Facial symmetry is an important component of a person's attractiveness, representing one of the determining factors in its evaluation.1 Patients' concern for their physical appearance and for a more pleasing facial aesthetic is growing.23 Thus, the orthodontist is more sought after than ever and often subjected to complex cases, which require a thorough examination to establish a correct diagnosis.4

The etiology of facial asymmetries is vast and in some cases remains mostly unknown.5 Morphology can differ under the prevailing environmental conditions under which the individuals’ development is taking place. 67

Aesthetic and functional problems associated with significant facial asymmetry can negatively affect the patient's orofacial area, nutritional and psychosocial development. Therefore, a critical assessment and accurate treatment planning is absolutely necessary.8910

Deviations from normal development, dento-maxillary anomalies in the transverse and vertical direction are studied in depth in the contemporary literature and, along with them, the correct evaluation of cephalometric measurements. 1112 As a complementary examination, the use of postero-anterior cephalometry was not introduced as early and used as widely as the lateral cephalogram.13141516 but patients are three-dimensional, and thus the two dimensions visible on the lateral cephalogram should not be acceptable as the sole evaluations determining an orthodontic diagnosis. Numerous PA cephalometric analysis systems have been recommended each having strengths and limitations. Internationally, in countries such as Saudi Arabia and Peru, there is growing interest in the correct use and evaluation of postero-anterior cephalometric analysis parameters.1718 In 2011, Perez et. al used the Ricketts PA analysis to evaluate the norms of the Hispanic population.17 In 2019, the Grummons analysis was used in the symmetry assessment of the Saudi population in a study by Kumar et. al. 18

The main aim of this study was to evaluate and establish statistical correlations between a combination of PA cephalometric analyzes parameters defined by Ricketts and Grummons in a group of Romanian patients.

This retrospective study included facial asymmetry Romanian patients aged 15 to 37 years, who presented at the Department of Orthodontics and Dento-Facial Orthopedics of UMF "Victor Babes", Timisoara between 2018-2020 and received orthodontic clinical examinations and complementary investigations such as: extraoral and intraoral photographs, study models, radiographs and cephalometric analysis. 50 patients had PA cephalograms made with the Cranex 3D device (Soredex). This research was conducted in the Orthodontic Research Center from Faculty of Dental Medicine, “Victor Babes” University of Medicine and Pharmacy, Timisoara.

The patient positioning system provided a head stabilization support, correct geometry and calibration ruler on each side of the image, teeth in maximum intercuspation position and a cephalometric light highlighting the Frankfurt plane that ensured the correct positioning of the patient. 35 PA investigations were excluded from the study due to not meeting the inclusion criteria: patients of Romanian origin (confirmed in the observation sheet) with no history of orthodontic treatment or trauma to the face, no craniofacial syndromes or skeletal deformities. The 15 remaining investigations were analyzed using the AudaxCeph software. Dental and skeletal cephalometric parameters described by Ricketts, Grummons and Kappeyne Van De Coppello were collected through linear, angular and volumetric measurements. The cephalometric reference points used in this study are presented in Table 1.

Dental parameters considered in this study are represented by molar relations on left and right (A6-B6 / 6B-6A linear measurement describing the vestibulo-lingual molar inclination), intermolar width and intercanine width. Skeletal relationships are defined by the following planes: right and left maxillomandibular width (JL / JR to ZL-AG / ZR-GA), maxilla-mandibular midline (angular measurement between the ANS-Me and ZA-AZ), maxillary width (JL-JR) and mandibular width (AG-GA). Dento-skeletal relationships were defined by the distance from the lower first molar to the JR-GA / JL-AG plane. The angle between the reference points ZA-AG-ZL / AZ-GA-ZR defined craniofacial relationship and the nasal width (NC-CN) described internal structures of the face. The reference points and planes of the Ricketts PA cephalometric analysis are illustrated in Figure 1.

To assess the discrepancies in facial asymmetry, three components of the PA analysis described by Grummons and Kappeyne Van De Kopello, which show left-right values, were generated by AudaxCeph version 6.0, after tracing the analysis, locating anthropometric points and tracing bone outlines: mandibular morphology - the analogous sides of the two mandibular triangles, formed between the points Co, Ag, Me and the two angles Ag (Co-Ag-Me) were compared; evaluation of linear asymmetries - the distance to the referencemidline and the difference between the vertical dimensions of the perpendicular projections of the bilateral landmarks (Co, NC, J, Ag) on the reference midline were calculated; maxillo-mandibular relationship - the distances from the vestibular cusps of the first maxillary molars to the projections of JL / JR to the reference midline were measured. In addition, the AG-AGplane, the ANS-Me and reference midline line were also drawn to reveal dental compensations for any skeletal asymmetry and the distances from the zygomatic-frontal sutures to the reference midline were measured and compared in order to evaluate the discrepancies in the upper floor of the face (Figure 2).

The MS Office Excel 16 was used to create the database and statistical analyzes were performed in the SPSS 24.0 program (SPSS, Chicago, IL). For the 15 selected measurements, the mean value and the standard deviation were calculated.

