Two subtypes of intervertebral disc degeneration distinguished by large-scale population-based study

Abstract

Background ContextLumbar disc degeneration (LDD) is a major cause of low back pain, and is a common and disabling condition worldwide. It has been defined and measured by multiple spine magnetic resonance imaging (MRI) features, but the heterogeneity among them has never been fully addressed. PurposeThis study examined the intercorrelations, risk factor associations, and single nucleotide polymorphism (SNP) heritabilities of lumbar disc MRI features in a large-scale sample to classify the different intervertebral disc phenotypes associated with LDD. Study DesignA cross-sectional study was conducted consisting of 2,943 volunteers of Southern Chinese origin (mean age: 41.1 years; range: 15–55 years; 59.6% women). Outcome MeasuresThe outcome measures were MRI phenotypic spinal patterns and their risk factor profiles in relation to developmental or degenerative origins of disc degeneration. MethodsSagittal T2-weighted MRI of the lumbar spine from L1 to S1 was assessed. The MRI features of lumbar intervertebral disc changes, such as disc signal intensity loss and disc bulges or extrusions, as well as additional imaging phenotypes of end plate changes, high-intensity zones, and bone marrow changes, were evaluated. Blood samples were taken for genotyping using the HumanOmni-ZhongHua-8 BeadChip. Subject demographics, environmental, and lifestyle factors were assessed by questionnaires. Multivariate statistical techniques were used for phenotype evaluation. Polychoric correlations and local regression statistical analyses were performed. The genetic components contributed by common SNPs were estimated by comparing genetic correlations and phenotypic correlations using the Genome-Wide Complex Trait Analysis (GCTA) tool. ResultsThe study noted that lumbar disc MRI features separated into two groups with differential patterns of risk factor associations. A subset of lumbar disc abnormalities, including end plate changes but also upper lumbar disc bulging and signal intensity loss, may have a developmental origin. Subsequent degenerative changes, typically affecting the lower lumbar discs, then emerge as individuals age and are associated with body mass index. ConclusionsThis is the first large-scale study to identify two distinct patterns of lumbar disc alterations, noting degenerative changes and a possible developmental component affecting the lumbar spine. This new classification provides a starting point for a more homogeneous phenotype definition, which may provide greater statistical power and precision in future genetic and epidemiologic studies. In addition, such insights may have direct clinical implications in the prevention, therapeutics, and prognostics of patients with disc degeneration.