The Journal of Bone and Joint Surgery 80:1515-20 (1998)
© 1998 The Journal of Bone and Joint Surgery, Inc.
Chronic Lumbar Epidural Hematoma in a Patient Who Had Spondylolysis at the Third Lumbar Vertebra. Report of a Rare Case Involving a Seventeen-Year-Old Adolescent*
KENSEI NAGATA, M.D. ,
MAMORU ARIYOSHI, M.D.,
KAZUMASA ISHIBASHI, M.D.,
SHOJI HASHIMOTO, M.D. and
AKIO INOUE, M.D., KURUME, JAPAN
Investigation performed at Kurume University, Kurume
 |
Introduction
|
|---|
A spinal epidural hematoma may result from one of many causes, including coagulopathy, trauma, a vascular lesion, iatrogenesis, and spontaneous occurrence2,4,6,7,11,12. Groen and van Alphen7 reviewed the cases of 333 patients who had a spontaneous spinal epidural hematoma; they had excluded patients in whom the hematoma had occurred after a traumatic episode involving spinal dislocation or fracture, after epidural anethesia, after diagnostic lumbar puncture, after an operation, or in association with a tumor in the spinal canal. Of the 333 patients, forty (12 per cent) had a lumbar epidural hematoma. Overall, fifty-five (17 per cent) of the patients were younger than twenty years old, and only four (7 per cent) of these young patients had a spontaneous lumbar epidural hematoma. Boyd and Pear reported that a hematoma caudad to the level of the conus medullaris was more likely to be chronic because the spinal roots appear to tolerate pressure better than the spinal cord does.
In a review of the literature, we found four cases of lumbar epidural hematoma in patients who were younger than twenty years old3,7,17,21, and we compared these cases with that of our patient. A chronic spinal epidural hematoma is rare in young patients; it occurred in only two of the patients, including ours. We report the case of a young patient who had a chronic lumbar epidural hematoma and spondylolysis at the third lumbar vertebra.
|
Case Report
|
|---|
The patient, a seventeen-year-old high-school student, had radicular pain in the left lower extremity when he was first seen by us in 1991. He was a rugby player (center prop position number three), and, in early December 1991, he had pain on the left side of the low back during rugby training; the pain was not associated with any specific injury. One week after the onset of this pain, he visited a local orthopaedic clinic because of persistent tingling in the left lower limb. Spondylolysis involving the third lumbar vertebra was diagnosed on the basis of radiographic examination. The patient was treated non-operatively with medication, pelvic traction, and thermotherapy. However, the symptoms worsened, and he was examined with magnetic resonance imaging, which revealed an epidural mass located posterolateral to the dura at the third lumbar level. He was referred to our outpatient clinic on January 17, 1992, and a tumor of the cauda equina was diagnosed.
The patient was admitted to the hospital on January 20, 1992, at which time he had pain on the left side of the low back, tingling in the left lower limb, and radicular pain. He had no history of coagulopathy, a previous lumbar operation, or an epidural puncture. The findings of the general physical examination were normal, with normal function of the bowel and bladder and normal laboratory findings. The patient could walk without limping, and the lumbar spine had normal alignment on visual examination. However, flexion of the spine was restricted, with a finger-to-floor distance of thirty centimeters. There were no abnormal cutaneous findings, such as a dimple, a hairy patch, or unusual pigmentation. The third lumbar spinous process and the left buttock were tender. Straight-leg raising produced radicular pain at 30 degrees of elevation of the left lower extremity. Hypoesthesia was present in the medial aspect of the left lower extremity. The deep tendon reflexes were normal in both knees and in the right ankle but were absent in the left ankle. The muscle strength of the lower extremities was evaluated with manual muscle-testing and was found to be normal.
