Early MRI in optic neuritis: the risk for clinically definite multiple sclerosis

• Published in: Multiple sclerosis 2010-02, Vol.16 (2), p.156-165
• Main Authors: Swanton, JK, Fernando, KT, Dalton, CM, Miszkiel, KA, Altmann, DR, Plant, GT, Thompson, AJ, Miller, DH
• Format: Article
• Language: English
• Subjects: Adult; Atrophy; Biological and medical sciences; Brain - pathology; Degenerative and inherited degenerative diseases of the nervous system. Leukodystrophies. Prion diseases; Disease Progression; Early Diagnosis; Female; Follow-Up Studies; Humans; Magnetic Resonance Imaging; Male; Medical sciences; Multiple Sclerosis - etiology; Multiple Sclerosis - pathology; Multiple sclerosis and variants. Guillain barré syndrome and other inflammatory polyneuropathies. Leukoencephalitis; Multivariate Analysis; Neurology; Optic Neuritis - complications; Optic Neuritis - diagnosis; Optic Neuritis - pathology; Predictive Value of Tests; Proportional Hazards Models; Prospective Studies; Risk Assessment; Risk Factors; Severity of Illness Index; Sex Factors; Time Factors
• ISSN: 1352-4585; 1477-0970
• DOI: 10.1177/1352458509353650

MRI brain lesions at presentation with optic neuritis (ON) increase the risk for developing clinically definite (CD) multiple sclerosis (MS). More detailed early MRI findings may improve prediction of conversion. The objectives of this study were to investigate the influence of number, location and activity of lesions at presentation, new lesions at early follow-up and non-lesion MRI measures on conversion from optic neuritis (ON) to CDMS. 142/143 ON patients, prospectively recruited into a serial MRI and clinical follow-up study, were followed-up at least once. Cox regression analysis determined independent early MRI predictors of time to CDMS from: (i) baseline lesion number, location and activity measures, (ii) three-month lesion activity measures and (iii) brain atrophy, magnetization transfer ratio and spectroscopy measures. 114/142 (80%) had abnormal baseline brain or cord MRI. 57 (40%) developed CDMS (median of 16 months from clinically isolated syndrome onset). Median follow-up of the non-converters was 62 months. Multivariate analysis of baseline parameters revealed gender, periventricular and gadolinium-enhancing lesions as independent predictors of CDMS. Considering both scans together, gender, baseline periventricular and new T2 lesions at follow-up remained significant (hazard ratios 2.1, 2.4 and 4.9, respectively). No non-conventional measure predicted CDMS. It was concluded that new T2 lesions on an early follow-up scan were the strongest independent predictor of CDMS.