Cephalometric Predictors of Treatment Outcome with Mandibular Advancement Devices in Adult Patients with Obstructive Sleep Apnea (PDF)

THE KOREAN JOURNAL of Review Article ORTHODONTICS pISSN 2234-7518 eISSN 2005-372X...

THE KOREAN JOURNAL of Review Article ORTHODONTICS pISSN 2234-7518 eISSN 2005-372X http://dx.doi.org/10.4041/kjod.2015.45.6.308 Cephalometric predictors of treatment outcome with mandibular advancement devices in adult patients with obstructive sleep apnea: a systematic review Giulio Alessandri-Bonettia Objective: The efficacy of mandibular advancement devices (MADs) in the Daniela Rita Ippolitob treatment of obstructive sleep apnea (OSA) ranges between 42% and 65%. Maria Lavinia Bartoluccic However, it is still unclear which predictive factors can be used to select suitable Vincenzo D’Antòc,d patients for MAD treatment. This study aimed to systematically review the Serena Incerti-Parentia literature on the predictive value of cephalometric analysis for MAD treatment outcomes in adult OSA patients. Methods: The MEDLINE, Google Scholar, Scopus, and Cochrane Library databases were searched through December 2014. Reference lists from the retrieved publications were also examined. English language studies published in international peer-reviewed journals concerning a Unit of Orthodontics, Department of the predictive value of cephalometric analysis for MAD treatment outcome Biomedical and Neuromotor Sciences, were considered for inclusion. Two review authors independently assessed eli­ University of Bologna, Bologna, Italy gibility, extracted data, and ascertained the quality of the studies. Results: b Department of Orthodontics, Fifteen eligible studies were identified. Most of the skeletal, dental, and soft University of Brescia, Brescia, Italy ti­ssue cephalometric measurements examined were widely recognized as not c Department of Neuroscience, prognostic for MAD treatment outcome; however, controversial and limited data Reproductive Science and Oral Science, were found on the predictive role of certain cephalometric measurements in­ University of Naples “Federico II”, Naples, Italy cluding cranial base angle, mandibular plane angle, hyoid to mandibular plane d distance, posterior nasal spine to soft-palate tip distance, anterior nasal spine to Department of Pediatric Surgery, Bambino Gesù Children’s Hospital, epiglottis base distance, and tongue/oral cross sectional area ratio thus justifying Rome, Italy additional studies on these parameters. Conclusions: Currently available evi­dence is inadequate for identification of cephalometric parameters capable of reliably discriminating between poor and good responders to MAD treatment. To guide further research, methodological weaknesses of the currently available stu­dies were highlighted and possible reasons for their discordant results were analyzed. [Korean J Orthod 2015;45(6):308-321] Key words: Obstructive sleep apnea, Cephalometry; Mandibular advancement Received February 25, 2015; Revised April 15, 2015; Accepted May 6, 2015. Corresponding author: Giulio Alessandri-Bonetti. Aggregate Professor, Unit of Orthodontics, Department of Biomedical and Neuromotor Sciences (DIBINEM), University of Bologna, Via San Vitale 59, Bologna 40125, Italy. Tel +39-0512088133 e-mail [email protected] The authors report no commercial, proprietary, or financial interest in the products or companies described in this article. © 2015 The Korean Association of Orthodontists. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. 308 Alessandri-Bonetti et al Cephalometric predictors of MADs treatment outcome INTRODUCTION response in adult patients with OSA. Obstructive sleep apnea (OSA) is characterized by repe­ MATERIALS AND METHODS titive episodes of complete or partial closure of the upper airway during sleep that lead to sleep fragmentation Search strategy and oxygen desaturation.1 This sleep-related breathing An electronic literature search was carried out on disorder is associated with daytime sleepiness, impaired the following databases: MEDLINE, Google Scholar, quality of life, poor work performance, neurocognitive Scopus, and the Cochrane Library. To identify the rele­ decline, increased risk of motor vehicle accidents and, vant studies the following search terms were used: in the long term, an increased risk of cardio-vascular “obstructive sleep" AND (apnea OR apnoea) AND disease and mortality.2 Nasal continuous positive airway (predict* OR outcome OR effect OR efficacy) AND (“oral pressure (nCPAP) maintains a positive pressure in the appliance" OR “mandibular advancement device" OR upper airway through a nose mask worn during sleep “mandibular repositioning appliance") AND (craniofacial and is currently the most effective treatment option for OR skeletal OR cephalometr*). The reference lists of OSA patients.3 Nevertheless, adherence to this therapy is all relevant publications were checked for additional low, with rates between 60% and 80%. The availability studies. Searches were updated to December 2014. of alternative treatment options is therefore of the ut­ most importance. 4 Mandibular advancement devices Screening and study selection (MADs), which hold the mandible forward with the aim In the first phase of selection, duplicates were removed of preventing collapse of the upper airway during sleep and irrelevant articles were excluded by reviewing the provide a less invasive, more comfortable, and less costly titles and abstracts from the search results. In the next treatment alternative for patients with mild to moderate phase, the full texts of potentially relevant papers were OSA who do not tolerate, do not respond to, or are not evaluated to determine if they met the eligibility criteria. appropriate candidates for treatment with nCPAP, or The inclusion criteria were: those who fail behavioral measures such as weight loss Type of study: Randomized or non-randomized or sleep position change. MADs can also be used in pa­ controlled trials, cohort or case-control studies tients with severe OSA who fail treatment attempts with (with a minimum sample size of 10 patients in nCPAP or who are not appropriate