T2 mapping of the IVDs showed promising results in previous studies.(1-4) Even histological examination showed to be coherent.(5) The purpose of this study was to determine diagnostic and predictive additional value of T2 mapping in assessing the IVDs, which play a leading part in disc degeneration and pathology. In this prospective study, T2 values of the IVDs were assessed via ROI analysis and compared with new diagnoses, RMDQ and VAS at follow-up to identify unfavorable T2 values of the disc as a means of prospective patient assessment and potential risk factor.Twenty-five symptomatic patients (mean age 39 years, range 18-60 at baseline; 13 male, 12 female) with low back pain and without radicular symptoms or previous surgery were assessed at baseline and follow-up. All patients were examined using a 3T MR scanner (Tim Trio, Siemens Healthcare) and a spine array coil. A sagittal multi-echo spin-echo sequence was used for T2 mapping at baseline and follow-up (six echoes; TE 13.8, 27.6, 41.4, 55.2, 69.0, 82.8 ms; TR 1200m; slice thickness 3mm, in plane resolution 256x256). For morphological imaging, standard sagittal T1- and sagittal, axial and coronal T2-weighted sequences were used.Radiological reports were compiled, morphologic grading according to Pfirrmann et al. conducted and questionnaires (VAS and RMDQ) surveyed in order to compare routine-based diagnostics, morphological grading, subjective patient estimation and quantitative imaging at the same time. Five rectangular equally-sized regions-of-interest were used for T2 value assessment on two central slides of the T2 maps. Unpaired t-tests were used for comparison of T2 values and new herniation, diagnoses and clinical parameters. Statistical analysis was done using SPSS 22.0 (SPSS Inc., Chicago, IL, USA).

From baseline to FU, the mean NP T2 values decreased from 109.0 ms (±32.4) to 103.5 ms (±37.2) whilst the mean AF T2 values AF increased from 57.8 ms (±10.7) to 58.7 ms (±11.3).Most significantly, there was a major difference (37.05 ms mean difference) between baseline T2 values of the NP grouped by new herniation at follow-up (p = .002) as seen in Fig. 1. Neither the mean AF (p = .634) nor the posterior AF region alone (p = .488) differed significantly when grouped by herniation at FU. The Pfirrmann score at baseline grouped by new herniation at follow-up also differed significantly between groups (p≤.001). The Wilcoxon Signed-Ranks Test indicated that the Pfirrmann score was significantly higher at follow-up than at baseline (Mdn=2.00 vs. 3.00; Z=-5.962; p≤.001).Spearman correlation coefficient was significant for baseline T2 values with Pfirrmann score at baseline (r=-.809; p≤.001) and FU (r=-.766; p≤.001) as illustrated in Fig. 2. Additionally, there was a significant difference regarding baseline T2 values grouped by new annular tear at FU for the posterior AF (p≤.001), NP (p≤.001) and age (p=.002). Interestingly, the posterior AF showed significant correlation, whilst the posterior 10 % (p=.072) and the mean AF T2 value (p=.882) did not. There was a significant difference for the NP T2 values grouped by new Modic marrow changes (p=.012) and foraminal and spinal stenosis (p=.015) at FU. An illustrated patient case can be found in the appendix (Fig.3, Fig.4, Table1).The minimal NP T2 value per patient showed a significant Spearman correlation coefficient with the respective RMDQ (r=-.517; p=.008), age (r=-.715; p≤.001), pain present (r=-.495; p=.012), pain past week (r=-.494; p=.012), pain duration over the past 5 years (r=-.442; p=.027), emitted leg pain duration over the past 5 years (r = -.417; p = .038), walking (r = -.482; p = .015), standing (r=-.468; p=.018), climbing stairs (r=-.513; p=.009) and sitting (r=-.414; p=.040). The posterior AF and the outermost posterior 10% did not show any of the mentioned significant correlations.

Our data suggests, that T2 mapping provides predictive potential and should be subject to future studies examining quantitative imaging for prospective patient assessment and risk prediction.