A Three-Year Follow-up of Lumbar Spine Endplate (Modic) Changes

Mari Kuisma, MD; Jaro Karppinen, MD, DMSc; Jaakko Niinimäki, MD; Mauno Kurunlahti, MD, DMSc; Marianne Haapea, MSc; Heikki Vanharanta, MD, DMSc; Osmo Tervonen, MD, DMSc


Spine. 2006;31(15):1714-1718. 

In This Article


The study population consisted of 60 unoperated sciatica patients (21 women and 39 men; mean age, 45 years; range, 23-76 years) with unilateral pain below the knee lasting from 3 weeks to 6 months. They had been referred to a randomized controlled trial, which evaluated the efficacy of periradicular nerve root infiltration for sciatica.[5] At 3 years (mean, 3.0 years; range, 2.6-4.0 years), these patients were called for a presceduled follow-up MRI assessment. The study protocol was approved by the ethical committee of Oulu University Hospital.

MRI scans were obtained at baseline and at 3 years with a 1.5-T imaging system (Signa, General Electric, Milwaukee, WI). At baseline, the imaging sequences included T2-weighted sagittal fast spin-echo (FSE) images with 4,000 milliseconds repetition time (TR) and 95 milliseconds effective echo time (effTE), echo train length (ETL) was 16, and T1-weighted axial spin-echo (SE) images with a TR/TE of 640/14 milliseconds. At 3 years, the imaging sequences included T2-weighted sagittal FSE images with a TR/effTE of 4,000/95 milliseconds, ETL 16 and T1-weighted sagittal SE images with a TR/TE of 400/14 milliseconds. The technical specifications included a slice thickness of 4 mm with interslice gaps of 1.0 and 0.5 mm, a field-of-view of 36 and 20 cm, matrix of 256 × 128 and 256 × 192, and number of excitations of 1 and 2 for sagittal and axial images, respectively.

MRI image analysis was performed at a workstation by comparing baseline and 3-year follow-up MR images from L1-L2 through L5-S1. MRI findings analyzed included classification of the lumbar vertebral endplate and subchondral bone marrow changes, and of intervertebral disc degeneration. Disc degeneration was graded as normal (no signal changes), Grade 1 (slight decrease in signal intensity of the nucleus on T2-weighted images), Grade 2 (hypointense nucleus pulposus on T2-images with normal disc height), and Grade 3 (hypointense nucleus pulposus on T2-weighted images with disc space narrowing).[6,7] Herniations were classified as contained herniations (focal extrusion of disc material through the anulus but not through the posterior longitudinal ligament) or extrusions (extrusion of disc material through the posterior longitudinal ligament).[7] Symptomacy of disc herniations was evaluated in the previous trial.[5]

The endplate and subchondral bone marrow (Modic) changes were graded into Types MI, MII, and MIII as previously described, and mixed Types MI/II and MII/III.[2,3,4] In addition to classification into different types, the involvement of one or both endplates, anteroposterior localization, maximal vertical depth (mm), and extent of Modic changes were also analyzed. Anteroposterior localization was defined as anterior, posterior, or centrally located. Centrally located changes also included Modic changes that continued the whole anteroposterior length of the disc margin. If there were Modic changes at both superior and inferior endplates, the two vertical distances were added for a sum score (mm). The extent of changes was estimated from sagittal or axial sequences as quadrants of the endplate area (1%-25%, 26%-50%, 51%-75%, or >76%; Figure 1).

Baseline MR images of a 33-year-old woman with Modic Type II changes (hyperintensity in both T1 and T2 sequences) demonstrate the classification criteria used: A, Sagittal T2-weighted image shows Type II change at both sides of the disc. Maximal vertical depth is 30 mm (a + b). Modic Type II change continues the whole anteroposterior length of the disc margin. B, Axial T1-weighted image shows changes involving 51% to 75% of the endplate area. There is Grade 3 disc degeneration at L5-S1.

The baseline and 3-year follow-up MR images were observed independently by two radiologists. A consensus reading was performed in cases of disagreement, and kappa statistics were used to analyze agreement between the observers. Interreader reliability is the mean of the kappa statistics for the baseline and follow-up imagings from L1-L2 to L5-S1. Intertester reliability was calculated similarly. Characteristics of Modic changes were illustrated by frequency tables and cross-tabulations. The paired t test and Wilcoxon signed ranks test were used to analyze the progression in size of Modic changes. Effects of physical work load and age were assessed with cross-tabulations and univariate analyses of variance. Physical workload was categorized as informative (sedentary), mixed (informative and physical), or physically demanding. P values less than 0.05 were considered statistically significant. SPSS software 11.0.1 was used to conduct analyses.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.