Radiographs and computerized tomography scans showed spondylolysis at the third lumbar vertebra with normal spinal alignment (Fig. 1). Spina bifida was not found. An anteroposterior myelogram showed a large defect on the left between the third and fourth lumbar vertebrae, and a lateral myelogram showed a posterior epidural mass compressing the thecal sac at this level (Fig. 2). Coronal and sagittal magnetic resonance images, made with a 1.5-tesla system (Gyro Scan T5; Philips Medical Systems International B.V., Veenpluis, The Netherlands), showed a large oval mass, approximately twenty-five by fifteen millimeters in size, occupying the posterior epidural space between the third and fourth lumbar vertebrae. The signal intensity was high but non-homogeneous on T1-weighted images and was high with a low-intensity rim on T2-weighted images. An axial image showed an epidural mass that appeared to be continuous with the left vertebral foramen between the third and fourth lumbar vertebrae. Magnetic resonance images of the intervertebral disc between the third and fourth lumbar vertebrae showed a normal shape and a normal intensity (Figs. 3-A, 3-B, and 3-C).
View larger version (89K):
[in this window]
[in a new window]
|
FIG1: Fig. 1 Lateral and bilateral oblique radiographs made on admission showed spondylolysis (arrows) at the third lumbar vertebra with normal spinal alignment.
|
|
View larger version (161K):
[in this window]
[in a new window]
|
FIG2: Fig. 2 An anteroposterior myelogram revealed a large defect on the left between the third and fourth lumbar vertebrae, and a lateral myelogram showed a severe posterior indentation.
|
|
View larger version (111K):
[in this window]
[in a new window]
|
FIG3-A: Figs. 3-A Magnetic resonance images made with use of a 1.5-tesla system. Midline sagittal images showed a large oval mass in the posterior epidural space between the third and fourth lumbar vertebrae. The signal intensity was high but not homogeneous on the T1-weighted image (Fig. 3-A) and was high with a low-intensity rim on the T2-weighted image (Fig. 3-B). Imaging of the intervertebral disc between the third and fourth lumbar vertebrae showed a normal shape with normal intensity (Fig. 3-B). An axial image showed an epidural mass continuous with the left vertebral foramen between the third and fourth lumbar vertebrae (Fig. 3-C).
|
|
View larger version (106K):
[in this window]
[in a new window]
|
FIG3-B: Figs. 3-B Magnetic resonance images made with use of a 1.5-tesla system. Midline sagittal images showed a large oval mass in the posterior epidural space between the third and fourth lumbar vertebrae. The signal intensity was high but not homogeneous on the T1-weighted image (Fig. 3-A) and was high with a low-intensity rim on the T2-weighted image (Fig. 3-B). Imaging of the intervertebral disc between the third and fourth lumbar vertebrae showed a normal shape with normal intensity (Fig. 3-B). An axial image showed an epidural mass continuous with the left vertebral foramen between the third and fourth lumbar vertebrae (Fig. 3-C).
|
|
View larger version (93K):
[in this window]
[in a new window]
|
FIG3-C: Figs. 3-C Magnetic resonance images made with use of a 1.5-tesla system. Midline sagittal images showed a large oval mass in the posterior epidural space between the third and fourth lumbar vertebrae. The signal intensity was high but not homogeneous on the T1-weighted image (Fig. 3-A) and was high with a low-intensity rim on the T2-weighted image (Fig. 3-B). Imaging of the intervertebral disc between the third and fourth lumbar vertebrae showed a normal shape with normal intensity (Fig. 3-B). An axial image showed an epidural mass continuous with the left vertebral foramen between the third and fourth lumbar vertebrae (Fig. 3-C).
|
|
An operation was performed on January 30, 1992. The unstable third lumbar lamina was removed from the pars interarticularis, and a partial laminectomy was performed at the fourth lumbar vertebra. After removal of the ligamentum flavum, the posterior epidural mass was visualized. The mass severely compressed the dural sac centrally to the left lateral edge in the epidural space. The mass adhered strongly to the dura and both the third and the fourth lumbar nerve root. The fourth lumbar nerve root was compressed more severely than was the third. The mass was completely extirpated under microscopic magnification. The capsule of the mass was hard and elastic and was filled with a dark red-gray solid and liquid material. No vascular abnormality was seen in the epidural space. The third lumbar lamina was replaced with bone grafts in the pars interarticularis bilaterally with use of a modified Scott wiring method15.
