To evaluate the effect of preoperative rod shape on 3D spinal deformity correction.
Summary of Background Data.
Differential rod contouring is a surgeon-dependent process based on an estimate of 3D deformity correction sought and the flexibility of the spine. An objective measure to this otherwise subjective scoliosis correction technique is lacking.
A series of adolescent idiopathic scoliosis (AIS) patients with right, thoracic major curves, preoperative rod contour tracings, and EOS imaging was evaluated. All patients underwent posterior spinal fusion with 5.5 mm steel rods contoured prior to insertion. 3D reconstructions were generated pre- and postoperatively using sterEOS software (EOS Imaging) and imported into MATLAB (Mathworks) for analysis. A new measurement of the maximum perpendicular distance from the preinsertion concave rod contour to the preoperative 3D sagittal spinal reconstruction was defined as rod to 3D spine distance (RSD). Linear regressions were used to identify relationships between pre and postoperative parameters, including RSD and 3D thoracic kyphosis.
Ninety-nine patients were included. Average preoperative concave rod angle decreased from (48 ± 10°) preoperatively to 26 ± 6° postoperatively (P < 0.001) for an average flattening of ∼20°. Average convex rod angle increased from 30 ± 6° to 34 ± 5° (P < 0.001). Average preoperative thoracic curve magnitude and apical vertebral rotation were 57 ± 8° and 16 ± 5° and decreased to 11 ± 6° and 5 ± 6° (P < 0.001). Average 2D and 3D thoracic kyphosis increased from 19 ± 14° and 2 ± 12° to 27 ± 6° and 22 ± 5° (P < 0.001). Preoperative RSD strongly correlated with 3D thoracic kyphosis change (P < 0.001, r = 0.796, R2 = 0.633).
On average, the concave rod flattened ∼20° after connection to the spine. Similarly, kyphosis increased ∼20° following rod insertion. A strong correlation was identified between RSD and 3D thoracic kyphosis restoration. This novel 3D parameter can guide the degree of concave rod bend necessary to achieve a desired increase in thoracic kyphosis in AIS.
Level of Evidence: 2