Low back pain due to disc degeneration is a major cause of morbidity and health care expenditures worldwide. While stem cell-based therapies hold promise for disc regeneration, there is an urgent need to develop improved in vivo animal models to further develop and validate these potential treatments. The objectives of this study were to characterize a percutaneous needle injury model of intervertebral disc degeneration in the mouse caudal spine, and compare two non-invasive quantitative imaging techniques, microcomputed tomography and magnetic resonance imaging (MRI), as effective measures of disc degeneration in this model. Percutaneous needle injury of mouse caudal discs was undertaken using different needle sizes and injury types (unilateral or bilateral annulus fibrosus (AF) puncture). Mice were euthanized 4 weeks post-injury, and MRI and microcomputed tomography were used to determine T2 relaxation time of the NP and disc height index, respectively. Disc condition was then further assessed using semi-quantitative histological grading. Bilateral AF puncture with either 27 or 29G needles resulted in significantly lower T2 relaxation times compared to uninjured controls, while disc height index was not significantly affected by any injury type. There was a strong, inverse linear relationship between histological grade and NP T2 relaxation time. In this study, we demonstrated that quantitative MRI can detect disc degeneration in the mouse caudal spine 4 weeks following percutaneous needle injury, and may therefore serve as a surrogate for histology in longitudinal studies of both disc degeneration and cell-based therapies for disc regeneration using this model. © 2018 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 36:2771-2779, 2018.