candidates for upper each group) addressing the research question of airway surgery.5,6 the predictive value of cephalometric analysis in Patients undergo a repeat sleep study with MAD in oral appliance treatment outcomes in adult OSA situ to determine its effectiveness, which usually ran­ patients. Studies had to be published in English in ges between 42% and 65%.7 Clinical prediction of MAD an international peer-reviewed literature treatment outcome would allow advanced selection of Population: Male or female adult patients (≥ 18 suitable candidates for this treatment before manu­ years old) with a polysomnographic diagnosis of facturing the device, thus avoiding inappropriate delays OSA (i.e., 5 or more respiratory events [apneas or in therapy and waste of resources. Accordingly, this topic hypopneas] per hour of sleep) has been defined by the American Academy of Sleep Intervention: Treatment with any MAD for OSA Medicine as an important area for future research.5,6 The exclusion criteria were: Previous studies have suggested that lateral cepha­ Lack of a clear description of inclusion/exclusion lometry can identify craniofacial characteristics that criteria could have an impact on treatment response.5,8-10 Cepha­ Previous and/or current surgical or pharmacological lometry is a low-cost, simple, and widely available interventions radio­graphic technique, and it is therefore suitable as Limitation to severe OSA patients a screen­ing procedure. Nevertheless, the clinical utility Two reviewers (DRI, SIP) independently screened paper of cephalometric measurements in the prediction of titles and abstracts, with access to full texts where MAD treatment outcomes in OSA remains controversial. necessary to select studies into the review. Disagreements A recently published review by Saffer et al.11 found no were resolved by discussion. Where resolution was not clear predictors of MAD treatment success. However, possible, a third reviewer (GAB) was consulted. The cephalometric and anatomical factors were not inves­ selected studies underwent data extraction and quality tigated because no randomized controlled trials add­ assessment. ressing this issue were included in the review. The aim of this study was to fill this gap by conducting a systematic Data extraction and quality assessment of selected review of published studies examining the ability of studies cephalometric parameters to predict MAD treatment Two review authors (DRI, SIP) independently performed www.e-kjo.org http://dx.doi.org/10.4041/kjod.2015.45.6.308 309  Alessandri-Bonetti et al Cephalometric predictors of MADs treatment outcome the data extraction. Extracted data included: first author, selected studies are listed in Table 3. The landmarks and year of publication, study design, sample size, patient reference lines necessary to define these parameters are demographic and clinical characteristics, MAD type, shown in Figure 2. degree of protrusion and vertical opening, time interval between polysomnographic evaluations, outcome (i.e., Cephalometric skeletal measurements cephalometric parameters that differed significantly between good and poor responders). Selected studies Cranial base were assigned to a class of evidence according to the 1-piece MAD: Svanholt et al. 24 found that the cla­ssification of study designs by Jovell and Navarro- distance between the sella turcica and the deepest Rubio (Table 1).12 point in the posterior cranial fossa was lower in the group that responded positively to MAD therapy. Data synthesis 2-piece MAD: Both cranial base and anterior cranial A narrative synthesis was carried out. Data were sorted base lengths were reported to be non-predictive of according to MAD type (1-piece or 2-piece appliances). MAD treatment outcome.5,8,22 Two studies addressed Given the lack of homogeneity in the study settings, a the predictive value of cranial base angle with con­ quantitative synthesis seemed inappropriate. Therefore trasting results; one found an increased cranial base no meta-analysis was performed. angulation to be predictive of MAD treatment su­ ccess22 and the other did not.5 RESULTS Sagittal jaw relationship The electronic database search and the review of the 1-piece MAD: Data concerning sagittal jaw rela­ relevant publication reference lists yielded 939 po­ tionship were controversial. SNA and ANB were tentially relevant titles and abstracts after duplicates recognized either as predictive of treatment success were removed from a total of 1,034 records. Following when decreased10,24 or as non-predictive10,14,18 of the first phase of evaluation, 907 publications were re­ treatment outcome. SNB was found to be non-pre­ jected based on the title and the abstract. One further dictive14,18 or predictive of treatment success when study was excluded because full text was not obtained increased10 or decreased.24 by searching paper and digital format sources, nor after 2-piece MAD: The majority of papers indicated that attempting to contact the authors by e-mail corres­ sagittal jaw relationship cephalometric parameters pondence.13 In the second phase, analysis of the full were not suitable for predicting treatment outcome text of the remaining 31 studies led to the exclusion of (SNA,5,8,9,19-22 SNB,5,19-22 ANB,5,17,20-22 Wits appraisal20). 16 additional publications. Fifteen studies were there­ However, a decreased SNB value8,9 or an increased fore selected for the systematic review. 