Out of the total PA cephalometric analyzes, 40% men and 60% women, aged between 15 and 37 years. The mean age was 23 +/- 1 year for the women group and 21 +/- 1 year for men. The subjects were divided in groups based on gender and asymmetry severity.

After interpreting the measurements resulting from the Ricketts PA analysis, we observed that 30% of patients had a significantly deviated mandibular midline. The difference of the parameter values in the group of patients with significant asymmetries of the face compared to the other group, is represented in Figure 3.

Comparison ​​between patients with mild asymmetry and those with severe asymmetry, was done through the independent T test. The results found statistically significant differences in mandibular midline deviation, facial symmetry and nasal width (p <0.05).The independent T-test was also used to compare the differences between the sexes. There were no statistically significant differences between women and men.

To study the correlation in the Ricketts analyzes, the Pearson correlation coefficients were calculated. The coefficient with the highest value was between the mandibular width and the left maxillomandibular width (r = 0.929), and the one with the lowest value was found between the maxillary width and the left maxillomandibular width (r = 0.01). Significant correlations were found between intermolar and intercanine distance (r = 0.607) at p = 0.05. For the level p = 0.01 the significant coefficients with the highest value were found between the mandibular width (AG-AG) and the right maxillomandibular width (r = 0.644), and the significant coefficients with the lowest value turned out to be between the width maxillary (JL-JR) and and with the left maxillomandibular width (r = -0.752).

In the PA analysis described by Grummons, the following measurements on the two sides of the face were compared to the reference median-sagittal plane: mandibular morphology (Co-AG, AG-Me, Co-Me distances, AG angle); the asymmetry evaluation was performed by comparing the distances of the following points to the mid-sagittal line: Z, Co, Za, NC, J, AG and the left and right maxillomandibular relationship. The paired t test was calculated between the left and right measured values, but none of these were statistically significant. Pearson correlation coefficients were calculated to study the correlations. The coefficient with the highest value was between the AG angle and the Co-Me distance, both on the left (r = 0.639), and the coefficient with the lowest value at p = 0.05 was observed between the left Co-Me distance and the right Co-AG distance (r = 0.521). Other statistically significant values at the same p level were between the coefficients Co-AG right with Co-AG left, Co-Me left with Co-Me right, Co-Me right with Co-AG right and AG-Me right. At p = 0.01 the coefficients with the highest value were between the distance J right and the distance J left (r = 0.993), and those with the lowest value were observed between the angle AG on the left and Co-Me on the the right side (r = 0.651).

Standardized methods of frontal cephalograms analysis have been around for decades most of them mainly used in surgical approach cases.131920

Postero-anterior digital cephalometric analysis is accurate and especially useful in diagnosing asymmetry patients with transverse and vertical anomalies.21 The earlier these abnormalities in the development of the maxillary bones are discovered, the more beneficial it is for patients, allowing the initiation of a rebalancing treatment. This can avoid the need for dental camouflage treatments or more complicated types of orthognathic surgery.22 If skeletal imbalances are detected in patients who are still growing, immediate initiation of orthopedic and functional treatment is recommended, as true correction could be possible during this period.2324 To enhance the amount of data collected and surpass the limitations of the individual analysis, our study combines parameters from orthodontically oriented PA analyzes: Ricketts, Grummons and Kappeyne Van De Coppello.

To date, research in the field of antero-posterios cephalogram analysis focuses mainly on verifying the applicability of certain reference values in different populations.1718 Few studies focus on the correlations between various parameters. 2526 Patients included in this study that met the inclusion criteria exhibited varying degrees of facial asymmetry. When correlating gender with the evaluated parameters in our study no differences between the sexes were statistically significant in either of the PA analyzes. Our literature search found studies that found female faces to be more asymmetrical while others found no statistical differences between genders, which is in accordance to our findings. 2728 Soft tissues that cover the skeletal components tend to minimize skeletal asymmetry, in our study gender differences in the asymmetry of skeletal and soft tissue components were not statistically significant. The Ricketts analysis revealed statistically significant differences between parameters that describe the degree of asymmetry, respectively the deviation of the maxillo-mandibular midline together with one of the parameters of the internal structures, namely that of the nasal width. The quantitative analysis of Grummons found a constant dominance of asymmetrical development on the right side in most of the compared parameters, meaning that all skeletal and facial structures were situated significantly higher on the right side than on the left. The lateral area of the jaw exhibited a higher degree of asymmetry than other components of the face in all subjects. The postero-anterior analysis described by Grummons and Kappeyne Van de Koppello proved reliable in the determination of the severity of the asymmetry, which is an influential factor in deciding orthognathic intervention and diagnosing a possible underlying TMD.29

Considering the results obtained from our study group, we conclude that our combination of postero-anterior cephalogram analyses provided data that showed some positive and statistically relevant correlations between the evaluated parameters in asymmetry patients useful in correct diagnosis and orthodontic treatment planning.

This research was conducted in the Orthodontic Research Center from Faculty of Dental Medicine, “Victor Babes” University of Medicine and Pharmacy, Timisoara.