Proton density (PD) and T₂-weighted images have been part of clinical MRI protocols for multiple sclerosis (MS) for many years. From these images, the quantitative transverse relaxation time (T₂) may be determined to serve as an absolute parameter to examine brain changes, rather than using weighted images. However, a simple two-point exponential fit is not appropriate and more advanced modelling is required [1]. We examine the feasibility of tracking disease progression and state in relapsing-remitting MS (RRMS) deep grey matter (DGM) over 7 years using retrospective standard clinical MRI obtained at 1.5 T, from which quantitative T₂, and brain volume changes are derived.

Imaging: Fourteen RRMS patients (age 36±7years, disease duration 9±3years and EDSS 3.0±0.9 after 7 years) were identified who underwent 1.5T MRI at baseline and 7-years follow-up scans after obtaining written consent according to institutional regulations. The clinical protocol consisted of PD and T₂-weighted fast spin echo (FSE) sequences, which were used to retrospectively calculate T₂. Typical scan parameters included: PD-weighted FSE sequence (TR/ESP/ETL/TE/N_av = 2000ms/6.5ms/5/13ms/2, refocusing flip angles 180° or 150°) and T₂-weighted FSE sequence (TR 4000ms or 3500ms, ETL 15 or 13, TE 91ms or 88ms, refocusing flip angles 180° or 150°, N_av 2). The clinical protocol also included: T₁-weighted sequence (TI/TR/ESP/TE/N_av = 2500ms/444ms/6.5ms/111ms/2, flip angle 80°). For each subject, 19 slices covering the whole brain were acquired (voxel size 0.45×0.45×5 mm³).

T₂ mapping: T₂ maps were computed on a pixel-by-pixel basis using a PD and a T₂-weighted image, and prior knowledge of flip angles [1]. This method extracts T₂ from the spin response from all echo pathways providing indirect and stimulated echo compensation (ISEC). Flip angle maps, averaged over the healthy subjects, were provided to the ISEC fitting, as previously described [1]. Additionally, T₂ maps were obtained from ISEC fitting of multiecho spin echo (MESE) data, and exponential fitting of two-echo data sets.

Method validation: Seven healthy controls (age 28±6years) were imaged with the same sequences as patients. Additionally, Flip angle maps were acquired using a double angle method [2] with correction for slice profile effects [1] from two spin-echo EPI images (TR/TE=7050ms/40ms, voxel size 1.87×1.87×2mm³) with excitation angles 60° and 120°. Since PD and T₂-wighted images of patients had different TR and ETL, a signal correction was made to the PD image, and the correction factor was experimentally determined from volunteers.

Analysis: Bilateral ROIs were chosen in iron-rich DGM. Global and regional brain volumes and atrophy were measured from 2D T₁-weighted images using FSL-SIENAX, and FSL-SIENA [3]. Statistical analyses were performed using SPSS. MS Severity Score (MSSS) were calculated from EDSS and disease duration [4].

Table 1 presents the mean T₂ values in DGM from healthy controls, obtained using the MESE-ISEC method including all echoes, two-point exponential fitting, and two-point ISEC method for FSE180° and FSE150°. The distribution of flip angles in DGM is similar between the subjects. Example PD and T₂-weighted images and the corresponding R₂ (1/T₂) map from an MS patient are shown in Fig. 1. Over 7 years, significant T₂ changes of 2-4% were observed when using two-point ISEC (Table 2), but no significant results were found when using the commonly misused standard exponential fit (not shown), which cannot account for imperfect refocusing. Multiple regression of change in T₂ of two DGM structures showed high correlation with final time point MSSS (r = 0.76, p < 0.05). The equation to predict disease severity: MSSS = -0.71*ΔT_2GP -0.48*ΔT_2CD +1.83; where ΔT_2GP and ΔT_2CD are the differences in T₂ over 7 years for globus pallidus and caudate respectively (Fig. 2). Changes in T₂ in globus pallidus correlated with change in EDSS score over time. Single time point as well as 7-year changes in T₂ in thalamus correlated with ventricular atrophy (r = -0.6, p < 0.05). We have retrospectively analysed 7-year T₂ changes obtained from standard clinical exams and relate these to disease severity and brain atrophy in MS patients. Though rate of percent change in T₂ over time is very low compared to that of high field [5], iron related change in DGM over time and its association with disease severity can be identified using two-point T₂ via ISEC at 1.5 T. Seven years difference measurements using two-point T₂ via ISEC from standard clinical MS exams correlated to disease severity in MS, as well as ventricular atrophy. Therefore, the long-term retrospective two-point T₂ and atrophy measurements at standard MRI can be used as markers to monitor disease course in DGM in MS.