Clinical and radiological correlates of severity and surgery-related outcome in cervical spondylosis

Abstract

The aim of this study was to determine if radiological features could be used to predict outcome in patients with cervical spondylotic myelopathy (CSM). The authors studied 69 patients consecutively referred to The National Hospital, Queen Square, for decompressive surgery. Data obtained from preoperative cervical spine magnetic resonance (MR) imaging studies were each analyzed on two separate occasions by two blinded radiologists. The parameters determined were signal change and the presence and severity of compression. Clinical outcome was determined by pre- and postoperative timed walks, as well as by evaluation of myelopathy disability index scores, Ranawat classification, and Nurick grades. There was good inter- and intraobserver reliability for determination of radiological data. A significant relationship was found between MR imaging signal change and surgery-related outcome, as reflected by improvement in walking parameters; however, this was confounded by the fact that signal change also related to preoperative walking parameters, and those patients for whom preoperative walking function was worse experienced greater functional improvement in walking postoperatively. The relationships between ambulatory-related data and severity or extent of spinal cord compression were less marked. Cervical cord compression and intrinsic MR imaging signal change correlate with clinical severity, and, in this population, the presence of signal change was correlated with better surgery-related outcome. However, confounding factors and the lack of strong correlation indicate that these radiological measurements are insufficient to be used as a reliable tool for predicting surgery-related benefits in individual patients.