頚椎症性脊髄症手術後の最小臨床的意義のある変化に対するMOLLI T₁マッピングイメージングの縦断的予測値

To investigate the predictive value of MOLLI T₁ mapping imaging for the longitudinal changes in Minimal Clinical Important Difference (MCID) after surgery for patients with Cervical Spondylotic Myelopathy (CSM).

A prospective study enrolled 163 surgically treated CSM patients (October 2022-October 2023). All underwent preoperative conventional MRI and MOLLI T₁ mapping. The MCID was determined based on the difference in Modified Japanese Orthopedic Association (m JOA) scores before and at 3, 6, and 12 months after surgery. Patients with an MCID less than 2 were defined as the poor prognosis group, while those with an MCID of 2 or higher were defined as the good prognosis group. Patients were classified into the good prognosis and poor prognosis groups at 3, 6,12 months postoperatively. The differences in conventional MRI features and T₁ values between the two groups at these time points were compared. Univariate and multivariate logistic regression identified recovery predictors, with forest plots generated. Receiver operating characteristic (ROC) curve analysis assessed diagnostic performance.

At 3 months, (27 good vs 136 poor). Only T₂WI hyperintensity differed between groups [Area under the receiver operating characteristic curve (AUC)=0.597, cut off = 0.1942; sensitivity = 40.74 %, specificity = 78.68 %]. At 6 months (60 good vs 103 poor), T₁ values showed significance (AUC = 0.690, cut off = 0.3091; sensitivity = 83.33 %, specificity = 47.57 %). At 12 months (100 good vs 63 poor), T₁ values demonstrated superior prediction (AUC = 0.875,cut off = 0.6389; sensitivity = 75.00 %, specificity = 88.89 %).

MOLLI T₁ mapping effectively predicts CSM recovery at 6,12 months postoperatively, with optimal diagnostic performance at 12 months.

This study first employs MRI T₁ mapping to predict MCID outcomes in CSM patients, demonstrating its high predictive value for favorable postoperative prognosis and providing a reliable imaging parameter for outcome assessment.