Modic changes and its association with other MRI phenotypes in east Anatolian low back pain patients

Abstract

Objective

Modic changes (MCs) are known to be associated with low back pain (LBP). Literature contains conflicting reports about the prevalence of MCs and other spinal phenotypes among different populations with LBP patients. We aimed to evaluate the prevalence of MCs in the lumbar spine and associated features in Eastern Anatolian chronic LBP patients.

Methods

The study sample comprised of 786 consecutive patients [(490 female, 296 male), (mean age 39.7; range 20–78)] with a history of low back pain for at least 3 months. Data about MCs involvement, Schmorl’s nodes (SN), disc degeneration (DD), disc displacement, disc height and osteophytes were obtained via MRI. Patients’ demographic characteristics, Oswestry disability index (ODI) and visual analog scale (VAS) scores were assessed using a questionnaire.

Results

MCs were present in 67.2% (528/786) of the patients. Of all evaluated lumbar-level changes, 86 (6.2%) were Type I, 991 (76.8%) were Type II, 11 (0.8%) were Type III, 47 (6.5%) were Type II/III, 89 (8.5%) were Type I/II, and 29 (1.2%) were Type I/II/III MCs. MCs were significantly associated with severe DD (p < 0.001), disc displacement (p < 0.001), SN (p < 0.001), and osteophytes (p < 0.001). In the multivariate regression analysis, BMI (for both ODI and VAS) and age (only for ODI) were the only independent predictors of clinical severity.

Conclusions

The present study is the largest cross-sectional study of adult members of the Eastern Anatolian population with chronic LBP. Modic changes were detected in 67.2% of patients with chronic LBP and the prevalence of other phenotypic features differed significantly between MCs and non-MCs disc levels. Nevertheless, the results of the current study do not support a causal relationship of MCs or any MRI changes with clinical symptom severity.

Keywords:

• Modic changes
• low back pain
• disc degeneration
• magnetic resonance imaging
• Schmorl's nodes
• disc displacement

Acknowledgements

The authors thank Dr. Serdar Cevik for his help with data acquisition and analysis, and Dr. Sahin Hanalioglu for his help with the statistical analysis and the revision of the manuscript.

Disclosure statement

There is no conflict of interest to disclose.