Published in:
Open Access 01-12-2015 | Research article
Are the presence of MODIC changes on MRI scans related to “improvement” in low back pain patients treated with lumbar facet joint injections?
Published in: BMC Musculoskeletal Disorders | Issue 1/2015
Login to get accessAbstract
Background
Modic changes (MC) have been linked with low back pain (LBP) and worse outcomes from some treatments. No studies have investigated the impact that MCs may have on patient outcomes from lumbar facet injections. Therefore, the purpose of this study is to investigate whether the presence of Modic changes is related to ‘improvement’ in patients undergoing imaging-guided lumbar facet injection therapy.
Methods
Outcomes from 226 patients with MRI scans within 3 months of their imaging-guided lumbar facet injections were investigated to determine whether MCs are related to ‘improvement’ post injection. At 1 day, 1 week and 1 month post injection the Patients Global Impression of Change scale answers were collected by postal questionnaire. This was the primary outcome measure. The numerical rating scale for pain data was collected prior to treatment and at the same post injection time points. The MRI scans were independently evaluated by two examiners for the presence/absence of Modic changes and the type of Modic change if present. Kappa statistics were used for reliability of diagnosis analysis. Chi-squared test and logistic regression analysis tested MCs with ‘improvement’.
Results
Intra- and inter-examiner reliability for the diagnosis of MCs was Kappa = 0.77 and 0.74. Intra- and inter-examiner reliability for categorizing MCs was K = 0.77 and K = 0.78.
At 1 month post injection 45.2 % of patients without MCs reported clinically relevant ‘improvement’ compared to 34.2 % of patients with MC I and 32.1 % of patients with MC II. However, this did not reach statistical significance. Logistic regression found that Modic changes were not predictive of ‘improvement’.
Conclusions
There was a tendency for patients without MCs to have better outcomes but this did not reach statistical significance. The reliability of diagnosing MCs was substantial.