The purpose of this study was to investigate the temporal evolution of type-1 end-plate changes on MRI in patients with degenerative disease of the lumbar spine and to evaluate whether any correlation exists between such evolution and the change in patients' symptoms. Forty-four patients with 48 Modic type-1 end-plate changes (low TI signal and high T2 signal) were studied. All patients had an initial and a follow-up non-contrast lumbar MRI with variable intervals between the studies (12-72 months). Severity of the end-plate changes was assessed by eyeball estimation. Correlation with patients' symptoms was studied with the help of the Visual Analogue Score (VAS), Oswestry Questionnaire Score (OQS) and patients' subjective assessment. Of the 48 disc levels with type-1 changes, 18 (37.5%) converted fully to type 2 (high T1 signal and intermediate to high T2 signal), 7 (14.6%) partially converted to type 2, 19 (39.6%) became worse (i.e. type 1 changes became more extensive) and 4 (8.3%) showed no change. Higher average VAS (5.7) and OQS (42.3) scores were noted in patients where there was worsening type-1 change and lower scores (3.8 and 27, respectively) were seen in those where there was conversion to type-2 change. These trends, however, did not reach statistical significance (P values 0.16 and 0.09 for VAS and OQS, respectively). The statistical relationship was stronger after exclusion of patients with confounding factors (i.e. changes in lumbar MRI other than end-plate changes that could independently explain the evolution of patients' symptoms) with P-values of 0.08 and 0.07 for VAS and OQS, respectively. Type-1 end-plate change represents a dynamic process and in a large majority of cases either converts to type-2 change or becomes more extensive. The evolution of type-1 change relates to change in patient's symptoms, but not to a statistically significant level.

With the methods presented here, MR imaging findings cannot predict ACI graft histologic features, and graft histologic appearance determined at biopsy was not related to graft signal intensity, graft thickness, overgrowth, surface smoothness, integration with adjacent cartilage or underlying bone, signal intensity change in underlying bone marrow, or underlying bone contour. Overgrowth and bone marrow changes underneath the graft were common.

No abstract

Reliance on MR imaging alone may lead to misdiagnosis. As the osteoid osteoma may be difficult to identify and the MR features easily misinterpreted, optimisation of MR technique is crucial in reducing the risk of missing the diagnosis. Unexplained areas of bone marrow oedema in particular require further imaging (scintigraphy and CT) to exclude an osteoid osteoma.

There was no significant asymmetry of the multifidus at spinal level above, same or level below the disc herniation. Instead, variations in muscle composition were observed, with greater fat infiltration on the side and at spinal levels adjacent to the disc herniation. Muscle asymmetry was not correlated with symptom duration.