Effect of the Mandibular Repositioning Appliance (MARA) on Posterior Airway Space (PAS)

Abstract

Aim of the study was to show the effect of skeletal Class II treatment with the mandibular anterior repositioning appliance (MARA) on the sagittal posterior airway space (PAS) diameter. A total of 53 patients were selected retrospectively: 26 male patients (median 13 years, age span 10–19 years) and 27 female patients (median 14 years, age span 11–47 years). All patients had lateral cephalograms taken at T1 (before MARA treatment) and at T2 (after MARA treatment). Average treatment took 13.1 ± 7.5 months (Group 1) and 10.5 ± 4.5 months (Group 2), respectively. The following PAS parameters were obtained at T1 and T2: TI (Tangent Point 1), Me/Gn (menton/gnathion), DW (dorsal wall). Additionally, Björk‘s sum angle, SNA, SNB and ANB were determined. The male patients showed a higher increase in the anteroposterior diameter of the PAS (+27.5%) compared to female patients (+11.6%). Male participants had a significantly higher PAS (p = 0.006) than female participants (p = 0.09). Although not significantly, Björk‘s sum angle decreased in both groups. In general, compared to female patients, male patients showed a greater decrease between T1 and T2. SNA and SNB exhibited varied behavior between T1 and T2, with some individuals reporting a decrease and others reporting an increase. SNA tended to decrease in general. In terms of ANB, the male participants displayed a decrease from T1 to T2. Treatment of a skeletal Class II malocclusion with the mandibular anterior repositioning appliance (MARA) caused an increase in the sagittal posterior airway space (PAS) diameter and, thereby, might be effective against obstructive sleep apnea.

1. Introduction

Clinical manifestations of mandibular hypoplasia range from mild esthetic discrepancies to functional imbalances and obstructive sleep apnea (OSA) [1].

The soft palate, the tongue, the hyoid and the associated soft tissues are directly and indirectly connected to the maxilla and mandibula. This means that the movements of both parts of the jaw directly cause both positional and tensile changes in those connected tissues. These changes in turn cause changes in the volume of the nasal cavity, the oral cavity and the PAS dimensions, depending on the direction and magnitude of the resulting force vector [2]. This is why one may expect ventral/anterior repositioning of the mandible to positively affect the sagittal PAS diameter. Cephalometrically, the PAS is defined as the metric distance between the root of the tongue (anterior margin) and the dorsal pharyngeal wall (posterior margin) and should be measured as an extension of the mandibular plane [3]. By definition, an obstruction is detected when this sagittal distance is smaller than 10 mm [3]. This, combined with the nocturnal phases of apnea during sleep, plus excessive daytime sleepiness and reversible cognitive degradation, indicates the presence of an obstructive sleep apnea syndrome (OSAS) [4]. OSAS is a risk factor for cardio-circulatory diseases, stress-related diseases and diabetes mellitus [5,6]. Newer research also suggests an increased risk for chronic inflammatory diseases (immuno-suppressive effects) and dementia, the latter of which seems to be related to a loss of function of the so-called glymphatic system [7,8,9,10,11]. A possible common denominator for all of these damaging effects is repeated nocturnal catecholamine- and cortisol-mediated emergency responses triggered by a drop in O₂ saturation [12,13]. In view of the potentially fatal outcome, especially in adults, OSAS is a condition in urgent need of adequate treatment [12].

Treatment with mandibular advancement devices (MADs), together with CPAP ventilation and orthognathic surgery, is the standard so far in OSAS cases [14]. Orthognathic surgery remains the only permanently curative option; however, it is highly invasive. Ventilation via CPAP (continuous positive airway pressure) or variants thereof are often poorly tolerated by some OSAS patients [15]. The search for a viable middle ground between those two has so far yielded MADs therapy. It has been established that the lateral cephalogram is a useful technique for confirming the possible significance of OSA in patients experiencing suspected symptoms [16]. They are part of the standard orthodontic diagnostic regime.

The mandibular anterior repositioning appliance (MARA) is a fixed functional appliance for Class II correction [17]. In this context, the MARA can be viewed as being analogous to MADs therapy. The MARA, however, produces permanent dentoalveolar and skeletal effects, as opposed to MAD treatment.

