Spine - 2026-08-12 - Journal Article
Impact of Preoperative Intramedullary T2 Signal Intensity Changes on Surgical Outcomes in Cervical Spondylotic Myelopathy: A Multicenter Prospective Study.
Horiuchi Y, Nagoshi N, Yamane J, Okubo T, Ikegami T, Ago K, Fukuda K, Fujii T, Kamata Y, Isogai N, Suzuki S, Ozaki M, Matsumoto M, Nakamura M, Watanabe K
Topics
Key Takeaway
In 721 CSM patients, binary preoperative intramedullary T2 signal intensity was not independently associated with worse JOA recovery rate, JOACMEQ effectiveness, or SF-36 at 2 years after multivariable adjustment.
Summary Depth
Choose how much analysis to show on this article page.
Summary
This 10-institution prospective cohort evaluated whether preoperative intramedullary T2 signal intensity (SI) predicts 2-year surgical outcomes in CSM across 721 patients (76.8% SI-positive). After multivariable adjustment for age, symptom duration, alignment, diabetes, and smoking, SI positivity was not independently associated with JOA score, JOA recovery rate, JOACMEQ domain effectiveness, or SF-36. Only preoperative C2-7 angle independently predicted SI presence (OR 1.02, P=0.023).
Key Limitation
Binary SI classification without accounting for signal intensity grade, lesion length, or presence of T1 hypointensity limits the ability to detect subgroup-specific prognostic effects that a graded or multi-parametric MRI analysis would capture.
Original Abstract
STUDY DESIGN
Multicenter prospective cohort study.
OBJECTIVE
To evaluate the impact of preoperative intramedullary T2 signal intensity (SI) changes on surgical outcomes in patients with cervical spondylotic myelopathy (CSM).
SUMMARY OF BACKGROUND DATA
Intramedullary SI on T2-weighted MRI has been considered a predictor of poor outcomes in CSM; however, its prognostic value remains controversial.
METHODS
We prospectively analyzed 721 patients with CSM who underwent surgical decompression at 10 institutions between 2019 and 2022. Patients were classified into SI (+) and SI (-) groups based on preoperative T2-weighted MRI. Outcomes were assessed preoperatively and at 2 years using the Japanese Orthopaedic Association (JOA) score, visual analog scale (VAS), JOA Cervical Myelopathy Evaluation Questionnaire (JOACMEQ), and Short Form-36 (SF-36). Multivariable analyses were adjusted for age, sex, symptom duration, preoperative C2-7 angle, cervical range of motion, diabetes mellitus, and smoking history.
RESULTS
Of the 721 patients, 554 (76.8%) were SI (+) and 167 (23.2%) were SI (-). Although preoperative upper extremity pain/numbness on VAS was significantly higher in the SI (+) group (64.7 vs. 58.9, P=0.037), this difference was not significant at 2 years. No significant between-group differences were found in preoperative or postoperative JOA score, JOA recovery rate, ΔJOA, SF-36, JOACMEQ effectiveness rates, or perioperative complications. SI was not independently associated with improvement in any JOACMEQ domain. Only the preoperative C2-7 angle was independently associated with the presence of SI (odds ratio, 1.02; P=0.023).
CONCLUSION
Preoperative intramedullary T2 SI was associated with greater preoperative upper extremity pain/numbness, and with a greater preoperative C2-7 angle. However, after adjustment for confounders, the binary presence of intramedullary T2 SI was not statistically significantly associated with worse 2-year neurological, functional, or patient-reported outcomes.
LEVEL OF EVIDENCE
Level II.