Radiographic predictors of residual low back pain after laminectomy for lumbar spinal canal stenosis: minimum 5-year follow-up.

STUDY DESIGN

Retrospective study of patients who underwent laminectomy for unification.

OBJECTIVE

To identify radiographic predictors of residual low back pain (LBP) after laminectomy for lumbar spinal canal stenosis (LCS).

SUMMARY OF BACKGROUND DATA

Residual LBP is a common complication of laminectomy and no radiographic predictors of its occurrence have been identified previously.

METHODS

Clinical results and radiographic findings in 49 patients (21 males and 28 females, minimum 5-year follow-up) who underwent single level laminectomy for LCS were retrospectively reviewed. Patients who had an improvement in LBP scores in the Japanese Orthopedic Association (JOA) scoring system during the follow-up periods were classified as the recovery group, whereas those without improvements were classified as the nonrecovery group. Patients' clinical data (sex, duration of symptoms, age at surgery, JOA scores) and radiographic parameters (including lumbar lordotic angle, lumbar range of motion (ROM) and the intervertebral rotational angle) were analyzed to detect the factors significantly related with the occurrence of residual LBP.

RESULTS

The average preoperative JOA score of 14.8+/-5.1 points improved to 21.6+/-5.5 points at the final follow-up providing an average recovery rate of 48.1+/-36.8%. Thirty-four and 15 patients were classified into the recovery and the nonrecovery groups, respectively. Binary logistic regression analysis revealed that significant predictors of residual LBP were preoperative lumbar lordosis angle and lumbar ROM. The mean preoperative lumbar lordosis and ROM in the nonrecovery group were significantly smaller than those in the recovery group (lordosis: 25.3+/-15.8 degrees vs. 37.8+/-13.6 degrees, P=0.006 and ROM: 22.1+/-10.6 degrees vs. 31.2+/-9.9 degrees, P=0.006). In addition, increase of the postoperative lumbar ROM was significantly larger in the nonrecovery than that in the recovery group (P=0.009).

CONCLUSIONS

Our results indicate that preoperative lordosis angle and lumbar ROM were the significant radiographic predictors for residual LBP after laminectomy for LCS. Patient with flatback and limited lumbar mobility before surgery are prone to suffer residual LBP. It is suggested that these sagittal radiographic parameters should be taken into account when choosing laminectomy as the surgical option for LCS.