It could be hypothesized that—considering its effect—the MARA be situated somewhere between MADs and maxillofacial surgery. The MARA is significantly more invasive than MADs, which is only worn at night, whilst simultaneously avoiding the potential risks and complications associated with maxillofacial surgery. There are multiple studies on changes in the PAS lumen after orthognathic surgery, but there is also a growing number of studies focusing on fixed functional appliances. However, such studies are still outnumbered by the former [18,19].

The present study aimed to show the effect of a fixed functional appliance (mandibular anterior repositioning appliance, MARA) on the posterior airway space. Furthermore, this study aimed to investigate the effects of the MARA on the PAS, taking the sagittal PAS diameter obtained from lateral cephalograms before and after MARA treatment. The H0 hypothesis was that MARA had no effects on the PAS and that no significant changes in the sagittal PAS diameter were found after MARA treatment.

2. Materials and Methods

2.1. Data Acquisition

The Landeszahnarztekammer Hessen’s Institutional Review Board gave its approval to the study, which was carried out in agreement with the Declaration of Helsinki (protocol code 03/2021). All records from patients undergoing MARA treatment due to a Class II malocclusion between 2006 and 2017 were checked for completeness. Thus, 53 patients (see

Table 1. Treatment composition of the groups.

Group	N	Mean Treatment Duration
1	26	13.12 ± 7.5 months
2	27	10.48 ± 4.5 months
	∑: 53	Δ: 2.64 months

and

Table 2. Age composition of the groups.

Group	Mean Age	Range	Median
1	13.46 ± 2.0 years	10–19 years	13 years
2	16.78 ± 8.2 years	11–47 years	14 years

) were selected from the collective.

2.2. Groups

The groups were composed of 26 male patients (Group 1) and 27 female patients (Group 2) who were selected for the study. Treatment lasted approximately 13 months in Group 1 and approximately 10.5 months in Group 2 (see Table 1). The average age in Group 1 was approximately 13.5 (median 13 years) and the average age in Group 2 was approximately 16.8 years (median 14 years) (see Table 2). A control group was not involved.

2.3. Digital Lateral Cephalograms

For digital lateral cephalograms, two lateral cephalograms were taken during treatment: one before the MARA and multibracket insert (T1) and one after the removal of all orthodontic appliances (T2). All lateral cephalograms were taken using the ORTHOPHOS XGPlus DS/Ceph X-ray machine (formerly, Sirona Dental Systems, Bensheim; presently, Dentsply Sirona Deutschland GmbH, Bensheim, Germany). Digital acquisition, storage and tracing as well as PAS measurements were performed with the software Sidexis XG version 2.56 (formerly, Sirona Dental Systems, Bensheim, Germany; presently, Dentsply Sirona Deutschland GmbH, Bensheim, Germany). To measure changes in the lumen, we selected three remarkable and reproducible markers (see Figure 1). The parameters used in this study were based on the parameters from Kinzinger et al. [20]. The authors selected parameters at the P4 level, focusing on the “mandibular PAS” according to Barrera et al. [21]. Measurements were performed by two clinicians: NP and ZS.

2.4. Cephalometry

Cephalometric tracing was performed by two clinicians (NP and CS) using the plugin DentalVision version 7.7.2. of the software ortho Express version 8.66.1. (Computer Forum, Elmshorn). Cephalometric analysis of the lateral cephalograms followed the use of the Hasund/Ricketts/Jarabak approach. The following parameters were selected:

• Björk’s sum angle (NSAr, SArGo, ArGoMe) [°];
• SNA [°];
• SNB [°];
• ANB [°].

2.5. Statistics

Statistical analyses, both descriptive and analytic, were performed using the corresponding functionality of Microsoft Excel (version 14.0.7268.500 for Microsoft Office Home and Business 2010, Microsoft Corp. Redmond, Washington, DC, USA). All averages are given with their standard deviation (SD); all changes are presented numerically and as relative difference (%). Differences are tested against a null hypothesis (no difference); p values are shown and interpreted as significant or non-significant at the 5% level (p ≤ 0.05) or at the 1% level (p ≤ 0.01).

