Pre-operative MRI Bone Texture is Associated with Complications After Spine Fusion

Bone quality is a critical determinant of outcomes following spine fusion surgery. However, patients with skeletal deficits are often only identified intra-operatively when there are unexpected problems with hardware placement, or post-operatively, when bone-related complications occur. MRI is routinely performed for diagnosis and surgical planning, making a technique based on clinical MRI ideal to opportunistically evaluate bone health. Texture analysis of trabecular bone from MRI has recently emerged as a biomarker of bone quality. This prospective study related pre-operative MRI-based texture from standard clinical images and areal BMD (aBMD) by DXA to post-operative outcomes following spine fusion. T1-weighted MR images were acquired from patients undergoing instrumented lumbar spine fusion. The following texture features were derived within the L1-L5 vertebral bodies to characterize the local distribution and spatial organization of signal intensity: contrast (variability) and entropy (disorder) for which higher values reflect high heterogeneity; angular second moment (ASM; uniformity) and inverse difference moment (IDM; homogeneity) for which higher values reflect low heterogeneity. Of 72 patients enrolled, mean age was 63 years (54% female). Patients were followed for a median of 13 months after surgery. Skeletal complications occurred in 27 patients (38%); proximal junctional kyphosis and screw loosening were the most common. There was no association between new complications and aBMD at any site. In contrast, patients who developed post-operative complications had greater pre-operative MRI-texture heterogeneity than those who did not, specifically higher contrast (26%) and entropy (12%), lower ASM (-13%) and IDM (-25%, p < 0.01 for all). In summary, greater heterogeneity of pre-operative bone texture from clinical MRIs was associated with skeletal complications following spine fusion. These findings support this novel method as an opportunistic screening technique to foster earlier identification of high-risk patients who would benefit from interventions to improve surgical outcomes.