Multiple sclerosis lesions and irreversible brain tissue damage: a comparative ultrahigh-field strength magnetic resonance imaging study


  • T. Sinnecker
  • P. Mittelstaedt
  • J. Doerr
  • C.F. Pfueller
  • L. Harms
  • T. Niendorf
  • F. Paul
  • J. Wuerfel


  • Archives of Neurology


  • Arch Neurol 69 (6): 739-745


  • BACKGROUND: In current clinical practice, T2-weighted magnetic resonance imaging (MRI) is commonly applied to quantify the accumulated multiple sclerosis (MS) lesion load, whereas T1-weighted sequences are used to differentiate edema, blood-brain barrier breakdown by contrast enhancement, and irreversible brain tissue damage (commonly called "black holes" owing to the loss of signal intensity in T1-weighted sequences). Black holes are histopathologically associated with axonal loss and severe tissue destruction. In addition, double inversion recovery techniques were developed to improve the sensitivity to cortical lesions. OBJECTIVE: To demonstrate the potential of ultrahigh-field 3-dimensional T1-weighted imaging using magnetization-prepared rapid acquisition and multiple gradient-echoes (MPRAGE) to detect and characterize white and gray matter pathology in MS. DESIGN: Comparative study. SETTING: The patients with MS were recruited from the outpatient clinics of the NeuroCure Clinical Research Center and underwent 7-T brain MRI at the Berlin Ultrahigh Field Facility, both of which are in Berlin, Germany. Patients  Twenty patients with relapsing-remitting MS and 14 healthy controls underwent 7-T brain MRI, using a 24-channel receive head coil, and a subgroup of 18 patients with relapsing-remitting MS also underwent 1.5-T brain MRI. The imaging protocol included 2-dimensional T2-weighted fast low-angle shot (FLASH) and turbo inversion recovery magnitude (TIRM) sequences. For 3-dimensional T1-weighted imaging, the MPRAGE sequence was used. Each sequence was initially examined independently in separate analyses by an investigator blinded to all other data. In a second study, all detected lesions were retrospectively analyzed in a side-by-side comparison of all sequences. RESULTS: By use of 7-T T2-weighted FLASH imaging, 604 cerebral lesions were detected in the patients with relapsing-remitting MS (mean, 30.2 lesions per patient [range, 2-107 lesions per patient]), but none were detected in healthy controls. Cortical pathology was visible in 10 patients (6 cortical lesions and 37 leukocortical lesions). Within the 7-T acquisitions, each lesion detected at T2-weighted sequences and/or double inversion recovery sequences was also clearly delineated on corresponding MPRAGE sequences in side-by-side analysis. However, at 1.5 T, the MPRAGE images depicted only 452 of 561 lesions visualized in T2-weighted sequences and/or double inversion recovery sequences. In contrast, when analyzing each sequence separately, we found that the 7-T MPRAGE depicted more lesions than the 7-T FLASH (728 lesions vs 584 lesions), and almost twice as many as the 1.5-T MPRAGE (399 lesions). The 7-T MPRAGE also improved the detection of cortical and leukocortical lesions (15 lesions vs 58 lesions). CONCLUSIONS: At ultrahigh-field strength, T1-weighted MPRAGE is highly sensitive in detecting MS plaques within the white and the gray brain parenchyma. Our results indicate structural damage beyond demyelination in every lesion depicted, which is in accordance with postmortem histopathological studies. The 7-T MPRAGE clearly delineated every cortical lesion that was visualized by any other MRI sequence at 1.5 or 7 T.