Assessment of facial asymmetry and establishment of threshold of sub-clinical asymmetry in Malwa population

Introduction: The word symmetry is derived from the Greek word ‘symmetries’ which means ‘of like measure’. Facial symmetry can be defined in numerous ways, one being associated with the state of facial equilibrium, in which there is a correspondence in size, shape, and arrangement of facial landmarks on the opposite sides. Materials & Method: The 1427 subjects in the present study were selected from the Out Patient Department of Government College of Dentistry, Indore(M.P), who presented with aesthetically pleasing faces over a period of one year (October 2013-0ctober 2014). Out of the 1427 patients examined 150 (17-30 years) subjects were randomly included. Photographs and orthopantomogram of all the 150 patients were obtained. Result: The photographs and orthopantomogram were analyzed & Absolute value of Asymmetry Index was taken for all the measurements. Comparison of absolute Asymmetry index of different parameters between male and female subjects was performed. Wilcoxon paired test showed right side predominance for the parameter corpus length, middle facial width, cheek length, lower facial width. Conclusion: In the present study an attempt was made to quantify sub-clinical asymmetries in clinically symmetrical faces. Minor asymmetries were observed in nearly all individuals taken up for the study. There is no association of gender with predominance of facial asymmetry. On assessment of side predominance of asymmetry it was concluded that the right side dominance of asymmetry for corpus length, middle facial width, cheek length, lower facial width. A threshold value of 6% for sub-clinical asymmetry was established from this study except for condylar and coronoid.

Soft tissue features are quantified by measuring frontal facial photographs, and the relationship of these measurements with cephalometric measurements has been reported. 4   Photographic procedure: The photographs were standardized being taken by a single operator who were unaware whether patients were included or not in the study. The photographic setup consisted of a tripod stand that held Nikon D 3200 camera with a 55-105mm macro lens and a primary flash. The photographs were taken from a fixed distance of 1.5 meter, from the patient; 30 cm from the lit box wooden stool was placed so that the patient can sit straight with his/her toes just behind a line drawn on the floor, looking straight at the camera with horizontal visual axis and remain expressionless. The stability of the elements and the easy adjustment of the tripod height allowed us to keep the optical axis of the lens horizontal during the recording. Patients were explained about the photographic procedure and were given identical instructions like tying back hair to allow visualization of the inferior ear insertions, removal of large earrings and glasses, so that the patient's forehead, neck, and ears were clearly visible during the recording. The photograph was printed with the same colored printer Konica Minolta Biz hub C224e using A4 size royal executive bond paper(thickness 85 gsm) to ensure that the photographs was identical throughout the study. OHP sheet was fixed on the printed photograph and vertical and horizontal reference line and landmarks were marked, vertical and horizontal measurements were measured and noted.
These photographs were subjected to scrutiny before a panel comprising of a lay observer, an orthodontist and a general practitioner for subjective evaluation of asymmetryscored '0' and '1' on the basis of prepared questionnaire, 60 patients were selected. OPG of the selected subjects was obtained and analyzed for percentage and severity of clinical asymmetry. The subjects were further divided into 'True symmetrical' (score 0) comprising of 27 subjects and 'symmetrical' (scored 1) comprising of 33 subjects.
Orthopantomogram procedure: The entire OPG procedure was explained to the subject prior to the procedure, so that the subject may not move from his/ her position after placement. The patient was asked to stand straight or sit erect in OPG machine; such that the patient's Frankfurt horizontal plane lies parallel to the floor and chin of the subject was placed on the chin rest. The subject was asked to place the tongue in such a way that it touches the palate. The subject was asked to bite on the bite block such that he/she may not be too far forward or backward. Any prosthesis, Jewelry, etc. was asked to be removed before the procedure. (Figure 3) Orthopantomograms were taken by using same machine Carestream CS8000c (formerly Kodak K8000c). Obtaining digital OPG from single machine by a single operator did Orthopantomograph standardization.
After obtaining the OPG from the machine, it was assessed for any errors, only those panoramic radiographs presenting no artifacts, the whole mandible fully captured on the radiographs, and the contrast on the radiographs sufficient to perform all intended measurements were chosen.If any errors were found, the procedure was repeated. Mandibular first molar mesio-distal dimension was checked on both sides to look for any magnification. If present, such radiographs were eliminated.
OPG of all individuals will be analyzed for vertical measurement of condylar process, coronoid process, ramus, body/ corpus length, orbit to mesio-buccal cusp tip of maxillary molar on the right and left side.
The distortionthat occurs in a panoramic radiograph6 on the linear dimensions of the mandible on both sides, a distortion factor for each half of the mandible was calculated. For this, 10 radiographs from the sample were randomly selected and the study model obtained from the subjects was used to make comparisons. The mesio-distal dimensions (also the occluso-gingival height) of the four permanent molars were measured on the study models. The same distance will be then measured on the panoramic radiographs.The "distortion factor" was calculated by:Mesio-distal length of the cast (four first permanent molars) divided by mesio-distal length of the molars on the radiographs multiplied by 100. This distortion factor is mainly applied to mesio-distal measurements taken for the study like corpus length but not for vertical measurements.
The magnification factor was assessed with the use of two plus shaped metal wire, fixed to both sides of a dry skull on the body and ramus of the mandible. The magnification or distortion factorsfor right and left sides, which show no asymmetric enlargement or asymmetric shortening of bilateral wire markers. A single operator manually traced OPG radiographswith use of 0.75 microns lacquered polyester,acetate tracing paper fixed on the radiograph using a 0.35 mm lead pencil. Asymmetry will be drawn for all four measurements.

