Extent of intramedullary short tau inversion recovery signal change predicts traumatic cervical spinal cord injury outcomes in surgically treated older adults: a multicenter study in Japan

Study Design: A retrospective multicenter study. Purpose: To determine whether the extent of intramedullary hyperintensity on initial fat-suppressed short tau inversion recovery (STIR) magnetic resonance imaging (MRI) predicts neurological outcomes in elderly patients with traumatic cervical spinal cord injury (CSCI) undergoing surgery. Overview of Literature: Prognostic stratification in elderly patients with CSCI remains challenging because neurological recovery varies widely. Studies focusing exclusively on surgical populations evaluating imaging predictors-particularly STIR signal changes-are limited. Methods: We retrospectively reviewed patients aged ≥65 years with traumatic CSCI treated at 33 institutions. All patients underwent surgery. Based on initial STIR MRI findings, patients were categorized as H0 (no intramedullary signal change), H1 (single-level change), or H2 (multilevel change). Neurological status was evaluated using the American Spinal Injury Association Impairment Scale (AIS) at injury and final follow-up. A good outcome was defined as AIS grade D/E at follow-up without deterioration or improvement of ≥1 AIS grade. A poor outcome was defined as AIS grade C or worse without improvement. After excluding cases with missing data, inverse probability of treatment weighting (IPTW) was applied for multivariate adjustment in 622 patients (H0=132, H1=350, H2=140). Results: Good outcomes were observed in 92% of H0, 80% of H1, and 68% of H2. Compared with H0, both H1 and H2 were independently associated with higher odds of poor outcomes (IPTW-adjusted odds ratio: 2.4 for H1 vs. H0 and 3.2 for H2 vs. H0). Conclusions: In surgically treated elderly patients with traumatic CSCI, the presence of intramedullary STIR hyperintensity at presentation-particularly across multiple levels-was strongly associated with poorer neurological recovery. The extent of STIR signal change may aid initial risk stratification and clinical decision-making.