Novel MRI study for lumbar spine degeneration

Abstract

Introduction Disc degeneration has been conventionally assessed by T2-weighted (T2W) Magnetic Resonance Imaging (MRI) based on subjective signal intensity. This method, however, does not consistently correlate with histological and clinical profiles. Our group utilized novel MRI imaging sequences including ultra-short time-to-echo (UTE) MRI and T1-rho MRI to further assess disc degeneration in addition to traditional T2W MRI. With UTE MRI, a new imaging biomarker – the “UTE disc sign”(UDS) - was identified. With T1-rho MRI, proteoglycan profiles for intervertebral discs were constructed to objectively quantify the level of degeneration with respect to disc displacement. The associations between these novel findings, disc degeneration, other MRI phenotypes and clinical profiles were assessed.

Materials and Methods 108 Southern Chinese subjects were recruited (50% male; mean age: 52 years old) for T2W, T1-rho and UTE MRI of the lumbar spine from L1 to S1 levels (n=540 discs). T2W MRI was used to assess disc degeneration, disc displacement and other radiological phenotypes; T1-rho MRI was implemented to obtain quantitative proteoglycan disc profiles. The UDS was detected on UTE as a hyper- or hypo-intense band across a disc. Subject demographics, low back pain and disability profiles were obtained.

Results For UTE MRI, the UDS was noted in 25% of the subjects. 80% of the UDSs occurred at the lower lumbar levels (L3-S1). Subjects with the UDS had significantly more disc degeneration, disc displacement, spondylolisthesis, and Modic changes (p<0.001). T1-rho values were lower in UDS discs than non-UDS discs (p=0.022). The majority of UDS could not be detected on T2W MRI. The number of UDS disc levels correlated significantly with worse disability scores (r=0.303; p=0.013), compared with T2W MRI (r=0.234; p=0.057). Chronic low back pain was noted in individuals with multi-level UDSs (p < 0.05) For T1-rho MRI, the median T1-rho values for overall lumbar non-displaced discs was 77.6ms compared with 64.5ms for displaced discs (p<0.001). Based on ROC analyses, the area under curve was 0.69 (95% CI: 0.63-0.74) for the association of T1-rho values and overall lumbar disc displacement, with an optimal threshold value of 64ms.

Conclusions For UTE MRI, this is the first study in human to report "UDS" , which is a novel imaging biomarker that is highly associated with the radiological and clinical profile of patients with degenerative disc disease, and is a more sensitive biomarker than conventional T2W MRI. The UDS may provide further insight into the pathology and diagnosis of degenerative disc changes and low back pain. For T1-rho MRI, this is the first study in humans to quantitatively assess the “proteoglycan profile” of disc displacement, and the findings have indicated that a decrease in proteoglycan concentration was noted in the presence of disc displacement at all levels. Level-specific values have been identified that may have predictive utility in clinical practice.