Computed tomography of herniated and extruded nucleus pulposus

Computed tomography (CT) is an accurate method of diagnosing herniated nucleus pulposus (HNP) in patients with back pain and radiculopathy. We evaluated 188 patients with 244 disk space explorations for the treatment of back pain and radiculopathy. In predicting HNP, CT was 92% accurate and myelography was 88% accurate. The only significant difference between the two modalities is at L5-S1. Here CT was 92% accurate in predicting HNP and myelography was 70% accurate, a difference that is statistically significant (p less than 0.01). The dural sac separation reduces the myelogram accuracy at L5-S1. In detailed evaluation of 188 patients with both CT-diagnosed and surgically confirmed HNPs, these HNPs were classified by location, calcification, size, and extrusion. If the maximum anteroposterior diameter of the HNP was less than one-half the anticipated normal sagittal diameter of the dural sac, the incidence of extrusion was less than 10%. If it occupied one-half or more of the anticipated normal sagittal diameter of the dural sac, the incidence of extrusion was greater than 90%. This difference is highly statistically significant (p less than 0.001). Thirty-five percent of the HNPs exhibited some degree of migration with the cephalic direction more common than the caudal direction. Computed tomography and myelography are both satisfactory methods of diagnosing HNP. The noninvasive CT gives direct anatomical information and is more accurate at L5-S1. If the physical examination correlates with the CT examination, then myelography may be bypassed and the patient treated with a high degree of confidence.