PDF The effectiveness of different mandibular advancement amounts in OSA patients: a systematic review and meta-regression analysis

Sleep Breath (2016) 20:911–919 DOI 10.1007/s11325-015-1307-7 SLEEP BREATHING PHYSIOLOGY AND DISORDERS ORIGINAL ARTICLE The effectiveness of different mandibular advancement amounts in OSA patients: a systematic review and meta-regression analysis Maria Lavinia Bartolucci 1 & Francesco Bortol...

Sleep Breath (2016) 20:911–919 DOI 10.1007/s11325-015-1307-7 SLEEP BREATHING PHYSIOLOGY AND DISORDERS ORIGINAL ARTICLE The effectiveness of different mandibular advancement amounts in OSA patients: a systematic review and meta-regression analysis Maria Lavinia Bartolucci 1 & Francesco Bortolotti 1 & Eugenia Raffaelli 2 & Vincenzo D’Antò 1 & Ambra Michelotti 1 & Giulio Alessandri Bonetti 2 Received: 12 August 2015 / Revised: 20 December 2015 / Accepted: 29 December 2015 / Published online: 15 January 2016 # Springer-Verlag Berlin Heidelberg 2016 Abstract higher than 50 % do not significantly influence the success Purpose The therapy with mandibular advancement devices rate (Q = 0.373, p = 0.541). According to the GRADE score (MADs) represents a treatment option for patients with ob- system, the quality of evidence resulted to be moderate. structive sleep apnea (OSA). The literature does not provide Conclusion The AHI improvement resulted to be not propor- evidence regarding the most effective mandibular advance- tional to the mandibular advancement increase. It is plausible ment; therefore, the aim of this systematic review with meta- that the success of the therapy is influenced by a combination regression was to investigate the effectiveness of different of variables that need closer study. mandibular advancement amounts in reducing apnea- hypopnea index (AHI) in OSA patients. Keywords Obstructive sleep apnea. Mandibular Methods An electronic search was performed in MEDLINE, advancement device. Sleep disorders. Sleep disordered Cochrane Database, Google Scholar Beta, ISI Web of Knowl- breathing. Temporomandibular disorders edge, Scopus, and LILACS to select randomized controlled trials (RCTs) investigating the efficacy of MADs in reducing AHI in adult OSA patients. Inclusion criteria were the diag- Introduction nosis of OSA and success evaluation performed with a polysomnography, follow-up of maximum 12 months, and Obstructive sleep apnea (OSA) syndrome is a common sleep- protrusion amount reported as a percentage of the maximum related breathing disorder affecting 2–4 % of middle-aged mandibular advancement. The quality of evidence was evalu- men and women [1–3]. It is characterized by the repetitive ated using the Grading of Recommendations Assessment, De- obstruction of the upper airway during sleep, which deter- velopment and Evaluation (GRADE) methodology. The suc- mines snoring, sleep fragmentation and is associated to sys- cess rate of each study was computed: [(mean AHI at temic hypertension, metabolic syndrome, heart failure, baseline-mean AHI after treatment)/mean AHI at baseline]. neurocognitive impairment [4, 5] and with a significant in- Results Thirteen RCTs performing advancements from 50 to creased risk of mortality [6, 7]. 89 % of maximum mandibular protrusion were included. The Forward repositioning of the mandible increases the upper meta-regression analysis showed that advancement amounts airway volume, widens the lateral dimension of the velopharynx, it stretches tongue muscles counteracting tongue’s retrolapse during sleep, and it moves the hyoid bone * Maria Lavinia Bartolucci [email protected] anteriorly and stabilizes epiglottis and soft palate preventing the posterior rotation of the jaw [8–10]. The therapy with a mandibular advancement device (MAD) represents a treat- 1 Department of Neurosciences, Reproductive Sciences and Oral ment option to obtain this anterior jaw repositioning; it is Sciences, Section of Orthodontics, University of Naples “Federico indicated in patients affected by mild to moderate OSA and II”, via Pansini, 5, 80131 