Several prognostic studies looked for an association between the degree of spinal cord injury (SCI), as depicted by primary magnetic resonance imaging (MRI) within 72 hours of injury, and neurologic outcome. It was not clearly demonstrated whether the MRI at any time correlates with neurologic prognosis. The purpose of the present study was to investigate the relationship between acute MRI features and neurologic prognosis, especially walking ability of patients with cervical spinal cord injury (CSCI). Moreover, at any point, MRI was clearly correlated with the patient's prognosis. Retrospective image study. From January 2010 to October 2015, 102 patients with CSCI were treated in our hospital. Patients who were admitted to our hospital within 3 days after injury were included in this study. The diagnosis was 78 patients for CSCI with no or minor bony injury and 24 patients for CSCI with fracture or dislocation. A total of 88 men and 14 women were recruited, and the mean patient age was 62.6 years (range, 16-86 years). Paralysis at the time of admission was graded as A in 32, B in 15, C in 42, and D in 13 patients on the basis of the American Spinal Injury Association (ASIA) impairment scale. Patients with CSCI with fracture or dislocation were treated with fixation surgery and those with CSCI with no or minor bony injury were treated conservatively. Patients were followed up for an average of 168 days (range, 25-496 days). Neurologic evaluation was performed using the ASIA motor score and the modified Frankel grade at the time of admission and discharge. Magnetic resonance imaging was performed for all patients at admission. Using the MRI sagittal images, we measured the vertical diameter of intramedullary high-intensity changed area with T2-weighted images at the injured segment. We studied separately the patients divided into two groups: 0-1 day admission after injury, and 2-3 days admission after injury. We evaluated the relationship between the vertical diameter of T2 high-intensity changed area in MR images and neurologic outcome in these two groups. This study does not contain any conflict of interest. In the group admitted at 0-1 day after injury, there was a relationship between the vertical diameter of T2 high-intensity area in MR image and the ASIA motor score at admission and at discharge, but correlation coefficient was low (0.3766 at admission and 0.4239 at discharge). On the other hand, in the group admitted at 2-3 days after injury, there was a significant relationship between the vertical diameter of T2 high-intensity area in MR image and the ASIA motor score at admission and at discharge, and correlation coefficient was very high (0.6840 at admission and 0.5293 at discharge). In the group admitted at 2-3 days after injury, a total of 17 patients (68%) recovered to walk with or without a cane. Receiver operating characteristic (ROC) curve analysis demonstrated that the optimal vertical diameter of T2 high-intensity area cutoffvalue for patients who were able to walk at discharge was 45.8 mm. If the vertical diameter of T2 high-intensity area cutoff value was 45 mm, there was a significant positive correlation with being able to walk at discharge (p<.0001). From our study, 2-3 days after injury, a significant relationship was observed between the vertical diameter of T2 high-intensity area and the neurologic prognosis at discharge. Zero to 1 day after injury, the relationship between the vertical diameter of T2 high-intensity area and the neurologic prognosis at discharge was weak. Neurologic prognosis is more correlated with MRI after 2-3 days after the injury. If the vertical diameter of T2 high-intensity area was <45 mm, the patients were able to walk with or without a cane at discharge. T2 high-intensity changed area can reflect the neurologic prognosis in patients with CSCI.