CHEN Zi-qiang, ZHAO Yong-fei, HE Shi-sheng, WANG Chuan-feng, ZHANG Jing-tao, ZHAO Ying-chuan , YANG Chang-wei, LI Ming

Background Recent studies have demonstrated that the Lenke system is relatively efficient and consistent in classifying scoliosis curves. Basically, fusion should include the main curve and the structural minor curve. The criteria for defining the structural minor curve were established to help guide these decision-making process. The present study was designed to investigate predictors of the structural curve, and see whether it was possible to prevent the formation of the structural curve by interfering with influencing factors to decrease the fusion level.
Methods Age, gender, Cobb angle, Perdriolle rotation, Risser sign and the number of vertebrae included in the curve, brace treatment, and curve location were recorded in 145 idiopathic scoliosis patients from July 2001 to January 2007. The patients were divided into two groups: structural and non-structural groups. Demographics and baseline characteristics were compared between the two groups as an initial screen. Logistic regression was used to analyze factors affecting the minor curve to become the structural curve.
Results Compared with the non-structural group, the structural group had a higher Cobb angle ((51.34±13.61)° vs. (34.20±7.21)°, P <0.001), bending angle ((33.94±9.92)° vs. (8.46±5.56)°, P <0.001) and curve rotation ((23.25±12.86)° vs. (14.21±8.55) °, P <0.001), and lower flexibility ((33.48±12.53)% vs. (75.50±15.52)%, P <0.001). There was no significant difference in other parameters between the two groups. The results of the Logistic regression analysis showed that the Cobb angle (OR: 9.921, P <0.001) and curve location (OR: 4.119, P=0.016) were significant predictors of structural curve in adolescent idiopathic scoliosis. Every 10° change of Cobb angle increased the possibility of turning the minor curve into the structural curve by 10-fold. And thoracic curve showed, on the average, the possibility of becoming the structural curve about 4-fold more often than did the thoracolumbar/lumbar curve.
Conclusions Curve severity and curve location affect the minor curve’s structural features in adolescent idiopathic scoliosis.