Naples, Italy in the ones with severe OSA who refuse continuous positive 2 Department of Biomedical Sciences, Section of Orthodontics, airway pressure (CPAP) treatment or surgery [2, 11–14]. The University of Bologna, via san Vitale, 59, 40125 Bologna, Italy MAD therapy generally improves the Apnea Hypopnea Index 912 Sleep Breath (2016) 20:911–919 (AHI) , the sleep quality, and also the work performances MADs’ protrusion amount, AHI at baseline, AHI after thera- [8, 12]. py, follow-up, and authors’ main conclusions. Despite many studies investigated the efficacy of the MAD therapy, no consistent data are provided about the most effec- tive mandibular protrusion amount: some authors suggest a Risk of bias in individual studies and across the studies mild advancement [15, 16], others prefer a high protrusion [17–19], and while others perform a progressive advancement To document the methodological soundness of each article, in order to avoid the masticatory side effects that the MAD the Quality Assessment Tool for Quantitative Studies by Ef- therapy brings about [15, 20]. fective Public Health Practice Project (EPHPP) was used. Since a common advancement protocol is lacking , the To evaluate the risk of bias in individual studies, the Cochrane aim of the present systematic review with meta-regression anal- Collaboration’s tool for assessing risk of bias in randomized ysis is to investigate which is the effectiveness of different man- trials was used. To evaluate the quality of body of evi- dibular protrusion amounts in reducing AHI in OSA patients. dence, the Grading of Recommendations Assessment, Devel- opment and Evaluation (GRADE) was performed. Two assessors (F.B. and G.A.B.) independently performed Material and methods the quality evaluations; when in disagreement, a conjunct evaluation was performed to reach a consensus. The present systematic review follows the Preferred Reporting The risk of bias across the studies was evaluated by means Items for Systematic Reviews and Meta-Analyses (PRISMA) of Egger’s test and Funnel plot; no information about system. reporting bias is presented because our evaluations are not To identify the studies to be considered for inclusion, de- based on raw data of individual studies. tailed search strategies were developed for the following elec- Statistical tests of heterogeneity were carried out to assess tronic databases: MEDLINE, Cochrane Database of System- whether the observed variability in study results (effect sizes) atic Reviews, Google Scholar Beta, ISI Web of Knowledge, was greater than that expected to occur by chance. The het- Scopus, and LILACS. Also gray literature was searched in erogeneity among studies was assessed using a χ2-based Q order to find unpublished data. Manual search of selected statistic test and I2 index; however, because of the moderate studies by looking at their references was conducted as well. insensitivity of the Q statistic, only an I2 index greater than Studies from 1 January 1990 to 30 April 2015 were ana- 50 % was considered associated with a substantial heteroge- lyzed, without language restrictions. neity among the studies. The tau2 was also calculated for the The search strategy used for MEDLINE including the heterogeneity assessment. MeSH and text words was (((Sleep Apnea, Obstructive [Mesh]) OR (Sleep Apnea) OR (Sleep Apnoea)) OR (Sleep breathing disorder OR sleep respiratory disorder))) AND Definition for success rate ((Mandibular Advancement Device) OR (Mandibular ad- vancement appliance)). Success rate for each protrusion amount was calculated as Only randomized controlled trials (RCTs) that investigated [(mean AHI at baseline - mean AHI after treatment)/mean the relationship between the MAD protrusion amount and the improvement of the AHI were included in the present review. Other inclusion criteria were the diagnosis of OSA performed with a polysomnography (PSG) (AHI ≥ 5) same as the evalu- ation of the success of the therapy, adult patients (>18 years), follow-up of maximum 12 months to limit the variation of