5,8-10,14-24 The ANB value8 were occasionally reported as predictive relevant data from each study are reported in Table 2. of treatment success. The PRISMA flow diagram (Figure 1) shows the number of articles reviewed in each phase of this systematic Vertical craniofacial dimensions review.25 1-piece MAD: Vertical craniofacial dimension para­ Cephalometric variables that were analyzed in the meters were usually recognized as non-predictive of Table 1. Levels of scientific evidence Level of evidence Type of study Strength of evidence Level 1 Meta-analyses of randomized controlled trials Good Level 2 Large-sample randomized controlled trials Level 3 Small-sample randomized controlled trials Good to fair Level 4 Non-randomized controlled prospective trials Level 5 Non-randomized controlled retrospective trials Level 6 Cohort studies Fair Level 7 Case-control studies Level 8 Non-controlled clinical series, descriptive studies Poor Level 9 Anecdotes or case reports Derived from Jovell and Navarro-Rubio.12 310 http://dx.doi.org/10.4041/kjod.2015.45.6.308 www.e-kjo.org  www.e-kjo.org Table 2. Summary of the main characteristics of the 15 studies selected for the systematic review Subjects, n Design of Level of First author, year (% male) Age (yr) BMI (kg/m2) bAHI (events/h) Country Inclusion criteria the study evidence* [withdrawals] Endo, 200310 Prospective 7 103 (-) GR: 51.2 GR: 24.5 GR: 21.1 ± 19.5 Japan AHI ≥ 10; no previous surgical operations case-control study [-] PR: 51.1 PR: 24.7 PR: 22.3 ± 19.7 Hoekema, 20078 Prospective 7 51 (-) 49.2 ± 9.5 32.2 ± 6.1 38.0 ± 29.8 Netherlands Age > 20 years; AHI ≥ 15 case-control study or AHI ≥ 5+symptoms Kim, 201414 Retrospective 7 86 (88.4) 51.5 ± 9.8 GR:25.98 ± 3.15 GR: 19.79 ± 17.32 South Korea AHI > 5; no sleep-related medication or case-control study [-] PR: 26.17 ± 2.78 PR: 30.08 ± 23.28 previous oropharyngeal surgery Lee, 200915 Retrospective 7 50 (92) 50.2 ± 9.8 GR: 25.5 ± 2.5 AI: GR: 37.6 ± 28.4 South Korea AHI > 5 case-control study [-] (21-69) PR: 26.9 ± 2.6 AI: PR: 34.6 ± 17.4 Lee, 201016 Retrospective 7 76 (89.5) 51.7 ± 0.3 < 35 38.9 ± 19.7 South Korea AHI > 5 case-control study [-] (21-69) Liu, 200117 Prospective 7 47 (89.4) 49.1 29.61 ± 6.49 CR: 44.21 ± 11.57 USA Mild to severe OSAS; case-control study [-] (25-80) PaR: 42.46 ± 19.33 suitability for MAS treatment NR: 28.69 ± 8.80 Marklund, 199818 Prospective 7 32 (100) Median: 57 Median: 28 Median: 23 Sweden AHI of at least 15 in the supine and/ http://dx.doi.org/10.4041/kjod.2015.45.6.308 case-control study [-] (37-72) or the lateral position Mehta, 20015 Prospective 7 28 (67.9) 48 ± 9 29.4 ± 3.1 27 ± 17 Australia AHI > 10/h; symptoms of OSA ≥ 2; case-control study (35-73) suitability for MAS treatment; no sedatives Menn, 199619 Prospective 7 29 (95.7) 53 ± 11 29 ± 5 RDI: 37 ± 23 USA Age > 18 yrs; RDI ≥ 10; case-control study (37-69) suitability for MAS treatment Milano, 201320 Prospective 7 23 (87.0) 54.0 ± 11.1 27.8 ± 4.4 24.5 ± 13.9 Italy Age > 18 yrs; AHI > 5; Alessandri-Bonetti et al Cephalometric predictors of MADs treatment outcome case-control study [-] (20-67) at least 2 OSAS symptoms Mostafiz, 201121 Prospective 7 53 (79.2) 49.5 ± 11.8 28.7 ± 4.2 CR: 28.4 ± 14.4 Australia Age > 18 yrs ; AHI > 10/h, case-control study [-] PaR: 34.8 ± 15.1 suitability for MAStreatment NR: 40.6 ± 12.7 No other sleep or medical disorders Ng, 201222 Retrospective 7 72 (76.4) 49.0 ± 11 30.2 ± 5.6 GR: 26.8 ± 14.4 Australia AHI ≥ 10/h; symptoms of OSA ≥ 2; case-control study PR: 28.1 ± 19.3 suitability for MAS treatment Rose, 200223 Prospective 7 57 (89.5) 56.5 ± 7.3 26.4 ± 2.0 22.0 ± 12.2 Germany Mild to moderate OSA; case-control study [-] (35-73) suitability for MAS treatment Shen, 20129 Prospective 7 52 (92.3) 43.4 ± 10.3 24.8 ± 2.6 GR: 34.1 ± 15.1 Taiwan Age > 18 yrs; AHI > 10; case-control study [-] (18-) PR: 38.6 ± 19.1 suitability for MAS treatment Svanholt, 201424 Prospective 7 27 (85.2) GR: 53.66 ± 7.27 GR: 27.13 ± 2.35 GR: 28.46 ± 15.17 Denmark AHI > 5 case-control study [-] PR: 52.02 ± 10.5 PR: 32.16 ± 5.42 PR: 43.52 ± 25.64 311  312 Table 2. Continued MAD type (protrusion) First author, year Follow up Success criteria Outcome (responders) [vertical opening] Endo, 200310 Monobloc type MAD (70 % of max protrusion) Within AHI red > 50% Larger SNB, lower ANB, shorter MP-H, [2 mm] 3 months shorter ANS- H, shorter ANS-Eb Hoekema, 20078 Thornton adjustable positioner type-1 (individual) 2/3 months AHI < 5/h or (AHI red > 50%) Higher ANB, smaller SNB, [-] and AHI < 20/h and no symptoms larger OJ, OB and N-ANS Kim, 201414 Monobloc type MAD (60 % of max protrusion) 3 months AHI red > 50% and AHI < 10/h Lower PNS-Go [minimum] Lee, 200915 Monobloc type MAD (60% of max protrusion) At least AHI red > 50% and AHI < 20/h No cephalometric predictors of [without open bites] 3 months treatment outcome Lee, 201016 Monobloc type MAD (60% of max protrusion) At least AHI red > 50% and AHI < 20/h Lower PNS-P [without open bites] 3 months Liu, 200117 MAD (two pieces) (2/3 max protrusion [for a start]) - GR: AHI red > 75%; MR: AHI red: 25% to 75%; Larger ratio of the vertical airway [minimum] PR: AHI red < 25% length to the cross-sectional area of the soft palate Marklund, 199818 MAD (one piece) (4–6 mm [for a start]) Up to AHI < 15 Normal (or low) SN-MP and [5 mm between incisors] 2 months lower ANS-Me Mehta, 20015 MAD (two pieces) (individual) 19.7 ± 8.8 Resolution of symptoms and AHI < 5/h Larger RPAS and larger SN-MP [3/4 mm] weeks Menn, 199619 MAD (two pieces) (max protrusion - 3 mm) Mean: RDI red ≥ 50% and RDI ≤ 20 No cephalometric predictors of [5-7 mm] 104 days treatment outcome Milano, 201320 Silensor (two pieces) (50/60% for a start) 2-3 months AHI < 5/h Lower MP-H and lower SN-MP < 29° [-] Mostafiz, 201121 SomnoDent (two pieces) (maximum comfortable limit of mandibular protrusion) 6-8 weeks GR: AHI red > 50%, AHI > 5/h Greater tongue/oral enclosure CSA ratio [-] MR: AHI red > 50%, AHI > 5/h; PR: AHI red < 50% Ng, 201222 SomnoMed (two pieces) (maximum comfortable limit of mandibular advancement) 6 weeks AHI red ≥ 50% (alone or in addition to Shorter PNS-P, increased Ba-SN; [-] AHI < 20/ < 10/ < 5/h) larger SNMP Rose, 200223 Karwetzky’s modified activator (two pieces) (individual) 6-12 weeks Age > 60: AHI < 5/h and AI < 2/h; Lower SN-MP; larger SGo:NMe; [individual] age < 60: AHI < 10/h and AI < 5/h. higher H-MP, lower H-Me Shen, 20129 IST-Appliance (two pieces) (individual) 6 weeks AHI red > 50% and AHI < 10/h Smaller SNB ; shorter N-Me; [-] narrower MinRGA Svanholt, 201424 Monobloc type MAD (75 % of max protrusion) 4 weeks AHI red > 75% Lower SNA, lower SNB, [-] lower SNPg, lower S-D http://dx.doi.org/10.4041/kjod.2015.45.6.308 Values are presented as number only, number (%), or mean ± standard deviation (range). *According to the classification of study designs by