Subjective efficacy of oral appliance design features in the management of obstructive sleep apnea: A systematic review

Introduction

The purpose of this study was to review available evidence on the efficacy of various oral appliances on subjectively perceived symptoms of obstructive sleep apnea syndrome.

Methods

A search of 4 databases was carried out. Articles were initially selected based on their titles or abstracts. Full articles were then retrieved and further scrutinized according to predetermined criteria. Reference lists of selected articles were searched for any missed publications. The finally selected articles were methodologically evaluated.

Results

Of an initial 1475 references, 14 studies were randomized controlled trials, which formed the basis of this review. Mandibular advancement devices (MADs) were compared with either inactive appliances (6 studies) or MADs with different design features (8 studies). In comparison with inactive appliances, the majority of studies showed improved subjective outcomes with MADs, suggesting that mandibular advancement is a crucial design feature of oral appliance therapy for obstructive sleep apnea syndrome.

Conclusions

There is no 1 MAD design that most effectively influences subjectively perceived treatment efficacy, but efficacy depends on many factors including materials and method used for fabrication, type of MAD (monoblock or Twin-block), and the degree of protrusion (sagittal and vertical). This review highlights the absence of universally agreed subjective assessment tools and health-related quality of life outcomes in the literature today. Future trials of MAD designs need to assess subjective efficacy with agreed standardized tools and health-related quality of life measures to guide clinical practicitioners about which design might be most effective in the treatment of obstructive sleep apnea syndrome with oral appliances.

Obstructive sleep apnea syndrome (OSAS) is the most common organic sleep disorder and is increasingly recognized as a serious public health issue. Its prevalence worldwide is estimated at 3% to 7% in men and 2% to 5% in women. Apart from having serious consequences for a patient’s physiologic health, untreated OSAS causes particular morbidity in terms of the health-related quality of life (HRQL) a patient experiences. The inability to sleep sufficiently at night causes excessive daytime somnolence, which impacts the ability to function at an optimum level, can lead to depression, contributes to traffic accidents, and disrupts social relationships.

The latest Cochrane review on oral appliances (OAs) for OSAS highlighted that they are increasingly recognized as a suitable and effective treatment option. Mandibular advancement devices (MADs) are the most commonly prescribed OAs in the treatment of OSAS. Although MADs are more effective than other types of OAs in treating OSAS, it has been suggested that design features of the various appliances can have an impact on treatment efficacy. The Cochrane review detailed some evidence that MADs improve subjective sleepiness compared with placebo appliances. A review by Ahrens et al evaluated the efficacy of various MAD designs on patients’ objective polysomnographic outcomes; however, it is unclear how the different design features of the various MADs impact patients’ subjective evaluations of treatment effect. Understanding whether a type or design of MAD is most effective in the subjective treatment of OSAS is imperative in informing patient-centered evidence-based practice. Therefore, in this review, we aimed to summarize the evidence on the efficacy of differently designed MADs on the subjective patient-centered outcome measures of OSAS.

Material and methods

To identify studies relevant to OA treatment for OSAS, a computerized database search was carried out by using MEDLINE, EMBASE, Cinahl, and the Cochrane Library ( Table I ). No language limitations were set, and the search was limited to human studies. If articles contained the search thesaurus anywhere, they were selected to constitute a list of potentially eligible studies to be included in this review.

Table I
Electronic search strategy
Obstructive sleep apnoea syndrome part
1 Obstructive sleep apnea (MeSH word)
2 Obstructive sleep (apnoea or apnea)
3 Sleep (breathing disorder ∗ or respiratory disorder ∗ )
4 1 OR 2 OR 3
Oral appliance part
5 Orthodontic appliances (MeSH word)
6 Oral (device ∗ or appliance ∗ or splint)
7 Dental (device ∗ or appliance ∗ or splint)
8 Orthodontic ∗ (device ∗ or appliance ∗ or splint)
9 Mandib ∗ advancement ∗
Final search syntax
10 4 AND (5 or 6 or 7 or 8 or 9)

MeSH , thesaurus word.

∗ Truncation of a text word.

