Does overweight affect the sagittal dimension of the posterior airway space in a non‐OSAS population? A case control study

Abstract Objective The null hypothesis was that, in a non‐obstructive sleep apnea syndrome population, overweight do not reduce the antero‐posterior dimension of the posterior airway space. Materials and Methods The author retrospectively reviewed the records of subjects evaluated at the Maxillofacial Surgery Unit, Department of Neurosciences, University of Padova Medical School, Padova, Italy, from 2016 to 2018. Only patients with complete demographic, anthropological and CBCT dataset were enrolled. OSAS patient were also ruled‐out. Enrolled patients were divided into overweight (28 cases) and non‐overweight (32 controls) groups according to the patient's Body Mass Index. Each two‐dimensional cephalometric radiography obtained from the cone‐beam computer tomography dataset was evaluated in order to measure linear and angular distances between standardized cephalometric landmarks. The two‐sample t‐test was the statistical test applied to compare the case and control data. Results There were no statistical differences between the two study groups for any of the evaluated variables: the null hypothesis was accepted. Conclusion This study showed that in a non‐obstructive sleep‐apnea population, overweight and class I obesity does not influence the airway space in the antero‐posterior dimension. Further investigation should focus on categorized overweight‐obese population. Accurate and reliable protocol for tridimensional airways assessment should be implemented.

Several hypotheses have been formulated in an attempt to explain a possible mechanism by which obesity affects OSAS. The main theory suggests that obesity may cause a decrease in the size of the upper airway by fat deposition in surrounding tissues, which contributes to hyperplasia and soft tissue collapsibility; another hypothesis suggested that mass loading of the anterior cervical region might increase upper airway resistance with consequent airflow reduction during breathing. Obesity may, also, indirectly affect airway size by causing a decrease in lung volume (Fritscher et al., 2007;Maciel Santos, Rocha, Laureano Filho, Ferraz, & Campos, 2009).
With regard to maxillomandibular hypoplasia, a large number of cephalometric and cone beam computer tomography (CBCT) or computer tomography (CT) studies have investigated the abnormal parameters that characterized subjects affected by OSAS in relation to age, sex, race and body mass index (BMI) (Abramson et al., 2011;Maciel Santos, Laureano Filho, Campos, & Ferraz, 2011;Paoli et al., 2001;Shigeta et al., 2008).
While in OSAS population a large amount of studies correlates the decreasing patency of the posterior airway space (PAS) with obesity, in non-OSAS population the effect of overweight-obesity and mandibular hypoplasia on the PAS are not yet defined.
The aim of the study was to compare, in a non-OSAS population, cephalometric airways measurements between overweight-obese and normal weight subjects; a possible effect of obesity on the reduction of sagittal dimension of upper respiratory airways was investigated.
The null hypothesis was that overweight do not reduce the sagittal dimension of the posterior airway space (PAS).

| MATERIALS AND METHODS
The authors retrospectively reviewed the clinical and radiological data of all patients referred from January 1, 2016 to December 31, 2018, to the Maxillofacial Surgery Unit, Department of Neurosciences, University of Padova, for dental disease or dentofacial deformity assessments. The study was approved by the Institutional Review Board and was conducted in accordance with the principles of the Declaration of Helsinki. All the enrolled patient gave permission to process personal data for scientific purpose.
The study design is reported in Figure 1. A total of 730 medical reports were screened. One hundred ninety-eight were found to be eligible for the study presenting adequate CT/CBCT scans and complete data regarding age, sex, height and weight.
The following exclusion criteria were applied: (a) confirmed diagnosis of OSAS; (b) subjects younger than 18 years of age, (c) craniofacial deformities, cleft palate, facial fractures, head and neck tumors, (d) previous surgical correction of dentofacial deformity. After screening with the exclusion criteria, 138 patients were excluded and 60 were included in the research.
For each enrolled patient, BMI was calculated (BMI = weight in kg/height 2 in cm) (Kushner, 2008): patients with BMI ≥ 25 were the overweight group (cases) and those with BMI < 25 were grouped in the non-overweight group (controls). Furthermore, patients in cases group were classified according to the accepted definition of overweight and obesity: patients with BMI between 25 and 30 kg/cm 2 were considered overweight, with BMI between 30 and 35 kg/cm 2 I class obesity, with BMI between 35 and 40 kg/cm 2 II class obesity, and subjects with BMI greater than 40 kg/cm 2 were considered III class obesity.
Each CT/CBCT dataset was imported to dedicated software with the aim of producing a lateral teleradiograph of the head (Dolphin Imaging 11.9 Premium, Chatsworth, CA). The same software was used by the same author (F.A.) to identify specific cephalometric landmarks and planes. As reported in Table 1, four cephalometric angular measurements were selected to assess maxilla-mandibular discrepancies with reference to the cranial base; moreover, according to Riley et al., 4 linear measurements were selected based on their clinical relevance in detecting the patency of the posterior airway space (PAS) (Riley, Guilleminault, Powell, & Simmons, 1985).
The two-sample t-test was the statistical test applied. A p value <0.05 was considered significant. The STATA™ 8.1 (StataCorp LP) statistical package was used for all analyses.

