Patients who had lumbar spine surgeries and experienced residual symptoms were prospectively studied using computed tomography/discography and enhanced magnetic resonance imaging.Objectives
Determining the sensitivity and specificity of gadolinium-enhanced magnetic resonance imaging versus computed tomography/discography in distinguishing scarring from recurrent disc herniation was the first objective. The second goal was to determine if there were any imaging characteristics unique to either scar or recurrent disc herniation.Summary of Background Data
The results of reoperation for recurrent disc herniation are uniformly good, whereas the results of reoperation for scar tissue are poor. There have been no studies comparing the ability of enhanced magnetic resonance imaging and computed tomography/discography to distinguish between scar and herniation material.Methods
Two neuroradiologists who were blinded to the surgical findings in 33 patients independently reviewed the imaging studies and labeled abnormalities as disc herniation or scar tissue. Their responses were compared with surgical findings to determine the accuracy, sensitivity, and specificity of each test. Also, imaging studies were used to determine if disc material or scar tissue had specific imaging characteristics.Results
Computed tomography/discography was more sensitive and specific in distinguishing between scar tissue and recurrent disc herniation. Information from both studies improved the sensitivity over either test alone. Characteristics associated with recurrent disc herniation included nonenhanced or rim-enhanced abnormality surrounding a low signal intensity lesion on magnetic resonance imaging and extension of contrast into the epidural space and an enhancing abnormality on computed tomography/discography.Conclusions
Using computed tomography/discography alone or with enhanced magnetic resonance imaging enables surgeons to distinguish between scarring and recurrent disc herniation so they can determine the ned to re-enter the spinal canal.