CLINICAL PHOTOGRAPHIC AND RADIOLOGICAL CORRELATION IN PATIENTS WITH SCOLIOSIS

ABSTRACT Objective: The study correlates angular radiographic measurements and measurements from photographs of the trunk of patients with adolescent idiopathic scoliosis. Methods: a cross-sectional study of 10 patients with adolescent idiopathic scoliosis and indication for surgical treatment in follow-up in the spine sector of HCRFMRP-USP. They were evaluated through measurements taken from clinical photographs of the patient’s torso and angles on panoramic radiographs of the spine. Results: only the correlation between the axilla angle (AHA) and the proximal thoracic curve and main thoracic curve was observed in the sample studied. No correlation was observed between the other compared variables. The study was carried out in a sample with a small number of patients, whose objective was to carry out a pilot study to explore this topic. Conclusion: the analysis of deformities should consider the aesthetic aspect and radiographic evaluation; however, in this study, no statistical correlation was observed between these parameters. Level of Evidence IV; Case Series Study.


INTRODUCTION
Spinal deformities are evaluated employing clinical parameters related to the asymmetry of the trunk surface references and radiographic parameters that evaluate angular measurements of the scoliotic curve. 1,2Scoliosis classifications have been carried out based on angular measurements and the location of the scoliotic curves, and in our midst, the classification proposed by Lenke stands out. 3 Assessment through photographs of the trunk has been used to evaluate posture and trunk deformities in idiopathic scoliosis. 4Shoulder and waist asymmetries have been evaluated utilizing photographs, and different parameters have been developed for the evaluation of deformities based on the parameters obtained by using trunk photography. 5,6[9][10]   The study aimed to correlate the angular radiographic measurements with the measurements derived from photographs of the trunk of patients with adolescent idiopathic scoliosis.

METHODS
This was a cross-sectional study of patients with adolescent idiopathic scoliosis and indication for surgical treatment, being followed up at the spine sector of the HCRFMRP-USP.The Research Ethics Committee of HCFMRP-USP approved the study under number: 68156423.5.0000.5440.Ten patients with adolescent idiopathic scoliosis and an indication for surgical treatment were randomly selected.The patients were evaluated by measuring the radiographic parameters of the scoliotic curves and the photographs of the patients.
The parameters selected for measurement on the radiographs were: proximal thoracic curve (PTC), main thoracic curve (MTC), thoracolumbar/lumbar curve (TL/L), T1 angulation (T1 tilt) (Figure 1).Panoramic spine radiographs were taken with the patient standing in AP and profile.The measurements were performed using the Surgimap imaging program with the "Coronal Wizard" tool to define the terminal vertebra, the projection of the curves, and the "Cobb Angle" option for measuring the analyzed values.
The photography of the patients was done on the same day as the radiographs, always by the same examiner, using the same camera and environment.The patients were positioned at 130 cm from the examiner, in the orthostatic position, with arms extended along the body in the anatomical position and with complete visualization of the shoulders, talus triangle, pelvic girdle, and lower limbs.
The angles selected for evaluation in the photographs were: shoulder height angle (SHA), armpit height angle (AHA), waist height angle (WHA), and area.(Figure 2) Shoulder height angle.(SHA): is defined as the angle between the upper border of the shoulder acromion, bilaterally related to the horizontal.(Figure 2) Axillary height angle.(AHA): corresponds to the angle between the upper edge of the axillary fold and a horizontal line.(Figure 2) Waist height angle (WHA): is defined as the angle between the line joining the apex of the right and left side carving tringulum and the horizontal.(Figure 2) Right and left waist angle (R. WHA and L. WHA): is defined as the angle between the line tangent to the lateral chest wall up to the concavity of the waist and another line from the apex of the waist tangent to the iliac crest on the respective right and left sides of the patient.(Figure 2) The anatomical landmarks: superior border of the acromion, axillary fold, and apex of the waist connected form the image of a polygon on which it is possible to calculate the occupied area.(Figure 2) The angles were measured based on the anatomical references according to the "Back Surface Metrics" tool of the Surgimap program.
Descriptive statistics were performed, the Kolmogorov-Smirnov test was used to evaluate the normality of the samples, and Sperman's test was used to evaluate the correlation between the study parameters.The significance level was set at 5% (p<0.05).