the BMI which could affect AHI’s results, and protrusion amount reported as a percentage value of the maximum mandibular protrusion. Two researchers independently selected the articles (M.L.B. and E.R.). Intra-examiner conflicts were resolved by the discussion of each article to reach a consensus. Data items and collection The following data items were collected from each study in- cluded: study design, sample size, mean age, mean BMI, Fig. 1 Flow diagram of the search strategy (PRISMA) Table 1 Description of the studies included in the systematic review Sleep Breath (2016) 20:911–919 Study Study Control Sample size Age BMI Protrusion AHI baseline AHI after therapy Follow-up Authors’ main design (Mean ± SD) (Mean ± SD) amount (Mean ± SD) (Mean ± SD) conclusion Bloch 2000 RCT crossover Different type of MAD 24 50.5 ± 1.5 27.4 ± 0.6 75 % 22.6 ± 3.1 7.9 ± 1.6 156 ± 14 days Both the Herbst and the 22.6 ± 3.1 8.7 ± 1.5 monobloc are effective Walker-Engstrom RCT UPPP 95 (37 with MAD) >20 and 0.10); if homogeneity was rejected (p < 0.10), a random Gauge System [20, 27, 31, 34] or Projet bite forks ). effects model was used to better aggregate the data. Ten papers described the type of PSG and the way it was Success rate and 95 % confidence interval (CI) of each treat- performed [16, 17, 20, 26–29, 31, 33, 34]. Few authors report- ed group within the studies were computed; regarding the stud- ed the total sleep time of the registration night [28, 32]. ies comparing the efficacy of different MADs at the same ad- vancement amount, the success rate was computed for each group. In order to evaluate the effect of the advancement Quality analysis and risk of bias in individual studies amount on success rate, a meta-regression analysis was per- formed. Calculations were carried out by means of Comprehen- The results of the EPHPP quality analysis are shown in sive Meta-Analysis software (Biostat Inc, Englewood, NJ). Table 2. Selection bias was weak in two studies [28, 31] because the authors did not report the way the par- ticipants were enrolled in the study. Confounders (age, Results gender, BMI, neck circumference, smoke, alcohol con- sumption, and AHI at baseline) were not controlled in From the 674 articles initially retrieved, 13 studies were in- four studies [16, 26, 31, 34]. cluded for the qualitative analysis. Figure 1 reports the selec- In only one study, both assessors and participants were tion process and Table 1 describes the studies included. blinded , in six studies assessors or participants were Among these, one trial compared the effectiveness of 25, 50, blinded [15–17, 26, 30, 32], and six studies did not use and 75 % of maximum mandibular protrusion and two blinding for measurements [20, 28, 29, 31, 33, 34]. Regarding studies compared the efficacy of 50 and 75 % [16, 17]; nine withdrawals, one study gained a moderate score because studies performed only one advancement amount (50 , 70 the rate of patients that completed the follow-up was of only , 74 , 75 [28–32], 83 , or 89 % ) in compar- 60–79 %. ison with other therapies or no treatment. Two studies [28, 33] The risk of bias analysis resulted low or unclear for all compared the two groups at the same protrusion amount wear- studies (Table 3). The main shortcomings are that allocation ing different devices: the analysis computed those groups concealment and sequence generation are not described in separately. most of the studies. Table 2 Quality evaluation in individual studies (EPHPP) Article Selection bias Study design Confounders Blinding Data collection Withdrawals and Global methods drop-outs ratings Bloch 2000 Weak Strong Strong Weak Strong Strong Weak Walker-Engstrom 2002 Moderate Strong Weak Moderate Strong Moderate Moderate Tan 2002 Moderate Strong Strong Weak Strong Strong Moderate Johnston 2002 Moderate Strong Strong Moderate Strong Strong Strong Tegelberg 2003 Moderate Strong Weak Moderate Strong Strong Moderate Walker-Engstrom 2003 Moderate Strong Strong Moderate Strong Strong Strong Blanco 2005 Weak Strong Weak Weak Strong