MRI of a syrinx: is contrast material always necessary?

Abstract

The workup of a syrinx often includes contrast-enhanced MRI to exclude the presence of an underlying mass. The diagnostic yield of performing these additional contrast-enhanced sequences is not well defined in the literature. We hypothesized that T2-weighted imaging alone could reliably exclude the presence of a syrinx-associated mass without the need for contrast-enhanced imaging sequences in all cases. Two independent readers retrospectively analyzed contrast-enhanced MRI studies of 87 consecutive patients with syringes. The presence or absence of an associated spinal cord mass was determined using only T2-weighted imaging. The imaging features considered positive for a possible syrinx-associated lesion on T2-weighted imaging were syrinx nodularity, syrinx septations, and a spinal cord signal intensity abnormality or a mass separate from the syrinx. The size of the syrinx was also recorded. Using contrast-enhanced sequences as the reference standard, statistical analysis was performed to determine the accuracy of T2-weighted imaging in detecting a syrinx-associated mass. Of the 87 cases of syrinx, there were 23 mass lesions, 11 Chiari malformations, three spinal cord contusions, and 50 idiopathic syringes. Using T2-weighted imaging alone, readers detected no findings suspicious for a syrinx-associated mass in 55 cases and detected findings suspicious for a mass in 32 of 87 cases. Reader sensitivity for an underlying mass lesion was 100%; specificity, 86%; positive predictive value, 72%; and negative predictive value, 100%. Interreader agreement was excellent (κ = 0.88). Syrinx size showed a positive correlation with the presence of a mass lesion (p < 0.0001). T2-weighted imaging alone appears to have a high sensitivity and high negative predictive value in evaluating for a syrinx-associated mass, and contrast-enhanced imaging may not be required for the workup of a syrinx.