Lumbar high-intensity zones on MRI: imaging biomarkers for severe, prolonged low back pain and sciatica in a population-based cohort

Abstract

Introduction: High-intensity zones (HIZs) of the lumbar spine can be visualized on magnetic resonance imaging (MRI). Their clinical relevance has been scrutinized and regarded as purely an incidental finding. However, previous studies largely consisted of small sample sizes, patient-based cohorts, and inadequate analysis of its association with other MRI phenotypes (e.g. disc degeneration, disc displacement, Modic changes) and the clinical profile. The purpose of this study was to assess the relationship of lumbar HIZs on MRI with LBP, sciatica, and back-related disability in a cross-sectional, population-based cohort study of Southern Chinese volunteers. Methods: A total of 1,214 subjects with T2-weighted 3T sagittal MRI of L1-S1 were assessed. Presence, single level, multilevel homogeneous (HIZs of the same morphological type and location) and multilevel heterogeneous (HIZs of variable morphological types and location) HIZs based on the authors’ classification scheme according to disc level, shape (round type, fissure type, vertical type, rim type, and enlarged type), and location of HIZs within the disc (posterior or anterior) as well as other MRI phenotypes were assessed at each level. Associations with prolonged/severe LBP (LBP lasting at least 30 days with the worst experience at least VAS 6), sciatica (sciatica lasting at least 30 days) and back-related disability (Oswestry Disability Index (ODI) scores ≥15%) were correlated with HIZ profiles. Results: There were 718 individuals with HIZs (59.1 %). Of the 718 HIZ subjects, 53.3% had single HIZ (n=383), 32.5% had 2 HIZs (n=233), 11.1% had 3 HIZs (n=80) and 3.1% had 4 HIZs (n=22). For the 335 multi-level HIZs subjects, 40.0% had homogeneous HIZs (n=134) and 60.0% had heterogeneous HIZs (n=201). Disc degeneration and disc displacement were more prevalent in individuals with HIZ (p<0.001). Based on adjusted multivariable regression analyses, individuals with homogeneous or heterogeneous HIZs were significantly associated with LBP (odds ratio (OR) range: 1.53 to 1.57; p<0.05). Individuals with homogeneous HIZs had a higher risk of sciatica (OR: 1.51, p<0.05), whereas other HIZ variables were not. No significant association was observed between HIZs and back-related disability (p>0.05). Conclusions: This is the first large-scale study to note that HIZs of the lumbar spine are clinically-relevant imaging biomarkers that are independently and significantly associated with prolonged/severe LBP and sciatica. Furthermore, various patterns of HIZs appear to be more related to symptoms. HIZs are important lumbar phenotypes that should be noted in the global imaging phenotype assessment of the spine, which may have immense clinical utility.