| RESULTS
Demographic and clinical features of the study population are reported in Table 2 The main cephalometric angular analysis regarding maxillomandibular discrepancies is reported in Table 3: no statistical differences were found between the obese and non-obese populations.
Similarly, the cephalometric linear analysis on PAS soft tissue patency is reported in:   of the correction of dentofacial deformity and many CT/CBCTs were already performed.
In this study the airway analysis have been made by the use of 2D cephalograms because, in the authors opinion, 3D software is not a completely reliable technology so far; some authors suggest that 3D viewer softwares, despite highly reliable in their airway assessment, gave poor accuracy thus suggesting systematic errors (Hakan & Palomo, 2010); another group found a significant difference in the airway volume definition up to 42% between the semi-automatic and the manual segmentation protocol in a widespread 3D software (De Water, Saridin, Bouw, Murawska, & Koudstaal, 2014); in another study a general underestimation of the upper airways volume have been assessed in three software packages (Chen et al., 2017). This variability seems to be correlated to the lack of a unique protocol in the dataset acquisition and processing, especially regarding threshold value selection: generally, the increase or decrease of the threshold results in a greater or smaller airway volume, respectively (Alves Jr et al., 2012); more variability is also correlated to the "partial volume effect" (Chen et al., 2017). Given the aforementioned problems in 3D PSA assessment, 2D lateral cephalogram is still consid-  Some authors suggested that the oropharyngeal airway space and the hyoid position may vary according to the antero-posterior position of the mandible (Tallgren & Solow, 1992). In order to avoid any cephalometric variability in the population, the authors assessed and compared SNA, SNB, and ANB angles between the two study groups: however, no significant differences were found between cases and control (Table 3); however, it has to be pointed out that SNA angles present a more pronounced trend toward statistical significance compare to SNB and ANB.
There is also evidence that flexion or extension of the head may influence the airway space (Hellsing, 1989). The authors assessed and compared the Cervical line-SN angle between the two-study group: again, there were no significant differences in the Cervical line-SN angle between cases and controls (

| CONCLUSION
This investigation showed that in a non-OSAS population, overweight and class I obesity does not influence the PAS sagittal dimension.
The study of non-OSAS population should be encouraged to better understand the OSAS pathophysiology.
Further investigation should focus on class II-III populations or, at least, should categorize the overweight patients according to BMI index.
Moreover, future research should consider different sample age.
Novel three-dimensional technologies should be considered in airway dimension assessment; future studies should be focused on the definition of a comprehensive and standardized 3D protocol for the evaluation of PAS. Reliability test should be performed to avoid lack of scientific significate in the results.
The main limitation of the present study is the retrospective study design: prospective multi-institutional studies are needed to assess if any relationship does in fact exist.

ACKNOWLEDGMENTS
The authors thank Andrea Giovanni Apolloni for the iconographic support. The authors thank John Feehan for correcting the English version of this paper.