3. Results

With regard to the PAS, it was found that, averaged over all the patients in the two groups, the two-dimensional, anteroposterior diameter of the PAS increased, but the increase was greater in the male patients (+27.5%) than in the female patients (+11.6%). A significant increase in the PAS was found in the male patients (p = 0.006) but not in the female patients (p = 0.09) (see

Table 3. (a) PAS, raw data; (b) PAS, without outliers.

Group	PAS T1	PAS T2	Δ PAS Abs.	Δ Rel.
(a)
1	10.53 ± 3.4 mm	13.42 ± 3.8 mm	2.9 ± 2.05 mm	27.5% p = 0.006
2	11.25 ± 2.7 mm	12.55 ± 2.9 mm	1.3 ± 0.92 mm	11.6% p = 0.090
Average	10.89 mm	12.99 mm	2.1 mm	-
[20]	13.27 ± 3.07 mm	12.86 ± 2.47 mm	−0.4 ± 2.58 mm	p = 0.5155
(b)
1 (N = 25)	10.36 ± 3.35 mm	13.39 ± 3.90 mm	3.03 ± 2.14 mm	29.2% p = 0.005
2 (N = 20)	10.95 ± 2.86 mm	12.90 ± 3.06 mm	1.95 ± 1.38 mm	17.8% p = 0.044

a).

Some patients (N = 7) showed a decrease in the PAS (maximum −1.2 mm). In the male patients, this only occurred in one case (−0.5 mm). This is probably due to the smaller average PAS increase in female patients. To verify this, decreases in the PAS (i.e., negative differences by subtracting the PAS measured in T2 from the PAS measured in T1) were considered to be outliers. This resulted in significant changes in the PAS for both groups: +29.2% (p = 0.005) and +17.8% (0.044) (see Table 3b). The following provides concise and precise descriptions and interpretations of the experimental results.

Additional cephalometric parameters were obtained from the lateral cephalograms.

Table 4. Sum angle [Björk’s sum angle].

Group	Sum Angle T1	Sum Angle T2	Δ Sum Angle	Δ Rel.
1	388.31 ± 7.0°	387.74 ± 7.9°	−0.57 ± 0.07°	−0.15% p = 0.780
2	393.18 ± 7.4°	392.83 ± 7.6°	−0.36 ± 0.25°	−0.09% p = 0.860
[20]	392.96 ± 9.02°	392.07 ± 8.85°	−0.89 ± 3.27°	p = 0.262

shows Björk’s sum angle. This parameter decreased in both groups but not significantly (p = 0.78, p = 0.86). Overall, the male patients showed a greater decrease between T1 and T2 than the female patients. Apparently, unspecified individual factors contributed to this development, other than the MARA alone, resulting in more heterogeneous inter-individual values.

Table 5. SNA.

Group	SNA T1	SNA T2	Δ SNA	Δ Rel.
1	81.98 ± 4.3°	81.34 ± 5.3°	−0.64 ± 0.45°	−0.78% p = 0.630
2	80.12 ± 4.2°	79.56 ± 4.4°	−0.56 ± 0.40°	−0.70% p = 0.640
[20]	80.15 ± 4.58°	80.47 ± 4.17°	+0.32 ± 1.44°	p = 0.363

and

Table 6. SNB.

Group	SNB T1	SNB T2	Δ SNB	Δ SNB Rel.
1	77.31 ± 4.8°	77.63 ± 4.7°	0.32 ± 0.23°	0.41% p = 0.810
2	75.98 ± 3.9°	76.08 ± 4.3°	0.10 ± 0.07°	0.13% p = 0.930
[20]	74.6 ± 4.99°	75.89 ± 5.12°	+1.29 ± 1.34°	p = 0.0008

display the results for SNA and SNB, and we once more compare them to Kinzinger al. [20]. In the present study, both SNA and SNB behave heterogeneously, with some individuals showing a decrease and other individuals showing an increase between T1 and T2. The general tendency for SNA was a decrease (−0.64 ± 0.45°, p = 0.630 and −0.56 ± 0.40°, p = 0.640).

As for ANB (see

Table 7. ANB.