Strong Weak Hoekema 2007 Moderate Strong Weak Weak Strong Strong Weak Petri 2008 Moderate Strong Strong Strong Strong Strong Strong Ghazal 2009 Moderate Strong Moderate Weak Strong Strong Moderate Deane 2009 Moderate Strong Strong Moderate Strong Strong Strong Campbell 2009 Moderate Strong Moderate Weak Strong Strong Moderate Aarab 2010 Moderate Strong Strong Moderate Strong Strong Strong  Sleep Breath (2016) 20:911–919 Table 3 Risk of bias in individual studies (Cochrane Collaboration’s tool) Study Sequence generation Allocation concealment Blinding Incomplete outcome data addressed Selective outcome reporting Bloch 2000 Unclear Unclear High risk (no blinding) Low risk (no drop-out) Low risk Walker-Engstrom 2002 Unclear Unclear Low risk (the assessor was blinded) Low risk (ITT) Low risk Tan 2002 Unclear Unclear High risk (no blinding) Unclear Low risk Johnston 2002 Unclear Unclear High risk (no blinding) Low risk (10 % of patients dropped out) Low risk Tegelberg 2003 Low risk Unclear Low risk (assessor was blinded) Low risk (missing data balanced across groups) Low risk (closed envelops) Walker-Engstrom 2003 Unclear Low risk (closed Low risk (dentist and assessor Low risk (ITT) Low risk envelope allocation) were blinded) Blanco 2005 Unclear Unclear High risk (no blinding) Low risk (missing data balanced across groups) Low risk Hoekema 2007 Unclear Unclear High risk (no blinding) Low risk (no drop out) Low risk Petri 2008 Low risk Low risk (allocation Low risk (patients, dentist and Low risk (missing data balanced across groups) Low risk (central generation) computer generated) assessor were blinded) Ghazal 2009 Low risk Low risk (central High risk (no blinding) Low risk (missing data balanced across groups) Low risk (computer generation) randomization) Deane 2009 Unclear Unclear Low risk (the assessor was blinded) Low risk (reasons unlikely to be related to true outcome) Low risk Campbell 2009 Unclear Unclear High risk (no blinding) Low risk (reasons unlikely to be related to true outcome) Low risk Aarab 2010 Unclear Unclear Low risk (Assessors, participants Low risk (15 % of patients dropped out) Low risk and personnel blinded) 915 916 Sleep Breath (2016) 20:911–919 The GRADE scores are shown in Table 4. The reason for Discussion lowering the quality of the evidence to the moderate score was the indirectness. The aim of the present systematic review with meta-regression analysis was to investigate the effectiveness of different man- Results of individual studies dibular advancement amounts to treat OSA patients with the MAD. The literature does not provide common guidelines: a Table 1 reports the results of individual studies. Aarab et al. large number of studies reported the efficacy of different de- compared the advancements of 0, 25, 50, and 75 % of the vices without explaining the rationale to choose the advance- maximum protrusion with a cross over study design showing ment amount. that the advancements of 50 and 75 % were effective in re- Thirteen RCTs were included for the qualitative analysis ducing AHI. Other two cross over studies [16, 17] compared and resulted to be of a medium-high quality on average. The different protrusion amounts (50 and 75 %) using the same meta-regression analysis was conducted to evaluate the effect MAD. Three studies compared different types of MAD at the of different amounts of mandibular advancement on success same protrusion amount [20, 28, 33]. Three authors compared rate defined as the improvement of AHI after treatment MADs with non-advanced appliances [27, 30, 31]. Two stud- weighted on the AHI value at baseline. The results show that ies [29, 34] compared the efficacy of the MAD versus CPAP the advancement amounts do not influence the success rate: in in reducing AHI. In one study , MAD was compared with fact, the regression line in the bubble plot remains almost flat, uvulopalatopharyngoplasty. without upward or downward trends (Fig. 3). The mean value of the success rate among the analyzed studies is 62.3 % (Fig. 2), and they include advancements