Titles and abstracts of study references on this list were reviewed by 2 independent researchers, who then determined whether they were relevant to the theme of this review: studies exclusively focusing on OA therapy for OSAS treatment ( Fig ). If the researchers disagreed about which articles were relevant, consensus was reached by discussion. To select articles that lent themselves to assessing the impact of appliance design on subjective treatment efficacy, 3 inclusion criteria were set. Only studies that investigated MAD vs other OA, MAD vs inactive OA, or the same MAD but with varying degrees of mandibular advancement or vertical bite opening were selected to remain on the list of potential studies for this review. The full texts of these studies were then obtained, and the reference lists were searched manually for additional relevant publications (reference linkage). All studies were methodologically appraised according to the American Association of Sleep Medicine’s levels of evidence ( Table II ) to identify “effective” articles.

Fig
Flow diagram of study selection procedure. OA , Oral appliance; OSAS , obstructive sleep apnea syndrome; CPAP , continuous positive airway pressure.

Table II
American Association of Sleep Medicine classification of evidence
Evidence level Study design
I Randomized well-designed trials with low alpha and beta error ∗
II Randomized trials with high alpha and beta error ∗
III Nonrandomized concurrently controlled studies
IV Nonrandomized historically controlled studies
V Case series

Adapted from Sackett.

∗ Alpha (type I error) refers to the probability that the null hypothesis is rejected when in fact it is true (generally acceptable at 5% or less, or P <0.05). Beta (type II error) refers to the probability that the null hypothesis is mistakenly accepted when in fact it is false (generally trials accept a beta error of 0.20). The estimation of type II error is generally the result of a power analysis. The power analysis takes into account the variability and the effect size to determine whether the sample size is adequate to find a difference in means when it is present (power is generally acceptable at 80%-90%).

Results

Initially, 1475 references ( Fig ) were retrieved from the primary database searches; among them, there were 470 duplicate references. An additional 467 references were excluded because they were not relevant for this review. Of the remaining 538 study references, a further 341 were excluded because they did not meet the criteria for inclusion ( Fig ). Full texts of the remaining 197 references were obtained, and an additional 3 articles were identified as potentially relevant by reference linkage. Among these 200 studies, 6 could be categorized as evidence level I, and 8 studies as level II of evidence. Fifty-nine studies reached evidence level III; 3 studies, level IV; and 124 studies, level V. Based on this classification of evidence, the 14 levels I and II randomized controlled trials were finally selected as the basis of this review. These studies were grouped according to the following outcome measures: (1) MAD vs inactive control OA, (2) studies comparing 1-piece MADs with 1-piece MADs, (3) studies comparing 2-piece MADs with 2-piece MADs, and (4) studies comparing 1-piece MADs with 2-piece MADs.

To assess patients’ subjective daytime sleepiness, all but 2 studies used the standardized and disease-specific Epworth sleepiness scale (ESS). To pool the data, 2 studies had to be excluded because only median values were reported. The ESS scores across the remaining 12 studies improved with the use of MADs and showed a mean reduction from 12.0 to 7.9. Gauthier et al used a fatigue severity scale, and Bloch et al used the sleep disorder questionnaire and a modified version of a sleep symptom questionnaire.

Four trials assessed patients’ HRQL by standardized tools. These were either generic—the medical outcome survey short form (SF-36)—or sleep-specific—the functional outcomes of sleep questionnaire (FOSQ). One study used both assessment tools.

Subjective treatment efficacy was assessed by nonstandardized (in-house) questionnaires or a visual analog scale (VAS) by 6 studies. Treatment compliance, treatment satisfaction, appliance preference, side effects, snoring, and sleep quality were generally self-reported by questionnaire. Across all studies, symptoms were reported to be mild to moderate and temporary, with temporomandibular joint pain, muscular, dental and gingival discomfort, dry mouth, and excessive salivation as the most common. Treatment compliance was generally high (dropout rates, 0%-26%). Average usage of individual appliances could not be calculated because this was reported differently across the reviewed studies, with some reporting nightly usage per week and others hourly usage per night. The studies’ durations ranged from 2 weeks to 12 months, and the sample sizes of the target study populations varied considerably from 16 to 93 subjects, but most studies specified a sample size between 20 and 30 ( Tables III and VI ).

