Objective. To create a multifactorial model for predicting the risks of developing clinically significant frontal imbalance in surgical treatment of severe idiopathic scoliosis based on the identification of predictors influencing the main clinical parameters of trunk asymmetry.Material and Methods. The results of surgical treatment of 288 patients with severe forms of idiopathic scoliosis with a primary thoracic scoliotic curve of types 1, 2, 3 according to Lenke (mean 97.6° ± 15.5° according to Cobb) who underwent surgery in 1999–2019 using posterior segmental instrumentation with hook, hybrid and transpedicular fixation, were analyzed. There were 243 female (84.4 %) and 45 male patients (15.6 %). The average age of patients at the time of surgery was 15.3 [10–39] years. The average postoperative follow-up period was 3.5 [2.0–19.5] years. The analysis included clinical and radiological data obtained in the preoperative, postoperative and late postoperative periods. Predictors of the occurrence of frontal imbalance (the distance from the plumb line to the navel and intergluteal fold more than 15 mm, the tilt of the shoulder girdles more than 5° and the tilt of the scapula more than 15°) were identified by building single- and multivariate logistic regression models.Results. In the total cohort, 41 (14.2 %) patients with clinically significant frontal imbalance were identified, including 10 (3,0 %) – with an increase in the distance from the plumb line to the umbilicus of more than 15 mm, 12 (4.2 %) – with an increase in the distance from the plumb line to the intergluteal fold of more than 15 mm, 8 (2.8 %) – with a shoulder girdle tilt of more than 5°, and 11 (3.8 %) – with a scapular tilt of more than 15°. A significant predictor of the risk of developing frontal imbalance was determined as postoperative thoracic scoliotic curve of more than 63°. Multiplicative predictors of the risk of frontal imbalance were identified: postoperative increase in the distance from the plumb line to the umbilicus by more than 15 mm and a tilt of the shoulder girdles by more than 5° with a sensitivity of 88.9 % and 100.0 %, and a specificity of 89.5 % and 100.0 %, respectively (p < 0.001).Conclusion. Identification of multiplicative predictors of the risk of frontal imbalance allows predicting the risk of increasing the distance from the plumb line to the navel by more than 15 mm and the risk of shoulder girdle tilt by more than 5°. To eliminate the risk of frontal imbalance, it is necessary to strive for maximum correction of the thoracic scoliotic curve. When planning surgical treatment using transpedicular fixation for the correction of severe thoracic scoliosis, it is necessary to take into account the patient’s gender and the presence of concomitant neurosurgical, cardiological and pulmonological pathology to prevent shoulder girdle imbalance.