Group	ANB T1	ANB T2	Δ ANB	Δ Rel.
1	4.87 ± 2.2°	3.70 ± 2.2°	−1.17 ± 0.82°	−24.0% p = 0.058
2	4.14 ± 2.6°	3.49 ± 2.0°	−0.65 ± 0.46°	−15.7% p = 0.311
[20]	5.54 ± 3.25°	4.57 ± 2.98°	−0.98 ± 1.34°	p = 0.0064

), the male patients showed a decrease between T1 and T2 (−24%, p = 0.058). The 5% level of significance was narrowly exceeded. The female patients displayed a decrease that was less clear than in the male patients (−15.7%, p = 0.311). ANB as well as SNA and SNB behaved differently between individuals, with some individuals showing a decrease and some individuals showing an increase. The overall tendency in the ANB angle was a decrease.

4. Discussion

The sagittal PAS diameter was measured in 53 retrospectively selected MARA patients (male: N = 26, female: N = 27) between T1 (before MARA) and T2 (after MARA). Overall, there was an increase in diameter (+27.5% and +11.6%, respectively). Significance could be established in the male and female patients. It was noticeable, however, in the female patients that some (N = 7) exhibited a decrease in the PAS. In the male patients, only one patient showed such behavior. The question arises here on whether real individual differences are reflected in these results or whether method errors or measurement errors played a significant role.

The increase in the PAS diameter was greater in the male patients than in the female patients. Compared to Kinzinger al. [20], the present study shows a proportional increase in sagittal pharyngeal depth that is mostly significant at the 5% level, while the comparative study, at the P4 level, using the FMA (functional mandibular advancer), found a non-significant decrease, i.e., an anti-proportional relationship at the same pharyngeal level as in the present study. While the changes resulting from MARA treatment were significant at the 5% level, the FMA resulted in a change that was non-significant.

Individual factors could have a negative influence on the PAS in spite of any growth happening. These negative factors could theoretically originate in the hard tissues or in the surrounding soft tissues. Soft tissues (muscles, fat, fibrous tissues and lymphatic tissue) are subject to substantial individual differences and also differences occurring in the same individual over time. Weight gain can increase pharyngeal tissue mass, particularly in the area of the root of the tongue, thereby causing the PAS to decrease in diameter [22]. However, we did not record the individual weight changes between T1 and T2, so we cannot make any assumptions here. Allergies and chronic upper respiratory infections cause an increase in lymphatic tissues, thereby negatively affecting PAS lumen and aerodynamic properties. When measuring the sagittal PAS diameter, both general methodological and individual measurement errors are possible. The posterior airway space is a complex elastic three-dimensional structure. In the lateral cephalograms, we are seeing two-dimensional projections of three-dimensional spatial relationships superimposed laterally [23]. A three-dimensional representation of the PAS would require CT or MRI imaging to obtain both sagittal and transverse measurements as well as volume measurements [24]. CTs come with the great disadvantage of ionizing radiation exposure. MRIs do not involve such radiation, but they are associated with the major disadvantage of substantial financial costs both for the imaging procedure and the contrast agent [24]. By using pre-existing lateral cephalograms, we avoided both problems by opting for a procedure that involves little radiation exposure and costs. Lateral cephalograms are routinely taken as part of orthodontic diagnostics. Becker et al. validated the lateral cephalogram as a predictor for intraoral splint therapy in OSAS [25]. They conclude that lateral cephalograms taken in a supine position are a valid, practical point of reference for the PAS while lateral cephalograms taken in an upright position do not represent the PAS situation during sleep. While they are correct in their conclusions, the immense requirements for a lateral cephalogram in a supine position are not suitable for daily routine practice and we therefore compromised by using lateral cephalograms already taken, without exposing our patient to more ionizing radiation or incurring the costs for MRI imaging.

Venza et al. demonstrated the significance of CBCT in the diagnosis and treatment planning of individuals with OSA [26]. As part of the overall examination, the authors advised evaluating CBCT images and estimating respiratory tract measurements [26]. Based on the outcomes of Vanza et al.’s study, our future study could examine the respiratory tract’s modifications on CBCT following MARA treatment [26].