Examination of histological sections of the mass revealed organizing hemorrhage with infiltration by surrounding fibrous granulation tissue. There was no evidence of neoplasm or infection (Fig. 4).
View larger version (148K):
[in this window]
[in a new window]
|
FIG4: Fig. 4 Examination of histological sections of the mass showed an organizing hematoma (bottom) with infiltration by surrounding fibrous granulation tissue (x 90).
|
|
The pain in the left lower limb was relieved immediately after the operation. The patient returned to playing rugby six months later. At one year after the resection, bilaterial osseous union at the pars interarticularis was confirmed radiographically and the screws and wires were removed. At six years, a slight narrowing of the intervertebral space between the third and fourth lumbar vertebrae was detected during a follow-up radiographic examination. (Fig. 5). At the time, the patient was working as an engineer and had no low-back pain or neurological deficits.
View larger version (108K):
[in this window]
[in a new window]
|
FIG5: Fig. 5 Radiographs, made six years after the operation, showed osseous union at the pars interarticularis at the third lumbar level (arrows) and slight narrowing of the intervertebral space between the third and fourth lumbar vertebrae.
|
|
|
Discussion
|
|---|
When we find a mass in the spinal canal on a magnetic resonance image of an adolescent, we confirm the mass with use of another imaging modality. The differential diagnosis includes a neutral tumor, nueral cyst, dermoid, granulation, lipoma caused by spina bifida, migration of a disc herniation, and epidural hematoma. In our patient, the epidural mass was located posterolateral to the dura and was confirmed with myelography (Fig. 2). Magnetic resonance imaging is generally used to diagnose an epidural hematoma. The signal intensity of a hemorrhage on magnetic resonance imaging changes over time after the onset according to the oxidation and deoxidation of hemoglobin, the hemolysis of erythrocytes, and phagocytosis5,19 Our patient had a high-intensity mass with isointensity inside the mass on T1-weighted images and high intensity on T2-weighted images (Figs. 3-A, 3-B, and 3-C). These findings suggested an old hemorrhage. A lipoma also shows a high-intensity signal on both T1 and T2-weighted images. However, the high-intensity mass on the magnetic resonance images of our patient was located inside the spinal canal and was not in continuity with the epidermis. Therefore, the mass was not a lipoma. The diagnosis of a chronic epidural hematoma was confirmed with both operative and histological findings.
A spinal epidural hematoma is rare in young patients. As stated, of the 333 spontaneous epidural hematomas in the study by Groen and van Alphen7, only fifty-five (17 per cent) were in patients who were younger than twenty years old. Twenty-eight (51 per cent) of these fifty-five hematomas were found to be at the cervical level, twenty-three (42 per cent) were at the thoracic level, and only four (7 per cent) were at the lumbosacral level. We found and reviewed seventy case of spontaneous lumbar epidural hematoma in the literature4,6-8,10,11,13,14,16 and compared them with the case of our patient. Of these hematomas, only five (7 per cent) (including the one in our case report) were in patients who were younger than twenty years old (range, eleven to seventeen years old). The cephalad level of the hematoma was the third lumbar vertebra in four of these patients and the fourth lumbar vertebra in one (Table I).
