Clinical significance of MRI/(18)F-FDG PET fusion imaging of the spinal cord in patients with cervical compressive myelopathy.

PURPOSE

(18)F-FDG PET is used to investigate the metabolic activity of neural tissue. MRI is used to visualize morphological changes, but the relationship between intramedullary signal changes and clinical outcome remains controversial. The present study was designed to evaluate the use of 3-D MRI/(18)F-FDG PET fusion imaging for defining intramedullary signal changes on MRI scans and local glucose metabolic rate measured on (18)F-FDG PET scans in relation to clinical outcome and prognosis.

METHODS

We studied 24 patients undergoing decompressive surgery for cervical compressive myelopathy. All patients underwent 3-D MRI and (18)F-FDG PET before surgery. Quantitative analysis of intramedullary signal changes on MRI scans included calculation of the signal intensity ratio (SIR) as the ratio between the increased lesional signal intensity and the signal intensity at the level of the C7/T1 disc. Using an Advantage workstation, the same slices of cervical 3-D MRI and (18)F-FDG PET images were fused. On the fused images, the maximal count of the lesion was adopted as the standardized uptake value (SUV(max)). In a similar manner to SIR, the SUV ratio (SUVR) was also calculated. Neurological assessment was conducted using the Japanese Orthopedic Association (JOA) scoring system for cervical myelopathy.

RESULTS

The SIR on T1-weighted (T1-W) images, but not SIR on T2-W images, was significantly correlated with preoperative JOA score and postoperative neurological improvement. Lesion SUV(max) was significantly correlated with SIR on T1-W images, but not with SIR on T2-W images, and also with postoperative neurological outcome. The SUVR correlated better than SIR on T1-W images and lesion SUV(max) with neurological improvement. Longer symptom duration was correlated negatively with SIR on T1-W images, positively with SIR on T2-W images, and negatively with SUV(max).

CONCLUSION

Our results suggest that low-intensity signal on T1-W images, but not on T2-W images, is correlated with a poor postoperative neurological outcome. SUV(max) of lesions showing increased signal intensity and SUVR measured on fusion MRI/PET scans are more sensitive parameters for predicting clinical outcome than signal intensity on the MRI scan.