Table III
Studies comparing MADs with inactive oral appliances: patient-centered and subjective outcomes
Patients (n) Oral appliance Advancement Assessment tool Result Statistical Subjective treatment satisfaction
Author Baseline Complete (OA) Sagittal Vertical used Pretreatment Posttreatment significance
Blanco et al 24 15 A: MAD
(1-piece soft elastic silicone positioner) 		75% of maximum
protrusion 		5 mm ESS MAD 14.7 ± 5.1 5.1 ± 1.9 P <0.05 Not reported
Inactive OA 16.3 ± 2.5 13.6 ± 6.7 NS
B: inactive OA – – SF-36
Physical MAD 70.7 (16.4) 74.1 (18.4) NS
functioning Inactive OA 71.5 (20.7) 78.8 (19.1) NS
Role MAD 83.4 (30.2) 87.5 (30.6) NS
physiological Inactive OA 81.2 (34.7) 87.5 (35.6) NS
Role emotional MAD 81.0 (37.7) 77.7 (46.6) NS
Inactive OA 80.0 (29.9) 87.5 (12.5) NS
Social MAD: 78.3 (13.6) 78.2 (12.4) NS
functioning Inactive OA 81.3 (18.8) 79.4 (26.9) NS
Mental health MAD 60.1 (19.3) 59.4 (19.2) NS
Inactive OA 52.0 (15.7) 56.0 (18.0) NS
Energy MAD 49.3 (18.8) 50.7 (8.4) NS
Inactive OA 55.2 (12.2) 56.2 (19.2) NS
Bodily pain MAD 70.3 (38.7) 67.0 (21.3) NS
Inactive OA 65.3 (37.4) 65.5 (19.2) NS
General health MAD 60.7 (22.0) 61.0 (20.7) NS
perception Inactive OA 57.4 (6.8) 58.4 (10.5) NS
FOSQ MAD 78.1 (22.6) 99.3 (14.4) P <0.05
Inactive OA 83.7 (20.8) 82.3 (13.9) NS
General sleep MAD 15.4 ±1.9 10.1 ± 3.2 P <0.05
questionnaire: Inactive OA 14.4 ± 3.0 14.6 ± 1.7 NS
snoring levels
Gotsopoulos
et al 		73 73 A: MAD
(custom-made 2-piece) 		Mean 80% ± 9%
(range, 50%-95%)
of maximum protrusion mean 7 ± 2 mm (range, 3-13 mm) 		3-4 mm ESS MAD 11 ± 5 7 ± 1 MAD vs inactive: MAD produced normal ESS in 60 (82%) patients compared with 45 (62%) on inactive treatment ( P <0.01)
Inactive OA 9 ± 1 P <0.001
B: inactive OA – – Symptom questionnaire: Snoring MAD Not reported 207 ± 20 MAD vs inactive: Treatment with MAD was significantly more satisfactory ( P <0.001)
frequency Inactive OA Not reported 366 ± 21 P <0.001
Snoring loudness MAD Not reported 48 ± 1 MAD vs inactive:
Inactive OA Not reported 51 ± 1 P <0.001
Improved sleep quality MAD Not reported Not reported MAD vs inactive:
Inactive OA Not reported Not reported P <0.001
Hans et al 24 18 A: MAD
(commercial thermoplastic 1-piece SnoreGuard) 		6-8 mm 8 mm ESS MAD 12.0 ± 3.9 8.2 ± 4.0 P <0.033 Not reported
B: inactive OA – 1 mm Inactive OA 13.0 ± 4.5 12.5 ± 5.7 NS
MAD vs inactive: NS
Johnston et al 21 20 A: MAD (customized 1-piece) 75% of maximum
protrusion mean, 5.7 mm (range, 4-9 mm) 		4 mm inter-incisal ESS MAD 13.90 ± 6.39 11.6 ± 6.7 MAD vs inactive: MAD produced posttreatment ESS of ≤10/h in 45% of patients; however, 60% of those showed ESS of ≤10/h at baseline
Inactive OA 12.6 ± 6.3 NS
B: inactive OA – 1.5 mm Sleep questionnaire MAD Not reported 2.58 ± 1.26 MAD vs inactive:
Inactive OA Not reported 3.16 ± 1.38 NS
Mehta et al 28 24 A: MAD (custom-made 2-piece) Not reported ESS MAD 10.1 ± 1.1 3.9 ± 0.6 Posttreatment vs pretreatment: Majority of patients reported substantial improvements in snoring (n =17, 70%) and sleep quality (n = 22, 91%)
Mean, 78% (range, 63%-89%) maximum protrusion mean, 7.5 ± 1.8 mm (range, 5-11.5 mm) P <0.01
Inactive OA Not reported
B: inactive OA – – Questionnaire: Snoring, sleep quality MAD Not reported Not reported Not reported
Inactive OA Not reported Not reported Not reported
Petri et al 93 81 A: MAD
(custom-made 1-piece acrylic) 		Mean protrusion, 5 mm in front ESS Posttreatment vs pretreatment: Not reported
74%
(range, 64%-85%) 		MAD 11.7 ± 4.3 8.4 ± 4.3 P <0.001
Inactive OA 10.8 ± 4.6 9.6 ± 4.2 P = 0.05
No intervention 10.7 ± 4.6 10 ± 4.8 NS
MAD
Inactive OA
No intervention 		8.4 ± 4.3
9.6 ± 4.3
10.0 ± 4.8 		Difference in means between groups: 0.044 †
B: inactive OA – Not reported SF-36: Posttreatment vs pretreatment:
MCS ∗ MAD
Inactive OA
No intervention 		47.2 ± 8.5
48.8 ± 10.0
50.2 ± 8.9 		51.1 ± 8.0
49.8 ± 8.5
51.2 ± 7.8 		P = 0.039
NS
NS
C: no intervention – – PCS ∗ MAD
Inactive OA
No intervention 		45.5 ± 9.5
48.1 ± 9.2
46.6 ± 9.6 		46.5 ± 8.0
47.5 ± 11.2
47.3 ± 8.7 		NS
NS
NS
General health ∗ MAD
Inactive OA
No intervention 		60.7 ± 21.9
66.6 ± 22.1
62.7 ± 19.8 		66.7 ± 20.8
66.0 ± 22.1
67.0 ± 19.5 		P = 0.011
NS
NS
Mental health ∗ MAD
Inactive OA
No intervention 		71.0 ± 14.7
78.4 ± 19.5
79.6 ± 15.2 		76.4 ± 13.8
80.4 ± 12.9
79.0 ± 15.4 		P = 0.016
NS
NS
Vitality ∗ MAD
Inactive OA
No intervention 		41.5 ± 23.4
47.8 ± 26.7
48.1 ± 24.3 		59.4 ± 24.7
47.0 ± 26.4
51.3 ± 23.4 		P <0.001
NS
NS
MAD
Inactive OA
No intervention 		59.4 ± 24.7
47.0 ± 26.4
51.3 ± 23.4 		Difference in means between groups: 0.001 †
NS , Not significant; SF-36 , medical outcome survey short form; FOSQ , functional outcomes of sleep questionnaire; ESS , Epworth sleepiness scale.

