RESUMO
PURPOSE: Aim of this study is to show the outcome of postoperatively measured lumbar lordosis in upright position in comparison to the intraoperatively estimated lumbar lordosis in prone position, as the lumbar lordosis is one of the most important factors for the clinical outcome after spinal fusion. MATERIALS AND METHODS: Eighty-two patients, receiving lumbar fusion were included in this retrospective study. Intraoperative radiographs were scanned. Then radiographs of the whole spine pre- and postoperatively, as well as 1 year after surgery were measured by a spine surgeon and a radiologist. The visible segment lordosis angles were measured and compared (L2-S1, L3-S1, L4-S1, L5-S1). In addition, the pelvic parameters pelvic incidence, pelvic tilt and sacral slope were measured pre- and postoperatively. RESULTS: The intraobserver reliability was almost perfect. The mean lordosis angle L4-S1 was 32.6° ± 7.8° intraoperatively and 29° ± 10.8° postoperatively. A linear correlation of these two measurements can be seen. In mean, the postoperative lordosis is 4° smaller than intraoperatively. This trend can also be seen in the level L3-S1. In levels L2-S1 and L5-S1 the postoperative values were slightly higher than intraoperatively, but without any significance. Also, 1 year after surgery there were no significant changes in global lumbar lordosis. CONCLUSION: Measuring lordosis angles intraoperatively resulted in almost the same values as measurements in standing plane radiographs postoperatively, despite prone position. These findings could especially be shown for the level L4-S1. The intraobserver reliability was almost perfect for both, intra- and postoperative measurements. In conclusion, the intraoperative measurement of a lumbar lordosis angle can perfectly predict the postoperative result.