Jovell and Navarro-Rubio.12 n, Number; -, not reported; GR, good responders; PR, poor responders; CR, complete responders; PaR, partial responders; NR, nonresponders; BMI, body mass index; AHI, apnea hypopnea index; bAHI, baseline AHI; h, hour; AI, apnea index; RDI, respiratory disturbance index; AHI red, AHI reduction; MAD, mandibular advancement device. Suitability for MAD treatment, good dental health, > 10 teeth/dental arch, no periodontal disease, TMJ dysfunction or exaggerated gag reflex. See Table 3 for the definition of cephalometric variables. Alessandri-Bonetti et al Cephalometric predictors of MADs treatment outcome www.e-kjo.org Alessandri-Bonetti et al Cephalometric predictors of MADs treatment outcome Identification Records identified through databases searching (n = 1,027) Additional records identified Pubmed Scopus Google Scholar Cochrane Library through other sources (n = 7) (n = 46) (n = 96) (n = 878) (n = 7) Records after duplicates removed (n = 939) Screening Records screened Records excluded (n = 907) (n = 939) Full-text article not available (n = 1) Eligibility Full-text articles assessed Full-text articles excluded, for eligibility (n = 31) with reasons (n = 16): Unsuitable sample (n = 7) Inadequate methodology (n = 5) Not published in the English Included language (n = 3) No pertinent/useful data (n = 1) Studies included in Figure 1. Flow chart of lite­ qualitative synthesis rature search and study selec­ (n = 15) tion. treatment outcome, 14,24 with the exception of treatment outcome. lower anterior face height, 18 lower posterior face 2-piece MAD: There was general agreement that height,14 and mandibular plane angle,18 which were maxillary and mandibular lengths, ramus height, occasionally identified as predictive of treatment and corpus length were not suitable for prediction success when decreased. of MAD treatment outcome.5,17,20-22 2-piece MAD: Saddle angle, articular angle, gonial angle, palatal plane angle, posterior face height, Hard palate and the ratio between upper anterior face height 1-piece MAD: Hard palate length did not appear to and lower anterior face height were not useful be predictive of MAD treatment outcome.10,16 for predicting MAD treatment outcomes.9,20-23 The 2-piece MAD: Ng et al.22 claimed that hard palate majority of the papers ascribed a non-predictive length was not predictive of MAD treatment out­ role to anterior face height, upper anterior face come. height, lower anterior face height, and the ratio between posterior face height and anterior face Hyoid bone height, but a lower anterior face height value9 as 1-piece MAD: Endo et al.10 found decreased values well as higher values of upper anterior face hei­ of MP-H, H-Me, and ANS-H to be predictive of MAD ght8 and of the ratio of posterior face height to treatment success, whereas Kim et al.14 supported a anterior face height23 were also occasionally found non-predictive role of H-Me and ANS-H. in good responders. Data on the predictive role of 2-piece MAD: There was general agreement that mandibular plane angle were conflicting: 3 out of 7 H-RGN, C3ia-H, H-Go, Go-H-Me were not suitable studies found it to be non-predictive of treatment for predicting MAD treatment outcomes.9,17,20,21 Rose outcomes, 8,9,21 2 studies indicated an increased et al.23 found that decreases in H-Me and ANS-H value as a predictor of treatment success,5,22 and 2 were predictive of treatment success, while Hoekema studies reported a decreased value as a predictor of et al.8 did not. Data on MP-H were conflicting. Five treatment success.20,23 out of 7 studies found it to be non-predictive,5,8,9,19,21 while the two others reported that decreased MP-H20 Maxillary and mandibular lengths and increased MP-H23 were predictive of treatment 1-piece MAD: Endo et al. 10 found that maxillary success. and mandibular lengths were not predictive of MAD www.e-kjo.org http://dx.doi.org/10.4041/kjod.2015.45.6.308 313  Alessandri-Bonetti et al Cephalometric predictors of MADs treatment outcome Table 3. Summary of cephalometric variables Variable Measurement Cephalometric skeletal measurements Cranial base Cranial base length (mm) Ba-S-N Anterior cranial base length (mm) S-N o Cranial base angle ( ) BaSN Distance between sella turcica and the deepest point S-D in posterior cranial fossa (mm) Sagittal jaw relationship Anteroposterior position of the maxilla (o) SNA o Anteroposterior position of the mandible ( ) SNB SN-Pg Anteroposterior relationship between maxilla ANB and mandible (o, mm) Wits appraisal (distance between A and B projections onto the occlusal plane) Vertical craniofacial dimensions Anterior face height (mm) N-Me Upper anterior face height (mm) N-ANS Lower anterior face height (mm) ANS-Me Posterior face height (mm) S-Go Lower posterior face height (mm) PNS-Go Posterior face height : anterior face height (ratio) S-Go : N-Me Upper anterior face height : lower anterior face height (ratio) N-ANS : ANS-Me Saddle angle (o) N-S-Ar Articular angle (o) S-Ar-Go o Gonial angle ( ) Ar-Go-Me o Mandibular plane angle ( ) SN-MP FP-MP PP-MP o Palatal plane angle ( ) SN-PP Maxillary and mandibular lengths Maxillary length (mm) Cd-A Mandibular length (mm) Cd-Gn Ramus height (mm) Ar-Go Cd-Go Corpus length (mm) Go-Me Go-Gn Hard palate Length (mm) ANS-PNS Hyoid bone Vertical position (mm) H-MP or Hy-MP (perpendicular) H-Go (vertical measure) 314 http://dx.doi.org/10.4041/kjod.2015.45.6.308 www.e-kjo.org Alessandri-Bonetti et al Cephalometric predictors of MADs treatment outcome Table 3. Continued Variable Measurement Horizontal anterior position (mm) H-RGN H-ANS (horizontal measure) H-Me or Hy-Me Horizontal posterior position (mm) H-C3ia o Hyoid angle ( ) Go-H-Me Cervical vertebrae Linear distance between RGN and C3 (mm) RGN-C3ia o Craniocervical angle ( ) C2C4-SN C2si-S-N Cephalometric dental measurements Maxillary incisor inclination (o) U1 (U1i-root apex)-PP U1 (U1i-root apex)-SN o Mandibular incisor inclination ( ) L1 (L1i-root apex)-MP o Interincisor angle ( ) IIAA: U1 (U1i-root apex)-L1(L1i-root apex) Maxillary molar inclination (o) U6 (U6c-mesial root apex)-SN U6 (U6c-mesial root apex)-PP o Mandibular