from 25 to 89 % of Synthesis of results the maximum mandibular capacity. All the studies included except one investigated the efficacy of a protrusion One study reported the AHI values as median and inter- amount of 50 % or higher. The mandibular advancement of quartile ranges: in order to compute the success rate, the mean 50 % is widely used as the minimum effective advancement to values were obtained contacting the authors. Heterogeneity of start the MAD therapy, but this parameter has not been ade- the groups was 8.7 % therefore a fixed effect model was used. quately determined on the basis of comparisons with lower Figure 2 reports the success rates and confidence intervals of mandibular advancements. In fact, the literature provides only each group included in the quantitative analysis. The meta- two studies that investigated the efficacy of mandibular ad- regression analysis showed that the mandibular advancement vancements lower than 50 % [15, 35]. What emerges from the amount does not significantly influence the success rate present analysis is that the success rate of the study with the (Q = 0.018, p = 0.892) as reported in Fig. 3. advancement of 25 % is not the lowest registered. As shown in Fig. 2, success rates lower than 50 % derive from investiga- Risk of bias across studies tions with mandibular advancements from 74 to 83 % [20, 27, 30, 33]. On the other hand, among the studies that registered a Concerning the publication bias, Funnel plot of healing im- success rate higher than 70 %, there is also one investigation provement (success rate) against its standard error is reported that performed an advancement of 50 %. It is also impor- in Fig. 4; results of Egger’s test (intercept 0.467, p = 0.623) tant to underline that in most of the analyzed studies the suc- demonstrate no significant deviation of the intercept from the cess rate has wide confidence intervals (Fig. 2) suggesting a symmetry, and consequently, studies with greater size are dis- high inter-individual variability in response to the MAD ther- tributed near the mean and studies of less size are spread. apy supporting the findings of a recent systematic-review Table 4 Quality assessment and summary of findings across studies (GRADE) Quality assessment Summary of findings Number of studies Risk of bias Inconsistency Indirectness Imprecision Publication Number Mean Success Quality (Design) bias of patients rate [95 % CI] Effectiveness of different advancements in reducing AHI Outcome: Success rate Thirteen (RCTs) Unclear/Low No inconsistency Serious indirectness No serious No serious publication 484 0.623 Moderate RoB for most for most of imprecision bias [0.562–0.678] of the studies the studies RoB risk of bias Sleep Breath (2016) 20:911–919 917 Fig. 2 Forest plot of success rates in the included groups (correlation = success rate). The present results seem to be in contrast with the current and temporomandibular joint ligament causing a strain in opinion that the more the mandible is stretched forward the those muscles and in the retrodiscal tissues, that could induce more the AHI improves. the onset of tenderness or pain in the temporomandibular The outcomes of the present meta-regression suggest that structures [39, 40]. As a consequence, a cautious approach the amount of mandibular protrusion could not be the unique with a lower mandibular advancement could control these factor that influences the effectiveness of a MAD therapy, but events, considering also that the American Academy of Sleep also other elements could contribute to determine the reaction Medicine guidelines recognize the development of temporo- of a single patient to the treatment and a careful control of mandibular disorders as the main reason for interrupting the predictive parameters would be crucial to personalize the therapy with the MAD. treatment. It is also important to consider that a significant intra- To date, the data provided regarding possible predictive individual night to night variability in AHI was demonstrated: parameters