Radiographic evaluation of ventral instability in lumbar spondylolisthesis: do we need extension radiographs in routine exams?

To determine the usefulness of acquiring extension radiographs for the evaluation of the degree of spondylolisthesis. Routine radiographs of the lumbar spine were retrospectively evaluated in 87 patients (mean-age 63, range 32-86) by two independent radiologists. All patients received radiographs in standing neutral, flexion and extension position. Vertebral body depth, sagittal translational displacement and lordosis angle were measured and slip percentage (SP) was calculated on standing neutral, flexion and extension radiographs. Statistical analysis was performed with a two-sided t test. Inter- and intraobserver reliability was assessed using the kappa-coefficient. There was no statistically significant SP-difference between neutral standing and extension images. Ventral instability was diagnosed in 25-34 % (cut-off > 8 % SP-difference) for neutral versus flexion comparison. The detection rate of flexion-extension radiographs representing the extremes of motion was lower with 15-22 %. Inter- and intraobserver reliability was good to excellent. Slip percentage in routine standing extension radiography ultimately does not differ from that obtained in a static neutral standing view. Extension radiography may therefore be omitted in a routine work-up of ventral instability in lumbar spondylolisthesis.