molar inclination ( ) L6 (L6c-mesial root apex)-MP Maxillary molar height (mm) MXMH (U6c-PP perpendicular) U6c-FP (perpendicular) Mandibular molar height (mm) MDMH (L6c-PP perpendicular) Maxillary molar distance to mandibular molar (mm) U6L6SN (U6c-L6c projected on S-N plane) Overjet (mm) Horizontal distance of Uli to Lli Overbite (mm) Vertical distance of Uli to Lli Cephalometric soft tissue measurements Soft palate Length (mm) PNS-P or PNS-Ut Thickness (mm) Maximum thickness of the soft palate (perpendicular to PNS-P) Depth (mm) PNS-P (horizontal) Cross-sectional area (mm2) SPXA (area confined by outline of soft palate through PNS and P) o Palatal angle ( ) ANS-PNS-P Tongue Tongue length (mm) TGL (TT-Eb) Tongue height (mm) TGH (maximum height perpendicular to TT-Eb) Tongue cross sectional area (mm2) Tongue CSA or TGXA or TNGXA (region within the outline of the dorsum of the tongue surface and lines that connect TT, RGN, H and Eb). TA (Triangle area constructed TGL as the base and TGH as the height) Tongue : oral cross sectional area (ratio) Tongue CSA : Oral CSA Epiglottis Linear distance between ANS and Eb (mm) ANS-Eb www.e-kjo.org http://dx.doi.org/10.4041/kjod.2015.45.6.308 315  Alessandri-Bonetti et al Cephalometric predictors of MADs treatment outcome Table 3. Continued Variable Measurement Upper Airway Width (mm) At the level of soft palate SPAS or SP-PAS or Superior Posterior (along parallel line to Go-B line) MAS or middle airway (along parallel line to Go-B line through P) MinRPA or narrowest palatal airway (minimal width perpendicular to posterior pharyngeal wall) RPAS (Phw-Spt) At the level of tongue base PAS or IAS or IAS1 (distance between posterior pharyngeal wall and the dorsal base of tongue surface, measured on a line intersecting Go and B Point) Posterior Inferior (distance between base of tongue and posterior pharyngeal wall) MinRGA or narrowest lingual airway (minimal width perpendicular to posterior pharyngeal wall) TB-PAS (at the level of the tongue base) PPW’-BT’ Through C3 IAS2 (along parallel line to Go-B line) Vertical length (mm) VAL (PNS-Eb) 2 Cross-sectional area (mm ) Nasopharynx cross-sectional area NASOXA (area outlined by line between R and PNS, extension of palatal plane to posterior pharyngeal wall and posterior pharyngeal wall) Oropharynx cross-sectional area OROXA (area outlined by inferior border of nasopharynx, posterior surface of soft palate, line parallel to palatal plane from point P to dorsal surface of tongue, posterior inferior surface of tongue, line parallel to palate plane through point Et, and posterior pharyngeal wall) Cross-sectional area (mm2) Hypopharynx cross-sectional area HYPOXA (area outlined by inferior border of oropharynx, posterior surface of epiglottis, line parallel to palatal plane through point C4, and posterior pharyngeal wall) Pharinx cross-sectional area PHYNXA (sum of NASOXA, OROXA and HYPOXA) Facial contours Facial convexity (o) N’-Prn-Pog’ Prominence of the nose (mm) Prn-S (vertical) Upper lip position (mm) Distance UL-E line Lower lip position (mm) Distance LL-E line See Figure 2 for definitions of landmarks and reference lines used in this table. Cervical vertebrae seemed to be predictive of MAD treatment outcome. 1-piece MAD: Svanholt et al.24 found that the pre­ valence of and severity of morphological deviations Cephalometric dental measurements of the upper spine were greater in the group of 1-piece MAD: Studies investigating dental para­ patients that failed MAD. meters agreed on the lack of predictive value of 2-piece MAD: Neither the linear distance between these measurements for treatment outcome.14,24 RGN and C3ia 21 nor the craniocervical angle 21,22 2-piece MAD: Overjet and overbite were widely 316 http://dx.doi.org/10.4041/kjod.2015.45.6.308 www.e-kjo.org Alessandri-Bonetti et al Cephalometric predictors of MADs treatment outcome outcome14,15 and predictive of treatment success, were controversial. 2-piece MAD: Shen recognized soft palate thickness S N N' SN as non-predictive of treatment outcome.9 The role of soft palate length was controversial. Two out of E line Po Cd R 6 studies found that decreased soft palate length Iop was predictive of treatment success9,22 while 4 did Or FP Ar Prn D Ba PNS ANS not.8,19-21 PP C2 Spt A Phw U6c Lli UL Tongue TT OP L6c LL PPW' P (Ut) Uli Go Et BT' B 1-piece MAD: There was general agreement that RGN C3ia Eb Gn Pog Pog' the cephalometric variables of tongue length, H (Hy) Me height, and cross-sectional area were not useful for C4 MP predicting MAD treatment outcomes.10,14,15 Figure 2. Diagrammatic representation of landmarks 2-piece MAD: Tongue-related cephalometric and reference lines. A, Subspinale; ANS, anterior nasal variables were widely recognized as non-predictive spine; Ar, articulare; B, supramentale; Ba, basion; BT’, of MAD treatment outcome9,17,22,23 with the sole base of tongue; C2, tangent point on the dorsal surface exception of tongue/oral enclosure cross-sectional of C2 vertebra to a line from C4; C3ia, C3 vertebra infe­ ratio, which Mostafiz et al.21 found was increased in roanterior; C4, C4 vertebra inferoposterior; Cd, condylion; complete responders. D, the deepest point in posterior cranial fossa; E line, Ricketts-E line; Eb, epiglottis base; Et, tip of epiglottis; Epiglottis FP, Frankfurt Plane; Gn, gnathion; Go, gonion; H (Hy), 1-piece MAD: Only 1 study examined the predictive hyoidale; Iop: internal occipital protuberance; L1i, lower role of the distance between ANS and Eb in incisor tip; LL, lower lip; Me, menton; MP, mandibular treatment outcomes, finding that the greater this plane; N, nasion; N’, soft tissue nasion; OP, occlusal distance, the less effective the treatment.10 plane; Or, orbitale; P (Ut), soft-palate tip; Phw, posterior 2-piece MAD: No study investigated the predictive pharyngeal wall; PPW’, posterior pharyngeal wall role of epiglottis parameters. interseption; PNS, posterior nasal spine; Po, porion; Pog, pogonion; Pog’, soft tissue pogonion; PP, palatal plane; Upper airway Prn, nasal tip; R, roof of pharynx; RGN, retrognathion; S, 1-piece MAD: Upper airway parameters were unani­ sella; SN, S-N