The purpose of this study is to identify if there are varying fracture-related marrow edema patterns on magnetic resonance imaging (MRI) in patients with acute vertebral body fractures, and whether patient age, gender, or the patient's underlying degree of degenerative changes may influence the appearance of marrow edema from a fracture. If there is no significant height loss or deformity in a vertebral body on MRI, then the presence of marrow edema is relied upon as a sensitive marker of an acute fracture. However, in our clinical practice we have encountered patients with known acute fractures identified on computed tomography (CT), but with minimal or no conspicuous associated marrow edema on MRI. This research would therefore help to establish how sensitive marrow edema on MRI is for the detection of vertebral body fractures in different patients.

A search of the radiology records will be performed using the department's Nuance mPower clinical analytics software to identify patients with proven acute vertebral body fractures identified on CT, who then also had a spine MRI performed as part of their imaging work-up. Patients who did not have both an MRI and CT will be excluded. Each patient's age and gender will be recorded. Each patient's EMR chart notes pertaining to the date of injury will be reviewed to learn the mechanism of their injury (e.g. motor vehicle collision, fall from a height, etc.) The included patients' names will not be recorded. A consensus review will be performed by the investigator and co-investigators to classify the type of fracture, and, if present, characterize the extent and degree of marrow edema. The patient's degree of underlying spinal degenerative changes will also be characterized. The marrow edema patters will be categorized (e.g. edema, nonspecific mild edema, extensive edema, etc.). . Data analysis will then be performed to assess for a significant effect on patient age, gender, and/or underlying degenerative changes on the marrow edema patterns observed. The date range of data obtained will range from 2010 to 2019.