Brain volume reduction over four decades in multiple sclerosis

Introduction: Multiple sclerosis (MS) is a common neurological disease causing intermittent or cumulative disability, predominantly among women of child-bearing age. MS is a chronic inflammatory disease of the central nervous system (CNS) characterized by early axonal damage and early atrophy progression. The disease has certain typical features, but disease progression is highly variable and has many atypical forms. In follow-up, the poor correlation between radiological findings and disability has sometimes been referred to as the clinicoradiolological paradox . The widespread effect on the whole CNS is a challenge to radiological evaluation. Atrophy of the brain and spinal cord that can be detected 6 months after symptom onset has been proposed as a better marker of disability. Longitudinal studies of MS-associated atrophy (over a decade) are few. Atrophy studies were often performed in association with clinical trials and technically difficult to compare. Three-dimensional atrophy measurements demand time, advanced software, and trained persons, making them less suitable for the routine radiological evaluation of MS. Volumetric changes can be indirectly evaluated by performing serial one-dimensional and two-dimensional measurements on a routine PACS workstation using a distance tool. Aim: Our aim was to evaluate the atrophy rate (reduction of brain volume and compensatory supratentorial ventricular enlargement) and its correlation to disability. We chose a cohort with widespread disease duration at baseline in order to study whether atrophy was accelerating or declining over time. The observed individual atrophy rates over one decade represented four decades of disease. We compared the 1D and 2D atrophy measurements to sophisticated 3D measurements to determine which of the measurements best correlate with the three-dimensional measurements. Materials and Methods: Thirty-seven MS patients aged 24-65 years with a disease duration of 1-33 years were consecutively selected at baseline in 1995-1996 and evaluated by MRI over a mean period of 9.25 years (range 7.3-10 years) from 1995-2005. Serial brain atrophy measurements (1D and 2D) were analyzed by one radiologist on a routine PACS workstation, and these results were compared to semi-automated 3D measurements (HERMES, neuroradiological rater). Results: Volumetric differences were found over shorter periods of time (1-7 months) but vanished at the end of the follow-up. At the end of the study, a uniform longitudinal decrease in brain volume and increase in ventricle volume was found. The 1D, 2D, and 3D measurements intercorrelated well. Frontal horn width (1D) correlated strongest with the 3D measurements. Atrophy progression was independent of the clinical course of MS. Atrophy measurements were associated with disability, and this association persisted during follow-up. Conclusions: Despite variable clinical courses, the destructive and degenerative effects of MS seem to progress uniformly over long periods of time. The volumetric changes can be detected using 1D and 2D measurements performed on a routine PACS workstation