RESULTS
The demographics of the patients are shown in Table 1.The values of the radiographic parameters measured in the ten patients selected for the study are illustrated in Table 2.
The values of the parameters measured in the photographs are shown in Table 3.
The correlation values between the parameters studied are shown in Table 4.
The correlation between shoulder angle (SHA) and radiographic parameters (proximal chest curve, main chest curve, thoracolumbar/ lumbar curve, and T1 tilt) are depicted in Figure 3.No correlation was observed between the parameters evaluated.
The correlation between axillary angle (AHA) and radiographic parameters is illustrated in Figure 4.Only a correlation was observed between the axillary angle (AHA) and the proximal thoracic curve (PTC), and the main thoracic curve (MTC).
The waist angle ranged from 1.1 to 26.7° and showed a statistical correlation with the main chest curve.(Figure 5) The area ranged from 0.59 to 1.1, with an average of 0.79.It showed a statistical correlation between the main thoracic curve.(Figure 6)

DISCUSSION
The results of the study showed that the radiographic parameters of patients with idiopathic scoliosis did not correlate with the parameters measured in the photographs.
In statistics, the term correlation denotes the situation in which two variables vary.The correlation analysis employs the correlation coefficient ranging from -1 to + 1.These two extreme values express perfect inverse correlation (-1) and perfect positive correlation (+1), while zero means no correlation.The analysis of the correlation between the two variables is also presented, considering the significance  level (P).The P value tests the null hypothesis, which considers that the correlation between the two variables is zero.We observed in the sample studied that there is only a correlation between the axillary angle (AHA) and the proximal thoracic and main thoracic curves.No correlation was observed among the other variables compared.
In spinal deformities, trunk asymmetry greatly impacts patients' aesthetics, representing the main complaint of young patients, unlike adults with deformities, whose complaint is related to pain and functional disability. 11Although the photographic record of patients allows only a two-dimensional analysis of trunk asymmetry, this evaluation method has been reported to be reliable in demonstrating spinal asymmetries and deformities. 6,12,13ccording to Lenke's classification, the different types of scoliosis showed different types of trunk deformity. 14The parameters analyzed did not show similar behavior.The asymmetry of the upper trunk was not specific for the Lenke classification curve types, and the parameters related to the lower trunk best allowed the curves' discrimination. 14he discrepancy between the radiographic and the cosmetic evaluation of adolescent idiopathic scoliosis has been reported in the literature, 15,16 contrasting with reports showing a correlation between radiographic and cosmetic evaluation. 14,17he study aimed to observe the correlation between radiographic and photographic parameters.The correlation between the parameters from the photographs and the radiographic ones showed a statistical difference in the correlation coefficient in only one of the parameters analyzed.However, the study was conducted on a sample with a small number of patients to conduct a pilot study to explore this topic.The observed results indicate the discrepancy between cosmetic and radiographic evaluation, already mentioned in other studies.Therefore, the analysis of deformities should consider the aesthetic aspect, which does not correlate with the radiographic evaluation that is highly valued in the surgical approach to deformities.

CONCLUSION
The analysis of spinal deformities should be evaluated using clinical parameters related to the asymmetry of the trunk surface  references and radiographic parameters that evaluate angular measurements of the scoliotic curve.However, the results observed in this study indicate no statistical correlation between the parameters measured in the clinical evaluation and the angular parameters evaluated in the radiographs of patients with scoliosis.

Figure 2 .
Figure 2. Photograph illustrating the parameters evaluated in the photographs of patients 1. L.WA corresponds to the angle of the waist to the left.R. WA to the angle of the waist on the right.WHA angle between the apex of the right and left waist.AHA to the angle between the axillary folds.SHA to the angle between shoulder heights.

Figure 3 .
Figure 3. Graphs illustrating the correlation of the shoulder angle (SHA) and the radiographic parameters.No correlation was observed between the parameters studied.

Figure 4 .
Figure 4. Graphs illustrating the correlation of axillary angle (AHA) and radiographic parameters.
The asterisk indicates the statistical relationship between the values evaluated in the radiographs: proximal thoracic curve (PTC), main thoracic curve (MTC), thoracolumbar/lumbar curve (TL/L), T1 angulation (T1 TILT) and the values evaluated in the images: SHA: shoulder-height angle.AHA: angle between the axillary folds, WHA: angle between the apex of the right and left waist, Trunk Area.

Figure 5 .
Figure 5. Graphs illustrating the correlation of waist angle (WHA) and radiographic parameters.

Figure 6 .
Figure 6.Graphs illustrating the correlation of area and radiographic parameters.

Table 1 .
Values of the individual parameters evaluated.

Table 2 .
Analysis of the radiographic measurements.

Table 3 .
Analysis of the photographic measurements.

Table 4 .
Correlation coefficient values between the parameters selected for the study.