The associations between magnetic resonance imaging findings and low back pain: A 10-year longitudinal analysis

PURPOSE: To conduct a 10-year longitudinal analysis of the relationship between magnetic resonance imaging (MRI) findings and low back pain (LBP). MATERIALS AND METHODS: Ninety-one volunteers with a history of LBP, but without current LBP were recruited between 2005 and 2006. Participants’ baseline demographics and MRI findings were recorded. All volunteers were invited for a follow-up MRI in 2016; of these, 49 volunteers (53.8%) participated in the follow-up. We enquired whether they had LBP history during the 10 years between the baseline and follow-up examinations. Sagittal T1 and T2-weighted MRI were used to assess the intervertebral space from T12/L1 to L5/S1. We evaluated the presence of disc degeneration by Pfirrmann’s grading system, disc bulging, high intensity zone (HIZ), spondylolisthesis, and any type of Modic changes in the follow-up MRIs. We compared the follow-up MRI findings with the baseline findings; the progress of each finding over the 10 years were also compared between the groups with (n = 36) and without (n = 13) LBP. RESULTS: Average age of the study participants at follow-up was 44.8 years; 25 were female and 24 were male. Average age, sex, body mass index, and smoking habits of those who did and did not participate in the follow-up study, as well as the demographic characteristics of those who did and did not have LBP history during the 10 years, were not significantly different. Compared with the group without LBP history, the group that had LBP history during the 10 years did not have a significantly increased prevalence of disc degeneration, disc bulging, and HIZ in the follow-up and baseline MRIs. Spondylolisthesis and any type of Modic changes were also not associated with LBP history during the 10 years. CONCLUSIONS: Follow-up MRI findings consistent with Pfirrmann grading ≥4, disc bulging, HIZ, spondylolisthesis, and any type of Modic changes were not associated with LBP history during the 10 years between the baseline and follow-up study. The progresses of these findings were also not associated with the LBP history. In addition, baseline MRI findings were not associated with LBP history during the 10 years; therefore, our data suggest that baseline MRI findings cannot predict future LBP.