The absence of a control group in this study could be a possible limitation. The control group could consist of patients with Class II malocclusion who were not treated. Since we analyzed cephalometric radiograms before and after orthodontic treatment, we feel that it would be unethical to exclude patients with Class II malocclusion (who have specific functional and aesthetic issues) from receiving orthodontic treatment in order to include them in the control group.

The parameters used in this study were based on the parameters from Kinzinger et al. [20]. The authors selected parameters at the P4 level, focusing on the “mandibular PAS” according to Barrera et al. [21].

Our PAS data were adjusted for outliers (for single negative PAS changes, see Table 3). The resulting PAS diameter increase between T1 and T2 was significant both for the male patients and the female patients (p = 0.005 and p = 0.044). Observed PAS changes were less pronounced in the female patients compared to the male patients.

Björk’s sum angle reflects the individual growth pattern: if it is less than 396°, the pattern is horizontal; if it is greater than 396°, it is vertical. In the male patients (388.31 ± 7.0°), the general growth pattern was more horizontal than in the female patients (393.18 ± 7.4°). Kinzinger et al. [20] found a greater, non-significant decrease (−0.89°) between T1 and T2. The standard deviation (SD) was 3.27, which was about 4.2 times greater than in the present study. Neither study found a significant change in this parameter. There was a tendency for Björk’s sum angle to decrease in the T1 and T2 interval as a result of MARA treatment. Changes in the vertical direction were the opposite of what was expected. The condyles moved down and forward. The pubertal growth spurt is the best time frame for treatment if skeletal effects are wanted.

The SNA as a sagittal parameter for the maxilla decreased; the upper jaw thus became more retrognathic between T1 and T2 but not significantly. Kinzinger et al. [20] found an increase (+0.32 ± 1.44°, p = 0.363). Both studies found non-significant changes in the SNA. The SNB as an indicator of the sagittal mandibular position increased non-significantly. The lower jaw became more prognathic between T1 and T2. The headgear effect on the upper jaw induced by the MARA occurs as a consequence of moving point A backwards. The vertical arm of the upper elbow produces pressure on the lower arm. This effect is more noticeable in adult patients in which growth intensity is limited. The MARA intensely stimulates lower jaw growth. The total increase in the length of the lower jaw was between 1.5 mm and 2.6 mm. In their study, Rizk et al. showed statistically significant decreases in SNA and ANB, which may have resulted from the maxillary growth restriction that previous researchers had discovered [27]. The amount of forward mandibular development may have been restricted by this “headgear effect”. Therefore, mandibular anterior posture in relation to the cranial base cannot be the only explanation for the notable increase in airway capacity, dimensions and hyoid bone position [27].

In comparing these results with Kinzinger et al. [20], it is notable that the ANB in the FMA group was higher (5.54) than the ANB obtained in the present study (4.87 and 4.14 respectively), with the SD being between 1.25 and 1.5 times greater than in the present study, which puts the ANB derived from the comparative study [20] still within the range that our study established. The ANB angle indicates the sagittal relationship of maxilla versus mandible: here, there was a non-significant decrease, which in turn corresponds with the fact that the lower jaw moves slightly anteriorly in relation to the upper jaw. The clear individual differences that can be observed with regard to SNA, SNB and ANB can be explained by slight individual differences in intermaxillary relation. However, the general tendency as explained here is evident.

Looking at all measured parameters together, a collective 57 patients undergoing MARA treatment (duration 13.12 ± 7.5 months and 10.48 ± 4.5 months) demonstrated a significant increase in the sagittal PAS diameter (+27.5% or +11.6%; adjusted: +29.2% or +17.8%). Jaw growth was directed more horizontally and more with regard to the mandible than the maxilla.

The findings and their implications should be discussed in the broadest context possible. Future research directions may also be highlighted.

5. Conclusions

Class II treatment with a fixed appliance (MARA) seems to be an effective means of increasing the sagittal lumen of the PAS and thus counteracting obstructive sleep apnea. Compared with the FMA [20], we showed a significant change during MARA treatment both for the male patients and the female patients. We hypothesize that the MARA can be a treatment option for Class II patients with mild to moderate OSAS. Also, the MARA could be used to help prevent the development of OSAS later in life.