Inter- and intraobserver reliability assessment of computed tomographic 3D measurement of pedicles in scoliosis and size matching with pedicle screws

Abstract

In patients with scoliosis, the morphology of the pedicles differs from those in normal spines. Preoperative three-dimensional information of these complex anatomic situations means a great advantage for the surgeon in order to assess which pedicles can be instrumented safely avoiding screw misplacement and for the decisions in choosing the appropriate screw size. The objective of this study was to measure pedicle dimensions in scoliotic spines on three-dimensional computed tomographic (CT) scans and to determine the intra- and interobserver reliability of the method. Additionally, the pedicles that cannot be instrumented safely by available screws were identified. All pedicles from T1 to L5 in 30 patients with scoliosis were measured by two independent observers. The pedicle width, height and length were assessed for two times with a 3-week interval. Intraclass correlation coefficients were used to determine the intra- and interobserver reliabilities. The diameters of the pedicles were matched with the dimensions of the smallest available pedicle screws of scoliosis implant systems. The intrarater reliability was strong in 60% of the parameters and moderate in 40%. The interrater reliability was strong in all parameters but in pedicle length. The concave-sided pedicles in the curve revealed to be smaller compared to the convex-sided ones. The highest percentage of size mismatch was found in levels T7 (31%) and T8 (33%). Measurement of pedicle dimensions on three-dimensional CT scans is a reliable but time-consuming procedure to assess pedicle dimensions. CT measuring should be reserved for special cases, where the anatomic situation remains unclear despite X-ray. In scoliotic spines, one-third of the mid-thoracic pedicles cannot be instrumented safely with pedicle screws.