The Position of the Aorta Relative to the Spine Before and After Anterior Dual Rod Instrumentation and the Precision of Anterior Screw Instrumentation in Idiopathic Right Thoracic Scoliosis

Background: There have been recent reports on a critical proximity of screw tips to the thoracic aorta in anterior single rod instrumentation of thoracic curves. While there exist data on the topography of the thoracic aorta in straight and in scoliotic spines, there is a paucity in the current literature on the changing topography of the thoracic aorta after anterior instrumentation in idiopathic scoliosis. The purpose of this study was to analyse the anatomical relationship between the aorta and the vertebral bodies in idiopathic right thoracic scoliosis before and after anterior dual rod instrumentation. Furthermore, the accuracy of the screws in relation to the spinal canal and the thoracic aorta was evaluated.

Methods: 15 patients with idiopathic adolescent right thoracic scoliosis have been included into the study. In all patients the primary thoracic curve was instrumented through a standard open approach using an anterior dual rod system, usually from end- to endvertebra. Preoperatively, axial thin slice T1-weighted magnetic resonance images (MRI) from the fifths thoracic to the first lumbar vertebra were required. After surgery each screw was imaged using sequential Computer Tomography (CT) scans with an angled gantry after positioning the patient on the left side. In the corresponding MRI and CT images the length and width of the vertebra, diameter of the aorta, the distance from the aorta to the closest point of the vertebral body cortex, and the aortavertebral angle were evaluated. In the postoperative CT the closest distance between the screw tips and the aorta and the amount of the of contralateral cortex penetration of the screws were measured. All measurements were performed on an on-line workstation by an orthopedic spine surgeon and a musculoskeletal radiologist in consensus. A total amount of 161 screws were evaluated.

Results: At the apex the closest distance between aorta and vertebral body changed significantly from 5.9mm (2.2-11.6mm) preoperatively to 3.4mm (1.4-5.1mm) postoperatively. The pre- to postoperative differences at the endvertebra, however, were only minimal. Preoperatively the aorta is positioned more laterally and posteriorly than postoperatively. At the curve apex aorta-vertebra angle changed from 70,7° (52-97°) preoperatively to 42,2° (27-68°) postoperatively. 96 of the 105 more posterior screws (91%) and 50 of the 56 more anterior screws (89%) had a bicortical purchase. The more anterior screw perforated the contralateral cortex by on average 3.2mm (-3.7-7.5mm) and the more posterior screw by 2.4mm (-2.4-5.6mm). The closest proximity of the screw tips to the thoracic aorta was found at the upper endvertebra (T5 or T6). At this level the distance between the more posterior screw tip and the aorta averaged 6.2mm (1.8-9.4mm) and the distance between the more anterior screw tip and the aorta averaged 4.4mm (1.1-7.4mm). The closest distance between the posterior screw and the spinal canal ranged between 0mm and 10.3mm (4.0mm on average), there were no screw perforations into the spinal canal.

Conclusions: Anterior correction of thoracic curves results in an anteromedial migration of the aorta towards the vertebral bodies, especially at the curve apex. Excessive bicortical screw perforation should be avoided in order not to endanger the thoracic aorta.