∗ Mental component summary ( MCS ) and physical component summary ( PCS ) and 3 domains, general health, mental health and vitality in SF-36

† Only vitality domain and ESS score differed significantly between intervention groups, with means in MAD group significantly different from means in inactive OA and no intervention groups (means of latter 2 groups did not differ significantly).

Table IV
Studies comparing 1-piece MADs with 1-piece MADs: patient-centered and subjective outcomes
Patients (n) Oral appliance Advancement Assessment tool Result Statistical
Author Baseline Complete (OA) Sagittal Vertical used Pretreatment Posttreatment significance Subjective treatment satisfaction
Tegelberg et al 74 55 A: MAD 50% of maximum
protrusion
mean, 4.5 ± 0.93 mm 		Not reported Questionnaire: MAD A Not reported Not reported Not reported Daytime sleepiness decrease
(1 piece) Daytime sleepiness,
apneas, and snoring 		MAD B Not reported Not reported Not reported MAD A: 45 (82%)
MAD B: 46 (84%)
B: MAD
(1 piece) 		75% of maximum
protrusion
mean, 6.4 ± 1.16 mm 		Not reported Decrease in apneas and snoring
MAD A: 48 (87%)
MAD B: 43 (79%)
Vanderveken et al 38 35 A: MAD (custom-made monobloc) 65% ± 10%
of maximum
protrusion 		Not reported VAS snoring MAD A 8 ± 2 2 ± 3 P <0.01 Satisfactory reduction in snoring
MAD A: 23 (80%)
MAD B: 18 (51%)
MAD B 8 ± 2 4 ± 3 P <0.01
B: MAD (thermoplastic Somnoguard Plus monobloc) 50% ± 20%
of maximum
protrusion 		Not reported ESS MAD A 7 ± 5 5 ± 4 NS
MAD B 7 ± 5 6 ± 4 NS
Walker-Engström et al 86 77 A: MAD A: ESS MAD A 11.7 ± 3.1 8.6 ± 2.8 P <0.001
(1 piece) 50% of maximum 2 mm MAD B 11.5 ± 3.1 7.5 ± 2.6 P <0.001
protrusion
mean, 5 mm
(4.8-5.3)
B: MAD B:
(1 piece) 75% of maximum 2 mm Questionnaire: MAD A vs MAD B: NS Satisfaction with MAD A:
protrusion Daytime sleepiness, apneas and snoring 69 (90%) patients at 6-mo follow-up were satisfied or very satisfied; 7 (9%) neither satisfied nor dissatisfied
Large decrease in daytime sleepiness:
MAD A: 33 (43%)
MAD B: 42 (54%)
Decrease in apneas and snoring:
MAD A: 48 (62%)
MAD B: 59 (77%)
mean, 7.2 mm
(6.7-7.6)
NS , Not significant; VAS, visual analog scale of patient self-evaluation of MAD efficacy; ESS , Epworth sleepiness scale.

