MRI in the acute phase of spinal cord traumatic lesions: Relationship between MRI findings and neurological outcome.

Purpose: To evaluate the role of emergency MRI in the diagnosis of acute spinal injuries, and to correlate the MRI pattern with the neurological outcome.

Materials And Methods: Thirty-eight patients with MRI-proven spinal cord injury were classified according to the Frankel classification. MRI was always performed within 8 hours from trauma. Frankel classification divides spinal cord injuries into 5 classes of decreasing severity based on the presence of motor and/or sensory function loss. On the basis of the MRI findings the patients were classified in 3 groups: group 1 (intramedullary haematoma), group 2 (multi-metamer oedema), group 3 (single-metamer oedema). All patients underwent neurosurgery and were clinically evaluated until the stabilization of neurological recovery. Mean follow-up time was 12 months. The MR images were retrospectively evaluated and correlated to the neurological outcome.

Results: Twenty-eight patients showed complete motor loss (Frankel classes A and B); of these 28 patients 12 (42.8%) had MRI evidence of intramedullary haematoma, 12 (42.8%) had multi-metamer oedema and 4 (14.4%) had single-metamer oedema. Of the 10 patients with incomplete motor loss, none had MRI evidence of haemorrhage, 4 (40%) showed multi-metamer oedema and 6 (60%) showed single-metamer oedema. Follow-up clinical assessment revealed that 14/38 patients (36,8%) had clinical improvement and 2/38 cases (5%) had a complete motor recovery, as demonstrated by the move to a higher Frankel class.

Conclusions: Our results, consistent with previous reports, confirm a strong correlation between the MRI appearance of traumatic spinal cord injuries in acute phase and long-term recovery of motor and sensory function: patients with initial haemorrhage had a poor prognosis, whereas those with spinal cord oedema had a good clinical outcome, as demonstrated by the passage to a higher Frankel class. MRI is particularly important in the initial evaluation of unconscious patients who cannot undergo a motor and sensory neurological evaluation, and to define the prognosis, which will influence the correct therapeutic choice.