MRI During Spinal Loading Reveals Intervertebral Disc Behavior Corresponding to Discogram Findings of Annular Fissures and Pain Provocation

Abstract

Study Design. Retrospective analysis of prospectively collected data. Objective. The aim of this study was to investigate whether spinal loading, depicted with magnetic resonance imaging (MRI), induces regional intervertebral disc (IVD) differences associated with presence and width of annular fissure and induced pain at discography. Summary of Background Data. Annular fissures play a role in low back pain (LBP) but cannot be accurately characterized with conventional MRI. Recently, annular fissures were suggested to influence different load-induced IVD behavior during MRI when comparing LBP-patients and controls. Thus, the loading effect could characterize behavior related to annular fissures noninvasively with MRI. Methods. Lumbar spines of 30 LBP-patients were investigated with MRI with and without loading, discography and CT. Five IVD regions were outlined on sagittal MRI images. Difference in normalized signal intensity (SI) with and without loading was calculated for each region. Eighty-three CT-discograms were graded regarding presence and width of fissures. Discograms were classified as pain-positive if a concordant pain response was obtained at a pressure <50 psi. Results. Comparing IVDs with outer fissures with IVDs without fissures, loading induced different behavior in the two ventral regions and in the posterior region. Higher SI increase in the central region was induced in IVDs with narrower fissures compared to IVDs with wider fissures. In the group of pain-negative discograms, a SI decrease was induced in the dorsal region whereas lack of such in the pain-positive group. Conclusion. The spinal loading-effect, depicted with MRI, reveals different regional behaviors between IVDs with outer fissures compared to those without, and between IVDs with narrow and broad fissures, as well as within posterior annulus between pain-positive and pain-negative discograms. Findings are of importance for future attempts to uncover phenotypes of painful IVDs. Level of Evidence: 2

Identification of potentially painful disc fissures in magnetic resonance images using machine-learning modelling

“…Blinded evaluation of CT-images of 10 patients was done by both the resident (after 3 months) and a senior radiologist to obtain measures for calculation of intra-and inter-rater reliability. An intra-rater agreement of 96% for labeling the fissure extension in the axial plane, grade 0-1 (p = 1.0) and 96% in the sagittal plane, grade 0-1 (p = 1.0) were found [15]. Also, the inter-rater agreement was found to be 100% for labeling the fissure extension in the axial plane (p N/A) and 100% in the sagittal plane (p N/A) for the current data set [15].…”

“…An intra-rater agreement of 96% for labeling the fissure extension in the axial plane, grade 0-1 (p = 1.0) and 96% in the sagittal plane, grade 0-1 (p = 1.0) were found [15]. Also, the inter-rater agreement was found to be 100% for labeling the fissure extension in the axial plane (p N/A) and 100% in the sagittal plane (p N/A) for the current data set [15].…”

“…The discograms were dichotomized into IVDs without (0) and with outer annular fissures (1), i.e. DDD ≥ 2/ DDD ≤ 1 and Adams ≥ D/Adams ≤ C, respectively [15]. Blinded evaluation of CT-images of 10 patients was done by both the resident (after 3 months) and a senior radiologist to obtain measures for calculation of intra-and inter-rater reliability.…”

Loading of the Spine in Low Back Pain Patients Does Not Induce MRI Changes in Modic Lesions: A Prospective Clinical Study

“…Furthermore, concordant pain has been shown to be induced by axial loading of the spine during MRI [ 28 ]. Previous findings of instantaneous load-induced regional behavior within the disc, with differences between LBP patients and controls, strongly suggest that annular fissures are associated with pain [ 19 , 29 , 30 ]. This, combined with a lack of image-based load-induced changes in MCs, probably reflects that MCs in general are not instantaneously influenced by such biomechanical stress.…”

Interpretation of Morphological Details of Nondegenerated Lumbar Intervertebral Discs on Magnetic Resonance Imaging: Insights From a Comparison Between Computed Tomography Discograms and Magnetic Resonance Imaging

“…Understanding how the MRI image correlates with detailed anatomy, such as the extent of the nucleus pulposus (NP) and details regarding the annulus fibrosus appearance, [2][3][4][5][6][7][8] is valuable to adequately interpret MRI data, especially since quantitative MRI markers are increasingly used within spinal research. [9][10][11][12][13] On T2weighted imaging (T2WI), the outer annulus fibrosus (OA) is readily apparent as a hypointense structure surrounding a central hyperintense zone. 14 The latter represents a combination of two structures that cannot be individually identified on MRI-namely the inner annulus fibrosus (IA) and the NP.…”