OBJECTIVETo compare the precision of the vertebral screw insertion in treating idiopathic thoracic scoliosis by thoracoscopy with that by mini anterior thoracotomy.

METHODSTen patients with an average Cobb angle of 52.9 degrees were operated thoracoscopically (group A), and twenty-one patients with an average Cobb angle of 45.4 degrees were operated by mini anterior thoracotomy (group B). In order to determine the accuracy of screw placement, several parameters were measured postoperatively on each instrumented level of CT scan, including the entry point and entry direction of screw, the relationships between the screw tip and aorta as well as spinal canal. The safety of screw placement was determined at the position of screw to the vertebral canal and the aorta, and the results of CT analysis were designated as; D, the screw tip was distant from the aorta (> or = 1 mm); A, the screw tip was adjacent to the aorta (< 1 mm); C, the screw tip was felt to be against the aorta and creating contour deformity. The screw which had a bi-cortical purchase but was distant from the aorta and vertebral canal was defined as a satisfied screw.

RESULTS73 screws were inserted into patients of Group A, and 162 into patients in Group B. Parameters measured on CT image and X-ray showed no difference between these two groups (P > 0.05). There is no difference (P > 0.05) between A and B group of percentages in D (89.0% vs. 80.2%), in bi-cortical purchase (89.0% vs. 87.0%), and in satisfaction of screw placement (74.0% vs. 66.0%).

CONCLUSIONThoracoscopic and mini-open thoracotomic anterior correction are safe and effective to correct idiopathic right thoracic scoliosis with satisfied correction. Vertebral screws placed by both techniques have the same satisfied accuracy, safety, bi-cortical purchase and the satisfaction rate either in total screws or in the corresponding area.