Table V
Studies comparing 2-piece MADs with 2-piece MADs: patient-centered and subjective outcomes
Author Patients (n) Oral appliance (OA) Advancement Assessment tool used Result Statistical significance Subjective treatment satisfaction
Baseline Complete Sagittal Vertical Pretreatment Postreatment
Gauthier et al 23 16 A: MAD (commercial two-piece Silencer) 50% of max protrusion average 10.5 mm 9-12 mm ESS MAD A 13.9 ± 1.3 9.9 ± 1.3 P ≤0.01 Subjects and sleep partners reported that both MADs significantly reduced snoring frequency, choking, cessation of breathing, number of arousals, daytime sleepiness, frequency of morning headaches, daytime aggressive or irritable reactions and decreased libido ( P <0.05 to 0.001).
MAD B 13.9 ± 1.3 9.3 ± 1.2 P ≤0.001
MAD vs MAD: NS
B: MAD (commercial two-piece Klearway) 66% of max protrusion average 12.5 mm 9-12 mm FOSQ MAD A 13.8 ± 0.7 16.8 ± 0.6 P ≤0.001
MAD B 17.2 ± 0.5 P ≤0.001
MAD vs MAD: NS
VAS MAD A Not reported 6.5 ± 0.5 MAD vs MAD: NS
MAD B Not reported 7.4 ± 0.4
FSS MAD A 45.4 ± 2.7 39.0 ± 2.6 NS Cessation of breathing and perception of choking showed greater reduction with Silencer than Klearway ( P <0.05).
MAD B 39.4 ± 3.6 NS
Lawton et al 16 16 A: MAD (twinbloc) Not reported Not reported ESS MAD A 10 (2-18) ∗ 8.5 (3-17) ∗ MAD A vs MAD B: NS Not reported
MAD B 10 (2-18) ∗ 8.0 (4-18) ∗
B: MAD (Herbst) Not reported Not reported SF-36 All categories ns
VAS: daytime sleepiness MAD A 3 (1-4) ∗ 2.5 (1-4) ∗ MAD A vs MAD B: P = 0.04 MAD B less sleepy than MAD A
MAD B 3 (1-4) ∗ 2.0 (1-4) ∗
VAS: snoring MAD A 4.0 (3.0-4.0) ∗ 4.0 (2.0-4.0) ∗ MAD A vs MAD B: NS
MAD B 4.0 (3.0-4.0) ∗ 3.5 (1.0-4.0) ∗
Pitsis et al 24 23 A: MAD (two-piece) 87 ± 4% of max protrusion mean 7.3 ± 0.5 mm 4 mm ESS MAD A 18 ± 1 12 ± 1 P <0.001 Not reported
MAD B 18 ± 1 12 ± 1 P <0.001
B: MAD (two-piece) 87 ± 4% of max protrusion mean 7.3 ± 0.5 mm 14 mm
NS , Not significant; VAS , visual analog scale of patient self-evaluation of MAD efficacy; SF-36 , medical outcome survey short form; FOSQ, functional outcomes of sleep questionnaire; ESS , Epworth sleepiness scale; FSS , fatigue severity scale.

∗ Median (range) values reported.

