Clinical Investigative Study MRI Shows Dorsal Lesions and Spinal Cord Atrophy in Chronic Sensory Neuronopathies Marcondes C. Franc ¸a Jr., MD, Anelyssa D’ Abreu, MD, Ver ˆ onica A. Zanardi, MD, PhD, Andr ´ eia V. Faria, MD, PhD, Iscia Lopes-Cendes, MD, PhD, Anamarli Nucci, MD, PhD, Fernando Cendes, MD, PhD From the Department of Neurology, Campinas State University (UNICAMP), Campinas, Brazil (MCF, ADA, AN, FC); Department of Radiology, Campinas State University (UNICAMP), Campinas, Brazil (VAZ, AVF); and Department of Medical Genetics, Campinas State University (UNICAMP), Campinas, Brazil (IL-C) Keywords: Sensory neuronopathy, magnetic resonance imaging, spinal cord. Acceptance: Received August 08, 2007, and accepted for publication September 02, 2007. Correspondence: Address correspon- dence to Fernando Cendes, MD, PhD, Department of Neurology, Universidade Estadual de Campinas–UNICAMP. Cidade Universitaria “Zeferino Vaz” Campinas, SP, Brazil-13083-970. E-mail: fcendes@unicamp.br. J Neuroimaging 2008;18:168-172. DOI: 10.1111/j.1552-6569.2007.00193.x ABSTRACT BACKGROUND AND PURPOSE Sensory neuronopathies (SN) represent a specific subgroup of peripheral nervous system diseases, characterized by degeneration of dorsal root ganglia (DRG) and its projections. We tried to estimate the frequency and extent of spinal cord MRI abnormalities in a group of patients with SN and correlate these with clinical and neurophysiological features. METHODS We performed spinal cord MRI scans in 16 chronic SN patients. Images were analyzed for the presence of posterior hyperintense lesions on T2WI and cord areas at C3 level were obtained using a previously validated method. A group of 14 healthy controls with similar age and gender distribution was used for comparison. ANOVA was employed for statistical analysis. RESULTS Posterior T2WI lesions were found in 13 out of 16 patients. Cord areas were significantly smaller in SN patients than controls (84.3 × 97.2 mm 2 , P < .05). Atrophy correlated with severity of sensory ataxia and neurophysiologic abnormalities but not with duration of disease. CONCLUSIONS These findings support volumetric spinal cord MRI as a useful tool in the assessment of chronic SN. Introduction Sensory neuronopathies (SN) represent a specific subgroup of peripheral nervous system diseases, characterized by degenera- tion of dorsal root ganglia (DRG) and its projections, both cen- tral and peripheral. 1 This leads to the concomitant involvement of short- and long-length sensory axons. On clinical grounds, the disease is thus manifested by patchy and asymmetric sen- sory deficits rather than the usual “stocking-and-glove” pattern of dying back axonal neuropathies. 1 Besides sensory ataxia, widespread areflexia and positive sensory symptoms are also frequently found. 1 As auto-immune and neoplastic conditions are closely re- lated to SN, 1 early recognition and management are essential. Diagnosis still relies upon clinical suspicion and abnormalities on nerve conduction studies (NCS). 2 The hallmark of the dis- ease is a diffuse and non-length-dependent axonal compromise of sensory nerve action potentials (SNAPs) with preserved mo- tor NCS and needle EMG. 2 In some patients, however, routine NCS are not conclusive. Spinal cord MRI has been performed lately in SN patients and appears to be a promising tool for diagnostic and prognostic purposes. 3,4 In this setting, the aim of this study was to estimate the frequency and extent of spinal cord abnormalities among chronic SN patients. In addition, we used quantitative MRI techniques to check if cord atrophy develops in these patients. Materials and Methods Subjects Selection Patients with chronic SN followed regularly at our Univer- sity Hospital were included. They met the following crite- ria: clinical—presence of sensory ataxia and preserved muscle strength; neurophysiological—widespread (ie, upper and lower limbs) reduction of SNAP amplitudes with normal conduc- tion velocities and latencies, preserved motor NCS, and nee- dle EMG findings. Individuals with demyelinating neuropathy and genetically confirmed Friedreich’s ataxia (FA) were not in- cluded in final analysis. Two patients were excluded because of disease duration shorter than a year (one with 6 months and the other 9 months). MRI scans with significant motion or CSF pulsation artifacts were not included in the study. The control group for MRI analysis consisted of 14 healthy volunteers with similar age (mean age 50 years, SD = 4) and gender (6 men:8 women) distribution.The Ethics Committee of 168 Copyright ◦ C 2008 by the American Society of Neuroimaging