A sixty-seven-year-old woman was admitted to the hospital because of low-back pain, sciatica on the left, and weakness of the left lower extremity. For two months she had had low-back pain, and for three weeks pain had radiated from the left buttock to the lateral aspect of the thigh, the anterolateral portion of the leg, and the dorsal part of the foot. She had slight difficulty walking because of muscular weakness. She had no bowel or bladder incontinence. The pain was aggravated with movement of the trunk, especially forward bending, and was relieved by bed rest.
The initial physical examination revealed diffuse tenderness over the low back and localized tenderness over the posterior portion of the left thigh and calf. A straight-leg-raising test was positive at 60 degrees on the left, and the femoral-nerve-stretch test was negative. No atrophy or fasciculation of the muscles was found. A neurological examination demonstrated weakness (grade 4 of 5) of the left tibialis anterior, extensor hallucis longus, and extensor digitorum longus muscles. There was no weakness of the iliopsoas or quadriceps muscles. The deep tendon reflexes in the lower extremities were normal, and the Babinski reflex was absent bilaterally. A sensory examination revealed decreased sensation to light touch and pain in the fifth lumbar dermatome, with normal proprioception. A rectal examination revealed normal anal sphincter tone. The laboratory findings were unremarkable. The plain radiographs showed slight osteoporosis and degenerative scoliosis of the lumbar spine, with no other abnormal findings. On the basis of the results of the physical and radiographic examinations, the provisional diagnosis was a herniated nucleus pulposus or foraminal encroachment with a radiculopathy of the fifth lumbar nerve root. The patient was managed conservatively with bed rest and oral administration of a non-steroidal anti-inflammatory drug (diclofenac, twenty-five milligrams three times daily), but the symptoms and neurological deficits did not decrease. Additional radiographic examinations were carried out.
An anteroposterior myelogram demonstrated mild compression of the fifth lumbar nerve root on the left at the level of the fourth and fifth lumbar vertebrae. A lateral myelogram showed indentation of the anterior portion of the dural sac from the first and second lumbar to the fifth lumbar and first sacral disc levels; the indentation was predominantly at the first and second lumbar level (Figs. 1-A and 1-B). A computed tomography myelogram demonstrated central herniation of the first and second lumbar disc, with the disc compressing the spinal cord slightly (Fig. 2-A); there was also herniation of the disc through the left foramen at the fifth lumbar and first sacral level (Fig. 2-B). There were no abnormal findings at the fourth and fifth lumbar level. Sagittal T1-weighted magnetic resonance imaging demonstrated protrusion of the first and second lumbar disc (Fig. 3), and the conus medullaris was found to terminate at the inferior end of the second lumbar vertebral body. A block of the fifth lumbar nerve root, produced by the injection of five milliliters of 1 per cent lidocaine, temporarily relieved the pain in the lower limb.
At this time, two lesions seemed to be responsible for the symptoms: a herniated disc between the first and second lumbar vertebrae and a laterally herniated disc between the fifth lumbar and first sacral vertebrae. On the basis of the physical and neurological findings, we concluded that the signs and symptoms were probably caused by the latter. The patient was informed of the possibilities of operative intervention for both lesions, and decompression of the fifth lumbar nerve root on the left was first achieved with a foraminotomy at the fifth lumbar and first sacral level. Although there was diffuse bulging of both the fourth and fifth lumbar and the fifth lumbar and first sacral discs, there was no substantial compression of the fifth lumbar nerve root. There was no diminution of the pain or neurological deficits after the operation. Two weeks later, anterior excision of the first and second lumbar disc was done through a retroperitoneal approach. Subsequently, an anterior arthrodesis of the first and second lumbar vertebrae was performed with an anterior Kaneda device and an autologous tricortical iliac bone graft. A bioactive ceramic spacer was placed in the donor site. Immediately after the operation, there was no sciatica and the straight-leg-raising test was negative.
Postoperatively, the patient wore a thoracolumbosacral orthosis for four months. Two years and one month after the operation, radiographs demonstrated a solid fusion and no breakage of the instrumentation (Figs. 4-A and 4-B). The patient was asymptomatic, had an intact neurological status, and had returned to full activities. She had no restrictions or difficulties.
