Study shows oblique MRI may aid precise lumbar foraminal stenosis diagnosis

Investigators showed that diagnosis was improved in 56% of patients with lumbar radiculopathy when a new type of oblique MRI fast spin-echo scan of the lumbar neuroforamen was used.

David B. Bumpass, MD, and colleagues developed the new oblique T1 MRI sequence, which is done in the plane of the foramen. Their goal was to obtain an en-face view of the patient’s nerve roots.

David B. Bumpass

A standard sagittal MRI approach can be inaccurate, according to Bumpass, because the nerve roots exit the foramen at an angle oblique to all three standard cross-sectional imaging planes.

“Inter-rater and intra-rater reliability were significantly better with the oblique T1 MRI, which provided en-face views of the lumbar nerve roots. The improvements were small in magnitude. Oblique scans were more likely to be useful to clinicians when more degeneration and foraminal stenosis were present,” Bumpass said.

New, experimental T1 MRI

From January 2012 to December 2013, 22 consecutive adults with leg pain, numbness and/or weakness were enrolled in a prospective study at a single academic, tertiary hospital. Patients were excluded from the study if they had lumbar scoliosis greater than 20°, were unable to undergo MRI or had previously undergone lumbar surgery.

Each of the patients underwent standard lumbar MRI in the axial and sagittal planes without contrast using T1 and T2 sequences. Then they underwent the experimental T1 fast spin-echo MRI sequence done oblique to the sagittal axis, and the same 1.5 Tesla MRI equipment was used for the scans.

Figure 1. The oblique T1 MRI slice on the left demonstrates en face view of lumbar nerve roots exiting the foramina. The L5-S1 foramen is narrowed, impinging the L5 nerve root. On the right, the axial slice at the level of the L5-S1 disc demonstrates the oblique plane orthogonal to the nerve root and foramen.

Source: Bumpass DB

The patients were 48.3 years old, mean. According to the abstract, the mean foraminal angle relative to the sagittal angle was 14°. Three patients had lumbar scoliosis with a Cobb angle at or equal to 20°.

Validated stenosis grading system

Four reviewers — three spine surgeons and one neuroradiologist — graded the foraminal stenosis of each lumbar nerve root on the sagittal and new oblique MRI scans.

“We used a validated 4-point grading system for foraminal stenosis,” Bumpass said.

Grade 1 was a nerve root completely surrounded by epidural fat, grade 2 was deformed epidural fat with the root touching one border of the foramina, grade 3 was no remaining epidural fat and no root deformity, and grade 4 was root compression, Bumpass said.

The investigators found the stenosis grades were better in the oblique series, which had an improved intra-class correlation coefficient (ICCC) score of 0.80 (P < .001, CI 0.76-0.83). The sagittal ICCC score was 0.68 (P < .001, CI 0.63-0.74). Bumpass said this showed significant superiority of the oblique scans over the sagittal scans.

Diagnosis improved with new MRI

The reviewers noted the oblique scans improved the diagnosis in about 56% of patients. For 11 patients, at least three graders stated the oblique scans improved their diagnosis. Eight of those patients had multilevel degenerative foraminal stenosis, according to Bumpass.

“The total number of yes responses for each patient correlated significantly with the mean stenosis grade. This was interpreted as meaning oblique scans were more useful when more degeneration was present,” Bumpass said.

A possible limitation of the study, he said, is it only evaluated the precision of foraminal stenosis grading from an MRI and did not evaluate the accuracy of diagnosis since the investigators could not correlate the radiographic diagnosis to actual symptomatic anatomic compression.

Bumpass said, “We feel we have demonstrated a way to improve the precision of radiographic diagnosis of foraminal stenosis, i.e., by reducing the spread of variability of both inter- and intra-rater reliability. Future research is needed to verify if this technique improves diagnosis accuracy.” – by Robert Linnehan

Reference:

Bumpass DB, et al. Paper #19. Presented at: North American Spine Society Annual Meeting; Nov. 12-15, 2014; San Francisco.

For more information:

David B. Bumpass, MD, can be reached at Washington University Orthopedics, 660 S. Euclid Ave., Saint Louis, MO 63110; email: bumpassd@wudosis.wustl.edu.

Disclosure: Bumpass reports no relevant financial disclosures.