We sought to conduct a systematic review to examine the role of magnetic resonance imaging in predicting outcomes after surgery and to evaluate the evidence currently available critically.Summary of Background Data.
Degenerative compressive myelopathy is a common clinical problem associated with adverse health outcomes. Although a number of studies have investigated the association between preoperative magnetic resonance imaging characteristics and outcomes after surgery for degenerative compressive myelopathy, the conclusions of these studies have often yielded differing results.Methods.
Articles examining the predictive value of magnetic resonance imaging were obtained from MEDLINE, EMBASE, and PubMed databases (1980–2011). Thirty publications that met the inclusion criteria were reviewed. Two reviewers independently assessed each study regarding the level of evidence (using the criteria proposed by Sackett) and methodological quality based on revised Cochrane quality assessment checklist.Results.
Three excellent, 1 good, and 10 poor quality studies assessed cord compression—transverse area (4), compression ratio (5), and anteroposterior diameter (1). Relationship between signal intensity (SI) changes and surgical outcomes were reviewed by 28 studies—8 excellent, 9 good, and 13 poor quality studies. SI changes within the spinal cord included the presence of SI on T2-weighted image (WI) (17), area of SI on T2WI (8), degree of SI on T2WI (5), presence of SI on both T1-/T2WI (2), SI ratio on T2WI (2), and the position of SI on T2WI (1).Conclusion.
Based on a combination of excellent and good quality studies, transverse area correlates with recovery ratio but not with postoperative functional score assessed by Japanese Orthopaedic Association/modified Japanese Orthopaedic Association scores. SI changes defined by (1) its presence on T2WI, (2) its extent (focal or multisegmental), (3) its brightness, and (4) its presence on both T1-/T2WI can predict surgical outcomes in degenerative compressive myelopathy.Conclusion.
Level of Evidence: 2