Detection of nociceptive-related metabolic activity in the spinal cord of low back pain patients using F-FDG PET/CT.

BACKGROUND

Over the past couple of decades, a number of centers in the brain have been identified as important sites of nociceptive processing and are collectively known as the 'pain matrix.' Imaging tools such as functional magnetic resonance imaging (MRI) and F-fluorodeoxyglucose (F-FDG) positron emission tomography (PET) have played roles in defining these pain-relevant, physiologically active brain regions. Similarly, certain segments of the spinal cord are likely more metabolically active in the setting of pain conditions, the location of which is dependent upon location of symptoms. However, little is known about the physiologic changes in the spinal cord in the context of pain. This study aimed to determine whether uptake of F-FDG in the spinal cord on positron emission tomography/computed tomography (PET/CT) of patients with low back pain (LBP) differs from that of patients without LBP.

METHODS

We conducted a retrospective review of F-FDG PET/CT scans of 26 patients with non-central nervous system cancers, 13 of whom had reported LBP and 13 of whom were free of LBP (controls). No patients had spinal stenosis or significant F-FDG contribution of degenerative changes of the spine into the spinal canal. Circular regions of interests were drawn within the spinal canal on transaxial images, excluding bony or discal elements of the spine, and the maximum standardized uptake value (SUVmax) of every slice from spinal nerves C1 to S1 was obtained. SUVmax were normalized by subtracting the SUVmax of spinal nerve L5, as minimal neural tissue is present at this level. Normalized SUVmax of LBP patients were compared to those of LBP-free patients at each vertebral level.

RESULTS

We found the normalized SUVmax of patients with LBP to be significantly greater than those of control patients when jointly tested at spinal nerves of T7, T8, T9 and T10 (p<0.001). No significant difference was found between the two groups at other levels of the spinal cord. Within the two groups, normalized SUVmax generally decreased cephalocaudally.

CONCLUSIONS

Patients with LBP show increased uptake of F-FDG in the caudal aspect of the thoracic spinal cord, compared to patients without LBP.

IMPLICATIONS

This paper demonstrates the potential of F-FDG PET/CT as a biomarker of increased metabolic activity in the spinal cord related to LBP. As such, it could potentially aid in the treatment of LBP by localizing physiologically active spinal cord regions and guiding minimally invasive delivery of analgesics or stimulators to relevant levels of the spinal cord.