Immediate improvements in side-to-side weight bearing and iliac crest symmetry after manipulation in patients with low back pain.

Background: Although there is good evidence that spinal manipulation is an effective treatment to improve pain and function for patients with low back pain (LBP), there is little evidence to support the mechanism by which manipulation works.

Objectives: To determine if iliac crest (IC) and weight-bearing (WB) symmetry improve after spinal manipulation and to determine if improvements in IC and WB symmetry are associated with improvements in pain and function in patients with low back pain.

Design: Single group, within-subjects, repeated measures design.

Method: Thirty consecutive patients (mean age = 40+/-13) who came to a spine specialty center for treatment of acute or chronic LBP and who would be receiving spinal manipulation participated in the study (14 male patients). Patients completed a series of self-report measures of pain and function and received a standardized physical examination, including the assessment of IC and WB symmetry. Patients received a standardized manipulative intervention, and immediate and 3- to 4-day follow-up examinations were performed by a blinded examiner. Paired t tests were performed to determine within-group changes, and Pearson product moment correlation coefficients were calculated to determine the relationship between improvements in IC and WB symmetry and improvements in pain and function. To control for the potential that an association between changes in IC and/or WB symmetry and changes in pain and function could be confounded by the baseline outcome measure, simultaneous linear regression was performed on any significant correlation. Partial F tests were used to determine if the additional explained variability was significant.

Results: Patients with LBP demonstrated significant improvements in IC and WB symmetry after manipulation (P<.001). Improvements in WB symmetry were associated with improvements in the patients' self-reported levels of pain 3 to 4 days after manipulation (r=.5, P=.007). Based on the significant association between improvements in WB symmetry and improvements in pain, the final pain score was regressed on the change in WB symmetry, after controlling for the baseline level of pain. The addition of the change in WB symmetry explained a 10% additional increase in variability in the patient's level of pain at the 3- to 4-day follow-up (P =.01). No relationship was found between improvements in IC and WB symmetry and improvements in function as determined by the Oswestry Disability Questionnaire 3 to 4 days after manipulation.

Conclusion: IC and WB symmetry improved immediately after spinal manipulation. Improvements in WB symmetry were related to improvements in the patients' self-reported levels of pain, even after controlling for the baseline level of pain. Improvements in IC and WB symmetry were not related to changes in function. The results of this study provide initial data to elucidate how manipulation may work to improve pain and function in patients with LBP.