Imaging Disc Endplates with DCE-MRI Reveals Modic Changes with Higher Sensitivity and Spatial Specificity

Abstract

Introduction One of the sequelae of intervertebral disc (IVD) degeneration is accompanying degenerative changes in adjacent subchondral bone (SB) of the vertebral bodies, which are classified as modic changes (MC). MC may represent vascularization and/or edema (Type I), proliferation of yellow fatty marrow (Type II), or sclerosis (Type III). Dynamic contrast-enhanced MRI (DCE-MRI) is a method to study blood perfusion in extracellular extravascular space (EES). Therefore, it could be sensitive especially to Type I MC. In this study, temporal and spatial characteristics of DCE-MRI in degenerative endplate changes were investigated. Materials and Methods This study was approved by the IRB and written consents were obtained from 31 participants (Age: median [min, max]: 36 [20–57 years]; 11 female patients; 23 asymptomatic controls, 8 patients with low back pain associated with degenerative disc disease). Images were acquired using a 3T MRI system with CTL-spine coil. DCE-MRI was acquired using dual-echo FSPGR (16-sagittal slices with 3 mm thickness, FOV = 31 cm, acquisition matrix = 310 × 300, TR = 4.0 ms, TE1 = 1.1 ms, TE2 = 2.2 ms, flip angle = 12 degrees, 23 frames with 28s frame rate). The contrast (Gd-DTPA 0.1 mmol/kg) was administered manually as a bolus via an antecubital vein at the end of the second dynamic frame. Conventional T2” and T1-weighted (T2W, T1W) images were also acquired and reviewed by two radiologists to assess MCs and disc degeneration. For analysis, an operator manually drew regions of interest (ROI) for endplates (EP) and adjacent SBs. Using these ROIs, planes parallel to the cranial and caudal faces of each IVD were generated and DCE-MRI intensity projection (IP) onto those planes were calculated for each EP and SB. A representative image slice with ROIs and projection planes (magenta dotted line) are shown in Fig. 1 . Results In three control patients, the radiologists reported two Type I MC (1.7% of controls' IVDs) and two (1.7%) Type II MC. In four patients, three Type I MC (12.5% of patients' IVDs) and one (2.5%) Type II MC were reported. Fig. 2 illustrates T1W, T2W, and average contrast enhancement projected onto EPs and adjacent SBs from two patients with Type I MC and one patient with Type II MC. Red areas in IP images show regions with high contrast agent uptake. Note that IP images are in oblique–axial orientation. [Figure: see text] Conclusion The results show that the SB regions with degenerative changes show much higher contrast agent uptake compared with normal SB regions. The enhancement was more prominent in Type I MC. Increased vascularization in Type I MC should lead to higher blood perfusion and edema should lead to increased accumulation of contrast agent. Interestingly, some Type II MCs had no noticeable enhancement while others showed some enhancement. It is possible that those enhancing endplates might be a mixed Type I/II. These preliminary results show that DCE-MRI might be a valuable tool to detect MC in earlier stages and also distinguish different phases of degenerative endplate changes with better specificity. Acknowledgment This study is supported by AOSpine research grant. [Figure: see text]