The clinical differentiation of the specific level of a herniated lumbar disc is primarily determined by the area of pain, the site of sensory dysfunction, the specific motor dysfunction, the changes in reflexes, and the results of the straight-leg-raising test3,11. The principal characteristics of a herniated disc between the first and second lumbar vertebrae are pain and decreased sensation over the anterior and medial aspects of the proximal part of the thigh, weakness of the quadriceps, a diminished patellar-ligament reflex, and a negative straight-leg-raising test3. A lesion at the thoracolumbar junction or at the first and second lumbar level, however, may produce more complex neurological findings that interfere with the diagnosis of the correct level. One reason for this is the proximity of the epiconus, conus medullaris, and cauda equina to the thoracolumbar junction and the first and second lumbar level. The other reason is that the level at which the spinal cord terminates differs among individuals. Approximately 90 per cent of spinal cords have been shown to end at or between the first and second lumbar vertebral bodies6,12,15. However, it can terminate at a level between the caudad third of the twelfth thoracic vertebral body and the middle third of the third lumbar vertebral body1,3,12,15. This complex intrathecal anatomy makes it difficult to diagnosis the correct level of a herniated disc that is at the thoracolumbar junction or between the first and second lumbar vertebrae. Kortelainen et al. evaluated the neurological symptoms and signs in 403 patients who had been managed operatively for a herniated lumbar disc, and they made special reference to the accuracy of the diagnosis of the correct levels11. They noted that pain in the distribution of the fifth lumbar nerve was a reliable indication of involvement of the fifth lumbar nerve root. The neurological findings associated with a herniated disc at a more cephalad level, however, were completely unreliable.
In the patient reported on here, we first performed a decompression of the fifth lumbar nerve root through the fifth lumbar-first sacral foramen because a monoradiculopathy of the fifth lumbar nerve root and positive radiographic findings had been demonstrated; however, no compression of the nerve root was found at this level, and the pain and neurological deficits were not alleviated by the first operation. Although computed tomography is useful for the diagnosis of lateral herniation of a lumbar disc, caution should be exercised when interpreting abnormal findings5,7,10,14,16. First, axial cuts should be obtained parallel to the corresponding disc space. Cuts that are not parallel to the disc space can produce an asymmetrical disc profile, which may show protrusion of the disc predominantly to one side. Because of the degenerative scoliosis of the lumbar spine in our patient, the computed tomography myelogram at the fifth lumbar-first sacral level may not have been made parallel to the corresponding disc space. Second, coexistent degenerative stenosis of the lumbar canal could influence the interpretation of lateral herniation of a lumbar disc7,14. Our patient also had multilevel degeneration of the disc and facet joint. Third, an enlarged or conjoined nerve root and enlargement of foraminal veins may mimic lateral herniation of the disc4. Grenier et al. noted that the interpretation of the foraminal area on computed tomography scans was sometimes difficult because of those findings4. Fourth, and most important, computed tomography scans of discs in asymptomatic individuals may demonstrate false-positive findings. Wiesel et al. evaluated computed tomography scans of the lumbosacral spine of seventy-two volunteers who had no history of pain in the low back or lower limbs; more than 35 per cent of these asymptomatic subjects were found to have abnormal findings on computed tomography scans on blinded review by radiologists. Discography-enhanced computed tomography and high-quality sagittal and coronal magnetic resonance imaging have been reported to be the most effective methods for the accurate identification of a lateral herniation of the lumbar disc, but even these methods do not always yield an accurate diagnosis4,5,7,14.
The herniated disc between the first and second lumbar vertebrae caused the monoradiculopathy of the fifth lumbar nerve root in our patient, as the anterior discectomy at this level relieved the sciatica and eliminated the limitation of straight-leg raising immediately after the operation. The herniated disc seemed to have compressed the left fifth lumbar nerve root only. The pathogenesis can be explained as follows. In our patient, the first and second lumbar level corresponded to the epiconus, from which the left fifth lumbar nerve root originated. At this level, the fifth lumbar nerve roots are located most medially to the epiconus15. In addition, the epiconus had already been shifted slightly to the right in the spinal canal because of the left-sided scoliosis. Consequently, the central herniation of the disc between the first and second lumbar vertebrae seemed to compress the left fifth lumbar nerve root directly at a point near its origin at the epiconus. Moreover, the left fifth lumbar nerve root may not have been mobile enough to avoid the compression by the herniated disc because of the constraint from its attachment to the epiconus.
We selected the anterior approach for the excision of the herniated disc at the first and second lumbar level, as this approach allows the protruded disc to be removed without undue manipulation of the spinal cord2. Anterior discectomy and arthrodesis provided a satisfactory result in our patient. We believe that the use of spinal instrumentation, such as the Kaneda device, can facilitate early walking, obviate the need for a postoperative cast, and improve the rate of fusion8,9,13.
Although partial facetectomy seemed to be enough to remove the laterally herniated portion of the disc7,10, a foraminotomy was performed to expose the fifth lumbar nerve root completely and to confirm the lesion in our patient. As a result, there was unilateral iatrogenic spondylolysis at the fifth lumbar level. Long-term follow-up is needed to observe whether instability at the lumbosacral junction occurs, although the left facet joint of the fifth lumbar and first sacral vertebrae was not damaged completely and the contralateral facet joint was kept intact.
The present report illustrates an unusual presentation of a herniated disc between the first and second lumbar vertebrae with a monoradiculopathy of the fifth lumbar nerve root. We emphasize the need for physicians to be aware of the possibility that a cephalad herniation of a lumbar disc can cause a monoradiculopathy at a caudad level.
NOTE: The authors thank Casey K. Lee, M.D., for his critical review of the manuscript.