of the efficacy of MAD treatment are quite few in a sample of 193 patients, Ahmadi and coworkers reg- and inconclusive. To the question of tailoring the treat- istered a difference of AHI > 5 between two consecutive PSGs ment according to the single patient’s characteristics, the video in 21 % of the subjects; White and colleagues registered a sleep endoscopy may represent a useful answer since during difference of AHI > 10 between the two consecutive PSGs in the investigation it could be possible to identify the effective the 35 % of their sample (n = 26). To this regard, Aarab protrusion amount controlling the progressive improvement and coworkers suggested to be cautious in diagnosing OSA on of the airway patency [8, 11, 12]. the basis of a single PSG registration. Consequently, fur- It could be interesting to better investigate the possibility of ther methodologically sound clinical studies considering the performing lower advancements in order to limit the side ef- intra-individual AHI variability are needed. fects: the MAD forces the mandible in a forward and down- The limitations of the present review are linked to the small ward position that elongates the fibers of jaw elevator muscles sample size of the studies. In addition, not all the included Fig. 3 Bubble plot of the relationship between success rate (reported as Fisher’s Z) and mandibular advancement (the size of the bubbles is Fig. 4 Funnel plot of the effect size of studies included against its proportional to the weight of the studies in the meta-regression) Standard Error 918 Sleep Breath (2016) 20:911–919 studies controlled the confounding variables such as BMI, 2. Epstein LJ, Kristo D, Strollo PJ Jr et al (2009) Clinical guideline for the evaluation, management and long-term care of obstructive sleep neck circumference, and AHI at baseline, which could be apnea in adults. J Clin Sleep Med 5:263–276 causally related to the outcome of interest. 3. Myers KA, Mrkobrada M, Simel DL (2013) Does this patient have The risk of bias of all the studies resulted low or unclear obstructive sleep apnea? The rational clinical examination system- (Table 3), and the EPHPP quality analysis showed that al- atic review. JAMA 310:731–741 4. Dong JY, Zhang YH, Qin L-Q (2013) Obstructive sleep apnea and though only RCTs were included, their quality resulted strong cardiovascular risk: meta-analysis of prospective cohort studies. in five studies, moderate in five, and weak in the other three. Atherosclerosis 229:489–495 The insufficient control of the possible confounders and the 5. Wang X, Bi Y, Zhang Q, Pan F (2013) Obstructive sleep apnoea and lack of blinding negatively influenced the scores attributed to the risk of type 2 diabetes: a meta-analysis of prospective cohort studies. Respirology 18:140–146 the studies. Regarding the body of evidence quality assess- 6. Young T, Finn L, Peppard PE et al (2008) Sleep disordered breath- ment (GRADE), the moderate score is due to the indirectness: ing and mortality: eighteen-year follow-up of the Wisconsin sleep most of the studies aimed at comparing different therapies or cohort. Sleep 31:1071–1078 different devices at the same protrusion amount, but not the 7. Wang X, Ouyang Y, Wang Z, Zhao G, Liu L, Bi Y (2013) Obstructive sleep apnea and risk of cardiovascular disease and efficacy of different mandibular advancements. all-cause mortality: a meta-analysis of prospective cohort studies. The present study revealed some weakness of the current Int J Cardiol 169:207–214 evidence; therefore, there is a need for further studies aimed at 8. Hoekema A, Stegenga B, De Bont LG (2004) Efficacy and co- detecting possible confounders able to influence the response morbidity of oral appliances in the treatment of obstructive sleep to the therapy and at comparing the efficacy of different man- apnea-hypopnea: a systematic review. Crit Rev Oral Biol Med 15: 137–155 dibular protrusion amounts also lower than 50 %, in order to 9. Chan AS, Sutherland K, Schwab RJ et al (2010) The effect of verify the results of this