plane; Spt, tangent point on a line parallel mously recognized as non-predictive of MAD treat­ to PNS-P on the dorsal surface of the soft palate at the ment outcome.10,14,16 maximum width; TT, tongue tip; U6c, maxillary first molar 2-piece MAD: Neither vertical length nor cross- point; U1i, upper incisor tip; UL, upper lip. sectional area were found to be predictive of out­ come.9,17,20 Concerning the upper airway widths, only 1 study (Shen et al.9) found significantly decreased re­c ognized as non-predictive of treatment retroglossal width in good responders. The majority success.9,17,20,22,23 However, Hoekema et al.8 fo­und showed a non-predictive role for retropalatal width, that increased overjet and overbite were progno­ although Mehta et al.5 and Liu et al.17 found an stically favorable. Increased maxillary molar height increased retropalatal space in good and poor res­ also seemed to be associated with a better chance ponders, respectively. of successful treatment.17 None of the other cepha­ lometric dental measurements exhibited predictive Facial contours value.21,23 1-piece MAD: Kim et al.14 found no differences in profile measurements between good and poor res­ Cephalometric soft tissue measurements ponders. 2-piece MAD: No profile measurements were found Soft palate to be predictors for treatment outcome with 2-piece 1-piece MAD: Soft palate depth and thickness and MAD.20 palatal angle were identified as non-predictive of treatment outcomes,10,14 whereas data concerning DISCUSSION soft palate length, which, when decreased,16 was recognized both as non-predictive of treatment MADs are increasingly used for treatment of mild to www.e-kjo.org http://dx.doi.org/10.4041/kjod.2015.45.6.308 317  Alessandri-Bonetti et al Cephalometric predictors of MADs treatment outcome moderate OSA because they provide a less invasive, predictive role of upper spine morphological deviations more comfortable, and less costly alternative to nCPAP. for 1-piece MAD was suggested. Further study will be Nevertheless, MADs are not as efficacious as nCPAP, and required to evaluate these findings. the treatment success rates range from 42% to 65%.7 With regard to dental cephalometric parameters, overjet As certain craniofacial characteristics, including reduced and overbite, although being reported as predictive of posterior airway space, abnormally long soft palate, and MAD treatment outcome by Hoekema et al. 8 alone, low position of the hyoid, are commonly found in OSA were mostly considered unlikely to carry any prognostic patients it seems reasonable to assume that the efficacy significance. Furthermore Liu et al. 17 found that in of MAD may relate to morphological factors.26 2-piece MADs, efficacy was decreased with greater Several modalities for assessing upper airway eruption of the maxillary molars. However, being as this morphology have been recommended, including mag­ study was the only one examining this parameter, the netic resonance, nasopharyngoscopy (in an awake state recognition of a predictive role for the distance between or during drug-induced “sleep”), computed tomography, the maxillary first molar and Frankfort plane seems and lateral cephalometry.8,9,27-29 Cephalometery is low- premature. Therefore, further research on this parameter cost, simple, and widely available, and these advantages is also warranted. may offset the disadvantages of the cephalogram being Cephalometric soft tissue measurements did not a 2-dimensional projection of a 3-dimensional structure seem to be very useful in predicting MAD treatment that is performed in an awake state and in an upright response. Soft palate length was recognized both as position whereas the pathology of OSA arises with the predictive of treatment success when decreased and patient lying down during sleep. Therefore, we aimed to as not significantly different between good and poor provide a systematic review of cephalometric parameters responders. Therefore, prediction of treatment outcome predictive of MAD treatment outcome. based on this measurement is not yet unjustified and will require further investigation. Greater ratio between Summary of the evidence tongue and oral cross sectional area 21 and shorter Within the limitations of the selected studies, some distance between Eb and ANS10 have been shown to con­sideration can be made with regard to cephalometric be associated with treatment success in 2- and 1-piece para­meters predictive of MAD treatment response. Data MAD, respectively; however additional studies will be on the skeletal cephalometric measurements were con­ needed to confirm these results. Although some studies flicting. Among the cranial base cephalometric values, have suggested a predictive role for retropalatal and cranial base angle and the distance between the sella retroglossal widths in treatment with 2-piece MAD, the turcica for 2-piece MADs and the deepest point in the results were controversial. The inconsistency of predictive posterior cranial fossa for 1-piece MADs were suggested values of upper airway cephalometric parameters was as possible prognostic factors. However, the data are still foreseeable because cephalometry is not the preferred too limited to draw any conclusion. imaging technique for evaluation of these anatomic The lack of predictive value of sagittal jaw relationships structures. Indeed, there appears to be a strong linear (SNA, SNB, SN-Pg, ANB, Wits appraisal) as well as li­ relationship between 2-dimensional cephalometric and near parameters related to vertical jaw dimensions (an­ 3-dimensional computed tomography reconstructions terior face height, lower anterior face height, lower for tongue, soft palate, and nasopharynx, while the con­ po­sterior face height, upper anterior face height, ratio figuration of oropharynx and hypopharynx appears to between posterior face height and anterior face height) be less consistent.30 was almost unanimously confirmed. On the other ha­ nd, the mandibular plane angle was associated with Limitations widely conflicting results, with some studies reporting a non-predictive role and other studies reporting Study level both increased and