The diagnosis of double-crush lesion in the L5 lumbar nerve using diffusion tensor imaging

A double-crush lesion is a condition in which the lumbar nerve is compressed both medially and laterally in the spinal canal, where diagnosis can be very difficult, and is a factor leading to poor surgical success rates.

Diffusion tensor imaging (DTI) was used to determine DTI parameter fractional anisotropy (FA) values and apparent diffusion coefficient (ADC) in both intraspinal column lesions alone and in double-crush lesions.

This study used a prospective study.

Of the 56 cases (mean age: 72.2 years) that underwent laminectomy for lumbar spinal stenosis at our clinic between April 2013 to March, 2015, 10 cases with L5 radiculopathy caused by L4-L5 stenosis (Intraspinal stenosis group (Group I); mean age: 74.7 years), and 5 cases with persistent symptoms caused by L5 foraminal stenosis despite L4-L5 decompression surgery (Double-crush group (Group D); mean age: 77.6 years) were targeted. One patient in Group D was diagnosed through microendoscopic intrapedicular partial pediculotomy and the remaining four cases by nerve root infiltration. Five healthy cases (mean age: 54 years) were studied as controls.

Intraspinal zone (Iz), nerve root (N), and extraforaminal zone (Ez) were established as the regions of interest, and the L5 nerve FA and ADC values were determined on the affected side.

Diffusion tensor imaging was performed prospectively by 1.5T magnetic resonance imaging before surgery, and DTI parameters of L5 nerve were evaluated in all patients and healthy volunteers. Student t test was used for group comparisons, and a p<.05 was considered statistically significant.

Fractional anisotropy values (Iz, N, Ez) were 0.415, 0.448, and 0.517, respectively, increasing as sites became more distal. Group I values were 0.335, 0.393, and 0.484, and Group D values were 0.296, 0.367, and 0.360. Compared with the healthy volunteers, Group D had significantly lower Iz (p<.05) and Ez (p<.001) values, while Group I had significantly lower Iz (p<.05) values. In Group D, Ez FA values were significantly lower (p<.001) than in Group I. Apparent diffusion coefficient values (Iz, N, Ez) in the healthy control group were 1.270 mm2/s, 1.151 mm2/s, and 0.937 mm2/s with values decreasing as sites grew distal. In Group I, the ADC values were 1.406 mm2/s, 1.184 mm2/s, and 1.001 mm2/s, while in Group D they were 1.551 mm2/s, 1.412 mm2/s, and 1.329 mm2/s. Compared with the healthy volunteers, Iz (p<.05) and Ez (p<.05) values were significantly higher in Group D. The N (p<.01) and Ez (p<.001) ADC values were significantly higher in Group D than in Group I.

Depending on where the nerve was compressed, changes in DTI parameters revealed nerve damage (low FA values and increased ADC) in the intraspinal canal in the Intraspinal Group, and over a widespread area in the Double-crush Group spanning the medial to lateral spinal canal. Our research suggests that in cases where double crush is suspected before surgery, failed back surgery syndrome may be prevented by evaluating DTI images.