View this table:
[in this window]
[in a new window]
|
TABLE I
REPORTS IN THE LITERATURE OF SPONTANEOUS LUMBAR EPIDURAL HEMATOMA IN PATIENTS YOUNGER THAN TWENTY YEARS OLD*
|
|
The cause of epidural hematoma in the lumbar spine is not clear. Gundry and Heithoff reported eighteen cases of spontaneous epidural hematoma in the lumbar spine that had been diagnosed with computerized tomography and magnetic resonance imaging. All of those hematomas were classified as type three according to the classification system of Hanley et al. This type of hematoma is submembranous and is contained in the epidural space; it occurs as a result of hemorrhage from the epidural venous system that collects beneath a fibrous membrane on the floor of the spinal canal. Gundry and Heithoff reported an association between the occurrence of an epidural hematoma and an underlying disc disruption. This led them to postulate that spontaneous spinal epidural hematoma resulted from tearing of the fragile epidural veins adjacent to a displaced annulus fibrosus or nucleus pulposus. Toyoma et al. described the concept of discal cyst on the basis of their experience. According to those authors, a discal cyst is associated with slight degeneration of the disc at the affected level, with severe radicular pain. Discography demonstrates a flow of contrast medium into the cyst. The cyst wall consists of fibrous tissue, and the fluid in the cyst is serosanguineous. The spontaneous epidural hematomas in the lumbar spine reported by Gundry and Heithoff appeared to be associated with annular disruption or small disc protrusions, or both. In contrast, in a report on 325 spontaneous spinal epidural hematomas by Groen and Ponssen6, almost all (99 per cent) of the hematomas were situated posteriorly in the epidural space. A posterior epidural hematoma in the lumbar spine appears to result from weakness of the walls of the radicular veins or the epidural veins without disc disease. This distinction between an anterior and a posterior lumbar epidural hematoma can allow differentiation between a disc disturbance and a vessel disturbance as the cause.
Acute spinal epidural hemorrhage frequently resolves spontaneously9-11,22,23. Therefore, non-operative treatment should be chosen whenever a patient demonstrates any tendency toward recovery. However, when a patient has progressive, severe neurological symptoms, an early decompression operation with evacuation of the hematoma should be performed. Operative treatment of a spinal epidural hematoma should be elected on the basis of an accurate diagnosis of the level of the hematoma with magnetic resonance imaging and a clinical evaluation of the neurological course of the patient.
Katsube et al. also reported the case of a patient who had an epidural hematoma associated with spondylolysis. However, the spondylolysis was not at the level of the hematoma. Spondylolysis is a common condition and usually occurs at the fifth lumbar vertebra; it is caused by a fatigue fracture in patients who have a lower lumbar index (trapezoidal lumbar vertebra)18. The cause of spondylolysis at a more caudad level in the lumbar spine is considered to be related to a history of trauma18. Waguri et al. reported the spontaneous resolution (as confirmed with magnetic resonance images) of a spinal epidural hematoma from the first lumbar vertebra to the sacrum that had occurred in a twenty-five-year-old man after a game of rugby. However, they did not discuss the cause of the hematoma. Our operative findings of no callus formation or hemorrhage from the lamina suggested that the spondylolysis was not acute in our patient. Also, we found no clinical evidence of severe trauma, and there was no acute low-back pain in the lumbar spine. Chronic stress in the lumbar spine could have developed over time, as the patient was a rugby player. We suspect that the hemorrhage occurred from the epidural veins under the pars interarticularis at the third lumbar level because of recurrent minor traumatic episodes during rugby training.
|
Footnotes
|
|---|
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
Department of Orthopaedic Surgery, Kurume University School of Medicine, 67 Asahi Machi, Kurume 830-0011, Japan. E-mail address: spine@med.kurume-u.ac.jp.
|
References
|
|---|
-
Boyd, H. R., and Pear, B. L.: Chronic spontaneous spinal epidural hematoma. Report of two cases. J. Neurosurg., 36: 239-242, 1972.[Medline]
-
Dauch, W. A.: Spinale Epiduralhämatome bei Kindern und Jugendlichen. Neurochirurgia, 29: 83-89, 1986.[Medline]
-
Dawson, B. H.: Paraplegia due to spinal epidural hematoma. J. Neurol., Neurosurg., and Psychiat., 26: 171-173, 1963.
-
Foo, D., and Rossier, A. B.: Preoperative neurological status in predicting surgical outcome of spinal epidural hematomas. Surg. Neurol., 15: 389-401, 1981.[Medline]
-
Gomori, J. M.; Grossman, R. I.; Goldberg, H. I.; Zimmerman, R. A.; and Bilaniuk, L. T.: Intracranial hematomas: imaging by high-field MR. Radiology, 157: 87-93, 1985.[Abstract/Free Full Text]
-
Groen, R. J., and Ponssen, H.: The spontaneous spinal epidural hematoma. A study of the etiology. Neurol. Sci., 98: 121-138, 1990.