Table VI
Studies comparing 1-piece MADs with 2-piece MADs: patient-centered and subjective outcomes
Patients (n) Oral appliance Advancement Assessment tool Result Statistical Subjective treatment satisfaction
Author Baseline Complete (OA) Sagittal Vertical used Pretreatment Posttreatment significance
Bloch et al 24 24 A: MAD (monobloc) 75% of maximum protrusion mean, 10 ± 0.4 mm 5-10 mm ESS MAD A 9.0 ∗ (6.5-10) ∗ MAD A vs no treatment: P <0.001 Not reported
MAD B 9.0 ∗ (6.5-11) ∗ MAD B vs no treatment: P <0.001
No treatment 13.5 ∗ (9.5-16) ∗ MAD B vs MAD A: NS
B: MAD (Herbst) 75% of maximum protrusion mean, 10 ± 0.4 mm 4-6 mm range of opening, >15 mm
Sleep symptoms questionnaire:
Interference with daily tasks MAD A 1.5 ∗ (1.0-2.0) ∗ MAD A vs no treatment: P <0.001
MAD B 2.0 ∗ (1.0-3.0) ∗ MAD B vs no treatment: P <0.03
No treatment 3.5 ∗ (2.5-4.0) ∗ MAD B vs MAD A: P <0.05
C: no treatment – – Perform ability MAD A 2.0 ∗ (1.0-2.0) ∗ MAD A vs no treatment: P <0.001
MAD B 2.0 ∗ (2.0-3.5) ∗ MAD B vs no treatment: P <0.03
No treatment 3.0 ∗ (2.0-4.0) ∗ MAD B vs MAD A: P <0.05
Energy level MAD A 2.0 ∗ (2.0-3.0) ∗ MAD A vs no treatment: P <0.001
MAD B 2.0 ∗ (2.0-3.0) ∗ MAD B vs no treatment: P <0.03
No treatment 3.0 ∗ (2.5-4.0) ∗ MAD A vs MAD B: NS
Snoring frequency MAD A 2.0 ∗ (1.0-3.0) ∗ MAD A vs no treatment: P <0.001
MAD B 2.0 ∗ (1.0-3.5) ∗ MAD B vs no treatment: P <0.001
No treatment 4.0 ∗ (4.0-4.0) ∗ MAD A vs MAD B: NS
Snoring loudness MAD A 1.5 ∗ (1.0-2.0) ∗ MAD A vs no treatment: P <0.001
MAD B 2.0 ∗ (1.0-2.5) ∗ MAD B vs no treatment: P <0.001
No treatment 3.5 ∗ 3.5 (3.0-4.0) ∗ MAD A vs MAD B: P <0.05
Rose et al 26 16 A: MAD (2-piece soft polyethylene Silencor) 75% maximum
protrusion 		5 mm VAS:
daytime sleepiness, 		MAD A
MAD B 		7.2 (1.7)
7.0 (1.5) 		5.4 (1.0)
4.1 (0.7) 		Both MADs reduced daytime sleepiness and snoring significantly while Not reported
B: MAD (acrylic 1-piece Karwetzky) 75% maximum protrusion 10-12 mm Snoring
Sleep quality 		MAD A
MAD B
MAD A
MAD B 		9.1 (0.8)
8.8 (1.0)
6.4 (1.8)
6.2 (1.2) 		3.2 (1.4)
3.4 (2.7)
4.1 (1.4)
4.5 (2.1) 		Enhancing sleep quality (no details given);
no differences were found in this respect between the 2 MADs
Only gold members can continue reading. Log In or Register to continue

Related posts:

  1. Self-ligating brackets: Looking back and going forward
  2. Correction
  3. Physical, mechanical, and flexural properties of 3 orthodontic wires: An in-vitro study
  4. Retainer wear and compliance in the first 2 years after active orthodontic treatment
  5. Three-dimensional ultrasound diagnostics of tongue posture in children with unilateral posterior crossbite
  6. It’s time we listened to our teeth: The SoundBite hearing system

Stay updated, free dental videos. Join our Telegram channel
Tags:
Apr 13, 2017 | Posted by in Orthodontics | Comments Off on Subjective efficacy of oral appliance design features in the management of obstructive sleep apnea: A systematic review

VIDEdental - Online dental courses
Get VIDEdental app for watching clinical videos