A Longitudinal MRI Study of Histopathologically Defined Hypointense Multiple Sclerosis Lesions Andreas Bitsch, MD, 1 Tanja Kuhlmann, MD, 2 Christine Stadelmann, MD, 2 Hans Lassmann, MD, 3 Claudia Lucchinetti, MD, 4 and Wolfgang Bru ¨ck, MD 2 Severe tissue destruction is the presumed histopathologi- cal correlate of hypointense multiple sclerosis (MS) le- sions. In this study we correlated changes of lesion hy- pointensity over time with initial histopathological features in 14 biopsied MS lesions. The extent of hypoin- tensity increased in initially demyelinated plaques and decreased in remyelinating lesions. The initial axonal loss determined the increase of hypointensity over time. In conclusion, both axonal loss and demyelinating activity determine the evolution of hypointensity over time. Ann Neurol 2001;49:793–796 Axonal loss in multiple sclerosis (MS) mainly contrib- utes to clinical disability and occurs early in lesion de- velopment. 1–3 In standard magnetic resonance imaging (MRI), cerebral atrophy and T1 hypointense lesions (“black holes”) are thought to reflect axonal loss. In combined histopathological and MRI studies, axonal loss appeared to be the correlate of lesion hypointen- sity. 4,5 N-acetyl-aspartate (NAA), an in vivo marker of axons, was strongly decreased in hypointense lesions. 6 Magnetization transfer ratio (MTR) also correlated with axonal density in a comparative morphological- radiological study. 7 Correlative MTR-MRI studies re- vealed a clear association between these two markers and their predictive value for disease progression or clinical disability in MS. 8 –12 However, the extent of hypointensity of an MS lesion is known to change over time. 13 A strongly decreased magnetization transfer ra- tio (MTR) and a long-lasting ring enhancement are predictive of persistent hypointensity. 13 However, the histopathological determinants affecting long-term de- velopment of lesion hypointensity have not been eluci- dated so far. In the present study, we longitudinally analyzed T1 lesion hypointensity of biopsy-defined in- flammatory demyelinating central nervous system (CNS) lesions. Materials and Methods Patients Brain biopsy specimen of patients who had been biopsied in the very early course of their disease were collected at the Institute of Neuropathology of the Charite ´, Berlin, Ger- many. These biopsies were performed to exclude treatable conditions other than MS. None of the authors was involved in decision making for biopsy. The attending physician ob- tained informed consent from each patient prior to biopsy. We analyzed 14 biopsies from 13 patients (9 women, 4 men, age 20 –52 years, median: 36 years; Table). Based on clinical follow-up data, 10 patients had a relapsing-remitting and 3 had a primary progressive disease course. At present, diagnosis is definite in 7 patients and probable in 6 patients according to Poser’s criteria. 14 Neuropathology Paraffin-embedded tissue was used. Routine neuropathological stains (H & E, Luxol-Fast Blue myelin stain, Bielschowsky’s silver impregnation for axons) were applied as well as immu- nocytochemistry for the following markers: antimyelin basic protein (MBP, Boehringer Mannheim, Mannheim, Germany), antiproteolipid protein (PLP, Dr. Piddlesden, University of Cardiff, UK), antimyelin oligodendrocyte glycoprotein (MOG, Dr. Piddlesden), anti-KiMlP (macrophages/microglia, Dr. Radzun, University of Go ¨ttingen, Germany), anti-27E10 and anti-MRP14 (activated macrophages, BMA Biomedicals, August, Switzerland), anti-CD3 (T cells, Dako, Denmark), anti-CD8 (Dako, Denmark), and anti-amyloid precursor pro- tein (APP; Boehringer Mannheim, Germany). An avidin bi- otin complex or an alkaline phosphatase/antialkaline phospha- tase technique was used with the appropriate controls. Biopsies were classified with respect to demyelinating ac- tivity and axonal loss as described in detail earlier. 3,15 Active demyelinating lesions (A) were diffusely infiltrated by macro- phages containing myelin proteins as markers of recent my- elin phagocytosis. Remyelinating lesions (R) were characterized by uniformly thin and irregularly arranged myelin sheaths. Inactive demyelinated lesions (I) were completely demyelinated without signs of active demyelination. Axonal loss was ex- pressed as the percentage of axonal density inside the lesions compared to axonal density in the periplaque white matter (PPWM). 3 MRI Analysis All MRI scans were copies of the original scans that had been performed at local institutions near to the patient’s place of residence. A baseline MRI scan close to biopsy was available for each patient (see Table). These were performed prior to biopsy in 12 patients and after biopsy in 2 patients (median: -4.5 days prior to biopsy, range -27– 45 days). A total of 32 follow-up scans was available for analysis (median time after From the 1 Abteilung Neurologie, Georg-August-Universita ¨t, Go ¨t- tingen, 2 Institut fu ¨r Neuropathologie, Charite ´, Campus Virchow- Klinikum, Berlin, Germany, 3 Institut fu ¨r Hirnforschung, Abt. Neu- roimmunologie, Vienna, Austria, and 4 Department of Neurology, Mayo Clinic, Rochester, MN Received Nov 20, 2000 and in revised form Feb 16, 2001. Accepted for publication Feb 17, 2001. Published online 25 April 2001. Address correspondence to Dr Bru ¨ck, Institut fu ¨r Neuropathologie, Charite ´, Campus Virchow-Klinikum, Augustenburger Platz 1, 13353 Berlin, Germany. E-mail: wolfgang.brueck@charite.de BRIEF COMMUNICATIONS © 2001 Wiley-Liss, Inc. 793