-
Groen, R. J., and van Alphen, H. A.: Operative treatment of spontaneous spinal epidural hematomas: a study of the factors determining postoperative outcome. Neurosurgery, 39: 494-509, 1996.[Medline]
-
Gundry, C. R., and Heithoff, K. B.: Epidural hematoma of the lumbar spine: 18 surgically confirmed cases. Radiology, 187: 427-431, 1993.[Abstract/Free Full Text]
-
Hanley, E. N., Jr.; Howard, B. H.; Brigham, C. D.; Chapman, T. M.; Guilford, W. B.; and Coumas, J. M.: Lumbar epidural varix as a cause of radiculopathy. Spine, 19: 2122-2126, 1994.[Medline]
-
Katsube, H.; Ri, K.; Shibata, T.; Nakajima, H.; and Nishitou, S.: [Acute lumbar epidural hematoma with spontaneous resolution.]. Rinsho Seikeigeka, 28: 625-629, 1993.
-
Klossek, H., and Hüller, E.: Zur Problematik der spontanen spinalen Epiduralhämatoma. Zentralbl. Neurochir., 45: 116-123, 1984.[Medline]
-
Lawton, M. T.; Porter, R. W.; Heiserman, J. E.; Jacobowitz, R.; Sonntag, V. K.; and Dickman, C. A.: Surgical management of spinal epidural hematoma: relationship between surgical timing and neurological outcome. J. Neurosurg., 83: 1-7, 1995.[Medline]
-
Lunardi, P.; Mastronardi, L.; Lo Bianco, F.; Schettini, G.; and Puzzilli, F.: Chronic spontaneous spinal epidural hematoma simulating a lumbar stenosis. European Spine J., 4: 64-66, 1995.[Medline]
-
Nehls, D. G.; Shetter, A. G.; Hodak, J. A.; and Waggener, J. D.: Chronic spinal epidural hematoma presenting as lumbar stenosis: clinical, myelographic, and computed tomographic features. A case report. Neurosurgery, 14: 230-233, 1984.[Medline]
-
Nicol, R. O., and Scott, J. H.: Lytic spondylolysis. Repair by wiring. Spine, 11: 1027-1030, 1986.[Medline]
-
Pear, B. L.: Spinal epidural hematoma. Am. J. Roentgenol., 115: 155-164, 1972.[Abstract]
-
Rao, B. D.; Rao, K. S.; Subrahmanian, M. V.; and Reddy, M. V. R.: Spinal epidural haemorrhage. British J. Surg., 53: 649-650, 1966.
-
Saraste, H.: The etiology of spondylolysis. A retrospective radiographic study. Acta Orthop. Scandinavica, 56: 253-255, 1985.[Medline]
-
Thulborn, K. R., and Atlas, S. W.: Intracranial hemorrhage. In Magnetic Resonance Imaging of the Brain and Spine, edited by S. W. Atlas. Ed. 2, pp. 265-314. New York, Lippincott-Raven, 1996.
-
Toyama, Y.; Kamata, M.; Matumoto, M.; Nishizawa, T.; Koyanaji, T.; Suzuki, N.; and Fujimura, Y.: [Pathogenesis and diagnostic title of intraspinal cyst communicating with intervertebral disk in the lumbar spine.]. Rinsho Seikeigeka, 32: 393-400, 1997.
-
Tsai, F.Y.; Popp, A.J.; and Waldman, J.: Spontaneous spinal epidural hematoma. Neuroradiology, 10: 15-30, 1975.
-
Wagner, S.; Forsting, M.; and Hacke, W.: Spontaneous resolution of a large spinal epidural hematoma: case report. Neurosurgery, 38: 816-818, 1996.[Medline]
-
Waguri, Y.; Takahashi, M.; Horie, Y.; Hara, M.; and Mizutani, Y.: Acute spinal epidural hematoma with spontaneous resolution. Seikeigeka, 42: 1750-1752, 1991.
 CiteULike  Connotea  Del.icio.us  Technorati What's this?
|
Top
Introduction
Case Report