decreased predictive values for The question of the predictive role of cephalometric treatment success. It is therefore impossible to draw parameters has not been addressed by randomized any definite conclusion on this parameter based on the clinical trials because the variables of interest are not available data. Hyoid bone position also showed widely under the control of the investigators. The inclusion conflicting results, with some studies reporting a non- of observational studies therefore appeared necessary. predictive role for MP-H and others reporting both The study design adopted in all of the studies selected increased and decreased predictive value for treatment for this review was case-control (either prospective and success. Available data for H-Me and ANS-H were ra­ retrospective) rather than cohort: 2 or 3 groups differing ther limited. Therefore, the prognostic value of hyoid in MAD treatment outcome (good and poor responders bone parameters needs further investigation. A possible or nonresponders, partial and complete responders) were 318 http://dx.doi.org/10.4041/kjod.2015.45.6.308 www.e-kjo.org Alessandri-Bonetti et al Cephalometric predictors of MADs treatment outcome prospectively or retrospectively identified and compared the earlier studies evaluated the respiratory disturbance on the basis of cephalometric parameters. As a result, index (RDI).19 The definition of sleep apnea was based the findings score low in terms of strength of evidence. either on AHI > 5, AHI ≥ or > 10, AHI ≥ or > 15, or RDI A selection bias may have occurred because the sample ≥ 10. Success criteria included RDI reduction of ≥ 50% was not randomly selected. Moreover in some of the and RDI ≤ 20, AHI < 5, AHI < 15, or AHI reduction of prospective studies patients were included after they > 50% alone or in association with AHI < 20, < 10, or had been allowed to choose whether or not they wanted < 5. A certain degree of variability was present with to participate or after they had decided entirely for respect to the type of MAD (monobloc or two pieces), themselves not to accept treatment with nCPAP (self- as well as to the amount of protrusion (from 50 or selection bias).18,19 Finally, a few studies reported that 60% of maximum protrusion to the most advanced some patients dropped out (attrition bias).5,8,19 position without causing any discomfort) or vertical MAD treatment success has been associated with fe­ opening (ranging from “minimum” to 15 mm). The male sex, a younger age, a lower body mass index time interval between diagnostic and with-appliance (BMI), and a lower baseline apnea-hypopnea index poly­somnography ranged from few weeks to several (AHI) but, due to the design of the studies included in months. The cephalometric parameters also varied sub­ this review, these baseline characteristics could not be stantially between different studies. Because of the lack properly addressed and may actually show systematic of homogeneity among the studies, any attempt to differences between good responders and poor res­ summarize the study results seemed unjustified. This, a ponders (confounding bias).17,22 In the 13 studies that narrative synthesis was carried out without performing a reported male-female ratio, males had considerably more meta-analysis. representation, ranging from 67.9% to 100% of the No “grey literature” or articles in non-English langu­ sample.5,9,14-24 With the exception of the study by Mehta ages were included in the present review. Therefore the et al.,5 none of the others investigated the differences review is also subject to publication bias and language in sex distribution between good responders and poor bias. responders. Six studies found no age difference bet­ ween good and poor responders, 5,10,14,15,19,24 while 4 CONCLUSION studies found age to be significantly lower in good res­ ponders.17,20-22 Eight studies failed to find a predictive This review found controversial and limited data on the role for BMI, 5,10,14,15,19,20,22,23 while 3 studies reported predictive role of certain cephalometric parameters for lower BMI in good responders.8,21,24 Only 1 paper add­ MAD treatment outcome. Therefore selection criteria for ressed the issue of weight gain or loss between the ini­ suitable candidates for MAD treatment by cephalometric tial and the final evaluations.19 Six out of 14 studies analysis are currently inadequate. Although no definitive did not account for potential differences in baseline clinical recommendations can be made, this systematic AHI,8,16,18-20,23 whereas comparison between good and review highlights the methodological weaknesses of the poor responders in terms of AHI or apnea index led to currently available studies, analyzes the possible reasons significant differences in 2 studies, with lower valu­es for their discordant results, and encourages and guides associated with treatment success,14,21 and no significant further research in this field. Prospective cohort studies differences in 6 studies.5,9,10,15,17,22 in large samples with cephalometric prediction made With regard to observer bias, only 3 studies reported prior to the MAD construction are required to clarify that the operator who performed the cephalometric the clinical utility of lateral cephalometry for adult OSA analysis or the polysomnography was blinded to the patients who are considering oral appliances as a thera­ treatment outcome.5,14,22 None of the selected studies peutic option. included a power analysis or a sample size calculation. The presence of potentially underpowered studies could REFERENCES lead to inconsistent or misleading results, because they may find no significant difference between study sam­ 1. Epstein LJ, Kristo D, Strollo PJ Jr, Friedman N, ples even in the presence of a real difference in the ge­ Mal­hotra A, Patil SP, et al; Adult Obstructive Sleep neral population. Apnea Task Force of the American Academy of Sleep Medicine. Clinical guideline for the evaluation, ma­ Review level na­gement and long-term care of obstructive sleep The findings of the present systematic review were also apnea in adults. J Clin Sleep Med 2009;5:263-76. limited by the heterogeneity in the methodology of the 2. Skomro RP, Kryger MH. Clinical presentations of ob­ included studies. Respiration was analyzed in terms of structive sleep apnea syndrome. Prog Cardiovasc Dis AHI in the majority of the investigations, but some of 1999;41:331-40. www.e-kjo.org http://dx.doi.org/10.4041/kjod.2015.45.6.308 319  Alessandri-Bonetti et al Cephalometric predictors of MADs treatment outcome 3. Sullivan CE, Issa FG, Berthon-Jones M, Eves L. Re­ et al. Determinants of treatment outcome after use versal of obstructive sleep apnoea by continuous of the mandibular advancement device in patients positive airway pressure applied through the nares. with obstructive sleep apnea. Arch Otolaryngol Head Lancet 1981;1:862-5. Neck Surg 2010;136:677-81. 