructive sleep apnea in adolescents. J Pediatr 1989;114:997–9.
apnea syndrome in children: a tale of inflammatory cascades. Pediatr Pulmonol [46] Tasker C, Crosby JH, Stradling JR, et al. Evidence for persistence of upper airway
2011;46(4):313–23. narrowing during sleep, to 12 years after adenotonsillectomy. Arch Dis Child
[27] Gozal D, Kheirandish-Gozal L, Bhattacharjee R, et al. C-reactive protein and 2002;86:34–7.
obstructive sleep apnea syndrome in children. Front Biosci (Elite Ed)
2012;4:2410–22.

Please cite this article in press as: Maria Pia Villa, et al., Rapid maxillary expansion outcomes in treatment of obstructive sleep apnea in children, Sleep Medicine (2015), doi: 10.1016/
j.sleep.2014.11.019