meta-regression. mandibular advancement on upper airway structure in obstructive Since some authors suggest to use adjustable MADs that sleep apnoea. Thorax 65:726–732 allow to perform a progressive advancement and to manage 10. Dieltjens M, Vanderveken OM, Heyning PH, Braem MJ (2012) Current opinions and clinical practice in the titration of oral appli- the possible symptoms during the titration [8, 14, 15, 17, 45] ances in the treatment of sleep-disordered breathing. Sleep Med and in light of the present results, it would be recommended to Rev 16:177–185 careful titrate the advancement of the MAD starting with min- 11. Ahrens A, Mcgrath C, Hägg U (2011) A systematic review of the imum protrusions in order to better individualize the therapy efficacy of oral appliance design in the management of obstructive sleep apnoea. Eur J Orthod 33:318–324 and to induce an adaptation of the masticatory system. 12. Lim J, Lasserson TJ, Fleetham J, Wright J (2006) Oral appliances for obstructive sleep apnoea. Cochrane Databases Syst Rev CD004435 13. Holty JE, Guilleminault C (2010) Maxillomandibular advancement Conclusions for the treatment of obstructive sleep apnea: a systematic review and meta-analysis. Sleep Med Rev 14:287–297 There is small body of moderate quality evidence to suggest 14. Kushida CA, Chediak A, Berry RB et al (2008) Clinical guidelines for the manual titration of positive airway pressure in patients with that increasing the mandibular advancement does not produce obstructive sleep apnea. J Clin Sleep Med 4:157–171 significant improvements in the success rate since there is a 15. Aarab G, Lobbezoo F, Hamburger HL, Naeije M (2010) Effects of high inter-individual variability in response to the MAD an oral appliance with different mandibular protrusion positions at a therapy. constant vertical dimension on obstructive sleep apnea. Clin Oral It would be cautious to begin the therapy addressing the Investig 14:339–345 16. Tegelberg Å, Walker-Engström ML, Vestling O, Wilhelmsson B minimum effective mandibular advancement. (2003) Two different degrees of mandibular advancement with a dental appliance in treatment of patients with mild to moderate obstructive sleep apnea. Acta Odontol Scand 61:356–362 Compliance with ethical standards The manuscript does not contain 17. Walker-Engström ML, Ringqvist I, Vestling O, Wilhelmsson B, clinical studies or patient data. Tegelberg A (2003) A prospective randomized study comparing two different degrees of mandibular advancement with a dental Conflict of interest The authors declare that they have no competing appliance in treatment of severe obstructive sleep apnea. Sleep interest. Breath 7:119–130 18. Gindre L, Gagnadoux F, Meslier N, Gustin JM, Racineux JL (2008) Mandibular advancement for obstructive sleep apnea: dose effect on apnea, long-term use and tolerance. Respiration 76:386–392 19. Gagnadoux F, Fleury B, Vielle B et al (2009) Titrated mandibular advancement versus positive airway pressure for sleep apnoea. Eur Respir J 34:914–920 References 20. Campbell AJ, Reynolds G, Trengrove H, Neill AM (2009) Mandibular advancement splint titration in obstructive sleep ap- 1. Young T, Peppard PE, Gottlieb DJ (2002) Epidemiology of obstruc- noea. Sleep Breath 13:157–162 tive sleep apnea: a population health perspective. Am J Respir Crit 21. Liberati A, Altman DG, Tetzlaff J et al (2009) The PRISMA state- Care Med 165:1217–1239 ment for reporting systematic reviews and meta-analyses of studies Sleep Breath (2016) 20:911–919 919 that evaluate health care interventions: explanation and elaboration. 34. Hoekema A, Stegenga B, Bakker M et al (2007) Simulated driving J Clin Epidemiol 62:e1–e34 in obstructive sleep apnoea-hypopnoea; effects of oral appliances 22. Thomas BH (2004) A process for systematically reviewing the and continuous positive airway pressure. Sleep Breath 11:129–138 literature: providing the research evidence for public health nursing 35. Kato J, Isono S, Tanaka A et al (2000) Dose-dependent effects of interventions. Worldviews Evid Based Nurs 1:176–184 mandibular advancement on pharyngeal mechanics and nocturnal 23. Higgins JP, Altman DG, Gøtzsche PC et al (2011) The cochrane oxygenation in patients with sleep-disordered breathing. Chest 117: collaboration’s tool for assessing risk of bias in randomised trials. 