4. Waldhorn RE, Herrick TW, Nguyen MC, O'Donnell 17. Liu Y, Lowe AA, Fleetham JA, Park YC. Cephalo­ AE, Sodero J, Potolicchio SJ. Long-term compliance metric and physiologic predictors of the efficacy of with nasal continuous positive airway pressure the­ an adjustable oral appliance for treating obstructive rapy of obstructive sleep apnea. Chest 1990;97:33-8. sleep apnea. Am J Orthod Dentofacial Orthop 2001; 5. Mehta A, Qian J, Petocz P, Darendeliler MA, Cistulli 120:639-47. PA. A randomized, controlled study of a mandibular 18. Marklund M, Franklin KA, Stenlund H, Persson M. advancement splint for obstructive sleep apnea. Am Mandibular morphology and the efficacy of a man­ J Respir Crit Care Med 2001;163:1457-61. dibular advancement device in patients with sleep 6. Kushida CA, Morgenthaler TI, Littner MR, Alessi CA, apnoea. Eur J Oral Sci 1998;106:914-21. Bailey D, Coleman J Jr, et al; American Academy 19. Menn SJ, Loube DI, Morgan TD, Mitler MM, Berger of Sleep. Practice parameters for the treatment of JS, Erman MK. The mandibular repositioning device: snoring and Obstructive Sleep Apnea with oral app­ role in the treatment of obstructive sleep apnea. liances: an update for 2005. Sleep 2006;29:240-3. Sleep 1996;19:794-800. 7. Ferguson KA, Cartwright R, Rogers R, Schmidt- 20. Milano F, Billi MC, Marra F, Sorrenti G, Gracco A, Nowara W. Oral appliances for snoring and obstruc­ Bonetti GA. Factors associated with the efficacy of tive sleep apnea: a review. Sleep 2006;29:244-62. mandibular advancing device treatment in adult 8. Hoekema A, Doff MH, de Bont LG, van der Hoeven OSA patients. Int Orthod 2013;11:278-89. JH, Wijkstra PJ, Pasma HR, et al. Predictors of ob­ 21. Mostafiz W, Dalci O, Sutherland K, Malhotra A, structive sleep apnea-hypopnea treatment outcome. Srinivasan V, Darendeliler MA, et al. Influence of J Dent Res 2007;86:1181-6. oral and craniofacial dimensions on mandibular ad­ 9. Shen HL, Wen YW, Chen NH, Liao YF. Craniofacial vancement splint treatment outcome in patients with morphologic predictors of oral appliance outcomes obstructive sleep apnea. Chest 2011;139:1331-9. in patients with obstructive sleep apnea. J Am Dent 22. Ng AT, Darendeliler MA, Petocz P, Cistulli PA. Ce­ Assoc 2012;143:1209-17. phalometry and prediction of oral appliance treat­ 10. Endo S, Mataki S, Kurosaki N. Cephalometric evalu­ ment outcome. Sleep Breath 2012;16:47-58. ation of craniofacial and upper airway structures in 23. Rose E, Lehner M, Staats R, Jonas IE. Cephalometric Japanese patients with obstructive sleep apnea. J analysis in patients with obstructive sleep apnea. Med Dent Sci 2003;50:109-20. Part II: Prognostic value in treatment with a mandi­ 11. Saffer F, Lubianca JF, Rösing C, Dias C, Closs L. bular advancement device. J Orofac Orthop 2002; Predictors of success in the treatment of obstruc­ 63:315-24. tive sleep apnea syndrome with mandibular reposi­ 24. Svanholt P, Petri N, Wildschiødtz G, Sonnesen L. tioning appliance: a systematic review. Int Arch Influence of craniofacial and upper spine morpho­ Otorhinolaryngol 2015;19:80-5. logy on mandibular advancement device treatment 12. Jovell AJ, Navarro-Rubio MD. Evaluation of outcome in patients with obstructive sleep apnoea: scientific evidence. Med Clin (Barc) 1995;105:740-3. a pilot study. Eur J Orthod 2015;37:391-7. 13. Liu Y, Lowe AA. Factors related to the efficacy of 25. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtz­ an adjustable oral appliance for the treatment of sche PC, Ioannidis JP, et al. The PRISMA statement obstructive sleep apnea. Chin J Dent Res 2000;3:15- for reporting systematic reviews and meta-analyses 23. of studies that evaluate healthcare interventions: 14. Kim YK, Kim JW, Yoon IY, Rhee CS, Lee CH, Yun explanation and elaboration. BMJ 2009;339:b2700. PY. Influencing factors on the effect of mandi­bular 26. Hoekema A, Hovinga B, Stegenga B, De Bont LG. advancement device in obstructive sleep apnea Craniofacial morphology and obstructive sleep ap­ patients: analysis on cephalometric and polysom­ noea: a cephalometric analysis. J Oral Rehabil 2003; nographic parameters. Sleep Breath 2014;18:305-11. 30:690-6. 15. Lee CH, Mo JH, Choi IJ, Lee HJ, Seo BS, Kim DY, 27. Sanner BM, Heise M, Knoben B, Machnick M, et al. The mandibular advancement device and pa­ Laufer U, Kikuth R, et al. MRI of the pharynx and tient selection in the treatment of obstructive sleep treatment efficacy of a mandibular advancement apnea. Arch Otolaryngol Head Neck Surg 2009; device in obstructive sleep apnoea syndrome. Eur 135:439-44. Respir J 2002;20:143-50. 16. Lee CH, Kim JW, Lee HJ, Seo BS, Yun PY, Kim DY, 28. Chan AS, Lee RW, Srinivasan VK, Darendeliler MA, 320 http://dx.doi.org/10.4041/kjod.2015.45.6.308 www.e-kjo.org Alessandri-Bonetti et al Cephalometric predictors of MADs treatment outcome Grunstein RR, Cistulli PA. Nasopharyngoscopic 2006;100:897-902. evaluation of oral appliance therapy for obstructive 30. Lowe AA. Can we predict the success of dental app­ sleep apnoea. Eur Respir J 2010;35:836-42. liance therapy for the treatment of obstructive sleep 29. Sam K, Lam B, Ooi CG, Cooke M, Ip MS. Effect of a apnea based on anatomic considerations? Sleep non-adjustable oral appliance on upper airway mor­ 1993;16(8 Suppl):S93-5. phology in obstructive sleep apnoea. Respir Med www.e-kjo.org http://dx.doi.org/10.4041/kjod.2015.45.6.308 321