The advantages of contrast myelography in the diagnosis of protruded intervertebral disk are considered by many neurosurgeons to be too slight to offset the possible (though unproved) risks of introducing lipiodol into the subarachnoid space, and they have preferred to base their indications for operation on careful evaluation of the clinical findings. Whether or not this attitude is justifiable, it is at least a signal for the radiologist to review his results more critically. Upon doing so, he cannot but confess that x-ray diagnosis of disk protrusion as generally practised has certain shortcomings which detract from its efficiency. Several investigators have sought to determine the cause of these shortcomings. Among the factors mentioned, the most important would seem to be the narrowness of the dural sac and lateral protrusion of the disk. That these do not account for all failures of the protruded disk to produce a characteristic defect in the oil column is, however, clear to anyone who has seen many of these cases in the operating room. With these considerations in mind, we sought to determine whether the dural sac is always in direct contact with the posterior surface of the vertebral bodies and the intervertebral disks, as has been commonly assumed. We found that, especially in the region of the fifth lumbar vertebra, where most protrusions occur, this is frequently not the case. For our investigation, the patient was placed in the prone position and examined with the horizontal x-ray beam and vertical film or screen (Fig. 1). In this position the best possible filling of the sac in the region under consideration is obtained, and at the same time opportunity is afforded to glance between the sac and the posterior surface of the vertebrae. It was found that the sac is, indeed, usually in direct contact with the body of the vertebra, but that in a certain percentage of cases (about 15 per cent) they are separated by a space averaging 2 to 4 mm. in width (Fig. 2), and in extreme instances (5 to 8 per cent) reaching as much as 10 mm. This space, the so-called epidural space, attains its maximum width between the fifth lumbar and the first sacral vertebrae. At this site the majority of spines show a sudden lordotic curve, and this angulation may give rise to an abnormal relation between sac and vertebra, as will be explained. So long as the length of the dural sac is ample, it can follow the ventral side of the canal, although this is the longer path, and it will do so because the nerve roots pull it ventrally. In cases of marked widening of the space, however, the dural sac appears to be stretched like a hammock between its attachment to the dorsal side of the sacral canal and its cranial attachment, following the dorsal rather than the ventral margin of the canal (Fig. 3). The significance of this observation is obvious.
Read full abstract