1065–1072 BMJ 343:d5928 36. Ngiam J, Balasubramaniam R, Darendeliler MA, Cheng AT, 24. Guyatt G, Oxman AD, Akl EA et al (2011) GRADE guidelines: 1. Walters K, Sullivan CE (2013) Clinical guidelines for oral appli- Introduction-GRADE evidence profiles and summary of findings ance therapy in the treatment of snoring and obstructive sleep ap- tables. J Clin Epidemiol 64:383–394 noea. Aust Dent J 58:408–419 25. Borenstein M, Hedges LV, Higgins JPT, Rothstein HR (2009) 37. Ferguson KA, Cartwright R, Rogers R, Schmidt-Nowara W (2006) Introduction to Meta-Analysis. Wiley, Chichester Oral appliances for snoring and obstructive sleep apnea: a review. 26. Walker-Engström M-L, Tegelberg A, Wilhelmsson B, Ringqvist I Sleep 29:244–262 (2002) 4-year follow-up of treatment with dental appliance or 38. Saffer F, Lubianca JF, Rösing C, Dias C, Closs L (2015) Predictors uvulopalatopharyngoplasty in patients with obstructive sleep apnea: of success in the treatment of obstructive sleep apnea syndrome a randomized study. Chest 121:739–746 with mandibular repositioning appliance: a systematic review. Int 27. Petri N, Svanholt P, Solow B, Wildschiødtz G, Winkel P (2008) Arch Otorhinolaryngol 19:80–85 Mandibular advancement appliance for obstructive sleep apnoea: 39. Bowley JF, Gale E (1987) Experimental masticatory muscle pain. J results of a randomised placebo controlled trial using parallel group Dent Res 66:1765–1769 design. J Sleep Res 17:221–229 40. Doff MH, Veldhuis SK, Hoekema A et al (2012) Long-term oral 28. Bloch KE, Iseli A, Zhang JN et al (2000) A randomized, controlled appliance therapy in obstructive sleep apnea syndrome: a controlled crossover trial of two oral appliances for sleep apnea treatment. Am study on temporomandibular side effects. Clin Oral Investig 16: J Respir Crit Care Med 162:246–251 689–697 29. Tan YK, L’Estrange PR, Luo YM et al (2002) Mandibular advance- ment splints and continuous positive airway pressure in patients 41. Kushida CA, Morgenthaler TI, Littner MR et al (2006) Practice with obstructive sleep apnoea: a randomized cross-over trial. Eur parameters for the treatment of snoring and obstructive sleep apnea J Orthod 24:239–249 with oral appliances: an update for 2005. Sleep 29:240–243 30. Johnston CD, Gleadhill IC, Cinnamond MJ, Gabbey J, Burden DJ 42. Ahmadi N, Shapiro GK, Chung SA, Shapiro CM (2009) Clinical (2002) Mandibular advancement appliances and obstructive sleep diagnosis of sleep apnea based on single night of polysomnography apnoea: a randomized clinical trial. Eur J Orthod 24:251–262 vs. two nights of polysomnography. Sleep Breath 13:221–226 31. Blanco J, Zamarrón C, Abeleira Pazos MT, Lamela C, Suarez 43. White LH, Lyons OD, Yadollahi A, Ryan CM, Bradley TD (2015) Quintanilla D (2005) Prospective evaluation of an oral appliance Night-to-night variability in obstructive sleep apnea severity: rela- in the treatment of obstructive sleep apnea syndrome. Sleep Breath tionship to overnight rostral fluid shift. J Clin Sleep Med 11:149– 9:20–25 156 32. Deane SA, Cistulli PA, Ng AT, Zeng B, Petocz P, Darendeliler MA 44. Aarab G, Lobbezoo F, Hamburger HL, Naeije M (2009) Variability (2009) Comparison of mandibular advancement splint and tongue in the apnea-hypopnea index and its consequences for diagnosis stabilizing device in obstructive sleep apnea: a randomized con- and therapy evaluation. Respiration 77:32–37 trolled trial. Sleep 32:648–653 45. Barthlen GM, Brown LK, Wiland MR, Sadeh JS, Patwari J, 33. Ghazal A, Sorichter S, Jonas I, Rose EC (2009) A randomized Zimmerman M (2000) Comparison of three oral appliances for prospective long-term study of two oral appliances for sleep apnoea treatment of severe obstructive sleep apnea syndrome. Sleep Med treatment. J Sleep Res 18:321–328 1:299–305