Accuracy of photogrammetry to detect thoracic changes in individuals with adolescent idiopathic scoliosis

Abstract Total radiography of the spine in patients with adolescent idiopathic scoliosis (EIA) is the gold standard for classifying curve severity by the Cobb method. These patients often require X-rays, thus being exposed to increased radiation. Therefore, studies have addressed alternatives for identifying vertebral alterations, as well as computerized photographic methods or photogrammetry. Thus, this study aimed to verify the accuracy of photogrammetry in detecting thoracic changes in individuals with AIS. A cross-sectional and all-time study was performed on patients diagnosed with Lenke I EIA and healthy adolescents. All patients had pictures taken of their whole body, with anatomical markers positioned at previously studied points. Roc curves were determined and the area under the curve was calculated to identify the best measurements of angles and distances in the Postural Analysis Software (SAPO). The combined points of greater sensitivity and specificity were set as the cutoff point for detecting postural changes. 78 volunteers were assessed, divided into two groups: scoliosis (n=47) and control (n=31). The angles and distances measured by SAPO showed differences between the groups. The angles and distances A2, A4D/A4E, A5D/A5E, A7, and D1D/D1E reached an area under the ROC curve above 0.7. In conclusion, the use of photogrammetry can benefit assessments by identifying thoracic alterations and screening the progression of the scoliosis curve, since the angles and relationships have good accuracy in identifying Lenke I scoliosis.