RESULT
The photographs and OPG were analyzed & Absolute value of Asymmetry Index was taken for all the measurements. Microsoft Excel (Redmond, Washington, USA) was used to compile the data. The data collected was analyzed using SPSS 18 (SPSS, Chicago, Illinois, USA). Alpha error was set at 0.05 and a 'p' value 0.05 was considered statistically significant.
Descriptive statistics (Mean, Minimum, Maximum, Stand. Deviation) was calculated for total 60 samples, 30 female 30 male samples separately (Table 1). The t-Test was used to compare the mean values of parameters between males and females. The result of t-test showed statistically insignificant differences between males and female subjects except for the mean right and left TRH, Gonial Angle, Middle facial width and facial height Based on the AI value of each parameter the subject was classified as left, right predominance or symmetrical.
The Chi-square test was used to see association of gender with predominance. No association was found between the gender and predominance. (Table 3) Wilcoxon paired test (Table 4) showed right side predominance for the parameter corpus length, middle facial width, cheek length, lower facial width.
( Figure 6) On the basis of severity of Asymmetry Index, high percentage of subject showed very mild-to-mild asymmetry with respect to total ramal height, orbit to 6, Gonial angle and corpus length while a high percentage of subjects showed moderate to severe asymmetry with respect to condyle and coronoid height.
On assessment of 95 % Confidence interval value highest value of AI in OPG was observed in coronoid    In Figure 7 when AI of True symmetry group (n=27), symmetry group (n=33) and total number of subjects(N=60) was plotted for each parameter it was found that 6 % was the threshold of underlying subclinical facial asymmetry. showing less asymmetry in maxillary region than in mandibular region (Table 6).

DISCUSSION
Facial asymmetry is frequently occurring in individuals, the so-called symmetrical faces represent a combination of numerous, minor-asymmetrical components. 9 In the present study AI was used to quantify facial asymmetry because AI is much better than measuring linear differences due to magnification occurring in OPG between right and left side. Measuring AI is easy and effective for calculating asymmetry and is not affected by positioning errors, distortion and magnification. 10 OPG is a reliable tool for determining mandibular asymmetry. 7,8,[10][11][12][13][14][15][16][17][18][19][20][21] It was used to assess AI because it offers a method to analyze the various structures of mandible separately on the right and left side. 7,13,14,20,21 Acceptable results can be achieved with OPG, which is non-invasive, have a favorable cost-benefit relationship and exposure of subjects to relatively low doses of radiation. 17 Reproducibility of vertical and angular measurements on OPG is acceptable if the patient's head is positioned Bharti C, Jain S, Bharti HV : Assessment of facial asymmetry and establishment of threshold of sub-clinical asymmetry in Malwa population properly in the equipment. 13,14,22 Habets et al 7,20,21 concluded that the head holder must be fixed well to the orthopantomograph, and the head has to be well centered in the head holder of the orthopantomograph when a clinical orthopantomograph is to be evaluated.
The method used for standardization in this study was easy to use and reproducible in any clinical setting and also does not require the use of any setup of accessory lighting equipment.
The findings of this study conclude that most of the   Similarly there was no statically significant difference in male and female subjects regarding to asymmetry index derived from all the parameters ( There is lack of scientific evidence to define a range of facial asymmetry that is aesthetically acceptable.

OJN
On the basis of severity of asymmetry at 95% confidence interval, all parameters showed very mild to mild asymmetry (3-5%) except coronoid and condylar height which showed moderate to severe degree of asymmetry between right and left side in symmetrical faces (Table 6). Lu (1965) 41 reported that only facial asymmetries greater than 3 % are clinically discernible. It is likely that the decision of the threshold value is rather subjective according to investigators or clinicians, particularly in the subject of asymmetry.
The reason for dissimilar results to our study could be because of selected subjects was having different facial form, age, racial origin, head orientation and methodology.

CONCLUSION
In the present study an attempt was made to quantify subclinical asymmetries in clinically symmetrical faces. Minor asymmetries were observed in nearly all individuals taken up for the study. There is no association of gender with predominance of facial asymmetry. On assessment of side predominance of asymmetry it was concluded that the right side dominance of asymmetry for corpus length, middle facial width, cheek length, lower facial width. A threshold value of 6% for sub-clinical asymmetry was established from this study except for condylar and coronoid. Present data can be used as baseline data for further research.